[From the U.S. Government Printing Office, www.gpo.gov]
X, 6L ,pp 7" "w '41 T" F @, A A,R,I',N,E,',,,,@,,,,I@N'7r,'E@R,ESTS" 'RT A PA NE S -8 -CH2585 vyc VOL 4 DINGS A /'gir p'; wo- 4- W tt a F '1,001* 1A :0,2 5 V' -.4 -7 COASTAL ZONE INFORMATION CENTER OCEANS CON FERENC_E_AoT7p_os)1T1ON PRESENTED BY MTS-OES-IEEE IN COOPERATION WITH THE PORT OF BALTIMORE B@LTIMORE CONVENTION CENTER, BALTIMORE, MARYLAND OCTOBER 31-NOVEMBER 2,1988 HONORARY CHAIRMAN DONALD SCHAEFER, GOVERNOR OF MARYLAND, AfMIRAL PAUL A. YOST, COMMANDANT UNITED STATES COAST GUARD, GENERAL CHAIRMAN OCEANS 88 A Partnership of Marine Interests Property of CSC Library PROCEEDINGS Conference Sponsored by Marine Technology Society IEEE Baltimore, Maryland October 31-November 2, 1988 U.S. DEPARTMENT OF COMMERCE NOAA COASTAL SERVICES CENTER 2234 SOUTH HOBSON AVENUE CHARLESTON , SC 29405-2413 IEEE Catalog Number 88-CH2585-8 Oceans '88 Proceedings Volume 1: Pages 1 to 2 74 Volume 2: Pages 2 75 to 718 Volume 3: Pages 719 to 1086 Volume 4: Pages 1087 to 1732 Copies of the Oceans '88 Proceedings are available from: The IEEE Service Center ". 445,Ffoes Lane Pis@cataway, NJ. 08854 and The Marine Technology Society 2000 Florida Avenue, N.W., Suite 500 Washington, D.C. 20006 Copyright and Reprint Permissions: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limits of U.S. copyright law for private use of patrons those articles in this collection that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through the Copyright Clearance Center, 29 Congress Street, Salem, Mass. 01970. Instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. For other copying, reprint or re-publication permission, write to Director, Publishing Services, IEEE, 345 E. 47th Street, New York, N.Y. 10017. All rights reserved. Copyright 1988 by The Institute of Electrical and Electronics Engineers. IEEE Catalog Number 88-CH2585-8 OCEANS )88 Proceedings Volume Four Table of Contents PLASTICS IN OUR OCEANS: WHAT ARE WE DOING MESOCOSMS AS TOOLS FOR COASTAL AND ESTUARINE ABOUTIT? ENVIRONMENTAL RESEARCH-1 Chairman: Chairman: B. Griswold G. F. Mayer OAR, National Oceanic and Atmospheric National Oceanic and Atmospheric Administration Administration E. Klos 1529 J. M. Coe and A. R. Bunn 1 An Experimental Estuarine Salinity Gradient Marine Debris and the Solid Waste Disposal Crisis S. W. Nixon and S. W. Granger 1604 D. Cottingham 6 Development of Experimental Ecosystems for the Federal Programs and Plastics in the Oceans Study of Coastal Lagoons X. Augerot 1711 J. G. Sanders and G. F. Riedel 23 Sea Grant Faces Oceans of Plastic The Use of Enclosed Ecosystems for the Study of Cyclingand Impact of Trace Elements K. J. O'Hara 12 Education and Awareness: Keys to Solving the Marine S. J. Cibik, J. G. Sanders and C. F. D'Elia 29 Debris Problem Interactions Between Insolation and Nutrient Loading and the Response of Estuarine Phytopla@nkton J. R. Whitehead 1507 Reducing Plastic Pollution in the Marine Environment: MESOCOSMS AS TOOLS FOR COASTAL AND ESTUARINE The U.S. Coast Guard and Implementation of Annex V ENVIRONMENTAL RESEARCH-H of MARPOL 73178 Chairman: CONTINENTAL SHELF ENVIRONMENTAL RESEARCH R. E. Turner Center for Wetlands Resources, Chairman: Louisiana State University W. W. Schroeder University of Alabama A. G. Chalmers 1652 Experimental Manipulations of Drainage in a Georgia R. Rezak and D. W. McGrail 1602 Saltmarsh: Lessons Learned Geology and Hydrology of Reefs and Banks Offshore Texas and Louisiana M. R. DeVoe, M. E. Tompkins and J. M. Dean 35 South Carolina's Coastal Wetland Impoundment W. W. Schroeder, M. R. Dardeau, J. J. Dindo, P. Project (CWIP): Relationship of Large-Scale Research Fleischer, K. L. Heck, Jr. and A. W , Schultz 17 to Policy and Management Geological and Biological Aspects of Hardbottom Environments on the LMAFLA Shelf, Northern Gulf of R. E. Turner 41 Mexico Experimental Marsh Management Systems in Louisiana W. W. Schroeder, R. Rezak and T. J. Bright 22 A. G. van der Valk, B. D. J. Batt, H. R. Murkin, P. J. Video Documentation of Hardbottom Environments Caldwell and J. A. Kadlec 46 The Marsh Ecology Research Program (MERP): The Organization and Administration of a Long-Term Mesocosm Study iv. TECHNICAL ADVANCES IN SEAFOOD TECHNOLOGY AND WATER REUSE ON ONSHORE MARICULTURE AND SAFETY PROCESSING FACILITIES Chairman: Chairman: D. Attaway R. Becker Sea Grant, National Oceanic and Atmospheric Louisiana State University Administration J. M. Fox and A. L. Chauvin 1536 R. R. Colwell 1606 Depuration of Oysters in a Closed Recirculating New Approaches for IndiceslMonitoring Microbial System Pathogens in Seafood M. P. Thomasson, D. G. Burden and R. F. Malone 70 J. Liston 52 Micro-Computer Based Design of Recirculating Microorganisms as a Cause of Economic Loss to the Systems for the Production of Soft-shell Blue Crabs Seafood Industry (Callinectes sapidus) G. J. Flick, Jr. 56 G. E. Kaiser and F. W. Wheaton 76 Sea Grant Advances in Seafood Science and Computerized Rapid Measurement of Ammonia Technology Concentration in Aquaculture Systems S. Garrett and M. Meyburn K. Rausch, W. H. Zachritz II, T. C. T. Y-Hsieh and Development of New Approaches to Seafood R. F. Malone 84 Inspection Use of Automated Holding Systems for Initial Off- Flavor Purging of the Rangia Clam, Rangia cuneata TECHNICAL ADVANCES IN SEAFOOD TECHNOLOGY AND SAFETY GULF OF MEXICO CHEMOSYNTHETIC PETROLEUM SEEP COMMUNITIES Chairman: G. J. Flick, Jr. Chairman: Virginia Polytechnic Institute R. Carney Louisiana State University R. C. Lindsay 61 Flavor Chemistry and Seafood Quality Factors 1. MacDonald, R. Carney and D. Wilkinson 90 Gulf of Mexico Chemosynthetic Communities at Oil H. G. Hultin 66 Seeps: Estimating Total Density Technical Problems and Opportunities Related to Utilization of Out Seafood Resources R. S. Carney 96 Emerging Issues of Environmental Impact to Deep-Sc2 J. P. Zikakis 1608 Chemosynthetic Petroleum Seep Communities A Biotechnological System for the Utilization of Waste Products of the Seafood and Cheese Mqnufacturing H. H. Roberts, R. Sassen and P. Aharon 101 Industries Petroleum -Derived Authigenic Carbonates of the Louisiana Continental Slope A. P. Bimbo 1513 The Production of Menhaden Surimi UNDERSEA VEHICLES AND PLATFORMS FOR SCIENCE OIL AND GAS INDUSTRIES CONFLICT APPLICATIONS Chairmen: Chairman: R. W. Middleton A. N. Kalvaitis Minerals Management Service National Undersea Research Program, M. Holliday National Oceanic and Atmospheric Administration National Marine Fisheries Service G. A. Smith and R. S. Rounds 106 J. Brashier 136 Scientific, Technological and Social Impact of NOAA's Coexistence of Fishing and Oil and Gas Industries in Mobile Undersea Research Habitat the Gulf of Mexico P. J. Auster, L. L. Stewart and H. Sprunk 1286 B. R, Clark 143 Scientific Imaging Problems and Solutions for ROVs Potential Conflicts Between Oil and Gas Industry L. L. Stewart and P. J. Auster 1610 Activities and Commercial Fishing Low Cost ROVs for Science R. M. Meyer 146 R. A. Cooper and 1. G. Babb 112 Information on Fisheries Risk Assessment in the Alaska Manned Submersibles Support a Wide Range of OCS Region Underwater Research in New England and the Great R. C, Wingert 150 Lakes Geophysical Survey and Commercial Fishing Conflicts, R. 1. Wicklund and B. L. Olla 119 Environmental Studies and Conflict Mitigation in the Field Research Programs at the Caribbean Marine Minerals Management Service Pacific OCS Region Research Center-National Undersea Research Program A. S. Knaster 156 The Use of Alternative Dispute Resolution in OCS FISHERIES AND RESOURCES Resolving Outer Continental Shelf Disputes Chairmen: CUMULATIVE ENVIRONMENTAL EFFECTS OF THE OIL AND R. W. Middleton GAS LEASING PROGRAM-I Minerals Management Service M. Holliday Chairmen: National Marine Fisheries Service J. Goll Minerals Management Service R. W. Middleton 123 J. M. Teal Oil and Gas Industry Conflicts on the Outer Woods Hole Oceanographic Institute Continental Shelf D.IChristensen 1624 D. V. Aurand 161 Outer Continental Shelf Fisheries and Resources in the The Future of the Department of the Interior OCS Northeast Region Studies Program R. J. Essig 127 T. Chico 166 Outer Continental Shelf Fishery Resources of the Air Quality Issues, Environmental Studies, and South Atldntic Cumulative Impacts in the Pacific OCS Region B. G. Thompson 1613 R. E. Miller 172 Outer Continental Shelf Fisheries and Resources in the Georges Bank Monitoring Program: A Summary Gulf of Mexico J. M. Teal 177 S. Koplin 132 The Role of the Scientific Advisory Committee, Outer The Outer Continental Shelf Fishery Resources of the Continental Shelf Program of Minerals Management Pacific Coast Service vi. CUMULATIVE ENVIRONMENTAL EFFECTS OF THE OIL AND OIL AND GAS EXPLORATION-H GAS LEASING PROGRAM-H Chairmen: Chairmen: J. R. Pearcy J. Goll Minerals Management Service Minerals Management Service C. Welling J. M. Teal Ocean Minerals Co. Woods Hole Oceanographic Institute C. A. Dunkel 208 R. M. Rogers 953 A Qualitative Assessment of the Hydrocarbon Potential Factors Contributing to Wetland Loss in the Coastal of the Washington and Oregon Continental Shelf Central Gulf of Mexico J.'M. Galloway and M. R. Brickey 1611 S. D. Treacy 180 The Hydrocarbon Potential of the Federal OCS, The Minerals Management Service Bowhead Whale Offshore Northern California Monitoring Program and Its Applications J. Kennedy and C. Grant 213 R. B. Clark 184 Impact of the Oil-bearing Monterey Formation on Impact of Offshore Oil Operations in the North Sea Undiscovered Resources of Offshore California J. P. Zippin 1615 S. Sorenson, C. Alonzo and M. Ibrahim 1612 Cumulative Environmental Effects of the Department Wilson Rock Field: A Case History of the Interior's Offshore Oil and Gas Program: 1987 Report to Congress OIL AND GAS RESOURCE MANAGEMENT OIL AND GAS EXPLORATION-1 Chairmen: R. V. Amato Chairman: Minerals Management Service J. R. Pearcy C. Welling Minerals Management Service Ocean Minerals Co. F. R. Keer 188 G. M. Edson 219 Geologic Characteristics of an Atlantic OCS Gas The Ancient Atlantic Reef Trend Discovery and Its Implications P. K. Rav 193 B. J. Bascle 223 Hydroca'rbon Potential of the Deepwater (600 Feet) The Effect of Exploration on Resource Estimates for Gulf of Mexico the Alaska Outer Continental Shelf W. E. Sweet and J. C. Reed 202 D. Mayerson 229 Correlation of Cenozoic Sediments-Gulf of Mexico Pr6-lease Geophysical Permitting for the Pacific OCS: Outer Continental Shelf Procedures, Problems, and Solutions D. A. Steffy 235 Post-Lease Sale Exploration of the Nivarin Basin, Bering Sea, Alaska vii. OFFSHORE DRILLING-ENVIRONMENTAL STUDIES ACOUSTIC APPLICATIONS-11 Chairman: Chairman: D. Cottingham A. I. Eller National Oceanic and Atmospheric Administration Science Applications International Corp. D. K. Fran@ois 241 L. C. Haines, W. W. Renner and A. 1. Eller 295 Environmental Studies and Impact Assessment on the Prediction System for Acoustic Returns from Ocean Atlantic Outer Continental Shelf Bathymetry R. B. Krahl and C. E. Smith 250 G. P. Vellemarette 298 Developing Technologies for Offshore Oil and Gas Programmable Subsurface Acoustic Recording System Structures in Frontier and Hazardous Areas D. F. McCammon 304 OCEAN LEASING AND DEVELOPMENT The Relationship Between Acoustic Bottom Loss and the Geo2coustic Properties of the Sediment Chairman: G. Pettrazzulo ACOUSTICS-NOISE Technical Resources Inc. Chairmen: S. Ashmore 259 D. J. Ramsdale Offshore Leasing Boundaries Along the Receding Naval Ocean R&D Activity Alaskan Coastline N. Miller T. J. Mac Gillvray 262 West Sound Association Development and Analysis of DCF Computer Models W. S. Hodgkiss 310 for EEZ Marine Mining Source Ship Contamination Removal in a Broadband M. E. Dunaway and P. Schroeder 268 Vertical Array Experiment Minimizing Anchoring Impacts During Construction of R. J. Lataitis, G. B. Crawford and S. F. Clifford 315 Offshore Oil and Gas Facilities A New Acoustic Technique for Remote Measurement ACOUSTIC APPLICATIONS-1 of the Temporal Ocean Wave Spectrum Chairman: ACOUSTICS-PROPAGATION A. 1. Eller Science Applications International Corp. Chairman: D. G. Browning W. Hill, G. Chaplin and D. Nergaard 275 Naval Underwater System Center Deep-Ocean Tests of an Acoustic Modem Insensitive to Multipath Distortion D. G. Browning, P. M. Schiefele and R. H. Mellen 318 Attenvadbn of Low Frequency Sound in Ocean A. Novick 1617 Surface Ducts: Implications for Surface Loss Values A Shallow Water Sonar Performance Prediction System W. J. Vetter 1540 On Ray Trajectories and Pithtimes for Acoustic R. L. Spooner 283 Propagation in a Medium with Velocity Gradients Signal Processing Using Spreadsheet Software D. K. Roderick 1619 J. M. Tattersall, J. A. Mingrone and P. C. King 1618 An Introduction to the Physics of Underwater Sound A VCR Based Digita] Data Recorder for Underwater and Their Application to Passive Anti-Submarine Acoustics Multipath Measurements Warfare L. Wu and A. Zielinski 287 Multipath Rejection Using Narrow Beam Acoustic Link viii. ACOUSTICS-SIDE SCAN SEA BOTTOM PROPERTIES Chairman: Chairman: R. Walker M. Cruckshank USCG R&D Center University of Hawaii A. St. C. Wright 323 R. B. Perry 366 The Wide Swath, Deep Towed SeaMARC Mapping the Slopes of Expanding Continental Margins R. G. Asplin and C. G. Christensson 329 C. de Moustier, T. Hylas and J. C. Phillips 372 A New Generation Side Scan Sonar Modifications and Improvements to the Sea Beam System On Board RIV Thomas Washington E. Kristof, A. Chandler and D. Schomette 335 Using a Sector-Scan Sonar to Hunt for Shipwrecks D. E. Pryor 379 Through Ice Theory and Test of B2thymetric Side Scan Sonar J. W. Nicholson and J. S. Jaffe 338 S. M. Smith, J. S. Charters and J. M. Moore 385 Side Scan Sonar Acoustic Variability Processingand Management of Underway Marine Geophysical Data at Scripps R. Gandy and S. Paulet 1620 Realtime Side Scan Sonar Target Analysis R. L. Cloet 1636 Implications of Using a Wide SWATH Sounding W. R. Abrams 344 System A Practical High Tech Advance in Side Scan Sonar Target Positioning and Analysis SEDIMENT STUDIES-[ ACOUSTIC DOPPLER CURRENT PROFILING Chairman: A. G. Young Chairman: FUGRO-McClelland H. R. Frey Office of Oceanography and Marine Assessments, S. K. Breeding and D. Lavoie 391 National Oceanic and Atmospheric Administration Duomorph Sensing for Laboratory Measurement of Shear Modulus G. F. Appell, J. Gast, G. Williams and P. D. Bass 346 Calibration of Acoustic Doppler Current Profflers D. Lavoie, E. Mozley, R. Corwin, D. Lambert and P. Valent 397 Y. Kuroda, G. Kai and K. Okunc, 353 The Use of a Towed, Direct-Current, Electrical Development of a Shipboard Acoustic Doppler Resistivity Array for the Classification of Marine Current Proffler Sediments D. Wilson, D. Bitterman and C. Roffer 359 P. F. Wainwright, B. Humphrey and G. Stewart 405 The Acoustic Doppler Current Profiling System at Sediment Contamination by Heavy Metals and AOML Hydrocarbons ix. SEDIMENT STUDIES-11 SATELLITE REMOTE SENSING Chairman: Chairmen: H. G. Herrmanrf III D. E. Weissman Naval Facilities Engineering Command Hofstra University J. Gallagher A. E. Hay, L. Huang, E. B. Colbourne, J. Sheng and Naval Underwater Systems Center A. J. Bowen 413 A High Speed Multi-Channel Data Acquisition System I M. R. Willard 1625 for Remote Acoustic Sediment Transport Studies Ocean Sensing Capabilities on Landsat 6 D. G. Hazen, A. E. Hay and A. J. Bowen 419 S. W. McCandless, Jr. and J. Curlander 479 Design Considerations for RASTRAN-System 2 The Influence of Packing Technologies on A. G. Young, L. V. Babb and R. L. Boggess 423 Environmental Application of Space-Based Synthetic Mini-Probes: A New Dimension in Offshore In Situ Aperture Radar Testing J. R. Benada, D. T. Cuddy and B. H. jai 473 K. L. Williams and L. J. Satkowiak 428 Adapting the NSCAT Data System to Changing Bounded Beam Transmission Across a WaterlSand Requirements Interface, Experiment and Theory W. B. Campbell and M. L. Weaks 1626 An Inexpensive Interactive Processing System for L. J. Satkowiak 433 NOAA Satellite Images Remote Sea Bottom Classification Utilizing the Ulvertech Bottom Profiler Parametric Source D. S. Bryant, A. M. Ponsford and S. K. Srivastava 485 A Computer Package for the Parameter Optimization THE GREAT LAKES AS AN OCEANIC MICROCOSM of Groundwave Radar Chairman: L. Pittman Merchant Marine and Fisheries Committee, U.S. Congress J. R. Krezoski 437 Particle Reworking in Great Lakes Sediments: In-Situ Tracer Studies Using Rare Earth Elements J. R. Krezoski 442 In-Situ Tracer Studies of Surficial Sediment Transport in the Great Lakes Using a Manned Submersible L. F. Boyer 443 Video-Sedimcnt-Profile Camera Imagery in Marine and Freshwater Benthic Environments L. F. Boyer, R. J. Diaz and J. D. Hedrick 448 Computer Image-Analysis Techniques and Video- Sediment-Profile Camera Enhancements Provide 2 Unique and Quantitative View of Life at or Beneath the Sediment-Waterface Interface X. OCEAN APPLICATIONS OF REMOTELY SENSED MICROWAVE WATER COLUMN MEASUREMENTS-1 TECHNIQUES Chairmen: Chairmen: R. S. Mesecar D. E. Weissman Oregon State University Hofstra University T. M. Dauphinee J. Gallagher National Research Council, Canada Naval Underwater Systems Center W. Kroebel 491 C. Bostater and V. Klemas 462 Results of Exact Investigations About the Remote Sensing of Physical and Biological Properties Characteristics of the Extremely Fast and Accurately of Estuaries Measuring Kiel Multisonde and Representations About Its Newest Performance D. E. Weissman 1546 The Dependence of the Microwave Radar Cross K.-H. Mahrt and C. Waldmann 497 Section on Ocean Surface Variables During the Field Proven High Speed Micro Optical Density Fasinex Experiment Proffler Sampling 1000 Times Per Second with 10- Precision W.-M. Boerner, A. B. Kostinski, B. D. James and M. Walther 454 R. Mesecar and C. Moser 505 Application of the Polarimetric Matched Image Filter Multi-Sample Particle Flux Collector (PMIF) Technique to Clutter Removal in POL-SAR Images of the Ocean Environment J. M. Moore, C. de Moustier and J. S. Charters 509 Multi-Sensor Real-Time Data Acquisition and D. L. Murphy 467 Preprocessing at Sea Radar Detection of Oceanic Fronts L. S. Fedor and E. J. Walsh 1697 WATER COLUMN MEASUREMENTS-H Interpretation of SEASAT Radar Altimeter Returns Chairmen: from an Ovefflight of Ice in the Beaufort Sea J. Jaeger L. S. Fedor, G. S. Hayne and E. J. Walsh 1704 Honeywell Hydro Products Airborne Pulse-Limited Radar Altimeter Return K. Hill Waveform Characteristics over Ice in the Beaufort Sea Honeywell Hydro Products J. Wagner and R. Mesecar 518 A Common XBTIPersonal Computer Interface D. I. Nebert, H. Saklad and G. Mimken 1627 CTD Data Acquisition Package H. Tremblay 522 Hydroball-A New Expendable: Uses and Issues Xi. COMMUNICATIONS OCEAN ENGINEERING-1 Chairman: Chairmen: R. A. Buddenberg, USCG C. A. Kohler Office of Command and Control USCG R&D Center R. A. Buddenberg and A. Givens 526 R. Geminder Shipboard Tactical Computer: The Coast Guard's Mechanic Research Inc. Combat Information Center Modernization R. L. Benedict 577 R. L. Moe 532 Destruction of Offshore Platforms by Accelerated Networking and Ship-to-Shore Ship-to-Ship Galvanic Corrosion Communication C. A. Kohler 582 S. C. Hall 537 Corrosive-Wear of Buoy Chain The Defense Mapping Agency's Navigation J. Larsen-Basse, B. E. Liebert, K. M. Htun and Information Network A. Tadjvar 1628 Long-Term Abrasion and Corrosion Damage to the COLD REGIONS OPERATIONS Hawaii Deep Water Power Cable Chairmen: M. Briere, K. C. Baldwin and M. R. Swift 588 8. Smith Collision Tolerant Pile Structures: Design Analysis U.S. Coast Guard Software E. Early T. Dowd 595 University of Washington United States Naval Experience with Antifouling Paints J. D. Crowley 543 Cold Weather Effects upon Marine Operations OCEAN ENGINEERING-11 S. M. Smith and D. Strahl 549 Chairmen: Articulated Lights in Ice J. R. Vadus Office of Oceanography and Marine Assessments, M. Gorveatt and M. C. Yee 555 National Oceanic and Atmospheric Administration Arctic Ice Island Coring Facility K. Okamura Special Assistant to the Minister of Science and COLD REGIONS MEASUREMENTS Technology, Japan Chairmen: A. Bertaux 598 S. R. Osmer Tapered Interface in Harsh Environment Connectors USCG International Ice Patrol W. E. Hanson J. F. Legrand, A. Echardour, L. Floury, H. Floch, J. USCG International Ice Patrol Kerdoncuff, T. Le Moign, G. Loaec and Y. Raer 602 Nadia: Wireline Re-Entry in Deep Sea Boreholes W. E. Hanson 561 P. K. Sullivan and B. E. Liebert 606 Operational Iceberg Forecasting Concerns Impedance Measurements of Biofouling in Seawater G. Steeves and S. Grant 567 Condensers: An Update An Autonomous Atmospheric Pressure Recorder for E. A. Fisher and H. P. Hackett 607 Establishing Polar Sea Surface Height World's First Rigid Free-Standing Production Riser T. K. Newbury and A. J. Adams 573 F. El-Hawary 291 Estimated Ice-Gouge Rates on a Manmade Shoal in the Compensation of Vertical Displacement Components Beaufort Sea in Marine Seismic, Applications Using the Coupled Hei ve an d Pitch Model Xii. INFORMATION SYSTEMS-1 INFORMATION SYSTEMS-III Chairmen: Chairmen: J. A. Smith C. D. Kearse USCG R&D Center Office of Marine Operations, G. Williams National Oceanic and Atmospheric Administration Texas A&M University D. White General Instrument Corp. H. Bhargava and S. 0. Kimbrough 1554 Oona: An Intelligent Decision Support System for the G. Samuels 648 U.S. Coast Guard A Shipboard Data Acquisition, Logging and Display System T. F. Pfeiffer 612 A Single Board-Computer Based Sail Controller C. V. Baker and W. T. Whelan 650 Offshore Oceanographic Applications for Battery- M. R. Nayak 615 Powered, High-End Microprocessors On the Knowledge-Based Expert System for Marine Instrumentation R. Findley 655 CIDS-A Shipboard Centralized Integrated Data R. J. Smith 618 System OPDIN-One Way the Ocean Community Informs M. Reynolds, R. Hendershot, M. jungck and INFORMATION SYSTEMS-U B. Reid 1560 The Zeno Alliance Network: A Dual-Loop Fiber Optic Chairman: Instrumentation Network for Ships P. Topoly Systems Planning NESDIS, MOORING National Oceanic and Atmospheric Administration Chairmen: D. Stamulis and M. P. Shevenell 623 K. R. Bitting The Use of WORM Optical Disks in Ocean Systems USCG R&D Center W. B. Wilson 629 R. Swenson A Method for Optimizing Environmental Observing Neptune Ocean Engineering Networks J. D. Babb 660 W. C. Sutherland 632 Validation of Computer Model Predictions of the A User-Friendly Multi-Functional CTD Software Large-Scale Transient Dynamic Towing Response of Package Flexible Cables D. Hamilton and J. Ward 637 H. 0. Berteaux, D. E. Frye, P. R. Clay and On-line Access to NODC Information Services E. C. Mellinger 670 Surface Telemetry Engineering Mooring (STEM) E. Voudouri and L. Kurz 641 D. R. May 681 Robust Sequential m-Interval Approximation Detectors New Technologies and Developments in NDBC Buoy with Q-Dependent Sampling and Mooring Design xiii. OPERATIONAL OCEANOGRAPHY INTERNATIONAL COUNCIL FOR EXPLORATION OF THE SEA Chairman: Chairmen: S. R. Osmer J. B. Pearce USCG International Ice Patrol ICES Marine Environment Quality Committee S. R. Osmer and D. L. Murphy 687 J. N. Moore International Ice Patrol Applied Oceanography Center Ocean Law and Policy, University of Virginia R. L. Tuxhorn 691 J. F. Pawlak 719 Oceanography on EAGLE Australia '88 Cruise A Review of the Origins, Responsibilities, Composition J. A. McNitt 696 and Main Activities of the International Council for United States Navy Operational Oceanography: the Exploration of the Sea (ICES) Fighting Smart with Oceanography Intelligence S''A. Murawski 726 An Evaluation of Shellfish Research in the OCEANOGRAPHY-MEASUREMENTS AND ANALYSIS International Council for the Exploration of the Sea Chairman: J. B. Pearce 732 W. D. Scherer The ICES Marine Environmental Quality Committee Office of Oceanography and Marine Assessments, (MEQC): Its History and Activities National Oceanic and Atmospheric Administration F. P. Thurberg 736 P. Clemente-Colon and J. Zaitzeff 1629 The ICES Working Group on Biological Effects of Upwelling Monitoring Off Western Sahara Contaminants: A Case Study K. Monkelien and T. L. Murrell 699 SEWAGE SLUDGE DISPOSAL AND MONITORING Windrose, PC Software for Wind Data Analysis M. Enomoto, T. Kawanishi and W. Kato 703 Chairmen: Measurement of Luminance Distribution on the Sea C. Dougherty Surface for Comformble Living Space Environmental Protection Agency G. Lotzic UNDERWATER PHOTOGRAPHY New York City Department of Environmental Protection Chairman: J. C. Swanson and K. Jayko 740 E. Kristof Modeling the Impacts of CSO Treatment Alternatives National Geographic Society on Narragansett Bay E. Kristof, J. Stancampiano and A. Chandler 709 H. M. Stanford and D. R. Young 745 Use of a Macro I-Hybrid Camera at National Geographic Pollutant Loadings to the New York Bight Apex E. Kristof, A. Chandler and W. Hamner 713 S.E. McDowell, C. S. Albro, W. R. Trulli, 3-D as an Underwater Too] W. G. Steinhauer and F. G. Csulak 1630 Optimum Techniques for Tracking Plumes in the Ocean: A Case Study of Sludge Plume Dispersion at the 106-Mile Site C. E. Werme, P. D. Boehm, W. G. Steinhauer and F. G. Csulak 1631 A Monitoring Plan for Disposal of Sewage Sludge at the 106-Mile Site C. D. Hunt, W. G. Steinhauer, C. E. Werme, P. D. Boehm and F. G. Csulak 1632 Monitoring Water Quality Characteristics During Dispoas] of Sewage Sludge at the 106-Mile Site xiv. MARINE MINERAL RESOURCES PROBLEMS IN OUR BAYS AND ESTUARIES Chairman: Chairman: B. Haynes V. K. Tippie Environmental Protection Agency Estuarine Program Office, National Oceanic and Atmospheric Administration R. J. Greenwald and H. F. Hennigar, Jr. 752 Designation of an Ocean Mining Stable Reference Area E. M. Burreson and J. D. Andrews 799 Unusual Intensification of Chesapeake Bay Oyster R. M. Mink. B. L. Bearden and E. A. Mancini 762 Diseases During Recent Drought Conditions Regional Geologic Framework of the Norphlet Formation of the Onshore and Offshore Mississippi, C. F. D'Elia and P. R. Taylor 803 Alabama, and Florida Area Disturbances in Coral Reefs: Lessons from Diadema Mass Mortality and Coral Bleaching T. J. Rowland 768 Availability of Minerals Offshore Virginia P. A. Tester, P. K. Fowler and R. P. Stumpf 808 Red Tide the First Occurrence in North Carolina C. E. McLain 777 Waters: An Overview Ocean Mining: An Opportunity for Public-Private Partnership B. L. Welsh 1633 Hypoxia in Long Island Sound (LIS), Summer of 1987 R. V. Amato 783 Recent Nonenergy Mineral Activity in the Atlantic P. Molinari 1609 Outer Continental Shelf EPA's Response to the Flotables Incidents of the Summer of 1987 TRASH ALONG THE COAST THE DOLPHIN DIE-OFF Chairman: L. Swanson Chairman: State University of New York N. M. Foster National Marine Fisheries Service J. B. Pearce 786 Events of the Summer of '87 D. R. Cassidy, A. J. Davis, A. L. jenny and D. A. Saari 812 L. Schmidt 790 Pathology of the Diseased Dolphins Impacts and Implications of the Summer of 1987, Newjersey Flotible Incidents J. Geraci 1634 Epidemiology of Bottlenose Dolphin Disease-U.S. R. E. Dennis, R. P. Stumpf and M. C. Predoehl 1569 Atlantic Coast, 1987-1988 Environmental Conditions in New York Bight, July- August, 1987 J. G. Meade, C. W. Potter and W. A. McLellan 815 Statistical Characteristics of the 1987 Bottlenose R. L. Swanson, R. Zimmer and C. A. Parker 794 Dolphin Die-Off in Virginia Meteorological Conditions Leading to the 1987 Washup of Floatable Wastes on NewJersey Beaches W. Medway 818 and Comparison of These Conditions with the Results of the Dolphin Epidemic Investigation as the Historical Record Disease was Presented in New Jersey Specimens of Bottlenose Dolphins in 1987 G. P. Scott, D. M. Burn and L. J. Hansen 819 The Dolphin Dicoff. Long-Term Effects and Recovery of the Population xv. SHIPWRECK ARCHEOLOGY OIL SPILL MOVEMENT Chairmen: Chairman: W. C. Phoel D. F. Paskausky NMFS, Sandy Hook Laboratory, USCG R&D Center National Oceanic and Atmospheric Administration J. Bondareff I. M. Lissauer 842 A Verified Model for Oil Spill Movement, Beaufort House Merchant Marine and Fisheries Committee Sea, Alaska J. D. Broadwater 824 M. Reed and E. R. Gundlach 847 Historic Shipwrecks: Resources Worth Protecting Hindcast of the Amoco Cadiz Oil Spill A. G. Giesecke 827 E. J. Tennyson and H. Whittaker 853 The Abandoned Shipwreck Act: A Context The 1987 Newfoundland Oil Spill Experiment: An P. J. A. Waddell 833 Overview Reburi2l of a 16th Century G211eon E. J. Tennyson 857 J. D. Broadwater 837 Shipboard Navigational Radar as an Oil Spill Tracking Supporting Underwater Archaeology with Ocean Too]: A Preliminary Assessment Technology C. M. Anderson and R. P. LaBelle 1673 R. W. Lawrence 1627 Update of Occurrence Rates for Accidental Oil Spills Consequences of the Abandoned Shipwreck Act: The on the U.S. Outer Continental Shelf North Carolina Example DRIFT MEASUREMENT J. Fullmer 1677 Myth and Management-The Shipwreck Management Chairmen: Act 1. M. Lissauer USCG R&D Center ART R.Q. Robe USCG R&D Center Chairman: H. B. Stewart, Jr. A. A. Allen and C. B. Billing 860 Old Dominion University Spatial Objective Analysis of Small Numbers of Lagrangian Drifters C. Olsen 1576 Art and Technology on 20th-Century Vessels M. J. Lewandowski 865 A Minicomputer Application to Graphically Display H. B. Stewart, Jr. 840 Tidal Current Drift Artists on Oceanographic Expeditions: A Neglected Partnership E. A. Meindl 871 Drifting Buoy Data Quality and Performance Assessment at the National Data Buoy Center P. J. Hendricks 1635 Drift Current Measurements from a Submarine Xvi. ECONOMICS OF MARINE OPERATIONS ENVIRONMENTAL POLICY Chairmen: Chairmen: F. Olson S. Bolton Environmental Consultant Office of Legislative Affairs, D. M. King National Oceanic and Atmospheric Administration ICF Inc. R. Dye House Merchant Marine and Fisheries Committee M. D. Aspinwall 876 Commercial Vessel Operations in the Exclusive R. W. Zeller 905 Economic Zone: Will the Jones Act Keep Up? Resolving the Environmental Decisionmaking and Research Dilemma M. W. Clark, Jr., D. P. Robinson and L. G. Antle 1690 Economic Impacts from Coal Exports: Through the C. A. Crampton and R. C. Helland 910 Port of Baltimore and the Port of Norfolk A Strategy for Program Implementation C. D. MacDonald and H. E. Deese 880 J. N. Leonard 914 Opportunities for Development: A Growth Scenario Updating the Stratton Commission: A Proposal for the and Situation Analysis of Hawaii's Ocean Industries U.S. Coast Guard Ocean Survey Corps D. L. Soden, J. D. Reighard and W. H. Hester 891 H. E. Schultz 920 Outside Influence on Port Operations: The Insider's National Response Mechanism Perspectives J. S. Hawkins 925 M. G. Johnson 896 Satellite Ocean Monitoring at Ten Years: Perceptions Use of Systems Analysis Techniques in Ocean and Realities Resources Development P. Stang and E. Turner 1616 EDUCATION AND TRAINING Legal and Policy Issues at Stake in the Current 5-Year Program Chairmen: Richard Asaro, USCG ESTUARINE STUDIES-I Office of Marine Safety, Security and Environmental Protection Chairman: Thoyer Shafer D. J. Basta Office of Oceanography and Marine Assessments, J. Morton 899 National Oceanic and Atmospheric Administration Marine Field Projects: Teaching is the Easy Part S. E. McCoy 930 S. Teel 1582 Monitoring the Estuary Maritime Training and Ocean Education L C. Sheifer 937 H. F. Trutneff 902 Climate, Weather, and Coastal Recreational Growth in The Impact of Marine Technology on Education and the Southeast U.S. in 1986 Training in Marine Transportation A. Stoddard 942 An Innovative Approach for the Synthesis of Large Oceanographic Data Sets with Pre-Processing and Post-Processing of an Ecosystem Model of the New York Bight J. Gerritsen 948 Biological Control of Water Quality in Estuaries: Removal of Particulate Matter by Filter Feeders Xvii. ESTUARINE STUDIES-11 OCEAN POLICY-A MATRIX OF FEDERAL, STATE AND INTERNATIONAL ISSUES Chairmen: S. E. McCoy Chairmen: Estuarine Program Office, E. W. Cannon National Oceanic and Atmospheric Administration USCG Governmental Affairs Staff D, Ashe K. U. Wolniakowski House Merchant Marine and Fisheries Committee State of Oregon Oral only E. W. Cannon 1717 The USCG: A Prototype for National and International FISHERIES AND RESOURCE ASSESSMENTS Ocean Policy Implementation Chairman: L. A. Berney R. Smolowitz The Unspoken Yet Vital Partnership Between the National Marine Fisheries, USCG Reserveand the Civilian Community at Large National Oceanic and Atmospheric Administration C. R. Corbett 992 F. L. Ames 961 International Oil Spill Liability and Compensation Improved U.S. Strategy for Fisheries Law Enforcement E. Hout, R. Bailey and K. U. Wolniakowski 994 G. Reetz 966 Ocean Resource Management in Oregon: Pushing California Sea Otter: Impact Assessment and Mitigation Open the Window of Opportunity D. Luo 972 J. S. S. Lakshminarayana 1000 Theoretical Analysis of Fish School Density Overview and Analysis of Coastal Zone Management in the Atlantic Provinces, Canada R. J. Smolowitz and F. M. Serchuk 975 D. C. Slade 1006 Marine Fisheries Technology in the United States: Coastal States and Marine Resource Development St2tus, Trends and Future Directions Within the United States Exclusive Economic Zone B. F. Beal 980 Public Aquaculture in Downcast Maine: The Soft-Shell OCEAN DRILLING PROGRAM Clam Story Chairman: P. H. Averill 1637 J. H. Clotworthy Development of Separator Trawl Technology Consultant FISHERIES-IMPACT STUDIES P. Brown, K. Lighty, R. Merrill and P. D, Rabinowitz 1012 Chairman: Collection and Quality Control of Marine Geological J. Chambers Data by the Ocean Drilling Program National Marine Fisheries Service D. Graham, B. Hamlin, B. julson, W. Mills, A. Meyer, H. A. Carr 984 R. Olivas, P. D. Rabinowitz, D. Bontempo and Long Term Assessment of a Derelict Gillnet Found in J. Tauxe 1018 the Gulf of Maine Shipboard Laboratory Support: Ocean Drilling Program A. E. Pinkney, L. L. Matteson and D. A. Wright 987 P. Weiss, G. Bode, C. Mato, R. Merrill, P. D. Effects of Tributyltin on Survival, Growth, Rabinowitz, M. Angell, J. Miller, P. Myre, S. Prinz, Morphometry and RNA-DNA Ratio of Larval Striped D. Quoidbach and R. Wilcox 1025 Bass, Morone saxatilis Core Curation: Ocean Drilling Program xviii. OCEAN ENERGY-[ MARINE MAMMALS RESEARCH AND MANAGEMENT Chairman- Chairman: D. Cotter D. Swanson CBI Industries National Marine Fisheries Service, National Oceanic and Atmospheric Administration P. Vauthier 1029 The Underwater Electric Kite: East River Deployment H. H. Armstrong and K. R. Banks 1073 Modern Eskimo Whaling in the Alaskan Arctic D. E. Lennard and F. A. Johnson 1034 British OTEC Programmes-] OMW Floating and G. H. Allen 1079 0.5MW Land Based Observations on the 1987 Subsistence Harvest of Northern Fur Seals on St. Paul Island, Pribilof Islands, R. K. Jensen 1039 Alaska Hydro Power from the Ocean R. T. Bennett 1083 A. Thomas and D. Hillis 1045 Endangered Species and Marine Mammal Protection First Production of Potable Water by OTEC and Its During Offshore Structure Removals in the Gulf of Potential Applications Mexico OCEAN ENERGY-II SHIP DESIGN AND REPAIR Chairman: Chairmen: L. Lewis M. S. Canavan Department of Energy USCG Office of Engineering and Development T. Colton D. C. Hicks, C. M. Pleass and G. R. Mitcheson 1049 Colton Company DELBUOY: Wave-Powered Seawater Desalination System M. S. Canavan and M. D. Noll 1087 U.S. Coast Guard's New Polar Icebreaker Design K. P. Melvin 1055 A Wave Energy Engine and Proposals for its D. W. Yu and J. H. Devletian 1098 Development and Usage Electroslag Surfacing for Construction, Restoration, L. Claeson 1638 and Repair of Ship Structures Recent Wave Energy Research in Sweden RESEARCH VESSELS K. Kudo, T. Tsuzuku, K. Imai and Y. Akiyama 1061 Wave Focusing by a Submerged Plate Chairman: W. Barbee Y. Masuda, M. E. McCormick, T. Yamazaki and University-National Oceanographic Laboratory Y. Outa 1067 System The Backward Bent Duct Buoy-An Improved J. A. Chance 1107 Floating Type Wave Power Device Conversion of Surplus Oilheld Supply Vessels to Research Vessels C. Hamlin 1111 Research Vessels: A Systems Engineering Approach B. L. Hutchison and S. jagannathan 1117 Monohull Research Vessel Seakeepingand Criteria R. J. Wilber, C. E. Lea and S. E. Humphris 1639 The SSV Corwith Cramer: Sea Education Association's New Sailing Research Vessel Xix. SALVAGE AND TOWING SWATH SHIPS-[ Chairman: Chairman: J. H. Boyd R. Dinsmore Booz, Allen & Hamilton, Inc. Woods Hole Oceanographic Institution J. K. Edgar 1640 G. R. Lamb 1131 Hazardous Materials in Marine Salvage Operations Relationship Between Seakeeping Requirements and SWATH Ship Geometry C. M. Kalro 1603 Launch and Retrieval of a 1,000 Ton Barge Shaped M. Rice, E. Craig, S. Drummond and Vessel from 2 Conventional Tinker C. junemann 1144 Conceptual Design of an Intermediate Size J. Strandquist 1124 Oceanographic Research Ship for the University- Removal of the Wreck of the Ex-USS TORTUGA National Oceanographic Laboratory System THE SMALL PASSENGER VESSEL INDUSTRY-I R. D. Gaul, A. C. McClure and F. E. Shumaker 1149 Design of a Semisubmerged SWATH Research and Chairmen: Survey Ship E. G. Sharf C. Kennell 1157 National Association of Passenger Vessels Tankage Arrangement for SWATH Ships H. Parker National Association of Passenger Vessels SWATH SHIPS-11 W. B. Hamner 1641 Chairman: The Future of the Tourist Submarine Industry K. W. Kaulum Oral only Office of Naval Research T. G. Lang, C. B. Bishop and W. J. Sturgeon 1163 THE SMALL PASSENGER VESSEL INDUSTRY-H SWATH Ship Designs for Oceanographic Research Chairman: E. Craig and S. E. Drummond 1169, E. G. Scharf SWATH CHARWIN-Range Support Ship National Association of Passenger Vessels A. Galerne 1573 T. MacRae 1125 Development of Deep Water Technology as It Relates The Realities of B2reboat Chartering to Future Salvage Oral only E. Craig Real World Experience with SWATH Design Xx. SHIPBOARD TECHNICAL SUPPORT AN INTERIM STATUS REPORT ON ORGANOTINS-11 Chairman: Chairmen: H. L. Clark P. F. Seligman National Science Foundation Naval Ocean Systems Center H. L. Clark 1644 M. A. Champ Shipboard Technician Program of the National Science National Science Foundation Foundation K. W. M. Siu 1716 J. D. Guffy, M. A. Spears and D. C. Biggs 1173 Analytical Chemistry of Butyltins Automated Analyses of Nutrients in Seawater with the T. L. Wade, B. Garcia-Romero and J. M. Brooks 1198 Technicon TrAAcs-800 Autoanalyzer System Tributyltin Analyses in Association with NOAA's D. J. Murphy, E. Wilson and E. Powell 1178 National Status and Trends Mussel Watch Program An Application of a Low Flow Current Meter to Broad C. M. Adema, W. M. Thomas, Jr., and Temperature Range Estuarine Current Measurements S. R. Mangum 1656 M. Maccio and C. Langdon 1181 Butyltin Releases to Harbor Water from Ship Painting Description of Conversion of an EG&G VMCM into a in a Dry Dock MVMS (Multi-Variable Moored Sensor) B. Cool Summary and Status Report of EPA's Special Review AN INTERIM STATUS REPORT ON ORGANOTINS-1 (PD14) Chairmen: WAVE MOTION P. F. Seligman Naval Ocean Systems Center Chairman: M. A. Champ R. H. Canada National Science Foundation National Data Buoy Center, National Oceanic and Atmospheric Administration P. F. Seligman, J. G. Grouhoug and C. M. Adema Field Monitoring of TBT Concentrations in Pearl L. J. Ladner, W. B. Wilson and P. J. Kies 1202 Harbor Correlated with Model Simulation Studies Lake Superior Winter Weather Station R. S. Henderson 1645 N. Lang Chronic Exposure Effects of Tributyltin on Pearl The Linear Properties of Spectra from a PitchlRoll Harbor Organisms Buoy M. H. Salazar and S. M. Salazar 1188 E. D. Michelena and R. Dagnall Tributyltin and Mussel Growth in San Diego Bay A Computer Controlled Signal Simulator for Buoy Motion Sensors M. H. Salazar and M. A. Champ 1497 Tributyltin and Water Quality: A Question of D. Smith and F. Remond Environmental Significance 3-Meter Directional PitchlRoll Buoy W. R. Blair, G. J. Olson, T. K. Trout, K. L. Jewett and F. E. Brinckman 1668 Accumulation and Fate of Tributyltin Species in Microbial Biofilms xxi. WAVE MEASUREMENTS UNDERWATER VEHICLES Chairman: Chairman: L. Baer R. Blidberg Office of Oceanography and Marine Assessments, University of New Hampshire National Oceanic and Atmospheric Administration M. Higgins and R. Whyte 1646 H. Brown 1205 Controlled Depressor Towed Sensor Platform-The Infrared Laser Wave Height Sensor U.S. Navy's Mk28 Search System G. Kontopidis and G. Bowers 1207 M. Higgins, B. Lawson and B. Field 1647 WavePro: An Autonomous Wave Processor with Long- Development and Testing of a Heavy-Duty Work ROV Range Telemetry for 10,000 Foot Service F. Ziemer, H. GUnther and E. Stockdreher 1212 H. Momma, K. Ohtsuka and H. Hotta 1253 Measured Transfer Functions for Shipmotions in JAMSTECIDeep Tow System Natural Seaways J. jalbert, M. Shevenell, S. Chappel, R. Welsh and D. W. Farrell 1587 R. Blidberg 1259 The Next Generation Water Level Measurement EAVE III Untethered AUV Submersible System: The Next Step in Real-Time Data for Navigation M. E. Cooke, S. Gittings, J. M. Brooks and D. C. Biggs WAVE ACTION ON SEA SHORES Texas A&M University Remotely Operated Oceanographic Vehicle (TAMU-ROOV) Chairman: M. Earle UNDERWATER VEHICLE SYSTEMS AND EQUIPMENT MEC Systems Corp. Chairmen: M. J. Briggs and P. J. Grace 1218 S. B. Cable Influence of Frequency and Directional Spreading on Naval Civil Engineering Laboratory Wave Transformation in the Nearshore Region R. Wernli D. D. McGehee and J. P. McKinney 1224 Naval Ocean Systems Center Tidal Circulation Data from the Los AngeleslLong F. Dougherty, T. Sherman, G. Woolweaver and Beach Harbors G. Lovell 1265 S. L. Da Costa and J. L. Scott 1231 An Autonomous Underwater Vehicle (AUV) Flight Wave Impact Forces on the Jones Island East Dock, Control System Using Sliding Mode Control Milwaukee, Wisconsin . M. L. Nuckols, J. Kreider and W. Feild 1271 J.Rosati III and G. L. Howell 1239 Thermal Modelling of Electro-Mechanical Cables for A Hierarchical Multiprocessor Data Acquisition System ROV Applications for Field Measurement of Structural Response in M. P. Shevenell and C. Millett 1276 Breakwater Concrete Armor Units A LISP Environment for Real-Time Ocean Systems J. P. Ahrens and E. T. Fulford 1244 S. B. Cable 1280 Wave Energy Dissipation by Reef Breakwaters A Guideline System for the Navy's Submarine Rescue E. H. Harlow 1250 Ship (ASR) Class Why Breakwaters Break W. J. Herr 1290 A UV Technology: Development and Demonstration Program Xxii. MANNED SUBMERSIBLES HAZARDOUS CHEMICAL IDENTIFICATION AND MANAGEMENT Chairman: Chairmen: R. W. Cook L. H. Gibson Harbor Branch Oceanographic Institute USCG Central Oil Identification Laboratory Oral only E. F. Batutis Phasesep Corp. DIVING OPERATIONS AND SYSTEMS W. R. Cunningham 1321 NOAA Fleet Hazardous Materials and Hazardous Waste Chairmen: Management W. C. Phoel National Marine Fisheries Service, L. H. Gibson and M. S. Hendrick 1326 National Oceanic and Atmospheric Administration U.S. Coast Guard Oil Identification System J. M. Wells T. J. Haas, J. J. Kichner and T. J. Chuba 1332 Office of Marine Operations, Course in Hazardous Materials National Oceanic and Atmospheric Administration J. K. Jeffries BUOY-BASED METEOROLOGY Standards and Procedures for Dry Suit Diving Education Chairman: R. Canada J. K. Jeffries National Buoy Center Thermal Guidelines for Diving Operations S. P. Burke and D. G. Martinson 1335 R. I. Wicklund 1614 An ARGOS Meteorological Oceanographic Spar Buoy An Inexpensive Mobile Self-Contained Habitat System for Antarctic Deployments for Marine Research D. B. Gilhousen 1341 J. W. Blackwell and C. D. Newell 1300 Methods of Obtaining Weather Data in Real Time Diving in Hazardous and Polluted Waters E. D. Michelena 1649 J. M. Wells 1305 The Measurement of Precipitation at National Data The Use of Nitrogen-Oxygen Mixtures as Diver's Buoy Center Stations Breathing Gas R. R. Miller and R. Canada 1650 J. P. Fish and H. A. Carr 1309 Mini-Drifter Test Deployment Data-Gulf of Mexico Integrated Remote Sensing of Dive Sites Spring 1988 S. L. Merry, S. L. Sendlein and A. P. Jenkin 1315 P. M. Friday, J. S. Lynch and F. S. Long 1344 Human Power Generation in an Underwater Interactive Marine Analysis and Forecast System Environment (IMAFS): The Oceanographic Workstation of the Future D. A. Storey and W. E. Woodward 1348 The Global Ocean Platform Inventory VESSELS OF THE 80s AND BEYOND Chairmen: E. K. Pentimonti American President Lines, Ltd. P. Mentz Advanced Ship Operations, MARAD Oral only Xxiii. FACILITIES IN SUPPORT OF MARINE FREIGHT NAVIGATION CHARTING TRANSPORTATION Chairmen: Chairmen: R. Vorthman M. J. Vickerman, Jr. USCG Operations Control Center, Atlantic Area Vickerman, Zachary, Miller M. Kumar R. Katims Defense Mapping Agency Container Transport Technology W. M. Maynard 1371 Oral only Cooperative Electronic Chart Development: The GAADS Project NAVIGATION SYSTEMS AND OPERATIONS P. W. Mushkat and C. Lamson 1589 Chairmen: Electronic Chart Display Information Systems: G. R. Perreault Operational, Policy and Legal Issues Office of Navigation, N. D. Smith 1374 U.S. Coast Guard Automated Nautical Data and Charting Development J. Illgen E. A. Soluri A. F. E. Fuentes 1651 Defense Mapping Agency's Navigational Information A Survey of Radionavigation System Users System L. Mehrkam 1352 NAVIGATION SYSTEMS Leading Lines for the Nineties G. R. Perreault 1356 Chairmen: Contract Service of Federal Aids to Navigation C. D. Kearse Office of Marine Operations, R. J. Weaver and R. M. Piccioni 1362 National Oceanic and Atmospheric Administration Marine Radionavigation of the Future P. Stutes John E. Chance & Assoc. Inc. L. V. Grant 1365 Federal Radionavigation Plan Overview J. L. Hammer III and W. R. Hoyle 1379 The Continuing Need for Accurate Positioning in J. Hammer III and W. R. Hoyle 1684 Naval Tactics The Continuing Need for Accurate Positioning in Naval Tactics E. F. Carter and J. Lewkowicz 1594 A Computer Navigation System Using Kalmaro Filter Smoothing R. Gandy and S. Paulet 1648 Design and Applications of SEATRAC, an Integrated Navigation and Data Management System D. C. Slade Solar Navigation A. E. Shaw III and T. E. Bryan 1384 Oceanographic Applications of the ARGOS System xxiv. AIDS TO NAVIGATION SYSTEMS AND EQUIPMENT SEARCH AND RESCUE-SURVEILLANCE EXPERIMENTS Chairman: Chairman: T. S. Winslow W. H. Reynolds Office of Engineering, USCG R&D Center U.S. Coast Guard D. Finlayson, D. Bryant, B. R. Dawe and T. S. Winslow, M. D. Dawe, K. R. Schroeder and A. J. Armstrong 1417 W. A. Fisher 1390 Results of an Experiment to Examine Certain Human High-Voltage Solar-Powered Navigation Range Design Factors Relating to Searches Conducted with Marine Radar J. McCaffrty An Alternative Hull Design for the U.S. Coast Guard D. Bryant, B. R. Dawe, D. Finlayson, W. Reynolds Bell Buoy and M. J. Lewandowski 1422 Results of Canadian Shipborne Night Search GLOBAL POSITIONING SYSTEM Experiments Chairmen: B. R. Dawe, D. Finlayson and D. Bryant 1427 K. Nakamura Results of a Canadian Shipborne Radar Search and Office of the Assistant Secretary of Defense Rescue Detection Experiment R. S. Warren D. Finlayson, B. R. Dawe and D. Bryant 1433 TASC Results of a Canadian Visual Search and Rescue Detection Experiment L. D. Hottram Relative GPS Kinemetric Surveying and Applications F. Replogle, Jr. 1436 for Marine Positioning A New Coast Guard Search Technique M. J. Mes 1395 SEARCH OPERATIONS Accuracy of Satellite Survey Measurements on Offshore Platforms for Monitoring Subsidence Chairmen: E. M. Geyer and R. S. Warren R. Q. Robe Mission Planning Issues and Answers for GPS Users USCG R&D Center B. Dawe SHIP OPERATIONS AND SCHEDULING Nordco Ltd. Chairman: R. W. Berwin 1439 C. Pritchett Alaska SAR Facility Archive and Operations System USCG R&D Center M. K. Kutzleb 1444 S. Cook, R. Benway, W. Krug, M. Nestlebush. A. The Search for South African Airways Flight 295 Picciolo, W. Richardson, P. Stevens and D. R. Paskausky, W. Reynolds, R. Gaines and V. Zegowitz 1400 R. Q. Robe 1605 Volunteer Observing Ships and the U.S. Improving Search Success; Real-Time Collection and Government-A Winning Partnership I Transfer to User L. C. Kingsley, K. S. Klesczewski, J. A. Smith and R. Q. Robe, D. F. Paskausky and G. L. Hover 1448 R. A. Walters 1405 Performance of Coast Guard Medium Range Comparing the U.S. Coast Guard Buoy Tender Surveillance (MRS) Aircraft Radars in Search and Performance Using Simulation Rescue (SAR) Missions K. S. Klesczewski 1411 J. B. Brewster 1454 Using Spacefilling Curve to Generate the Feasible Sea Based Aerostats (SBA): Effective Surveillance for Routes for the Set Partitioning Problem Maritime Interdiction S. F. Roehrig 1643 Scheduling Patrols Usinga Hybrid Integer Programmingl Rule-Based System Approach xxv. PORT MANAGEMENT AND SECURITY Chairmen: T. Robinson Port Safety and Security Division, U.S. Coast Guard D. Smith House Merchant Marine and Fisheries Committee D. J. Evans, R. W. Owen and P. R. Farragut 1457 Innovative Technology Applied to Maximize a Port's Lifeline: A Case History for the Sea Lanes of the Chesapeake Bay N. A. Marziani 1463 The Multi-Agency MOU on Port Security: A Model for Conflict Resolution D. J. Sheehy and S. F. Vik 1470. Mitigation Planning for Port Development J. J. Zagel, R. T. Kilgore and S. M. Stein 1642 Hydrodynamic and Mass Transport Modeling of Navy Harbors MARINE SAFETY Chairmen: C. L. Hervey USCG R&D Center S. Steele House Merchant Marine and Fisheries Committee F. H. Anderson 1598 Awakening the Consciousness of the Boating Public Regarding Pollution, Intoxication, and Common Sense Safety of the Nation's Waterways A. Colihan 1476 Coast Guard Recreational Boating Product Assurance Program C. L. Hervey 1482 Determining Horsepower Limits on Recreational Boats S. Johnson and J. Veentjer 1487 Regulation of Passenger Carrying Submersibles G. L. Traub 1493 Recreational Boating Accidents in Ocean Waters Manuscript unavailable for publication xxvi. Authors List Abrams, W. R ............... 344 Boehm, P. D ......... 1631, 1632 Clark, M. W., jr ............ 1688 Adams, A. j ................. 573 Boerner, W.-M .............. 454 Clark, R. B ................. 184 Adema, C. M ............... 1656 Boggess, R. L ............... 423 Clay, P. R .................. 670 Aharon, P .................. 101 Bonetempo, D ............. 1018 Clemente-Colon, P .......... 1629 Ahrens, J. P ................ 1244 Bostater, C ................. 462 Clifford, S. F ................ 315 Akiyama, Y ................ 1061 Bowen, A. j ............ 413, 419 Cloet, R. L ................ 1636 Albro, C. S ................ 1630 Bowers, G ................ 1207 Coe, J. M ..................... I Allen, A. A ................. 860 Boyer, L. F ............. 443, 448 Colbourne, E. B ............. 413 Allen, G. H ................ 1079 Brashier, j .................. 136 Colihan, A ................ 1476 Alonzo, C ................. 1612 Breeding, S. K .............. 391 Colwell, R. R .............. 16o6 Amato, R. V ................ 783 Brewster, J. B .............. 1454 Cook, S ................... 1400 Ames, F. L .................. 961 Brickey, M. R .............. 1611 Cooper, R. A ................ 112 Anderson, C. M ............ 1673 Briere, M ................... 588 Corbett, C. R ............... 992 Anderson, F. H ............. 1598 Briggs, M. j ................ 1218 Corwin, R .................. 397 Andrews, J. D ............... 799 Bright, T. j .................. 22 Cottingham, D ................ 6 Angell, M ................. 1025 Brinckman, F. E ............ 1668 Craig, E ............. 1144, 1169 Antle, L. G ................ 1688 Broadwater, J. D ........ 824, 837 Crampton, C. A ............. 910 Appell, G. F ................ 346 Brooks, J. M ............... 1198 Crawford, G. B .............. 315 Armstrong, A. j ............. 1417 Brown, H ................. 1205 Crowley, J. D ............... 543 Armstrong, H. H ............ 1073 Brown, P ...........I..... 1012 Csulak, F. G. . . 1630, 1631, 1632 Ashmore, S ................. 259 Browning, D. G ............. 318 Cuddy, D. T ................ 473 Aspinwall, M. K ............. 876 Bryan, T. E ................ 1384 Cunningham, W. R .......... 1321 Asplin, R. G ................ 329 Bryant, D ........ 485, 1417, 1422, Curlander, j ................ 479 Augerot, X ................ 1711 1427, 1433 Da Costa, S. L .............. 1231 Aurand, D. V ............... 161 Buddenberg, R. A ............ 526 Dardeau, M. R ................ 17 Auster, P. j ........... 1286, 1610 Bunn, A. R ................... I Davis, A. j .................. 812 Averill, P. H ............... 1637 Burden, D. G ................ 70 Dawe, B. R ........... 1417, 1422, Babb, I. G .................. 112 Burke, S. P ................ 1335 1427, 1433 Babb, J. D .................. 66o Burn, D. M ................. 819 Dawe, M. D ............... 1390 Babb, L. V .................. 423 Burreson, E. M .............. 799 Dean, J. M ................... 35 Bailey, R ................... 994 Burroughs, R. H ............ 1607 Deese, H. E ................. 880 Baker, C. V ................. 650 Cable, S. B ................ 1280 D'Elia, C. F .............. 29, 803 Baldwin, K. C ............... 588 Caldwell, P. j ................ 46 de Moustier, C .......... 372, 509 Banks, K. R ................ 1073 Campbell, W. B ............ 1626 Dennis, R. E ............... 1569 Bascle, B. j ................. 223 Canada, R ................. 1650 Devletian, J. H ............. 1098 Bass, P. D .................. 346 Canavan, M. S .............. 1087 DeVoe, M. R ................. 35 Batt, B. D. j .................. 46 Cannon, E. W .............. 1717 Diaz, R. j ................... 448 Beal, B. F .................. 980 Carney, R. S .............. 90, 96 Dindo, J. j ................... 17 Bearden, B. L ............... 762 Carr, H. A ............. 984, 1309 Dougherty, F .............. 1265 Benada, J. R ................ 473 Carter, E. F ................ 1594 Dowd, T ................... 595 Benedict, R. L ............... 577 Cassidy, D. R ............... 812 Drummond, S ........ 1144, 1169 Bennett, R. T .............. 1083 Chalmers, A. G ............. 1605 Dunaway, M. E .............. 268 Benway, R ................ 1400 Champ, M. A .............. 1497 Dunkel, C. A ................ 208 Berney, A ................. 1725 Chance, J. A ............... 1107 Echardour, A ............... 602 Bertaux, A .................. 598 Chandler, A ........ 335,709,713 Edgar, J. K ................ 1640 Berteaux, H. 0 .............. 670 Chaplin, G ................. 275 Edson, G. M ................ 219 Berwin, R. W .............. 1439 Chappel, S ................ 1259 EI-Hawary, F ................ 291 Bhargava, H ............... 1554 Charters, J. S ........... 385, 509 Eller, A. I ................... 295 Biggs, D. C ................ 1173 Chauvin, A. L .............. 1536 Enomoto, M ................ 703 Billing, C. B ................ 860 Chico, T ................... 166 Essig, R. j .................. 127 Bimbo, A. P ............... 1513 Christensen, C. G ............ 329 Evans, D. j ................ 1457 Bishop, C. B ............... 1163 Christensen, D ............. 1624 Farragut, P. R .............. 1457 Bitterman, D ................ 359 Chuba, T. j ................ 1332 Farrell, D. W ............... 1587 Blackwell, J. W ............. 1300 Cibik, S. j .................... 29 Fedor, L. S ........... 1697, 1704 Blair, W. R ................ 1668 Claeson, L ................. 1638 Feild, W .................. 1271 Blidberg, R ................ 1259 Clark, B. R ................. 143 Field, B ................... 1647 Bode, G .................. 1025 Clark, H. L ................ 1644 Findley, R .................. 655 xxvii. Finlayson, D .......... 1417, 1422, Hazen, D. G ..........I ..... 419 Kerdoncuff, J ............... 602 1427, 1433 Heck, K. L., jr ............... 17 Kichner, J. J ............... 1332 Fish, J. P .................. 1309 Hedrick, J. D ............... 448 Kies, P. J .................. 1202 Fisher, E. A ................. 607 Helland, R. C ............... 910 Kilgore, R. T ............... 1642 Fisher, W. A ............... 1390 Hendershot, R ............. 1560 Kimbrough, S. 0 ........... 1554 Fleischer, P .................. 17 Henderson, R. S ............ 1645 King, P. C ................. 1618 Flick, G. J., jr ................ 56 Hendrick, M. S ............. 1326 Kingsley, L. C .............. 1405 Floch, H ................... 602 Hendricks, P. J ............. 1635 Klemas, V .................. 462 Floury, L ................... 602 Hennigar, H. F., jr ........... 752 Klesczewski, K ........ 1405, 1411 Fowler, P. K ................ 808 Herr, W. J ................. 1290 Klos, E ................... 1529 Fox, J. M .................. 1536 Hervey, C. L ............... 1482 Knaster, A. S ................ 156 Fran@ois, D. K .............. 241 Hester, W. H ................ 891 Kohler, C. A ................ 582 Friday, P. M ............... 1344 Hicks, D. C ................ 1049 Kontopidis, G .............. 1207 Frye, D. E .................. 670 Higgins, M ........... 1646, 1647 Koplin, S ................... 132 Fuentes, A. F. E ............ 1651 Hill, W .............. I ..... 275 Kostinski, A. B .............. 454 Fulford, E. T ............... 1244 Hillis, D .................. 1045 Krahl, R. B ................. 250, Fullmer, J ................. 1677 Hodgkiss, W. S .............. 310 Kreider, J ................. 1271 Gaines, R ................. 1605 Hotta, H .................. 1253 Krezoski, J. R ........... 437, 442 Galerne, A ................ 1573 Hout, E .................... 994 Kristof, E .......... 335, 709, 713 Galloway, J. M ............. 1611 Hover, G. L ............... 1448 Kroebel, W ................. 491 Gandy, R ............ 1620, 1648 Howell, G. L ............... 1239 Krug, W .................. 1400 Garcia-Romero, B ........... 1198 Hoyle, W. R .......... 1379, 1684 Kudo, K .................. 1061 Gast, J ..................... 346 Htun, K. M ................ 1628 Kuroda, Y .................. 353 Gaul, R. D .................. 1149 Huang, L ................... 413 Kurz, L .................... 641 Geraci, J .................. 1634 Hultin, H. 0 ................. 66 Kutzleb, M ................ 1444 Gerritsen, J ................. 948 Humphrey, B ............... 405 LaBelle, R. P ............... 1673 Gibson, L. H ............... 1326 Humphris, S. E ............. 1639 Ladner, L. J ................ 1202 Giesecke, A. G .............. 827 Hunt, C. D ................ 1632 Lakshminarayana, J. S. S ..... 1000 Gilhousen, D. B ............ 1341 Hutchison, B. L ............ 1117 Lamb, G. R ................ 1131 Givens, A .................. 526 Hylas, T ................... 372 Lambert, D ................. 397 Gorveatt, M ................ 555 Ibrahim, M ................ 1612 Lamson, C ................ 1589 Grace, P. J ................ 1218 Imai, K ................... 1061 Lang, T. G ................ 1163 Graham, D ................ 1018 Jaffe, J. S................... 338 Langdon, C ................ 1181 Granger, S. W .............. 1604 Jagannathan, S ............. 1117 Larsen-Basse, J ............. 1628 Grant, C ......... ......... 213 jai, B. H ................... 473 Lataitis, R. J ................. 315 Grant, L. V ................ 1365 Jalbert, J .................. 1259 Lavoie, D .............. 391, 397 Grant, S ................... 567 James, B. D ................. 454 Lawrence, R. W ............ 1627 Greenwald, R. J ............. 752 jayko, K ................... 740 Lawson, B ................. 1647 Guffy, J. D ................ 1173 Jenkin, A. P ............... 1315 Lea, C. E .................. 1639 Gundlach, E. R .............. 847 jenny, A. L ................. 812 Legrand, J. F ................ 602 Gunther, H ................ 1212 Jensen, R. K ............... 1039 Le Moign, T ................ 602 Haas, T. J ................. 1332 Jewett, K. L ................ 1668 Lennard, D. E .............. 1034 Hackett, H. P ............... 607 Johnson, F. A .............. 1034 Leonard, J. N ............... 914 Haines, L. C ................ 295 Johnson, M. G .............. 896 Lewandowski, M. J..... 865, 1422 Hall, S. C ................... 537 Johnson, S ................ 1487 Lewkowicz, J .............. 1594 Hamilton, D ................ 637 Julson, B .................. 1018 Liebert, B. E ........... 606, 1628 Hamlin, B ................. 1018 Junemann, C ............... 1144 Lighty, K .................. 1012 Hamlin, C ................. 1111 jungck, M ................. 1560 Lindsay, R. C ................ 61 Hammer, J. L., III ..... 1379, 1684 Kadlec, J. A .................. 46 Lissauer, M ................. 842 Hamner, W ................. 713 Kai, G ..................... 353 Liston, J .................... 52 Hamner, W. B ............. 1641 Kaiser, G. E ................. 76 Loaec, G ................... 602 Hansen, L. J ................ 819 Kalro, C .................. 1603 Long, F. S ........ ........ 1344 Hanson, W. E ............... 561 Kato, W ................... 703 Lovell, G .................. 1265 Harlow, E. H .............. 1250 Kawanishi, T ............... 703 Luo, D .................... 972 Hawkins, J. S ............... 925 Keer, F. R .................. 188 Lynch, J. S ................ 1344 Hay, A. E .............. 413, 419 Kennedy,J ................. 213 Maccio, M ................. 1181 Hayne, G. S ............... 1702 Kennell, C ................ 1157 MacDonald, C. D ............ 880 xxviii. MacDonald, I ................ 90 Moser, C ................... 505 Reetz, G ................... 966 Mac Gillvray, T. j ............ 262 Mozley, E .................. 397 Reid, B ................... 1560 MacRae, T ................. 1125 Murawski, S. A .............. 726 Reighard, J. D ............... 891 Mahrt, K.-H ................. 497 Murkin, H. R ................. 46 Renner, W. W ............... 295 Malone, R. F .............. 70, 84 Murphy, D. j ............... 1178 Replogle, F., jr ............. 1436 Mancini, E. A ............... 762 Murphy, D. L ........... 467, 687 Reynolds, M ............... 1560 Mangum, S. R .............. 1656 Murrell, T. L ................ 699 Reynolds, W ......... 1422, 1605 Martinson, D. G ............ 1335 Mushkat, P. W ............. 1589 Rezak, R ............... 22, 1602 Marziani, N. A .............. 1463 Myre, P ................... 1025 Rice, M ................... 1144 Masuda, Y ................. 1067 Nayak, A R ................ 615 Richardson, W ............. 1400 Mato, C ................... 1025 Nebert, D. L ............... 1627 Riedel, G. F ................. 23 Matteson, L. L ............... 987 Nergaard, D ................ 275 Robe, R. Q ........... 1448, 1605 May, D. R .................. 681 Nestlebush, M .............. 1400 Roberts, H. H ............... 101 Mayerson, D ................ 229 Newbury, T ................ 573 Robinson, D. P ............. 1688 Maynard, W. M ............. 1371 Newell, C. D ............... 1300 Roderick, D. K ............. 1619 McCammon, D. F ............ 304 Nicholson, J. W ............. 338 Roehrig, S. F ............... 1643 McCandless, S. W., jr ......... 479 Nixon, S. W ............... 16o4 Roffer, C ................... 359 McClure, A. C .............. 1149 Noll, M. D ................. 1087 Rogers, R. M ................ 953 McCormick, M. E ........... 1067 Novick, A ................. 1617 Rosati, J., III .............. 1239 McCoy, S. E ................ 930 Nuckols, M. L .............. 1271 Rounds,R.S ................ 106 McDowell, S. E ............. 1630 O'Hara, K. j ................. 12 Rowland, T. j ............... 768 McGehee, D. D ............. 1224 Ohtsuka, K ................ 1253 Rusch, K .................... 84 McGrail, D. W ............. 1602 Okuno, K .................. 353 Saari, D. A .................. 812 McKinney, J. P ............. 1224 Olivas, R .................. 1018 Saklad, H ................. 1627 McLain, C. E ................ 777 Olla, B. L .................. 119 Salazar, M. H ......... 1188, 1497 McLellan, W. A .............. 815 Olsen, C .................. 1576 Salazar, S. M ............... 1188 McNitt, J. A ................. 696 Olson, G. j ................ 1668 Samuels, G ................. 648 Meade, J. G ................. 815 Osmer, S. R ................ 687 Sanders, J. G ............. 23, 29 Medway, W ................ 818 Outa, Y ................... 1067 Sassen, R ................... 101 Mehrkam, L ............... 1352 Owen, R. W ............... 1457 Satkowiak, L. j .......... 428, 433 Meindl, E. A ............ 629, 871 Parker, C. A ................ 794 Schiefele, P. M .............. 318 Mellen, R. H ................ 318 Paskausky, D. F ....... 1448, 1605 Schmidt, L ................. 790 Mellinger, E. C .............. 670 Paulet, S ............. 1620, 1648 Schomette, D ............... 335 Melvin, K. P ............... 1055 Pawlak, J. F ................. 719 Schroeder, K. R ............ 1390 Merrill, R ............ 1012, 1025 Pearce, J. B ............. 732, 786 Schroeder,P ................ 268 Merry, S. L ................ 1315 Perreault, G. R ............. 1356 Schroeder, W. W .......... 17,22 Mes, M. j .................. 1395 Perry, R. B ................. 366 Schultz, A. W ................ 17 Mesecar, R ............. 505, 518 Pfeiffer, T. F ................ 612 Schultz, H. E ................ 920 Meyer, A .................. 1018 Phillips, J. C ................ 372 Scott, G. P ................. 819 Meyer, R. M ................ 146 Picciolo, A ................ 1400 Scott, J. L ................. 1231 Michelena, E. D ............ 1649 Piccioni, R. M .............. 1362 Sendlein, S. L .............. 1315 Middleton, R. W ............. 123 Pinkney, A. E ............... 987 Serchuk, F. M ............... 975 Miller, j ................... 1025 Pleass, C. M ............... 1049 Shaw, A. R., III ............ 1384 Miller, R. E ................. 172 Ponsford, A. M .............. 485 Sheehy, D. j ............... 1470 Miller, R. R ................ 1650 Potter, C. W ................ 815 Sheifer, 1. C ................ 937 Millett, C .................. 1276 Powell, E ................. 1178 Sheng,j ................... 413 Mills, W .................. 1018 Predoehl, A C ............. 1569 Sherman, T ................ 1265 Mimken, G ................ 1627 Prinz, S ................... 1025 Shevenell, M. P. . . 623, 1259, 1276 Mingrone, J. A ............. 1618 Pryor, D. E ................. 379 Shumaker, F. E ............ 11149 Mink, R. M ................. 762 Quoidbach, D .............. 1025 Siu, K. W. M ............... 1716 Mitcheson, G. R ............ 1049 Rabinowitz, P. D ........... 1012, Slade, D. C ................ 1006 Moe, R. L .................. 532 1018, 1025 Smith, C. E ................. 250 Molinari, P ................ 1609 Raer, Y .................... 602 Smith, G. A ................. 106 Momma, H ................ 1253 Rausch, K ................... 84 Smith, J. A- ** .... ** ....... 1405 Monkelien, K ............... 699 Ray, P. K ................... 193 Smith, N. D ................ 1374 Moore, J. M ............ 385, 509 Reed, J. C .................. 202 Smith, R. j ................. 618 Morton, j .................. 899 Reed, M ................... 847 Smith, S. M ............. 385,549 xxix. Smolowitz, R. j .............. 975 Valent, P ................... 397 Zaitzeff, j ................. 1629 Soden, D. L ................ 891 van der Valk, A. G............ 46 Zegowitz, V ............... 1400 Sorenson, S ............... 1612 Vauthier, P ................ 1029 Zeller, R. W ................ 905 Spears, M. A ............... 1173 Veentjer, j ................. 1487 Zielinski, A ................. 287 Spooner, R. L ............... 283 Vetter, W. j ................ 1540 Ziemer, F ................. 1212 Sprunk, H ................. 1286 Vik, S. F .................. 1470 Zikakis, J. P ................ 1608 Srivastava, S. K .............. 485 Villemarette, G. P ............ 298 Zimmer, R ................. 794 Stamulis, D ................. 623 Voudouri, E ................ 641 Zippin, J. P ................ 1615 Stancampiano, j ............. 709 Waddell, P. J. A ............. 833 Stanford, H. M .............. 745 Wade, T. L ................ 1198 Stang, P. R ................ 1616 Wagner, J .................. 518 Steeves, G .................. 567 Wainwright, P. F ............ 405 Steffy, D. A ................. 235 Waldmann, C ............... 497 Stein, S. M ................ 1642 Walsh, E. j ........... 1697, 1704 Steinhauer, W. G ........... 1630, Walters, R. A ............... 1405 1631, 1632 Walther, M ................. 454 Stevens, P ................. 1400 Ward, j ..................... 637 Stewart, G ................. 405 Weaks, A L ............... 1626 Stewart, H. B., jr ............ 840 Weaver, R. j ............... 1362 Stewart, L. L .......... 1286, 1610 Weiss, P .................. 1025 Stockdreher, E .............. 1212 Weissman, D. E ............ 1546 Stoddard, A ................ 942 Wells, J. M ................ 1305 Storey, D. A ............... 1348 Welsh, B. L ................ 1633 Strahl, D ................... 549 Welsh, R .................. 1259 Strandquist, J .............. 1124 Werme, C. E .......... 1631, 1632 Stumpf, R. P ........... 808, 1569 Wheaton, F. W ............... 76 Sturgeon, W.. j ............. 1163 Whelan, W. T ............... 650 Sullivan, P. K ............... 606 Whitehead, J. R ............ 1507 Sutherland, W. C ............ 632 Whittaker, H ................ 853 Swanson, J. C ............... 740 Whyte, R ................. 1646 Swanson, R. L ............... 794 Wicklund, R. I ......... 119, 1614 Sweet, W. E ................ 202 Wilber, R. j ................ 1,639 Swift, M. R ................. 588 Wilcox, R ................. 1025 Tadjvar, A ................. 1628 Wilkinson, D ................ 90 Tattersall, J. M ............. 1618 Willard, M. R .............. 1625 Tauxe, j .................. 1018 Williams, K. L ............... 428 Taylor, P. R ................ 803 Williams, R. G .............. 346 Teal, J. M .................. 177 Wilson, B ................. 1202 Teel, S .................... 1582 Wilson, D .................. 359 Tennyson, E. j .......... 853, 857 Wilson, E ................. 1178 Tester, P. A ................. 808 Wilson, W. B ............... 629 Thomas, A ................ 1045 Wingert, R. C ............... 150 Thomas, W. M., jr .......... 1656 Winslow, T. S .............. 1390 Thomasson, M. P ............. 70 Wolniakowski, K. U .......... 994 Thompson, B. G ........... 1613 Woodward, W. E ........... 1348 Thurberg, F. P .............. 736 Woolweaver, G ............ 1265 Tompkins, M. E .............. 35 Wright, A. St.C .............. 323 Traub, G. L ................ 1493 Wright, D. A ................ 987 Treacy, S. D ................ 180 Wu, L ..................... 287 Tremblay, H ................ 522 Yamazaki, T ............... 1067 Trout, T. K ................ 1668 Yee, A C .................. 555 Trulli, W. R ................ 1630 Y-Hsieh, T. C. T ............. 84 Trutneff, H. F ............... 902 Young, A. G ................ 423 Tsuzuku, T ................ 1061 Young, D. R ................ 745 Turner, E ................. 1616 Yu, D. W ................. 1098 Turner, R. E ................. 41 Zachritz, W. H., Il ............ 84 Tuxhorn, R. L ............... 691 Zagel, J. j ................. 1642 xxx. U.S.COAST GUARD'S NEW POLAR ICEBREAKER DESIGN LCDR Michael S. Canavan CDR Mark D. Noll Office of Acquisition Office of Engineering U.S. Coast Guard Headquarters U.S. Coast Guard Headquarters 2100 Second St., S.W. 2100 Second St., S.W. Washington, D.C. 20593 Washington, D.C. 20593 Abstract 1. Introduction The United States Polar The Coast Guard has completed Icebreaker fleet operated by the a Contract Design for a new class U.S. Coast Guard will be reduced of polar icebreaker intended to to two ships, the Polar Star and replace the last of the aging the Polar Sea, in November 1988 WIND Class icebreakers which are upon the decommissioning of the scheduled to be decommissioned Northwind, the last of the WIND during 1988. This design Class icebreakers. The Coast represents a multi-mission Guard has recently completed a icebreaker capable of operating Contract Design for a Polar in both polar regions providing Icebreaker Replacement(PIR) to escort, logistics support and take the place of the scientific research support as decommissioned WIND Class well as traditional Coast Guard icebreakers which were built in missions such as search and the 1940's. This design rescue and enforcement of laws represents a multi-mission and treaties. icebreaker which incorporates The evolution of this design capabilities defined by the needs began in 1984 with the Polar of DOD, NSF, NOAA, UNOLS, MARAD Icebreaker Requirements Study[l) and the USCG. The design which was a joint effort of the incorporates proven technology to Coast Guard, Navy, National provide a reliable platform to Science Foundation, NOAA and perform escort, resupply and Maritime Administration. This scientific missions in both polar study recommended that the Coast regions. The extensive science Guard immediately begin work on support incorporated in this the design of a new polar vessel represents a major step icebreaker and that consultations forward in providing science be held with the Polar Icebreaker capabilities for polar regions User's Council in considering all research. Other interesting aspects of the design. The study aspects of this vessel include a also recommended that the new bow thruster, AC cycloconverter design contain "enhanced research propulsion system and helo support" capability as well as securing and traversing systems. 11 essential escort and logistics The characteristics and support capabilities" and an capabilities of the vessel as "icebreaking capability between a well as a brief synopsis of the WIND Class and a POLAR Class" design process completed at the icebreaker. time of the preparation of this paper will be discussed. United States Government work not protected by copyright 1087 Work on the design began in were prepared. Extensive model the fall of 1984 with the tests and analytical studies were completion of the Sponsor's conducted during Preliminary Requirements Document(SRD), which Design to determine the optimum detailed the essential hull form, power requirements, capabilities and characteristics superstructure and antenna of the PIR as a set of design arrangements and icebreaking requiements representing the capability. The PIR design needs and desires of all U.S. process was concluded in January icebreaker users. A Feasibility 1988 with the completion of the Study, completed in December Contract Design phase which 1984, was then conducted by the included resolution of the Coast Guard to determine if the remaining engineering problems requirements of the PIR could be and developement of the contract satisfied by utilizing the specifications and drawings. existing POLAR Class icebreaker Throughout the design process, hull. The Feasibility Study comments and recommendations were concluded that the use of the solicited from the User's Council POLAR Class hull would, by on a regular basis. Many of limiting the size of the ship, these became requirements were greatly restrict the ability to incorporated in the design where provide all of the capabilities feasible. desired in the PIR and a new design was recommended. A 2. Characteristics parallel study, conducted under contract to the Coast Guard, to The PIR has an endurance of determine the suitability of 37,000 miles at 9.25 knots open existing foreign polar ocean steaming. It will have the icebreakers concluded that while capability of breaking 4.5 feet many design features of modern of level polar ice at 3 knots foreign icebreakers should be continuous speed or 8 feet of considered in developing the PIR polar ice by backing-and- design, no foreign icebreaker ramming. To achieve these goals, design had the capability to the PIR design includes a hull satisfy the requirements for the form similar to the Polar Class PIR. Based on the results of icebreaker hulls (CGC POLAR SEA these studies, a Conceptual and POLAR STAR), twin screws, Design for the PIR was initiated twin rudders, LOA of 460 feet, with completion in September LWL of 401 feet, full load 1985. The Conceptual Design displacement of approximately Report concluded that the 17,700 tons, maximum beam of 94.5 required capabilities of the PIR feet, and 30,000 SHP. The liquid including endurance, icebreaking load capacity includes 1,410,000 and science facilities could be gallons of marine diesel fuel and incorporated in a single design 46,000 gallons of JP-5 fuel which and a basic hull form was will provide an endurance of 80 developed. This was followed by days underway. Accommodations a Preliminary Design[21, are provided for 17 officers, 120 completed in September 1986, enlisted, 30 scientists and 1 during which the majority of the visitor. engineering problems were solved, The PIR is designed to operate the final hull form was developed in the extreme environmental and general arrangements plans conditions which will be 1088 encountered during equatorial will be mounted to isolate any passages and in polar regions. vibration. A quiet ship Air temperatures ranging from 95 condition can be achieved for deg F (summer) to -50 deg F science operations by securing (winter) and sea water. the main generating plant and temperatures ranging from 85 deg supplying power with the port F (summer) to 28 deg F (winter) generator. are the operating extremes With twin screws, twin rudders designed into the PIR. The and a bow thruster, vessel is designed to survive 100 maneuverability and steerage for ,knot beam winds, severe topside the PIR is greatly enhanced. icing and meet a two compartment This improves the operations of floodable length standard for the ship for docking evolutions damaged stability. Table 1. in port, for operating in contains a comprehensive list of restricted waters and for the PIR characteristics. icebreaking evolutions. The ability to maneuver near 3. Machinery Plant navigational hazards and to cast ship within a polynia(open water The main machinery plant area in the ice pack) will consists of a diesel-electric increase the number of scientific power generation plant that activities that the vessel may be utilizes four medium speed diesel tasked with during her AC generator sets with two AC deployments. main motors which drive two fixed pitch propellers through AC/AC 4. Structure cycloconverters. The main generators provide power to the The superstructure consists of main motors through the AC/AC six levels and includes a cycloconverters and a common bus helicopter hanger and flight deck allowing any combination of main on the 02 level. The hull generators to power either or structure includes six decks both of the main motors. Power consisting of tank top, lst at 4160 volts is also provided platform, 3rd deck, 2nd deck, through a distribution board to main deck, and 01 level. handle large auxiliary loads such as bow thrusters, hydraulic 5. Aircraft systems, deck machinery and cranes. 4160 volt power is The design includes a flight transformed to 450/230 volts and deck and hangar as well as provided through a ship service securing, traversing and fueling power distribution system for systems for helicopter auxiliary machinery and operations. These features allow lighting. This arrangement the PIR to launch, recover and eliminates the need for a hangar two HH-65 Short Range separate diesel generators to Recovery(SRR) helicopters. The provide ship service power. A flight deck has also been port generator is provided to designed to permit the launch, supply electrical power for recovery and fueling of an SH-60 emergency use or when in port heavy lift helicopter which can with no main generator provide excellent cargo and operating. The port generator is personnel transfer capabilities. located in the superstructure and The TALON securing system, which 1089 is the Coast Guard version of the to open and maintain clear ice Navy RARPOON helicopter securing channels and to escort vessels system, will allow the safe through these ice channels or operation of the HH-65 through the ice pack. These helicopters with reduced flight scientific missions span the deck personnel and in a wider range from counting birds, seals range of ship motions. A cable and krill to taking bottom and winch traversing system will samples and measuring the provide a much safer, quicker and geo-magnetic spectrum. To more efficient means of moving, accomplish the variety of the helicopters between the. scientific research that may be flight deck and the hangar with scheduled for these vessels, the reduced personnel. The flight PIR design has been greatly deck arrangement allows influenced by the scientific scientific operations to take community through several design place during flight operations, meetings during the development interrupted only by the actual of the contract design. passage of the helicopter over The dedicated science the stern during launch and,@. facilities aboard the PIR include recovery. an aft science conning station, two wet labs,,one dry lab, an 6. Electronics electronics lab, a computer/ navigation lab, a high-overhead The communications suite.on. vestibule, refrigeration the PIR will be similar to.that facilities, communications on the Coast Guard 378 Foot high center, a combination endurance cutters and will.: library/conference room, provide all military and ; , recompression and dive gear commercial communications, locker, five portable van capabilities, including.satellite laboratories, a bow boom, an communications External,voice instrument room, winch room, an communications are provided from accommodation ladder with all scientific spaces. handling gear, improved scientist Navigation systems,include.a berthing, two oceanographic Global Positioning System(GPS), winches, two trawl/core winches LORAN, OMEGA and inertial.@ and a cable maintenance system. navigation. These systems The science equipment and spaces provide the PIR with exceptional incorporated into the PIR are far navigational and positioning superior to the older icebreakers data. Science electronic systems which the scientific community will be discussed in a later have utilized in the past. These section. features are described in more detail below. 7. Science Facilities Aft Science Conning Station: The PIR science facilities and During science operations over .capabilities are a significant the stern, control of the ship improvement over past U. S. can be transferred to the aft icebreaker designs. Deployments science conning station to allow to the Arctic and to Antarctic precise maneuvering of the vessel regions very often have during.science ops. This station scientific missions combined with is located on the 01 level aft of the Coast Guard primary mission the deckhouse and extends the 'V90 L/I full width of the vessel. This the vessel is hove to in ice for space has a conning station, science ops, relative motion three winch/crane control between the ship and the ice floe stations, a winch/wire data will not create excessive stress processing station, a chart table on the accommodation platform and a computer work station. The because its shape allows the conning station provides platform to slide easily across propulsion and steering control the ice. Safe and easy access to while providing the operator with and from the ice will be maximum visibility of the work possible. In an emergency, the area astern. Minor ship system is capable of lifting the positioning adjustments to accommodation ladder with a 180 account for wire lead, sea lb person standing on the lower conditions or personnel safety platform. can be readily made from this improved aft vantage point. Science Labs: More than 2800 sq ft of deck area is dedicated to Bow Boom: A portable bow boom science labs, staging vestibule holds 200 lbs of atmospheric and and walk in freezer. The two wet acoustic equipment 10 feet and one dry science labs have a forward of the bow rail at a 0 to combined deck area of 1600 sq ft 15 degree angle above and are located on the main deck horizontal. The boom provides aft. To improve the adaptability the advantage of positioning of the spaces for differing instrumentation ahead of the science projects, the vessel to be able to record laboratories can be subdivided to icebreaking reactions forward of adapt to different requirements the vessel and to measure sea on each deployment. The wet labs conditions forward of the vessel include one 20 cu ft upright prior to vessel interactions with freezer and one 20 cu ft upright seaway. A catwalk capable of refrigerator along with other supporting two 180 lb persons at miscellaneous furniture and the tip of the boom in sea state science equipment. The 4 conditions or less will be accessibility of the labs for integral with the boom for access retrieved samples is greatly to equipment without removal of improved via a vestibule staging the bow boom. area. A walk-in freezer of approximately 200 sq ft will Accommodation Ladder: In the provide storage for core past, access and egress from the samples. A computer/navigation ice was via a Jacob's ladder or lab and anelectronics lab are scramble net or a cumbersome also provided and will be accommodation ladder. The discussed in the science accommodation ladder for the PIR electronics section. will be integrated with a To complement these lab handling system and two portable spaces, area has been set aside ice platforms which will enhance to accommodate 5 portable science the utility of the accommodation vans, three on the stern under ladder in port and for science the flight deck and two in the ops on the ice. The platforms forward cargo hold on the second attached to the ladder are dished deck. These vans can become a shape on the bottom to minimize temporary laboratory or storage ice resistance when in use. When space for the scientist and will 1091 receive ship service power, water complete, a dedicated science and heat. A standard size van(20 computer, anticipated to be at ft ISO) can be pre-outfitted by a least as capable as a VAX 11/750 scientific group prior to arrival mini-computer, will be provided at the ship and then simply together with terminals, printers loaded into position, secured and and plotters throughout the hooked up to facilities. The science spaces. A Precision vans could be loaded at home port Bathymetric Survey System(PBSS) or at a foreign port depending on will provide detailed contours of the designated logistics for the the sea floor. Current speed and vans. direction to approximately 600m Approximately 7700 sq ft of will be available from a Doppler weather deck space is available Current Profiler. Voice and as working space for conducting data(56 kilobyte) communication science. Both the main deck will be provided via INMARSAT. A working area aft and the interior Ships Data Video Monitoring science labs will a grid tiedown System(SDVMS), with video system for securing various displays throughout the ship, equipment not normally carried will display information on time, onboard. position, course, speed, distance, weather and general Vestibule: Often, the in past, informational messages. A bottom sample grabs or rossettes computer/navigation lab of with sea water samples had to be approximately 250 sq ft, located removed from the samplers on the as close as posible to the center open deck in a harsh environment of ships motions to minimize before moving to the laboratories problems with the computer for storage or analysis. This is equipment, and an electronics lab not necessary in the PIR design of approximately 300 sq ft are because of the vestibule which is provided for science uses. accessed by a roller curtain door so that scientific equipment can Library/Conference Room: This be moved from the weather to the area is a dedicated space for science labs via the vestibule science oriented discussions or (or staging area). The meetings. This space is located high-overhead of 16 ft in the on the 01 deck aft, in close vestibule with overhead trolley proximity to the science labs, then provides an enclosed work and is equipped with a conference area with a workbench and a sink table, file cabinets, TV and protected from the elements. chalkboard. This space is far Preparation of scientific superior to that provided for use experiments for use over the side as a science office on past or for storage and testing in the icebreakers. Scientific teams laboratory can be handled safely aboard the icebreaker will be and comfortably within the able to meet undisturbed to vestibule. organize their experiments without interferring with the Science Electronics: An ship's operations or daily extensive suite of elctronics has routine. been provided for scientific purposes. While selection of a Scientist Berthing: The chief computer system will not be made scientist's stateroom will be until construction is almost located on the same deck and in the vicinity of the CO's cabin. 1092 It will provide over 400 sq ft of For the oceanographic winches: living and office space. The High speed, low pull: remainder of the scientists will 250-800 ft/min live aft where there is one Low speed, high pull: one-person stateroom and 14 155-500 ft/min two-person staterooms located in the vicinity of the library/ For the trawl/core winches: conference room. Each of these High speed, low pull: staterooms provides approximately 150-500 ft/min 160 sq ft of living space. Low speed, high pull: 95-300 ft/min Cranes: The PIR will be equiped with cranes capable of servicing The winch consoles include a approximately 95 percent of the processor, color monitor and weather deck area. In addition joystick control. The wire to the normal ship operations instrumentation monitoring system such as lowering small boats and and a Data Acquisition and handling cargo, the cranes will Control System (DACS) with system be utilized in place of A-frames control and data recording for handling the oceanographic functions reside at the console. wires. The cranes will also be The monitoring system operates in able to handle loads up to 15 real time with a data update 10 tons at approximately 70 ft from times per second via a data the side of the ship. multiplexing system. There are seven remote displays in the Winches: To accommodate the science conning station, myriad of oceanographic programs pilothouse, vestibule, wet labs 1 that may be scheduled, some of & 2, dry lab and the which are performed computer/navigation lab. simultaneously, the PIR will have four science winches: two Cable System Maintenance: This oceanographic winches and two system consists of a wire brush trawl/coring winches each having cable cleaner, pressurized water 10,000 meters of wire rope or wash, air dry unit, and pressure electromechanical (EM) cable. lubricator which can be used at The ocean winches are all cable speeds of 60-100 ft/min. identical and have eight The wire brush removes dirts, interchangable drums. The rated salt, scale, marine growth; the pull is 10,000 lbs. The fresh water and air washes and trawl/core winches have a rated dries the cable; the lubricator pull of 40,000 lbs. Each winch applies a liquid lubricant of is controllable from either the varying viscosities at 100 psig. winch room or remotely from the This system will reduce the total science conning station or from wear experience by the gear and their respective cranes. The should greatly extend its useful winch station includes: (1) CCTV, life. This system also includes (2) VDT display and (3) crane a wire rope and cable emergency controls. Each winch has a severence system to cut the wire variable speed motor to allow in an emergency using a adjustments for different guillotine device. equipment loads, line speeds and depth of water. An example of the varying speeds available is: 1093 Diver Support System: A dive vessel with polar icebreaking gear locker with a minimum deck ability and marine research area of 450 sq ft will be located facilities which have not been on the main deck. This area will available in any previous U.S. provide enough space for the polar icebreaker. The design of installation of a portable the PIR is a compromise which decompression chamber while responds to a diverse group of deployed. A washroom, specific and in many cases watercloset and shower will be conflicting desires. It integral with this space. Clear satisfies all sponsor deck area is provided near the requirements which represent the dive locker and into the needs of the U.S. icebreaker decompression area to accommodate users. a recompression chamber. This With only two polar area will serve as an excellent icebreakers remaining in the U.S. staging area for any dive fleet as we are entering the operations required during 1990s, the addition of the PIR is science operations. The ability of great importance to U.S. to carry a decompression chamber interests in the Polar regions, will add a degree of safety never especially the polar research before available on a U.S. polar community. The extensive science icebreaker. capabilities of this ship will give it the ability to make major 8. Summary contributions toward our knowledge of the polar regions The Coast Guard is currently well into the next century. modifying the PIR Contract Design Specification into a performance References: oriented document which is being called a Design-Based Performance [11 "United States Polar Specification. This will allow Icebreaker Requirements Study", industry more latitude in their Interagency Report, July 1984, ability to provide alternative U.S. Coast Guard, U.S. Department design solutions and an of Transportation. opportunity to incorporate innovative ideas in their [2] "Polar Icebreaker construction proposals. It is Replacement Preliminary Design anticipated that a draft request Report", September 1986, Naval for proposal will be issued in Engineering Division, Office of the spring of 1989 for industry Engineering, U.S. Coast Guard comment, followed by a request Headquarters. for proposal to build in the summer of 1989. This should result in a contract award in July of 1990 and an operational icebreaker in late 1994. This paper has attempted to provide a broad overview of the Coast Guard Polar Icebreaker Replacement design characteristics and capabilities as well as the design process. The PIR design represents a 1094 2ABLE 1. POLAR ICEBREAKER REPLACEMENT CHARACTERISTICS LOA 459.5 FT LWL 401.0 FT BEAM(MAX) 94.5 FT BEAM(WL) 88.1 FT DRAFT 31.6 FT DISPLACEMEJNT(FULL LOAD) 17,700 LONG TONS SHAFT HORSEPOWER(SHP) 30,000 SHP LIQUID LOADS DIESEL 1,410,000 GAL FRESHWATER 24,000 GAL JP5 46,000 GAL BALLAST/TRIM(SW) 335,000 GAL L/O 24,000 GAL ICEBREAKING, PERFORMANCE 4.5 FT CONTINUOUS AT 3 KTS 8. 0 RAMMING TRANSIT SPEED (OPEN WATER) 12.5 KI!S SERVICE &'4VIRON_MFA4T AIR TEMP -50F TO 95F SEA WATER TEMP 28F To 85F BEAM WIND 100 KTS ENDURANCE 80 DAYS (23 FULL POWER) 180 DAYS PROVISIONS MEETS TWO COMPARLAENT FLOODABLE LENGTH STANDARD FLOODING ALARMS ON BRIDGE POSITIONING HEADING +/-5 DEGREES WINDS 35 Krs STEADY UP 450 OFF THE B04 SURFACE CURRENT 3.0 KTS SEA STATE 5 HABITABILITY ACCOMODArION PROVIDED FOR 17 OFFICERS (INCLUDES CO) 12 CPO 109 CREW 30 SCIENTIST 1 GUEST INCLUDES BERTHING MARGIN OF 13 PROVIDED--BARBER SHOP, LAUNDRY, SEABAG LOCKER, FOUL WEATHER GEAR ST04AGE, C"ONFERENCE ROOM, LIBRARY, GYM, SELF SERVICE LAUNDRY,,CLASS 2A MEDICAL FACILITIES 1095 'TABLE 1. (CONT) FLIGHT OPERATIONS PROVIDED--HANGER FOR 2HH-65A HELOS, HELO SECURING AND TRAVERSING, 450 HOURS OF FUELr HELO CONTROL STATION, AUTOMATIC FIREFIGHTING, AND NON-INTERFERENCE WITH OPERATIONS ON FANTAIL MAIN PROPULSION AC-CYCLOCONVERTER-AC DIESEL ELECTRIC PLANT (SHIP SERVICE ELECTRICAL POWER PROVIDED BY MAIN GENERATORS) 2 FIXED PITCH PROPELLERS 2 RUDDERS AUXILIARY SYSTEMS TRIM TANKS ANTI-ROLL TANK (PASSIVE) HEELING SYSTEM INPORT GENERATOR LOW FRICTION HULL COATING BOW THRUSTER TOWING WINCH 1 ARCTIC SURVEY BOAT 1 LCVP 2 RHIB'S LIFERAFTS FOR 219 DIVING LOCKER W/AIR COMPRESSOR PORTABLE HYPERBARIC CHAMBER NAVIGATION CONNING STATIONS PORT, STBDI MIDSHIPS ON BRIDGE, OCEANOGRAPHIC DECK, ALOFT MAX STERN, FLIGHT DECK AND ENGAGED SIDE VISIBILITY LORAN, OMEGA, GPS AND INERTIAL NAVIGATION EXTENSIVE COMMS SUITE WHICH INCLUDES INMARSAT SCIENTIFIC SUPPORT 1 DRY LAB-512 SQ FT 2 4ET LABS AND VESTIBULE-1568 SQ Fr 1 INSTRUMENTATION LAB-354 SQ FT 1 COMPUTER/NAV LAB-412 SQ FT BOW INSTRUMENT BOOM 4 WINCHES (HYDROGRAPHIC, CORING TRAWLING) OPEN DECK SPACE-1770 SQ SIDE -5950 SQ FANTAIL TRANSDUCER SPACE SCIENCE OFFICE/LIBRARY 'SEA BEAM' BATHYMETRIC SONAR ICE ACCESS VIA SPECIAL ACCOW40DATION LADDER CRANES -CAPACITY 15 TONS -COVERAGE 95% OF DECK SPACE, OVERSIDE 15 TONS AT 64 FT REFRIGERATED SPACE-FREEZER 320 SQ FT DARKROOM 1096 TABLE 1. (CONT) SCIENTIFIC SUPPORT (CONT) COMPUTER SYSTEM CARGO CAPACITY -1300 CU FT APT -34,700 CU FT FWD HOLD W/O VANS -30,000 CU FT WITH VANS ACOUSTIC DOPPLER CURRENT PROFILER SHIPS DATA VIDEO MONITORING SYSTEM CLOSED CIRCUIT TELEVISION WARTIME CAPABILITIES 6 MACHINE GUN MOUNTS ARMORY DEGAUSSING CBR WASHDOWN 1097 ELECTROSLAG SURFACING FOR CONSTRUCTION, RESTORATION, AND REPAIR OF SHIP STRUCTURES D. W. Yu and J. H. Devletian Oregon Graduate Center 19600 N.W. Von Neumann Drive Beaverton, OR 97006-1999 ABSTRACT (2,3). This has created a fiercely competitive environment for the dwindling U.S. Naval con- With construction of new commercial ships in U.S. tracts. With construction of new commercial shipyards at an all-time low and Congressional ships in U.S. shipyards at an all-time low and appropriations insufficient to maintain a U.S. Congressional appropriations insufficient to fleet of 600 ships, the priorities of the surviv- maintain a U.S. fleet of 600 ships (4), the pri- ing U.S. shipyards are changing from that of ship- orities of the surviving U.S. shipyards are building to ship rebuilding, restoration and re- changing from that of shipbuilding to that of pair. ship rebuilding, restoration and repair. This paper presents a review of the international Various surfacing processes have been utilized to literature on the most recent developments in repair and rebuild corroded or worn ship compo- thick section surfacing by electroslag surfacing nents. For many years, SAS with strip electrodes (ESS) using strip or wire electrodes. The advan- was considered the most cost-effective method to tages of this newly-developed technique from Japan overlay large components, such as ship propeller are explained in comparison with the conventional shafts, and now still prevails in the United surfacing processes, such as submerged arc surfac- states. The Japanese and Soviet shipbuilders, in ing (SAS). A number of innovations and applica- particular, have developed highly cost-effective tions in this area are introduced to emphasize the methodologies to rebuild large ship components substantial economical advantage of strip ESS for using an innovative concept known as "Electroslag ship repair and manufacturing. Surfacing". ESS with strip electrodes is capable of overlaying In 1980, Kawasaki Steel (5) of Japan first devel- a wide variety of corrosion and/or wear-resistant oped a reliable strip'ESS process, which rapidly deposits on structural ship components with half spread throughout Japanese industries (6). Sev- the dilution level and twice the deposition rate eral Western European countries also adopted this of its closest competitor, SAS. Because of its process and are commercially manufacturing stan- significant economical merits, strip ESS has al- dard ESS equipment (7). ready become the dominant thick-section surfacing process in many industrialized countries, particu- Strip ESS exhibits substantial advantages over larly in Japan, the Soviet Union and parts of strip SAS in the areas of process control, sur- Europe. facing quality and economic productivity. It has completely replaced the less economical surfacing methods in Japan and the Soviet Union, but has 1. NOMENCLATURE yet to be "discovered" in U.S. shipyards and man- ufacturing industries. ESS Electroslag Surfacing ESW Electroslag Welding The purpose of this study is to critically review ESR Electroslag Remelting the international literature on ESS and strip SAS Submerged Arc Surfacing ESS. Of particular emphasis will be the flux SMAW Shielded Metal Arc Welding chemistries and surfacing parameters that are GMAW Gas Metal Arc Welding associated with this processing innovation. The advantages of surfacing with the strip ESS method 2. INTRODUCTION are reported. The future requirement for new ships forecast by 3. CHARACTERISTICS OF ELECTROSLAG SURFACING the Association of West European Shipbuilders (1) implies over a third of the world's shipyard ca- Although the strip ESS process is new, the funda- pacity active in 1985 will have to close if it is mental, principle of ESS is similar to that of the to be brought into line with demand. Unfortunate- Electroslag Welding (ESW) and Electroslag Remelt- ly, international competition and foreign labor ing (ESR) processes. Heat is generated by ohmic rates have pot virtually all commercial ship@uild- heating of a resistive slag by the passage of an ing contracts out of reach for U.S. shipbuilders electric current through a strip electrode, which CH2585-8/88/0000- 1098 $1 91988 IEEE is continuously fed into the molten slag pool. more deeply into the workpiece to increase the Figure 1 shows a schematic diagram of the ESS dilution ratio. On the other hand, the strip ESS process. process allows the use of almost double the cur- rent density to produce.a much higher deposition rate while still maintaining a lower dilution level. This desirable combination of a high depo- sition rate and a low base metal dilution was the main incentive for Japanese industries to elimi- Flux delivery Feed rolls nate SAS in favor of ESS. Strip electrode The third important feature of ESS is the feasi- bility of single layer deposition. By virtue of its low dilution and high deposition rate, sur- Contact shoes facing can be most economically attained for the desired thickness of a corrosion or wear resis- tant layer with a designed chemical composition. Flux Solidified Since the dilution level for strip ESS is almost slag half that of strip SAS, the strip ESS process can M it .;'Weld deposit more likely eliminate the necessity for multiple la layer deposits and result in greater cost effec- tiveness. Furthermore, thin overlays (about 3 mm L or 1/8 in. thick) are far more advantageous by e =-et-l ESS because dilution decreases with overlay thickness for ESS but increases by SAS. Further economical advantage is gained by the use Figure 1. Diagram of the electroslag surfacing of wide strip which deposits a greater surface process. area per unit time. Large strip widths (> 60 mm E2.4 in.]) are particularly more difficult to ap- ply by SAS than ESS. In the SAS process, the arc The process appearance of strip ESS is nearly is struck at one corner of the strip and then identical to that of strip SAS, except SAS is pri- starts traversing the entire width of the strip marily arc-functioning while ESS is arcless and (5,9). The strip is consumed by the oscillatory produces heat by 12R (ohmic) heating of the mol'ten movement of the arc across the strip. However, slag. However, strip ESS exhibits a series of in ESS the strip is consumed uniformly across its advantages over strip SAS in providing low dilu- entire width. The movement of the arc in SAS is tion deposits, high deposition rates and better not necessarily uniform and leads to inconsistent productivity. penetration and lack of fusion. For this reason, SAS has been limited to a strip width of 75 mm, The first important feature of strip ESS is low whereas using strips as wide as 300 mm (11.8 in.) dilution in the deposits. In any surfacing proc- is not uncommon in ESS. A comparison between the ess, a critical factor requiring precise control strip ESS and SAS process is presented in Table is the dilution ratio, expressed as: I. % Dilution = B/(A + B) x 100 4. FURTHER INNOVATIONS IN ESS FROM JAPAN where A is the cross sectional area of reinforce- a. External Magnetic Field for ESS ment of deposits above the base metal surface, and B is the cross sectional area of the melted base In 1980, Nakano and his colleagues in Japan (5) metal below the workpiece surface. In terms of first developed an electromagnetic controlled surfacing, it is necessary to keep low dilution strip ESS method called the MAGLAY process. Dur- levels because the surfaced layer has to maintain ing surfacing with wide strips (> 60 mm [2.4 its desired inherent properties, like wear or cor- in.]), the formation of undercutting and lack of rosion resistance. In SAS, the arc tends to pene- fusion defects were found to be related to the trate more deeply and melt more base metal in com- flow pattern of molten slag and metal, which is parison to ESS. For example, the strip SAS proc- driven by the electromagnetic force induced by ess typically produces a dilution ratio of 18% at the high values of the surfacing current. Elec- a current density of 25 A/MM2 (16.1 kA/in2), com- tric current, flowing parallel from the strip to pared to approximately a 9% dilution ratio for the the bottom of the molten pool, makes both slag strip ESS process at 41 A/mM2 (26.5 kA/in2) (8). and metal move from the edges of the pool toward Thus, the chemical composition of the overlay de- the center. To counteract this force in the posited by ESS will more closely resemble that of MAGLAY process, two direct current coils are the filler metal. mounted adjacent to the edges of the strip elec- trode resulting in counterbalancing magnetic The second important feature of ESS is its high forces. The use of an external magnetic -field deposition rate, which is a function of the cur- effectively (a) avoids undercut at the bead tie- rent density. The use of a high current density ins, (b) eliminates slag entrapment, and (c) pro- in the SAS process will effectively make the arc duces a more uniform thickness of overlay. o en S g hotter and stiffer, thus causing it to penetrate 1099 Table I. Comparison between submerged arc to submerged arc due to an increase in slag re- surfacing and electroslag surfacing with stainless sistivity with decreasing slag superheating. The steel strip electrodes (8). inclination angle of the electrode permits molten metal to enter the gap between the base metal and electrode and produces a buffer by preventing SAS ESS deep penetration into the base metal, and reduces the dilution of the surfacing layers. 5. INNOVATIONS IN THE SOVIET UNION dimension (mm) 60 x 0.5 60 x 0.5 carbon content M 0.015 0.015 The ESS process is also widely used in the Soviet Union and Eastern Europe. A great amount of in- Parameters: novations were frequently reported in their tech- I TAT 750 1250 nical journals. Although most articles are often V (V) 26 24 lacking technical details, their basic designs v (cm/min) 10 16 and functions could still be reviewed. Current density (A/mm2) 25 41.7 a. Multi-Strip Feeding Heat input (KJ/cm) 117 112.5 The Paton Welding Institute started studies on (KJ/cm2) 19.5 18.7 ESS with two electrode strips in the late 1970's (11-13). The advantage of this dual strip feed- Bead thickness (mm) 4.5 4.5 ing process is a substantial increase in deposi- tion rate. When two strips are fed in parallel, Dilution W is 9 the molten slag may rise between the two strip electrodes and directly contact with air, causing Deposition rate (Kg/h) 14 22 considerable convectional agitation. Thus, the distance between two strips has become another Flux consumption 0.65 0.5 important parameter to be taken into account. Carbon content of single ESS with more than two wire electrodes has also deposit layer (base been reported by V. Melikov (14). As many as 15 metal: 0.18% C) 0.045 0.030 stainless steel electrodes (all 3 mm [1/8 in.] diameter wires) were simultaneously deposited over the entire 340 mm (13.4 in.) width of the workpiece in the downhand position with artifi- b. "PZ" Arc-Facilitating Process cial cooling. Strip surfacing at the Japan Steel Works also Shvartser (15) claimed two hardfacing processes utilizes the electroslag mode of deposition but with a group of plate electrodes. In one case, without the aid of magnetic devices (6). Their the high Mn steel electrodes were deposited on process is called "PZ". The important feature of worn dredger buckets. In another case, the high this process is that an arc is always maintained Cr casting iron electrodes were deposited on worn at the strip extremities while most of the strip steel blades, as shown in Figure 2. The absence tip is still in the electroslag mode. The auxi- of cracks and formation defects made it possible liary arc facilitates bead tie-in and penetration, to greatly increase the service durability of but avoids excessive dilution at the center of the hardfaced components and reduce the production bead caused by the Lorentz force. The 150 mm (6 cost in comparison with brazing expensive alloy. in.) wide strips used in the "PZ" process provide a uniform overlay surface and a low dilution level Ve in each bead. c. "HS Process" Kobe Steel Ltd. developed the high speed overlay welding technique called the HS Process (High 4 F77 Speed Strip Overlay Welding Process) which util- izes a strip that can be applied to an actual ves- sel even as a single layer process [10J. In terms of efficiency, the HS technique is claimed to utilize only a 75 mm (3 in.) wide electrode but competes attractively with the ordinary ESS proc- ess using a 150 mm (6 in.) wide electrode. The key points of this technique are a high travel speed and a forward electrode inclination angle. Figure 2. The hardfacing of buckets: the dia- UsualT, when the electrode travel speed exceeds gram of the process 1) component, 2) electrode, 200 mm/min (7.8 in/min), the electrical transfer 3) deposited metal, and 4) solidification mold through the slag pool shifts from electroslag (15). 1100 b. Plasma-Electroslag Deposition 2 A plasma-ESS method was reported by Batakshev et al. (16) to deposit high purity copper on low al- loy carbon steels. The requirement of an auxil- 3 iary plasma is to counteract the high conductivity of the copper overlay. In this process, the pilot arc is initially ig- nited in the plasma torch, followed by ignition of the plasma arc between the workpiece and the elec- trode of the plasma torch. A special flux is fed into the acting zone of the plasma jet. This flux contains elements with low ionization potentials (Ca, Na, Ba, etc.), which increase the stability of the plasma jet due to a decrease in electrical 4 resistance. The flux soon melts and forms a.slag pool. When the base plate is heated to a suffi- ciently high temperature, the copper filler metal 5 is fed info the slag pool, and the plasma torch Z and the mold are moved at the same time, resulting in the surface overlay. The plasma ESS process provides a means to control Figure 3. Diagram illustrating the ESS of an ex- the time dependence of heating the parent metal cavator shovel tooth with varying chemical compo- without the use of consumable electrodes. It also sition metal: 1) tooth blank, 2) standard head, prevents contamination of the deposited copper. 3) consumable electrode, 4) mold base, and In a steady-state operation with the optimum pa- 5) frame. rameters of 450-50OA/55-60V and a 30-40 mm (1.4- 1.6 in.) deep molten slag pool, the process could produce a 2-3 mm (0.1 in.) thick and 15-20 mm thermal conductivity and also high resistance to (0.6-0.8 in.) wide deposit in a single pass. the action of molten slag and metal pools. In this case, copper and its alloys do not ensure c. Surfacing of Shaped Parts the required thermal efficiency. He reported an application using a damping heat conducting layer A variety of examples could be found in the Soviet in the solidification molds. The inner contact technical journals, reporting the use of ESS for surface of the mold was made of a less thermally the restoration of worn components having complex conductive alloy steel, followed by a damping shapes. The surfacing of those shaped -parts is heat conducting layer made of pure copper, and performed by a modified electroslag welding proc- finally by the water-cooled structural steel ess. A specially designed water-cooled mold is base. This method provides proper control of the used to confine the molten slag and metal pool cooling rate of the deposited metal, and ensures into the desired shape. Figure 3 illustrates the a uniform temperature distribution in the molds. use of a shaped mold for surfacing the teeth of excavator buckets. d. Surfacing of Thin-Walled Components As Valits indicated (17), to restore a complicated Multi-electrode surfacing usually makes it possi- shape, the energy of the melt must be sufficient ble to deposit, in a single pass, a layer of to ensure both the transfer of the melt to the metal having the required thickness and width remote part of the mold and the complete fusion at nearly equal to that of the components. Although that location. His experimental results verified its use for thin-walled components risks the pos- that an increase in the voltage or the current sibility of burning through, a report from the density improved the quality of the deposited Tashkent Institute of Railway Transport Engineers metal. However, when the current density was ex- claimed the development of a successful example cessive, the slag pool could be "thermally satu- for the ESS of the friction wedge of the damper rated". The thickness of the deposited layer no of wagons whose maximum wall thickness was only 5 longer increased, and the risk of short-circuiting to 6 mm (=- 0.2 in.) (19). was eminent. In this process, nine electrode wires (each 3 mm For large-scale parts, the ESS operations were [1/8 in.] in diameter) are deposited simultane- reported lasting more than 30 hours. To ensure a ously, and an AC (not specified in that paper, high quality of the deposited layer, the process but believed) power source with a hard external must be stable and maintained without interrup- characteristic is used. The main parameters of tion. Stepanov (18) pointed out that the non- this process include: 32 volts, an electrode uniform adhesion of molten metal and the presence feed rate of 0.51 m/hr (20 in/hr), and a surfac- of a slag skull in long term ESS could cause rapid ing speed of 1.8 m/hr (70 in/hr). A few factors abrasive wear and local superheating in the tail are critical to prevent burn-through defects, part of the shaped mold. Thus, the inner surface including the slag pool depth, the electrode ex- of the mold has to be made of materials with high. tension, and the stationary (without longitudinal 1101 displacement) feeding time of electrodes in the chemical composition. They constructed a model initial stage of surfacing for inducing the slag of the molten surfacing pool and proposed ways of pool and the final stage of surfacing for filling reducing the difference between the required and the crater. By increasing the initial stationary actual compositional variation. A programming feed time, the molten filler metal spread ahead of device was also designed to facilitate the gra- the electrode tip thermally protecting the base dient method of surfacing (23). metal. A wear-resistant layer of 6-12 mm (0.24- 0.47 in.) thick and 135 x 180 mm (5.3 x 7 in.) in 6. PROCESS DETAILS size is reported being deposited in a single pass on the surface of mild steel. a. Power Source e. Surfacing Layers With Compositional Gradients ESS is always carried out with direct current, constant voltage (DC-CV) power sources using re- In many cases of service, the different portions versed polarity (the strip electrode is connected of an individual hardfaced work piece experience to the positive terminal of the power source) in different degrees of wear. The geometrical loss order to ensure adequate fusion to the base met- due to the uneven wear reduces its life premature- al. Since the optimal current density for ESS is ly. The rational solution to this problem is to around 40 A/MM2, the output rate of power sources make the working surface from composite metal, at a 100% duty cycle should meet the following whose wear resistance changes gradually to accom- minimum load handling requirements: 1250A for 60 modate the differences in the severity of wear at x 0.5 mm (2.4 x 0.02 in.) strips; 1800A for 90 x different locations on the workpiece. By produc- 0.5 mm (3.5 x 0.02 in.) strips; and 2400A for 120 ing a part that wears uniformly, the functional x 0.5 mm (4.7 x 0.02 in.) strips (8). In prac- life of the part is lengthened. tice, such high current levels are usually ob- tained by connecting two power sources in paral- Shviartser (20,21) developed a special surfacing lel. process to provide a wear gradient for an excava- tor shovel, which is illustrated in Figure 3. In b. Flux Chemistry service, the abrasive wear on its rear face in- creased substantially from the tail end of a tooth In strip ESS, it is very critical to establish to its apex. In order to extend its life, a pre- stable ohmic (arcless) conduction of electricity scribed variation in chemical composition of the through a shallow slag pool of only about 20 mm deposited metal was obtained by depositing a spe- (0.79 in.) depth. To maintain a stable electro- cial composite electrode which consisted of two slag mode through a shallow slag, a special flux dissimilar metals (a high Mn steel and a high composition had to be developed. Such fluxes chromium iron) meeting along an inclined plane must provide greater electrical conductivity than (Figure 4). The surfacing deposits adjacent to would be needed for normal electroslag welding of the front face of the teeth were a wear-resistant the same plate material. Adding large quantities Cr iron, changing (towards the rear face) into a of fluorides, mainly CaF2 and NaF and/or semicon- high Mn steel. In service, those hardfaced teeth ductors, such as TiO, and FeO, can greatly raise maintained a consistent geometry (21,23). the electrical conductivity of molten slag with- out risk of generating arcs. However, large quantities of Ti02 in slag cause a deterioration in the detachability of the slag. Therefore, Metal A Met'al A additions of fluorides are more preferable (5,8). The level of electrical conductivity of slag is closely related to the fluoride content in the flux, as illustrated in Figure 5. The IN (In- ternational Institute of Welding) Document XII-A- 4-81 (24) indicated that in the 3CaO-3SiO,-AI,O, ternary system, when the fluorides were less than 40% (balance ternary), the, submerged-arc mode Metal B prevailed; and when more than 50% fluorides, the electroslag mode prevailed. In terms of the electrical conductivity of the slag, this corres- ponded to a transition range of 2 to 3 2-1 cm-1. Metal B V 17 Above 3 0-1 cm-1, a stable electroslag mode is easily achieved. However, to restrict the gener- ation of fluoride type gases (due to a reaction: Figure 4. Two composite electrodes designed to 2CaF2 + S'02 - 2CaO + SiF4), additions of CaF2 provide surfacing layer with compositional gra- were usually held at slightly less than 50%. dient (20,21). In Japan, fluxes for stainless,steel overlays are principally supplied by Kawasaki Steel and Kobe e'i Recently, Gulakov et al. (22) analyzed both the Stee,l (6). The Kawasaki KFS-150 is a fused flux buffer effect of the weld pool and the element with an electrical conductivity of about 3 2-1 transfer from the base metal (or the previous cm-1 at 1700*C. The composition of Kawasaki's deposited layer) on the final gradient of the patented flux (25) contains 50-60% CaF21 10-20% 1102 Table II. Fluxes Available From Transition European Sources (7,S) Arc Slag Sandvik Content EST 122 EST 201 37S 1 3 - M Alkaline & 50 50 50 E Cie alkaline earth oxides (CaO, 2 - MgO, K20, Na2O) 05 0 CaF, (5) Amhoteric & 25 25, alkaline earth 0 NaF C) oxides (Al 03, TiO,, Zr02 CaO:SiO :A'20,. 2 Z: M M Silicon 10 max 5 max 10 >1 0.8- =3 3 1 dioxide (SiO.) 0 1 (%) Fluorides 30 25 30 (expressed in F-) 0.5 C) Flux density 0.85 0.85 0.85 0 0.4 00 (kg/dM3) 0 0.3 Flux consumption 0.5 0.5 rate (kg flux/kg strip) L4 0.2 0 20 40 60 80 100 Fluoride content(Yo) was once again that of arc conduction. As the percentage of the CaF2 in the flux increased, Figure 5. Effect of fluoride content in flux on there was a corresponding increase in the conduc- electrical conductivity and current conduction tivity level. This, in effect, raised the cur- type during Strip ESS (24). rent at the same wire feed speed, and gave rise to burn back problems. Hence, at higher CaF2 percentages, arcing could be visible on the sur- Si02, 5-25% CaO and 10-30% A1203 in a ratio of face of the molten slag. On the other hand, be- S'02/CaF2 of at least 0.20 and a ratio of CaO/S'02 low 40% CaF21 the observed arcing noticed was the of at least 0.50. submerged-arc type as illustrated in Figure 6. In the Soviet Union, a series of fluxes were de- 8.0 veloped for ESS. The ANF series fluxes are of high fluoride contents (> 50%) and high electrical conductivities (26). The AN-series fluxes, which 7.0- were originally used in ESW, are also used for the thick layer build-up. Their fluoride contents are 4S.0 - below 25% and electrical conductivities are com- paratively low (26). Some new fluxes were occa- sionally reported being developed for certain spe- 5.0- cial ESS processes (27). However, no concrete compositional information was presented. 4.0- In Western Europe, fluxes EST 122 and 201 are com- monly used [8]. Some characteristic data of these 03.0- Submerged Exposed two agglomerated fluxes are given in Table Il. -Arcing table ESS Arcing The flux EST 122 is specifically designed to be 2.0- used for the depositions of all types of 300 and 400 series stainless steel strips. The flux EST 1.0 201 is designed for the deposition of the Ni-base 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0 alloys, such as 825, 600, 625 and 400 (7). X CaF2 In a recent study (28) of ESS with stainless steel wire electrodes, fluxes of the CaF2-CaO-Al2O3 Sys- Figure 6. Effect of CaF2 content in flux on tem were studied. It was noticed that at higher electrical conductivity and current conduction gel table @SS CaF2 percentages, (i.e. beyond 70%), the process mode during wire ESS (28). 1103 Usually, CaO, S'02 and A1203 are common additions and geometry of the surfacing layer as shown in for optimizing the conductivity level of a CaF2_ Figure 7. By increasing voltage, the rising heat based flux. The practice at the Oregon Graduate input increases the volume of base metal melted, Center (28) indicated that optimizing the viscos- thereby increasing the level of penetration and ity level of molten flux is of the same impor- altering the geometry (width/ thickness) of over- tance. Since conductivity and viscosity have an lays. Nevertheless, the dilution level still re- inverse relationship, adjusting the flux composi- mains essentially constant or only slightly de- tion becomes a complex problem. Si02 is one com- creases with increasing voltage. pound which has a major influence on slag viscos- ity and slag flow (25). To maintain the desired Dilution - viscosity, it is necessary to control a Si02/CaF. M ratio of at least 0.2 and to avoid evolution of 15.- toxic gases like SiF4 by the reaction: 10-- 2CaF + Sio = 2CaO + SiF 5- 2 2 4 PenetLation In addition, as specified by the Kawasaki patent (MM) - (25), the minimum requirement for CaO/SiO, ratio 1.5 should be about 0.5 for stable ESS. c. Strip Electrode Sizes 1.0 The thickness of the strip electrode is always Bead width expected to be thin enough to facilitate coiling (MM) 70 into rolls, in order to conveniently feed cladding during ESS. The Japanese appear to have standard- ized the 0.4 mm (0.016 in.) thickness for all 65 strip widths. This differs from the European practice where a 0.5 mm (0.02 in.) thickness is Bead most common. thickness 6 The ESS process favors the use of wide strip as (MM) 5 long as the capacity of the power supply is ade- quate to provide 1000-2000 amps, typically. That 4 is because, at a given layer thickness, the most marked effect of increasing the strip width is a j- decrease in dilution and penetration. Usually, 24 26 28 the penetration of over-lay deposits is always more accentuated at the sides of the bead. However, Voltage (V) the relative importance of this localized higher penetration is lessened when (a) the strip width Figure 7. Influence of voltage for strip elec- is increased, and (b) bead overlapping is con- trode, size 60 x 0.5 mm at 1250A surfacing cur- sidered (8). rent and 150 mm/min travel speed (7). d. Voltage e. Current Voltage is perhaps the most critical controlling ESS has been reportedly used (7) only with DC parameter in the ESS process. In most ESS prac- reverse polarity (electrode positive). At a tice, the working range of voltage values is quite given voltage and surfacing speed, increasing the narrow, because of the shallow depth of the molten ESS current directly increases penetration, bead slag pool (5-10 mm [0.2-0.4 in]). Forsburg (7) width and thickness, but has little effect on reported that, for a fluoride-based flux, the dilution. stable range is usually 26-28 volts. When the voltage is below 24 volts, it is difficult to ini- Stable and quiet welding conditions can be tiate the process, and the strip tends to stick to achieved within a given range of ESS current. the base metal resulting in short circuiting. On The optimum current density for strip ESS is the other hand, above 28 volts, the process starts around 40-45 A/mm2 (26-29 kA/in2). At the higher arcing on the surface of the flux, and slag spat- values of current density, the amount of slag ter becomes violent. Therefore, an accurate con- spatter increases and the depth of the slag pool trol in voltage is extremely important. Yu, et al. has to be raised to stablize the operation. (28) found that the optimum voltage was closely related to the actual depth of the molten slag f. Travel Speed pool , and a stable ESS process could be performed at 22-24 volts. In addition, it was also shown At a given welding current and voltage, increas- (28) that an intentionally increased open-circuit ing the travel speed tends to increase dilution voltage is beneficial to the initiation of ESS. and penetration, while decreasing bead width and thickness (7). Increasing the travel speed in Even within the stable voltage range, fluctuations effect reduces the heat input and, thereby, de- in voltage also affect the dilution, penetration creases the electrical conductivity of slag. The 1104 ESS process can only be stable when sufficient systems have been commercially manufactured in contact area between the molten slag pool and the the United States for many years for submerged melting strip is maintained. An excessively fast arc welding applications, particularly in ship- surfacing speed may cause the strip to be in con- yards. tact with cold flux or insufficiently heated slag, thus resulting in sporadic arcing and process in- 8. CONCLUSIONS stability. Based on a computerized search of the interna- In general, the travel speed should be optimized tional technical journals on the subject of elec- for both economy (fast speed) and an adequate troslag surfacing, a critical review was per- thickness of surfacing layer (about 4-6 mm [=- 0.2 formed and the following can be concluded: in.]) (8). Excessive travel speed results in not only a bead thickness less than 4 mm, but also in 1. ESS with strip electrodes is the most eco- the risk of the formation of undercutting. On the nomical and productive method to overlay a other hand, too slow a travel speed results in a wide variety of corrosion and/or wear resis- bead thickness above 6 mm. Then, the wetting an- tant deposits on structural ship components, gles of beads become too steep and slag entrapment such as propeller shafts. may occur at the overlaps. In general, the opti- mum travel speed range is about 160-200 mm/min 2. The highest deposition rates combined with (6-8 in/min), which results in about a 10% dilu- the lowest base metal dilution are charac- tion level, and consumes about 0.15 kJ/MM2 (96 teristic of ESS with strip electrodes com- kJ/in2) heat input (28). pared to conventional surfacing methods, such as strip SAS, GMAW and SMAW. g. Other ESS Parameters 3. The dominant thick-section surfacing process The strip extension, i.e. the conventionally in Japan, the Soviet Union and several Euro- called "stick-out" (the free length of the strip pean countries is ESS. extending from the contact jaws to the slag pool), is not critical in this process. Usually it may 4. Neither U.S. shipyards nor U.S. manufactur- vary from 25-40 mm (1-1.5 mm) (7). ing industries have adopted the ESS process. Conventional surfacing methods are still The limitation of parent metal thickness depends utilized in the U.S. on the heat input duriM ESS and the width of strips used. Fi - reported that in order 5. Technically, the key difference between the to ensure sound ESS with a 60 x 0.5 mm (2.4 x 0.02 newly-developed ESS. process and other simi- in) strip, the minimum parent metal thickness is lar processes, such as SAS and ESW, is the 40 mm (1.6 in.). The minimum diameter of curved flux chemistry. surface for ESS with the 60 x 0.5 mm strip elec- trodes is 250 mm (10 in.) for external surfacing, 9. ACKNOWLEDGMENT and 450 mm (18 in.) for internal surfacing. This is ideally suited for the rebuilding of ship pro- The authors are grateful to the U.S. National peller shafts. Practices at the Oregon Graduate Coastal Research Institute for financial support Center (28) have found the above limitation was under Contract 51-01-38RO-D. The authors wish to relatively conservative. For example, with a 60 x thank Mr. Tom J. Maginnes and Dr. Earle N. 0.5 mm strip, ESS could be performed on 25 mm (1 Buckley of NCRI for their helpful advice. Also, in.) thick plates. the assistance of Mr. Sin-Jang Chen is appre- ciated. 7. APPLICATIONS OF STRIP ESS 10. REFERENCES Presently, strip ESS is entirely foreign technol- ogy, which has further widened the construction 1. P. A. Milne, "High Technology in Shipbuild- cost gap between the Asian shipyards and U.S. ing", Proc. Inst. of Mech. Engrs., Vol. 201, shipbuilders. However, utilization of this for- No. B2, 1987, pp 59-72. eign technology and the substantial improvements 2. L. M. Thorell and T. Watanabe, "Technical in strip ESS anticipated at the Oregon Graduate Collaboration Between Mitsubishi Heavy In- Center will enhance the economic position of U.S. dustries and Todd Shipyards", in the Pro- shipyards to rebuild worn, eroded or redesigned ceedings of the National Shipbuilding Re- structural ship components, such as large propel- search Program, 1986 Ship Production Sympo- ler shafts, rudder horns, strut shafts, deeply sium, Paper No. 5, Sponsored by SNAME Ship corroded portions of the hull, hawse pipes and Production Committee, 1986. leading edges of rudder castings. 3. T. S. Upham and W. M. Crawford-; "Decentrali- zation--The Management Key to Effective Ac- This process, though fully automatic, is also curacy Control", in the Proceedings of the portable in the shipyard when a conventional (and National Shipbuilding Research Program 1986 inexpensive) carriage system is used to mobilize Ship Production Symposium, Paper No. 4, the strip ESS system in remote locations. A typi- Sponsored by SNAME Ship Production Commit- cal carriage system can handle 1500 amps and can tee, 1986. pull power cables 30 m (100 ft .) long while being 4. `600 Ship Navy Won't be Sustained", Marine either track or manually guided. These carriage Log, January, 1988, pp 49-51. 1105 5. S. Nakano, N. Nishiyama, T. Hiro and J. 22. S. Gulakov and B. Nosovskii, "Special Fea- Tsuboi, "Maglay Process-Electro-Magnetic Con- tures of Production of Welded Joints with trolled Overlay Welding Process with ESW", Regulated Distribution of Chemical Elements Kawasaki Steel Technical Report, No.2, March Along The Length", Welding Production, July (1981), p 31. 1982, p 8. 6. "Welding Technology in Japan", Welding Re- 23. S. Gulakov and B. Nosovskii, "A Programing search Council, N.Y., (1982), p 16-22. Device for Depositing a Layer of Metal with 7. S. Forsburg, "Resistance Electroslag (RES) a Varying Chemical Compos i ti on Welding Surfacing", Welding Journal, August 1985, p Production, July 1985, p 52. 41. 24. S. Nakano, T. Hiro, N. Nishiyama and J. 8. "Electroslag Strip Cladding"--A draft, Der- Tsuboi, "The New Strip Electroslag Overlay likon Industries, (1985). Technique with Electro-Magnetic Control--The 9. V. Pilous and R. Kovarik, "Experimental De- Maglay", IIW Document, XII-A-4-81, (1981). termination of The Metallurgical and Techno- 25. J.Tatershi, T. Ishikawa, S. Nakano and N. logical Parameters in Surfacing with Austen- Nishiyama, United States Patent 4,437,406, itic Strip Electrodes", Welding Internation- (1984). al, No.1, (1987), p 84. 26. B. Paton, (ed.), "Electroslag Welding & Sur- 10. 0. Tanaka, K.'Takeba and Y. Matsushita, "High facin Vol.1, MIR Publishers, Moscow, Speed Overlay Welding With Strip Electrodes", (1983@','Ip 172, Stainless Steel, Nov. , (1985), p 58. 27. K. Val its, A. Shvartser and V. Stoiko, "A 11. V. Mastenko, E. Starchenko and V. Khodakov, Flux For Electroslag Hardfacing of High- Stability of The Electroslag Process in Two Manganese Steel Adv. Spec. Electrometall., Electrode Strip Surfacing", Welding Produc- Vol. 1, No. 4, @'11985), pp 162-165. tion, March, (1982), p 23. 28. D. Yu, Y. K. Oh and J. Devletian, Unpub- 12. V. Mastenko and E. Starchenko, "Effects of lished research, Oregon Graduate Center, Variation in Welding Parameters on Weld Sur- (1987). facing With Two Strip Electrodes", Automatic Welding, July 1984, p 50. 13. P. Blaskovic, S. Lesnak and J. Zajac, "Elec- troslag Surfacing in the, Horizontal Position Using Two Strip Electrodes", Surfacing J. Int. 1, (2), (1986), pp 75-78. 14. V. Melikov, E. Sheinman and M. Brodyanski, "Special Features of the Multielectrode Wide- Layer D 'eposition of Austenitic Layers on Low- carbon Steel", Welding Production, Feb. 1982, p 7. 15. A. Shvartser, "The ESS of Components With a Group of Electrodes", Welding Production, Jan. 1981, p 27. 16. A. Batakshev, S. Berezhnitskii, A. Lamzin and 0. Steklov, "Plasma-electroslag Deposition of Copper on Steel", Welding Production, July 1984, p 17. 17. K. Valits, A. Shvartser, V. Dorokhov and B. Matveev, "The Effect of the Conditions of ESS of the Quality of the Deposited Layer", Weld- ing Production, May 1981, p 26. 18. B. Stepanov, V. Yakovlev, G. Surovtsev and I. Izyur'ev, "Using a Damping Heat Conducting Layer in Solidification Moulds for Electro- slag Surfacing", Welding Production, May 1984, p 39. 19. E. Ashkinazi and E. Sheinmass, "Factors Af- fecting the Quality of Bonding in Wide-Layer Surfacing of Thin-Walled Components", Welding Production, Feb. 1986, p 24. 20. A. S 'havartser, V. Shvarts and Nikitenko, "Calculating the Amounts of Components in a Varying Chemical Composition Alloy During Electroslag Surfacing", Automatic Welding, Feb. 1980, p 17. 21. A. Shvartser, V. Stoiko, Z. Nikitenko and I Morgachev, "Excavator Shovel 'Teeth Recondi: tioned by Surfacing with Varying Composition Metal", Automatic Welding, March.198 5, p'43. 1106 CONVERSION OF SURPLUS OILFIELD SUPPLY VESSELS TO RESEARCH VESSELS James A. Chance Survey Boats, Inc. There are still a great number of oilfield ser- as a supply boat, they can,be used as oceangoing vice vessels held by financial institutes and tugs. Four-point boats have four mooring winches the Maritime Administration. These vessels can with 2,000 to 7,000 feet of cable. They are used be bought at reasonable prices, but the great for diver support and coring on a four-point an- difference in mission requirements between an chor spread. oilfield service vessel and a research vessel can make the conversion difficult. Design goals Utility boats are smaller steel vessels in the 90' and constraints of OSV's are discussed in their to 120' range. They are used for field production relationship to the conversion of th6se'vessels support, diver and ROV support, surveying, main- to research use. The conversion of a:150.' sup- tenance, sandblasting and painting. They are de@ ply Vessel to a research vessel is discusEed in signed to carry much smaller amounts of cargo, 12 detail. to 20 passengers, and to function as work plat- forms in addition to transportation, block coef- ficients tend to be near .60. Because they must The United States oilfield support vessel fleet meet damage requirements, they have more free- is made up of four classes of vessels built to board than supplyboats. The extra freeboard is perform various jobs. Despite being built to usually taken advantage of by reducing the initial very different criteria than typical research stability, which gives a gentler roll. The rear vessels, their low acquisition cost can make cargo deck is usually.in the 50' to 60' range. them interesting candidates for conversion. Some utility boats have focsle cabins giving very good space utilization. Others have a deck house The largest class of vessels and the most avail- and a low bow giving a more practical deck height able for conversion are the supply boats. They forward for diving work and increase visibility. are of steel construction and range from 150' to Because the boats are smaller, tonnage is less of 180' in length. The overriding emphasis in their a problem, and the below deck areas are often used design is deck cargo capacity. They are very for accomodations. Staterooms are not as spacious full in shape with a typical block coefficient as on supply boats, but are of reasonable size. A being in the range of .65 to .70. Because they typical boat would have accomodations for 16 to 22 do not have to meet damage criteria, they have including the four-man boat crew. very limited freeboard. The initial stability is very high to give adequate stability with limited Crewboats are used principally for the transporta- freeboard. The rear cargo deck is large and flat tion of personnel and have very limited cargo capa- and typically runs two-thirds (2/3) the length of city. They range in size from 65' to 150' with the boat. The forward one-third (1/3) consists the vast majority at or near the 100' mark. Un- of a focsle cabin with a deck house and pilot like supply boats and utility boats, the hulls are house above. The volume of the ship below decks made of aluminum and are designed for planing, is usually either deepframed or in ballast to rather than displacement. The cargo decks are reduce tonnage, except, of course, for the ma- similar in size to that of a utility boat, but chinery space. Some vessels are equipped to load capacity is much reduced, often to less than carry bulk mud or liquid mud below the main deck. 35 tons. Passenger capacity is much higher with In any case, a supply boat with no deck load 50 to 60 passengers being typical. Accomodations needs to carry substantial ballast to maintain for these passengers, however, usually consists of submergence of the wheels and rudders while oper- a seat and access to a head. No attempt is made ating in a seaway. This limits the use of below to berth or feed the passengers. Modest accomo- deck spaces. Typical accomodations are for 16 dations for a four-man boat crew are usually pro- men, including the ship's complement. vided. Crewboats have considerable freeboard, but because of their low polar moment of inertia, low Two common variations of the common supply boat mass and shallow draft, they tend to have a stiff, are the tug/supply and the four-point boat. A harsh ride. Their higher speed.(20 knots) is tug/supply is very similar to a normal supply achieved partially by lower weight and partially vessel, but is equipped with much more power and by horsepower 50% to 100% higher than a comparable a large towing winch. In addition to operating displacement hull. Because the poweris squeezed CH2585-8/88/0000- 1107 $1 @1988 IEEE out of smaller engines, maintenance and repairs to passenger boats. A research vessel over 300 tons the propulsion units are considerably higher. must meet subdivision and damage stability require@ ments much greater than a 300 ton cargo vessel. Geophysical vessels range from 160' to 300' and There is no practical way to do this. If the sub- are used to collect low frequency seismic re- division and stability problems could be met, - flections from the seafloor. The primary design there are a host of other rules (listed in 46 CFR characteristic of a geophysical vessel is a huge Subchapter U), that make the conversion imprac-@. interior volume to house computer rooms, compres- tical. If for some reason a vessel over 500 gross sor spaces, workshops, accomodations for 30 to 50 tons is invisioned, then S.O.L.A.S. regulations people, and a vast amount of storage for equip- would come into play. ment, stores, and supplies. A focsle cabin comes well aft of midship leaving a small back deck Because the rules for uninspected vessels are con- covered overhead by a continuation of the focsle siderably less stringent, virtually any inspected deck. The focsle deck holds a deck house forward cargo or passenger vessel will meed the require- with a pilot house above. ment for an uninspected research vessel. Many geophysical boats are conversions of basic Some research vessels under 300 tons are still supply boats. subject to manning requirements. Those over 100 tons are required to have a three watch system for Geophysical boats are much more diverse than licensed crew members and deckhands. Two-thirds other classes of boats and it is difficult to of the deckhands must have an A.B. ticket. Un- refer to them as a class. Because they are de- licensed crew members are not allowed to work signed for geophysical research, they are obvious- both in the engine room and on deck. Those ves- ly suited for that type of research. Because of sels over 200 tons must, in addition to the above, their relative scarcity and diversity, they will have a licensed master, first mate, and second only be dealt with peripherally. mate, and engineer. Government vessels and ves- sels in the service of a public authority are The first step in acquiring and converting a ves- exempt from these licensing requirements. sel is determining your needs. Not only must you define what your needs are, you must determine Blocking the tonnage openings, removing or modi- what they aren't. The biggest threat to a con- fying framing, or a reduction of ballast capacity version is that it is inflicted with the Swiss can all increase tonnage. Most boats are built army knife syndrome (SAKS). The symptomstare a within one ton of a step in the rules, so a change crippling addition of paraphernalia that makes of even a hundredth of a ton may cause you to fall the maintenance and operation of the vessel much in a higher tonnage class. like trying to drive screws with a Swiss army knife. Other things that should be considered before pur- chasing a vessel are the freeboard, the position Another point to remember is that flexibility is of blowers and exhaust trunks, tank vents (tank what gives a boat the ability to adapt to the scantlings are determined by the height of the future and weather the onslaught of SAKS. In vent), access to engines after modifications, practical terms, flexibility means extra deck galley size, generator capacity, availability of space, cabin space, machinery space, electrical below deck space (usually mud tanks) and ability power, electrical circuits, winch and A-frame to trim vessel after modification. capacity, and bouyancy. For our conversion project, we chose a 150' x 36' In our preliminary plans to purchase and convert x 12.5' supply vessel which admeasured at 195.5 a vessel for research use, much thought was given gross tons. There were two mud tanks below that to a vessel capable of coring operations, support- could be converted to quarters and an additional ing the larger sidescan systems, such as the EDO mud tank aft that could be brought into ballast Deep Tow and the various SeaMarc systems, and 48 for tonnage reduction. The additional ballast trace high resolution geophysical surveys. The could also counteract the forward trim due to mooring and/or dynamic 'Positioning system required modifications. for coring, along with the large deck space needed were determined to be incompatable with the eco- A typical conversion of a supply boat to a research nomical pursuit of our sidescan and geophysical vessel entails an increase in accomodations and survey requirements. Coring capability was delet- stores, the addition of laboratories and workshops, ed in order to optimize these other capabilities. the installation of deck gear and scientific in- struments, and the modification of the propulsion When a firm concept of needs is developed, the and steering system to meet special low speed con- search for a vessel begins. In addition to the trol requirements. practical problems of making the modifications fit the tonnage of the vessel should be noted. Our conversion followed these lines. Accomodations were increased from 16 persons to 24. 613 ft.2 of A research vessel under 300 gross tons is not sub- lab space were added along with 298 ft.2 of stor- ject to Coast Guard inspection. A research vessel age, and 130 ft.2 of workshop. An A-frame and over 300 gross tons is inspected, and by a set of several winches were added along with a well for rules different than those for cargo boats or an ultrashort baseline acoustic positioning system, 1108 a retractable subbottom profiler, and mounting was under four seconds. After a brief period, we points for various navigation antennas. A bow determined that there was not an economical way to thruster was installed and the rudders were en- reduce and still maintain dry decks. Stabiliz- larged to aid in slow speed maneuvering. Genera- ers are not effective at such low speeds and flume tor capacity was increased and isolated circuits tanks designed for such a short roll period were were provided. of an impractical shape and would have little, if any, effect on roll period. It was decided to add A careful analysis of the requirements to operate bilge keels not only to increase dampening through the sidescan and geophysical equipment gave us our eddy making, but to increase the entrained mass design goals. In addition to the above, it was and the effective polar moment of inertia about necessary for us to leave adequate deck space for the longitudinal axis. Bilge keels on supply boat our intended mission, comply with ABS loadline hulls do not usually meet with much success. Be- requirements, and Coast Guard Subchapter C re- cause the boats are hard chined, the bilge keels quirement. tended to reduce eddy loss at the chine, almost as much as they increase it around their own peri- During the search for a suitable conversion candi- phery. In addition, the fullness of the hull date, preliminary design concepts had been devel- usually hampers effective placement of the bilge oped. Although we need 88' of clear deck, we only keels. Rather than place the bilge keels below needed the full width near the stern. This gave the chine in a near vertical position, we put them us room to build storage rooms, lab space, and a about two ft. above the keel and in a horizontal workshop on the perimeter of the deck. It also plane. This gave us a better angle of attack for gave us a foundation for a 21' x 23' electronics eddy loss and also appeared to give a better bite lab on the upper deck. to increase entrained mass. These modifications exceeded our expectations. The roll period was When we began the conversion, the boat admeasured increased to above five seconds and complaints 199.5 tons. The addition of a walk@in cooler and dropped to zero. We were able to increase the freezer, transducer wells, deck lockers, and ad- freeboard to about two and one-half (21@2) ft. to ditional waste holding tanks increased the ton- give reasonable deck dryness. Check flaps on the nage to well above 200. The conversion of the aft freeing ports further reduced deck wetness. mud tanks to ballast, in theory, brought us just below our 200 ton goal. Unfortunately, the Coast A second problem encountered was an inability to Guard officer measuring the boat decided that two easily maintain slow speeds. In clutch, the boat companionways were excessive by about 25 cubic ran about four knots. Our desired low speed was feet each. This-..brought us above the 200 ton two and one-half (2@) knots. Trolling valves were goal. The officer didn't believe me when I told installed on the marine gears to remedy this prob- him that the freezer and cooler were portable and lem. The valves work by modulating the oil pres- not a part of the structure. A study was made of sure to the clutch engagement cylinder. This all tonnage in the vessel to try to find a way to allows the clutches to slip in inverse proportion exempt a small amount. It was found that a linen to the oil pressure giveing control of shaft RPM locker had a wall surface large enough for a ton- below idle speed. This allowed us to meet our nage opening. A tonnage opening was installed, low speed requirement. the vessel was readmeasured at 199.95 tons, and the vessel was exempt from licensing requirements. The vessel is currently being used to operate Shell Offshore's EDO deeptow sidescan system in the Gulf Because the additional cabin on the main deck form- of Mexico. We are in our second year of operation ed three sides of a well, insufficient bulwark without any downtime due to vessel operations. area was left for ABS-required freeing ports. To In addition to the two EDO sidescans, the vessel alleviate this problem, a raised watertight floor is equipped with an EDO narrow beam fathometer, an was used in each cabin aft of the main cabin with O.R.E. 30 KHz subbottom profiler with E.P.C. re- freeing port area below. corder, an EDO Nav Track V acoustic positioning system, and a Sonardyne Compatt long baseline A further problem developed with ABS loadline re- acoustic positioning system. Topside positioning quirements. ABS requires a Coast Guard-supervised is obtained through a John E. Chance & Associates stability test. The Coast Guard is only authoriz- Starfix system with cubic argo, GPS, and loran ed to supervise stability tests on inspected ves- supplying backup data. sels or vessels in the process of being inspected. When the local Coast Guard office discovered the The real question is not whether a conversion is bottle neck, they managed to bend the rules enough possible, but whether it is economically practical. to accomodate us. The SEIS SURVEYOR cost us $1.9 million, included the cost of bringing existing machinery and struc- When the vessel was delivered and put into service, ture to like-new condition. Our preliminary cost several problems arose. The most noticeable was a estimate for a new vessel was $2.7 million. Job sharp, snappy roll. The seakindliness could be performance is identical to a new boat, but main- improved by pumping the vessel down, but then the tenance costs are slightly higher. Altogether, we decks would stay too wet for comfortable work. appear to be doing better by about $9,000 per The violent movement was caused by the extreme month. stiffness of the vessel. In our normal operating state, we had a U-M of over 10 ft. Our roll period 1109 The way we chose to modify our vessel is only one of many approaches. Standard utility boats can be made quite sufficient for light research by adding a "doghouse" to the back deck. Their general nature is much more akin to research needs than a supply boat. Since they are under 100 tons, they can be operated in an economical manner. Supply boats cry out for containerization. There large flat decks can easily hold six forty-footers with deckspace leftover. Crewboats are fast and have surprising room for use on protected waters where their limited endur- ance and poor seakindliness is not a problem. Any conversion, however, takes careful planning. The problems are more plentiful and the solutions are fewer. Only through careful planning, innova- tive thinking, and a willingness to compromise can success be achieved. 1110 RESEARCH VESSELS: A SYSTEMS ENGINEERING APPROACH Cyrus Hamlin, NA. Kennebunk, Maine 04043 USA ABSTRACT instead will join in a common effort at keeping the world a good place to live.) This paper presents the case for the application of p While we need not perhaps plan now for the year systems engineering methodology to coordinating 3000, we should and can start along that path by existing research vessel operations and planning for planning for 25 or 50 years hence, well within the future construction and operation of the research fleet lifetime of many of us. A vast anount of data will be An essential part of such systems engineering is required for colonizing the oceans, and much if not modeling research vessels. most of those data will be gathered by ships. A preliminary research vessel model is described. For Ships are long-term projects, they take several years a unique set of inputs, output is given which includes from conception to commissioning and have a useful the physical characteristics of the vessel, Its operating life of 20 to 30 years. They are expensive to build, and patterns, and its construction and operating costs. A unlike land structures are very difficult and costly to sketch design is included which Illustrates and verifies alter when once built, Finally, the rigors of the them outputs. It demonstrates what can be done with environment in which they operate, and their a minute or two of computer time. mechanical complexity, make vessel operating costs a large proportion of the expenses of any enterprise which depends upon them. BACKGROUND So the effort now being made, through UNOLS At some time in the future, perhaps five hundred to a (University-National Oceanographic Laboratory thousand years from now, the world must become a System), to coordinate the construction and operation no-waste system. All organic and industrial processes of the U.S. fleet of research craft must be greatly must eventually return their products and byproducts expanded to meet the financial, personnel, and into the global resource bank. The independence organizational restraints of the future. enjoyed to a large degree by manned satellites, resulting from recycling, will expand to global ATTACKING THE PROBLEM proportions. Considering the inevitable increase in population and the vast resources contained in sea Systems engineering has long proven its worth as a water, there should be little doubt that the oceans win, tool in planning and decision-making for complex perforce, play a major role in that self -sufficient systems. Few land-based systems can match marine future condition. research operations for complexity and indeterminateness. The variables are almost limitless The seas will then be teeming with activity. The in number, and in size range from the swirling global extraction of protein, minerals, energy, various metals weather patterns causing El Nino, down to the degree and chemicals, the location of habitats on and in the of seasickness among a team of scientists rolling about ocean, and the puruit of life's pleasures will serve to at sea. These variables all combine to complicate the populate the oceans with cities, pleasure places, and design of the fleet of ships which for the foreseeable industrial facilities. (We must assume that mankind future must be the primary data-gathering will have by then given up fighting each other and oceanographic instrument, CH2585-8/88/0000- 1111 $1 @1988 IEEE This paper describes a simple working model for use as BUILDING THE RESEARCH VESSEL MODEL a tool in planning the research fleet of the future, or any vessel in it. With easily ascertained inputs, it can In constructing a model, the important parts are the in a matter of minutes delineate a vessel in terms of its beginning and the end. Of them it is vital to select the physical characteristics, functional capabilities, and correct end product because that is the part which is construction and operating costs, thus providing all the fed into the decision-making and planning processes. major bits of information for planning marine research The "bottom-line" of this present model is: operations. Fig. 1. Block Diagram for Research Vessel Model - INPUT INDEPENDENT VA S F NUMBER 0 TASKS DISTANCE TO STATION TRIP SCHE NO. OF SCIENTISTS DULE ANNUAL SCIENTIST i NO. OF SCIENT. CREW HOURS ON STATION '"@ANNUAL S C WFU-L -E] --I p DRY L NO. OF WATCHES (1, 2, 37) ICREW SIZE WET LAB SI SHAFT HP ACCOMODATIOT-- HOLD/STORE @@IZE AREA AUX. HP TOTAL P [TOTAL UTILITY AREA MACH'Y AREA CUBIC NUMBER FUEL CAPACITY SPECIAL CO S TOTAL VESSEL COST VESSEL PROPORT IONS CREW COS VESS i ANNUAL VESSEL COSTI EL DIMENSIONS VARIA15LE Ed@TTS ANNUAL OPERATING COST PRELIMINARY SKETCHE COST PER SCIENTIST HOUR ON STATION J7 i I i INCORPORATION IN OMPARISON WITH PROJECT BUDGETING FLEET PLANS OTHER VESSELS PREPARATION NO, F@SC o 0 SCI ENO F AB SIZE @B SIZE N A @LE R A U op TI C 0ST PER SC IE HOUR ON ST A 1OPERATION1 1112 The annual SJUMst-hours on station are calculated -Cost per Scientist-Hour on Station- from the number of scientists carried and the hours on station taken from the annual schedule. Total This output probably best expresses the accomodation area is summed from the living areas cost-effectiveness of a research vessel, which, after an, assigned to the scientists, the scientific crow, and the exists for the purpose of putting scientists out where ship's crew. (The area derived in the program is they can do their work. However, all the other nominal, to match other areas used in vessel size Outputs derived from operating the model can be determination, and represents about half the actual useful in the pinning process. These peripheral habitation area.) outputs can include: a full description of the vessel; the sin of staff required to man it, vessel construction cost;- The shaft horesp-ow-er (SHP) to give the required speed Vessel operating costs. While the values of the is approximated from the total utility area and the 0outputs must be considered estimates only, they can accomodation area (this will be chocked later on and provide a very precise basis for comparing different adjusted upwards or downwards as necessary), vessels, and a reasonably accurate first-pass estimate NARIMMY h0rSftKw_e_r for generators, refrigeration, etc., Of the Major physical and financial elements of the is taken as a function of SHP. From the total vessel system. kQ=23o_r is derived the fuel cap4dty needed and the size of the Mggoea spa2k. As a starting point for the model, I have selected two primary parameters. The first is the number of From the three areas - utility, accomodation, and different tasks the vessel is to be engaged in, on which machinery - can be determined the cubic will depend the number of scientists on board. The hence size, of the research vessel. Using preset but second primary parameter is the average distance to Variable vessel PLORLEUgns (length to beam, beam to the vessel's research station from the base port, which depth, etc.) the vessels dimensions are easily will largely determine the size of the vessel. calculated. Finally, to check the vessel delineation portion of the, program and to help visualize the Fig. I is a block diagram illustrating the major vessel, a scale re *gk= sketch can be made. elements Of the Model and their interrelationships which connect the start and end described above. The From the cubic number, total horsepower, and any general design of the model is based on considerable �pecial costs such as unusually extensive electronics, experience the author has had In modeling fishing complex riggin& etc, the t;otal vewl cost of vessels 111, 12). A number of assumptions are construction can be reasonably accurately derived. employed in order to make the- model as simple and The annual vessel ie. debt retirement is easily operated as possible. These assumptions are dependent on the particular financial arrangements realistic, but if changes are desired, they are easily made for paying for it, made in the program. The annual gR2Lating cost is composed of the crew A RUN -THROUGH OF THE MODEL costs (wages, insurance, etc-), fuel Cost Und costs such as insuranM variable costs like port fees, etc, and the Looking at the left side of Fig. 1, on the number Of annual vessel cost The annual operating cost divided kala depends the number of scien from which is - by the annual scientist-hours on station gives the final derived the SiZe of the scientific Mg (that portion of criterion - Cost RK scientist hour on station the ship's company which actually handles and maintfts the scientific gnT), the size of the dry and 11 the model is operated with a range of values Net laboratories, and the - ce for storM of assigned to one of the variables, such as 4, 6, 8 1pecimens and gear. scientific tasks, the outputs describe a number of vessels with differing characteristics and performance, On the right side, having to do with the vessel, the The criteria for the specifc application can be applied distance to the station (an average) sots the kJR to this series of vessels to identify the one with that Schedule and hence the annual mbedule The trip mix of values which is most advantageous. length dictates the number of watches stood per day which determines the crow size. 1113 DISCUSSION OF SOME OF THE PARAMMS Vessel Speed limitations Non-naval architects often have trouble with some of Virtually all research vessels are of the displacement the factors with which naval architects deal on a daily type which seem to travel along in a hole they dig in basis. Following are discussions of two of then: the water, as opposed to planing craft, which appear to be skipping along the very surface of the water. Vessel size Whereas the top speed of planing craft is limited only How big is a vessel? To the layman, length is usually Table I considered an adequate description of size. but Is a Computer-generated Research Vessel destroyer, with a length of 400 ft, beam 40 ft, and depth 25 ft. the same size as a 400 ft tanker with beam Number of tasks assigned 5 and depth of 75 ft and 50 ft? Obviously not. Avg. distance to research area 200 n. mi. For the naval architect size is most accurately stated as Average round trip steaming 2 days "displacomenC, which Is the actual weight of the vessel Average time on station 10 days and its contents, equaling the weight of the water It Average time between trips 3 days displaces. But this is an arcane quantity, difficult to Days at sea per year 250 come by, and constantly changing as cargo and fuel are Number of trips per year 17 taken aboard or discharged. Number of scientists 10 Fig. 2. Cubic Number Definitions Number in sdentific crew 5 Number in ship's crew 8 Area of dry laboratory 160 sq ft Area of wet laboratory 160 sq ft Storage for gear and specimens 320 cu ft r.9 T Length over all 131.1 ft DEAK- Length waterline 120.6 ft Beam 30.44 ft C-vee L-WL DeArl Draft, max. 12.43 ft Depth of hull 17.76 ft Loaded displacement 600 Tons Fortunately, for general use in and out of the Cubic number 65,210 profession, there is a simple number, known as the Shaft Horsepower 575 Xubic Number*, or CUBE, which equals LWL x beam x Auxiliary horsepower 345 Depth. LWL is the waterline length, Beam is the Fuel capacity, U.S. gallons 16,000 maximum width, and Depth is the vertical distance Steaming speed I I knots from main deck to the juncture of the plating with the keel (see Figure 2). These dimensions are usually Total vessel cost $2,860,000 listed in descriptions of vessis (but don't confuse Depth Annual operating costs $1,640,000 with Draft!). Annual scientist hours on station 13,333 CUBE is a most useful number. For vessels in similar Cost per scientist-hour on station $123 service, the displacement and cost of building and operating will be very nearly proportional to CUBE. It is independent of the proportions of the vessel and of by the size of engines which can be Installed, or by the its condition of loading. Unless the vessel is rebuilt it foolhardiness of the driver, the top speed of will accurately define the real size of the vessel displacement craft has a well-defined limit This limit @13 @IA I A 5-- throughout its life. can be expressed as: 1114 Upper Speed Limit, knots - 1.2 to 1.3 x LWL A SAMPLE COMPUTER RUN For eliample, by this rule a 100 ft vessel will have a A preliminary program was written incorporating the top speed of 12 to 13 knots. Above this speed, the elements shown in Fig. 1. The resulting data are given average displacement huff will require a In Table I - The primary inputs are: the number of disproportionate amount of additional horsepower to tests the ship must support, and the average distance realize a very small gain in speed. (Oceanographers from the ship's base to the research area. The primary will recognize the above equation as similar to the output is the cost per scientist hour on station. To got formula for the speed of a deepwater wave.) from the input to the output the program has a good Fig. Sketch Design Based on Computer Output A .Go f4dpb 4 Irtev., ccd e- a VR -1 ( 0 0,.*S -r& re --5, -1-A 4c. 5 o &4 We sx. V- C- be- + A -0 L A -------------- AM, =D41 W_ KAIIJ ------------- 1115 many constants and semipermanent variables relating Nevertheless, the ability of the program to be quickly the various blocks as indicated in Fig. 1. and the revised, and to translate a new set of inputs into a new elements within the blocks, the values of thew are hypothetical vessel in a minute or twq, should make taken from past experience of the author. this type of software a useful tool to the marine research planner and administrator, and to the A sketch is included (Fig. 3) to illustrate how the designer of research vessels. computer output can be verified by a simple scale drawing. Using the power of the computer, this very simple program can be expanded to incorporate the finest When the program is run with six tasks instead of five, level of detail which could be desired. hence more scientists, etc, the vessel becomes larger, costing more to build and to operate. Nevertheless, the REFERENCES cost per scientist-hour drops from S 12 3 to $119, This gain in economy must, however, be balanced against 1. C. Hamlin and J Ordway. Desigg Study a An QRU= the difficulty of efficiently scheduling six instead of Ei" Vessel for the Georgta Bank Groundfish five projects from one vessel. EJW=. Springfield, VA.: National Technical Information Service, 1972. SUMMARY 2. C. Hamlin and J. Ordway. DgWM Study: An Og=um The quantifications used in this program even though Vessel for the Northern ShrimRZOM Unpublished. reasonable, are subject to further verification. 1116 MONOHULL RESEARCH VESSEL SEAKEEPING AND CRITERIA Bruce L. Hutchison and Sridhar Jagannathan The Glosten Associates, Inc. 600 Mutual Life Building 605 First Avenue Seattle, Washington 98104-2224 Motion Program SMP was available by 1981 f5) offering improved treatment of forward speed dependent terms such as roll damping. ABSTRACT Probabilistic treatments of the analysis of vessel risk and operability in Seakeeping performance of several classes of existing and proposed temporally and spatially varying climatologies were described in 1981 [6]. monohull research vessels is compared. Performance was estimated using the U.S. Navy's Standard Ship Motion Program (SMP). Other vessels All of these accomplishments could be regarded, in the parlance of modem considered include the AGOR 9/10 and the proposed UNOLS medium- risk analysis, as progress in the definition of the demand functions, endurance monohull research vessel. All of these vessels are compared to Progress in defining the corresponding capability functions has lagged. the U.S. Navy's COR seakeeping criteria for the new AGOR 23. The Dr. St. Denis compiled one of the early volumes addressing the limits of Navy's COR seakeeping criteria for the AGOR 23 are discussed and operability for various activities [7]. Included in [7] are discussions of the suggestions for increasing the clarity and usefulness of such criteria are threshold of malaise, safe footing, slamming and shipping of green water provided. on deck. Spouge [81 published criteria for deck wetness, slamming and roll which he applied to a British fisheries protection vessel and Kennel, et. al. [9] have published criteria for these functions, pitch and vertical acceleration. INTRODUCTION These criteria are summarized in Table 1. Modem ocean research missions increasingly demand high standards of ship motion performance from oceanographic research vessels. Most of the current U.S. fleet of research vessels were commissioned before the advent of seakeeping criteria and modern analysis capabilities for ship motions. TABLEI The synthesis of hull forms exhibiting improved seakeeping performance Seakeeping Criteria is still an activity guided by judgement and intuition. However, once a design has been postulated modern analysis methods permit efficient British performance assessment during the design development process. Fisheries U.S.N. Protection Combatant These new analytical capabilities have created the opportunity for the Vessel Mobility oceanographic community to express their desires for seakeeping performance in the form of seakeeping criteria which in turn become Ref. [81 Ref. [91 objective goals for the designers. As the custom of promulgating such Vert. Accel. criteria is a relatively recent development, this paper seeks to contribute to on Bridge (g's) 0.4 the evolution of such criteria by reporting on the performance of existing and proposed research vessels as compared to criteria currently in use. Deck Wetness 36/hr 30/hr Slamming 20/hr 20/hr BRIEF HISTORY Roll, Signif. Modern ship motion analysis had to await the development of modern Amplitude 10* 8- models for the ocean wave environment. The seminal event was the landmark paper by Dr. W.J. Pierson, Jr. in 1952 [1]. This was followed in Pitch, Signif. 1953 by a paper marking the collaboration of Drs. Pierson and Manley St. Amplitude 3- Denis [2] which presented a ship motion theory describing the ship as a tensor operator acting on the seaway. Ship motions therefore became subject to the same statistical descriptions of second-order stationary random processes which have been successfully applied to the seaway. The most comprehensive and recent seakeeping criteria applied to research The thirty-five years since the Pierson/St. Denis paper has seen remarkable vessels are those specified for the acquisition of the AGOR 23. These growth and success in the application of analysis to ship motions. A criteria are presented in Table 2. These criteria may be seen to be much practical analysis of coupled pitch and heave was developed as early as more sophisticated than those of Table 1. Many more seakeeping 1967 [3]. A complete analysis of the six-degree-of-freedom rigid body processes are considered and the criteria are parameterized by vessel motions was available by 1971 [41 and the U.S. Navy's Standard Ship operation, speed and sea state. CH2585-8/88/0000- 1117 $1 @1988 IEEE TABLE2 AGOR 23 COR - Seakeeping Criteria Condition Sea State Ship Speed Heading Roll Pitch Wetness Slam Lat'l Accel Vertical Accel (g's) (Feet) (Knots) (Deg.) (Deg.) (Deg.) (NumA-1r) (Num/Hr) Bridge Bridge Stem Midship TRANSIT 8.2' 12.0 All 8.00 3.0' 30 20 .2 .4 OPERATING 12.2' 6.0 Best 8.00 3.00 30 20 .2 .4 .4 ON STATION 11.01 0.0 Best 5.00 3.00 5 5 .2 .4 .4 .4 Notes: 1) Roll, pitch and accelerations are specified as the maximum (with respect to sea state modal period) significant single amplitude values. 2) Best heading is not defined in the AGOR 23 COR. These criteria represent, from the designers point of view, a significant VESSEL PERFORMANCE advance towards a design for seakeeping process. There is however, in our opinion, room for further improvement. In this paper we offer what The performance of four monohull research vessels was predicted using we hope is constructive criticism along with some suggestions regarding the Standard Ship Motion Program SMP [5]. Two classes of existing fruitful directions for improved seakeeping criteria development. research vessels, the AGOR 9/10 and AGOR 14/15, were examined. Also examined were the AGOR 14/15 class following a planned lengthening and a proposed new medium endurance research vessel (MERV) under the CRITIQUE OF AGOR 23 SEAKEEPING CRITERIA UNOLS fleet replacement program [101. The conceptual design for the proposed new vessel was developed with particular attention towards The seakeeping criteria of Table 2 are prescriptive rather than functional improved seakeeping performance. Table 3 presents the particulars of criteria. It is not known to the authors whether these prescriptive criteria each of these vessels. are in fact based on an underlying consideration of functional limits but any such consideration has not been carried through to the final criteria in such a way that the designers can benefit from the underlying functional TABLE3 background. The limits are not related to any activity which might assist the designer in developing a solution to the real problems faced by the Monobull Research Vessel Particulars scientists aboard the research vessel. It would be an improvement to state activity related functional criteria such as the limiting acceleration Proposed environment in which a circuit board can be soldered, the conditions under Existing Modified New which a biological sample can be dissected, the limits for handling AGOR 9/10 AGOR 14/15 AGOR 14/15 Vessel [101 chemical glassware and the limits on launching and retrieving packages of Length 196'-0" i2_0'0- 254'-6" 212'-0" different sizes and types. An example of how such functional activity Beam 39'-0" 46'-0" 46'-0" (W-011 based information could benefit the design process would be the selection Draft 15'-3" 151-01, 15'-6" 15'-2" of a space with minimal accelerations for a critical activity sensitive to Displacement 1490 1948 2685 2469 acceleration. Year(s) Built 1965 1968/69 1989/90 Similarly the prescriptive criteria do not provide any insight into the relative importance of various.seaceprng processes. Stated another way, Table 4 presents the seakeeping performance for each of these four what is the relative value of increased performance margins (margin is the research vessels and compares their performance to the AGOR 23 positive excess of criteria limit over seakeeping performance) in roll seakeeping criteria. compared to a similar margin in pitch (or some other mode)? Condition (C) in Table 4 is not an AGOR 23 COR criteria condition. A closely related criticism concerns the fact that the criteria are all posed Condition (B) is an AGOR 23 COR criteria condition but it can be seen as deterministic limits acting as step-function capability. Presumably that none of the vessels examined can satisfy the criterion for pitch in many of the the science activities affected by the seakeeping performance condition (B). Condition (C) has been added to indicate a reduced sea undergo gradual degradation of performance. Even concerns such as state where all of the vessels excepting the AGOR 9/10 can satisfy the structural failure can be regarded equally well from a probabilistic pitch criterion. perspective subject to a distributed capability function. Certainly those activities which rely on human performance are subject to capability This situation illustrates one of the most troublesome problems with the functions reflecting human variability. AGOR 23 COR seakeeping criteria, that being the lack of definition regarding "best heading". All of the vessels examined could satisfy the Another criticism regarding unstated relative importance can be lodged pitch criterion in condition (B) at headings in the neighborhood of 90* with respect to speed and sea state parameters. For instance, what is the (beam seas) but they would concurrently violate the criterion for roll. The relative value of increased performance margins at (for example) 12 knots AGOR 23 COR is not definitive regarding the extent to which the criteria in sea state 4 versus 8 knots in sea state 5? If the designer has the are joint. Table 4 was constructed under a particular interpretation opportunity to increase the performance margin in one case at the expense regarding "best heading" which will be described in the following of the other, which is to be preferred? paragraph. 1118 TABLE4 Comparison of Seakeeping Respons@ AGOR 23 COR, Existing AGOR 14/15, Modified AGOR 14/15, AGOR 9/10 and 212' ME, RV Vessel Sea State Ship Speed Heading R oil Pitch Wetness Slam Lat'l Accel Vertical Accel (g's) (Feet) (Knots) (Deg.) (Deg-) (Deg.) (Num/Hr) (Num/Hr) Bridge Bridge Stem Midship (A) TRANSIT CONDITION AGOR 23 COR 8.2' 12.0 All 8.0* 3.0* 30 20 .2 .4 EXISTING AGOR 14/15 4.78* 2.98* 0 0 .079 .152 .241 .152 MODIFIED AGOR 14/15 4.54' 2.63* 0 0 .064 .143 .216 .131 AGOR 9/10 10.15o 4.09* 2 0 .111 .338 .366 .21 MERV 5.91* 2.70* 0 0 .158 .169 .185 .20 (B) OPERATING CONDITION AGOR 23 COR 12.2' 6.0 Best 8.0* 3.0* 30 20 .2 .4 .4 - EXISTING AGOR 14/15 60o 5.69* 3.63* 0 0 .106 .151 .233 .156 MODIFIED AGOR 14/15 60o 4.28* 3.23o 0 0 .085 .150 .225 .142 AGOR 9/10 60* 11.4o 4.57o 0 0 .145 .283 .293 .206 MERV 150o 5.72o 3.27o 0 0 .090 .072 .096 .11 (C) OPERATING CONDITION [Note: not an AGOR 23 COR Condition] AGOR23COR 8.2' 6.0 Best 8.0* 3.0* 30 20 .2 .4 .4 EXISTING AGOR 14115 60' 4.04* 2.44* 0 0 .073 .117 .185 .119 MODIFIED AGOR 14/15 60o 3.08o 2.17* 0 0 .057 .114 .175 .108 AGOR 9110 60o 8.16o 3.35* 0 0 .104 .238 .246 .16 MERV 150o .4.52* 2.20* 0 0 .084 .053 .07 .096 (D) ON STATION CONDITION AGOR23COR 11.01 0.0 Best 5.0* 3.Oo 5 5 .2 .4 .4 .4 EXISTING AGOR 14/15 135' 6.10* 3.33* 0 0 .081 .086 .178 .107 MODIFIED AGOR 14/15 135* 4.32* 2.88* 0 0 .061 .085 .168 .092 AGOR 9/10 135* 11.0* 4.05* 0 0 .107 .150 .221 .142 MERV 050 7.11o 3.011 0 0 .125 .096 .147 .16 Best heading is chosen as the heading which minimizes both normalized roll and normalized pitch response. Bold lettering indicates an exceedence of die AGOR 23 Circular of Requirements. TABLE5 Comparison of Seakeeping ResDonse With Alternate "Best Heading" AGOR 23 COR, Existing AGOR 14/15, Modified AGOR 14/15, AGOR 9110 and 212'MERV Vessel Sea State Ship Speed Heading Roll Pitch Wetness Slam Lat'l Accel Vertical Accel (g's) (Feet) (Knots) (Deg.) (Deg.) (Deg.) (Num/Hr) (Num/Hr) Bridge Bridge Stem Midship (B) OPERATING CONDITION AGOR23COR 12.2' 6.0 Best 8.0* 3.Oo 30 20 .2 .4 .4 EXISTING AGOR 14/15 105o 6.81*' 2.99o 0 0 .108 .123 .172 .141 MODIFIED AGOR 14/15 105* 5.181 2.68o 0 0 .086 .123 .171 .124 MERV 180* 4.05o 3.45o 0 0 .052 .057 .085 .064 Best heading is chosen as the heading which maximizes the sum of the normalized margins for mll and pitch response. Bold lettering indicates an exceedence of the AGOR 23 Circular of Requirements. value. Table 5 shows the "best heading obtained as the heading Pitch and roll as functions of heading angle were normalized by their corresponding to the maximum of the sum of the normalized margins in respective greatest values, thus producing functions with ordinal roll and pitch responses. The headings so obtained are clearly different magnitudes ranging from zero to one. The "best heading" was determined from those in Table 4. With the "sum of margins" approach it is seen that to be that heading corresponding to an intersection between the normalized the existing and modified AGOR 14/15 now meet both roll and pitch pitch and roll functions which possessed the least ordinal value among all criteria whereas they had violated the pitch criterion in Table 4. However intersections between these normalized functions. the proposed medium -endurance research vessel still violates the pitch criterion under this definition of best heading. Even the "sum of margins" Consider, for example, an alternate "best heading" based on the margin approach is not satisfactory since it can be shown that-tim proposed between the actual response and the criterion, normalized by the criterion medium-endurance research vessel satisfies both the roll and pitch criteria 1119 FIGURE I FIGURE 2 Speed-Polar Diagram Response-Polar Diagram AGOR 14/15 Existing Ship Zero Trim Comparison of Seakeeping Short-crested for Existing and Modified AGOR 14/15 Significant Wave Height = 12.20 Feet Short-crested Pitch Angle Significant Wave Height = 12.20 Feet (Degrees) Pitch Angle Statistic: Maximum Significant Single Amplitude (Degrees) Statistic: Maximum Significant Single Amplitude HERn HE 0 S= F' EFS SEAS 12 k, 5 3 k 3 2 so 27Z V-2 1 _7@ 3 .7@ /1 4" 4 4 K. ISO ISO I.... Pitch Angle = 2.00 DEG 6 Knots: Existing AGOR 14/15 2.... Pitch Angle = 2.50 DEG - - - - - 6 Knots: Modified AGOR 14/15 3.... Pitch Angle = 3.00 DEG 3 DEG AGOR 23 COR Criterion 4 ....Pitch Angle = 3.50 DEG 5.... Pitch Angle = 4.00 DEG 6.... Pitch Angle = 4.50 DEG ....7 .... Pitch Angle = 5.00 DEG An example of a new kind of polar plot for response is introduced in Figure 2. We denote this type of presentation as a response-polar plot. Figure 2 shows the response-polar plot for pitch corresponding to the (and the remaining criteria as well) at a best heading around 120* where speed-polar plot of Figure 1. Again the polar angles correspond to roll is 7.88* and pitch is 2.88* . heading but the radii now correspond to significant pitch amplitude. The contours are now iso-speed lines. The region outside the three degree radii It is obvious that these definitions of best heading are arbitrary and may is shaded to indicate that operating conditions in that region exceed the not reflect the intent of the authors of the AGOR 23 COR seakeeping AGOR 23 COR pitch criterion. criteria. Clearly other definitions are possible and an alternative definition may be in some way "better" than those discussed above. Alternative Figures 1 and 2 are mappings of the same response process in the polar approaches to the formulation of seakeeping criteria which resolve this plane. The speed-polar plot is a particularly useful presentation for the issue will be discussed in a later section of this paper. ship operators while the response-polar plot is perhaps more useful for analysis during the design process. One of the attractive features of the response-polar plot is the simple way in which the region violating a stated POLAR DIAGRAMS criterion can be indicated. Another practical advantage is the topologically simpler regions defined by the iso-contours in the response- Another perspective on vessel performance is achieved by producing polar plot. These simpler regions make both plotting and interpretation speed-polar and response-polar plots. Figure 1 is an example speed-polar easier. plot of pitch for the existing AGOR 14/15. Speed-polar plot grids directly present the operational parameters of speed and heading which are under One measure of operability which can be derived from the speed-polar the control of the ship's officers. The polar angle corresponds to the diagram is the ratio of the "operable" domain to the total domain. For the heading angle, the radii correspond to speed. The contours are iso- diagram shown in Figure I this measure of operability is 0.281. Assuming response lines. The shaded regions represent combinations of speed and that every speed and heading combination is equally desirable in a sea heading where the 3' pitch limit of the AGOR 23 COR is exceeded in a state with significant wave height of 12.2 feet, then the operability fraction short-crested sea state characterized by a 12.2 foot significant wave height. subject to a step function pitch criterion of 3' is 0.281. 1120 FIGURE 3 the absence of adequate criteria. The current state of criteria development Speed-Polar DIAGRAM could be characterized as a series of performance gates. At this level of development the role of analysis is limited primarily to confirming the Superposition of Significant Pitch success of a proposed design, but cannot play an effective role in guiding on Significant Roll (Degrees) the designer towards new and more efficient designs. The proposal for in Short-crested Sea State 6 future criteria development contained in this section would result in the codification of both subjective and objective owner/operator requirements HEAD in such a way that designers could make use of modem analysis tools to SERS IJ'6 knots seek improved designs. POWER LIMITED PEED In the course of discussing the AGOR 23 COR seakeeping criteria and the :@@ . . . ...... .... S PITCH evaluation of vessel performance with respect to those criteria a number of criticisms have been set forth. These criticisms are summarized as follows: There is no metric for the relative importance of each criteria. There is no metric for the relative importance of each operating condition (i.e. speed, heading and sea state). . . ..... . There. is no clear definition of "best heading". The criteria are prescriptive rather than functional. The criteria are provided as step-function deterministic limits where in many instances the true capability function is distributed. 270 - An underlying problem is that the criteria exist in a vector space where each individual criterion may be regarded as a component of a vector. The design problem is satisfied if each criterion is satisfied. Any individual criterion is either satisfied or it is not, there is no provision for partially ROLL satisfying a criterion. Stated another way, there is no basis for stating a preference between one satisfactory solution and another. The conventional mathematical solution to a problem of this character is to 6 compose a vector dot product between a design solution vector and a static weighting vector. Such a dot product produces a scalar valued objective 180 function which can then be the subject of optimization procedures. FOLLOWING SERS It is not possible to compare vector A with vector B directly. No statement such as "vector A is better than vector B" is possible. However, if a weighting vector, W, is stipulated it is possible to state that scalar a = A similar measure can be constructed from the response-polar plot in W*A is greater than, equal to, or less than scalar b = W *B. This is the Figure 2. Taking as our example the modified AGOR 14/15, the polar arc property which recommends the dot product as the means to formulate through which the 6 knot iso-speed contour lies within the operable objective functions for vector processes. domain represents 0.491 of the potential arc (360* ). Assuming that every heading is equally desirable when operating at 6 knots in a sea state with Application of this approach to generate scalar seakeeping rank functions significant wave height of 12.2 feet, then the operability fraction subject to is associated with the lateNathan Bales (111 who applied the concept to a step function pitch criterion of 3' is 0.491. the optimization of destroyer seakeeping. More recently van Wijngaarden [12] has applied this technique to small ships operating in the North Sea Through superposition of speed-polar or response-polar diagrams which more closely resemble a typical research vessel. The Glosten operability subject to joint criteria can be evaluated. Figure 3 is a Associates have recently applied this technique to an investigation of superposed speed-polar diagram showing the operability of the proposed surge, heave and pitch stable spar buoy hull forms under consideration for new medium-endurance research vessel (101 subject to joint roll and pitch an RIV FLIP replacement [131. criteria of 10* and 5' respectively in sea state 6. Installed power limits the vessel speed in this sea state to 11.6 knots so operability is measured over Seakeeping operability of a research vessel must be evaluated in an the zero to 11.6 knot speed domain. The operability, measured as environment subject to the following environmental parameters and described above, is 0.672 for roll alone, 0.859 for pitch alone, and 0.531 operational factors under the control of the bridge: subject to the joint criteria. This approach to evaluating operability could be extended in the obvious manner to any number of joint criteria. Environmental and Operating Factors TOWARDS A GENERAL CRITERIA STRUCTURE Hs Significant Wave Height Prior to the development of modern analysis methods for seakeeping TO Modal Wave Period performance designers relied solely on judgement and experience to V Ship Speed achieve seakindly forms, with occasional objective feedback from model X Heading Angle tests. Owners and operators of research vessels communicated their desires for seakeeping performance to the designer, primarily through subjective and qualitative descriptions. Each process or activity critical to the success of a particular scientific mission takes place at some location in ship coordinates. As there may be The advent of modern ship motion analysis has created the opportunity to many critical activities there will, in general, be many locations of interest. objectify this communication process between owner/operator and To make a criteria most general and useful, these locations should be designer. However, modern computer analyses can generate more data defined functionally (e.g. chemical lab sink; biological dissection table; than either the designer or the owner/operator can effectively evaluate in electronics repair workbench; etc.). This will permit the designer to vary 1121 the selection of space for these activities in order to obtain the highest The joint probability for significant wave height and modal period can be overall operability. obtained from climatologies such as [14]. The conditional probability for speed and heading can be developed from mission requirements and vessel speed and power relationships in a seaway. Location in Ship For the global set of N missions there are K locations, @j(j=1,2,...,K), =(71,72,73) Location in Ship within the ship where mission critical activities must take place. For any given mission only a subset of these K locations may be of importance. There are many seakeeping responses potentially of interest. In general There are twenty-four response processes listed in the example global the local accelerations and relative orientation of the virtual gravity vector response list above. For any given mission, Mi, and each location of will be most representative of the environment experienced by shipboard interest, @j, a subset of the global response list will be of mission personnel, instruments and apparatus. The local relative motions (between importance. For each such response a weight, w(Mi,xk(@j)), should be the ship and sea surface) and shipping of green water are expressive of the assigned and a criterion value, c(Mi,xk,(@j)). environment for launching and retrieving gear over the side. Local accelerations (for instance at a boom tip) are associated with fluctuating forces which may limit operations, while local velocities would relate to The weight, w(e), should be composed in such a way as to reflect both the the required performance of motion compensating mechanisms. relative frequency with which the given activity at the given location must Emergence and slamming relate, among other things, to sonar be preformed and the relative importance of the activity to mission performance, emitted noise and forward speed maintenance. success. The final value of the scalar objective function, ( ), is obtained from a probability weighted summation over all missions, sea states, operating Response Factors conditions, locations and responses; of the products of the weighting function with a function developed from the response and the associated Xn( ), n= 1,2,3,4,5,6 Displacements criterion value. An example equation showing one form which such an Xn( ), n= 7,8,9,10,11,12 Velocities objective function could assume is shown in the box at the bottom of this Xn( ), n=13,14,15,16,17,18 Accelerations page. Xn( ), n=19 Relative Displacement Xn( ), n=20 Relative Velocity Xn( ), n=21 Relative Acceleration There are many possible forms for the function f(xk(@j),c(Mi,xk(@j))). Xn( ), n=22 Slamming Three possible forms will be discussed. Xn( ), n=23 Shipping of Green Water Xn( ), n=24 Relative Orientation of First form: Virtual Gravity Vector c(-) - x(-) Mission Profile f(-) = ----------------- c(-) A mission profile consists of N discrete missions, Ml,M2....MN. For Second form: each mission the number Mi is a weight expressing the relative importance f(-) = Pc (n) dn of that particular mission with respect to all the other missions considered. x(@) Each mission is characterized by a climatology and operating profile which may be described using discrete probability as follows: where: Pc (n) is the probability density for P(Hs,To) Joint Probability of Significant the capability function Wave Height and Modal Period Third form: P(VXIHsTo) Conditional Probability for f (.) = 1.0 if x(.) < c(.) Speed and Heading Given = 0.0 if x(.) 2 c(.) Significant Wave Height and Modal Period The first of these functional forms is a linear function with the property P(V,X,Hs,TO)=P(Hs,To)P(VXIHsTo) Joint Probability of Wave that -oo< f(e) < 1.0. A response of zero will produce a function value of Height, Period, Speed and Heading one, a response equal to the criterion value, c(*), will produce a value of zero, and responses greater than the criterion value will produce negative function values (a penalty). Example Equation for Objective Function 1122 The second functional form may be nonlinear and reflects a modern REFERENCES perspective on risk and operability where capability is a distributed probability function. In this instance the function f(e) can take on values 1. Pierson, Dr. W. J. Jr., "A Unified Mathematical Theory for the between zero and one where zero means that x(o) entirely exceeds the Analysis, Propagation and Refraction of Storm Generated Ocean capability and one means that x(o) is entirely below the capability. Surface Waves," Parts I and II, N.Y.U. College of Engineering, Prepared for the Beach Erosion Board, (1952). The third and final functional form shown may be regarded as a special case of the second form, where capability is a step function occurring at 2. Pierson, Dr. W. J. Jr. and St. Denis, Dr. Manley, "On the Motions of the criterion value, c. Ships in Confused Seas," SNAME Transactions, Volume 61 (1953). All of these functional forms share the property that the best criterion 3. Korvin-Kroukovsky, B. V. and Jacobs, W. R., "Pitching and score is one. Since the criterion score contributes to the final value of the Heaving Motions of a Ship in Regular Waves," SNAME objective function, 92, as a product with the weight, all of the relative Transactions, Volume 65 (1957). importance is embodied in the weight function, Mw(*). The authors preference is for the first functional form as it is simple, linear, and offers 4. Salveson, N., Tuck, E. 0., and Faltinsen, 0., "Ship Motions and Sea a penalty for criterion exceedence. Loads," SNAME Transactions, Volume 78 (1970). All of these functional forms are appropriate for non-exceedence type 5. Meyers, W. G., Applebee, T. R. and Baitis, A. E., "User's Manual criteria. In the case of exceedence type criteria there are fairly obvious for the Standard Ship Motion Program, SMP," DTNSRDC (Sept. modifications which could be made to each of these functional forms so 1981). that they would properly participate in the scalar objective function. 6. Hutchison, B. L., "Risk and Operability Analysis in the Marine EnvironmenC, SNAME Transactions, Volume 89 (1981). SUMMARY 7. St. Denis M., "On the Environmental Operability of Seagoing Seakeeping criteria as currently applied to the development of new Systems,"'@NAME T&R 1-32 (1973). research vessel designs represent a significant improvement over the situation of only a few years ago. Such criteria serve as both goals and 8. Spouge, J. R., "The Prediction of Realistic Long-Term Ship guides for design development. However, the current criteria lack clarity Seakeeping Performance," North East Coast Institution of Engineers regarding the concept of "best heading". They also do not provide any and Shipbuilders Transactions (1985). guidance regarding the relative importance of various criterion or operating conditions. Likewise there is no guidance embodied in the 9. Kennel, C., White, B., and Comstock, E. N., "Innovative Naval criteria which would assist the designer in determining the proper location Designs for North Atlantic Operation," SNAME Transactions, on board ship for critical activities. Volume 93 (1985). It is the authors' opinion that most oceanographic research activities are 10. Hutchison, B. L. and Laible, D. H., "Conceptual Design of a sensitive to either local accelerations or relative motions, or both. Rational Medium-Endurance Research Vessel Optimized for Mission seakeeping criteria of the future should emphasize these processes. Flexibility and Seakeeping," SNAME Marine Technology, Volume 24, No. 2, (April 1987). For many oceanographic research activities lateral and longitudinal accelerations are better expressed in vessel rather than earth coordinates. li. Bales, N. K., "Optimizating the Seakeeping Performance of Accelerations in vessel coordinates should include the first-order Destroyer-type Hulls," Proceedings, 13th Symposium on Naval contributions of gravity which arise from roll and pitch. SM? does not Hydrodynamics, Japan, (1980). compute vessel coordinate accelerations but the addition of this feature should be considered for future program enhancements. An alternative 12. van Wijngaarden, A. M., "The Optimum Form of a Small Hull for approach achieving this objective would be the use of post-processors such the North Sea Area," International Shipbuilding Progress, Vol. 3 1, as those discussed in reference [15]. No. 359, (July 1984). Proposals have been set forward for a general criteria structure which 13. "Study of Large Spar Buoy Hull Forms for Motion Minimization - would address the identified concerns. Implementation of such a general Phase I - Linear Motion Analysis of Cylindrical Stepped Buoys," criteria would require effort on the part of the scientific and naval The Glosten Associates, (June 1988). architectural communities. Such effort would however be rewarded by better designs and greater research productivity. 14. Lee, W. T., Bales, S. L., and Sowby, S. E., "Standardized Wind and Wave Environments for North Pacific Ocean Areas," DTNSRDC Report No. SPD-0919-02, (July 1985). 15. Hutchison, B.L. and Bringloe, J.T., "Application of Seakeeping Analysis," SNAME Marine Technology, Vol. 15, No. 4, (October 1978). 1123 41 REMOVAL OF THE WRECK OF THE EX-USS TORTUGA LCDR Jack Strandquist, USN Philadelphia Naval Shipyard Philadelphia, PA 19112 In December 1987, the ex-USS TORTUGA (LSD-26) was set adrift off the coast of Southern California as a target for a missile exercise. Instead of being sunk as planned, the ship went aground on the southeast coast of San Miguel Island during a storm. The stern section broke off before wreck removal operations could begin, leaving the fore section approximately 355 feet long, weighing about 4400 tons, grounded on a hard bottom and impaled on a very large boulder. Salvage operations commenced with the piece-by-piece removal of the superstructure. This lowered the center of gravity and, most importantly, reduced the ground reaction to a reasonable level. Most spaces below the well deck were holed and open to the sea. Bulkheads and piping were repaired and welded over where possible and low pressure air was piped to each space for dewatering purposes. On-site computerized salvage computations were utilized to predict ground reaction, stability and remaining hull girder strength. Work was accomplished with consideration for environmental concerns. Finally, on 20 August 1988 the wreck was pulled from its strand, towed to sea and scuttled in a designated deep water site. CH2585-8/88/0000-1124 $1 @1988 IEEE THE REALITIES OF BAREBOAT CHARTERING Terry MacRae Hornblower Dining Yachts NAPVO (National Association of Passenger Vessel Owners) There is a growing problem in the commercial passenger vessel industry associated with the use of private yachts and uninspected vessels being chartered for the purpose of carrying passengers for hire. Owners and agents of these vessels are inducing businesses and private parties to sign "bareboat charter agreements" which in theory transfer the ownership and control of the vessel to the charterer. Many of these private parties and corporations then utilize these vessels for business purposes. They risk violations of federal regulations as well as other local regulations if more than six passengers are on board on the vessel. The National Association of Passenger Vessel Owners, a group whose members operate U.S. Coast Guard inspected passenger vessels - which allow them to carry passengers for hire in accordance with regulations - are concerned for the well-being of charterers and passengers who may unknowingly or unintentionally use a vessel not licensed for the intended purpose of the charter. NAPVO supports a position of public awareness, self-regulation, and increased enforcement activity of the regulatory agencies responsible for vessel charter activities. The concept of a bareboat charter was originally developed to handle business relationship issues associated with commercial cargo carriers. For example, when "Texaco" has a surplus of oceangoing tankers which they wish to utilize, they may "bareboat charter" them to "Exxon", generally for a five or ten year period. Both companies are experienced in the management of vessels; including the crewing, fueling, provisioning, insuring, berthing and navigation of this type of vessel. Once a true bareboat charter agreement or demise charter is consummated, the charterer becomes the effective owner of the vessel for all but title or legal ownership purposes. He is responsible for the use, care, control and loss of the vessel during the charter period. "Bareboat" literally means the charterer has leased a bare vessel. If you own a private yacht or other vessel which does not meet the appropriate standards set by the U.S. Coast Guard or other regulatory agencies to enable it to carry passengers for hire, you might perceive that there is a lucrative market in turning that vessel into a part-time business enterprise. As a result of this perception, there are many owners who charter their vessels on a part or fulltime basis to defray maintenance, berthing and ownership costs. There are also a large number of agents operating privately-owned vessels under contract for the same purpose, CH2585-8/88/0000- 1125 $1 @1988 IEEE Some of these vessels could be inspected and licensed in accordance with the regulations if the owners would make the decision to spend the additional money required to make the vessels comply with safer standards. However, many of them are foreign built or owned vessels or vessels which are not constructed according to the safety or engineering standards required by U.S. Law. A great number of owners and agents utilize the concept of "bareboat charter" to circumvent the requirement to use vessels which are in compliance with all the appropriate regulations. In effect, they're passing some of the risks associated with safety, liability and propriety along to the charterer. This strategy results in the creation of additional risks for both the owner and charter. Table I shows some of these risks. There is certainly no intent to infer that there are not legitimate bareboat charters - they obviously have a place within the marine industry. Extended duration trips, small (6 passengers) groups on small vessels and charters to industry professionals fully aware of the risks and responsibilities are certainly legal and appropriate examples of legitimate bareboat charters. There are many reputable companies offering services which include legal bareboat charters. Charterers, however, should be careful to avoid situations where an otherwise legitimate bareboat charter may be structured so that the charterer violates passengers -for-hire regulations. The definition of a passenger used by the U.S. Coast Guard in this context is shown in Table 2. Unlike the example cited previously with the oil companies, most individuals or companies chartering a vessel are interested in a 3 to 4 hour cruise activity with a group of friends, business associates or family. They generally have no experience in operating vessels, have no knowledge of captains and crews which they must directly hire, nor are they skilled or knowledgeable in the provisioning, fueling and insuring of vessels, and generally have limited, if any, experience in the direction or supervision of passenger-carrying vessels. In many cases these are multi-million dollar vessels over 100 feet long with complex machinery and navigation equipment. Under the concept of "bareboat chartering", to be a true bareboat charter the charterer must take control and possession of the vessel for the duration of the charter. Over the years, the regulatory agencies and courts have established guidelines as to what constitutes transfer of control and possession. Some of these guidelines are shown in Table 3. In order to constitute a true bareboat charter, most or all of these elements of control must be the charterers direct responsibility. It is worth noting here that the execution of a legal document alone will not create a valid bareboat charter. If the parties do not conduct the business relationship in a way that transfers the possession and control of the vessel to the charterer for the duration of the charter, and take all the risks associated therewith, it most likely will not be construed as a valid bareboat charter. Is it reasonable to believe that the owner of a large passenger vessel or multi-million dollar private yacht is willing to totally transfer the control and possession of a vessel to an inexperienced charterer for a short period of time for a small charter fee or percentage of an agents charter revenues? The vessel owner is unlikely to be. able, in a short period of time, to evaluate the charterer's capability to accept control of the vessel. The charterer of the vessel is certainly not in a position, without the proper experience, to evaluate the risks and provide the services necessary to truly take control of the vessel during the charter period. 1126 Any reasonable evaluation of the bareboat charter concept will tell you that typical charter users such as a company introducing a new brand of perfume, a convention group out for a night on the town or a bride and groom wishing to be married at sea, are not likely to be qualified - nor interested - in accepting the risks associated with taking control and possession of a complex vessel for an evening with business associates or friends. In many cases the charterers of the vessel are not told of their risks and responsibilities until time of departure, at which time often a charter agreement is presented to them as "only a formality". An agent may have arranged for the crew, provided the food and beverage, assured them that the vessel will be filled with fuel, and otherwise attended to all details which an owner would normally have to manage. The agent cannot legitimately maintain control of the vessel and is not responsible for the crew or the liabilities associated with the vessel. -Even a small problem such as an overactive bilge pump in an engine room could provide a small oil slick (water pollution) and expose the charterer to a $5,000 fine. In many cases, agents give the appearance that their vessels are inspected and suitable for the appropriate purposes. For instance,, some West Coast companies ask the U.S. Coast Guard Auxiliary to provide a' safety inspection of their vessels so they can advertise that they are "U.S. Coast Guard inspected". Obviously these vessels do not meet the USCG standards, as the U.S. Coast Guard Auxiliary can only provide a complimentary safety inspection which they would provide to any non-commercial or recreational boater. There are many other regulatory influences associated with the chartering of vessels and the carrying of passengers. Some of these are listed in Table 4 This is by no means a comprehensive listing, as each area has its own specific regulations pertaining to commercial charter activities. Many of these regulations apply only to vessels that are licensed to carry passengers-for-hire. If an owner is using the bareboat charter to circumvent these regulations, he will generally be unaware or unconcerned about the need to comply with many other regulatory requirements. One of the greatest concerns of NAPVO members pertains to the risks associated with operating unsafe vessels. Any vessel carrying a large number of passengers which has a problem will be viewed negatively by the public at large. Other transportation related industries, such as trains or airlines are more heavily regulated and thus have very few private carriers attempting to circumvent regulations. The boating industry does not fit into the same category. Can you imagine the potential problems with airline safety if airplanes could be chartered to carry passengers without compliance with Federal Aviation Authority Standards! The National Transportation Safety Board has recently become aware of the potential for problems associated with illegal and unsafe bareboat charters and may become more active with bareboat charter issues. Owners and bareboat charterers are subject to a number of fines and penalties should they be cited and proven guilty of operating outside of the regulations. The fines for carrying passengers -for-hire go up to $2,000 per passenger, and further, have resulted in seizure of vessels. In addition, operators and crew which have U.S. Coast Guard licenses or certificates may have these suspended or revoked. Table 5 shows the civil 1127 penalties which can be assessed for violation of the passengers-for-hire regulations. This does not include any of the local county or state regulations, which could also result in fines, penalties, and/or suspensions. While the enforcement of these regulations is not the U.S. Coast Guard's highest priority, they have recently increased their efforts to make vessel owners, operators, and the boating public more aware of the risks and penalties associated with violation of federal regulations. Clearly their greatest mission is marine safety, and as a result, the huge amount of activity throughout many U.S. ports associated with the chartering of unsafe vessels is of great concern to them. In many areas where the Coast Guard marine safety office has adequate staff and appropriate priorities they have dramatically reduced the number of unlicensed charter vessels and unscrupulous agents attempting to deceive the unknowing public with sham bareboat charter agreements. However, on the West Coast alone there are still over 75 vessels representing a capacity of over 3,000 passengers offered for charter. A list of some recent USCG enforcement activities and their probable results is included in Table 6. While the U.S. Coast Guard has not been completely successful in halting illegal charters, they are continuing to aggressively enforce regulations pertaining to passengers-for-hire on unsafe vessels. They recognize that the risk of a disaster associated with continuation of such unsafe practices increases daily. This is especially true when you consider special events such as the Statue of Liberty celebration and the Americas Cup race viewing. The U.S. Coast Guard certainly understands that a company or individual without appropriate experience does not fully intend to take possession and control of a large vessel in such a high risk environment. In addition to the U.S. Coast Guard's efforts to more aggressively enforce charter regulations, insurance companies have begun to reevaluate their charterer's liability coverage. They too recognize the risks and costs associated with operating vessels that are not suitable for their intended purpose. In many cases they are not told that the owner is transferring control and possession of the vessel to novices. In addition, marina operators and owners have begun to question whether bareboat charter activities within their marinas are safe. Claims associated with the operation of a charter vessel will certainly go beyond the charterer of the vessel if there is a major problem. In most cases, the charterer of the vessel has not been properly insured, and is sailing from docks owned by others. Most berthage and dock use leases prohibit operation in violation of local, state and federal regulations. In other words, marina operators and owners have the ability and the obligation to stop bareboat charterers illegally carrying passengers-for-hire without the help of other regulatory agencies. If you are associated with these types of activities, and at risk, or you are interested in keeping the commercial passenger vessel industry safe, we would encourage you too to take action to stop illegal charter activities and violation of passenger vessel regulations. The U.S. Coast Guard encourages calls from the concerned public regarding these issues. A listing of the appropriate Marine Safety Office phone numbers is included in Table 7. Like any regulatory agency, the U.S. Coast Guard's resources are limited, and their efficiency will be much greater with more accurate information. Table 8 provides some guidance to assist the public in providing the required information about bareboat charter violations. 1128 The National Association of Passenger Vessel Owners (NAPVO) also encourages the public's support of NAPVO member sponsored chartering and commercial vessel activities. Ultimately, when agents and owners who are attempting to operate in violation of the regulations face the realities of bareboat chartering, they will have to reevaluate their business practices. Recent changes in the tax law have already made it less practical for owners to continue to operate vessels as marginal businesses, with large depreciation losses providing tax benefits. NAPVO membership supports increased enforcement by regulatory bodies which will keep passenger vessel industry safety records intact. NAPVO further continues to support the U.S. Coast Guard, and commends them on their continued efforts to keep the passenger vessel industry safe. Table 1 - Risks for Owners and Table 2 - Definition of a Passenger Charterers in Illegal Bareboat Charters On an uninspected passenger vessel, means an individual E Fines and Vessel Seizure carried on the vessel except- 0 Lack of experience of user in: i) the owner or representative of the owner; Procuring insurance, fuel and provisions ii) the managing operator; Hiring and supervising crew iii) a crewmember engaged in the business of the ves- M Liability sel who has not contributed consideration for carriage on Personal injury board and who is paid for services on board; or Vessel damage (hull insurance) iv) a guest on board a vessel that is being operated Protection & Indemnity insurance only for pleasure who has not contributed consideration for 0 Jones Act compliance carriage on board. E Vessel safety issues Title 46 CFR �2101.21D Table 3 - What Constitutes a Table 4 - Regulatory Bodies Concerned Valid Bareboat Charter? with Illegal Bareboat Charters The vessel's owner must relinquish =1 control and possesion 0 United States Coast Guard of the vessel, including : 0 United States Customs Service � Hiring and supervision of Captain and Crew 0 Alcholic Beverage Control � Fueling 0 Internal Revenue Service IN Provisioning (catering) 0 Public Utilities Commission � Docking and port fees 0 Local law enforcement agencies � Navigation and route 0 County health & sanitation departments (charter must be free to take vessel on any route) � Survey on delivery and return Table 5 - Penalties for Table 8 - Information You Should Have Illegal Passengers-for Hire Operations To Report an Illegal Bareboat Charter Violations found proven carry a maximum civil penalty of 0 Vessel name 0 Location of departure $5,000 per incident (per passenger) and could potentially M Planned departure and return time. lead to forfeiture of the vessel. It should also be realized that should an injury occur on a vessel operating in violation, Also helpful are the names of the Captain(s), crew, insurance policies carried by the vessel's owner would passengers, owner and/or agent. probably be void and the parties involved (including the Time is of the essence. Advise the Coast Guard agent) would carry much more liability in civil litigation. immediately. 1129 Table 6 - Selected Charter Violation Case Histories Proposed Vessel Name ilia Location Alleaed Infraction EaMILV DULCINEA May 1986 San Francisco, CA Illegally carrying 136 passengers $27,000 Violation of Jones Act $34,000 KIALOA October 1987 Richmond, CA Improper Bareboat Charter $52,000 PARADISE 11 July 1988 Erie, PA Carrying passengers-for-hire Vessel Seized without a proper license. CAROLANN July 1988 Erie, PA Illegal Charter Operation $43,000 Table 7 - Investigative Division or Marine Safety Off ices USCG LISCG M= 0 Phone 2iiLkd "I ELQa First Boston, MA 617-835-9040 Eighth New Orleans, LA 504-682-6251 Portland, ME 207-833-3251 Morgan City, LA 504-687-0515 Providence, RI 401-528-5335 Baton Rouge, LA 504-687-0271 Houma,LA 504-868-5595 Second St. Louis, MO 314-425-4657 New Iberia, LA 318-369-6221 Peoria, IL 309-360-7195 Corpus Christi, TX 512-529-3162 St. Paul, MN 612-725-7452 Brownville, TX 512-529-2583 Davenport, IA 319-322-6297 Galveston, TX 409-527-6638 Huntington, WV 304-529-5524 Houston, TX 713-526-7558 Marietta, OH 614-373-5476 Mobile, AL 205-537-2998 Louisville, KY 502-582-5194 Port Arthur, TX 713-527-8333 Evansville, IN 812-335-6275 Lake Charles, LA 318-687-7226 Cincinnati, OH 513-684-3295 Memphis, TN 901-222-3941 Ninth Cleveland, OH 216-522-4405 Greenville, MS 601-497-2400 buffalo, NY 716-846-4168 Nashville, TN 615-852-7362 Alexandria Bay, NY 315-953-8482 Decatur, AL 205-355-7158 Chicago, IL 312-353-5080 Paducah, KY 502-442-1621 Detroit,.Ml 313-226-7777 Pittsburgh, PA 412-644-5808 Milwaukee, WI 414-291-3788 Sturgeon Bay, WI 414-743-9448 Third New York, NY 212-664-7859 Duluth, MN 218-720-5285 New London, CT 203-645-6003 Toledo, OH 419-979-6372 Philadelphia, PA 215-597-4337 St. Igance, MI 906-643-8086 Gloucester City, NJ 609-488-5199 Muskegan, MI 616-759-7508 Fifth Norfolk, VA 804-827-3276 Eleventh Long Beach, CA 213-499-5500 Baltimore, MID 301-922-5145 Santa Barbara, CA 805-965-0407 Wilmington, NC 919-671-4881 San Diego, CA 6119-895-5848 Morehead City, NC 919-670-2438 Twelfth San Francisco, CA 415-536-3`149 Seventh Miami, FL 305-350-4524 Key West, FL 305-296-6825 Thirteenth Seattle, WA 206-399-7510 Jacksonville, FIL 904-791-2648 Portland, OR 503-422-0312 Tampa, FL 813-826-2190 Savannah, GA 912-944-4347 Fourteenth Honolulu, HI 808-551-2070 Charleston, SC 803-677-4392 San Juan, PR 809-725-2697 Seventeenth Juneau,AL 907-586-7307 Ponce, PR 809-766-3497 Anchorage, AL 907-271-5137 St. Croix, VI 809-842-2368 1130 RELATIONSHIP BETWEEN SEAKEEPING REQUIREMENTS AND SWATH SHIP GEOMETRY G. Robert Lamb David Taylor Research Center - Code 1235 Bethesda, MD 20084-5000 ABSTRACT Existing SWATH Craft and Ships Even though in 1987 there were only a dozen SWATH In 1987, to the author's knowledge, there were 12 craft and ships afloat around the world, word of their SWATH craft and ships operating in the world, spanning the markedly superior seakeeping performance is spreading size range from 22 LT to about 3500 LT. Thus far, 5 rapidly. This paper reviews the characteristics of existing SWATH craft have been built in the U.S. and 7 have been SWATH craft and ships from the standpoint of the stated built in Japan. Japan has both the largest SWATH, the seakeeping objective. Hullform differences between four KAIYO [1], and the smallest, the MARINE ACE [2]. The SWATH craft and ships, including the Navy's SSP U.S. Navy has the distinction of having the oldest SWATH, KAIMALINO, are analyzed and interpreted. Important the 220 LT SSP KAIMALINO [3). The TWIN DRILL [4] considerations for the early stage design of a SWATH ship (formerly the DUPLUS) was not included in these totals are discussed. Differences in the range of feasible hullform. because, while it started out life in the mid-1960s in Holland geometries for coastal areas and unrestricted ocean as a SWATH, since then blisters have been added which operations, and for low speed vs. moderately high speed increased the waterplane area substantially. The TWIN applications, are pointed out. DRILL, which displaces 1400 LT, is the only known example of an MWATH (Medium-Waterplane-Area Twin-Hull) ship. The 3500 LT diving support ship KAIYO was delivered Introduction to the Japan Marine Science and Technology Center in June 1985. It was designed and built by Mitsui Engineering and Naval architects know they can increase the seaworthiness Shipbuilding Co. In the U.S. commercial development of the of a surface ship design and reduce its pitch motions in SWATH concept has been limited to craft of less than 150 moderate seas by making the ship bigger and, in particular, LT. However, in November 1986 the Navy awarded a longer. Increased waterline length is advantageous because contract for the construction of the lead ship of a class of long steep waves occur relatively infrequently in the world's 3360 LT SWATH ocean surveillance ships (T-AGOS 19 class) to McDermott Shipyard in Morgan City, LA. The T-AGOS oceans. 19 was designed in-house by the Navy [5]. Looking ahead, the Another factor that contributes to the reduced motions of Navy has plans for other classes of SWATH auxiliary ships. large monohull surface ships is that the wave exciting forces There is also growing activity in the U.S. in commercial acting on them are smaller, relative to the ships' mass. The applications of SWATH ships. Two new U.S. SWATH craft first-order determinant of wave exciting forces is the ships' are currently under construction. waterplane area (area of the horizontal cross-section through the hull at the design waterline). Waves also act on the SWATH SEAKEEPING OBJECTIVE submerged portions of the hull, but the wave forces decrease exponentionally with depth of submergence. This paper summarizes what the author has learned, SWATH (Small-Waterplane-Area Twin-Hull) ship while working at the David Taylor Research Center (DTRC), configurations make it possible for the naval architect to raise about hullform. selection for SWATH ships for good the periods of maximum pitch and roll response significantly seakeeping performance. The approach and judgments higher than for a large monohull ship, while simultaneously contained herein are simplified by considering heave, pitch bringing about a very marked decrease in wave exciting and roll motions to be uncoupled. Nevertheless, it is believed forces. The underlying reason is that, typically, a SWATH that the information presented provides a basis for early stage ship will have no more than one-third of the waterplane area design by naval architects with little previous familiarity with of a monohull of equal displacement. the SWATH concept. It should be acknowledged that there is a price to pay for Designing a SWATH ship is different than designing a these benefits. Comparative design studies have shown that semi-submersible drilling rig or work platform because adopting the SWATH configuration results in a larger full SWATH ships are so much smaller. Drilling rigs are sized and load displacement to carry the same payload weight as a designed to provide very little deck motion even in severe monohull. Consequently, SWATH hullforms should be storm seas. As a consequence, semi-submersibles have 50 or considered only for applications requiring excellent seakeeping 60 ft. of clearance from their operating waterline to the performance for the size of ship in question. CH2585-8/88/0000- 1431 $1 @1988 IEEE underside of the working platform. Moreover, they displace .:-,-., 20,000 to 40,000 LT. !U, The objective of SWATH ship design usually is not near- 0 zero deck motion. Instead, the objective is to reduce ship SSP KAIMALINO motion to an acceptable level. For Navy designs acceptability is usually defined as the following motion limits [6] for onset of degraded human performance: significant single amplitude pitch 3 degrees significant single amplitude roll 8 degrees significant single amplitude vert. accel. 0.4 g's Automotive engineers have developed electrically controlled suspension systems that can be changed from a "highway" setting to a "sport" setting at the flick of a switch. The designer of a SWATH ship must select a combination of strut and lower hull geometry, and appendages, which has an inherent capability for smooth, steady "highway" performance in the design operational sea conditions and "sport" performance in severe sea conditions. In naval architectural terms, the former is called "platforming" while the latter is "contouring" behavior. Table 1. Summary of design operating conditions for existing SWATH craft and ships. Design Design Clearance Design q Displ. Max. Operating to Wet Operating Craft Name (LT) Operational Speed Deck Draft KOTOZAKI Sea State (kts) (ft) (ft) MARINE WAVE 25 3 16 2.7 (Est.) 5.2 HALCYON 57 4 18 4.5 7.0 it SSP KAIMALINO 217 4 0-5 6.0 15.3 KOTOZAKI 232 4 0-5 4.0 10.5 Fig. 1. Comparison of the hullforms of the SEAGULL 338 5 23 7.7 (Est.) 10.3 SSP KAIMALINO and the KOTOZAKI. KAIYO -3500 4 0 11.0 (Est.) 20.7 IMPORTANCE OF DESIGN OPERATING SPEED design operating condition is Sea State 4. In contrast, the AND SEAWAY CONDITIONS KOTOZAKI was designed to carry out water and bottom surveys in the Seto Inland Sea of Japan [9]. The KOTOZAKI Table I summarizes published information on the design went into service in March, 1981 and as of January, 1985 had operating conditions for a number of existing SWATH craft accumulated more than 700 operating days [9]. The largest and ships. It is interesting to note that the KAIYO was observed wave height during these operations was only 8 ft. designed for operations through Sea State 4 [1] while Thus, the prevailing sea conditions for the two craft are SEAGULL was designed for operations in Sea State 5 [7]. markedly different. Since both craft are reported to be very Given that the displacement of SEAGULL is smaller by a successful in providing small ship motion response, it is factor of 10, how is this possible? Two key factors are the instructive to compare their dynamic and hydrostatic difference in their design operating speeds and the difference characteristics. in seaway conditions where the two ships operate. How those Figure 2 is a plot of available data on the measured pitch differences affected these two SWATH designs will be natural periods for the SSP KAIMALINO [10] and discussed later in this paper. KOTOZAKI [7], as well as three other SWATH craft designed by Mitsui [7, 11]. The pitch period of the SSP Comparison of the SSP KAIMALINO and the KOTOZAKI KAIMALINO is 9.7 sec., while the KOTOZAKI's pitch period is 8.9 sec. Also shown on Figure 2 is the pitch period The SSP KAIMALINO and the KOTOZAKI are similar trend line for hypothetical geosims of the KOTOZAKI. (A in size and both were designed for good seakeeping at 0 to 5 geosim is a hullform that has exactly the same proportions as knots in Sea State 4. In addition, they are both capable of the base ship because all dimensions have been multiplied by transit speeds of about 20 knots. Yet Figure I shows these the same scale factor.) The pitch periods of two of the three two craft are very different in appearance. Why is this so? other Mitsui SWATH craft fall on this trend line. The The SSP KAIMALINO was designed by the Naval Ocean MARINE ACE has a substantially shorter pitch period, but is Systems Center (NOSC) as a work platform for their Hawaii an early experimental prototype. Figure 3 is a similar plot for Laboratory [8]. The seas in the vicinity of the NOSC Lab on the roll natural periods. The roll period of 15.8 sec. for the the island of Oahu are often rough. Sea State 5 and Sea State SSP KAIMALINO is substantially longer than the 6 conditions characterized by long modal periods are KOTOZAKI's roll period of 10.7 sec. Two of the three other 1,0- experienced frequently. The necessity of coping with rough Mitsui craft have roll periods that are close to the trend line seas influenced the configuration of the craft, even though the for geosims of the KOTOZAKI. One reason why the roll 1132 period of the SEAGULL is somewhat shorter than the trend line is that the buyer specified that the transverse metacentric 9.0- height had to be large enough to resist the heeling moment SSP KAIMALINOO that would result from all 450 passengers crowding along the 00 SEAGULL rail on one side of the ship. 0 KOTOZAKI 2 6.0 - Cc -0- 15 LU LU Q TREND LINE FOR GEOSIMS SSP KAIMALINO 3.0 - MARINE LU OF MARINE ACE 10 MARINE 0 X ACE 0 ___0-__ - WAVE KI SEAGULL UJ KOTOZA (L 0 X 5 0 TREND LINE FOR GEOSIMS - 0 50 150 2W 350 MARINE OF KOTOZAKI FULL LOAD DISPLACEMENT (L. Tons) ACE I I I Fig. 4. Heave period trend line for existing SWATH craft. 0 50 150 250 350 750 FULL LOAD DISPLACEMENT (L. Tons) Cr GEOSIMS OF MARINE ACE Fig. 2. Pitch period trend line for existing SWATH craft. A 4 1\1 - - uJ 500 20 1 4 LU z CHUBASCO... SSP KAIMALINOO 5 250- I-Avo OSSP KAIMALINO- 15 IL HALCYON (Est.) co UJ O'MARINE ACE a KOTOZAKI 0 MARINE -0- 0 1 w 10 - WAVE SEAGULL- 0 50 150 250 350 UJ CL 0 FULL LOAD DISPLACEMENT (L. Tons) _J o _J TREND LINE FOR GEOSIMS 0 Fig. 5. Waterplane area trend line for geosims of the Ir 5 -0 OF KOTOZAKI MARINE ACE. MARINE ACE The greatly reduced waterplane area of the SSP 01 KAIMALINO further reduces wave exciting forces in a given 0 50 150 250 350 seaway relative to the forces that would be acting on the FULL LOAD DISPLACEMENT (L. Tons) KOTOZAKI. Because of its very small waterplane area, the Fig. 3. Roll period trend line for existing SWATH craft. SSP design needs a 10-ft. wider hull spacing to satisfy intact roll stability requirements. In addition to increasing deck area and enclosed volume, the wider hull spacing results in greater Figure 4 is a plot of available data on the heave natural roll inertia. This is one of the reasons that the roll period of periods for existing SWATH craft. The SSP KAIMALINO's the SSP is substantially longer than the KOTOZAKI's roll heave period of 8.5 sec. is considerably longer than the period. KOTOZAKI's period of 5.8 sec. The heave periods of the The exceptional seakeeping performance of the SSP at Mitsui-designed craft appear to follow a trend line for low speeds is shown in Figures 6-8. These full-scale trial geosims. In this case, the trend line was based on the measurements [14] were taken near Oahu at a nominal ship MARINE ACE because there is data available on the total speed of 5 knots. The measured sea spectra during these trials strut waterplane area of this craft [2]. The trend line for the had significant wave heights between I I ft. and 13.5 ft, The waterplane area of geosims of the MARINE ACE is shown in measured spectral wave energy peaked at a wave period of Figure 5. It was a surprise to see that the estimated about 12 sec. This is termed the modal period of the wave waterplane area of the RMI-designed HALCYON [12, 13] spectrum. The Navy criteria for onset of degraded personnel falls on this trend line. This is also the case for the performance are also plotted on Figures 7 and 8. Only in CHUBASCO's waterplane area, which was furnished courtesy following seas is the limit of 3 deg. single amplitude pitch of SWATH Ocean Systems, Inc. Since all of the Mitsui exceeded. At this heading, assuming linearity of motion SWATH craft have similar oval hull cross-sections, and since responses, the pitch limit would be reached in seas with a sig. their heave periods fall on the trend line for geosims, it is wave height of about 8 ft., i.e., high Sea State 4. This was reasonable to infer that the waterplane areas of the the original design objective. On the other hand, Figure 8 KOTOZAKI and SEAGULL would fall on the trend line shows that the measured roll motions are far below the limit shown in Figure 5. This inference leads to an estimated total of 8 deg. single amplitude even though the trial conditions waterplane area of 536 sq. ft. for the KOTOZAKI, while the correspond to high Sea State 5. Similarly, the vertical SSP KAIMALINO has only 247 sq. ft. of waterplane area. accelerations are only one-fourth of the criterion of 0.40 g's Even after making an adjustment for the small difference in sig. single amplitude. These very low acceleration values are the displacements of these two craft, their waterplane areas the result of the SSP's long heave and pitch periods. The differ by a factor of 2. measured vertical acceleration values are also well below the 1133 15 1 1 1 1 more stringent criterion of 0.20 g's which is often used for 'a applications requiring the handling of equipment on deck. 0 V In the author's judgment, it would have been possible to satisfy the Sea State 4 seakeeping objective with a SWATH 10- design that had substantially more waterplane area than the SSP if low speed performance was the only requirement. Z - CRITERION However, NOSC also desired the ability to transit along the 5- coast at a speed of 20 knots or more. Given the common occurrence of Sea State 5 and 6 conditions, it is advantageous Z CD for the SSP to have a long heave period Tor best transiting us 01 1 1 1 1 performance. 0 50 100 150 200 Figure 9 shows the effect of a ship speed of 20 knots on SEAWAY HEADING (deg) the ship/wave encounter period for a ship heading into waves t t of various periods. It can be seen that the encounter period FOLLOWING HEAD with the waves of maximum energy in a seaway having a Fig. 7. Variation of significant pitch double amplitudes 12-sec. modal period is 7.7 sec. The ratio of the SSP's heave with seaway heading for the SSP KAIMALINO period (8.5 sec.) to the wave encounter period is called the at a nominal speed of 5 knots [141. heave tuning factor. In this case the heave tuning factor is 1.10. Figure 10 shows schematically the variation in nondimensional heave response for a SWATH operating at low speeds as a function of the tuning factor value. The least 0.3 1 1 1 1 heave response occurs when the tuning factor is greater than 1.20. This region is termed the supercritical region and it is UJ characterized by platforming motion behavior. The subcritical region covers values of up to 0.75 for the tuning factor. In LIJ Z 0.2 the subcritical region the ship moves up and down in phase 2 with the wave, and with a heave amplitude about the same as Z the wave amplitude. The greatest heave motion response cc LU LL -J 0.1 occurs when the value of the tuning factor is nearly unity, LU i.e., heave resonance. When the encounter period is equal to the heave period, a SWATH ship operating in regular waves can experience heave motions as large as twice the wave 0 50 100 150 200 amplitude. Moreover, at resonance the ship's heave motion SEAWAY HEADING (deg) will be out of phase with the wave, thereby increasing the t t incidence of wave impacts on the wet deck. In a more realistic FOLLOWING HEAD irregular seaway the amplification of heave responses in resonant wave lengths will be smaller, but will still be very Fig. 6. Variation of significant heave acceleration double evident. amplitudes with seaway heading for the SSP KAIMALINO at a nominal speed of 5 knots [141. 20 20 1 1 1 1 Z CRITERION 15 .0 -J 2 -J LU 0 IL W cc VK 0/ 10- LU Z 10 - Z 0 Q VK 20 Z 5 Z LU LU 0 50 100 150 200 -t SEAWAY HEADING (deg) t FOLLOWING HEAD 0 6 10 is 2D Fig. 8. Variation of significant roll double amplitudes WAVE PERIOD (sec) with seaway heading for the SSP KAIMALINO Fig. 9. Effect of ship speed on wave encounter period in @R LTE@111@ VK 0 Al VK 20 at a nominal speed of 5 knots [141. head seas. 4134 I I I Figure I I compares the regions of platforming and RESONANT resonant response at a ship speed of 20 knots for two J REGION different heave periods: 6.0 sec. and 8.5 sec. The former is 0. approximately the same as the heave period of the 2 3 SUBCRITICAL SUPERCRITICAL_ 4 KOTOZAKI. For a ship moving at 20 knots in head seas, a UJ (CONTOURING) (PLATFORMING) 8.5 sec. heave period extends the region for platforming behavior through wave periods of I I sec., while a 6 see. 2 period limits platforming behavior to wave periods of less than 9 see. A SWATH craft having head sea platforming CL capabilities limited to wave periods of less than 9 sec. would UJ have been unacceptable for operations off of Oahu. Passenger Ferry SEAGULL Figure I I is also helpful for explaining why the passenger 0 0.6 1.0 1.5 2.0 ferry SEAGULL, which displaces only 338 LT, is able to sustain operations in Sea State 5. The SEAGULL has a heave TUNING FACTOR = THEAVE /T ENCOUNTER period of 6.2 sec. and a pitch period of 9.5 sec. [7]. In head Fig. 10. Typical variation in SWATH ship heave response seas, at its normal transit speed of 23 knots, t 'he SEAGULL at low speeds as a function of tuning factor. will exhibit platforming heave response in waves having periods shorter than 9 sec., and will exhibit platforming pitch response in wave periods shorter than 12 sec. According to Narita [71, the most probable wave period on the SEAGULL's route is 6 sec. Thus, the SEAGULL has natural heave and pitch periods that provide minimal heave and pitch 5 10 15 20 responses on its design operating route. With platforming behavior it is essential that there is an adequate amount of clearance to the underside of the wet deck. At its design 3 HEAVE PERIOD 6.0 sec waterline the SEAGULL has about 7.7 ft. of wet deck C learance amidships and about 9.4 ft. of clearance at the bow and stern. Usually, the SEAGULL operates at a draft about 6 PLATFORMING BEHAVIOR inches deeper than the design value, so the wet deck clearance is about 7 ft. amidship and about 9 ft. at the bow and stern. 2 The ferry company which operates the SEAGULL has a RESONANT policy of cancelling operations only when seas with wave REGION heights in excess of 3.5 m (11.5 ft.) are forecast [9]. CONTOURING Typically, if a given seaway has a significant wave height of BEHAVIOR 11.5 ft., the'average height of the one-tenth highest waves will be about 14.7 ft. The corresponding wave amplitude is 7.35 0 ft. Since there is little ship heave motion at its design operating speed, the SEAGULL has about the minimum acceptable amount of wet deck clearance amidships for its 0 operating requirements. Z D 3 HEAVE PERIOD 8.5 sec Diving Support Ship KAIYO LU In contrast to the SEAGULL, the seakeeping PLATFORMING BEHAVIOR requirements of the KAIYO apply to operations which are -the ship at rest. It is judged that the most carried out with 2 X demanding of the KAIYO's operations is lowering one of the RESONANT submersible diver decompression chambers into the water. REGION This is carried out through the ship's center well using four vertical guide rails. During this operation only a minimal CONTOURING BEHAVIOR amount of pitch and roll motion can be tolerated. In addition, the relative vertical motion between the ship and waves must also be small. The latter requirement can be met by ensuring a platforming-type heave response in the design -sea conditions. It is estimated that the KAIYO has a heave period of 00 6 10 15 20 about 10.5 sec. Figure 12 shows the wave period regions for WAVE PERIOD (sec) platforming and contouring response for zero ship speed. The limiting wave period for platforming is about 8.8 sec. Fig. 11. Comparison of the regions of platforming and Referring to Figure 13, it can be seen that modal periods as contouring response at 20 knots in head seas for ship short as 8 sec. have a high probability of occurrence in 2.5 m heave periods of 6.0 and 8.5 seconds. 1135 seas (high Sea State 4). It is for this reason that high Sea State 4 is considered to be the limiting condition for the 15 diving support portion of the KAIYO's mission. This example also illustrates why zero speed, open ocean operations are KAIYO (Est.) among the most demanding seakeeping requirements that a SWATH designer must satisfy. 2 10 -SSP KAIMALINO 6A MODEL 0 0- Uj 0 CL NTREND LINE FOR SELECTION OF WATERPLANE AREA FOR LU SEAGULL 5 k SMALL SWATH CRAFT LOW SPEED OPERATIONS LIJ KOTOZAKI MARINE ACE Figure 14 compares the estimated heave period of the I I I I I I I _j KAIYO with the trend line developed in Figure 4 based on 0 0 1000 2000 3000 existing small SWATH craft. The heave period of the KAIYO FULL LOAD DISPLACEMENT (L. Tons) Fig. 14. Comparison of the estimated heave period for the KAIYO with the trend line for small SWATH craft. 3 V 0 kts is 15 percent above the trend line. It appears that a conscious decision was made to raise the heave period. Also shown in PLATFORMING BEHAVIOR Figure 14 is the data point for the DTRC 6A model, which was designed as a candidate configuration for a 2900 LT 2 RESONANT oceanographic research ship. The measured heave period for Z REGION the 6A model is 10.1 sec. [15]. This data point is also about CONTOURING 15 percent above the trend line. Results of the seakeeping Uj BEHAVIOR experiments with the 6A model were documented by Kallio [15]. Uj The heave natural period of a surface ship can be estimated using the equation shown below: I V (1 + A '331 01 T HEAVE 2n + 0 5 10 15 20 r_@g7A_w p WAVE PERIOD isec) Fig. 12. Regions of platforming and contouring response where V = the displaced volume of the ship in head seas for a ship heave period of 10.5 seconds. Af 33 = the hydrodynamic heave added mass divided 20 by the ship mass g = gravitational acceleration GENERAL NORTH PACIFIC - ANNUAL AWP = total waterplane area For a simple SWATH ship with lower hulls having a circular 4 10.0 rn cross-section, such as the 6A model, the A 33 term has a value Uj 10 - < 5.0 rn of about 0.70. This means that the hydrodynamic heave 0 < 2.5 rn added mass is 70 percent of the ship's mass. For a monohull, Z the heave added mass is typically equal to the ship's mass. Cr 'The SWATH designer has only two ways of increasing the ship's heave period. Either the value of A 33 must be 0 0 increased orthe total amount of waterplane area must be U. 20 reduced. It has been determined from model experiments that 0 GENERAL NORTH ATLANTIC - ANNUAL the -heave added mass of a SWATH ship with elliptical lower Z hulls having,a horizontal axis 1.8 times the vertical axis LU L) amounts to 120 percent of the ship's mass. For a given design W ul displacement and waterplane area, selection of 1.8:1 elliptical CL 10 4 10.0 M lower hulls would result in a heave period that is 14 percent 4 5.0 rn longer than the equivalent configuration with circular lower < 2.5 m hulls. However, in most cases practical design considerations will preclude the use of such highly elliptical hulls. Recent Navy designs -have had oval cross-section hulls with a 0 horizontal axis 1.4 times the vertical dimension. The added 0 10 20 30 mass of a SWATH with this type of lower hull is almost 90 SPECTRAL MODAL PERIOD (see) percent of,the ship's mass. Fig. 13. Variation in probability of occurrence of different In addition to the structural considerations, the designer's spectral modal periods for several significant wave ability to reduce waterplane area is constrained by the need to height bands [6]. maintain adequate static roll stability. For a SWATH ship, 1136 the amount of transverse waterplane inertia, IT, is is a 50 percent increase in the measured heave damping factor approximately equal to the following: as ship speed is 'increased from zero to 3 knots. At 11 knots, the measured damping factor is seven times greater than with I (hull spacing 2 (2) the ship at rest. Analyses have determined [16] that the T = (waterplane area) x 2 amount of heave damping is largest when the total fin area is distributed between forward and aft pairs of fins, similar to Equation (2) shows that, if the amount of waterplane area is the distribution shown in Figure 15. Making the aft fins larger decreased by 50 percent, the hull spacing must be increased by than the forward fins also serves to counteract the the square root of 2, i.e., 41.4 percent. So large an increase in destabilizing Munk moment [16]. hull spacing usually is undesirable from an overall design standpoint because it will result in a large increase in structural weight and enclosed volume. Effect of Panama Canal Beam Limit 18.60' 1754 MEAN STABILIZER CANARD M AN For Naval designs there is a reluctance to accept the 12.4- 8,77'--] 77.0' concept of a relatively small displacement ship that cannot pass through the Panama canal. Consequently, the overall beam is not allowed to exceed 105 ft. The 2900 LT SWATH 6A configuration has a hull spacing of 75 ft, and an overall beam of 90 ft. [15]. A feasible alternative geometry is a Fig. 15. Plan view of 1981 model of notional 3000 LT SWATH with 20 percent wider hull spacing, resulting in an overall beam of 105 ft. With 20 percent wider spacing, it is SWATH ship. possible to reduce the amount of waterplane area to 70 While the amount of heave damping at 3 knots is small percent of the 6A value without reducing transverse stability. even for the fully appended model, the resulting reduction in This approach would increase the heave natural period to maximum heave response is significant. Figure 16 shows the about 12.1 sec., as compared with 10.1 sec. for the 6A measured nondimensional heave response at a full-scale speed geometry. However, referring back to Figure 13, there is of 3 knots in regular head waves. The peak heave response essentially the same probability of encountering Sea State 6 was only 1.6 times the wave amplitude, and it occurred in spectra with a 12 sec. modal period as with a 10 sec. period. waves with a period of about 13 sec. Contouring behavior is The amount of waterplane area would have to be decreased to evident in wave periods of 15 sec. or greater, since the a very small value to raise the heave period sufficiently to measured heave responses are equal to the wave amplitude. ensure platforming heave behavior at zero speed in the open Below a wave period of 12 sec. there is little heave motion. In ocean. Faced with the impracticality of ensuring platforming these shorter period waves the model shows platforming heave responses, what the SWATH designer must do is to behavior. provide an adequate amount of heave damping. Low Speed Heave Damping J 3 1 1 1 1 1 1 a. 0. Figure 15 shows the plan view of a 1981 DTRC model of 2 2 2- a notional 3000 LT SWATH designed for low speed LU operations. This model has elliptical lower hulls with the > horizontal axis 1.8 times the vertical axis, thereby increasing the heave added mass and damping. Even so, it was determined from free oscillation tests that the bare hull heave 00 200 400 600 8W 1000 1200 1400 1600 damping factor for this model at zero speed was only one WAVE LENGTH (ft) percent of the critical damping value. Addition of the bow and stern fins shown in Figure 15 increased the measured heave damping factor to 3.6 percent of the critical value. The 7 9 11 13 15 17 measured heave damping factor values at other speeds are WAVE PERIOD (sec) shown in Table 2. Fig. 16. Measured nondimensional heave response at 3 knots in regular head waves for the 1981 model of a Table. 2. Effect of stabilizing fins on the measured heave 3000 LT notional SWATH ship. damping factors for the model of a 1981 notional 3000 LT SWATH ship. Implications for Ship Heave Accelerations Full-Scale Speed, Kts 0 3 11 Due to the lightly damped, tuned nature of a SWATH Bare Hull Damping Factor 0.010 0.012 - ship's heave response curve at low speeds, the peak heave Appended Damping Factor 0.036 0.057 0.263 acceleration occurs at the heave natural frequency. Consequently, the ballpark value of the ship's peak heave It is apparent from the experimental data that the four fixed accelerations in the design operating Sea State can be fins on this model had a dominant effect on total ship heave computed using a simplified method. For any single-degree-of- damping at zero speed. Moreover, with the fins in place there freedom oscillation, the acceleration amplitude is equal to: @@l 7j5 _.-@@@STA@BILIZE@RCA@W 18 Mj6()' JM A@N 1137 (frequency of oscillation in radianS)2 X oscillation amplitude. degrees significant single amplitude. An underlying reason is Assuming that the ship's heave motions increase linearly with that its pitch period of 9.7 seconds is shorter than the increasing wave height, the peak heave response for the measured wave spectra modal period of 12 seconds. While it notional 1981 design for a 3000 LT SWATH ship would be can be seen in Figure 2 that the SSP's pitch period is longer 1.6 times the design wave height. As an example, it will be than the typical value for a SWATH craft of its size, the seas assumed that the design maximum operating condition is high around Oahu often have substantial wave energy in the range Sea State 6, defined by a significant wave height of 19.7 ft. of periods near 10 seconds. The measured heave frequency of the appended model at 3 knots is 0.546 rad/sec. The significant double-amplitude heave 3000 acceleration for this ship in its design limiting sea condition is I I estimated as shown below: TREND LINE FOR GEOSIMS OF,..- SMALL SWATH CRAFT 00 Sig. double-ampl. accel. = (0.546)2 X 19.7 ft. x 1.6 .01 = 9.41 ft/sec.2 TWIN DRILL 0 6A MODEL = 0.29 g's UJ 2000 - X The corresponding significant single-amplitude UJ acceleration value of 0.145 g's is well below the standard Z Navy criterion of 0.4 g's. While the estimate assumes heave motion only, and so applies only to the acceleration value at cc LU the ship's center of gravity, worst case assumptions were used 1000- for heave response period and amplitude. Confidence in this TREND LINE FOR GEOSIMS estimating methodology is gained from published full-scale OF 6A MODEL trials data for the KAIYO [1]. The highest reported significant single-amplitude acceleration value for the KAIYO in high Sea State 6 is 0.16 g's. The estimated heave period of the KAIYO OSSP KAIMALINO is 10.5 sec., while the heave period of the 1981 DTRC 3000 1 _J LT Model is 11.5 sec. Thus, the two ships are roughly 0 0 1000 2000 3000 comparable. FULL LOAD DISPLACEMENT (L. Tons) The amount of waterplane area for the 3000 LT model is essentially the same as for the 6A model. However, the highly Fig. 17. Comparison of the waterplane area trend lines for elliptical lower hull of the former results in a measured heave SWATH craft and geosims of the 6A model. period that is 1.4 sec. longer. In the light of the very low heave acceleration values, it is judged that the 6A model's Two factors that affect the zero speed pitch period of a heave period of 10 sec. is sufficiently long. Using the same surface ship are the ship's length and the value of the assumptions and methodology as in the previous example, the longitudinal metacentric height (GML). The GML is a measure estimated significant single-amplitude heave acceleration for a of the ship's longitudinal waterplane inertia. The equation for SWATH with a 10 sec. heave period is 0.19 g's in high Sea estimating a ship's pitch period is shown below. State 6. A personnel limit of 0.20 g's is frequently assumed for the handling of gear on deck. A shorter heave period is advantageous for ensuring contouring behavior in the long TPITCH @ 2ni If g GML 55) (3) period waves which have the greatest. energy in State 7 and higher seas. where Figure 17 compares the previously developed waterplane L = length of the ship area trend line for small SWATH craft with the trend line for kP = the pitch gyradius divided by the ship length geosims of the 6A model. It can be seen that the 6A model, A, 55 @ the hydrodynamic pitch added inertia divided at 2900 LT, has only 73 percent as much waterplane area as a by the ship mass times the ship length squared scaled-up SWATH craft would. The reason is that the g = acceleration of gravity planned SWATH oceanographic ships have especially GML = longitudinal metacentric height stringent seakeeping requirements. However, there may be other applications where an amount of waterplane area close In the case of the SSP, equation (3) gives an estimated pitch to the small craft trend line would be acceptable. For period of 9.3 sec., as shown below. informational purposes, the estimated waterplane area of the TWIN DRILL is plotted on Figure 17. The TWIN DRILL Z8)2 + 0.060] has almost 60 percent more waterplane area than a 6A model TPITCH = 2 geosim. 'IV 32.174 ft/sec.2 X 12.6 ft. TPITCH = 9.3 sec. SELECTION OF PITCH PERIOD FOR LOW SPEED OPERATIONS The SSP's GM L of 12.6. ft., which is large for a SWATH " "'Oe '0' ,@@TREND LI NE FOR GEOSIMS OF 6A MODEL el craft of its size, resulted in a shorter pitch period than is The previously discussed full-scale trials data for the SSP optimum for Hawaiian waters. The relatively high GML is KAIMALINO, Figures 6-8, consists of measured motion and due to the tandem-strut configuration of the SSP, combined acceleration levels far below all but one of the standard Navy with the desire for access to the lower hulls. Due to its fins, motion limit criteria. The sole exception is the pitch limit of 3 the pitch period increases to 14 sec. at 15 knots (10]. 1138 Figure 18 compares the author's estimate of the KAIYO's pitch period with the extrapolated trend line for small 50 SWATH craft. Surprisingly, the estimated pitch period of 14 seconds for the KAIYO falls right on the trend line. The measured pitch periods for two DTRC models designed for 40- .1981 MODEL low speed operations are considerably longer. Referring to Figure 13, it is apparent that there is considerable likelihood 6A MODEL of encountering Sea State 6 spectra with modal periods as long as 16 seconds. In order to minimize pitch motions in 30- such a seaway, the pitch period of a SWATH designed for AUTHOR'S TREND LINE low speed operations should be at least 20 percent longer, or 0 19.2 seconds. Since this long a pitch period is achievable with 20- SSP KAIMALINO UJ a suitably designed SWATH ship of 3500 LT, a trend line is shown in Figure 18 for geosims of the assumed design. cc I- lo- TREND LINE FOR DTRC 6A MODEL 20- LOW SPEED SWATH MODELS 0 0 0 1981 MODEL 0 1000 2000 3000 15- FULL LOAD DISPLACEMENT (L. Tons) Fig. 19. Trend line for the ratio of strut length to the square 0 KAIYO (Est.) root of GML for low speed SWATH ships. cc SSP T 10 - N, L , SEAGULL TREND LINE FOR SMALL KOTOZAKI SWATH CRAFT (L 57 MARINE WAVE Longitudinal Metacentric Height MARINE ACE SWATH ship designers are often concerned about the 01 effect of their GML selection on pitch performance, but a 0 1000 2000 3000 more meaningful parameter is the ratio of strut waterline FULL LOAD DISPLACEMENT (L. Tons) length to the square root of GML' Figure 19 shows a trend Fig. 18. Comparison of the estimated pitch period for the line developed by the author for the value of this ratio as a KAIYO with the trend lines for small SWATH craft function of ship displacement. This trend line is intended to and DTRC models. apply only to 6A type SWATH configurations being designed for low speed applications. For informational purposes, the It is judged that there are two reasons why the KAIYO's data point for the SSP is also shown. Strut length is used, pitch period is shorter than the author's SWATH ship trend rather than lower hull length, because most of the ship's mass line. The first reason is that, in order to minimize the strut is concentrated within the strut length. Strut length is thus the length and avoid shifting the longitudinal center of buoyancy best measure of the ship's pitch gyradius. too far forward, the lower hulls of overhanging strut SWATH Within the small community of SWATH ship designers configurations are relatively short. Typically, the lower hull of there has been much discussion of the relative merits of a SWATH with an overhanging strut at the stern is 15 percent configurations with a single strut per side and those with a shorter than the lower hull of an equivalent 6A type pair of struts, in tandem, on each hull. In 1976 DTRC carried configuration [15]. This decrease in lower hull length results out seakeeping experiments on a series of SWATH models in a decrease of roughly one-third in the hydrodynamic added called the SWATH 6 series. In the initial series of experiments pitch inertia, compared with a 6A configuration. Since for a [15], motions and accelerations were measured with three 6A configuration the hydrodynamic added pitch inertia is different strut configurations on the same lower hull. The first approximately equal to the pitch mass moment of inertia, the two, struts A and B, were single-strut configurations. The reduction in total pitch inertia for the overhanging strut third configuration, strut C, was a tandem-strut configuration hullform is about 17 percent. The resulting decrease in the having the same waterplane area as struts A and B. The pitch period would be 9 percent. In reality, the shorter lower design displacement of the SWATH 6 models was 2900 LT. hulls also reduce the pitch mass moment of inertia, with the The previously mentioned 6A configuration had the shortest result that the reduction in pitch period is closer to I I or 12 struts and the lowest GML at 20 ft. The 6B struts were percent. A second factor that comes into play is that both longer, resulting in a GML of 38 ft. The 6C configuration, ends of the struts of overhanging strut configurations tend to with tandem struts, had a GML of 45 ft. Experimental data be fuller than with 6A configurations. Consequently, for for the 6B and 6C models showed similar heave and pitch equal strut lengths the GML of a SWATH with overhanging motion responses at all headings and speeds in both regular struts will be greater than for a 6A type design. A larger and random waves. Figure 20 compares the heave and pitch GML value further decreases the ship's pitch period. One responses of all three strut configurations at zero speed in solution to this problem is to adopt a shorter strut for random head waves. Based on these results, it was concluded @6A MODEL KAIYO type designs. Obviously, there are practical limits on that the strut hydrostatic characteristics, namely GMLI have a how far the naval architect can go in this direction for a given greater effect on SWATH seakeeping than whether each hull design. has one or two struts. 1139 lies midway between the trend line for small craft and the SIGNIFICANT HEAVE trend line for DTNSRDC models. While the KAIYO's roll period is very long by monohull standards, if the ship had 0 STRUT A been designed for Sea State 6 operations a roll period of at LU 0 STRUT B least 20 seconds would have been chosen. 40 - *STRUT C Uj 1981 TREND LINE FOR DTFIC 0 MODEL- z 20 -LOW SPEED SWATH MODELS 20 - @6AMODEL z 0 OSSP cl;@ 15 - 0 KAIYO (Est.) 0 cc 0 00 SEAG 01 LU 10 ULL TREND LINE FOR SMALL - 20 a. KOTOZAKI SWATH CRAFT _j _j NARINE WAVE 0 SIGNIFICANT PITCH 5 @r_MARINE ACE CL 10 - *0 z 0 01 0 0 1000 2000 3000 00 FULL LOAD DISPLACEMENT (L. Tons) z 0 0 0 - Fig. 21. Comparison of the estimated roll period for the 0 10 20 30 KAIYO with the trend lines for small SWATH craft SIGNIFICANT WAVE HEIGHT (ft) and DTRC models. Fig. 20. Significant double amplitudes of motions in random The two dominant factors that affect the roll period of a head seas for SWATH 6 series models at 0 knots surface ship are the waterline beam and the transverse 1151. metacentric height (GMT). The equation for estimating a Importance of the Separation Between LCB and LCF ship's roll period is shown below. The maximum buoyant force exerted on a SWATH by T (4) the crest of a long wave tends to be centered about the ROLL @ 2 g GMT centroid of the waterplane area, which is termed the where longitudinal center of flotation (LCF). Ship weight, on the other hand, is centered near the longitudinal center of B= waterline beam buoyancy (LCB). Consequently, if there is substantial kr= the roll gyradius divided by the waterline beam separation between the LCF and the ship's LCB, the wave A'44 = the hydrodynamic roll added inertia divided by buoyant forces will result in a pitch moment. Since the small the ship's mass times the waterline beam GML of a low speed SWATH results in a small hydrostatic squared pitch restoring moment, substantial pitch motion can be 9= acceleration of gravity caused by the LCB/LCF separation. This problem can be GMT = the transverse metacentric height avoided if the naval architect selects strut and lower hull Using the 6A model as an example, the above equation gives geometries which result in minimal LCB/LCF separation. an estimated roll period of 17.9 seconds, as shown below. SELECTION OF ROLL PERIOD FOR 487)2 + 0.20] LOW SPEED OPERATIONS IIROLL = 2 32.174 ft/sec.2 x 11.06 ft. It was explained earlier in this paper that the wide hull spacing of a SWATH ship results from the need to satisfy TROLL = 17.9 sec. static stability requirements. The wide beam makes it possible for the designer to largely detune the roll motion In this sample calculation the value used for the ship's roll characteristics of the ship from the wave periods of concern gyradius is the value assumed for the 6A model experiments. in the design Sea State. Figure 21 compares the author's Subsequent studies have determined that a more typical value estimate of the roll period for the KAIYO with the for the roll gyradius of a SWATH is an amount equal to 46 extrapolated trend line for small SWATH craft. Also shown percent of the waterline beam. in Figure 21 are the measured roll periods for the 6A model and the 1981 model of a notional 3000 LT design. Based on Transverse Metacentric Height the average roll period for these two models, a trend line was developed for low speed SWATH ship designs. Figure 21 For monohulls, a rule of thumb value for the GMT is 10 shows that the KAIYO's estimated roll period of 18 seconds percent of the ship's waterline beam. This percentage is also 1140 reasonable for a SWATH ship and has the virtue of making minimum acceptable clearance to the wet deck would be an the SWATH sensitivity to heeling moments about the same as amount equal to the average amplitude of the one-tenth a monohull. On the other hand, from the viewpoint of highest waves in Sea State 6, i.e., 12.5 feet. Figure 24 shows designing a SWATH for seakeeping, the most meaningful the measured relative bow motion response of the 6A model parameter is the ratio of waterline beam to the square root of at several speeds in regular head waves. At 20 knots the GMT' Figure 22 shows a trend line developed by the author relative bow motion is about 1.7 times the wave amplitude for for the value of this ratio as a function of ship displacement. wave periods between 14 and 15 seconds. This type of The basis for the trend line is the average value of the response leads to a required wet deck clearance of 21 feet, B/VrGMT ratio for the 6A and 1981 models. For comparison using the same criterion of the average of the one-tenth purposes the data point for the SSP is also shown. Usually, highest waves. Little benefit can be gained from active control the minimum acceptable value for GMT for a Navy design is in this situation because the ship/wave encounter period at 20 established by the requirement to resist the heeling moment knots is less than the ship's heave period of 10 seconds. It is due to a wind speed of 100 knots [17]. Because SWATH ships very difficult to force a SWATH ship to move faster than its typically have relatively high GMT values, the intact stability heave natural period. Since the 6A model was designed for requirements are easily satisfied. However, damaged stability transiting at 20 knots in Sea State 6, it had 20 feet of considerations may dictate a higher GMT* clearance. The situation with regard to pitch responses is 40 ------- considerably different. Because forward speed decreases the period of encounter with waves of a given length in head 1981 MODEL seas, platforming pitch responses can be maintained in longer 30 - 0 period waves. Alternatively, the previous goal of platforming SSP KAIMALINO in wave periods up to 16 seconds can be achieved with a 0 0 shorter pitch period. It follows that a higher GML is 6A MODEL acceptable for SWATH ships that normally operate at 10 to 20- LU 15 knots than for a SWATH designed for very low speed M AUTHOR'S TREND LINE operation. Figure 25 compares the author's trend line for the Uj Z ratio of strut length to \/GML for medium-speed SWATH _j lu - ships to the trend line shown in Figure 19 for low speed ships. LU WAVE LENGTH Ift) 0 1 1 1 1 1 0 400 800 1200 1600 2000 0 1000 2000 3000 1 1 1 1 1 FULL LOAD DISPLACEMENT (L. Tons) LU Uj 0-0 0 kn 00 Fig. 22. Trend line for the ratio of waterline beam to the M :) 0--* 10 kn square root of GM for low speed SWATH ships. t 2 - 0-0 20 kn T =i CL Lr---6 28 kn 22 Generally, the naval architect has more latitude in 4 < lnrx@A&6 selecting the GMT value than for other hydrostatic LU LU parameters. Care should be taken to ensure that the resulting Uj < roll period differs from the pitch period so that coupled pitch mi and roll motions are minimized. Similarly, the roll period 01 L should not be exactly twice the heave period, to minimize 8 10 12 14 16 18 20 coupled heave-roll motions. The easiest solution is to select a roll period that is more than twice the heave period. WAVE PERIOD (sec) Fig. 23. Effect of forward speed on the heave response of the EFFECT OF FORWARD SPEED IN HEAD SEAS 6A model in regular head waves [151. For a monohull the effect of forward speed in head seas WAVE LENGTH Ift) LU 0 400 800 1200 1600 2000 is to shift the peak pitch response to somewhat longer waves. For a SWATH the trend is similar but the shift is more pronounced. Figure 23 gives the measured heave response of LU 0-0 0 kn CL 0----o 10 kn the 6A model at various speeds in regular head waves. At 20 2 knots the peak heave response occurs at a wave period of < P 2 - C)--0 20 kn - Z _j inr- - 28 kn 15.5 seconds, whereas at zero speed the peak occurred at the 6 CL ship's heave period of 10. 1 seconds. It follows that the ship/wave encounter period at 20 knots with the 15.5 sec. LU 1 wave must be about 10 seconds. Figure 9 confirms this. Since > wave periods as long as 16 seconds have a reasonably high likelihood of occurrence in Sea State 6, a speed of 20 knots is J1 0 not high enough to ensure platforming behavior for the 6A LU model in all Sea State 6 conditions. 8 10 12 14 16 18 20 WAVE PERIOD (sec) <P Kn '0 '0 k n 2' kn 28 k. Perfect platforming behavior implies there is zero heave motion. In such a situation the relative bow motion response Fig. 24. Effect of forward speed on the relative bow motion is 1.0 because it is equal to the wave amplitude. A reasonable response of the 6A model in regular head waves [15]. 1141 AUTHOR'S TREND LINE 20 - 6A MODEL + FOR OPS AT 0-5 kn 1981 MODEL LU AUTHOR'S TREND LINE FOR 9' 40- U Z PLATFORMING MODE OF 6A MODEL cc OPERATION IN MODERATE SEAS 30- UJ _J 1981 MODEL KAIYO AUTHOR'S TREND LINE HALCYON LU 10 - + FOR OPS AT 10-15 kn Z 20 - SSP KAIMALINO UJ LU SEAGULL_ _J S E@P, D 5 AUTHOR'S TREND LINE /+KOTOZAKI FOR CONTOURING MODE 10- -MARINE WAVE IN MODERATE SEAS 10 00 20 00 30W 0 1000 3000 FULL LOAD DISPLACEMENT (L. Tons) FULL LOAD DISPLACEMENT (L. Tons) Fig. 26. Comparison of trend lines for midship wet deck Fig. 25. Effect of forward speed on the recommended ratio clearance for contouring and platforming modes of of strut length to the square root of GML for operation in moderate sea conditions. SWATH ships. The preceding discussion assumes that it would be EFFECT OF FORWARD SPEED IN FOLLOWING SEAS deemed acceptable for a 3000 LT SWATH to slow down in Sea State 7 and higher seas, thereby reducing wave impacts by Up to this point there has been relatively little discussion contouring the waves of greatest energy. If the SWATH of pitch motions, which everyone knows are a serious problem design of interest is exclusively for low speed operations, then with monohulls. The reason is that the author has assumed a reduced amount of wet deck clearance is acceptable because that a sufficiently long pitch period has been chosen to detune the ship will contour all but the shortest period waves in Sea the ship from the wave periods of concern at low or moderate State 6. The 1981 low speed 3000 LT design had a wet deck speeds in head and beam seas. In following seas the pitch clearance of 15 feet. The T-AGOS 19 also has 15 feet of response of a SWATH can be detuned only for very low speed clearance at the forward end of the cross-structure, but only operation. Figure 27 shows that at moderately high speeds in I I feet of clearance amidship. According to the author's following seas, wave encounter periods of 20 to 40 seconds are estimate, the KAIYO has I I feet of wet deck clearance. very likely. Moreover, these very long encounter periods occur Presumably, the operators of the KAIYO will slow down in over a wide range of wave lengths. For this reason, wave very severe seas. The instinctive response to slow down, encounter conditions corresponding to pitch resonance cannot learned from operating monohulls in severe seas, is also the be avoided. There are similarly long wave encounter periods at best response for a SWATH ship. However, this is not always moderately high speeds in stern quartering seas. The likelihood the case with SWATH craft. of experiencing wave encounter periods corresponding to either Figure 11 shows the regions of platforming and pitch or roll resonance is high. contouring response for a typical SWATH craft with a heave Fortunately, these very long encounter periods also make period of 6 seconds, at a speed of 20 knots. It can be seen it possible to derive considerable benefit from active control. that contouring behavior will occur in waves of 13 seconds or Figure 28 shows the results of a recent analysis of the effect of longer. In these long waves it will be possible to gain active control on the limiting significant wave height for a substantial benefit from active motion control, since the wave notional 1600 LT SWATH transiting at 12 knots [18]. As has encounter periods are longer than the craft's heave period. been explained in more detail by McCreight and Stahl [6), the Even though these same conditions are likely to be close to limiting wave height is the lowest wave height at which one of the pitch resonance conditions, this would not pose a problem the motion limit criteria is exceeded. Figure 28 shows that because the pitch motions of such a craft are well damped at active control raised the limiting wave height considerably in 20 knots. stern, quartering and beam seas, but was of no benefit in bow Figure 26 compares trend lines developed by the author or head seas. In this analysis active control was used to reduce for the minimum recommended amount of midship wet deck both pitch and roll motions. Interestingly, the improvement clearance for SWATH ships designed to platform or contour with just the aft fins active was found to be the same as with moderately severe seas for the ship's size. Small SWATH both pairs of fins active. craft will usually be designed with a platforming philosophy for such conditions, while larger SWATH ships will usually be SUMMARY designed with a contouring philosophy. In either case, it is recommended that somewhat more clearance be provided at In this paper the author has described the principles and the forward end of the cross-structure than in the midship key considerations in SWATH hullform selection for excellent region. It probably bears repeating that all SWATH ships seakeeping. The importance of strut hydrostatic characteristics "I'R '0 N "M M 0 DEL K 11 A 0 ALCYO N LL SEAG U SS, A I@ E should be designed to contour extreme seas at a low speed, and damping provided by the stabilizing fins was explained in similar to a monohull. considerable detail. The necessary differences in approach for 1142 2. Oshima, M., Narita, H. and Kunitake, Y., "Development of the Semi-Submerged Catamaran (SSC)," NAVAL ARCHITECTURE & OCEAN ENGINEERING, Society 28 of Naval Architects of Japan, Vol. 18, 1980. UJ 3. Hightower, J.D. and Seiple, R.L., "Operational Experiences 40 - with the SWATH Ship SSP KAIMALINO," Paper No. 0@ UJ 78-741, AIAA/SNAME Advanced Marine Vehicles 20 z Conference, San Diego, CA, Apr. 1978. n 0 101 4. -IUC's New Multipurpose Vessel 'Twin Drill' Combines Q Geophysical, Geotechnical, ROV Technologies," SEA i@ TECHNOLOGY, Feb. 1985. U. 0 0 5. Covich, P.M., "SWATH T-AGOS: A Producible Design," a I I I Paper No. 86-2384, AIAA 8th Advanced Marine Systems 0 0 FX Conference, San Diego, CA, Sep. 1986. LU 7. McCreight, K.K. and Stahl, R.G., "Recent Advances in the Seakeeping Assessment of Ships," NAVAL ENGINEERS JOURNAL, Vol. 97, No. 4, May 1985. 28 7. Narita, H., et al., "Design and Full Scale Test Results of SPEED Semi-Submerged Catamaran (SSC) Vessels," Proceedings, 20 (kn) First International Marine Systems Design Conference on 10 Theory and Practice of Marine Design, London, Apr. -40 1982. 8. Lang, T.G., -SSP KAIMALINO; Conception, Developmental History, Hurdles and Success," Paper No. 86-WA/HH-4, ASME Winter Annual Mtg., Anaheim, CA, 0 400 1200 2000 Dec. 1986. WAVE LENGTH (ft) 9. Mabuchi, T., Kunitake, Y. and Nakamura, H., "A Status Report on Design and Operational Experiences with the Semi-Submerged Catamaran (SSC) Vessels," Proceedings, 7 9 11 13 15 17 19 RINA International Conference on SWATH Ships and WAVE PERIOD (sec) Advanced Multi-Hulled Vessels, London, Apr. 1985. 10. Fein, J.A., Ochi, M *D. and McCreight, K.K., "The Fig. 27. Effect of ship speed on wave encounter period in Seakeeping Characteristics of a Small Waterplane Area, regular following seas [151. TWin-Hull (SWATH) Ship," 13th ONR Symposium on Naval Hydrodynamics, Tokyo, Oct. 1980. 11. Komoto, M., Nishimura, K. and Nakamura, H., "Mitsui E - Sea Saloon 15 'Marine Wave'," Paper No. A[AA-86-2368, 8.0 - AIAA 8th Advanced Marine Systems Conference, San - UJ Diego, CA, Sep. 1986. 12. Luedeke, G., et al., "The RMI SD-60 SWATH LU 6.0 X - ACTIVE Demonstration Project," Paper No. 10, RINA International LU 4.0 Conference on SWATH Ships and Advanced Multi-Hulled UJ 5 co Vessels, London, Apr. 1985. 62.0 INACTIVE 13. Lewis, Q.M., Jr., "Trials and Tribulations-Operational Fn Experiences with the HALCYON," Joint SNAME/ASNE 0.0 Mtg., Philadelphia, PA, Dec. 1985. FOLLOWING BEAM HEAD 14. Fein, J.A., "Low Speed Seakeeping Trials of the SSP SHIP HEADING TO THE WAVES KAIMALINO," David W. Taylor Naval Ship Research and Fig. 28. Predicted effect of active control on the limiting Development Center Report No. SPD-0650-04, Mar. 1978. significant wave height at various headings for a 15. Kallio, J.A., "Seaworthiness Characteristics of a 2900 Ton notional 1600 LT SWATH transiting at 12 knots Small Waterplane Area Twin-Hull (SWATH)," David W. 1181. Taylor Naval Ship Research and Development Center Report No. SPD-620-03, Sep. 1976. 16. Lee, C.M. and Curphey, R.M., "Prediction of Motion, the design of small SWATH craft and larger oceangoing ships Stability, and Wave Load of Small-Waterplane-Area, Twin- were discussed. Rather than being somehow intrinsic in the Hull Ships," Trans. SNAME, Vol. 85, 1977. SWATH concept, excellent seakeeping performance was shown 17. "Design Data Sheet-Stability and Buoyancy of U.S. Naval to be the result of a careful, engineering-based approach to Surface Ships," DDS 079-1, Naval Ship Engineering Center, SWATH hullform design. Aug. 1975. 18. McCreight, K.K., Hering, J.A., and Waters, R.T., REFERENCES "Seakeeping and Maneuvering Assessment of Swath [email protected] IV E AGOR 23 Configurations;' David W. Taylor Naval Ship 1. Saeki, M. and Nakamura, H., "Motion Characteristics of Research and Development Center Report No. the KAIYO," OCEANS '86, Washington, D.C., Sep. 1986. SPD-1198-01, July 1986. 1143 CONCEPTUAL DESIGN OF AN INTERMEDIATE SIZE OCEANOGRAPHIC RESEARCH SHIP FOR THE UNIVERSITY-NATIONAL OCEANOGRAPHIC LABORATORY SYSTEM Mark Rice Ed Craig Maritime Applied Physics Scott Drummond corporation Carolyn Junemann Washington, D.C. SEACO, A Division of SAIC Alexandria, VA - 2 knot current - 12 foot seas Payload: Specified Mission Equipment ABSTRACT Plus 50 Ton Itinerant Load Accomodations: 30 persons in 2 person staterooms 2 persons in 1 person In response to requirements set by the UNOLS Ship staterooms Improvement Committee, SEACO and Maritime Applied Deck Live Load: 1200 psf Physics corporation completed the conceptual design Centerwell: 15' x 30' of an intermediate-size Small Waterplane Area Twin Seakeeping: Sea State 4 Sea State 5 Hull (S%IATH) general purpose oceanographic research Heave: 2.2 ft. heave 4 ft heave ship. After consideration of alternatives, a 150 Pitch: 3 deg. pitch 4 deg. pitch foot, 1070 ton, tandem strut, diesel-electric SWATH Roll: 3.5 deg. roll 4.5 deg. roll was conceived. This paper reviews the requirements and describes the resultant design. mission Equipment: Cranes SHIP REQ(]IREMFNTS - 20,000 lb offload (side and/or stern) - 10,000 lb working close The general. ship requirement, as provided by the in to stbd side UNOLS Ship Improvement Committee is quoted below: - 5,000 lb at 30 ft from stbd side "The ship is to serve as an intermediate size A-Frames general research ship. The overriding - Stbd side, 30,000 lb required characterisiic is that the ship lift provide the most stable environment possible - Stern, 30,000 lb lift, in order to allow both over-the-side and 151 x 25' throat laboratory work to proceed in greater capacity Portable Laboratories and in higher sea states than is now possible. - 2 ea., 8'x 81x 201 vans, other general requirements are for large 12,000 lb ea. scientific parties and greater flexibility in winches use of laboratory/deck spaces than is now 1 ea oceanographic available aboard intermediate size ships." w/40,000 ft 1/2 inch wire A portion of the detailed performance requirements 1 ea hydrographic are listed in Table 1. w/30,000 ft 3/8 inch wire SEABEAM, navigation & TABLE 1 - UNOLS PERFORMANCE REQUIREMENTS stationkeeping systems Laboratory Space: 2,000 sq. ft. minimum Draft: 18 feet maximum Damaged Stability: 1 compartment flooding Speed- 14 knots for 2 hours Deck Tiedowns: 1 inch threaded sockets on 2 12 knot in sea state 4 foot centers 10 knots in sea state 5 Towing Loads: 12,000 lbs @ 6 knots 6 knots in sea state 6 25,000 lbs @ 2.5 knots Endurance: 6,000 Nautical Miles, or Noise: continuous acoustic doppler 15 days @ cruise plus 15 days profiling at 12 knots towing, or SEABEAM continuous operation 15 days @ cruise plus 15 days at 8-10 knots stationkeeping Stationkeeping: 15b foot watch circle at 45 degrees from specified heading under the influence of: 35 knot wind CH2585-8/88/0000- 1144 $1 @1988 IEEE DESIGN FEATURES General characteristics of the resulting design are provided in Table 2. A profile drawing is found at Figure 1. The present work represents one trip around the classic "design spiral" in a funding constrained effort. The design configuration presented in this paper will change as refinements are made in subsequent design iterations. Principal features of the design and the rationale which led to their selection are discussed below. M 03 113 113 Gag@_v A im m cnnn EMCMC :Fi@ ll II Im' 3,h 11 66& ----j 0 0 0 0 0 0 0 0 0 -IN 0 0 0 - - - - - - - - - - - -- -- - - - - - - - - - - - - - - - - - - - - - - - - - FIGURE I - UNOLS 150 PROFILE TABLE 2 - GENERAL CHARACTERISTICS OF UNOLS 7,000 nautical miles @ 181-4" SWATH-150 departure draft Cruise Speed: 12 knots Basic Dimensions Maximum Speed: 14.5 knots Length Overall: 1501 Stationkeeping: 150 foot watch circle under Beam on Main Deck: 75' UNOLS conditions. Beam at Lower Hulls: 741-1.7" Mission Features: Draft at Full Load: 171 Accommodations: 32 persons Construction Material: HSLA-80 Steel Laboratory Area: 3,650 sq. ft. Forward Strut Length: 55' Open Deck Working Area: 5,000+ sq. ft. Aft Strut Length: 54' Winches: Center Well Diminsions: 151 x 301 - Markey-8 (30,000 lbs/30,000 ft/100 ft/min) Basic Hydrostatics - Pinehill Aluminum Waterfall Type (2x3O,OOO Full Load Displacement: 1070 Long Tons ft) Vertical Center of Buoyancy: 8'-3.62" ABL Vans: 2 ea. 12,000 lb (8x8x2O') Waterplane Area: 961 ft^2 Main Machinery: Transverse Waterplane Inertia: 820,000 ft^4 Main Engines: 2.ea., 900 kw @ 1200 rpm Tons Per Inch at DWL: 1.64 Auxilary Engines: 2 ea. 250 kw Vertical Center of Gravity: 24.651 Main Propulsion: 2 ea. Azimuthing Thrusters, GMT: 5.6 feet 94" at 1,000 shp KML: 47 feet Bow-Thrusters: 2 ea. 34w Tunnel Thrusters Longitudinal Center of Buoyancy: 75.51 from FP Driveline: AC/DC Electric Longitudinal Center of Flotation: 811 from F? Propulsion motors: DC Motor @ 1,000 shp/1150 Vessel Performance rpm Endurance: 5,600 nautical miles @ 171 Total Electrical Capacity: 2190 kw departure draft 1145 Fuel Consumption at Cruise: 750 lbs/hr (125 gph) Full Fuel Load: 188 long tons @ 1814" draft Stern Configuration UNOLS committee members expressed a preference for designs in which the upper hull structure is the aft-most extension of the hull. In such designs, oceanographic personnel are able to work over the stern without concern about directly fouling the propellers or steering gear. Although superior seakeeping, resistance and mission load capacity could result from an extension of the lower hull beyond the transom, such a hull form was not considered to be operationally acceptable. The present concept design employs tandem struts with the trailing edge of the aft struts in the same vertical plane as the transom. FIGURE 2 PERSPECTIVE OF HULL FORM Waterplane can be fabricated by rolling only one radius. The UNOLS specified stringent motion requirements midbody has principal dimensions of 17 feet (width) including the desire for heave motions of 4 feet or by 12 feet (height) while the 4.5 foot radius is less, a significant roll limit of 4.5 degrees and a maintained at the four corners. The resistance is significant pitch limit of 4.0 degrees all in sea given as a function of speed in Figure 3. state 5 . The significant wave height for sea state five was defined as 12 feet with a modal TOTAL DRAG VNWUT THRUSTIM period of 9.2 seconds. Early in the conceptual design it was concluded that a tandem strut design 6060ML would provide a vessel with minimum waterplane area while affording the required stability and lower 50.000- hull access. The present design has a waterplane area of 961 square feet, a transverse metacentric height of 5.6 feet, a longitudinal metacentric height of 47 feet. With these characteristics, a 20,000 pound load handled 10 feet over the side 130.000- induces a heel of 3.7 degrees while a 50,000 pound load handled 5 feet over the stern causes a trim of 20AW 2.1 degrees. Resistance 10.0001 a 9 to U 12 13 14 13 16 The design requirements included a 12 knot cruising YMM BPM - EMM speed in sea state 4 and a maximum speed of 14 FIGURE 3-PREDICTED RESISTANCE VERSUS SPEED knots. The selection of a tandem strut arrangement made it necessary to vary lower hull geometry as a function of longitudinal station to obtain Propulsor desirable wavemaking drag characteristics. Resistance predictions for a number of hull forms Requirements for stationkeeping included a 150 foot were compared before the hull form shown in Figure watch circle under the combined influence of a 35 2 was selected. The forward section of the hull knot wind, 2 knot current and 12 foot seas. Under consists of a cylinder with diameter of 9 feet. the these forces, the ship must be capable of Between the forward section and the expanded maintaining heading within +/- 45 degrees of any midbody lies a semi-conical transition. In this specified heading. This requirement was analyzed transition, the 4.5 foot radius of the forward to determine the maximum transverse thrust and section is maintained and triangular flat plate is pivoting moment required. Following analysis of used to fill the space left as the four quadrants various combinations of loads at various headings, are expanded. The resulting seini-conical expansion stationkeeping performance was collapsed to a single requirement for 50,000 pounds of athwartship thrust. Under all conditions investigated, the wide spacing of the propulsors provided adequate pivoting moment; however, the athwartship thrust requires special propulsion machinery. Two approaches were considered. The first used transverse tunnel thrusters fore and aft with conventional propellers behind the lower hull while 1146 the second employed azimuthing thrusters suspended TOTAL LIGHTSHIP 1764511 76.5 below the aft struts. The size of the four tunnel 27.4 thrusters needed to produce 50,000 pounds of transverse thrust led to a substantial increase in Fuel & Lube Oil 340480 70.0 machinery weight. For this reason, the use of 5.0 azimuthing thrusters was assumed in the baseline Water 21000 12.5 design. An open 94 inch propeller was selected for 30.0 the zee drive input from a 1,000 horsepower D.C. Foodstuffs 15000 56.3 electric motor. When used in conjunction with 33.3 modestly sized bow thrusters, this arrangement is Personnel & Effects 15000 70.0 capable of producing the required transverse. 45.0 Itinerant Equipment 112000 110.0 Structure 40.0 margin (7.3% of lightship) 128811 60.0 HSLA steel was selected as the hull material while 21.0 superstructure above the weather deck was designed in 5456 aluminum. Athwartship wave-induced bending FULL LOAD DISPLACEMENT 2396802 75.5 forces of 963 tons were assumed to act at 8.5 feet 24.6 above baseline to compute the primary load in the hull structure. Secondary loads (normal loads), including slam loads on the underside of the upper hull, were based upon accepted Navy structural Endurance criteria. Primary and secondary loads were combined, again using Navy criteria, to determine The endurance requirement was established not by the design stresses. Parametric summaries of the overall range specification but by the plate- stiffener combinations were used to select requirement for 15 days of transit followed by 15 the combination which produced minimum panel days of oceanographic towing. The fuel required weight. Overall structural weight estimates were for this mission is 188 long tons with a 6 percent made by performing detailed structural design at tankage margin. A decision was made to allow draft more than 30 locations and expressing results as a to increase beyond 17 feet to teinporari.ly panel weight per unit area. These results were accommodate the situation in which this.fuel is then used to extrapolate all panel weights to yield required in conjunction with the specified the overall structural weight estimate. The itinerant mission payload of 50 tons. Since the resulting structural design employs transverse clearance between the upper hull and the water line frames on 1.8 inch centers and exhibits a structural is large at the 17 foot draft (10.5 feet), the weight to full-load displacement ratio of 0.57. additional draft does not result in a severe degradation of seakeeping capabilities. To Weight accommodate the 188 ton fuel load and the full itinerant load, the vessel must sail at a draft of P modified Ship work Breakdown Structure (SWBS) was 18'-4". The vessel will return to its design draft used to account for ship weight. The feasibility gradually during the subsequent 112 hours of .of the design is closely tied to accurate transit. Fuel tanks were located in the lower prediction weight and center of gravity location. hulls amidships to minimize trimming moment as fuel For this reason, disproportionate attention to is used. weight accounting was given in this conceptual design phase. The resulting three digit breakdown Arrangement vields a fully loaded displacement of 1070 long tons as suffmrized in one-digit form in Table 3. Factors which determined arrangement included specification of a large center well, the need for quiet laboratory space, a desire for excellent TABLE 3 - ONE DIGIT WEIGHT BREAKDOWN habitability, a specified capability for load handling over the starboard side, the stern and WEIGHT WEIGHT LCG through the center well, and the general VOG hydrostatic requirement to move the longitudinal CATEGORY Ubs) center. of gravity as far forward as practical. The ----------------- - ----------------------------- resulting arrangement has machinery spaces forward, ------------------ laboratory spaces aft on the centerline and Bull Structure 1366134 77.6 berthing space outboard along the aft two-thirds of 27.4 the upper hull. The arrangement works well with Propulsion Plant 125794 75.7 the bulkheads required for damaged stability (one- 18.4 compartment criteria). The arrangement of Electric Plant 30490 45.0 bulkheads allows an open transverse dimension of 41 29.7 feet along the centerline thereby providing for Command & Surveillance 11319 47.1 relatively large laboratories which have large- 33.1 hatch access to the center well and stern porch. Auxiliary Systems 181824 75.6 A compromise in this design is the placement of 32.8 messing facilities above the auxiliary machinery Outfit & Furnishings 48950 76.4 space; however, it was felt that sufficient 30.1 1141 acoustic treatment could be provided to minimize the noise level in the messing area. Seakeeping Results of seakeeping predictions indicated a capability to met UNOLS requirements at most headings. However, the heave requirement in head and following seas proved troublesome. The longitudinal GM was increased to make room for propulsion machinery in the aft strut. This increase led to excess vertical accelerations at the extreme ends of the vessel. It is anticipated that geometric changes in the aft strut thickness and GKL in conjunction with fixed bilge keels for lower hull damping will correct the vertical acceleration. SUMARY As a first-pass conceptual design, the current effort demonstrates the feasibility of meeting UNOLS requirements for an intermediate-size oceanographic ship with SWATH technology. The present design offers realistic estimates of weight, modest powering requirements, excellent mission capabilities and good seakeeping. Future efforts will be centered on the improvement of at- rest seakeeping performance in head and following seas. 1148 DESIGN OF A SEMISUBMERGED SWATH RESEARCH AND SURVEY SHIP Roy D. Gaula , Alan C. McClure b and Fred E. Shumakerc Blue Sea McClured Suite 317 14300 Cornerstone Village Drive Houston, Texas 77014 ABSTRACT stabilizing fins, usually leads to a higher capital cost than for a rougher riding, but Variable draft is shown to be an essential otherwise equivalent, monohull. feature for a research and survey SWATH ship large enough for unrestricted service worldwide. The distinguishing feature of the semisubmerged An ongoing semisubmerged (variable draft) SWATH SWATH ship is variable draft. Sufficient can be designed for access to shallow harbors. allowance for ballast transfer is made to enable Speed at transit (shallow) draft can be the ship to vary its draft under all load comparable to monohulls of the same power while conditions. The shallowest draft is well within assuring equal or better seakeeping usual harbor limits and gives the lower hulls a characteristics. Seakeeping with the ship at slight freeboard. It also permits transit in deeper drafts can be superior to an equivalent low to moderate sea conditions using less SWATH that is designed for all operations at a propulsion power than is needed by a constant single draft. The lower hulls of the draft SWATH. The semisubmerged SWATH gives more semisubmerged SWATH ship can be devoid of fins. design flexibility to provide for deep draft A practical target for interior clear spacing conditions that strike a balance between between the lower hulls is about 50 feet. operating requirements and seakeeping Access to the sea surface for equipment can be characteristics. Intermediate "storm" drafts provided astern, over the side, or from within a can be selected that are a compromise between centerwell amidships. one of the lower hulls can seakeeping, speed, and upper hull clearance to be optimized to carry acoustic sounding avoid slamming. A discussion of these and equipment. A design is presented in this paper other tradeoffs in semisubmerged SWATH ship for a semisubmerged ship with a trial speed in design for oceanographic applications is given excess of 15 knots, a scientific mission payload in a paper by Gaul and McClure 5 . A more of 300 tons, and accommodations for 50 personnel. general discussion of design tradeoffs is given in a later paper6. 1. SEMISUB14ERGED SWATH TECHNOLOGY The semisubmerged SWATH technology gives rise to some notable contrasts with constant draft A single draft for the full range of operating SWATH ships. For any propulsion power applied, conditions is a common feature of typical SWATH the semisubmerged SWATH has a range of speed ship designs. This constant draft characteristic that depends on draft. Highest speeds are is found in the SWATH ships built by Mitsuil,2 obtained at minimum (transit) draft. Because 3 4 the lower hull freeboard is small at transit most notably the KAIYO , and the SWATH T-AGOS which is now under construction for the U.S. draft, seakeeping at service speed ban be made Navy. The constant draft design for ships of equal to or better than an equivalent this size (about 3,500 tons displacement) poses monohull. The ship is designed for maximum two significant drawbacks. One is that the draft speed at transit draft so the lower hull form must be at least 25 feet to satisfy seakeeping is more akin to a surface craft than a requirements. This draft is restrictive for submarine. This allows use of a nearly access to many harbors that would be useful for rectangular cross section for the lower hulls research and survey functions. The second is which provides damping of vertical motion. For that hull and column (strut) hydrodynamics moderate speeds at deeper drafts with the highly generally result in the SWATH being a larger ship damped lower hull form, the ship need not be and having greater power requirements than for equipped with stabilizing fins. Since maximum an equivalent monohull. The ship size and hull speed is achieved with the columns (struts) out configuration, together with the necessity for of the water, it is practical to use two a. President, Blue Sea Corporation b. President, Alan C. McClure Associates, Inc. c. President, Omega Marine Engineering Systems, Inc. d. Joint venture of Blue Sea Corporation and Martran Consultants, Inc. CH2585-8/88/0000- 1149 $1 @1988 IEEE columns, rather than one, on each lower hull. ference with sonar transducers that are The four column configuration at deep drafts positioned exclusively in the starboard side minimizes the variation of ship motion response lower hull. The diesel generators are in the with change in course relative to surface wave upper hull to minimize acoustic noise radiation direction. The width of the ship and lack of underwater. At normal operating draft, the appendages on the lower hulls increases the greater depth of the propellers compared to a utility of a large underside deck opening monobull suppresses cavitation and associated (moonpool) amidship. acoustic radiation. The underside moonpool in the centerwell gives favorable access for launch The basic Semisubmerged SWATH Research and and retrieval of equipment at a position close Survey Ship design has evolved from requirements to the center of ship motion. The three deck first stated by the Institute for Geophysics of upper hull and crane arrangement gives good the University of Texas (UTIG) in 1984. Blue Sea access to winches and other equipment. The McClure provided the only SWATH configuration in broad expanse of flat main (weather) deck gives a set of five conceptual designs procured ample space for installation of vans, retention competitively by the University. Woods Hole of off-line equipment, landing of helicopters, Oceanographic Institution, on behalf of the and installation of antennas and above water University-National Oceanographic Laboratory instrumentation. System, subsequently contracted for a revision of the UTIG design to meet requirements for an 3. CONFIGURATION AND ARRANGEMENTS oceanographic research ship. The design was further refined to meet requirements posed by The Semisubmerged SWATH Research and Survey Ship the U.S. Navy for an oceanographic research consists of a rectangular upper hull supported ship. The intent of this paper is to use this by two slender columns on each of two lower generic design to illustrate the main features hulls. Table 1 is a list of principal of semisubmerged SWATH ships. particulars. General arrangement drawings for the bow elevation, outboard profile and working 2. DESIGN APPROACH (third) deck are given in Figures 1, 2 and 3, respectively. The design discussed in this paper is for a general purpose oceanographic research and Four columns (struts), two port and two hydrographic survey ship to be used in both starboard, extend upward from the lower hulls to coastal and deep ocean environments. The design provide support for the upper hull. Each column is readily adaptable to other applications such is 60 feet in length, 9 feet 6 inches in width, as geophysical research and exploration. The and 23 feet 6 inches high. The columns are overriding design goal is good seakeeping, both offset outboard on the lower hulls. dead-in-the-water and underway. Typical missions which were assessed in developing the design are: The ship has a maximum beam of 104 feet 11 inches which encompasses the lower hulls and 1) Physical, chemical and biological allows passage through the Panama Canal. Each oceanography. of the lower hulls have a width of 27 feet 11 2) Multidiscipline environmental inches and a depth at midship of 17 feet 6 investigations. inches. The clear space between the inboard 3) ocean engineering and marine acoustics. sides of the lower hulls is 49 feet 1 inch. 4) Hydrographic surveys. There is no cross bracing between the hulls. 5) Gravimetric and magnetometric surveys. 6) Geophysical surveys. The ship is designed to have a transit draft of 16 feet 8 inches (approximately 10 inch The basic design approach is to provide a freeboard amidship) when fully loaded and with variable draft (semisubmerged) SWATH ship to the ballast tanks empty. Under all load suit the complete range of anticipated operating conditions, the ship can be ballasted down to a conditions. The ship is designed to float on the normal operating draft of 27 feet. It may be lower hulls fully loaded. This is required to operated at intermediate drafts as dictated by obtain a suitable draft for entry into shallow sea conditions. This feature is especially water harbors and to reduce propulsion useful during transit to permit higher speeds requirements while transiting in mild to and to save fuel by operating at the shallowest moderate weather conditions. The ship can be draft that is consistent with seakindliness operated at deeper drafts for transiting in heavy requirements. To allow for ease of deploying weather and while conducting mission related and retrieving equipment, the ship can be activities at low speeds. The design is ballasted to a special operating draft of 34 optimized to have minimum response to wave feet. At this draft, the clearance from the motions in this mode of operation. water to the third deck work area is 9 feet 6 inches. Unique features of the semisubmerged SWATH ship configuration have been used advantageously in The upper hull has three continuous decks: main positioning research equipment and improving (weather), second, and third, reckoning from the operations. For example, a single bow thruster top down. In the middle of the upper hull is an is in the port side lower hull to avoid inter- open centerwell bounded vertically by the main 1150 Table 1. Principal Particulars Ship Length, Overall on Lower Hulls 225 ft Beam, Overall 104.9 ft Height, Keel to Main Deck at side 61.5 ft Draft, Transit, Full Load 16.7 ft Draft, Storm 23 ft 19 Draft, Normal Operations 27 ft Draft, Special operations 34 ft Lower Hulls (2) Length 225 ft Beam 27.9 ft Depth, Midship 17.5 ft Spacing, Transverse Centerline 77 ft Columns (4) Length, Overall 60 ft Beam, Overall 9.5 ft Figure 1. Bow elevation. Height 23.5 ft Spacing, Transverse Centerline 90 ft Spacing, Longitudinal Centerline 110 ft and third decks. The centerwell at the third deck level is comprised of a moonpool (opening UpTer Hull in the third deck) which is surrounded by work and staging areas. There is adjacent access to Length, Centerline at Third Deck 162 ft all laboratories, stern and starboard side work Beam, Midship 90 ft areas, change room, storage rooms, and other Beam, Maximum 99.5 ft mission spaces. The centerwell is serviced by a Depth to Main Deck 20.5 ft 10-ton capacity overhead traveling gantry crane for handling miscellaneous loads. The Displacement centerwell overhead is formed by the main deck with a flush hatch access and affords an Light Ship 2522 LT interior clear height of 17 feet. Transit Draft (16.7 ft) 3530 LT Storm Draft (23 ft) 3989 IT The third deck also contains a work platform Normal Operations Draft (27 ft) 4179 LT with an area in excess of 2,000 square feet Special operations Draft (34 ft) 4511 LT located across the stern of the ship and continuing around the starboard side. Direct Power access is provided from this deck to the main and biochemistry laboratories. The stern Main Diesel Generators (5) 6030 KVI platform is serviced by an A-Frame on centerline In-Port/Emergency Generator 300 KW and a 20-ton telescopic boom crane on the Propulsion Motors (4) 6400 HP starboard quarter. The work area on the Bow Thruster (1) 500 HP starboard side is serviced by one winch and a 12 ton J-frame in addition to 20-ton and 10-ton Speed deck cranes. The work area has access to the various laboratories. Scientific storage areas Clean Hull Trial at 16.7 ft Draft 15.6 kt are arranged adjacent to the laboratories and In-Service Transit at 16.7 ft Draft 14.6 kt work areas. The remaining part of the third Clean Hull Trial at 27 ft Draft 11.8 kt deck forward contains the main and auxiliary machinery for the ship. Range The second deck contains quarters, ship storage Transit Draft at 12 knots 10,100 nm spaces, and lounge and recreational facilities, as well as the galley and mess room. The aft Endurance open deck platforms at this level can carry winches or other mission equipment. Two vans Stores and Supplies (Minimum) 60-120 dy can be connected to the second deck accommo- dation space on the port side. Accommodations The main deck and second decks i6corporate two Marine Crew 20 hydraulically powered, 20-ton telescopic boom Technical Crew 30 cranes with a 6,000 pound lift capacity at a 1151 ,tA.4 1A .7-1 i as - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - r ------ ---------------- - -------- ------ - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Figure 2. Outboard profile. ACCESSINT WT 910 CIN11016SORS SCIENTST's X.AND;B;. MAIN LAB MO. 2 CONFEF04M AD I. I'.. AJBRARTAGGM e - A.. " -/I ------ AT &111 1 : IT SLIDING OR IT IT W N, 2 L SCtEWM C110ME ROOM FUM SCI OFFICE TAS AND alTO"E 7 - N"' ROOM OF r,,@ FI) up WT r1% IN, L JEUP WE MAIN LAB NO. I STAGING Is 7 AREA 0.0PNPO 7 1 If 10 CMx1FK IN STORAGE UP LKR-a-U ADZ WET LAB E"L DECK \Ij 'ZW LA LAB FIEUM REF a -i IT OF NO. 5 IT IT --- 13ARKRDW 'j IT SLIDING SC IENT[FIC SCNO. I No. 3 ------- 6TORAGE IE.TIFIC SCIENTIFIC STORAGE sTORAGE Odw BIOCHEM B .. .. 21B P LA L.. ANT E A r", "Ago WORK DECK ACCESSIVNT SERVICE PLATFORM Figure 3. Third deck general arrangement. 1152 radius of 50 feet. There is one hydraulically Intact stability is adequate for safety and to powered 10-ton articulated boom crane with a permit moderate load transfers on board. Loads 4,500 pound lift capacity at a radius of 28 up to about 20 long tons can be moved trans- feet. The wheelhouse is located above the 01 versely from the center of the moonpool to the level on the starboard side. A remote docking deck edge or from the deck edge to 16 feet station is provided on the port side. The outboard without the need of simultaneous helicopter hovering area is located on the port ballast compensation. Transfer of ballast or side. Ship masts are forward and aft on the main fuel oil can be used to counterbalance larger deck. They are configured to accommodate load shifts. The upper hull is designed to be antennas and instrumentation for research and watertight to the main deck, except for the survey applications. centerwell, while maintaining strength with light weight. The buoyant upper hull thus The five main diesel engine generator sets, power provides a large margin of safety in the event conversion system, main electrical switchboard, of partial flooding of compartments in the lower various auxiliary machinery, and transformers are hulls and columns. Furthermore, the lower hulls located on the third deck. The DC propulsion and columns are subdivided into watertight motors and gears are located in the lower hulls. compartments to control the volume that is One fixed pitch propeller per hull, fitted in a likely to be flooded in any single incident. fixed propulsion nozzle, is driven through a The design complies with USCG regulations for reduction gearbox. A 300 KW inport/emergency certification for single compartment flooding generator is located in the deckhouse at the main and damaged stability. Some special deck level. Ballast water tanks are in the lower interpretations have been approved by USCG to hulls and columns. Fuel storage tanks are adapt the conventional ship rules to the located in the lower hulls only. realities of the semisubmerged SWATH ship. 4. STABILITY 5. SHIP MOTION RESPONSE The stability of the semisubmerged SWATH is much The dominant advantage that the semisubmerged higher than that of a comparable monchull. At SWATH ship offers is drastic reduction in ship the light ship and transit operating draft motions. The improved seakeeping is provided in conditions, the lower hulls are at' the surface both the transit mode and, more importantly, and have some nominal freeboard. Initial intact during on-station research operations. This stability is very high because of the large motion response reduction allows the waterplane area. At a transverse inclination semisubmerged SWATH ship to operate in much angle of a few degrees, one hull goes awash, higher seas than can be tolerated with an reducing the waterplane area and the slope of the equivalent monobull. At transit draft, the righting arm curve. Stability in this condition freeboard of the lower hulls is only about one is, however, very high compared to a monohull. foot. Wave crests wash over the lower hulls and As the heeling moment is increased, the ship increase resistance to heave. The radical tends to rotate about the lower hull that decrease in waterplane area above the lower hull remains on the surface. At an angle of about 15 reduces buoyant forces caused by wave action and degrees, the upper hull begins to enter the water ameliorates motion response. It should be noted and stability of the ship increases dramatically. that this is a key difference between a semisubmerged SWATH ship and a conventional As ballast is added to the loaded transit draft catamaran vessel. condition, the lower hulls are submerged and the waterplane area is reduced to only the cross Model tests of several designs have confirmed section of the columns. Minimum initial that the semisubmerged SWATH ship, even at stability occurs when both hulls are just awash. transit draft, will have lower motion response However, at this point the application of a than a comparable monchull. Figure 4 heeling moment will cause one hull to broach the illustrates relative motion responses for a surface, rendering the stability characteristics conventional monohull and a semisubmerged SWATH similar to those described above for the shallow ship (at normal operations draft). In beam draft condition. seas, roll of the monohull can be more than five times higher than that of the semisubmerged At still deeper drafts, the ship floats with only SWATH. In pitch, the monobull response the columns piercing the water in both the typically is two to three times higher. This upright and inclined conditions. This is the feature alone prompts the consideration of desired operating draft range which gives minimum semisubmerged SWATH ships in place of monohulls motion response to the seaway. The transverse for research and survey applications. stability (GM) in the fully loaded condition at the normal operations draft is about five (5) The response of the ship to a seaway is highly feet. This allows for a 0.5 foot margin for dependent on wave period or, more precisely, on vertical center of gravity. The vessel has been encounter period which is a function of wave designed with greater stability in the lon- period, ship speed, and relative heading. At gitudinal direction than transversely to improve zero speed, encounter period equals wave period. motion characteristics while underway. The semisubmerged SWATH ship characteristics are such that ship heading at zero speed 1153 5 ample design margins. Propeller-nozzle efficiencies were calculated for the design i Vonohul I based on input hull related data developed from (ARCO RESOLUTION) previous self-propulsion model test results of similar designs. Model tests previously done indicate that these estimates will be well Semisubmerged within five percent of actual values. The range SWATH in calm water at a ship speed of 14.6 knots is 4 5,700 nautical miles with ten percent fuel reserve. The range at a speed of 12 knots is 10,100 nautical miles. The estimate of shaft horsepower versus speed ?-Rol l (beam seas) for the semisubmerged SWATH at transit draft is 70. shown in Figure 5. The curve is based on clean 3 hull and calm seas. Installed horsepower and machinery ratings reflect the SHP needed to propel the ship at a clean hull trial speed of 15.6 knots. Model test data also were acquired at transit draft in head and following seas to allow accurate calculation of performance in Sea State 4. The design gives in-service ship speed of more than 15 knots for the average of all 2 headings in SS-4. The worst condition for propulsion is in head seas (Figure 6). Cc 7. SUMMARY The semisubmerged SWATH hull form has been adapted to the requirements for an oceanographic Pitch Qhead seas) research and hydrographic survey ship. The design provides for unrestricted operations in ice-free areas worldwide. The ship can be brought to a full load draft that allows access Rol I (beam seas)-'-, to harbors that serve major oceanographic and survey organizations. Performance of the ship Pitch (head seasl will match or exceed comparable monohulls. 0 Seakeeping will allow operations in one or two 4 6 8 10 12 14 16 sea states higher than can be done with Wave Period (seconds) equivalent monchull ships. Figure 4. Typical ship motion response. The design of the Semisubmerged SWATH Research and Survey Ship incorporates some noteworthy features that either are not found in equivalent relative to wave direction causes little change monohulls or are significant improvements in in motion response. If the operation requires mission capabilities. The objective throughout that the ship lie in the trough of the waves, the design has been to adapt inherent attributes this can be done as comfortably as lying head to of the semisubmerged SWATH configuration to the the wind. , When underway in head seas, the particular requirements for research and survey apparent period of encounter is shorter than the applications. Notable special features are wave period, so ship motions are very small. In summarized below. following seas, the encounter period is longer (1) The ship is designed to operate over a than the wave period, so motion amplitudes tend variable (selectable) range of drafts (and to be greater than at zero speed. However, at corresponding clearance between the upper hull these long periods the accelerations are very and sea surface). The shallow draft of less small. Deck motions rarely will exceed the than 17 feet fully loaded allows access to most acceleration threshold for seasickness discomfort harbors with facilities appropriate to this (about 0.2 g). class of oceanographic research ship. Ship motions in the seaway are minimum at the normal 6. SHIP PERFORMANCE operations draft of 27 feet which typically would be used in exposed deep water locations For purposes of calculation, stated in-service for conducting technical activities. At the SHP has been converted to installed BHP by special operations draft of 34 feet, the level allowing for total mechanical' and electrical trim elevation of the stern and centerwell work systems losses of 15.7%. Total fuel consumption decks is less than ten feet above still water. has been calculated based on operationally A slight adjustment of trim (by shifting required BHP with appropriate fuel conversion ballast) can change or even close the air gap to bea s V_1,',,'@ @Pi@th=mse@arj@@ (h rates specified by the engine supplier and with the aft work deck. 1154 (2) The four-column design, coupled with 7 1 1 1 1 1 1 lower hull geometry and overall weight distribution, gives ship motion response at deep drafts and zero speed that essentially is 6 *4EAVailable 15.6 _ independent of heading relative to surface wave Sha ft Horsepower direction. (3) The design has been developed and model tested to achieve superior propulsion and motion 5 - performance at the transit draft in light to moderate sea states, including Sea State 4. The predicted speed in Sea State 4 in the head seas Shaft Horsepower condition (worst case) is 14.8 knots. In 2'4 Sea Trials - following seas, the speed is 15.9 knots. A major advantage of the variable draft SWATH configuration is the ability to transit at higher speeds in moderate sea states with less & 3 power than for monohulls. This is achievable because the incremental increase in resistance due to waves is significantly less than for a monohull. 2 (4) A key issue for evaluation of performance at transit draft is seakeeping of the EHP semisubmerged SWATH compared to an equivalent Calm Water monohull. At this shallow draft, the semisubmerged ship does not behave like a conventional catamaran because the lower hulls are ballasted to a freeboard of only about one foot. Figure 7 shows pitch response based on 0 comparative model tests of a semisubmerged SWATH 4 6 8 10 12 14 16 and a comparable monohull. Ship speeds Speed (knots) nominally were 10 knots in head seas of Sea State 4. The full scale displacement of the Figure 5. Transit speed in calm water. SWATH at transit draft is 2025 LT and of the monobull is 2285 LT. The SWATH pitch response is less than for the monohull across the entire range of wave periods from 5 to 16 seconds. 7 (5) The deck service cranes are arranged such that a load in excess of 5,000 pounds can ,-Available PrDPUlsion be transferred between the centerwell and over Motor Horsepower 14.8 either side or the stern. This gives 6 flexibility for loading supplies and equipment as well as for supporting oceanographic operations. 5 (6) A capability for load transfer up to ten tons is provided within the entire centerwell by a horizontal gantry crane mounted on the underside of the main deck. 4 (7) With its 77 foot transverse propeller separation, the semisubmerged SWATH has enormous turning capability for precise heading control Motor Power in in adverse conditions. The long lower hulls and SS-4 slender struts have high directional stability 93 for good tracking. The conventional rudder t is designed to give maximum heading arrangemen EHP in SS-4 control at slow speed. Course keeping in a 2 seaway is much better than for a monchull. (8) The starboard lower hull is designed as a "quiet" hull with sonar transducers located forward and away from ship service equipment. A 1 single bow thruster is provided and is located in the port side lower hull. (9) The centerwell containing the moonpool is enclosed with clear headroom of more than 17 0 1 1 1 1 -1 feet. A removable hatch in the main deck 8 10 12 14 16 allows entry into the centerwell of objects up Speed (knots) to 25 feet long. The hatch is positioned over the moonpool to allow free-fall access from the Figure 6. Transit speed in SS-4 head seas. 1155 1.4 1 1 -1.2 14onohull a - SWATH 1.0 1p, 0- 0.6 0- 0- _01 0.4 0.2 0.0 4 6 8 10 12 14 16 Period (seconds) Figure 7. Underway comparison of pitch response for monohull and semisubmerged SWATH. main deck to the water between the columns and lower hulls. (10) A large open main deck area is provided which could be upgraded to work deck, storage area or helicopter landing area. (11) Laboratory space of 4250 square feet is provided. All laboratories are located in the aft portion of the third deck and are contiguous to the centerwell, the aft work area, and the starboard side work area. 8. REFERENCES 1. Narita, H., T. Mabuchi, Y. Kunitake, H. Nakamura, and M. Matsushima, "Design and Full Scale Test Results of Semisubmerged Catamaran (SSC) Vessels," Paper presented at First Int'l. Marine Systems Design Conf., London, England, April 1982. 2. Oshima, M., H. Narita, Y. Kunitake and H. Nakamura, "Semi-submerged Catamaran (SSC) Application in Marine Transport and Ocean Development," Proceedings PACON 84, Marine Technology Society, Symposium at Honolulu, Hawaii, April 24-27, 1984. 3. Saeki, Muneharu, and Hiroshi Nakamura, "Motion Characteristics of the KAIYO," Proceedings of Oceans, Symposium at Washington, D.C., September 23-25, 1986. 4. covich, P. m., "SWATH T-AGOS, a Producible Design," Paper AIAA-86-2384, AIAA Sth Advanced Marine Systems Conference, San Diego, California, September 22-24, 1986. 5. Gaul, R. D., and A. C. McClure, "Development of the SWATH Ship Concept for Research Ship Design," Proceedings of Oceans, Symposium at Washington, D. C., September 10-12, 1984. 6. McClure, A. C., and R. D. Gaul, "Let SWATH 7 Mon 'U" ,,- SWATH be SWATH,- Proceedings, U. S. Naval Institute, April 1987. 1156 TANKAGE ARRANGEMENT FOR SWATH SHIPS Colen Kennell Naval Sea Systems Command ABSTRACT for SWATH designs than for the monohulls normally encountered with values typically less than 20 SWATH ships are.being developed for a variety of percent of the monohull MT1". Consequently, SWATH applications requiring superior seakeeping in rough shi s will be much more sensitive to the variations seas. Certain aspects of the design and operation of in @ongitudinal center of gravity (LCG) encountered SWATH ships combine with the inherent hydrostatic during design and while operating. characteristics of this type of hull to magnify the importance of tankage arrangement to the successful Designs for SWATH ships are produced by NAVSEA design and operation of these ships. This paper using the same design approach used for monohulls. describes the hull form basis of the problem in This process includes inaccuracies and uncertainties comparison to conventional monohulls. The major that are routinely absorbed by the margins used as parameters affecting tankaqe arrangement are well as the checks and balances inherent in the discussed including uncertainty during design, process and the monohull form. SWATH desig@s topside ice accumulation, and the role of mission produced with this process are expected to contain equipment and loads. An approach to the solution of the same inaccuracies and uncertainties and the the.problem is described that is compatible with the rocess is expected to provide the same checks and design process. Application of the proposed approach galances. However, the checks and balances inherent to the arrangement of tankage on a large in the SWATH form differ from those of monohulls in oceanographic ship is described. terms such as the M171", a measure of the trim sensitivity of the hull form. As a result, the design process must be modified to reflect this less 1. INTRODUCTION forgiving hull form. The U. S. Navy has actively developed SWATH ship For example, the design process corrects misalignment of a monohulls LCG and longitudinal technology since 1970. The Navy's first SWATH ship center of buoyancy (LCB) through changes in designed for open ocean operations, the 3,400 ton arrangement and hull form. These chanqes are usually T-AGOS 19 (Figure 1), is currently under small. Past experience with many similar designs construction. A 5,400 ton oceanographic research allows selection of a hull form and arrangement ship, the T-AGS(OCEAN), is under design in NAVSEA. likely to succeed. This generally assures a ood Studies completed as part of these efforts indicate starting point. Convergence of the design spiral is that certain aspects of ship design and operation also enhanced by the high moment to trim of the combine with the inherent hydrostatic characteristics monohull form which makes the design relatively of the SWATH form to magnify the importance of insensitive to small changes in longitudinal moment. tankage arrangement in the design of SWATH shM. This paper describes some aspects of the SWATH s ip tankage problem and a proposed solution that is compatible with the NAVSEA design process. 2. BACKGROUND SWATH ships are designed for superior seakeeping in high sea states. This improved hydrodynamic performance is derived from the characteristics of the hull form, particularly the small waterplane area. The small waterplane area of SWATH ships also results in hydrostatic characteristics that are significantly different than those normally encountered in ship design. SWATH ship hydrostatic characteristics are compared with monohull hydrostatics in reference 1. Figure 2 (reproduced from reference 1) indicates than the moment to change trim one inch (MT1") is much lower FIGURE 1. T-AGOS 19 1157 United States Government work not protected by copyright 4000 - Operation of SWATH ships is also expected to be affected more than comparable monohulls by the 0 MONOHULL change@s in LCG due to consumpt-ion of fuel, stores, A SWATH 0 and mission expendables. Mission expendables are particularly important since large weights may be 0 0 deployed from the ship in a short time. Fuel and 3000 - stores are expected to be less of a problem because of the slower rate of consumption. Z 0 0 The factors discussed above suggest that better LU control over longitudinal weight distribution is Z 0 0 0 0 2000 - required during the design and operation of SWATH 0 ships than commonly used in monohulls. Several methods are known for achieving the desired level of cc control including: 0 - use a trim control system 1000 - - lengthen the design period to obtain LU convergence 2 0 - provide excess tankage and balance the ship by 0 0 shifting fuel and ballast water 2 - use solid ballast 0 10000 20000 30000 Some of these methods may provide only part of the re7uired control over trim. For instance, solid DISPLACEMENT (LTON) ba last can be used to counteract trim encountered during design and construction but provides no relief FIGURE 2. MOMENT TO TRIM ONE INCH from trim encountered at sea due to expenditure of loads or ice accretion. The SWATH design.data base is sparse by comparison The excess tankage approach has been used in the and the design is much more sensitive to fluctuations T-AGOS 19 and T-AGS(OCEAN) designs because it is of the LCG. As a result, the risk associated with compatible with design periods of different length, converging to a balanced design solution is much does not require introduction of additional systems greater. into the design, and provides solutions to both the SWATH hull forms are likely to be selected for design problem and the operational problem. The those missions requiring sustained operational major design limitation of the approach is the amount performance in rough seas, typically sea state 5 and of excess volume required. An operational above. Such sea conditions are often encountered in consequence of the excess tankage approach is the the northern parts of the Atlantic and Pacific Oceans presence of fuel tanks in a ship that should be only that are strategically important. Ships operating in partiall filled or empty. Pressing up all available these areas must contend with environmental factors fuel tal prior to getting underway greatly reduces other than rough seas. The occurrence of winter trim control and may violate other naval storms is common in these areas. These storms often architectural constraints. The discussion that result in a significant accumulation of ice on the follows reflects this method for controlling trim. upper parts of ships. 3. TANKAGE ARRANGEMENT PHILOSOPHY This added topside weight can be a hazard to monohulls due to the associated reduction in The approach taken is to estimate the forward-most stabilit The effect on the stability of a SWATH and aft-most locations of the ship's LCG likely to be ship is j@ss dramatic due to the much greater reserve experienced. All components of the ship's full load buoyancy of the SWATH form. T-AGOS 19 calculations weight are included as well as the weight of topside showed that the design could withstand 100 knot beam ice, if applicable. The required range of fuel winds with a uniform layer of ice 12 inches thick on weight LCG is then derived to provide the desired the main deck. This result indicates that the design degree of trim control. This deqree of control is has an adequate intact stability margin of safety. nominally assumed to be that necessary to achieve However, realistic ice storms do not produce uniform zero trim. Finally, the range of LCG locations layers of ice on ships. Ice build-up tends to be required for the compensating sea water ballast is most severe on those parts of the ship to windward. assumed to equal or exceed that of the fuel. The bow of a ship is Particularly likely to receive heavy ice build-up since ships often.head into the The approach is strongly de endent on the estimates re 1 at i ve wi nd. Sh ip generated spray a I so adds to the of the expected variation oT lightship LCG and the rate of ice accumulation near the bow. weights and moments of topside ice. Relevant data exists for only the T-AGOS 19 design at this time. The weight and bow-down moment due to this Use of this limited data base implies a degree of asymmetric accumulation of ice on a SWATH ship can risk in the approach. However, the approach result in significant operational limitations if compensates for this risk by attempting to estimate uncompensated The reductions in cross-structure the worst case situations likely to be encountered. clearance due to the added weight of the ice and the For instance, the aft-most LCG limit results from the trim due to its asymmetry will increase the rate of lightship LCG being the maximum expected amount aft cross-structure slamming in waves. The resulting of the estimated location, forward loads being hull poundin is expected to limit operations and may expended, all of the ship's fuel onboard, and the damage the Nip. aft-most topside icing condition expected. This 1158 "worst-of-the-worst" approach was adopted to address standard practice at NAVSEA, schedule constraints this problem throughout the design spiral. As a precluded following this approach on the T-AGOS 19 design proceeds, increased confidence in the program. lightship LCG location or toTsi@e icing studies for the configuration being esigned should allow reduction of these allowances. 5. EFFECT OF VARIABLE LOAD The variable load (excluding fuel and sea water 4. LIGHTSHIP LCG VARIATION DURING DESIGN ballast) is divided into two groups to reflect the variable payload nature of some ship missions. The The lightship LCG of a SWATH design should be firsttgroup contains those weights that remain on the accurately estimated as soon as a stable ship size ship roughout a given voyage but may vary from one and arrangement are available. The lightshi@ LCG voyage to another. Laboratory equipment and will change as the design progresses throug the accommodation vans are examples of such weights. The design process due to inaccuracies within the process second group contains those weights that may be and design changes. The variation of the lightship expended during a single voyage such as disposable LCG during the T-AGOS 19 design is shown in Figure 3. anchors and seismic explosive charges. The range of The left-most data point is the Feasibility Study LCG values expected for these two groups of loads estimate and the right-most point is the Contract should be estimated for all loadout conditions Design point. The lightship LCG estimates generally expected throughout the life of the ship. fell within a band about two feet wide. This band represents 1.05 percent of the length between perpendiculars (LBP) for this design. The lone data 6. EFFECT OF TOPSIDE ICING point that is outside this band is the first Contract Design estimate. A major piece of the mission Little data exists on the effects of ice storms or equipment was located near the bow for this iteration frozen spray on SWATH ships. The effects of topside and subsequentl moved to the stern. This ice accumulation were studied during the T-AGOS 19 configuration re?ated LCG shift combined with start- design (reference 2). Expected ice accumulation was up problems associated with the complex weight estimated for operating areas of interest in the accounting procedures used account for this anomalous North Atlantic, North Pacific, and the South Bering data point. Sea where heavy icing conditions are known to occur. Head and beam ship eadings relative to the storm 8- were considered. U_ Z 6- Ice accretion data was produced by a method that incorporates analytic and empirical techni ues. This gproach was selected because of the nee@ to model t e different, complex shape of a SWATH ship which is 2 4- 2 outside the conventional ship data base. The 2 T-AGOS 19 geometry was modeled including topside 0 2- details such as rails, masts, and deck equipment. M LL FEASIBILITY 'CONTR.ACT Air and water temperatures, wind conditions, and the 0 properties of wind and ship generated spray were defined. The icing model includes the relevant dynamic and thermodynamic aspects of the ice 2 formation and accretion processes. Size and liquid 0 10 20 30 40 content of the expected spray droplets, the dynamics DESIGN TIME IN WEEKS of the flow of droplets around the ship, and the thermodynamics of the icing process were modeled. These factors were then used to estimate the volume FIGURE 3. T - AGOS 1'9 and density of ice for each of the ship components LCG VARIATION DURING DESIGN modeled. Finally, the weight and centers of gravity of the accumulated ice were calculated. The results Figure 3 represents data for one ship that is not are tabulated in Table 1. f et built. Data for other Navy SWATH ships desiqned TABLE I or open ocean operations does not exist. T-Ar- 1Q Consequently, the method described below will assume that lightship LCG may vary by as much as 1.05 ICE LCG AFT AVG ICE AVG ICE SHIP ICE LOAD DECK EDGE LOAD/BEAM LOAD/MN DK LEN percent of the LBP of the design in question. This LOCATION HEADING LB/1000 rT LB/FT LB/FT variation may be on either side of the estimated LCG. W N ATL BEAM 414 90.8 - 2,179 Some reduction in the expected amount of lightship HEAD 426 88.0 5,325 - LCG variation should be possible during the later C N ATL BEAM 324 90.1 - 1,705 design stages after the design process has HEAD 371 92.9 4.638 - stabilized. Figure 3 shows that progressively E N ATL BEAM 235 91.3 - 1,237 sTaller adjustments were made to the lightship. LCG HEAD 252 89.3 3,150 - with each successive estimate. The additional design iterations eliminate errors and produce design W N PAC BEAM 925 91.7 - 4,868 information such as system drawings that improve the HEAD 881 qO.7 11,013 - accuracy of the estimated LCG. Use of a Preliminary W/C N PAC BEAM 85 @1.9 - 447 Design between Feasibility Studies and Contract HEAD 82 65.0 1,025 - Design should also reduce the expected lightship LCG S BER SEA BEAM 29 95*'8 - 153 variation. While use of a Preliminary Design is HEAD 31 61.0 388 - LFEASIBILITY 1159 This data must be scaled to reflect configurations 7. SAMPLE PROBLEM - T-AGS(OCEAN) that differ from the T-AGOS 19. A simplistic scaling method has been used in the absence of more rigorous Application of the approach described above is @nalysis. A fundamental assumption in this approach illustrated by an example. The hull form used is the is that the geometry of the ship is similar to the T-AGS(OCEAN) as it existed midway through Preliminar T-AGOS 19 including such key variables as cross- Design.. Weights were taken from the th2 structure clearance, arrangement of topside Preliminary Design weight report. Characteristics of eguipment, and deckhouse configuration. The flow of the design are: air and spray around the ship is expected to be similar for such similar ships. Ice accretion Length Overall 279 ft properties are also expected to be similar since ice Between Perpendiculars 230 formation is strongly influence by the dynamics of Main Deck 220 spray when it first encounters a ship. Beam at Main Deck 84 Weight - Full Load 5,361 tons - Lightship 3,766 The scaling method assumes that the average weiqht - Fuel 1,256 - Other Loads 339 of ice per foot of main deck beam is constant when Longitudinal Center of Buoyancy Ill ft the ship is pointed into the wind for the different icing cases. Similarly, the same average weight of (at full load,displacement) aft of FP ice per foot of length is assumed when the wind is Loads were broken down into the following groups for off the beam. The distance from the forward deck the purpose of assessing the limits of LCG variation: edge to the LCG of the ice is assumed to be the same as on the T-AGOS 19. Table 1 shows the average Weight LCG weight per foot of ice accumulated for the different (tons) (ft aft FP) locations and headings from the T-AGOS 19 study. The ME -Mission Expendables 100 194.74 table shows that the average ice load ma be as much MNE -Mission Non-Expendables 150 159.87 as 11,000 pounds per foot of main deN beam when FW -Fresh Water 27 115.00 running into the storm and as much as 4,900 pounds FDT -Fuel in Day Tank 34 85.00 er foot of main deck length with the storm on the MLl -Miscellaneous Loads 1 35 93.33 Eeam. The weight and LCG of ice expected for a ML2 -Miscellaneous Loads 2 20 56.85 similar SWATH configuration can then be estimated using the main deck beam and length for the design. The ML1 group includes those load items that would These results can then be combined with the ship's remain with the shi throughout a voyage (crew & LCG to estimate the largest bow and stern trim effects lube oil, Phydraulic fluid, and residual moments due to topside icing expected for the design. ballast ' water). The ML2 group might be expended during a Voyage (ship ammunition, personnel stores, general store's). This simplistic ice scaling method is not expected to be valid for configurations that are si nificantly This data can be used to estimate the forward-most different from T-AGOS 19. Major variagles to be and aft-most locations of the LCG of the ship without considered are main deck height, deckhouse size and fuel or ballast water. The calculations for the aft location, and type and location of topside equipment case are summarized in Table 2 and the forward case since these parameters are known to influence ice in Table 3. The results for both cases are accumulation. illustrated in Figure 4. TABLE 2 TABLE 3 STERN TRIMMING MOMENTS ROW TRIMMING NEW NEW ITEM WEIGHT LCB-LCG MOMENT LCB-LCG ITEM WEIGHT LCB-LCG I MOMENf LCB-LCG LS+LSM 3765.93 - 3.51 - 13218 LS+LSM 3765.93 0.90 3389 MNE + 150.00 - 48.85 - 7328 MNE + 150.00 -48.85 - 7328 ---------- ---------- ---------- ---------- SUB 3915.93 - 20546 - 5.25 SUB 3915.93 - 3939 - 1.01 MLl + 35.39 17.69 626 MLI + 35.39 -17.69 - 626 ---------- ---------- ---------- ---------- SUB 3951.32 - 19920 - 5.04 SUB 3951.32 - 4565 - 1.16 ME +100.00 - 83.72 - 8372 ML2 + 20.2B 54.17 + 1098 ---------- ---------- ---------- ---------- SUB 4051.32 - 28292 - 6.98 SUB 3971.60 - 3467 - 0.87 FW + 26.88 - 3.98 - 107 FDT + 34.00 26.02 + 885 ---------- ---------- ---------- ---------- TOT 4078.20 28399 - 6.96 SUB 4005.60 30.32 - 2581 - 0.64 TSIl + 412.97 + 12521 ---------- ---------- TOT 4418.57 9939 2.25 ME - Mission expend-lbles FDT - Fuel in Day Tank TSI-1 - Topside Ice 1 MNE - Mission Non-Expendibles ML1 - Miscellaneous Loads 1 LS - Light Ship FW - Fresh Water ML2 - Miscellaneous Loads 2 LSM - Light Ship Margin 1160 down trimming moment results from heading into a storm in the western North Pacific. These conditions Z result in an accumulation of 413 tons of ice with an 0 17 associated bow down trimming moment of 12,500 foot- 20 - tons. The effect of this weight and moment can be seen in Figure 4 (marked F2). The tankage 10 - F2 arrangement must provide adequate moments to MNE compensate for the forward and aft moments at ints Al and F2 in Figure 4 to maintain an even ke 0- Xwith topside ice. DESIGN F1 .10 - Z The limiting moments reVred are shown in Tables W 2 - 2 and 3. These moments an the fuel LCGs required to 0 -20 NINE assure zero trim using the 1229 tons of fuel in the ME Al design are: -30 - MOMENT FUEL LCG -40 3500 4000 4500 (FOOT-TONS) (FT from FPI LCG for ice) 9940 119.1 WEIGHT IN TONS LCG for no ice 2581 108.9 FIGURE 4. T-AGOS (OCEAN) LCG aft @no ice@ --28399 87.9 LONGITUDINAL MOMENT SUMMARY The tankage arrangement must allow the fuel LCG to be placed anywhere from 88 ft to 119 ft aft of the FP to assure zero trim with the specified amount of fuel on The departure poi .nt is the estimated lightshi'p board. weight and LCG. T-AGOS 19 experience indicates that this LCG estimate ma@ vary as much as 1.05 percent of ThisoCapproach was used as guidance during the the LBP(+/- 2.4 ft or the example). Consequently, T-AGS( EAN) design. The locations of fuel and the. li@htship LCG is expected to be between 109.7- ballast tanks selected at the end of the Preliminary 114 5 t aft of the FP. The weights and moments of Des i gn are shown on the i nboard prof i 1 e (F i gure 5) of the loads expected to be in place throughout a voyage the T-AGS(OCEAN). Most of the forward half of the (MNE & MLI) are then added. Next, weights and hulls and lower parts of the struts have been used moments of the loads that may be expended during a for fuel and ballast tankage. Other smaller tanks voyage are added (ME & FW for the aft case, ML2 & FDT have been located in the aft half of the hulls and for the forward case). Since the weights and moments struts. Additional tankage in the aft half of the excluded from the two tables (and the two branches of sh%was not practical because of main propulsion Figure 4) are on the opposite side of the LCB in each mac nery and steering gear volume requirements case, the two points (marked Al and F1 in Figure 4) combined with an operator preference for horizonal represent the maximum bow and stern trimming moments access between the machinery spaces. of the ship without fuel, ballast water, or ice. These moments must be offset by the moments of the This configuration is a compromise that reflects fuel and ballast water to maintain an even keel in machinery requirements and operator preferences as the absence of ice. well as trim control requirements. The consequences of this compromise are illustrated by the The centroid of the topside ice is well forward of fuel/ballast polygon for the design shown in Figure the LCB for all of the locations and headin s covered 6. This plot defines the range of trim moments that in the T-AGOS 19 stud . As a result, the a9dition of can be generated by the fuel and ballast systems on ice can only reduce Ne stern trimming moment. Only the ship Any trim moment inside the polygon can be bow down trim moments are expected. The maximum bow generate@ through a suitable distribution of the fuel H H ............. .......... ............. ....... ........ ....... ............ ........ ............. ....... .......... ........ .......... ....... ............. ........ ................ ........... ......... ............. ................... .......................... ....... .. . .......... ............. ....... ............. ............. . . . ......... ....... .... ............. ............. ........... 3a FUEL 0 BALLAST El MACHINERY MN F1 NINE ME FIGURE 5. T-AGS (OCEAN) TANKAGE ARRANGEMENT 1161 and seawater ballast onboard. Although trim moment most tanks will provide 8,400 foot-tons of stern is plotted against fuel weight, the sum of fuel and trimming moment with a resultant four inch increase ballast weight is held constant for all cases. For in draft. example, the 80,000 foot-tons of bow trimming moment is generated by 400 tons of fuel and about 800 tons of ballast water in the forward tanks. 8. CONCLUDING REMARKS The LCB of the ship at the design draft is shown on The approach to SWATH ship tankage arrangement the plot for reference. The.3uidelines produced by outlined above is intended to assist the designer in this method (tabulated abov are also indicated. producing a balanced design. It is a first-step The plot shows that the tankage arrangement chosen by approach that has not been thorouqhly validated by the end of Preliminary Design can generate larger bow design application. However, it is based on sound trimming moments than recommended by this method at principles and parallels the approach used in the the beginnin? of Preliminary Design. In general, the T-AGOS 19 design. The T-AGOS tankage arrangement design can a so generate adequate moments to trim the seems to be adequate although this design is not yet ship by the stern. Slightly less than the build. recommended trimming capacity is provided when the full fuel load is onboard and when the fuel tanks are The dominant factors in the approach are: almost empty. These conditions are not considered significant due to the maturity of the design and the - variation of lightship LCG during design nature of the intended operations. - large variable mission loads - location of LCB relative to lightship The trim polygon represents the design at the end LCG and ice LCG of Preliminary Design while the guidelines were - topside icing formulated to compensate for inaccuracies that are encountered throughout the design process. In 'Accurate estimates of these factors should be particular, the variation in the lightship LCG of the vigorously sought as part of the design. design following Preliminary Design is expected to be much less than anticipated early in a design. Consequently, the 8,300 foot-ton allowance for variation in the lightship LCG could be reduced. Furthermore, evolution of the design has shown that ACKNOWLEDGMENT the lightship LCG has moved aft slight] durl fo in' The contributions of Christos Kasselas of NAVSEA in Preliminary Design. As a result, the neeV r the tankage system to provide the recommended capacity performing the above analysis is greatly appreciated. for trimming the ship by the stern is reduced. REFERENCES 1400 - GUIDELINES 1. Kennell, C.,"SWATH Ship Design Trends", presented at the RINA International Conference on U) 1200- z SWATH Ships and Advanced Multi-Hulled Vessels, 0 London, 17-19 April 1985. 16- 1000 - Z 800 - 2. Zahn, P. B., et al "Considerations for Northern Latitude Operations of the SWATH T-AGOS", ARCTEC TRIM By TRIM BY Report 1259C, March 1986. 600 LU STERN BOW @: 400 - _j Uj :D 200 - CB ILL 60 40 20 0 -20 -40 -60 -80 -100 FUEL/BALLAST MOMENT (FT-TONS/1000) FIGURE 6. T-AGS (OCEAN) FUEL/BALLAST POLYGON Operational factors also reduce the probability of needing the stern trimming moments recommended by the guidelines. Heavy icing conditions and the expenditure of mission equipment are unlikely when the full fuel load is being carried. Figure 6 shows that the maximum recommended stern trimming moments can be provided by the tankage system after 100 tons of fuel have been burned. Should mission equipment be expended and heavy ice be encountered upon leaving port, unwanted trim can be eliminated by taking on additional ballast water at the expense of a slight increase in draft. Forty tons of ballast in the aft 1162 SWATH SHIP DESIGNS FOR OCEANOGRAPHIC RESEARCH Thomas G. Lang Charles B. Bishop W.J. Sturgeon Semi-Submerged Ship Corp. Scripps Institution Semi-Submerged Ship Corp. 417 Loma Larga Dr. of Oceanography 417 Loma Larga Dr. Solana Beach, CA 92075 San Diego, CA 92037 Solana Beach, CA 92075 ABSTRACT studies using collections of current meters to long range topography observations using networks This paper describes several different SWATH of acoustic sources and receivers. Chemical vessel designs with relation to their value for oceanographers collect samples to study the oceanographic research operations. All of these diffusion of chemical constituents throughout the vessels follow the same SWATH design rules, but ocean volume, and in concentrated areas such as are of different size and provide different newly located hot vents on the ocean floor. capabilities to cover a wide range of research Biologists collect samples from the surface down requirements. The vessel lengths vary from 64 ft to the seafloor sediments using devices ranging to 247 ft. A common feature is low motion, both from nets and traps to manned undersea vehicles. at rest and underway, which permits research work Ocean engineers participate in all of these to be conducted in 1 to 2 sea states beyond that endeavors in order to develop more capable of a similar-sized monobull. instruments and vehicles to support the needs of scientific research. The ability of SWATH vessels to maintain transit speed and provide a steady working platform in While not all research ships can support all of rough sea conditions gives them high potential for the different scientific tasks, their value is increasing the efficiency, reliability, and safety enhanced by their ability to support several of seagoing research operations. tasks, particularly during the same cruise. There are other factors which affect the value of a 1. INTRODUCTION research ship. Can it handle the winches and wires needed for deep ocean investigations? Does Herein we describe several different SWATH vessels it have the endurance and comfort for long in relation to their value for oceanographic voyages, both in fuel and consumables and in research operations. Oceanography consists of the personnel efficiency? Does it need payload search for knowledge of the physical and chemical capacity for large and heavy deckloads or for properties of the oceans and their boundaries, the computer equipment and tape storage? Is it naLure and extent of their living organisms, the capable of supporting long time-series nature of the seafloor and its substructure, and measurements or just short term data collection? the interactions of all of these aspects of the Can it handle several overside instruments? How world's oceans with each other and with human about safe small boat operations? Is it a good intervention into the environment. The value of a platform for communications with other ships, research ship relates to how much of this search shore and satellites? it can support effectively and economicaYy. The UNOLS Fleet Replacement Committee report of June rhe basic requirements for research ships are to 1986 states that the overriding requirement for be able to go to the desired locations, andto new vessels is improved seakeeping to allow both- conduct- the desired research operations, in a safe overside and laboratory work to proceed in higher and economical manner. Existing ships come in sea states than is now possible. different sizes to meet the needs of different research operations, but they all suffer from The instruments used by oceanographers are many restrictions on transit speed and on research and diverse, reflecting not only the different effectiveness caused by rough sea conditions. For scientific disciplines but also the different example, 12 knots is commonly given as the objectives of particular research cruises. cruising speed of larger research ships, although Geology and geophysics require contact with the not often attained. A speed of 9-10 knots in many seafloor, either with mechanical devices or by transits is normal, which is about a 20% remote sensing systems. Piston- and box-corers reduction. Since transit is often 30-50% of the are lowered and recovered on site, while total voyage, this means a decrease of 6-10% in seismographs are deployed and later recovered. time available for research work. Heavier seas Acoustic mapping of seafloor topography, on the will cause a greater reduction, and on station other hand, is done both in concentrated areas and they can cause delays in preparing, deploying and while conducting long transits. Physical ocean- recovering equipment; they have also been the ographers study the dynamics of the ocean cause of loss or damage to equipment, and injury circulation in many scales, ranging from local or death of personnel . Even in "normal" CH2585-8/88/0000-1163 $1 @1988 IEEE conditions at sea, ship motions have a degrading Still another feature of SWATH vessels is their effect on human performance, ranging from total low wavemaking drag relative to monobulls due to wipe-out by sea sickness to the less noticeable the deep submergence of their lower hulls. This but more prevalent increase in fatigue. Since wavemaking drag can be even further reduced by scientific organizations pay for each day at sea, shaping the hulls and struts so that their whether they are getting data or not, the ratio of wavemaking is partially canceled. In the case of working time to total time is a measure of the four-strut SWATHs, their lower hulls are typically value of the ship to their research. Reduced ship bulged out in the region between struts to reduce motions at sea provide a multiple payoff. drag. In moderate to high sea states, SWATH ships exhibit much lower drag over a wide range of 2. FOUR STRUT SWATH SHIPS speeds than monohulls because rolgh seas have very little effect on SWATH ship drag The oceanographic research vessel designs dis- cussed herein are all four-strut SWATH (Small The characteristics of significantly different Waterplane Area Twin Hull) vessels, shown sche- SWATH designs will be significantly different. matically in Figure 1. Each design has twin This is true of any design field: large differences in design form will produce large differences in design characteristics. One of the greatest differences between SWATH designs is whether they have four or two struts. A primary advantage of four-strut SWATHs is that they typically have less motion at rgst or when moving slowly than two-strut SWATHs . This is mostly due to the gap between struts on each side that permits waves to pass through, thereby reducing the side pressures and side loads that cause roll and sway in beam seas. It is also due to their waterplane areas being concentrated near the four corners. A third contributor to reduced motion is the longer natural periods in roll and heave which are characteristic of the four-strut SWATHs relative to two-strut SWATHs. A fourth Figure 1. Schematic illustration of a four-strut contributor is increased hull damping in the SWATH. vertical direction due to the absence of struts in the midsection, especially when this midsection is submerged hulls that support a cross structure bulged out to reduce drag. Reduced motion at rest above water by means of four streamlined struts. or at low speeds is especially important for Canard fins are placed near the forward ends of vessels whose primary missions are carried out at the submerged hulls, and larger stabilizing fins rest or at low speeds. are placed near the aft ends. The fins dyna- mically stabilize and trim the vessel at moderate Another advantage of the four-strut SWATHs is that to high speeds. Optionally, the fins can be auto- they do not require the automatic f in control at matically controllable when underway to further moderate to high.speeds in order to re..duce-motioq, reduce the already small motions in waves. broaching, or diving in large waves, as do some two-strut SWATHs. (The four-strut SSP KAIMALINO Several s?u ,5c operated for its first year in the rough Hawaiian ,%a provide basic information on SWATH vessels . Both two-strut and four-strut waters without automatic control, and still SWATH vessels have greatly reduced motion in waves operates between islands much of the time without relative to monohulls. Motion is reduced because: automatic control. In the higher sea states (1) the small waterplane area minimizes the operators of the SSP prefer to turn off the buoyancy changes when waves pass, (2) the automatic system and to control the hydraulically- submerged hulls and fins damp motion, (3) SWATH driven fins manually because they have more vessels have long natural periods of resonance in precise control over motion). heave, pitch, and roll due to their small water plane area. and (4 ') at higher speeds, their fins Other advantages of four-strut SWATHs are less can be controlled to reduce motion. The motion of strut weight and cost due to their reduced strut a SWATH vessel in waves is typically around one surface area, better lower hull form for enclosing fourth of a similar-sized monohull, or conversely, engines because of the drag-reducing mid-section � SWATH ship will have the motion characterist gf bulges, and a better structural arrangement for � monohull about four times its length @S' . inclusion of center wells. On the other hand, Another feature of SWATH vessels is their ability potential advantages of two-strut SWATHs are to reduce their freeboard height through ballast- slightly less beam, and possibly a simpler strut- ing with sea water. Also, well-designed SWATH to-cross-structure structural arrangement. ships will not exceed the heel of equal-sized monobulls when handlingloads over the side. 1164 3. 64-FOOT SWATH This smallest SWATH, as designed * by the Semi- (RV) Submerged Ship Corporation (SSSCO) , would serve TkBLE I. SWATH RESEARCH VESSEL the needs of researchers interested in coastal .41 .4 observations which require less than 7 days at RV1 RV2 RV3 RV4 sea. Figure 2 shows one of many possible cabin arrangements and sizes, since the main deck is BASIC CRARACIERMCS designed to carry all structural loads. A SWATH vessel's ability to operate with more speed and LEWIH (FT) 64 88 150 247 BEM (FT) 35 46 75 95 DRAFT (FT) 5-7 13-15 10-16 19-24 FREEECARD (LO/HI-Fr) 2-7 8-13 4-14 4-20 DISPLACE1,1ENT (IT) -99 228 688. 2489 BULL MATERIAL A[M ALLWST ALLWST SM REL CAPACITY (GAL) 1260 6000 34100 151000 RU am (KIS) 16 19 14 18 CRUISE = (IGS) 14 13-15 13 15 4 CRUISING RANGE (N M) 638 300 7000 11400 EMMANCE (DAYS) 7 14 3D 40 7, PRPLSN FOWER (HP) 2X300 2X2200 2X625 4XL500 ENGINE LOCATION L.HLUS U.HU1L L.HM U.HLU GENERATmS W 3D 25D 466 4XIO90 All GERRAIM (KW) 30 30 2XL50 250 THRUSM (HP) 0 0 =0 2YA70 10 12 25 CFFICERS AND CRIEW 2 -24 -40 Sam= 10 6 2D 35 nTiERANT LOAD (IT) 11 16 50 10D Figure 2. Artist's rendering of SSSCO's 64-f t multi-purpose SWATH design, shown outfitted for M AREA (SQ Fr) 123D 3120 6000 13782 one of many possible applications. Cam WHIL (FT*FT) 0 12X23 15X30 15X30 steadiness than an equivalent monohull should HID DUE @ 2 FT - YES YES reduce the at-sea time needed to obtain data, LOADING CRANE 1 2 2 collect samples or recover instruments. Its ample TVM CRANE 0 0 1 1 deck space for its size should make it possible to OVERSIDE CRANE 1 1 2 2 OVERSIDE A FRAME 0 1 2 2 carry more instruments, and to conduct training S= RAMP 0 0 1 1 sessions for students in instrument preparation, STERN A FRAME I 1 1 1 deployment and recovery. Such a craf t would be H= WINCH 2 3 2 2 particularly useful for the physical, chemical and HEAVY WINCH 0 0 1 1 biological measurements associated with coastal VANS-@ FT 1 2 3 4 environmental monitoring. Inflatable workboat NCK BOATS 1 1 2 3 operations and scientific diving explorations GANIRY CRANE 0 1 1 1 would benefit from its stability and maneu- verability, since its motions when lying-to are LAB AREA (SQ FT) 337 300 25DO 4797 strongly damped, and its width between propellers. provides an effective turning moment. Hydro- UNISIRUTS YES NO YES YES graphic surveys of the bottom could be easily CLEAN F%a (1W) 0 3D 75 75 conducted using standard sonar equipment. This FUME HXDS NO NO YES YES small SWATH would even be capable of conducting HVAC OFT CFT YES YES seismic surveys, using air guns and towed acoustic UF AIR (CLEW CPT YES YES YES streamers, with accurate course control in rough SEA WATER (CLEAN) YES YES YES YES seas. It would be equipped with a stern-mounted REFR SIULALE (CU FT) 30 60 100 10D A-frame, siderail davit posts, maindeck tiedown SCIENCE Sim (CU FT) VAN VAN 6000 15000 fittings, maindeck electrical power boxes, and a DATA SIULAGE (CU FT) VAN VAN 1500 3000 suitable crane. Its physical and performance characteristics are shown in Table I, together with those of the other SWATH designs. Three@i vessels in this size range have already been built in the United States under a patent license from the lead author. These are SUAVE LINO (renamed wEM OF OCEANOGRAFBIC B:@L@ RR RVI AND RV2 IS BETSY, the tender f or the 12-meter America's Cup OMIDERED PART OF ITINERANT PAYLOAD. RV2 IS ME winner STARS AND STRIPES), HALCYON, and CHUBASCO. SSP K04ALINO, AS aMMMY CONFIGURED. UE LEFT AND RIGHT VALUES REFER ID RILL PAYLOAD, BUT WITH HALF AM FM FUEL LOADS, RERIEMELY. + ME LEFT AND RIGHT VAUES REFER 70 BALLASIED-EM AND SSSCO, Design team: T.G. Lang, P.V.H. Serrell, NDN-BMIAM RILL LOAD camm, RESPECTIVELY. W.J. Sturgeon, and U.W. Hird. 4 ME 3AALLM VALLE REFERS TO DAYrM OPERATIONS. 1165 4. 88-FOOT SWATH To learn first-hand about performing scientific research aboard a SWATH, the National Science This vessel is the SSP KAIMALINO 9110,11, a 227- Foundation and the Office of Naval Research ton SWATH built in 1973 by the U.S.C.G Shipyard, sponsored four short test cruises aboard the KAI- Curtis Bay, Maryland for the Naval Ocean Systems MALINO, and nine scienWic projects were Center (Figure 3). The KAIMALINO has operated in conducted in February 1985 . The following are the rough seas of f Hawaii since 1975 as a range excerpts from Ref. 12. "During one storm, winds support craft, and is capable of coastal gusted between 45 and 60 kts and seas were 6-8 ft. operations for periods of up to two weeks Even in these fairly heavy seas, ship motions were duration. This relatively old SWATH design has never great enough to shift gear on the main deck or equipm ent on the countertops in a van on the main deck." "The stability and steadiness that were experienced was called simply extraordinary." 11i)u - ring the storm, participants who usually get seasick (in less severe weather, they said) did not." "KAIMALINO's most critical trials, which it passed with consistently high marks, were instru- ment deployments and recoveries." "Operating from ,101, the center well was especially attractive. There was no need to steady instruments when they were % ni jM suspended near the surface. Participants repeat "The edly expressed their surprise at this." usual rolling of conventional ships can make it at least difficult, and sometimes impossible, to prevent instrumentation deployed over the side of a vessel from swinging against the hull. This was not a concern on the KAIMALINO b'ecause of the ship's extremely slight roll." "Without excep- Figure 3. Navy photograph of the 227-ton SSP tion, participants of the KAIMALINO evaluation KAIMALINO operating in the waters off Hawaii. cruises were enthusiastic about the SWATH. All felt that the SWATH design is well worth pursuing performed exceptionally well, and has required for an oceanographic vessel." only minor hydrodynamic and structural changes. However, changes made by a new design would great- ly improve the SSP characteristics in Table 1. 5. 150-FOOT SWATH In addition to the capabilities of the 64-footer, The SSSCO 150-foot SWATH design (Figure 4) would the KAIMALINO has demonstrated the ability to serve as an intermediate size general research support large ROV operations through its center ship and permit open ocean operations of up to @O well, and to deploy and recover sizable hardware days duration. Its range is 6000 miles, and it can over the sides and stern. This size of craft handle the standard suites of oceanographic should be capable of handling ROVs and AUVs, and instruments with overside cranes, davits, and A- scientific diving groups. Its open-ocean speed frames, and with a hydraulic A-frame at the stern. should make it valuable for rapid deployment and The stern has a ramp or vertical lift deck section recovery of instrument packages from the surface, to expedite launch and recovery of scientific floating in mid-water, and from the seafloor. M M M 20L 29 *4 IF =@ E--a:::7 UFE? FEE E-5 M M M',@, Figure 4. Profile drawing of SSSCO's 150-ft SWATH M IJAM im@ @m %M oceanographic research ship design. 1166 equipment. The 15 x 30 ft center well provides This design has a diesel-electric propulsion for launch and recovery of ROVs, AUVs and DSVs of system with four Caterpillar 3516 diesel ALVIN size. This SWATH is capable of towing generators located in the cross structure. sensor packages such as DEEP TOW, GLORIA, SEAMARK, Electric motors drive two Kort nozzles mounted at and seismic streamers, and can deploy seafloor the ends of the lower hulls and two thrustors work vehicles such as RUM. Its 2500 sq ft of mounted near their bows. Thus, underwater noise laboratory area and 6000 sq ft of usable deck area should be very low, both underway, when station can support several scientific groups. The lab keeping, and when traveling at low speed. space consists of a subdividable main lab, hydro lab, wet lab, electronics/computer lab, climate Normal research work should be possible through controlled reefer, and freezer. In addition, 1200 S.S. 6, and limited work through S.S. 7. The sq ft of science storage space is available. Two estimated natural periods in pitch, roll, and standardized 8' x 8' x 20' portable deck vans can heave are 11.6 sec, 21.0 sec, and 11.0 sec. be carried to provide additional laboratory, Predicted motions in S.S. 6, having a significant berthing, storage or specialized use space. This wave height of 18 ft, are +/- 2.8 deg pitch, 3.4 vessel can maintain station and work in sea states deg roll, +/- 4.1 ft heave, and +/- 0.12 g of .through 5 having significant wave heights to 12 vertical acceleration. Human factors data ft. Its pitch and roll will not exceed 5 degrees, indicate that roll angles should be limited to and vertical accelerations will not exceed 0.10g. 7 deg to maintain 80% effectiveness, and that the tolerance threshold for vertical acceleration 6. 247-FOOT SWATH varies from +/- 0.08g for 4-sec periods to about +/- O.18g for periods above 10 sec or below I sec. This large SWATH 13 (Figure 5), was des .igned by Since SWATH ships typically exhibit about 1/4 the SSSCO under contract from the Woods Hole motions of similar-lengtb monohulls, this SWATH's Oceanographic Institution. It matches the seakindliness greatly exceeds those of a monohull. university ships MELVILLE and KNORR in length, 7. SUMMARY exceed-- them in laboratory space and usable deck space, speed, low motions in rough seas, and in The cost of research operations at sea emphasizes cruising range, but falls short in payload weight the need for efficiency, reliability and safety. and endurance. This SWATH's range is 11,400 run at Risks to personnel and instruments on research 15 kts, or 16,880 nm at 12 kts, each with a 15% ships are greatest when equipment is being handled on deck in rough sea conditions. Research groups are made up of some experienced people and sometimes novices, including new graduate students % and volunteers. Even the older hands have few opportunities to keep their seagoing skills sharp. The potential for injury or accident is always resent. A SWATH's inherent advantage in stability 8 p and low motions, both underway and when stopped, certainly will contribute to greater saf ety. In addition, the large deck area reduces the need for crowding of equipment and the difficulty in fairleading wires and cables, all of which increase the hazards of operations on deck. Reliability of research operations is degraded a..... . . . . when the ship cannot meet its scheduled transit Figure 5. Artist's rendering of SSSCO's 247-f t schedules, and when it must curtail operations on oceanographic research vessel SWATH design. station because of rough seas. The ability of a SWATH to maintain speed in higher sea states than can a comparably-sized monohull ship, and to fuel reserve. It carries 336 LT of oceanographic continue research operations while stopped in equipment, including a 100 LT itinerant payload, rough seas should contribute to greater and has an endurance of 40 days. The large center reliability. The large deck area, well above the well of this SWATH permits saf e handling of large waterline, improves the capability to continue packages in sea conditions that would cancel working in unfavorable conditions. operations of the MELVILLE and KNORR. In addition Efficiency and productivity is affected by a vari- to the standard suite of winches, A-frames, davits ety of factors. Cost of operations is significant and cranes, this SWATH has a large gantry crane to sponsors of research programs. Expeditions mounted at the center well. This SWATH carries 35 which are aborted, or which return with less A scientists and 25 crew. Its large usable deck results than planned because of ship limitations space of 13,800 sq ft provides space for all deck have a depressing effect on sponsors, particularly equipment and for the location of four vans for additional berthing, labs and storage. at times of new budget preparation. Loss of equipment because of difficulties in recovery in rough seas is not uncommon. Even when the seas do not present hazardous conditions for deck 1167 operations, their motions af f ect personnel with 10. Lang, T. G., Hightower, J. D., and Strickland, varying degrees of discomfort, ranging from A. T., "Design and Development of the 190-Ton increased fatigue to debilitating sea sickness. Stable Semisubmerged Platform (SSP), @L Eng. There is an associated cost in time, related to Ind., 1974, Pages 1105-1111. mistakes and slowness in accomplishing tasks.- that are normally easily done in calm conditions. 11. Hightower, J. D. and Seiple, R. L., "Oper- ational Experience with SWATH Ship SSP The outstanding characteristics of these SWATH Kaimalino",AIAA/SNAME Advanced Marine Vehicles designs are their low motions in rough seas and Conference, San Diego,Apr. 1978, Paper 78-741. their large areas available for research use. 12. Kaharl, V., "SWATH: Calm Seas for Both of these factors contribute to the increase Oceanography", The Oceanography Report, EOS, in efficiency, reliability and safety of research operations. Vol.66,No.36, American Geophysical Union 0096- 3941/6636-0625,September 1985, Pages 626- 627. ACKNOWLEDGMENTS 13. Dinsmore, R.P., and Lang, T. G., "Replacement We thank P.V.H. Serrell, a member of the SSSCO of the University Research Fleet and a 2,500 design team, for reviewing this paper. Ton SWATH Ship Candidate", AIAA 8th Advanced Marine Systems Conference, San Diego, CA, REFERENCES September 1986, Paper 86-2378. 1. A Plan for Improved Capability of the University Oceanographic Research Fleet, UNOLS Fleet Replacement Committee Report, June 1986. 2. Lang, T. G., "The SWATH Ship Concept and its Potential", AIAA/SNAME Advanced Marine Vehicles Conference, San Diego, April 1978, Paper 78-736. 3. Lang, T. G., Sturgeon, W. J., and Bishop, C. B., "The Use of Semi-Submerged Ships to Support New Technology at Sea", MTS/IEEE, Oceans 79 Conference, San Diego, Sept. 1979. 4. Lang, T. G. and Sloggett, J. E., "SWATH Developments and Comparisons with Other Craft", RINA-International Conference on SWATH Ships and Advanced Multi-Hulled Vessels, Paper # 1, London, April 1985. 5. Woolaver, D. A. and Peters, J. B. "Comparative Ship Performance Sea Trials for the U.S. Coast Guard Cutters Mellon and Cape Corwin and the U.S. Navy Small Waterplane Area Twin Hull Ship Kaimalino", DTNSRDC-80/037, March 1980. 6. Narita, H., et al., "Design and Full Scale Test Results of Semi-Submerged Catamaran (SSC) Vessels", IMSDC 82, London, 1982, Paper 11. 7. Sloggett, J. E. and Lang, T. G., "SWATH: T be Past, the Present, and the Future", The Institute of Marine Engineers, Bicentennial Maritime Symposium,Sydney,Australia, Jan 1988. 8. Sturgeon, W.J., "Analysis of Motion Picture Data on Motions in Beam Waves for Single and Tandem Strut SWATH Ship Models Scaled to 1800 Long Tons", Memo, Code 631, Naval Ocean Systems Center., September 1977. 9. Lang, T. G., "SSP KAIMALINO; Conception, Developmental History, Hurdles and Successit, ASME Winter Annual Meeting, Anaheim, California, December 1986, Paper 86-WA/HH-4. 1168 SWATH CHARWIN - RANGE SUPPORT SHIP Ed Craig Scott E. Drummond SEACO, A Division Of SAIC Alexandria, Virginia ABSTRACT CHARWIN - History and Background The SWATH CHARWIN has the following characteristics: The small waterplane area twin hull (SWATH) CHARWIN was built in 1984 for commercial fishing. This 80 Length 80 feet foot steel SWATH successfully harvested scallops Beam 40 feet off the East Coast of Florida. In 1985 the fishing Draft 9feet gear was removed and the CHARWIN was marketed to Displacement (Full Load) 193 Lt the offshore oil industry. Because of the Speed (max) 10 Kts depressed state of offshore oil no orders were confirmed and the ship was laid up. In 1987, the CHARWIN is a unique vessel in several ways. At the Naval Coastal Systems Center investigated using the time of its inception and building during 1983/84, CHARWIN as a range support ship for instrumentation it was the first all steel SWATH to be built in the development. CHARWIN was modified for this mission United States. It is unique in that its design and in May of 1988, and has been-successful in that construction was a private commercial venture which role operating out of Fort Lauderdale and Panama utilized standard shipbuilding practices and City, Florida. This paper describes the CHARWIN modular construction techniques. Further, she was and its conversion and use as a range support built as a work boat for the conmrcial fishing vessel. industry. In these ways it differed from other el, N"@ A A FMr, FIGURE 1 SWATH CHARWIN CH2585-8/88/0000-1169 $1 @1988 IEEE U.S. SWATHS, such as the Navy's Kaimalino which is Activation and Modification a steel/aluminum research vessel built by the Government; or the Suave Lino a smaller all The Naval Coastal Systems Center, Panama City, aluminum boat built as a privately owned yacht. Florida is charged with the development of advanced ocean systems. Up until mid 1987 NCSC utilized the CHARWIN is a rugged work boat which was outfitted USS FIDELITY MSO 443, one of three active vessels as a scalloper as shown in Figure 1. Construction of this class, as a test platform for systems under was started in January 1984 and sea trials were development. FIDELITY was part of a class which completed in May 1984. The modular construction resulted from the Navy's mine warfare experience techniques used were important factors in this gained during the Korean War. She was commissioned short delivery schedule. in 1953 and subsequently modernized in the mid 1960's. This class vessel carries a crew of 70 As a scalloper her large deck area allowed her enlisted and 6 officers. catch to be spread out over the deck. This was in contrast to the conventional method of "pilingn Due to limited availability of the FIDELITY, NCSC used by monohull trawlers. Spreading the catch in obtained CHARWIN as a uniquely qualified this manner prevented heat build up in the lower replacement. She was selected because as a SWATH, layers which resulted in less product spoilage and she could provide good seakeeping, had the large higher yields. open deck area required, had the capability to handle the heavy deck load, and could accommodate The SWATH design allowed CHARWIN to run against the center well required for installation of strong tides which would force other conventional support equipment. However, significant monohull trawlers to resort to tacking along the modifications were required. trawl run. This forced tacking results in allowing a higher sand and debris content to enter the nets Having been laid up for almost three years made it to another factor in quality and profitability. necessary that CHARWIN be hauled out of the water and given a thorough inspection and overhaul. A major factor in the design of CHARWIN was her cargo capacity. At 104 long tons CHARWIN could Once the decision was made to utilize CHARWIN NCSC handle about three tires as large a catch as a personnel worked closely with the owner Charles standard monohull. Additionally, by distributing Rains, President of WINCHAR, Inc., his naval this load over the large deck both fore and aft, architect De Jong & Lebet, Inc., and the shipyard the vessel avoided heavy deck loading on the stern where the work was to be accomplished, St. which has caused the loss of many conventional Augustine Trawler Inc. In order to save as much trawlers. time as possible, since the system development schedule was of paramount importance, this close The efficiency of the SWATH design resulted in a cooperation was vital. Many decisions were taken lower fuel and oil consumption than originally on a day-to-day basis and design changes were anticipated. of course the seakindliness of the implemented as the.work progressed. design made life much easier on the crew during heavy weather. In addition to re-installing new engines, decisions had to be made regarding their type and location. The lower hulls and struts of CHARWIN were designed original plans to install direct drive diesels in to meet both the fishing mission load,requirements the lower hull were called into question because of and ease in construction. The lower hulls are 61' the possibility.of acoustic interference. The use 4" in diameter. The struts are four feet wide and of a hydraulic drive was investigated and 75 feet long. Although this is a large waterplane subsequently rejected in favor of a deck mounted area for a SWATH of this size, and it did belt drive system. In its final configuration the compromise some of the seakeeping performance vessel is powered by two 485HP GM 8U9TI main .people have come to expect from such craft, it was engines located on the after deck in small engine outstanding in meting the fishing mission compartments on either side. The belt drive requirements. Even with this large wAterplane required several modifications owing to the limited area, the vessel when rated by its two most availability of proper component parts. experienced Captains proved to be far more seakindly than monohull t 'rawlers of similar size. The existing deckhouse with the spartan These observations cited operations in which sea accommodations found on most fishing vessels was conditions included waves up to twenty@five feet. inadequate for the new mission. The scientific party to be embarked had to be housed and When the scallop catches declined in 1985 new facilities provided for the instrumentation that business was sought from the oil industry which would be brought aboard. One half of the existing unfortunately was also suffering a sever decline at deckhouse was converted from a ships control center the same time. Although interest was expressed in to a housing compartment with six berths, two small CHARWIN from several sources the coincidence of heads, and a galley/messing area. The other half these two unhappy events forced the owner to lay up of the deckhouse was converted into the scientific CHARWIN. operations center for use by the systems operators. 1170 These spaces were air conditioned and supplied with The seachest design ran through several iterations the appropriate power required to support the from a simple swingdown system originally thought testing operations. to be adequate to a much more complicated structure. The seachest, extending from the main The modifications to these spaces were kept as deck to keel depth, a distance of approximately simple as possible since the requirement for the sixteen feet, is protected by a seven by vessel is limited primarily to day work with only twenty-eight foot fairing. This fairing was the occasional overnight operations anticipated. object of considerable discussion not only because of the size but because of concerns over the effect A pilot house with modest berthing facilities for on the sensor electronics. It was finally mandated the crew was added on top of the existing that the structure be constructed of stainless deckhouse. A crew of three, supplemented by a steel to minimize any interference with the system. fourth member for prolonged operations, was provided for. The after portion of the deck was This large structure was permanently secured given over to housing the necessary power between the lower hulls with large horizontal foil generation equipment and the large winch and reel bracings. Inactive fins were added forward on the required to handle the scientific package. Space lower hulls to provide additional motion dampening was also available to handle three instrument vans while underway. Figure 2 shows the seachest as which carried the associated systeins electronics. finally constructed. The most significant modification was the addition Activation and modifications began February 25, of a center well and seachest to accommodate the 1988 and sea trials for both CHARWIN and the instrumentation. The well was cut in the main deck installed system were completed May 20, 1988 when and a shelter to provide protection for the the CHARWIN sailed. Needless to say this time instrumentation package, and to accommodate a period was extremely short. Labor and weather Special support structure for raising and lowering problems had their part to play in complicating the the instruments was built over the well. Because work of the overhaul. of the weights and strains anticipated these structures were of significant size. 4", -X, 51 J@ . .. . . ..... "fffl . . . . . . .. . . . .-11192, 1! &N . ....... . FIGURE 2 SEACHEST MODIFICATION 1171 Material and equipment problems required ad hoc incurred by a military manned vessel engaged in decisions on the spur of the moment but finally all similar activities. Use of the FIDELITY with her was completed. Figure 3 shows CHARWIN just after 76 man complement will be necessary for final sea trials with all equipment on board. operational evaluation, but CHARKIN with her small crew has provided the Navy with an inexpensive OPERATIONS platform during this part of the systems development cycle. After sea trials of both vessel and scientific equipment were completed in the St. Augustine area Lessons Learned where the modifications took place, CHARWIN began operations out of Fort Lauderdale. After SEACO, the vessel's operator, has had extensive completing operations in the Fort Lauderdale area, prior SWATH experience, as operators of the Navy's the vessel transited to Panama City in 3 1/2 days. SWATH Kaimalino and having participated in the She subsequently has returned to Fort Lauderdale early operations of Suave Lino. Our current work for additional operations in that area. with CHARWIN confirms the outstanding utility of yet another SWATH variant. The vessel has proved to be completely successful The installation of the large seachest housing in that she provides a stable platform equally presented the possibility of a significant change capable of transiting at ten knots or towing in the seakeeping characteristics of the original equipment through the seachest at two to five vessel. Since this was an item of such concern knots. during the early stages of planning, the owner, WINCHAR, constructed a six foot model for empirical In operating in both the Atlantic Ocean and the testing in the St. Johns river. NCSC subsequently Gulf of Mexico under various sea conditions, it is requested that the model be tested at the Stevens clear from coments from the onboard operators that Institute of Technology tow tank facility. The CHARWIN compares favorably with the FIDELITY, more result of this test showed satisfactory than twice her length, in regards to both vessel performance. and equipment operation. Finally, the transition of the CHARWIN from fishing In addition to these attributes CHARWIN is a vessel to test platform can be considered highly comparatively low cost platform. Her crew of three successful, and demonstrates the versitility of or four slashes the major operating cost (manpower) the SWATH design. q, -r7@ 7 77 -V d .7f X, z-_ 41 1.,:4 4 :4 Z 7" Z . ... . ...... All FIGURE 3 CHARWIN UNDERWAY 1172 AUTOMATED ANALYSES OF NUTRIENTS IN SEAWATER WITH THE TECHNICON TRAACS-800 AUTOANALYZER SYSTEM J.D.Guffy, M.A. Spears, D.C. Biggs Technical Support Services Group, Dept. of Oceanography Texas A & M University, College Station, Texas 77843 ABSTRACT AUTOANALYZER AND METHODS During 1987-88, we adapted methods for The TrAAcs-800 System is a computer automated analyses of nutrients by controlled, segmented flow analysis, wet Technicon's newest generation industrial chemistry analytical system. The small- autoanalyzer so that the TrAAcs-800 bore components allow samples to be run could be used to analyze low levels of three to five times faster with longer nutrients in open-ocean water. The steady-states for each peak and with a TrAAcs-800 system runs three to five better separation between peaks than did times more samples with longer steady the "older generation" Technicon II and states and better separation between Alpkem IWA-6 Analyzers. peaks than can older generation autoanalyzer systems like our Technicon The Traacs System soft-ware prints real AA-II or our Alpkem IWA-6. On completion time data next to the peaks during the of a run the TrAAcs operating program run and upon completion of the run automatically calculates sample automatically corrects for baseline concentrations from the linear drift, any changes in the standard regression slope obtained from the absorbances, and carryover. It then standards which are run and corrects for recalculates the concentrations from the baseline drift, changes in absorbance, corrected peak heights and prints this and carry-over. corrected data under the original run (Fig. 1). The raw data and calculated concentrations are stored on a floppy disk from which one can make x-y plots of the standards, take a closer look at INTRODUCTION each of the results, correct for errors and/or anomalous peaks, and recalculate the run. From June 1987 to May 1988 the Technical Support Services Group of the Department The flow rate for the TrAAcs System must of Oceanography at Texas A & M be within certain limits, depending upon University adopted oceanographic the length of the flowcell, so that the nutrient methods to allow the de-bubbling software will cause the determination of low concentrations of bubbles to be ignored. The optimum flow phosphorus, nitrogen and silican species rate is about one-half what it is for in seawater using a "new generation" earlier systems, so we have adapted our TrAAcs-800 auto analyzer developed by nutrient methodologies to fit the slower Technicon Corporation. We evaluated its flow rate. performance in meeting our needs for accuracy and precision of the four The nutrient methodologies we are using nutrients P04, N02, N03 and SI(OH)4 both at TAMU are modified from those in the laboratory and at sea aboard our pioneered by Dr. Lou Gordon's research vessel. autoanalyzer group at Oregon State University (E.L. Atlas, et al. 1971). The purpose of this note is two-fold: The recipes, manifold diagrams and other first, to summarize the results of our pertinent information concerning the Technical Report 88-02-T published in methods modification for the TrAAcs-800 April 1988, and second, to update our can be found in our Technical Report 88- findings subsequent to that report. 02-T (M.A. Spears et al. 1988). CH2585-8188/0000- 1173 $1 @1988 IEEE PRELIMINARY LAB RESULTS Property-property plots of the average nutrient concentrations versus Preliminary results from the TrAAcs-800 temperature and versus sigma-t density gave limits of detection as follows: were quite similar to results obtained 0.15 ug-at/l for phosphate; 0.05 ug-at/l with a Technicon AA II used on cruises for nitrite; 0.5 ug-at/l for nitrate; 87-G-11 and 87-G-12 (Spears et al., and 0.5 ug-at/l for silicate. 1988). Both the 1987 and 1988 data Laboratory accuracy and precision data agree with historical data for the (derived from a run consisting of 45 Western Gulf of Mexico (Morrison et al., shots of 13 different standards) are 1983). givin in Table 1. The first column gives the number of the standard (roman DISCUSSION numerals designate the working standards used to form the calibration curve, and Linear standard curves and flat ordinary numbers designate other known symmetrical peaks with good separation concentrations used as samples). The indicated that the TrAAcs-800 can second column gives the number of times support accurate nutrient analyses of each was analyzed. The succeeding f our open ocean seawater. However, noisy sets of three columns give the known baselines and poor sensitivity at low concentration, the mean TrAAcs level concentrations still have to be calculated concentration, and the improved. standard deviation for each respective nutrient. The improvement in our latest TrAAcs (SEE TABLE 1) data is the result of several factors. All lines, joints, and butt joints were LATEST LAB RESULTS replaced. Large pump tubes for the samples were replaced with smaller Our latest TrAAcs data were compiled in matched pairs. This enabled the orange- May of 1988 from fourteen runs yellow debubbler pump tubes to better consisting of 350 samples. With some remove air which enters when the sampler modifications to our analytical methods moves from position to position. The (detailed in the discussion which flow-rates were altered slightly to follows) , we have been able to improve better fit the optimum flow-rate for the limits of detection to 0.04 ug-at/1 each length flow cell. The sampling for phosphate, 0.02 ug-at/l for nitrite, rate was reduced from 75 samples per 0.1 ug-at/l for nitrate, and 0.2 ug-at/l hour to 60 samples per hour, and the for silicate. sample to wash ratio was changed from 5:1 to 3:1. The mean standard deviations and coefficients of variation are as The following is our evaluation of follows: TrAAcs-800 System strengths and NUTRIENT S(UG-AT/L) Cv weaknesses. P04 +0.03 0.013 N02 +0.01 0.011 STRENGTHS N03 +0.09 0.003 SI(OH)4 +0.18 0.006 1. Has the capability to run samples three to five times faster than older autoanalyzers. FIELD RESULTS 2. Peak height measurements and calibration factors produce Field data were obtained from two 24- concentration data that is almost bottle CTD casts taken in the Gulf of identical to that computed by hand. Mexico on 31 January and 1 February 1988 during cruise 88-G-02 aboard R/V GYRE. 3. Prints real-time data next to the Samples from each depth were drawn into peaks. three nutrient bottles and replicate samples then drawn from each of those 4. Automatically recalculates samples bottles. This provided three distinct with a linear regression slope from analyses with replicates for each CTD all the standards and corrects for cast. baseline drift, any change in standard absorbance and carryover. At sea, subsampling precision within each distinct run averaged + 0.02 ug- 5. Stores raw data and calculated at/1 for phosphate, + 0.2 ug-at/1 for results on a floppy disc. nitrate and + 0.1 ug-at/1 for silicate (Spears et al., 1988). 6. Prints relative absorbances. 1174 7. Numerous options available for 6. Technicon Instruments Corporation. recalculating and plotting data. TrAAcs-800 Industrial Methods 811-86T (Silicates in water and 8. Compact and easily transported. seawater); 817-87T (Nitrite in water and seawater); 812-86T (Orthopphosphate in water and sea- WEAKNESSES water); 818-87T (Nitate in water and seawater). 1986-1987. 1. The system is fragile. 7. Whitehouse, M.J. & V.R. Woodley. 2. The flowcells in particular are Automated Seawater Nutrient extremely fragile yet must be Analysis, British Antarctic Survey- flushed on a regular basis for best National Environment Research results. Council, High Cross, Madingley Road, Cambridge CB30ET, U.K., 1987, 3. Technicon provides minimal 41 pages. information about software and instructions for field servicing a 8. Morrison, J.M., W.J. Merrell Jr., defective module. R.M. Key and T.C. Key, Property Distributions and Deep 4. Reliable initial baselines are Chemical Measurements Within the often difficult to obtain or keep. Western Gulf of Mexico. J.Geophys. Res., 88(C4); 1983, 2601-2608 5. Sample cups are too small and easily contaminated. 9. Spears, M.A., J.D. Guffy and D.C. Biggs. Automated Analyses of 6. The system is not as sensitive as Nutrients in Seawater with the the older generation AA-II. Technicon TrAAcs-800 Autoanalyzer System. Technical report 88-02-T Department of Oceanography, Texas A & M University, College Station, REFERENCES TX 77843, 1988, 68 pages. 1. E.L. Atlas, L.I. Gordon, S.W. Hager & P. K. Park A Practical Manual For Use of the Technicon Autoanalyzer in Seawater Nutrient Analyses (Revised). Tech Report 215, Department of Oceanography, Oregon State University, Corvallis, OR 97331, 1971, 49 pages. 2. Technicon Instruments Corporation, AA-II Industrial Methods 155-71W (Orth-phosphate in water and sea- water); 186-72W (Silicates in water and seawater) ; 161-71W/B (Nitrite in water and seawater); 158-71W (Nitrate and Nitrite in water and seawater). 1972-1976 3. Alpkem Corporation, Manual for the IWA-6 Industrial Water Analyzer. 1981 4. Glibert, P.L. & T.C. Loder. Automated Analysis of Nutrients in Seawater; A Manual of Techniques. Tech Report WHOI-77-47. Woods Hole oceanographic Institution, Woods Hole, MA 02543, 1977f 46 pages. 5. Whitledge, T.E., S.C. Malloy, C.J. Patton & C.D. Wirick, Automated Nutrient Analyses in Seawater. Tech Report BNL 51398, Brookhaven National Laboratory, Upton, NY 11793, 1981, 216 pages. 1175 FIGURE 1: a) Chart of a TrAAcs-800 analytical run with five calibration standards and one sample which was analyzed sixteen times to evaluate analytical accuracy and precision. b) Gives the corrected results for the run. The known concentrations of the four nutrients in the mixed standard analyzed sixteen times were 1.32 ug-at/1 for phosptrate, 0.72 ug-at/l for nitrite, 20.5 ug- at/l for nitrate and nitrite, and 19.5 ug-at/l for silicate. The high cv for phosphate (71.0%) is the.artifact of a baseline shift after the run. I h er t-m, ar (--? 2"',-"1 U8 (b ytes I" ee of-, d i 5@ k . Fi le D: DU.C'HFN' will talle aj@)proximaLL@,ly V5t@2 (bytes). Channel s 1 2 3 4 Dase: 24 31 -16 23 G a i n 75 75 27 7 75 P04 1403 Si (014) 4 4 LIM LIM UM LIM BzLse dri Ft correction made Carryovpr correction made. Gain dri Ft correction made Init base 42LI E14 214 11 V 0 -F . . . L + + + + CO c. S06 ()9e 1 -1 ,. 222e+(X' 2. 737e+0('.) ...... K 1 typ P 1.125 N 0. 8=3 N 1.7.66 N 16.61 Pr, I CP - n 1 e-0 1 1.7 1-1 0 . -3, 4 5 7 Q. 1167 Rel Ab5 0. 1 o-7 ...... . ... Fl-'* 2 CP -2 typ C 1.162 0. 6799 N 17.96 N 16. 93 V-1 3 typ C 0.9746 N C). 7 14.99 14. 38 CP (:i 4 FT 12 ..... j Pl.': 4 CP 4 typ C 0. 7860 (1. 56416 N 12.07 11 . ZI 7 + + + + PK 5 CP 5 typ C (1. 5742 N 0. 47166 N 9. 079 9 . 5, 6 9 K. 6 CP 6 typ C 0. 3708 N 0. 2e66 6.164 5.685 ........ P 7 C P 7 typ C 1.127 N 0. 67-?0 N 17.74 N 16.67 -5 Ill F,f, 0 CP B typ H 0.17 0. 1779 M 3 . -3, 8 6 M 2.766 11 ......... P [-,'. 9 CP 9 typ L 0. 1471 N 0.1662 11 3. 227 M 2. 6 10 N 16 FT'* 10 CP 10 typ L 1 . 7- 1@ 5 N 0. 7204 N 2o. 1-6 N 19.29 + + + + + P@::: 11 CP 11 typ S I -.2o N 0. 7235 2o.51 N 19.61 F'f:,' 12 CP 12 typ S I . 3 22' N 0.7296 20.51 N 19. 6 PKI 13 CP 13 typ S 1.318 0. 7'@'.-.7 20.55 N 19.66 PP'. 14 CP 14 typ S I . 3 71 5 N 0. 77@@ 6 '20. 47 19. 68 ..... ......... F'[;..' 15 CP 15 typ S - 20.49 N 19.j5 1. -@'16 N 0.7262 N - ...... -'K' yp S 1.317 0. 7254 19.5C.) 20 F 16 CP 16 L 20. 4o + + + 4- F'K' 17 CP 17 typ 3 1 . 0 B (I. 7@109 N 2 0. 29 19. 5'@ 10 CP 18 typ S 1 . 3. 11 0. 72-50 '2@ 0 . 4 e N 19. SO PK 19 CP 19 typ S I . :!'C) 7 N Cj . 7 2 @- 3, 220. 51 N 19. is P 1%'. 20 CP D-) typ S 1. 304 N 0. 7292 2 0. 4 6 N 19.57 ....... 21 C P 21 typ S 1.296 0. 72t53 N '20 . 5 2 11 19.61 PK - 22 C P 22 typ S 1 . 2 03, N 0.7271 N '20. 41 19.62 24 F'[::' - - - - + + + Pk: 23 CP 23 typ S 1 . 234 N o. 7'207 N --o.4S N 19.61 1-4 CP '24 typ S 1 . '206 o.7155 N 2;-J. q Cj 19.67 - 1 . '122 N 0. 71194 N 20.52 N 19.83 P1 25 CP 25 typ S - 2 6 C P -7'6 typ 17. ?8 17.22 F'K. G I . 100 N 0. B-1@6 N Last base 796 269 2o7 54C.) 3 ..... Conc. 1. FJ06e-0 1 1. 709e-0 1 3. 222e+00 2. 737e+(-.* =I, + 4 4, + + 4 +-j sample statistics A n a I -v e I J 1L C Lt P S 8'Vg std dev cv -32 1 all 15 1..@X),52 0. 0154 0. (".)1 1 S + + + + + + + + 2 al 1 15 0. 7245 0. ()044) 0. ()055 all 15 2c.-) . 4 6 6 6 0. 0667 1. QO'-33 4 all 15 19.6174 0. 09f.)4 0. 0046 Anal 1: 1 i near fit 1.5t5e-o2 1. B 1 (".)F---o2. Carryover factor: 0. Anal '2: 1 i near fit 9. 153"e-o3 1. 074e-01. Carryover factor: 0. 00%. Anal ::: linear fit 1. 9SOe-01 2.137e+00. CEtrt--yover factor: I:). C)C)7. Anal 4: 1 i near fit 2. 07.1)e-O I 074e-0 1. Carryover f':'ctor: 0. 00%. 1176 TABLE 1: Summary of laboratory precision for our 4-channal TrAAcs-800 System from Summer 1987 methods development work. Accuracy can be evaluated by comparing known concentrations of each nutrient (first column in each set), with the calculated means (X) and standard deviations (S) in columns two and three of each set. n rP041 X S rN021 x S rN031 x SrSI(OH)41 X S 1 7 1.88 1.88 .02 1.02 .98 .03 31.1 31.1 .1 30.3 30.2 .2 11 3 1.25 1.22 .02 .68 .68 .02 20.8 20.8 .3 20.2 20.4 .1 111 3 .63 .63 .01 .34 .38 .03 10.4 9.9 .1 10.1 10.7 .2 IV 5 0 .08 .01 0 .06 .02 0 .2 0 0 .5 .3 5 3 .20 .22 .03 .07 .07 .03 2.7 2.7 .1 2.6 3.2 .3 6 3 .30 .30 .01 .13 .13 .01 5.4 5.2 .2 5.3 5.7 .1 7 3 .65 .64 .01 .25 .24 .01 10.7 10.2 .1 10.5 11.0 .1 8 3 1.30 1.30 .02 .50 .49 .01 21.4 21.0 .2 21.1 21.4 .2 9 3 1.96 2.10 .05 .75 .73 .01 32.1 32.1 .1 31.6 31.7 .2 10 3 .13 .14 0 .07 .14 .01 2.1 2.2 .1 2.0 3.1 0 11 3 .31 .32 .01 .17 .28 .09 5.2 5.2 0 5.1 6.2 .1 12 3 .80 .81 .02 .90 .89 .01 10.4 10.4 .1 10.4 10.5 .2 13 3 .68 .67 .01 .22 .22 .02 10.4 10.5 .1 10.4 10.7 .1 .78 .79 .02 .39 .41 .02 12.5 12.4 .1 12.3 12.7 .2 1177 AN APPLICATION OF A LOW FLOW CURRENT METER To BROAD TEMPERATURE RANGE ESTUARINE CURRENT MEASUREMENTS David J. Murphy, Eric Powell and Elizabeth Wilson Deparment of Oceanography, Texas A&M University, College Station. TX 77843 Ai3STRACT flow and will change in resistance proportionally to the Instruments utilizing heated thermistors as sensors change in ambient temperature. The voltage across the can measure currents over the range of 5 to 50 cm/s. Owing bridge determines the voltage supplied to the bridge by a to their small size and great sensitivity, such devices are ideal difference amplifier and a power supply transistor. This for the estimation of benthic boundary layer thicknesses, stirring rates in in-situ incubation experiments and currents circuit provides approximately 10% precision over a 5' in and around sessile creatures. Such an apparatus is temperature range (MacIntyre, 1986). presented and a means of removing ambient temperature effects from data discussed. Ambient temperature effects may also be corrected for mathematically. A simple bridge with a heated thermistor INMODUCTION and a constant supply voltage is used to deterrnine the power A small device for measuring fluid flow rates in the dissipated across the thermistor for a given flow speed and range of 5 to 50 cm/s has application in a variety of ocean ambient temperature. science endeavors. This sensor can measure benthic This paper is an extension of the mathematical boundary layer thicknesses, flow rates in and around sessile treatment of flow measurements, described above, applied creatures, such as sponges or anemones, or stirring rates in over a temperature range of 15 to 35' as would be found in in-situ incubation experiments. Heated thermistors, of the southern estuarine environments. type described in this paper, provide a good means for measuring low flow rates. Their small size allows placement METHODS in confined areas and they exhibit highest sensitivity in the A bridge is constructed with one leg consisting of a range of 0 to 20 cm/s. 100 K resistor and a 1 M potentiometer and the other a 1 K Thermistor properties that facilitate measurement of thermistor and resistor selected to match the resistance that fluid flow, unfortunately, make the instrument highly the thermistor will have at a temperature approximately 10' sensitive to changes in ambient temperature. Strategies to above the highest expected ambient temperature. overcome this problem include a self compensating bridge T'hermistors of nominal resistance higher than 1 K were tried (Reidl, 1972; LaBarbera and Vogel, 1976; and MacIntyre, and found to require too high a power supply voltage while 1986) and mathematical treatments to generate a temperature- thermistors of lower nominal resistance were unstable at independent power dissipation coefficient (Rasmussen, higher temperatures. 1962; Katz, 1987; and Lents, 1986). Power to the bridge is provided by an adjustable A self compensating bridge consists of a low voltage regulator. The proper supply voltage is determined resistance value heated thermistor on one leg of the bridge as follows. With the thermistor in water of the highest and a high resistance unheated thermistor on the other leg of temperature in the desired range of operation, the voltage is the bridge. The unheated thermistor is kept out of the fluid slowly increased until the voltage drop across the resistor CH2585-8/88/0000- 1178 $1 @1988 IEEE R1 (Figure 1) and the thermistor are equal. For a I K Where Vo is the output voltage of the bridge, Vc is the thermistor and a desired operating temperature of 45' the voltage across the compensating resistor, Vb is the bridge supply voltage was approximately 16.3 V. The thermistor is supply voltage and R1 is the resistor in series with the then placed in water of the highest expected temperature and thermistor. the I M potentiometer is adjusted to give a bridge voltage of approximately 0.5 V. The bridge output voltage is buffered Finally, bead temperature can be derived from with a unity gain differential amplifier and converted to a 0 to thermistor resistance data supplied by the manufacturer. A 10 KHz frequency for the sensor output. simplex optimization algorithm fitted the inverse of bead temperature to resistance as: I Tb = C1 + C2 Ln(Ro + C3 Ln(R03 5) R, v Where C1, C2 and C3 are coefficients determined by the V, to F minimization of the sum of the squared residuals in the optimization routine. T RC The above equations and knowledge of the ambient temperature permit the calculation of a power dissipation coefficient for the thermistor sensor. Figure 1. Simplified Flowmeter Circuit Diagram. The sensors are calibrated in a I M3 temperature- As a fluid moves over the sensor, heat is removed controlled tank. A turntable, fitted with the sensor is placed from the bead and the power dissipation coefficient in the tank and the drive motor and slip rings positioned over increases. The power dissipation coefficient may be defined a cutout in the tank lid. When a desired temperature is as: reached, the stirring motors are shut off, and the tank temperature monitored at the level of the sensor. The K Pb turntable is rotated once while speed and power dissipation (Tb-Ta) data are collected. Calibrations are then made in 50 intervals Where K is the power dissipation coefficient, Pb is the over the range of 15 to 350. power supplied to the sensor bead, Tb is bead temperature RESULTS AND DISCUSSION and Ta is ambient temperature. Equation (1) is applicable in situations of small ambient temperature range or small range Five calibration data sets were generated over the 15 in flow rate. In the case of large ambient temperature range to 35' range. Each temperature curve was fitted, with the and/or flow rate, equation (2) gave more use ful results. previously mentioned optimization algorithm, to the K' Pb 2) equation: Tb K = A+BLnU 7) Power supplied to the thermistor bead is defined as: Vt2 Where A and B are coefficients arrived at by the same Pb = Rt 3) method as the thermistor bead temperature coefficients, and U is the flow speed. This equation is inverted in order to Where Vt is the voltage drop across the thermistor and Rt is calculate flow speed, U , from power dissipation, K, as: the theTMistor resistance. The voltage drop across the FK - A] thermistor and the resistance of the thermistor are, in turn, U = exp 8) L -BJ defined as: Vt = V0+ Vc and R, - Vt R1 4) Figure 2 is a plot of the power dissipation , <R, >RC @::::" Vb - Vt coefficient, derived from equation (1), for each temperature. 1179 For speeds less than 15 cm/s or an ambient temperature owing to their uniform slope and spacing. Values for A and range of 15' or less, the effects of ambient temperature B at temperatures other than those calibrated may be changes are corrected for by equation (1) One aspect of this estimated by a piece-wise linear approximation or fitted to a calibration method is a slightly higher power dissipation polynomial equation. The curve coefficients from Figure 2 coefficient at 300 than at 25'. When a bridge was constructed do not have a linear relationship to ambient temperature and to operate 20' above the highest temperature in the desired thus do not lend themselves to simple approximation for range, its calibration produced similar results. The temperature other than those calibrated. Table 1 lists thermistor manufacturer wams of long term instability for optimized values of A and B for each calibration curves in operation at temperatures above 50', prompting the selection Figure 3 of the lower operating temperature. Ambient temperature A B 70 35 3.352 0.0443 30 3.741 0.0429 60 25 4.137 0.0535 50 20 4.628 0.0746 40 15 5.355 0.1474 '0 30 20 REFERENCES 10: Katz,l., and E.J. Shaughnessy, 1987. Digital temperature 0 compensation of a thermistor flowmeter. J. Phys. E. Sci is 16 17 18 19 20 Instrum. 20: 561-564. Power Dissipation Coefficient (mW/'C) Figure 2. Speed versus power dissipation coefficient as LaBarbera, M. and S. Vogel, 1976. An inexpensive derived with equation (1). thermistor flowmeter for aquatic biology. Limnology and Oceanography, 21: 750-756. Figure 3 shows speed versus power dissipation Lents, C., F. Incropera and R. Viskanta, 1986. Application coefficient, derived from equation (2), for the same data. of heated thermistors to speed measurements in Equation (2) provides a family of calibration curves that thermohaline solutions, Solar Energy, 36(2): 179-196. increase regularly in power dissipation as ambient MacIntyre, S., 1986. A flow-measuring system for use in small lakes. Limnology and Oceanography, 31(4): 900- temperature decreases and have very little scatter in data. 906. Although equation (2) provides curves of very steep slope, Rasmussen, R., 1962. Application of thermistors to the uncertainty in the measurement is small enough to permit measurements in moving fluids, Review of Scientific a fl.ow speed resolution of + 10%. Instruments. 33(l): 38-42. Reidl, R. and R. Machan, 1972. Hydrodynamic patterns in 10- lotic intertidal sands and their bioclimatological implications., Marine Biology, 13: 179-209. 60 50 E 40 30 20 - 10 0 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Power Dissipation Coefficient (mW/'C) Figure 3. Speed versus power dissipation coefficient, K', as derived with equation (2). U Over wide ambient temperature and speed ranges, the calibration curves shown in Figure 3 are more desirable 1180 DESCRIPTION OF CONVERSION OF AN EG&G VMCM INTO A MVMS (Multi-Variable Moored Sensor) Miguel Maccio and Christopher Langdon Lamont-Doherty Geological Observatory of Columbia University Palisades, NY 10964 ABSTRACT irradiance (PAR), beam transmission, chlorophyll fluorescence, conductivity and dissolved oxygen (DO). A standard EG&G VMCM has been modified to Two of the sensors, the SeaTech fluorometer and the acquire data from external sensors. The external ENDECO type 1133 oceanographic DO sensor, were sensors are mounted on a compact backpack which developed in cooperation with the respective companies attaches to the VMCM's cage. Some simple changes specifically for the Biowatt mooring. The SeaTech to the VMCM's A/D board opened up eight analog fluorometer was designed to have lower power voltage channels. Changes to the VMCM's firmware consumption and better noise characteristics than other were made to allow control over the period of time available fluorometers. The ENDECO DO sensor was between toggling external power on and the A/D designed to employ a pulsing technique (Langdon, sampling cycle. Some sensors with low power 1984) which improves long term accuracy. Also, consumption were powered continuously and their because the output is independent of the membrane, signals passed through a low pass filter (TC 8 min) in anti-foulant wax can be directly applied to the order to match time constants and minimize aliasing. membrane surface. Sixteen VMCM's were modified and three deployments of the MVMS's were made in 1987 with durations of A field proven unit, the VMCM (Weller and 3 to 4 months as part Biowatt II, funded by the Davis, 1980), was chosen to serve as the primary Office of Naval Research. Data return was better data-gathering and recording instrument in the than 80%. Reasons for loss of data include tape system. The other sensors for irradiance, fluorescence, breakage, underwater connector failure, pressure case beam transmission, conductivity and dissolved oxygen flooding, and fouling. were mounted on a backpack which was in turn mounted on the pressure case of the VMCM (Figure 1). The following sections describe the physical, 1. INTRODUCTION electrical and programming modifications to the VMCM necessary for converting the the VMCM into a Multi-Variable Moored Sensor (MVMS) capable of Ocean processes are highly variable in space and logging nine 0-5v analog channels with power toggling time. Improved understanding of these processes of sensors and anti-alias filtering. requires higher resolution sampling than is possible from shipboard. For any given project there are a 810-OPTICAL PHYSICAL MOORED MEASURING SYSTEM suite of variables that are of interest, and the ability VMCM. to monitor many of these variables concurrently is LEFT VIEW FRomr RIGHT VIEW highly desirable. Our interest in the development of a multi-variable moored sensor was fueled by the requirments of the program Biowatt (Bioluminescence and Optical Variability in the Sea), Briefly, the major ORTHOGONA CURSEN goal of Biowatt is to predict patterns of oceanic ROTOR37 bioluminescence and relate these to variability in optical properties over appropriate spatial and 'THERN'ST" temporal scales (see Marra and Hartwig,1984). The FLUOROMETER @ - _aEA best hope for meeting these objectives is to MeSSURE TUM36AS30ACM 'Nousam Pon supplement the classic shipboard sampling with a CLECTRON" =ED BATTERIES. OXYGEN mooring that would obtain data on bio-optical and TAM OML SENSOR ETC. physical variability over an annual cycle, and resolve MAN TRANS- CONOUCT'Virr-" V01:0HOUCTIVITY NISSOMETER SENSOR SENSOR, time scales as short as a few minutes (see Dickey, Hartwig and Marra, 1986). For the Biowatt II mooring, the variables selected for measurement were: currents, temperature, photosynthetically available Fig. I Physical layout of sensors on MVMS. CH2585-8/88/0000- 1181 $1 @1988 IEEE 2. DESIGN CONSIDERATIONS 2.2.2 Matching the time constants The VMCM had to be modified to accept at One of the parameters measured was the salinity. least rive external analog signals. Given the desired This parameter is calculated from the temperature and deployment times of 3-4 months data storage conductivity of the water. The conductivity sensor limitations set the maximum sampling frequency at used was the Sea Bird Model SBE-4 whose response once every 4 min. The 4 min sampling rate in turn time is very small compared to the temperature signal created the necessity of low-pass filtering key signals, from the EG&G VMCM which has a response time of such as irradiance, beam transmission, and approximately 240 s. In order accurately to compute conductivity in order to avoid aliasing. The salinity it is desirable to match the time constants of fluorometer and DO sensor are pulsed output devices the conductivity and temperature sensors. Therefore and therefore are not conveniently filtered. Therefore the conductivity signal was passed through a low-pass these sensors were power toggled to come on just filter with a time constant of 240 s. before the A/D sampling cycle. Details of all the modifications necessary to meet the design objectives 2.3 Data storage are given in the following subsections. There is no need of any modification to the data 2.1 Data input storage, capability of the EG&G VMCM. The instrument is already prepared to write all the data The standard EG&G VMCM uses a sampled by.the A/D to the tape. time-multiplexed scheme to gather temperature data, Each data record consists of the following: auxiliary voltage and current consumption of each circuit board. Using this idea it is possible to substantially increase the instrument capabilities -by Description of bits simply replacing the front end of the A/D board and doing minor modifications to the instrument firmware. Preamble 8 This instrument already logs all data that is converted Record count 16 by the A/D board into the cassette storage system. North vector 16 Therefore, with very simple changes the instrument is East vector 16 open to acquire signals from external sensors. For our Rotor 1 count 16 application some signal processing is necessary to Rotor 2 count 16 interface these sensors to the A/D board, as is Compass 8 explained in the next paragraphs. Temperature 16 Pressure 16 2.2 Signal processing Auxiliary PAR 16 CM1 Battery Voltage 16 2.2.1 Anti-ahassing filters CM2 Not used 16 CM3 Dissolved 0. 16 The spectrum of the signals sampled by the A/D CM4 Conductivity 16 do not have significant energy levels above half the CM5 Transmissometer 16 sampling rate. Ilowever,the vertical movement of the CM6 Dissolved 0, Temp. 16 MVMS through the water column due to mooring CM7 Not used 16 oscillation, combined with the signal change along that CM8 Fluorometer 16 path, make neccessary the use of anti-aliasing filters. Parity 4 Just as a quick example, consider the signal from the PAR sensor. Mooring motion will cause the water TOTAL Data bits 276 column above the sensor to oscillate approximately one Inter record gap 76 meter with a period varying from 5-12 s. - The Total # of bits per record 352 output of this sensor will vary from 7-14% m-1 450 ft tape= 17.25 106 bits depending on the clarity of the water. Therefore, the Therefore, the data capacity is about 49,000 records. signal that is being sampled will contain a significant high frequency component that will cause aliassiDg Each record contains: the record number, east and noise. north components of the, current, upper and bottom rotor counts, compass, and the extra channels used to Since the sampling rate is 4 min a first order measure the signals from the external sensors. The low-pass filter with a cutoff frequency of 10 min. and deployment periods were slighty shorter than four a 20 dB/dec attenuation slope after the cutoff will months and since the tape capacity is 49K records,the sufficiently attenuate the movement- induced high fastest sampling rate possible is 4 minutes. frequencies to a level smaller than the quantization step. In other words, the noise due to abassing will be smaller that one bit. 1182 2.4 Power consumption board functions. The F/V uses a constant charge pump capacitor technique, which has a very linear Special consideration was given to this area of response, very low drift with temperature, and requires the design. The relatively laxge amount of sensors only 1.0 mA. After the F/V conversion the signal is and small volume available for the batteries made it low-pass filtered (240 s time constant) to equate its necessary to switch off some of the power-hungry response time to the temperature sensor's. sensors while the system was not sampling. The power-toggling was used only with those sensors that 2.5.2 Isolation requirements of the DO sensor do not need anti-aliassing filters. The power-toggling effectively decreases energy requirements. As an Proper operation of the ENDECO DO sensor example, the DO sensor consumes 11 mA-h if requires that the instrument be electrically floating operated continuosly. Switching power on only during with respect to the seawater. In contrast, the VMCM the sampling period, which is about 14 s every 240 s and PAR sensor ground to the seawater via their reduces the power requirements of this sensor to 0.64 pressure cases. To isolate the DO sensor from the mA-h. The standard EG&G VMCM turns the A/D VMCM during the 2 s every 4 min that the DO board on and off using a control line. We used that sensor was actively sampling, there is a circuit line to gate the solid state power switches that feed consisting of two relays which switch the DO sensor the above mentioned sensors. A simple modification to internal power and isolate electrical grounds for 2 s to the firmware was performed in order to have this whenever external power is sensed to come on. After control line turn on and off at the appropiate times 2 s, electrical connection with external power and giving the sensors enough time to settle their outputs. common ground is re-established so that the analog DO and DO temperature voltages could be sampled by Using the above scheme for the DO sensor, the the VMCM. This circuit proved to be unreliable and fluorometer, and the analog output buffers, the power was the primary cause of low data return by this budget of the instrument is the following: sensor. A better approach for the future would be to INSTRUMENT DUTY CYCLE VOLTAGE STD BY ACTIVE ENERGY CURRENT CURRENT (4 MONTHS) % V mA mA A-h VMCM electronics 100 6 4.5 3 9.1 Signal Processing Bd. 5 6 6 12 19.3 Tape Recorder 5 14 - 350 4.2 Conductivity 100 14 13 - 37.4 PAR 100 14 1 2.8 Dissolved Oxygen 5 14 - 11 1.9 Transmissometer 100 14 10 - 28.8 Fluorometer -0.4 14(*) - 2.5 A 23.7 Separate 14 V supply for the fluorometer. The battery pack was composed of lithium cells float the VMCM and PAR pressure cases or rind some (Electrochem DD size Model BCX72-3B76), since the other means of isolating the DO sensor. alkaline types available did not have enough capacity to supply the external sensors during a four month 2.5.3 Adjustable input range for PAR sensor deployment. The mean value of the output signal from tha The cells were distributed as follows: PAR sensor is highly related with the depth. In order to have a good dynamic range not using logarithmic 6 Volts 60 A-h for the VMCM and S.P. board)14 amplifiers (rejected for their problems with Volts 30 A-h for the fluorormeter)14 Volts 90 A-h temperature), the anti-aJiassing filter used for the for the rest of sensors and tape recorder. PAR signal is designed to have three different switch-selectable gains. This permits us to field set 2.5 Special interfacing requirements the gains according to the instrument depth or the signal range for that deployment site. 2.5.1 Conductivity F/V conversion 3. DESCRIPTION OF THE MVMS A Sea-Bird SBE-4 instrument was used as a conductivity sensor. Its output is a frequency signal. 3.1 Signal processing board The MVMS is prepared to measure analog voltages, therefore a frequency to voltage conversion is necessary The signal processing board contains three and is performed as part of the signal processing very-low-cutoff frequency, first-order, low-paw 1183 filters, a low-power frequency- to-voltage converter, indefinitely those values to obtain any large value of and several solid state switches. The solid state time constant. However, there are practical limitations. switches are used to turn on and off the fluorometer For the capacitor, there will be a maximum size, and (which requires switching 14 volts at 2.5 A) and the also the leakage currents which am proportional to power fine that feeds the DO sensor (14 volts at 12 their capacitance and produce undesirable unstabilities mA). Both swicthes are gated by the same control line in the output. For the resistor, large values are that turns on and off the A/D board of the EG&G undesired because the bias currents needed for the VMCM. The frequency signal from the conductivity operational amplifier will produce a very large (and sensor is transformed into a voltage signal using the temperature dependent) voltage offset at the output. frequency- to-voltage converter. After the conversion, The solution suggested to obtain very large time -the signal is passed through a low-pass filter with a constants consists of using Miller's Theorem to time constant of 240 s to equate the sensor res ase the effective capacitance of a capacitor. The Tonse incre to the temperature sensor. The transmissometer and technique is implemented using a simple inverter PAR sensor signals are passed through filters with a amplifier circuit as shown in Fig. 3, where a capacitor 600 seconds time constant to reduce aliasing caused by the vertical movement of the instrument through the C water column due to the mooring oscillation. Power consumption requirements make neccessary the use of very low power operational amplifiers. I NPUT -A OUTPUT A 3.1.1 Voltage-to-frequency converter L4 The converter is based in a switched capacitor C ef = (I +A) C technique described in the Linear Technology switched capacitor technical literature. It based on charging a Fig. 3 Block diagram of low-pass filter implementing capacitor to a precise reference voltage level, then Miller effect. tranferring that charge to the input of a low-pass filter. The frequency signal drives the charge transfer has been put in parallel with the output and the input from the capacitor to the filter, the higher the of the amplifier. frequency the laxger the charge transferred. Since the filter output is proportional to the average current The effective capacitance (with respect to provided by the capacitor this voltage will also be ground) seen at the input of the amplifier is the ratio proportional to the frequency that switches the charge. between the differential increment in current and the This technique yields a very linear transfer function differential increment in voltage with respect to ground and has low power consumption. The circuit shown in when a given differential voltage increment is applied Fig. 2 consumes 1.2 mA and the frequency range is to that input. Since the amplifier will pull the voltage from 0 to 13 Khz. up and down at the other end of the capacitor, the current through the capacitor is proportional to the voltage difference across it. This current is proportional to the gain of the inverter and thus the effective capacitance is proportional to the gain. aioF IAS Therefore, the effective capacitance can be increased, Ur-10" increasing the gain and without increasing the real a valu e of the capacitor. 17 a The formulation that shows quantitatively the capacitance change is as follows. For a given dVi: Fig. 2 Scbematic diagram of low-power V/F circuit for conductivity sensor. Ceff= dl/dVi 3.1.2 Low-pass filters and dl (dVi-dVo) C where C is the real capacitance The fact that the three filters plus the voltage converter and switches should fit in a 3 by 4 in. also dVo -A * dVi where A is the Gain of circuit board made impossible the use of conventional the inverter low-pass filter designs which need large capacitors to achieve large time constants. In standard first-order replacing it in the second equation results in: low-pass filter designs, the value of the time constant is given by the product RC, where R is the resistance dl = (dVi - (-A*dVo)) * C and C the capacitance of two critical components of the circuit. It would be very simple to increase dl = (I+A) * dVi 1184 that is Ceff = (I+A) * C 3.3 Modifications to the A/D board The main limitation of this scheme is that the The standard EG&G VMCM has an A/D board dynamic input range is decreased proportionally to the used to convert the temperature sensor signal, an increase in effective capacitance. To solve this auxiliary voltage input and the current consumption of inconvenience, an attenuator prior to filtering is eight interface boards. The eight current consumption neccessary and an amplificqation after the filtering channels are grouped in one of two eight-channel restores the dynamic range of the signal. The final multiplexers. . The current measurement is circuit for the low-pass filter is similar to a simple accomplished using a resistor network pulling down the "RC" configuration except that a very large power supply and feeding the corresponding interfaces, capacitance is artificially achieved through the use of To make these eight channels available for external Miller's effect. It is interesting to note, that the measurements, this resistor network is replaced by a amplification stage used for increasing the effective half socket connector which is used as the signal input capacitance, also serves to amplify the filtered signal, for the signals from the Signal Processing board. In restoring the dynainic range previous to . the order to have appropiate gain values, some resistor attenuation section. Figure 4 shows the final circuit values need to be changed. diagram for the frequency to voltage converter and the 3.4 Modifications to the firmware three low-pass filters. The first stage of the filter is an inverting amplifier with a gain of 0.45 for the The original firmware of the EG&G VMCM had conductivity channel and 0.1 for the PAR and beam to be changed to allow extra time for some of the transmission channnels. The second stage is a high sensors to stabilize their output before being sampled. input impedance non-inverting amplifier with a gain Changing the contents of 456EH (EPROM U3) on the of 1. This stage serves as a buffer to avoid memory board from 06H to FFH analog power control deteriorating the "RC" of the final stage. The 4.75 signal 11VSW" on time was stretched to 12.7 seconds. MOhms resistor and the 10 uF metallized polyester The following list shows the current meter (CM) capacitor in the final stage of the filter form the channels in the order in which they were sampled, the RC(I+A) time constant. With the above values, and using above equations the resultant time constant of time relative to power-on that they were sampled and the circuit is about 550 s. the sensor assigned to that channel. 27t -V XK 17 WK CM Sampling Time (see) Sensor 0.7 Battery Voltage 7 2.1 (unused) 5 3.5 Transmissometer QX swi 4 4.9 Conductivity I 2=1 2 6.4 (unused) 2 2 8 7.8 Fluorometer 6 9.2 DO temperature 3 10.6 DO Ilk I t5K f." M= Note that the fluo 'rometer and DO sensors, which both 14 have stabilization times of 7 seconds, were sampled 7 last. 1K 10S. ISK 4.75 ma I 7 2, _19 4. EVALUATION OF PERFORMANCE non 4.1 Sample data set Fig. 5 shows a representative data set from the second deployment for the instrument located at 10 in. This deployment spanned the period from mid May to mid August 1987. The data indicates that fouling was not a problem with the PAR, fluorescence or Do sensors. The anti-foulant treatment on the Fig. 4 Schematic diagram of signal processing board transmissometers was not as successful. In one case a circuit including: V/F, three low-pass filters barnacle was observed growing in the light path. The and power toggling transistors. data show a steady decline in transmission with time puncuated by occasional abrupt increases which are 1185 @P- 0 p tz@ :n H 0 p tl @ = r, @@ v -C@ ow - 0 r- 9: ra@ to p p p P CD 'A 0-1 0 CD P. oq CP --1 -4 a 00 00 -4 0 cim 01 CZ 00 00 t") rA 2 W on 0 0 @;: % OD C@ M -4 qD 0 CD CD tZ, !@* , J6 -4 -4 -4 -4 to eb to 9) oq @-h rt @o tj m 0 0 m 0 oq " ct rt P. W @31 0 N CL CD 0 0 m C. o cn -1 w w w 0 0 cn 2; . 4 rt w t7l a. X r, - F@ 0 W tj 0 5z, so oQ @j PW3 H. 0 UQ w CrQ @r H 00 w 0 ol "0 C) 4 rD 9) P3 0 IV (D rt = rt F@ @D a, 0 00 @-f = D) -i F-4 Mm In Oq F,*,, ID. > 0 0 rt tl 0 (D gr 03 = @:r M @h o (a -.@ 0 (D 0 to 0 M go w rt H rr M M o CA @l v m W P- (D g CA Oq m t-' OQ C:) 0 *.4 0 go M W aq tv :r El 0 2-4 ci rt I-j. El .4 0 9L - CA r, En (D 4 0 m p m o It @i = 0-0 0 03 H. 0 ts (A o M :3 P. @:$ CA n 8 CL GQ N 11 8 0 m rt im EL 0 0 ::1- rt E3 -O'Q o rA < 0 a- 0 0, 0 aq =-. 0 r- -- - 0. Im tm-- I 0 G@ 9 21 0, r. -0 V :@o CD o m = (D in. n 0'0 0-0 tr - 0 er-0m 0 0 D0 4:4 1 (M M aq w o 0-5 -4 t=l N !0 t:l 0 111 p p U2 p p tv, 0 0 M -0 0 W 0 0 t to t- C2 Ji, 0 c+ Im" v m 9L $0. 0 m processing techniques in conjuction with widespread power switching to realize longer deployment times. Attention will also have to be paid to longer acting anti-foulants for the optical instruments. 6. ACKNOWLEDGEMENTS The conversion of the EG&G VMCM into a MVMS instrument was funded by the Office of Naval Research contract no. N00014-84-C-0132 under the Biowatt II program. The Biowatt Mooring Group consisted of principal investigators and technical support staff at Lamont-Doherty Geological Observatory (LDGO) and the University of Southern California (USC). At LDGO the members were: I. Bitte, C. Langdon, I Luther, M. Maccio, J. Marra, and L. Sullivan. At USC: A. Bratcovich, T. Dickey, D. Manov and J. Scott. 7. REFERENCES Dickey, T., E. Hartwig and J. Marra. (1986). The Biowatt bio-optical and physical moored measurement program. EOS Trans. AGU. 67:650-651. Langdon, C. (1984). Dissolved oxygen monitoring system using a pulsed electrode: design, performance and evaluation. Deep-Sea Res. 31:1357-1367. Maccio, M., C. Langdon and J. Basna. (1986). Signal Processing Board for VMCM's. Manual. Lamont-Doherty Geological Observatory Technical Report Marra, J. and E.O. Hartwig. (1984). Biowatt: A study of bioluminescence and optical variability in the sea. EOS Trans. AGU. 65:732-733. Weller, R.A., and R.E. Davis. (1980). A vector measuring current meter. Deep-Sea Res. 27: 565-581. 1187 TRIBUTYLTIN AND MUSSEL GROWTH IN SAN DIEGO BAY Michael H. Salazar and Sandra M. Salazar Naval Ocean Systems Center Environmental sciences, Code 522 San Diego, California 92152 ABSTRACT therefore not environmentally realistic (11). Third, growth rate may be affected by JTBTJ in During three San Diego Bay field tests (1987- tissues. Previous field studies have shown that 1988), juvenile mussels (Mytilus edulis) were from 60 to 112 days may be necessary for TBT to exposed to mean ambient tributyltin (TBT) reach equilibrium in tissues of mussels and concentrations from 7 to 500 ng/I for 12 weeks. oysters, respectively (12, 13). Mussel lengths and weights, seawater [TBT] and various physical/chemical parameters were measured Stephenson et al. (7) avoided some of these weekly. TBT bioaccumulation was measured at the end problems by conducting a 3-month field study of of each test. At mean seawater JTBTJ L200 ng/l juvenile mussel growth in San Diego Bay and significant reductions in mussel growth rates were reported significant growth reductions for observed. Other environmental factors appear to transplanted mussels in the Shelter Island yacht modify the effects of TBT on juvenile mussel growth basin at 230 ng/l TBT. However, TBT measurements below 100 ng/l TBT. No clear relationship between were extremely limited, and as in most field growth rate and [TBT] in mussel tissues at studies, the control site selection and subsequent concentrations below 1.5 ug TBT/g tissue was found. comparisons were misleading. Reductions in mussel Although there appears to be a linear relationship growth rates attributed to TBT exposure may have between TBT bioaccumulation and seawater [TBT), the resulted from unmeasured stress factors (11, 14). relationship between bioconcentration factor and Most field assessments of TBT effects on bivalve seawater [TBT] appears to be inverse and growth (7, 15, 16, 17t 18) have not adequately exponential. No dose-related mortality was defined natural TBT variability or measured other observed. stress-inducing physical/chemical factors that could affect growth. Site-specific, flow-through bioassays INTRODUCTION potentially provide more meaningful results because they combine the advantages of controlled Mussel growth can be an effective and dynamic laboratory dosing with realistic field test indicator of environmental stress because it conditions. In our previous PETS experiments in represents an integrated response of biological San Diego Bay, juvenile mussel growth was processes to the environment (1). The effects of identified as the most sensitive TBT bioindicator tributyltin (TBT) on mussel growth have been tested, but the results were conflicting (1, 12). studied in the laboratory (2, 3, 4, 5, 6), the In one test, no effects occurred after 56 days at field (7) and in a Portable Environmental Test 160 ng/l TBT. In another, significant reductions System (PETS) for a site-specific, flow-through in mussel growth were found after 63 days at 70 bioassay (1). In laboratory tests with juvenile ng/l TBT. TBT effects were overestimated due to mussels, significant reductions in growth have been system-induced stress and the inability to provide reported at 240 and 400 ng/l TBT (2, 3, 4, 6) for realistic conditions for juvenile mussel growth tests lasting 49 and 7 days, respectively. In (12). another test with adults, significant reductions were reported at 310 ng/l TBT after a 66-day exposure period (5). Important issues in predicting environmental impact remain unresolved due to inconclusive These tests have limited environmental results from laboratory, field and site-specific significance for several reasons. First, high TBT studies. Part of this deficiency can be attributed concentrations ([TBTJ) are only characteristic of to the paucity of bivalve growth measurements at enclosed basins with poor tidal exchange and many TBT exposures <100 ng/l that are characteristic of organotin-painted vessels (8, 9, 10). Most harbor most harbor environments. We conducted extensive environments have much lower [TBTJ and field studies to monitor weekly changes in seawater environmental impact is more difficult to assess. [TBTJ and juvenile mussel growth rates to clarify Second, mussel growth under laboratory conditions some of these issues. Since our previous work is significantly different than field growth and has suggested other environmental factors can 1188 United States Government work not protected by copyright .................-........... ............ .............. .................. modify TBT effects in juvenile mussel growth, HARBOR..:@::::@::::::::::::::::@@:::@::::::::::: .............. ISLAND . . . . ...... ......... ...... ::2::::::::::::: ..............I........... .......................... -:: .............................................. ........ ........ various physical/chemical factors were monitored at =R111111111111111111111 ..........................I................ each site to assess their separate and combined ............. ................... .......... ..... ..... ........................ effects on growth. This report summarizes the HIW ................................... ........... results of these studies and discusses their ..................... ............. ..................... .................... ................................................ si significance in relation to previous work.. ............................... ............... si ................................................ 0SC, SC ........... METHODS AND MATERIALS .............. ........................... Three separate tests were conducted over a 12- ................... month period at 10 different sites in San Diego Bay ....................... .................. ..... ............... (Figure 1). Test I lasted 18 weeks (August to December 1987). Tests II and III lasted 12 weeks; EOD .. . ........... NAV January to April 1988, and April to July 1988, %SURV respectively. All mussels (Mytilus edulis were SITE TEST TESITV%TEST collected from a large population at a relatively SUB (lost) 83 89 clean site near the mouth of San Diego Bay (Figure 1) where [TBT1 in seawater and mussel tissues were NOSC 94 83 100 low (-30 ng/l TBT, -0. 12 ug TBT/g tissue, NCSCD 94 SI 89 94 94 respectively). Animals were initially selected by 94- --i-00 SID length. Each test began with 18 animals (10-12 mm SAIL 100 CAYS 78 in length, X -= 11.0 mm) per tray. All test HIW 56 so mussels were suspended from floating piers and EOD 94 100 89 continuously submerged 1 In below the surface or 3 NAV 89 100 100 m below the surface at "deep" sites. Whole animal CAYS 100 100 wet weights and lengths, seawater [TBT], chlorophyll-a, temperature, salinity, dissolved oxygen and pH were measured weekly. During Tests 1 2 3 n II and III, temperature measurements were made NAUTICAL MILES continuously at 30-minute intervals with a field- deployed, battery-operated Tempmentor (Ryan Figure 1. Percent mussel survival (n-18) at San Instruments). TBT accumulated in mussel tissues Diego Bay test sites for Test 1, 11 and III. was.measured at the end of each test. Copper was only measured in mussel tissues from selected sites in Test III. Each test was limited to 12 weeks to minimize the effects of gametogenesis on growth measured by AA after acid digestion. Chlorophyll-a rate. was measured by fluorometry after cold extraction (21). Ten different sites were monitored (Figure 1): For each site, cumulative percent increases in six Navy sites and four marina sites. The six Navy lengths and weights were calculated to normalize sites were selected for low [TBT] (8) and varying size effects and estimate relative growth rates for levels of boating activity. Subase San Diego (SUB) graphic presentation. Statistical analyses were and Naval Station San Diego (NAV) had many large conducted only on survivor data. Serial ANOVAs (p vessels. The Explosive Ordnance Disposal Pier (EOD) < 0.05) were performed on weight and length data at had the fewest boats and was the test site in each sampling interval to test for differences previous PETS experiments (1, 12). Navy Sailing among sites. If differences were detected, Club Recreational Marina (SAIL) and Naval Ocean Duncan's new multiple-range test was used to Systems Center (NOSC) were selected for proximity determine which sites were similar. A series of to yacht basins with high TBT levels. NOSCD was linear regression analyses were performed to approximately 3 m below the surface at NOSC. The compare slopes of calculated growth rates. Slopes four marina sites in three yacht basins were differing by more than two standard errors (p < selected for high seawater [TBTj (8). The yacht 0.05), were considered significantly different. basins studied were Shelter Island (SI), Harbor Island (HIW), and Coronado Cays (CAYS). SID was Although changes in mussel weights and lengths approximately 3 m below the surface at SI. Deep were used to assess TBT effects on growth, weight sites were included because previous measurements measurements appeared to be more accurate and had shown that [TBTj was lower near the bottom sensitive indicators. The relationship between (19). weight changes and [TBTj in seawater and tissues will be emphasized here. Length data are included TBT concentrations in seawater and mussel. for comparisons with other work. tissues were measured according to Stallard et al. (20). For seawater, purge and trap hydride RESULTS derivatization was followed by atomic absorption spectrophotometry. For tissues, acid/solvent Percent survival by site and test are shown in extraction was followed by grignard derivatization Figure 1. Survival was generally high at all sites and . gas chromatography with flame photometric except HIW, where it was very low (50-56%). Figure detection. All TBT measurements are reported to two 2 includes relative growth estimates from changes significant digits. Copper in mussel tissues was in mussel weights, calculated mean growth rates . 1189 1500 TEST I (Aug-Oct 1987) Seawater Tissue Growth Rate X[TBT] X[TBT] 1000- Site (mg/wk) (mm/wk) (ng/t) Wg/g) BCF NOSC (0) 253 1.42 80 1.4 17,500 EOD (0) 246 1.43 10 0.32 32,000 NAV (A) 175 1.22 8.4 0.37 44,000 500 SUB (0) - - 19 - - S1 (T) 17 0.21 530 4.5 8,500 m_j 0 1200 TEST 11 L) (Jan-Apr 1988) SAIL (V) 158 1.20 56 0.33 .5,900 EOD (0) 141 1.22 11 0.08 7,300 NAV (A) 134 1.19 9.0 0.10 11,100 800 SUB (11) 132 1.10 18 0.23 12,800 NOSC (0) 91 0.94 59 0.54 9,200 HIW W 95 0.92 130 0.69 5,300 CAYS (A) 78 0.83 80 0.75 9,400 400 SID (0) 55 0.66 200 1.0 5,000 Sl (V) 17 0.31 360 1.7 4,700 __j 0 1'00 TEST III (Apr-Jul 1988) SAIL W) 246 1.62 28 0.53 18,900 NAV (A) 205 1.54 6.5 0.12 18,500 NOSCDW 204 1.44 15 0.45 30,000 1000 EOD (N) 197 1.39 7.0 0.12 17,100 SUB (0) 180 1.32 10 0.34 34,000 NOSC (0) 178 1.29 27 0.79 29,300 HIW W 157 1.25 72 0.56 7,800 500 CAYS (A) 153 1.17 42 0.78 18,600 SID (0) 147 1.15 64 1.1 17,200 ISI (V)j 70 0.69 1 200 2.3 11,500 5 2 4 6 8 10 12 TIME (weeks) Figure 2. Mussel growth expressed as cumulative % change in weight for Test 1, 11, 111. Mean mussel growth rates (mg/wk, mmlwk) paired with mean TBT seawater concentrations (ng1l), final TBT concentrations in tissues (ug1g) and calculated bloconcentration factors (BCF). 1190 (mg/wk and mm/wk), mean [TBT] in seawater (ng/1) 10:0 and tissues (ug TBT/g tissue wet weight), and bioconcentration factors (BCF) (calculated as the 9 O@ o NO ratio of TBT in tissue to TBT in seawater) for z; 800- Tests I, II and III. Test I values only include - data for the first 12 weeks for comparisons with 700- Tests II and III. Since tissue [TBT] was measured cc W 600- at the end of the test, the values are 18-week I'- 4 means. Growth rates (� 95% CL) are shown in Figure 3: 500- 3. Figures 2 and 3 show that there were distinct 4 UJ differences among sites in each test and that some 0) 400- growth rates in each test were significantly z 300- different. 200- Copper concentrations in the tissues of 100 - mussels from SAIL, EOD, NOSC, SID and SI were 3.3, 3.4, 3.4, 9.9 and 21.4 ug Cu/g tissue wet weight, 0 ---------- 1--13 respectively. There was little variability among -7@--28 2 '20 ' Test III sites in salinity, dissolved oxygen and pH, so AUG Test I OCT JAN Te;t it APR JUL these factors will not be presented or discussed. 87 87 88 88 88 Means, minima and maxima for [TBT] in seawater, temperature and chlorophyll-a are given in Table 1. Figure 4. Natural weekl3r variability in seawater The highest [TBT1 with the most extreme natural [TBT] for Tests 1, 11, 111. variability was measured in the vicinity of marinas and is presented in Figure 4. Most test sites had lower and less variable [TBT]. [TBT1 decreased at The graphs in Figure 5 were generated to most sites, particularly SI, from Test I to Test clarify some of the relationships suggested by the Iii. data in Figure 2. A comparison of mussel growth rate and seawater [TBT] (Figure 5A) suggests no clear relationship below 100 ng/l TBT. At 80 ng/l TBT growth rates ranged from 78 to 253 mg/wk. At 200 ng/l TBT and above growth rates were much OSI 192 lower. A. comparison of growth rate and [TBT] in mussel tissues (Figure 5B) shows that there is no 1 0 NAV 192 clear relationship below 1.5 ug TBT/g tissue. U) Between I and 1.5 ug TBT/g tissue, mussel growth LU t-@ EOD 204 rates varied from 64 to 253 mg/wk. Growth rates NOSC F-41-A 204 were much lower above 1.5 ug TBT/gtissue. 0SI 204 There appears to be a linear relationship WSID 204 between [TBT] in mussel tissues,and mean [TBT] in HN CAYS 216 seawater (Figure 5C). However, the relationship between mussel BCF and [TBT1 in seawater (Figure @-*ANOSC 180 5D) appears to be inverse and exponential. The F-0-1 H IW 120 relationship between BCF and mussel growth rate is LU 1,-@ SUB 180 also exponential (Figure 5E). The highest BCF's 1-@ NAV 216 are all associated with growth rates above 170 1--@EOD 216 mg/wk. Nearly all sites with the slowest growing SAIL 1L68 mussels were associated with BCF's below 15,000, I while the majority of fastest growers were 204 associated with BCF's above 15,000. F-*-A S I D 216 CAYS 216 DISCUS SION F-*-iHIW 108 The results of three 12-week field studies (n 0 NOSC 216 LU indicate that TBT and other highly variable natural SUB 192 factors affect juvenile mussel growth. Although 1--@ E01D 192 the lowest growth rates were at the highest [TBT1 !--@ NOSCID 204 C>200 ng/1), all other growth rates were not well [-*--I NAV 216 correlated with [TBT]. Growth rates were highest SAIL 1 216 1 in the summer when both temperature and 0 100 200 300 chlorophyll-a were high. There was no correlation GROWTH RATE (mg/wk) between [TBT1 and survival, even at high [TBT). Two important questions to be answered are 1) at what concentration does TBT become the dominant factor affecting growth and bioaccumulation, and 2) Figure 3. Mean mussel growth rates (mg/wk) by site what other factors modify mussel growth rates in for Test I, II, III. Error bars indicate 959 San Diego Bay? The literature indicates that confidence limits L+2 standard errors). N temperature and nutrition are the most important number of measurements. [ o NOSC V S, 0 SID in 1191 Table 1. Means, minima, maxima for [TBT] in seawater, temperature and chlorophyll-a. SUB NOSC NOSCD SI SID EOD SAIL NAV HIW CAYS [TBT1 (ng/0 Test I mean 19 so 530 10 8.4 min 9.0 15 270 1.0 2.8 Max 48 210 860 18 12 Test II Mean 18 59 360 200 11 56 9.0 130 80 Min 6.0 18 150 36 4.9 27 3.8 20 22 Max 47 160 940 620 36 100 16 320 210 Test III Mean 10 27 15 200 64 7.0 28 6.5 72 42 Min 4.4 8.0 3.4 130 17 3.8 9.7 2.3 36 18 Max 19 70 42 430 160 10 75 12 140 140 TEMPERATURE 00 Test I Mean 19.6 19.8 21.1 22.7 23.5 Min 18.1 18.4 19.0 20.9 21.4 Max 21.5 20.7 22.9 25.1 25.0 Test II Mean 15.5 16.0 16.5 17.3 17.0 16.2 17.1 16.4 17.9 Min 12.8 13.0 13.4 16.2 14.1 13.1 13.4 13.2 12.8 Max 19.6 19.8 20.2 19.1 21.8 19.9 21.4 20.3 23.5 Test III Mean 17.0 17.7 17.3 18.9 17.4 20.4 18.7 20.8 19.2 22.5 Min 13.0 13.9 13.4 16.0 14.6 16.8 15.3 18.3 16.7 18.8 Max 21.1 21.5 21.8 22.7 18.8 23.9 22.2 24.0 22.6 25.8 CHLOROPHYLL-A (Ug/0 Test I Mean 2.10 2.-75 2.95 1.78 2.05 Min 0.76 1.53 2.04 1.23 1.29 Max 3.44 3.45 4.24 2.45 3.08 Test II Mean 2.33 2.37 1.61 3.01 2.01 1.90 1.92 2.31 2.28 Min 1.22 1.49 1.10 .1.82 1.07 1.04 1.11 1.32 1.12 Max 6.39 5.41 2.31 4.89 3.29 3.35 2.90 .5.41 4.87 Test III Mean 3.89 3.27 3.68 2.36 5.43 1.52 2.54 1.40 2.48 1.74 min 1.50 1.46 1.12 1.60 1.89 0.79 1.53 0.60 1.57 0.95 Max 9.92 5.86 6.72 3.50 13.16 2.08 4.41 2.49 3.56 3.17 factors controlling mussel growth and that mussel growth (1). However, the highest growth temperature,,nutrition and mussel grdwth have a rates were not at the lowest [TBTJ. Mussels at high degree of temporal and spatial variability NOSC, exposed to a mean of 80 ng/l TBT, grew faster (22, 23). Site and seasonal differences in these than all others. Growth rates of mussels exposed factors seem to be important in the present study. to 80 ng/l TBT at NOSC and other sites ranged from 78 to 253 mg/wk. Below 100 ng/l TBT the potential Effects on Growth Rates TBT effects on juvenile mussel growth may be reduced by environmental factors that enhance The lowest growth rates were measured at the growth or amplified by factors that retard growth. highest [TBTJ in a *11 three tests (Figure 5A). The This may explain the wide range of mussel growth four lowest growth rates were from the Shelter rates measured in this study. Island yacht basin, characterized by very poor tidal flushing and many other factors that affect 1192 5.0- 240- SIO 4.0- ul 0 *SID In 3.0- LU M 120- W -'SI 2.0- z 0 0SI cc 0S] 0 60 *SID 1.0. eSID *SID *SI 6SI 0 i 0.0 0 260 460 600 0 200 400 600 [TBT1 IN SEAWATER (ng/9) [TST] IN SEAWATER (ng/9) 01 240- B LL D L) 40,000 - 3 0 '6180- % U 30,000 - U. LU *SID z 2 120- 1: 20,000 - cc z 0 Ui cc U SI 4SI z 10,000- 60 SID 1 0 0SI *SID 0 SID 6SI 2 0SI S11 ca 0 0 0 200 40'0 660 0.0 1.0 2eO 3.0 4.0 [TBT1 IN SEAWATER (ng/E) [TBT] IN TISSUES (ug/g) 240- E At concentrations >200 ng/l TBT there is a more obvious reduction in growth. This was -11180- previously suggested by Thain and Waldock in a laboratory study (2, 4) and Stephenson et al. in a E field study (7). Stephenson et al. reported LU 0 significant reductions in mussel growth after 150 <120- days exposure to 230 ng/l in Shelter Island. X - Although their data are consistent with our 3: - *SID findings of decreased mussel growth at >200 ng/l est 0 TBT, their results are misleading because of poor W60- *SID experimental controls and limited TBT measurements 0 - *SI OSI 01 Thain and Waldock (2, 4) reported much lower 6 10,60 20,000 30,0'00 40,000 growth rates in the laboratory than in the present BIOCONCENTRATION FACTOR (BCF) field study at similar mean TBT concentrations and exposure periods. At SI in Test III, mussels increased in weight approximately one order of Figure 5. magnitude more than their laboratory control A) Mussel growth rate and seawater [TBT], animals (220 vs 20%). SI mussels in Test I were B) Mussel growth rate and bioaccumulation. C) Mussel bloaccumulation and seawater [TBT]. exposed to TBT concentrations an order of magnitude D) Bioconcentration factor and seawater [TBT]. higher than laboratory animals (530 vs 50 ng/1) E) Mussel growth rate and bioconcentration factor. and increased in weight by about the same amount (70 vs 65X). Only Shelter Island sites are indicated (SI, SID). The vertical dashed lines are for emphasis only. 1193 In previous PETS bioassays at the EOD site Previous flow-through bioassays (12), Hawaii (12), we found significant reductions in mussel microcosm studies (Seligman, 1988, pers. comm.), growth rate at 70 - 80 ng/l TBT, concentrations San Diego Bay monitoring studies (Dooley, 1988, much lower than reported in other laboratory or pers. comm.) and the present three field studies field studies (2, 3@ 4t 5t 67 7), but the bioassay have shown a similar relationship between bivalve results were conflicting. This inconsistency BCF1s and seawater [TBT]. Results from the suggested that TBT effects on juvenile mussel present study suggest the BCF is not a very precise growth were artificially enhanced and overestimated predictor of [TBT] in mussel tissue since other due to stress on the test animals (1). The present factors can affect growth rates and TBT study clearly shows that 80 ng/l TBT does not bioaccumulation. It also casts doubt on the affect mussel growth unless the animals are utility of extrapolating environmental impact from stressed by other factors. tissue levels (4, 9, 16), extrapolating seawater TBT concentrations from tissue levels (13, 33, 347) TBT Bioaccumulation or using field data to verify laboratory data (16, 17). The early laboratory work suggests bivalve growth rates are directly affected by accumulated Natural Variability TBT (3, 17, 24). Figure 5B shows that the lowest growth rates are associated with the highest JTBTJ There is a wide range of mussel growth rates in mussel tissues and that below 1.5 ug TBT/g in San Diego Bay and a wide range of factors tissue there is no clear correlation between TBT in affecting growth. Daily maximum temperature mussel tissues and mussel growth rates. Test I changes of up to 50C and overall temperature ranges mussels at NOSC accumulated 1.4 ug TBT/g tissue. of over 80C were measured. There were order-of- Excluding the Shelter Island yacht basinp this is magnitude shifts in seawater [TBT1 and chlorophyll- the highest tissue level measured in San Diego Bay a concentrations at some sites (Table 1). Previous mussels (Seligman, 1988, pers. comm.). Yet, the measurements have shown that [TBT1 near the bottom growth rate for these mussels (253 mg/wk) was the are significantly lower than near the surface (19), highest measured in all tests. It appears that and tidal fluctuations in [TBT] vary by a factor of high tissue levels of TBT (1.4 ug TBT/g tissue) do 20 near Shelter Island (35, 36). Water bodies not reduce mussel growth rate unless the animals cannot be adequately characterized without are stressed by other factors. accounting for this kind of spatial and temporal variability (37). Davies et al. (13) have shown that TBT accumulation by oysters in the field is generally The site with significantly lower growth rates higher than in the laboratory and it takes longer In all three tests was Shelter Island. Growth to reach equilibrium. Laughlin et al. (25) suggest rates at SID were significantly higher in each similar differences in accumulation of TBT by test. Juvenile mussels near the bottom in Shelter mussels based on a laboratory study. They also Island grew significantly faster than those near report a significant difference between measured the surface. They also had less TBT and copper in BCF1s (-5,000) and the BCF predicted (260) by TBT their tissues. Mussel growth rates and TBT in partitioning coefficients. Recently, it has been mussel tissues generally decreased between Tests I suggested that oysters may have BCF1s near 500,000 and II and increased between Tests II and III, (Huggett, 1988, pers. comm.). In the present while [TBT1 in seawater generally decreased between study, most mussels with growth rates above 170 Test I and III. Higher growth rates are generally mg/wk had BCF1s above 15,000. This suggests a attributed to higher temperatures and more relationship between mussel growth rate and phytoplankton (22, 23). If phytoplankton increase bioaccumulation. The present study clearly shows TBT bioavailability to mussels (25), and higher that mussel BCF1s below 10,000 can result from temperatures increase filtration.rates (22) mussels field exposures as low as 11 ng/l TBT or as high as will accumulate more TBT at higher phytoplankton 530 ng/l TBT. These differences are attributable to concentrations and higher temperatures. stressed animals. In the laboratory, stress can be caused by maintenance under unnatural conditions In Test I summer temperatures at some sites (26, 27, 28). In the field this stress could be reached 250C, which may reduce growth because they attributed to TBT, other contaminants or approach levels that adversely affect mussel physical/chemical factors (I). physiology (38). However, Page and Hubbard (39) suggested that water temperature is not a Kiorboe et al. (29) suggested growth rates in significant factor in regulating mussel growth in optimum laboratory studies do not approach growth California waters. San Diego Bay is very near the rates in the field primarily because M. edulis southern limit of the range for M. edulis (40) and derives additional nutrition from sus@e_nded high temperatures may be a limiting factor at some particulates. They may also accumulate additional sites. The highest growth rates were not found at TBT from these suspended particulates. Laboratory the warmest sites. This implies that San Diego Bay studies have shown that TBT rapidly binds to mussels may have adapted to higher temperatures or suspended sediment particles, algal cells and potentially higher growth rates were reduced by bacterial cells (25, 30, 31, 32), and that temperature stress. We did not find significant suspended particulates enhance TBT bioaccumulation temperature or chlorophyll-a effects even though we in mussels (11, 25) and oysters (11, 24). believe they are important factors, as suggested from other work (22, 23). 1194 Copper REFERENCES Field studies provide realism but generally 1. Salazar, M. H. and S. M. Salazar. 1987. TBT lack the control necessary for establishing cause- effects on juvenile mussel growth. In: and-effect relationships, particularly with TBT Proceedings, Oceans 1987 Conference, Halifax, Nova (11). There are many variables other than TBT Scotia, Canada, 28 Sept - 1 Oct 1987, Organotin which may affect mussel growth. For example, Symposium. Vol. 4 pp. 1504-1510. copper concentrations between 3 and 10 ppb have been shown to reduce mussel growth rates in 2. Thain, J. E. and M. J. Waldock. 1985. The laboratory studies (41, 42). Copper concentrations growth of bivalve spat exposed to organotin in Shelter Island between 8 ppb and 12 ppb have leachates from antifouling paints. ICES Paper CM been reported (43, 44). Copper alone could reduce 1985/E: 28 , International Council for the growth in mussels. Shell deformities have also Exploration of the Sea, Copenhagen. been reported at these copper concentrations (45). TBT effects can be overestimated if contaminants 3. Thain, J. E. 1986. Toxicity of TBT to like copper are ignored. None of the field studies bivalves: Effects on reproduction, growth and attributing growth anomalies or depressed growth in survival. In: Proceedings, Oceans 1986 Conference, bivalves to TBT (7, 9, 18) have included Washington, D. C., 23-25 Sept 1986, Organotin measurements of other factors that could affect Symposium, Vol. 4, pp. 1306-1313. growth rates. 4. Thain, J. E. and M. J. Waldock. 1986. The Copper concentrations were more than an order impact of tributyl tin (TBT) antifouling paints on of magnitude higher than measured [TBT] (44). molluscan fisheries. Wat. Sci. Tech. 18:193-202. Measurements from Test III show mussels also accumulated copper in their tissues to levels an 5. Valkirs, A. 0. , B. M. Davidson, and P. F. order of magnitude above TBT tissue levels. Seligman. 1987. Sublethal growth effects and Mussels near the surface not only accumulated more mortality to marine bivalves from long-term TBT, but more copper as well. Observed differences exposure to tributyltin. Chemosphere 16(l):201- in growth at SI and SID could be attributed to a 220. variety of factors, including TBT, copper, temperature or chlorophyll-a. This type of 6. Stromgren, T. and T. Bongard. 1987. The variability is seldom included in laboratory effect of tributyltin oxide on growth of Mytilus bioassays or measured in field tests. edulis. Mar. Poll. Bull. 18(l):30-31. Summary 7. Stephenson, M. D. , D. R. Smith, J. Goetzl, G. Ichikawa and M. Martin. 1986. Growth Environmental impact of TBT cannot be predicted abnormalities in mussels and oysters from areas until TBT levels in seawater and tissues have been with high levels of tributyltin in San Diego Bay. correlated with sublethal responses in nature to In: Proceedings, Oceans 1986 Conference, clearly demonstrate a cause-and-effect Washington D. C., 23-25 Sept 1986, Organotin relationship. An extensive field study was Symposium, Vol. 4, pp. 1246-1251. conducted here but the relationship remains unclear. The results of three 12-week field 8. Grovhoug, J, G., P. F. Seligman, G. Vafa and R. studies suggest that mean seawater [TBT] 2_200 ng/1 L. Fransham. 1986. Baseline measurements of caused significant reductions in mussel growth butyltin in U. S. harbors and estuaries. In: rates. Other environmental factors appear to modify Proceedings, Oceans 1986 Conference, Washington D. the effects of TBT on juvenile mussel growth below C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, 100 ng/l TBT. As in most field studies, absolute pp. 1283-1288. conclusions about TBT impacts cannot be made from these results. However, it has been demonstrated 9. Valdock, M. J. , J. E. Thain and M. E. Waite. that the apparent effects of TBT on juvenile mussel 1987. The distribution and potential toxic effects growth and bioaccumulation are significantly of TBT in UK estuaries during 1986. Appl. different than suggested by previous laboratory and organomet. Chem. 1:287-301. field studies due to other environmental factors that can reduce or amplify TBT effects. 10. Maguire, R. J. 1987. Environmental aspects of tributyltin. Appl. Organomet. Chem. 1:475-498. ACKNOWLEDGMENTS 11. Salazar, M. H. 1986. Environmental We wish to thank Team Hydride (S. Frank, M. significance and interpretation of organotin Kram, M. Stallard, P. Stang, A. Valkirs) for bioassays. In: Proceedings, Oceans 1986 measuring over 300 water samples in support of this Conference, Washington D. C., 23-25 Sept 1986, work. We also thank S. Cola for measuring all the Organotin Symposium, Vol. 4, pp. 1240-1245. tissue samples. Special thanks are due to B. Davidson for technical assistance in all phases of the research. This research was funded by the Naval Facilities Engineering Command. Most of the analytical work was supported by the Office of Chief of Naval Research. 1195 12. Salazar, S. M., B. M. Davidson, M. H. Salazar, 23. Newell, R. C. 1979. Biology of intertidal P. M. Stang and K. Meyers-Schulte. 1987. Field animals. 3rd ed., Marine Ecological Surveys Ltd., assessment of a new site-specific bioassay system. Faversham, Kent, 781 pp. In: Proceedings, Oceans 1987 Conference, Halifax, Nova Scotia, Canada, 28 Sept - 1 Oct 1987, 24. Waldock, M. J. and J. E. Thain. 1983., Shell Organotin symposium, Vol. 4 pp. 1461-1470. thickening in Crassostrea Rigas: organotin antifouling or sediment induced? Mar. Poll. Bull. 13. Davies, I. M. , J. C. McKie and J. D. Paul. 14(11):411-415. 1986. Accumulation of tin and tributyltin from anti-fouling paint by cultivated scallops (Pecten 25., Laughlin, R. B. , Jr. , W. French and H. E. maximus and Pacific oysters (Crassostrea gigas Guard. 1986. Accumulation of bis(tributyltin) Aquaculture 55(2):103-114. oxide by the marine mussel Mytilus edulis. Environ. Sci. Tech. 20:884-890. 14. White, H. H. and M. A. Champ. 1983. The great bioassay hoax, and alternatives. In: R. A. 26. Bayne, B. L. and R. J. Thompson. 1970. Some Conway and W. P. Gulledge (Eds.), Hazardous and physiological consequences of keeping Mytilus Industrial Solid Waste Testing: Second Symposium. edulis in the laboratory. Helgo. Meeres. 20:526- ASTM STP 805. American Society for Testing and 552. Materials. pp. 299-312. 27. Bayne, B. L. , P. A. Gabbott and J. Widdows. 15. Alzieu, C. 1986. TBT detrimental effects on 1975. Some effects of stress in the adult on the oyster culture in France - Evolution since eggs and larvae of 'Mytilus edulis L. J. Mar. Biol. antifouling paint regulation. In: Proceedings, Assoc. U. K. 55:675-689. Oceans 1986 Conference, Washington, D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, pp. 1130- 28. Bayne, B. L. , D. L. Holland, M. N. Moore, D. 1134. M. Lowe and J. Widdows. 1978. Further studies on 16. Waldock, M. J. 1986. TBT in UK estuaries the effects of stress in the adult on the eggs of 1982-86. Evaluation of the environmental problem' Mytilus edulis. J. Mar. Biol. Assoc. U. K. 58:825- In : Proceedings, Oceans 1986 Conference; 841. Washington, D. C., 23-25 Sept 1986, Organotin 29. Kiorboe, T. F. Mohlenberg and 0. Nohr. 1981. Symposium, Vol. 4, pp. 1324-1330. Effect of suspended bottom material on growth and 17. Thain, J. E. , M. J. Waldock and M. E. Waite. energetics in Mytilus edulis. Mar. Biol. 61:283- 1987. Toxicity and degradation studies of 288. tributyltin (TBT) and dibutyltin (DBT) in the 30. Yamada, J., K. Tatsuguchi and T. Watanabe. aquatic environment. In: Proceedings, Oceans 1987 1978. Uptake of triproplytin chloride by Conference, Halifax, Nova Scotia, Canada, 28 Sept - Escherichia coli. Agric. Biol. Chem. 42:1867-1870. 1 Oct 1987, Organotin Symposium, Vol. 4, pp. 1398- 1402. 31. Blair, W. R., G. J. Olson, F. E. Brinckman and W. P. Iverson. 1982. Accumulation and fate of 18. Wolniakowski, K. U., M. D. Stephenson and G. tri-n-butyltin cation in estuarine bacteria. S. Ichikawa. 1987. Tributyltin concentrations and Microb. Ecol. 8:241-251. Pacific oyster deformations in Coos Bay, Oregon. In: Proceedings, Oceans 1987, Halifax, Nova 32. Dooley, C. A. and V. Homer. 1983. Organotin Scotia, Canada, September 28 - October 1, 1987, compounds in the marine environment: Uptake and Organotin Symposium, Vol. 4:1438-1442. sorption behavior. 17. Naval Ocean Systems Center 19. Seligman, P. F. , J. G. .Grovhoug and K. E. Technical Report #9 . Richter. 1986. Measurement of butyltins in San 33. Thain, J. E. , M. J. Waldock and M. Helm. Diego Bay, Ca. In: Proceedings, Oceans 1986 19 8 6 . The e f f e c t o f t r i - buty-1-- t fin-o-n- -th-e- Conference, Washington D. C., 23-25 Sept 1986, reproduction of the oyster Ostrea edulis. ICES Organotin Symposium, Vol. 4, pp. 1289-1296. Paper CM 1986/E:14, International Council for the 20. Stallard, M. 0. , S. Y. Cola and C. A. Dooley. Exploration of the Sea, Copenhagen. 1988. Optimization of butyltin measurements for 34. Bryan, G. W., . P. E. Gibbs, G. R. Burt and L. seawater, tissue and marine sediment samples. In G. Hummerstone. 1987. The effects of tributyltin press: Appl. Organomet. Chem. (TBT) accumulation on adult dog-whelks, Nucella lapillus: long-term field and laboratory 21. Venrick, E. L. and T. L. Hayward. 1984. experiments. J. Mar. Biol. Assoc. U. K. 67(3):525- Determining chlorophyll on the 1984 CALCOFI 544. Surveys. CaICOFI Rep., Vol. 25:74-79. 35. Clavell, C., P. F. Seligman, P. M. Stang. 22. Seed, R. 1976. Ecology. Marine mussels: 1986. Automated analysis of organotin compounds: Their ecology and physiology, pp 13-65. Edited by a method for monitoring butyltins in the marine B. L. Bayne. Cambridge: Cambridge University environment. In: Proceedings, Oceans 1986 Press. Conference, Washington D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, pp. 1152-1154. 1196 36. Stang, P. M. , D. R. Bower and P. F. Seligman. 1988. Vertical and temporal variability of tributyltin in San Diego Bay. (In prep.) 37. Carpenter, J. H. and R. J. Huggett. 1984. Meaningful chemical measurements in the marine environment - transition metals. In: H. H. White (Ed.), Concepts in Marine Pollution Measurements. Maryland Sea Grant Publication, University of Maryland, College Park, pp. 379-403. 38. Bayne, B. L. , D. A. Brown, K. Burns, D. R. Dixon, A. Ivanovici, D. R. Livingstone, D. M. Lowe, N. M. Moore, A. R. D. Stebbing, and J. Viddows. 1985. The effects of stress and pollution on marine animals. Praeger Special Studies, Praeger Scientific, New York. 39. Page, H. M. and D. M. Hubbard. 1987. Temporal and spatial patterns of growth in mussels Mytilus edulis on an offshore platform: relationships to water temperature and food availability. J. Exp. Mar. Biol. Ecol. 111:159- 179. 40. Soot-Ryen, T. 1955, A report on the Family Mytilidae (Pelecypoda). Allan Hancock Pacific Expeditions, Vol. 20(l):261 pp. The University of Southern California Press, Los Angeles. 41. Manley, A. R., L. D. Gruffydd and P. C. Almada-Villela. 1984. The effect of copper and zinc on the shell growth of Mytilus edulis measured by a laser diffraction technique. J. Mar. Biol. Assoc. U K. 64:417-427. 42. Stromgren, T. 1982. Effect of heavy metal (Zn, Hg, Cu, Cd, Pb, Ni) on the length growth of Mytilus edulis. Mar. Biol. 72:69-72. 43. Krett Lane, S. M. 1980. Productivity and diversity of phytoplankton in relation to copper levels in San Diego Bay. Naval Ocean Systems Center Technical Report #533. 44. Johnston, R. K. 198B. The response of marine fouling communities to a pollution gradient in San Diego Bay, CA. Master's Thesis, San Diego State University (in prep;). 45. Sunila, I. and R. Lindstrom. 1985. Survival, growth and shell deformities of copper- and cadmium-exposed mussels (Mytilus edulis L.) in brackish water. Estuar. Coastal. Shelf Sci. 21:555-565. 1197 TRIBUTYLTIN ANALYSES IN ASSOCIATION WITH NOAAIS NATIONAL STATUS AND TRENDS MUSSEL WATCH PROGRAM Terry L. Wade, Bernardo Garcia-Romero and James M. Brooks Department of Oceanography, Texas A6,M University, Geochemical and Environmental Research Group 10 S. Graham Rd., College Station, TX 77840 ABSTRACT concentration, rapidly reaching an equilibrium plateau and slowly depurating". The concentration of butyltins including tributyltin (TBT), dibutyltin (DBT) and In order to evaluate the current status of U.S. monobutyltin (MBT) were determined in bivalves coastal ecosystems with regard to TBT contamina- and sediments from 36 U.S. coastal sites. tion, bivalves and sediments were analyzed for Bivalves (oysters and mussels) from 97% and butyltin content. Additional oyster samples@from sediment from 75% of these sites were all Gulf of Mexico sites from 1988 and selected contaminated with butyltins. The average bivalve Gulf of Mexico sites for 1986 were also analyzed total butyltin concentration (640 ng/g as Sn) was to provide information on temporal variations of 18 times higher than the average sediment butyltin concentrations. concentration (36 ng/g as Sn). The most toxic butyltin, TBT, accounts on average for 74% in bivalves and 80% in sediments of the total 2. METHODS butyltins present. Sediments may be a long-term chronic. source of butyltins to estuarine Mussels and oysters were collected from 3 ecosystems even after TBT inputs cease. Temporal stations at 36 coastal sites distributed on the concentration changes were found for Gulf of Atlantic, Gulf of Mexico, and Pacific coasts, Mexico oysters, but no seasonal trends were including one site from Hawaii. Individual apparent. Continued temporal monitoring will be stations at each site were generally from 50 to required in order to determine what effects bans 10.00 m apart. An attempt was made to avoid known or limitation of TBT uses have on environmental point-sources of contamination in site selection. levels of these contaminants. Bivalves are Sample locations have been previously reported12. excellent indicator organisms for this purpose. Mussels (Mytilus edulis) or oysters (Crassostrea 1. INTRODUCTION virginica or Ostrea sandwichensis) were either collected by hand, tonging, or dredging. The National Status and Trends Program for Marine Sediments were collected with a box corer or by hand. The top I cm. was sampled with a Teflon Environmental Quality was established to define coated scoop. Rigorous protocols were followed the geographic distribution of contaminant con- to avoid sampling contamination13. The tongs and centrations of organisms and sediments, determine dredges were steel. The oyster shells were temporal changes in those concentrations, and cleaned by scrubbing with seawater before document biological responses to contamination. opening. The animal tissue was removed from the The Mussel Watch component of the NOAA Status and shell and a combined sample consisting of 15 to Trends (S&T) Program employs bivalves (oysters 21 individuals was stored frozen in combusted and mussels) as sentinel organisms of contaminant glass jars until analysis. In the laboratory, loading. Sediments are also monitored as a the sample was homogenized with a Tekmar longer term integrator of contamination to the tissumizer. The animals from each station at a ecosystem. site were either analyzed individually (Galveston Bay) or a composit sample for each site was Bivalves have proven to be valuable sentinel analyzed. Dry weight was determined on a 1-g organisms for various pollutantsl,2,3 and reflect aliquot. the current contamination burden of an ecosystem, while sediments reflect long-term contaminant The analytical method described here is a modifi- loading of an ecosystem4,rl,6. The increased cation of previously reported methods' 4,15,16. concern about butyltins, especially tributyltin Analytical sample sets consist of eight samples, (TBT) from antifouling paints"119110, led to the a blank, and a spiked blank. The blank consists decision to include the analyses of butyltins, of all the glassware, reagent, and the internal including TBT, as part of the ongoing NOAA Status standard (I.S.) but no sample. The spiked blank and Trends Mussel Watch Program. Oysters are is the same except for the addition of reported 'to bioconcentrate TBT from water by tributyltin chloride (2.28 Ag as Sn), dibutyltin factors of 2,300 to 11,400 times the water dichloride (DBT, 2.38 Ag as Sn), and monobutyltin CH2585-8/88/0000- 1198 $1 @1988 IEEE trichloride (MBT, 2.50 pg as Sn) as reference flame photometric detector (FPD). The CC standards. Samples spiked with these reference temperature was programmed from 600C, at a rate standards were also analyzed. of 120C/min, to a final temperature of 3000C with Bivalve extraction consists of weighing a 15-g a hold time of 10 min. The injector temperature was 3000C and the detector temperature was 2500C. sample into a 250-ml centrifuge tube, adding The FPD had a hydrogen-rich flame consisting of tripropyltin chloride (2.26 mg as Sn) as the I.S. hydrogen (80 ml/min) and air (100 ml/min). The followed by sodium sulfate (40 g) and 0.2% carrier gas, helium, had a flow rate of 1 ml/min. tropolone in CH2CI2 (100 ml). The sample was A 610 run filter was used in the FPD in order to extracted for 3 min with a Tekmar tissumizer. be specific for tin-containing compounds. The The sample was then centrifuged and the presence of the butyltins was confirmed by gas supernatant was decanted into a 500-ml flat chromatograph/mass spectrometry 4nalysisl2. bottom flask. The extraction was repeated two more times using 0.2% tropolone in CH2C12 (100 ml) each time and the extracts combined in the 3. RESULTS AND DISCUSSION 500 ml flat bottom flask. The total butyltin bivalve concentration in ng Sediment extraction consisted of weighing 20 g of Sn/g dry weight of tissue are reported in Table freeze-dried sediment, adding tripropyltin 1. The concentrations range from less than 5 chloride as internal standard, and 100 ml of 0.2% (the limit of quantitation) to 3760 ng Sn/g. The tropolone in methylene chloride in a 250 ml amber bivalve concentration of total butyltin is bottle with a Teflon-lined screw cap. The bottle highest in Hawaii and the West Coast, with the was capped and rolled for 3 hrs. The sample was Gulf and East coasts having lower and similar centrifuged, and the organic phase was decanted mean concentrations (Table 1). The ranges of into a 500-ml flat bottom flask. The sample was concentrations, however, show considerable extracted two more times using 100 ml of 0.2% overlap. The mean total butyltin concentrations tropolone solution and rolling for 16 and 3 hrs, for mussels and the two species of oysters respectively. The organic phase was collected in collected are also given in Table 1. Mussels the 500-ml flask. collected for this study had higher mean concentrations than oysters, C. virginica, Hexylatio although the ranges have a large overlap. All of After the tissue or sediment samples were the West Coast samples analyzed were mussel ' and extracted, they were concentrated to 10-15 ml by all the Gulf of Mexico samples were oysters (C. means of a water bath at 600C in a 500-ml flat virginica). The butyltin comcentrations bottom flask connected to a three -ball -Snyder determined for the five mussel and four oyster column. The samples were transferred to 50-ml samples from the East Coast generally fall within centrifuge tubes, the flat bottom flasks were the range of the West and Gulf Coast rinsed twice with about 10 ml of hexane and the concentration ranges. Therefore, the higher West rinses were added to the centrifuge tubes. A Coast concentrations may be due to sampling only clean Teflon boiling chip was added and the mussels which may have higher bioconcentration samples were heated in a water bath at 600C until factors or the West coastal sampling sites may vigorous boiling stopped (at this point, no contain higher water concentrations of butyltins. methylene chloride remains). The samples already The composition of the butyltins is important in hexane were purged with nitrogen and 4 ml of because TBT is more toxic than the other 2M hexylmagnesium bromide (Grignard reagent) was butyltins". The mean percent of TBT and ranges added. The hexylation reaction was carried out of TBT percent are also reported in Table 1. The overnight at room temperature. Five ml of 6M concentration of TBT accounts for 25 to 100% of hydrochloric acid was added in order to all the butyltins present with a mean of 74%. neutralize excess Crignard reagent. The organic The mean percent TBT is highest for the East and fraction was decanted, and the aqueous fraction Gulf coasts and lower for the West Coast and was extracted twice more with 15 ml of pentane. Hawaii samples. Oysters (C. Virginica) have a The combined organic phases were concentrated on higher mean percent TBT than mussels (M. edulis) a rotary evaporator to about 5-10 ml and were or the one sample of 0. sandwichensis. Again it further concentrated to 2 ml in a concentrator tube. The hexylated organotin compounds were separated from other polar organics using a TABLE 1. BIVALVE TOTAL BUTYLTIN CONCENTRATION AND % TBT column containing 15 g of combusted alumina and 10 g of combusted silica. The column was eluted with 50 ml of pentane. The pentane fraction was SITES MEAN(ppb) RAW--r= % TBT RANGE further concentrated to 500 pl in a concentrator ALL(36) 640 <5-3760 74 25-100 tube in a water bath at 600C. EAST(9) 340 50-770 78 64-100 GULF(l 7) 310 <5-1180 84 50-100 Cas chromatography analysis WEST(9) 1210 200-2820 55 25-70 Samples were spiked with tetrabutyltin as HAWAII(1) 3760 41 recovery standard and then analyzed by gas ORGANISMS chromatography on a HP 5790 gas chromatograph M. EDULIS(14) 890 170-2820 60 25-100 (GQ equipped with a capillary column (DB-5, 25 m C. VIRGINICA(22) 328 <5-1180 85 50-100 x 0.25 mm ID x 0.25 mm coating. thickness) and a 0. SANDWICH ENS IS(1) 3760 41 1199 should be noted that 'the ranges of the means considerably higher in 1987 and nearly the same overlap. This may reflect differences in the in 1986 and 1988. Concentration at some sites ability of these bivalves to take up or decreased, while others increase between 1987 and metabolize TBT or difference in the butyltin 1988. The mean butyltin concentration for all composition of their respective growing waters. the Gulf sites was 310 and 290 ng Sn/g for 1987 and 1988, respectively. From this Culf of Mexico The sediment total butyltin concentration ranged data there is no apparent change in the mean from less than 5 (the limit of quantitative) to butyltin concentration between 1987 and 1988 and 282 ng Sn/g with a mean of all sites of 36 ng large changes would not be expected until the Sn/g (Table 2). The Gulf of Mexico sediments had input of TBT is limited. These temporal data the lowest mean total butyltin concentration, will allow future investigations to determine while the East and West Coast and Hawaii sites recovery time from TBT contamination when limits had similar concentrations. Eight of the 17 or are placed on the use of TBT-containing anti- 46% of Gulf of Mexico sites contained no fouling paints. detectable butyltins. The mean total butyltin concentration for the remaining 9 sites (22 ng 2000- Sn/g) is still lower than the concentrations for Bivalves Total Butyltin 1987 the other areas. Concentrations 1988 1500- TABLE 2. SEDIMENT TOTAL BUTYLTIN CONCENTRATIONS AND % TBT CA 1000 SITES(#) MEAN(ppb) RANGE % TBT RANGE ALL(33) 36 <5-282 77 42-100 EAST(8) 64 5-117 64 48-100 500 GULF(I 7) 12 <5-89 93 54-100 WEST(7) 58 6-282 77 50-100 L HAWAII(l) 66 42 0 Ain 02 cc 0 M 0 @k C4 U C6 co M Z @0 0@ @Z 02 &OD CO M En 'A MUE.-4P.VM=E- The percentage of TBT in the sediment samples Z Station ranged from 42 to 100% with a mean of 77%. The lowest percent of TBT was found in sediments from 800 Hawaii, while Gulf coastal sediments had the 1986 highest mean percent TBT. Of the 25 sites where Bivalve Total Butyltin 0 1987 percent TBT could be calculated, 11 were 100% TBT 600, Concentrations 0 1988 (7 of 9 for the Gulf Coast). Most, of the butyltins found in sediments are in the form of undegraded TBT, which is also the most toxic of the butyltins. This is consistent with a mean 400 partition coefficients determined for TBT and DBT of 4.0 X 104 and 4.5 X 102, respectively17. Therefore, only 4.1% and 0.044%, respectively, of 200- TBT and DBT total water column concentration are associated with particles. This indicates that only a small percent of the butyultins found in 0 4 k the water column will reach the sediments. It U also indicates that if TBT and DBT concentrations 1= in the water column were equal and equilibrium U were reached, that the sediments would initially Station contain 99% TBT and only 1% DBT. This is consistent with the findings of this study that Figure 1. Gulf of Mexico Bivalve (oysters) Total 44% of the sediment samples analyzed contained Butyltin Concentration in 1986, 1987 100% TBT. Even though the sediment butyltin and 1988. concentrations are low, they may be a long-term chronic source of TBT to the ecosystem. The Gulf of Mexico bivalve total butyltin 4. CONCLUSIONS concentrations reported in Table I were for samples collected in 1987. Bivalve samples Butyltins, predominantly TBT, were found in 97% collected in 1988 at all of the Gulf sites and in of bivalve and 75% of sediment samples analyzed. 1986 at 6 of the sites were also analyzed. The Considering that TBT has been found to have a total butyltin concentrations for 1986, 1987, and half-life of a few days in the water column"I", 1988 for each site are plotted as a bar graph in the detection of an average of 640 ng/g as Sn of Figure 1. Generally the concentrations remain butyltin in bivalve samples was unexpected. approximately the same over this time period. These results indicate that bivalves must be very The exception is site CCNB (corpus Christi, efficient at bioaccumulating TBT from the water Neuces Bay) where the concentration was column. Based on reported bioconcentration 12,00 factors of 2,300 to 11,40011, TBT must be a 8. Oceans '87 Proceedings, International consistent contaminant of these bivalve gowing Organotin Symposium, 4, Marine Technology regions at mean concentrations in the range of 50 Society, Washington, D.C., 1987, pp. 1295- to 200 ng Sn/l. Sediments may be a long-term 1524. chronic source of TBT to growing waters even after limits on TBT uses are instituted. The 9. Thompson, J.A.J., Sheffer, M.G. Pierce, temporal data indicates no large increase or R.C., Chau, Y.K., Cooney, J.J., Cullen, W.R. decrease in TBT concentrations between 1986 and 1988. Determination of TBT concentration in and Maguire, R.J. Organotin Compounds in the oysters at these sites after limits are placed on Aquatic Environment: Scientific Criteria for TBT uses will provide valuable information on Assessing their Effects on Environmental recovery time of these ecosystems. Quality, National Research Council of Canada, Ottawa, Canada, 1985, pp. 1-284. 5. ACKNOWLEDGMENTS 10. Interagency Workshop on Aquatic Monitoring and Analyses for Organotin Compounds, Eds. Funding for this research was provided by Landy, R.B., Holm, S.E. and Conner, W.G. 1986 the National Oceanic and Atmospheric NOAA/National ocean Services National Marine Administration Grant Number 50-DGNC-5-00262 Pollution Program Office, 1986, pp. 1-53. (National Status and Trends Mussel Watch Program) and Texas A&M University's Sea Grant Program 11. Hall, L.W., Jr. and Pinkney, A.E. Acute and (R/ES-18). sublethal effects of organotin compounds on aquatic biota: An interpretative literature 6. REFERENCES evaluation. CRC Critical Reviews in Toxicology, 14, 1985 pp. 159-209. 1. Farrington, J.W., Risebrough, R.W., Parker, 12. Wade, T.L., Garcia-Romero, B. and Brooks, P.L., Davis, A.C. deLappe, B., Winters, J.K., J.M. Tributyltin contamination in bivalves Boatwright, D. and Frew, N.M. Hydrocarbons, from U.S. coastal estuaries, Environ. Sci. polychlorinated biphenyls, and DDE in mussels Technol., 1988 (in press). and oysters from the U.S. Coast, 1976-1978, The Mussel Watch. WHOI Tech. Report, WHOI- 13. Brooks, J.M., Wade, T.L., Atlas, E.L., 82-42, 1982, 106 p. Kennicutt, M.C. II, Presley, B.J., Fay, R.R., Powell, E.N. and Wolff, G. Analyses of 2. Farrington, J.W., Goldberg, E.D., Risebrough, Bivalves and Sediments for Organic Chemicals R.W., Martin, J.H. and Bowen, V.T. U.S. and Trace Elements from Gulf of Mexico "Mussel Watch" 1976-1978: An overview of the Estuaries, 1987, 355 pp. trace metal, DDE, PCB, hydrocarbon and 14. Maguire, R.J. Butyltin compounds and artificial ratio-nuclide data. Environ. Sci. Technol., 17, 19837 pp. 490-496. inorganic tin in sediments in Ontario. Environ. Scl. Technol. , 18, 1984, pp. 291- 3. Farrington, J.W., Albaiges, J., Burns, K.A., 294. Dunn, B.P., Eaton, P., Laseter, J.L., Parker, 15. Unger, M.A., MacIntyre, W.C., Greaves, J. and P.L. and Wise, S. Fossil Fuels. In: The International Mussel Watch: Report of a Huggett, R.J. GC determination of butyltins Workshop Sponsored by the Environmental in natural waters by flame photometric Studies Board Commission on Natural detection of hexyl derivatives with mass Resources. National Research Council, 1980, spectrometric confirmation. Chemosphere, 15, pp. 7-77. 1986, pp. 461-470. 4. Wakeham, S.G. and Carpenter, R. Aliphatic 16. Rice, C.D. , E,spourteille, F.A. and Huggett, hydrocarbons in sediments of Lake Washington: R.J. Analysis of tributyltin in estuarine 1976. Limn. Oceanogr. , 21, 1976, pp. 711- sediments and oyster tissue, Crassostrea 723. virginica.Appl. Organometallic Chem., 1, 1987, pp. 541-544, 5. Van Vleet, E.S. and Quinn, J.G. The contribution of chronic petroleum inputs to 17. Hinga, K.R., Adelman, D. and Pilson, M.E.Q. Narragansett Bay and Rhode Island Sound Radiolabled butyltin studies in MERL enclosed sediments. J. Fisheries Res. Board. Can. , ecosystems. Proceedings Oceans '87, Vol. 4, 35, 1978, pp. 536-543. International Organotin Symposium, IEEE, New York, 1987, pp. 416-419. 6. Wade, T.L. and Quinn, J.G. Geochemical distribution of hydrocarbons in sediments 18. Lee, R.F., Valkirs, A.0. and Seligman, P.F. from Mid-Narragansett Bay, Rhode Island. Fate of tributyltin in estuarine waters. Org. Geochem., 1, 1979, pp, 157-167. Proceedings Oceans 187, Vol. 4, International Organotin Symposium, IEEE, New York, pp. 7. Oceans '86 Conference Record, Organotin 1411-1415. Symposium 4, Marine Technology Society, Washington, D.C., 1986, pp. 1101-1330. 1201 LAKE SUPERIOR WINTER WEATHER STATION Lloyd J. Ladner and W., Brett Wilson Phillip J. Kies National Data Buoy Center (NDBQ Bay Technical Associates, Inc. Stennis Space Center, MS 39529-6000 Bay St. Louis, MS 39521-2040 ABSTRACT The National Data Buoy Center, under the auspices of Jhe National Weather Service, has, maintained a network of eight environmental data collection buoys on the Great Lakes since 1979. A critical consideration is the severe winter weather on the Lakes, requiring retrieval of the buoys every autumn and their redeployment in the spring after the end of the "ice" season. Deployment and retrieval of all eight buoys every year are major operational and logistical undertaking@ for botk NDBC and the Coast Guard, which provides the ship support for the NDBC moored buoy program. To preclude working under- unsafe weather conditions, the buoys are normally deployed after the start of the shipping season and retrievedprior to its end. Consequently, several weeks of the shipping season each spring and fall are without the benefit of the reports from W g these weather buoys. e@- All. V, le Faced with limited Coast Guard resources and uncontrollable weather AFI@! 4, during the winter months, NDBC decided to establish a buoy station in Lake Superior as an experimental "ice buoy" to report meteorological J and oceanographic data year-round. The station was established on November 7, 1985, for long-term evaluation, which is still ongoing (Figure 1). A second ice buoy station wasplannedfor a September 1988 deployment in northern Lake Michigan in order to broaden the scope R, of the experiment.. All our experience to date looks promising; however, final conclusions cannot be drawn until additional winter experience with these buoys has been obtained. Figure 1. NDBC Ice Buoy 12D07 at Station 45001, February 1986 of buoys, which had been ruggedized to withstand ice, and the possible 1. INTRODUCTION methods of instrumenting these buoys to provide basic meteorological parameters (i.e., wind speed and direction, barometric pressure, and air Since 1979, the National Data Buoy Center (NDBC), under the auspices and water temperatures). One initially attractive concept involved a new of the National Weather Service (NWS), has maintained a network of type of "ice buoy," developed by the Coast Guard, 2.2 meters in diameter eight environmental data collection buoys on the Great Lakes (Figure 2). and 6.1 meters in length. This buoy is designed to "ride up" on light The network employs three buoys on Lake Superior, two buoys on Lake ice and to submerge under the ice during a " freeze-over. " NDBC took Huron, two buoys on Lake Michigan, and,one buoy on Lake Erie. The a serious look at "throw-away" sensors to instrument this type of buoy, standard method for operating this network is to deploy the buoys after with the understanding that the sensors would be lost each winter. An the start of the shipping season and to retrieve them before the end of additional consideration if this type of "ice buoy" were to be used would the season. This typically results in a period of several weeks during the be the loss of data collection in the spring after the buoy reappeared but shipping season when no buoy data are available. Each year, it is riot uncommon for the Lake Carriers' Association and NWS Forecast Offices in the Great Lakes Region to request NDBC to extend the buoy time on 95 90 85 80 75 the Lakes. a I I 45001 The fundamental aspect of the problem of accomplishing earlier 7:111 45004 deployments and later retrievals is that these operations require major operational and logistical efforts by NDBC, and the Coast Guard, which 45 6 45003 -45 provides excellent ship support. However, available ship services are necessarily confined to function within the framework of existing Coast Guard resources and operational requirements in the Lakes. Consequent- 45002 ly, means to increase buoy station endurance time on the Great Lakes 45008 were investigated, and the challenge quickly evolved into the concept of 45- a survivable "ice buoy." 45007 2. ICE BUOY SYSTEM SELECTION 45005 In early 1985, NDBC undertook a series of. engineering design analyses 9,5 90 85 go to investigate alternate "ice buoy" concepts and to select the most likely candidate for further field testing. These analyses evaluated various types Figure 2. Buoy Station Locations in the Great Lakes prior to a service visit by NDBC/USCG technicians to restore operation of the buoy revealed only minor damage to the rubber fender that extends of the meteorological sensors. around the perimeter of the buoy deck. The fender is required to prevent metal-to-metal contact between the buoy and a ship during servicing Another candidate for possible use as an "ice buoy" was a 12-meter, operations. Upon completion of the physical inspection, a new suite of 100-ton discus buoy - a venerable hull built in the early 1970's that has electronics equipment, sensors, and power supply batteries was installed. survived many years of operation in the North Pacific and North Atlan- tic Oceans. NDBC engineers performed several computer simulations of Early in May 1986, the SUNDEW towed buoy 121307 back to station "worst-case conditions," which indicated the buoy should survive the 45001 and moored it using an all-chain mooring. The all-chain mooring harsh Great Lakes winter under all except the rarest of extended, severe, was expected to provide an improved holding capability in the event a arctic-type conditions. Various parameters, such as topside icing, ice floes, moving ice field interfered with buoy stationkeeping. Our previous fast ice, reserve buoyancy, ballasting techniques, and freezing precipita- experience with a nylon line, inverse-catenary mooring at station 45001 tion, were considered in selection of the 12-meter hull as a test platform. indicated that abrasion from either dragging on the take bottom or rubbing An additional consideration for using the 12-meter hull was that this class against large sections of ice caused mooring failure. of buoy was scheduled for removal from the "fleet" due to high annual maintenance costs when used in a saltwater environment. Next, NDBC personnel conducted interviews with several Great Lakes 4. 1986-1987 AND 1987-1988 WINTERS ice "experts." They all agreed the 12-meter buoy should have a good to excellent chance for survival if all precautions were followed and the buoy In order to gain further general operational experience over several winters were located in an open lake. That is, the buoy should not be deployed and to test the all-chain mooring, the "ice buoy" 121307 experiment con- in a bay or harbor that could make it susceptible to being driven ashore tinued through.the winters of 1.986-1987 and 1987-1988. During the during the spring ice breakup. relatively mild 1986-1987 season, the all-chain mooring performed satisfac- torily and the buoy maintained its original position. There was very little ice on Lake Superior that winter. The performance of the buoy payload 3. INITIAL ICE BUOY DEPLOYMENT, 1985-1986 and sensors was also satisfactory throughout the winter. In early 1985, buoy hull 121307 was available for immediate use as an The next winter turned out to be more harsh than the 1986-1987 season. experimental "ice buoy." It was readied for deployment during that sum- In early February 1988, both barometers on the buoy began reporting mer, was towed from NDBC in mid-September, and arrived in Chicago erratic data, apparently as a consequence of ice accretion on the pressure in early October. The Coast Guard Cutters KATMIA BAY and MOBILE ports. On February 10 the buoy began to drift, dragging over 1,000 feet BAY shared in towing the buoy up the Mississippi River to Sault Ste. of chain and a 20,000 pound concrete sinker-type anchor. The drift stop- Marie, the site of final outfitting. ped on February 18, but began again at the start of March. The buoy After all sensors, power supplies, and electronics equipment were installed, finally came to rest at 47035'N, 87004'W, some 37 miles from its initial the buoy was deployed in mid-Lake Superior at station 45001 in early position at 48*01'N, S7*43'W. The final location was in water over 200 November 1985. The existing NOMAD buoy on that station was retrieved. feet deeper than the original position. Since the all-chain mooring was On February 7, 1986, the Coast Guard provided on overflight of the buoy designed with a scope (length-to-depth ratio) of approximately 1.4, suf- and took pictures of its condition (Figure 3). The buoy had considerable ficient chain was present in the mooring to accommodate the additional ice accumulation, but there was no lake ice in the immediate area. On depth at the new.location. February 12, position data received from the buoy indicated it was mov- ing slowly off station. An overflight conducted by the Coast Guard on February 15 confirmed the buoy was moving, and was amid considerable "brash ice" (Figure 4). However, all systems continued to work. It was difficult to determine whether the mooring was still attached and being dragged about the take bottom, or if it had been severed. Figure 5 gives one an idea of the considerable movement (a total of 300 miles) experienced by the buoy for the months of February, March, and April 1986. Because the buoy moved into shallow water during early April, the mooring was assumed to have parted. On April 10, the buoy came to rest in a small bay on the east side of Manitou Island. Buoy 121307 was recovered by the Coast Guard Cutter SUNDEW in late April 1986, and towed to Houghton, Michigan. A complete inspection Figure 4. 12D07 Amid Brash Ice, February 15, 1986 On.May 15, 1988, a Coast Guard cutter attempted to free and retrieve the mooring, but the effort was not successful. Since the mooring would not budge, it was assumed that it was fouled on an underwater obstruc- tion of some sort. Consequently, the buoy was left where it was rather than risk breaking the mooring to return it to the original station 45001 position. Figure 6 shows 121307 while it was drifting in February 1988. Curiously, the ice coverage on the lake does not appear to be severe, yet substantial drift is occurring as is evident from the buoy wake. Note that the buoy freeboard is only about six inches. This means the buoy is carrying about 120 long tons of extra weight. If the entire mooring and concrete sinker (combined weight of 38,000 pounds) were suspended, as was likely when Figure 6 was taken, the buoy still was holding almost 230,000 pounds Figure 3. Overflight Photo of 12DO7, February 7, 1986 of ice on deck (corresponding to an average thickness of about 3 feet). 1203 STA 45001 7 FEB 15 FEB 27 FEB 13 MAR, 10 APR MANITOU ISLAND Figure 6. 12D07 Adrift in Feburary 1988 Figure 5. Track of the Movement of 12DO7, February 1986 5. CONCLUSIONS NDBC will continue the "ice buoy" experiment during the 1988-1989 winter with buoy 12DO7 at station 45001 in Lake Superior. In addition, NDBC has replaced the 3-meter buoy at station 45002 in upper Lake Michigan with a second "ice buoy," hull 12DO8. This will enable addi- tional winter experience to be obtained in somewhat different lake conditions. In general, the technical considerations involved in year-round operation of large discus buoys on the Great Lakes are well understood as a result of the "ice buoy" deployments to date. The issue of long-term cost- effectiveness is not as clear. Operational factors affecting cost- effectiveness include the frequency of mooring loss and the rate of damage to buoy hulls. It is hoped that the 1988-1989 deployments at stations 45001 and 45002 will further help in the determination of these factors. Never- theless, the technical feasibility of long-term, year-round operation of "ice buoys" on the Great Lakes has been clearly demonstrated. 1204 INFRARED LASER WAVE HEIGHT SENSOR Hal Brown National Data Buoy Center (NDBC) Stennis Space Center, MS 39529-6000 ABSTRACT ting. The external electrical connection consists of one waterproof elec- trical connector, with six conductors for electronic circuit power, heater The National Data Buoy Center (NDBQ has a requirement for an ac- power, and signal output. A thermostat-controlled heater, made up of curate, reliable, economical method of measuring ocean wave activity at a quartz window with a high-transparency conductive heater coating on its Coastal-Marine Automated Network (C-MAN) meteorological sta- the inner surface, covers the optics and prevents the buildup of ice or tions. This paper describes a diode-laser-based ocean wave height sensor moisture. developed to fill this need. The device, called the Infrared Laser Wave Height Sensor (IR L WHS), was designedfor use in a remote, unmanned, The pulse repetition frequency is 1500 pulses/second. When the signal marine environment. is passed through a low-pass (IO-Hz bandwidth) filter, the analog output signal is effectively the result of averaging 150 range measurements. This averaging technique greatly improves the measurement accuracy. 1. INTRODUCTION An RC differentiator circuit is used to prevent false returns from aerosols (i.e., fog, mist, ocean spray). The backscatter from aerosol has a slow The National Data Buoy Center (NDBC) operates a network of remote, rise time relative to the transmitted light pulse due to the extended volume unmanned meteorological data acquisition/transmission stations, the of the aerosol (backscatter signal energy from an aerosol is a function Coastal-Marine Automated Network (C-MAN). The C-MAN stations are of the aerosol volume). Therefore, the RC differentiator is an effective located in various coastal areas, primarily at USCG light stations and signal processing technique for discrimination against aerosols. Also, since on fixed, offshore platforms. The C-MAN stations monitor various en- near backscatter from aerosols give the largest returns, the receiver is time- vironmental parameters, including wave height and wave period. The out- gated such that signals from less than 6 meters (the required minimum puts of the various sensors are processed by the C-MAN data acquisition range) are ignored. system and are transmitted to the National Weather Service via the GOES satellite system and landlines. Table I lists the NDBC specifications for the IR LWHS. The Schwartz Electro-Optics, Inc., sensor meets or exceeds all requirements. As part of its waves development program, NDBC has investigated various sensing methods, including accelerometers, wavestaffs, and reflective devices (i.e., radar, acoustic, laser). The IR laser distance-measuring Table 1. NDBC Specifications for the IR L WHS method has proven to be the most adaptable to NDBC requirements for installation, operation, and long-term maintainability; therefore, it is the PARAMETER RANGE preferred method for those coastal locales for which a favorable look angle is achievable (i.e., within approximately 15 degrees of vertical). OPERATING VOLTAGE 10.5-15.0 VDC, 12-15 VDC NOMINAL SENSOR CURRENT :5 1.0 A AT 12 VDC 2. INTENDED USE HEATER CURRENT :5 1.0 A AT 12 VDC The IR LWHS was designed to operate in a continuous mode (24 FREQUENCY RESPONSE FLAT FROM 0 TO 1 HZ hours/day) in a marine environment, including marine growth, oil films, ACCURACY, STATIC -t1% FULL RANGE sunglint, ocean spray, rain, snow, and fog. The unit had to be small and light enough that it could be installed and removed by a single technician ACCURACY, DYNAMIC �3% OF TROUGH-TO-PEAK WAVE HEIGHT, OR � 10 CM, WHICHEVER working in awkward positions, such as the underside of offshore plat- IS GREATER forms. Because the only power available at some C-MAN locations was battery/solar charger units, consumption had to be held to a minimum. RESOLUTION :s 0.1% OF FULL RANGE SENSOR RANGE MINIMUM :s 6.0 M, MAXIMUM Schwartz Electro-Optics, Inc., of Orlando, Florida, was selected to design 50 M and build the IR LWHS from NDBC specifications derived in part from OUTPUT SIGNAL 0-5.0 VDC the above requirements. SCALE FACTOR LINEAR, TYPICALLY 0.1 VDC/METER 3. DESIGN LOAD IMPEDENCE 10 KOHMS BEAM SIZE 1* FULL ANGLE The IR LWHS employs a pulsed GaAs/GaAlAs diode laser and an AIR TEMPERATURE 150C TO + 500C avalanche photodiode detector along with the appropriate transmit- ter/receiver optics in a side-by-side (as opposed to,a coaxial or concen- WIND 0-60 M/S tric) configuration. Range is determined by measuring the round-trip pro- HUMIDITY TO 100%, CONDENSING OUTSIDE pagation time of a 10 nanosecond, 850 nanometer laser pulse to the water SENSOR HOUSING surface and back, generating a continuous voltage proportional to the RAIN MODERATE TO HEAVY, TO 13.0 distance from the sensor to the surface of the water. The change in voltage CM/HOUR, FREEZING IN WINTER level is proportional to wave height. The device is housed in a cylindrical, SNOW NONE TO FREQUENT AND HEAVY, TO %tainless steel housing approximately 9 inches (22.86 centimeters) in 60 CM IN 24-HOUR PERIOD diameter and I foot (30.48 centimeters) in length. Total weight is 15 MEAN TIME BETWEEN FAILURES 2 YEARS pr)ijrid% (0.8 kilogram@). A flange on the rearof the unit is used for moun- 1205 United States Government work not protected by copyright 4. SENSOR EVALUATION resolved. It is not believed to be originating with the IR LWHS. Testing will continue until the observed anomalies are resolved. During the past six months the IR LWHS has undergone evaluation at the Chesapeake Light Platform. A wavestaff is being used as a reference. Data are taken synoptically for 20-minute periods once each hour. Data 5. CONCLUSIONS from both the IR LWHS and the wavestaff are sampled and processed by separate Data Acquisition, Control, and Telemetry (DACT) payloads. Tests conducted at NDBC and at Schwartz Electro-Optics, Inc., have The data are sampled and converted to digital at the rate of 256 samples shown no basic problems with the IR LWHS sensor. The Infrared Laser per 100-second segment. The segments overlap, with the last 50 seconds Wave Height Sensor will provide a reliable, accurate, and economic of one segment comprising the first 50 seconds of the subsequent seg- method for measuring wave activity in areas where the sensor can be pro- ment. Data from all segments comprising the 20-minute acquisition period perly mounted. The sensor must be mounted in a stationary position above are processed into one data set. The data from each segment are process- and perpendicular to the surface of the water. Experimentation has shown ed using an FFT to produce a 129-element data set representing the 0.00- that data can be obtained from a laser pulse for angles up to 15* from to 1.28-Hz energy spectrum. Although data are acquired and processed the perpendicular. However, for greater angles, the return signal for the entire 0- to 1.28-Hz spectrum, only the 0- to 0.40-Hz data are diminishes, and the wave geometry makes accurate data acquisition used. Following some further processing, the data are used to calculate difficult. three sea state parameters: significant wave height, probable maximum wave height, and spectral peak period. WAVE STAFF - Wave data to date have shown good agreement between the wavestaff and the IR LWHS. The difference between the two sensors is usually less than 0.1 meter. This difference may be due to the fact that the two sen- sors are located in different positions on the platform, with the IR LWHS being on the outboard corner, and the wavestaff being inboard, where the support structure of the platform could affect wave action. Figure I shows a time-series plot of the significant wave height from 0600 UTC June 9, 1988, to 0600 UTC June 11, 1988. Figure 2 shows the average 7- wave period for the same time frame. Two different problems have been observed during the evaluation period. On two occasions during the winter months,when local temperatures dropped to approximately - 10*C, data dropouts of about 12 hours oc curred. When the temperature increased, data transmission resumed. Heat ------- generated by the quartz window heater, along with heat from the elec- tronics, should have been sufficient to keep the optics free of ice or moisture and the unit operating. Subsequent temperature chamber tests at - 151C failed to duplicate the malfunction, and it has not yet been determined if the problem is related to the IR LWHS or to some other 1. part of the system. Another problem encountered has been a "spike" in the significant wave height reported by the IR LWHS, relative to the wavestaff, of about 0.5 meter. This problem has appeared for short periods of time, almost always 6 12 12 is 0 6 around 1800 UTC, on about one-half of the days during the evaluation 6/ 9 511D 6/10 6/11 1 988 period. Figure 3 shows a time series plot of the significant wave height UTC HOURS from 0000 UTC June 4, 1988, to 0000 UTC June 6, 1988. As in Figure I there is good agreement between the wavestaff and the IR LWHS, wit@ Figure 2. Time-Series Plot of Average Wave Periodfor 0600 UTC June the notable exceptions of "spikes" in the IR LWHS data around 1800 9, 1988, to 0600 UTC June 11, 1988. UTC on each of the three days. This anomaly, as yet, has not been WAVE STAFF - WAVE STAFF - IR LWHSa C 2- o. 12 18 0 6 12. 18 a 5 a 6 12 is a 6 12 18 a 6 12 is 0 Via 6110 6/11 61 4 614 W 5 61 5 6 6/ 6 61 UTC HOURS ) 1988 uTc HOURS 1999 Figure 1. Time-Series Plot of Significant Wave Height from 0600 UTC Figure 3. Time-Series Plot of Significant Wave Height from 0000 UTC June 9, 1988, to 0600 UTC June 11, 1988., June 4, 1988, to 0000 UTC June 6, 1988. 1206 WavePro: AN AUTONOMOUS WAVE PROCESSOR WITH LONG-RANGE TELEMETRY George Kontopidis Gary Bowers Sea.Data Inc., A Pacer Systems Company one Bridge Street, Newton, MA 02158, USA, ABSTRACT A real-time wave processing system with long-range The National Data Buoy Center (NDBC) deployed the first telemetry capabilities has been developed. The system is weather/wave processing bouy in 1971 and since then, at least housed in an accelerometer wave-measuring bouy and consists seven generations of bouys have been developed taking of a wave data acquisition module, a versatile processor/ advantage of technological improvements. Each new generation controller module, and a satellite navigation and telemetry improves the performance, reliability, and deployment time, module. Real-time information including the wave height and reduces the overall experiment cost [9]. power spectrum, the longitude and latitude of the bouy is computed in situ and transmitted via the telemetry link. IFREMER has developed an electronic module, compatible Selected data products and raw data are archived on magnetic with Waveriders, whichcomputes the ornnidirectional spectrum tape at regular intervals or when the significant wave height of sea-state on board and transmits the results using the exceeds a predetermined threshold. ARGOS positioning and transmitting system.. Deployment results have been reported in [4]. The WavePro is useful in applications where wave data must be taken in remote locations, which are lacking line-of@ Although it is impossible to cover here all related sight telemetry paths required by existing RF systems. works in the area (and we apologize for unintentional Extended deployment periods are made possible by transmitting omissions), it would be a mistake not to give a reference to only statistically significant frequency domain data the remarkable engineering work of the WAVESPEC [10] in 1984 products. Despite the fact that the WavePro was developed independently of the WAVESPEC, the similarities of the two systems are This paper focuses on the system architecture and noticable. The improvements/advances of the WavePro are: implementation details. A 60 MB recording option with data play back capability. Field programmability to parameter and algorithm level INTRODUCTION without opening the hatch cover. Usage of the SATNAV navigation system. Sea Data has developed a self contained wave processing Safer and less expensive battery system using alkaline system called the WavePro. The system is based on the batteries instead of lithium. Datawell Waverider bouy with an integral accelerometer. It Minimal modifications to the Waverider bouy electronics. processes wave information in real time, and transmits the Reduced module assembly time. It typically takes under complete wave spectrum via the GOES system. A polar orbiting one hour to remove and put back together a WavePro from satellite navigation system (SATNAV) is used to determine the a Waverider bouy. coordinates of the bouy which are appended in the GOES record. The WavePro also includes a recording device to Evaluation of initial WavePro field data are encouraging archive all data products along with important storm events so far. It is expected to improve the performance/cost ratio in the form of raw data. The system is controlled by two and make WavePro a valuable research tool to the scientific low-power microcomputers, one of which controls the data community. acquisition and system timing, and the other which performs the necessary wave signal processing computations. The latter is built around a Vector Processing Language (VPL) SYSTEM ARCHITECTURE which assures software robustness and field programmability. All system submodules are powered by internal alkaline Figure I shows a block diagram of the WavePro. The batteries supplying power for 9-12 months depending on the major subsystems are: sampling and transmission schemes. a. The Waverider bouy and the Accelerometer In the following, some background work is mentioned and b. The Data Acquisition and Timing module references to related/similar systems are given. c. The Wave Processor and System Controller d. The GOES transmitter Real time transmission of wave information has been the e. The SATNAV module focus of scientists and engineers for at least two decades. f. The T/R. switch and the Antenna. Datawell has set a de facto industry standard with their g' The Archival Recorder. bouys (Waveriders) featuring RF transmission of analog real- C. The Power Supply Subsystem time wave measurements. Successful deployment results and world wide experience have been reported and published [3]. Wave height is sensed by measuring the vertical acceleration of the bouy. This signal along with temperature measurements are fed to the Data Acquisition and Timing CH2585-8/88/0000- 1207 $1 @1988 IEEE Module (DAS- 18a) which conditions, digitizes and qualifies Modifications of the Waverider bouy (and the integral the sensing parameters. The DAS-18a includes a very stable electronics) by Sea Data are kept to a minimum with the clock which is used to synchronize the operation of the exception of the battery system. The original Leclanche Fntire WavePro (except the GOES transmitter). At given cells were replaced by potted alkaline battery packs which intervals, DAS-18a turns on the Wave Processor and System provided three times more energy for the same volume. Controller (WPC-18) to process the "raw" data. The WPC-18 Alkaline batteries also reduce the risk of hydrogen keeps its own timing and controls the operation of the GOES generation and gas leaking which is inherent to Leclanche transmitter, SATNAV receiver, and the Archival Recorder. cells. The Datawell's modulation circuit and RIF transmitter Depending on the time and the past sea conditions, the WPC-18 are disabled. A new antenna mounting plate has been can perform wave processing, control of data archival, fabricated to accept the 40OMHz antenna used for GOES comparison of current sea state with previous ones, error transmissions and SATNAV receptions. recovery, positioning correction, timing correction or a combination of the above. These WPC-18 functions are programmed in VPL, which allows the end user to alter them b. The Data Acquisition and Timing module to best suit the needs of the experiment. A robust power distribution system involving a "battery booster" is used to The Data Acqusition and Timing module (DAS-18a) is provide for a wide range of current demands, from a few manufactured by Sea Data and performs three major functions: milliamperes to several amperes. data collection, preprocessing and precise overall system timing. It consists of a waveform shaper and frequency The first two WavePros were built in 1987. Since then, multiplier to improve the resolution of the wave height the design has been refined and substantially-improved. The (frequency) output from the Waverider, a 12 bit integrating technology used in all submodules has been in use for several A/D to measure air temperature and water temperature years and 95% of the WavePro firmware is identical to (options), a I ppm time base circuit over the operating previously well-tested wave processing systems which have temperature range, and a microcontroller. The microcontrol- been used offshore. ler handles all data conversions and formulates "records" of data to be sent for futher processing to the Wave Processor SYSTEM DESCRIPTION and System Controller (WPC- 18). The DAS- 18a is built exclusively from CMOS components a. Waverider bouy and the Accelerometer and the microcontroller has the capability to be placed in a virtually zero power consumption state during periods of The Datawell Waverider bouy contains an integral heave inactivity. A regular interrupt given by the system time sensor to measure vertical acceleration. The associated base "wakes up" the microcontroller every 30 see to check electronics convert the acceleration signal to vertical dis- what function has to be performed. Note that WPC-18 is held placement by integrating it twice. The accelerometer is in standby mode by the DAS- I Sa until a record of data is protected from temperature gradients by foam insulation and ready to be processed. This reduces the power consumption of electronic compensation is used to reduce the effect of the electronics to less than ImA in standby (75% of the temperature variations. The bouy dynamics have been modeled deployment time) and 90mA in Data Processing mode (under 5% precisely and a second order transfer function is available of the deployment time). f2]. The output of the Waverider electronics is a square wave frequency signal (lv pp) proportional to vertical dis- placement at a rate of 0.98 Hz per meter. 400 Mz Antenna Air Temperature Data T/R Acquisition wi Water T_mp_r@tur- (DAS-18a) RxDl TxDl Waverider WPC_ON RxD2 Accelerometer IFQ2 TxD2 GOES Wave GOES_ON Transmitter Processor Switch and Controller Voltage Sense 3 RxD3 .............................. A/D FFQ TxD3 SATNAV Power Suppl@ RxDO System TxDO Q07 SATNAV-ON-i user Pwr Switch Terminal Attn (in the LAB) (WPC-18) st%am.Bing Tape Recorder Fi gure 1: The WavePro Block Diagram 1208 BBBB MMDDHH 15 7ss F7TTT [ Vj EE 22 x DOD ... UNNYYZZC B FF Bouy ID Hs Normalized Longitude DSP Wave Energy NNN degrees Td Status YY minutes ZZ 1/100ths Battery C Est/Wst Voltage GOES Status FF forward RF Latitude Month, Day, Hours RR reverse RF AA degrees EE errors YY minutes ZZ 1/100ths SATNAV C Nth/Sth Status Figure 2: The GOES Message Data Record c. The Wave Processor and System Controller available for transmission. GOES transmissions take place at specific times, which are programmed into the Master Control The Wave Processor and System Controller (WPC-18) is a Module. The Master Control Module includes the GOES-SOFT, low power modular microcomputer (based on the C-44 bus) which is a high level real-time firmware package. This manufactured by Sea Data. The hardware consists of a CPU, package provides time keeping facilities and the necessary 256KB of EPROM, 32kB of EEPROM, 256KB of RAM, four H/W and S/W handshake with the data source (the Wave UARTS, calendar timer, watchdog circuitry and numerous Processor and System Controller). The format of the trans- external 1/0 interface components. mitted record is shown in Figure 2. The real power of this system is in the operating e. The SATNAV module software called the Vector Processing Language (VPL). VPL is a combination of a real-time environment with an embedded The SATNAV module is manufactured by Mars Electronics compiler/interpreter executing pre-compiled FORTRAN and [6] (VIGIL RX). It receives position information on a UHF .C.1 modules. The language syntax is modelled after "C", where channel (40OMHz) from the "Transit" Navigation System of - all pre-built FORTRAN and "C" routines are callable, and satellites which has been well-proven with over 90,000 mili- executable at full CPU speed. The design philosophy of VPL is tary and commercial users worldwide. At present, the five similar to the "modular decomposition" approach recommended in operational satellites in various polar orbits provide at [8] which is well recognized by professional software writers least one "pass" every 70 minutes for users located at 50 for oceanographic applications [7]. degrees latitude. This time is improved at higher latitudes. The SATNAV module not only computes the latitude and long- The most significant feature of VPL is that it can be itude but it also computes when the next satellite pass is customized to meet the most demanding system requirements by expected. This information is (optionally) used by the WPC- the end user. The end user has access not only to "operating 18 to control power switching of the SATNAV to conserve parameters" of the WavePro (such as the sampling rates, the battery energy. The SATNAV is accurate to 200m approximately; conditions for recording on tape, the windowing parameters an error indicator is provided to qualify the validity of the etc) but also to the operating algorithms. For example, a reported coordinates. standard RS-232 terminal (or a portable computer) can be attached to the WavePro and without opening the hatch cover, f. The T/R switch and the Antenna one can instruct the WPC-18 to change the spectrum processing methodology by including the effects of the bouy dynamics. The GOES transmitter and the SATNAV receiver share the same antenna. Since GOES transmits 40W signals and the In the lab, short data collection cycles can be initiated SATNAV can receive no more than a few microwatts, a UHF and executed in several minutes instead of waiting for hours Transmit/Receive switch has to be inserted. This switch is or days. VPL also allows simulation and testing of the fabricated by Sea Data and the GOES transmitter is adapted to behavior of the system in fault conditions, such as GOES control the position of the switch. Futhermore, the WPC-18 malfunctions, SATNAV errors, low battery voltages etc. In can (optionally) keep track of GOES transmissions and enable fact, the VPL code can change the functionality of the system the SATNAV at non-overlapping time intervals. in case of a severe error. For example, if the battery voltage level decreases a below certain point, VPL can decide to "shut The Synergetics Model 14A half-wave quadrafilar helix down" the SATNAV and make GOES transmissions less often. antenna was used, tuned at 40IMHz and capable of transmitting 50 watts. It was found that this antenna was suitable not d. The GOES Subsystem only for GOES transmissions but also appropriate for SATNAV The GOES transmitter used in the WavePro is manufactured receptions. by Synergetics [I I] and consists of the Master Control Module g. The Archival Recorder (340113) and the GOES Transmitter Module (3426A). The transmitter synthesizes a 40OMz signal at a 40W power level. A small capacity (2MB) or a high capacity (60MB) It also includes a very stable TXCO time base with Ippm long recorder can be installed in the WavePro. In both cases, the term stability over a -40 C to 55 C temperature range. The recorder is used to record data products, raw data and system Master Control Module is a microprocessor based controller information. The data products consist of the band averaged with 4-8KB of RAM. Data transmitted asynchonously to this wave spectrum and its moments, and the position (coordinates) module are stored temporarily in the RAM buffer and become of the WavePro. The raw wave data are recorded either 1209 conditionally or at fixed time intervals. The system inform- the average current drain of the system cannot be more than a ation records include timing, battery voltages, operating few milliamperes for deployments of 9-12 monthi. To achieve temperature, SATNAV positioning error indicators, GOES these requirements, a rechargable battery was used to provide forward/reverse RF power, software error codes etc. the "peak" currents of the system. A WPC-18 controlled charger was used to charge this battery from the primary The Wave Processor and System Controller computes the batteries, calculating the amount of energy taken out and significant wave height and decides when and for how long to (optionally) compensating for temperature. A high efficiency record raw data. More complicated recording conditions can micropower switching regulator is used to charge the l2v be easily programmed (in the field). As an example, the user battery accepting input voltages from 8v to 24v. A "watch might program the Wave Processor to record data according to dog" timer and additional direct voltage sensing circuit was the following scheme: used to monitor the state of the battery. It was found experimentally that under these operating (charging/dis- If the wave energy in the frequency band 0.311z to charging) conditions, the efficiency of the battery (charge 0.5Hz is more than 20% of the total wave energy, in/charge out) is better than 95% for a temperature range of record data until the significant wave height is - 15 to 40 degrees C. less than 5 meters. Independently of the above conditions, store I hour of raw data every 2 days The GOES module requires l2v at 5mA in standby mode, starting at 10:00 at night. 250mA when data are transmitted to its buffer, and 8A during UHF transmissions. It is powered directly from the recharg- The High Capacity Recorder (HCR-660) is a low power able battery since it turns itself ON/OFF asynchonously with streaming recorder, recording data in the industry standard respect to the rest of the system. QIC-24 format on 600' tape. It contains its own tape drive, formatter (manufactured by Tandberg and adapted for low power The SATNAV module contains its own switching regulator operation by Sea Data), a microprocessor based controller and requiring 9-30v input with a current drain of 600 mA (at a 128-256K solid state data buffer. The recorder module 12V). The SATNAV is powered from the primary or the includes its own switching power supply and can store 60MB of secondary (rechargable) batteries via an electronic switch data, consuming only 6AH of battery energy (at 15V). The which is controlled by the WPC-18. recorder includes both write and read electronics, incorpo- rating an automatic read -after- write correction mechanism. The Waverider heave sensor electronics are powered from More details about the HCR-660 can be found in (5]. the primary cells by an electronic switch under DAS- t8a control. h. The Power Supply Subsystem The power supply system of the WavePro was a very challenging electronic design. The various system components require a wide range of supply currents ranging from 0.8mA for the heave sensor, to 200mA for the main processing CPU, and up to 85OOmA for the GOES transmitter. At the same time, 5 WO J Figute 3: The DAS-18a ready to be inserted in the Waverider 1210 @ p@ 'P 6", I-A Figure 4: WavePro Electronics (GOES, T/R. Switch, SATNAV and WPC-18) CONCLUSION The concluding remark of this paper is similar to the [6] Mars Electronics. VIGIL RX SATNAV Manual. Berks, last sentence of reference [10]: The only way to appreciate England 1986. the difficulties in putting together the prototype WavePro system is to ask the technicians and engineers who had to [7] McClintock J.D., Davidson L.W., "Advanced Ocean Wave work over the hull with one hand through a single 8" hole for Synthesis, Analysis, and Generation Software," in Proc. several hours. IEEE Oceans'87, Halifax, Canada, Sept. 1987, The WavePro is a stand alone, in-situ wave processing pp. 554-557. system with long range telemetry and positioning [8] Parnas D.L, "On the Criteria to be Used in Decompositing capabilities. It is intended to be used to monitor real time Systems into Modules." Comm. ACM, vol. 15, no. 12, wave conditions in remote areas for deployment periods of pp. 1053-1058. 9-12 months. The WavePro is engineered to be modular and adaptable to a broad class of experiments involving [91 Prine G.D., Timko R.E., "Acquisition of a Value customized processing. Engineered Environmental Payload," in Proc. MDS'86, New Orleans, April 1986, pp. 602-604. REFERENCES [10] Smith R.D., Lawson J.L., Campbell E.C., "WAVESPEC -- A New Wave-Measuring and Processing Bouy," in Proc. [1] Bowers G., Kontopidis G., "A Novel Approach to Real-Tirne MDS'86, New Orleans, April 1986, pp. 402-408. Wave Processing," in Proc. IEEE Oceans'87, Halifax, Canada, Sept. 1987, vol. 3, pp. 1142-1148. [11] Synergetics International. Model 3401B Master Control Module (03401-94102) and Model 3426A GOES Transmitter [2] Datawell BV. Operation and Service Manual for the Wave- Module (03426-94101). Colorado, 1986. rider. Haarlem, Netherlands, 1980. [3) Draper L. (editor) "The Waverider Discussion," Proceedings, Institute of Oceanographic Sciences, May 1975. [4] Ezraty R., Racape J.F., "Directional and Ornnidirectional Sea State Information Obtained from Instrumented Bouys," in Proc. MDS'86, New Orleans, April 1986, pp. 315-321. [5] Kontopidis G., "Design and Implementation of Streaming Tape Data Loggers," in IEEE Proc. 4th Symposium Oceanographic Data Systems, San Diego, Feb 1986, pp. 156-165. 1211 MEASURED TRANSFER FUNCTIONS FOR SHIPMOTIONS IN NATURAL SEAWAYS F.Ziemer, E.Stockdreher, H.Giinther GKSS - Research Center GmbH 2054 Geesthacht, West Germany ABSTRACT natural seaway, assuming linear beha- viour. Now the required power and the In cooperation between the Institute fuel consumption can be determined. The for Shipbuilding of the University of optimization has to find the route of Hamburg, the German Hydrographic minimum total costs. It is an optimiza- Institute, the Max Planck-Institute for tion in space and time and delivers not Meteorology, the German Weather service only the coordinates of the optimum and the GKSS-Research Center a numeric track, but also the recommended speed. model was developed to determine the This implies good cooperation between Optimum route for a ship, which means the ship's management and the routing minimum costs. institution. To verify different moduls of the The new method of wave measurements model, measurements of the ship's from board a fast going ship [21 pro- motions, the ship's operational data vides the possibility to test the pro- and the actual wave field were taken. grams which have been developed for the The wave measurements, using a method routing. Full scale measurements of the developed from GKSS, were carried out ship's seakeeping were carried out from by means of the marine radar on board January until March 1988 on board the the ship. container vessel 'Stuttgart Express' of The measurements of the ship's beha- the HAPAG LLOYD AG. The particulars of viour in the seaway are compared with this ship are given in table 1. Simulation runs of the routing model under assuming the measured wind and wave conditions. 1. INTRODUCTION Table 1. Main particulars of the To find the route of most favourable 'Stuttgart Express' costs it is evident to know about the future winds and seastdtes and about length over all 240.5 m the behaviour of the ship under these ..length between Perpendiculars 233.0 m conditions. breadth 32.2 m The routing model [11 Proceeds from draught 9.0 m the service speed 21.0 kn ship's motion in the regular sea- way. With the Pitching and heaving motions the increase of the resistance is determined and then transferred to I Date, time, position, course, draft, trim, loadiDg conditions 2 Torque, number of revo- lutions 3 Doppler-log, echo- sounder 4 Average rudder angle 5 Vertical accelerations 6 Horizontal acceleTations 7 Relative wind velocity and direction Fig. 1. Ships profile with measuring locations R`2 "', =3_=@@j CH2585-818810000- 1212 $1 (@)1988 IEEE For this purpose a picture camera was 2. THE EXPERIMENT installed on top of the radar screen (fig. 2.) to photograph the pattern of The 'Stuttgart Express' was sailing each rotation of the antenna (fig. 3.). it's regular track across the North Thus the samples were taken in a space Atlantic Ocean between North Europe and time box. the east cost of North America. The ship was equipped with the 'Marine Voyage Recorder' [3). This is a 'Black Box' which provides most of the rele- vant data (such as ship's speed and course, torque and number of revolu- tions, wind speed and direction) for the experiment. 4: Measurements were taken in routine every three hours during the local day. The ship's motions were recorded by four rigidly installed accelerometers. Two of them are sensitive in vertical direction and two in horizontal direc- tion [4]. Figure 1. shows all measuring locations. The accelerations were sampled during a period of 10 minutes with 10 cps and stored digitally. For each measurement the operational data were stored additionally. "N -U&).kk tk, The used method of wave Measurements by means Of d marine radar is described in [2), [51 and [6]. Each wave measurement is composed by a series of 32 consecutive photographs. They are done by taking Photos from the pattern of the rddarbackscdtter field as it is displayed on the PPI, when no ..sed clutter" suppression is switched on. Fig. 3. Photo of radar screen during one rotation of the antenna Table 2. The ship's and environmental data at time of measurement. Date: 3/11/1988 Time: 13:29 UTC Position: 51*N, 27*W Course: 270* Speed: 16.5 kn Draught: 9.5 m Trim: 0.5rn Volume of displacement: 45750 m3 Wind and wave conditions from bridge Zz@ observation. Wind: Bft 6-7 Direction- 160* Waves Hs=4m Direction: 160' Swell T=9sec Direction: 180* Fig. 2. Camera mounted on radar tubus 1213 3. DATA ANALYSIS The purpose of the analysis of the experiment is to obtain the motion A A A A A A AAA AA H AAA spectra and the wave spectrum in- order AVAV AVAV@@V VVVVXJV- to get a 'measured' transfer function. ffVVVV_V1VVM The results are compared with computer simulations. In the further analysis the propulsive data will be analysed 0 60 lio 180 240 and compared with computer calculations Time (see) (especially the increase of resis- tance). c! A A qV VqV V V V V VV V V V V V V VV VV VV A AIV. t C! AX. A, kkd A, V: V -V -Bow Stern 0 so do Time (see) igo 240 300 0 io 120 60 240 Time (sec) A _V_TV V V V V V A J 6. A A A IN VN V@ V V 0 a10 lio tio 2@0 360 deck Time (sec) 0 so 190 lio 240 30. nme (sec) Fig. 5. Computed time-series of the rolling, pitching and heaving motion Fig. 4. Measured horizontal and vertical accelerations In this paper one first example is given. The choosen measurement was 0 taken on march 12, 1988 at 13:29 UTC. -Heave (M-2/cps) The observed wind and wave conditions are g-iven in table 2. ---Pitch (deg--2/cps) The time series of the four accelera- ..... Roll (deg--2/cps) tions are shown in figure 4. From the difference of the two vertical acceler- ations the pitch angle and from the difference of the horizontal accelera- tions the roll angle con be calculated. These calculations are described in [4). For the calculation of the heaving motion the measured accelerations are used together with the before calcula- ........ ted pitching and rolling angle. Figure 0.00 0.05 0.10 0.15 0.20 5. shows the computed series of rol- fe (cps) ling, pitch and heave in the time domain. Fig. 6. Power spectra of the rolling, pitching The spectra of motion are computed by a Fast Fourier Transformation over the and heaving motion hole sample period. The power spectra shown in figure 6. are averaged over 9 frequency bins. 1214 part. Therefore the resulting wave- number spectra are unambiguous. short waves (< 10 cm) 40 M) Time series taken from board a sailing ship are Doppler shifted caused by the radarbackscatter modulation of radar velocity a of the ship. backscatter a If: o = Vgl@ltnh(J@Jh) (2) Y time series of 32 pictures is the frequency of the wave with the from the PPI wavenumber E, this wave will be seen under the shifted frequency: X I digitization and three-dimensional FFF (3) W. = Wo + I from board the sailing ship. Therefore po rspectra the frequency in (1) has to be replaced by the frequency of encounter: F(3) ky,wo) (4) F(3) (k, k,, w,) ime Here the next advantage becomes evi- .9ta-, and transformation dent. The speed of the ship displaces the wave energy parallel to the fre- quency axes. Thus by integrating along the frequencies the Doppler shifted F(')(k_k,) EOI(f,..) F("(f) AW energies are projected to the wave- number plane k,;,,ky unaffected by the Doppler shift: Fig. 7. Flow diagram showing the data reco- (5) F(2) (k., ky) F(')(kr ding and analysis procedure f. 00 , k,, w,)dw,. All the other spectral presentations of the wave spectra, that are calculated The flow diagram (fig. 7.) shows the out of (5) are unaffected by the steps of the used wave measurement and Doppler shift as well. analysis. The sea wave measurement is 3W FU composed by a series of photos. For analysis the pictures were digitised STUTTGART EXPRESS and sent to a computer. 320 11.03.88 13:32 The most important step in the analysis FILM NR.: 31 ZA is the three dimensional FFT. This results in wave spectra defined over 28C SHIPS HEADUP the two wavenumber components and the frequency: 240 F (3) (kx I k1l) wo) e2vsp) (3) 2W ,The index in (1) gives the dimen- 0 sion of the Fourier domain that is used. Two of the three dimensions are 3 WND FROM 0 1? given by the wavenumber components k.,ky and the third dimension is given by the f requency wo. 120 The three dimensional spectral presen- tation is the most complete presen- tation of a wave measurement. Therefore so other kinds of presentations may be calculated from (1) by integrations and transformations (see [51 and [61). To get the two dimensional image spectra the energy (A) has to be inte- grated along the frequencies. Here one d.03 dio dm big advantage of the image time series FREOUENCY @PPlf' - 4@09@@ ou becomes evident. If the integration is done only for positive frequencies, the Fig. 8. Two-dimensional wave spectrum energy is separated from it's mirror 1215 Figure B. shows the two dimensional, The transfer function is defined as the normalized wave energy spectrum as a square root of the spectrum of motion function of frequency and direction. devided by the inciting wave spectrum. The distance of the isolines is 1/10 of For the determination of the frequency the maximum. The auxiliary line is to of encounter (fe) in this case a mean find the Doppler frequency for energy direction for the wave propagation of with the frequency 'FM'. 190* (up to north) was taken. The The one dimensional spectrum is shown ship's course was 270*. This results in in figure 9. As this is the integration an angle of encounter of 100*. of energy given in fig.8. over all The dotted line in figure 10 is the directions for each frequency bin, result of the measured transfer func- there is no Doppler shift in this tion from the choosen example. The spectrum. The absolute scaling for one solid line gives the transfer function dimensional spectrum was estimdted by for pitch and heave calculated with the aid of spectral parameters (i.e.: the modul of the program used in the peak frequency and the Phillips routing model. For these computations constant). the actual trim and loading condition of the ship on it's regular seaway were fitted to the program. For the reason of better comparison the calculations were not only made with the angle of encounter 100* but also with it's 15* neighbours 85* and 115*. H 1/3 6.70 rn Figure 10 shows the comparisons between ansfer the predicted and measured' tr functions. For this one example the 0 ON- -Heave ..... Measurement Prediction 0.00 0.05 0.10 0.16 0.20 frequency (cps) 100* 115* N Co C! ....... 0.04 0.08 0'12 0.16 0.20 0 fe (cps) CQ Pitch ..... Measurement Prediction 0.00 0.05 0.10 0.15 0 .20 frequency (cps) 115' Fig. 9. One-dimensional energy spectrum 100' and mean direction of the seaway 85' 4. RESULTS 0.104 0.08 0.12 0.18 0.20 For a first comparison between the fe (cps) measurements and the computer simu- lation, a 'measured' transfer function for the pitch and heave motion was Fi& 10. Measured and predicted transfer calculated. Pitch and heave are the function for the pitching and motions which are mainly responsible heaving motion for the increase of resistance and thus 10, 115 85. for the loss of speed in the seaway. Therefore the rolling motion is neglec- ted here. 1216 comparison is good for frequencies (21 ZIEMER,F. and W. ROSENTHAL (1987), higher than 0.1cps. In the low On the Transfer Function of a frequency range (up to 0.07cps), where Shipborne Radar for Imaging only a low wave energy was detected, Ocean waves. the measured transfer function Proceedings of IGARSS'87 Symposium, overshoots. But this phenomenon Ann Arbor, May 1987 pp. 1559-1564. continues up to 0.09cps. This result may be a hint that the used [3) WITT, J. (1987), Bergungsfdhiger parameterisation is too rough. The Schiffsdatenschreiber different solutions for the prediction (Marine Voyage Recorder). show how sensitive the transfer Hansa 1987, Nr. 17, pp 1029-1030. function is on the angle of encounter. In this example the direction of the [4] STOCKDREHER,E. and F.ZIEMER (1988), wave field was characterised by only Ubertrdgungsfunktionen fUr Schiffs- one mean value. Figure S. showing the bewegungen im Seegang. Berichtsheft two dimensional wave spectrum zum 9. Duisburger Kolloquium demonstrates how rough this Schiffstedhnik/Meerestechnik. simplification is. If the parameters of encounter would be calculated out off (51 YOUNG, I.R., W. ROSENTHAL and F. the two dimensional presentation, ZIEMER (1985), A three dimensional results should be more coincidentally. analysis of marine radar images for the determination of ocean wave directionality and surface currents. J. Geophys.Res., Vol. 90, 6. REFERENCES pp. 1049-1059. (11 STOCKDREHER, E. (1986), Vergleich [61 ZIEMER, F. (1986), Untersuchung zur gemessener und berechneter Fahrt- quantitativen Bestimmung zweidimen- verluste eines Schiffes im Seegang. sionaler Seegangsspektren aus Mes- Diplomarbeit, Institut fUr Schiff- sungen mit ndutischem Radar. bdu der UniversitAt Hamburg, GKSS-Externer Bericht 87/E/10,151 pp. September 1986 1217 INFLUENCE OF FREQUENCY AND DIRECTIONAL SPREADING ON WAVE TRANSFORMATION IN THE,NEARSHORE REGION Michael J. Briggs & Peter J. Grace Research Hydraulic Engineers Coastal Engineering Research Center USAE Waterways Experiment Station Vicksburg, MS 39180-0631 has been proposed to restore the j etty to its original length. Thus, th@dimensional sta- bility tests of this entrance jetty were con- ABSTRACT ducted in the directional spectral wave basin of -the USAE Waterways Experiment Station's A three-diremional, physical model study of a Coastal Engineering Rese=h Center (CERC) to harbor entrance jetty in Yaquina Bay, Oregon verify this design. Site-specific extreme using spectral and directional spectral waves storm events were hindcast for a nearby deep- was ccapleted. The 1:45 scale model consists of water site and numerically transformed to a an entrance channel protected by two rubble shallow water depth corresponding to the loca- mound jettys with actual bathymetry offshore of tion of the wave generator in the model. Six the entrance. Site-specific wave spectra, rep- unidirectional and six equivalent direc-tional resentative of the most severe hindcast storms spectra were selected from the 20 year hindcast in a 20 year period, were selected for study. A period as representative extremal events. Three range of spectral parameters, including frequen- gages were set up in the basin to measure inci- cy and directional spreading, were modeled. Peak dent and jetty head wave conditions for the sta- prototype periods were 12.5, 14.3 and 16.7 sec. bility study for two water depths. Thus, the Prototype zero-normnt wave heights ranged from effect of depth and frequency and directional 14.4 to 23.0 ft. Surface water elevations were spreading on wave transformation at the jetty measured offshore and near the jetty head for head can be quantified. two water depths. The effects of depth and fre- quency and directional spreading on transforma- This paper includes a brief description of the tion of wave period and.height in the nearshore physical model and test setup. Next, a descrip- region are presented and compared. tion of the six unidirectional and directional spectra test conditions is given. The model wave generation procedure is then discussed. Finally, results are presented showing the ef- 1. IRIR0V=0N fects of transformation on peak wave period and zero-moment wave height. Irregular waves are routinely used in physical model tests of port facilities and breakwaters. Real ocean waves are short-crested, however, and 2- MEEL DESCRIFTICK AND TEST SETUP have directional spreading which spreads energy over many directions. Sand et. al. (1983) meas- Yaquina Bay is an estuary located on the Oregon ured diffracted wave energy in the lee of an en- coast, approximately 110 miles south of the trance breakwater for unidirectional and direc- mouth of the Columbia River (Figure 1). Two tional spectral waves. They found larger waves rubblemound jetties protect the 40-ft deep, 400- for the directional cases. In model tests of ft wide entrance channel, which passes through a wave transformation aver a submerged elliptical narrow opening in an offshore reef. This basal- round, Vincent and Briggs (1988) found signifi- tic reef lies 3500 ft seaward of the channel en- cant 'differences between monochromatic, spec- trance and extends northward for 17 miles. tral, and directional spectral waves. Kirkegaard et. al. (1980) measured motions of a moored A three-dimensional physical model (Figure 2) of vessel near an open jetty. They also found the.Yaquina Bay entrance channel jettys was con- differences in response between spectral and structed in CERCIs 96-ft long by 121-ft wide di- directional spectral waves. Thus, it appears rectional spectral wave generator (DSWG) basin. that the inclusion of directional spreading can The electranechanical DSWG is 90 ft long and have a significant effect on the design of port consists of 60 paddles in four portable modules facilities and breakwaters. of 15 paddles per module (Outlaw and Briggs 1986) - The origin of the local coordinate sys- The north jetty of the Yaquina Bay, Oregon en- te. is at the edge of paddle 1. The y@axis and trance channel has a history of repeated deteri- the north axis are aligned with the DSW.. Three oration and rehabilitation. A 11placed-stonell capacitance wave gages (Figure 2) were used to construction technique with larger armor units measure surface water elevations in the basin. 12W United States Government work not protected by copyright one was used to measure incident wave conditions respectively. The directional spreading for near the DSWG and the other two flanked the head the directional cases ranged between 10 to 30 of the north jetty near the reef. Table 1 lists deg (Briggs, Grace, and Jensen 1988). The dif- the gage coordinates. ferences between the unidirectional and direc- tional spectra are due to variations in input and boundary conditions used in SHAIW. Table I Gage Coordinates Table 2 Gage X-axis Y-Axis Prototype Wave Spectra No. ft ft 1 5.0 45.0 storm TP 11410 E) 2 46.0 42.0 ID sec in deg 3 46.0 45.0 unidirectional series 1969 16.7 22.3 15 An undistorted model scale of 1:45 was selected 1970 14.3 19.0 9 based on several factors including a) DSWG size 1972 12.5 14.4 19 and capabilities, b) required nearshore batby- 1973 14.3 16.7 15 metric area, c) available stone sizes, and d) 1974 16.7 19.0 21 preclusion of stability scale effects. Thus, 1983 16.7 18.7 14 it was not possible to model the entire entrance channel, nearshore bathymetry, and inner harbor. Directional Series Since the main focus of the study was on the 1969 16.7 23.0 10 jetty head area, only 32 ft of the north jetty 1970 14.3 19.0 2 and 21 ft of the south jetty were modeled. The 1972 14.3 15.4 11 nearshore bathymetric features and offshore reef 1973 14.3 17.1 4 were duplicated to the extent that further wave 1974 16.7 20.3 21 transformation prior to interacting with the 1983 16.7 20.3 9 channel entrance was properly modelled. Thus, the DSWG was located at a depth corresponding to 58 ft in the prototype (1.29 ft model). To quantify the effects of depth changes, tests 4. MCCEL WAVE GENERATION were also conducted at a water depth of +10.0 ft ml1w, representative of a storm tide and set-up. The twelve SHAIW storms were converted for The corresponding water depths were 68 ft proto- use in the DSwG using the 1:45 scale. Table 3 type and 1. 51 ft model. Incoming wave enexgy sumnarizes the target model wave parameters. was dauped with passive absorbers at the back end of the entrance channel. The interested reader is referred to reports by Grace and Table 3 Dubose (1988) for detailed descriptions. Target Model, Wave Parameters 3. PIVIUTUIE WAVE SPECIRA storm TP Hipo E) ID sec in deg The Wave Information Study (Corson, et. al.1987) data base was used to hindcast the initial deep- Unidirectional Series water spectra. They were multi-modal in fre- 1969 2.49 5.95 15 quency and direction with locally-generated 1970 2.13 5.07 9 wind-sea and distant swell components. The 1972 1.86 3.85 19 numerical model, SHAIW, (11ughes and Jensen 1973 2.13 4.46 15 1986) was used to transform six "Worst case" 1974 2.49 5.16 21 directional spectra to the shallow water depth 1983 2.49 4.99 14 of the DSWG. This model included source/sink and finite depth mechanisms such as refraction, Directional Series shoaling, wave bottom interactions, and depth- 2.49 6.12 10 related breaking. Six unidirectional and six 1970 2.13 5.07 2 directional spectra were selected from these 1972 2.13 4.11 11 numerically transformed SBAIW results. 1973 2.13 4.55 4 1974 2.49 5.42 21 Table 2 summarizes peak period Tp, zero-moment 1983 2.49 5.42 9 wave height Hmo, and overall mean-wave direc- tion 9 for each case. The angle 9 corresponds to the swell direction and is measured counter- The spectra were converted into control signals clockwise from east. All waves approach from at each of the 61 WwG paddle drives using a the southwest quadrant. Figures 3 & 4 illus- double summation, deterministic anplitude, . ran- trate the range of frequency spreading for the dcm phase, frequency domain method (Briggs, six unidirectional and six directional spectra, Borgman & outlaw 1987). Since the D/A rate for 1219 the DSWG is 20 Hz, time series of 12,000 points to a higher unidirectional period - In general, or 600 sec were generated. An initial calibra- the periods increased more for the unidirection- tion phase was conducted to measure and correct al than the directional cases. The average in- the control signals. A response amplitude cper- crease for the unidirectional cases was 1.08 and ator (RAO) was calculated for each of the twelve 1. 05 for the two depths, respectively - The di- control signals to compensate for observed vari- rectional cases were invariant to depth change, ations in peak period, wave height, and spectral maintaining a constant 1. 01 increase for both shape (Briggs and Jensen 1988). depths 5. EPUR 0011EMON & ANAINSIS Table 4 Measured Peak Periods Wave elevation data was sampled at 10 Hz. A min- imum of 200 waves at the peak period was col- Gage 1 Ave 2 & 3 lected for each case (Goda 1985). The data rec- storm sec sec Normal. ords were zero-meaned and tapered by a 10% cos- ine bell window. For the frequency spectral an-- a) Depth alysis, the data was band averaged with a band- width of 0.0671 Hz within lower and upper cutoff Unidirectional Series frequencies of 0.2013 and 1.476 Hz, respective- 1969 2.46 2.50 1.02 ly. A typical value for degrees of freedom is 1970 2.19 2.18 0.99 68. A Gaussian smoothing technique (Borgman 1972 1.79 2.69 1.50 1984) was used in the directional spectral anal- 1973 2.53 2.68 1.06 ysis within the same frequency limits to calcu- 1974 2.56 2.51 0.98 late directional spreading. 1983 2.68 2.73 1.02 Directional Series 1969 2.49 2.57 1.03 6. WAVE TRANSFUR-TiMCN RESUlaS 1970 2.01 2.08 1.03 1972 2.54 2.47 0.97 only the effect of wave transformation on peak 1973 2.53 2.50 0.99 period and zero-rcment wave height are presented 1974 2.56 2.52 0.98 in this paper. The interested reader is re- 1983 2.47 2.57 1.04 ferred to a report by Briggs, Green, and Grace (1988) for-coirplete results from all tests. b) Depth = 1.51 ft Table 4 lists the measured peak periods for the Unidirectional Series unidirectional and directional cases at the two 1969 2.46 2.50 1.02 depths for the incident gage I and the average 1970 2.19 2.18 0.99 of gages 2 and 3 at the jetty head. The normal- 1972 2.15 2.73 1.27 ized value is the ratio of the average period to 1973 2.53 2.65 1.05 the incident period. 1974 2.56 2.47 0.97 1983 2.53 2.67 1.06 Figure 5 shows the influence of depth and direc Directional Series tional spreading on peak period transformation. 1969 2.49 2.58 1.04 Normalized directional peak period is platted 1970 2.01 2.11 1.05 against normalized unidirectional peak period 1972 2.54 2.47 0.97 for five of the storms. The 1972 storm is not 1973 2.51 2.50 0.99 plotted because it lies off the edge of the 1974 2.56 2.52 0.98 plat. The left-most symbol represents the lower 1983 2.47 2.56 1.03 water depth. The 45 deg line is shown to indi- cate equivalence between unidirectional and di- rectional periods. A value above the line im- Table 5 lists the measured and normalized zero- plies that the directional period is larger than moment wave height for the twelve storms for the the corresponding unidirectional period for the two depths. Figure 6 shows the influence of same depth. Conversely, a value below the line depth and directional spreading on wave height indicates the opposite relationship. A horizon- transformation for the two depths. The same tal line between points on the graph (i.e. 1973, relationships described for the wave periods in 1974) indicates a constant directional period Figure 5 apply for the heights in this figure. which is uneffected by changes in water depth. Also, a line parallel to the 45 deg line indi- Likewise, a vertical line between points (i.e. cates no change in the relationship between 1969, 1970) indicates a constant unidirectional unidirectional and directional storms due to period. The increase in depth for the 1972 depth (i.e. 1969). A decrease in the slope of storm caused the unidirectional period to de- the line connecting two depths indicates an crease while the directional period remained increase in the unidirectional height relative constant. This storm was the only one to e>per- to the directional height due to an increase in ience a significant increase in period due to depth (i.e. 1970, 1972). Similarly, an increase transformation in shallow water. The 1983 stonn in the slope of the line shows an increase in is interesting because the increase in depth directional height relative to the unidirection- caused a shift froin a higher directional period al value (i.e. 1973, 1974, 1983). The trans- 1220 formed wave heights at the jetty head were smal- type). These storms had peak periods of 12.5, ler than the incident conditions at gage 1. As 14.3, and 16.7 sec and zero-nKment wave heights expected, the heights increased for all cases ranging from 14.4 to 23.0 ft. The twelve storms due to depth increase. The unidirectional cases were calibrated and tested at two water depths: were higher than their corresponding directional 0 and +10.0 mllw. cases except for the 1970 storm at low depth. The average normalized height for the unidirec- Results from this limited series of tests with tional cases was 0.83 and 0.94 for the two one incident gage and two gages near the jetty depths, respectively. For the directional head confirm previous investigators findings cases, this average increased from 0.75 to 0.87 that directional spreading can have a signifi- due to the increase in water depth. cant effect on the modeling of wave environments in the nearshore region. In general, the per- iods shifted more for the unidirectional than Table 5 the directional case due to wave transforma- Measured Wave Heights tion. The period was not affected by depth changes for the directional cases. wave heights Gage 1 Ave I & 2 for all cases were reduced at the jetty. In- storm in. in. Normal. creases in depth produced increases in the transformed wave heights. Except for the 1970 a) Depth = 1.29 ft storm at Oft mllw, the transformed unidirection- al heights were always larger than their corres- Unidirectional Series ponding directional values. 1969 5.87 4.13 0.71 1970 5.15 4.08 0.79 1972 3.89 3.96 1.02 8. ACPU14LEDGEMENTS 1973 4.35 4.20 0.97 1974 5.09 4.10 0.81 The writers wish to acknowledge the office, 1983 5.12 4.07 0.80 Chief of Engineers, USAE, for authorizing publi- Directional Series cation of this paper. It was Prepared as Part 1969 5.46 3.84 0.70 of the Laboratory simulation of spectral and 1970 4.83 4.04 0.83 Directional Spectral Waves work unit, Coastal 1972 4.19 3.93 0.94 Flooding and storm Protection program of the 1973 5.23 3.93 0.76 Civil Works R & D program. We would also like 1974 5.65 3.87 0.68 to thank R. E. Jensen, D. L. Green, D. A. Daily, 1983 5.93 3.98 0.67 W. G. Dubose, and W. D. Corson for their help. b) Depth = 1.51 ft 1 Unidirectional Series 1969 6.79 5.65 0.83 Borgman, L. E. 1984. "Directional Spectrum 1970 5.89 5.59 0.95 Estimation for the Sxv Gages I " USAE Waterways 19-72 4_5_3 5.16 1.14 Experiment station, ILE -Borgman, Inc., November. 1973 5.23- 5.43 1.04 1974 5.89 5.00 0.85 Briggs, M.J., Borgman, L.E. and Outlaw, D.G. 1983 5.99 5.42 0.90 1987. "Generation and Analysis of Directional Directional Series Spectral Waves in a Laboratory Basin," OTC 1969 6.07 4.93 0.81 5416, Offshore Tecbnology Omference, Houston. 1970 5.55 5.04 0.90 1972 4.89 5.06 1.03 Briggs, M.i., Grace, P.J., and Jensen, R.E. 1973 6.11 5.43 0.89 1988. "Directional Spectral Wave TransfOrma- 1974 6.32 5.01 0.79 tion in the Nearshore Region, Volume I: 1983 6.74 5.58 0.83 Directional Spectral Performance CharaCteri- stics," USAE Waterways Experiment Station, Technical Report (in pub.), Vickdmirg, MS. 7. CONCIUSIONS Briggs, M.J., Green, D.R., and Grace, P.J. 1988. "Directional Spectral Wave Transforma- The effects of depth and frequency and direc- tion in the Nearshore Region, Volume II: Jetty tional spreading on wave transformation in the Head Couparisons," USAE Waterways Experiment nearshore region were investigated using a 3D Station, Tech. Report (in prep-), Vicksburg, Ms. physical rcdel of Yaquina Bay, Oregon. The 1:45 scale model consisted of an entrance channel Briggs, M.J. and Jensen, R.E. 1988. "Simulation protected by two nibble mound jetties with an of Extreme Storm Events in Coastal Structural irregular bathymetry and submerged reef offshore Models," OSDS 88, Corvallis, OR, September. of the entrance. Six unidirectional and six equivalent directional spectra, representative Corson, W.D., et. al. 1987. "Pacific Coast Hind- of the most severe hindcast storms in the past cast Phase II Wave Information," WIS Report 16, 20 years, were numerically transformed to a USAE Waterways Experiment Station, Vicksburg. water depth of 58 ft (all units proto- 1221 Goda, Yoshimi. 1985. RVa= SEAS AND EESIGN OF NU=ME SHIMMUS, University of Tbkyo Press, Japan, pp. 1-323. Grace, P.J. and Dabose, W.G. 1988. "Jetty Rehabilitation Stability Study, Yaquina Bay, Oregon," Tech. Report (in pub.), USAE Waterways Experiment Station, Vicksburg, MS, pp 1-57. Hughes, S.A. and Jernen, R.E. 1986. "A User's Guide to SHALW: NLmierical Model for Simulation of Shallow Water Wave Growth, Propagation, and Decay," IR CERC-86-2, USAE Waterways Experinient Station, Vicksburg, MS. Kirkegaard, J., Sand, S.E., Ottesen-Hansen, N.E. and Hvidberg-Knudsen, M. 1980. "Effects of Di- rectional Sea in Model Testing," Forts 180, Norfolk, VA. Outlaw, D.G. and Briggs, M.J. 1986. "Direction- al Irregular Wave Generator Design for Shallow Wave Basins," 21st ATX, Washington, DC, Aug. Sand, S.E., Kirkegaard, J., Larsen, J. and Rodenhuis, G.S. 1983. "Numerical and Physical Modeling of Directional Diffraction of Waves," Coastal and Port Erx3ineering in Developing Cmmtries, Colombo, March. Vincent, C.L. and Briggs, M.J. 1988. "Refraction Diffraction of Irregular Waves Over a Mound," ASCE WPCOE Journal (In press), New York. 0 Q L HILL CO RIVER POLK CO SILETZ 011-1-.1 N@ BA Y z 0 DALLAS 4# ROOF SUPPORT 43.9' SALEM POLES 7, DEPOE BAY A POLK BENTON CO IVER YAOUINA BA Y-@ ALBANY cc CORVALLIS X Cn ..Cl) ALSEA __j a BA Y x ALSEA ---------- LINCOLN CO BENTON CO y L E WAVE MAKER SCALE 5 0 5 10 MI ABSORBER IROOF P7ORT SUP POLES 120.8' Fig. 1. Yaquina Bay, Orelalon Fig. 2. Physical Model Setup 1222 1000 1000 STORM YEAR STORM YEAR C3 1969 E]= 1969 C) 1970 0 = 1970 & 1972 A,= 1972 750 + 1973 14 750 + = 19@3 ?@4 x = 1974 X 1974 1 o = 983 Cu 0 1983 500 u- Soo 250 250 0 0 0.-0 0.1 0.2 0.3 0.0 0.1 0.2 0.3 FREQUENCY, HZ FREQUENCY, HZ Fig. 3. Six Prototype Unidirectional Storms Fig. 4. Six Prototype Directional Storms C3 1. 10 1.2 @TDRM @ERR STORM YEAR LLJ ui E]= 1969 (L ID= 1969 X: 0 = 1970 (1)=1970 a: 1.05 + = !Q73 = 1972 Lu Cr += 1973 X 1974 :9 1983 1.0 X= 1974 a: a: q5 DEG 1 0= 1983 z z --= 45 DEG 1-00 A Lj Lu cc C@ A im 0. 8 Cy- 0.95 uj j Cr a: X: X: o 9 o L 2 0.6 z 0.90 0.95 1.00 1 . 105 1.110 0.6 0.8 1.0 1.2 NORMALIZED UNIDIRECTIONAL PERK PERIOD NORMALIZED UNIDIRECTIONAL WAVE HEIGHT Fig. 5. Peak Period Transformation Fig. 6. Wave Height Transformation '9- '9- '9' 1223 TIDAL CIRCULATION DATA FROM THE LOS ANGELES/LONG BEACH HARBORS David D. McGehee and J. P. McKinney Department of the Army Waterways Experiment Station, Corps of Engineers P.O. Box 631 Vicksburg, Mississippi 39180-0631 ABSTRACT Coastal Enaineer Research Center (CERC) of WES. This report describes the methodology and results A synoptic data collection effort was undertaken of the data collection effort. between June and October 1987 at the Ports of Los Angeles and Long Beach, CA. Three months of II: DATA COLLECTION tidal data from offshore and within the harbor, 1 month of current data and half tidal cycles of The general requirements and schedule for each current profiles throughout the harbor were task in the program was prescribed in the Manage- obtained. The data are to be used for model ment Plan for the Model Enhancement Program. verification and calibration. Data collection was divided into three subtasks: tidal data, in situ current data, and current Tidal elevations were referenced to MLLW by anal- prof ile data. Data were collected for varying ysis of average water surface elevations at each intervals between 10 June 87 and 14 October 87. site and adjustments for variance in monthly mean Measurement intervals for each subtask varied, sea surface elevations from MSL. Excellent but deployment times were nested to provide a agreement is obtained between this method and period of simultaneous data from all three ele- traditional surveying from a benchmark. ments from 6 to 14 August 1987. Requirements of the final data sets and the instrumentation and The method of collecting and analyzing the data techniques used to obtain them are discussed in is described. Samples of plotted data are pre- this section. Figure 1 is a site map showing sented, and hydraulic gradients are compared with deployment locations of all instruments. resultant water velocities. Tidal Data The requirement of this subtask was to obtain time series of water surface elevations of two I: INTRODUC-TZOX outside and two inside locations for a minimum of 90 days. Each data set was to be a continuous The Ports of Los Angeles and Long Beach, Cali- record of elevations relative to a Mean Lower Low fornia are conducting planning studies for harbor Water (MLLW) datum at six minute intervals with a development in coordination with the Los Angeles total error of no more than tO.05 ft. Eight tide District of the US Army Corps of Engineers (COE). gages, consisting of a bottom mounted pressure The COE is charged with responsibility for pro- transducers with self contained power and inter- viding deeper navigation channels and determining nal data storage capacity (Sea Data 635-11) were effects of the harbor expansion on the environ- deployed. Each gage was attached with stainless ment. To upgrade the Corp's capability to deter- steel bolts to a vertical mount which was welded mine these effects based on state of the art to a 600 lb. railroad wheel. A subsurface buoy modeling technology, the US Army Engineer Water- was attached to the wheel with a length of re- ways Experimient Station (WES), is executing the treival line coiled in a cannister. The cannis- Los Angeles/Long Beach Harbors Model Enhancement ter, in turn, was attached to the wheel with a Program. transponding accoustic release. The release served the dual purposes of a transponding The program is separated into two major studies. beacon, for locating the instrument package, and The first will address the question of long as a means of releasing the buoy and recovery period wave energy in the Harbors and its effect line at the end of the deployment. on moored vessels. The second will provide improved tidal circulation modeling with a more In addition to the 8 tide gages specifically efficient numerical model system which will deployed for this study, data were available from couple hydraulics rind water quality variables. a National Ocean Service (N03) primary control The prototype data for model calibration and tide station (#33) located inside Los Angeles verification has been collected by the Prototype Harbor, and from four pressure sensors located Measurement and Analysis Branch (PMAB) of the around the harbor perimiter. These four pressure 1224 United States Government work not protected by copyright LOCATION MAP 7 0 r S4N SA 0 7 3 3 5 PORTOFLB 6 6 A OF LM @2 4 BAY PEDRO SA14 LEGEND 0 TIDE GAGE 0 CURRENT METER STRING -0- CURRENT PROFILE RANGE & NOS TIDAL STATION & CE FIC WAVE GAGE Figure 1. Site map with instrument locations. sensors were installed by CERC to obtain long Velocity is measured by counting impeller revolu- term measurements of harbor surge events under a tions over the selected averaging interval, and separate subtask of the Model Enhancement Program an instantaneous direction taken at the end of entitled Wave Data Acquisition. They were origi- that interval. Both values, along with tempera- nally configured as low frequency wave gages, but ture and conductivity are recorded on a magnetic post processing of the time series produced con- tape. The averaging interval selected was two tinuous tidal data. minutes, allowing a total tape capacity of 34 days. These data were processed to provide additional boundary conditions for the model and in quality Each meter was calibrated prior to deployment at control checking of the primary tide data sets. the National Space Technology Laboratory, NSTL, MS, calibration facility by tank tow at known Currents velocities through still water. The calibration coefficients for each impeller /bearing combina- The requirement for this subtask was a 30 day tion were used in post-processing the data. record of the vertical velocity profile at the Compasses were bench checked and accepted if major tidal Pxchange openings inside the harbor within manufacturers tolerances without indivi- and at the harbor complex perimeter. Desired dual corrections. accuracy in speed and direction was �0.1 knot and 2 degrees, respectively. Samples were needed at In a typical current meter string (CM), a spher- a similiar frequency as the tidal data, that is, ical 36 in. diameter steel buoy is attached to a continuous time series at average intervals of 900 lb. railroad wheel with a 1/4 in. diameter three minutes (Fig. 1). mooring cable. A taut moor is maintained in spite of the tidal variation by including a The vertical profiles were obtained by deploying length of 1 in. diameter rubber cord below the a total of 19 meters, with up to three current buoy. The cable was attached at either end with meters on a string supported by a surface buoy on 3/8 in. screw-pin shackles. Each buoy supported a taut mooring. To reduce the risk of collision, an 8 ft. mast with a radar reflector and amber the gage sites were moved to the sides of major marker light. For additional visibility, two or entrances and channels. Spatial coverage was three "guardian" buoys of similiar design (but obtained at 9 sites by reducing the number of without instruments) were placed around the meters to two or one at those sites where strong instrued buoy approximately 200 ft. away. vertical gradients were not expected. To allow the meter to rotate around the mooring The current meters were ducted impeller type with as current directions changed without fouling on an internal compass for direction (Endeco 174). the mooring, a split teflon bearing sleeve was 1225 attached to the 1/4 in. cable at the appropriate conclusion is that the meters were stolen by depth for each string. A hinged stainless steel persons using dive gear. clamp went around the bearing and locked with a captive hinge pin. The meter was attached to a The buoy from CM site #2 was located on board a ring on the clamp with 3/8 in. screw-pin shackle, commercial derrick barge moored adjacent to the safety wired after closure. This arrangement, meter site. The elastic section of the mooring permitted rotational freedom and a means for had parted when the contractor attempted to move diver changeout of the gage during inspection the buoy to accomodate repair work on a nearby without recovering the entire mount. wharf. Divers were able to recover the three meters, which were still attached to the mooring Profiling meters were of a similiar type to the cable and the weight at the deployed location. in situ meters (Endeco 174 SSM), but equipped with a cable f or telemetry of the data to the For the current profiles, each range was profiled surface. Profiles were taken from a 26 ft. on the scheduled stage of tide to include a peak vessel between 6-14 August 87. A vertical pro- f lood or ebb and a tidal reversal. When a range file consisted of 3 measurements, taken near contained only three stations, each station was bottom (B), at mid depth (M), and near surface profiled at approximately one hour intervals, as (S) at one location, called a station. A range planned. The one hour return interval proved is a transect across an entrance or interface less attainable when a range contained five sta- between major harbor sections with three to five tions, and return intervals varied from 1-1/2 to stations. Profiles were scheduled to be taken at 2 hours. Increased travel time between stationst hourly intervals for 13 continuous hours for each increased anchoring time in deeper water, and the range. occasional instrument malfunction all contributed to the overall increase in return intervals. Data Recovery Overall data recovery from the current meters was -75% of the measurement plan goal. Positions were obtained for all measurement sites with LORAN, supplemented when possible by visual III: Data Analysis and radar bearings to identifiable shore fea- tures. The LORAN was calibrated for local vari- Tidal Data ance by occupying known locations with the boat and recording both time differences (TD) in Raw data from the recovered tide gages were pro- microseconds and the LORAN's automatic conversion cessed using SEA11.FOR, a program which converts to latitude and longitude. The offset from true raw ASCII data into tide, time, and wave record position was entered into the microprocessor con- files. The tide file at this stage is a time trolled LORAN for automatic adjustment of calcu- series of absolute pressure values. Atmospheric lated position. After this calibration, known pressure taken at Long Beach Airport at hourly positions were again occupied on each day of intervals, was used to recover sea level pressure operation. Thus, a correction factor for each and the resultant time series converted to a region of the harbor, valid for each day's atmos- depth time series through multiplication by a pheric conditions, was obtained that allowed an constant of 2.246 ft/psi. improvement from LORAN's typical accuracy of t1OO meters to �10 meters. The required datum for all tidal time series was MLLW. Since it was impractical to level each Of the 8 tide gages deployed, 7 were recovered. gage from shore using traditional surveying tech- Of these, 2 experienced electronic failures and 1 niques, an assumption was made that the average experienced a tape drive failure. Since four of free surface elevation over the deployment inter- the gages were redundantg tidal data recovery was val was constant throughout the harbor region. 100%, in regard to the requirement for two out- This is valid over a limited area with insignifi- side and two inside gages operating during the cant fresh water inflow and where no net trans- period when the in situ meters and the profiling port can occur within the area over the time meters were operating. interval. Of the 18 in situ current meters deployed at Mean Sea Level (MSL) is defined as the arithmetic 9 sites, 13 were recovered between 9 & 12 Sept. mean elevation of the sea surface over a specific The surface buoy at site #9 was found adrift on 19 year metonic cycle.' Means calculated over 21 Aug. A dive team failed to locate the meter shorter intervals will vary, and are calculated or its mooring. Due to the proximity of ship and by the National Ocean Service (NOS) for monthly tug traffic in the area, this buoy and mooring and annual departures f rom MSL at each primary were likely pulled off station by a vessel. The control tide station.2 Thus an average free sur- remaining unrecovered gages were the upper meter face elevation for the deployment interval is not of CM site #4, both meters at CM site #5, and the necessarily MSL. single meter at CM site 418. The buoys at these sites were intact, the moorings undamaged, but Each tidal depth time series had the average the meters were removed at the shackle connecting depth of that time series subtracted (demeaned), the meter to the swivel. Since the shackles and converting it to a time series relative to the their safety wires had been individually in- average free surface at that site over the spected a week after deployment, the most likely deployment interval. If the exact elevation of 1226 the free surface at any time and place (relative recovery are apparent from pressure records and to MLLW or to any arbitrary datum) was of primary shut off time from the last time word in the importance, then the average surface from each series. A.arepment within the 3.75 minute sam- gage over a selected interval (for eKample, pling interval was considered adequate monthly) could be compared with the average verification. departure of the nearest NOS primary control tide station during the same interval. In fact, a Tidal accuracy is addressed in more detail under primary control tide station, #33, is located Part IV Discussion. Figure 2 is an example of a inside Los Angeles Harbor (see Fig. 1). tidal elevation time series from gage #1, located outside of the harbor breakwater for the period In the numerical model, the elevation of the free 4 August to 9 September. surface of each tide gage relative to other tide gages is the forcing function that drives the circulation of the model. Since the monthly LA/LB TIDAL CIRCULATION STUDY. TGIS average varies, its use as a datum would produce 8.0- 8/4/87 00000 - 9/9/87 00000 discontinuities in the records each month. A constant datum, such as MLLW, for all gages over the entire interval is appropriate and produces a 6.0- continuous record. 4.0 To convert the datum to MLLW, the monthly mean LU sea levels relative the MLLW for tide station #33 0 were obtained from the Tides Analysis Branch of < 2.0 3 NOS for June through October. These monthly levels were again averaged to obtain a single 2 us offset representative of the deployment interval, x 0.0 3.09 ft. This was added to the demeaned depth time series to obtain the tidal elevation above MLLW at each site. -2.0- 4 8 12 16 20 24 28 1 6 9 To validate the supposition of constant average 4 AUG - 9 SEP 1987 sea level throughout the harbor, monthly means Figure 2. Tidal elevations, site gage 1. relative to MLLW were also calculated for each tide gage. Means, the average mean and the dif- Current Data ference in means between months (A) are listed in Table I for the NOS tidal station #33 and the For the in-situ meters, raw data in ASCII for- CERC tide gages. Direct comparison is only valid mat is converted to separate velocity (feet per for thos-e months when th-e gage was operational second), direction (degrees from magnetic north), for nearly the entire month. temperature (degrees C), and conductivity files. Direction data words were converted to true north The cumulative average water levels at each sta- by adding the magnetic variation specific to the tion are within.-O-03 ft. The individual monthly LA/LB area, -13.55 decimal degrees., Clock accu- averages show more variation, though the differ- racy was verified and adjusted as required by ences, between months indicate the same trends comparison with recovery and shut-off times. One are occurring at each site. meter had a significant clock error and was not processed, leaving a total of 12 useable data To verify clock accuracy (nominally t1ppm), de- sets. ployment, recovery and shut-off times, as re- corded in the field, were checked against the The current profile data required no analysis data sets indicated times. Deployment and since it was read directly in engineering units Table 1. Monthly Mean Sea Surface Levels June-Oct 87, Sea Level Variations and Differences (A) Elevations in ft. rel to MLLW Month A TG-1 A TG-3 A TG-6 A -TG-Y-- A June 2.88 0 21 2 95 0.15 2.90 0.22 July 3.09 0:05 3: 10 0.05 3.12 0.03 Aug 3.14 3 .15 3.15 2.99 2.99 Sept 3.23 0.09 3.14 0.15 3.15 0.16 3. 0 7 Cum. hver. 3.09 3.07 3.06 0 3. 7 1227 velocity were visually averaged during observa- two gages by subtracting one time series from tion of the output. another and is a plot of the instantaneous hy-. draulic head existing between them. Since each IV. DISCUSSION data set was de-meaned independentlyp the depar- ture from zero of the residual from two gages at Before examining implications of the data col- the same location is an indication of the overall lected, certain investigations should be con- accuracy of the instruments. The beat approxima- ducted regarding its confidence level. Due to tion to this condition occurs with sites 1 and 6, time constraints, a rigorus statistical analysis located outside the harbor breakwater at approxi- will be postponed until subsequent reports, but mately the same depth. direct observations of trends and selected samples will provide some criteria for evaluation The residuals between sites 1 and 6 are plotted of the validity of the measurements. General in Fig. 4, The mean of the residual is included, characteristics of the observed harbor circula- which would theoretically be equal to zero over a tion patterns will be discussed. sufficiently long time interval if the assump- Tidal Data tions made in selecting the datum are valid. Perhaps the most basic concern is the overall LA/LB TIDAL CIRCULATION STUDY. RESIDUAL- TGIS - T068 shape of the tide curves over the deployment. 814/87 00000 - 9/9/87 00000 Classic semi-diurnal behavior is evident in Fig. 2. Another obvious test is to compare the measured tidal data with the predicted tide for the same period. Exact, agreement is never ob- tained, but given the proximity of the tide station, and by selecting periods with low atmos - Z 0A pheric anomolies, a close agreement can be 0.0 expected between observed and predicted tidal elevations. Figure 3 shows the predicted tide for 7-8 Augus t4 (average atm. press. 29.8 in Hg, average wind speed < 7.5 kt) overlaid with measurements from MEAN - -0.00111 wave gage LA4, located several hundred ft. away. A shift upward of the measured data, on the order 4 a ig io i4 is 'I a 9 of 0.3 ft.j_ is evident while the overall range of 4 AUG - 9 SEP 1987 Figure 4. Residual time series; tide LA/LB HARBOR STUDY, STATION LA4 gages 1 to 6. 8/7/87 00000 - 8/9/87 00000 The means of all residual pairs, which range from 0.0001 ft to 0.0038 ft, are the result of can- 6.0- cellation of variance somewhat larger at any one time, but indicates an overall accuracy commen- 2 4.0- surate with the stated specifications of the ku and well within the experimental > sensor 0 requirements. co 0 2.0- 0 X Unlike the primary and back-up tide gages, the a M wave gages were installed nearshore on harbor 0.0- structures, and were accessible (via diving rodmen) to standard leveling.5 Wave gage LB4 was -2.0 PREDICTED TIDE surveyed in on two occasions to a nearby bench- 7 a 9 mark. The average of the two surveys -is 7 - 8 AUG 1987 17.60 ft. below MLLW. Figure 3. Predicted tide versus measured tide, Two data sets, A and B, cover the period of con- gage LA4 for August 7 and 8. sideration. A simple average of their two mean 8.3 ft is- matched exactly. The average August water depths gives departure from the annual sea level for NOS tide [-20.65 + (-20.77)] / 2 = -20.71 station #33 between 1963 and 1981 is + 0.20 ft. It would be impossible to accurately distribute Adjusting to ML1W by the average difference used the remaining 0. 1 ft of difference between gage in the previous analysis: error and predicted tide error. -20.71 + 3.09 = -17.62 A better indication of the final accuracy is obtained by comparing two tide gages at the same The agreement within 0.02 ft is less than the location. A residual is calculated between any variance of the two su rveys. 0 4228 Current Data residual abruptly goes from positive (flood) to negative (ebb). The current reverses at the same Rose plots, which display current velocity and hour, increasing to maximum ebb 8 hours later, direction as polar vectors, with shading repre- when the residual reaches an extreme of -0.4 ft. senting percent of occurance, provide the most condensed display of the current data for sta- tistical purposes. The plots for meters #3B and 2M (Fig. 5) indicate predominant flow in line with the orientation of the nearby retaining SCALE: 1.0 FPS VECTOR POINTII IN DIRECTION Of TRAVEL walls. Additionally, a net mass transport, or I N circulation, in a counter clockwise direction 1A through the Cerritos Channel is apparant. CM2M: AAA, w6 .111@1 JA@@Al' p, q1 -.row Ef -0.0 -0.4 7 7 - 4 AUG 1987 Figure 6. Residual time series tide gages 1 to 3 LEGEND PERCENTAGE and resultant current vectors, meter 2S on 0.26 FPS OF SAMPLES August 7 and B. ... 0-02% 2-10% -------- 10-20% Further evidence of the net clockwise flow in the > 20% 111111111 Cerritos Channel is evident, since the flows out- CM30: ward at the Western opening clearly surpass the flow inward. 111 Conclusions I w Et A synoptic data collection effort at Los Angeles/ Long Beach Harbors was completed that provides adquate data to calibrate and verify a three dimensional numerical model of tidal circulation. Three months of tidal data, 1 month of current data and half tidal cycle current profiles were obtained throughout the harbor. Project require- VECTORS IN DIRECTION OF TRAVEL ments and schedules as directed in the Management Plan were fully met. Figure 5. Current vector rose plots meters ?,'I and 3B. Conclusions resulting from the study are: The skewness observed in the rose plots are the Use of a common mean water surface as a datum for result of statistical representation of cumula- synoptic tidal measurements over limited space tive events, and do not imply a flow occuring at and duration provided reasonable results in this any one time. To see the instantaneous currents, situation and is a cost effective alternative to time series representations are required, as independent leveling of offshore gages. shown in the next section. Tidal circulation in LA/LB Harbor during the col- To observe details of the harbor flow patterns lection period was characterized by sma .11 scale and to illustrate the correlations between sites spatial and temporal variations, including strong at simultaneous times, currents must be observed vertical stratification. Currents are shown to at a smaller scale. Two nested windows were be in correct orientation and phase with the mea- selected for detailed observation: the 7th and sured surface elevation slopes. 8th of August, and 0000 hours to 0400 hours on 8 August. At these times scales, it is also con- Flow in the Cerritos Channel was basically venient to combine the velocity and direction divergent/ convergent from the two openings, but information into a single vector time series sufficient amplitude and phase differences plot. Figure 6 shows the 2 day residual between existed to result in a net circulation counter- tide gage 1 and 3, with the simultaneous vector clockwise. A migrating node existed at the back plot of current meter 2S. At 1800, 7 August the of the channel. 1229 V: ACKNOWLEDGEMENTS 3. Scherer, Wolfgang. Personal Conversation, Feb. 1988, US Department of Commerce, NOAA, The tests described and the resulting data pre- Natioal Ocean Service, Sea and Lake Levels sented herein, unless otherwise noted, were Branch. obtained from research conducted under the Los Angeles and Long Beach Harbors Model Enhancement 4. "Tide Tables 1987," 1986, US Department of Program of the United States Army Corps of Engi- Commerce, NOAA, National Ocean Service. neers by the US Army Engineer, Waterways Experi- ment Station. Permission was granted by the 5. Hicks, Steacy D.; Morris, Phillip C.; Chief of Engineers to publish this information. Lippincott, Harry A.; and O'Hargen, Michael C. "User's Guide for the Instal- VI: REFERENCES lation of Bench Marks and Leveling Require- ments for Water Leveling Stations," 1. Harris, D. L. "Tides and Tidal Datum in the Oct. 19879 US Department of Commerce, NOAA, United States," Special Report No. 7, Feb. National Ocean Service. 1981, US Army Engineer, Waterways Experiment Station, Coastal Engineering Research Center. 2. Harris, D. L. "Tides and Tidal Datum in the United States," Supplement to Special Report No. 7., Feb. 1981, US Army Engineer, Water- ways Experiment Station, Coastal Engineering Research Center. 1230 WAVE IMPACT FORCES ON THE JONES ISLAND EAST DOCK, MILWAUKEE, WISCONSIN Steven L. Da Costa and Joseph L. Scott CH2M HILL, INC. P.O. Box 91500 Bellevue, Washington 98009-2050 ABSTRACT level but subjected to impact of large storm waves. A brief and inadequate discussion of uplift forces On December 15, 1987, a moderate storm on Lake Mi- is given in Design Manual E)M-26.2 [9] . Only ten chigan generated waves characteristic of a 2- to papers (in English) were found in the research 4-year return period. These waves propagated literature (since 1960) applicable to the problem. through the breakwater entrance at Milwaukee, Wi- None of them address the specific configuration of sconsin, and moved partially constructed concrete interest--a cantilevered platform.backed by a ver- slabs at the Jones Island East Dock. The wave tical curvilinear wall. characteristics of the December 15, storm were hindcast and compared to calculated design waves Recent experience with an extended dock platform for the structure. The uplift forces due to the has suggested a method to approach such a design storm waves were also hindcast and compared to com- problem and has also illustrated the requirements puted wave force envelopes. The comparison sug- for further research needed. Much of the, admit- gested that the slowly varying uplift forces for tedly scarce, literature considers only the slowly which the dock was designed would not have produced varying forces to be important for design purposes. the motion. It was concluded that the rapidly However, it has become apparent that the short du- varying vertical impact forces, amplified by the ration, rapidly varying, forces resulting from the generation of standing waves, caused the observed impact of the wave on the underside of the platform motion. The results show that further research is can be an important design criteria. required regarding impact forces encountered by cantilevered horizontal platforms. The need to The nature of these forces may be similar to the consider these forces for future designs of hor- more familiar horizontal breaking wave forces on izontal platforms is indicated. vertical planar structures, but some significant differences appear to exist. The wave need not be breaking in order to generate rapidly varying, INTRODUCTION vertical impact forces. the effect is more related to the slamming forces on a ships hull. Further- The engineer involved with coastal structures and more, if the horizontal platform is backed by a shoreline facilities is almost always faced with vertical wall, standing waves are formed. This the problem of predicting and designing for the will significantly increase the uplift forces (both impact forces of incident waves. There exists a slowly and rapidly varying). The efficiency of the large amount of information on wave forces on ver- vertical wall in reflecting wave energy and the tical structures subjected to breaking, broken, and manner in which it is reflected is an important nonbroken waves. This information is readily a- design consideration. vailable to the design engineer in standard sources such as the Army Corps of Engineers Shore Protec- Available Literature tion Manual (11 and the Naval Facilities Engineer- ing Command Coastal Protection Design Manual As previously mentioned there appears to be little DM-26.2 [9]. Many reference texts and a large or no useful information in the design manuals or amount of published literature also address such handbooks available to the designer of coastal forces. However, little information is available structures concerning rapidly varying uplift regarding wave forces in the vertical direction on forces. There also appears to be little cross-over overwater horizontal structures. from the field of naval architecture to coastal engineering. Information from studies on hull The lack of information about vertical or uplift slamming forces is the most likely avenue to fur- forces on horizontal structures makes it difficult ther research targeted for the coastal engineer. for the engineer to consider the widest possible range of options when attempting to select the best The type of information available, in English, is design for a particular application. The case con- presented in Table 1. This list is not meant to be .sidered in this paper is that of a cantilevered exhaustive but it does appear to he representative. horizontal platform extending seaward of a vertical Table 1 shows that most of the available informa- curvilinear wall; the platform is above stillwater tion is from model scale experiments and only two CH2585-8/88/0000- 1231 $1 @1988 IEEE Table I REPRESENTATIVE RESEARCH ON WAVE UPLIFT FORCES Platform Problem Forces SWL-Platform Structure Treatment Studied Depth Distance Vertical Scale Prototype Slowly Rapidly Date Author Title - OPen Wall Model __@cale Analytical Varyin Varyin Constant Variable Constant Variable 1963 Elghamry Wave Forces on X X [31 a Dock 1966 Furudol and Wave-Induced X X X X X X X Nurota Up-Lift Forces [7] Acting on Quay-Aprons 1967 Wang Wave Pressures on X [13] Horizontal Pier 1967 Verhagen Impact of Flat X X X [12] Plate on Water (Cross-over from Naval Architecture) 1969 French Wave Uplift X X X [51 Forces on Horizontal Platforms 1970,Wang Water Wave X [14] Pressures on Horizontal Plates 1971 Elgbamry Uplift Forces X X X X X X [4] on Platform Decks 1978 Tanimoto and Wave Forces on X X X X X X Takahashi a Horizontal [11] Platform 1979 French Wave Uplift X X X X X (61 Pressures on Horizontal Platforms 1986 Lee and Lai Wave Uplift on X X X X X X [8] Platforms or Docks papers consider a cantilevered horizontal platform (JIED). The facility is located on Lake Michigan backed by a vertical wall. An examination of these but is protected to some degree by a breakwater, as papers clearly illustrates that much more research shown in Figure 1. The dock is subjected to waves and experience, utilizing cross-overs to related transmitted through the harbor entrance. fields, is required. It is expected that the in- formation presented in this paper, regarding proto- The JIED consists of a north-south wall of alter- type scale experience, will provide some guidance nating cell and arc coffer dams, as shown in Fig- to the design engineer as well as indicate the need ure 2. The coffer dams are made of vertical sheet for future investigations. piles arranged into 13 circular cells and 12 inter- vening arcs. The cells are approximately 54 feet Description Of Dock in diameter and the centerline distance from cell to arc is about 32 feet. The sheet piles penetrate New docking space for the Port of Milwaukee is be- the mudline at an elevation of 547 feet, referenced ing constructed at the eastern edge of a lakefill to the International Great Lakes Datum (IGID) and on Jones island, Milwaukee, Wisconsin (Figure 1). extend into mud to an elevation of about 534 feet The dock is called the Jones Island East Dock (IGLD). The lakeward half of the cells and arcs 1232 are capped by cantilevered concrete slabs as shown ANALYSIS OF PLATFORM MOVEMENT in Figure 2. The bottom of the slabs are at ele- vation 583.3 feet (IGLD), which is nominally about On December 15, 1987, the partially completed slabs 4 feet above still water level. of the JIED were observed to move vertically during a moderate winter storm (Figures 3 and 4). Accord- ing to an eye witness, the slabs tended to rotate 0 500 1000 1500 Feet with the seaward edge of the slabs rocking upward 6 to 8 inches. The slabs moved up and down in re- sponse to the movement of the storm wave crests from north to south along the JIED wall (Figure 4). The purpose of the investigation presented here was .......... ...... ............ to report the design wave characteristics, deter- .......... . ..... Breakwater mine the severity of the December 15 storm waves, JONES ISLAND hindcast the resultant wave characteristics, calcu- -i*@ .... .. ........ EAST DOCK late the applied uplift forces using the most re- _XX cent research literature, compare these recently .......... computed uplift forces with those used in the ini- Lake tial design, compare the calculated movement of the Fill dock platform due to these forces to the observed S6 movement, and determine if and where further re- search is required. -: ... . ...... Incident Wave Characteristics .......... . ...... ..... .. ... .... . . ... .. . ........ .... .. .............- .... .. .............- .... .. ........... :@xxxj:j:j* ant wave characteristics for engineering Signi ic :@xx: ............... design in the Great Lakes have been hindcast by .. ........ MILWAM Resio and Vincent [10). They hindcast wave height and period for locations approximately 12 miles offshore for recurrence intervals of 5 to ........... ... 100 years, on a seasonal basis, and for three dif- ferent shoreward approach directions. Waves af- Figure 1. Jones Island East Dock at Milwaukee, fecting the JIED must approach Milwaukee from the Wisconsin. east north-east direction, pass through the en- Bullrail Ira b Slab@_, Slal:@/J@-' ab'j '1a b @Iab Slab 5 J, 5 )10, 11 2 JA 3 3@ 4 6 7 8 9 )9@"' 10 12 111A 13 @rc Sla 0 100 Cell 1 Feelt Plan of Jones Island East Dock EL 591 Concrete Fence EL. 588.3 Bullrail F EL. 585.3 Fill Slab A. EL 583.3 I - EL 582.3 bute, Limit of Cell Fill Outer Limit of Cell -w- Outer Limit of Arc Outer Limit of Arc L*- Inner Limit of Arc Inner Limit of Arc ---w- L Section Through EL. 547 ones Island East Dock -Mudline- ELEVATIONS AT IGLD EL 534 Figure 2. Plan and Section of the Jones Island East Dock. Slab 2 _j 2 Slab 'Ce 1233 trance to the Milwaukee breakwater and impinge on Table 2 the JIED. DEEP WATER DESIGN WAVE INFORMATION FOR LAKE MICHIGAN AT MILWAUKEE, WISCONSIN To obtain design wave characteristics at the JIED, these deep water wave characteristics off Milwaukee Return Significant Highest 1% Significant Deep Water (Table 2) have been propagated shoreward, shoaled, Period wave Height a Wave Height Wave Period Wave Length refracted, diffracted through the breakwater en- (years) (feet) (feet) (seconds) (feet) trance, and shoaled to the face of the JIED. The b analysis was made using the,methods outlined 'in the 6.6 11.0 6.8 237 Shore Protection Manual [1). Winter conditions are 9.4 15.7 7.6 296 the most extreme and will result in the highest 5 13.0 21.7 8.6 379 waves. Although ice conditions may modify the 10 15.6 26.1 9.4 452 effects of waves on the shoreline, open water win- 20 18.4 30.7 10.2 533 ter conditions were used as the example conditions. .50 22.0 36.7 11.2 642 throughout for the following discussions. The re- 100 24.6 41.1 12.0 737 sults of the analysis are shown in Table 3. December 15, 1987, Storm aData is interpolated between Resio and Vincent (1976) shore- On December 14, 1987, winds were 4 to 8 knots from line grid points 38 and 39, for winter seasonj and for waves the northwest,sector throughout the morning. Dur- approaching normal to the shoreline. ing the early afternoon the wind began..to shift to bExtrapolated values. the north and increase in speed. The lake level was 579.0 feet (IGLD). By late evening the speed had increased to '15 to 18 knots out of the east Table 3 northeast. By about 3:00 a.m. on December 15 the DESIGN WAVE INFORMATION wind speed had increased to over 20 knots and blew AT THE JONES ISLAND EAST DOCK from the northeast; the speed increased to about High Lake Level at 582 Feet (IGLD) Return Significant Highest 1% significant Deep Water Period Wave Height Wave Height Wave Period Wave Length (years) (feet) (feet) (seconds) (feet) 1 2.2 3.5 7.2 187 2 3.1 5.1 8.1 216 5 4.4 7.1 9.1 253 10 5.0 8.4 10.0 280 20 5.9 8.9 10.7 312 50 6.7 9.1 11.8 350 0", 100 7.1 9.3 12.6 373 High Lake Level at 582 Feet (IGLD) Return Significant Highest 1% Significant Deep Water Figure 3. Jones Island East Dock on December 15, Period Wave Height Wave Height Wave Period Wave Length 198,7.,. Photograph shows the partially (years) (feet) (feet) (seconds) (feet) completed slabs at cells 12 and 13. 1 2.2 3.5 7.2 171 2 3.0 5.1 8.1 195 5 4.4 7.1 9.1 227 10 5.0 7.4 10.0 251 20 5.9 7.5 10.7 280 50 6.7 7.5 11.8 315 100 7.4 7.5 12.6 339 30 to 35 knots from the northeast until approxi- mately 8:00 a.m. At this time the wind speed began to drop and the wind shifted quite rapidly to the north. and then. to the northwe 'st. Maximum gusts, were observed at 6:00 a.m. and were 51 knots out of the northeast. Heavy snowfall was associated with the storm; thunderstorms were observed. The lake level had risen to 580.6 feet (IGLD). Barometric Figure 4. Jones Island East Dock on December 15, pressure drop and rise was relatively rapid (small, 1987. Photograph shows the pa .rtially "tight" storm). completed slabs at cells 2 through 7. 1234 Hindcast Waves from Storm Condition Annual Probability 1 0.5 0.2 0.1 0.05- 0.02 0.01 Depending on a variety of factors, a best estimate I I I of deep water waves generated by the storm indi- 100,000 cates significant wave heights (H ) of 10 to 12 feet, with periods M of 8 seconsds, and wave 50,000 lengths (L) of 328 feet. These deep water waves were attenuated by shoaling and refraction as they moved toward the breakwater channel. Lake levels at the time of the storm resulted in a water depth 20,000 of approximately 40 feet at the breakwater gap. At T & T @ High Lake Level this depth for wave periods in the range of 8 seconds, the shoaling coefficient is less than 11-- 10,000 one (about 0.92). The refraction coefficient is I, I a T @ Low Lake Level between 0.92 and 1.00 depending on the direction of M wave approach in deep water; 0.98 provides a rea- 5,000- sonable estimate. Thus, the wave at the breakwater gap is estimated at H = 9 to 11 feet; T = 8 se- T & T @ High Lake Level conds; L = 250 feet. s 2,000- _.F. -T Accepting the calculation technique used [21, the CL 49 @_o T & T @ Low Lake Level major uncertainties in this estimate involve defi- CL nition of the meteorological conditions. The most 1,000-- significant condition is the size of the storm and - Level - V: @ the resulting fetch of the wind. The assumption of ke Level a duration limited wave generating event would 500 yield the highest deep water wave heights, 12 to -Low I- 14 feet, for sustained 30 knot winds for 12 hours. .4e\ Atmospheric stability and the uncertainties in- 200 volved with wind data collection and the conversion of airport wind to over-water wind are other fac- tors. Estimates of all of these factors based on 100 an examination of their effect were made. The evi- dence indicates a rather "tight storm" that would 1 2 5 10 20 50 100 decrease the fetch. The calculations assumed a Return Period (Years) wind speed toward the high end of the observations. There is a reasonable confidence level of charac- terizing this storm as generating 10- to 12-feet Standing Wave deep water waves. - - - Progressive Wave T&T Tanimoto & Takahashi (1979) The waves at the breakwater were then diffracted through the gap and propagated to the JIED. The F French (1979) resulting incident progressive wave height at the N NAVFACENGCDM (1982) dock face is estimated to be 3 to 4 feet (H s -sig- Figure 5. wave uplift pressure envelopes for the nificant wave height) and 6 to '7 feet (HI-one per- Jones Island East Dock. centile wave height). This corresponds to a return period of between 2 and 4 years .(see Table 3). Based on the observations of the December 15 storm Wave Forces we suspected that the very short-term impact force may have been important. This force is frequently Because of the geometry of the cell-arc wall at the ignored in conventional design and techniques to JIED, some portion of the impinging progressive evaluate it are not published in recognized design waves will be transformed into standing waves at manuals. A computer search of the technical lit- and near the wall. The theoretical limit of wave erature was made to determine if research infor- mation was available that would explain the heights and associated vertical velocities of the movement (Table 1). standing wave are twice those of the incident pro- gressive-wave reflecting from a planar wall. This There are experimental methods for estimating the fundamental'change in the wave characteristics from uplift pressures due to sudden impact' based on progressive to standing wave must be considered in these references. Envelopes can be developed for the force calculations. both the slowly varying (SV) sustained pressures, After the movement was reported, the stability of called "BOURRAGE" and the rapidly varying (RV) the completed dock was studied using the methods shock or impact pressures, called "GIFLE." Using ,outlined in readily available references [6][9]. the information reporting experimental data by Tan- s This analysis indicates uplift forces up to imoto and Takahashi [113, the pre sure curves for _,_ ,4P. M 1,000 psf as shown in Figure 5. Forces of this both purely progressive and purely standing waves magnitude will not cause movement of the dock, at high (582 feet) and low (575 feet) lake levels which will resist uplift forces of over 1,500 psf. were computed and are shown in Figure 5. Impact Thus, we concluded that the standard references pressures (RV) are one to two orders of magnitude were insufficient to explain the observed movement. higher than quasi-static (SV) pressures., 1235 The actual impact pressures will be between the Angular velocity and angular deflection can then be limits represented by purely progressive and stand- found from: ing waves. we cannot estimate bow much of the im- pact can be attributed to standing waves and how ot = fe tt dt and e = f8 tdt much attributed to progressive waves. The effi- ciency of wave reflection from the complex shape of The maximum deflection (8max) occurs when the angu- the vertical structure is difficult to assess. For lar velocity becomes zero (0 t = 0). An expression simplicity we consider the uplift force per slab is for 6 twas derived by evaluating the integral of a combination of progress ive- standing pressure att which can be expressed in terms of the maximum times the slab area exposed. moment due to wave impact (P ), the magnitudes of the slowly varying componmeaft (SV) and weight Calculation Of Platform Deflection moments (W), the moment of inertia (I), and the rise time of the rapidly varying component (tr ). Estimates of the motion of the dock platform in These terms are illustrated in Figure 6. The rise response to the wave-induced forces were based on a time is estimated as one half of the impact dura- number of assumptions to facilitate simple calcu- tion of the rapidly varying impact force given by lations. It is assumed that the platform is a French [61: rigid body, and is simply connected (pinned) at the -1/2 landward bottom corner. It is also assumed that t. 0.1 (H/g) the platform motion is two-dimensional and in the where I vertical plane. H incident wave height, and In addition to the above assumptions and data, it g acceleration of gravity was necessary to know or assume the distribution of The expression of the slowly varying force and re- the wave-induced forces along the dock and the time history of the moments. The wave forces that tend sultant moment as a constant is justified since the to uplift the dock are of two kinds: (1) Slowly times of movement of the slab to maximum displace- varying (described as sustained, bourrage, etc.,) ment was much less than the waver period. dynamic forces that are associated with the veloc- using the expressions derived for Elt, the time of ities (nonsteady) of the wave, and the hydrostatic maximum deflection of the platform was found. forces, and (2) rapidly varying (described by var- Using this time and the expression of displacement ious terms including impact, slapping, slamming, (the integral of e ) the angular deflection of the shock, gifle, etc.) forces that have high but very 1 t short duration peaks. platform was ca culated. Small angle approxima- tions were invoked throughout the analysis. The The method outlined in the standard literature ap- vertical displacement of the free end of the dock pears to address the slowly varying dynamic form platform was calculated from the angular displace- (6],(9]. The estimation of the rapidly varying ment. .The derived expressions for 6 8 , and 6 forces is still in an experimental or research were incorporated in a BASIC program wthtich twas used stage. A schematic representation of the time P max 0 history of the uplift pressure on the dock platform is shown in Figure 6. RV (t) 0 + Simplifying assumptions were necessary to proceed a C with computations based on rapidly varying forces. LU Given the uncertainty inherent in estimating the 0 SWL magnitude of this force, and its duration, a tri- J2 angular shape of the force curve variation in.time is probably reasonable. Since the platform over- E 0 hang is much shorter than a wavelength, a uniform 2 spacial distribution (average pressure) is reason- - able. Note, however, that there may be time-of- SV (t) peak variations from cell-to-cell section along the dock. i w Time The assumption of a simply connected rigid body Ts tend allows straightforward application of conservation !@'_ @Tr of angular momentum to estimate the movement of the wt platform slab. On initial contact with the wave the platform remains motionless until the wave pressure moment exceeds the weight moment. At this EXPLANATION point angular motion (rotation about the backedge) begins and is described by: P_ =Maximum Moment Due Te = Time Between Initial Impact to Wave Forces and Start of Movement M = 1 0 W = Weight Moment Tr = Rise Time to P max tt RV = Rapidly Varying Moment t_ = Time of Maximum Deflection where M net moment SV = Slowly Varying Moment tend = Time of Return to Ground I moment of inertia e angular acceleration Figure 6. Time variation of moments for the Jones tt Island East Dock slabs. 1236 to find the time and value of the platform. dis- N=4 N=3 placement. The program was written to account for 100- the discontinuity in the RV(t) function. The estimated uplift based on these calculations is 50 - N 2 1 L40 shown in Figure 7 for both the progressive and 'O'l I standing wave cases. The results shown in this figure are for the dock platform as completed with, 20 - all backfill and road surface material in place. The observed movement and wave height estimated for. N =4 December 15, 1987, storm are plotted on Figure 8. /"OBSERVED This figure represents movement calculated for only 10-1 CONDITIONS-- P_......:.... ;. I the bare slab, without fence, bullrail, or backfill R I A I C r portion of the structure as it existed during the 5 storm. C '001, OF '00, E The length of dock platform involved in the impact M of the wave is also taken as a parameter, shown as 9 3 N = 1 to 4 in Figures 7 and 8. The value of-N rep- 2 resented the number of cell-to-arc (centerline to centerline; see Figure 2) segments of dock platform 1 - "hit" by the wave. The restoring force is, in > every case, taken as due to the weight of the en- 11 0.5 tire section of platform (N 4). __N=2 100- __LLj ;----N = 4 ol 0.2 50--- 0" 0. 1 00 N=3 1 2 5 10 20 50 100 000. 7@ IF1 20- Tr Return Period (Years) Standing Wave C 10- N 2 Progressive Wave J I 'EXPECTED MOVEMIENTf.. N Number of Segments Involved _1 YTY ...... ..... ::T:: Figure 8. Observed movement for slab only compared 1::j: .... ....... . . to calculated moment for a range of tions. assump 2 r- 2 IF:- The results of the analysis of the bare slab (Fig- X:. @::[@ ure 8) show that the rapidly varying impact forces as formulated here can account for the observed T.- N 4 movement. The effect of standing waves is also evident and must be present to some extent. It is 0.5 ITI ..:X not possible to quantify the effect of wave re- flection in increasing the force since the area of the slab involved at any one time is not known. In 02 this case, the expected effect of impact1forces on the dock as completed (Figure 7) is estimated by N 3 assuming the same position on the "maps" of deflec- 0.1 LLI _@@9 tion, number of segments, and wave climate (return 1 2 5 10 20 50 100 period). A preliminary examination of the expected lengthwise distribution of forces indicates that T, = Return Period (Years) approximately half (N = 2) of the slab will be in- volved based on the assumed angle of wave approach. Standing Wave This is consistent with visual observations. - - - Progressive Wave If these dynamic wave forces are of importance, as the research data tend to indicate, by Figure 5, N Number of Segments Involved the dock slab can be expected to experience upward movement of 3 to 6 inches every few years (see Fig- Figure 7. Expected movement of completed dock based ure 7). The effect of the movement of the dock on observed movement of the Jones Island under various storms is difficult to estimate. East Dock slabs. Civil engineering works of this kind and size are NII NJI 1237 seldom, if ever, designed for this magnitude o.L 4. u]21ift Forc2s on Platform motion. The downward motion may be dampened by the Decks. Offshore Technical Conference, Pap. water under the slab and between the underside of No. OTC1381. May 1971. Pp. 1-538 through the slab and the cellular cofferdam and the upward 1-548. motion of adjacent slabs. If this dampening is sufficient, then there may be no structural effect 5. French, Jonathan A. Wave Uplift Pressures on from the movement. On the other hand, a sharp drop Horizontal Platforms. w. M. Keck Laboratory, of the slab onto the cell could possi . bly.cause set- Report No. KH-R-19, Hyr. and Water Resources, tlement of the crushed rock fill in the cells, Division of Engineering and Applied Sciences, structural damage to the slabs and structural dam- California Institute of Technology, Pasadena, age to the sheet piles of the cells and arcs. The California. July 1969. potential for lateral movement also exists. 6. Wave Uplift Pressures on Hori- z6ntal Platforms. ASCE Civil Engineering in CONCLUSIONS Oceans III. September 1979. Pp. 187-202. The December 15, 1987, storm at Milwaukee, Wiscon- 7. Furudoi, Teruaki and Akira Murota. Wave- sin, generated waves with a 2 to 4 year return Induced Up-Lift Forces Acting on Quay-Aprons. period. The wind-generated waves associated with Technology Reports of the Osaka University. the storm, while not especially large at the Jones Vol. 16. No. 734. 1966. Pps. 605-616. Island East Dock, were partially transformed into standing waves. These standing waves imparted to 8. Lee, Jiin-Jen and C. P. Lai. Wave Uplift on the partially completed dock slabs a rapidly vary- Platforms or Docks in Variable Depths. Proc. ing force greater than current design manuals pre- 12th Coast Engineering Conference. Taipei, dict. These rapidly varying forces caused the Taiwan. November 1986. slabs to move. 9. NAVFACENGCOM. Coastal Protection, Design Man- Further research is needed to define the uplift ual 26.2. Department of the Navy. Naval forces (pressures) on horizontal platforms backed Facilities Engineering command. 200 Stovall by vertical walls. The vertical walls should con- Street. Alexandria, VA 22332. 1982. sist of both planar and curvilinear form. When the 356 pages. results of the research is complete, it should be made available to the design engineers. This will 10. Resio, D. T., and C. L. Vincent. Design Wave allow the analysis of a wider range of potential Information for the Great Lakes, Rej2ort No. 3, designs for particular applications and permit a Lake Michigan. United States Army--Corps of better evaluation of overwater horizontal Engineers, WES, Technical Report H-76-1. platforms. 1976. 11. Tanimoto, Katsutosbi and Shigeo Takahashi. ACKNOWLEDGEMENTS Wave Forces on a Horizontal Platform. Proced. 5th Intern Ocean Dev. Conf. Tokyo, Japan. The authors wish to acknowledge the assistance of September 1979. Pp. Dl-29 to D1-38. Mr. Dick Horning for the preliminary derivations of the angular deflection equations used in the de- 12. Verhagen, J. H. G. The Impact of a Flat Plate flection analysis; the help of Ms. Mary Landsteiner on a Water Surface. Journal Ship Res. Decem- for assistance in coding the computer routine used ber 1967. Pp. 211-223. for the parametric studies leading to Figures 7 and 8; and comments from Mr. Jonathan French on the 13. Wang, Hsiang. Estimating Wave Pressures On a approaches taken in the analysis and the theoreti-, Horizontal Pier, Technical Report No. R546. cal basis behind these techniques. Naval Civil Engineering Laboratory, Port Hueneme, California. October 1967. REFERENCES 14. Water Wave Pressure on Horizon- tal Plates. c. ASCE, V. 96, No. HY10. 1. Coastal Engineering Research Center. Shore October 1970. Protection Manual. U.S. Army Waterways Exper- iment Station. Vicksburg, Mississippi. 2 Volumes. 1984. 2. Computer Program: JONSWAP (MACE-12) Deep Water Wave Forecasting. U.S. Army Waterways Experiment Station, CETN-I-34, 1985. 3. Elghamry, Osman A. Wave Forces On A Dock. University of California, Institutional Engi- neering Research, Technical Report HEL-9-1, Berkeley, California. October 1963. 1238 A HIERARCHICAL MULTIPROCESSOR DATA ACQUISITION SYSTEM FOR FIELD MEASURMENT OF STRUCTURAL RESPONSE IN BREAKWATER CONCRETE ARMOR UNITS James Rosati III Gary L. Howell U.S. Army Corps of Engineers, Waterways Experiment Station, Coastal Engineering Research Center Abstract constructed breakwater was rehabilitated with new lavers of 42 ton dolosse. Because of the severe wave conditions A custom. high performance, field data acquisition sys- at Crescent City and the desire to acquire dynamic struc- tern for measurement of wave loads, motions, and dynamic tural response data, new measurement technology had to stresses on large (42 ton) concrete armor units is described. be developed. To meet the technical objectives the system Twenty 42 ton dolos armor units were instrumented and had to have the capability of continuously recording data installed as part of the rehabilitation of the breakwater resulting from measurements with an aggregate sampling at Crescent City Harbor, California. A data acquisition rate greater than 66,000 samples per second. This paper computer system was designed using a hierarchical, paral- describes the dolos data acquisition system developed for lel architecture involving 47 cooperating computers includ- Crescent City and its performance during a two year mea- ing microcomputers permanently cast within the concrete surement period which included numerous major storms. dolosse. The data acquisition system has sufficient process- ing capability to make decisions about the type of incoming 2 Dolos Configuration data and perform appropriate action. Successful operation of the system during the period 1987-1988 resulted in data Selection of the test section of instrumented dolosse was, being successfully acquired from full scale concrete armor made after a careful evaluation of different proposed strate- units for the first time. gies. Participants at the Workshop on Measurement and Analysis of Structural- Response in Concrete Armor units 1 Introduction (Howell, 1985) discussed the trade offs among various op- tions and arrived at a recommended consensus. Sixteen The lack of adequate full scale field dat a has slowed progress dolosse were placed in an approximate four by four matrix in applying physical and numerical models to the problem near the mean water line with six accelerometer equipped dolosse in tbe center. Four dolosse were placed in the bot- of assessing the loads on concrete armor units. Progress in solving this problem requires the modeling of complexinter- tom layer under the top layer. This is test section was placed actions of physical processes and because of the severe envi- 0n the breakwater to permit all of the dolosse to receive ronment of breakwaters, very little prototype scale data are substantially the same incident waves. By limiting spatial available. Hydraulic models of breakwaters using concrete variability of the incident waves, statistical analysis of the armor units are routinely used for most designs. However, prototype data will be comparable to physical modeling contemporary hydraulic scale models consider only stability studies. and assume that all armor units have sufficient structural strength if stability conditions are maintained. Experiences 3 Measurement Plan with breakage of concrete armor units have shown that armor unit structural strength must be considered in the The measurement plan for this study was developed af- structural design process. No method is presently available ter consideration of many alternative plans and several ap- to predict, the loads. forces and stress required to establish proaches to the problems of instrument survivability and structural strength requirements. As part of an effort to data acquisition. A key decision was to define the funda- develop a structural design procedure, the US Army Corps mental measured parameters as the two moments and the of Engineers. Waterways Experiment, Station (WES) has torque taken about the sliank-fluke interface (Figure 1). concluded a program to acquire data on the structural re- This measurement could be taken by internally mounted sponse of dolos armor units in the prototype. Tqhe site for strain gages, therefore providing protection for the instru- the,study was Crescent City, California where a previously merits. Experiments with finite element method (FEM) nu- 1239 United States Government work not protected by copyright inerical models indicate(] that given ineasureinents of ino- per sec 'Isec angular servo accelerometers. The combination Ments and torques the FEM models could give stress distri- of six accelerometers allows the determination of motion butions throughout, areas of interest in the dolos (Howell, of any point on the dolos. assuming rigid body dynamics. 1986). The ranging of the accelerometers was chosen to determine an accurate kinematic velocity vector just before impact, because impact energy can be calculated from the veloci- ties. The dolos processor can be commanded to select the strain signals as well as ampli 'fier gain. The data rate is reduced by having the Dolos Processor combine selected strain signals to @onipute resultant, moments and torques, reducing six channels of strain to three channels of nionient and torque data. The dolos processor was implemented DOLOS AG DOLOS PROCESSOR f ACCEL. BREAKWATER CAP Figure 1: Dolos Instrumentation PRESSURE XDC DAT ATORS CONCENTR 4 Dolos Instrumentation All dolos instrumentation was permanently scaled inside the concrete arnior unit. The internal microcomputer sys- tem for each dolos digitizes the strain at a 500 Hz sample OFFSHORE rat e., and accelerometer dala. at a. .50 Hz sample rate. This WAVE GAGE rate is sufficient to capture all high energy dynamic niodes resolvable with the straiii gages located at the sliank-fluke TEL LINK interface. A rosetle of sleel reinforcing rods upon which TO CERC strain gages were mounted is located at the four faces of Figure 2: Major Data Acquisition System Components the sectioii through the shank-fluke interface, The rosettes are located 10 cm below the surface of the concrete. Only with a CMOS STD bit- microconiptiler syslent using the 1 lie.verl Ica] b)i a it k - fl it he i it t erface wa s I it st ru in ented as a cost Hitachi 64180 inicroproccessor operating al 6.1 MlIz. Both saving Ineasure. Algebraic conibiiia-tions of strains allow es- tiinatcs of the horiz .oual and vertical ])ending moments, tile the microcomputer board and the CMOS A-D board were axial thrust. and the torque about, the sliank-fluke Iinterface. manufactured by WINSyslems, of Arlington. Texas. Various methods of communicating data front the do- los were investigated. The high data rates eliniiiiated from 5 Dolos Processor consideration most types of low bandwidth RF and acous- tic telemetry systems. A high data rate telemetry systen) The first, element in the data. acquisition System is the Dolos would have required batteries that were impractical due to Processor, a microcomputer system which is permanently cost and size limitations. Use of a cable allowed power to sealed inside the trunk of the dolos (Figure 2). The do- be supplied to the dolos instrumentation and data to be los processor performs signal conditioning., analog to dig- retrieved at high rates. The dolos data, were formatted as ita.) conversiou of the strain gage da.ta.. and houses the a serial binary data stream allowing the number of coiiduc- accelerometer platform for the six dolosse which contain tors lo be limited to eight,. By minimizing the number of accelerometers. The accelerometers assembly is composed ('011duclors, a highly reliable cable was constructed consist,- of three lG Iniear servo acceleroinelers and three 100 rad ing of the core cable, an extra thick polyurethane jacket, 1240 CONTROL TAPE CONSOLE BANK MASTER TO TELEPHONE VAX 11/750 CERC LINK ARBITER ARBITER PDPII/23+ F PDP11/23+ MTU To Concentrator A To Concentrator B Figure 3: Shore Based Data Acqui Sitioll SySteln covered b 'v a. double layer of steel arnior, with a final jacket, wire pair for the command serial link froin the shore coin- of polyethylene. ielding a total outside diameter of 2.2 puter which was redistributed by the CoriceiAra,tor to each ' y fill. 11' a cable were allowed to oscillate with the high ve- individual dolos. All dai a connections were RS422 and were locit.N, oscillatory flows associated with the arnior layer of buffered and regenerated in the Concentrator. The cable a breakwater then it would fail due to fatigue. A systein distance form the Concentrator to the sbore based conl- employing a modified 3-inch stud link anchor chaiii was de- puter was 5,500 feet. veloped that, increased the density sufficiently to prevent Each Concentrator contained a single board computer flow induced motion. named an SAIT (Serial Analog Unit). The SAU is hosted The data from each dolos was sent. to the shore based oil the saine CMOS STD bus platform at the Dolos Pro- computer using a R-S422, asynchronous biiiary. protocol at cessor. The SAU sampled 12-bit pressure da,ta at 5 Hz and 38.4K Baud. Dal a, were frained in one secoii d packet - which transmitted it asychronously directly to the shore computer included a. synclirc ir of wi mizatiou preamble and a, unique I.D. for oil a dedicated physical pa ires also at 38.4K Baud. each dolos. The dul '% c'Nrcle over the link was 94"'(. and was Additionally each SAU contained fuuctions which allowed Inaliaged by a. CAIOS@ version of ille Z-80 S10 operating remote control of power for each dolos. permitting flie se- ill DX1 A mode. The dolos processor was programmed ill a, lective powering up of dolos. and I)re\,ei)lii)g the failure of coinbiiiatioii of Pascal for high level routines and careful] 'v One dolos cable from draining power froin the Whole S. Nlstein. coded. inhue a.,seinbi.y language routines for the lilne criti- The Coucentralor wa., supplied power al 4,." VDC ai)d ral finiclions. A real lime mull)-tasking executive was used current ranging from 0.5 Anip to 3.5 Anips depeii(lilig oil lo coordinate execulioil of six task,, whicli perforined the the number of dolosse powered oil. This corresponded to required functions. a shore ciid supply voltage which ranged from r;6 VDC to over 150 V DC. A K EPCO model I OE] 50 power supply with 6 Concentrators aiialog voltage programming was used at the shore end wit li a special feed back circuit to automatically adjust the out- Each dolos cable terminated at, one of two Data Conceil- put voltage to compensate for the shore cable drop over tralors which were located in holes cast oil the harbor side varying current requirements. of the breakwater (using the breakwater itself as a I protec- tive s1ructure). Up to ten dolos cables were terniiiiated at. 7 Data Acquisition System each 0--mceutrator. The Concentrators distributed power i o each dolos and combined the serial links fron-1 each dolos A senil-trailer van was modified to house the shore based into ()lie shore cable, maiulaining a separate physical wire data acquistioii components. The trailer cout ained the data pair for each dolos. The shore cable provided one pbysical acquistion computers as well as a 25 kw diesel powered 1241 generator and a power line conditioner. The trailer was loaded, booted, and configured by the arbiter. While one outfitted with an air conditioner and heater unit, various packet (representing one second's dat.a) was being received, graphics and text terminals, telephones, and office space. the previous second's data was timestamped and format- Figure 3 shows the block diagram of the computer sys- ted in order to allow for interpretation by post processing tern used for the shore-based data reduction and storage data analysis routines and sent to the host computer via systern. Since all data were formatted in packets they could the Ethernet link. be processed independently for each dolos. Since many si-nall computers are less expensive than a large computer of 7.3 Host Computer equivalent. power, a hierarchical, parallel architecture was selected. The last layer of processing was performed by a DEC VAX 11/750 computer running the VMS operating system. The 7.1 KXT-11 VAX had 4 high density 500 MB cartridge tape drives, man- ufactured by Mega,tape Corporation, and a. 456 MB RA81 Each Dolos communicated with a Digital Equipment Cor- disk to record data. Dat.a was initially recorded on disk and poration (DEC) KXT-11 single board computer that per- then backed up to tape offline in order to avoid exceeding formed the lowest level processing tasks. The KXT-11 con- the bus bandwidth of the VAX computer. The VAX was sists of a DEC T-11 microprocessor supported by several the development platforna for all data acquisiton software powerful 10 modules and a DMA controller. Each KXT- and was also a server, responding to boot requests from the 11 was configured as a slave processor residing on the bus arbiters. Asynchronous serial interfaces were used to coin- of an arbiter processor (a DEC PDP-11/23+). The KXT- municate to the Concentrators to configure dolos processors 11 has a section of memory which is dual ported (acces- and control power to the instrumented dolosse. The VAX sible by both the slave and the arbiter processor) which communicated to SAU's that sent data from breakwater helped to synchronize DMA transfers between the arbiter pressure sensors and from the moored wave buoys. and the KXT-11.Each KXT-11 ran a prograrn created by the DEC MicroPower Pascal realtime executive and re- ceived 3612 bytes per second (of serial data transmitted at 8 Waveriders 38.4K Baud) from each dolos. The KXT-11's were pro- grammed lo receive the serial data under the direction of Three wave measuring buoys were deployed offshore of the the DMA controller. During system startup, the incoming dolos test section of the breakwater. The buoy nearest the data were sychronized so that each second of data started breakwater was .1 mile offshore, the next buoy was 0.6 miles with the synchronization preamble embedded in the data offshore in a line perpendicular to the breakwater. The stream. The KXT- I I program checked each second of data third buoy was located 6 miles offshore. During the fall to look for an impact (an impact was defined to be t.wo of 1987, The buoys were moved so that 2 were about 0.3 successive instances of the difference between the current miles offshore and the third 3 miles offshore. The buoys and previous sample exceeding a programmed threshold). telemetered data to the data collection trailer where the If the strain data did not contain an impact then they were data were digitized. decimated by a factor of ten. Otherwise the data were not decimated. A program option allowed the KXT-11 to send only undecimated data to the arbiter. 9 Development Problems 7.2 Arbiters While developing a da.ta acquisition system as complex as qhe one required for Crescent CitY some development prob- The next layer of processing was done by 3 DEC PDP- lems were encountered. The orginial design data sampling 11/23+ computers. Five KXT-11's resided on the bus of rate was 400 Hz for strain and 40 Hz for accelerometers. Af- each PDP-11/23+ computer. The number of KXT-11's per ter a more thorough analysis, it was determinined that 500 Hz arbiter was determined by the bus and network bandwidth. was required to accurately represent dynamic loads. The The Q-bus used by the PDP-11's of the has a rate(] band- Baud rate had to remaian the same since 38.4K Baud 'Is the width of 3.3 MB/sec, however in practice only about 20% of maximum asynchronous speed supported by the KXT-11. that bandwidth was found to be useable due to system over- This limitation imposed a high duty cYcle upon the com- head and network controller throughput limitations. The municating processes, and severely limited allowable sys- arbiters were also controlled by a program written iii Mi- tem overhead time. The dolos processor used DMA char- croPower/Pascal. The arbiters were booted and downline acter output while the KXT-11 used DMA character in- loaded via Ethernet physical link using the DECNet soft- put. The KXT-11 could in fact not interpret the received ware protocol. Then they were sent the current system time data quickly enough to maintain synchronization with the and configuration, and then a copy of the KXT-11 program incoming data flow when using the conventional interrupt was copied into arbiter memory. Each KXT-11 was then per received character technique of acquiring the incoming 1242 data. The Micropower/ Pascal software used in the KXT- accelerometers (Howell, 1988). Several storm events were 11 did not support DMA input, although the hardware had recorded with breaking waves in excess of 10 meters. the capability. So, the necessary support software had to be developed for DMA input from a serial port. The com- mand links from the VAX to the Data Concentrators and Conclusions the dolosse were required to operate at a Baud rate of 38AK also. Although the asynchronous communication controller Availability of low power, high performance rnicrocomput- the VAX supported the 38AK Baud rate, the VMS oper- ers, and distributed 10 processors allows the development ating systein di(I not. After consulting with DEC, a patch of large, high performance data acquisition systems, pre- was developed for the appropriate VMS driver so that the viously uneconomical. Microcomputer technology is suf- 38.4K Baud rate was available. ficiently reliable to permit a no maintenance design phi- losophy as long as sufficient redundancy is incorporated into the total design. Development. of such systems re- 10 System Performance quires sophisticated real time, inultitasking software. Per- forniance analysis of hardware/software data acquisition Although the data acquisition system was designed to col- systems remains a difficult task and actual performance lect data from 20 dolosse, it never bad to perform under full was not determined until I lie system was fully operational. load. While 20 dolosse were instrumented, only 17 were op- Real tirne data reduction which is possible with the hierar- erational when data acquisition began. One instrumented chical. parallel architecture allows significant cost savings dolos was broken during placement, and two others had in the overall system. Special packaging of instrumentation their cables damaged during the early nesting phase of the and low power microcomputers permit data acquisition in breakwater. By the end of experiment, only eight dolosse extremely severe environnients, previously considered too were operating. The failures of the dolosse were primarly hostile. Through the development of these new measure- due to problems with their cables being overtensioned and ment techniques, prototype data that had been previously crushed. unobtainable have been acquired. During one tliuiidcrst,orni in April 1987, the Data Cou- centrators suffered damage to their RS422 receivers due to lightning. On December 2nd 1987, an intense storm 12 Acknowledgement produced lightning that. not only damaged the Data, Con- centrator's but, also every KXT-11 's RS422 receivers. Due The work described in this paper was conducted as part of to overtopping storni waves. access to the breakwater to the Crescent City Prototype Dolosse Study sponsored by repair the Data Concentrators was not possible uiitl*l the the U.S. Army Corps of Engineers. Permission to publish waves subsided 3 days later. Fortunately the Data Conceii- this paper was granted by the Chief of Engineers. trators successfully protected the dolosse from the lightning induced voltage surge. Subsequently, additional lightning protection and isolation equipment was installed. References On two separate occasions, the waverider buoy closest jo the breakwaler broke loose and was rendered unusable Howell, G. L.,ed. 1985. Proceedings of the Workshop on after being battered upon the breakwater rocks. Although Measurement and Analysis of Structural Response in it is not coninion practice to place wave buoys in such a Concrete Armor Units., Jan 23-24; Waterways Exper- bazardwis shallow water environment, it was essential to inieut Stalion. Coastal Engineering Research Center, the project to Irv to measure the incident wave statistics Vicksburg, Mississippi, 421 pp. as accuratelY as possible. and this buoy was considered sac- Howell, G. L. 1986. A System Ifor the Measurement, of rificial. the Structural Response of Dolos Armor Units in the Two of four pore pressure transducers buried in the Prototype.The Dock and Harbor Authority, vol. 67, breakwafer cap failed during the summer of 1987. Since number 779, May 1986. they were grouted in place, if, was impossible to cleteriiiiiie their mode of failure. Howell, G. L. 1988. Measurements of forces on dolos ar- During the data collection period of 19k7-88 over 20 Gi- mor units at prototype scale., Proceedings 21st In- gaBytes of raw data were collected. The collected data are ternational Conference on Coastal Engineering, June free of noise and transmission errors, due the digitization 1988, Costa del Sol.. Spain, American Society of Civil of the data inside the dolosse. Preliminary analysis of the Engineers. data. has shown significant static loads which vary with tide and wave loads of the same order of magnitude as static loads. Very few impact events were observed and there was negligible dolos inotion within the resolution of the of the 1243 WAVE ENERGY DISSIPATION BY REEF BREAKWATERS John P. Ahrensi and Edward T. Fulford2 lCoastal Engineering Research Center, S. Army Engineer Waterways Experiment Station, Vicksburg, MS and Chief Coastal Section, Planning Division, Baltimore District, U.S. Army Corps of Engineers, Baltimore, MD ABSTRACT parallel manner. Generally these functions can allow, or possibly benefit from, larger trans- Shoreline erosion has become a chronic problem mitted wave heights than can be tolerated in a around much of the U.S. coast. Fulford (1985) has harbor. This suggests considerable cost savings shown using segmented reef breakwaters is one of could be achieved by using lower crested struc- the most effective strategies for providing shore- tures. In a discussion of a variety of typical line protection. Reef breakwaters are essentially coastal erosion problems, Fulford (1985) concludes homogenous piles of stone without the multi-layer reef breakwaters represent the most satisfactory cross sections of traditional breakwaters. The general approach to shoreline stabilization. effectiveness of a reef is partly due to its ability to dissipate wave energy and additionally This paper will not discuss the stability of reefs due to the bulk of the salient which accumulates other than to Mention their low profile and high behind the structure. Extensive laboratory model porosity makes them surprisingly stable to severe tests have been used to define performance charac- wave conditions. Stability of reef breakwaters is teristics of reefs. Results from these tests have discussed in Ahrens (1987a, 1987b). The purpose been used to develop equations that predict wave of this paper is to discuss the performance transmission and reflection characteristics which characteristics of wave transmission, wave re- can then be used to determine the ability of the flection, and energy dissipation by reefs. Per- structure to dissipate wave energy (Ahrens 1987b). formance characteristics were determined by an The equations fit the data well, approach logical extensive series of laboratory model tests. limiting values, are easy to use, and are consis- tent with the physics of the interaction between 2. LABORATORY SETUP, CONDITIONS, AND PROCEDURES waves and rubble structures as it is currently understood. Reef breakwater model tests were conducted in a 61 cm wide channel within CERC's 1.2 m high by 4.6 m wide by 42.7 m long wave tank. Water depths 1. INTRODUCTION at the reef were 25 cm for most tests and 30 cm. for a few tests. Water depths at the wave gene- Erosion is a chronic problem along U.S. coasts. rator were 25 cm greater than at the reef and Properly designed riprap revetments can prevent waves shoaled over a 1 on 15 slope to reach the erosion, but can also interfere with the use of concrete platform the reef was built on. This the shoreline. Block revetments have not proved setup insured very severe wave conditions could be to be reliable against wind wave attack. Using produced at the structure. The testing channel offshore breakwaters has been recognized as an was open to a wave absorber area on the landward effective way to cope with erosion, and at the side of the reef so there could be very little same time, provide access to the shoreline for ponding effect behind the breakwater to complicate recreational purposes (Fulford 1985, Dally and evaluation of stability. Pope 1986). The concept of reef breakwaters is another step toward more effective methods to All tests were conducted with irregular waves. combat coastal erosion. Spectra were JONSWAP type in the deeper portion of the wave tank prior to wave breaking. Incident A reef breakwater is a low-crested rubble-mound and reflected spectra were resolved with three breakwater without the traditional multi-layer parallel wire resistance wave gages in front of cross section. This type of breakwater, in es- the reef by using the method of Goda and Suzuki sence, is a homogeneous pile of stone with in- (1976). Period of peak energy density of the dividual stone weights similar to those used in spectra Tp I ranged from about 1.45 to 3.60 see- the armor and first underlayer of conventional onds an@ the range of incident zero-moment wave breakwaters. Because of their high porosity, reef heights, HMO I was about 1.1 to 18.2 cm. Two breakwaters are stable to wave attack and, at the gages were used behind the reef to measure trans- same time, can dissipate wave energy effectively., mitted wave heights. See Figure 1 for a profile view of test channel. It is anticipated reef breakwaters would be used primarily for beach stabilization or to protect Two basic types of tests were conducted during the eroding shorelines in a shore parallel or near laboratory investigation. Most tests fell into a 1244 United States Government work not protected by copyright SCALE 0 1 2 M 30 SWL SEASIDE NI@IDL 11VITIAL 20 00 PROFILE EOUILIBRIUM 3 -4-- PROFILE '0 TRAINING WALL WAVE GAGES WAVE GAGES CONCRETE PI FORM 19 METERS TO WAVE 8RE4KWA TER - GE;E7R;-T@ - - - - - - - REEF FROM END OF TRAINING WALLS 15 0 10 20 30 40 50 ED 70 80 90 GUNCHETE PLA FFORM DISTANCE ALONG CHANNEL, CM 110 120 1 2 3 4 5 a 7 a 9 10 11 12 13 Figure 2. Initial and equilibrium reef profile DISTANCE ALDNG CHANNEL (M) caused by severe wave conditions. Figure 1. Profile view of test channel category where the primary objective was to deter" profile caused by severe wave conditions. After some stability tests, previous damage tests would mine stability of the reef fo a specific wave con- be conducted. Previous damage tests always used dition and wave transmission and reflection data wave conditions less severe than the preceding were obtained as a by-product. These are referred stability test so there was almost no change in to as stability tests. The other category was the profile during this type of tests. tests conducted at the completion of a stability test to determine transmission and reflection Two sizes of stone were used in this study. The characteristics of a damaged structure to a smaller stone was an angular quartzite with a variety of less severe wave conditions. These are median weight of 17 grams. The other stone was a called "previous damage" tests. Table 1 organizes blocky diorite with a median weight of 71 grams. the tests into subsets and gives important charac- Stone characteristics are given in Table 2. The teristics of each subset. In Table 1, stability stone used had shapes and surface texture typical tests have odd subset numbers and "previous of the type of stone which might be used for a damage" tests have even numbers. For stability reef. Further details on laboratory setup and Tablel stone characteristics are given in Ahrens (1987b). Basic Data For Each Subak Table 2 Crest Area of stone and Gradation Characteristics No. Water Height Median reakwater Subset of Depth, as Built, Stone Weight, Cross section No. Tests he (Cm) ho (gr.) At (cm2) Characteristic Quartzite Diorite 2% weight (9r) 7.0 14.0 1 27 25 25 17 1170 Median weight, W50 (gr) 17.0 71.0 2 3 25 NA 17 1170 98% weight (gr 28.0 139.0 3 29 25 30 17 1560 Density (grlcm@) 2.63 2.83 4 12 25 INA 17 1560 Porosity 44% 45% 5 41 25 35 IT 2190 6 it 25 NA 17 2190 7 38 25 32 71 1900 3. PERFORMANCE OF REEFS 8 26 25 NA 71 1900 9 13 30 32 71 1900 10 5 30 NA 71 1900 Performance, as used in this paper, refers to NA denotes not applicable to ,previous damage- test series. characteristics of wave transmission over and through a reef, wave reflection from a reef, and tests, the stone was dumped in the dry testing wave energy dissipated by a reef. Wave energy channel and pushed around primarily by foot to dissipation cannot be directly measured, but can conform to a template outlining the desired pro- be infered from measurements of wave transmission file of the structure. This procedure was used to and reflection. It has only been relatively re- prevent overly careful placement of the stone, but cently that reliable methods to measure the re- at the same time to insure the initial profile of flection of irregular waves have been available all reefs within a subset would be reasonably con- (Goda and Suzuki 1976). This means that reliable sistent. Desired initial profile for a stability estimates of wave energy dissipation by coastal test is a trapazoid with a crest width of three structures are rather unusual, see Seelig and stone diameters and having both seaward and shore- Ahrens (1981) for a related study. The shortage ward slopes of 1 on 1-1/2. When the profiles were of information about the nature and magnitude of surveyed, initial crest elevation of reefs within energy dissipation is unfortunate since for many a subset could vary as much as +1.5 cm from the situations this ability of rubble-mounds could be desired height. Duration of wave action for sta- regarded as their most important measure of per- bility tests was one and one-half hours for tests formance. This paper will show how the wave dia- with T p of 1.45 sec and up to three and one-half sipation process can be parameterized for reefs hours for tests of T p of 3.60 sec. This dura- and will present quantitative values of dissipa- tion gives a total of between 3500-4000 waves tion for specific reef configurations. A method based on T pwiand the reefs appeared to be at to predict wave dissipation will be discussed and equilibrium th wave conditions well before the compared to laboratory data. completion of a test. At the completion of a EI@ID SWL! ,@71V 1 71 A 'R stability test, there was a final survey to docu- Wave Transmission ment the equilibrium profile of the reef. Figure 2 shows initial and typical equilibrium To calculate wave energy dissipated by a reef it 1245 is necessary to know the wave transmission and crest, either because the crest was submerged or reflection characteristics of the structure. The by overtopping. wave transmission coefficient used in this study was defined K Ht high reefs, F > Hmo t H c Kt f (void size, reef width, relative wave where Ht is the zero-moment transmitted wave length height and Hc is the zero-moment wave height measured at the same location as Ht but with the td A L (2) reef not present in the testing channel, i.e., a f 150 t 1-2 VHmo ' Td d calibration wave height. The transmission co s50 8 efficient given by Equation 1 has the advantage of not including natural losses of energy due to wave where 'If" indicates a functional dependence and breaking between the incident and transmitted d50 is a typical stone dimension given by gages; these losses can be significant for severe conditions. There is however, no shoaling between W 1/3 gages since the floor of the channel is horlzontal d 50 in this section (see Figure 1). Using Equation 1 50 A_@T) definition of Kt makes it easier to quantify the relative ability of various reef configurations to and W 0 is the median stone weight and W r is reduce wave transmission, but tends to give values the un7t weight of the stone. Surprisingly the somewhat higher than the traditional transmission above parameterization can be consolidated rather coefficient. The definition actually compounds well for reefs into one variable (Ahrens 1987b). the tendency for high transmission coefficients which would be expected because of the high low reefs, F < H mo porosity of the reef and its lack of an impermeable core. Kt f (reef height, reef width, void size) One of the most important variables necessary to f( Fo hc At At 3/2 define the performance of a reef is the relative U__ - - 1 (3) freeboard, F/H mo where F I the freeboard is m d8 dsLp d 2 given by 50 Lp 0 s The functional relations given above for wave Figure 3 is a conceptual sketch showing how dif- transmission and below for wave reflection were ferent modes of wave transmission can be identi- used with regression analysis to develop pre- fied as a function of the relative freeboard. diction equations for Kt and K r as shown in 1.0 Ahrens (1987b). Z Z Wave Reflection ,2 0.8 The method developed by Goda and Suzuki (1976) was D Uj OrA Z> used to resolve wave spectrum into incident and Z D 0 < cr 0 reflected components. Following Goda and Suzuki ).-Z the reflection coefficient is defined 0.6 W TRANSMISSION R THROUGH E H E Z REEF K r 0 r NCT-Ei- to mo TRANSMISSION 0.4 where E r and E, are the reflected and incident OVER CREST Z wave energy of the spectrum respectively, and H r is the zero-moment reflected wave height. Com- 0.2 pared to wave transmission, it was relatively easy to develop an effective parameterization for K r Kr (relative wave length, reef height, reef shape) RELATIVE FREEBOARD, F/Hmo L p h c F At (4) Figure 3 Conceptual sketch of dominant modes of d d 2) ( S s mo h wave transmission for a reef as a function of relative freeboard. Co:nservation of Wave Energy Using the conservation of energy gives the '2z> 'U'D 0 'Z < TRANSMISSION \.I OVE@ R CREST To develop formulas to predict wave transmission following relation Ahrens (1987b) found it necessary to divide reefs into two catagories. High reefs had little or no incident wave energy transmitted wave energy + wave transmission by overtopping and low reefs had reflected wave energy + dissipation by natural substantial amounts of wave energy going over the breaking + dissipation by reef 1246 Based on the above relation, the following equa- 90 tion can be written to account for additional 80 energy dissipation due to the presence of the reef 70 Hc 2) 2 3 + D = Kt - K (5) A 60--_ y s iT- r mo + 50 where the ratio of H to H was determined 1 c mo t prior to normal testing of the structure during 12- 40 ----1 calibration of the wave channel. It was found the wave height calibration ratio, HC/Hmo was 30 + largely a function of HMO/ds and the following 20@ 0 equation could be used to estimate the ratio, 0.10 0.14 0.18 O@22 0.26 0.30 0.34 0.3B 0.42 HC 1.0 Relative Reef Width, Ajd@_, Hmo C 1 (6) Figure-4. Percent of incident wave energy 1.0 + (Hmo H dissipation as a function of relative reef width @i_ exp 0 + C2 (dmo)] for he/ds- 1.2. s IC illustrate influence of a second independent vari- where CO , C1 , and C2 are dimensionless cali- able by using both the ordinate and abscissa for bration coefficients with the following values: independent variables relative freeboard, FlHmo and relative reef width At/dsLP , respectively Co 3.16 and with values of the Lpendent variable, percent C1 = 0.86 wave energy dissipated, denoted by single digit C2 = 3.56 integers. The integer represents the percent dissipation divided by ten and rounded off to the 4. WAVE DISSIPATION BY REEFS 1.80 To discuss wave energy dissipation by reefs it is 7 convenient to refer to the processes of wave 1.40- transmission and reflection which can be measured directly. Since dissipation by reefs can be re- 5 garded as a function of all the variables shown in Equations 2, 3, and 4, it is obviously a very com- 1.00- plex process. To simplify the discussion three a heights of reef will be analyzed in detail but a 7 using the insight gained from analysis of all 0.60- tests. The three reef heights to be considered 4 have the following characteristics: one is sub- 3 233 merged, one has its crest near the still water 0.20 level, and one is non-submerged. The.relative 0.10 0.18 0.26 0.34 0-42 height, he/ds I is used to define the heights of Relative Reef Widttx [email protected]. the reefs. Ranges of relative height of the reef tests to be used in the following analysis are: Figure 5. Percent dissipation divided by ten and hc rounded to nearest integer value shown as a T 1.20 +/-0.05, typical high reef function of relative reef width and relative h freeboard for he/ds= 1.2. c T = 1.00 +/-0.05, water level reef, nearest integer value, eg. 11511 represents a test s where the observed dissipation of wave energy by hC the reef is about 50 percent. Figure 5 shows that T_ = 0.80 +/-0.05, typical submerged reef. for condition where wave overtopping is strong, s i.e., F/Hmo < 0.50, the Influence of the relative width of the reef is quite important in dissipat- Figure 4 shows the percent of incident wave energy ing wave energy. For conditions where there is dissipated by typical high reefs as a function of little wave overtopping, i.e. F/Hmo > 1.0 , the the relative reef width, At/dsLP . For these relative reef width, as formulated, is not very tests there is a reasonably strong trend for wave important and the relative wave length, L /d , and dissipation to increase with increasing reef the reef transmission variable (L d5 02)/(P R t ) I width.. The substantial data scatter shown in Fig- are important to the dissipation @rocess. MThis is ure 4 is characteristic of situations where the because for high reefs wave reflection is strongly 2 dependent variable is a function of other indepen- dependent on the relative wave length and wave dent variables in addition to the one shown. It transmission through the structure is strongly de- is clear from Figure 4 that the scatter is not pendent on the reef transmission variable (Ahrens random because predicted dissipation follows the 1987b). From another perspective it can be noted observed values very well. Figure 5 helps that for high reefs that a substantial portion of 1247 the dissipation occurs within the structure and that after accounting for reef width effects that this mechanism is quite dependent on the using At/dsL p there is a small tendency for stone size. In addition both wave reflection and dissi;ation to decrease as stone size increases. transmission through the reef increase with in- The influence of stone size tends to decrease as creasing wave length. The above mechanisms in- the height of the reef decreases since an in- dicate that wave energy dissipation should de- creasing portion of the transmitted wave energy crease as wave lengths and stone sizes increase goes over the crest. Since this data subset spans with the influence of wave length rather pro- transition between where the dominant mode of nounced on non-overtopped reefs. Figure 6 shows transmission is over the crest and where it is by the relative wave length and relative freeboard as runup and overtopping, there is a reversal in the the independent variables and the dissipation trend of wave transmission with wave height in denoted as an integer, using the same format as this interval. The reversal in the role of wave Figure 5. It can be seen in Figure 6 that the height causes difficulties in parameterizating relative wave length has substantial influence on dissipation for reefs near the still water the dissipation process especially for infrequenly level. It should be concluded then that for reefs overtopped reefs, i.e., F/Hmo > 1.0. with their crest near the still water level, that 23 it is the width as measured by the relative reef 3 4 5 width which has the most influence on dissipation. _j 20 - 2 Figure 8 shows wave energy dissipated by a sub- merged reef as a function of the relative reef 17 - 3 width. As with higher reefs, there is a trend of increasing dissipation with increasing relative 3: 14 - width. The submerged reef also continued the trend of decreasing dissipation with decreasing relative crest height. 65 8 7 60 0.20 0.50 1.00 1.40 1.80 55 Relative Freeboard, F/H_ c: .o 50 to Figure 6. Percent dissipation divided by ten and 45 rounded to nearest integer value shown as a 0 40 function of relative freeboard and relative wave - 35 + + length for h C/d s @-1.2. 30 Test data for reefs with their crest near still 25- water level are shown in Figure 7. The trend of + ++ o C)b-.d 20 increasing dissipation with increasing reef width C]o + Predi-ed is similar to that for typical high reefs (Fig- 15- ure 4) except that levels of dissipation are 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 slightly lower. For tests shown in Figure 7 the Relative Reef Width, A@d@_, @ 70 Figure 8. Percent of incident wave energy dissipated as a function of relative reef width for h /d -:-- o. 8. 60 c s + + I Submerged reefs tend to cause waves to break 04- oo t+ + prematurely; this process is referred to as wave .6 tripping. For a fixed submerged freeboard, T EP 1 0 40 - --- Eh tripping is more effective the larger the wave. This characteristic can be seen in Figure 9 where o the independent variables relative freeboard and b 30 - 4 1 Cl- L.g-d relative reef width are shown on the ordinate 20 + - ------ l31L_Ob-ed_. and abscissa, respectively, and the percent dissi- El + Predicted pation divided by ten is shown as in an integer. 10 0.08 0.12 0.16 0.20 0.24 0.28 0.32 There is another aspect of energy dissipation Relative Reef Width. A@d.[_. which affects reefs of all heights although its effect is somewhat more noticeable in Figure 9. Figure 7. Percent of incident wave energy If wave conditions become too severe in relation dissipated as a function of relative reef width to the size of the reef the tripping process can for h C /ds =1 .0. become rather ineffective. The ability of the reef to disrupt wave action can be parameterized range of the relative freeboard is too small for using the variable, A t /H mo L p . This variable this variable to have much affect. About the only seems to dominate the dissipation process for other variable which has substantial influence on At/H mo L < 0.5 . Within its range of dominance a this subset of data is another width parameter P surprisingly simple formula relates percent which includes the stone size. Analysis indicates dissipation to disruption effects, i.e. 1248 -0.2 22 a The considerable range of dissipation for each 4 5 characteristic height in Table 3 was shown in 1 3 4 5 the paper to be partly due to the relative width T -0.7 5 of the reef A /d L However, the dissipation LL process is t S p , 'd 4 3 ;uite complex and it is impossible to 5 show simple graphs of dissipation versus an 0 4 independent variable that do not show considerable LL apparent data scatter. Much of the scatter can be explained by using the model for wave trans- mission and reflection given in Ahrens (1987b), in the conservation of energy relation, Equation 5, 2 4 and correcting for natural losses of energy due -2.1 to wave breaking not associated with the presence 0.05 0.15 0.25 0.35 0.45 of the reef using Equation 6. Energy dissipation can be estimated very well by the above prediction Relative Reef Widtli Ajd@_, method, as shown in Figures 4, 7, and 8. The Figure 9. Percent dissipation divided by ten and range of conditions for which this method can rounded to nearest integer value shown as a be applied is wider than used in this paper; function of relative reef width and relative see Ahrens (1987b) for details relating to ranges freeboard for hcds @z-o.8. of the various conditions tested. For severe wave conditions a small reef can become relatively At ineffective in disrupting wave action. Equation 7 percent dissipation 100 (7) can be used to define and directly estimate per HmoLp cent energy dissipation for these conditions. for At, < 0.5 ACKNOWLEDGEMENTS H L - mo p The support of the Office, Chief of Engineers It is wave disruption effects which cause the Civil Works Research and Development Program for somewhat confusing data scatter in the upper left funding and granting permission to publish this hand corner of Figure 9 and lower left corner of paper is greatly I appreciated. Figure 5. REFERENCES 6. SUMMARY AND CONCLUSIONS 1. Ahrens, J. P. "Characteristics of Reef Break- An approach to evaluating effectiveness of reef waters," U.S. Army Corps of Engineers Coastal breakwaters based on their ability to dissipate Engineering Research Center, TR 87-17, Vicksburg, wave energy is presented. While this approach is MS., Dec 1987b. unusual, it provides a method to judge performance 2. Fulford, E. T., "Reef Type Breakwaters for which is directly related to the reef's function. Shoreline Stabilization," Proceedings of Coastal As a tentative criteria, the Baltimore District of Zone 85, Baltimore, MD, Sept, 1985. the Corps of Engineers considers dissipating 50 percent of the incident wave energy during 3. Goda, Y. and Suzuki, Y., "Estimation of storm events with a 5 year return interval to be Incident and Reflected Waves in Random Wave satisfactory performance for a reef breakwater. Experiments," Proceedings of the 15th Coastal To simplify and focus the discussion three dif- Engineering Conference, Honolulu, Hawaii, July ferent heights of reef are treated in this paper: 1976. (1) a non-submerged reef, (2) a reef with the 4. Dally, W. R. and Pope, J., "Detached Break- crest near the still water level, (3) a submerged waters for Shore Protection," U.S. Army Corps of reef. Using the relative height , hc/ds I to Engineers Coastal Engineering Research Center, define typical reef heights, dissipation by reefs TR 86-1, Vicksburg, MS, Jan 1986. was found to be in the range shown in Table 3. Table 3. Percent of incident wave energy 5. Ahrens, J. P., "Reef Breakwater Response to dissipated by reefs. Wave Attack," Proceedings ASCE Conference on Berm Breakwaters: Unconventional Rubble-Mound Range of Wave Energy Breakwaters, Ottawa, Canada, Sep 1987a. Characteristic Height Dissipated, Percent 6. Seelig, W. N. and Ahrens, J. P., "Estimation h- c o.8 17-56 of Wave Reflection and Energy Dissipation Co- T efficienta for Beaches, Revetments, and Break- h s waters," U.S. Army Corps of Engineers, Coastal c 1.0 17-68 Engineering Research Center, Fort Bel-voir, VA, Feb d 1981. 3 h c 1.2 22-76 1249 Why Breakwaters Break E.H. Harlow Soros Associates turned goo. The part that turns I. The Forces of Nature upward causes a geyser. The part that II. The Works of Man is turned downward causes damage. The III. Mechanisms & Dynmics intensity of the pressure is limited IV. Maintenance mainly by the quantity of air (bubbles) in the water. For example, if the portion of breakwater being struck is impervious, and there is no air in the water, the virtual incompressibility of the water can produce near--infinite instantaneaus pressure. Whenever this A breakwater by definition is a work of happens, some portion or particle in the man built in the water for the purpose structure must move. of breaking down the waves of the sea, so as to shelter certain objects to the B. Current lee. In other words, it is a confrontation between a possibly When a breakwater is built, it will irresistable force and what is hoped surely deflect natural current. In will prove to be an immovable object. It doing so, the natural flow is nearly is hardly surprising that although man's always increased. The result is usually efforts may succeed for a while, to increase wave energy, change its eventually natural forces will prevail. direction, increase undermining, and to The question is, for how long can we carry away the eroded particles to a succeed? down--current location away from the breakwater. I. The Forces of Nature C. Abrasion There is an old saying that water will wear away a stone, and it is literally As loose particles are detached from a true. There is no material that can breakwater or are carried against it by withstand indefinitely the abrasion and moving water, the result is aki,n to solvent action of unlimited amounts of sandblasting. It becomes more intense moving water. And the ocean is almost with time, as the supply of sand always moving. in response to winds and increases. Rock pieces and concrete currents on the spinning shpere of the shapes become rounded. They are jostled earth. But breakwaters break long into new positions under the effects of before the materials have been reduced gravity and wave action. The large to grains of sand. The forces they must contact pressures at the three points resist are of four types: impacts, where each piece is supported causes currents. abrasion and piping. more sand particles to be worn off. and add to the supply of the sand-blasting A- _j_Uacts. material. The crest of a large breaking wave D. PLpinq striking a stationary object might be likened to several dozen fire hoses This is the term applied to the ejection being turned against it simultaneously. of particles from a mass when water Or. worse, if the face of the crest is exiting at a free surface has sufficient parallel to the surface of the object, velocity to carry sediment particles its kenetic energy is stopped in with it. It occurs frequently in near-zero time, causing instant shock embankments, along the top of a less pressures, as the water velocity is pervious layer, causing a series of CH2585-8188/0000-1250 $1 @1988 IEEE holes that undermining the slope above. diameter concrete piles or caissons It occurs in the seaward slope of driven into the bottom like sheetpiling breakwaters, as a wave recedes, and to form a continuous wall. water in the voids rushes out into the trough. If there are particles free to Caissons are often combined with rubble escape, the velocity may carry them out, mounds, particularly in Japan, with the too. concrete caissons generally on the landward side, and large cast concrete 11. The Works of Man units, known as "armor," on the seaward side. Perhaps the oldest type of breakwater is the rubble mound. It may be as long Partial mounds are often used on the as needed to project a harbor, or an seaward side of sheetpiling, to help entrance to a harbor, in which case it guard against undermining and to help may be called a jetty. There may be reduce overtopping. two, on either side, with an opening just large enough to exclude the worst III. Mechanisms and_j@yRamics, storm waves. It may be as narrow and steelp-sided as possible without slope When the forces of nature meet the works failure, (to save costs.) "Possible" of man, we see how devious is the former involves the art and science we are in circumventing the ingenuity of the discussing, although with the selection matter. of sizes, arrangement and kinds of pieces of stone and possible cast A. Mac@qp4ysj_5@s concrete shapes, of which there are a great many. The largest pieces are A breakwater will break whenever the placed on the outside of the seaward forces on a part of it are too large for slope in the hope that the biggest waves the forces holding it together. This will not roll them down the slope. has happened many times, for example: Blocks are often cast, if of concrete, A stone or piece of armor in a rubble or cut, if of stone, and piled in mound is lifted out of its position; ordered tiers, with an almost vertical face toward the sea. They are more A caisson is shifted, bounced or broken; often used as a revetment, with land backfill on the shoreward side. Each A hole is eroded in the seabed in f ront block must be heavy enough to withstand of the structure, causing it to be .the worst waves without being undermined; dislodged. The slots between and around blocks become the water wedges during A block or caisson is toppled, forward wave action, tending to pry them loose. or back; Caissons of concrete are often so large A whole section of a rubble mound is as to be the full height of the virtually liquified and flattened; breakwater, and are placed close together to form a wall. The slots A whole section is fractured and between the caissons, and the foundation collapses. materials upon which they rest become the water wedges. Various means of Events like this can occur when nature sealing these slots and protecting the produces waves much larger than foundation from undermining have been expected, or waves with much longer devised. period, such as tsunamis, where the volume of water and the mass of a single Sheetpiling is often driven in a single wave are much larger than for normal row with structural support or in double conditions. Intersecting wave crests rows with backfill between, to form a from different directions can cause breakwater. This latter is often done monsters. in cellular configuration, particularly to form temporary cofferdams for deep During a storm the level of the sea may excavations. The sheetpiling, generally rise many feet higher than anticipated, of steel but potentially of concrete. owing to hurricane (typhoon) conditions, forms a highly impervious wall embedded combining barometric rise and storm in the sea bottom to prevent surge, along with big waves. undermining. Catastrophic events can also occur when. Hybrids of various kinds have been owing to cost constraints, optimism or built. One example is the use of large ignorance, man's works fail utterly 1251 because we do not understand or allow Breakwaters break because we are unable for extreme natural forces. An example to forecast and design for all the of this might be the building of a road conditions they will meet during their and 24 lovely vacation houses on lifetime. If we -were able to do this, cost constraints would prevent it in Wapatree Point, 'Rhode Island, all most cases. Therefore, breakwaters must totally wiped out in one storm in 1938. be carefully watched, particularly after each storm, and repaired. B. Microphvsics IV. Maintenance The detailed abrasive mechanics of water and stone, or concrete are not well The first and most obvious repair is the documented. Little is known about the replacement of components. For rubble rate of attrition, the degree of mounds, a supply of spare outer armor rounding of particles, loss of Eines and under layers should be kept on from core materials, shifting of pieces shore, ready for deployment. For as their edges wear and become sand. blocks, depending on whether the damage Neither do we know much about the is displacement or rupture, new blocks pressure changes in a micro-seale within and means for patching, sealing, a rubble mound or between caissons or concreting. For sheetpiling, additional stones. sheeting, piles and materials for sealing and concreting. We know that changes in position and elevation of marked points on To avoid long delays in making repairs, breakwaters are sometimes measured. planning must provide for early Long-term records are scarce. Most mobilization of floating or land significant chariges occur during severe equipment to place the materials. storms, as they do in the case of beaches and shoals. The effects of the B. Reinforcement minute particle reactions ace concealed beneath these major events, but it is It may be prudent in some cases to obvious that as each piece becomes reinforce the damaged structure in a smaller and more rounded, the water major revision. For example, the forces will more easily move it. The breakwater at Sines, Portugal, a rubble angle of internal friction of the mass, mound with dolos outer layers and a or of the base. will become smaller, heavy concrete capwall, was reinforced chinks or breaks in block, caisson or after its storm damage in the late 170's sheetpile walls will become larger, and by adding an extensive seaward shelf of water will squirt through them faster very heavy (90 t) concrete cubes. and with greater force. Other caisson-type breakwaters have been Rubble mounds tend to become flatter reinforced with wide rubble-mound beams with time. It is argued that if the to break up wave energy over some seaward slope is built with a berm, distance, preventing the clapotis creating an ogee contour like that action, the impacts and the undermining encountered in beaches or old that sometimes occurs with vertical-wall structures, the stability will be structures. improved. Yet, for each sea level there is a beat-shaped ogee slope. If all storms achieved the same level, the shape would be well adapted to it. But eventually a super-storm will reach a higher level. Then the berm will be too deep to trip the breakers. and they will strike against the upper portion, as if it were part of a uniform slope, with similar devasting effects. This phenomenon was demonstrated in model tests for a proposed massive breakwater to protect nuclear floating power plants. Designed for the "probable maximum hurricane.11 it withstood all wave action when water level was at predicted elevation. But when some observer insisted that a still higher elevation be tried. The model breakwater unraveled at the top. 1252 JAMSTEC/DEEP TOW SYSTEM Hiroyasu Mam, Kiyoshi Ghtsuka, Hiroshi Hotta Deep Sea Research Department Japan marine science & Technology Center (JAMSTEC) 2-15 Natsushima-cho Yokosuka 237 Japan ABSTRACT expanding the visibility of the camera, by adding a small ROV which could be launched from the deep tow, The development of JAMSTEC/,Deep Tow system began in and by exchanging a coaxial cable with a fiber-optic the late of 1970s and presently two systems, which cable. In this paper, we would like to introduce the have depth capabilities of 6000 and 3000 meters, are system configuration, operation, significant results operated. The characteristic features of the and future developments of the JAMSTEC/Deep Tow JAMSTEC/Deep, Tow system are as follows: system. (1)Real time visual observations are possible by black and white or color TV which is used with 7800 or 2. SYSTEM DESCRIPTICNS 4500 meters of tow cable respectively. (2)Selective bottom sampling could be carried out by The JAMSTEC/Deep Tow system comprises of (1) towed visually observing the sea floor with the use of sled equipped with underwater instruments such as grab and dredge equipped on the deep tow. sonar and camera, (2) on board units of underwater (3)Quantitative monitoring of the tow cable is instruments which are housed in a container van, (3) possible by a special deep tow sheave named "Gimbal launch/ recovery and on board towing gears such as Sheave". A-frame, Gimbal Sheave, winch and tow cable, (4) (4)A delicate ship control in the deep towing and acoustic navigation system for precise underwater operation in the rough sea conditions are possible positioning, and (5) research ship which is equipped by a semi-submerged catamaran(SSC) "Kaiyo". with basic scientific instruments such as precision In present, the JAMSTEC/Deep Tow is indispensable for depth recorder or Sea Beam. In the following sections the site survey of manned submersible and the deep sea basic features of the sub systems or system ccrqponents research in JAMSTEC. will be described. 1. INTRCWCTICN 2.1 Sonar The first JAMSTEC/Deep Tow system has been developed There are two sonar systems, one of which comprises of and operated since 19771 . The second system was a 100 kHz side scan sonar and a 3.5 kHz sub-bottm developed in 1982. The first one was developed for profiler2, and the other comprises of a 70 kHz side finding artificial objects such as low level radio scan sonar and a 4.8 kHz sub-bottom profiler. The 100 active wast packages on the deep ocean floor. Its kHz sonar, wtiich has 300-meter maximum range per side, depth capability is 6200 meters. The second one was is used primarily for searching artificial objects in- developed for the site survey in the dive areas of the ocean floor, because its resolution is better than Japanese manned submersible "Shinkai 2000". Its depth the 70 kHz sonar. On the other hand, the 70 kHz wide capability is 3500 meters. For convenience, we will range sonar, which has 1000-meter maximum range per call these systems as 6000-meter system and 3000-rreter side, has great advantage in geomorphological and system respectively. The 6000-meter system comprises geological studies. of a 7800-meter long tow cable, a black and white TV system and a 100 kHz side scan sonar. The 3000--meter The 100 kHz sonar was purchased from ORE in 1978, and system ccuprises of a 4500-meter long tow cable, a lost in 1983 during deep towing in the steep wall of color TV system and a 70 kHz side scan sonar. Each submarine caldera. After the accident we purchased coqponent of the both systems such as still camera and side scan transducers frm Klein and modified on board CID is primarily exchangeable. The tow cable is a sonar unit. The 70 kHz sonar was manufactured by double armored coaxial cable (FGB/U) of which outer Nippon Electric Company (NEC) in 1982(Fig.1). Sonar diameter is approximately 17 mm in both systems. data in each system is displayed on a dry paper In these five years, the primary role of the graphic recorder and stored in a 1/4-inch analog data JAMSTEC/Deep Tow system is the site survey for manned recorder. submersible. Many significant engineering and scientific results were obtained by the JAMSTEC/Deep 2.2 Still and TV Camera Tow. In present, the deep tow is indispensable for the deep sea research in JAMSTEc. As Japanese Benthos model 372, 35 mm cameras and model 382, 100 ws 6500-meter manned submersible will be in operation strobes are used in-both-systems. A real time black from 1990, it is necessary to establish a new site and white TV by NTSC system is used in the 6000-neter survey system. we are planning to increase the system because of great attenuation of the video capability of the 6000-meter deep tow system by signal through the RG8/U coaxial cable 7800 meters in CI-1258s-ms/oooo. 1253 $1 @1988 IEEE length. The minimum lumious intensity is 0.35 lux. A significant factors to influence the operating real time color TV by NTSC system is used in the efficiency of the total system at sea. There are two 3000-meter system. 4.5 mz band width is required to typical cranes for launch and recovery. The first one transmit color video signal. As it is difficult to is an articulated boom crane which is used by Scripps amplify homogeneously from D.C. to 4.5 MHz, the MPL/Deep Tow. The advantage of this crane is that it transmit center frequency is shifted to 6.2 MHz, thus is possible to launch and recover the deep tow very the video signal is frequency modulated with plus close to the sea surface. Therefore, it is possible minus 4.2 Miz shift frequency and transmitted via to operate the system in rough sea conditions. Also, 4500--meter coaxial cable. The mininum luminous as the crane rotates with a ship changing her course, intensity for practical use is 150 lux using a this minimizes the possibility to injure the tow cable three-tube image intensifier. Recently, we replaced at the top sheave of the crane. The second one is an the camera with the latest CM carera of which rainimum A-frame which w use in JAMSTEC and is ccmicnly used luminous intensity is 32 lux. by research ships. In this case, as the distance between the top sheave and the sea surface is greater Lightings are two 250 W quarz iodide lamps for black than 4 or 5 meters, the towed sled is likely to swing and white TV, and four 250 w quarz iodide laqX for by pendulum wtion and is apt to collide with the ship color TV. It is possible to select either black and hull which will damage the instruments. white or color TV by a comend signal from on board unit in the 30OG-veter system. There are two more In order to avoide the collision, we put four comand signals for the actuation of still camera and unti-swing ropes on the towed frame. These ropes are strobe and also grab sampler. CM data are holded by four deck men to stop swinging during launch transmitted with the video signal. operation. The other end of each rope is attached to the tow cable with attatchment during towing. In The original black and white TV was purchased from recovery, the ropes are detached from the tow cable Hydro Products in 1973 and totally modified by JAMSTSC and are used as in the launching. In general, the and Shonan High-Frequency Laboratory (SHL) 3. The recovery is more difficult than the launch, because it color TV was developed by JAMSTEC and SHL in 1982 is more difficult to attach a rope on the towed frame (Fig.2)4. The video data are recorded in 1-inch than to detach it. As the above method is very simple cassette(U-iotic) and 3/4-inch cassette(VHS). As the but reliable, the JAmSTEC/Deep Tbw can be launched and video data are replayed easily on board, it is not recovered at any sea state. necessary to develop the photo film on board. 2.6 winch and Tow cable 2.3 Sampling Devices There are two winches for the deep tow, i.e. a During the deep tow camera observation, it is possible 8000-aeter winch with 7800 meters long tow cable and a to carry out sarrpling by use of grab sampler and box 4000-ceter winch with 4500 meters long tow cable. The dredge. The grab sampler could be dropped at a 8000-meter winch, which comprises of a 200 KVA sanpling point by watching TV monitor. The box electric-hydraulic unit, a traction winch and a dredge, which is named the "Deep Tow Dredge", could be storage winch, was built to be used on the ships of used any time by simply lowering camera frame close to opportunity. The 4000--meter winch is aimed to be used the bottom. The sizes of the grab and dredge are 50 both on the mother ship "Natsusuima" of manned an cube and 40K x 15H x 50L an respectively. The grab submersible "Shinkai 2000" and semi-submerged sampler is suitable to collect sedirrents, on the other catamaran (SSC) "Kaiyo". The tow cable is wound hand, the dredge is useful both for sediments and directly in the 4000-mter winch drum and its rocks. hydraulic power is supplied by both ships. The 8000-meter winch was modified from direct winding to 2.4 Towed Frame the present system because there was a difficulty in level winding of the tow cable. As far as the level The JAMSTSC/Deep Tow adopts a direct towing rrethod winding is maintained, we consider that the traction which suspends a towed fram(sled) directly to the winch is not always necessary. cable stop(termination). The advantage of this method is that it is easy to keep the towed frame very close The tow cable is a double armored coaxial cable(RG8AJ) to the sea floor even in very rough topographies. On of which outer diameter is approximately 17 mm, the other hand, a depressor towing method, which a breaking strength is 19 tons and weight in water is neutral buoyant towed fraw is towed behind a 0.85 t/km. The maximum length which this cable breaks depressor or a heavy weight, has an advantage that the by its own weight in water is 22000 meters, while the towed frame is very stable because it is free from the deepest water is 11000 meters in reality. Here we up and down motion caused by a surface ship. As the will estimate the safety factor of the tow cable when towing speeds of sonar and canera are approximately 2 it is used in the depth of 6000 meters. we will kt and 1 kt respectively, an open fram is more stable assurne that the weight of towed frame in water is 1 than a faired topedo shape as for towed body. Also, ton and cable length which will be payed out is 6500 open fraw is stable by vertical motions, while the meters, then the static tension at the winch is 6.5 faired topedo shape is feasible to cause pitching tons, thus the safety factor of the tow cable is 3. which degrades sonar records and is haratful for camera According to our measurements of cable tension during observations. deep towing, the dynamic tension does not exceed plus minus 50 % of the static tension. Therefore, we 2.5 Launch and Recovery System consider that there is not serious problem on the strength of the tow cable in the deep towing at a A launch and recovery system is one of the most depth of 6000 meters. 1254 It seems that there is no quantitative standard on a times higher than the mono-hull ships, there is no cable life. As electrical deterioration of the problem in launch and recovery of the deep tow, coaxial cable is faster than mechanical deterioration because "Kaiyo" is very stable and four anti-swing of the armor strands, we measure periodically a high ropes attached on the towed frame prevent swinging. frequency characteristics of the cable for estimating The maximum deep tow operating wind speed is cable life. approximately 20 mls in "Kaiyo", wiiile it is less than 15 mls in "Natsushima". 2.7 Gimbal Sheave and Deep Tow Display 2.10 operation of the JAMSTEC/Deep Tow 8 Gimbal Sheave is an unique tool which was specifically designed for the deep towing(Fig.1) 5 . The The JAMSTC/1Deep Tow is operated by Deep Sea characteristic features of the Gimbal Sheave is that Reserch(DSR) group organized by four to six members on it is possible to eliminate damages of the tow cable board and no supporting technitian is required. All at the sheave and also to measure all towing data such on board check, maintainance and repair are carried as cable length, tension, towing angle and rotation of out within the DSR group. Launch and recovery of the three gimbal axes(Fig.3). These data are processed by deep tow, deplayraemt of transponders, operation of a micro conputer and displayed on a color CRT named acoustic positioning system and winch control are Deep Tow Display(Fig.4). This is useful not only to carried out by ship's crew. understand visually and quantitatively the status of deep towing but also to maneuver the ship. The deep tow sonar is towed at 60 to 120 meters above Approximate positions of the towed frame which is the sea floor and at speed of appoximately 2 kt. obtained by the deep tow display is used as a back up According to our experience in operating the deep tow of acoustic positioning. system, it is very important to keep the cable inclination as vertical as possible in rough 2.8 Precise Underwater Positioning topographies. If the cable inclination is very large, the towed frame is apt to collide with the steep A precise underwater positioning is to the utmost slope, because the depth of the towed frame will not important for the deep tow surveys. A long base line change rapidly with cable hoisting. Usually, the (LBL) acoustic navigation system was purchaced from towing speed of the sonar is faster than the camera, ORE in 1977 6 . After several trials and errors the damage of the sonar by collision is greater than the system was almost in practical use in 6000 meters of camera. water depth by 1982. From 1983 to 1986, we improved software and updated the onboard units such as In 1983,we lost the deep tow sonar in the steep wall computer, displays, receiver etc.. we also modified a of submarine caldera of which average gradient was 35 release mechanism of the ORE transponder to improve degrees. As the cable inclination at that time was 27 release reliability in the depth of 6000 meters7. degrees relative to horizontal plane, the tow cable This LBL system was intended to be used on the ships touched the slope. Without knowing the fact, the tow of opportunity. cable.was hoisted at full speed to keep the sonar away from the slope. Shortly, the sonar touched the bottom Taking these experiences into account, new underwater wbile the tow cable continued hoisting. Finally, the positioning systems, which couprise of LBL and SSBL, cable stop cam out and the sonar was lost. This by Oki Electric Company (OKI) -were equipped on the accident was a valuable lesson not to hoist the tow mother ship "Natsushima" and SSC "Kaiyo" in 1982 and cable by force but to watch the cable tension 1985 respectively. The most advatageous feature of carefully and to keep the cable inclination as the SSBL (Super Short Base Line) acoustic positioning vertical as possible. system is that there is no need to put transponders on the botom if the ship positions could be obtained by The maximum operating sea state for the deep tow is up LORAN C or GPS. As an accuracy of the SSBL system is to 5 by "Natsushima" and up to 6 by "Kaiyo". Although 1.5 % of slant range, the accuracy degrades with launch and recovery would be possible in more rough increasing water depth. Although the SSBL is not so sea conditions, ship control and camera observation is accurate as LBL in the deep water, it is useful in difficult and sonar record degrades. surveying wide area by the deep tow. 3.SIGNIFICANT RESULTS AND FINDIMS 2.9 Ship Capability and maneuverability BY THE JAMSTEQ/DEEP TOW As the deep tow itself has no mobility, the capability 3.1 Experiments for Finding Artificial Objects to guide the deep tow sonar or camera to the desired in the Deep Ocean point or on the intended line depends on a ship maneuverability and positioning accuracy. Both mother Experiments to search for small artificial objects by ship "Natsushima" and SSC "Kaiyo" have excellent use of deep tow sonar and camera were carried out in maneuverabilities at low speeds, and equipped with 1983 and 1986. precise surface and underwater positioning system. Especially, the SSC "Kaiyo" has a dinamic positioning In 1983, 22 drum cans, 200 1 in volume, 60 cm. in system (DPS) and has a capability to steer any diameter, 90 cm. in length and 200 kg in water weight, directions at low speeds by use of two main were deployed over the side to a depth of 5700 meters propulsions and eight side thrusters. It is usually in the North West Pacific Basin. These targets were very difficult to steer the ship in a side wind at low deployed within a diameter of 50 meters at the speeds. The hight of the work deck of "Kaiyo" is 7 surface. According to the sonar survey, they were meters above the sea level. Although this is three scattered approximately within 300 meters on the 1255 bottom. This mans that the drum cans dropped almost fiber-optic cable instead of a coaxial cable (Fig.8). vertically through 5700 meters of water colum. After Already we finished factory and at sea tests of the the sonar survery, the deep tow camera was guided to fiber-optic tow cable. the target area by precise LBL acoustic navigation. Finally, drum cans were observed and photographed for 5. swomy three times. A real time color TV, which is installed on the In 1986, almost the same experiment was carried out in 3000-meter deep tow camera, is one of the most the Hachijo Depression, 150 nautical miles south of characteristic survey equipment in the JAMSTEC/Deep Tokyo. Fifteen drum cans were deployed to a depth of Tow system. it is very useful both for geological and 1400 meters and they were searched by sonar and camera biological investigations. as in 1983. The drum cans were detected by sonar and observed by camera for five times (Fig.5&6). The A manned submersible is highly sophisticated importance of this experiment is that the sonar echo man--machine system wiuch requires marry operating was identified as the drum can, because the position personnel and lots of money to operate the system. On of the sonar echo coincided with that of the camera the other hand, the deep tow requires less people and where the drum can was sighted(Fig.7). Another mmey than the manned submersible, and could be important finding is that the same drum can was operated in rough sea conditions. observed for three times and the distribution of each position was approximately within plus minus ten Sea Beam, JAMSTEC/Deep Tow and manned submersible meters(Fig.7). This is a measured realtive accuracy "Shinkai 200011 are the basic tools for the deep sea of the LBL positioning system. research in JAKSTEC. A heavy duty ROV "Dolphin 3K", which has a depth capability of 3300 meters, joined 3.3 Significant Scientific Firklings the deep sea research from 1988. Between 1977 to 1988, more than fifty five deep tow 6. ACKNJALEDGEMERM cruises for experiments and site surverys were carried out. Singnificant scientific findings so far are we acknowledge the contributions of several colleagues listed below. in JAMSTEC, especially in the Deep sea Research Department, who assisted in developing and improving Min 1982, remarkable ripple marks and sand dunes at the JAMSTEC/Deep Tow system. We are grateful to Dr. a depth of 2000 meters, which indicate strong F.N. Spiess and Mr. D.E. Boegeman of Scripps bottom current, were found in the Suruga Trough, 70 Instituiton of oceanography for cordial advice and nautical miles south west of Tokyo. encouragement. Part of this work was supported by (2)In 1983, a big submarine earthquake(M[7.7) occured Bureau of Atomic Energy of Science and Technology in the central sea of Japan of which water depth Agency. was around 3000 meters. many cracks, small trenches, teared fishes and yellowish deposits were 7. REFERENCES found on the sediments9. (3)In 1986, biological canunities by giant clam and 1 Hotta,H. et al. "Preliminary Sea trial of Deep tube worm originated by cold water spring were Sea Floor Survey System!', JAKSTECTR (5), P.27-44 found in the Sagami Trough, 40 nautical miles south 1980 (in Japanese) west of Tokyo. 2 Chtsuka,K., Nakanishi,T. and Hotta,H. "Deep Tow (4)In 1987, hot springs and biological camninities by Sonar System", JAMSTECTR (8), P.1-28 1982 (in giant clam, white crab etc. were found in the Japanese) submarine volcano Kaikata Seamount in the Bonin 3 Tsuchiya,T. and Chtsuka,K. "Development of TV Islands" System for the Deep TbW', JAMSTECTR (11), P-13-29 (5)In 1987, a hydro-thermal activity and associated 1983 (in Japanese) biological camunitites were found in the North 4 ohtsuka,K., Tsuchiya,T. and Hashimoto,J. Fiji Basin, south of triple junction. "Development of Color TV System for the Deep Tow", (6)In 1988, hydro-thermal activities and associated jAMSTECTR(19), P.249-262 1988 (in Japanese) biological ccffm-mities were found in the Okinawa 5 Momma,H. and Hotta,H. "Development of a Gimbal Trough. Type Sheave for Deep Towing " Proc. Oceans 183, vol.I, P.270-273 1983 4.FUTURE DEVELOPMENTS 6 Nakanishi,T. et al. "Laboratory and at Sea Tests on Long Base Line Acoustic Navigation Systed', As a Japanese 6500-neter manned suhmrsible will be in JAMSTECTR (3), P.39-49 1979 (in Japanese) operation from 1990, it is necessary to establish a 7 Momma,H., Tsuchiya,T. and Hotta,H. "Improvement new 6000-meter site survey system. The deep tow of Release mechanism for Transponders", JAMSTECTR survey efficiency, i.e. coverage per day, decreases (9), P.67-74 1982 (in Japanese) with increasing water depth because of great 8 Momma,H. and Hotta,H. "Operating Techniques of hydrodynamic resistance of the tow cable. At present, Deep Towing and Accurate Positioning", Proc. 8th the observation swath width by the deep tow camera is ocean Engineering Symposium, P.57-62, 1987 (in 6 meters -square--at maxirma. Also, it is very Japanese) difficult to stop and observe at a point of interest. 9 Hotta,H. et al. "Visual observations in the In above reasons, we are planning to increase the Central Area of the 1983 Central Sea of Japan visibility of the camera and to add a small ROV which Submarine Earthquake", JAMSTECTR (14), P.37-53 1985 could be launched from the deep tow. For these (in Japanese) purposes it is by all means necessary to use a 1256 10. Deep Sea Research Group "Preliminary Report of Deep Tow Survey in Kaikata Sea Mount (DN87-3-MCS) -IJAMSMDCM (19), P.223-230 1988 (in Japanese) 7 77 7@,FTr T;;F I FIII -X-* MWO, VIO, Z, X FA hr V UK -V j Fig.l. A-frarre, Gimbal Sheave and deep tow sonar on Fig.2. Deep tow camra in recovery. Four unti-swing board Semi-Subfferged Catamaran "Kaiyo". Kevlar ropes are visible in the comers of the towed fraire. S-- S ... d I kt 2kt 3 kt Fig.3. All the deep tow cable data are neasured by Gimbal Sheave. This figure shows relation between ship 0 -0 speed and cable tension. 7 C.bl. 0 C.bl. A,gl, C.@I. L-gth L L 0 15 30 45 14!.00 1 L5 30 45 15:00 1 5 45 16@00 T i.. .1 0., .. . ..... Fig.4. Deep Tow Display is useful for visually recognize the towing "711: status and also for ship .... ... . . maneuvering. 1257 0 U) Fig.5. Sonar echoes (in circles) of 2001 drum cans Fig.6. One of fifteen drum cans on the sea floor of filled with concrete by a 100 kHz side scan sonar. 1400 meters of water depth. This No.7 drum can was The echoes are intentionally enhanced for visual sighted by camera for three times. convenience. I 'DRUM N.. 2 8,12, 13, 4,15 3 5, 6 (D 9 , 11 *4 IRON PLATE 0 0 SONAR ECUO 0 CAMERA CONTACT 0 0 0 2 3 0 0 0 000 Or-,* 4 6GAr 1 0 No. 7 N.. 2 N.. 10 0 0 0 100 200 300 400 500 M Fig.7. Positions of saiar echoes (circles) and camera Fig.8. Fiber-optic tow cable which was manufactured contacts (dots) are illustrated. Numbered stars are by Ocean Cable Company (OCC) for testing. positions where drum cans were deployed over the side. They drifted approximately 60 meters toward the south during descent. 1258 EAVE III UNTETHERED AUV SUBMERSIBLE J. Jalbert, M. Shevenell, S. Chappell, R. Welsh, R. Blidberg Marine Systems Engineering Laboratory University of New Hampshire Durham, New Hampshire 03824 ABSTRACT future for various advanced technology development concepts. The Marine Systems Engineering Laboratory of the University of New Hampshire 2. PROGRAM GOALS AND OBJECTIVES designed and constructed two new untethered AUV submersibles in 1987. The primary goal of the program described These vehicles were designed as testbeds herein was to design and develop two AUVs for the investigation and demonstration of with state-of-the-art processing and data several concepts including: multi-vehicle storage capability, navigation, and sensor cooperation, inter-vehicle communication, systems. The electronic hardware and and automated decision making. These software architectures are designed such vehicles are designed such that additional that knowledge-based concepts could be sensors and software can be included to developed and tested on the vehicles. augment the system and make it an This involves designing a system in which adaptable autonomous testbed. Al languages such as Lisp can be ported This paper describes the basic system and run with a real time operating system hardware and software architecture as well on the vehicles' computer hardware. The as the existing sensor subsystems. it system is also designed such that a high describes results of underwater tests baud rate (4800) acoustic link can be used conducted in the fall of 1987 with respect for inter-vehicle and vehicle to launch to preliminary sensor and vehicle communication. This link is also used to performance. This paper also describes transmit real-time compressed video plans for future developments and imaging data between the vehicle and the important concept tests of these vehicles launch site, and can provide tele-presence which are of significance to the long capability. range application of autonomous submersible technology. 3. EAVE III SYSTEM DESCRIPTION The EAVE III is an unmanned untethered 1. INTRODUCTION submersible. It is designed as a highly maneuverable, sophisticated system which The Marine Systems Engineering Laboratory can be easily modified to perform various of the University of New Hampshire has types of missions and experiments. A been involved in developing state-of-the- drawing of this vehicle is shown in Figure art technology for autonomous underwater 1, and a vehicle specification is shown in vehicles (AUVs) since 1977. This Figure 2. laboratory has previously developed two The vehicle consists of two buoyancy tubes generations of unmanned AUVs. which contain the system computers, two The third generation system, described in battery tubes which house the lead calcium this paper, was developed as a generic batteries, and six DC brushless thrusters testbed with an increase of several orders used to maneuver the vehicle vertically, of magnitude in processing power and horizontally (forward and backward), memory capability. It was developed for laterally as well as in heading (yaw ). the purpose of testing knowledge-based The system sensors include an acoustic guidance and control concepts (AI), inter- obstacle avoidance system operating at 300 vehicle acoustic communication and cooperation, and bandwidth compression and kHz, an,acoustic depth (uplooking) and transmission of imaging systems data. it altitude (downlooking) sonar both will serve as a versatile testbed in the operating at 200 kHz, a pressure depth sensor, a water temperature sensor, a CH2585-8/88/0000-1259 $1 @1988 IEEE flux-gate compass and an acoustic long and 4. SYSTEM STATUS AND SUBSYSTEM TEST short baseline navigation system which RESULTS operates at either 30 kHz or 100 kHz. The two vehicles were designed and A work package installed on the EAVE III fabricated during the course of 15 months, vehicle was an underwater video system * and underwent extensive laboratory and The video camera used for the system was a field tests. A total of approximately 200 Sub Sea Systems, Inc. SL99/WA silicon untethered mission tests were conducted intensifier target (SIT) camera, capable over the late summer and fall of 1987 at of operating from full sunlight down to the Lake Winnipesaukee test facility in 0.0001 ft. candles. It offers a 110 New Hampshire. A one mile square area of degree angle of view, with auto iris and the lake was surveyed acoustically prior motorized focus. A Kodak MVS-500 8mm VCR to testing to develop maps of the lake handled video recording, which provided a bottom for use in the software world light-weight, compact power solution to model. video documentation. The field tests progressed from sub-system An RF telemetry system was installed on tests to vehicle performance tests to the vehicle for use in fresh water mission demonstration tests of various environments. The telemetry system was types. The demonstration tests included: used to transmit data from the vehicle to (1) cooperative testing utilizing the the launch barge and to a shore station. National Bureau of Standards RCS The system consists of Joslyn Defense controller as the high level processing Systems RFM-2A RF modems operating at 27 system, (2) conducting a barrel search, MHz, and standard C.B. antennas. detection, and video taping mission using the MSEL system, and (3) employing Lisp The computer system architecture (Figure software in the MSEL high level software 3) is time ordered, modular, and system on the VME computers to conduct a hierarchical. The low level consists of mission similar to (2) above. three 68000 processors which handle all of the fast (<l second) system requirements. Integrating all of the subsystems into a This lower level effectively can run the total vehicle system presented the vehicle in a preprogrammed scenario predictable problems normally encountered without the upper level. This is in such an effort. These problems important in determining basic vehicle generally involved the categories of performance and diagnostics of subsystems. interface definition, interprocessor The low level processors perform the communications, acoustic signal filtering, functions of controlling and reading etc. All of the problems of any sensors, preprocessing data, monitoring significance were solved or are understood vehicle status parameters, and controlling and are being corrected at this time. vehicle motion by constantly modifying the six thruster speeds as a result of The software in this system, for both the processing the sensor data through the low level and high level architectures software control algorithm. provides significant capabilities for an AUV. This software proved to be very The interface to the upper level is reliable as evidenced by the number of designed such that either the MSEL high test missions which were accomplished. The level knowledge-based architecture, or any most dramatic and significant other architecture (such as the National accomplishment in this area was the Bureau of Standards (NBS) RCS controller) development of a means of running a could be tested with this system. symbolic language (Lisp) on a real time operating system (pSOS) on an untethered AUV submersible. This advance not only The high level computers consist of 68020 demonstrates feasibility, but also makes processors housed in a VME bus system. possible the actual implementation of The VME bus has been chosen f or use in symbolic processing concepts and provides EAVE as the framework upon which to build a development environment for further a testbed for AUV knowledge based systems research in automated decision making. development. Each module in the high level model is allocated to a separate 5. EAVE III PERFORMANCE processor, in our case a Motorola 68020 .CPU. These processors all share a global The EAVE III vehicle has two modes of bus and have significant dual port vertical control. It is capable of either memories to allow local use by the CPU, running in the depth or altitude (terrain as well as remote communication by other following) control modes. The modes can CPUs sharing the bus. be changed as the vehicle is underway by simply changing one parameter in the 1260 command string. The data indicates that telemetry system on the barge then relayed it controls equally well in either mode. the data to the shore facility for Throughout the mission, the vehicle plotting of vehicle position. remains within +.6 meters of its altitude or depth. The NBS team was able, over the testing period to demonstrate RCS control by The EAVE vehicle is capable of motion in 3 running the following types of missions: dimensions (x, y, z) as well as yaw. In order to determine performance a series of 9 Straight line motion tests were conducted to determine the * Diamond pattern motion vehicle's response in each axis. A * Pseudo circular motion tabulation of results is summarized in * Raster search notion Table 1. The data indicates that for the control gain settings used the vehicle 7. MSELINUSC DEMONSTRATION TESTS typically moves in the forward direction with an average velocity of .5 Mps and The purposes of these field tests were to: little, if any, overshoot. Backward motion velocity is .35 Mps with little * Demonstrate EAVE III search capability. overshoot. The backward velocity is * Demonstrate bottom object detection slower than forward due to the decrease in capability. prop efficiency in the reverse direction. * Demonstrate modification of vehicle Similarly the parameters for slide motion, control as a result of detection, and vertical motion and rotation can be read control of a camera to videotape a directly from the tabulation. it is bottom object. important to note that the vehicle possesses an oscillation in yaw, of about Divers from the NUSC facility laid three 10 degrees with a period of oscillation of objects in the test area- These objects about 20 seconds. This problem is consisted of 2 sets of 55 gallon drums currently being addressed in a new welded together on end to form an object controller designed for this vehicle. with dimensions of approximately 8 feet in TABLE 1 length and a diameter of approximately 30 EAVE III Performance Summary inches. The third object was an 8 foot aluminum drum of approximately 20 inches MOTION TYPE AVE. VEL. OVERSHOOT OSCILTATION in diameter. VEL. (KPS) DEG. SEC. Forward .5 none 10 - 20 Barrel Detection and Video Taping Backward .35 none same The vehicle was to detect the barrels, Slide .20 none same halt its forward progress, turn on a video Climb .35 none same camera and VCR, perform some maneuver over the barrel, turn the camera off, and then Dive .25 none same continue with the original mission. This deg/sec sort of mission interruption is composed Rotate 6 12 same of two parts: the detection of the,bottom object and the reaction to that detection. NOTE: Average velocities can be increased by changing gain settings in software. Experiments with various barrel detection algorithms were tested in the Data 6. MSEL/NBS COOPERATIVE TESTS Assessment and Situation Assessment modules of the laboratory emulation A series of tests were conducted both in system. The detection of a bottom object the laboratory as well as in the field in is based on changes in the altitude order to properly interface the NBS RCS readings. The Data Assessment algorithm high level controller software to the EAVE deals with the computation of the slopes III low level system. Most of the between the various readings, and the problems were resolved, and NBS tested Situation Assessment algorithm looks at their software on the vehicle at the Lake sequences of slopes. Certain sequences Winnipesaukee test facility. The software are taken to mean an object was just system was able to control the vehicle passed. through the NBS 11EMOVE11 level software. A raster search of the area was made to NBS also designed an interface to the RF determine the location of the objects. telemetry system, described earlier, in This was successfully accomplished. The order to have a means of telemetering planned line spacing between tracks was 2 vehicle position data in real time to the meters. The data suggests that the barge and island terminals. An vehicle usually was within I meter of interference problem was avoided by planned track even considering the yaw transmitting data at 10 second intervals problem discussed earlier. At the ends of for a 1 second period to the barge. A UHF each track in the raster search the 1261 vehicle performs a 180 degree turn and 9 Interrupts the default plan simultaneously slides 2 meters to the next 9 Formats a series of new vehicle path track position. commands e outputs path series to the IV-1624. The objects were detected using the downlooking acoustic altitude sonar system A task on the 1/0 board then handles the described earlier. This system has a 10 sending of path segments to the Effector degree beam and operates at 200 kHz. It Manager computer. When the last default has a resolution of approximately 4 cm and path segment has completed, the path which proved adequate to the task. the vehicle was executing before the barrel was detected is continued. Further The concept was to fly over the objects, execution follows the preplanned path set and when an object was detected, turn on a in the Effector Manager. downlooking video camera, (also discussed The Planner and Lisp environment performed earlier). The vehicle would then back up as expected when a message was sent from to the object location, climb to some the Sensor Manager stating that a barrel preselected altitude above the object, had been detected. Upon detection of a descend along the original track path and barrel, the default plan was to: then resume its mission plan. In conjunction with this mission, NUSC e Hover at current location; turn on provided divers to videotape the vehicle camera and light as it maneuvered over the objects. A e Vertical climb of 10 meters videotape of these missions was attained 9 Vertical descent 10 meters; turn off and produced by NUSC. camera and light. Figure 4 is a plot of raw position data A plot of the successful mission is shown for one of the object (barrel) detection in Figure 5. This figure shows a plot of missions. The location of the objects are x, y, z, altitude and bearing as a easily identified by the very dark areas function of time. The vertical hash lines which are created by the vehicle backing represent command changes of the up over the barrel, and then climbing and preplanned mission contained in the descending to resume the planned path. Effector Manager. The time scale shown Notice also that one track veers off for represents about 15 minutes. The data approximately 10 meters before returning indicates that the Sensor Manager sent a to track. The vehicle missed the barrel barrel detect message to the upper level on this track. It appears that a system shortly after command #6 started. multipath was read by the navigation system and believed as true for 9. FUTURE PLANS approximately 10 meters of track. The lane spacing for this mission was planned A current major effort of this laboratory at I meter for the three sides of the is aimed at fully implementing the EAVE figure. vehicle's high level knowledge-based 8. LISP PLANNER ON REAL-TIME architecture on the EAVE III vehicles now OPERATING SYSTEM IMPLEMENTATION that feasibility has been proven. This will allow for further research and Field tests were conducted with the demonstration in the area of automated reduced version of the hardware decision making which is a key development architecture described earlier. The area for untethered AUV submersibles. purpose of these tests were to use the high level MSEL system in conjunction with We are also currently evaluating and the low level system to perform implementing an acoustic communications intelligent missions. Of particular link for both image transmission and tele- interest was the testing of the Lisp presence capability. Planner module which would demonstrate the feasibility of placing a Lisp or symbolic Current plans also include flying both processing environment on a real time vehicles simultaneously in the near operating system on an AUV. future. This will provide a realistic environment in which to demonstrate The mission begins with the low level developments in the area of communications vehicle executing a preplanned search to and between vehicles, as well as mission which takes it over a submerge ,d concepts of multi-vehicle cooperation. barrel. This preplanned mission continues with the upper level monitoring all data Recent technology advances in power which is passed from the Sensor Manager sources indicate that this vehicle can be and Effector Manager. Once a barrel is fitted with a new power supply system to detected in the Sensor Manager it passes lengthen its mission time to 80 hours in this information along with current the same physical package that currently vehicle position. Once the barrel.data is exists on the vehicle. This is also received, the planner: planned for the near future. 1262 EAVE III VEHICLE SPECIFICOONS Overall Dimensions Height - 51 inches Width - 41 inches EAVE III Length - 51 inches VEII"TiCAL Weight 1000 lbs (air) THRUSTERS RF ANTENNA N ..... .... ON Speed/Endurcnee :1 knot / 4 hours AC USTIC S'ONAR H DEPTH SONAR 9013"Veo A Max Speed 1.5 knots CONT90L JVUP RESSURF SENS A Power (battery) lead calcium 2KW-HRS K 09 COMPUTERS Payload 50 (bs. ALTITUDE OBSTACLE SONAR AVOIDANCE SLIDE Thrusters DC Brushless (5) SONAR tHAUSTPRS Degrees of Freedom 5 Depth Rating 500 feet PROPULSION -ORWA BATTERIES rIHMUSPEDRS Navigation System Acoustic - 30 KHz or 100 KHz ELEMETnY COMPUTER COMPA S Long Baseline / Short Baseline ELECTRONICS BATTERIES (BOTTOX CENTER) Sensors Obstacle Avoidance Altitude Sonar Figure 1. Water Temperature Flux-Gate Compass Pressure Transducer System Monitors Communications Acoustic Telemetry RF Telemetry Computers 68020 based system on VME Buss 68000 based system (MSEL-CPU) General Characteristics Highly Maneuverable Autonomous Unmanned Untethered Knowledge Based System Figure 2. SUPERVISOR ASSESSOR PL NMER DATA ASSESSMENT GUIDANCE SYSTEM SENSOR EFFECTOR MONITOR MANAGER MANAGER UN LEWL KBS/EAVE SOFTWARE ARCHITECTURE DATA SU P PL E A R N .UIB, A SSE _.E.T SENSOR EFE VOR A UAGERAN Figure 3. 4263 Barrel I-Un video Barrel Barrel Barrel ems 02 y c m Figure 4. 6 Lisp mission v 8 a 110 ?4M.44 X MO vM3. y 23097- aloes. 1993. z Ism. A Isle. L Iles. 4" B me. E too. 14. R I Figure 5. Barrel '0 @V_ 1264 An Autonomous Underwater Vehicle (AUV) Flight Control System Using Sliding Mode Control Frank Dougherty, Tom Sherman, Gary Woolweaver, Gib Lovell Martin Marietta Aero & Naval Systems Baltimore Md. ABSTRACT 3. AUV FLIGHT CONTROL SYSTEM This paper discusses ongoing work at Martin Marietta in the design of Introduction an advanced AUV control system. This work focuses on the MUST is capable of high speed transit using its control surfaoes and application of a nonlinear control technique known as sliding mode main propeller, and hovering near zero speed with its four thrusters control (SMC) to an AUV testbad being developed at Martin Marietta. and main propeller. In addition there is a third operating region AUIPs present new challenges in the area of dynamic closed loop which is not pure transit or hover. This hybrid region we have control. AUV control systems must be able to robustly control defined as transition. Transition requires the use of all available movement at all angles of attack or sideslip, adapt to changing force effectors - control surfaces and thrusters. The significant mission payloads and dynamics, and provide a tolerance to control achievement of this IRAD has been the design of a robust FCS for element failures. SMC provides a theoretical framework for the the hover and transition regimes. This FCS allows the AUV to use its design of controllers that can meet these requirements. The AUV full dynamic operating range, thereby increasing its mission control system has been designedusing SMC andis being prepared capabilities and effectiveness. for at sea testing onboard the Martin Marietta AUV testbed in early 1989. In addition a SMC trajectory transit controller was developed in conjunction with the Woods Hole Oceanographic Institute's Deep 1. BACKGROUND Submergence Lab (2). The design of the the transit, hover, and transition controllers will be discussed in detail as well as the modelling In 1984, Martin Marietta Aero and Naval Systems began a program required for these model based controllers. Two different models of independent research and development (IRAD) directed were developed and used for the controldesign and testing effort: a 6- towards the development of critical AUV technologies. As part of DOF constant coefficient nonlinear model valid for alphas to +/- 20 0 for this program an IRAD was begun which focused on the problem of the transit,regime and a 6-DOF nonlinear model valid for all alpha, for dynamic control of an AUV. The primary goal is the development of the transition and hover regimes. an AUV flight control system (FCS) that can increase the mission capabilities of an AUV, and that is also robust to varying Sliding Mode Control and Trajectory Tracking environments, dynamics, and control element failure. SMC has been proposed for both robotic and air/spacecraft applications (4-7). SMC has excellent robustness, stability, and To demonstrate critical AUV technologies, development was begun in disturbance rejection properties. SIVIC has been studied 1985 of the Mobile Undersea Systems Test (MUST) Laboratory (1 &2). extensively in the Soviet Union over the past twenty years(8). The The MUST vehicle (fig 1) is a modular underwater vehicle capable of development of this FCS follows the SIVIC development of Slotine serving as an open ocean test platform for a wide range of R&D (9) for robotic applicaitons. projects. The baseline vehicle will be 30 feet long, 4.5 feet in diameter, capable of speeds from 0-8 knots, with an operating depth Tt*ectoty Tracking Using SMC of 2000 feet. MUST will undergo sea trials in late 1988 and early 1989, A trajectory for an underwater vehicle defines its position and and then will be available to support government and Martin Marietta associated velocity at any instant in time. The use of trajectory R&D programs. The MUST lab will serve as the AUV FCS testhed. tracking controllers allows the path of the vehicle to be explicitly defined and tracked. A sliding surface is a manifold in the phase 2. INTRODUCTION plane. This surface is chosen to be time varying such that it always contains the commanded states (Fig. 2). The dynamics of the AUV Proposed missions for AUV's require that they have a wide range of can be represented by scalar equations of the form: dynamic maneuvering capabilities. AUV missions will require the vehicle to be capable of tracking complex trajectories X, (t)=f(X;t)+b(X;t)u(t) (1) orrinidirectionally. This includes the ability to operate in regions of high angles of attack (alpha), where the dynamics are highly nonlinear and as yet cannot be predicted with a high degree of where: X =: (x ix for (61,n) = actual state vector accuracy. The application of nonlinear model based control f (X-,t) = nonlinear dynamics techniques significantly increases AUV capabilities and mission b(X-,t) =control gain effectiveness. This is because these techniques use all available u(t) =control setting information about the nonlinear dynamics directly. A single nonlinear controller can be designed that will allow uniform operation over the For simplicity, the subscripts will be dropped. entire range of dynamics with robust performance and stability guarantees. CH2585-8/88/oooo- 1265 $1 @1988 IEEE The sliding surface for this system can be defined as (9): system by definition will always remain inside the boundary layer then some level of tracking performance of the system can be S =X +X 'X (2) guaranteed. The tracking errors will be bounded. There is a trade- off for SMG between bandwidth, tracking precision and modelling where: X x- X d uncertainty (110): V 2 * Uncertainty (6) Tracking precision = (Bandwidth) X=X- X d Hover Control )(d=(Xd, Xd, Xd =desired state vector Hover has been defined as that operating region where the thrusters are used to control the dynamics of the AUV. MUST has Lambda in the context of SMG will determine the bandwidth of the five thrusters: two vertical, two lateral, and the main propeller along closed loop system (9). the long axis of the vehicle. Nominally the hover region is limited by forward speed as the two vertical and horizontal thrusters become The resulting sliding control law is: ineffective at speeds greater then 1.5-2.0 knots. The thruster arrangement gives MUST the ability to control five degrees of u f k sgn(s)+ id - ;@x_) / lo' (3) freedom: yaw, pitch, sway, surge, and heave. Roll is the only degree of freedom that cannot be actively controlled. During hover where: sgn sign function the roll will be passively controlled by the gravity buoyancy separation along the Z-axis. The hover regime is characterized by k Time varying feedback gain nonlinear hydrodynamics which cannot be modeled with a great lo', f '= estimates of nonlinear dynamics deal of accuracy. This requires the use of nonlinear model based control to meet proposed performance requirements. The time varying feedback gain is determined by the level of Control System Structure modelling uncertainty and neglected dynamics and is defined as: The SMC hover control system that has been designed is a single trajectory controller that is valid for all alpha and all speeds within the f (XI) -f '(X;t) I < F(X;t) (4a) hovering region. The sliding surface for each loop has been chosen as in eq. 2 with an additional integral term to eliminate b = (bmax / bmin) 1 @2 (4b) steady state errors. k = b(F+ il )+( b -1) f +id- X71 (4c) Commands are issued to the controller in a three dimensional cartesian inertial reference frame. The commands consist of initial Tl = positive constant and end point positions (X,Y,Z), attitudes (pitch, heading), associated rates, and times at the beginning and end points. One of the major problems in the design of trajectory tracking systems is where: F(X,-t)= bound on the additive uncertainty the generation of suitable trajectories to meet system and b= bound on control gain uncertainty operational requirements. This problem is further compounded when required to do this on-line. To address this problem we have bmax, bmin = bounds on b' come up with a simple, and effective way to generate trajectories The control law (eq. 2) is defined so that the following sliding on-line. The trajectory generation portion of the controller will be discussed in more detail later. Once the trajectory has been condition is always satisfied: generated in the inertial reference frame it is transformed into the 1/2 d/dt(s ) 2 < - 11 IS1 (5) body fixed coordinate system. The body referenced trajectories are tracked using five single The sliding condition ensures that all trajectories not on the input, single output (SISO) SIVIC controllers. The dynamics of the sliding surface will tend towards the sliding surface. The objective of system are second order as in eq.1 with the state vector (X) SIVIC is to design a control law such that the system will remain on consisting of the vehicle attitude and translational rates, b(X;t) being the sliding surface (S=O) in which case the tracking errors decay the gain of the thruster along each DOF, u(t) is the percentage of exponentially. If the system is not on the sliding surface (SYCO) then available static thrust along each DOF. The nonlinear it will be driven back to the sliding surface because the control law hydrodynamics and inertial effects are reflected in f(X;t) for the body has been designed to satisfy the sliding condition (eq 5). The and b(X;t) for the thrusters. The system inertia consisting of inertia control law has two components. First, there is a component that and added inertia is a matrix which has been diagonalized for the compensates for the known nonlinear coupled dynamics (f'). controller and is included in f ' and W. The inertial crosscouplings Second, there is a feedback component (k) that compensates for are generally small and are accounted for in the feedback portion of model uncertainties or simplifications, disturbances, etc., making the controller. The thruster arrangement along with the the resulting control robust. diagonalized inertia matrix allows us to effectively decouple each control loop. Although the resulting SIVIC system is SISO all The switched control (eq. 3) will cause the system to chatter along crosscouplings between the degrees of freedom are accounted for the sliding surface at a high frequency as it approaches the desired in the estimates of f ' and W. The uncertainties and neglected state. To prevent chattering a thin boundary layer around the dynamics in the model have been bounded and are included in the sliding surface is defined across which the feedback is interpolated. SMC feedback. The uncertainties and disturbances which are used to define the boundary layer thickness are dynamic. Therefore the boundary The controller as described requires full state feedback The layer should also be dynamic in order to always fully realize the best baseline MUST system has only sensors for angular position, and tracking performance within the control bandwidth (10). Since the angular rates, and a simple water relative forward speed sensor. A 1266 second hover controller was designed using SIVIC that only used end conditions and is subject to velocity, acceleration, and no- this partial state information. This second hover controller does not overshoot constraints. The no-overshoot constraint is imposed in have nearly the performance of the full state feedback hover order to prevent vehicle collisions withobstacles. controller. The second controller is only intended to be used in midwater where precise maneuvering is less of a requirement. The end conditions for each degree of freedom are specified initial When the vehicle is operating in proximity to the bottom in a and final positions and velocities: pi , pf , vi , and vf. The trajectory precision hovering task the full state feedback hover controller must generation problem is to determine a position function p(t) with t E be used. In this mode it is logical for the vehicle to station itself (0,11 such that p(O) =pi , p(T) =pf , d/dt(pi) =vi, and d/dt(pf) =vt and relative to an object on the bottom. To do this MUST will be outfitted such that velocity, acceleration and no-overshoot constraints are with a high frequency acoustic navigation system. This system satisfied. The type of scalar trajectory that is used for each degree consists of a set of acoustic transceivers on the vehicle and an of freedom is of the form constant acceleration-constant velocity- acoustic transponder on the bottom. The system, known as constant acceleration. A velocity profile corresponding to this type SHARPS (Marine Telepresence, Inc.) was originally designed for of trajectory is illustrated in fig 3. This profile, of course, reduces to use by underwater archioligists but has been modified for use in the commonly used trapezoidal profile when the end velocities vi this application. The system operates at 300 KHZ thus avoiding and vf are both zero. When the starting acceleration a and the the multipath problem, with accuracies of several centimeters. ending acceleration b are of maximum magnitudes, then this is called a candidate velocity profile, and a trajectory corresponding to Control Analysis such a profile is called a candidate trajectory. Candidate trajectories The control analysis for the hover region consisted of; are important in addressing the trajectory existence problem. determination of lambda(control bandwidth) in eq. (2); tradeoffs between model uncertainties or simplifications and tracking The velocity and acceleration constraints for a scalar trajectory are performance; determination of open and closed loop response; specified by the parameters vmax, amax, and bmax and can be sensitivity to time delays etc. graphically represented by a constraint polygon in the velocity vs. time plane as illustrated in fig 4 Trajectories In order to analyze the bandwidth of the vehicle as well as the are constrained such that: closed loop characteristics, the full nonlinear simulation has been used. The analysis consisted of frequency response analysis of the I' I..;__vmax, J'pl..@_amax for t E [0, V) (7) full nonlinear dynamics. A sine wave of varying amplitude and P frequency was input into the command input of one loop at a time. A fourier analysis was done on that particular output state. For a given amplitude several frequencies were run and the results for lpl.-@_brnax for te (t', T] that amplitude were used to create frequency response diagrams. This frequency analysis can also be used during the at sea testing where: V= the time at which the constraint boundaries to evaluate the closed and open loop characteristics of the real due to the acceleration bounds intersect vehicle, but more importantly to verify the model and simulation. This kind of real time frequency analysis has been done for flight An important property for velocity profiles that satisfy the velocity testing aircraft (12). and acceleration constraints is that they must lie within the polygon. In the figure, the maximum velocity for all permissible Selection of lambda is based primarily on the following: profiles is labelled vu (in this case vu=vmax) and the minimum (1) nonlinear frequency response velocity is labelled vl (in this case, vt is determined by the (2) digital loop closure constraint acceleration constraint boundaries). (3) neglected time delays (4) actuator dynamic response For a candidate velocity profile, it is clear that Jal=amax and (5) sensor dynamics lbl=bmax. In addition, once the constant velocity vc is specified, (6) model fidelity (uncertainty) the candidate velocity profile is determined. A candidate velocity profile is represented by vvc(t) and a corresponding position The open loop frequency response of MUST was determined for function is given by: the yaw, pitch, surge, and sway loops. The response was generated with.one loop open at a time. The other degrees of t freedom were run closed loop. These open loop bandwidths give a P(t)=Pi+fO vVC (t) dt (8) good estimate of where the closed loop bandwidths (lambda) should be set. The selection of lambda has a significant effect on Now, for a given set of end conditions and velocity and acceleration the sensitivity to parameter uncertainty. The tracking error is constraints, it can be shown that there is a unique candidate inversely proportional to lambda. The sensitivity to parameter trajectory, if it exists, that satisfies these end conditions. The uncertainty can be reduced by using a bandwidth that is several candidate trajectory is very useful for the trajectory generation times the open loop natural frequency. Extensive parameter problem because of the following: sensitivity studies have been conducted with the 6 DOF model to (1) It the candidate trajectory does not exist, then there is no other determine the significant parameters.The determined values of trajectory that satisfies these end conditions and constraints. lambda are conservative. They will limit the effects of parameter (ii) Suppose a candidate trajectory exists that satisfies the specified uncertainty yet are below the upper constraints. The vehicle will end conditions and velocity and acceleration constraints. Then, if have the potential for improved tracking by increasing lambda, yet this trajectory overshoots, then all trajectories that satisfy the end the stability margin will be large and the control activity should not conditions and velocity and acceleration constraints be excessive. overshoot. The proofs are omitted due to their length. Trajectoty Generation The trajectory generator first attempts to find scalar trajectories that The trajectory generator creates five scalar position functions are candidate trajectories when a move command is given, the corresponding to each degree of freedom when a vehicle move is corresponding position functions of which would be of the form ( commanded. Each position function must satisfy certain specified eq. 8). If some of the candidate scalar trajectories cannot be found, 1267 then, by virtue of the above two statments, no trajectory exists for eliminate steady state errors. The control law resulting from these the given move command. If all candidate scalar trajectories exist, assumptions takes the same form as eqs 2 & 3 where b(X;t) is the then these trajectories can be used for the move or other scalar gain per degree of sternplane deflection, and u (t) is the sternplane trajectories of the form constant acceleration-constant velocity- deflection. constant acceleration may be used provided that the end conditions and constraints are satisfied. The estimation of the pitch forces and moments is based on a low alpha coefficient model of the vehicle dynamics (3). This is Transition control adequate because the vehicle will be operating in a low alpha In this paper we have distinguished between the hover and region where constant coefficient models are valid. transition controller although these will actually only be a single controller. Control of the vehicle in the transition region requires the Alternate control structures such as control from depth to use of both thrusters and control surfaces. The transition region is sternplane were tried but have significant drawbacks such as important because it will allow the vehicle to operate in the very low requiring the measurement of accel e ratio n(2). The depth to speed ranges as will be required to effectively accomplish some sternplane dynamics are nonminimum phase so blindly applying proposed missions. The limitations of the transition controller are SMC will make the system unstable. This can be easily remedied by that the vehicle must operate in a region where the fins can neglecting the nonminimum portion of the depth dynamics in the contribute to the total control. Nominally this will require that the fins computation of f ' in the control law.These dynamics would be to operate at alphas less than 451 in both forward and reverse bounded and included in the feedback gain. motion. Above 451, the fins generate little lift. A requirement for transition control is that there is sufficient overlap between the Closed loop control of heading is accomplished in a similar manner thrusters and fins. The MUST vehicle has little overlap, thus making to the inner pitch loop. Commanded yaw angle and yaw rate are transition control difficult. For purposes of transiton control functions of the desired track. The measured states are u , psi and development, larger thrusters were modeled. r. The sliding surface is composed of yaw rate errors and heading angle errors. The control law takes the same form as the Control System Structure longitudinal equation with b(X;t) defined as the control gain of the The desired control input is determined using a combined control rudder, and the other terms correspond to yaw control. ga:in, which accounts for the reduced thruster performance and the low lift developed by the fins at low forward speed. This combined ControlAnalysis gain is then used to apportion how much of the control force is to be Determination of transit control parameters is analagous to the developed by each of the thrusters and control surfaces. The input hover analysis. The same characteristics must be examined about command structure and trajectory generation are the same as in speeds of 2 to 8 knots. The primary complication is that the hover mode. frequency response becomes a function of forward speed. The nonlinear simulation has been the final analysis too[ for traditional During hover, MUST uses only its four thrusters to provide checks such as parameter sensitivity, stability margin, frequency maneuvering control. As forward speed increases, the thrusters response and expected transient response. begin to lose effectiveness. When this happens, the controller determines whether the control surfaces can provide sufficient lift to Vehicle Dynamic Model augment the thrusters. If the local angle of attack on the rudder or The nonlinear model based controller being developed for MUST sternplane is greater than 45 degrees in either forward or reverse required the inclusion of a hydrodynamic model in its algorithms. motion, the fins are not used for control. If the angle is less than 45 Accordingly, a model of the MUST vehicle dynamics was developed degrees, the fins are commanded to move to a deflection angle that which is valid for all alpha(3), which can accurately model highly provides the required force, assuming it is achievable. Once the nonlinear maneuvers such as are seen during hover. A detailed fins have been deflected, the controller will first check to see model of the thruster dynamics is included in the model, as are whether the fins can produce sufficient control force, and will representative models of the vehicle sensors and actuators. These reduce the thruster command if possible. This logic provides a models comprise the AUV simulation that is being used as part of smooth transition from using only the thrusters to using only the the MUST lab development. control surfaces for maneuvering control. The uncertainties and neglected dynamics in the model have been bounded and are The hydrodynamic model is based upon body-buildup techniques included in the SMC feedback. and relies on semi-empidcal methods for the prediction of forces and moments. The model computes the forces and moments of Transf controlt each component individually, and then sums them together with The transit mode for the MUST vehicle consists of forward motion at interference effects to form the total forces and moments. The speeds greater than 2 ft/sec. The vehicle is controlled in this region advantage to this approach is that fluid phenomena such as stall or by four independently actuated tail fins each with a 60 degree vortex shedding can be properly modeled. The thruster model range of motion. The thrusters are not used for control. Forces splits the developed thruster forces into four components, two along the X-axis are supplied by the main motor. In the transit related to the open water characteristics of the thruster, and two mode, the vehicle states that are controlled are depth, pitch, pitch representing the degradation of thrust due to vehicle motion. rate, heading, heading rate, forward speed, and roll angle. These motion interference effects can be substantial, especially for forward speed. The sensor models include the effects of threshold, Control System Structure limits, and random noise. The control surfaces are modeled as Depth control is achieved with an outer depth guidance loop second order systems with deadband, voltage and slew rate limits, supplying a pitch command to an inner pitch angle control loop. The hysteresis, friction, and hydrodynamic loading. measured states are u, q, and theta. W is estimated from the depth rate and vehicle attitude. The commanded pitch angle and pitch Results rate are functions of the desired depth, depth rate, depth The design of the hover and transit FCS have been completed and acceleration and forward speed. The sliding surface is first order in are being integrated into the MUST vehicle. For purposes of control terms of pitch rate error and pitch attitude error. The sliding surface validation the controllers have been run in the full AUV nonlinear is defined the same as in eq 3 with an integral term added to 1268 simulation. This has allowed for various sensitivity studies to be 1 O.Asada, H., Slotine, J-J., Robot Analysis and Control, New York, NY, conducted. John Wiley Sons, 1986. To show the capability of the FCS, typical hover and transit maneuvers 1 1.J. Gera, J.T. Bosworth, *Dynamic Stability and Handling Qualifies have been selected. Figure 5 shows a hover maneuver where the Tests on a Highly Augmented, Statically Unstable Airplane," Guidance vehicle is moving around an object while staying pointed at the object. and Navigation Conference, Monterey, CA, 1988. This would be a maneuver required to do some close up inspection with a forward looking sensor. Figures 6&7 show the resulting tracking errors throughout this maneuver. These plots show the robustness of THRUSTERS (4) UPLOOKING VARIABLE 3 HP IBM RPM ACOUSTIC MAIN PRESSURE VESSEL BALLAST (2) OA the SMC FCS because the model used in the control design has COMMUNICATION \ =NSOKAR (1) underestimated all forces, inertias, and control gains by 50%. MAIN UTR TRANSDUCER --V I- . RECOVERY MAST CRY a HP IN ENTE Ism RPM TRANSPONDER A sample transit which consists of a low speed turn of 90 degrees coupled with a 100 foot depth change. The resulting tracking error in GIR heading is shown in fig 8 , and for depth in fig 9. The performance of the transit controller is comparable to a linear gain scheduled ID controller. The advantage is that the SMC transit system is a single NI UDE ArTUAtRS @AL T controller with specific stability and robustness guarantees. In addition (4) BOMAR RA ERY DROP SHOT Do BATTERY 24 VOLT GUIDANCE HOPPER the performance should be more consistent with a single controller PPLER. 120 VOLT 20 KWH SENSORS over the entire speed range. MISSIO Do* CHLOOKING BOMAR 20 KWH N A OUSTI CORDER COMM 4,r HIGH PAYLOAD UNICATION ELECTRONICS PACK TRANS UC IN 4. CONCLUSION O.UL:. ELECTRONICS 24V LOAD 24V PAYLOAD RACKS CENTQR BATTERY The use of SMC for the development of AUV flight control systems 24V LOAD EMI BANK (2) provides increased system capabilities and mission effectiveness. A CENTER BARRIER SMC FCS has been designed for an AUV testbed being built at Martin VOS 12% BATTERY Marietta Aero & Naval Systems. The goal is to take this FCS to sea and ANK fully explore and document its capabilities. To date the SMC FCS AFT HULL CROSSECTION VATH TYPICAL HULL CROSSVCTIOM WITH system has shown excellent performance in all operating regimes and PROPULSION BATTERY MODULE INSTRUMENTATION BATTERY MODULES at all angles of attack This performance has been verified through the use of a high fidelity 6-DOF simulation that accounts for dynamic and FIGURE 1. MUST VEHICLE hardware nonlinearities. References Sliding Surface in Phase Plane l.Herr,W.J., CollinsX., "MUST, A Large Versatile AUV Technology X Testbed System," Underseas Defense '87, San Diego Ca., Oct. 1987 2.Herr, W.J., " AUV Technology, Development and Demonstration Program," OCEANS'88, Baltimore, Md., Oct. 1988. trajectory Xd ------------ 3.Dougherty, F., Sherman, T., Woolweaver,G., "Modeling and I Simulation of an AUV," Modeling and Simulation Symposium at C.S. sliding surface Draper Labs, Cambridge, Mass, June 1988. 4.Stalford, H., Garrett, F., "Robust Nonlinear Control for High Angle of X:dlambda X Attack Flight," AIAA 25th Aerospace Sciences Meeting, Reno, NV, January 1987. FIGURE 2. SLIDING SURFACE 5.Vidyasagar, M., Sprong, M., "Robust Linear Compensator Design for Nonlinear Robotic Control," IEEE Transactions on Automatic Control, Vol RA-3, No. 4, pp. 345-351. p 6.Yoerger, D., et al., "Robust Trajectory Control of Underwater Vehicles," Proceedings of the Fourth International Symposium on Unmanned, Untethered Vehicle Technology, Durham NH, June 1985. VC Vf 7.Slotine, J-J,, "Robust Control of Robot Manipulators," Journal of V Np, p Robotics Research, Vol. 4, No. 2, 1985. t 8.Utkin, V. , *Variable Structure Systems: Present & Future," t t 2 T Automation and Remote Control, February 1984. 1 9.Slotine, J-J., "Tracking Control of Nonlinear Systems Using Sliding FIGURE 3. CONSTANT ACCELERATION - CONSTANT VELOCITY- Surfaces," PhD Thesis, MIT Dept. of Astronautics and Aeronautics, CONSTANT ACCELERATION Cambridge, MA, May 1983. 1269 T=60 *ARROW DENOTES VEHICLE 60 T-30 HEADING p T-90 V y 40 a v p 0 20 s t I T.0 @@OF t T 0 I's T-120 T T=240 N .20 ------ *\\ (Fn -40 INTEREST T=210 T.180 T-150 -60 -20 0 20 40 60 80 100 120 X-POSITION (Fr) FIGURE 4. CONSTRAINT POLYGON FIGURE 5. SAMPLE HOVER MANEUVER o3 '101 * 2 *00s f . 1 11 I)l r 04, t 0, r , I a -. oos 141 '001 -00is Z 0 o -35 -0, t i me( sec t i me( sec ) FIGURE 6. ERRORS IN TRANSLATION FOR HOVER MANEUVER FIGURE 7. ERRORS IN HEADING FOR HOVER MANEUVER so 7s d 60 f so e 40 e 2S t 20 0 01 -2SI so 100 1 so zoo 100 150 zoo r time(sec) time(sec) FIGURE 8. HEADING AND HEADING COMMAND VS. TIME FIGURE 9. DEPTH AND DEPTH COMMAND VS. TIME FOR A TRANSIT COURSE CHANGE FOR A TRANSIT DIVE 1270 THERMAL MODELLING OF ELECTRO-MECHANICAL CABLES FOR ROV APPLICATIONS M. L. Nuckols* John Kreider+ William Feild+ U.S. Naval Academy +Eastport International Annapolis, MD Upper Marlboro, MD ABSTRACT Figure 1: Cross-section of an electro- mechanical cable with an inner and A method for predicting the steady-state outer sheath, a strength member, and and transient temperature distributions for electrical conductor insulation. electro-mechanical cable reels is shown. A r generalized, finite difference analysis 3 technique is introduced which gerterates r2 two-dimensional steady-state solutions and one-dimensional transient solutions for K cable temperatures. The method allows a variable number of layers. turns per layer, cooling methods, cable resistive heating K r levels, and ambient conditions. The paper 2 includes art example application of the model to an ROV umbilical cable. r4 K 3 NOMENCLATURE K Cable Characteristics Ambient Conditions Kj.: Thermal conductivity of conductor tco Ambient temperature insulation h@ Convective heat transfer coefficient Km: Thermal conductivity of inner on inside of drum reel sheath material. h'@ Convective heat transfer coefficient KZ: Thermal conductivity of strength on outside of drum reel member material ht Convective heat transfer coefficient K.: Thermal conductivity of outer on drum flange surfaces sheath material r' Electrical core radius r Iriside. radl.us of inner sheath r3: Outside radius of inner, sheath TNTRODUCTION rA: Outside radius of strength member r5: Outside radius of cable Recent trends in the ROV industry have Eg: Resistive heating loss from cable been toward deeper and more powerful Interstitial Space Characteristics vehicles. Because of the critical role of C : Fractional contact area between the umbilical cable in these systems, adjacent cables efforts have been made to optimize cable Ct: Fractional contact area at the design and reduce cable diameter. drum flange surface Unfortunately, the design goals of more Kx@,,: Thermal conductivity of power and smaller cable diameter can lead interstitial medium to cables with a high resistive heat loss. X..-,: Mean thickness of interstitial If the vehicle is operated with a gap significant quantity of cable wrapped on the winch drum, overheating and cable Reel Characterlstics damage may occur. Because of a large X@: Thickness of reel flange variety of cable types and operational xiz@: Thickness of reel conditions, rules-of-thumb should be used with caution. Experience with @3 r2 Kr@: Thermal conductivity of drum reel material double-armored cables and low resistive beat loss may not be appropriate for KEVLAR-reinforced cables with high heat loss. 1271 United States Government work not protected by copyright This paper describes a computer model developed to evaluate overheating concerns and to assist in cable design. A Reel Midline generalized, finite difference analysis technique in used which provides two-dimensional solutions for cable reels having a variable number of layers, turns per layer, cooling methods, cable resistive h heating levels, and ambient conditions. f The analysis technique is particularly beneficial in conducting cable sensitivity analyses. which identify the critical design and environmental parameters related to cable core temperature. The approach N, provides greater accuracy and versatility than a simple cloged-form solution, and h. helps the designer to specify acceptable cable designs. EFFECTIVE THERMAL CONDUCTIVITY Nodal For modelling purposes, the cable Poi nts central core is assumed to be surrounded by h0 a homogeneous material having an effective thermal resistance equal to the sum of the Figure 2: Sketch of cable reel showing resistances of each layer. Based on the cable stacking arrangement and core surface area shown in Figure 1, this cooling paths. can be written in cylindrical coordinates Resistance @ Rxn + Rx@ + Rsm + R(.>s where rl@ 172 STEADY-STATE MODEL R.. Insulation resistance In --- KI riL When wound onto the drum reel as shown in Figure 2, heat is conducted between P3. r3 cable layers. Excess heat is ultimately Rx. Inner sheath resist In --- dissipated through convective cooling on K2 r2 the inside of the drum reel, characterized by convective heat transfer coefficient h,; rx r@ at the outside cable layer surface, R.. = Strength member resist= -- In --- characterized by h-; and at the exposed 1<3 P3 drum flange surfaces, characterized by h,. All Convective heat transfer coefficients r, re were calculated assuming natural convection Ros @ Outer sheath resist -- In --- (a worst case). K@ r4 For modelling purposes, nodal point locations are defined at the core of each The sum of the resistances can be cable composite. Due to the presumed equated to an overall resistance based on syrometry of the cable stack about the drum an effective thermal conductivity of the midline. only one-half of the drum is composite to give analyzed. Following the conduction through the ra_ P'a rz cable composite, heat travels to adjacent KICIrv @ In - ___ + --- In ---- cables through direct contact with Pi 1K, r, K2 r2 neighboring walls and by conduction through air in the interstitial spaces. as shown in Ps Figure 3. Heat conduction through cable + In + In -.- contact can be approximated (between nodes K3 r3 K4 P4 m,n and m,n+l) as where t_'_ t-,_.l Kzvv C & Z ---------------- Kzpv is the effective thermal conductivity 6X 2 r, of the cable composite 1272 where C is the fractional contact area ratio between cables qn,J+1 t-.- is the core temperature at node m,n AX,AZ equal the outside cable q 9 q M.M+I X On diameter mo n For heat transfer through the ip.erstitial spaces, we assumed an average space thickness exists for the region not q in contact with the adjacent cables. The n- 1 In n heat transfer through this interstitial space will be written K'mpw(l-C) &Z ------------ AX - 2 r, M M M + 1 where fer between adjacent K'=,, is an effective thermal cond Figure 3: Heat trans of the flow path including cables via direct cable contact and the interstitial space through the interstitial air gaps. With these definitions, an energy balance wag written for each node of an N x M @tystem of nodes (N is the number of TRANSIENT ANALYSIS layers on the reel; M is the number of turns). Appendix A summarizes the nodal A solution for the transient thermal equations resulting from this energy behavior of an electro-mechanical cable balance at various positions on the cable reel was also developed. This solution was reel. A sample application of this set of originally developed as a means to validate equations for a reel having 4 cable layers theateady-state analysis described above. and 8 turns per layer, resulting in a 4 x 4 An initial transient model used homogeneous system of nodes (a symmetrical temperature boundary conditions (the temperature at the distribution about the reel midline is reel boundaries was assumed to be equal to assumed) is shown in Figure 4. A Fortran the surrounding ambient temperature), which program was written to solve this set of made an analytical solution manageable. nodal equations to obtain the steady-state temperature distribution on the winch drum The utility of this simplified transient during prescribed operational conditions. model highlighted the benefits of having a complementary transient analysis technique to be used with the steady-state model. In a later effort, an analytical solution wag first attempted with convective boundary conditions. These conditions resulted in FIGURE 4: MATRIX REPRESENTATION FOR 4 X4 SYSTEM the solution of a nonhomogeneous differential equation with nonhomogeneous boundary conditions. Initial attempts at Z 3 4 5 6 7 8 9 10 11 12 13 14 15 16 solving this problem analytically resulted .! Jt 0 1 . - in failure. 3 C 112 tg,,/K .1 '13 ti;13/K Although not described in detail in this 0 .1 '14 (t914+h0alt.)/K paper, a successful numerical solution was 1 -1 0 0 0 3+./@ 0 0 -t t2l (t921-.)/K developed using finite difference -, ' -, ' * -1 '22 t92 21K 0 -1 4 0 123 h23/K techniques. This solution uses the Crank- 1 0 0 -1 3-Lof-@ 0 0 0 -1 '24 tg24+h0edZt.)/K Nicolson algorithm for approximating the -1 0 0 0 30/K .1 0 0 -1 '31 tg3l+Ht.@/K temperature derivative within the reel [1]. -1 0 0 .1 4 -1 0 0 -1 '32 432/K Good agreement is seen between the maximum -1 0 0 .1 4 .1 0 0 -1 t33 t'33" temperatures predicted by the transient -1 * I I 34V 0 0 0 -1 '34 f9'.h..AZ W /K model and the temperatures predicted by the -1 0 0 0 Z-V -1 0 0 t4l @941*(H"f) WJK .1 0 0 -1 ].Hf/K -1 0 142 642'"tdix above steady-state model at the drLm -1 0 0-1 3.Hf/9 -1 3] (f943 -Hft.J/K midline. 1 2 'n -1 00 P4 r944'(Hf1h0MZL@/K n All - klq n 1,1n 1273 Figure 5:Max Cable Temperature Figure 6.-Max Cable Temperature Resistive Heating Level: 1.16 W/M Resistive. Heating Level: 1.84 WIM 100 140 go- 130-- A A 110- X 100 T 70- T go - E E 80-- M co- M p 50 P 70 60 C 40 50 2 3 4 5 6 7 a 9 1 2 3 4 6 B 7 8 9 CABLE LAYERS CABLE LAYERS AIR COOLING @- WATER COOLING AIR COOLING WATER COOLING KEFF-03W W,t,l-i@, Amblent TOMP-376 0 KEFF-03V W,-M-C, Amble@t TS@P-378 0 SAMPLE ANALYSIS Figure 7: Transient Response Figures 5 and 6 demonstrate the Resistive Heating Level: 1.16 W/M application of this analysis technique to predict maximum steady-state cable 100 - temperatures on a sample winch drum with two resistive heating levels. Note that M the maximum temperature occurs at an A X 80-- intermediate layer on the drum. In these T 70 examples the Gemini ROV cable was analyzed. E As would be expected, cable temperatures M 60 P will increase as more cable layers are 50 wrapped on the winch drum. Cable 40 temperatures exceeding recommended operational temperatures of 82 degrees 300 100 200 300 400 Celsius are seen when more than 6 layers of TlIvIE, HRS cable are on the drum at the lower heating - 2LAYERS -4LAYERS -SLAYERS -SLAYER$ level, and when more than 4 layers of cable are on the drum at the higher heating K(D@ Reel)-0 66' W,M-O.Alrhe-2 38800WHR levol. Using water cooling methods, acceptable operational temperatures are predicted for more than eight layers at both power levels. Additionally-, the The sample analysis described above is transient behavior of cables can be only one way in which the cable thermal simulated to identify worst-case time model can be used. Existing cable designs constraints for operating vehicles at high can be evaluated with this model to power levels. Figures 7 and 8 demonstrate determine the chances for mission success that cable failure can be avoided even at under specific vehicle requirements, or to high resistive heating levels for many jobs help establish the acceptable operational that do not require continuous envelopes for that vehicle. Alternative around-the-clock operations. Figure 6 shows cooling methods for the drum reel can be that maximum temperatures can approach 130 investigated. But perhaps the most value degrees Celsius when 8 layers of the Gemini from this model is seen in helping the cable are wrapped on the winch drum and cable designer to specify acceptable operated at the higher resistive heating thermal properties for new cable level. However, Figure 8 indicates that it composites. In so doing, new cable designs will take in excev,-- of 400 continuous hours can be developed to power advanced ROV's of operation to reach this level. during prolonged missions without Additionally. Figure 8 shows that the overheating or cable damage. Gemini cable should perform succevsfully for up to 75 hours under these severe conditions before the recommended design REFERENCES limit of 82 degrees Celsius is reached. 1. Myers, G.E.. "Analytical Methods in Conduction Heat Transfer'. McGraw-Hill, New York, 1971. 1274 APPENDIX A Figure 8: Transient Response MODAL EQUATIONS FOR STEADY-STATE IDDEL Resistive Heating Level: 1.84 W/M General Equation 140 130 + 4 t... - t.,..l t-1'. ------ m 120 x A 110 x i0o T go Boundary Conditions E M 80 P 70 1) Outside (n = 1) 1 60 hoe&Z ig.,.-hoeAZ t. C 50 -t- t.'w-1 +(3 - ------ t.'V - t-I's ----------------- 40 K K 300 60 100 180 200 250 300 360 400 TIME, HRS 2) Inside Boundary (rk = 1) - 2 LAeERS - 4 LAYERS e LAYERS -- 6 LAYEAS K(Cl@ Peel)-0 687 W'1M-C,A10ha-2 38SOG@VHR -t- , + (3 + H/1) t.', - t.,2 t- I'l --------------- x 3) Reel Flange WK) igu.. + -tu-I.. - tu'.-l + (3 + RIM tu'. - tu,..l ---------------- h.. = ------------------------ K I/bo + (re-rx)/Xz" 4) Reel Midline W-1) -ti,.-i + 3 ti,. ti,..i - t2,. = ------ hi. = ---------------------------------- K I/hi + 5) lode 1,1 # H t.0 -tl,2 + (2 4 H/K) t1,1 - t2,1 = -------------- Wi. = -------------------------------------------------------- K 1/hi f )6/Kst + XjvT/2XxxT + I(ra-rJ-XxwT/21/Kxr, 6) lode V, I H+H' hu, I + (H+Hf) tw -tai-i.x + (2 - ----- ) t1l, I tM, 2 = --------------------- H @ hi. C 6 Z + b 0-046 Z K K 7) lode V,l &+boed Z i9v,u+(H,+hoe4 Mao hf. --------------------------------- -tM-1'w + (2 - ----------- t.'. ----------------------- I/bg 4 Xi/XR + (re-ri)/KRFP K K 8) lode 1.1 1 hoeA Z i9,,w+boeAZ too h't. -------------------------------------------------------- -ti,w-l + (2 - ------ ) tl,m - t2,N = ----------------- I/bi + Xt/XP + X,mT/2XluT + I(rs-r.)-X,xT/21/XzFir K x DEFINITIONS Rf = br. CiA X + b'f. (I-W.6 X re I r2 rz N Ear], = In--- ln --- - --- In --- - --- In --- ri r, X2 r2 Ks r3 X4 P4I iga,n is the resistive beating logs from the cable core, watts/ft [,*T 2(rt,-r,)-XxNT] ----- - -------------- XINT Kzyr X z I Karl? C + X'STY 0-01 / (I - P@/r.) 1275 A LISP ENVIRONMENT FOR REAL-TIME OCEAN SYSTEMS Michael P. Shevenell and Charles Millett Marine Systems Engineering Laboratory University of New Hampshire Durham, New Hampshire 03824 ABSTRACT approaches and algorithms which more closely approximate intelligent behavior, There is a constant demand for increased i.e., have greater symbolic reasoning. capabilities and . intelligence in Throughout this time LISP has been used underwater autonomous vehicles and ocean extensively as the implementation language instrumentation. One way to implement because of its symbolic processing these new capabilities is through symbolic capabilities [3]. Another reason why LISP processing systems which use the LISP is an appropriate choice for knowledge- language and its extensions. A previous based systems and expert systems is that hindrance to placing LISP in real-time it is interpreted and allows fast ocean systems was due to poor performance, prototyping. A unique feature of the large memory requirement, and virtual language is that both functions and data implementations of LISP. are represented in the same way, namely, I's-expressions". This similar This paper describes a unique system which representation enables functions and data integrates a high performance LISP to be handled by the same mechanisms: this environment and a real-time operating enables functions to modify the knowledge system, PSOS. High level tasks can be base by creating new functions. The written in common LISP and Flavors and implementation of expert systems with LISP controlled by the real-time operating is made simpler by the availability of system. This system executes on a single object-oriented programming. VME bus board with a 32 bit CPU and 4 megabytes of memory. Successful field In the past several incompatible dialects tests were conducted on the EAVE of LISP were used. The recent development autonomous vehicle which performed simple of common LISP is aimed at increased vehicle path planning. portability and consistenc_y of applications. Presently, Common LISP is running on myriads of computers from PCs to mainframes, and expert systems build 1. INTRODUCTION tools are available for all these environments. The advantage of Common Over the past ten years the focus at MSEL LISP is further increased by the has been to develop intelligent ocean introduction of Common LOOPS, a compatible systems which effectively deal with the object-oriented programming package. environment, unexpected events, and time constraints. The continuing research As expert systems and symbolic processing program, EAVE (Experimental Autonomous capabilities continue to enter the domain VEhicle), has been the testing ground for of real-time reasoning and control, new autonomous underwater systems. Since the methodologies to deal with time vehicle is completely autonomous, all constraints must be developed. Previous intelligence must be placed on board. work in this area has produced some useful Current directions include a knowledge- techniques, but there are no widely based system for guidance, control and accepted procedures for the design of mission planning. As the need f or more real-time expert systems. The testing and capable ocean systems grows, the development tools available at MSEL importance of symbolic processing to include the two languages, Common LISP and emulate some intelligent behavior becomes Standard LISP, and the object-oriented vital [1,2]. programming package, Flavors, running on a conventional real-time operating system, The goal of AI research over the past 25 PSOS [4]. PSOS is a product of Software years has been to develop programming Components Group. CH2585-8/88/0000-1276 $1 @1988 IEEE The techniques used in this work exploit The LISP environment is a result of a the connection between a real-time multi-year research program at the operating system, PSOS, and a multitasking University of Utah on high performance LISP environment. The event driven portable LISP systems [7]. This package capability of PSOS is needed in a real- was purchased and ported to the Ironics time knowledge-based system. This will Performer 32 VME development system allow knowledge sources to be triggered running Unisoft Unix System V.2. Since asynchronously and executed in a fixed the operating system in EAVE is PSOS, it time domain. This combination enables was necessary to port the LISP environment LISP tasks to use the same event to PSOS. This was accomplished by mechanism, message passing and signaling designing a PSOS to Unix interface for the as C tasks. In addition, the priority LISP kernel; this allows the LISP system feature can be used to schedule tasks or to make PSOS system calls for 1/0 and knowledge sources for a given context or control. One portion of the interface is emergency condition. spawned by the root task on system initialization and remaps all Unix systems The ability for applications written in C calls to PSOS. The other function of the and LISP to communicate through PSOS interface provides parameter passing interprocess communication primitives has compatibility with PSOS for each system two important functions. Most call. This environment, plus interface importantly, this provides a foundation on and PSOS, occupied about 600 Kbytes with which to implement the symbolic to numeric additional memory required for LISP heap interface. The second purpose enhances space. system perf ormance. Since symbolic and numeric tasks.can exchange information, 3. RESULTS device interfaces and numeric processing may be done more efficiently in C with In the fall of 1987 successful field tests results being passed to tasks in LISP. were conducted on EAVE [8]. A mission was programmed to detect submerged barrels 2. SYSTEM DESCRIPTION with acoustic sensors and plan a new path. The sensor processing functions of the low The system architecture of the MSEL EAVE level vehicle detected a barrel and sent a vehicle is a hierarchical control system message to the planner module through the shown in Figure 1, this is used as a 1/0 controller. The goal of the planner framework for knowledge-based systems was to interrupt the currently executing development [5]. It is implemented with path and generate a new path near the four 32 bit microprocessors (Motorola barrel; in addition, a video camera and 68020) operating on the industry standard lights were activated by the vehicle to VME bus (Ironics IV-3201 and IV-3204). photograph the barrel. once the exception These processors communicate with 68000 path was completed, the normal survey based sensor and effector manager continued. In this test the planner used processing systems through an intelligent had a simple algorithmic design, yet serial 1/0 controller. The f our upper written in LISP. The importance of this level modules communicate over the VME bus demonstration is the fact that a LISP through dual port memory which uses environment was placed in an autonomous mailbox interrupts to enhance message underwater vehicle for the first time. throughput. 4. CONCLUSION The Supervisor module shown in Figure 1 is a rule-based planner which will accept As expert systems and symbolic processing real-time inputs to plan a new path. capabilities continue to enter the domain since this is implemented as a rule-based of real-time reasoning and control, new system, symbolic processing capabilities methodologies to deal with time are necessary at that level [6]. A constraints must be developed. The unique comparison -of the software architectures symbolic development environment at MSEL of the symbolic and numeric systems is is a tool for development and testing of shown in Figure 2. The Supervisor module concepts as well as a run time system for which utilizes symbolic processing is working underwater autonomous vehicle. shown on the right of Figure 2. It is This system will be used to implement a important to note the similarity between new symbolic processing operating system the symbolic and numeric architectures in which will be the foundation for current Figure .2; these similarities make the and future development in intelligent time communication and control between C and constrained systems. The system described LISP tasks possible. The only difference in this brief paper utilizes the framework between the two architectures is: of a conventional real-time operating LISP application -- planner system to provide a novel solution to. this LISP environment important problem. LISP to PSOS interface 1277 It is recognized that future efforts in the design of real-time symbolic systems requires work in the following areas: �Intelligent Scheduling Algorithms �Rule-Base Design Applicable to Real- Time Applications �Understanding Impact of Garbage Collection on Performance �Computer Architecture to Real-Time Applications �Low Latency Communication Protocols 5. REFERENCES 1. Russel, G.M. and Lane, D.M., "A Knowledge-Based System Framework for Environment Perception in a Subsea Robotics Context", IEEE Journal of Oceanic Engineering, July 1986, pp. 401-412. 2. Kaisler' Stephen, "Expert Systems; An overview", IEEE Journal of Oceanic Engineering, Vol. OE-11, No. 4, Oct. 1986, pp. 442-448. 3. Kaisler, Stephen, "LISP: An Al Programming Language", IEEE Journal of Oceanic Engineering", Vol. OE-11, No. 4, Oct. 1986, pp. 468-473. 4. Shevenell, Michael P., "Hardware and Software Architectures for Realizing a Knowledge-Based System on EAVE", Fifth International Symposium on Unmanned Untethered Submersible Technology, June 1987, University of New Hampshire, pp. 220-237. 5. Blidberg, D. Richard., "Time-Ordered Architecture for Knowledge-Based Guidance of an Unmanned Untethered Submersible", oceans 84 Conference Record, Sponsored by MTS/IEEE, Sept. 1984, pp. 571-575. 6. Chappell, Steven G., "A Blackboard- Based System for Context Sensitive Mission Planning in an Autonomous Vehicle", Fifth International Symposium on Unmanned Untethered Submersible Technology, June 1987, University of New Hampshire, pp. 467-476. 7. Griss, Martin L., Benson, Eric and Maquire, Gerald 0., 11PSL: A Portable LISP System", 1982 ACM Symposium on LISP and Functional Programming, Pittsburg, PA. 8. "Intelligent Control for Multiple Autonomous Undersea Vehicles", Marine Systems Engineering Laboratory, MSEL Report #88-03, University of New Hampshire, Jan. 1988. 1278 SITUATION SUPERVISOR ASSESSMENT I DATA GUIDANCE ASSESSMENT SENSOR EFFECTOR MANAGER MANAGER ENVIRONMENT FIGURE 1. CONTROL ARCHITECTURE APP N C APP N C APP 1 UNP Planner IMP Environment APP 1 IMP to PSOS Inter Global Com- Global Commm PSOS Real TLme OS Psos Real Time Os IMON Monitior IMON Monitior NUMERIC SYMBOUC FIGURE 2. SOFTWARE ARCHITECTURE 1279 A GUIDELINE SYSTEM FOR THE NAVY'S SUBMARINE RESCUE SHIP (ASR) CLASS S. Brian Cable Naval Civil Engineering Laboratory Ocean Structures Division, Code L44 Part Hueneme, Ca. 93043 ABSTRACT This report documents the development and centerwell to the operating depth. During testing of a Guideline System (GLS) for the the mission, surface conditions (wind and Navy's two Auxiliary Submarine Rescue (ASR) current) can move the ASR away from its ships - the USS Pigeon (ASR-21) and the USS original position and/or sub-surface Ortolan (ASR-22). The ASR Guideline System current can cause the capsule to move away was developed as a result of positioning from the target. In many situations, this related problems during operation of the results in movement of the capsule (and ASR deep diving Personnel Transfer Capsule divers) away from its original position. (PTC). The GLS provides guidance for the If the capsule is moved more than 90 to 100 PTC during descent and ascent and provides feet, the target will no longer be within the capability to maintain relative the working radius of the PTC. If this position of the capsule during saturation occurs, the PTC and divers must be returned diving operations at or near the seafloor. to the ASR and the ship must be Development of the ASR Guideline System was repositioned (warped) in its mooring before initiated in 1985 by the Naval Civil continuing diving operations. Engineering Laboratory (NCEL) under the sponsorship of the Naval Sea Systems In many cases, deployment of the GLS will Command (NAVSEA) and is scheduled for allow uninterrupted operation of the PTC. installation on ASR-21 and ASR-22 during regardless of surface and sub-sea effects their routine overhaul periods. on the ship and/or the PTC. 1.0 INTRODUCTION. 3.0 GUIDELINE SYSTEM DESCRIPTION The USS Pigeon and the USS Ortalan provide Two GLS Guidewires restrain the PTC in the saturation diving capabilities for the Navy 'horizontal direction during deployment, with their onboard Mark 11 Deep Diving operation and recovery as shown in Figure Systems (DDS). Both the USS Pigeon and its 1. A pair of deck mounted air winches are sister ship the USS Ortolan utilize DDS used to deploy and recover a Guideweight hardware of the same design. and provide motion compensation to de- couple vessel motion when the 6uideweight The Personnel Transfer Capsule is the is on the seafloor. manned transport capsule used for saturation diving missions. The PTC is suspended from the surface ship by a During motion compensation, GLS winches Strength, Power and Communications Cable provide constant tension to the Guidewires (SPCC) and is also connected via an Air to allow the system to respond to ship Supply Hose Bundle (refer to Figure 1). heave motion. There is essentially no The PTC is essentially a 7-ft diameter limit to the amount of ship heave the sphere capable of transporting divers to a winches can compensate. However, the depth of 850-feet. Once at depth, divers system is limited to the same operating have an operating radius of approximately conditions as the PTC and associated ASR 100-feet from the PTC due to their DbS hardware: Sea State 3 and 2 knot umbilical link to the capsule. current. The GLS interfaces with the ASR's existing 2.0 SUMMARY OF PTC OPERATION. complement of DDS hardware and provides an extended capability far the ship's Prior to deployment of the PTC, the ASR saturation diving system. However, the GLS deploys a 4-point mooring and positions is an optional capability. i.e., The PTE herself over a target on the seafloor. The may be deployed with or without employing PTC is deployed through the ship's the GLS. 1280 United States Government work not protected by Copyright OVERBOARD SHEAVE (P&S) SPC CABLE AIR WINCH (P&S) ASR MAIN DECK FWD HOSERNOLE PTC (ROTATED 90 DEG FOR CLEARITY) GUIDEARM GUIDEWIRE LIFT PLATFORM GUIDEWEIGHT a 4M % FIM I PTC GUIDELINE SYSTEM GENERAL ARRANGEMENT 4.0 GUIDELINE SYSTEM COMPONENT CONTROL PANEL ASSEMBLY. Both GLS winches DESCRIPTION AND SHIP INTERFACE are operated from a single control station located at the starboard forward corner of The major components of the ASR Guideline the ASR centerwell. A series of gauges and System are as follows: Winch Assembly; pilot operated valves at the control panel Guidewire; Control Panel Assembly; Fairlead are used to monitor and control air Assembly; Guideweight; Buideweight Lack pressure to each winch. Assembly; and Guidearm Assembly. GUIDEWEIGHT. The Guideweight "anchors" the WINCH ASSEMBLY. The GLS uses two Ingersol- two Guidewires at the seafloor to provide Rand Model K5UL air winches to raise and horizontal stability to the system. The lower the Guideweight and provide motion weight is filled with approximately 6vOOO compensation. Air winches were selected for pounds of lead ballast. GLS due to their inherent low maintenance, high reliability, and constant tension GUIDEARM ASSEMBLY. Two Guidearms are capability. mounted an each PTC. During operation, the Guidearms are extended and locked in place WINCH BOOSTER RELAY VALVE. A Pilot by surface divers. A capturing device at operated high volume booster relay with the end of each arm is opened and closed back-flow capability is mounted an each around the Guidewire. winch inlet port. Pilot air pressure to the booster relay is controlled by the Constant Tension Pressure Regulator mounted 5.0 GUIDELINE SYSTEM OPERATION on the GLS Control Panel. GLS DEPLOYMENT. The first step to deploy GUIDEWIRE. Each winch has 1000-ft of 0.5- GLS is to release/unlock the Guideweight inch diameter torque balanced wire rope on from the Lift-Platform and lower it to the its drum. A 3x46 wire rope construction seafloor. The descent a+ the weight is was selected for the Guidewires because of accomplished in the CONSTANT TENSION made. its flexibility and excellent torque-free Constant tension insures a Smooth and level (non-rotating) performance under load. (horizontal) deployment of the weight by C K TMWELL % 1281 maintaining a constant and equal tension on winch air cylinder/piston interface. each Guidewire. Guidewire tension is a During NORMAL operation, the GLS winch function of angle "theta" and can be functions as a conventional air winch. Line described as follows (refer to Figure 2): tension and speed are proportional to air pressure and flow rate respectively. The system is in equilibrium when, During CONSTANT TENSION operation, the same Tldl = T2d2 (1) relationships between air pressure/line- tension and flow rate/line-speed described Where, above apply. However, there are additional T = line tension relationships that take affect when torque di = r*cos(theta) h*sin(theta) (2) is externally applied to the winch drum. d2 = r*cos(theta) h*sin(theta) (3) In this case, the external torque is due to heave motion of the ship via the 6uidewire. Substituting (2) and (3) into equation (1): When torque is externally applied to the Tllr*cos(theta) + h*sin(theta)] drum, the air winch is converted into an T2Er*cos(theta) h*sin(theta)]. air compressor. In general terms, when the externally applied torque exceeds the torque imposed by the system-supplied air pressure, the winch operates in reverse TI T2 (out-haul) and compresses the cylinder air to force the air back through the inlet WMEM part and out through the GLS winch booster I-A relay exhaust port. The result is Q h 1. constant" line tension. This of course is the ideal case. In actuality, line tension dl is never constant. It fluctuates between maximum and minimum values due to friction and inertial effects as the winch C hauls-in and pays-out during operation. r However, as long as the maximum and minimum values are within acceptable limits, motion compensation performance is considered adequate. PTC DEPLOYMENT. Once the Bijideweight is on h r the seafloor, the Lift-Plat-Form and PTC are deployed. Surface divers extend the Guidearms and capture the Guidewires. The PTC is then lowered to.its operating depth 13JIMEaff W while being restrained between the two 6uidewires. FISK 2 - BUIDENEW AK61 -VS-- TMION GLS RECOVERY. The Guideweight is recovered after the PTC and Li+t-Platform have been T is proportional to h*sin(theta) or recovered and secured on deck. The GLS more simply sin(theta). As angle "theta" recovery procedure is accomplished with the increa ses, the distance (dl) decreases, and Lift-Platform secured in tile ASR to maintain equilibrium, tension (TI) centerwell. The weight is raised in the increases. This increase in line tension CONST-ANT TENSION mode to insure a level causes winch #1 to pay-out Guidewire #1 ascent. As the weight makes contact with until Tj is equal to T2 and the Guideweight the bottom of the Lift-Platform, four is level. This relationship applies during lock-ing- devices an the Lift-Platform are ascent and descent of the Guideweight. forced open. The recovery operation is essentially a "hands-off" operation at the GLS CONSTANT TENSION/MOTION COMPENSATION Lift-Platform/Buideweight interface. OPERATION. The two GLS winches provide motion compensation to de-couple the motion 6.0 GUIDELINE SYSTEM TEST PROGRAM of the ship from the stationary Guideweight on the seafloor. This is due to the A series of tests were conducted on land constant line tension feature of the GLS and at sea to investigate the performance winches. of an air winch as a constant tension device and to evaluate the feasibility of Motion compensation or constant tension is integrating a new system into the ASR's a result of air pressure changes at the present inventory of.deck hardware. In addition to those tests listed here, @dl 1282 0 M CE - 4J C Ln in -C a) 4- a) 4- 0L(D '0 -Y (D .6j4 Ln tn D"a E +Ul+ C M +>@ 4-J c c 4J -j 0fd 0CU In aj u u 02aj -CCLn C- M cc ri 0H0 C G) 0UC rUL ra -1 4L olc a) a, -Cc al -WC 4.J ra .4 Cr, OJ W 3 Hra L C -i L CM L0 Ln U-H I U ro c41 L "13-W C L 4-1 41 .61 0; kn ai 41 41 m >Ln3 aj c 4-1 Xra aj0 aj Ln to n al rd 0 o E-HL c Gj (U Ul 4-JM -W H Ln lp > -C c 3). c -C 13 -W a) cpC E -,i Lu0 0Qj 4J L 4-d tnC c'0 1 L3 H q4-t 4-d 0Ul aj W00 u4j c C_ C 4@ Ln W@4 CL I> rd Gi > u L 41 Qj Qj CL 001 13 LZ-1 ID ajC mm c a) (a m-C C 4-J 4-E 4-JC ,q fa :3 aj CLN -4 -0 1 C u-CZ L -W '-f3 "4 E L LC 0 +-H - IL -W 't- E4 .0C cCL 4j 01 4.J -iW-W -61 -4 z0 30 0>@ al =ro u cW .0 "a w 0 a0 1M 00 LM C Oj @H 13C uIa LOH Ln L0 > 'DaC C - -i m a) 0 E U 1 4jc H ajC -W C-H3 aj ai 4-1 LU > = Crd - to toz41 u Lq cC Le u to Qj Ul o-w0 4.J0 -W Aj LnC +-0 3 L4J u- '0a C-W c -LLn -H @H u L Qj (n L U -j c-H ra 41 M 4- aj0-0 +-q (U w a) a) ra caj W > -@q Ll OjWQj M4J a) 0 - Li ICC cLO -QJT u ci ci0 C0Ul (U :1 -W Cl Z>L CUC MM C iTu 3 ra L-C- "0 Ln C2 1- 0 Ul cl ra 0 muc 41 LC 4-J CL 61 ED 4-100 ,j ric 3 0V fa , H - L E to F: "s 41 JA 'D- c0 c ulM 13 U3 u4- 4-M Z 41 EHul x oj Ix 3 CnL -W COJ c:M .-1 0 Wul-L 0 -4 ED uj 0-W u-W C.,4 LCIt +Q) cl Ld c 0 0 0*0 1 W-WL,4 Lrt CT al c L -j 4-L0 .1 IUC W rj -u L al Gi3 aLn -4 6+- r- NGi MIt 4j -HWLC W 4-J =1 U- 04j @4H ,4 CP L) wul @H -riu C Ld QJ - H C 41 3 L C- .,q cmc c(D Lnur. ro 3Ul -C :2 --1 Ul -CC M a -0 - H Cra CL .@4 z 3"a+ M c -W Qj U, W- L c :3 Ln C aj IP L rO L C A u a)w4j00 14j0 w CE LE @u C 4-JC+ to LCL Ua 0) 0 W CL N C fa rjMCH 'D Lnc u a: > >' ft 'H -W ]t E wnw E :1 -WW M,-, I -H 4 3 4-1 1L I Ln c 70 -HT- .6-1 c0 MW x 100 L LLc -WC w -W 0 0 c m n.61 H0 CA c Za;0u .@j 041 -Wm 0 CLL - C -C c0 0 - 0 m -1 4.C .. -'0C all uCr-i LOL4J Ll rgWC0 9) _j -4-1 al U C4 LL 0% a) c 4- > -H fa " -W 4-j -i L "4 > 11) ra zz lucc uIW -WC Ul 14C c:I X , Z U L -W0 >al in ip in i c w Gi 4-1 ra H- 1W 41 uu .14 11 `4c 0cL al 0) m MGj .0 CZ NM w ro0ol c cu -Hc aC x ci v m 4a z M 10 IU -C bi 41c -w rv 3,q ID -4 X:C30 c MEra 0all A CL LnH H CL -J X(U tV 4J W 41 LO m 4-10 m c0c ,IH Lqm 41=Iu m-Ww c @> 4J iM L+ Li4C. 3 0al 41 0 3 c 0i to-a M3 -1L 0 mra rjC aj 0CL 00 00 0: ra" .,q0 CL Ql (U Ln ciu Q@ Ul Li 41 vi0 jom E -ci cc -C La 3to C LaE 'D 4.J - V M w aj CL rAj 3 41 Ln a - CCCL a) CLU aj ID 0 z 0 0Eft Qj H > U 3 4jE ftU at C-j 0 to 141VI@j a - - -j -4 aju .4-j +Gi C) > a@j L .,4 '+-0L IP L n Ln fa 4 z 0 m 0 C LMc ol@ LO L Eul cl 0 0) CT C It 'a 11 N C) Ln " U C4J '-I 4J u ajL *V L Z Lnu cra a) -40 -4 (U -W 'JI WQ; tTT LWL C aj r1i (0> Ln c rac IV I waj Cr1l3 Qj E c:1 Qj 0 2@N0H C0C u 0 1 z L4u inZL z tuw lp -WF -4 Ai r N m 41c L 0Lm 4-1 3 c ci C7m C @+ 0 5Cr@C :1 rd a; cL ajW H41 C edj4J L CL .c z0 -4 CPI al 4W @q a) 0)cci -W Ln -fu aic -4 Nc,4 LnL Gi> uaj aj n M 0 Ln u+ UlCL -0 jc - 3CLLrl 01 -0 -WLCj @q -4 ,4C Ln= L0 m0L a cr 4-JE c0 Z cn > 4-1 CL CVJ -,AW @H al LMal (7 -w -w ,jL r- Ln aj X LL4-j CjU Cr 4U E +Ln CLW E 0) 00 1C ci w c 0 -H (U ID -C 41 CL nZ tCL C WCC:3-( M,qZ + C) 0J-i LU a) QJ C: 4- ri Z 0 'a >L Z:W H- C pq <C 4 E 3CL 4-1 -WZ0 to 41 = -WMra .0W,-, -W3 Z13 4W0 M Z3 a. rj afoL M CiL .6J U 161+ 4J -W0 CU Ln 4-1 QjuCc LnL -W CL cGiaQj aj "a ajw -C aj -w -i aj in 0) 3Q) -C0 L4r -i lp,Cc-Wu m0 ro 1-1 a)c0L CCZ IP U C ip =c cZ 0L u0 oj IN E!) aj 4j -4c -C >41 -,4W 4j ro tlj C 4d aj x41 0 .,4 c0 3D, 4-1L 4 0 .,4 -W CL aj m .,4 3 -P4M =L 3 Ul a0E L I Ln 3ol uj 4-J 4j 4- 13 0) Q) (u 'a -EM U (r 4j cC a Xin 0 Ln c Ln wc wu 0 z a In Ln Ln0v a :j al >ai 490 \ \ CU 'D m L -@4CL0 -4C 4J CLwaj Lnma .-i cra C 0)0 OD uc -.4 CajUC'.4 rd ro ra m:3L CLc + ai c 0Ln -i -C 4J C- 2 c.10 u + 4-1ucC L > Lna ajj 0 - Ln 41 al 31 M 4J ir a C aj C.,4 U . 4.J rd+LM n C 41 CL N L EL C 0 LU u Im ra3c E- cE m(U0> 0 cl c r= L Qj CL :3 a) 0) 1 4- 4- (U *E al fe 0-H ELL N-0 W a) 1 0 (D 6-W> LC 00 U. m C 0LL 4- 4Uu -H Ln Ln:ZC 0 U L x + L 4L0 @q rq w aj 0 w cCt. 41 0@q 0 Ln --q IP a)u0 >.10 L :1 a) L c0- > LU LO ai 4- M - >a cal :1oc.,4 to0co 01 LA 0 a) -.4 CL u(U CL CC w inLc> Ln> 0)c -41 .-1 -Ln ulI CL ra Efl 0 E(I + Ad ccr Wc Ln 0a)-04-J -W 4-J '@4 ra3UlC aj al o (U z0 .11c wc M CL CL -j -4 -4 L >*0caj .6jW13 e LA m -i H- :1 lp 1 0-.4 .,4 to 'D IL .,4 . I u L ZW -14 CU tM L 0)041 m0 c Do ai 13 1 w z3 w :I a$ inm-4 c c'o Ul CLnL@4 fta 0 c 0 LnL LM C4- -0 ix a 4-1 U) -W CL CYLn -I at -4c 0:1 '1 41E Lnw- 114 m 41 w :)> @t 4-J0 A n 4- (L aj Ul 4J 0)M c Lna -4MM= '0Ln 0 Im lp mm 4- z aj 0) @4 CLU Gi -4 ai P4 Ln (UuE CuC C a) 4C (U (U CL C al rd 4- -j -W Ic 4L :13,4 ajcx 0-,1 41 f9 L 13 .14 :1 (3 ;0 al r(D Uj0-1 oE 6L4.1 c.0 .@4H :3c3 (U 61 a) L LL0 E4-J CL 0 UJ (1) 0W z rd0C M :3 (D ELn '0 4-JCMCL M CL ai 0 ino cUl EOjL 4J Cn U (C "a Di6CCL IM 4J @4 LnC ra CLe 4Jq at LL ui Ln -4C -HC fa to z LnZ m -i aj .-H '.4M 0) cW0 Eum 0cc Z m a) ai x41 Ci aC L z 0 al I U? 4-1 4J -Ll 70 -,4LU H- C- 00) ro 41 (U L "o AJ ML cu WL 4f - -H 4.J -I U) to) LA UI -0Cal x ECL wc:1 MUL Ul CL w E-4 roc >di Jo _j m aj 0)c0iz rn0 C LCr= L :1 3 NJ c3 -.4 to> z Ln LO 41>m-w 41 C4-1 oLC x.10-4 (VC 4- c -l =) tu --3 0 W 4-j aj cm C-H -.4 mi Ln >. -HcCL0 0 > )l 0 L4: Q)CQJ -C ai Ul 13u 73 -4 1 41 rd to 41 9UlU4J "q aj u u H fuI >3-C u C a) c C 'a) -W oj CC wcm vi n c a Ln e 1 -.4 41 cLn m .HJ 14 4-J aJ M L00C41 Mt E0E SLW0 Mcc W a) ci C EZ 4J EL MU L 10 -1 m-W0mfa -W ,j a-icai 0 w 0 JU :3" ULn -0 3LD N Ln Gi 3L-W0 L E0W EUOJ -C -4 4J L u u -W Ln waj ai .m w10 4- 'a a) Ln- C) 4-4> P4qCL 4J 0Ul al -.4 CL :1 r@ cic A3N310WA3 't (961 V >c a) a) L0Cr C I LrL XL10 CL4-E 0CCU > * at v UT - U7 aj Cl Ln .1 aj0 rr z ramc0 > 0E mrj 01 -H -H 61.. 0 +0ajcc + c 0c wmn -i -C 0M0 LnU c3-u w + > ta a,uP-4 - aj a# -H fV L>CL ED 4J L3CL TEST RESULTS compressor operating, deployment and motion compensation are unaffected and a GUIDEARM INTERFACE. The prototype Guidearm de-rated or emergency recovery is design was a "bolt-on" configuration. The possible at a reduced line speed. bolt-on Guidearms did not require structural modifications to the PTC and for TWO SS COMPRESSORS ON LINE. Air this reason did not effect certification of consumption during weight recovery was the PTC for unmanned use. There were no below the predicted rate and line speed significant inter-face related problems was well above the minimum required speed encountered at the Guidearm/PTC area. of 35 ft/min. The air consumption rate was approximately 190 SCFM and a OPERATIONAL TESTS. A critical part of the continuous recovery of the weight from success of deploying GLS is the 6tjO-foot depth was possible at an aver-age Guidearm/Guidewire capturing procedur@e. line speed of 60 ft/min. Two surface divers were used to extend and retract the PIC Guidearms as planned. OPERATIONAL LIMITATIONS. As a result of Diver feedback indicated that alignment the At-Sea test the following GLS (rotating) of the PTC a few degrees to operational limitations were defined. capture the Guidewires was not difficult. 6.4 ASR-21 AT-SEA TEST (Sept 87) 1) MAXIMUM HORIZONTAL MOVEMENT OF SURFACE VESSEL. If the ship moves The GLS At-Sea Test was required to fully horizontally on the surface to the extent validate the GLS/ship hardware interface that I-Vie quidewire makes contact with the and test the operational compatibility of ship's hull, wire chafe at the hull will the new system. The test location was occur (refer to Figure 4). The maximum approximately 7.5 miles northwest of Point allowable horizontal distance is a Loma, San Diego, Ca. Water depth at this function of water depth and can be location was approximately 650---feet. The approximated as follows: main objectives of the test were . to Maximum allowable angle of the Guidewire demonstrate the capability of GLS during as it leaves the overboard sheave is a PIC operation and to verify that the f unc tion of ship geometry and is existing ASR SS air system was adequate to support GLS operation. approximately 19 After deployment of the GLS and PTC, the Therefore, tan 190 = Wmax/D USS Pigeon was repositioned (warped) 250- or, Wmax @ 0.34*D feet from its original position in its Where Wmax @_ maximum allowable warp mooring to simulate the effects of surface conditions on the ship with the GLS and PTC and, D water depth. deployed. TEST RESULTS HORIZONTAL. STABILITY OF GUIDEWEIGHT. The U FA" M weight remained horizontally level within a few degrees during all phases of deployment and recovery. Guideweight stability performance was identical to performance during the NCEL Shallow Water Test of May 1986. #A WL WINCH PERFORMANCE. The At-Sea Test was an opportunity to con'firm the predicted air consumption/+low rate for the GLS winches during all phases of GLS operation. The predicted air flow rate and line speed for each winch was 200 SCFM and 50 ft/min. A line speed of 35 ft/min is considered the Ron minimum acceptable speed for the GLS winch during weight recovery. The ASR has 2 air compressors rated at 150psi and 250SCFM each. ONE SS COMPRESSOR ON-LINE. As was expected, GLS operation was limited at best. One compressor was not adequate to HIM 4 M OF AM FM US ORM maintain a minimum of 35 FPM during GLS recovery. However, with a single 1284 2) During conditions where the surface vessel has moved horizontally in, the mooring, the distance between the 9.0 REFERENCES Buidewire and the Guidearm may be too 1)Technical Report R-703, Naval Civil great to allow the divers to connect. disconnect)the tensioned Guidewire at Engineering Laboratory, Technical DYNAMIC the Guidearm. If this condition exists, STRESS RESPONSE OF LIFTING LINES FOR the GLS winch operator must slack each OCEANIC OPERATIONS, by Dr. Francis Liu, Guidewire to allow the diver to connect November 1970. (or disconnect) the Guidewire at the 2) Technical Note N-1099, Naval Civil Guidearm. Engineering Laboratory, TAUT GUIDELINE FOR OCEAN I-DAD HANDLING SYSTEM, EXPERIMENTAL 7.0 SUMMARY RESULTS by, Dr. F. Liu and H. Kusano, June 1970. The development of the ASR Guideline System 3) Technical Note N-1365, Naval Civil was in direct support of fleet operations Engineering Laboratory, GUIDELINE SYSTEMS and was a result of the ship's request for FOR DEEP-SEA DEPLOYMENTS, Dr. Francis Liu, this added capability.The ASR GLS project December 1974. started at the conceptual level at NCEL in mid-FY85.The following year and a-half was spent. desiging,fabricating and testing GLS components.The final developmental test (At-Sea Test) was successfully conducted in September 1987. The first article GLS is scheduled for installation on OSR-21 this year as part of the ship's routine overhaul program. One of the most challenging aspects of this project was to integrate the new system into the strip's existing inventory of deck hardware with minimal impact and complexity. With this in mind, the GLS may have had a different confiquration if it had been included in the initial design of the ASR class. 8.0 ACKNOWLEDGMENTS The author- would like to thank Captain Appleby,Lieutenant j.q. Harper,and the crew of the Uss Pigeon for their cooperation and assistance with the design and testing of the prototype Guideline System. Their efforts made it possible to bridge the gap between the Navy's fleet and R&D assets and make the transition from product development to end product user in the most efficient manner possible. A special word of thanks is extended to Master Diver GMCM Powell, the ASR-21 dive team, and the COMSUBDEVGRU-ONE Unmanned vehicle Detachment (UMV) for their assistance and input during the development and testing of the ASR-21 system. 1285 SCIENTIFIC IMAGING: PROBLEMS AND SOLUTIONS FOR ROVs Pete r J. Auster, Lance L. Stewart, and Helga Sprunk NOAA's National Undersea Research Center (NURC-UCAP) The University of Connecticut at Avery Point Groton, CT 06340 USA ABSTRACT We consider the ROV as a mobile platform. Imaging systems and Imaging systems utilized on ROVs are components for specific scientific primarily for vehicle piloting and needs are designed and added to the positioning (i.e. tool use and vehicle. High resolution data which instrument emplacement). The use of was previously unavailable with ROVs for science missions requires standard sampling methods, due to their records of images for future averaging nature, is now possible due qualitative and quantitative to the fine-scale maneuverability and evaluations. The scientific use of placement capabilities of the system. standard ROV imaging systems is often In this paper, we describe systems difficult due to: (1) the need to which have been designed (or are in the change the field-of-view of the video design stage), and procedures developed camera for piloting, and (2) the often to accomplish specific scientific standard configuration of the still tasks. (film) camera. Precise calibration for areal or volumetric coverage is a VIDEO TRANSECTS necessity and several variance minimizing procedures can be utilized. Video transects for enumerating biota Considering the ROV as a mobile are probably the most easily platform, imaging packages for specific accomplished and standard procedure for science missions can be added which ROV systems. Problems exist, however, take advantage of the small-scale for determining area of coverage (path maneuverability and placement width x distance). capabilities of the system. Area of coverage for single frames (one INTRODUCTION field) -or overlapping areas can be determined using standard still Low-cost remote operated vehicles photography calibration techniques. (ROVs) are. slowly becoming standard The most direct method is to: (1) instrumentation for a variety of place a calibrated frame in the field oceanographic disciplines. Off-the- of view of the camera, (2) set the shelf units provide the ability to camera to a pre-determined angle, and perform reconnaissance of sampling (3) delineate the useable area of sites and verify sampling procedures coverage on the viewing screen. once in-situ. These systems are, however, the area of coverge is known, limited in their ability to perform overlapping areas can be cummulatively more complex imaging tasks for sampling added along a transect by using and observation. The use of RoVs for reference points in the field of view complex science missions requires to create a strip of adjacent fields records of images for future (see Stewart et al. 1985 for an qualitative and quantitative application of this technique). evaluation. Imaging systems utilized on Rovs are primarily for vehicle Camera systems which are on pan-tilt piloting and positioning. These heads can be calibrated for a systems are difficult to use for. continuous series of tilt angles. The scientific purposes due to: (1) the area of coverage for a series of tilt need to change the field-of-view of the angles is determined, then a regression video camera for piloting, and (2) the model is constructed with tilt angle as often standard horizontal configuration the independent variable and area of of the still (film) and video cameras. coverage as the dependent variable. CH2585-8/88/0000- 1286 $1 @1988 IEEE Another procedure for determining path essentially a sophisticated microchip width and linear distance covered capable of electronically manipulating Utilizes a set of parallel lasers an image. The typical CCD chip (Tusting 1986) mounted on the vehicle contains an array of pixel .s which make- (Figure 1). The parallel beams up the picture elements recorded. As (typically spaced at 10 cm) delineate light reaches a pixel, a charge is path width. Linear distance covered is accumulated which is representative of either determined with a navigation the object color and intensity. The system or by flying the ROV out a known charge in each pixel is accumulated length of tether from a downweight. through the period since the last scan or "reading" of the chip. Typically, Parallel lasers also allow estimation this is 1/30th second. of vehicle stability during a transect (combined pitch and height off bottom). CCD chips have a unique property, When the vehicle is flying in proper though, that will allow faster configuration over the bottom, the "shutter" speeds. if one wishes the ratio of the distance between the laser frames recorded at 1/500th second or spots to the screen width will be a set I/1000th second, the charge in each value. When pitch or vehicle height pixel is dumped 1/5001h second or exceeds some predetermined limit, the 1/000th second.respectively, before the ratio will exceed the standard and the chip is "read." The chip effectively subsequent images will be edited from yields these shutter speeds while the the data set. system continues at standard field and frame rates. Since charges are Ideally, a camera dedicated to transect accumulated at rates equivalent to the, imaging would be added to the vehicle. amount of light presentf the faster This configuration would allow shutter speeds require more light to be unimpeded movement of the standard effective. The advantage is well vehicle camera for piloting and a defined edges of objects requiring standard fixed mount for the survey image analysis in frame grabing or camera. motion analysis systems'. USE OF CCD CAMERAS VOLUMETRIC IMAGING Video images are often needed for Studies of particles, zooplankton, and single frame or frame-by-frame larval fish in the mid-water require analysis. Utilizing CCD (or similar) the ability to survey a known volume of imaging technology pan solve a variety water. While. nets and trawls can of problems inherent to specific survey large volumes, they miss small- applications. scale patterns of patchiness and association between organisms. A review of how CCD imaging compares to Photographic techniques have the standard tube pick-up. devices is inherent ability to discern small-scale appropriate. Tube cameras, regardless patches and associations. of typef scan at the NTSC standard of 60 fields per second. A single video Hamner and Carlton (1979) first frame is made up of two fields described a-technique which enumerated interlaced. Essentially, a video zooplankton from calibrated volumes in frame when viewed in real time is a focus in a photograph. Using a 1:1 1/30 second representation of the real extension tube and 80 mm UW Nikkor lens world. When viewed in single frame - on a Nikonos camera, they photographed modef howeverf the image is a single aggregations of reef zooplankton and field (1/60th second), not a complete computed densities of organisms. frame (1/30th second). Working distances of 35 to 48 cm were possible with the system so as not to The imaging problem inherent to this disturb the individuals under technology is image blur or ghosting. observation. This same system could be Often, fast-moving phenomenon are adapted to any 35 mm format system. recorded that move rapidly through the Variation of lens focal length, focus field of view. Because each frame is distance, extension tube distance, and only 1/60th second exposure, the object aperature will yield a wide series of moves rapidly within and between volumes in-focus for enumerating fields, causing indistinct images. The various types of organism. This system undefined edges of subjects in each has been used successfully from a field make motion analysis or single manned submersible and is being adapted frame analysis difficult. for ROV use. CCD technology can alleviate these The same system described for a still problems by "shuttering." The CCD is camera can also be adapted to a video 1287 system. Alternatively, close-up exposed properly. Attenuation away diopter lenses could be added to the from the light source reduces the primary lens. Though not as easily reflectance back to the camera so calibrated or viewed, this system could little in the distance can be be used for real-time verification of discerned. vehicle position within patches of small plankters and to position other When a light source is required, sampling devices. several options are available. Use of a Kodak 1A (red) filter over a white Honjo et al. (1984) described a towed light source can reduce the impact on system for enumerating particulates in crustacean behavior (Lawton 1987). the deep sea. The system consisted of This is particularly important when a 35 mm deep-sea camera, electronic using the collimated light source for flash and stack of fresnel lenses plankton enumeration in low current beneath the flash which produced a environments where attraction is a collimated beam of light. Knowing the factor. Due to the constraints of red depth of the beam and the length light transmission in water, and width photographed by the camera, a observations must be made relatively known volume of water is illuminated in close to the camera and., light source. each frame. The advantages of this Again, the vehicle here would be used system are that there is no need to as a mobile platform to deliver the determine what is in and out of focus, low-light camera to a study site or the and particulates in the water between camera used primarily for search the camera and light column are not tasks. illuminated (reducing masking of objects in the zone of the beam). CONCLUSIONS Youngbluth (1984) and Tusting (1986) describe how this system was adapted The preceding discussion demonstrates for use on manned submersibles. how current imaging technology can be Although requiring more hardware, and used to solve unique sampling problems. hence weight, than the previous A number of custom imaging systems and system described, this also could be variations of present systems are adapted (for smaller volumes) for ROV available for specific scientific use. tasks. These tasks range from relatively simple strip transects to A continuous light source focused locating and enumerating small patches through a fresnel lens stack to produce of mid-water macro zooplankton. a collimated beam can be used with Matching specific scientific objectives video systems. Tusting et al. (1988) to available technology and hardware built a small system for use on NURC requires close collaboration of the ROVs (Figure 2). Initially, this science party with ROV operations system will utilize the standard personnel to ensure successful cameras in our ROV systems. We plan to missions. have a dedicated CCD video system, framework, and independent tether for REFERENCES the entire system so it can be fitted Hamner, W.M., and J.H. Carlton. 1979. to any ROV vehicle. The CCD or other Copepod swarms: Attributes and chip type video system is critical to role in coral reef ecosystems. reduce ghosting of particles moving Limnol. Oceanogr. 24: 1-14. through the beam. Ghosting will be a Honjo, S., K.W. Doherty, Y.C. Agrawal, problem when using motion-analysis and V.L. Asper. 1984. Direct systems or single frame image analysis optical assessment of large systems for enumeration. amorphous aggregates (marine snow) in the deep ocean. Deep Sea Res. LOW-LIGHT IMAGING 31: 67-76. The intensity and spectral composition Lawton, P. 1987. Diel activity and of light can affect the behavior of foraging behavior of juvenile organisms, especially in deeper water American lobsters, Homarus where light is of limited spectral americanus. Can J. Fish. Aquat. composition or absent. Using available Sci. 44: 1195-1205. light, an ICCD or SIT camera can reduce Stewart, L., P. Auster, and A. Shepard. the impact of the vehicle on organism 1985. Remote operated vehicle behavior. Low-light imaging will (RECON - IV) survey of benthic increase the depth of the field-of-view conditions at dredged material due to the even light distribution in disposal sites off New England. front of the camera. A strong light ROV185 Conference proceedings. source near the camera lens causes the Marine Technology Society. aperature to close so the foreground is Washington, D.C. p. 137-145. 1288 Tusting, R.F. 1986. Non-conventional techniques for sampling and collecting marine organisms. PACON 1986. MPM 1/12-18. Tusting, R.F., F.M. Caimi, and L.D. Taylor. 1988. Collimated illumination system user information manual. Harbor Branch oceanographic Institution. Technical Note No. 25. 13 p. A Youngbluth, M.J. 1984. Water column ecology: In-situ observations of marine zooplankton from a manned submersible. Occ. Paper in Biology, Mem. Univ. Newfoundland. 9:45-57. Figure 1. NURC's Minirover MK II with the parallel laser scaling unit attached over the camera dome. Also shown is the skid system/manipulator which elevates the vehicle off the bottom to reduce vertical thruster prop wash. The downweight serves as a sample and tool receptacle. opg 4t__ Figure 2. Mounted above the camera dome is a continuous collimated light source for zooplankton enumeration. The camera is tilted perpendicular to the axis of the light. A rotary current meter (center, below dome) is used to determine the volume of water passing through the field-of-view. 1289 AUV TECHNOLOGY DEVELOPMENT AND DEMONSTRATION PROGRAM William J. Herr Martin Marietta Aero & Naval Systems Baltimore, Maryland 21204 (301) 682-0431 ABSTRACT performance. Additional work is directed toward improved materials and energy systems. This paper presents an overview of Martin To test and demonstrate these technologies, a Marietta Aero & Naval Systems's Mobile Undersea requirement was identified for a testbed system to Systems Test (MUST) Laboratory and the plan for AUV provide validation of the technology at sea under actual technology development and demonstration. The environmental conditions. An evaluation of available projects for technology development are focused on operational systems indicated that no suitable system capabilities critical to attaining desired Autonomous exists. Therefore, development began in 1985, of the Underwater Vehicle (AUV) performance. MUST is Mobile Undersea Systems Test (MUST) Laboratory. being developed to test and demonstrate these MUST, shown in figure 2, consists of a modular technologies in the infinitely variable ocean underwater vehicle capable of serving as a test platform environment. The MUST Lab encompasses a general for a wide range of systems in the open ocean, and the purpose, modular unmanned underwater vehicle components required to operate and maintain the including required support systems, and a land based system from a ship at sea. To optimize the use of MUST, Simulation and Integration Lab. The baseline vehicle is an Integration and Simulation Facility has been 30 feet long, 4.5 feet in diameter and capable of diving constructed in Baltimore. to 2000 feet. MUST will have completed sea trials and be operational by January 1989. MUST and the SYMBOLIC COMPUTING technology development projects are funded by Martin COMPUTER ARCH HIGH-LEVEL OPS ANALYSIS Marietta but the MUST system will be made available to CONTROL SUPERVISORY CONTROL support other industrial, academic and government test DATA FUSON "CONTROL requirements. I SLIDING MODE SIGNAL PROCESSING DATA NON-LINEAR BEAM FORMING AUTOPILOT SENSOR VEHICLE PROCESSING YNAMIC CTL 1. INTRODUCTION LIDAR HIGH FREQUENCY SONAR Since Martin Marietta entered naval systems MODELING development in the early 1980's at Aero & Naval SENSING ACTUATION Systems in Baltimore, a primary focus has been Anti- COMPOSITES Submarine and Undersea Warfare. For this area, 7 DOF ANTHRO MANIPULATOR advanced teleoperated and autonomous underwater REAL TIME CONTROL SYSTEM (RCS) vehicle (AUV) systems can potentially meet many critical Figure 1. AUV Technology Focus at Martin Marietta national defense requirements through the 1990's and into the 21st century. This conviction is based upon an; understanding of Navy requirements, maturity of ocean MUST is being tested in San Diego Bay and will engineering and foundation technologies, and undergo trials at sea in 2000 foot depths in December of readiness of new technologies to provide the advanced 1988. Initial Operational Capability (IOC) is set for early sensory and control performance required for AUVs. 1989, when Martin Marietta research projects are In 1984, an internal program of research and scheduled for installation on MUST. development was implemented for critical AUV technology. That work is directed toward new and 2. MUST SYSTEM OVERVIEW improved sensors and sensor processing (Ref. 1), robust vehicle dynamic control (Ref. 2 and 3), and automated 2.1 REQUIREMENTS high level vehicle control and planning that builds on the first two technologies to attain advanced teleoperated The potential operational impact of AUVs was and autonomous performance as shown in figure 1. analyzed and results indicated that a wide range This plan was chosen because it focuses on the ofmissions, characterized by the generic list which technology critical to attaining autonomous follows, can be performed or augmented by AUVs. CH2585-8/88/0000- 1290 $1 @1988 IEEE Advanced Submarine Target vehicle to monitor the vehicle's position. The support Submarine Tactical Decoy ship will maneuver to follow the vehicle and stay within Submarine Strategic Decoy range of communication using a low bandwidth acoustic Submarine Minefield Negotiation data link. The acoustic link will be used to supervise the Seabed Search and Survey vehicle by monitoring onboard parameters, sending new Reconnaissance commands, modifing the program active on the vehicle Surveillance and activating other stored programs. Mine Delivery An extensive mission analysis was done by r @T -70 TW@T TTWTTTTTF7 -Ft, 77, -i-@777-,ol decomposing each mission into segments and establishing performance boundaries for parameters including speed, range, duration, payload, sensors, and processing. The requirements for a testbed to test and demonstrate the critical technologies for each segment requiring technology development were treated a operational requirements for the testbed. However, MUST is not required to perform the complete evolution of all phases of any one mission and thus become an operational system. As general requirements for a successful testbed it was agreed that MUST shall be: � a complete system that can be installed on a ship of opportunity in a short time and perform repeated dives at sea without returning to port � reliable, easy to operate and maintain, and have minimum development risk tall =1@ L � modular and versatile, with payload capacity and performance required to be useful through the 1990's. 2.2 SYSTEM COMPONENTS ayn A The centerpiece of the MUST System is a modular underwater vehicle. The baseline configuration is 30 feet long, 4.5 feet in diameter, and contains 53 cu ft and an electronics payload power capacity of 7 Kw. The vehicle is supervised during dives and between dives maintained from the Control/Maintenance Center that contains the Control Consoles and related electronics. As the requirement for additional space for research Figure 2. Mobile Undersea System Test (MUST)' payload support electronics grows a separate control Laboratory at sea. van will be added. The Deployment system consisting of vehicle launch and recovery system and Hangar provide for vehicle handling during all phases of operation at sea. An AUV Integration and Simulation Laboratory in Baltimore provides a wide range of PRINTER NTENA services from payload checkout to training and dive data INTEGRATED MA CI ON$OL NECE analysis. NAVIG TION COMPUTER CONTROL "ACKER CON1*1ff 2.3 SYSTEM OPERATION L: W@ EE A OR 10HIFT DOCK31DE MAINTENANCE A During initial sea trials, the baseline vehicle.will be CONTROL CENTER CA.It CENTER operated in a supervised control mode. A block diagram VEHICLE of the control and communication architecture is shown TACCUSTI'CEN .. ........... infigure3. Extensive vehicle software is permanently SE... RANSCEI LT@XL CONTROLLER stored on an optical disc and loaded into random access F S`E@ NE.E. ON VEHICLE ------- ------ DATA .=TH. memory over a SCSI port when the vehicle is powered- (Vos) PROCESSOR up. Before launching, initialization of the vehicle's S'.N ...... ; Z-;C@@U ;SKAKE onboard program for the planned dive is done at 9600 LBL ACOUSTIC LO. A....' rIIIT,R,18UTE,D LOOK LWAR 3TE 2 Baud, over a 1000 foot cable. The cable can be lRocasn.. SIDE SCAN SYSTEM _7_71 fl IT E.-I --------- RECOVERY removed before or after launching by disconnecting a SYSTEM (VR$, ------ -------------- ANDSE MSONS r I. RESEARCH PAYLOAD wet make/break connector. . After the vehicle has _--_ completed an initial dive maneuver, the operator in the Control Center will use an ultra-short baseline acoustic Figure 3. MUST System Control and Communication tracking system with a transponder mounted on the Block Diagram C"T IIA1111 C.NT..L A.;.7 ...... AND CPS A @5111. ACOUSTIC MIC oVVAVE ACOUSTIC -NIC1.- -, O.L.E.T. . M- E ------- ;IGV.OCNA L., ACOUSTIC PT't. I R @.LCoo K .C._ r 1291 As research payloads are installed, such as the 3.2 CAPABILITIES non-linear sliding mode autopilot, LIDAR, etc. and eventually the high-level autonomous controller as Operating depth of the vehicle is 2000 feet. It can shown in figure 3 with dashed lines, the operational be launched and recovered in a sea state 3. Redundant architecture will evolve. systems on the vehicle ensure that it returns to the surface for recovery. Forward speed of the baseline vehicle with the 10 HP main motor is 0 to 8 knots. 3. VEHICLE DESCRIPTION Typically, underwater vehicles are developed for a specific purpose or mission, and optimized to minimize 3.1 CONFIGURATION the size and cost of the system. In contrast, MUST is designed to meet a wide range of requirements and thus The general arrangement of the MUST vehicle is it has an unprecedented payload capacity and flexibility. shown in figure 4. Typical transverse section views are This has been accomplished with an innovative shown for segments with battery modules configured for pressure vessel design. Payload space in the main minimum and maximum separation of the centers of pressure vessel is configured to permit the use of buoyancy and gravity in the verticle axis (ZBG). The standard 19 inch wide rack mount equipment (per EIA- main pressure vessel is the central structure of the RS-310C) to avoid the cost and schedule impact, on vehicle and the fore and aft fairings,that support all each research project, of custom packaging. It was external sensors, motors and actuators, are attached to impossible to determine all potential payload each end of the pressure vessel. Power is derived from requirements for MUST at the outset of the design batteries in the main pressure vessel for the onboard process. However, a suite of payload subsystems was control system, propulsion motors and research generated that represented the worst case requirements payloads. The extensive heat generated by the large to serve as the strawman for sizing the vehicle. The volume of onboard electronics is dissipated to the sea strawman included extensive computing and signal by conductive heat sink mounting of the motor processing equipment, and all relevant sensors and controllers and forced air circulation through the navigation equipment. The results of an analysis of the electronics racks and circumferentially along the inside strawman payload power and duty cycle requirements wall of the aluminum hull. for the generic missions resulted in a range of connected payload power from 3.5 to 7.5 Kw. THRUSTERS (4) UPLOOKING VARIA13LE The total payload, weight and volume of the dry 3 HP 1600 RPM ACOUSTI HALLAST (2) C M C electronics rack space based on the lead acid battery 0 UNICATION MAIN PRESSURE VESSEL MAIN MTR TRAN DUCER 2N4V LOAD RECO ERY MAST RECOVERY system for the various potential configurations of the 10 HP 600 CENTER MECHANISM RPM TRANSPONDER vehicle are listed in Table 1. Note that the 5 section, 30 foot baseline vehicle configuration which has been built MTR GYR has over a ton of payload capability and 53 cubic feet of TLS x" payload volume available in the main pressure vessel. K fl.413111 NTUDE ACTUA ORS AL .NO so NAA BATTERY GUIDANCE DN'0`PPER H DOPPLER 2A.,,,'LT" 3' VOLT SENSORS E.TF..ED Table 1. Vehicle total payload weight and volume of DOWN OOKING SOMAS 20 KWH Zo KWH H)AR EN CO ACOUSTIC RUNSOSRIDO RDMMU ICATION 4C HIGH PAYLOAD dry electronics (* Baseline vehicle TRANSDUCER ELECTRON100S RACK configuration). ADAYS 24V LOAD (MIN ZaTOR 1204 CENT BA ERY BANK 12) 24Y LOAD EMI Vehicle No. of Payload Payload CENTER BARRIER Length Hull Sections Weight Volume VOS MAXZ. BA BANK (feet) (pounds) (cu. ft) AFT HULL CROSSECTION TYPICAL HULL CROSSECnON WITH MIN Z., BATTERY MODULES 35.0 7 3100 77 32.5 6 2760 65 Figure 4. MUST Vehicle Configuration. 30.0* 5 2350 53 27.5 4 1940 41 25.0 3 1530 29 Testbed versatility and adaptability are realized 22.5 2 1120 17 with this modular design that permits: �varying the vehicle length from 22.5 to 35 feet by changing the number of hull sections from 2 to 7 The vehicle was designed to have endurance on �changeout of components within the freeflooded the order of a normal working shift, with full propulsion fairing sections to add research payloads such as a and payload power requirements. The number of towed array attachment interface, fiber optic tether battery modules connected to the propulsion and payout system, or different sensor packages payload power busses can be set in between dives to �reconfiguration of the vehicle by swap out of hull maximize endurance or range depending on planned sections that have been assembled with energy speed and amount of payload electronics power and storage and research payload electronics weight. If the payload electronics do not require the full �changeout of motors and/or propellers for increased volume, an optional 20 batteries can be installed in the speed, thrust and efficiency. center cylinder, enabling the vehicle to operate with 3.5 1292 Kw of payload for 28 hours at slow speeds or for 8 hours constructed with 5 segments, however the system is at the maximum speed of 8 knots. The maximum capable of accommodating 2-7 segments, resulting in a achievable range with that configuration is 100 miles at vehicle length from 22.5 to 35 feet. The forward and aft 4.5 knots. When the pressure vessel is full of payload sections are free flooding fiberglass fairings. The electronics, which uses 7.5 Kw, the maximum forward fairing is hemispherical in shape to provide the endurance is 12 hours with a maximum range of 50 maximum volume for various payloads such as forward miles. See figure 5, for vehicle endurance as. a function looking sensors with a minimum impact on the drag. of speed for these various battery and payload The aft fairing has a smooth transition to a 17 degree configurations. taper for good hydrodynamic performance. 3 D BASELINE BATTERY 3.3.2 POWER 3.5 KW PAYLOAD WIOPTIONAL BATTERY MODULE X The power source on the baseline vehicle is 100 20- car-battery size,.. rechargeable, lead-acid suspended OPTIMUM 3.5 KW PAYLOAD electrolyte batteries, that are low risk, low cost, and that LU ENDURANCE OPTIMUM 0 satisfy the requirement for range and duration. An Z 6.5 KW PAYLOAD < - - - - - - - - - - - - - - - - - - additional battery module can be added for optimized 10- endurance. The sysiern can be upgraded with silver- Q zinc or higher energy density batteries. Z The 12 volt batteries can be connected to a 120 LU 06 VDC buss for propulsion or a 24 VDC buss for 1 2 3 4- 5 6 7 8 instrumentation. The reconfigurability of the battery VELOCITY (KTS) system allows the vehicle to be optimized for mission Figure 5. Vehicle Endurance as a Function of Speed for profiles ranging from high thrust-short duration to low Various Battery and Payload Configurations. thrust-long duration by allocating the batteries between the two power busses, as explained in Section 3.2. The power busses are floating so the power systems are 3.3 SUBSYSTEMS tolerant of a single point ground fault. The propulsion power bus is centralized in the 120 3.3.1 STRUCTURE VDC load center located in the aft hemisphere of the main pressure vessel. The instrumentation power bus is The central structural element of the vehicle is a made by interconnecting the 24 VDC load centers cylindrical pressure vessel with hemispherical endbells. located in each hull section. All the load centers contain An exploded view is shown in figure 6. The pressure remotely operated breakers, and current and voltage FNDSELL sensors which give the vehicle control over the battery and load configuration during a dive. 3.3.3 PROPULSION AND ACTUATORS The vehicle's propulsion system consists of a 10 BAND CLAMP STRUCTURAL PING HP motor that drives a 3 bladed propeller at 0-600 RPM for transiting and four 3 HP motors driving tunnel FOR RD thrusters from 0-1600 RPM for hovering and low-speed FAI HULL SECTION SEC I N vertical and lateral excursions. The tunnel thrusters are arranged in pairs forward and aft of the main pressure TRUCTURALRING vessel, perpendicular to the long axis of the vehicle, one (REMOVED) vertical and one lateral. The motors are 3 phase brushless AC motors electronically commutated by DC power supplies located in the main pressure vessel. Control of vehicle attitude during transit is accomplished Figure 6. Vehicle Main Pressure Vessel with four control fins on the aft fairing, driven Exploded View. independently by four stepper motor and worm gear actuators. vessel is optimized to provide the maximum payload volume, minimum diameter and weight, and has a 3.3.4 VARIABLE BUOYANCY modular structure for ease of payload and battery access. The design is a bolstered shell multi-segment The vehicle is designed to be launched and structure made of 6061-T6 aluminum. The removable operated near neutral buoyancy. Final corrections of the support rings allow the payload to occupy the full inside vehicle weight and center of gravity are made using SPEED shell diameter. Each cylindrical segment is 2.5 feet fixed ballast lead in calm shallow water. Independent long, which accommodates 10 boxes with two batteries variable buoyancy systems (VBS), one located in the arranged end-to-end and standard electronic rack bow fairing and one in the aft fairing, provide overall mount equipment. The baseline configuration is being buoyancy and pitch trim adjustment. 1293 The VBS systems are sized to accommodate 3.3.6 ACOUSTIC COMMUNICATION variations in water density and hull compressibility. Each system is capable of 60 pounds change in Although the vehicle is capable of operating buoyancy in sea water, for a total capacity of 120 autonomously, it has an acoustic half duplex pounds. The systems operate by pumping one cubic communication system that allows the operator in the foot of oil from a hard tank to a soft bladder. This design Control Center onboard the support ship to obtain offers high reliability and low maintenance without status information and send commands to the vehicle requiring plumbing penetrations into the main pressure during the mission. This advanced system, which can vessel. The valves and motor pump are operated by the operate in multipath environments, was developed in vehicle control system, controlled manually from the conjunction with the Naval Ocean Systems Center (Ref. Control Center or by automatic vehicle control in 4). It is called the Adjustable Diversity Acoustic response to the thruster force required to hover level at a Telemetry System (ADATS). constant depth. Each end of the link implements multi path/doppler resistant transmission techniques with several custom 3.3.5 CONTROL circuit boards and an industrial IBM PC controller. The operator onboard the ship can adjust the amount of The vehicle is controlled by two Motorola 68020 frequency and time diversity based on the performance processors which are the heart of the Vehicle Operating of the link, local environmental conditions, and the System (VOS). These processors and all of the memory relative positions of the vehicle and the support ship. As and 1/0 boards are off-the-shelf VME buss circuit boards, the amount of diversity increases, the data rate and are housed in a standard VME chassis in the aft decreases. These features permit robust pressure vessel section. The VOS runs a commercial communication at 200 Baud in the presence of multipath multiprocessor real time operating system that provides and doppler shift or up to 2400 Baud when favorable the VOS with an extensive expansion capability. The conditions exist. Because of its small size and VOS will respond to commands stored on the vehicle multipath performance of the underwater components, before launch, commands sent to the vehicle after this system should adapt easily to a sonobuoy for launch via the acoustic communication system, and aircraft-to-vehicle communication or be installed on commands from onboard research payloads. Onboard other vehicles for vehicle-to-vehicle communication. research payloads will control the vehicle via an interface to the VOS, designed to allow access to all 3.3.7 RECOVERY SUITE system data and input of any commands at the fastest rate of the system, 10 Hz. Supervisory control by the Highly reliable vehicle surfacing is accomplished operator is always possible since the VOS prioritizes by a multiply redundant weight release system. During commands based on their source. Commands received normal operations, the VOS operates an independent over the acoustic data link are handled with the highest electrical release to drop the recovery weight. An priority. electrically isolated low power processor-with dedicated An efficient, modular, event-driven state table sensors and' battery pack called the Vehicle Recovery process system was created to allow for easy software System (VRS), ensures reliable return of the vehicle to maintenance, application software development, and the surface in the event of a hazardous condition. The simulation. This design has the fundamental faulty conditions it monitors include lack of VOS activity, architecture of and is compatible with the Real-time water leaks, high temperatures and excess depth. If a Control System (RCS) being used at the National fault is detected, the VRS causes the the release of the Bureau of Standards and Martin Marietta. recovery weight with an independant actuator. Two The VOS interfaces to most of the sensors and other independent mechanical releases can drop the actuators over a serial bus using a 4 conductor cable to ballast to initiate a recovery: a corrosive link works after run through the vehicle. This minimizes the cabling and 36 hours in sea water and an over-pressure device inter-segment connector requirement. The control actuates before the vehicle reaches crush depth. When sensors include depth, magnetic heading (flux gate), the 250 pound drop weight is released the rudders are forward speed (mechanical flow meter), pitch and pitch put to one side and the vehicle spirals to the surface at rate, roll and roll rate, yaw rate and altitude above sea speeds approaching 8 knots due to positive buoyancy. bed. Provisions have been made for a gyrocompass Once the vehicle is near the surface a 10 foot mast and doppler sonar. with RIF beacon and strobe light, is mechanically raised. The vehicle dynamic control system design was The recovery line is deployed with a high visibility, radar accomplished using frequency domain design reflective buoy attached. Once floating on the surface in techniques with a 6 degree-of-freedom linear model of this fail-safe mode, the vehicle can be located by the MUST and then verified and refined with time domain support ship. techniques by evaluating the vehicle dynamic performance for specific trajectories, using the model 3.3.8 RECORDING AND PROGRAM STORAGE with a full non-linear hydrodynamically accurate representation of the vehicle. The autopilot, operating at A 400 Mbyte write-once-read-many (WORM) 10 Hz, will provide stable vehicle transiting trajectories optical disc drive logs all control data at 10 Hz and and automatic control of attitude during hover and communications as it occurs. At a nominal recording transiting. rate of 790 bytes/sec, 68 MBytes are used per 24 hour dive. The WORM drive can be commanded to play back 1294 for any specified time interval. While the vehicle is bridge enables the ship's captain to steer the ship to underwater, playback can occur over the acoustic link or stay within communication and tracking range of the tether and when on deck over the tether, or the disc can vehicle. be removed. This system will be used for dynamic The Control Console was implemented with a Sun performance evaluation and troubleshooting and workstation in the C language under UNIX, with the functions as a traditional black box recorder. The Sunviews windowing system. The entire operator vehicle program is also written onto the WORM disc at interface has been implemented in fixed windows on the the control console before the disc is installed in the color workstation display. The operator inputs vehicle. A small program in PROM boots the VOS over commands using the mouse, and keyboard. a SCSI port. This reduces the time to install new The primary window for vehicle operation, shown versions of the program into the vehicle which is very in figure 8, has a fixed area for status display, mode helpful during a development project. control switches, messages received from the vehicle, and messages transmitted to the vehicle. The large 4. CONTROUMAINTENANCE CENTER window in the lower left corner displays windows tailored to specific operator functions such as vehicle The Control/Maintenance (C&M) Center, shown on command generation, enhanced status monitoring, or figure 7, is a 40 foot long by 8 x 8 foot van outfitted to dive log editing. Commands available include: provide self-contained operations control when the - Hover: hold or change heading and depth or altitude vehicle is in the water and maintenance and while compensating for ocean currents refurbishment when the vehicle is onboard in the - Waypoint: transit at a given speed and maximum Hangar. The C&M center is connected to the Hangar by pitch angle to a new position defined in latitude and a 20 foot long by 4.6 foot wide companionway separated longitude or in relative or absolute northings and by heavy plastic curtains, providing a buffer for the eastings or polar coordinates in meters, feet, or yards. operations area. This command also has a glide distance parameter defined which lets the vehicle coast or not when it HANGAR HEAT EXCHANGE S(4) reaches the new position . Stop and Shutdown: provides decelerated and "dead A stick" stopping capabilities . Navigation Update: for periodic updates of the vehicle's dead reckoning position algorithms (based EYE wASH on surface tracking system onboard the mother ship.) DECK - Status: requests transmission of parts or all of the FILES UMBIWCAL COMPANIONWAY VA C. -DR-t N2 onboard sensors and status buffers. This can be set PUMP PRIMARY- _M;INT@NA@C1 t:G9_41 up for automatic repeats CTL CONSOLE CONSOLE@-- DC PWR I- Power: turns on or off any of the breakers for the BACKUP I AV PWR SUPPLIE C L T CONSOL '!NS@ LE DIST. payload electronics or battery banks IWORK BENCHIJ, 7tf @ru 71PS,/,4@s@ I Manual Thrust: operates any of the thrusters, "I "`;E@ I actuators or variable buoyancy pumps for testing or FILE CO NTROUMAINTE NANCE VAN BATTERY emergency propulsion (if sensors for control loops are SERVER CHA.RGERS(3) inoperational) Figure 7. C&M Center and Hangar Plan View Services: includes other functions such as acoustic communication data rates, clock synchronization, 4.1 CONTROL setting of default control gains and timer values, flight recorder data playback. The vehicle is controlled from the Primary Control All of the commands can be stacked in the Console, which is a Sun Microsystems Series 3 terminal command queue, which allows command groups to be server with a color graphic display, keyboard and stored and acted on as a single complex command. mouse. A fully-redundant Sun Microsystems Series 3 workstation is located adjacent to the Primary Control 4.2 REFURBISHMENT AND MAINTENANCE Console and configured as a Backup Control Console. Another rack houses the integrated navigation system The C&M Center supports refurbishment of the (INS) that computes ship and vehicle geodetic position vehicle between dives which involves battery recharging using Loran C, TransiVGPS and Microwave receivers to and drop shot, corrosive link, recovery float compressed measure ship's position and an ultrashort baseline gas, and VRS battery replacement. During (USBL) acoustic tracking system with depth telemetry refurbishment, the vehicle 24 VDC instrumentation can transponder on the vehicle and ship's gyro repeater to be powered by a set of power supplies and a Sun measure the vehicle's position relative to the ship. The Series 3 color workstation located in the C&M Center INS sends the vehicle geodetic position in latitude and near the opening to the Hangar used as a test console. longitude to the control console which then transmits this This allows the operator to work close to the vehicle and information to the vehicle for use in way point technicians during pre-dive checkout, fault isolation, navigation. This system defers the requirement for an and installation of new subsystems. This work can be inertial navigator onboard until requirements for done at the same time post-dive data analysis and accurate, fully autonomous navigation are established mission planning are being done on the primary and for the testbed. An INS remote display on the ship's backup control consoles. 1295 EMERGENCY CONTROL MAIN FIXED FORMAT WINDO VEHICLE STATUS- ... ESSAGES FROM VEHICLE FUN; 7--- 31 WORKING WINDOW CONTROL "I .. . . ...... .... -- TRANSMISSION QUE 7C WINDO .. ........ WORKING POWER IDENTIFIER CODE: PWR. IMTHR FWD WORKING THRUSTER COMMANDS IDENTIFIER CODE MAN. USER ---- DEAN ED NONE_ t t TOP MEW DWA- W PT7- 9M all MOE VIEW INSERT IN Q SEND TO Q SIDE MEW SEND TO0 EXIT ,FF EXIT 4" .-T 'POWER CONTROL COMMAND PROPULSION COMMAND GENERATION WINDOW GENERATION WINDOW Figure 8. Control Console Window System, With the Main Fixed Format Window and Two Working Windows. 5. DEPLOYMENT SYSTEM end, that transports the vehicle over the stern of the ship where it pivots into the water over wheels mounted on a The Deployment system is designed to maintain trunnion which permits the ramp to yaw. This three positive control of the vehicle during all phases of degree-of freedom- mechanism (translation, pitch and launch, recovery, refurbishment, and maintenance, with yaw) launches and recovers the vehicle without ever minimum manpower and rigging requirement on deck, picking it but of the water and suspending it from an without using divers or men in small boats and with overhead structure. This approach eliminates any minimum impact on ship installation. The system condition where the large mass of the vehicle (>18000 consists of a launch and recovery ramp and trunnion, lbs With entrained water) is able to act as a pendulum on vehicle dolly train and tracks, powering system and the rocking ship. The three degrees-of-freedom hangar. The system is being installed on a 194 foot provides flexibility to decouple the motion at the transom offshore oil mud boat which will fill it's tanks to attain 4 of the ship from the surface of the ocean, even when the foot freeboard. waves and the ship are not in phase. Our design builds on the concept and experience of the Advanced 5.1 LAUNCH AND RECOVERY RAMP AND TRUNION Undersea Search System (AUSS) Launch and Recovery system developed at the Naval Ocean The central feature of the design is a ramp connected to Systems Center (NOSC) which was used successfully a drive cart by a strong arm with universal joints at each for over 80 dives. 1296 The recovery sequence, illustrated in figure 9, directions. The cables are 1-3/8 inch diameter extra commences when the vehicle surfaces and deploys the improved plow steel selected for the systems peak load recovery buoy and line. The buoy is snagged from the of 41,000 pounds. The onboarding cable runs straight ship and removed, and a line that has been fed down to the drive cart which pulls the ramp onboard while the the ramp through the docking cart is attached to the overboarding cable rounds a sheave located just vehicle recovery line. The ship gets underway at forward of the trunnion and experiences maximum load minimum forward speed, 1-3 knots, and then the vehicle, of 10,000 pounds. which is now in tow behind the ship, is pulled up to and mated with the docking cart. The vehicle and docking 5.4 HANGAR cart are pulled up the ramp by a drive cart linked to the main power drive system. A latch at the top of the ramp, The Hangar is an extensively rebuilt 40 foot on the forward fairing dolly, holds the vehicle in place on offshore container. The base was replaced with a rigid the dolly train, setting in the ramp. At this point the structure with 18 inch I-beams to support the dolly vehicle is out of the sea and safely held. The drive cart tracks. The rigidity of the structure and the non- is next used to pull the ramp and vehicle onboard. This compliant dollies provide a reliable stackup for the concept provides a safe, reliable method of recovering interface between vehicle hull sections to permit the vehicle in rough seas. Launching is accomplished alignment during opening and closing of the pressure by releasing the vehicle from the ramp and letting it slide vessel. The hangar is mounted to the ship using proven into the water and clear the ramp. deck fittings but with sufficient freedom of translation so that deck flexure is not transmitted into the hanger base. STIFF ARM HANGAR DRIVE CART 6. AUV INTEGRATION AND SIMULATION VEHICLE IN TOW RAMP LABORATORY TRUNNION WINCH While MUST is designed to be readily mobilized VEHICLE MATED onto a ship of opportunity, an AUV Simulation and WITH DOCKING CART Integration Laboratory (Dry Lab) has been built in Baltimore to facilitate making the most efficient use of MUST while it is operating a sea. The Dry Lab is intended to allow verification of interfaces and functional VEHICLE PULLED UP RAMP AND LOCKED checkout of payloads without tying up MUST while it is ------- in an operational configuration onboard a ship. This capability minimizes the operational time required to RAMPIVEHICLE complete research tests and demonstrations. This HAULED ON BOARD approach, which is shown in figure 10, is being used to stage the integration and testing of research developments at Martin Marietta in an efficient serial VEHICLE TRANSFERED manner. INTO HANGAR Y RAIN DOLL SONARS ON-LI R LIDAR DA AUTO I T FUSI N Figure 9. Vehicle Recovery Sequence IN 5.2 VEHICLE DOLLY TRAIN AND TRACKS MOCKUP & MOCKUP& MOCKUP & SIMULATE SIMULATE SIMULATE The vehicle rests on a train of dollies, one per vehicle faring and hull section, and slides into and off of the dolly train on DeIrin pads. When the ramp is flat on DEVELOP INSTALL INSTALL INSTALL the deck, the dolly train carrying the vehicle is pulle AND EST AND TEST AND TEST AND TEST d M from the ramp into the hangar along a set of tracks which hold the dolly train captive and that are continuous from ANALYZE AN LYZE (ANDALYZE within the ramp to within the hangar. Once inside the E _.ULTS hangar the vehicle is opened at one of the hull joints at a RESULTS R: ULTS time and the dolly train is separated at that same point by anchoring one part while the other is translated to Figure 10. System Use Plan provide the desired separation for access into the vehicle for refurbishment and maintenance. The Dry Lab shown in the block diagram in figure 11 consists of a physical mockup of the vehicle which 5.3 POWERING SYSTEM supports installation of rackmount electronics inside the pressure vessel and sensor/actuator packages in the The powering system for all phases of operation fairings. Power to the payloads is provided by a variable consists of a 75 HP hydrostatically driven winch located DC supply which permits simulation of the voltage range forward of the hangar. This winch has two steel cables as the vehicle batteries go from the fully charged to fully running off in the aft direction but wrapped in opposite discharged state. The power distribution, VOS and 1297 distributed input/output system are functional duplicates Diego Bay and offshore in varying sea states before of the real system. The Dry Lab VOS is connected to a operation begins with the real vehicle. Sun System 3 terminal server configured exactly as the The vehicle test program has included hardware surface control console over a cable but at a Baud rate subsystem tests such as hydrostatic pressure test of equivalent to that of ADATS. the main pressure vessel and software component and I i interface tests. Before hardware and software CONTROL CONSOLE SIMULATION CONSOLE integration began the complete software system was VEHICLE tested using test software written to permit efficient INS S STATUS N IS repetitive execution of complex scenarios to exercise LATION PLAY ACTUAL I III many of the elements of the system software. The H/W & S/W C-- OCP fidelity of the tests were ensured by testing the system ON SIMULATION LATI COTTROLLER IN E;IFACE against the hydrodynamic model of the vehicle. This test configuration has been expanded upon in the ACOUSTIC DECK implementation of the Dry Lab described in Section 6. A LINK CABLE simple version of the simulation has been embedded in VO SI.MULATION the VOS so that these complex system software tests INTERFACE_ can be repeated when vehicle hardware and software ACTUA.L integration is complete and after IOC for pre and post HN & SIVY dive checkout and recertification after system DISTRIBUTED 110 SYSTEM modifications. Additional early validation of the system software architecture and function is being obtained in a cooperative 'effort with Applied Remote Technology (ART) by installing MUST software and control Figure 11. Dry Lab Block Diagram@ electronics and sensors into ARTs 21 inch diameter vehicle, XP-21, which can be operated from a small ship Simulation of the system's response is done in a in the bay and shallow waters offshore. second Sun System workstation by running the 6 Following shallow water verification of the MUST degree of freedom non-linear hydrodynamic model of vehicle mass properties and functional checkout, the the vehicle (including sensor and actuator models) vehicle will be recovered for the first time by the interfaced to the-Dry Lab VOS by a dedicated RS-232 Deployment System in San Diego Bay and MUST sea serial data link and ship Integrated Navigation System trials will commence. After verification of system function model connected to the Surface,C6ntrol Console by an in near shore depths of 75 to 200 feet, qualification dives RS-232 link configured just as the actual INS interface. in depths of 2000 feet will be made. After this checkout The VOS command outputs go to the model in the period, extensive sets of maneuvers will be performed to simulation workstation which provides sensor feedback tune the autopilot, determine the dynamic capabilities of to the VOS and the surface control console. Operating the vehicle and validate the hydrodynamic model. This conditions such as. bottom profile, water currents and work is scheduled for completion by the end of January acoustic transmission parameters are controlled at the and constitutes IOC of the MUST Laboratory. simulation workstation. Additional important uses for the Dry Lab include independent verification and validation 7.2 AUV TECHNOLOGY TEST PROGRAM -(IVV) of MUST software, operator training, mission planning and post dive data analysis and mission As soon as the IOC milestone is reached, reconsfruction., installation and test of AUV technology developments will commence. The first effort will add a non-linear 7. MUST AND AUV TECHNOLOGY TEST model base (sliding mode) autopilot which will provide PROGRAMS significant performance benefits over the standard linear autopilot with speed scheduled gains (Ref. 3). 7.1 MUSTTESTPROGRAM Following tests of transit and mid-water hover performance, a high frequency short baseline The MUST test- program consists of parallel efforts positioning system (SHARPS) with fraction of an inch on,the Deployment System and Vehicle, followed by accuracy and a range of 100m will be used to measure system integration and sea trials. position with reference to a bottom mounted transponder The Deployment System test sequence began with inorder to provide feedback for hovering control near the stress tests of cnitical load path components and function seabed. This autopilot should be able to hold position of critical mechanisms in the factory. Next the and-attitude even with the vehicle at an angle to a Deployment System is being installed o.n the support current. ship and functional validation of all phases of operation Next the LIDAR (Light Detection And Ranging) will be done except for vehicle refurbishment (opening developed at Martin Marietta will be installed and tested. and closing the Vehicle) by operating with a test vehicle The LIDAR determines the range to an object by which is a low cost replica with the same shape, mass measuring the time for a laser pulse to reflect from the properties and recovery mechanisms as the actual object. The laser is a Nd:Yag which emits 40mJ pulses vehicle. Deployment tests will be performed in San 13 nsec in duration at 20 Hz. The infrared output is 1298 frequency doubled to a green, 532 nm color which is in have an unprecedented capability to test and the water transmission window. The laser can be demonstrate AUV technology at sea. The modular scanned over �45 degrees in pitch and yaw, with 9 vehicle design has the versatility to support a wide mrad resolution at 400 deg/sec. The return signal is range of test requirements. Beginning in 1989, Martin detected by a photomultiplier with active grid control and Marietta research equipment will be integrated digitized at 1.3 Gigahertz. The maximum range for according to established interface control specifications target detection depends primarily on the optical and operation of MUST as a testbed will begin. properties of the water and the type of target. Based on Additional requirements for the use of MUST can be experiments we performed in the Scripps Visibility Lab accommodated because modules for the vehicle can be (Ref. 6) and predictions from our laser propagation built, debugged, and interfaces checked in the AUV Dry model, detections at 70-100m in clear ocean water is Lab at Martin Marietta without using the operational expected. Building on the performance of the near- system. The success of MUST will be measured by its bottom high-accuracy hover capability of the sliding operational reliability during at sea tests and the degree mode autopilot, the Vehicle will be maneuvered to scan to which it meets the requirements for future research targets of interest on the seabed to collect LIDAR projects. The AUV technology development underway performance data. at Martin Marietta when combined with capabilities The LIDAR is a testbed for optical sensor systems being developed, by others such as fault tolerant technology in a similar way that MUST is a testbed for computers, high density energy sources and improved AUV technology. The LIDAR will accept improved lasers inertial navigation will provide the foundation for and support imaging and communications operational AUVs in the 1990's. configurations of the same basic system. Additional projects will add obstacle avoidance and 10. REFERENCES imaging sonars, a sensor data fusion processor and a high-level controller which builds on the advance 1. Chande, A. M'., and K. M. Noon, "Obstacle dynamic control and sensing package to demonstrate Avoidance and Navigational Sensing for an robust autonomous control. Autonomous Underwater Vehicle", Proceed'nas Qi the 1986 SPIE Sy , i Optical and 8. PROGRAM ORGANIZATION Optoelectronic EngineerinaL October 1986. Martin Marietta is developing MUST as a capital 2. Dougherty, F., T. Sherman, and G. Woolweaver, facility, funding the entire project internally. However, "Modeling and Simulation of an AUV.", Proceeding the program organization is set up with all the discipline of the Modeling and Simulation Symposium fo and formality associated with a major government Underwater Systems and Subsystems, C.S. Draper project. We have established the system requirements Laboratories, Cambridge, MA., June 1988. and developed a plan combining inhouse engineering and review with a strong team of contractors. 3. Dougherty, F., T. Sherman, G. Woolweaver, and G. A team lead by Ken Collins at Applied Remote Lovell, "An AUV Flight Control System Using Sliding Technology (ART) in San Diego was the lead designer Mode Control.", Procceeding of the MTS/IEEE of the vehicle and is building the vehicle and VOS and Oceans 88 Conference, November 1988. surface control console software. ART will perform shallow watGr testing in San Diego Bay. Aeronautical 4. Barbera, Anthony J. et. al.; A Language Indel2endant Research Associates of Princeton (ARAP) developed the Superstructure for Implementing Real-Time Control 6 degree-of-freedom hydrodynamic coefficients to model Systems. Proceeding of the International Workshop the vehicle dynamics, evaluated the benefits of various on Highlevel Computer Architectures, 7.28-7.39, fairing shapes and developed the linear autopilot control San Diego, CA., May 1984. equations and actuator requirements. Perry Offshore Inc. (POI) did preliminary design of the vehicle power 5. Mackelberg, J., L. Bodzin, H. McCracken, T. and propulsion and deployment studies. We designed Diamond., Naval Ocean Systems Center Navy Case and fabricated the Launch and Recovery System and Number 71243 patent application. Hangar at Martin Marietta in Baltimore. Martin Marietta will perform MUST at-sea trials offshore San Diego with 6. Martin Marietta, Laser Pulse Propagation Test support from ART and other subcontractors. Summa-ry-, Report Number IRB21DO186-1, July 1986. 9. SUMMARY 7. Herr, W.J. and K. Collins., "MUST - A Large Versatile When the Mobile Undersea System Test (MUST) AUV Technology Testbed System", Proceedings of facility reaches IOC in January 1989, Martin Marietta will Underseas Defense Conference., October 1987. 1299 DIVING IN HAZARDOUS AND POLLUTED WATERS LT John W. Blackwell, NOAA NOAA Diving Program, Rockville, MD Clifford D. Newell NOAA Diving Program, Seattle, WA ABSTRACT NOAA first became concerned with the protection of our divers in polluted Diving units today are concerned with waters when the Northeast Fisheries Center factors other than just blowing bubbles. became active in environmental monitoring Among potential diver hazards, the efforts in the New York Bight dump site consequences of exposure to water-borne areas. NOAA, the U.S. Naval Medical contaminants are of particular concern. Research Institute (NMRI), and the Direct skin contact with certain University of Maryland, cooperated on a pollutants is obviously serious business, pilot study of the effectiveness of yet more insidious dangers may result from routine diving equipment in protecting the exposure to pathogens during normal diver from pathogen `1c microorganisms found respiration, depending on the equipment in estuarine waters . The nature of the worn. Standard SCUBA gear is NOT adequate beast began to emerge. protection. For several years, the NOAA Diving Program has coordinated research It was soon apparent that the diversity of and development to confront diver exposure unsafe conditions to which NOAA divers had to waters containing hazardous substances. exposure potential was not limited to The magnitude of this problem is microbial hazards. Working divers today significant, and exposure pathways are routinely dive with petroleum products, varied. NOAA's evolution of protective toxic chemicals, and radioactive materials diving equipment and operational as "buddies". An informal working group guidelines provide a basis for further evolved among representatives of NOAA and education and development. the Environmental Protection Agency (EPA), U.S. Navy, U.S. Coast Guard, and commercial diving industry. A workshop under the auspices of the Undersea Medical INTRODUCTION Society brought this group togsther to exchange ideas and experiences and an Diving units today, whether scientific, interagency agreement between NOAA and EPA commercial, or municipal, have several secured funding for the development of safety factors under scrutiny when procedures to envre diver safety in planning operations. The management of contaminated waters potential diving accidents and rescue evacuation protocols is a popular topic. Since every diver likes their toys, the The diving environment is a pqrt of every logical first step in solving this problem basic diving class, where hypothermia, involved the modification of diving venomous organisms, and other hazards are equipment. The early methods employed the required reading. But as man-made "shotgun" approach with commercially contaminants pervade our coastal waters, available apparatus. Several suits and another significant concern should be on breathing systems were evaluated for the diving supervisor's mind. Divers effectiveness in providing as complete an working in areas of potential, biological isolation from the environment as or chemical pollution must be protected possible. Minor modifications were made against the medical consequences of to some items to increase protection, but exposure to water-borne hazardous a consensus swiftly evolved towards a materials. The abundance of these system of a drysuit mated with a full-face pathogens may be surprising since mask or helmet. detection techniques are often limited., There are also several pathways of This initial decision was the last simple exposure which require prophylaxis. Since one encountered. The significance of the mid-1970's, the NOAA Diving Program disinfection methods on materials, the has pursued efforts to define equipment system's swimmabi 1 ity/ range of motion, and modifications and operational procedures hyperthermia resulting from total to confront this problem. encapsulation were immediately recognized. 1300 United States Government work not protected by copyright EXPOSURE PATHWAYS Demand Systems As the diving apparatus increases in Band masks and some lightweight helmets, complexity, so does the surface support while giving the rugged appearance of necessary to conductasafe operation. increased protection, are really SCUBA Protection of tenders and topside personnel regulator components mounted in a rubber or is essential and often logistically more fiberglas hat. These systems must be complex. It is possible that the safest modified to create a one-way gas flow and person during management of a toxic spill positive internal pressure to keep water may be the diverl out. The normal exhaust port can be blocked off and the secondary Itchin" As if that is not enough to contemplate, exhaust valve brought into primary use. It routine diving tasks were recognized to may also be possible to create a separate risk exposure to substances of at least exhaust valve location using an alternate equal consequence to well-publicized oil helmet penetration. Various off-the-shelf spills and other obvious situations. products were modified with prototypes to Underwater hull cleaning involves fully develop these ideas. Figure 1 positioning a diver's head inches from illustrates an early design. toxic paints as they are introduced into the water column. Jetting of pipelines may re-suspend contaminants which have been buried in the sediment. Injuries to Driger Polluted Water Diving Hood Mod. I commercial divers working in polluted water environments, and the prevalence of adjustable assoc4ated equipment malfunctions are well exhaust valve known . Newspaper accounts and personal communications have reported cases of ear equalization police rescue divers in metropolitan areas pad who have experienced medical problems after working dives. Why did gastrointestinal problems cause an 80% absenteeism rate Series Exhaust Valve among divers one week after a training Microphone Mod. 11 exercise in harbor waters? Possibly a case of bad taco dinners after the dive, but maybe a lot more. It is important to understand these more insidious pathways to exposure as the proper configuration of Demand Regulator Mod. 11 protective equipment is developed. Although pollutants in water are generally thought to be dispersed or diluted, prudent 'planning would have to include the chance encounter with a pocket of full strength or Figure 1. Initial modification of Drager concentrated contaminant.. Diving Mask BREATHING APPARATUS Requirements An exception is the AGA Divator full-face mask, which has an excellent gas flow Second only to direct innoculation, the configuration and a built-in positive most effective way to infiltrate the body internal pressure. Unfortunately, the AGA with a contaminant is to breath an aerosol. is a mask rather than a helmet, limiting Standard SCUBA is woefully inadequate its use to less than maximum levels of protection against contaminants in the protection. ambient water since the same mechanical conduit is used for both inhalation and A useful modification to an existing exhalation. During a normal respiratory exhaust port can be accomplished by placing cycle using SCUBA, water droplets an additional mushroom valve in line with frequently congregate on the exhaust valve the original (Figure 2). This series of as it opens and closes. The next exhaust valves (SEV) provides greater inspiration atomizes these droplets, which reliability and eliminates 11splashback". contain any ancillary substances as well, The SEV concept has proven worthwhile, and The importance of maintaining a dry is now available on several off-the-shelf breathing apparatus becomes paramount. The helmets. best choice is a full-face mask or helmet which minimize the number of , penetrable The diaphragm separating ambient water from seals and allows for separate inlet and the regulator demand lever is a thin, exhaust channels. flexible membrane and a possible weak link 1301 in the protection of the diver. Various with toxic substances and must also methods have been attempted to completely withstand the rigors of post-dive cover the diaphragm and reference ambient decontamination. Foam neoprene drysuits pressure from inside the helmet, thus fall to meet these needs. Suits with minimizing entry problems. Unfortunately, smooth outer surfaces such as the Viking this disrupts the helmet design enough to and Drager have demonstrated the best cause breathing resistance. performance., The suit MUST mate dry to the helmet used for optimal protection. Dry suits containing air are subject to significant pressure differentials when Series Exhaust Volvo Mod. I 18reakdown) submerged, dependent on diver posture. if a tear occurs in the diving dress, this pressure imbalance contributes to entry of ambient water and any ancillary materials. Diver protection was considerably- upgraded by an offbeat approach: design an inner, IExhaust valve body form-fitting suit for wear under the 2. 0-RIrg (01)c 1 14@ W=1/4) drysuit, seal the two suits together at the 3. Check valve b@dy neck and wrists, and fill the space between 4. 0. Ring (00 = i W: W = Va) S. Check valve body with water, as shown in Figure 3. G. 0. Ring (00 = I W., W = 7. Cap B. Exhaust valve body cap MK-12 Loaren Breach Ring FRONT Figure 2. Series Exhaust Valve (S EV) Jacking H...... Front Joc ISir It WoW Exhaust Valve Free-flow Helm6ts A functionally simpler configuration is wmw In Part inherent to f ree-f low helmets . Inlet and exhaust valves are separate, and the steady gas flow creates the positive internal pressure required to prevent water entry. Free-flow means a surface-supplied Crotch Jacking Strep Outer Dry Suit configuration, which has significant advantages for contaminated water diving: Inner My Suit i Outer Restricting Demand Increased gas supply allows duration to Outw Dry Suit of extreme significance be increased if the diver must undergo surface decontamination . while still,suited up. Attached A Water Exhaust Vol" Diver-to-surface communications - and a direct tether to the surface are, essential.to safety in hazardous areas. Several commercially available free-flow Figure 3. Dual layer, water-filled hats are acceptable, provided they are Suit-Under-Suit (SUS) capable of dry mating to an appropriate exposure suit. Some brands of drysuits may be ordered with modified neck yokes, comprising an integrated system which is This "suit-under-suitit (SUS) design "well-suited" for polluted water diving. eliminates the tendency for suit infiltration, and can actually reverse the EXPOSURE SUITS pressure differential in favor of the diver. Clean surface water is circulated Certain criteria are vital for exposure via umbilical through a penetration in the suit selection. From the outset, only drysuit and out exhaust valves installed variable volume drysuits which permitted near the ankles. This provides an outward reasonable range of motion were evaluated. gradient to minimize contact of toxicants A minimum number of "breakable" seals and with the diver's skin in case of a tear in zippers is desirable, and boots should be the outer suit. The elimination of air attached. The exterior surface must be spaces makes the entire dress neutrally resistant to degradation when in contact buoyant and much easier to swim. 1302 Thermoregulation is now possible, evaluated the efficiency of the SUS, controlled from the surface and with a resulting in a maximum safe operating limit diver by-pass valve. This was a major step of 50 degrees Centigrade (122 degrees in solving the danger of overheating a Fahrenheit). Air-filled suits were totally encapsulated diver, especially comparatively evaluated, yielding a limit working in a warm water environment. This of 40 de 9T ees Centigrade (104 degrees hyperthermia can overcome a diver with Fahrenheit) . Additional tests at the NOAA m i rTi7-m a 1warning, and must be monitored Experimental Diving Unit in 1987 closely by the diving supervisor. incorporated inner cooling suits. Field tests at the NOAA EDU and in Seattle, WA The U.S. Navy MK-12 Surface Supplied Diving successfully integrated diving operations System, configured with a modified Viking with the surface containment precautions SUS drysuit, has been selected as an necessary for a full-ocale hazardous optimal diver protection system for use by materials response effort . The systems NOAA, EPA, and other agencies in were ready for field operations. contaminated waters (Figure 4). OPERATIONAL EVALUATION The opportunity for the first operational evaluation of NOAA polluted water diving systems occurred in . the spring of 1987. BACK Scientists at the Florida State University Communication Whip were planning phosphate mineral formation Air Whip studies in an South Pacific island lake Water Exhaust Valve containing toxic levels of hydrogen sulfide. Diver assistance would be required to position a coring device in the Roar Jocking Straps lake bottom, where the H S concentrations Restricting Lacing approached 85 PPm. NOAA ;greed to provide personnel and equipment to assist the Air to Diver Water Inlet To port scientific effort. A Superlight 17B demand breathing system was chosen since the transport of gas bottles was thought to be a major difficulty at the lake's remote location. A Viking SUS with 'a third outer jocking garment provided exposure Restricting Lacing protection. Since the diver's rising air bubbles might carry hazardous concentrations of H 2S up to the surface personnel confined on araft above, Water Exhaust Val" considerable pre-planning in medical advice and evacuation protocols was prudent. Protective breathing apparatus and monitoring devices-were on s tation for all Attached Soots those topside. Unfortunate delays in the arriv al of integral diving system components and Figure 4. Detail of SUS Design several on-scene equipment failures for USN MK-12 frustrated the attempts to conduct a progression of exploratory dives on schedule. Lots of determined effort finally effected the, culmination of pre- TESTING mission activities. NOAA has sponsored a dynamic testing and A six-inch, six-foot long sediment core wa's evaluation program since this project was successfully extracted from the lake bottom undertaken. Initial suit and breathing during a climactic 110 FSW, 24-minute dive. apparatus studies were conducted in clean Diver hyperthermia was asignificant waters, and actually began as a spin-off factor, primarily caused by the excessive from other efforts. As design and time necessary for dress-in with this implementation progressed, the testing system. A simpler mating configuration of environment was manipulated with trace suit to helmet, such as that of the@ MK-12 chemicals 5 and dyes to evaluate system breech-ring assembly, would ha've reduced integrity . Range of motion studies this complication In hindsight, chilling conducted with NMRI and UCLA showed no of the SUS wate; input would have been- significant decremen@ in range of motion beneficial, although this was not indicated for underwater work Thermal studies by previous test data. 1303 CONCLUSIONS 4. McLellan, S.A., and R.F. Busby. 1982. Evaluation of the Use of Divers and/or NOAA's development of a safe protective Remotely Operated Vehicles in system for diving in contaminated waters Chemically Contaminated Waters. USEPA has been fortunate to yield many Contract No. 68_03-@3113, Task 42-m3. commendable results in both techniques and equipment modifications. The work is by no 5. Nash, J. 1983. Chemical Tank Testing of means complete, and remains a dynamic Modified Commercial Diving Helmets and investigation. Here are a few salient Dress. USEPA Contract No. 68-3-3056. points for any unit desiring to operate safely in hazardous or polluted waters. 6. Bachrach, A.J., G.H. Egstrom, and J.M. Wells. 1985. Biomechanical Assessment Standard diving equipment (especially of MK-12 Dive System NOAA Modification SCUBA) is NOT adequate diver protection for Anti-Contaminant Use in in contaminated water. Contaminated Waters. Eighth Joint Meeting of the Panel on Diving Positive pressure, surface supplied Physiology and Technology. Washington, breathing apparatus is optimal. D.C. and Honolulu, HI. The level of protection needed requires 7. Wells, J.M. 1985. Diving Operations in increasingly complex equipment. Heated/Contaminated Water. Eighth Joint Consequences like unanticipated Meeting of the Panel on Diving hyperthermia may result. Physiology and Technology. Washington, D.C. and Honolulu, HI. Surface personnel may be exposed to at least equal levels of hazards as the 8. Stringer, J. 1985. Surviving Underwater diver. Coordination of topside Deathtraps. National Oceanid and operations is a significant Atmospheric Administration. NOAA undertaking, especially if these folks Magazine, Vol-15 - No. 1, Winter 1985. must wear protgctive clothing and breathing systems 9. Traver, R.P., and J.M. Wells. 1984. "Summary of On-Scene-Coordinator The diseases to which a diver is Protocol for Contaminated Underwater exposed may not be acute. Long term Operations. International Diving incubation periods may exist. Divers Symposium 184. New Orleans, LA. should maintain current immunizations, and should be monitored by supervisors for any unexplained illnesses. Further information is available from: NOAA Diving Program NOAA, N/MO15 6001 Executive Blvd., Rm. 304 Rockville, MD 20852-3806 301-443-8007 REFERENCES 1. Craven, J.P. et al. 1981. Special Issue on Microbial Ha,zards of Diving in Polluted Waters. Marine Technology Society Journal. Volume 15 - No. 2. 2. Wells, J.M., and M. Heeb. 1983. Protection of Divers in Water Containing Hazardous Chemicals, Pathogenic Organisms and Radioactive Material. Undersea Medical Society, Report No. CR 60(CW) Z-1-83. 3. U.S. Environmental Protection Agendy, and National Oceanic and Atmospheric Administration. Interagency Agreement. 1983. Diver Safety in Contaminated Waters. AD-13-F-1-826-0. 1304 THE USE OF NITROGEN-OXYGEN MIXTURES AS DIVERS BREATHING GAS J. Morgan Wells National Oceanic and Atmospheric Administration NOAA Diving Program Rockville, Maryland 20852 ABSTRACT it can be breathed pure or as a component of a nitrogen-oxygen (NITROX) mixture. If both "The Use of Nitrogen-Oxygen Mixtures as the toxic properties of oxygen and its Divers Breathing Gas" discusses the use of decompression-obligation-reducing properties nitrogen-oxygen mixtures as a substitute for are taken into account, an "ideal" gas air by divers, detailing their advantages in mixture for any depth/time combination can terms of reduction of decompression be produced. Such a mixture would offer the obligation and extension of "bottom time". maximum decompression advantage without the risk of oxygen toxicity. Advantages of such The paper concentrates especially upon the a mixture relative to air are: use of such mixtures by the NOAA Diving 1) extension of "no-decompression" time Program and the experience gained thereby. limits, 2) reduction of decompression time The NOAA-developed mixture known as NOAA if no-decompression limits are exceeded, and NITROX I, in use since 1978, has been 3) reduction of residual nitrogen in the gaining increased acceptance in the diving body following a dive. The latter would community. A new mixture, NOAA NITROX II, either increase the time allowable on is still in the process of development. The repetitive dives or reduce the surface NOAA"developed Continuous NITROX Mixer is interval required to make repetitive dives, also discussed. or both. The decompression procedure which must be INTRODUCTION followed when NITROX is used is based on the concept of "equivalent air depth" (EAD) . Air has been used as a breathing gas by This procedure equates the inspired nitrogen divers since the beginning of diving. Its pressure of a NITROX mixture at one depth to principal advantage is that it is readily that of air at another depth, the EAD. This available and inexpensive to compress into procedure has been used for over 20 years cylinders or use directly from compressors with semi-closed and closed-circuit mixed- with'surface-supplied equipment. It is not gas underwater breathing apparatus. Such the "ideal" breathing mixture because of the equipment is both very expensive and decompression liability which it imposes. complicated. since decompression obligation is dependent on inspired nitrogen partial pressure and time, not "depth and time", this obligation NOAA NITROX can be reduced by reducing the nitrogen content of divers breathing gas and In 1978, NOAA introduced diving procedures substituting for the removed nitrogen a gas and decompression tables for a standard which is metabolized away by the body, i.e mixture of 68% nitrogen, 32% oxygen called oxygen. NOAA NITROX I (NNI). It can be used with normal SCUBA or surface-supplied equipment, and is the best general-purpose NITROX NITROGEN-OXYGEN MIXTURES mixture for use in the.30- to 130-foot depth The toxic properties of oxygen at elevated range. The NNI tables (Appendix E of the NOAA Diving Manual) are identical in format pressure limit the depth and time to which and use to the U.S. Navy Standard Air 1305 United States Government work not protected by copyright Decompression Tables, thereby eliminating the necessity of learning to use a new set of tables and providing a means of mixing air and NNI dives. During the last f ew years, there has been an increase in the acceptance and use--and misuse of NITROX by the diving community. Figure I compares the no-decompression limits of air and NNI. The use of NNI approximately doubles the no- Single Dive Time Comparison decompression limits at most depths. One "not-so-obvious" advantage is the extension M Air M NOAA of useful "on-the-bottom" time. At 130 Nitrox I feet, the no-decompression time for air is ..,4o-Decompression Limit (min) 10 min. while the NNI no-decompression time is 20 min. If two minutes are required for descent, air gives the diver 8 minutes of 180 useful time at depth while NNI will allow 18 minutes. Figure 2 shows the repetitive groups for air 12D and NNI for dives to the same depth and time. The repetitive groups, and thus the residual nitrogen times, are significantly lower for NNI. This is of great value in reducing the surface interval between dives o on -a and/or extending the time of repetitive 6o 60 70 8D %0 1@ 110 1Z) 130 dives. The net result of all of the above Depth (ft) is more bottom time per day. Figures 3 and 4 show the advantages of NITROX use in repetitive diving to several Figure 1 depths and with different surface intervals. Economics and time savings are key factors with respect to diving operations. Breathing gas mixtures are normally prepared by mixing oil free, pure gases in appropriate proportions and checking the Repetitive Dive Group Letter mixtures with oxygen analyzers. The After Maximum Air No-Decompression Dive expenses associated with the purchase of -1--- Air Nitrox I gas, equipment, and utilization of qualified personnel can be significant. So can the consequences of shortcuts and/or poor L procedures. For example, traces of oil in K a high pressure oxygen mixing or pumping J system can lead to an explosion. Too much or too little oxygen in the mixture could ii 2 H lead to oxygen toxicity or decompression 0 sickness. The extreme case of accidental use of pure gases (N 2 or 0 2) has proven to be F fatal. E Personnel time and ship time associated with D- diving operations can also be significant. so so 70 80 90 100 110 120 130 The increase in useful "bottom time" which (loo (00) (50) (40) (80) (16) (211) (151 (10) is provided by the use of NITROX mixtures Depth (ft.) (Bottom Time (min,)) often makes it very cost-effective. The reduced nitrogen pressure at depth slightly reduces nitrogen narcosis. During ascent I and decompression stops, it also reduces Figure 2 the probability of decompression sickness. The subjective post-dive feeling of "well- being" (reduced fatigue) reported by some users suggests that sub-clinical DCS is reduced. subtle effects of increased oxygen pressure on nitrogen uptake and elimination remain to be determined. 1306 Another NITROX mixture, NOAA NITROX II (NNII) is currently being evaluated. it contains more oxygen than NNI, thereby reducing the decompression obligation even more, but its use will be limited to depths of 100 feet or less. THE NOAA CONTINUOUS NITROX MIXER Repetitive Dive Time Comparison (with 120 min. surface intervals) The NOAA Diving Program recently developed M A,r EM NOAA M sci-n-i a system for mixing air and oxygen at Nitrox I atmospheric pressure, and compressing it to ISO qo-Deco. Bottom Time Imin.) pressures of over 3,000 psig. oxygen UO - &ddwk- concentrations from 21-40% can be easily and 1W - accurately mixed. The system utilizes a 120 1. F"t Diva to NO-0@. LIFNI Of Air. "non-oil-lubricated" compressor, and the 110 loo 2. T- H- S@f- IM-1. design is such that both large storage QO 3, S@ ova @ hb-Dow. L"It of P@11- Gas Ar or cylinders and SCUBA cylinders can be filled 130- and/or cascaded simultaneously. This 70 - feature eliminates the requirement for a gas 60- 50- booster pump, and significantly reduces the 40- cost and time required to prepare NITROX. 30 - 20 - 11 jj IR Ig 12 Three NOAA Continuous NITROX Mixer units 10 - OL (Fig. 5) are currently in use at NOAA diving 6 Q 1 60 1 70 1 so I QO 1100 1110 1120 11.3Q units, and Hyperbarics International Depth (ft) recently installed a system on Key Largo. For the preparation of large quantities of NITROX (in oxygen-clean systems), mixing of I pure oxygen and oil-free air is probably the Figure 3 most "time effective" method. The use of pure oxygen and nitrogen is an obsolete method of preparation. NOAA has established a color coding and labeling system for NITROX mixtures (Fig. 6), and uses dedicated cylinders for NITROX. Air is never put in Repetitive Dive Time Comparison NITROX cylinders and NITROX is never used in (with 60 min. surface intervals) air cylinders. 00 Air EM NOAA M Both Nitrox I A final analysis for oxygen is conducted by 40-Deco. Bottom Time (rnin.@ the diver prior to using NNI. 90 1. F- dr,as are pla-L saparated by 60 mir, @Faoa imerval. TRAINING So 2. Mrst diva 1-W . d- WIt F .11 .1rig b.1h 9- 70 3 If a-d dve 1. or@ mpoaslbl@ Training of NOAA working-level divers in the do 1-20 .1 -f- r'la'al La`@ @ aao-' use of NITROX in open circuit SCUBA requires 50 4. It fmrth dive impractcal or Imc)ossible, It (a omitted one day; a four-hour lecture period and a 40 four-hour practical session. The lecture 30 portion includes a heavy emphasis on oxygen toxicity. The potential consequences of 2 isuse of NITROX, decompression and 1 repetitive diving procedures, and mixing air I 1,N UP m 60 60 0 SO 9 0 1 0 0 1 1 0 1 1 0 and NITROX repetitive dives are covered. The Depth (ft) practical sessions include safe handling of oxygen-rich gas mixtures, gas sampling, and oxygen analysis. Brief written and practical exams are given in the afternoon. IFigure 4 These divers are "users" of NITROX. Training in the preparation of mixed gases is considerably more extensive, and generally limited to unit diving supervisors or those divers responsible for diving lockers and breathing gas preparation. Due to the extra cost, equipment, complexity, and training required for the use of NITROX, its use in NOAA is currently limited to those diving units to which it can make a 1307 significant contribution to cost effectiveness. As its advantages become better known, we anticipate more requests to establish mixing systems at diving units. OTHER CONSIDERATIONS In addition to its relatively recent use in, open-circuit SCUBA, NITROX has long been NOAA CONTINUOUS NITROX MIXER O.yg.. A..fy used in closed-circuit mixed-gas underwater 1 0 breathing apparatus (UBA) These units Qua Y maintain a constant partial pressure of oxygen rather than a fixed percentage of oxygen as do open circuit units. Akintk. G 9;u 0: When the oxygen pressure is set at a level oxygen 01-froo KP. kdeallon Coanpressor where the percentage of oxygen at a 3Y.t_ particular depth is greater than that in air, the EAD is less than the actual depth, providing a decompression advantage over air. Helium-oxygen (HELIOX) mixtures are normally WTFI X 4 used on deep dives to eliminate nitrogen narcosis. At shallow depths (less than 150 fsw) HELIOX offers a significant decompression advantage over NITROX. However, the cost of helium and the complexity of breathing gas preparation make HELIOX impractical for use in shallow diving with open-circuit UBA. In closed-circuit Figure 5 UBA, where exhaled gas is recirculated, HELIOX becomes very economical, and is actually the breathing gas of choice. Currently, the cost and complexity of closed-circuit UBA limits its use to the military and highly special 'ized diving yellow units. NOAA is currently working with green Band private industry, scientific diving <@_ starts at the 4 Inch curve at the organizations, and other government agencies Oxygen Gureen ----C> ink to maximize the utilization and availability Band of mixed gases. and closed-circuit UBA at reasonable cost. CONCLUSION I Inch Oxygen Qroon lettering X >. <3-- yellow Mother Nature provided the planet Earth with 5 Inches from bottom of tonk a NITROX atmosphere known as air. She never Zo said that air was the best breathing medium for divers. Here, as in many other fields of endeavor, human beings have ' used their knowledge of natural laws to go one step Back View of NITROX Tank beyond what Nature has provided for them. Figure 6 Xa For example, the U.S. Navy Mk 15 and Mk 16 and the Biomarine CCR 155. 1308 INTEGRATED REMOTE SENSING OF DIVE SITES John P. Fish Iand H. Arnold Carr 2 2 Oceanstar Systems Inc. American Underwater Box 768 Cataumet, MA USA 02534 Search and Survey LTD Box 768 Cataumet, MA USA 02534 ABSTRACr Planned survey methodology and proper employment of the tools selected is also necessary to A par avant survey of dive sites using remote maintain the quality of the remote sensing data sensing equipment increases the efficiency and recorded. Often, when the exact location of a success of any dive program performed for dive target is not known, a search is required ecological, geological, archaeological or other prior to dive operations. For recently lost specific search and survey purposes. The type of targets, thorough research of wind, deep water equipment deployed can include side scan sonar, currents, wind driven surface current, last sector scanning sonar, magnetometers, profilers, known location of target as well as potential very high accuracy positioning systems or a target location on the seabed is necessary to combination of this equipment depending upon the. provide the highest probability of success in focus of the dive program and general the shortest amount of time. For older targets hydrography and geology of the area. Pre-dive research is necessary of historical loss survey and analysis whether through the use of a records, bathymetric charts, and target single piece of remote sensing equipment or a construction details. combination is further enhanced by integration of data 'on computer. The use of remote sensing, The tools used to pinpoint a dive site are often the integration of data and its use on dive chosen as a result of such research. The su*rveys is demonstrated by four examples of at environment in a specific dive area may preclude sea surveys. These examples relate to programs the use of any of the survey tools. Rugged of specific search, fisheries research and geology, for instance, complicates the use of nautical archaeology. side scan sonar in both operations and interpretation of the sonar record. On the other INTRODUCTION hand, high current velocities, while having Remote sensing underwater was in its. infancy little effect on most sonar systems, make the until the development of the acoustic use of many smaller, less powerful, ROV systems bathymetric sounder in the 1930's. Since the impractical. Searching for small targets using first third of this century, every decade has a proton precession magnetometer in areas of produced great strides in the development.of geological formations containing magnetite can equipment and techniques to remotely sense and be difficult at best. High noise environments image both natural and man made structures often complicate the use of acoustic mapping underwater. In the late 1980's the use of this systems. When the environment is not optimal for equipment and the development of integrated common remote sensing systems, a combination of remote sensing techniques have greatly increased systems will often provide the data needed to our ability to predict the environment and complete thedive operation with minimal time existing conditions before commencing diving and cost. operations whether by manned submersibles or by METHODS divers alone. Without a full understanding of the conditions Often in cases of marine survey using towed surrounding a dive site, dive operations become systems, the water depth, type of operation, and more time consuming and often have less bottom type will dictate the best methods to predictable results than if the site is surveyed use. However, for the described examples, the beforehand. methodologies included the following parameters: Acoustic imaging with both accurate high For Side scan sonar operations, scanning ranges frequency bathymetric sounding equipment and were between 25 m and 150 m per side. On the side scan sonar is usually the first equipment shorter ranges, in order to gain the best chosen for remote sensing operations but both possible resolution, 500 kHz was the primary the magnetometer and ROV's can provide frequency used. At the longer ranges, 1001diz was information that may not be detected by either used in order to minimize attenuation over the of these imaging systems. longer acoustic travel paths.1 Tow speeds ranged CH2585-8/88/0000- 1309 $1 @1988 IEEE from two knots for specific target mapping to used to determine a path to the source. six knots for large object search operations. Limeburner modified conventional wreckage drift During search operations, lane spacing was analysis by factoring the wind drift speeds by between 100 meters per side to 250 meters per the amount of windage or keel the flotsam side. For all sonar operations, primary support possessed. vessel positioning was achieved by using LORAN C. MATERIALS During magnetometer searches, tow speeds ranged f rom 0. 5 to 4. 0 knots. Speeds were determined All the magnetometers used for the surveys were by target mass and the water depth at the dive of a proton precession design. The equipment site. Lane spacing was five meters for small used included a Geometries model 806, a targets and up to 30 meters for large targets. Geometries model 866 and a Barringer model M- Three types of positioning were used to position 123. All magnetometers were interfaced to a the support vessel during magnetometer computer using custom interfaces designed and deployments. For operations using wide lane manufactured by Marine Systems-Technology. These spacing, LORAN C was adequate.enough to maintain interfaces allowed magnetometer and positioning vessel track. For operations to map or locate data to be logged to a single RS-232 computer smaller targets, either a radio transponder net port. Because of it's portability, the computer was set up on shore or visual surface markers used was a Zenith model 182 laptop. The data was were used to maintain vessel track. The latter then transferred to a 386 based desktop computer technique is difficult unless the search area is and.processed into contour and isometric small, the water depth shallow and current projections. almost non-existent. SHARPS, manufactured by MTI Pocasset, MA During ROV operations, the surface vessel was consists of four small transducers each capable usually positioned into a multi-point moor over of transmitting or receiving high frequency the dive site. Often at depths where an ROV is acoustic signals. A 386 based desktop computer required and especially if there is any current, controls system operation including acquisition, a downweight system is married to the ROV processing display and output of position data. Each transceiver is connected to the system umbilical. A length of free umbilical off the controller through a coaxial cable which downweight, line allows for ROV maneuvering. The provides power and a data path. length of the free tether is dependent on the maneuvering required and the hydrographic The side scan sonar used for the pre-dive conditions encountered at the site. surveys include an EG&G Image Correcting, dual frequency model 260 and a Klein model 401. Detailed mapping of a specific shipwreck was These systems provided high resolution images at done with a Sonic High Accuracy Ranging and ranges of 100-200 meters from the tow path as Positioning System (SHARPS) . Deployment of well as being capable of short range close-up SHARPS was accomplished by placing the surface inspection of targets. Positioning was performed vessel in a 3 point moor over the dive site. In by either a Northstar model 800 LORAN C or a order to calibrate the acoustic grid on the site Motorola MR III positioning system. of the LRRRT.ET 8 JACKSON, the team measured the speed of sound in the water using two of the RESULTS AND DISCUSSION receiving stations. Three transceivers were Search then set up on the 1st or 2nd futtock of the frames on the wreck. Extreme care is necessary In 1908 the steamship YANTiEF sank off the coast when positioning the transceivers to insure that of New Bedford, Massachusetts. Subsequent they are absolutely stable during the mapping salvage efforts were unsuccessful and the ship process. These transceivers were set in a w7aS extensively dynamited to minimize it as a triangle forming a plane to which 3-D hazard to navigation. measurements are referenced. The sea's surface was the reference plane for locating the vertical 'position of the receivers and the In a pre-dive survey of the wreck the tools distance measured from this datum and recorded. chosen were a side scan sonar and a When this was accomplished, three divers ' magnetometer.-Because of the large size (>100 m) working in rotation, mapped the site using a of the target and its probable location in trigger fired acoustic "gun".2 shallow water (10 m) on a moderately flat seabed, a side scan sonar was the first tool For planning search programs for recently lost used to search for the site. Ranges used were targets, computerized drift studies were 150 m per side and lane spacing was 250 m required to narrow the search areas. The providing 33 per cent overlap during the search. northeastern coastal region is very tidally Tow speeds of 6 knots were deemed appropriat active and flots 9M can often be tracked back to for a target of this size at 150 meters range. a point of loss. This work was performed by Dr. Once located the target was imaged at Richard Limeburner of The Woods Hole progressively lower speeds and shorter sonar oceanographic Institution. To back track flotsam ranges. from a terminal location drift was broken into the three components of wind, tide and low The outline of the shipwreck (A) and surrounding frequency current. All three components were seabed is shown in Figure 1. In normal sonar 1310 77@ 4 MR i,,I "P. "S ';'A T-uyj T@@' -g 1@3,,,k "Itz A _,12 IM 94 46 ''I IS _FLA io @@ff, C 7z 4 A 'Ix3 -a -4,4' I, v, if,g J,@ 44 T A 1 C PPI V EI WAK 4p If", JW# v A cH p"t py ,XI jkf 14, i, vt N, tfU4, ti"A4'_ 4, Ra Its", 'A, 4 32 U 14 A4 j l@ . 9 Ve tl IS z ki VIE, 11 fIV III ex 1,11 WWW" 95 V -$J kJ, of 41, Figure 1: Side scan sonar image of the remains of the steamship YANKEE (A). Due to dramatic changes in seabed make-up and reflectivity, an isolated target (B) is barely recognizable in relation to the primary target. si55 Rl V711 771, P -W Figure 2: An isometric projection of magnetic data at the site of the steamship YANKEE. A magnetic anomaly (C) separate from the main wreckage has appreciable ferrous mass. 1311 operations the smsl 1 target (B) may easily go essential, tool. @ The possibility of search unnoticed considering the large size of the operation success, particularly in aircraft primary target and the difference in bottom search, can be increased by integration of a reflectivity of the seabed surrounding the two variety of data from several sources. @This can objects. best be illustrated by the following two examples. The aircraft in each example was A magnetometer survey was made at the site to tracked by FAA radar and lost without warning. determine the amount of buried debris around the The search plan relied on the radar plot prime target. The data was processed into an information and computer generated current and isometric projection for easier visual wind drift of aircraft debris. The radar plot interpretation (Figure 2). The magnetic survey would normally be the location to center the shows a large amount of scattered debris that search. Debris from the aircraft can be used to probably occurred during the demolition. More generate a current and wind drift model that may importantly, the position of the magnetic further confirm the center of the search area. anomaly (C) shown in Figure 2 was determined to The degree of error is obviously a function of be the same as target (B) in Figure 1. It was the duration between actual loss and the time subsequently scanned with sonar at close range the debris was recovered. Error is also a and high frequency. A 500 1diz sonar record of function of the quality of data available on the target shows that it had considerable relief current and tide. As the error enlarges, the from the seabed (>1m) as evident by the size of most probable search area increases in size. the acoustic shadow. (Figure 3) Dive results on this target identified it as being a large The first aircraft search, conducted off Block section of hull plate from the main body of the Island, RI, was based solely on what was shipwreck. believed to be accurate radar information. A sonar search of the probable impact area did not result in any likely targets. Twenty three days after the aircraft disappeared, small pieces of debris confirmed to be from the subject aircraft, was located 60 miles from the search X_r- area. At this point wind, tide and low frequency current data were assembled and processed into a drift plot and search area. 'T' Because of the time between the aircraft loss and the finding of the debris, the new search area was large. The area was also well-removed from the area searched on the basis of the radar information. But based on this new data and the redirected search a likely target was located with sonar f ive days later. The target, once W, @TA located at long sonar ranges and high tow speeds, was repeatedly seamed at close ranges. A Since these high resolution sonar runs did not @C' TA exclude this as the suspect aircraft, a small 6@j,@rA ROV was used to video the target. The ROV N'T' 9 confirmed that this was indeed the badly broken remains of the lost aircraft. 4 Ye q W, In the second example,, a search was conducted 6@',A or Ad P for a light aircraft lost off the island of AU, A J11 Nantucket, Massachusetts during a landing approach. Other than a portion of landing gear found approximately 10 hours after the incident, no flotsam was ever located. For this search, wind drift computer analysis of flotsam in the Figure 3: A short range sonar scan of targets B Nantucket Shoals areg was also provided by Dr. and C (Figures 1 and 2) shows a sizable section of Richard Limeburner. His results provided a debris with relief from the seabed of over 1 drift of the debris over the previous 10 hours. meter. Divers identified this debris as sections Side scan sonar was used as the primary search of hull plate, probably blown away from the main tool with divers confirming sonar targets. The wreckage during demolition. computer generated drift plots, coupled with the FAA radar tracking.data, were used as a primary By combining both sonar and magnetometer data it basis for the sonar search plan. was possible to determine the integrity of the main target and the distribution of any other Although many sunken aircraft are found near- anomalies around the main site. This made dive intact, our analysis of the FAA radar and operations and identification of the debris control to aircraft communications suggested field far easier and less time consuming. that this aircraft impacted the sea surface at high speed. We therefore analyzed all targets In searches for downed aircraft, sonar is an very carefully. 1312 A pre-sonar search included a fly over at low activity that has been reported on the site altitudes by a fixed wing search aircraft. From (Morris Johnson, Yarmouth Department of Natural the air, one target was located and subsequent Resources - personal communication). sonar inspection and research proved it to be a steel shipwreck approximately 20 years old. The reef still must face the forces of a severe hurricane. Several storms with southerly winds A number of other targets were found and up to 50 knots have hit the area. One hurricane positioned using the side scan sonar. The area with winds between 60 - 80 knot winds also surrounding Nantucket Shoals has moderate to impacted the area in 1984. None of these winds high tidal currents and for this reason have caused sea conditions that appear to have positioning the target accurately before the changed the tire reef distribution. dive operation was important. One target in particular provided an acoustic reflection much Nautical Archaeology like that of a fixed wing aircraft with proper dimensions and shadowing. The target position In another example, a shipwreck having sunk in was carefully determined and the surface vessel 1893 began to emerge from the sands of the high moored in a one point moor. Divers were deployed energy eastern coastline of Cape Cod and was to the site. The target proved to be a mooring located by a local fisherman. Our historical block and chain assembly discarded or lost from records indicated that the wreck was the a offshore navigational buoy. The divers barkentine HARRIET S JACKSON. We made a inspected the site carefully to be sure that the reconnaissance dive on the site and found a lost aircraft was not in the region or being remarkably well-preserved sailing vessel in this masked from the sonar by the old mooring. They shallow marine shoreline. The keel and many of found that the mooring block and portion of the the frames and inner timbers werepresent. This chain made up part of the target while the suggested that the vessel may have quickly remainder of mooring chain had fallen to the buried after it came ashore and may, as many seafloor perpendicular to the block. This shipwrecks in this vicinity do, become orientation gave the entire assembly the periodically exposed for short periods of time. appearance on the sonar record of a light That possibility could best be answered by a aircraft. This particular search was suspended multidisciplined remote survey of the site. We during the winter of 1987-88 without locating further decided to determine how discreet the the aircraft, but further work is planned in the site was and to utilize SHARPS to record the region indicated by the drift plots. construction detail of the exposed remains. Fisheries Research First, the site was acoustically imaged using an EG&G model 260 Side Scan Sonar and dual- In 1978, the town of Yarmouth, Massachusetts frequency towf ish. The sonar image supported obtained approval to construct a tire reef in the divers reconnaissance observations but also adjacent Nantucket Sound. The.approval was provided. spatial relationshipEl of the site conditioned on a monitoring program proposed and components. The divers noted that the foremost undertaken by thip Massachusetts Division of frames of the vessel were gone and that the Marine Fisheries. The Division combined diving forward part of the keel and keelson were slowly observations with sonar imaging to assess the separating from the body of the ship. potential of the site, the construction of the reef as to density and distribution of tire The side scan sonar record was the first large units, and.stability of the tire units. scale, two dimensional plan view of the wreck The typical tire unit placed on the reef site which provided an important quantification consisted of four tires secured together by non- of the vessel.remains on the site. metallic bands, vented at the top, and ballasted A magnetometer was used to survey the site to with concrete. The tire units were placed at determine if any anchors, chains or other iron random within a 200 meter diameter circle. fittings were present but buried around the The reef,was located in shallow water, about 30 site. Although the vessel's hull was primarily feet. The area was therefore subject to the fastened with copper and bronze, she had iron energy of any major storm out of the south. sister ribs installed sometime after her After 515 tire units were deployed, a sonar construction. These sister ribs were not record was made of the reef. The record allowed observed during the initial reconnaissance dive; divers to more efficiently survey the tires they did present a significant read-out on the magnetometer and this construction detail and because they could better plan their dive source for the unexpected magnetic anomaly was survey. Underwater visibility was ten feet later confirmed by divers. The magnetometer maxiTmnn, so divers were unable to accurately log survey indicated that the portions of the vessel distribution. that were missing for the wreck were not in the A comparison of two sonar records of the tire immediate area of the wreck. The absence of reef, one taken in 1981 and the other in 1988, anchors and chains led the team to believe that indicate that the reef is generally stable. the wreck had been quickly salvaged between the Changes in the distribution of the tire units in grounding and several weeks before the upper some isolated areas of the reef, we believe to decks were destroyed and the remaining hull be caused by some commercial fish trawling buried by a storm. 1313 Since there was a considerable amount of relief through the use of these two instruments and to the site the team realized that that a three SHARPS. The fisheries research which utilized dimensional representation would be required to divers and the sonar continues as the fully document the site. The SHARPS was distribution of the tire units in successive deployed on the site to map the wreck in three sonar images are being compared with the use of dimensions. Aboard the support vessel the survey a computer. The use of computer generated drift team was able to fully monitor the progress of plots for debris from downed aircraft coupled the mapping. The computer screen displayed the with sonar and ROVs will assist dive operations position of the probe as the diver traced out to recover submerged aircraft. the lines of the keel and frames where they protruded from the sediment. Approximately one Integration of instruments and onboard third of the way forward from the stern some computers, especially on small craft and through ceiling planks were intact and these were relatively inexpensive means will allow the plotted by the diver swimming along the joining researcher or group with a modest financial edges and triggering the transmitter at capability to acquire data that will justify and intervals of three centimeters. satisfy their honest effort. Also the combination of data that can be gathered Each of these data points were automatically increases an overall understanding and recorded by the computer. The total time it took interpretation of the conditions at the site. for the diver to map the site was just over The operation may not always be successful, as three hours. In that time, the system recorded in one of the one of the aircraft searches 10,000 data points directly to the computer. mentioned above, but the quality of the operation and the chance of success is greatly Since the three axis data is accessed increased. sequentially, the resulting line drawing is a three dimensional representation displayed on a The ability and means to integrate remote CRT and printed on a plotter. A further sensing instruments and the computer will remain advantage of the 3-D data is that it can be dynamic and relate to (even more) modest rotated on the CRT by the computer and printed financial demands and increasing personal or examined from a variety of perspectives. innovation. The integration of remote sensing and detailed REFERENCES mapping of this site provides us with complete information about the wreck and its surrounding 1. Fish, J. P. State of the Art Sonar Images. Skin environment. Not only is the map of the site Diver Magazine, January 1983 excellent documentation with which further deterioration of the site can be assessed but 2. Fish, J. P., Carr, H. A. and Geisel, F. A. also later work on the site by divers can be SHARPS: High Accuracy Mapping Of Submerged greatly aided by the application of these Shipwrecks. Sea Technology, March 1988 instruments. Work In Progress 3. Butman, B., Beardsley, R.C., Magnell, B., Frye, D., Bermersch, J.A., Schlitz, R., Limeburner, R., Wright, W.R., Noble, M.A. Recent Observations Of A project still in progress is a fisheries the Mean Circulation on Georges Bank. Journal Of investigation and survey of one ghost gillnet Physical Oceanography, 12 (6), p.569-591 that was lost on Jeffries Ledge, in the Gulf of Maine. The net was first found during a three 4. Fish, J. and Carr, H. A., (unpublished year study of derelict demersal nets using manuscript) Seabed Target Detectability Using Wide submersibles and subsequently re-surveyed every year since 1984.7 Recent surveys have been with Area Search High Speed Sonar Tow Methods. an ROV during months when a submersible was not Technical Note No. 12, 1987, (available from: available. Efforts to map the net with either a American Underwater Search and Survey) side scan sonar or sector scanning sonar have 5. Limeburner, R., Beardsley, R.C. Seasonal been, frustrated, by weather or operational Hydrography and Circulation Observations on problems. However, the investigation is Mantucket Shoals. Journal Of Marine Research documenting the impact of the net, especially Supplement 2,40, 1982 P.371-406 during winter months, which is of particular importance becausg important groundfish, such as 6. Carr A. and Amaral E. H., Review of the cod, are present. Potential for Artificial Reefs along Coastal Massachusetts. Proceedings MTS OCEANS 81, 1981 CONCLUSION 7. Carr, H. A. and Cooper R. A. Submersible and All of these examples integrate one or more ROV Assessment of Ghost Gillnets in the Gulf of combinations of instruments that give the Maine. Proceedings MTS OCEANS 87, 1987 (p. 622- researchers program managers or diver means to 624) better document, search, or understand the subject of their work. The shipwreck site 8. Carr, H. A. 1988 Long Term Assessment of a information was better described with the use of Perelict Gillnet In The Gulf of Maine. t1TS OCEANS the combination of the magnetometer and sonar or 88, 1988 (Manuscript in process) 1314 HUMAN POWER GENERATION IN AN UNDERWATER ENVIRONMENT Stephanie L. Merry, Scott L. Sendlein and A. Paul Jenkin Department of Ocean Engineering Florida Atlantic University Boca Raton, Florida 33431 ABSTRACT phenomenon is well known to racing bicyclists and much work has been done to determine the optimum A pilot study of SCUBA divers pedaling an cadence [ 2 1 . Maximum power is achieved at underwater bi@yclelergometer is described. The approximately 90 rpm. ergometer design and experimental procedure used for tests on 14 male and female subiects in the The optimum cadence for a SCUBA diver moving his age group 22-46 is presented. The av;rage useful limbs against water resistance should be lower power output for subjects who are regular than fo; the bicyclist in air and the 'useful' bicyclists or who participate. in a regular power transmitted to the pedals will also be exercise program was 0.12 HP at the optimum lower. Some research in this area has been done cadence of 50 rom. Non-bicyclists and less fit by the Navy Experimental Diving Unit in Panama subjects tended towards a lower cadence for City, Florida, who conducted a series of tests on maximum performance. These results, and the test diver work performance between 1971 and 1978 f3, subjects' reactions to the tests, are discussed in 41. Their data was collected under conditions of relation to the design and safety of a human- high pressure and below normal temperatures, using powered submersible. a bicycle-type ergometer. Trained Navy divers were tested in a flooded byperbaric chamber at simulated depths of up to 1600 feet of sea water INTRODUCTION (.fsw). The Ist Annual International Submarine Races will. The Navy studies were primarily concerned with the be held in West Palm Beach, Florida on June 23-25, effects of extreme pressure and cold. Since the 1989. The competition is for small subsea human-powered submarine races will be conducted in vehicles, propelled solely by human power, which semi-tropical waters off the coast of Florida, at will be judged on their speed, cost-effectiveness depths less than 20 fsw, this earlier research and innovation of design. Entries will be limited does not apply to the conditions that the race to free flooding su@marines, having two crew competitors will experience. Theref ore the members sustained by SCUBA equipment. Sponsored by present project was undertaken, with the following the H. A. Perry Foundation and the Department of objectives: Ocean Engineering at Florida Atlantic University, the competition is intended to inspire innovative a.) to measure the maximum sustained useful power research and development in the areas of output of a SCUBA diver when pedaling hydrodynamic efficiency, propulsion, and life b) to assess the optimum underwater pedaling 5upport systems for subsea vehicles. With the cadence restriction of human power, the designer's c) to gain experience dealing with SCUBA divers attention must focus on the optimization of these exercising at high workra@es and appraise critical parameters. the safety hazards. In addition to its organizational involvement, This paper describes the ergometer design and Florida Atlantic University will also be entering experimental procedure used for a series of tests a vehicle in the competition. Background research on fourteen SCUBA divers. The results are into previous human-powered vehicle projects discussed in relation to the design and safety showed that pedaling is the most efficient method aspects of a human-powered submersible. of converting human power to mechanical power [11, so this type of transmission was chosen for.the TEST EQUIPMENT vehicle design. The experience of earlier research in the area of There is generally a certain pedaling rate human-powered vehicles suggests that a recumbent (cadence) which is most comfortable for an bicycling position is the most efficient [5). individual bicyclist on land and which is easiest Therefore, a recumbent frame was salvaged from a to sustain for long periods of time. This previous man-powered land vehicle project and was CH2585-8/88/Q,000- 1315 $1 @ @1988 IEEE Q converted into the underwater ergometer shown in A, Figs. I and 2. L7 e Z @Z, 1 4C Adjustab@e Seat Front Forks Fiywhee@ tj A17 5M-Tank Stand S p r o-c 1k @eett TEIP VIEW rks Instruments C,bl,e Tensioner-"J),-Brakes Spring Scate ZE.r 0 Fig. 3. Flywheel and Brake Mechanism CUP Heart rate was measured using an optical pickup, which is clipped to the subject's ear lobe and '--Reor Stand Lain Flywheet shines a lig@t through the lo@e to a photocell on the other side of the clip. Blood pulses obstruct SIDE VIEW Crankset \-Front Stand some of the light and the photocell records a pulsing light source. The computer compares these Fig. 1. Schematic Diagram of the Bicycle/Ergometer pulses with its internal clock and displays the subject's heart rate in beats per minute. The electronic pickups which were exposed to water e am, A *1 & . . . . . . . were encased in epoxv and therefore perfectly sealed water tight. The instrumentation was inside a commercially available, clear vinyl, plastic bag (Fig. 4 The bag was sealed by means of a bolt-down metal closing rail. The pickup wires were sealed as thev exited the bag by placing them between layers of an adhesive putty between the closing rails. The instrument bag was then tied to the el@gometer and allowed to float at the eye level of the subject diver and the test operator (Fig. 5). This allowed the subject to monitor and control his cadence while the test MR, operator monitored the subject's heart rate. A F, X @g Fig. 2. Side View of the Bicycle/Ergometer ne. The ergometer utilizes an inertial flywheel which is connected via a chain drive to a standard bicycle crank set and pedals. Attached to the axle of the freewheel is a fork which is free to rotate independently about the flywheel axle (Fig. C 3). Mounted on this fork are a set of bicycle caliper brakes, with a device for adjusting the amount of contact force of the brakes. The fork is restrained from rotation by a spring scale V calibrated In lbf, which is attached between the LA W R, X I fork and the frame of the test stand. Since the brake Dads are positioned at a radius of exactly one foot from the axle of the flywheel, the force indicated on the spring scale is numerically equal to the resisting torque generated, measured in Fig. 4. Heartrate and Cadence Instrumentation units of foot-pounds. The seat on the ergometer was adjustable to Pedal cadence and subiect heart rate were accommodate variations in leg length. There was a monitored with a standard bicycle computer system simple seat-belt arrangement to secure the subject which provides these functions. The cadence was in the -seat (Fig. 5.) and toe-clips on the pedals. measured by a magnetic pickup near the crank arms The ergometer was placed in a pool at the FAU of the pedals. Department of Ocean Engineering, which caused the 1316 subject's chest to be at a depth of ten feet of fresh water (.ffw). ------ -7T-117 In;tiat Torque 2.0 f t-tbf 4, ATT Peclat For 2 Mins, HR < 60% MHR 60% < HR < 70% MHR HR > 70% MHR JUm J P ff 41, - RX -W Increase I Ft-kbf -7 Increase 2 ft Continue To Ri- d; 2 Min. Intervo Fig. 5. Seating Arrangement and Instrument eady Heart (+/- 5 BP Location STOP A SCUBA regulator was selected based on its ability to supply large quantities of air to the Fig. 6 Test Procedure Flow Chart subject [6] and the same regulator was used for all subjects. The regulator was attached to a As an aid to data analysis, the subjects filled standard SCUBA tank of 80 cubic foot capacity. out a questionaire after'the tests, giving details which was installed next to the seat of t@e of their SCUBA and bicycling experience, and ergometer. frequency of participation in other exercise or sport. They were also asked to comment on several TEST PROCEDURE aspects of the test equipment, such as the seating arrangement, belt constraint and air supply. Testing was performed in accordance with the sit Finally, they listed any forms of discomfort f widely accepted Monark bicycle fitness test (3). during' testing, which ranged from nervousness and This procedure is summarized in the flow chart of anxiety, through headaches and aching legs, to Fig. 6. The subject is required to pedal at a claustrophobia, lack of air and dizziness. fixed cadence against an applied load, which is gradually increased until the heart rate reaches 70% of maximum. Each subject's maximum allowable heart rate (MHR) is computed by the formula: MHR 220 - (subjects's age) 4. 4,& . The optimum underwater cadence was thought to be in the range 40 - 60 rpm. This range was selected intuitively after several trials on the ergometer by one of the researchers, who is an. experienced racing bicyclist. Three test cadences of 40, 50 N. Q ;: and 60 rpm were then chosen. The order of the d - cadences was randomized, so that the results would 6W not be biased by a "learning effect" and subjects "o elf 11 0-7 W- rested for at ieast one hour between trials, to @4 "g, -41 eliminate the fatigue factor. 4 k M The pressure in the SCUBA tank that supplied the air to the subject was noted at the beginning and Fig. 7. Underwater Testing end of the test and the difference was recorded. This data was used to give an approximate Figure 7 illustrates underwater operation and indication of the air con s umntion rates that can testing on the ergometer P 'A be expected in the operation of a buman-powered submarine. 1317 TEST SUBJECTS the Navy Experimental Diving Unit which suggests that onlv 50% of a diver's sustained gross power Test subjects were male and female volunteers from output (approximately 0.33 HP for an extremely fit the Department of Ocean Engineering. They were male) can be converted to useful mechanical all certified divers, but their ages range@ from energy. The remaining 50% is wasted through 22 to 46 and there was a wide variation in level resistance to motion underwater [7]. of fitness. As a safety precaution, all subjects were required to undergo an independent fitness Note that for subject #1, the 40 rpm data point is evaluation, prior to the underwater tests. Details slightly low, since the limit of the applied load of the test subjects are presented in Table 1. an the ergometer was reached when his heart rate was only 65% MHR (rather than 70%). TABLE I. Details of Test Subjects a) The second highest power output was from subject 4, who was a novice SCUBA diver. This suggests that the data is not a function of a subject's 0 a) 1@4 previous SCUBA experience, provided that he or she $4 has sufficient confidence to complete the tests. cd W @4 Power output from the female subject (#6) is predictably lower than for the males. Although W relatively fit, she was a non-bicyclist, and lack W) U > 0 @E: of technique may also be a factor here. I M 29 E R Y 0.17 - 2 M 24 E T Y 0.16 - 3 M 24 E T Y 0.15 - 4 M 26 N T Y 0.14 - 5 M 34 E N N 6 F 38 E N Y o.13 7 M 46 E 0 N o.12 8 M 26 E N N E o.11 0 1 9 M 26 E N o.1 10 F 22 N N N o.og 11 F 32 N T Y o.oo 12 M 37 N 0 Y .13 F 28 E N N 0.07 14 M 23 E N N o.o6 o.o5 - Key: SCUBA Experience: N - novice 0.04 E - experienced 40 so 6o Bicycling Experience: R - racing 0 2 C.d-- (RPIM) T - regular 0 #1 & . #4 x #6 0 - occasional Fig. 8 Power Output for Subjects Peaking N - non-bicylist at 50 rpm. Regular exercise program: Y - yes N - no The data for the subjects who peaked at 40 rpm is RESULTS presented in Fig. 9. None of these male subjects were regular bicyclistsi nor did they partake in a There was a wide scatter in absolute values of output power, due to the variation in fitness and OAL - bicycling technique between the subjects. However, a significant downward trend in power o.1 output was apparent in all cases at 60 rpm (Figs. 8 and 9.). o@0 Figure 8 presents the data for those subjects o.o8 whose peak power occurred at 50 rpm. These people were all regular bicyclists and/or participated in o.o7 a regular program of fitness training. Four of the five subjects were male (#6 was female) and it 0.06- is interesting to note that, despite the wide spread in their maximum power (0.09 - 0.17 HP ') at o.o5 50 rpm, the data for these males converges to 0.09 HP at 60 rpm. 0.04 The highest output of 0.17 HP was generated by 40 so 6o c.d..C. (RPNO subiect #1, who is an experienced SCUBA instructor C! 19 .6 o .7 and a competitive racing bicyclist in training. This fizure correlates well with information from Fig. 9 Power Output for Subjects Peaking at 40 rpm. 1318 regular exercise program. The maximum power TABLE Il. Summary of Discomfort Symptoms ranges from 0.08 HP to 0.10 HP, with a similar Experienced During Testing spread of data at 50 rpm and some convergence at 60 rpm, although this is less marked than in Symptom Subject Number Figure 8. 11 2 3 4 5 6 7 8 9 10 11 12 13 14 Mean values of the data are presented in Figure 10, where the upper line represents those subjects Headache during who partake in regular exercise (bicycling or test Ix otherwise) and the lower line represents the Headache after I subjects who do not. The optimum pedaling cadence test Ix X X is about 50 rpm for the former group, whereas the Leg ache I X X X X X less fit subjects prefer a lower cadence, of 40 Claustrophobia Ix X rpm or less. Mean peak values of useful output Tightness in chastlX X X X power are 0.12 HP fo@ the 'fit' group and 0.08'HP Lack of air Ix X X X X X for the others. Dizziness Ix X Tunnel vision I X 0.12 - Potential black- I out I X Nervousness/ I OAL - anxietv X X OA-I Dry mout@ X Neck strain X 0.09- 0 0.08- Equipment section this suggests that development of a high flowrate SCUBA system would be 0.07- advantageous for the human-powered submarine. One subiect complained of a dry mouth, so a humidifier 0.06 should also be incorporated. 0.05 Four subjects suffered from tightness in the 40 so 60 chest. The maiority of these were-tall people and a [email protected] E--. C.d-- (RPM) E..,,i.e we believe tWat the oroblem mav be overcome bv extending the length of the bicycle frame and Fig. 10. Effect of Regular Exercise on redesigning the seat to permit a wider range of Mean Power Output adiustment. The neck strain problem for subject #4 would also be solved by this modification. INCOMPLETE TESTS Five people, all non-bicyclists, complained of lee Subjects #10-13 failed to complete the tests, due ache, wLch was simply caused by the unaccustomed to physical or mental discomfort. Of these, three exercise. Two subiects, who were inexperienced were inexnerienced SCUBA divers. The fourth SCUBA divers, confessed to anxiety during the subiect was unable to attain the 70% MHR, because tests, but the maiority were relaxed. of tired leg muscles. The reason for headaches during and/or after Subject #14 successfully completed the tests, but testing is not fully understood, but we hope to the data was suspect due to equipment malfunction gather more data on this subject during future and has been discarded. projects. Finally, claustrophobia is an individual problem, which would probablv be exacerbated in the enclosed environment of a DISCOMFORT DURING TESTING buman-powered submarine. Information provided by the test subjects on the The experience of subject #12, who suffered all questionnaire, regarding any discomfort they the symptoms of a potential blackout, emphasizes experienced during the tests, is summarized in the hazardous nature of these tests. This project Table II. This information has been used to would not have been attem-oted without: identify specific safety hazards for the operator of a human-powered submarine and potential areas a) the background of an established divine program where equipment development or a special training for the training of scientific divers, regime would alleviate operator discomfort in the within the Ocean Engineering Department. stressful environment of the submarine races. b) the collaboration of the Universitv's Department of Professional and Human Services, The most common complaint from the test subjects who certified all subjects as physically fit was an inadequate air supply. Since the regulator and advised on the test procedure. was selected on the basis of its reported ability c) an experienced SCUBA instructor to conduct to deliver large quantities of air (see Test the tests. 1319 AIR CONSUMPTION ACKNOWLEDGEMENTS By the nature of the Monark bicycle test (see Fig. The authors,wish to thank Messrs. T. Johnson, E. 6) the work rate is not constant, but gradually Hendra, K. Houggy and K. Nelson for assistance in increases. Therefore, the air consumption rate constructing the ergometer. The advice from Dr. (which is a function of work rate [8]) will also Michael Whitehurst and the contribution from Miss gradually increase. The subjects' air consumption Felicia Ricci (who conducted the fitness was monitored by noting the reading of the evaluations), both from the Department of contents gage at the beginning and end of each Professional and Human Services at Florida test and thus the most meaningful data was Atlantic University, is gratefully acknowledged. gathered from subjects whose maximum work load was equal to the initial applied load of 2 lbf. This Boca Seaadventures generously donated SCUBA occurred for subject #s 5, 6 and 8 at 60 rpm. equipment for this project ana discount bicycle parts were provided by Boca Schwinn Cyclery. The respiratory minute volume for each of these subjects, averaged over a four minute period of pedaling against a constant resistance, was 75 REFERENCES liters � 15%. The accuracy of this figure is limited by the precision oi the contents gage 1. Brooks, A. N., "Some Observations on the and by the short pedaling time, but it validates Energy Consumption of Human-Powered Vehicles", the maximum respiratory minute volume at high work Proc. lst H.P.V. Sci. Symp., Anaheim, CA, loads of 100 liters, quoted in Ref. [8). Nov. 1981, pp. 42-53. Further research is required in this area, which 2. Spinnetti, R., "Backward Versus Forward will be the subject of a future project. Pedaling: Comparison Tests", Human Power - The Technical Journal of the IHPVA, Vol. 6, CONCLUSIONS No. 3, Fall, 1987, pp. I and 10-12. This project has demonstrated that the maximum 3. James, T.W., "Modified Collins Pedal-Mode power output and optimum cadence of a SCUBA diver Ergometer: Development and Medical Tests", pedaling underwater depends upon fitness and/or Navy Experimental Diving Unit Report No. bicycling expertise, but not SCUBA experience. l-A, 15 June, 1976. The latter affects only a subject's ability to complete the tests successfully. 4. Middleton, J.R. and Miller, B.E., "Evaluation of Kinergetics Breathing Gas Heater", Navy The maximum measured power from a racing bicyclist Experimental Diving Unit Report No. 9-78, was 0.17 HP at 50 rpm. Underwati;r bicycle April 1978. training of the subject and improvements to the air delivery system may increase the power by 5. Wilson, D.G., Forrestall, R. and Henden, D., perhaps 10 - 15%. This data will be used for "Evolution of Recumbent Bicvcles and the optimization of the propulsion and transmission Design of the Avatar Bluebell", Proc. 2nd devices on the FAU human-powered submersible. HPV Sci. Symp., CA., Oct. 1983, pp. 92-103. Subjects experienced discomfort such as tightness 6. Morson, P.D. "Evaluation of Commercially of the chest and neck strain after relatively Available Open Circuit Scuba Regulators short periods (i.e. - 10 minutes) of working on Navy Experimental Diving Unit Report No. the ergometer. Since these symptoms were caused 8-87, August 1987. by a non-ideal pedaling posture, it is important that the interior of the submarine is customized 7. Schwartz, H., Navy Experimental Diving Unit, for the intended operator. Panama Citv, FL., Private communication, Dec. 1987. The testing program demonstrated the potential for blackout when a SCUBA diver is working at maximum 8. British Sub-Aqua Club Diving Manual, p. 84, rate underwater. This justifies the safety Stanley Paul Co.. London. guidelines for the buman-powered submarine races, which stipulate that a deadman switch must be installed for both occupants, and that only one of the two may generate power (the other being responsible for vehicle control and safety). Fruitful related areas for further research include: (a) SCUBA diver air consumption under maximum sustained workload (b) modifications to air delivery systems, to orovide unrestricted flow on demand (c 1) cause and prevention of the headaches experienced by test subjects. 1320 NOAA FLEET HAZARDOUS MATERIALS AND HAZARDOUS WASTE MANAGEMENT William R. Cunningham National Oceanic and Atmospheric Administration National Ocean Service, Office of Marine Operations Rockville, Maryland 20852 ABSTRACT hazard if handled and stored properly. These materials are used for numerous purposes, including degreasing, as "NOAA Fleet Hazardous Materials and Hazardous Waste additives to fuels and lubricants, cleaning agents, Management" details the objectives, development, and solvents, and as paints or coatings. Some NOAA Fleet implementation of a hazardous materials and hazardous employees encounter one or more of these substances on waste program for NOAA hydrographic, oceanographic, a routine basis through the normal course of their work and fishery research ships. while others are infrequently, if ever, exposed to the same hazards." The paper discusses key elements for development of the program, namely, conducting shipboard and ship base OBJECTIVES hazardous materials and hazardous waste surveys, reviewing EPA and OSHA regulations, developing a The objectives of the NOAA program are twofold: 1) to hazardous materials and hazardous waste manual for protect the health and safety of personnel and the shipboard use, and conducting hazardous materials and environment, and 2) to ensure compliance with hazardous waste awareness training. applicable EPA and OSHA regulations. Through interpretation of applicable regulations with emphasis on The paper also discusses implementation of the resulting protection of personnel and the environment, the NOAA hazardous materials and hazardous waste policy, content program has satisfied the combined objectives. of the manual, training information, shipboard and shoreside procedures, and maintenance of the program to To gain the support of those individuals who lacked an reflect changing hazardous materials and hazardous awareness of the problem, OMO cited medical concerns, waste regulations and shipboard needs. legal concerns, regulations, directives, and the overall direction in which hazardous material and hazardous waste issues were heading. However, like any program where benefits, improved productivity, or reduced costs INTRODUCTION are difficult to measure, obtaining adequate funding and support was not without obstacles. We were fortunate The Office of Marine Operations (OMO), a component to have been mandated to comply with EPA and OSHA of the National Ocean Service within the National regulations at about the same time that concern over Oceanic and Atmospheric Administration (NOAA), hazardous material handling and hazardous waste manages and operates a fleet of 23 hydrographic, disposal was being voiced by fleet and shoreside support oceanographic, and fishery research ships. The ships personnel. perform a variety of missions and serve as mobile platforms for visiting scientific parties conducting NOAA's early commitment to meeting the objectives has oceanographic and atmospheric related projects. saved a great deal of time and effort. NOAA has managed to keep abreast of regulatory requirements and As trustee of the Nation's oceans and In response to at times has taken the lead in developing standards for shipboard concerns, legal requirements, and federal, the marine industry. The organization is still benefiting state, and local environmental regulations, OMO has from the approach that was taken as new developments developed and implemented a successful NOAA fleet and related tasks become evident. hazardous materials and hazardous waste program which is centered around a hazardous materials and hazardous APPROACH waste manual prepared as a guidance document for fleet personnel. OMO initially developed a statement of work emphasizing the objectives and detailing specific tasks The NOAA Fleet Hazardous Materials and Hazardous that included: reviewing applicable health, safety, and Waste Manual states, "Hazardous materials and environmental regulations; surveying NOAA ships and hazardous wastes are encountered by NOAA Fleet fleet facilities for hazardous materials and hazardous personnel both aboard ship and at shore facilities. wastes; preparing a guidance document for fleet These materials/ wastes come in many forms with personnel; and conducting awareness training for varying degrees of hazard. Some are highly flammable, employees who routinely work with hazardous some are toxic, while others are considered a minimal materials/chemical products. NOAA was fortunate to 1321 United States Government work not protected by copyright have in-house expertise on the subject, and the information and training. In short, all chemical Hazardous Materials Response Branch in Seattle, containers must be labeled. Chemical labels are to be Washington worked out the plan. applied to containers by the chemical manufacturers or Importers. The labels must Identify the contents of the Regulations container and warn of any specific hazards of the contents. Secondly, OSHA requires that employers Due to the simultaneous development of the program and maintain an MSDS for each hazardous chemical product the lengthy regulatory process, the Resource purchased or used in a facility. An MSDS is a form that Conservation and Recovery Act (RCRA) Amendments of describes the characteristics and properties of a 1984 and OSHA's Hazard Communication Standard were chemical product. Specific physical and health hazard of great concern during the development of the program. data is given. Employees may review MSDS's of We reviewed the regulations to determine which portions chemicals in their work areas. Thirdly, every employer of RCRA and the OSHA standard were applicable to shall provide hazard communication information and shipboard operations. We also considered how future training to all employees. Employees will be made developments would impact the program and tried to aware of the chemicals in their work area, the hazards stay one step ahead of the regulatory process. OMO's of those chemicals and the availability of information interpretation of the legal language and complexity of about hazardous chemicals at their workplace. The the regulations has been in keeping with NOAA OSHA standard requires that training be conducted once objectives to protect the health and safety of the for existing employees, once for new employees, and personnel and the environment. Key provisions of the again should new hazards be Introduced into the regulations are described below. workplace.4 RCRA Amendments of 1984 Surveys Prior to the RCRA Amendments of 1984, generators of Surveys were conducted aboard NOAA ships and at less than 1000 kg (2200 pounds) of hazardous waste per shoreside support facilities to determine the types and month were exempt from regulations governing the amounts of hazardous materials, chemicals, and wastes disposal of.hazardous wastes. According to Attorney handled along with current stowage, disposal, and Michael A. Brown, former Enforcement Counsel for the handling procedures. Materials or products that could EPA, "The specter of up to 2200 pounds per month of be considered hazardous due to their toxic, irritating, unregulated hazardous wastes being placed into the reactive, flammable or polluting nature were identified environment by each of the' estimated 130,000 small and inventoried. From a report of the results of the generators caused Congress great concern."' surveys comes the following, "We Identified and inventoried [over 300] substances which pose varying in short, as a result of Congressional enactment of the degrees of hazard ranging from chlorinated solvents RCRA amendments, on March 17, 1986 EPA promulgated (which have a known carcinogenic potential) to regulations governing small quantity generators. The relatively innocuous products such as laundry detergents rules, now similar to those imposed on large quantity .... Ninety-five percent of these products were found generators, require that generators of more than 100 kg to be commercially prepared and packaged compounds but less than 1000 kg of hazardous waste per month used as ships' stores (required to maintain or operate (small quantity generators): the ship). These products were used for a wide range of activities including; degreasing, washing, lubricating, � Obtain a U.S. EPA Identification Number. fuel additives, antifreeze, Ice melting, water � Use the full Uniform Hazardous Waste Manifest conditioning, and for scientific purposes. The remaining when shipping hazardous waste off site. five percent was comprised of reagent chemicals that � Submit wastes only to hazardous waste transporters had apparently been left on board by previous scientific with U.S. EPA Identification Numbers. parties. These products and chemicals were stored In a � Accumulate waste on site for no more than 180 days, variety of locations, with most in the original product or 270 days if the waste is to be shipped more than container supplied by the manufacturer." 200 miles, unless generators obtain a hazardous waste permit. There was no information about the chemical products o Ensure that generators' hazardous waste is managed found aboard the ships other than what was labeled on at a hazardous waste facility with interim status or the products' containers by the manufacturers. Material a permit under RM.' Safety Data Sheets of the majority of products inventoried were obtained. Certain products were no OSHA Hazard Communication Standard longer being made and some manufacturers were no longer in business, thus some MSDS's were unavailable. Similarly, the Hazard Communication Standard has been viewed as a much needed but long debated rule that in addition to the materials mentioned above, several 55 became bogged down by technical detail, political gallon drums were found at the marine centers (major interference, and state and local regulations. As NOAA OMO ship bases) containing used solvents, lubricating forged ahead, trying to determine to what degree the oils, antifreeze, and bilge slops. standard was applicable to the marine industry, NOAA recognized the benefit's of incorporating the OSHA DEVELOPMENT standard into the program. The outcome of the surveys indicated that the type of The Hazard Communication Standard is a three-tiered hazardous materials used aboard NOAA ships was limited program comprised of 1) chemical labeling, 2) mostly to chemical products in relatively small amounts. material safety data sheets (MSDS), and 3) employee However, NOAA realized the need for inventory control 1322 of those products, improved procedures for chemicals A simple means of avoiding problems and providing for brought aboard NOAA ship's by visiting scientific proper handling of a varied noncontinuous waste stream parties, proper procedures for the disposal of shipboard- is to label all wastes, funnel the waste through a single point of contact at each shoreside facility, provide for generated hazardous waste, and hazardous materials temporary storage at the facility, and set up a purchase awareness training. agreement for routine disposal with a licensed disposal Inventory Control contractor. In order to reduce the numerous types and amounts of chemical products aboard NOAA ships and to provide Trainin some sort of control over inventory, NOAA investigated The nature of NOAA shipboard employment caused the the existence of a list of products approved for development of an atypical approach to meet the hazard shipboard use. We found out that the U.S. Coast Guard Commandant Instruction for listing products approved for Communication standard's "one time" training use as ships' stores was not appropriate. The Coast requirement. Accessibility of personnel, ship/shore Guard was In the process of limiting its approval rotation of officers, winter in-port personnel turnover, program to list only those industrial type chemicals that And costs were all factors in determining how the OSHA would be encountered aboard ship in bulk. NOAA was training requirement would be met. informed that commercial chemical products would be regulated by the Consumer Products Safety Commission The logistics of training shipboard personnel is such that (CPSC) and that commercial chemical products readily OMO holds one training session at each marine center available for home use would be acceptable for use as per year for key shipboard personnel. Due to the above ships' stores. factors, OMO realized that annual training Is required during the first several years to ensure coverage of We have found that the CPSC review process, the OSHA those key employees going from shore to ship standards, and the liability concerns associated with the assignments, and that annual refresher training is use of hazardous materials have had a self-regulating required for previously trained key personnel to ensure effect on chemical product manufacturers. "Non- the subsequent individual training of new employees as hazardous" alternatives are being developed and they report aboard ship. Specific policy concerning marketed. By using less hazardous ingredients in smaller training and other issues is discussed in the section that amounts, a chemical product manufacturer can produce follows. an effective less hazardous alternative, for example a cleaning compound formerly containing hydrofluoric acid IMPLEMENTATION may now use citric acid as the effective ingredient. Based on the surveys and review, of the regulations, Visiting Scientist OMO issued an instruction titled "Hazardous Materials and Hazardous Waste; policy, guidance, and training. ,6 The majority of pure chemicals found during the surveys The purpose of the Instruction is to promulgate the were left aboard ship by visiting scientific parties, and NOAA Fleet Hazardous Materials and Hazardous Waste to a certain degree, they posed the greatest potential Manual and to establish policy and training requirements hazard. This caused somewhat of a dilemma since OMO for procurement, storage, use, and disposal of hazardous needed to develop procedures for handling these materials and hazardous waste aboard all NOAA ships, materials without restricting the scientists from marine centers, and ship bases. In summary, this performing their projects. instruction requires that: In keeping with OSHA's Hazard Communication o Vessel commands and Facility Directors make every Standard, shipboard personnel are required to be effort to reduce quantities of hazardous materials, use apprised of the specific hazardous materials in the alternative materials wherever possible, inventory workplace, and since the chemicals brought aboard by hazardous materials quarterly, and obtain, maintain, and the visiting scientists vary from project to project and make accessible to all personnel MSDS's for all ship to ship, responsibility for providing information hazardous materials inventoried. about the chemicals lies with the scientists. In addition, based on RCRA requirements, responsibility for proper o Visiting scientific parties provide an inventory list deposition of the chemicals at the completion of a and MSDS's of all hazardous materials brought on board project also rests with the scientists and the NOAA ships to the vessel command. On departure from laboratories they are affiliated with. the ship, visiting scientific parties shall provide an Inventory of hazardous materials showing that all Disposal Procedures hazardous materials brought aboard have been properly used up or removed in suitable waste containers. Laboratory chemicals notwithstanding, shipboard generated hazardous waste such as used solvents, o Each marine center and ship base obtain an EPA degreasers, and antifreeze now must be accounted for Hazardous Waste Generator Identification Number for and handled in accordance with RCRA requirements. disposal of hazardous wastes at its facility. Unlike that of a shoreside production facility, shipboard Arrangements shall be made with a licensed hazardous generated hazardous waste is varied and noncontinuous. waste disposal contractor for routine removal of it is the accumulation of this varied and often hazardous waste from the facility and from NOAA ships unidentified waste stream which will pose a hazard and that arrive in port. NOAA ships will dispose of their be in violation of RCRA requirements. hazardous waste while in port. 1323 o Hazardous waste from outside facilities Is not to be Fleet Inspection Checklist - Lists the criteria that will accepted for storage or disposal at our facilities and be used by the NOAA Fleet Inspection Team, which spills of hazardous materials that threaten human health performs annual shipboard inspections for material or environment, or exceed quantities listed in 40 CFR condition and safety, to evaluate shipboard compliance 117.3, shall be immediately reported to the marine with the program. centers and the National Response Center (800-424- 8802). Emergency Response Procedures - Guidance on what to do in case of a spill. o Training be taken by all Commanding Officers, Disposal Guidance - Provides guidelines for disposal of Executive Officers, shipboard department heads, ship specific wastes from NOAA ships and ship bases. Lists base port captains, and marine center safety officers. alphabetically everything from aerosol cans to waste oil, Personnel receiving this training will be responsible for the preferred disposal method, and the waste hazard. the subsequent training, once a year, of all shipboard States several cardinal rules: avoid unknowns, segregate and shoreside personnel under their supervision who may nonhazardous from hazardous, if you buy it use it up, if be exposed to hazardous materials. you think it might be hazardous it probably Is, and avoid buying hazardous materials because their wastes are o Annual hazardous material training be scheduled and usually hazardous. Lists points of contact and telephone conducted by each marine center for all newly appointed numbers for waste disposal agencies and disposal personnel listed above and for those in need of refresher contractors at NOAA ship base locations. training. Records shall be kept of all shoreside and shipboard hazardous material training. Material Safety Data Sheets - Discusses information that can be found on material safety data sheets, how The 11HAZMAT Manual" to obtain them, and provides a location to keep them on file. The NOAA Fleet Hazardous Materials and Hazardous Waste Manual is the vehicle supporting the policy and Reportable Quantities - Lists the quantities of hazardous objectives explained in the preceding sections of this substances per 40 CFR 117.3 that if spilled need to be paper. It is at the center of the NOAA program and has reported to the National Response Center. been used successfully as a guide for responding to past and present problems and changes associated with Reference Library hazardous materials and hazardous waste management. In addition to the manual, the following reference books A description of each chapter and some of the highlights were distributed to the ships and ship bases. of the manual are discussed below. NIOSH/OSHA Pocket Guide to Chemical Hazards - General Questions - The manual defines a hazardous Presents condensed chemical and physical information on material as a product or material that is flammable, generic chemicals. reactive, toxic, or stored in a hazardous form unless It can be readily purchased by the public through a Clinical Toxicology of Commercial Products, Fifth commonly found commercial outlet and is stored and- Edition - Provides chemical, physical, toxicological, and used in the same way a typical household consumer proprietary information on commercial products. Useful would use it. The manual goes on to define and give in determining the constituents of a product and then examples of hazardous waste, the dangers hazardous determining its properties. Also provides a listing of materials and hazardous wastes pose, and where in the manufacturers of commercial products. manual more specific information on disposal guidance, emergency response, and reference material can be Hazardous Materials Injuries. A Handbook for Pre- found. This section answers additional questions Hospital Care - Provides first aid and toxicological concerning which materials are considered "safe" for use response information to help first responders treat aboard ship and their storage requirements. victims of hazardous material exposure. Sources of Information - Provides information, telephone Handbook of Poisoning, Eleventh Edition - Provides first numbers, and guidance on when and whom to call in the aid and toxicological information to help physicians or case of medical or environmental emergencies due to medical technicians treat victims of hazardous material personnel exposures or spills. The section mandates that exposure. spills exceeding the amounts listed in 40 CFR 117.3 or threatening human health or the environment be reported The Condensed Chemical Dictionary, Tenth Edition - to the National Response Center. Provides chemical and physical information on numerous chemicals. Transportation - Discusses Department of Transportation (DOT) requirements for manifesting, labeling, and 1984 Emergency Response Guidebook, DOT P 5800.3 - placarding of hazardous materials and hazardous waste Keyed to the Department of Transportation placarding during its transportation and provides a list with phone system, this reference provides information to help the numbers of qualified shippers. first responder mitigate a chemical spill. Human Health - This section is a comprehensive guide Glossary for Hazardous Materials, Hazardous Chemicals, for shipboard personnel covering the risk of exposure, Hazardous Substances, Hazardous Waste - A the effects, and the precautions associated with comprehensive glossary of terms and definitions. hazardous materials encountered aboard ship. 1324 Trainin In keeping with NOAA's mission as trustee of the Nation's oceans, the program satisfies the objectives of Training sessions are conducted in accordance with protecting the health and safety of personnel and the Reference 6 and OSHA's Hazard Communication environment from the risks associated with hazardous Standard to impart the policy, procedures, and materials and hazardous waste. information discussed throughout this paper. The training lasts approximately 2 hours per session and is REFERENCES Intended to Increase the hazardous material awareness of fleet personnel. Commanding Officers are instructed 1. Kummerlowe, David and Harris, Lori, NOAA Fleet concerning requirements of the program, namely: limit Hazardous Materials and Hazardous Waste Manual quantity and use; inventory quarterly; maintain MSDS's; December 1986, p. 1. Inform and instruct the crew; and dispose of wastes properly. 2. Brown, Michael A., 111984 RCRA Amendments (Non- UST),11 Environmental Laws & Regulations: 1986 Update RECENT DEVELOPMENTS Government Institutes, Inc., April 24-25, 1986, pp. 1-12 to 1-13. Several additional requirements and improvements in the program are being investigated based on feedback from 3. U.S. Environmental Protection Agency, "Small the fleet and ship bases. We are developing more Quantity Hazardous Waste Generators: Questions and specific guidance on the handling and disposal of Answers on the Final Requirements for 100 to 1000 radioactive materials by visiting scientific parties. We kilograms/month Generators," March 24, 1986. have obtained EPA Small Quantity Generator Numbers for NOAA ships to facilitate hazardous waste disposal at 4. Miller, Marshall Lee, "OSHA's Hazard Communication non-NOAA ports and are responding to recommendations Regulation," Environmental Laws & Regulations: 1986 that resulted from a recent OSHA inspection. Update, Government Institutes, Inc., April 24-25, 1986, pp. XII-1 to XII-8. Radioactive materials are strictly regulated by the Nuclear Regulatory Commission (NRC) and are the 5. Kummerlowe, David, "Update on Hazardous responsibility of the designated laboratory radiation Materials/Waste Management Program," January 24, 1986 safety officer licensed by the NRC to procure controlled (internal memorandum). radioactive sources. OMO has reviewed the NRC regulations as published in 10 CFR 19, 20 and 34 to 6. Office of Marine Operations, '1OMO Instruction 6280, ensure compliance by visiting scientific parties. Hazardous Materials and Hazardous Waste; policy, Interpretation of the regulations and a policy statement guidance, and training," May 13, 1987. concerning use of radioactive materials will be provided as an update to the HAZMAT manual. Several NOAA ships have attempted to dispose of hazardous materials while In port at non-NOAA facilities. Disposal contractors have refused to accept the shipboard generated waste without an EPA generator number. In accordance with EPA recommendations, OMO has obtained generator numbers for those ships requiring hazardous waste disposal at non-NOAA facilities. OSHA recently inspected NOAA's Pacific Marine Center for compliance with the Hazard Communication Standard and found that the NOAA program has been implemented properly. Procedures are being written to inform employees of the hazards of non-routine tasks and to apprise contractors of potential chemical hazards. CONCLUSION The Office of marine Operations, acting on behalf of the National Ocean Service within the National Oceanic and Atmospheric Administration, has developed and implemented a successful NOAA fleet hazardous materials and hazardous waste program. To briefly summarize, components of the program include: o the OMO Instruction based on applicable regulations, o the HAZMAT manual for use by fleet personnel, o the reference library containing additional technical Information, o a current MSDS inventory, and o annual awareness training. 1325 U.S. COAST GUARD OIL IDENTIFICATION SYSTEM CDR Larry H. Gibson and Dr. Martha S. Hendrick U.S. Coast Guard Central Oil Identification Laboratory Avery Point Groton, Connecticut, 06340-6096 ABSTRACT The Coast Guard Oil Identification Coast Guard System has been supporting f ield inves- personnel respond to spill tigators enforcing the requirements of It y s t e r yconduct Investigation the Federal Water Pollution Control Act -Take oil iamples- FF1._1d for more than ten years. At the heart Oil - -I of the Oil Identification System is the S P Oil Central Oil identification Laboratory Samples (COIL), the Coast Guard's and the United --am I I rce - District Office States' only forensic laboratory dedi- don Ifled [Administrative Hearing, cated to the analysis of petroleum Central Oil Settle samples for oil spill source identifica- dentification Lab Commandant tion. This paper gives a synopsis of Oil Sample (Administrative Appeal Hearing the development of the oil Identifica- Anal) tion System, an overview of the current Settle analytical techniques used to identify oil samples, and descriptions of several Department of Justice recent cases of special interest. F-plor prosecutIon FIGURE 1 INTRODUCTION U S Coast Guard Oil Identification System Flow Diagram The Coast Guard oil identification HISTORY System (OIS) , which is managed by COIL, serves as a powerful tool to aid field In 1972, the Federal Water Pollution investigators in determining the source Control Act (FWPCA) assigned general of "mystery" oil spills as mandated by responsibilities to the U.S. Coast Guard federal law. COIL uses several comple- for the protection of the marine envi- mentary chemical tests that exploit the ronment including enforcement of the unique, intrinsic properties of petrole- nation's anti-pollution discharge laws um oil and make it possible to match a and regulations in U.S. waters. To spilled oil with its correct source. carry out these responsibilities, the Thus, OIS provides the means to fix oil development of a system to identify pollution responsibility, assess penal- pollutant sources was necessary. The ties and help recover federal pollution Coast Guard Research and Development cleanup funds expended during an inci- Center was tasked with this project in dent. OIS serves as a deterrent to 1973. Over the next four years many deliberate oil pollution discharges and analytical tests and procedures were encourages the reporting of and accep- evaluated for their ability to distin- tance of responsibility for accidental guish among all types of petroleum oil. spills. Figure 1 is a flow diagram In 1977, the R&D Center published its depicting the operation of the OIS. final report detailing the "Oil Spill Identification System". I COIL was OIS methodologies are continuously established as the principal operating reviewed and updated to keep pace with arm to implement the system at the R&D changing technology. The information on Center facilities in Gr oton, Connecti- analytical techniques and laboratory cut. Formal operations began on Novem- statistics presented in this paper is ber 17, 1977. current through the end of 1987. 1326 United States Government work not protected by copyright One of the first steps for COIL and the crude oil field is readily distinguisha- new OIS was the setting of legal prece- ble from another, differences in the dent for it's floil fingerprinting" makeup of oils from the same crude oil technique. This occurred in December field can often be observed as well. 1978 at a federal criminal jury trial, Refined oils are fractions from crude under the Federal Water Pollution Con- oil stocks, usually derived by distilla- trol Act, involving spilled oil. In tion processes. Two refined oils of the this case, U.S. vs DISTLER, Judge same type differ because of dis- Charles M. Allen ruled that "chemical similarities in the characteristics of evidence" would be admissible thus their crude oil feed stocks as well as establishing the necessary precedent. variations in refinery processes and any subsequent contact with other oils mixed On April 1, 1979, administrative control in during transfer operations from of COIL was transferred to the Coast residues in tanks, ships, pipes, hoses, Guard Oceanographic Unit in Washington, etc. Thus, all petroleum oils, to some D.C. and plans were made to relocate the extent, have chemical compositions laboratory to that facility. A new different from each other. laboratory was constructed at the Wash- ington Navy Yard Annex from existing The Coast Guard OIS uses these unique, Oceanographic unit space. A brief shut intrinsic properties of petroleum oil down period from June through October which make it possible to match' a 1980 occurred to facilitate the .move, spilled oil with the correct source. set up equipment and train new 'person- The system is designed around five nel. COIL operations under the parent analytical methods: Thin-Layer Chroma- Oceanographic unit were short lived. tography, Fluorescence Spectroscopy, The Oceanographic Unit was closed for Infrared Spectroscopy, Gas Chromatogra- budgetary reasons in April of 1982. At phy, and High Performance Liquid Chroma- that time COIL became the fifth branch tography. Each of these analytical of the Port and Environmental Safety techniques measures different chemical Division, Office of Marine Environment properties of an oil sample and has been and Systems, Coast Guard Headquarters. shown to produce results independent of .the others. The results of analysis by In 1984, planning began again to relo- any of these methods can be presented in cate COIL to a less costly, more suppor- graph form. Data produced for spill tive location. The present location on samples and suspected source samples the University of Connecticut's Avery analyzed on the same instrument, on the Point Campus was selected and COIL moved same day, and under identical conditions into newly refurbished spaces during are visually compared. If two oil March of 1986. samples thus compared are chemically identical, they are said to be from a On October 1, 1986, COIL was established common source. In nearly every case, as an independent Headquarters Unit oils from other suspected sources will operating under the direction of Chief, be simultaneously eliminated from office of Marine Safety, Security and consideration as the pollutant source Environmental Protection. COIL is because they are chemically different manned by a Commanding officer, one as determined by the test methods. civilian chemist, one yeo"T n, and nine marine Science Technicians. CURRENT ANALYTICAL METHODS At the time of preparation of this The following sections provide an over- paper, yet another administrative reor- view of the current oil identification ganization is in progress. COIL will be methods used at COIL. A more technical merged with the Marine Fire and Safety description of the OIS is contained in Research Staff resulting in a new Head- Report CG-D-52-77 "Oil Spill Identifica- quarters Unit named Marine Safety Labo- tion System", Final Report, June 1977, ratories. U.S.C.G. R&D Center (reference 1). This document is available through the Na- WHAT IS OIL IDENTIFICATION? tional Technical Information Service, Springfield, VA 22161. Petroleum is a complex mixture of thou- sands of different organic compounds. Thin-Layer Chromatography (TLC). it is formed from a variety of organic This method separates the different materials, chemically converted under components of an oil on a silica gel differing geological conditions over plate and presents a colored spot pat- long periods of time. The infinitely tern for comparison. Methanol is used variable nature of these factors results to selectively extract components from in distinct chemical differences between an oil sample. Then extracts from the oils formed under dissimilar conditions spill and suspected sources are spotted and/or environments. While oil from one near one end of a pre-coated glass 1327 plate. The plate is then placed in a analyzing each sample IR continues to development chamber where solvents are be the most labor intensive technique used to separate the components along used at COIL. the length of the plate. After the separation has been completed, the plate Gas Chromatography (GC). This is viewed under ultraviolet light and method separates an oil's components .the colored spot patterns are visually primarily on the basis of boiling tem- compared. Permanent records of these perature. The separation of the compo- spot patterns are recorded with photo- nents is carried out under controlled graphs. conditions so that the same component will be eluted from the gas The TLC method has been changed very chromatographic column at the same little since its original concept de- relative point in time for all samples. scribed in reference 1. This procedure The separation is based on the parti- provides a rapid and easy to perform tioning of molecular species between an comparison of oil samples. However, inert carrier gas and a stationary because of problems encountered with liquid phase by repeated absorption- evaporation of light oil samples and the desorption. The components of the method's sensitivity to impurities, its injected sample are separated by passing current use has been restricted to cases the volatilized sample, in helium, involving heavy, viscous oils. through a fused silica capillary chromatographic column coated with a Fluorescence Spectroscopy (FL). stationary liquid phase of methyl sili- Spectroscopic methods rely on chemical cone. As the separated components are absorption and emission of light. The eluted from the column, they are sensed fluorescence method is based on the by a flame ionization detector (FID). emission of light in the ultraviolet Simultaneously, a quantitative measure- wavelength region. Aromatic compounds, ment of the components is recorded on a which may comprise only a small percent- strip chart, or chromatogram, which age of a petroleum oil, have the ability provides a record for comparison of to absorb ultraviolet light energy and different samples. then re-emit this energy over a number of different wavelengths. The re-emit- The GC method has probably benefited ted light is detected and recorded as a more from advances in technology than fluorescence spectrum. This spectrum any of the other OIS methods. The provides a pattern for comparing dif- introduction and widespread use of ferent oil samples. capillary chromatographic columns has occurred since the OIS was established. The efficiency of the FL method has been These have greatly enhanced the degree improved by the addition of new, state- of resolution obtained in GC separa- of-the-art instrumentation. The Perkin- tions. New generation robotic auto Elmer MPF-66 Fluorescence Spect- samplers provide reliable overnight .rophotometers, computer controlled operation of the instruments. instruments, currently in use at COIL Chromatographic data reconstructions are much less labor intensive and less using a computer data station and digi- susceptible to operator inconsistencies tal plotter have reduced the need for than previous instruments used. time consuming reanalysis of samples when closer examination of particular Infrared Spectroscopy (IR). The regions of an original chromatogram is infrared method looks at absorption of required. All of these advances have infrared light over a spectral region improved the effectiveness and efficien- which corresponds to the bond vibrations cy of the method. of the molecules that form the oil. The sample is scanned with a beam of light High Performance Liquid Chromatog- over the infrared frequency range and raphy (HPLC) . This method is used to the absorptions at each frequency are separate the components of a mixture on recorded on a chart which provides a the basis of polarity of the individual permanent record of the analytical components. it relies on the measurement. A number of absorptions partitioning of molecular species be- are common to all petroleum oils. These tween a carrier liquid and a stationary absorptions allow the analyst to identi- liquid phase by repeated adsorption- fy the sample as a petroleum product. desorption. A dilute solution of petro- Other absorptions are used for uniquely leum oil in acetonitrile and dichloro- identifying individual oil samples. methane is injected into a reverse phase liquid chromatographic column and devel- The IR method has also benefited from oped using acetonitrile as the eluent. improvements in analytical instrument The effluent of the column containing technology. However, since the sample the separated components is passed cells require careful cleaning after through both ultraviolet absorbance and 1328 fluorescence detectors. The detector reference oil is analyzed on each of the ,responses are simultaneously recorded on instruments and the results compared to a strip chart which provides a permanent the grior week's data to detect any record for comparison with other sam- changes in performance. As an addition- ples. al check on instrument reliability and oil sample preparation technique, a The performance of the HPLC method has duplicate aliquot of one sample in each been improved over the years by various case is independently analyzed for sample handling and filtration tech- consistency and reproducibility. These niques and by the addition of a state- and other internal checks monitor all ,of-the-art solvent handling system. procedures used in the laboratory. Who Spilled the Oil? The Oil sample preparation, testing, and chromatograms and spectrograms produced storage are conducted in accordance with by the methods discussed above represent the controlling American Society for different aspects of an oil sample's Testing and Materials (ASTM) concensus inherent characteristics. This standards or U.S. Coast Guard Standards. graphical data, or "fingerprint", for The Coast Guard actively participates in each spilled oil sample is qualitatively the work of ASTM Committee D-19 on Water compared to the data for each suspected and Subcommittee D-19.31 on Identifica- source sample in a particular case to tion of Waterborne Oils to establish new determine if.they are related. standards and update existing ones to reflect "state-of-the-art" practices in From the moment oil enters the environ- oil identification. Current appli 'cable ment, evaporation, dissolution, photo- ASTM Standards include D3415 - Identifi- chemical oxidation, biodegradation and cation of Waterborne Oils, D33.@6 - other forces begin to alter the oil's Preparation of Samples for Identifica- characteristics or "fingerprint". These tion of Waterborne Oils, D3650 - Compar- combined processes are termed weathering ison of Waterborne Petroleum Oils by and can significantly complicate the Fluorescence Analysis, D3328 - Compari- data interpretation. For these reasons son of Waterborne Petroleum Oils by Gas data interpretation in oil spill source Chromatography, D3414 - Comparison of identification is not straight forward. Waterborne Petroleum Oils by Infrared It is fundamentally different from that Spectroscopy, and D3325 - t reservation of ordinary chemical analyses. The of Waterborne oil Samples. Work is experienced oil spill analyst is famil- progressing on a new standard for iar with the complexities of the weath- Comparison of Waterborne Petroleum Oils ering processes and their impact on the by High Performance Liquid test methods. The analyst must be able Chromatography and updates of standards to distinguish real differences between D3328 and D3415 to address the latest two oils from those apparent differences improvements in technology. ASTM resulting from weathering alterations. approval of the analytical techniques Interference from contaminants can enhances their credibility and legal usually be recognized as such and dis- acceptance. counted when weighing the test results. However, at times, severe weathering CASES OF SPECIAL INTEREST and/or contamination can mask most of the inherent differences between similar Thousands of oil pollution investiga- type oils. In such cases, comparison of tions have been processed through the the test results may be inconclusive. Coast Guard OIS over its eleven year Only nine percent of the cases analyzed history of operations. Although OIS over the past four years have resulted results are designed primarily for use in indeterminant conclusions, and many in legal proceedings stemming from of these were due to the unavailability pollution incidents, on occasion, oil of a sufficient quantity of spilled oil spill source identification can be very sample for complete analysis. useful during the response phase of an incident. most analyses are complete in Laboratory Operations. A rigorous four days or less. In critical cases, a quality assurance program monitors both request for priority analysis usually instrument performance and sample prepa- results in a determination within 24 ration technique to ensure 3 results are hours. In addition to performing analy- accurate and reproducible. Personnel ses for Coast Guard pollution investiga- undergo extensive training programs at tions, as workload permits, COIL makes COIL and leading civilian institutions its expertise available to other feder- with expertise in the analytical method- al, state and local government agencies ologies used in the laboratory. Proce- for a variety of problems involving oil dures are continually monitored and identification. Several recent cases of labwork reviewed at several levels. At special interest are described below. the beginning of each week, a standard 1329 Was that really my oil? One note- the roadway and the wheelwell of the worthy case began with the grounding of wrecked car with samples from a nearby an oil tanker in the Deleware 5 River residence where the suspect in the case causing a major crude oil spill. The had recently changed the motor oil in vessel's owners initially accepted his automobile. The analysis linked the responsibility for the spill and began residue samples from the oil change cleanup efforts. But, when large a- operation to the samples from the mounts of oil appeared more than a mile wrecked car and roadway while simultane- upstream from the spill site, they ously eliminating samples collected from refused to accept responsibility for other possible sources in the area. The this "new" oil. In their view, the oil analysis results and testimony of the was not from their tank vessel. Time COIL analyst were presented to the Grand was critical and a decision to open the Jury for the ca'se to help secure an federal Pollution Fund to continue the indictment of .the defendant on cleanup operations was imminent. Oil manslaughter charges. : This was the samples from the river and the tank fifth homicide investigation assisted by vessel were collected by CG Captain of COIL. the Port personnel and flown to COIL for testing. Overnight analysis linked all Arson! In a different type of law of the spill samples to.the,ta'nker. The enforcement action, COIL assisted the ship's owners accepted the Coast Guard Connecticut State Fire marshal in the findings and completed the cleanup. As investigation of a suspected arson case. a result, an estimated three to four A store owner was suspected of trying to million dollar obligation from the torch his business establishment to Pollution Fund was avoided at a time collect insurance proceeds. An old fire when the fund was critically low. extinguisher filled with kerosene was used to spread the accellerant through- Which tank? In another case, a out the attic space of the store, and a mystery spill of heavy oil was discov- fire ignited. The fire was discovered. ered in the water adjacent to the load- very quickly and extinguished by the ing dock of an oil terminal in Boston. local fire department before the build- Just prior to the discovery, a tank ing was destroyed. The fire extinguish- vessel had completed off-loading a cargo er was recovered with some of the kero- of fuel oil to the facility and depart- sene still inside. COIL analysis ed. Coast Guard investigators took matched this kerosene with samples samples of the spill and samples from seized from the suspect's residence. the shore facility storage tanks. They also obtained composite samples of the Old Oil. Another interesting case tank vessel's cargo which are routinely analyzed by COIL involved a laid-up tank taken for customer quality assurance vessel in Portland, Oregon. In order to purposes. These samples were hand conduct some needed maintenance, the delivered to COIL for testing. Analysis vessel was ballasted down by the bow. of the samples revealed that the spilled An oil slick appeared in the water near oil, the cargo from the tank vessel, and the vessel. ' Coast' Guard investigators produc t from one of the shore facility collected samples of the spill, from the storage tanks were all the same oil. bunker tanks of the vessel, and from oil However, small but significant differ- trapped in a seachest which had been ences, probably introduced by residues exposed by the ballasting operations. in the shore facility tank, enabled COIL Analysis matched the spilled oil to the to distinguish between the facility oil trapped in the vessel's seachest and samples and the spilled oil. The spill indirectly linked it to the oil from the samples were identical to the tank vessel's bunker tanks. When MSO Port- vessel cargo samples. When the vessel land personnel received these results, arrived at its next port of call in they requested COIL to re-analyze sam- Houston, Coast Guard investigators ples in storage at COIL f rom a mystery obtained additional samples from the spill that had occurred in the same area vessel's tanks and forwarded them to a year before. These old spill samples COIL for comparison. This subsequent also matched the oil trapped in the analysis confirmed the result that the vessel's seachest, enabling the Coast mystery spill had originated from the Guard to seek reimbursement of the tank vessel. federal pollution funds expended on cleanup of the previous spill. Homicide? Responding to a com- pletely different situation, COIL as- sisted the District Attorney for Queens County, NY investigate a fatal automo- bile accident. The accident was caused by oil poured onto a roadway. COIL compared samples of oil collected from 1330 rt rr 5 o W- 0) 0(D rr P) w- D) (D C) 0 :00 0) 0 0 r. 0w- 0) 0 0 0 (D H 9) I-h H. 0PC :3 0 @-i w U) w D) 0rn t-h 0 (1 to 0) m ;q* (D r PO U) (D Z PO (DDrl i., ril 0 ft rt @- @< EG (1) H- ti 0,10 (D (D (n 0 0) (D (A 0) rr o o -B "@O (D 0m W. rt- (D F- (D0M"0 :3 0 0 -4 M r@ 0 (D CO tD :3 0,0 aj rt,071 03 F, 000rr P- (DWU)5 Ul 0 ti C@ 0 o-4 l< (D0(D0:) (Dc -4 ".0 Dj ko I'< Ea pi<t-h oj rr 01- th co (D P 0(D rr (a m 0 0 0U) t-h rt.0 m (D 0- " 0 ai 3rr rr :3 @r 0o ra (D PI (D OD En 0, :y (A ct VI (D W. rr ,, " m& n mmaj0 U) 1- -J (D tj0t-h (n0 W U, :r 0 (D 000Wt-h &< rt- 0Z rr (D 1j. 0mrr (n H 0 Cl (D 0rl- H.00 "L4 0 rt, 0H o H. rr (a w. (D tlHrra- W. H0) EQ =1 cr (DWWrt. OD rt.< :1 EnZ(n to 10 H :y r%) 0 (D @C r- lb (D 0 (D 0ul0w(D0 (D (D 01 0<:%"EO0" 0 w Ul (Dw z 0aF, EA (D tQMrr0(D CD 0 En CD CA) 0 0 Ja, rr En 0) 0rt. 10 tp (DH6 0 H (n B. @-0:C `0 (D (D 0 '0 (D rA (D t-h 0Zrr (D a En (D 1-0rt-X CL t-h Z W rr rr (D tn r, rt. M rr M I.. 0 rt, c 0(D En0 OD -h0 ::J (D rl- ril rt* (D (4wL< V @ (-n D' (D a' *(110m(D 0 M(D ftJ PO<0En k< M@ X Da OD rt, @e En9(D 10 rr 010w01 CY) @r :r 0 tr :E (D W- En 0pi (D"ct,0 (D ID rt' rr0 (DD0 0 <W@3"(DvCD rr (D M V V 0 rt 9D 0) (D Q@r W3(D Z (n ::r OD iol (D M (D CA (D M tO@ (D M tOLL w CA3 :2@ F- 0 LF1 0 Gi 41 Al W 0W SD W 0 - w w0. kDw.. E4D14 @m LO (D -j mr ol ID :3 -j W. -n P...0rr-0 C3 Z) p 0 0 a3> 21--" 00)0 LO %D CL H0I-w r- 0 N0) 0 Crt (D '-4 rr co 0) 0F-- w tn 0 %D-W. VOD 0-11 D 0mrn Go a1.4 rr rl cn 00 rt* Mrt* M ON En r4 " -0Hm-4 - .l<0ft00 Z0rt (n rr :%Dw- rn .CA) (D 0m F.Mq nrA0 W00000 rt @<Q OD C:"Z0 @or to1 0 F-ort %., -3HQM.tzj 0 Mt.000.00 ?v OD ctmCA 00 rr aVIF - 100rt* @j fm > (D < 0t-h (D co %0 (A-pqD0 -< " m LQ0Nz Icr rt, "m:@M M(D @w 0) 02 in (D 00 rt- to 0rn (Dn(D00H- 0 r- a @< 11h0 0) 0) (D F,0@ (A OD F-Jm@a CD -Pl F- Dw (D > po W. :3 OD Q(D tj0 0) 00M (40rr ISD> - OD tzo-cm Qj Vr4 @-i U) (D rtart, wC4 "i CF) >A)00 wc m t-Iwz0 00 (DI CD @4 F A COURSE IN HAZARDOUS MATERIALS LCDR T. J. Haas, LCDR J. J. Kichner and LT T. J. Chuba Department of Science United States Coast Guard Academy New London, CT 06320 Todays industries use materials unknown as The chemistry of hazardous materials at first recent as a decade or two ago: synthetic plastics and look seems to be as varied as the number of fibers, explosives, corrosive acids, pyrophoric chemicals one may encounter in industry. Upon liquids, combustible metals, toxic and radioactive closer examination however, the hazards can be materials, highly flammable petroleum fuels and sorted into three basic categories: chemical many more. Much of this material is transported reactivity, fire/explosion and health. These are in both large and small quantities over water. Since commonly referred to in the course as the "HAZ the Coast Guard is tasked with regulating the MAT BUGS." transportation of these materials in ships, an The introduction into chemical reactivity is understanding of the inherent dangers and proper given as a review of chemical terms and chemical handling of hazardous materials is of paramount families: acid/base, oxidizers, organic compounds importance to individuals charged with the enforcement of these regulations. The need to such as solvents and other petroleum products, and provide training for future officers of the Coast radioactivity. Each of these topics is discussed in Guard in the area of Hazardous Material class using examples to illustrate the chemical Transportation is being met at the Coast Guard principles. One area described in detail is water Academy. reactivity. Other types of chemical reactivity are also presented. These include self-reactivity (polymerization), air reactivity, reactivity with materials of construction and reactivity with other substances. These lectures prepare the students for exercises in cargo compatibility, which involves the proper stowage of hazardous materials for safe transportation. The students are provided general The course, "Hazardous Materials in Marine guidelines and a list of cargoes. Using this Transportation," is designed to give cadets the information they must complete a proper loading theoretical base'upon which their future decisions, and stowage plan for these materials on board a as officers, can be made regarding various aspects of hypothetical vessel. hazardous material transportation including containment and safety systems, research and The second major element for examination development, environmental protection, and is fire/explosion hazards. An unwanted by- domestic and international regulations. Three product, in some instances, of chemical reactivities major areas - Chemical Hazards, Marine and improper stowage of non-compatible Transportation and Environmental Response are chemicals, is fire and/or explosion. The theory discussed in detail. Lectures are presented behind both fire and explosion is presented. addressing specific topics within these areas. Guest General characteristics and hazards of fires lectures, audiovisual materials, demonstrations involving solids, liquids and gases are extensively and field trips are employed to enhance the covered. The burning characteristics and hazards of understanding of course content. Additionally, flammable metals are also briefly mentioned. during a drill based upon a realistic spill response Thermal transport by conduction, convection and scenario, the cadets must plan and execute the radiation is introduced. The student is exposed to proper response procedures learned in class. This the concept of the fire tetrahedron and the allows the cadets to apply the course material in a interrelationship of the components that make up practical, real-world situation. this model. CH25185-8/88/0000- 1332 $1 @1988 IEEE These concepts are then used for applying codification of these laws and general organization extinguishing theory to the different classes of fires. of the overall applicable regulations. The content Best methods of extinguishment and use of of each part of the regulation is presented briefly in different extinguishing agents are discussed, as they order to provide the student with some reference apply to different situations involving the fires of point in the event of future involvement. Three liquids, gases and solids, both normal and metal. areas of particular interest to the Coast Guard, Movies and slides are utilized to supplement the namely the packaged, bulk and pollution course lecture. regulations, are covered in more detail. The last of the "Haz. Mats Bugs" to be Under the "packaged" hazardous material discussed is health. A large amount of time is regulations, the problems involved in offering a spend describing toxicity. Acute and chronic hazardous material for shipment through toxicities are defined and examples of each are classifying, packaging, marking, labeling/ placarding given. The measurement of toxicities using and describing the material on shipping papers, are parameters such as LD 50 (Lethal dose that kills 50% extensively covered. Use of the Hazardous Material of a test population) and Draize testing (skin Table in the regulations is introduced. A irritancy/ corrosion) are discussed. Transportation comparison of International and U. S. Regulations health guidelines such as the different is made and an assignment is given requiring the classifications of poisons (Poison A, B, etc.) are also students to properly prepare a hazardous described. After a general introduction, the focus is commodity for shipment in compliance with the turned to the toxicity of certain substances. As an applicable regulations. example, solvents and pesticides are studied for their acute toxic action as well as the chronic effects In covering the "bulk" transportation of that one may experience with an exposure over a hazardous materials, time is spent on the general long period of time in fairly low concentrations. design of bulk carriers - petroleum Benzene and other substances are used to illustrate tankers,chemical ta 'nkers, and liquified gas tankers. the differences in the toxicities. Specific design areas covered include cargo containment, vessel arrangement, cargo transfer The course then turns to the specific area of systems, safety equipment and any "special occupational health (industrial hygiene). hazards" associated with the transportation of cargo Threshold limit values (TLVs) are related to both on each type of vessel. In these areas, concepts from qualitative and quantitative fashions. Controls the first half of the semester are related to actual used to minimize health risks from exposure to engineering practices applied in vessel construction chemical substances are also described. to minimize the "Haz Mat Bugs." Reference to Engineering, administrative and personal applicable regulations in the construction and protective measures are prioritized for the student. operation of these vessels is made. Unique cargo hazards as specifically outlined in the regulations Lectures dealing with the chemistry of are also related. Comparisons between petroleum, hazardous materials utilize approximately one-half chemical and liquified gas tankers pointing out the of the semester. Once a student is familiar with necessary "high technology" sophistication are these concepts the course shifts to the problems made. Photographs and blueprints of vessels associated with transportation of hazardous showing the various configurations are utilized in materials. the classroom. Field trips to chernical, petroleum and gas tankships, as available, are scheduled to The Coast Guard is the primary government enforce classroom- learned concepts. Literature agency concerned with the transportation of supplied by the chemical industry and movies are hazardous material over water. This transportation combined in the lectures to supplement and occurs in two separate and distinct modes; bulk and reinforce the material presented. packaged. Inevitably associated with hazardous materials, are the laws and regulations that govern An overview of the Coast Guard's role in their transportation. Although not a "regulations Commercial Vessel Safety, Marine, Environmental intensive" course, some knowledge of regulations Protection and Port Safety and Security is presented. is required to obtain a better overall understanding Coast Guard Organization from the Headquarters of the transportation of hazardous materials. Laws level to field Marine Safety Offices is outlined. The and Treaties empowering the Coast Guard with the Coast Guard responsibility and role in hazardous regulation of hazardous material transportation are materials spills is covered. Applicable, introduced. Time is devoted to defining the "responsible" laws "are introduced. 1333 As an available tool for assigning hazardous planning, writing, knowledge/use of material mishaps the students are introduced to the engineering/ scientific/ technical information that Coast Guard CHRIS (Chemical Hazard Response results in accomplishment of "real world" Coast Information System) system. The use of this Guard mission - something the cadets will be system for hazard assessment, personnel protection required to do many times in the future as Coast and containment, treatment and recovery, is Guard officers. stressed. Briefly introduced is HACS (Hazardous Assessment Computer System) and its potential use in spill situations. Spill scenarios are then presented and students utilize the information discussed to determine appropriate organization and remedial action. The topic of hazardous waste management is covered briefly. Different methods of waste disposal from landfills to "state of the art" plasma incineration are introduced. Coast Guard involvement in the approval of vessels for the incineration of waste at sea is also discussed. In order to pull the varied course material together as the semester nears completion, a chemical spill drill is conducted by the students and members of the Coast Guard Strike Team. Each .member of the class is assigned a position on the spill response team ranging from On-Scene Commander to decontamination team member. Department of Science faculty and the local media are invited to play various roles during the exercise to increase the realism. The students are required to begin organizing two weeks before the drill and develop a general contingency plan. On the day of the drill, Strike Team members brief the students concerning the material spilled, weather conditions, and available materials. The students then take over and make all safety, clean up, and public relations decisions reacting to information provided by the instructor and role players. The entire event is videotaped for an in-class critique. This exercise provides the students with an opportunity to utilize their knowledge, don equipment, and think on their feet just as. a response team would in the field. The "Hazardous Materials in Marine Transportation" course is, in essence, a "capstone" course for those students majoring in Marine or Applied Sciencs and is also of great importance to .naval and other engineering disciplines., Management majors find the course a useful and practical technical elective. It's a contemporary and practical course for the technically oriented 'individual. At the same time, however, certain management principles are strategically introduced to.make the course complete. Cadet evaluation indicates that the course is well received. We believe it's a practical course involving research, 1334 AN ARGOS METEOROLOGICAL OCEANOGRAPHIC SPAR BUOY FOR ANTARCTIC DEPLOYMENTS Samuel P. Burke Douglas G. Martinson Polar Research Laboratory, Inc. Lamont-Doherty Geological Observatory Carpinteria, California and Department of Geological Sciences Columbia University Palisades, New York ABSTRACT structure of the water column. He was able to detect the passage of an eddy based on data from a A specially configured spar buoy with a 150 pair of depth 'sensors on the cable. Sensor data meter kevlar multielement sensor string has been is sampled every hour and sent out over a designed for deployment in the Antarctic Ice telemetry link to the ARGOS data collection system Fields. The buoy collects a full suite of on board the currently active US NOAA Polar .meteorological data combined with hourly samples Orbiting Satellites. from the oceanographic sensor tether. Water column sensors consist of ten medium resolution Buoy development for three drifters was multiplexed temperature and pressure sensors along started in the fall of 1985 with the first with six pairs of high resolution conductivity and deployment during the Antarctic winter of 1986 temperature sensors to provide an estimate of the followed shortly afterwards with loss of the buoy salinity structure below the buoy. Data relay is during a storm. Following the short-lived carried out using the ARGOS capability of the two deployment of the first AMOS Buoy, the remaining orbiting NOAA Satellites. A pair of developmental two drifters were returned to PRL for extensive buoys were deployed this past winter in the open s.!nsor recalibration and long term burn-in waters of the Weddell Sea, Antarctica from the [email protected]. During the spring of 1988 the two German Oceanographic Research Vessel POLARSTERN. ii.maining AMOS Buoys were deployed into the open Recent data shows the rate of mixed layer cooling ocean approximately 480 nautical miles north of prior to freeze-up. Dronning Maud Land Antarctica from the the ice strengthened German research vessel, POLARSTERN. The results of these recent deployments will be INTRODUCTION reviewed. The Antarctic Meteorological Oceanographic SCIENTIFIC OBJECTIVES Spar Buoy or the AMOS Buoy was developed by Polar Research Laboratory for Lamont-Doherty Geological The Southern Ocean has long been recognized Observatory as a means of collecting environmental as a.major component in the ocean/atmosphere data from the Antarctic Seas. The buoy design system. Oceanographically, the region serves as borrows design concepts previously used in PRL's the dominant area for deep and bottom water SALARGOSO) Arctic buoy and the open ocean ASID(2) formation and deep ocean ventilation. buoy. The Meteorological measurements consist of Atmospherically, the region serves as an important Air Temperature, Barometric Pressure and Wind climate modulator through the insulating effect of Speed. The Oceanographic measurements consist of the extensive seasonal sea-ice cover. This sea- Temperature and Conductivity along a 155 meter ice, in turn, is intimately coupled to the electromechanical cable which hangs below the 4.5 ocean/atmosphere system. Its temporal and spatial meter long spar buoy hull. The underwater sensors distribution is determined by the nature of the are spaced along the cable at 10 meter intervals atmospheric forcing and the ocean's density with four of the six C/T pairs positioned in the structure. The atmospheric forcing though, is first 35 meters of the water column and the ten itself modified by the presence of sea-ice, while low resolution temperature sensor modules the ocean's density structure is modified by the positioned from 45 meters down. The three cable growth and decay of the ice, which in turn, depth sensors are positioned to break the cable affects the magnitude of the deep ocean into thirds so that cable catenary shape can be ventilation and deep water formation. The determined thereby making it possible to make air/sea/ice system is thus highly interactive and depth corrections to the individual sensors. While understanding of this interaction is necessary to using an open ocean PRL drifting buoy with a 300 determine the nature of such important phenomena meter string Picket(3) demonstrated that pressure as deep water formation, oceanic heat ventilation, sensors on subsurface cables have the potential to sea-ice distribution and climate modification. provide information on the vertical current CH2585-8188/0000- 1335 $1 91988 IEEE To obtain this requisite understanding tapered' section of polyurethane rubber called a requires extensive modeling efforts and , the carrot is used to reduce bending stress on the 155 collection and analysis of the appropriate data. meter ocean sensor cable where it attaches to the To date, however, the models are dependent upon base of the buoy. The cable conductors enter the numerous parameters and forcing functions of buoy hull through a compression seal at the base. unknown character and the data necessary to refine Sensors are positioned along the cable with these unknowns are virtually nonexistent. As a conductor breakouts every 10 meters as illustrated consequence the models are overly simplified, in Figure 1. The sensor fairings act to protect highly tuned and lack the observational framework the delicate sensor modules during handling and at against which to gauge their success. Even the deployment. Both the high resolution basic data required to test simple conceptual conductivity/temperature and low resolution ideas are lacking due to the inherent difficulties multiplexed faired modules are shown in the photo of collecting data in the Antarctic sea-ice field. of Figure 2 taken during an ice bath test. Underwater connectors are spliced to the cable In the austral winter of 1986, POLARSTERN triples at the breakout points. made an expedition to the Atlantic sector of the Southern Ocean. There, the vessel sailed through the sea-ice field along the eastern margin of the DEPTH SCALE Weddell Sea to the Antarctic continent and back out over the winter season. The expedition, the POLYURETHENE 0- first of its kind in the winter Antarctic, was FAIRING CTI 5 - multidisciplinary in nature while focused on TEMP SENSOR HIGH RESOLUTION obtaining those data necessary to advance our TEMPERATURE/ CT2 15- general understanding of the air/sea/ice system. CONDUCTIVITY SENSOR - ICT3 25- CELL-. - A specific component of that program was the SHROUD CT4 - deployment of three prototype CONouc-nATy TI d 45- meteorological/oceanographic salargos type SENSOR 50- drifters (AMOS Buoys) capable of operation over UNDERWATER TZ - the full range of conditions from open water to CONNECTOR fully formed sea ice. The drifters, described in T3 detail below, were designed to provide time series - data (as opposed to spatial series provided by CT5 75- ship measurements). These time series would T4 reveal the nature of the seasonally varying TEM atmospheric forcing and upper ocean CAP ENSOR T5 d (thermal/saline) response to this forcing as the Ta 100- MORRISON system evolves through an annual cycle. These SEAL T7 data would thus provide fundamental observational evidence pertaining to the interactive nature of SERIAL MULTIPLEXED TS the air/sea/ice system while providing a PC BOARD SENSOR MODULE T9 125 foundation against which to test, evaluate and T10d improve the existing models. PRESSURE TECHNICAL DETAILS HOUSING CTS 150 PROTECTIVE The AMOS Buoy uses a standard 8 inch diameter FAIRING 1/8 inch wall Arctic aluminum type hull to house .3 WRE BUSS d. INDICATES TEPTH SENSOR the electronics and batteries. This hull is encased with a 12 3/4 inch diameter 0.3 inch wall glass fiber sheath with the interspace filled with Figure 1. AMOS Buoy Sensor String Schematic. polyurethane flotation foam. The length of the hull is 15 feet. The sheath was used in an The 155 meter sensor cable contains 13 attempt to protect the thin walled aluminum twisted triads and 5 twisted pairs of number 24 cylinder from blocks of sea ice during freeze up stranded (19/36) wires with low density and during ice breakup. The buoy 'electronics are polyethylene insulation. A kevlar braid passes housed within an internal rack which slides into over a low density rod filler. The conductors are the hull. Fourteen alkaline battery packs are wrapped around the kevlar with a protective layer located at the base of the rack. A 63 inch long of mylar tape between. A 0.10 inch polyurethane ARGOS antenna housing is positioned at the top of jacket is added for protection of the conductors the hull. A special barometric pressure port is from rafting ice below the buoy, the cable is free situated inside the housing. The port includes a flooding. The overall diameter is 0.82 inches. special water trap to prevent water from getting to the pressure sensor during periods of complete IVE buoy submersion. The air temperature ventilator and the ruggedized wind speed sensor are located at the top of the antenna housing. A heavy 1336 7! , tt" A0, T- T-.- M Y 44@ - F-Ift 3a ft wa WODLU wwr@ C1 7A T2 co-WRwrr is T4 UM REX@ r VIC Go lie- A Figure 2. Sensor Fairing Clusters. t The measurements made by the AMOS Buoy are we listed in the table below showing ranges covered, resolutions, averaging period and accuracy. Figure 3. AMOS Buoy Functional Block Diagram. MEASUREMENT RANGE RESOLUTION AVERAGING ACCURACY The sensors are sampled once every hour Barometric starting at the buoy reset time. Two ARGOS ;'ressu re 950 - 1050 MB 0.1 MB 50 sec 1.0 mB transmission groups, each with a full complement Air and Nutt of eight frames, are required to accommodate a one Temperature -21 to +29*C 0.2*C 10 sec 1.0*C hour sample. Three hours of data are held in a buffer consisting of six separate ARGOS groups. Wind Speed 4 to 51 m/s 0.2 m/s 0.2 sec 1 m/s or These data are sent out over a six minute period sea Subsurface 10% and then repeated. The ARGOS data format is Temperature outlined below. Low Resolution -2.0 to 3.12*C 0.02*C 20 sec 0.1% High Resolution -2.0 to 25% 0.001% 10 sec 0.03*C FRAME WORD NUMBER 1 2 3 4 cable 0epth 1 Barometer Battery Air Temp, Hull Temp DI 35.25 to 48.0 m 0.05 m 0.1 sec 0.2 m 2 Blank Wind Spd Blank Heading D5 74.5 to 100.0 m 0.1 m 0.1 sec 0.5 m D10: 16.5 to 152 m 0.14 m 0.1 sec 0.7 m 3 ClH ClL C2H C21, 4 C3H C31, C4H C41, Conductivity 14 to 59 mmho/cm .0003mm/cm 0.1 sec .0003mm/cm 5 C5H C51. C6H C61, Battery Voltage 12.8 to 19.1 vdc 0.1 vdc 0.1 sec 0.1 vdc 6 TlH TlL T2H T2L 7 T3H T31. T4H T41. 8 T5H T51. T6H T61, Irf.-HD-- 1337 1 Barometer Battery Air Temp Hull Temp A sun shield and air ventilator is located at 2 Blank Wind Spd Blank Heading the top of the ARGOS antenna 2.5 meters from the 3 TI T2 T3 T4 buoy water line. A Yellow Springs Instruments 4 T5 T6 T7 T8 thermistor network is located within this housing 5 T9 T10 Dl D5 to make the air temperature measurement. The 6 D10 BLANK BLANK SAMP sensor and the associated electronic signal 7 SAMP SAMP SAMP SAMP conditioning have a combined accuracy of better 8 SAMP SAMP SAMP SAMP than -+0.2*C.However, errors associated with solar heating effects on the ventilator add to the total The barometer words are ten bits to the system error accounting for an overall accuracy of accommodate the 100 millibar range with the 0.1 �1.0*C. The same sensor and signal conditioning millibar resolution. The battery word is six bits technique are used to make the ice/water and the remaining words are all eight bits. The temperature measurement except that the sensor is sixteen bit high resolution conductivity and located on a spring loaded holder in contact with temperature data are indicated as ClH and ClL the inside surface of the hull one meter below the where the high and low bytes are indicated by the buoy water line. Data from this sensor is also letters H and L. The low resolution temperature used to correct the barometric pressure sensor for data words are indicated as Tl through T10 and the temperature effects. depth data is indicated as DI, D5 and D10. SAMP indicates the hour that the sample was taken. The wind speed sensor is located above the air temperature ventilator. A three bladed The system block diagram of Figure 3 defines savonius rotor turns on a sapphire jewel bearing the major functional blocks making up the AMOS under the influence of the wind. The rotation buoy electronics. A card cage holds the f ive rate of the rotor is measured using a magnetic system printed circuit boards as shownin the photo pickup to sense the passage of six small ferrous of Figure 4. Each of the boards communicate over slugs at its base. The rotor is 2-1/4 inches in a common 50 conductor buss. The boards consist of diameter and is housed in a 4 inch diameter an encoder/processor, an analog to digital housing. This sensor represents many years of converter, a power conditioner, a serial multiplex development to provide a robust, yet relatively module interface and the 16 bit multichannel low threshold device. The approximate transfer period counter. The sensor sampling is carried function of the sensor is WS - 0.7 + 0.05 * F out under control by an RCA Cosmac 1806 8 bit where WS is the wind speed in meters per second microprocessor while executing a controller and F is the frequency in hertz. Sensor program written in assembly language. calibrations are carried out using the wind tunnel at the University of Southern California at Los Angeles. The measurement is made by accumulating the electrical pulses from the sensor in an eight bit counter. The eight bit low resolution temperature and t -N depth sensors are individually multiplexed on one of two three wire data busses. The design is such that up to 64 modules can communicate on the same buss, in this case each buss holds five modules ]L 5@ thus providing isolation so that a partial cable failure or a flooded unit will not affect the operations of the second. Each sensor module is A interrogated with two, 11 bit serial data words. WMN One of the data words contains the address, the other is a command word. When the address matches that of the programmed module address the module samples the sensor data and transmits it back up the line in two, 11 bit data words. Each 11 bit word consists of eight data bits , a parity bit, a start bit and a stop bit. The multiplexed signal output from the sensors is delivered to the buoy processor via the interface. A simplified block Figure 4. AMOS Buoy Electronics. diagram of the module electronics is shown in Figure,5. The major components consist of an 8 Barometric pressure is measured using a bit analog to digital converter, a power Paroscientific Digiquartz transducer which regulator, analog signal conditioner, line produces a frequency output which varies with isolation components, a pressure sensor and a atmospheric pressure as described by a series of temperature sensor. In its normal operating mode, third order equations. The frequency is measured each module's Asynchronous Receiver Transmitter using a 50 second gate period to open a 16 bit (ART) processes all serial data appearing on the accumulator to the input pulse train from the cable pair. When the identification code word sensor. matching a module's preprogrammed code word 1338 appears on the cable pair, the module's ART oscillators were suspect so they were sent to the accepts the next 8 bit word as a command word. Northwest Regional Calibration Center in Bellevue, Washington for recalibration. As a group the twelve sensors were placed in large temperature baths. At each calibration point, the bath TO BUSS - - - - - - - - - - - POME. Co. POWER 10 SEMAL DATA 01 UART OSC LwE TEMPERATURE DFJVM CONIPOL DATA N ANALOG A/D C014WERTER CONWTL@HER L- - - - - - - - - - - - - - - -- J DEPTH SENSOR 3 WK Buss TO OTHER MODULES 64 MAL Figure 5. Three Wire Multiplexed Sensor Module Block Diagram. After the command word reception, the ART signals the A/D to process the temperature or depth signal and return the resulting 8 bit data to the ART. Upon receipt of the end of conversion pulse from the A/D, the ART sends two 11 bit words back up Figure 6. AMOS Buoy 6440 Deployed in 40 cm of Ice the cable pair. The digital words appearing on the at 62*S 1*W on 28 July 1986. cable pair are formatted so that no responding module could cause another module to respond. Primary power is supplied from a bank of + + + +61.5. S fourteen each 18 volt alkaline battery packs each with a capacity of 18.4 amp-hours. The average current drain is 23.5 milliamps with 62% of the load due to the processor and 33% due to the ARGOS + + +62-S transmitter which uses 500 milliamps for 0.92 Dy 205 Np%V@t- D.V 201 seconds every minute, the remaining 5% of the R.1 V@Wim - 0" 220 power consumption is attributed to sensor power and sampling. With this load the battery pack D" 213 will provide power for the buoy for 456 days. + + + +62.3-S 2.0- V 1.0. 0.0. 1.01 E DEPLOYMENT The first prototype drifter was deployed on Figure 7. Drift Track of AMOS Buoy 6440. July. 28, 1986 at approximately 62 S and I W in newly consolidated pancake ice 40 cm thick as temperature was stabilized to within one shown in the photo of Figure 6. Unfortunately, millidegree Centigrade and the temperature was this first prototype proved defective and on board measured using a Standard Platinum Resistance testing revealed the problems to be common with Thermometer. The output frequency from each the remaining two drifters which were returned to sensor was read 12 times giving a mean which was PRL for repair and later deployment. The used to arrive at that calibration point. A total defective drifter was tracked for a total of 19 of seven different points were taken in the days before it stopped transmitting during a process to make a least squares fit to calculate period of heavy weather. The 19 day drift is shown coefficients to a third order equation. The in th'e plot of Figure 7. Later testing of the quality of the calibration was verified by remaining two drifters suggested that the failure calculating the temperature using the equation and was due to leakage through the pressure sensor taking the difference between the measured and housings via faulty connector seals. The sensors calculated terperatures, the resulting average were returned to the manufacturer and refurbished. standard deviation was .003*C,. The calibrations on the 16 bit temperature sensor + 1339 The refurbished drifters were deployed from ACKNOWLEDGMENTS the POIARSTERN on March 5, 1988 at about 62 S, 0 W within approximately two nautical miles of one This project benefited from: Rosemary Xacedo another under open ocean conditions. The close who oversaw the 1988 drifter deployment; the chief proximity of the two drifters allows replication scientist, Dr. D. Fuetterer who graciously of data and provides a consistency check on sensor accommodated the drifter program; and from the readings. Over the first month of transmitted excellent assistance of the crew and Captain of data, the temperature data show excellent the POLARSTERN. The extra effort given by Kevin consistency. Later data has not yet been Brooks and Noley Baker of Polar Research processed in one drifter due to recently resolved Laboratory and Miguel Maccio of Lamont-Doherty problems with Service Argos magnetic tape during development and testing of the drifters is distribution. Conductivity/temperature/depth gratefully acknowledged. measurements made from the ship at the time of deployment were unreliable so inter-sensor This research was supported by National comparisons were not possible. Temperature Science Foundation Division of Polar Programs sensors all tested within specifications on the grant DPP 85-01976. ship just before deployment. Both drifters experienced significant.sensor dropout shortly after a smooth deployment into calm waters for reasons not yet diagnosed. It is expected, however, that improved string design techniques developed by PRL over the past year, E which make the in-water system much more rugged, 9! will result in greatly improved reliability. This is currently being successfully verified in sea tests on other programs. In one drifter, all high resolution/conductivity sensors failed as did 5 of the 10 low resolution temperature sensors. In the 41. is to too 11. 11. ta. - other drifter, 4 of the 6 temperature conductivity 0"Sovow pairs failed as did 1 low resolution temperature Figure B. Temperature measurements from sensors at 45 and 65 sensor. No additional failures were experienced meter depth. The shallower sensor is in the warm subsequent to those occurring at deployment. In summer mixed Layer and the deeper one in the temperature minimum layer. The water at the 65 general, the meteorological data, low resolution meter depth is entrained into the mixed layer by day temperature and depth sensor data return is 104 during the fail cooling period. this is excellent. The drifters are presently locked in revealed by the convergence of temperatures and the decrease in magnitude and frequency of the 65 meter the sea-ice field which developed in late June. temperature fluctuations. This reflects the increased thermal inertia of the mixed layer as its CONCLUSIONS volume increases. A substantial amount of information concerning the upper ocean evolution is obtained from the temperature data. Interesting observations collected thus far reveal the rate of mixed layer cooling and expansion prior to ice onset as shown in the temperature time history plot of Figure 8. Data indicate that there is a strong coherency of temperature fluctuations across the entire pycnocline, possibly indicating passage of a warm core eddy through the region as to 1. to suggested in Figure 9. The data reveal the D";'Y- easurements from the Figure 9. Comparison of the tempe ature m linearity of the thermocline as it increases from two drifters, located approximately 2 nautical mites the freezing point in the mixed layer to apart, at the 105 meter depth level. Note that the approximately .4*C at the 155 meter depth near the sharp temperature increase is probably due to the passage of a warm core eddy through the region from temperature maximum. The temperature sensors show days 68 to 78. slight, but consistent, offsets from the absolute temperature (approximately .32*C; this amount has REFERENCES been added to the temperatures in Figures 8 and 9). This is corrected by comparison to the 1. Burke, Samuel P. and Morison, James, "An improved SALARGOS Buoy for deployments in Polar Seas01 oceans 87 Conference, functioning high resolution sensors in the mixed September 1987, Halifax, Nova Scotia, Canada, pp 49 - 53 layer. If possible, the sensors will be recalibrated upon recovery during the austral 2. Kozak, R. and Anderson, J., "Overview of the NOAA Date Buoy Center Drifing Buoy Development Programs" NOAA NDBC summer of 1989. Report WP219, ARGOS Users Conference May 21, 1984, Seattle, Washington 3. Picket, R. L. and Lee, J., "Estimating Vertical Current structure From Sets ttite-Tracked Drifters- Marine Technology Society Journal Volume 18 Number 4, pq 4 - 8 1340 METHODS OF OBTAINING WEATHER DATA IN REAL TIME David B. Gilhousen National Data Buoy Center (NDBC) Stennis Space Center, MS 39529-6000 NY 5 ABSTRA CT CATE- CANA A 0 7 N.S 0 INEIEL8 - @ N Small craft owners who have access to a personal computer now have a OTHER AGENCIES wls@ - 0 DEVELOPNIENTAL NDBCk 0 several methods of obtaining real-time meteorological data before ven- 2@ MICH STATUS turing out to sea. These data include high quality NDBC buoy and coastal & ES7AWSH,5D marine observations, cooperative ship observations, coastal marine 0 PLANNED -V forecasts, wind forecasts for coastal marine locations, and graphical 0 weather maps. These observations andforecasts are typically not available on a timely and continuous basis via radio or television. They are available 0 by dialing into a variety of on-line meteorological data bases or receiving @0. the data directly from a commercial satellite. Strengths and weaknesses 0 of these methods are discussed. 0 11. 0 0. 0 1. INTRODUCTION Mariners who have personal computers can obtain a variety of weather observations and forecasts which will help them decide whether they should venture out to sea. Unfortunately, obtaining this data will usual- ly cost them money. So, before describing these data access methods, we Figure 1. NDBC Buoy Locations should first ask the question: "Why use a PC when you can get weather information via NOAA Weather Radio or the Weather Channel?" become invalid. Because they are available hourly, they provide a good indication of deteriorating weather conditions. They also give an indica- There are several good answers. First, some data are'not available via tion of how well the marine forecasts are verifying. One disadvantage these traditional sources. For example, water temperatures from the of the data is that they are usually available only in coded form. However, National Data Buoy Center (NDBC) buoys and tide levels from coastal with a little effort, most people can mentally decode an observation quick- stations are typically not reported. These observations may be very im- ly. Information on station identification and codes is available from portant to fishermen. Second, because of manpower constraints, NOAA NDBC's Data Systems Division at (601) 688-2836. Weather Radio reports are updated about once an hour, except under Another source of observations consists of reports from coastal airports emergency conditions. Because buoy observations arrive later than the regular airport observations, it is not uncommon to have observations and other hourly observing stations near the coast. These stations main- I to 2 hours old being broadcast via the Weather Radio. Observations tain a continuous weather watch and quickly transmit a report when are available in a much more timely manner via a PC, typically arriving thundershowers are in the area, or the ceiling and visibility measurements within 5 minutes of issuance. In bad weather, timeliness can be important. quickly change. The primary advantage of these data are their visibility and current weather information. A few stations report tide data. The Now that I've covered the "Why," I'll address the "What" and the biggest disadvantage is that the winds often don't represent the winds "How." Namely I'll discuss what products are available that would in- reported in the marine environment. They are frequently lighter than terest the mariner and then how to access them. marine winds because of land-induced turbulence. A third source of observations is cooperative ship data. These are reports 2. THE WEATHER PRODUCTS from small craft operators or commercial fisherman who radio the in- formation back to shore. Confidence in the quality of these observations 2.1 Observations is generally low, however they can provide a good indicator of weather tendency. For example, the absolute accuracy of a wave height report As you might guess from my affiliation, I'm "high" on observations pro- may not be that great due to human estimation, but the observation that duced by NDBC. These observations are transmitted hourly from about seas are building rapidly is invaluable. A big disadvantage is that the obser- 55 moored buoys and 40 Coastal-Marine Automated Network (C-MAN) vations are made infrequently, usually after everybody has reached the stations. The locations of these stations are shown in Figures I and 2. fishing grounds! Also, more observations are available during good Many of the stations are located within 20 miles of the coast. All stations weather. report winds, sea level pressure, and air temperature. All buoys and a 2.2 Forecasts few C-MAN stations report significant wave height, dominant wave period, and water temperature. A few C-MAN stations also report tide In addition to observations, various forecast products are also available. data. An overview of NDBC data is given by Hamilton[I ]. The network Obviously, the most important is the coastal marine forecast for your of stations is slowly expanding and NDBC is expecting to be able to local area issued by the National Weather Service (NWS). This provides measure relative humidity at a number of stations in the near future. the expected wind and wave conditions for waters extending out to 20 miles offshore from a stretch of coastline 100 to 200 miles in length. The main strength of these data are their excellent quality[21. Scientists Similar forecasts are also available for tidal bays. If you obtain nothing and weather forecasters regard these reports as much superior to ship data. else, obtain these forecasts before making your decision to venture out. All data are monitored daily and removed from distribution if the data One disadvantage of these forecasts are their "broad brush" approach, 1341 United States Government work not protected by copyright W-1 W.5 0 W-6 W-2 B-1 0 B-1 C,;. NO. STATION NO. STATION '1jr @,V C-4 W-3 W-1 SMITH ISLAND, WA E-1 DUNKIRK, NY C-5 E-6 W-2 WEST POINT, WA E-2 SOUTH BASS ISLAND, OH C-2 E-8 W-7 W-3 NEWPORT, OR E-3 GALLOO ISLAND, NY B-2 W-4 POINT ARGUELLO, CA E-4 FOLLY ISLAND, SC W-5 TATOOSH ISLAND, WA E-5 CHESAPEAKE L.S., VA E-3 7 B-5 W-6 DESTRUCTION ISLAND. WA E-6 MOUNT DESERT ROCK, ME -8 W-7 CAPE ARAGO, OR E-7 ISLES OF SHOALS, NH B-4 B69 W-8 POINT ARENA, CA E-8 MATINICUS ROCK, ME E-1- E-13 E-9 FRYING PAN SHOALS L.S., NC C-1 ROCK OF AGES, MI E-10 CAPE LOOKOUT, NC E-1 2 C-2 DEVILS ISLAND, WI E-11 DIAMOND SHOALS L.S., NC E-12 E-12 AMBROSE L.S., NY B-7 W-8 C-3 SHEBOYGAN,Wl J1- -t:.^ B-3 C-4 PASSAGE ISLAND, WI E-13 BUZZARDS BAY L.S., MA NO. BUOY B-6 C-5 STANNARD ROCK, MI E-14 THOMAS POINT, MD B-i COLUMBIA RIVER, OR "( B-2 PORTLAND, ME E-14 0 E-5 S-1 CAPE SAN BLAS, FL B-3 SAN FRANCISCO, CA S-2 SABINE, TX B-4 NANTUCKET,MA W-4 S-3 SOUTHWEST PASS, LA B-5 ST. GEORGE REEF, CA *E-11 S-4 PORT ARANSAS, TX B-6 DELAWARE BAY, DE E-10 S-5 ST. AUGUSTINE, FL B-7 FIVE FATHOM, NJ S-6 LAKE WORTH, FL B-8 BOSTON,MA S-7 GRAND ISLE, LA B-9 BLUNTS REEF, CA 0 E-9 S-8 MOLASSES REEF, FL B-10 GRAYS HARBOR, WA S-9 SAVANNAH L.S., GA E-4 S-1 0 VENICE, FL S-1 I SETTLEMENT POINT, GBI CENTRAL@ OS-9 S-12 DAUPHIN ISLAND, AL S-13 SOMBRERO KEY, FL PEPOT S-14 CHEVRON MP133, LA S-1 S-12 S-1 S-5 NO. STATION S-2 S-7 .31 A-1 FIVE FINGER, AK S-4 S-10 S-6 es-11 E - EA ERN S - SO T ERN S-13**S-8 A-1 C - CENTRAL W - WESTERN A - ALASKAN 1@ B - LNB/ELB Figure 2. C-MAN Site Locations sometimes specifying a large range of wind and wave conditions because data. First, they could be able to more accurately locate and time frontal they cover a large section of coastline for a 12-hour period. passages. Second, regional sea-surface temperature maps are transmit- ted twice each week. These maps could be of interest to fishermen if they A second product that can be very helpful is Model Output Statistics know how to correlate the location of certain species with a certain range (MOS) Offshore Wind Forecasts. This little-known product provides of water temperatures and/or gradients. forecasts of wind direction and speed for 3-hour intervals for the next day for specific offshore locations. The locations are sites where coastal observations are available, and are typically spaced about 50 miles apart. 3. DATA DELIVERY METHODS These wind forecasts are available twice a day and are produced directly by a computer using the N40S method[3). An example forecast is given So much for the "what," how about the "how?" I will consider four in Figure 3. Because the forecasts are point- and time-specific, they are possible data delivery methods. The first method is simply continuing better than the winds contained in the coastal marine forecasts in to rely on NOAA Weather Radio. Then, I'll explore the possibility of 11 average" weather conditions. During unusual weather, for example, dialing into an NWS computer, which is only available if you live in cer- when the coastal marine forecast calls for winds higher than 25 knots tain areas. Third, I'll discuss how you can dial into private meteorological or thunderstorms, then the winds in the coastal marine forecasts should firms' on-line data bases. Finally, I'll explore the Cadillac of data delivery be relied on. methods: using a satellite dish and software to continually monitor the weather yourself. Table I lists which products are available for each data The MOS forecasts should be particularly helpful to sailboat operators. delivery method. Sailboaters may not want to venture out if the winds are forecast to be mostly less than 5 knots. Six to twelve knots may be ideal, but beyond First, I'll briefly mention the old standby, NOAA Weather Radio. As that it may be a rough ride for some of the small sailboats in unprotected mentioned previously, not all observations are reported and the infor- waters. The MOS Wind Forecasts help provide this kind of resolution. mation is updated hourly, causing late observations to be broadcast. In Because of improvements in computer weather forecasting and the addi- addition, MOS wind forecasts are not available. This is partially because tional NDBC observations in the coastal zone, these forecasts are expected of time constraints, and partially because of the perception that these to improve considerably in the next few years. One disadvantage of these forecasts serve as guidance for only trained meteorologists. Of course, forecasts is that wind speeds for a few locations exhibit a high or low NOAA Weather Radio is free and you can take it with you anywhere. bias, because of the observational data set that these forecasts are based on. Some sites had very poor anemometer exposure, while others were Next, if you live in either the Cleveland, Ohio; Washington, DC; Boston, located on offshore towers 100 or more feet above the water. With a lit- Massachusetts; or Portland, Maine area, you will be able to dial into an tle experience, you can spot these biases and adjust for them. These biases NWS computer for free and retrieve certain products if you are willing should be reduced when MOS forecasts use NDBC observations. The loca- to provide cooperative weather observations when you are at sea. Local tions are given in NWS Technical Procedures Bulletins Nos. 332, 339, marine forecasts are available, and this is the only source of cooperative and 340, available by calling (301) 427-7462. marine weather observations. However, all the systems, except for the one in Cleveland, are updated infrequently. Therefore, they do not con- Finally, graphical weather maps and analyses are available. There ap- tain "perishable" products such as the buoy, airways, and radar obser- pear to be only a few reasons why small craft operators might want this vations. Because these NWS computer systems are PCs or small 1342 Table 1. A list of which products are available through several data FZLIS40 i,:.'@JBC 281200 delivery methods. Y means that the product is available, N F',-,.'U S'4 () cs-ri_ wra) FcsT-s_&.,,m 7129188 120o GMT means that it is not. P means that some of the product is available. D/GMT 20 1 @1-3 28,*21 2900 2963 2906 2909 2 9 1. '2 P. 9 1. E; 2918 P921 3C)CM) 3C)CJ3 3006 30C.)9 30 1. 2 DELIVERY METHODS 19 0,3 1603 9999 99 `-@ 9 17 0 1 050P 07CA 1 15 C) 3 1. 8 (),@% 19 C) 2! 9999 9999 Fj. 2 0 1. 020'3 lIol, NOAA DIAL INTO DIAL INTO RECEIVE WEATHER PRODUCTS WEATHER NWS PRIVATE FROM NPA 1906 2007 2 1 C)4 2/tC)3 1402 0603 0303 C)9o3 RADIO COMPUTER DATA BASE SATELLITE "1606 170'5 23C.)6 0602 0401 o2,..)4 02o2 NDBC OBSERVATIONS P N' Y Y 3 F 2302 2103 2 @3 0 3 2o02 0000 22'-)2 0000 0000 11?02 190J 20()3 2 10 1. 0000 0o00 0000 AIRWAYS OBSERVATIONS Y Y Y Y BRZJ 13C.)j 1509 170E] 1607 9999 9999 1700 1110 COOPERATIVE SHIP 1007 14141 1311 1312 9999 9999 1412 OBSERVATIONS N Y N N Figure 3. MOS Wind Forecasts issued at 1200 UTC, July 28, 1988, for COASTAL MARINE the Gulf Coast. Station NPA is Pensacola, Florida. The first FORECASTS Y Y Y Y forecast calisfor winds blowingfirom 190 degrees (S) at 6 knots. SPECIAL MARINE The time for this forecast is found at the top of the column, WARNING AND RADAR the 28th at 1800 UTC. Forecasts are then given at three-hour REPORTS Y N' Y Y intervals. MOS WIND FORECASTS N N Y Y minicomputers, access is limited. Typically, only one phone line into these systems exist. If you are interested in providing cooperative observations GRAPHICAL MAPS N N Y Y and dialing into these systems, contact these local NWS forecast offices. The third option consists of dialing into a data base maintained by a *THESE PRODUCTS ARE AVAILABLE THROUGH THE CLEVELAND, OHIO,, NWS private meteorological company. Users can obtain the most recent ver- COMPUTER FOR THE GREAT LAKES. sion of all products except for the cooperative ship observations. Com- puter graphic displays are also available if you have an appropriate ter- minal. Some services will decode coastal airways and buoy observations, Table 2. Private Meteorological Companies Offering Data Bases for and place the data in a nice tabular format. Previous observations are Public Access also available if a user is interested in determining a trend. The cost is about $60 for each hour of connect time. Typically, the user will dial a local TYMNET phone number to access the remote computer, thereby The Weather Network Oceanroutes, Inc. Weatherbase, Inc. eliminating long distance charges. One possible problem with this method 3760 Morrow Lane, Suite F 2185 South 3600 West is the occasional garbled messages you'll receive due to noisy phone lines. Chico, CA 95928-8865 Salt Lake City, UT 84119 This is more of a problem in remote, rural areas. Another disadvantage (916) 893-0308 (801) 973-3131 is that in severe weather, you'll need to frequently redial in order to keep abreast with the latest information. A list of companies that I know of Acces-Weather, Inc. WSI Corporation which currently offer this service is given in Table 2. 619 W. College Ave. 41 North Road State College, PA 16801 Bedford, MA 01730 Finally, a user can receive and store his own weather data via a satellite (814 237-0309 dish and software running on his own PC. The two-foot-diameter dish (617) 275-5300 will be pointed to a geostationary commercial satellite where a continuous stream of weather data is available. The data stream will be passed through a controller, which is leased from a private meteorological firm, and on- to a PC. Software continually monitors the receipt of the data and stores 4. CONCLUSION whatever products you indicate. Products can also be automatically printed upon receipt. All products are available within a couple of minutes Mariners who own PCs have a wealth of real-time weather data available of their issuance, so this is a great way to maintain a continual weather for them. Though the choice of data delivery methods and costs vary watch. The disadvantage of this system is the price. The initial price for considerably, the mariner should be able to find a service to meet his needs the dish, controller, and software runs between $3,000 and $5,000, and and pocketbook. As the price of PCs continues to fall, more mariners the monthly data service fee costs $95-$230 each month, depending on should be turning to them for obtaining their weather data. The aviation your exact data needs. Though this is out of the price range of all but community has already begun to use PCs to obtain more of their weather the most "well heeled" individuals, it is not out of the question for groups briefings, and I predict that the marine community will follow this like fishermen cooperatives and yacht clubs. These groups could even pro- established trend. vide a phone line so that members can dial in from home. I know of two companies that currently offer this type of service - Contel 5.REFERENCES (703) 790-2094 and Zephyr Weather Information Services'(617) 898-3511. The Zephyr system uses software written under MS-DOS and can run I Hamilton, G.D., "National Data Buoy Center Programs," Bull. on any IBM PC/XT or compatible with a hard disk. The Contel soft- Amer. Meteor. Soc., 67, 411-415, 1986. ware is written under the Xenix operating system and an IBM PC/AT or compatible is needed. The Zephyr software is less expensive and was 2. Gilhousen, D.B., "A Field Evaluation of NDBC Moored Buoy designed for a single user. The Contel software permits multitasking. In Winds," J. Atmospheric and Oceanic Tech., 4, 94-104, 1987. other words, if you want to have mariners at a remote location dial into your computer, or run word processing and spreadsheets at the same time, 3. Carter, G.M. "Automated Prediction of Surface Wind from you will be concerned about multitasking. Numerical Model Output," Mon. Wea. Rev., 103, 866-873, 1975. 11343 INTFRACTIVE MRINE ANALYSIS & FORECAST SYS= (DWS) THE OCEAWGRAPHIC WOMMMON OF THE FLUURE P.M. Friday, J.S. Lynch, and F.S. Long Office of Ocean Services National Ocean Service National Oceanic & Atmospheric Administration U.S. Department of Commerce ABSTRACT Towards this end, NOAA is implaTenting a number of new activities aimed at The National Oceanic & Atmospieric providing timely and useful information to Administration (NOAA) and the U.S. Navy decision-makers. In order to provide have begun development of the Inter-active these timely services, NQAA is developing Marine Analysis & Forecast System (DWS). an interactive workstation and The'system will provide forecasters, communications capability to Support the research scientists, and data managers marine programs. with the means to access and integrate interdisciplinary data sets f-ra, several sources, from archived to real-time data. 2. OBJECTIVE The system will support NQAA's programs, including: climate and global chanje, The objective of this project is to fisheries productivity, environmental develop and nplement an Interactive quality, coastal hazards, and marine Marine Analysis and Forecast System weather. The interactive analysis of (IMUS). nie system is an interactive oceanographic (biological, chemical, and (man-machine mix) analysis and forecast physical) and atmospheric data sets is tool, based on existing technology, with fundamental for understanding and supporting communications capabilities. predicting the environment. The system will be used for preparing analysis and forecast products, product quality assurance, and database management at NQAA1s field sites. The program builds upon an ongoing malti-Line office effort 1. INTROWCTTON to establish a computer workstation which when combined with program-specific The Nation is facing a number of critical software can be applied in a mmiber of environmental issues, including marine disciplines, including global pollution, fisheries depletion, and global environmental monitoring, climate and climate change. NGAA has begun an global change analysis and prediction, aggressive program to address these and marine weather forecasting, and coastal related issues by focusing existing ocean analysis and prediction. resources and prnposing new initiatives Uirough the Climate and Global Change Program and the Coastal Ocean Program. 3. RATIONAI[E AND BACYGROM The Climate and Global Change Program is a Ocean data sets and products are only of systematic effort to understand the minimal value for climate research and processes of oceanic and atmospheric ocean applications if they are not change (seasonal to interdecadal) so that generated and integrated into a useful scientific predictions can be developed. form and made available to those who are The Coastal ocean Program is an integrated trying to understand ocean variability. approach to understanding biological, Therefore, a significant need exists to chemical, and physical processes and implement an interactive analysis and variability in the Nation's coastal ocean applications s@,@stem which will provide (the Exclusive Economic Zone). the means to develop new and improve existmig analysis and forecast products of the upper ocean; the interactive capability to display, manipulate, and disseminate the products; and the ability to manage the associated databases. CH2585-8/88/0000-1344 $1 (D1988 IEEE Recognizing the success that the OAR The cmmunications capabilities of the Program for Regional Observing and system are being designed to utilize a Fbrecasting Services (PROFS) has had in wide-area network to distribute data and applying this interactive system approach generalized large-scale products from to weather forecasting, NOS has recently appropriate cextral databases to the stimulated some NCAA inter-Line office DUOS workstations (see Figure 1) - cooperative activity to beging applying coammications ports at the IKkFS sites this technology to ocean needs. Design will be used to establish local-area specifications of imAFS are being prepared networks to support other federal and by Nos. The hardware coaponent of IMAFS, state offices using ocapatible PC the prototype workstaticn, is being systems. The resulting local-area developed by PROFS with funding provided networks will ensure that: by NOS, National Marine Fisheries Service (NNES), National Weather Service (NWS), 1) data and regionally produced and the Navy to meeet specific programmatic products can be accessed at other requirements, including a I'demonstration" federal and state offices; and capability to illustrate the type of support possible for global change, 2) data collected'and products fisheries, and coastal ocean programs. produced by the federal and state offices will be accessible via the IMAFS is being designed to store, process, same local system to the IMAFS for and display conventional observations, distribution to other users. gridded fields, digital satellite data, and climatologies. The system will rftm telecomumnications network will: allow integrate the overlay of multiple data and access to databases at a variety of product sets, and will include some places; disseminate data/products to interactive and cooputational applications regional DMFS workstations; and provide capabilities, such as: annual mean and two-way transfer of data and products from anomalous SSTs based on monthly SS`r regional imus sites to other analyses; interannual and interdecadal federal/state offices. IMUS is being trends; upper ocean (0 - 400m) heat designed to utilize existing databases content and heat transport from [eg. the Coaprehensive Ocean-Atmosphere BATHY/TEsAc and current data; and Data Set (COADS) and NOS/CMA status and estimates of upper ocean kinetic and trends data] and communications systems. potential energy from in situ and rerDtely sensed data. A mmber of additional applications could also support the global change, fisheries, and coastal ocean programs. U@ese applications include, but axe not limited to: Climate and global change applications COA/FNOC PCs OM (El Nino/Saxthern Oscillation) D < E> Global Change (WOCE, GOFS, TOGA .... ) I - water level arsomalies/variability S T - sea ice ancmlies/variability R - ocean circulation anomalies U T -Q D, PCs - mass/enexgy transport E Coupled air-ocean model verification D Pattern separation D A T Reanalysis A B A @pC Fisheries applications S E ra B@3 PCs Global fisheries monitoring S Species migraticWtrends IMAFS .5o E, Species populatials/trends AW Resource assessment Coastal environmental applications Coastal zone & estuarine studies Pollution marLitoring Figure I. Distributed database network. Coastal hazards mitigation Habitat monitoring other ocean-related applications Data archiving and retrieval Data quality control Applied research 1345 4. THE FUTURE 5.IMPLEMENTATION This project is linked to the NOAA Climate FY89 ACTIVITIES: IAFS workstations will and Global Change Program, the NOAA be procured by NOS and installed at the Coastal OCean Program, and the NWS NOAA Ocean Products Center(OPC,Camp modernization program. These activities Springs, MD)and the NOAA Center for Ocean will incorporate the IMAFS approach into Analysis and Prediction (COAP,Monterey, their programs, building upon the CA). These workstations and additional workstation and generic software developed systems in place at OAR will be used to by OAR PROFS. The parallel development develop new and innovative techniques and efforts can provide considerable cost applications software specifically related savings, such that each Program can focus to seasonal, interannual, and interdecadal on specific applications pertinent to variablility of the ocean-atmosphere system their activities. addressed by the Climate and Global Change Program. The Climate and Global Change Program will provide for the development of specific applications software related to ocean FY90 ACTIVITIES: Additional IMAFS climate processes, particularly as related workstations will be installed at other to seasonal, interannual, and interdecadal NOAA facilities. Several systems will be variablility of the global ocean. installed at the NOAA archive centers (National Oceanographic Data Center The Coastal Ocean Program will provide for Washington,D.C.),National CLimate Data the development of specific applications Center (Asheville, NC), and the National software related to forecasting daily, Snow & Ice Data Center (Boulder,CO)so weekly, monthly, and seasonal variability that: in the coastal ocean (the U.S.Exclusive climate-related data and operational Economic Zone) pertaining to fisheries products generated with IMAFS during productivity, environmental quality, and the 1990s can be archived and made coastal hazards. Also, it is important to available to the retrospective user note that due to the broad spectrum of community in a timely manner; and expected participants, the Coastal Ocean Program will need to establish a the data archive centers can begin substancial communications network among on-line services to make archived data NOAA's laboratories and regional centers. sets available to NOAA users on the IMAFS network. The NWS Modernization Program intends to use IMAFS technology as a prototype marine Systems will be installed t the Regional weather fore-casting system (similar to the Fisheries Centers (Woods Hole,MA;Miami, prototype regional forcasting system at FL;Seattle,WA; and LaJolla,CA)for the Denver,CO)to demonstrate the potential development of region-specific fisheries benefits of this capability in the 1990s. applications based on national guidance Specific applications software related to products prepared and distributed by the marine weather forecasts, nowcasts, and Center for Ocean Analysis and Prediction, warnings will be developed through the NWS the Ocean Products Center, the Navy-NOAA Modernization Program. Joint Ice Center, and the Climate Analysis Center during the 1990s. IMAFS is being designed for maximum compatibility with other NOAA Software development for climate, workstations, including systems at OAR's fisheries, and coastal ocean applications marine laboratories, NMFS"S regional will continue at OPC,COAP,and PROFS, and centers and laboratories, NESDIS's archive new development efforts will be initialed facilities, NWS's Great Lakes Forecast by the Fisheries Centers. Center, and NOS's status/trends and ocean services programs. IMAFS builds upon an COMMUNICATIONS: Two existing open architecture approach, providing the communications capabilities can be capability to expand memory capacity and utilized in the early phase of this upgrade the hardware configuration. project:the communications system at Software is being designed for modularit, COAP; and the NASA Space Physics Analysis such that:(1)the generic workstation can Network (SPAN)with existing nodes at CAC accept, displlay, and process virtually any and the NOAA archive centers. Real-time data set, gridded field,or satellite data sets will be disseminated via the image; and (2)seperate applications COAP. Archival data sets will be module will be developed for climate and communicated via NASA/SPAN, the COAP,or global change, coastal ocean, marine CD-ROM (Compact Disk-Read-Omly-Memory). weather, and possibly other activities. Both COAP and SPAN systems utilize a Packet Assembler/Disassembler approach, 1346 wtuch sorts and schedules the wide variety of available NOAA products, and disseminate only those pro&Y-tz requested by users. As presently envisioned for the early phases of this project, ODAP will serve as a centralized distribution site which will access real-time and some archival data and products from existing databases, merge and create new climate and fisheries products, provide unrestricted aooess to NOAA IMAPS users, and disseminate data/products to appropriate federal and state users. 6. BENE= Benefits of this program include: Increased availability and enhanced utility of real-time and archived data, analyses, products, and model output by users (NOAA, federal, and state) for specific ocean applications; Ability to incorporate and combine a wide variety of data and product types, including the ability to merge the various data/products to generate new products to meet specific ocean applications; and Ability to provide tw@way ccomunications to exchange data/product sets generated at national processing centers, national archive centers, regional DWS facilities, ocean research laboratories, fisheries regions, and local PC workstations. Because multiple NOAA program are using or expect to use the IMAFS technology, significant leverage can be applied within each participating program to access camxm system elements at minimal or no ocat. 1347 THE GLOBAL OCEAN PLATFORM INVENTORY Deborah A. Storey and William E. Woodward National Oceanic and Atmospheric Administration Washington, D.C. ABSTRACT The development of a comprehensive opera- tional capability for ocean analysis and By initiating a comprehensive inventory of prediction requires that every available in situ ocean platforms that routinely observation be used to the maximum extent. collect oceanographic and marine meteoro- This is especially important because the logical data, the National Oceanic and spatial and temporal coverage of the Atmospheric Administration's Office of existing network is sparse and expensive Ocean Services (OOS) has taken the first when compared with the coverage of equiva- step toward improving the coordination of lent measurements available over land the globally dispersed elements of the masses. As the Nation's civilian agency system of platforms. The resultant engaged in environmental monitoring and database should form a basis for improving prediction, NOAA is addressing both near- the measurements per se, eliminating and long-term improvements aimed at redundant sensors, and revealing geograph- providing major improvements in forecast ic areas that are in need of increased accuracy and in the efficiency of both coverage. OOS is exploring advanced data sensor systems and human effort in communications techniques for making this collecting data that will be fully database available to the scientific utilized. community. INTRODUCTION NOAA's Office of Ocean Services has taken a first step toward improving the Our present source of information for coordination of a globally -based ocean understanding and forecasting the ocean's observation system by initiating a survey structure, variability, and dynamic inter- of in situ platforms that routinely action with the atmosphere is a loosely collect oceanographic and marine meteoro- organized network of global satellite and logical data. The resulting computerized conventional observation platforms. This database describes the number and types network is a composite of platforms used of federally owned and sponsored in situ for both operational and research needs, observing platforms. Additionally, the operated by a varied group of agencies, Office is presently working with U.S. each with different missions and objec- coastal and Great Lakes states to tives. Consequently, the network is a inventory the monitoring systems they costly, poorly coordinated mixture of support in nearshore regions. An observing systems. Thus some ocean inventory of international programs will regions may have a significant amount of soon be initiated. The federal inventory data being collected while other areas will be expanded to include the informa- have no observational coverage at all. tion derived from these state, local, and Furthermore, incompatibilities in platform international inventories. type and location, measurement instrument- ation, data format, quality control When fully completed, the database will procedures, and communications links, provide a tool for rational decision combined with inadequate national or making regarding quality control of data, global data assimilation capability, possible joint procurement of observing preclude optimum exploitation of the systems within and among agencies to valuable data that are being collected include operation and maintenance, and a worldwide. This is a costly, inefficient, coordinating service vital to making unacceptable situation for operational effective and efficient use of all data. forecasting. The problem is exacerbated Furthermore, the database will provide by the inability of existing systems to opportunities for those interested in handle the increasing volume of data from deploying new ocean sensing platforms to both conventional and remote observation review similar systems that already exist platforms. in a particular region of interest. OOS' 1348 United States Government work not protected by copyright is now exploring ways to make -this database readily available to federal and non-federal entities. PROGRAMS DESCRIPTION OF THE OCEAN PLATFORM DATABASE CLASSES USED (12J The file structure of the ocean platform database is depicted in figure 1. It is organized into four interrelated files,, each containing a different set of facts ASSOCIATED about the platforms. In the figure, PROGRAMS (4) ovals represent database files. The directed lines between the ovals denote key fields (i.d. numbers) in one file that identify related records in the PLATFORM other f ile. Labels identify the nature a-ASSES ASSOCIATED of the relationship and numbers in PROGRAMS (5) parentheses represent the number of fields available for each record in PARTICIPATING ORGANIZATIONS CLASS support of the relationship. For (1) example, each platform record may identify as many as six programs related OWNER (I by common factors such as sponsorship, HARDWARE (5) ownership, maintenance responsibility, DESTINATION (4) and access to data. Central to f igure 1 and to the database ORGANIZATIONS PLATFORMS is the platform class file . ( The platform class file for VOLUNTEER OBSERVING SHIPS is shown as figure 2.) Fields of the platform class file are clustered in five qroups: A heading Figure 1. Database File Structure Class ID 0001 ClA Name VOLLu?IEM OBSERVING SHIPS organization ID 0024 Assoc. programs 0004 platform Type Surface Ship jAmber in Class 745 AppllCatICn Forecast Cmuient TBIS IS THE MST BASIC VOS CIJLqS, WME KWUAL F=RDING AND REpc=;G OF MET. OBS. VIA RADIO. Nor AINAYS SYNOPTIC. Telemetry Real Time Synoptic GIs APOS yes Satellite Radio Telephcm Messenger Mail Other no yes no no yes no Teleretxy Descr. Pora METEOR. OFFICER MAILS REPCEE Recorded format STANDARD FORK Transmit format SHIP SYNOPTIC CODE How often sent Mk= CN ARRIVAL. RADIOED NM24A= 3 HOURS- Destination ID 0020 Oceanographic Measurement Capability parameter Loc. sensor description Accuracy Frequency Temperature Teuperature S Con./Salinity Con./Salinity S Current Current S Waves VISUAL 1-8 HOURS Water Level other s indicates sub-surfaoe sensor. All others are surface. Figure 2. Sample Platform Class File (continued'next page) 1349 Meteorologic Measurement Capability @r ld@ i1L Parameter ::"tilm Accuracy Frequency Temperature V13cm 1-8 HOURS Pressure BAROGRAPH 1-8 HOURS Wind Speed ANEM34ETER (POSSIBE 1-8 HOURS Wind Direction VANE 1-8 HOURS Dew Point /R.H. PSYCffirbZTER 1-8 HOURS other ICE, VISIBILITY, CLOUD - VISUAL 1-8 HOURS Hardware Management Responsibility Contact person org. Phone System operation NWS - Vince Zegowitz 0024 (301)-427-7724 calibration NWS EMS 0024 Maintenance NWS PMOS 0024 Quality Control PRIM?W QC BY NMC 0020 Data Archive NCDC, IN ASHMILE, NC 0017 Figure 2. Sample Platform Class File group includes class i.d. number and name buoys had been released on the Arctic ice as well as references to associated pack for research purposes. We believe programs and organizations Other that the data from these buoys could have groups identify telemetry, measurement benefited operational polar ice forecast- capabilities, and individuals/agencies ing. Similarly, it was discovered that responsible for the class and its data. there are both drifting and moored buoys deployed in Equatorial Pacific programs, The platform file is a list of all class collecting data that could be of major members giving, for each platform, its benefit to the work of NOAA's Climate i.d. number, associated class, associated Analysis Center in improving its realtime programs, and location parameters. ocean models. At the engineering level, The program file contains identifying the inventory shows that certain sets of data on platforms associated with major useful data are not being ingested by government oceanographic programs. it NOAA's weather and marine forecasting includes five information groups listing models because of incompatible data associated organizations, approximate foremats. Workshops are planned to level of program funding, platform coordinate and pool the efforts of all classes used, platform counts by class, agencies that stand to benefit from the and users of platform products. The sharing of data. organization file is a simple list of participating organizations and their FUTURE DIRECTIONS addresses. Now that the initial inventory has been INVENTORY RESULTS accomplished and the database estab- lished, the inventory will be extended The primary data sources for the inven- while increasing access to the database. tory consist of studies of three dozen The remaining elements of the ocean national programs which include over 500 observation enhancement effort are platforms categorized into five dozen envisioned in the following paragraphs: classes. The studies include annual reports, technical reports, catalogs, INTERAGENCY COST REDUCTION PROGRAM - This brochures, charts, tables, records of effort builds upon the substantial telephone conversations, and detailed information gained during the inventory notes taken during technical conferences. phase. The inventory provides a useful tool for planning and coordinating system Although the computer database created standards, e.g., reliability statistics; with these records has been populated organizing government-wide procurement only with data from the U.S. federal strategies; promoting cost sharing in inventory, it is demonstrating that there system operation and maintenance; and are certain areas of the world's oceans providing management information. that contain very few in situ observing A maj or Iobjective of this activity will platforms. At the other extreme, the be to identify and eliminate redundancies database indicates that there are several and system deficiencies, providing ocean monitoring platforms collecting substantiation of critical needs and significant amounts of data (at great optimization of ultimate "mixes" of expense) that are not being used to the platforms, sensors and measurement maximum possible extent. For example, it techniques. was found that some expendable drifting 1350 REAL-TIME QUALITY CONTROL SYSTEM - This Informing the system opera- is an effort that can be initiated tors, performance monitor, and immediately. Since the majority of datausers when there is marine observations still come from ships missing or degraded data from at sea participating in the Volunteer Individual platform(s) and Observing Ship Program (now over 7,700 1monitoring of corrective vessels worldwide), and these data actions needed to restore the undergo only cursory editing at our .,platform to full operational National Prediction Centers, this effectiveness. database can benefit immediately from several basic procedures (listed below). The same techniques, once they undergo COORDINATION AND INTEGRATION - Because operational testing and are in place, can the entire.global ocean data set must be then be applied to other platforms and uniformly accessible in both the near - sensor systems: real time and retrospective modes, � Evaluation/Assessment Procedures NOAA's' global database must be operated and managed.as an integral unit. We are - Comparing to a range of values aware that compatible technologies in or climatology measurement systems, data communications, - Comparing data with previous and data processing are already in use and later observations in NOAA or that their procurement should - Comparing data with nearby be planned and programmed systematically observations so as to gain maximum benefit from scarce - Checking for internal fiscal resources. Therefore we must act consistency decisively and move ahead on both fronts � Action Procedures in a logical, fully coordinated manner. Maintaining corrector values for each platform and/or .@ACKNOWLEDGMENT sensor - Correcting errors, where Creation and population of the Inventory feasible Database 1was .,accomplished by EG&G, - Editing data before the Washington Analytical Services Center, archival process Inc., under contract to NOAA. 1351 LEADING LINES FOR THE NINETIES Dr. Lester Mehrkam United States Coast Guard Washington, D.C. 20593-0001 is on the channel's center line. Thelaser ABSTRACT line is visible from the horizon to the zenith because the atmosphere scatters light out of the laser beam. Multiple Navigation in a channel by observing sector lights do not give continuous leading lines is discussed in terms of feedback but may be useful in acquiring operational requirements and new signals the entrance to the channel or for Once off the channel's center line navigation in short segments of the conventional leading lines provide n@ channel. inherent clues to the ship's lateral position in the channel. Alternative methods of navigating in a channel using SINGLE STATION LEADING LIGHTS vernier lines of light, laser lines in the sky and multiple sector lights are described. The ideal leading line will Single station leading lights with a mark the center line and edges of the luminous range over a mile can provide channel as an inherent characteristic of continuous feedback for navigation only the method, and will be usable by with a temporal signal, because the navigators who do not have previous signaling apparatus must be enormous to experience in the channel. Methods that obtain a resolved spatial display. provide continuous feedback from spatial Interpretations of temporal signals such signals are preferred. Field tests of as counterrotating pencil beams of a these alternative methods are described. light, flash rate changes of a sector light and alternating colored flashes from a sector light are poor navigation tools INTRODUCTION compared to spatial parallax. Off center line navigation is not an inherent feature of single station leading lights. Light Staying on the center line of a channel is signals that provide an instantaneous, not always possible or desirable. Once continuous spatial signal are desirable @off the channel's center line, the for navigation in a channel. navigator relies on experience from previous transits and buoy fences for confidence. The need for of f center line CONVENTIONAL LEADING LIGHTS navigation is an established fact. Leading lines are now being provided for inbound and outbound lanes in some The observer looks toward the horizon when channels to overcome the difficulty of off using conventional leading lights. The center line navigation. Use of center line of a channel is operationally conventional leading lines (lights and located when two point light sources, dayboards) is a learned experience for called "leading lights", appear to be on a each navigator in each channel. Vernier vertical line. A line connecting the lines of light (extended light sources) "leading lights" appears oblique when the have operational limitations similar to observer is off the center line. The those of conventional lights (point angle between a vertical line and the sources), but may have fewer engineering slanted line through the "leading lights" limitations. Laser lines in the sky increases with the observer's distance off uniquely mark the center line and both the channel's center line. Experience in edges of the channel as an inherent a specific channel is needed to make use characteristic of the method, not as a of the information provided by "leading learned experience. Laser lines originate lights" that appear to be on an oblique f rom a single point on the horizon and line. Feedback is continuous but no pass through the zenith of an observer who definite channel edge or lane parallel to 1352 United States Government work not protected by copyright the center line can be located uniquely with "leading lights". VERNIER LINES OF LIGHT The operational use of "leading lights" is a result of spatial parallax. To achieve parallax the "leading lights" are placed The center line of a channel is on structures that are built on the operationally located when two vertical channel's extended center line. The upper lines of light that are at different light is farther away from the end of the heights appear to be colinear. This channel than the lower light. Use of the vertical line may have a gap at the "leading lights" to navigate on an off junction for some channel positions. The center lane is a learned capability of the horizontal separation of the two vertical operator and not an inherent feature of lines increases with the observer's the "leading lights" method. Buoys enable distance off the channel center line with mariners to navigate with confidence when the same convention as for point lights. off the center line. Only the center line of the channel is inherently marked by use of vernier lines Two point light sources must have an of light. Navigation off the center line angular separation of 1.3 milliradians to is a learned experience. Lines of light be easily resolved as separate lights. will be more conspicuous than points of The line connecting the point lights will light and will make leading line lights have a detectable slant from the vertical unique from other aids to navigation when the horizontal angular separation of lights and commercial lighting. the point lights is 0.3 milliradians. This minimum detectable slant depends on the relative intensity of the lights and The two lines of light must have a the vertical separation of the lights. geometrical configuration similar to the Airport Reach Channel on the Columbia conventional point lights to obtain River has a width of 300 feet and a f ar spatial parallax for navigation in a end of channel to f ront light separation straight channel. An improvement in of 10,190 feet and an acceptable off navigation can be expected if the lateral center line awareness distance of 40 feet. alignment sensitivity of two vertical The required horizontal and vertical lines, vernier acuity, is greater than the separations of the two lights are 1600 lateral sensitivity for vertical alignment feet and 14 feet, respectively. of two point sources. Difference in how the eye responds to points and lines with Limitations of conventional leading lights different separations and apparent are reviewed: intensities and the difference in atmospheric scintillation effects on point (a) Only the center line of the channel is sources and extended sources are reasons uniquely determined. Off center line for expecting increased sensitivity. navigation is a learned capability and not Another advantage to using extended rather an inherent property of the system. than point light sources is that when viewed through binoculars the lateral (b) Buoy fences are needed to give alignment sensitivity is increased for mariners confidence for off center line lines of light and decreased for points of navigation. light. Since binoculars are frequently used this advantage should not be ignored even though leading lines are designed for (c) The two point light sources must be the unaided eye. separated by several thousand feet for a typical range, and the light's locations The same operational requirements for the are critical. Airport Reach Channel can be achieved with lines of light separated horizontally by (d) Access cost for the land and the cost 426 feet or less, approximating a single of bringing power to the locations can be station serving as a leading light. The high. length of each lighted line should subtend a vertical arc of 1.3 milliradians, which (e) Lights several thousand feet apart are is 15 feet for this example. The tugboat hard to synchronize. captains like the vernier range installed at Airport Reach. Their only operational (f) Rear tower heights of 50 feet or more complaint is that the range is too are common. Towers must support the sensitive which requires decreasing the servicing men and hardware. 426-foot separation of the lines of light. (g) Residential and industrial lighting A vernier range is planned for Curtis often hinders detection and use of point Creek Channel in the Chesapeake Bay. source leading lights. 1353 A line of light. can be achieved by LASER LINES IN THE SKY illuminating the interior of a clear acrylic pipe which has a rectangular cross section with a searchlight. The optical axis of the searchlight and axis of the A laser beam passing through the sky is pipe are made to coincide. Total internal visible to an observer because the reflection of light at the outside surface atmosphere scatters light out of the laser of the acrylic walls is achieved for a beam. The parallax between several horizontal laser lines which are at narrow pencil beam of light that is different heights is usable for navigation emitted by the searchlight. A mirror is in a channel. Polluted air in heavily placed on the end of the pipe opposite the industrialized ports increases the light searchlight to reflect the light. The scattered which makes the line more acrylic pipe appears as a uniform lighted visible. Heavy background lighting from surface for lengths of 20 feet. Only a the city is no longer a problem because passive vertical acrylic pipe must be the laser line is aimed over the supported. Maintenance on the searchlight observer's head and is seen against the is done at the base - of --- the tower, night sky. Standard Coast Guard searchlights ar usable for this application. The lamp can be flashed to produce a pulsating line of light, and f ilters can be used on the The Coast Guard performed a laser line in searchlight to make colored lines of the sky experiment on the St Mary's River, light. Michigan in 1972. A single horizontal laser line was aimed over the center line The limitations (a) and (b) for of a six mile long channel. The laser conventional point sources still apply for beam was 100 feet above river level, and extended line sources. Lines of light the entire length was visible from must appear as extended sources which scattered light on clear and hazy nights. requires light source lengths, L =0.0013R, The beam was visible from all points in where R is the distance from the light to the six mile channel. The spatial the end of the channel. Typical lengths parallax of the line with the stars and are 30 to 60 f eet. These lengths can be clouds at the observer I s zenith was used reduced considerably, by a factor of 7, if to navigate on the channel ' s center line. binoculars are used. Extended light However, the stars and clouds are so f ar sources will be less energy efficient than away that the laser line shifted radically point light sources. away from the zenith as the observer walked a few feet across the deck away Advantages of vernier ranges are: from the channel's center line. The laser line was easy to see but the- lateral (a) Front and rear lights are close sensitivity was too great to be useful. together, 100 to 1,000 feet, so a COMMON This experiment demonstrated that light POWER SOURCE ( commercial line, engine scattered out of a laser beam by the generator or battery ) can be used, and atmosphere can be used as an aid to lights can be hardwire synchronized. navigation. This laser beam had a diameter of 8 inches (23 cm. ) and power (b) EXPENSIVE TOWERS are not needed. output of 1 watt in the wavelength only the vertical acrylic light pipe must interval of 458 to 514 nm. Eye damage be supported. Servicing personnel need risk was low because the power per unit not climb the pipe. area was reduced to 2.4 milliwatts per square centimeter. The effort was (c) All servicing is done on the abandoned because of technical problems searchlight at the base of the pipe and at with the 1 watt output lasers available in ground level, for a significant increase 1972., in SAFETY. (d) AIMING of a diffuse, extended light The observer looks at the zenith, the spot source is not critical. in the sky directly overhead, to use these lines. The center line of a channel is (e) Extended light sources are operationally located when two horizontal CONSPICUOUS and more easily recognized in lines seen in the sky are equidistant from a background of residential and commercial a third horizontal line which is at the lights than point lights. observer's zenith. The display is symmetrical about the observer's zenith. (f) Binoculars increase the lateral The three horizontal lines in the sky are alignment sensitivity for lines of light. parallel to the channel's center line. The two horizontal, outside lines are in 1354 the same plane, and the line through the (g) The 1972 Coast Guard experiment showed zenith is above this plane and in the that the laser line has the same apparent vertical plane intersecting the channel's brightness over its entire length, and center line. Off axis navigation is an that strong forward scatter of light by inherent feature of the laser line method. the atmosphere occurs since the beam The heights and horizontal distance expands from 23 to 46 cm. in 0.75 miles. between the lines can be fixed to provide This is a divergence of 0.29 degrees. exact identification of the channel's edges and the of f axis center lines of (h) Laser lines are not obstructed by inbound and outbound traffic lanes. ,No other ships. experience is needed. Once the observer is off the channel's center line the display is asymmetrical and shifts away from the zenith toward the port or MULTIPLE SECTOR LIGHTS starboard horizon. At the edge of the channel one outside line coincides with the center line so only two lines are Two colored sector lights can be used to seen. The two lines are below and on the navigate in a segment (about one mile) of same side of the zenith. The spatial a straight channel. Use of more sector parallax between the lines and the spatial lights can extend the navigable segment. parallax of the lines against the stars The lights must be separated by 1.3 are effective aids to navigation. milliradians to be resolved. Spatial parallax plays no part in this method, the There are two possible objections to laser sector lights need only be resolved to be lines in the sky. A stable platform is useful. Two sector lights will be needed so the horizontal direction of the installed to mark Craighill Entrance beam does not wander causing variations in Channel in the Cheseapeake Bay. Entering lateral sensitivity, and the laser beam the channel the mariner sees one or two can blind a pilot who is in the beam and lights changing color from single red, looking toward the source with binoculars. red-red, red-white, white-white, white- The laser beams will be 75 to 500 feet green, green-green, single green as the above sea level which is not an altitude lateral position in the channel changes. where planes fly. Helicopters would have The white-white signal has a constant to be restricted f rom the area but the width equal to the horizontal separation danger is minimal. The Coast Guard's of the lights. The other signals are demonstration showed that a laser line is diverging sectors. The alignment of the visible when the power density is safe for sectors with the channel's center line and the naked eye. angular width of the sectors determines the sensitivity. The sector lights will be at least 100 feet apart for this 500 foot wide channel. The lateral uncertainty at a sector boundary is Advantages of laser lines are reviewed between 5 and 17 feet. This method here: inherently marks the channel's center and (a) Off center line navigation is edges over a limited segment, but does not provided and edges of the provide continuous, instantaneous inherently feedback. channel are uniquely marked for the operator. CONCLUSION (b) Buoy f ences are not needed to give operators confidence. (c) Background lighting is not a problem New technology may now be available to since the operator looks at the zenith of implement significant improvements in channel navigation. The possible the night sky. operational advantage of the laser line (d) Position of lasers from the near end method is easily recognized. The use of of channel is not important. extended light sources as range lights may provide improved conspicuity of the (e) Tall towers are not needed since lights and more localization of hardware. lasers can be aimed upward; this decreases Potential cost savings from reducing tower lateral sensitivity at the far end of the heights and loads and from reducing the channel but does not change the inherent number of buoys used may be a side benefit off axis navigation capability. of the proposed navigation systems. (f) Laser lines are conspicuous and easily acquired by approaching ships. 1355 CCMI?ALV SERVICE OF FEDERAL A TO NAVIGATIM George R. Perreault United States Coast Guard 2100 2nd Street SW Washington, DC 20593 - Discrepancy response ABSTRACr - Establishment and relocation of SRA - Ccnversicn of primary battery (i.e. not Recently, the United States Coast Guard began a rechargeable) powered aids to solar power series of contracts for servicing federal aids Conversion of SRA to comply with the to navigation. These contracts constitute International Association of Lighthouse replacement of the previous government servicing Authorities (IALA) markings personnel and craft with privately owned and operated vessels. This paper describes the Normal on-station time for a buoy is six years. waterway selection criteria, the services being During that time, it is inspected annually for contracted and the results to date. position accuracy and general operating conditions. Biennially, it is hauled aboard the servicing unit to allow the underwater components and mooring system to be checked and XXXJMCN serviced as necessary. Periodically, depending on the light characteristic and hours of Today, one of the largest single missions which operation, the batteries are recharged or the Coast Guard performs is the establishment, replaced. At the end of six years, another buoy placement, and maintenance of federal aids to is put in its place (relieved), and the old navigation. This system includes an extensive buoy is sent to be completely overhauled. In system of buoys, beacons, and lighthouses. most waterways, this process is staggered to There are approximately 49,000 government owned allow the servicing unit to balance its and serviced short range aids to navigation workload. During each visit, the service (SRA) which mark this nation's waters, serving routine includes touchug up the exterior paint, military, commercial, and recreational needs. replacing retroreflective numerals, repairing Additionally, the Coast Guard regulates the minor damage, and testing the light- and power establishment and maintenance of approximately source. In waterways where the winter climate 44,000 private aids to navigation. is extreme, many of the buoys are also removed during the fall and reset in the spring. BACMUUND The buoys require a lift capability of the The Coast Guard Authorization Act of 1982 servicing vessel, based on the buoy size. The mended portions of the legislation governing following table breaks down the buoys by type. aids to navigation (14 USC 81) allowing the U.S. The terminology used indicates the diameter and Coast Guard to contractually establish, overall length of the buoy. For example, an maintain, and operate aids to navigation. one 8x26 is 8 feet in diameter by 26 feet long. The of the outgrowths of that change was a program letters after the bucy size connote type (e.g., to test the feasibility of private contractors LBR is lighted, bell, radar reflector). The to service aids to navigation. A Coast Guard following abbreviations are used: lighted (L), Headquarters task grx)up was formed in the Office bell (B), whistle (W), gong (G), radar reflector of Navigation to develop a strategy to evaluate (R), ice (I). unlighted buoys are classified contracting aids to navigation service, which according to: class (i.e. 1 thru 6 meaning large includes: thru small); can (C) or nun (N), and construction material, if other than steel: - Annual inspection plastic (P), foam (F). - Biennial mooring inspections - Relief of buoys Size Nominal Weight (tons) - Recharge of battery powered lighted aids (no mooring or power unit) - Seasonal aid changes - Building beacons (i.e. SRA structures) Lighted - Lighthouse maintenance 8x26 ILBR 6 8x26 LWR 6 7x17 LR 4 1356 United States Government work not protected by copyright Size Nadnal. Weight (tons) 2. Identify the Coast Guard's ability to (no mooring or power unit) contract and administer annual contracts. This includes the logistics requirements, contracting 6X20 LBR 3 expertise, evaluation process, administration, 5xll LR 1.5 and monitoring workload. Unlighted 8x26 GR 6 3. Determine the actual cost of contractor 8x26 BR 6 servicing and compare it to the historical cost INR 3 of the coast Guard to service the contracted 1CR 3 aids to navigation. 2C 1.5 3C 0.5 SELECrION 5CI 0.5 6CR 0.1 To accomplish these objectives, a task group developed procedures for selecting the trial Beacons (i.e. structures) are serviced similarly contracting candidates - Since all federal to buoys, except the biennial mooring and six waterways are categorized under the Coast year relief periods are not Performed, for Guard's waterway Analysis Management system obvious reasons. in their stead, the dayboards (WAMS), this system assisted in targeting and hardware associated with the structure are eligible waterways. In general, all waterways routinely inspected and periodically fall into one or more of the following refurbished. Also, the visibility of the categories: structure is maintained by brushing-out weeds, trees and other plants that impair the aid's Militarily Critical: Waterways which serve visibility from the waterway. military or military essential facilities and are deemed critical for national defense. Although the preventive maintenance program is extensive, discrepancies still occur. In Environmentally Critical: Waterways where a addition to equipment failure, aids are degradation of the short range aids to vandalized, damaged due to the weather or struck navigation system would present an unacceptable by ships passing too close. The unit primarily level of risk to the general public safety responsible for the aid responds when a failure because of the transport of hazardous materials is reported. Normally, that unit repairs the or dangerous cargoes, as defined in 46CFR and reported discrepancy, as well as performs the 49CFR. annual service, if appropriately timed. occasionally, the unit responsible will not be Navigationally Critical: Waterways where a available and a secondary unit responds. In degradation of the short range aids to areas where there is extensive damage to the SRA navigation system would result in an system, due to acts of nature such as floods or unacceptable level of risk of a marine accident, hurricanes, several servicing units respond. due to the physical characteristics of the This minimizes the time durIng which the waterway, difficult navigation conditions, aid waterway has limited availability for marine establishment difficulties, or a high aid traffic. discrepancy rate. CEOBCrIVES Non-critical: Waterways which serve conTercial and recreational interests, where the disruption With all this in mind, the Coast Guard embarked or degradation of an aid system . beYcnd the on a three year trial of comnercially servicing normal level of discrepancies will not increase short range aids to navigation (SRA). The the risk from a marine accident to an objectives of the trial are: unacceptable level. 1. Determine the ability of a contractor The selection process for the trial waterways to provide equivalent service as provided by the was a three step process. A list of each Coast Coast Gjkird unit replaced. Guard district's units and the waterways 1357 they service was carpiled. The list was broken The original contract (RFp) was advertised on 28 down by critical waterways/aids versus non- February 1986 with 9 lighted buoys, 117 critical waterways/aids. With this breakdown, unlighted bx)ys, 7 light structures, and 8 the units best suited to contract were daybeaccns- determined. Since a very large percentage of the Coast Guard buoy tender SRA are in critical The Coast Guard received two responses to the waterways, rx:) trials affect these units. RFP. A technical evaluation board was conducted on 10 July 1986 to evaluate the two candidates. Certain factors were of paranount concern for Of the two,. only one bidder was found to be the final selection: the ability to resume qualified, and price negotiations began. On 3 service to any trial area should the contractor September 1986, the bidder submitted his best default; the ability to continue the other and finni offer. Although the contract cost rrmssions which the servicing unit performs; and greatly exceeded the estimate of the Coast the desire to minimize the risk to the maritime Guard' s servicing cost, a decision was made to community. The five waterways selected were: award this trial contract. This was done because of the difficulties encountered in - The Merrimack River and Ipswich Bay in trying to apportion the marginal cost of the Massachusetts (First Coast Guard District) sRA, and because of a desire to gain experience in contract service of. Award was made in late - The Intracoastal Waterway in New Jersey December 1986 with work starting on 1 March (NJICW in the Fifth Coast Guard District) 1987. - The Inside Passage of the Virginia A summary of the quarterly reports submitted by Intracoastal Waterway (VIP in the Fifth Coast 'the First Coast Guard District indicates- Guard District) The contractor is performing at a satisfactory - The Sacramento and San Joaquin Rivers in level, however the contractor requested and California (Eleventh Coast Guard District) required a great deal of Coast Guard guidance to initiate his operations. Performance was The Snake and Willamette Rivers in Washington unsatisfactory for the first four months of the and Oregon (M-iirtaenth Coast Guard District) contract. A key premise of the trial contracting The contractor required approximately three initiative was that contracts were only to be months to become accustomed to procedures and awarded when the cointractor's bid was less than reports.@ This is important since changes to the the Coast Guard's current cost of SRA service. aid system are broadcast immediately, and This agrees with the privatization principle to published weekly. The contractor remains slow contract only where it makes good economic and in submitting required reports and paperwork- management sense. In order to permit reasonable cost comparisons, an amB circular A-76 type cost - The contractor has not paid the same degree of comparison was used in order to determine if a attention to detail as the Coast Guard servicing contract should be awarded. unit (i.e. sloppy work, crooked lettering). STATUS - The contractor has been unable to work several aids at different times. This necessitated MERRIMACK RIVER/IPSWICH BAY IN MASSACHUSETTS Coast Guard servicing of those aids. This cmtract involves a portim of the SRA - Several complaints were received concerning assigned to Aids to Navigation Team (ANT) placement of seasonal aids. The contractor had Boston, the Cutter PENDANT and the Cutter WHITE relocated the aids to mark the best water BEATH (141 of 721 total SRA). location. The complainants stated the aids 1358 were not in their historic locations as set and - No public comments have been received. published by the Coast Guard. THE INTRACOASTAL WATERHAY IN NEW JERSEY THE SNAKE AND WILLAI-01TE RIVERS IN WASHIWrON AND OREGON This contract covers aids contained in the New Jersey Intracoastal Waterway which were This contract area consists of two tributaries historically serviced by Aid to Navigation Team of the Columbia River which were serviced by two (ANT). Hornbeam. The contract service area separate Coast Guard units. The contract covers consists of only a portion of that unit's SRA 186 of the 588 aids assigned to Coast Guard and includes: 24 lighted buoys, 260 unlighted Cutter BLUEBELL, and 35 of the 150 aids assigned buoys, 89 daybeacons, and 89 light structures to ANT Bluebell. A total of 127 structures, 4 (462 of 745 total). Of these, approximately lighted buoys, 60 unlighted buoys, 12 half require seasonal replacement. daybeacons, and 20 survey monuments were included in this contract. All of the lighted Four proposals were received by the close of buoys are 5xlls, and the majority of unlighted bids on 2 October 1986. Best and final bids buoys are 5NPR or 5CPR. were received on 13 March 1987. Based on the dollar amounts and scope of work, the contract A REP was advertised, , and closed on May 15, was negotiated as'a fixed firm price contract. 1986. Five bidders responded to the. RFP, of Although the contract costs greatly exceeded the which three were deerned qualified to perform the estimate of the Coast Guard's current servicing work. Best and Final bids were received in cost, a decision was made to award this trial September 1986. Although the best and final contract. Again, this was done because of the offers were all higher than the Coast Guard's difficulties encountered in trying to apportion cost, the decision was made to award the the ANT's marginal cost to the contracted SRA, contract. Since the contract included aids from and because of the need to acquire experience more than one unit there was difficulty in with contracting. Actual contract maintenance assessing the Coast Guard's cost. The contract began in August 1987. was awarded in December 1986, with work beginning in March of 1987. The contractor is a female-owned, small company which is trying to establish a marketplace in The winning bidder is a medium sized barge central New Jersey. Prior to this contract, company which is also a primary user of the the company was conducting local marine waterways. The contract is performed using construction in the area around Tcms River, N.J. existing equipment, so the initial mobilization To satisfy the terms of the oontract,. the and marginal contract costs were small. rrb contractor has procured a 65 foot vessel and execute the contract, Tidewater euploys several expanded operations. Since the bulk of the former Coast Guardsmen with aids to navigation effort for the trial contract is in the spring experience. Buoy work is accomplished with a and fall, the company can employ 10 to 15 people company owned crane barge and tug. The project yew round. Although they did not,have any manager for the contractor is a retired, Coast badqjround in aids to navigation, they have Guard E-7, and the project supervisor is a obtained the training necessary to properly retired Coast Guard Warrant Officer. To date, position and service the aids. To some extent, the work has been excellent; however the number this has been through trial and error using the Of discrepancies reported has risen Coast . Guard provided Computer Assisted significantly, with the majority being reported Positioning (CAP) program. The contractor was by the contractor. provided the program as part of the contract, but had to procure necessary hardware. The A summary of the quarterly reports submitted to contractor presently uses a commercial distance date indicates: measuring system for positioning. The contractor is performing quality work. 1359 A summary of the quartly reports submitted by November 1987, and found both bidders qualified the Fifth Coast Guard District indicates the to perform the work. following: At the final analysis,the Coast Guard cost was -The contractor has performed most work to the considerably less than the lowest bid so the same standards as the Coast Guard. solicitation was cancelled. -The contractor has not been as diligent as the THE INSIDE PASSAGE OF THE VIRGINIA INTRACOASTAL Coast Guard in reporting the location or time WATERWAY for changes to the aid system. This contract area is serviced by Coast Guard -Initially slow in responding to discrepancies, Aid to Navigation Team(ANT) Chincoteague. Some SRA discrepancies were acknowledged and allowed to exist for two months when the terms Initially, the trial contract included 389 SRA of the contract state all discrepancies must be in non-critical waterways. THis represented 93% corrected no later than seven days after report. of the unit workload (7personnel). The original contract included 5 lighted buoys 32 -The contractor required approximately three unlighted buoys, 147 light structures and 205 months to become accustomed to procedures and daybeacons. The RFP was scheduled for release reports. Slow in submitting complete paperwork. in April of 1987. As a result of the earlier contracting experiences, a decision was made in -Only one public comment was received. The early 1987 to contract the entire unit's SRA inquiry concerned the removal of seasonal aids workload. This allows the cost for an entire as listed in the Light List. The complaintant unit to be compared ith the bid costs,similar stated the contractor had removed aids at the to the ANT Rio Vista contract. Due to this scheduled time vice leaving them until they were decision, and the Coast Guard's district threatened by ice (which was the OG practice). reorganization, the RFP was delayed until July 1988. As of the date of this article, no THE SACRAMENT AND SAN JOAQUIN RIVERS IN details regarding response to the RFP are CALIFORNIA available. The contract for this waterway offered a clean SUMMARY analysis of the entire work performed by a unit From the beginning, the contract included the The time necessary to prepare, consider and total area covered by ANT Rio VIsta (four award each contract is summarized below: persons). Sixteen additional buoys were added from those serviced by the buoy tender BLACKHAW Timelines in order to form a more representative vross- section of the aid population. The contract District Contract includes 216 aids consisting of 5 lighted buoys, 16 unlighted buoys, 184 structures, and 11 Initiation Opt.1 Opt.2 Termination daybeacons. 1 Jan '87 Oct '87 Oct '88 30 Sep '89 This contract was delayed due to a reorganization of Coast Guard districts, and 5 (NJ)Aug '87 Oct'87 Oct'88 30 Sep '89 specifically, the dissolutiion of the Twelfth Coast Guard District, which has been handling 5(VIP)Nov'88 Oct'89 Oct'90 30 Sep '91 the RFP. Bids closed in October 1987. Three initial responses were received, however only 11 Solicitation canceled two bidders provided the required technical information. An evaluation board convened 17 13 Jan '87 Oct '87 Oct'88 30 Sep '89 The costs associated with each contract, except the Virginia Inside Passage, which is 1360 currently being caTipeted, are sLmuLarized below: FUTURE POLICY Annual Cost Data General results show contractors can perform the work, but it is also apparent that private District OG Service Best & Final Actual contractors accomplish the work less econcinically. In order to gain additional data, 1 148,608 343,516 368,448 this privatization initiative will continue and more Coast Guard units stand to face ccmpeti- 5 (NJICW) 416f694 704,000 704,000 tion for servicing our federal aids to naviga- tion. 5 (VIP) CURRENTLY IN COMPETITION 11 529,750 657,700 Not awarded 1Thomas B. Darr, Pondering Privatization May Be 13 259,693 268,023 273,310 Good For Your Ck@vernment, Gmmrnirxj, November 1987, pp. 42-50. Total lf472,856 To Be Determined Total for 3 Waterways 824,995 1,315,539 1,345,758 In general, it's become apparent there tradeoffs when a Coast Guard unit departs a waterway and is replaced by a private contractor. Search and rescue, ice-breaking, law enforcement, port security and support for national defense operations all diminish when a Coast Guard unit leaves. Coast Guard aid to navigation servicing units perform more than the singular tasks involved in maintaining this nation's aids to navigation. A benefit of contracting is the continuity which a private organization can bring by not continually transferring personnel, as the Coast Guard does. However, the corollary to this benefit is the negative aspect that the Coast Guard loses touch with the citizenry and the waterway user's needs. Perhaps the biggest benefit resulting from privatization was best stated by Mark menchik, a senior analyst for the Advisory Commission on Intergovernmental Relations: "Often one of the best results of privatization is that it forces government to reconsider the nature of the service, the rationale for providing,,lit and the means by which it's provided- 1361 MARINE RADIONAVIGATION OF THE FUTURE CDR Robert J. Weaver, USCG LT Richard M. Piccioni, USCG United States Coast Guard Washington, D.C. ABSTRACr techniques, and the use of new power sources such as solar, wind, wave and even nuclear power to name What will the mariner of tomorrow use to navigate a few. While more refinements in short-range aids with? He will use some of the same basic elements will certainly be made in the future, this paper he does today; however, technology will change the will only focus on the current trends in the way they are used. Improvements and availability evolution of long-range radionavigation. of the long-range aids will play an ever-increasing role. Radiobeacon coverage will be limited to The Transportation Systems Center is conducting a single line availability and used mainly for radionavigation user survey for the Coast Guard. homing. omega may be phased out by 2005 unless This survey will help us determine what the future retained as a backup to satellite navigation. radionavigation user needs will be, what systems Loran-C will be around until 2015 and maybe longer. mix will best meet those needs and when we can Transit will be phased out by 1996. When the phase out older systems. However, until the survey Global Positioning System becomes operational, a is complete we can make some educated guesses about new era will begin and may prove to be the answer the future of radionavigation. obviously, new to all of the mariners' navigation needs. However, satellite systems will have the greatest impact, there is international sentiment to retain a but existing systems such as Radiobeacons, Loran-C terrestrial system to complement the emerging and Omega will not disappear over night. What is satellite systems. Will the prudent navigator the status of these radionavigation systems? What require both terrestrial and satellite systems? can we expect of radionavigation in the early 21st The navigator will continue to find his century? destination. The navigation systems of the future will help him get to his destination more 2. RADIOBRACONS economically and safer. Marine radiobeacons operate in the 285 to 325 kHz band. While their use as a primary radio aid may be declining, they are still used extensively 1. INMDUCTION around the world. Employed usually as a back-up navigation system on larger vessels, they are Modern technology promises a broad range of sometimes the only radionavigation equipment on improvements in marine navigation... accuracy, smaller recreational boats. When first introduced, availability, and reduced operator costs.. : to name their long-distance, all-weather capability, a few. What are the current trends in marine provided tremendous improvements in marine navigation and what will be the mix of marine navigation. Today there are a number of other navigation aids beyond the year 2000? As we look systems available that have made radiobeacons at the evolution of marine navigation we see that rather insignificant in comparison. Still, they we have come a long way from crude beginnings. provide a cheap and reliable radionavigation aid Even so, many of the old techniques are still meeting the needs of many mariners and can be used effective and used today. As new methods and as a back-up for other, more sophisticated systems. systems were developed, many of the useful aspects of the older methods were retained, new techniques The Coast Guard currently operates approximately often augmenting existing ones rather than 200 marine radiobeacons. A number of changes to replacing them. While some of the old marine the U.S. radiobeacon system are being made to navigation techniques eventually disappeared, improve efficiency. Recent studies have shown that others continue to serve the mariner well. Today radiobeacons are now used mostly for homing. The we have a broad range of marine navigators, from system will be reconfigured with emphasis on the very small pleasure boater with little or no providing a homing service to the user. Sequenced navigation equipment to very large merchant vessels radiobeacons will be eliminated. The operating and military ships with an array of modern frequencies.of some radiobeacons will be changed. navigation equipment at their disposal. The output power of some will be reduced and many hard to service remote sites will be eliminated. There have been many developments in the last few With these changes the U.S. radiobeacon system will decades in short-range aids to navigation. New be more reliable and cheaper to operate. Since the materials used in buoy construction, new mooring receivers are relatively inexpensive, we can expect 1362 United States Government work not protected by copyright that there will be continued user demand for the Loran-C/Chayka Joint Chain radiobeacon system. Additionally, the use of radiobeacons to transmit differential information At the U.S./Soviet summit in May of 1988, Secretary for other radionavigation systems is being of State Schultz and Foreign Minister Shevardnadze considered. We expect to see radiobeacons to be signed an agreement for our two countries to create part of the radionavigation mix for some time to and jointly operate a Loran-C/Chayka come. radionavigation chain in the Bering Sea. Chayka is a Soviet radionavigation system where the operation 3. LOVAN-C and signal format are similar to Loran-C. This joint chain will improve marine navigation in the Initially a military navigation system, over the Bering Sea and Northern Pacific Ocean. It will years Loran-C has gained a high degree of user demonstrate the utility and compatibility of the acceptance within the maritime community. As two systems, and the service it will provide will coverage increased and receiver costs dropped, the be available to all who wish to use it. There number of Loran-C users grew quickly. The remain differences between the two systems which reliability, accuracy, cost and coverage advantages must be resolved if the chain is to be useful. The of Loran-C have lead to a continued, steady growth agreement calls for subsidiary arrangements to within the marine community and, more recently the create coordinating committees and to develop aviation community and land users have discovered operation procedures and other working its advantages. Today there are over 300,000 arrangements. Marine and 50,000 Aviation users world-wide. Creation of the the joint chain is to take place in Loran-C is now firmly in place around the world and two stages. Stage one, will link the U.S. Loran-C provides extensive coverage in the northern station at Attu, Alaska with the Soviet Chayka hemisphere. The Coast Guard operates fifteen stations at Petropavlosk and Aleksandrovsk. Loran-C chains in the U.S. and overseas. Some of Petropavlosk will be the master station, our overseas stations are operated by the host Aleksandrovsk and Attu will be the secondary nations under bilateral agreements. Other stations. Our goal is to complete stage one by the countries, including; Canada, Egypt, France, Saudi end of 1990. Stage two, to be completed within two Arabia and the Peoples Republic of China, are years after stage one, will replace the Soviet operating their own Loran-C stations. The Soviets secondary station at Aleksandrovsk with a station have a system they call "Chayka" which is very at Kurilsk. (The Kurilsk station is currently, a similar to Loran-C. In fact, the U.S. and the small, tow-power station which the Soviets will U.S.S.R. recently signed an agreement to create and upgrade before adding it to the joint chain). jointly operate a Loran-C/Chayka chain. One goal of the agreement is to demonstrate the compatibili- Improved Loran-C Timing ty of the two systems. The "Airport and Airway Safety and Capacity While the Coast Guard is planning to cease Expansion Act of 1987" was signed by President operation of our overseas Loran-C stations in 1994, Reagan on December 30, 1987. This act requires a number of nations are considering retaining the that U. S. Loran-C master stations maintain stations when we leave and even adding new ones to synchronization to Coordinated Universal Time (UTC) increase current Loran-C coverage in their area. within 0.1 microseconds. Currently, the master This is particularly true in northern Europe, but stations are held to within 2.5 microseconds of other nations have shown an interest in retaining UTC. The act also calls for a study of the impact or expanding Loran-C coverage as well. Thus, the of synchronizing secondary stations to within 0.1 outlook for Loran-C in the foreseeable future, is microseconds of UTC. Both the synchronization of bright. Even in light of the newer satellite the master stations and the study of the secondary systems, the present Loran-C user population is stations are to be completed by September 30, 1989. large and is expected to continue to grow. The Improved master station timing may benefit Precise Coast Guard currently has several active projects Time and Time Interval (PTTI) and aviation users, which could have significant impact on the future but may have little effect on marine navigation of Loran-C. users. The full user-impact of this law cannot be determined until the study is completed. Mid-continent Expansion Project 4. OMEGA The Mid-continent Expansion Project (MEP) is currently the major Coast Guard Loran-C project. Omega is an all-weather, global radionavigation Two new Loran-C chains, called the South Central system consisting of eight transmitting stations U.S. (SOCUS) and the North Central U.S. (NOCUS) are located throughout the world. It is the only being built to provide Loran-C coverage in the system that currently provides worldwide coverage. central United States. This is an aviation-related The Coast Guard operates two of the stations. The project and will have little impact on marine other six are operated by partner nations. While users. Funding for the project is being provided the world-wide coverage of Omega is the major by the Federal Aviation Agency. Both of the new advantage of the system, the 2-4 mile accuracy chains will be a mix of new stations and existing limits its utility. The accuracy provided by Omega stations. The new stations will be located in is sufficient for most open ocean navigation but Havre, Montana; Las Cruces, New Mexico; Gillette, not for coastal navigation applications. Accuracy Wyoming and Boise City, Oklahoma. can be improved using differential techniques but 1363 there has been little desire for their development 6. SUMKARY and implementation. The Department of Defense plans to phase out military use of Omega by the end Many of the radionavigation systems used today will of 1994. Complete phaseout of Omega will probably be around well into the 21st century. Radiobeacons occur by 2005 unless there is renewed emphasis by might be used to transmit differential information civil users or it is identified as a terrestrial to improve the accuracy of other systems. Loran-C back-up for satellite systems. In the meantime, no is well established as the preferred terrestrial significant changes to the current Omega operation based system. Many countries are planning to and eight station configuration are anticipated. continue operation of existing overseas Loran-C Considering the near world-wide coverage of Omega, stations. Cooperative international efforts to it is a relatively inexpensive system to operate. build and operate new Loran-C chains are being However, existing Omega transmitters and antennas discussed. Recapitalization and user demand for are over 20 years old. The system will become Omega will determine the future of the system. increasingly difficult to maintain without Transit will be phased out by 1996. GPS, when recapitalization. The Recapitalization issue is fully implemented, will provide the mariner with currently being studied. The results of this study complete world-wide coverage, 24 hours a day. We will bear significantly on the systems longevity. can expect that as has happened in the past with other systems, with a drop in receiver cost, the 5. SATELLITE SYSTEMS number and type of users of GPS will grow quickly. Clearly, the radionavigation needs of the Mariner Space-based navigation systems currently hold the will be well served into the 21st century. greatest promise for the future. Older systems, such as Transit, are being phased out in favor of 7. REFERENCES newer, more accurate systems. Several countries, including the U.S. and the Soviet Union are 1. Federal Radionavigation Plan, Department of building new satellite systems. Commercial systems Defense and Department of Transportation, DOD are also being implemented. Global Positioning 4650.4, DOT-TSC-RSPA-87-3, Washington, D.C., 1986. System (GPS), the new U.S. satellite system, is expected to be declared operational within the next 2. Captain Nelson H. Keeler, USCG, Maritime Future five years. Navigation Needs and Plans, NAVIGATI-0-N, -Journa-lof the Institute of NaviiiCi-on, Vol. 34, No. 2, Winter Transit 1987, pp. 290-296. Initially established as a military system primarily for support of the Navy Fleet Ballistic Missile Submarines, civilian users of Transit now greatly outnumber DOD users. Today there are approximately 80,000 commercial sets in use. The system is operated by the Navy and currently consists of six active satellites and one spare. Transit provides worldwide coverage but is limited by periodic availability. As GPS becomes operational, Transit will be slowly phased out. There are no plans to continue Transit beyond 1996, either as a Navy operated system or as a civilian system. Global Positioning System The Global Positioning System is being touted as the navigation system of the future and indeed, promises to revolutionize navigation. When fully operational, in about 1992, it will provide accurate, full-time, world-wide coverage. Operated by DOD, GPS was designed and is being implemented for military use but will be available to civil users. However, for reasons of national security the accuracy available to civil users will be limited to less than the system is capable of providing. The accuracy provided to civil GPS users may not be adequate for some marine applications such as, harbor and harbor approach and inland waterway navigation. Studies are being done to determine what those accuracy requirements are and how they can beat be provided. Methods being explored include differential Loran-C and differential GPS. 1364 FEDERAL RADIONAVIGATION PLAN OVERVIEW Larry V. Grant Research and Special Programs Administration DeparbTent of Transportation Washington, D.C. 20590 Abstract Significant changes to the Federal Radionavi- operations. The Federal Aviation Administration gation Plan (FRP) were made in the 1986 edition. and the U.S. Coast Guard ccrnpleted studies in 1969 A brief review of FRP history will be presented to that eventually led the Departrrent of Transporta- establish the foundation for the current federal tion to the development of a national navigation radionavigation plan policy statement. Special plan. The first edition of this plan was the DOT emphasis will be given to the Global Positioning National Plan for Navigation (NPN) and was pub- System (GPS) and its impact on other systems, lished in May 1970. Military navigation require- especially long-range, ground based systems. ments were published in the Joint Chiefs of Staf f Highlights from the radionavigation users confer- Master Navigation Plan and were not included in .ence held in Washington, D.C. in March 1988 and the NPN. The first NPN set national policy, other user oriented actions will be discussed. established requirements and presented a basic Current and future federal requirements for operating plan for civil radionavigation systems. radionavigation systems will be addressed. The last NPN (third edition) was published in 1977 and was a major refinement of the 1970 and 1972 NPN's and the 1974 NPN annex. 1. INIRODUCrION The General Accounting Office (GAO) report to The formation of the Department of Transpor- the Subcarmittee on Coast Guard and Navigation, tation in 1967 provided the impetus to develop a Carmittee on Merchant Marine and Fisheries, U.S. coordinated national navigation plan. With the House of Representatives, in March 1974, titled Federal Aviation Administration and the Coast "Summary of GAD Study of Radionavigation Systems: Guard providing navigational services in the Meeting Maritime Needs" endorsed Coast Guard plans aeronautical and maritime environments respective- that would designate: ly, the planning and studies needed to formulate a national plan proceeded fzat the humble beginnings 0 Omega as the most cost-ef f ective of the National Plan for Navigation of the 1970's solution to the medium accuracy and to Federal Radionavigation Plan that we know worldwide coverage requirements for all today. Early long range radionavigation systems users on the high seas. were all ground based and the Transit system's enormous success showed the potential for a 0 Loran C as the radionavigation system satellite based full coverage system. The that can best satisfy the precision emergence of the Global Positioning System navigation requirements resulting from provided the push for better planning and control heavy traffic in the coastal confluence over the presumed prolif eration of radionavigation and harbors and estuaries, thereby, systems and emerged into the force that produced eliminating the requirement for Loran A. the Federal Radionavigation Plan. With the Department of Defense setting the pace for the Additionally, this report noted that space satel- Global Positioning System for national defense lite systems were not economical or readily purposes, users and manufacturers began to rally available for civil users. With Ioran-C satisfy- and demand a mix of radionavigation system that ing civil maritime navigation requirements, Loran- would include ground based as well as satellite A would be phased out, writh a period of two years system. dual Loran-A and Loran-C operations being reasona- ble for equipment amortization and changeoverl 2. HISTORY OF THE FEDERAL This GAO study of radionavigation systems was RADIONAVIGATION PLAN undertaken because of the presumed proliferation of such systems. Recognition of the differences In the later half of the 1960's the Depart- in requirements between civil and military ment of Transportation and the Department of aviation and maritime ccmmunities and the poten- Defense saw an increased need for better coordina- tial number of land users highlighted the need to tion of racLionavigation systems development and standardize on the minimum number of long-range radionavigation aids. 1365 United States Government work not protected by copyright The NPN Amex of July 1974, announced the ment-wide navigation plan to the Congress. An designation and implementation of Loran-C as the interagency working group, already studying government provided radionavigation system for the radionavigation planning, developed the first U.S. coastal/confluence zone and the deactivation edition of the FRP, jointly published by DOD and of the Loran-A radionavigation system. The annex DOr in 1980. This new national plan contained provided for a five year phase out period for policy for national defense as well as civil Loran-A and a two year dual operation of Loran-A. policy for radionavigation systems. The most and Loran-C. significant goal of the plan called for a 1986 decision an optimum radionavigation system mix. The last edition of the NPN was promulgated. This 1986 decision was to be based on the GPS in Novenber 1977. Based on reconTendations in the becoming operational in 1987. The system mix March 1974 GAD report and the wrk being done by decision was to be made based on coordination and the Office of Telecommunications Policy, this last consultation with all groups affected by the NPN came closer to being a national radionaviga- planned 1986 decision. tion plan for civil users than those previously. The plan was prepared and approved by the Depart- The GAD published another report in September ments of Transportatim, Defense and Camerce and 1981 titled, "DOT Should Tlerminate Further Loran-C the National Aeronautics and Space Administration. Development and Modernization and Exploit the The emeiging technology of the Global Positioning Potential of the Global Positioning System." This System was included and showed great prcmise as an report was highly critical of Coast Guard plans all weather truly worldwide radionavigation for operating and improving the Loran-C system to system. the year 2000 when the GPS was being deployed to provide coverage by 1986. The March 1978 GAO report "Navigation Plan- ning--Need for A New Direction," was more ccapre- The second edition of the FRP was published hensive than the 1974 study and recommended a in 1982 and recognized the need for input from government wide radionavigation plan along with a sources other than the Federal Government. reduction in the number of federally operated Liaison with the private sector was initiated by systems. The report relied heavily on the poten- the Research and Special Program Administration tial of the DOD Navigation System using Timing and (RSPA) of DOT. Ranging (Global Positioning System) to replace most existing long-range radionavigation systems. The 1984 edition of the FRP provided the following preliminary policy on an optimum system The International Maritime Satellite Telecom- mix: munications Act, Public Law 95-564 of November 1978, required the President to sub-nit a govern- -DOD phase out military air use of Omega and THIS PHASE OF THE PROCESS DEALS WITH ACTIONS TO BE TAKEN BOTH INDIVIDUAL SYSTEMS THIS PHASE DEALS N THIS PHASE DEAL WITH AND THE SYSTEM MIX WITH THE SYSTEM MIX ;NDIVIDUAL SYSTEMS Intergovernmental Consultation & Agreement Considerations qui,ed : CAD as. :Oystems :MO NATO/Allies Upgrade as Needed Use, Req"lemonlsl National Decision current Systems Status FRP --- W-1 Review FRP Non Exc as systems Plan Phase Out/Over Intragovernmental ___N_ Consultation I & Approvals A national decision occurs upon approval of requested budget authority FIGURE I DOD/DOT RADIONAVIGATION SYSTEMS PLANNING PROCESS 1366 overseas Loran-C by 1992, VOR/DME and land-based a need for redundancy will be taken into consi- TACAN by 1997 and cease TRANSIT operation in 1994 deration when warranted by user requirements. -Civil user phase out of Loran-C and Omega Although not explicitly stated in the FRP, after certain problem with GPS were resolved redundancy of radionavigation systems, strongly -A 15 year transition period for phase out of expressed during the 1986 and 1988 users confer- Loran-C and Omega as GPS became operational ences, is considered extremely important for -Resolution of international commitments. safety of navigation. Figure 1 is noteworthy also because of the lack of a tine line driving the 3. THE 1986 FEDERAL RADICNAVIGATION PLAN distinct phases of the planning process. The planning process now extends over the two year The 1986 FRP, the fourth edition, is struct- period between subsequent editions of the FRP. urally a major revision of the 1984 edition. The 1986 FRP consists of one volume containing four The current policy is: chapters and several appendices. The FRP is better organized and easier to read. Redundant materials -DOD phase out military air use of Omega and have been eliminated. However, the basic content overseas Loran-C by 1994, VCR/DME and land-based of the FRP is consistent with previous editions. TACAN by 1997, and cease TRANSIT operation in To encourage wider audience participation in the 1996. Additionally, DOD will phase out military civil review process of the FRP, the request for use of ITS private sector input has been placed in the pre- -Civil user phase out of VCR/DME, Wran-C and face. Omega continues to depend on resolution of certain problem with GPS In the discussion to follow, I will cover -Establishment of a 15 year transition period only those areas where FRP content has changed significantly from the 1984 edition. Editorial -Resolution of international commitments. changes or differences in presentation not accompanied by substantial information additions Factors Affecting Selection of the System Mix or deletions will not be addressed. Theref ore, the reader should consult the FRP directly to The selection of the system mix has not been clarify specific areas of interest or concern. made for several reasons. The GPS is not opera- tional and will not be operational for. several Objectives more years. Since a decision on whether to phase out a radionavigation system can have serious The need to consolidate and reduce the number consequences, including increased costs for the of radionavigation systems is still recognized as users of the system, a delayed decision is the major objective of the FRP. However, due to obviously the best choice. When other factors are delays in the GPS operational status and comments considered, such as GPS satellite launch schedule from radionavigation system users, a recommenda- delays, changes in user profiles, changes in tion on the optimal mix of systems has been dynamic radionavigation technology, and input deferred until after GPS becomes operational and received during the 1986 civil radionavigation significant questions relating to GPS operation users conferences it becomes readily apparent that have been satisfactorily resolved for all classes government concern for maintainable, economical, of users. The 1986 FRP has updated the 1984 and accurate navigation system that meet current preliminary decision based on current system user requirements would force a delay in selection plans. Thus, improvements in existing system of the system mix. Since any reccmTendation on an operating characteristics is not precluded since optimal system would be premature before GPS system lifetimes are most probably in excess of operation for all user classes can be verified, fifteen years. there will be no optimal system recommendation until the GPS is operational. Policy Considerations GPS satellite launch delays are a direct A major change in selecting systems to be result of the shuttle accident. The changes in part of the system mix is shown in f igure 1. users, profiles are attributable to: an expansion Previous editions of the FRP referred to a DOD/DOT of the role of Loran-C in aviation, vehicle f inal recommendation on selection of navigation location, and timing synchronization applications. systems of the future. The current policy is more Dynamic electronics technology has resulted in the dynamic since there is no f inal recommendation but largest increase in users because of the ease of rather each revision of the FRP provides an operating neweer receivers. Automation of receiver updated decision on the current system mix. This functions, dramatic decreases in receiver costs, process now allows system to be selected based on internal receiver databases, and use of software user need rather than forcing an artificial to provide additional capabilities (automatic selection of systems based on Global Positioning calculation of latitude and longitude, way point System availability. calculations, course to steer, etc.) have resulted in a significant increase in the number of users, Indeed, the 1986 FRP refers to a "long-term especially in the marine and land community. goal to establish, through an integrated DOD/D01T Navigation management systems have introduced planning and budgeting process, a cost-effective, additional complexity into the system mix selec- user-sensitive, mix of systems for the post-2000 tion process, especially for aviation applica- time frame. " Thus, a mix of systems implies that tions. Digital signal processing has resulted in 1367 easier receiver operation, less need arid demand location with good results. Other commercial ac- for operator training,. and a quantum hq=vment tivities are engaged in the monitoring of vehicle in receiver performance. Strong support for re- position using private or federal systems. taining Loran-C was evident at the civil users conferences held in 1986, especially from the Uses Other than Navigation marine camunity. The FRP recognizes applications other than navigation: these applications include radioloca- Role of the Private Sector ticn in two form (surveying/site registration and automatic vehicle monitoring/ location) and Private sector involvement in radionavigation time/frequency dissemination. These applications services is briefly discussed in the 1986 FRP. will further enhance transportation safety, There are no recommendations on the role of the especially in applications dealing with shipments private sector in providing radionavigation of hazardous cargoes and improved coordination and services, but several factors are mentioned for response to accidents or disasters. other areas consideration. The Federal Ccmmuiications Ccnzrds- of radionavigation use showing definite economic Sion (FCC) has authorized commercial Radiodetermi- benefits include wildlife migratory, forestry nation Satellite Service (RDSS) and such a service conservation and crustal motion studies. could provide navigation capability, although it is unlikely that RDSS will replace the federal 4. CURRENr AND FU`ITJRE STATUS government's role in providing radionavigation OF NAVIGATION SYSTEMS services for basic safety of navigation. RDSs systems differ from federal radionavigation Loran-C system in that a one or two way ccrmmnicaticns channel is generally supplied by the RDss system, The domestic Loran-C system expansion in the thus providing fleet management capabilities, the mid-continent continues on schedule, with full major RDSS advantage over a pure radionavigation operation planned by 1990. Military requirements system. Sam of the f actors remaining to be for overseas Looran-C will cease in 1994. Present examined to ensure the appropriate future roles of government plans are to turn operation of these private and federally operated radionavigation overseas stations over to host nations where there systems am: is such interest and to shut down the Hawaiian Loran-C chain. The domestic Loran-C system should 0 Privately-operated service's inipact on be in operation well into the next century. usage and resultant demand for federally Loran-C user population is expected to increase operated services. substantially by 1991 as a result of new land 0. Resolving competition related questions based applications and an increased acceptance by between private sector services and a the aviation commuiity. , Comments from the 1986 federally provided free service. and 1988 users conference indicates an increased � Possible private sector operation of interest and acceptance of the Looran-C system, phased-out federally-operated services. especially in land based and timing applications. � RDSS liability in providing navigation Differential Loran-C system are in use today. services and the federal government's The Coast Guard is using differential Loran-C to regulatory role. set aids to navigation. State governments are using such systems to monitor ocean dumping These factors could have a significant impact on practices and procedures. f uture radionavigation policy and a detailed examination of the relationship between federal Omega systems and RDSS must. be undertaken. However, since RDSS is not presently fully operational, The Omega system will see Sam modernization questions relating to these factors cannot be of station equipment, but no significant change in properly answered at this time. the system configuration is planned. Use of the aviation trust f und to pay for Omega system Land Navigation improvements was mentioned during the 1988 users conference. Projections of the Omega user popula- Land navigation applications are recognized tion shows no change or only a slight increase in for the first time, however requirements are still aviation users through 1992. Military air in the developmental state. Also there is no requirement for Omega will cease in 1994, but some designated federal agency to represent the land naval receivers will remain in operation after navigation users. The Research and Special 1994. Any decision to phase out the Omega system Programs administration is coordinating these is not expected to be announced until GPS has been requirements with other federal agencies and civil operational for at least three years and only if users. The major application is in vehicle all requirements currently being met by the Omega navigation system, however requirements are st.111 system are satisfied by GPS. During the 1988 in the developmental stage. users conference, a transition period of less than Automobile manufacturers are studying the use fifteen years was mentioned. of GPS or Looran-C based systems for intra-city GPS navigation of automobiles. A differential Omega system has been used for automatic vehicle The first GPS operational satellite is 1368 scheduled to be launched in November 1988. The Cnly me user conference was held to obtain 21-satellite constellation is scheduled to be in civil user input on the 1986 FRP. This public place by 1992. Thus GPS should be fully operation- hearing was held in conjunction with the annual al in 1992 in three dimensions for military and FAA user conference in Washington, DC during March civil users. Studies have shown that a constella- 1988. CtnTrents from civil users of radionaviga- tion of 24 operational GPS satellites is one means tion system can have a significant impact on the to meet civil aviation requirements for sole-means contents of the FRP and accordingly, active navigation in the national air space. Funding for participation by the civil sector in providing additional satellites is being considered. A input into the FRP is highly encouraged. Although Department of Commerce study estimates a 1995 private sector input cannot be guaranteed to worldwide civil GPS user population of one-half directly influence the national decision on million, a most optimistic figure, unless a low radionavigation policy, consideration of civil cost GPS receiver is available. A request' from user comments and opinions are very important in the timing and frequency community of GPS users to the overall planning process to ensure that all leave two GPS satellites on orbit with no selected user requirements are properly satisfied. availability is being considered by the GPS Steering Committee. A 24 satellite constellation Projection of user populations can markedly will, solve the GPS coverage problem for aviation affect decisions relating to continued operations users. of any radionavigation system. 'User equipment costs will continue to have a direct impact on any 5. CURRENT AND FUTURE RADIONAVIGAMCN PIANNIM future reccmTendations relating to an optimal radionavigation system mix. The ef fect of an GPS will become operational in 1992 when 18 operational GPS on radionavigation systems will satellites and three spares are on orbit. Since not be fully known until after several years of GPS receivers are not currently economical for GPS operation; even then, redundancy requirements most general aviation and the majority of marine could preclude the shutdown of particular system. users, civil use of GPS will be limited to special user groups. When economical GPS receivers are available, probably within three to five years of 6. CONCLUSICNS GPS being declared operational, the number of GPS users is expected to increase dramatically. The 1986 FRP did not select an optimum mix of Discontinuance of federal funding for dcaTestic radionavigation systems because of the rapid Loran-C or worldwide Omega service, in lieu of GPS changes occurring in today's technology. This service, is not likely to occur until sometime in rapid technological change is evident in Laran-C the twenty-f irst century and not until the receiver costs, some now available below f ive following issues are resolved: hundred dollars. For GPS to successfully compete as a civil radionavigation system, receivers -Resolution of GPS accuracy, integrity and costing less than me thousand dollars mast be financial issues readily available to the marine community. The -GPS meeting the needs of civil air, marine, decrease in receiver costs has been of significant and land users'currently net by existing systems importance in fostering the large increase in -Economical GPS receivers becoming available Loran-C users. Reduction in receiver costs for -Resolution of international commitments the other radicnavigation systems is probably -Designation of an appropriate transition imminent. Economical receivers will attract more period for any phased out system. users to a particular system, causing further difficulties in the decision making process to Another matter which must be reconciled is the reduce the number of radionavigation systems. selection of radionavigation system by DOT to With the selection process calling for a mix of meet civil requirements. Although the FRP policy systems to meet user requirements, the number of. statement has, for several issues, indicated that users of a particular system will be a major DOT might decide to discontinue federal operation obstacle in deciding on the final system mix. of loran-C and Omega for the reasons cited above, recall that the selection process calls for a With a dynamic process for fonitalating policy cost-effective, user-sensitive, mix of systems for decisions for selection of the optimal mix of the post-2000 tine frame. Thus one cannot rule federally funded radionavigation, systems, civil out a civil user requirement for redundant systems system user populations will play a major part in based on the fundamental rule that the prudent shaping the biennial FRP policy statement for navigator not depend on one system or method for domestic systems. The FRP WIll continue to be all navigation needs. updated every two years, allowing for civil review and influence in the radionavigation system The selection process provides for biennial selection process. Decisions on the phasing out reviews of the current national decision, with any of current radionavigation systems will not occur changes being promulgated in a revised FRP. This until after GPS has been operational for several process was applied to the preliminary decision in years and a solid base of GPS users is in place, the 1984 FRP to reach the decision on the current along with a concurrent decrease in users of other mix of system . Because the questions relating to systems. Thus, the middle to late 1990's is the GPS issues cannot be fully answered until the earliest possible date to validly change the early 19901s, no change in the current mix of current radionavigation system mix for the post system is anticipated prior to 1992. year 2000 time fraire. 1369 7. REFERENCES 1. National Plan for Navigation, May 1970, Department of Transportation. 2. National Plan for Navigation, April 1972, Department of Transportation. . 3. National Plan for Navigation Annex, July 1974, Department of Transportation. 4. National Plan for Navigation, November 1977, Department of Transportation. 5. "Summary of GAO Study of Radionavigation Systems: Meeting Maritime Needs, " B-180715, March 26, 1974, General Accounting Office. 6. "Navigation Planning--Need for a New Direc- tion," LCD-77-109, March 21, 1978, General Accounting Office. 7. "Radionavigation Action Plan Summary," DOT- RSPA-DPB-20-79-1, April 1979. 8. "NAVSTAR Should Improve the Effectiveness of Military Missions--Cost has Increased," PSAD-80- 21, February 15, 1980, General Accounting Office. 9. "DOT Should Terminate Further LCRAN-C Develcp-- ment and Modernization and Exploit the Potential of the NAVSTAR/Global Positioning System," MASAD- 81-42, September 18, 1981, General Accounting Office. 10. Federal Radionavigation Plan, July 1980, Volumes I-IV, DOT-TSC-RSPA-80-16, DOD-NO. 4650.4-P. 11. Federal Radionavigation Plan, March 1982, Volumes I-IV, DOT-TSC-RSPA-81-12, DOD-4650.4-P. 12. Federal Radionavigation Plan, 1984, D0T-TSC- RSPA-84-8, DOD-4650.4. 13. Federal Radionavigation Plan, 1986, DC)T-TSC- RSPA-87-3, DOD-4650.4. DISCLAIMER This paper expresses the views of the author and are not necessarily those of the United States Department of Transportation. 1370 COOPERATIVE ELECTRONIC CHART DEVELOPMENT THE GAADS PROJECT William M. Maynard Nautical Charting Division Charting and Geodetic Services National Ocean Service, NOAA Rockville, Maryland 20852 ABSTRACT GAADS, an Electronic Chart Display System (ECDIS), is menu driven and easy to operate. It allows the The National Ocean Service (NOS) is currently user to display and analyze multiple vessel supporting the development and implementation of positions, aids to navigation, way points, search the Graphical Analysis, Archiving and Display patterns, and composite data against digitized Station (GAADS) of the United States Coast Guard nautical charts with scales ranging from 1:20,000 (USCG) through the provision of electronic chart to 1:30,000,000. These charts are loaded using prototypes. NOS is working closely with the USCG standard chart numbers for easy identification. and is assisting them in the determination of data GAADS uses an algorithm to display the latitude- requirements and feature display density by produc- longitude graticules on a Mercator Projection.3 ing a number of digital nautical chart specific data files. In addition, NOS is developing an 2. COMPUTER SYSTEMS efficient method for producing digital data files for the entire suite of NOS nautical charts, if In order to understand the compilation process of a required. This paper will describe the GAADS GAADS nautical chart, it is necessary to identify system, data preparation, and prototype production and describe the computer systems that support the techniques employed in the compilation of GAADS process. The Intergraph Digitizing System is used charts. to compile the graphic portions of a nautical chart digital data file. This System consists of a PDP 11/70 central processor, three 300 megabyte disk 1. INTRODUCTION drives, and seven graphic workstations. Digital data collected at the Intergraph is processed using In 1986, the United States Coast Guard (USCG) a Univac 1100 computer system located in Suitland, Office of Research and Development in Washington, Maryland. The software available on the Univac 1100 D.C., requested assistance from the National Ocean gives NOS the capability of converting Intergraph Service (NOS) Office of Charting and Geodetic x and y table coordinate positions to geographic Services in Rockville, Maryland, to develop a positions (G.P.'s) or latitude/longitude. The method of producing digital navigational charts for Univac 1100 software also produces a file which is a USCG electronic piloting and buoy positioning used to plot the digital data on a Calcomp 748 system. Their prototype system, known as the flatbed plotter. These plots are used to verify Automated Aids Positioning System (AAPS), had the accuracy of the digital data collected at the proven successful according to their Office of Intergraph. An IBM-AT personal computer with a Navigation and was therefore being considered for 30-megabyte hard disk is used as the collection production. NOS, in its support of research and point for the certified data files. A contemporary development of new technology, responded to the word processing software package on the IBM-AT USCG request by agreeing to utilize its digital allows for last-minute editing of the data prior to shoreline and aids to navigation data files as a transfer to a Hewlett Packard 330 Computer. The base to fulfill the USCG requirement of producing GAADS software runs on a Hewlett Packard series digital navigational charts for the AAPS Project. 9000, model 330 computer, which has a 12-inch color The USCG agreed to provide funding for the NOS monitor and a resolution of 512 by 400 pixels. The development effort along with a loan of all associ- computer utilizes 4 megabytes of RAM and a ated hardware and software. In return, NOS would 20-megabyte hard disk. The computer has the capa- be responsible for developing, producing, and bility to quickly store, manipulate, and retrieve certifying the nautical chart data files required display screen images. The GAADS software is to produce an initial suite of digital navigational written in HP BASIC 4.0 and contains 200 separate charts. While these negotiations were underway, subprograms offering flexibility and maintenance.2 the AAPS Project was replaced by a more sophisti- The keyboard has 10 menu keys, a numeric key pad, cated project undergoing development at the USCG and a knob for positioning the cursor on the Research and Development Center in Groton, Connec- screen.1 ticut. This new project, the Graphical Analysis, Archiving and Display Station (GAADS), not only 3. GAADS DATA FILE STRUCTURE included AAPS, but also supported several other applications. In 1987, technical representatives from the USCG R&D Center in Groton, Connecticut, and NOS met in 1371 United States Government work not protected by copyright Rockville, Maryland, to discuss data structure, an Intergraph vector file containing x and y table specifications, and compilation procedures for the coordinates, is loaded on the Univac 1100. A nautical chart data files required to support printout of the shoreline file is obtained to verify GAADS. A result of this meeting was an agreement that all shoreline polygons are closed. Any that each GAADS chart compiled by NOS contain four additional editing is performed using the Univac separate and distinct data files: a shoreline 1100 text editor. The file is downloaded from the file, a linear features file, a discrete point Univac 1100 to the IBN-AT over a data line. A hazards file, and an aids to navigation file. The contemporary word processing package is used to data in the shoreline and linear features files remove unwanted Univac 1100 commands such as end of contain the x and y table coordinate positions, file marks and line feed carriage returns from the compiled directly from the Intergraph, of all shoreline file. The file is written to a text file points required to adequately depict these line in ASCII format on the IBM-AT. A vector file features. The data in the discrete point hazards containing all shoreline data with polygons closed and aids to navigation files contain the G.P.'s is complete and ready for transfer to the Hewlett required to adequately depict all point data in the Packard 330. file. All four files are structured in a vector format. The four files are combined at the Hewlett The linear features file contains a selection of Packard 330 Computer, utilizing GAADS software, to lines on the chart other than shoreline. Presently, produce a GAADS chart. In addition, all data files this file consists of 6, 12, 18, and 30-foot depth are in ASCII to facilitate the transfer of digital curves, all channel lines, spoil and anchorage data between the IBN-AT and Hewlett Packard 330. areas. This file contains nautical chart line features that do not require color fill on the GAADS 4. GAADS CHART COMPILATION monitor. This data does not reside in the NOS digital data base, so it must be digitized from the A description of each data file, its current most recent published nautical chart drawing at the specifications, and its compilation will provide Intergraph System. This digital data is processed some insight into the actual production of a GAADS on the Univac 1100 where it is converted from x and chart. The shoreline, as defined by NOS, is the y table coordinates to G.P.'s and plotted. The plot Mean High Water line on a nautical chart and is compared to the chart drawing and edited as includes both natural and man-made features. For necessary at the Intergraph workstation. The file GAADS chart compilation, this file must support is filtered to eliminate unnecessary points and color fill of all land areas displayed on the GAADS reduce the size of the file. The file is processed monitor. For the majority of nautical charts at again on the Univac 1100 and plotted to verify the NOS, a shoreline file is readily available in the accuracy of the digitizing against the most recent form of NOS/Exclusive Economic Zone (EEZ) digital published nautical chart drawing. A copy of the shoreline data sets. These sets contain the large- completed Intergraph file, a vector file containing scale digital shoreline of the United States in x and y table coordinates representing the linear G.P.'s. The first step in the compilation of a features digitized, is loaded on the Univac 1100. shoreline digital data file for a GAADS chart is to Any additional editing of the data files is per- identify which chart is to be compiled and where formed using the Univac 1100 text editor. From this the data exists within the EEZ digital shoreline point, the file is essentially treated like the data sets. The shoreline data is partitioned from shoreline file. The file is downloaded to the the data set which contains the chart, then IBM-AT, edited, and placed in a text file in ASCII converted to x and y table coordinates on the format. A vector file containing all linear Univac 1100, and downloaded onto the Intergraph features required for GAADS chart compilation is System. A raw plot of this file is compared to the complete and ready for transfer to the Hewlett most recent published nautical chart drawing, and Packard 330. the shoreline data is edited at the Intergraph workstation to bring the file up to date. A filter The discrete point hazards file consists of rocks, routine is run to reduce the size of the file by piles, islets and other hazards to navigation found eliminating all unnecessary points. The shoreline on the nautical chart outside the 6-foot depth file is processed on the Univac 1100 converting the curve. This point data does not exist in the NOS Intergraph x and y table coordinate data to G.P.'s. digital data base so it must be digitized from the This conversion is not a require-ment of GAADS but most recent published nautical chart drawing at the is used solely by NOS to produce plots to check the Intergraph. Upon completion of the digitizing, the accuracy of the data. The converted file is Intergraph data file is processed on the Univac plotted and checked against the most recent 1100, where it is converted to G.P.'s and plotted. published nautical chart drawing. Any additional The plot is compared to the chart drawing and edited editing required is performed at the Intergraph as necessary at the Intergraph workstation. The workstation. Upon completion of all editing, it is file is processed again on the Univac 1100 and essential to close all shoreline polygons because plotted to verify the accuracy of the digitizing. A this file must support color fill of all land areas copy of the G.P. file, a vector file-containing the displayed an the GAADS monitor. To close all geographic positions of all discrete point hazards, shoreline polygons, the line strings are connected is loaded on the Univac 1100. Any additional on the Intergraph. The file is once again editing is performed using the Univac 1100 text processed on the Univac 1100, plotted, and checked editor. From this point, the file is essentially against the most recent published nautical chart treated like the shoreline and linear features file. drawing. A copy of the completed shoreline file, The file is downloaded to the IBM-AT, edited, and 1372 placed in a text file in ASCII format. A vector 7. REFERENCES file containing the G.P.'s of all discrete point hazards for the GAADS chart is complete and ready 1. Lamb, Michael J., GAADS User's Manual, Prepared for transfer to the Hewlett Packard 330. for U.S. Coast Guard Research and Development Center, September 15, 1986, p. 1-1. The aids to navigation file consists of all charted aids to navigation along with the characteristics 2. Poltilove, David L., Development of Digital of all lighted aids and references to the source Chart Products, Proceedings, U.S. Hydrographic document which established the position of the aid. Conference '88, April 12-15, 1988, Baltimore, This file does exist in the NOS data base on the Maryland, p. 190-193. Univac 1100 and is known as the Discrete Independ- ent Point File (DIPFILE). It is updated as Notice 3. Vorthman, Lt. Cdr. Robert G., Jr., U.S. Coast to Mariners and other source documents are received Guard Alumni Association, The Bulletin, August/ by each nautical chart compilation area team. A September 1986, p. 32-35. software routine is run to partition from the DIPFILE, all aids to navigation which fall on the chart. The file is processed on the Univac 1100, and a plot is made of all the aids that fall on the chart. The plot is compared to the most recent published nautical chart drawing and edited to bring it up to date. All discrepancies are resolved by checking the chart history and Notice to Mariners. The characteristics of all aids are also edited and updated using the USCG Light Lists and/or Notice to Mariners. All editing of the file is performed on the Univac 1100 text editor. Upon certification of the file, the vector file contain- ing the G.P.'s of all aids to navigation for the GAADS chart, is downloaded from the Univac 1100 to the IBM-AT, edited, and placed in a text file in ASCII format. The aids to navigation file is complete and ready for transfer to the Hewlett Packard 330. 5. TRANSFER OF THE DATA FILES The shoreline, linear features, discrete point hazards, and aids to navigation files are trans- ferred to the Hewlett Packard 330 via an RS-232 interface or floppy diskettes. Upon transfer, the USCG assumes control of the NOS nautical chart data files. The vector files are transformed by the GAADS software for display on the HP-330 color monitor. The GAADS software offers the flexibility to recall the data by individual files, or in any combination, depending upon the functional require- ments of the user. The NOS nautical chart data files support the USCG GAADS chart requirements by providing the USCG with a highly accurate product which conforms to paper nautical chart production standards. 6. CONCLUSION The development of nautical chart data files for the compilation of GAADS chart prototypes continues at NOS. The initial prototype, chart 13213 cover- ing the New London - Groton Connecticut area, has been transferred to the USCG and transformed for display on their GAADS color monitor. At USCG request, NOS is currently compiling additional nautical chart data files for charts covering the St. Marys River, Michigan area. The NOS/USCG cooperative effort continues to provide the USCG with the nautical chart data files required to fully realize the capabilities of the GAADS System. 1373 AUTOMATED NAUTICAL DATA AND CHARTING DEVELOPMENT Norman D. Smith NOAA Charting Research and Development Laboratory Charting and Geodetic Services National Ocean Service, NOAA Rockville,.Maryland 20852 ABSTRACT a whole new generation of digital products would be available. The National Ocean Service (NOS) is beginning the development and implementation of an automated The NOS developed a digital plotting capability in system of nautical data evaluation, data base the late 1960's and early 1970's. In 1975, the administration, and nautical chart production. design of the Automated Information System (AIS), a This system, the Automated Nautical Charting storage,retrieval and update system for nautical System II (ANCS II), represents NOS' next genera- charting data, was begun. This system was developed tion of automation of the cartographic processes and integrated with data collection and processing involving nautical data. capabilities, and digital plotting functions, to implement the Automated Nautical Charting System The ANCS II will be utilized to evaluate and select (ANCS) in 1978. data from the many types and forms of source information received and to interactively apply The ANCS includes all the functions to process that data to a geographic, product-independent data incoming digital and analog data into the AIS data base. Data will be drawn from this data base and base and produce NOS nautical charts. A digitizing applied to a logically separate data base which capability enables analog source documents 'to be will be used to produce approximately 1,000 NOS completely or selectively converted to digital form paper nautical charts. In addition, the system for processing along with NOS digital hydrographic must support requests for digital chart products. data into the AIS system. Once loaded into the AIS, .data can.be interactively attributed to reflect While the initial product load of ANCS II will be application to NOS nautical charts. This data base chart products, the nautical data base will service is geographic and logically continuous. Data is the expected growing demand for custom digital data then extracted from the AIS according to parameters requests. Additional logically separate data bases describing the dataset or the nautical chart may be established to service new products. desired. For nautical charts, offline batch processing then constructs a plot dataset compatible 1. INTRODUCTION with one of many devices available. The ANCS includes electrostatic, flatbed vector, and laser The NOS is required by statute to publish nautical raster plotting capability to produce paper, scribe charts and related publications, and otherwise make film, and photographic film plots. available information necessary for the safe and efficient transit of the Nation's coastal waters While the production level for ANCS never reached and inland waterways. This information is neces- the initial desires, the concepts and value of the sary to support domestic and foreign commercial digital data base and automated chart production shipping, private recreational boating, government have been demonstrated and a modest level of chart regulation and defense of the Nation's coastal production has been achieved and maintained. areas, offshore resource exploration and extrac- Currently, the system fully maintains 70 published tion, and coastal zone planning. nautical charts on the AIS system with digital data collection and plotting support for all chart In the late 1960's, NOS recognized the advantages production except minor corrections and reprints. of automating the production process of the tradi- tional nautical product line and the value of The ANCS II is designed to provide the capability creating and maintaining a digital data base of the and capacity to handle nautical chart production and information represented by these products. The requests for digital data. The system will take application of digital plotting technology offered advantage of the tremendous advancements in auto- the ability to reconstruct current nautical chart mated mapping and charting hardware and software products to improve and standardize the construc- technology, and experiences with the current ANCS. tion of the chart base and the depiction of the A Nautical Information Data Base (NIDB) of data will chart symbology, and offer the capability to serve as source data for all nautical products reconfigure the chart format as desired. In addi- including nautical charts and digital nautical tion, new chart constructions could be made in a information requests, and a logically separate Chart much shorter calendar period. If a multipurpose Graphics Data Base (CGDB) derived from the NIDB will digital data base were made first to produce data specifically support nautical chart production. for this digital plotter output, then a source for 1374 United States Government work not protected by copyright 2. REQUIREMENT allows a cartographer to assess the impact of the change on the nautical chart and safe navigation Data for nautical chart application is received during the compilation process. After this from more than 60 sources including NOS, in the assessment a "Notice" may be issued. form of NOS hydrographic and photogrammetric surveys, the U. S. Coast Guard, U. S. Corps of The primary long-term goal of the ANCS II system is Engineers, and numerous other public and private to maintain the NIDB as the data source for the CGDB organizations. ANCS II must provide for the evalu- and all nautical products and directly service the ation, extraction and application of this data to growing number of requests for digital data to the NIDB according to all product requirements. support scientific and economic studies, electronic Approximately 35,000 new source documents must be charts, and any other navigational uses. A major processed annually. Of these, about 9,000 are demand for digital nautical charting data will come typically determined by an initial evaluation to from the Department of Defense (DOD) which plans to contain information affecting the NIDB. Since only store nautical data in an all digital form in the NOS hydrographic surveys, and in the near future early 1990's. The DOD and other government agencies NOS photogrammetric surveys, are received in use about half of the approximately 2 million NOS digital form, this number represents physical charts distributed each year. documents that must be processed. Data and product archiving will be tracked and The legal nature of current and anticipated NOS reported during the process. Data to assist in products necessitates maintaining a complete planning and supporting the acquisition of photo- history of the application of source data. The grammetric and hydrographic data within NOS will ANCS II system must provide tracking and status feed back to those acquisition activities. reporting to assure that data is processed and applied on schedule and provide this history of The ANCS II must contain the batch processing processing activity and disposition of the data. capability currently provided offline on offsite This capability goes far beyond current capabili- mainframe and onsite micro computers. Data is ties in maintaining an inventory of all source currently offloaded from independent systems and documents received and providing management transferred by magnetic tape or floppy disk for tracking of the progress of application of the processing. The ANCS II will have the capability to document. batch process this data as it passes between logical production steps within the ANCS II Local Area Once data is applied to the NIDB according to Network (LAN). product specifications, data to produce products or satisfy data requests can be extracted. Additional processing steps supply special overlays and graphic elements for NOS products which include The immediate production goal of ANCS II is to electronic positioning lattices, magnetic informa- support the production of nautical charts and tion, services, and boundaries of all types. Other related publications, and to provide critical processing supports data evaluation and chart compi- feedback to the U. S. Coast Guard and the Defense lation processes such as line generalization, sound- Mapping Agency (DMA) in the form of the Notice to ing selection, and hydrographic surface modeling. Mariners publications. All batch processing requirements will be satisfied within the networked ANCS II system. Data will be extracted from the NIDB to produce NOS nautical charts according to a regular printing 3. ANCS II FUNCTIONAL OVERVIEW schedule. This data enters the CGDB where it is interactively applied to previously published The ANCS II system will process selected naviga- information to yield a new chart edition. This tional data from digital and analog source form for step also requires a complete history of the inclusion into the NIDB and provide the tools to application of changes. extract that data for a multitude of products. - Currently NOS publishes approximately 980 nautical Figure 1 is a conceptual overview of the ANCS II charts to fulfill its nautical charting require- system. Two major processes take place within the ment. The design goal for ANCS II is a capability system. Data from all sources is evaluated, to produce about 500 revised chart editions each compared to existing data in the NIDB, and applied year. Nautical charts may include extensions, to the NIDB in the "manage data" process while larger scale insets, and be divided into several updating appropriate control files to assure data panels in order to adequately represent an area. and task tracking through the system. The "produce Because of this, approximately 2500 individual charts" process, logically separate from the "manage chart panels must be maintained. data" process, extracts data from the NIDB to maintain the CGDB in order to compile digital chart The reporting of critical conditions, or "Notice," images. through the U. S. Coast Guard and the DMA Notice to Mariners publications, takes place both during data In the "manage data" process an initial evaluation application to the NIDB and the CGDB. A critical will determine if source information received should change noted during source application and NIDB be added to or might affect the NIDB. Digital update may be reported immediately, but more information will be processed into a compatible typically would generate a CGDB application which format and input into the system. Analog documents 4375 source Pejected SOURCE i.0 ORIGINATOR task MANAGE DATA old supersede data data parameter parameters new data parts FEATUR ARCHIVE SOURCE DOCUMENT NAVIGATION PARAMETER INFORMATION DATA document statue task event parameters deletions TASK 2.0 additions EVENT HISTORY PRODUCE task task CHARTS event published old graphic critical feature graphi features status CHART GRAPHIC NOTICE TO MARINERS ARCHIVE HOLDING new graphic notice CHART GRAPHICS DATA USCG chartlet REPRODUCTIO DMA Figure l.--Automated nautical charting system II conceptual overview. 1376 selected will then be processed by the system. will be queried to prioritize possible CGDB updates Processing of a document entails a comparison of necessitated by chart printing schedules and the data contained to current NIDB data and data application of information critical for the safety specifications to determine changes requiring an of navigation. NIDB addition or update. This process will incorporate most of the same tools Each change or addition normally requires conver- and the same system components used in the NIDB sion of the data involved to digital form and an update but be controlled by different product speci- interactive application to the NIDB. A number of fications, procedures and rules. This process, too, the system workstations will be equipped to digi- will maintain control files to provide for manage- tize source information from analog documents. ment tracking and a permanent history of data This information is then immediately available for application or change to the CGDB. application to the NIDB. A small percentage of documents will require bulk digitizing that would The logical separation of the CGDB from the NIDB unduly occupy a workstation. This will be done on will insure that cartographic modifications to the a peripheral system or on contract. data for printed or electronic chart depiction of nautical features will not affect NIDB data usage The current system requires data to be collected but be confined to the CGDB. This concept is offline after selection and then applied to a extremely important to maintain the integrity of the database in a separate session. The ANCS II will NIDB for multiple uses. eliminate the multiple handling of documents, breaks in the process flow, and excess data These completed chart compilations contained in the collection inherent in the offline process. CGDB are then available for processing electronic chart and NOS chart graphic requests. Textual and Methods to quickly scan these sources into a raster graphic output are necessary for requirements such digital format for display and interactive visual as the Notice to Mariners publication, and graphic comparison to existing data will be investigated. chart output to support NOS traditional paper This method is very attractive as it allows a charting production and electronic chart data comparison of the entire source document before requests. resources are expended to convert and attribute individual features that may not be needed. While 4. HARDWARE it is not anticipated that entire documents will be bulk converted by this method, the majority of the The ANCS II system architecture, currently in the documents received which have relatively small planning stage, may vary from a highly centralized numbers of features selected from many could be system with database functions, batch processing and efficiently handled. Other selection methods much graphics processing residing on one or two typically require more features to be converted large processors networked to single or clustered than will be used. limited capability workstations, to a highly distributed system with many processors networked to once this process of source document assessment, very powerful workstations where database functions data selection, and data collection is complete, a and batch processes are distributed and workstations workstation session will commence to interactively operate independently. A likely configuration is apply the data to the current NIDB contents. one in which very capable workstations connected by Processing power available to the ANCS II work- a standard LAN ate distributed in the cartographic stations will allow the interactive comparison of work areas. The databases would reside on larger source data to data residing in the NIDB, selection processors on the LAN along with batch processing of data which changes NIDB contents, and applica- capability. Microcomputer "smart terminals" would tion of this data. ANCS II will contain the full be connected to the LAN for management control and range of compilation and graphic tools to assist reporting and peripheral operations such as word the cartographer in this process. Application processing, software maintenance, and routine data software such as sounding selection, automatic input. contouring, surface modeling and,line generali- zation will be used for certain applications. When The current system is hampered by obsolete system the process is complete and the necessary quality hardware. The ANCS II will benefit from tremendous checks and tracking data are complete, the data is advancements now commonplace to mapping systems. applied to the NIDB through the ANCS II Data Base The system will be coupled by a LAN allowing great Management System (DBMS). flexibility in system configuration and placement of equipment while maintaining an integral system. The "produce charts" process is the product depend- Logically the NIDB, CGDB, and any other required ent process to compile and maintain the suite of data bases will be centrally supported while work- NOS nautical charts. The ANCS II system will pro- stations distributed in the cartographic areas will vide for extracting data from the NIDB and process- function essentially autonomously. The capability ing the application of the data to the CGDB. of workstation hardware to function within typical office environments and the flexibility of physical The CGDB update task will be independent from NIDB placement afforded by LAN's allows the integration update, initiated by a cartographer/manager. of the system functionality into the workplace and Control file entries established during the NIDB the modular growth and reconfiguration to adapt to update process, and the chart production schedule production and organizational changes. 1377 The system will consist of up to 40 graphics The required applications for the ANCS II system workstations, smart terminals, batch data process- will require the bulk of the development effort ing resources, data base processing resources, and since these are highly influenced by specific NOS peripherals locally networked together. As many as specifications and policies. 15 of the 40 workstations may include a precision digitizing capability for direct data entry from 6. IMPLEMENTATION graphic source documents. Each workstation will be capable of performing all data entry, NIDB data A Request for Proposals (RFP) to procure the ANCS II base, and CGDB chart production functions. Smart system was released on January 19, 1988, and closed terminals will be used for a variety of functions on April 28, 1988. Responses to the RFP are being including data entry, development, and management evaluated as of this writing with contract award inquiry and reporting. expected in September 1988. 5. SOFTWARE The system will be developed in phases over a period of 3 to 4 years and include a prototype development The software for the system will include a DBMS, and demonstration phase, a trial production phase, Operating System, mapping routines, and applica- and a full production implementation phase. The tions. It is anticipated that the majority of the system hardware configuration will be expanded software will be commercially available, proven during the phases in four steps culminating in the products. full configuration at the end of the final phase. The DBMS will manage the NIDB, CGDB and other data 7. SUMMARY bases required for the system. While it may be possible that different DBMS software may manage The NOS plans to procure and implement the ANCS II the different data bases, one system is more system as part of the modernization of the nautical likely. charting system. The result will be a product independent data base capable of supporting a range Data will be stored in a topological structure to of products. Separate databases derived from this take advantage of relationships between features. product independent data,base will support specific Line features in particular can be interrogated product lines including NOS nautical charts and according to these relationships for automatic electronic digital charts. update, error checking, and extraction in logical groupings according to purpose. 1378 THE CONTINUING NEED FOR ACCURATE POSITIONING IN NAVAL TACTICS John L. Hammer, III Wayne R. Hoyle Qubit North America, Inc. ABSTRACT differences. Fundamentally, the word 1positioning' implies a more accurate Modern naval warfare has, over the years, knowledge of,one's position than is resulted in the relegation of accurate required in navigation. The act of positioning and navigation to a second navigation is more dynamic in that it class role. The modern naval officer includes movement from one place to worries about positioning usually only another. Positioning reflects a greater when concerned with navigation. During interest in one's position as accurately the battle he most often conducts his fixed on the surface of the earth. It is operations in a relative motion sense very often used in association with using relative ranges and bearings. The activities requiring close tolerances -once widely-respected title of navigator such as in surveying or the offshore oil has been replaced by that.of the Tactical industry. Positioning is necessary where Action Officer. Despite these lamentable the mariner wishes to accurately mark a trends, accurate positioning remains as place for return or some future action. important today as it ever was. -The trends toward over-the-horizon DECLINE IN NAVIGATIONAL AND POSITIONING targeting/battles and strict Emission EXPERIENCE Controls (EMCON), coupled with the more traditional tactics in Anti-Submarine The term 'Navigator' within the U.S. Navy Warfare (ASW) and Mine Warfare, continue has lost some of its luster over the to require good geographic positioning as years because of two factors. One, a matter of paramount importance. improved techniques and equipment have made navigation less difficult to practice but at the same time more accurate. Secondly, the evolution of INTRODUCTION modern warfare has shifted emphasis from true positioning to a relative frame of Naval warfare, as with all other reference. A navigator is generally less operations at sea, depends heavily on the interested in precision positioning positions of one's own vessel and those because his ship's motion does not allow of other contacts in the area. While him time to take full advantage of much activity is conducted in the multiple or more complex positioning relative frame of reference using strategies. bearings and ranges, it is still necessary to have accurate and current Less than a hundred years ago, the title knowledge of one's position on the Navigator was one which inspired surface of the earth. admiration. Now navigation is something that must be done to keep out of trouble. In this discussion on determining one's It's still possible to ruin one's location we need to review two reputation through bad navigation. For fundamental definitions. The first is this reason, navigation is maintained as for navigation, which is the art or a necessary but not altogether desirable science of conducting one's self from one skill. Unfortunately, the imperatives of place to another. The second is for today's fast-moving tactical and battle positioning, the act of determining one's situations have diluted the emphasis place on the earth to a specific placed on traditional navigation skills. tolerance. The modern naval officer must concentrate While the two words mean the same thing more on threat indicators and less on to many people, there are subtle finding his position. This lamentable 13@9 United States Government work not protected by copyright state of affairs appears in both the lack position to aid in the visualization of of strong and comprehensive navigation the tactical situation. Computations may training and in the lower priority given be made within the equipment to solve for its actual practice. the maneuvering board solutions of targetst courses,and speeds thereby The evolution of modern electronic reducing relative data to true motion. navigation systems has led to the design Similarly, bearings and ranges from one of relatively simple-to-use equipment target to another can be determined by with direct reading displays. One needs moving the cursor origin to a desired. only take latitude and longitude readings blip and rotating the cursor to a third. from the equipment and plot them. This may be well and good in the open sea Modern tactical operations are normally where the hazards of running into a shoal conducted within a relative motion are less unlikely. However, in framework because most weapons systems restricted waters, or where an accurate deal in range and bearing data elements. position is required, these systems are When contact is made, the target is not always as accurate as manufacturers generally identified by its range and would like users to think. Potential bearing from the detector. The resulting problems range from what we call range evolution in tactical thinking, holes, to system failures to poor pattern therefore, has concentrated almost solely geometry. Navigation in well-travelled on the relative motion solution. This is waters, is generally easy because a dangerous trend because, with the flood provision has been made for complete and of information now available from all our accurate electronic navigation signal sensors, there is a higher probability of coverage. However, in less frequently losing the 'big picture.' That's why, travelled waters, it becomes more when you see a picture of the Combat problematic. It is into these waters Information Center of an Aegis ship, you that our navies must sail and fight. How see a bank of very expensive screens. many of you, ten years ago, would have Unfortunately most ships cannot carry thought that there would be major actions such an expensive suite of equipment and in such out-of-the-way places as the must make do either with small and very Falklands and Grenada? In order to limitecd screens or with the more position oneself well, one must pay full traditional and underdeveloped plotting attention to all the variables. tables. Nevertheless, one of the best ways for There are still many operations personnel keeping a clear picture of what,is going who find it difficult to adjust to a on is to plot the action on a table. digital environment because they feel This effectively results in conversion of they've lost touch with the real world. relative motion into a true motion They find it difficult at times to picture on a geographic plot. The interpret the information presented. resultant plot then directly relates the It's not surprising that we still find action to the real world. the requirement aboard U.S. Navy ships to keep a hard-copy (paper) geographic track The fact that such solutions are still plot. available, dramatically points to a definitive need for tactical operations officers to get their target information RELATIVE AND GEOGRAPHICAL POSITIONING on a true or geographic display. Movement is then unambiguously portraye. One need only look at any tactical radar Mariners feel at home working tactical screen today to get an appreciation for solutions in true motion where action is how operational tactics have developed depicted in real world terms. It after World War II. Before that, they significantly reduces the complexity and were always played out on plotting tables allows for faster decision-making. using geographic references. Virtually every navigational and tactical radar The major advantage is that one gets a displays targets in the Plan Position true appreciation of the interaction of Indicator (PPI) format. With PPI, the all the participants in an action without radiating ship is positioned in the having to work out a relative motion center. Contacts are located on the solution in one's head. This is " screen along radials, at some true or particularly advantageous when there are relative bearing, extending from the multiple contacts to contend with both radiating vessel's blip. Some radars true and relative data flows being allow radar operators to displace the received. origin of the cursor to an off-center 1380 LARCE AREA BATTLES which radiates from a location far removed from its task force. One of the Modern weapons systems have longer and greatest sources of error in the Link longer stand-off ranges. The result is system arose from the fact that the that tacticians have had to consider ever helicopter had taken its relative extended detection horizons. The only position from the carrier which was in good method for adequately referring turn 'not well-positioned. The result was target positions to scattered elements of that all the contacts illuminated by the a task force is on some sort of grid or helo,.which was the only radar in the geographic coordinate system. This is EMCON-shrouded force, were in as much because not all vesssels can support error as that of its guide, the carrier. complex, large and expensive detection She was, interestingly, the element which and tracking sensor suites, or their had appeared in triplicate (i.e., in sensors are limited by EMCON restraints. error) on the Link screens of the force Ships must rely on oth er sources of earlier in the exercise. information to learn the whereabouts of the enemy. As with most detection When enemy targets are detected by such a systems, the detection is made in the mis-positioned helo's radar, then the relative mode but converted into error is propagated throughout the force. geographic positions for transmission to, Any ships needing to plot those targets and plotting by, other elements. would be in error as would any weapons which were to be fired. NTDS AND THE LINK SYSTEMS During a similar exercise transit, the Much of these data communications are Link control ship role was given to one performed through the Navy Tactical Data of the screening escorts. It was easy to System (NTDS) or Link 11 or 14 Systems. see that the destroyer, which was also These systems pass disposition the screen commander, was soon overloaded information on a number of contacts trying to run the ASW problem against automatically between computers. One three submarines whilst trying to weak link in these systems is the maintain the tactical data base. When frequent inability of data generating the going got hot, it was obvious which ships to position themselves accurately. effort was left to languish. We Any errors in their positions are passed observers noted an early breakdown in the through the system and multiplied by any navigation function as disorientation other errors present at the receiving became the rule on the Link system. We vessel. believe this occured because the navigation function is not considered If every ship and aircraft was properly critical when compared with the positioned, the problems could be priorities of the battle. This might significantly reduced. In a recent possibly be considered acceptable in exercise, the Link system was observed at battle there is no need for this to one instant to have three contact symbols happen. There are systems available portraying the same vessel in three which will automatically track the different places on the screen. Two of geographic plot for low cost and minimal these positions were more than five miles effort.. Positioning information was of distant from each other. The force was paramount importance to other ships in in radar silence and at darken ship. The the force when in EMCON. positions provided the only information by which the ships maintained their ATTACK SCENARIOS stations in the main body and screens. The ship from which we made these Attack scenarios are most often played observations was carrying a navigation out on geographic plots because of the integration system which combined three clarity of their presentations. This is navigation system inputs with an particularly true in situations where automated'tactical plotting table. We there is a 'time-late' or datum to deal were able to obtain the finest with. The solution to such problems is positioning information of any ship in best found on a geographic frame of the force. Since we knew the location of reference because the "lost" contact can our ship to within a hundred meters, we be fixed to a position on the earth where were easily able to see the failings in it was last detected. In this way its other ships not possessing good projected movement can be easily navigation or position information. portrayed. Similarly, it is far easier to set up search and attack patterns on a A common tactic during periods of EMCON geographic grid than it is to work in silence is to send a radar aircraft aloft relative motion. This fact becomes even 1381 more true when there are multiple vessels for error. Many of the larger equipment and helos involved in the action. manufacturers are quick to provide very large and technologically advanced suites Traditional Anti-Submarine Warfare (ASW) of equipment by which they claim.,to solve attack scenarios are traditionally played many of the mineman's problems. We've on a tactical plotting table. The datum found that a far better and less or target is plotted with the positions expensive approach is to increase the of other participating ships and precision in solving the aircraft. The plot is most critical for navigation/positioning challenges. This the Attack Coordinator who must be able is a relatively low cost solution with to see the entire situation and call in leveraged benefits surpassing those of his forces to maximum effect. the 'more and larger' equipment syndrome. Most modern plotting tables still require Q-ROUTES the concerted efforts of several persons to keep the plot current. The ship's Mine warfare begins with the complete position is projected by a light source survey of Q-Routes, i.e., the intended from below and a person must mark its routes through restricted waters down position manually at fixed intervals. which vessels may steam. This concept Similarly, all other contacts and targets limits the amount of area needing to be must be similarly tracked. The result is continually searched. The objective is that many persons must lean over the to condition a channel by accurately table while hand plotting. Meanwhile, surveying it and then resurveying it the Attack Coordinator must look over frequently to ensure that no new mine- everyone's shoulders to get a clear like ob.jects have appeared. When the picture of the unfolding attack. Not initial surveys are undertaken, all only is this method manpower-intensive, objects of interest on the bottom are but it is prone to human error in the accurately positioned and identified. plotting. Very little development has These become the reference objects taken place in this area. against which future contacts will be compared. It is important that all Few equipment manufacturers have objects are located in order to reduce attempted to automate this function. future requirements for classification of While there are a number of systems ?new' contacts. attempting to consolidate all this information on a geographic plot, most Q-Routes should be regularly equipment tends to be very expensive and 11conditioned"' so that, should war break complex. Additionally, many operators out, the number of new contacts needing tend to feel alienated from a strictly to be classified would be limited. digital solution. The plotting table Positioning during these conditioning still has an intimacy that is not found surveys must be as accurate as possible from viewing a video screen. to provide accurate positions for future comparison with objects previously MINE WARFARE detected and classified. The need for good geographical Much of this kind of survey work is done positioning is even more critical in the with the aid of side scan sonars. These area of mine warfare. It is of paramount allow operators to sweep a swath of importance that mine warfare vessels, bottom between the survey run lines. aircraft, Q-Routes and objects on the There are inherent inaccuracies in the seabed be accurately positioned. Each of side scan sonar principle which must be the elements in the mine warfare equation considered before the positioning of must be fixed on a common reference frame objects far from the survey run line may on which all players may track and be regarded as accurately positioned. display objects of interest. The most common grids are those developed around Side scan sonar traces are good geographic coordinate systems, whether reconnaissance tools but there are a they be in Latitude/Longitude or number of variables which are often not Universal Transverse Mercator. Without a taken into account when accurate common reference frame, the ability to positions are desired. A number of avoid or re7acquire the contacts is products on the market-and in development significantly reduced. project the paper side scan trace onto a video screen. This should allow the Precise positioning is of even greater operator to rapidly classify and importance in this critical area of naval 11position" objects detected. Many of operations. There is very little margin these systems, however, fail to address 1382 such problems as a display (waterfall) solution was to turn off one of the three scroll rate tied to ship's speed. This stations in order to reduce any is common on paper recorders but causes ambiguity. With two lines of position, distortion on video, leaving doubt as to they had no doubt where they were and where the object really is. Other they had a point fix. We have no interesting features touted are 'freeze knowledge what criteria they used in framing' and 'zooming,' which allow the selecting the deselected station, but it operator to stop the scrolling to better points to the fact that they really view a contact or blow up its scale. needed more than three lines of position Both these functions serve a useful to get a statistical evaluation of their purpose but obscure the cascading display positioning systems. There are too many currently being collected. lives and High Value Targets riding on their efforts. There is an ever greater Positioning objects detected by side scan need to provide for the integration of a sonar is possible, but there are a number number of positioning systems in order to of variables which must be addressed. come up with a statistically weighted The principal consideration is the answer. If nothing else, you will know position of the towfish at the time of how bad your position is. each ping. This is no small problem. Just knowing the position of the towing There is an old mappers' axiom that says, vessel/aircraft is not good enough when "The next battle will be always be fought obtaining positions for use in mine at the junction of at least two map sheet warfare. One of the other papers at this lines." The extension of this rule in conference on the Bathyscan system (in mine warfare is that the next mine will the Acoustics - Side Scan Session) be seen at the outer limits of your provides an interesting and accurate radiolocation chain pattern. Minemen approach. will always be on the fringes; out where the coverage is weak and support is weak. MINE HUNTING AND NEUTRALIZATION That's why they need systems.which can be counted upon to give the very best Return to a previously-detected mine-like solutions for positioning themselves and object for neutralization is an extension the objects they're looking for. of the problem discussed above. It's no good just having the objects accurately positioned and charted. The mine hunter must be able to navigate back to the CONCLUSION object. Without proper navigation, useless time is wasted looking for the We've tried to point out that the object to be neutralized and the risk is evolution of modern naval warfare has always present that the wrong object will drawn us away from the careful and be destroyed. Therefore, as much assiduous application of navigation and attention must be paid to positioning the positioning information. The ease of hunter and disposal vessels/aircraft as obtaining a 'fix' and the imperatives of in the initial survey. tactical threats have relegated this aspect to a lesser role. Nevertheless, The point being made here is that the importance of accurate positioning, positioning is the key to good mine whether it be for navigation or some warfare tactics from beginning to end. warfare application cannot be dismissed. Mine warfare operators cannot be content Only with good positions may naval forces to be haphazard in their approach to op.timize their power. navigation. The best solution is to use more than one positioning system and Without good positions, tracking the mathematically evaluate the inputs to battle is made difficult. The lack of a reduce the errors. We are not even sure plot makes decision-making difficult. that the GPS will be the final solution Too much effort is put into making large to all these problems, though it is systems which do everything, with too supposed,to give very accurate positions. little attention being paid to the key Mine warfare personnel cannot afford to element, the positioning of the force. be cavalier in their approach to Neither can we let the speed of the positioning. battle detract from maintaining our positions because they are what we need Recently we heard a story about some mine to deconflict our scenarios in the end. forces in the Gulf. They had come up with the traditional "Cocked Hat" problem where three lines of position resulted not in a point but in triangle. Their 4383 OCEANOGRAPHIC APPLICATIONS OF THE ARGOS SYSTEM Archie E. Shaw III, Executive Vice President Katherine R. Swenson, Assistant, Marketing & Promotion Service Argos, Inc. Terry E. Bryan, Meteorologist Office of Climatic and Atmospheric Research National Oceanographic and Atmospheric Administration processing and distribution of the instru- ABSTRACT ment data to the user. Since 1978, the Argos Data Collection and Development of the Argos system was Location System (DCLS), operating in bi- achieved through a Franco-American agree- lateral cooperative use of NOAA's TIROS ment among CNES (Centre National d'Etudes series of polar orbiting satellites, has Spatiales - the French space agency), NOAA, played a leading role in most major world- and NASA, and officially began with the wide oceanographic data collection signing of a memorandum of Understanding programs. This paper describes the Service (MOU), in 1974. The Argos system, which is Argos/NOAA satellite system, its' policy reserved for programs involved with the and operation, and recent Argos system collection of environmental data, became technological advances. Information con- operational in October of 1978 with the cerning past, present, and future major successful launch of NOAA's proto-flight United States, North American, and global satellite, TIROS-N. oceanographic programs is also presented, along with other types of oceanographic Outstanding performance of the Argos system applications possible through the use of was demonstrated during the Global Weather the Argos system. Experiment, also known as the First GARP (Global Atmospheric Research Program) Global Experiment (FGGE), in 1978 and 1979. NOAA was the lead U. S. organization in this international program which proved the INTRODUrTTON value of using the relatively new concept of oceanographic data collection and The need to acquire oceanographic data to transmission by satellite. use as a basic tool towards a broader understanding of the global ocean's climate The success of FGGE, in a sense, "turned has never been more pressing than it is on" the ocean research community, and other today. The impact of the southern Pacific global scale ocean research efforts came oscillations (El Niho) on the world's into being, such as the recent Tropical climate is an example of the importance of Ocean Global Atmospheric (TOGA) program understanding oceanographic processes. directed towards study of the El Niho/ Today there are satellite system tech- Southern oscillation. The World's Oceans nologies available that support, to an ever Circulation Experiment (WOCE), now in the increasing extent, the requirement to design phase, is another. gather in-situ ocean and atmosphere data. One of these satellite systems, the Service INITIAL PROGRAM APPLICATIONS Argos Data Collection and Location System (Argos DCLS), directly supports NOAA's First CAP? Global Ex-periment (FGGE) charter to monitor and predict the The scientific objectives of the FGGE re- environment. quired observations of pressure, tempera- ture, humidity, and wind velocity over the The relationship between NOAA and the entire globe from the earth's surface to French based Argos system is unique. While the low stratosphere. Existing surface ob- other national and international agreements serving land stations and ships at sea pro- exist to provide other than United States vided adequate base for satellite data instruments for flight on NOAA's satel- applications in the middle and high lites, no other agreement includes the latitudes of the Northern Hemisphere. instrument provider in the operational However, in the Southern Hemisphere, a CH2585-8/8810000-1384 $1 @1988 IEEE special observing system was required to logical and oceanographic programs, a provide the sea surface temperature and format and code for drifting data buoys atmospheric pressure fields to "anchor" the (DRIBU), was established through the World satellite vertical temperature and humidity Meteorological Organization (WMO) to trans- profiles. mit the data from FGGE, POLEX, and EPOCS over the Global Telecommunications System The FGGE Southern Hemisphere drifting buoy (GTS). For the first time, reliable in- system provided the required surface meteo- situ information was available from some of rological information through deployment of the more remote areas of the Earth. 300 drifting buoys, instrumented with atmo- spheric pressure and sea surface tempera- The Australian Meteorological Service found ture sensors. The buoy data were transmit- that there were more numerous, and more ted and processed through the Argos system intense, low pressure systems in the and made available for operational use in Southern Hemisphere than previously be- near real-time. These buoys were con- lieved. Post analysis of the Arctic buoy tributed by seven nations, including 56 by data showed a different location and the United States. The international strength to the Arctic high. These discov- drifting buoy deployment plan, utilizing eries, and others, resulted in a rethinking ships from 12 different nations and U.S. of climatology which had been heavily aircraft, was developed and coordinated by relied upon in some areas for producing Canada. weather analysis and forecasting. Arctic Buoy Proaram In 1980, Canada began a program to study The Polar Experiment (POLEX), a sub-program the feasibility of using drifting buoys as of FGGE, saw 15 ice buoys deployed to mea- part of a system to replace Ocean Weather sure atmospheric pressure and ice drift. Station PAPA, a ship located in the Gulf of Preliminary examination of the POLEX buoy Alaska, the spawning ground for many of the data showed substantial variation in the intense storms that affect North America. ice flow along the Canadian Archipelago and The PAPA Alternative Data System (PADS) was north of the Soviet Union, and yielded a a part of the (STREX) Storm Transfer and better understanding of the upper ocean Response Experiment, an extensive joint currents in the Arctic. Use of these data United States and Canada field operation to made a positive impact on short term fore- study North Pacific storms. casts. Because of the success of this pro- gram the National Academy of Sciences Although STREX was primarily a research recommended continuing it at least 20 more program, 3 buoy groups, in addition to years to establish a climatological data PADS, were deployed on different space base. With major contributions from Canada, scales to determine the usefulness of Norway, and several U. S. agencies, the drifting buoys, and to test new instrumen- network has now grown to 30 buoys and the tation for operational weather analysis and technology has advanced to include new sen- forecasting. sors for measuring wind, temperature, and sub-surface parameters. Research While GARP was being conducted, plans were Ecruatorial Pacific Ocean Climate Studiea being formulated for a World Climate (EPOCS1 Research Program (WCRP). The WCRP was for- The EPOCS is a continuing NOAA program mally established in 1980 by an agreement implemented during FGGE to study the between the WMO and the International processes in the near surface and deep Council of Scientific Unions (ICSU). The ocean responsible for the changes in the WCRP encompasses a wide range of inter- temperature of the equatorial sea surface. action between the atmosphere, ocean, ice, In this program a number of observational and land surface. systems are used, which include the deployment of about 30 drifting buoys Tropical Ocean and rlobal Atmosphere (TOGA) annually to determine surface current and Building directly on the FGGE, TOGA is a the sea surface temperature field in the program within the WCRP that began in 1985 eastern equatorial Pacific. Early EPOCS and is scheduled to last for 10 years. Its data became a part of the FGGE Data Set, aim is to explore the predictability of the and the program is now an integral part of coupled tropical ocean- atmosphere system the Tropical Oceans and Global Atmosphere and its impact on global climate on time (TOGA) Program. scales of months to years. TECHNOLOGY TRANSFER The observational domains for the TOGA Operational Analysis and Forecasting Program are the tropical oceans (from about In recognition of the potential of drifting 20ON to 200S) , and the global atmosphere. buoys for obtaining data so vital to sup- The Argos DCLS is being used extensively to port the implementation of major meteoro- meet the TOGA observational requirements. 1385 Moored and drifting buoy networks are de- The Coordinated/Ope rational Arctic Buoy ployed in the tropics to measure surface program (the continuation of POLEX/Arctic winds, sea surface temperature, and temper- Basin Buoy Program) and the Ice Island ature as a function of depth. A Southern Trajectories Program collect meteorological Hemisphere drifting buoy network, similar and oceanographic data to support opera- to the FGGE network, measures surface pres- tional forecasting and climate research sure, air temperature, and sea surface programs in the Arctic Ocean. Physical temperature. Some ships of opportunity use process and mesocale studies in the the Argos system to relay surface meteoro- marginal ice zone are conducted in the logical and expendable bathythermograph Beaufort, Bering, and Chukchi Seas. (XBT) data. Several oceanographic research programs with emphasis on monitoring and developing the technology to measure subsurface param- OTHER OCEANOGRAPHIC APPLICATIONS eters, including acoustics, are also being conducted. In addition to the vanguard FGGE Program, its subsidiary EPOCS and POLEX Programs, Drifting buoys are air-deployed in the and the STREX and TOGA Programs which were Antarctic region to collect meteorological formulated and built upon the FGGE experi- and oceanographic data to support oper- ence, Argos technology has been used by the ational and research programs. Another United States and Canada for several other program collects upper ocean and tempera- oceanographic research and operational pro- ture and salinity data in the southern grams. In the last ten years, hundreds of ocean from ice-locked drifters. Several platforms, mainly drifting buoys, have been programs are conducted in the U.S. Coastal deployed in nearly every ocean area of the Areas to study near shore surface circula- world. tion and processes off both coasts, the Great Lakes, and the Gulf of Mexico and to Programs in the Atlantic Ocean have in- test and evaluate new systems. cluded: studies of the North Equatorial Countercurrent, the Gulf of Maine, the Agulhas Current, the development of the ARGOS SYSTEM OPERATION Saragasso Sea thermocline, warm core rings, and the effects of off-shore currents on The Araos Subaystems waste disposal; collection of meteorologi- The Argos system is viewed as five (5) cal and oceanographic data to support oper- basic components; these are: ational programs such as the International Ice Patrol and weather analyses and fore- MThe PTTs (Platform Transmitter casting; and the short and long term Terminals), which are used on moving monitoring of various physical properties objects such as buoys, ships, balloons, of the ocean for research activities. animals, or on fixed environmental data Use of the Argos DCLS has allowed new sensing stations. research into the vast data-sparse regions of the Pacific Ocean. Large research pro- A PTT transmits without satellite interro- grams to study ocean-atmosphere interac- gation at preset regular intervals. tions and their influence on weather and Individual PTT transmission repetition climate have been undertaken. In addition periods are set between 60 and 200 seconds. to TOGA and EPOCS, the Tropic Heat Program Message transmission duration is less than studies the heat flux and storage in the one second. Each message may contain up to Central Equatorial Pacific and the Ocean 256 bits of sensor data. All PTTs uplink Storms Experiment observed upper ocean to the satellite at 401.65 MHz. Since there responses to severe storm forcing. Circu- are normally two satellites in operation, lation and physical oceanography experi- seven contacts per day are obtained from a ments include studies of the Gulf of PTT at or near the equator, and up to 28 Alaska, processes in the Yellow Sea, contacts per day in the polar areas. In the California fronts, particle fluxes in the mid-latitudes, ten per day are provided. Southern California Bight, the evolution of synoptic and mesoscale eddy fields, and (2)The onboard Argos Data Record Unit (DRU) vertical transport and exchange. system package included in the TIROS polar orbiting, sun-synchronous satellite pay- load. Two programs deployed over 40 drifting buoys in the Indian Ocean. These programs, The DRUs receive the messages transmitted closely related to TOGA, were designed to by those PTTs within the satellite(s) in- describe the surface current field, define stantaneous field of view; (approximately a statistically the time and space scales of 5000 km diameter "footprint"). Onboard DRU the surface current variability, compose processing consists of: model and drifter results for dynamic stud- ies, and study processes which influence -identification of the PTT number; sea surface temperature variability. 1386 ,asynchronization of the transmission time The recorded data processed by the Argos through use of the different repetition system are available in computer files ac- periods; cessible by telephone or telex through telecommunication networks, generally ,measurement of the Doppler shift on the within 4 hours after satellite reception. transmission frequency; The "real-time" data is made available within 20 minutes. Archived data is avail- *"hand-off" of these data, coupled with the able on computer compatible tapes, "floppy" platforms' sensor data,to the TIROS disks, or printouts, fortnightly or Information Processor (TIP) instrument. monthly. (3) The three telemetry receiving ARGOS SYSTEM ADMINISTRATION stations located at Wallops Island, Virginia, and Fairbanks, Alaska, in the The Argos system is administered by Service U. S., and Lannion in France. Argos, Inc., within the United States, and by CLS/Service Argos, S. A., in France. Each time one of the satellites passes over Argos system usage and performance is moni- one of the telemetry receiving stations, tored by the Argos Operations Committee, a the recorded TIP data are transmitted to bilateral (France/USA) organization which that station on 1698, 1702.5 or 1707 Mhz at meets once a year. This committee, 2.66 mbps. comprised of members from CNES, NOAA, and NASA, is responsible for ensuring that the system is managed according to the Each PTT transmission is also rebroadcast agreement. in "real-time" by the satellite as soon as it is received. Users with a downlink Service Argos is operated on a cost- receiver within range of the satellite can recovery basis with users contributing a receive data from PTTs that are in the fraction of the running costs proportional footprint of the satellite at the same to the amount of service required. time, through either the S-band on 1698 or Government users from the same country can 1707 MHz at 665.4 kbps, or the VHF beacon reduce their financial contribution by par- on 136.77 or 137.77 MHz at 8.32 kbps. ticipating in the "Joint Tariff Agreement". (4) The NOAA National Environmental Joint Tariff Aareement (JTA) Satellite, Data and Information Service The Franco-American Memorandum of Under- (NESDIS) data processing center, located in standing allows Service Argos, as an agent Suitland, Maryland. of CNES, to recover the costs associated with processing the platform sensor data and determining the location of the trans- Both the recorded and "real-time" data re- mitting platform. ceived from the U. S. receiving stations are then sent via domestic satellites to Since 1980, the United States has negoti- the NESDIS facility for decommutation. The ated a reduced tariff rate for the data data from Lannion, France, is rebroadcast processing provided by Service Argos. This through NOAA's geostationary satellite reduced rate is predicated on the large use (GOES east) to Wallops Island, Virginia. of the system by United States Government All of these data are then transmitted to sponsored programs, and the ability to the Service Argos Processing Centers at guarantee a minimum amount of processing. Landover, Maryland, and Toulouse, France. In 1982, other countries desiring to use (5) The two identical processing centers the system were invited to participate in perform the following: the JTA. The philosophy was that with lower processing costs, more countries -decoding of the PTT sensor messages and could deploy data platforms in remote conversion of data into physical units, as areas, thus increasing the global data base specified by the end user. for operational and research activities. Participation in the JTA is limited to -very accurate computation of the those users who are funded by their satellite(s) orbits. national government. Other user organiza- -PTT location computation using orbital tions which are considered nonprofit-making data and Doppler shifts. may also participate. -storage of all these results in computer To negotiate and administer the Agreement, files for user access. each country or group of countries has a designated Representative Organization for -generation of the archive data base for a Country (ROC) . The ROC serves as the each program and PTT. unique focal point between Service Argos 1387 and that country's users on all matters Argo Is system users from the Argos data concerning the JTA. The NOAA Office of dissemination computers. Climatic and Atmospheric Research and the Institute for Ocean Sciences, are the ROC's for the U. S. and Canada, respectively. The complete XBT microcomputer based system includes the launcher, the microcomputer RECENT AMOS SYSTEM ENHANCEMEN= interface and computer, the acquisition, bathy-message, and encoding software, and Alaorithm Modification the GTS data processing software resident In April 1987, two software algorithm modi- in the Argos data processing system. fications were made to reduce systematic location errors: 1) all the orbitography The Institut Franrais de Recherche pour platforms which are used as fixed precise llExploitation de la Mer (IFREMER - the reference points for location calculations French oceanographic agency) has ordered 15 were referenced to the same geocentric sys- complete XBT shipboard units which will be tem, (World Geodetic System 1984); and 2), progressively implemented by the TOGA net- the complete Goddard Earth Model 10 (GEM work. Forty (40) World Weather Watch (WWW) 10) was incorporated, to take resonance ships of opportunity will be equipped with coefficients into account. these units by 1989. Araos' Accurate Location System In the mid-1970's, during the original FUTURE REQUIREMENTS/PROGRAMS Argos system design stage, the targeted system location accuracy was approximately World Climate Research Proaram (WCRP) one kilometer. However, since the begin- World Weather Watch (WWW) Systems ning, far more accurate locations were The WCRP must rely on the operational WWW routinely achieved. data acquisition and management to provide longterm time series of global data Today, the system supports three (3) describing the components of the climate categories of location accuracy: system. It is essential that WWW be main- tained at least at the present level of (1) Quality, 150m at 1 sigma* performance or, for some critical compo- nents, upgraded to the level of FGGE. (2) Standar , 350m at 1 sigma The WCRP atmospheric sampling requirement (3) "Not Guaranteed", lkm at 1 sigma for drifting buoys calls for: *(l sigma: meets the criteria for accuracy 1) A network of about 100 buoys equipped approximately 68% of the time to within a with, at least, sea-level pressure and sea radius of described meters from true surface temperature sensors position) 2) expanding the present polar programs and A project by Service Argos to determine the the development of new atmospheric, maximum achievable system location accuracy oceanic, and sea ice sensors. was conducted during 1987. This is in addition to the oceanographic buoy and float requirements for specific The project, conducted over a period of programs, such as the World Ocean time sufficient to generate quantitative Circulation Experiment. conclusions, proved that with a few enhancements, and using retrospective pro- World Ocean Circulation Ex3periment (WOCE) cessing, the Argos system could achieve The WOCE is designed to provide the data absolute location to within 20 meters, and base for the empirical description required relative location to within 10 meters. to support long term climate modeling. This new location system was implemented This will be possible due to the emergence during the first quarter of 1988. of new global observing systems such as satellite altimetry and advances in in-situ Ex1pendable Bathythermograph (XBT) System methods such as drifting floats, expendable in conjunction with the TOGA Experiment, profilers, fast hydrographic/geochemical Service Argos developed an end-to-end XBT sampling and analysis, acoustic tomography, system. This system enables the results of and Doppler current profiling. The WOCE an XBT sounding to be processed by the includes an intensive five year monitoring Argos system and automatically sent to the period that is critically dependent upon Global Telecommunication System (GTS). The the successful development and implementa- data are sent as messages in the required tion of these observing techniques. The World Meteorological Organization (WMO) core projects of WOCE will depend upon sur- code format BATHY. The data acquired from face drifting buoys and subsurface floats the XBT probes are also made available for to determine the global ocean velocity 1388 field. It is envisioned that a network in platforms for special purposes. Various excess of 1000 drifting buoys will be oper- thermistor chain and drogue configurations ating at any given time and nearly 3000 have been developed and tested. Prototype subsurface floats are anticipated during drifting buoys have been successfully de- the observational years. These data will ployed in the paths of hurricanes. Deep be available on the GTS through Argos for drifter or "pop-up" type buoys which float operational as well as research use. at a given depth have been developed. Data relay systems are being developed for Oceanoaraiphic and Meteorological Research future application in major research Many of the oceanographic research programs programs. now underway or planned, particularly those dealing with circulation and flux measure- New Satellitp Tnatrumentation ments, may be encompassed by the WCRP/WOCE. In cooperation with NOAA and NASA, Service It is envisaged that specialized projects Argos increased the DCLS data rate between such as marine biological studies, minerals the four (4) Argos DRUs and the Tiros management programs, and sea and ice pro- Information Processor (TIP) instrument cess studies, will continue to grow and use onboard TIROS series models H, I, and J. the Argos system. Large numbers of buoys This modification was successfully demon- are also planned for deployment in support strated with the launch of TIROS H, (NOAA of the Navy's Project ERICA, a program to 11), in September, and increased by 20% the obtain new scientific understanding of the number of PTTs able to be received simulta- rapid intensification of storms at sea. neously onboard the satellite. Operational Programs TIROS satellite series K through M will see The present operational networks are yet another increase in the DRUs-to-TIP planned to continue. These include the data rate, and, the DRU complement will be Arctic Buoy Program, the Canadian Atlantic doubled. These latter modifications will and Pacific networks, NOAA's rapid response increase the onboard PTT reception capacity and hurricane drifters, and the Coast by another four-fold. Guard's International Ice Patrol. The Navy's experimental data buoy networks in the Atlantic, Pacific, and Arctic Oceans Service Argos, acting with CNES, and in have become operational and are expected to cooperation with NOAA and NASA, has plans increase dramatically in the next few to continue its strong leadership role in years. These data will be used to deter- the area of environmental data collection mine surface, subsurface, and boundary and platform location by satellite, through layer conditions for inclusion in opera- possible current generation add-on's, and tional analyses and environmental models to the next generation polar orbiting space identify, forecast, and provide early warn- platforms. Clearly, the Argos/NOAA cooper- ings of dangerous phenomena, such as ative arrangement has provided, and will tropical and other severe storms. continue to provide a technically proven and financially attractive DCLS service available to regional and worldwide governments' scientific centers, and the TECHNICAL ADVANCEMENT commercial sector. The increasing demand for more intelligent and sophisticated sensor-PTT configura- tions, fueled to a large extent by the requirements of animal tracking applica- tions, has dramatically affected technical design in terms of solar power, battery life, miniaturization (weight), and other factors. However, costs are down, princi- pally due to the increased demand for transmitters, plus manufacturing efficien- cies, and reduced chip and component costs. The incorporation of microprocessor tech- nology into PTTs can be of tremendous value, allowing platforms to be customized to meet specific requirements of the users. Advances in buoy platform technology are proceeding at a swift pace, and "expendable" air-deployable units are available in the marketplace. Development, test, and evaluation programs are continu- ously underway to improve on the buoy technology and to produce new sensors and 1389 HIGH-VOLTAGE SOLAR-POWERED NAVIGATION RANGE DESIGN LCDR Timothy S. Winslow, USCG LT Michael D. Dawe, USCG Mr. Karl R. Schroeder, USCG LT Wayne A. Fisher, USCG United States Coast Guard Washington, DC 20593-0001 ABSTRACT PROJECT BACKGROUND Design, development, and testing of As part of the project to establish a lighting and photovoltaic power systems Trident submarine support facility at for a high-voltage offshore navigational Kings Bay, Georgia, the Navy asked the range system are discussed. Operational Coast Guard to construct two ranges to requirements and solar power limitations mark the coastal approach to the St. dictated the selection of lighting Mary's River. The two basic operational components. High-efficiency metal-halide requirements for both ranges were 24-hour lamps_were chosen for the 120-volt daytime service and effectiveness over six-mile system. Photovoltaic array and secondary channel segments. Design and construction storage battery configurations were of the towers was turned over to the Coast designed to accommodate lighting system Guard Facilities Design and Construction power requirements and expected climatic Center (Atlantic). The Acoustics and conditions. System design emphasized Optics Section of Headquarters' Ocean personnel safety and component reliability Engineering Branch maintained overall in the marine environment. The paper project control and designed the power discusses these aspects in detail, giving system and signal control scheme. particular attention to photovoltaic array Integration of system components and pre- design and testing. deployment testing were accomplished by the Coast Guard Research and Development Center. INTRODUCTION @DAYTIME POWER AND LIGHTING Navigational ranges provide visual signals SYSTEM CONSIDERATIONS to guide mariners down narrow or hazardous channels. A conventional range consists To be effective (visible) over the same of two lights (and/or dayboards) mounted channel lengths, daytime range lights at separate locations, such that an typically must produce 2,000 to 5,000 extension of an imaginary line drawn times the luminous intensities of through the two locations defines the nighttime range lights. For the Kings channel centerline. Since the rear range Bay Range systems, this dictated selection light is higher than the front light by of 120-volt lighting equipment since design, the mariner simply attempts to available 12-volt equipment could not keep the lights lined up vertically as the achieve the required daytime intensities. ship proceeds along the channel. If the The logical choice of optic was the ship deviates right (left) from the standard Coast Guard 24-inch range centerline, spatial parallax causes the lantern, which uses a 24-inch parabolic rear (higher) range light to appear mirror to project an intense pencil beam, shifted to the right (left). Mariners and is normally equipped with two 1,000- judge conventional ranges to, be very watt tungsten -halogen lamps (one burning effective aids to navigation. and one back-up). Range design computations called for two 24-inch lanterns on each rear range tower and one on each front tower. 1390 United States Government work not protected by copyright Coast Guard lighthouses and other high- A. Procurement and installation of power aids to navigation have solar hardware could be traditionally relied on submarine cables accomplished quickly and easily. or on-site diesel generators for B. Each tower would have an electrical power. These familiar options, independent, self-contained power along with a relative newcomer, solar system, without reliance on photovoltaics, were the only systems vulnerable cables or refueling seriously considered for the primary vessels. daytime source of power. Thermoelectric C. Troubleshooting and maintenance generators were given some thought as would be much simplified, compared back-@up power sources, then dismissed as to other power systems. an unnecessary and costly luxury. D. Towers would not have to be specially designed to accommodate Coast Guard diesel generator installations heavy loads. at automated lighthouses have not enjoyed high reliability; rather, they have To accommodate selection of solar power required (on average) several scheduled and keep array size (and required tower maintenance and unscheduled discrepancy deck space) to a minimum, a nonstandard visits per year per lighthouse. At best, lamp was chosen for the 24-inch lantern. generator installations are expensive! The 175-watt metal-halide lamp delivers Two diesel generators are required for approximately the same luminous output as redundancy, towers must be sturdier to the 1,000-watt tungsten- halogen lamp at accommodate the additional weight of 1/5 the input power. Metal-halide lamps generators and fuel tanks, and maintenance also have longer rated lifetimes. costs are high. Trips to remote generator Selection of this lamp necessitated sites (for maintenance, troubleshooting, modification to the 24-inch range lantern or refueling) cost $600/hour and lampchanging mechanism. As mentioned $1,600/hour by buoy tender and helicopter, previously, no provision is made for back- respectively. Diesel generators were up daytime power. Daytime power and judged to be the least reliable and most lighting systems are expected to be very costly of the three power system options. reliable. Large fluorescent dayboards on each tower will provide passive daytime Submarine cables have generally been more range signals. These were designed to reliable than diesel generators for provide adequate range signals when viewed automated sites, often providing service through ordinary binoculars. over 15 to 25 year life spans. However, they are vulnerable to damage from anchors 12-VOLT POWER AND SIGNAL SYSTEMS and fishing nets, dredging operations, and natural causes. Repairs may take days or Each tower will also have independent even weeks to complete. Other major low-power 12-volt solar systems for: drawbacks with respect to this project were high projected installation cost and A. Nighttime range lights (both the potential for lengthy delays while ranges). seeking approval of the required B. Sound signals (offshore environmental impact statement. The range). offshore range system would have required C. Obstruction lights (offshore cable runs of up to 10 miles and likely range). would have been prohibitively expensive. D. Passing light (inshore range, The possibility of extended periods front tower). without power ( while troubleshooting or repairing the cable) was a grave concern. These systems will be similar to standard Coast Guard 12-volt solar installations. Solar photovoltaic technology was selected to provide daytime power for a number of reasons. It has had remarkable success in 120-VOLT SOLAR SYSTEM DESIGN the Coast Guard, with over 10,500 aids to navigation converted to solar power since The.heart of the Kings Bay daytime power 1983. Solar modules manufactured to system design is the array of rugged solar rigorous Coast Guard specifications have modules constructed to Coast Guard proven to be extremely reliable. Emphasis specifications. To ensure environmental on system design simplicity has produced integrity of solar modules, Coast Guard solar systems that can operate for long engineers developed the very rigorous periods of time with little or no Pressure Immersion Temperature (PIT) test maintenance. Life-cycle costs are much several years ago, (Figure 1 depicts the lower than for diesel generators or PIT test system). Modules are placed in submarine cables. Solar power was the test chamber and subjected to considered well suited to the Kings Bay repetitive 30-minute cycles of hot (45- project for several reasons: degrees C) and cold (5-degrees C) salt 1391 --------------------------------------------------------------- by a 45:1 ratio. It has been used effectively to reveal module design problems such as frame corrosion, junction Thermostat SALT WATER SALTWATEF 250GALLONTANK 250GALLONTANIt box defects, and substandard cell encapsulation. Figure 2 depicts the 120-volt daytime aremoo. system design for the rear tower of each range system. Five parallel strings of Drain firm 9L ten series -connected 35-watt photovoltaic modules provide power for twin 24-inch range lanterns. Surplus energy charges the 120-volt, 585 ampere-hour battery bank consisting of 60 series -connected 2-volt lead-calcium flooded cells. The initial an'shves Air pressro. arvitch solar design specified four parallel 45pi gul.to, module strings. An additional 120-volt Ai,-.P-I:td PUMP Ai,..p ... led COCUNCI our UNIT at. valve. string was added to maintain the battery PUMP bank at a higher average state-of-charge ME IN (thereby prolonging battery service life). Strains, . . . . . . . . . Float le The extra power will also allow th valve ulato, PrrCHAMBER battery bank to recover much more quickly from long periods of inclement weather. Theoretically, a similar daytime-demand solar system array should be tilted South as ev".... about 20-degrees (from horizontal) in this .. .......... .... -------------------------------------------------------------- location to optimize energy balance Figure 1; Pressure Immersion Temperature (PIT) test facility. throughout the year. In practice, however ' a fairly steep slope is required to discourage birds from roosting/nesting water immersion, with several superimposed on solar modules at sea (a significant 5-psig pressure cycles. The PIT test concern in this vicinity). Compromising simulates the unique combination of at a 30-degree angle provided an stresses present in the marine environment acceptable energy balance as well as a and accelerates solar module degradation reasonably effective deterrent. A row of 10 X 35 STRING BYPASS 30.degree tilt OOMBINER 10 x 35-w STRING art 12OVDC BYPASS 30 degree tilt COMBINER POWER EQUIPMENT 10 It 35-walt STRING (TYP BYPASS GROUP DISTRIBUTION PANEL d6r.-I .... ...... ...... .. .. COMBIN DOWNER STRING BYPASS & COMBINER., @i@:] EQJIPIMENT::':@:'*@. - 10 X _@all STRING BYPASS 30-degree tilt OOMBINER SOLAR PANELS 12VDC SOL Three @ 12VOC PAN 0. to- BATTERY 76-degree tilt 12VDC OBSTRUCTION OBSTRUCTION cne 20'eaft LIGHT LIGHT 0.. to-. 60-degree tilt (Master) (Slave) 12VDC :=]D 0m. 20@vratr L 75-degres tilt L__Zff102M1DCW F----MBATTERYM Figure 2; Kings Bay rear range design. 1392 Nixalite bird spikes was attached along individyal modules, for added personnel the top edge of each solar module string safety. To increase personnel to prevent birds from perching. Extreme protection, all 120-volt negative care was taken to position the solar components are grounded directly to the arrays so that no shading would occur from structure. other parts of the tower. Figure 3 depicts a typical system layout of the Each string bypass, combiner's 120-volt tower platform. output was connected in parallel by the group combiner. After considering advantages and disadvantages of blocking 0OWMM"UGHT _,J18CRANE diode placement, attachment was made on the array side of the parallel -------------------- connections. The primary advantage of this method is reduced current flow in the bypass circuit. The extent of bypass protection required is directly N 12VDC BATTERIES BATTERY ROOM proportional to circuit current. Most manufacturers advocate one diode for at least each 20 cells because standard I-GROUPCOMMER -parallel 2 @ POWER IMMIRUMN PANEL Elm practice involves series 3-NAL CONTROL LW _Sr 4 STRINGBYPASSCOMBINER connections where high currents are F]----ANa- possible. Employing one bypass diode for each 32-cell module under single-string 4 4 Q --r-Em- (low predictable current) operation was SUPPORTNUNG considered conservative enough for this 4 system. The possibility of an open- circuit fault disabling a string is 4 probably very slight considering the demonstrated reliability of our modules. Single-string design also simplifies servicing procedures; covering 10 modules mTRucncN unw--00, deenergizes the circuit. Figure 3; Main deck arrangement Three module strings are paralleled at the group combiner and are electrically hardwired to the charging/power system at Blocking diodes were removed from the power distribution panel (PDP). The individual modules to increase string two remaining strings are similarly voltage; a single blocking diode was paralleled and fed to a switched input of substituted at the termination of each the PDP. To prevent significant battery 120-volt string inside the group combiner overcharge, these strings are shed box (blocking diodes prevent the batteries (disconnected) as average battery voltage from discharging through the solar modules rises; lowering the charge rate to C/78. during periods of low illumination). Battery banks are expected to last 10 Longer special-jacketed output cables were years with-minimal annual maintenance. installed on the modules to eliminate splices, decrease the potential for corrosion, and minimize resistance. All LABORATORY TESTING OF SYSTEM COMPONENTS system wiring was run through teflon coated metal conduit and terminated in Prior to installation of the range towers, Nema-4X enclosures sealed by stuffing all lighting and power system components tubes. were tested thoroughly in the laboratory. Five 175-watt metal-halide lamps and Bypass diodes were installed in parallel ballasts were subjected to 4-hour on / 1- with each module to reduce power losses hour off cycles for 40 consecutive weeks and prevent damage caused by localized to test reliability. The 24-inch range reverse biasing (hot spots). Shading lantern lampchangers were modified to (caused by bird droppings or other operate on DC rather than AC power, and obstructions) causes the affected cells to reconfigured to accept screw-base metal- dissipate, rather than generate, halide lamps rather than bi-post tungsten- electrical power. Without bypass halogen lamps. Lamp out sensing circuitry protection, a single shaded silicon cell was also redesigned to detect reduced (in systems over about 50-volts) can cause current flow rather than an open circuit, a significant (30-60%) reduction in power since metal halide lamps do not normally production. Dissipated power is expelled create an open circuit during failure. as heat and may cause various degrees of This function was incorporated into the damage to the shaded cell(s). The bypass Ballast & Lampchanging Equipment Modules diodes were hardwired into the string (BLEM's), which were tested extensively combiner boxes instead of in the for proper switching to the standby lamp 1393 in case of simulated operating lamp failure. Proper day/night switching of CONCLUSION the Signal Control Units (SCU's) was confirmed by exposing the photodetectors The Kings Bay range systems will be to periods of light and darkness. String operational by Fall of 1988. Af ter all Bypass Combiner boxes and Group Combiner power and lighting hardware is installed Boxes were thoroughly inspected for on the towers, the battery banks will be continuity of all circuits and proper brought up to f ull charge by portable operation of blocking and bypass diodes. generators. Systems will then be energized for a detailed final inspection MAINTENANCE AND TRAINING and necessary troubleshooting. Frequent inspection trips in the early months will Basic system indoctrination was recently ensure that all systems are operating as provided to servicing personnel by designed. representatives from Coast Guard Headquarters and the Research & Development Center. Very little experience with 120-volt DC solar power exists within the Coast Guard. Servicing personnel regularly work with 12-volt solar/battery combinations, and to a 1 "Design of High Voltage Arrays;" lesser extent 120-volt AC systems. The P.A. Lawson and R.R. Addiss, Jr. dangerously high potentials associated Fourteenth IEEE Photovoltaic Specialists with these inherently live DC sources Conference 1980, Pgs. 512 517. require more safety consciousness than many other systems. An Operation and Maintenance Manual will be published in the near f uture wherein normal operation, maintenance and troubleshooting of each component will be described. In the interim, servicing personnel will accompany Headquarters and R&D designers to the towers whenever possible. Generally, the system should require only minimal maintenance to ensure reliable operation. Battery specific gravity will be initially monitored quarterly as the primary indication of overall system integrity. This requirement should be slackened after electrolyte depletion and battery performance trends are established. The only additional preventative maintenance requirements expected are: A. Replacement of metal-halide lamps semiannually. B. Replacement of photoresistors annually. C. Retorquing battery intercell connections and reapplying corrosion preventative annually. D. Annually examining circuit board terminal strips for corrosion and retorquing connections. E. Examining Nema-4x enclosure cover gaskets annually. F. Coating terminal strips and cover gaskets with silicone lubricant before resealing enclosures. 1394 ACCURACY OF SATELLITE SURVEY MEASUREMENTS ON OFFSHORE PLATFORMS FOR MONITORING SUBSIDENCE Dr. M.J. Mes Phillips Petroleum Company Norway ABSTRACT TOR Phillips Petroleum Company Norway measured ALBUSKJELL the progression of subsidence of the Z Ekofisk Field with a NAVSTAR GPS satellite 7.5 MILES surveying system. The uncertainty of GPS WEST EKOFISK 8 MILES satellite measurements onshore is compared FISK to the uncertainty of similar offshore PLEX subsidence measurements. It was found 8 that the uncertainty of offshore satellite measurements may be slightly higher than EDDAZ for onshore measurements. Computed uncertainties are stated INTRODUCTION ELDFISK Ekofisk is an oil and gas field in the southern portion of the Norwegian North Sea. it is one of several such f ields GREATER EKOFISK MAP VALHALL operated by Phillips Petroleum Company Norway in that area, such as the nearby Tor, Edda and Albuskjell fields. Figure 1 Map of the Greater Ekofisk area. Late in 1984, it was noticed that the Ekofisk platform decks were closer to the BACKGROUND OF THE GPS SYSTEM sea surface than when the platforms were installed. Subsidence was the only logical The GPS is a satellite supported radio explanation for these observations. After navigation system which provides 3- the subsidence phenomenon was recognized, dimensional position and velocity an accurate measurement method was needed information as well as high precision time to measure progression of subsidence and information for an unlimited number of the associated subsidence rate. One suitable equipped users. In the near available system for which no further future the GPS will be available world development was needed, is the NAVSTAR'GPS wide 24 hours per day. Currently, the (NAVigation @jystem with Time And Ranging system is in the late stages of ,Global Positioning @jystem). The development' phase where the design of measurements started in March 1985 with operational satellites, the operational Blom A/S as the survey contractor. These control segment and prototypes of surveys were continued after January 1987 receivers are tested and finished. At the by GPS Services A/S. Aero Service provided end of 1990, the system should be equipment and specialized expertise to completed. In space, 24 satellites should Blom A/S and GPS Services as required be orbiting with four satellites in each after October 1985. of the six orbit planes. They will For a period of almost three years this operate at approximately 20,000 km height system was the back bone of the methods with corresponding orbit period 12 hours used to derive subsidence rates and the sideral time. progression of subsidence. At approximately monthly intervals measured The satellites are transmitting signals at the elevation difference between three two , frequencies (Ll and . L2). Two bench mark platforms which do not subside different codes, the P-code (P for (Albuskjell, Edda and Tor) and the precise) and the C/A-code, (C/A for subsiding Hotel (2/4H) platform were coarse acquisition) are modulated on the measured. Figure I shows their relative Ll frequency. The P-code is modulated on locations. the L2 frequency only. CH2585-8/8810000-1395 $1 @1988 IEEE SINGLE POINT.POSITIONING The phase measurements are only useful if the carrier beat phase difference between With knowledge of one of the codes, it is different stations and between satellites possible to get clock information of the is known as a time history. The cycle transmissions from the satellite. With differences representing an integer number that, it is possible to calculate the so- of wave lengths can only be calculated called pseudorange between the satellite after the receiver has continuously and the receiver. The pseudorange is tracked a satellite signal and the phase different from the true range because of difference history for typically over half the receiver clock offset from system time an hour. When the signal is lost for a (see Figure 2). By placing one antenna in short time, so-called cycle slips may an unknown position, the pseudoranges can occur. The receiver starts developing a be measured simultaneously to four phase difference history again which may satellites with one receiver. This contain an error equal to an integer information is used to determine the three number of signal periods. The phase difference history can only be corrected in the relative sense for these cycle slips by computation., This causes that this method can only provide a relative position of one station to the other, as absolute, biasfree corrections are impossible. EQUIPMENT USED For the subsidence measurements, four Macrometer (R) V-1000 GPS receivers are used. Macrometer is a registered trade mark of Aero Service Division, western SINGLE POINT Atlas International. The Macrometer is an POSITIONING interferometric GPS receiver which uses I the L-Band frequency of 1575 MHZ to Figure 2 measure the phases. The field unit The principle of single point positioning consists of one antenna and a field by using pseudoranges. receiver/terminal that are connected by coaxial cable of up to 30 meters length during observations. The data have been position coordinates as well as the clock processed on a special purpose, Model P- offset of the receiver clock and the 1000 Data Processor. This is a desk top satellite clock. Without further system built around the DEC LSI-11 micro refinement, this method today gives computer. results of 30 m uncertainty for the C/A- code and 10 m uncertainty for the P-code. Control measurements by the manufacturer It is mostly used for navigation of moving (Aero Service), and independently by the military objects, where real time US National Geodetic Survey in the USA, positions are needed. The uncertainty have provided relative vertical quoted in this paper equals three standard uncertainty estimates expressed in the deviations of,the data.concerned. distance between receivers of one to four millionths of this distance. Over a RELATIVE POSITIONING distance of ten kilometers between receivers, a relative uncertainty (three To obtain a much more accurate position, standard deviations) of + 45 mm is at least two GPS receivers Used relatively expected on land under -ideal and to one another are necessary. The basis controlled conditions. of this method is to have one reference station , with accepted geographical MEASUREMENTS ONSHORE - OFFSHORE coordinates, and to determine the coordinates to one or more stations By a simultaneous measurement at the relative to this reference station. Ekofisk Center and an onshore bench mark reference station called Eikeberg near For this type of Precision geodetic Stavanger (Norway) , the relative position surveying, the phase of the carrier signal of the Ekofisk Center was determined with transmitted by the satellite is measured an estimated uncertainty Of + 1.0 meter. and interpreted. The reason for the During the first nine- surveys, higher accuracy is that the carrier signal measurements were taken onshore as well as has a much higher frequency than the code. offshore to check the system accuracy and For performing carrier phase measurements, repeatability. For all following offshore it is not necessary to have knowledge of measurements, the position of the Ekofisk the code signal. Center was used as reference. 1396 The Ekofisk Hotel platform has shown the predictive information about the largest amount of subsidence. it is anticipated satellite position and path as located nearest to the center of the a function of time, the so-called reservoir structure. Three other platforms ephemeris. This predictive information are used as bench mark references. These about the satellite's position in it's are the Edda, Tor and Albuskjell orbit is not very accurate, i.e. the platforms. on the Ekofisk Center and on estimated uncertainty in the anticipated the three bench mark reference platforms, position is + 20 - 100 m. , That measurements are carried out to obtain a corresponds to a relative uncertainty of measure for the relative subsidence of the up to 5 ppm by a distance of 20,000 km Ekofisk Hotel platform. These between the satellite and the user. measurements have been conducted at approximately monthly intervals from March The Macrometer V-1000 data are interpreted 1985 through April 1988. by using measured satellite trajectories provided by Aero, Service in the United States. Such data usually become FACTORS INFLUENCING THE UNCERTAINTY available a few days after the measurements via a telephone modem link. Atmospherical Influence These have an uncertainty of about 20 - 50 m, i.e. 1 - 2.5 ppm. Confirmed post- The GPS phase measurements could be processed ephemerides are available at corrected for differential atmospherical present from the US Department of Defense influences such as signal delays caused by and at least one commercial agency. tropospheric refraction and ionospheric refraction. The tropospheric refraction Multipath and Reflection correction would be based on actual surface weather information such as The so-called multipath effect occurs when humidity, barometric pressure and some satellite radio signals reflect off temperature at each measurement station an object before they reach the receiver interpreted with a Troposphere model. antenna. Thus the original signal may Corrections for surface weather conditions reach the antenna phase center via are not generally applied in GPS different paths at different times. This processing over distances where weather introduces ambiguities in the measurements conditions can be considered as uniform. which lead to an inaccurate solution. To avoid such effects, a carefully chosen The ionospheric correction is not critical. antenna type and antenna position is to the uncertainty of the solution for important. One has to pay attention, base lines 10 km long. It is difficult or especially to metal objects and other near impossible to determine this conducting surfaces in the vicinity of a correction when using single frequency GPS GPS antenna that could reflect the radio receivers . The ionosphere is situated at signals. a height of approximately 40 - 400 km above the earth surface. For relative A special antenna design reduces the measurements and short distances such as multipath effects to a minimum. For 10 km, the ionospheric delays are equal instance, the Macrometer V-1000 has an for both stations. A correction for the aluminum plate below the actual antenna to ionospheric refraction delays are only shield the antenna from reflections from possible if the measurement system relies below and provide a very stable phase on two different frequencies. When antenna. measurements rely on one frequency, only approximate and non-exact corrections can Available Field Data and Data be made. Interpretation The ionospheric refraction delay Data recorded in the field, consist of variations are largest near polar regions. phase measurements taken during a five It increases with sun spot activity. The hour long observation period with sun spot activity is expected to be above simultaneous measurements on four average large in 1988 to increase to a platforms. The Macrometer clocks are cycle maximum in 19.90-1991. The data calibrated before and after the analysis must account for these delays as measurements with a portable atomic clock. well since possible invisible systematical This gives the connection between the errors will be introduced in the computed internal Macrometer clock and UTC. results. Following the measurements, the quality of the data is checked by using the The Satellite's Position Macrometer as a computer. For processing of the raw data, knowledge The post-processing is done later by GPS of the satellite's position is necessary. Services A/s, using special purpose The coded satellite message contains software developed by Aero Service who 1397 manufactured the system. The data Eikeberget 1 and 2 are located only processing is done independently for every 9 meters apart. Therefore, for the baseline. Three dimensional relative assessment of elevation differences from coordinates with their co-variance matrix Eikeberget and J&ttAnuten (10 km apart) are the results. The uncertainty and the these measurements to JdttAnuten were adjustment are used to optimize the combined. The deviation of the computed solution by iterative means. elevation difference to the official difference in elevation is plotted in Figure 4. The standard deviation of these The Network Adjustment differences is 15 mm, giving an estimated uncertainty in the results of 45 mm. After processing each baseline separately, the results are entered in a so-called network adjustment program. The purpose EIKESERGET 1-EIKEBERGET 2 of that is to connect all stations and all MEASUREMENTS measurement periods from several (2 - 4) survey days to one consistent solution. DEVIATION (MM) The network analysis mainly consists of a 30- least square adjustment of the relative 20- coordinates. The sum of the squared residuals is adjusted to a minimum. At 10- S=4mm least one station has to be chosen as reference station. Its coordinates are 0- kept fixed in the network adjustment -10- process. The calculations provide all adjusted coordinates, and the uncertainty of the information. 13 118 210 @ Tio 7 IV suayw EVALUATION OF ACCURACY Figure 4 onshore Onshore GPS survey results. As a check on the system uncertainty and In Figure 5 the computeddeviation between repeatability, onshore measurements were the computed and the official difference taken during the first 19 surveys I in in elevation between Eikeberget 1 and 2 is parallel to the offshore measurements. shown. The standard deviation of these These were taken on the three onshore differences is 4 mm. Because of the, stations: Eikeberg 1, Eikeberg 2, and proximity of these bench marks, it is, J&tt&nuten (see Figure 3). These stations believed that most of the spread in their Are part of the geologically stable computed elevation differences are caused standard Norwegian Geodetic Grid System. by ephemerides errors and operational influences. 5* 30' 5*45' EIKEBERGET - iATTANUTEN 59*OC@- MEASUREMENTS DEVIATION ST IMM) 30- 20- Jitta. 10- 10 0- -10- SANDNES Eigeberg 1,2 -20- S= 15MM -30 MEASUREMENTS ARE + I. 3ppni. 5 10 is 20 SURM Fi'gure 3 Figure 5 The onshore stations Eikeberget and Onshore GPS survey results.for the height iAtt&nuten near Stavanger. difference between the Eikeberget 1 to Eikeberget 2 bench marks. 1398 During the initial surveys, personnel similar onshore measurements. This operating the equipment and setting up the corresponds to an uncertainty in the antennas were inexperienced. Thus these results of 60 mm. After survey no. 18, results may include errors caused by some of the antenna positions on several inexperience. Nevertheless, we feel that platforms were moved because of multipath the above results reflect the accuracy problems. Therefore, a better estimate of attainable under operational conditions the uncertainty in the measurements can be very well. obtained from the data of survey no. 19 through 30. offshore a) Tor - Edda (12 + 10)mm/year The survey results of the 2/4-E Tor, 2/4-F b) Edda - Albuskjell (14 T 7) mm/year Albuskjell, and 2/7-C Edda benchmark C) Tor - Albuskjell (5 + 13) mm/year platforms are shown relative to each other in Figure 6: The standard deviations of individual a) Tor platform 2/4-E relative to Edda measurements 'from the regression line are platform 2/7-C. found as 11, 8 and 14 mm, respectively. b) Edda platform 2/7-C relative to The mean of these standard deviations is Albuskjell platform 2/4-F 11 mm. This indicates that the C) Tor platform 2/4-E relative to uncertainty of Ekofisk offshore GPS Albuskjell platform 2/4-F. measurements is approximately 33 mm. We have used an assumed uncertainty of 40 mm in our own evaluation. COMPARISON OF CONCLUSIONS BENCHMARK ELEVATIONS M LUMETERS- The uncertainty of offshore GPS satellite zZMATIVEw '7 measurements of elevation differences of 1314 ?.122624 ?.62' 40 - -W -2-' survey points approximately 10 km apart 3 @ ._0 3 4- 9 6 @ 8 0.111 1 51-W 7.19 20 7 and derived from all measurements is TOR-WDA 60 mm. This uncertainty reduced markedly 300 to 3 3 mm for the latest eleven measurements when different, presumably better, antenna locations were used and 200 (b) EDDA-ALBUSKJELL the operators had gained experience with the system. 100 o Similar onshore measurements show an (c) ToR-ALBusrJELL uncertainty for all available measurements 0 19" 1986 1 1997 119as of 45 mm. This compared to the equivalent uncertainty for offshore measurements of Figure 6 60 mm indicates that offshore measurements Comparison of survey references for all 30 may have a slightly higher uncertainty satellite surveys. than onshore measurements. ACKNOWLEDGEMENT There one can see systematic trends in the The author acknowl .edges permission to elevation differences as well as seemingly publish the above paper from Phillips random variations. It was found difficult Petroleum Company Norway and Partners, to separate the random variations from any including Fina Exploration Norway Inc., systematical influence. The data are Norsk Agip A/S, Elf Aquitaine Norge A/S, analysed for the apparent uncertainty. A Norsk Hydro A/S, Den Norske Stats linear regression analysis is used to Oljeselskap A/S, and Total Marine Norsk estimate random variation in the three A/S. elevation differences between the three benchmark platforms. The standard deviation in the measurements is computed from offsets from the regression line. The standard deviation of individual measurements from the regression lines are respectively 19, 17 and 25 mm. The mean of these standard deviations is 20 mm. This value includes the effects of possible systematic influences. This is o higher than the standard deviation f 1399 Volunteer Observing Ships and the U.S. Government A Winning Partnership 2 3 4 4 S. Cook R. Benway , W. Krug , M. Nestlebush , A. Picciolo , W. Richardson 5 6 P. Stevens , and V. Zegowitz National ocean service, NOAA2National marine Fisheries service, NOAA3Office of Oceanic and Atmospheric Research, NOAA4National Environmental Satellite Data, and Information Service, NOAA5Fleet Numerical Oceanography Center, U.S. Navy 6National Weather Service, NOAA ABSTRACT This paper focuses on the VOS Program - - a partnership program where The U.S. Government continues to the maritime industries provide the enjoy excellent partnerships within the platforms and personnel and the U.S. National Oceanic and Atmospheric Government supplies equipment, logistics Administration (NOAA), private support, and training. Through this industry, universities, the U.S. Navy, Program the Government gains timely, and other Government agencies in the accurate data from data-sparse ocean collection of marine meteorological and areas. Shippers share in better marine oceanographic data from the world's forecasts for safer, speedier at-sea oceans. A partnership exists where the operations. The following presents a maritime industries provide the platforms brief history and background of the VOS and personnel and NOAA and Navy provide partnership; describes the present training, equipment, and logistics coordination of the VOS Programs in the support. Data gathered by this National Oceanic and Atmospheric Adminis- partnership and received at the National tration (NOAA) and the Navy; and outlines Prediction and Archive Centers aid the the future direction of the Program. entire maritime industry by improving forecasts and services thus allowing HISTORY AND BACKGROUND ships to make safer voyages while saving time and money. Ship weather reports have histori- This partnership is not limited to cally been the backbone of the ocean data U.S. ships but extends to ships of other observational system. The worldwide VOS countries. This international coopera- Program has approximately 7,000 ships tion makes possible the collection of from 47 countries. Ships vary in size ocean data on a global scale, allowing from large containers/tankers, to the world to better understand global university vessels, to small fishing environmental changes and their interac- boats. The meteorological reports from tions with the oceans. these vessels contain up to 19 elements, including sea level pressure, wind INTRODUCTION velocity and speed, wave height, sea surface temperature, etc. Some ships are The level to which we understand the also used as platforms for making subsur- marine environment and the role it plays face measurements such as temperature, in our weather and climate depends conductivity, and currents. directly on our ability to observe its Within the U.S. Government the VOS structure and variability. Advances in Program is a coordinated partnership science and technology in the past few among NOAA's major elements. These years - - particularly in satellite-based elements include the National Weather sensing and computers - - have given us Service (NWS); the National Ocean Service the capability to collect and transmit (NOS); the Office of Oceanic and Atmo- marine meteorological and oceanographic spheric Research (OAR); the National data in real-time by an array of space Environmental Satellite, Data, and based and conventional platforms. Information Service (NESDIS); and the Existing and future satellite systems National Marine Fisheries Service (NMFS). have increased the need for conventional Outside NOAA, the partnership extends to data to provide surface truth for the U.S. Navy, U.S. Coast Guard, the satellite measurements. These data Environmental Protection Agency, the U.S. (sea-level pressure, winds, waves/swell, Geological Survey, and universities such ice, water levels, surface and subsurface as the Scripps Institution of Oceanogra- temperature, and currents) are collected phy (SIO) . Key to the U.S. VOS effort by drifting and moored buoys, water level are the nearly 1,500 volunteer ships that gages, research vessels, and Volunteer collect and transmit timely, accurate Observing Ships (VOS). marine observations from the world's oceans. These ships account for nearly 30,000 marine observations each month. 1400 United States Government work not protected by copyright The number of ships and the monthly marine observations. Unique to the contribution of each U.S. VOS Program Navy's Program of 85 vessels are its range in size from 1,400 ships and 15,000 strategically stationed support personnel observations to 3 ships and 100 observa- in the U.S., Guam, Japan, Philippines, tions. Table 1 lists the U.S. VOS and Spain. The NOS VOS Program differs Programs. Also tabulated are the number from the other Programs in that all 120 of ships, the geographical coverage, and ships in this Program have Shipboard the data transmission capabilities of Environmental [Data] Acquisition System each program. (SEAS) equipment. Data entered into the SEAS unit are automatically checked, Table 1. U.S. VOS Programs formatted, and transmitted through the Geostationary Operational Environmental Number of Trans. Satellite (GOES) system and passed to Program Ships Coverage capability NWS via NESDIS. Data collected by all U.S. programs NWS 1,400 Global Radio/SAT are transmitted to National Prediction NOS 120 Global SAT Centers (National Meteorological Center NAVY 85 Global Radio/SAT and Fleet Numerical Oceanography Center) SIO 34 Pacific Radio/SAT and National Archive Centers (National OAR 10 Global SAT Climatic Data Center and National NMFS 3 Atlantic SAT Oceanographic Data Center). The key NOAA partner in this data transfer is NESDIS While the common goal of each who makes available and coordinates GOES program is to collect and transmit transmission assignments, and oversees accurate, timely marine data, some the archive of retrospective data. programs have unique features. For Real-time and retrospective data are used example, the NWS Program employs Port in support of high seas, offshore, and Meteorological Officers (PMO's) in major nearshore marine forecasts; climate and U.S. ports. These PMO's are responsible global change studies; environmental for recruiting ships into the program, monitoring activities; and satellite administering to and maintaining partici- calibration/validation. Figure 1 depicts pating ships, installing and calibrating the locations of VOS field support instrumentation, and instructing ships' personnel and a schematic of data flow. officers in observing and recording A WINNING PARTNERSHIP MARITIME INDUSTRY U.S. GOVERNMENT Platform Training Personnel Equipment Logistic support National Prediction and Archive LEGEND Centers Field support f- Personnel � Marine Forecasts � Climate and Global Change Studies � Environmental Monitoring Activities Figure 1. U.S. VOS Partnership, Locations of Field Support Personnel and a schematic of data flow. 1401 PRESENT COORDINATION AMONG PROGRAMS personnel on new procedures and data gathering and transmission equipment. At A high degree of coordination exists these meetings logistics support in the VOS partnership as evidenced by personnel from other VOS Programs are the joint authorship of this paper. on- invited. going coordinated activities involve user For ship greeting and retraining, requirements, equipment design and the preference is to have the same person procurement, training and logistics greet his/her assigned vessel. This support, and system monitoring. allows the greeter to build a good working relationship with the ship's User Requirements personnel. Because of changes in ship's schedule this is not always possible. In Selection of VOS vessels for SEAS these situations training is provided by equipment in the NOS VOS Program is the most geographically convenient decided by the SEAS Advisory Committee. trainer, regardless of Programs. The committee is composed of a represen- tative from each U.S. VOS program, System Monitoring Canada, and Australia. Based on user requirements (operational and research) Shipboard equipment must be this committee recommends and votes an monitored to spot problems as soon as the deployment of SEAS units by electron- they occur. When ship's data are trans- ic mail. Factors considered are: data mitted by the GOES system, monitoring is need, logistics support capabilities, extremely important. A transmitter satellite communication coverage, vessel malfunction could cause another ship's time at sea, area of operation, and data not to be received. User agreements existing shipboard equipment. with NESDIS require transmitter problems to be corrected within 24 hours. Equipment Design and Procurement Data transmitted by the GOES system are monitored by a microcomputer which Coordinated efforts exist on the allows VOS support personnel from NOS, design and procurement of equipment. NMFS, and OAR to dial-in and monitor Equipment is state-of-the-art, their assigned vessels. During off-the-shelf hardware with standard installation the system's transmission interfaces that are compatible with capability can be tested via the dial-in equipment throughout the VOS Program. monitoring mode. Monthly statistics Interfaces which must be designed are compiled by the system are shared with coordinated among the various programs. participating Vessels in a monthly This is a must to ensure the cross newsletter. utilization of logistics support people among the programs. FUTURE DIRECTION Joint procurements have been shared among the programs. For example, the The planned course for the U.S. VOS Navy and NOS normally share procurements Program is continued coordinated, shared of expendable bathythermographic probes support; deployment of more shipboard (XBT's) and the purchase of the central environmental measuring systems with processor for SEAS. NWS and NOS are automated oceanographic and meteoro- jointly funding 20 "Lap Top" SEAS units. logical sensors; increased use of satel- The automated pressure sensors used on lite telemetry methods; increased effort some of the SEAS units also constitute a to interface Government developed data shared activity. The pressure sensor collection software with commercially purchased with NOS funds are calibrated developed and deployed hardware; and and routinely checked by NWS PMO1s. fostering a coordinated international VOS Program. Training and Logistics Support Shipboard Environmental Measuring Systems Because of rotating schedules, ship personnel often change. The key to a More equipment with automated and successful VOS program is frequent ship state-of-the-art, off-the-shelf, oceano- visits and retraining of ship personnel. graphic and meteorological sensors will NWS and NOS schedule annual meetings for be deployed. All systems will have briefing and training VOS support satellite telemetry capability. The move 1402 to fully automated systems will requir e Table 2. Planned Direction for less time of ship's personnel to take and VOS Shipboard Systems transmit meteorological and oceanographic Equipment observations. For example, winds can be measured by anemometers and fed directly into the ship's central processor. Winds Deploy more Systems will be automatically adjusted for (Less Expensive and Smaller) ship's heading and speed. Automated Interface with more Sensors systems will also allow for more in-line Work with Private Industry quality control of measured data. Plans for automated oceanographic New Sensors and meteorological sensors include acoustic doppler current profilers; Expendable Conductivity - expendable conductivity-temperature-depth Temperature - Depth (XCTD) sensors; and automatic meteorological Acoustic Doppler Current packages that measure wind speed and Profilers direction, relative humidity, air and Automatic Meteorological surface water temperature. Automated Packages sensors and , satellite telemetry method o Wind Speed and Direction will allow for more frequent and more o Relative Humidity timely observations. The overriding 0 Air and Surface Water requirements for the equipment are price, Temperature compactness, and flexibility for add-on, state-of-the-art sensor packages. The Satellite Telemetry interfacing of the packages will be handled by system software. Interface with INMARSAT Continue Coordinated Satellite Telemetry Methods Deployments with Polar To take advantage of the more than orbiting French Reporting 6,000 ocean-going vessels with Interna- System tional Marine Satellite (INMARSAT) Partnership with Commercially Developed capability, software for shipboard data and Deployed Systems collection systems is being modified. The INMARSAT capability will allow for the transmission of large volumes of data Representatives of the U.S. VOS such as subsurface current data collected Program and private industry are working by acoustic doppler current profilers. to interface ocean data collection INMARSAT transmission, which is two-way, software with commercially developed and will make it possible to send marine deployed systems. This software will run forecasts and value-added products back in the "background" of these system and to the ship. send data via a ship's INMARSAT The U.S. VOS Program is working with transmitter. the French on coordinated deployments of The U.S. Program is contracting equipment to collect subsurface tempera- private industry to upgrade (add ture data. The French system - - Service subsurface data collection capability) to ARGOS System - - transmits data via the "MET ONLY" systems. The program is also polar orbiting TIROS satellite. Rather looking to the private sector to inter- than modify U.S. shipboard systems to face off-the-shelf, state-of-the-art transmit via the TIROS satellite, the hardware into SEAS-like systems. U.S. and French are coordinating deployments to take advantage of their Foster An International, Coordinated VOS respective satellite coverages. Program Summarized in Table 2 is the planned direction for the design and depoloyment The U.S. VOS Program contributes of VOS equipment. supplies, logistics support, and 1403 equipment to a number of international will be timely data sets that will enable VOS Programs. These Programs include: us, as a nation, to better understand Scientific Industrial Research global climate change, to develop better Organization in Australia, Office De La predictive models, and to increase our Richerche et Technique Outre-Mer, forecast skills thereby saving life and France; Ministry of Agriculture and property. If you are interested in Fisheries, New Zealand; and ocean Science taking part in this partnership please and Surveys/ Institute of Ocean Sciences, contact either Steve Cook, NOS Canada. In addition, cooperative data (408-893-7103), Paul Stevens, U.S. Navy collection efforts are underway with (408-647-4351), or Vince Zegowitz, NWS, Chile, China, Ecuador, Japan, Mexico, (301-427-7724). Peru, Portugal, Uruguay, and West Germany. Data collected by this ACKNOWLEDGMENTS international partnership are very important to global programs such as the Special appreciation is expressed to Tropical ocean Global Atmosphere (TOGA) all personnel who provide data monitoring and the World Ocean Circulation and field support to the U.S. VOS Experiment (WOCE) programs. Programs. To further the international effort, the U.S. Program shared in the design and REFERENCES distribution of the Integrated Global Ocean Services System (IGOSS) Ship of Benway, R., 1987: Water Column Thermal Opportunity Programme brochure. This Structure Across the Shelf and brochure describes this worldwide service Slope Southeast of Sandy Hook, NJ for rapid collection, exchange, analysis, in 1987. Northwest Atlantic and dissemination of meteorological and Fisheries Organization, Ser. No. oceanographic data and products. A N1441, 7 p. tear-out, addressed panel of the brochure is included for those who would like'to Brown, S., 1987: Watch Out For Nelson receive more information about the Eddy. Lamp, Vol. 69, No. 1, 12-17. Program and for those who are interested -in participating in the Program. Cook, S. and V. Zegowitz, 1983: An Automated Data Acquisition and SUMMARY Satellite Transmission System. Sea Technology, (Oct. issue), 48-49. The VOS partnership with the U.S. Government and private industry is Saur, T. and P. Stevens, 1972: continuing to grow. This partnership, Expendable BT Observations from where the maritime industries provide Ships of Opportunity. Mon. Wea. platforms and personnel, is growing to Rev., 16, 1-8. the stage where private industry will more fully share in satellite telemetry Stevens, P. and D. McLain, (unpublished and intergrating shipboard data manuscript), 1987: Cooperative collection equipment. This partnership Oceanographic Observations Program, is headed toward a totally international 18 P. program. Continuous monitoring of the data collected, formatted, and Szabados, M., C. Roman, and B. Taylor, transmitted by sensors aboard vessels 1987: Transmission of Real-Time will ensure the collection of accurate, oceanographic and Meteorological real-time data in the most cost-effective Data From Ships. Proceedings of manner. Oceans 87, Vol. 3, 863-869. The net result of this U.S. Government Private Industry Partnership 1404 COMPARING U.S. COAST GUARD BUOY TENDER PERFORMANCE USING SIMULATION Leonard C. Kingsley Joe A. Smith Kenneth S. Klesczewski Roberta A. Walters PSI International, Inc. USCG R&D Center 510 N. Washington St. Avery Point Falls Church, VA 22046 Groton, CT 06340 ABSTRACT Island Sound region. The Coast Guard plans to replace its Servicing aids to navigation and aging fleet of buoy tenders with newly correcting discrepancies is a constructed vessels. Many of the vessel complicated activity. otherwise well- designs being considered have never been planned servicing trips are often built, therefore they have no track altered to accommodate reported record. Moreover, the Coast Guard discrepancies in the area or because the requires a way to compare these vessels weather creates unsafe working against one another for each operating conditions. At times, several aids district of consideration in order to become discrepant simultaneously due to choose the best vessel possible. A weather, collisions, or vandalism, simulation has been developed which demanding rapid and extraordinary models a single tender working a given response from servicing units. field of buoys. The purpose of this model is to evaluate alternative buoy The present fleet of buoy tenders used tender designs of differing to service these aids are aged. Thus characteristics in realistic operational they require ever-increasing amounts of environments. maintenance and are no longer cost efficient. Over the next decade the 1. INTRODUCTION Coast Guard plans to replace these tenders with a new fleet, possibly of a A major responsibility of the Coast new design (class). This means that Guard is to maintain a system of many of the designs to be considered navigational aids to mariners within have never been constructed and thus territorial waters. These aids, have no historical track record by which otherwise known as buoys and the Coast Guard can evaluate their lighthouses, are used to mark navigable operational performance. waters for various sizes of vessels and also to warn of submerged dangers such To make performance comparisons of the as rocks, sandbars, and sunken vessels. various tender designs is cost prohibitive. In order to do this, each An essential feature of overall buoy alternative design would need to be tending operations is that the buoys are built and tested in each area. As a maintained in a variety of environments result, a simulation model has been which can affect the way in which the developed. The purpose of this model is tender performs service. An example of to evaluate alternative buoy tender this would be the differences between designs in realistic operational the tenders that service the coastal environments. The model (see Fig. 1) is area of the Pacific around Alaska versus basically a SIMSCRIPT 11.5 simulation those that service the Long Island Sound which can be used to realistically region between New York and Cape Cod. simulate buoy tenders with differing The tenders servicing the Alaskan area characteristics while they work a field must travel great distances to reach a of buoys in a given geographical area. particular buoy, whereas tenders in the This allows the analyst to measure Long Island Sound area travel much relative performances of various buoy shorter distances. In addition, the tender alternatives (new designs). aids they service are placed closer Since no data or experience would be together and tend to be in groups. In available for new ship designs, the comparison with the Alaskan region, the model provides the only means of aids are smaller and the weather is measuring operational performance of susceptible to quick changes in the Long these ships. CH2585-8/88/0000-1405 $1 0c1988 IEEE area being simulated. This data was obtained from the NOAA Data Buoy Center DISCREP WX and Army Wind and Wave Summaries. The Sea State model takes into account total time at a wave height interval as well IS OY STAT as the average time at that interval. D T SM Visibility was generated using actual TIME STAMP readings taken from local airports. SHIP DATA Importantly, surface visibility affects PORT RPM RULES the tender"s ability to position buoys after servicing. Their proper positioning is critical to mariners. In positioning a buoy, the tender's crew must take measurements that require certain reference points. Visibility that falls below the minimum level at which the tender can make an accurate placement, means that the tender cannot Figure 1. Simulation Model complete its work until the visibility returns. to acceptable levels. This 2. OVERVIEW increases the downtime of the buoy and tender under-utilization, thereby The Simulation essentially provides the degrading performance. "driver" for moving the tender around, servicing the field of buoys. The speed Geographical limitations (navigable of the tender changes as a function of waters) are dependent on land obstacles sea state, thereby providing a realistic and water depth for a given area. Each setting for buoy tender operations. The tender may have characteristics which simulation also keeps track of limit its ability to travel in some statistics over time, such as operating waters. The difference between hours, response times, buoy and tender navigable waters and land and non- downtimes, etc. Resource usage such as navigable water is represented by a fuel and material are also accounted for boundary polygon. A route consists of a by the model. sequence of line segments lying entirely within this polygon. The polygons are The model addresses a select group of derived from a nautical geographical environmental factors that were felt to data base f or the region to be modeled have an influence on the performance of (see Fig. 2). Importantly, the polygons the tender. For each of these serve to delimit the area for travel and environmental considerations the tender define the navigable region for the has attributes that designate the ship. Therefore an individual tender operational limits of the tender with has its area of navigable waters respect to each of these factors. The specified by boundaries defined by height of waves, and the speed and polygons which can be used to route the direction of prevailing winds, have a tender properly. significant effect on the cruising speeds of the tenders. Moreover, in summary, the simulation model can be visibility is a factor when positioning used to simulate buoy tenders with a buoy. If reference sites are not different attributes such as speed, visible because of haze or fog, then lifting capacity, draft, length, weight maintenance and repositioning of the limitations, deck space, etc. This buoys could not be performed. allows the analyst to measure the effectiveness of a new buoy tender Uncontrollable environmental factors design. The measures of effectiveness handled by the model are Navigable will be related to output data which Waters and Weather Conditions.' reflects quality of service to the Navigable waters are those waters in mariner, e.g., how long a buoy is out of which the tender can safely travel. Any service. aid outside the tender's navigable waters cannot be directly accessed by 3. ROUTE PLANNING MODULE the tender and thus cannot be serviced. Two .weather-related uncontrollable A significant aspect of modeling is that factors are wave height and surface the needs of a ship within a region visibility. These two f actors have a depend critically on its particular direct bearing on tender's performance. geography. Possible travel routes,vary according to which ships 'can success- The Sea State was modeled using wave fully travel them, due to depth of data from weather buoys local to the water or weather exposure. Therefore, U AA DATA ,1406 T 4 1 2 CA 4 p.; EP C.: S F! F U 10 U'S R J CF@ .0 LEi 41 A C N ............. .................................... ................. .. ................ ......................... ..... .................................... .................... 6: TE 4 f Z 41 ............ ....... ....................... ............ ........ .............. "Nov V, 7 21 5 0 7 2 4 0 72 3 0 7 2 2 C-1 72-D i I]D Figure 2. Navigable Waters the modeling process must take into account geographical constraints such as shoals and islands. 3.1 Subjective Elements of Route Planning The Route Planning Module (RPM) is @alled by the simulation model when it One important set of variables that @s required to plan the tender's travel influence the routes are those derived itinerary. The RPM will select routes from human judgement. ship captains based on a multitude of inputs f rom the differ in their preferences and their simulation model. These include vessel decisions can depend on many variables. attributes such as; geographic To include judgement or preference in limitations, buoy attributes, weather the model requires a framework which limitations, and resource requirements will reflect the preferences or utility such as fuel f or the ship. Moreover, of a particular ship captain. the RPM is called when the simulation model generates interrupts of the normal Two significant factors are thought to buoy servicing for such things as be most important in constructing a unscheduled buoy maintenance and route based on judgemental factors: @esponding to emergencies. , Upon timeliness and minimizing the time to interrupt, the ship must be reassigned complete a route of buoys. Basically, temporarily and then rescheduled from a the captain is asked to construct two new starting point to efficiently utility functions which give a measure service the remaining buoys. of timeliness and time to complete service. Judgemental Independence is concisely, the route-planning problem assumed and the two utility functions can be stated as follows: A base has a are combined additively. By obtaining single ship and N buoys, where N is too the most preferred attribute and the large to be visited on a single trip. scaling weights for the functions, allow The navigable space is defined as the one to combine the attributes into a 2- interior of a specified polygon. dimensional utility function. Once a Islands are represented as polygon set of alternative, routes are "holes" within the outer specified constructed. The most preferred route, boundary polygon. The capabilities of can be selected using the utility the ship affect the routing. Therefore, function. the model must find the least expensive and timely set of trips that visits each This leaves the problem of searching the buoy once for the length of time nodes for feasible sets of routes. This required to perform the speci fied main- is performed using two linear evaluation \.Xv tenance. functions. The search is performed 1407 depth first and is controlled by using BITTERSWEET scaling weights on the evaluation functions. A Search tree is created for Top economical speed: 10.0 knots each buoy. Within a predefined cluster Depth restriction: 15.0 feet of buoys the next buoy visit selection crane capacity: 40,000 lbs is made based on applying the evaluation Payload (buoys on deck): 99,999 lbs function. The buoy with the highest Night buoy operations: No value resulting from the application of Endurance: 120 days the evaluation function is selected. Fuel capacity: 28,000 gals This process is continued until all the Fuel usage: 39.47 gals/hour feasible routes have been generated. Maximum wave height Finally, the utility function is for safe tender calculated for each feasible route, operation: 3.2 meters resulting in the route's combined utility. The route selection is based Discrepancy downtime was used as a on maximizing utility of the routes. surrogate for a measure of mission degradation. The response policy 4. EXPERIMENTS currently being used within the simulation gives discrepancies highest This section will describe an experiment priority. That is, unless the tender is running the model in the Long Island currently performing work on a buoy or Sound region. This was a comparison is in port to satisfy a constraint, the between a coastal tender (REDWOOD) and tender will immediately head for the an oceangoing tender (BITTERSWEET). discrepant buoy. There were 251 buoys assigned to each alternative tender. Eight ports were it is possible for buoys to be serviced designated as places the tender can behind schedule due to such factors as; dock. Of the eight, only the homeport weather, buoy discrepancies, constraint was designated with the capability to satisfactions required by the tender, store buoys and refuel the tender. For and maintenance of the tender. Any these comparison runs, a period of time buoys that are worked beyond an from April to June, 90 days, was acceptable scheduled maintenance date selected as indicative of peak workload. are output in a report in addition to the amount of time beyond the scheduled Wave height data from a NOAA data buoy, maintenance date. The acceptable range which is located just south of Cape Cod, of time used in current runs was 14 was used in the generation of wave days. heights for these simulation runs. This buoy data was selected for its proximity Buoy time stamps for routine buoy to the mouth of Long Island Sound and maintenance are obtained from the its completeness of data. Because the operational offices. These time stamps data buoy location is in an exposed indicate the type of work required area, simulated wave height data was (i.e., replacement) and the date the biased to rougher wave conditions. work should take place. During the course of the simulation run, the time The'characteristics of the tenders were stamps are updated to the next time the based on engineering estimates. The work needs to be done according to tender's characteristics were set as policy. follows: 4.1 Results of Experiment REDWOOD AsIpreviously mentioned, the measure of Top economical speed: 10.0 knots buoy tender performance was the average Depth restriction: 10.0 feet down time of the discrepant buoys. Crane capacity: 20,000 lbs Since the primary objective of buoy Payload (buoys on deck): 40,000 lbs tending is to keep buoys in service and Night buoy operations: No operating, it is desirable and of the Endurance: 5 days utmost importance to keep these buoys Fuel capacity: 17,620 gals functioning to reduce the danger to Fuel usage: 60.00 gals/hour mariners. Importantly, the objective Maximum wave height with regard to mission effectiveness is for safe tender to minimize the down times of the operation: 1.2 meters discrepant buoys. The test data was taken in 90-day intervals. Twenty repetitions of this interval were performed with different samplings of weather, service times and buoy failures. The Measures of Relative 1408 Effectiveness (MORE) was represented by would be required to service these a sample mean and variance of average buoys. Incidentally, since the REDWOOD discrepancy downtimes. A 99% confidence has been operating in this area for interval of the mean was calculated years, one would expect it to be the using the Standard Normal Approximation. better choice when tested against Along with the above statistics, a current designs. histogram is generated to give a visual idea of the distribution of the MORE. In order to show that the model does indeed distinguish among tenders, Figure 3 below presents the MORE results sensitivity runs were performed using of the experiment. As expected, the the REDWOOD as the baseline tender operating in Long Island Sound. Figure SHIP REDWOOD 4 on the next page presents the MORE SAMPLE SIZE 20 results of changing the Maximum MEAN 14 876961 Allowable Sea State f rom calm to VARIANCE 46:423405 99% CONFIDENCE INTERVAL 10.946231 18.807691 moderate seas. When the tender is 40 restricted to operate in calm seas, the 38 MORE is 18, while if allowed to operate 36 34 in moderate seas the MORE is 10. The 32 30 tender with the smaller MORE is more 28 capable of operating in the established 26 24 environment. Since the confidence PERCENT 22 20 levels do not overlap, there is a 18 significant decrease in the MORE when 16 14 Maximum Allowable Sea State is 12 increased. This implies that the less 10 8 capable vessel would allow for more 4 equipment outages and may reduce 2* 4** 7** 4** 1** 1* 1- 2* operational safety compared to the more 7 10 13 16 19 22 25 26 MORE capable vessel. of course, tradeoffs TO LESS THAN 10 13 16 19 22 25 28 31 must be made with other MOREs such as tender utilization and operating cost Figure 3a. REDWOOD Run before any concrete conclusions could be reached. SHIP BITTERSWEET SAMPLE SIZE 20 S. SUMMARY AND CONCLUSIONS MEAN 15.455328 VARIANCE 59.827816 99% CONFIDENCE INTERVAL 10.993053 19.917602 A simulation model for evaluating 40 various buoy tender designs operating in 38 36 selected geographical regions has been 34 32-**** developed and has produced results which 30 ... could be used with other supportive 28 26 evidence to decide on the best design 24 PERCENT 22 for a particular region. Although the 20 model does not directly use engineering 18---- 16 parameters such as heave, pitch and 14 roll, the results indicate that 12 10 environmental effects and approximations 8 6 (or surrogates), such as Maximum 4 Allowable Sea State, are able to 2* 7** 5** 3** 3** 1- 1* 7 11 15 19 23 27 31 35 MORE distinguish significant differences TO LESS THAN 11 15 19 23 27 31 35 39 between tender designs. Of course, it is essential that sound engineering Figure 3b. BITTERSWEET Run, judgement is used to select these engineering characteristics. MORE's show that both tenders would Finally, it is important to note that perform relatively the same with regard the simulation results are only valid to mission effectiveness. However, the when making comparisons between BITTERSWEET, as an oceangoing tender, alternative designs within fixed weather would be extremely more costly to patterns, discrepancy rates, service operate than the REDWOOD in this area time distributions, and geography. No given its larger manning and fuel attempt should be made to extrapolate requirements. Moreover, additional any of these results beyond their output revealed that the BITTERSWEET was intended purpose. As with all models of unable to service 10 buoys in the 90-day this nature, other information (or interval due to draft restrictions. models) must accompany the analysis This would further add to the cost of before conclusions can be reached operating, because additional resources concerning the best tender design for a 1409 given region. If viewed in this light, SHIP REDWOOD (2 wea) the simulation model can produce results SAMPLE SIZE 20 MEAN 18.076307 which would not otherwise be available VARIANCE 61.595390 to the decisionmaker, thereby providing 99% CONFIDENCE INTERVAL 13.548595 22. 604020 a valuable tool for evaluating competing 40 38 buoy tender designs. 36 34 32 6. REFERENCES 30 28 26 14 1. Law, A. M. , and C. S. Larmey, "An PERCENT 22 Introduction to Simulation Usin 21 18 Simscript 11.5,- CACI, Inc-.. 16 Federal, 1984. 14 12 10 8 2. "Winds and Wave Summaries for 6 Selected U.S. Coast Guard Operating 4 Areas," Report No. CG-D-11-83, April 2 10 13 16 19 22 25 28 31 MORE 1983. TO LESS THAN 13 16 19 22 25 28 31 34 3. "Diskette Documentation for NCDCI-s Figure 4a. Sensitivity output - Element Archive Format," Release B - Limited to Calm Seas condensed, November 1985. 4. Kuipers, B., "Modeling Spatial SHIP REDWOOD (5 wea) SAMPLE SIZE 20 Knowledge," Cognitive Science, 2, MEAN 10 226603 VARIANCE 5:737406 pp 129-153, 1978. 99% CONFIDENCE INTERVAL 8.844747 11.608458 40 5. "Kaun, D.T., "Terrain Map Knowledge 38 36 Representation for spatial 34 Planning," Proceedings of First 32 30 Conference on Artificial 28 21 Intelligence, Applications, Denver, 24 CO, December 5-7, 1984, pp 578-583. PERCENT 22 21 18 16 6. Kuipers, B. and Alan Cline, "Coast 14 Guard Route Planning," U.S. Coast 12 10 Guard R&D Center Report, December 1, 8 1986. 6 4 2 7** 7- 3** 2- 1* 7 9 11 13 15 17 19 21 MORE TO LESS THAN 9 11 13 15 17 19 21 23 Figure 4b. Sensitivity Output - Moderate Seas 1410 USING SPACEFILLIWG CURVE TO GENERATE THE FEASIBLE ROUTES FOR THE SET PARTITIONING PROBLEM Kenneth S. Klesczewski university of New Haven Foundation Groton, Connecticut 06340 The heuristic for,generating feasible ABSTRACT routes that is described in this paper, uses a method of route estimation based on The Space Filling Curve. Route estimation The Vehicle Routing Problem, which with the Space Filling Curve is fast and is a variation of the classical Traveling relatively accurate. Salesman Problem, is and has been a topic of much interest. The basic premise of The basic premise of the vehicle the problem is that of a vehicle that must routing problem that this paper will visit a large group of nodes but is examine is that a vehicle that must visit limited by the distance it can travel in a large group of nodes but is limited by one trip. Therefore, several routes must the distance it can travel in one trip. be scheduled in order for the vehicle to Therefore, several routes must be visit every node. scheduled in order for the vehicle to visit every node with the vehicle Set Partitioning algorithms are a returning home in between each trip. An good approach to solving this problem. Up example would be a buoy tender who must to now, the feasible routes were only able service all the buoys in.a,given area. to be generated if the routes were already The tender obviously could not service all defined, such as an airline flight the buoys at one time so several trips schedule. with use of Space Filling Curve must be planned. Therefore, the entire Heuristics, feasible routes can now be schedule for the buoy tender could be generated for a random field of nodes thus generated. making Set Partitioning a viable option for the general vehicle Routing Problem. Ii. SET PARTITIONING INTRODUCTION This section contains a brief desdiiption of the Set Partitioning problem. It is presented as background The Vehicle Routing Problem, which for the other sections of this paper. is a variation of the classical Traveling This discussion will not explain any Salesman Problem, is and has been a topic algorithms,but will only describe the set of much interest. one approach that is partitioning problem. favored by several researchers is the use of set partitioning.. In a paber by The set partitioning problem can Marsten and Shepardson[l), several be written as an integer program[41 with successful applications are presented. As the basic formulas described in a result, there already exists good equations(l, 2, and 3) as they apply to algorithms for solving the set this particular problem. These can be partitioning problem. Two examples are expanded as in figure(l). the Branch and Bound algorithm(21 and the EnumerAted algorithm[31. However, very In these equations x represents a little has been written on the method for particular feasible route while c generating the the feasi 'ble routes for the represents the length of the route. The set partitioning problem. In some ith rows represent a particular node while specific applications,the feasible routes the jth columns stand for a particular can be efficiently generated but there route of nodes, thus the intersection a lacks a general algorithm or heuristic for is used to show if a particular node is nodes randomly located in an area. contained in a particular feasible route. The a values can be better visualized in The next step, therefore, is to matrix form as seen in figure(2). develop a method to generate the feasible routes for the set partitioning problem. CH2585-8/88/0000- 1411 $1 @1988 IEEE min(z)= cj xj (1) In brief, heuristics based on Spacefilling Curves is a method by which a*,, x@= i (2) N-dimensional space is mapped to a 1-dimensional line. The space in which X@= O'l j 1,...,n (3) the nodes are mapped from tend be a unit square or unit cube etc. Therefore, all values for dimensions are be scaled to where unit length. Since vehicle routing occurs on a plane, a 2-dimensional space, a unit square, is used. I if feasible route j is used in the partition The Spacefilling Curve heuristic xi creates a path through the unit square @O otherwise which visits every space in the plane as seen in figure(3). Where a node is mapped 1 if node i is in feasible on the 1-dimensional line is dependent on route j where it is positioned on the Spacefilling a; Curve. Using the Spacefilling Curve 0 otherwise heuristic described by Bartholdi and Platzman, the path is drawn and the node located by dividing the square into c@is the length of the feasible route quarters and then finding which quarter the node lies. After the node is located, that quarter is divided again into four and the process repeated. This process, FIGURE 1. which is recursive, continues until the node is in the center of the new divided min(z) = c,x,+ cax,+ caxa+ ... + ChxrJ square. Then the nodes location on the 1-dimensional line is calculated by where subject to it was at each recursion level. The nodes then receive a value from 0 to 1 according a,, x,+ a,,xa+ a,Ix-j+ + a,, x,,@ 1 to its location on the 1-dimensional line. These nodes are then sorted by value and a,,, x, + aa,,x.,+ a,;xa+ + aa,, xn@ the route is thus generated. The path the Spacefilling curve takes at the first three recursions are demonstrated in figure(3). a,,, x, + a,,x,+ a,,axl,+ + a,,,xn FIGURE 3. FIGURE 2. -@--l F-h a,, a,, als ... a J aal a., & aA3 ... aa, aa, asa a33 ... a LP I 1 11 1 ... a;S ... LEVEL 1 LEVEL 2 a, a... a., ... a r -i I r-h r-h r -i Li-rr@' III. SPACEFILLING CURVE L M r r Spacefilling Curves was first introduced by Peano, Hilbert, and Sierpinski in the late 19th and early 20th LEVEL 3 century. Since then, they have interested many mathematicians for its simplicity, speed of execution, and recursive quality. This heuristic creates routes one popular application of Spacefilling which circle back on themselves. The path Curves is their use by Mendelbrot in that navigates through the unit square generating Fractal Curves. Spacefilling begins and ends at the origin, coordinates curves along with the $pecifics used in (0,0). This lends itself quite well to this'heuristic are discussed extensively this problem since the vehic14 must return by Bartholdi and Platzman[51. at the end of each trip to its base. 1412 For this application,.two Now that the nodes have been additional procedures are added to the sorted with the Spacefilling curve Spacefilling heuristic to create a heuristic,, the generated route can be Modified Spacefilling Curve Heuristic. optimized. The inefficiencies are These steps are designed to improve the primarily caused by the convoluted path by results of the Spacefilling Heuristic. which the Spacefilling curve can take. They are framing, and flip-sort optimizor. This is seen in the example in figure(3). The framing step is used to translate the At this point, the Flip-Sort is utilized. nodes into the unit square for the This new optimizer is similar to the Spacefilling Curve heuristic while the R-optimal method which has been used by flip-sbrt optimizer is used to improve the Christofedes[6]. In actuality, the results of the heuristic. R-optimal method may be used in its stead, however, the Flip-Sort optimizer is much The framing procedure is faster with an order of growth, 0(n). particularly useful when the nodes that ate being routed are only located in one The Flip-Sort Optimizer differs portion of the plane, e.g. just on one from the R-optimal method in that the @side, as seen in figure(4). This internal order of the nodes which the generally occurs when nodes that are being Optimizer looks at stays the same, the considered for a route are all close only decision made is whether or not to together or when all the nodes considered reverse the entire order of the nodes (see are on one half of the plane. In these figure(5)). The assumption is that the circumstances, framing can be useful in Spacefilling curve will give a good route creating more efficient routes by for the nodes being examined. The problem .increasing the circle back effect of the. lies with the nodes association with its Spacefilling curve. neighbors, therefore, the R-optimal method will not gain you very much for the extra time used. FIGURE 4. 1.0 FIGURE 5. 0 N-4 EXAMPLE A 0.0 r 3 INITIAL RESULT The framing step transposes a JVr section of the plane which contains all the nodes to be routed into a unit square. This is done by finding an adjusted minimum(eq. 4) and maximum(eq. 5@ of the x and y coordinates and then inserting the location of the nodes into.equations(6 and 7) to get the new coordinates in the Original Path Resulting Path transposed unit square. These new coordinates are then inserted into the IF Spacefilling curve heuris,tic for the new route. 1DIST(NX,NX.1 + DIST(Nx,, Nx.,,., (8) is greater then adj.min(-() = .95 * MIN( (4) adi-max(-() = 1.05 * mAx(o,(, ... ve, (5) [DIST(NX Nx,, ) + DIST(Nx,, Nx..., (9) THEN Xi = __.xi - adj.min.(x) (6) adj.max(x) - adj.min.(x) Nx,,.through Nx,,.are "flipped" Yi = --- yi - adi.ir n.(y) (7) FROM N, , N., , Na , N, , Nr adj.max(y) - adj.min.(y) TO Ni , N, , NS , Na , Nr 1413 The nqdified Spacefilling Curve its own cluster but this would cause a Heuristic is efficient and fast. The large number of feasible routes to be Framing Step simply finds the smallest and generated for"the set partitioning step. largest x and y coordinates and then a It is likely that The best percentile simple transposition. This has a linear should be obtained through experimentation order of growth, 0(n). The Spacefilling in the particular application. Curve also has a linear order of growth since each node is processed separately. Each pair of nodes with a The last step, the Flip-Sort, has a linear difference less than -or equal to the order of growth, 0(n), because the nodes clustering difference are permanently that are examined are not reordered. clustered together. The order of the Therefore, the largest order of growth for nodes in each cluster is not important and this algorithm is that of the sorting is not be maintained. However, since these algorithm which is used once the nodes are close enough together they will Spacefilling curve has given the values to always be together when generating each node. This order of growth is 0(n feasible routes. log(n)) which is standard for sorting algorithms. As a result, this heuristic The next step is the feasible has an order of growth of O(n log(n)) due route generation. This step uses the to the sorting algorithm and is therefore order of the clusters as they were formed quite fast. in the clustering route. The same principal is used in that if nodes that are close together are clustered together IV. FEASIBLE ROUTE GENERATION then clusters that are close together should be routed together. The algorithm for routing the Starting with the first cluster vehicle is executed in three steps. The and later adding the succeeding clusters first step is to cluster the nodes, the in order, feasible routes are generated. second is to generate the feasible routes The length and path of the new route is using the clusters, and the third is to calculated using the Modified Spacefilling partition the feasible routes into sets. Curve Heuristic. Here is where the The first two steps use the modified framing procedure is most useful. The Spacefilling Curve and are the subject of nodes in the clusters being considered may this paper. be only in part of the.plane, and therefore a more efficient route may be Clustering is accomplished by calculated then using the path of the more using the modified Spacefilling Curve general clustering route. The length of discussed in the last section on all nodes the route represents the values(c ) used to be visited. The path that is generated for the columns in the set partitioning will be called the clustering route. Each problem. At this point, the next cluster node will then have the distance to its in order is added to the previously next neighbor calculated(see equation 10). examined nodes with the length again These differences(A;) are then sorted and calculated and another feasible route is with an arbitrary percentile, the generated. This process continues until clustering difference(Z) is found the 1'ength of the route being examined is (see equation 11). larger then the vehicle can handle in one trip. This length represents the maximum distance that the vehicle can travel in 6i = Distance between Ni and Ni+l (10) one trip. After this first group of routes is ge 'nerated, the next cluster in order is selected as the first cluster in = P% of the order Ai (11) the next group of routes and the process is repeated. This continues until everv cluster is used to start a group of The arbitrary percentile is feasible routes. decided upon by several factors, The important ones includes the maximum size Now that all the feasible routes of the feasible routes and the number of are generated, the set partitioning nodes to be visited. Even more important algorithm is applied. Although the is the optimization of the results and the algorithm is not discussed, it may be speed of the calculations. The smaller desirable to include certain steps to the percentile the fewer the number of improve its computing performance. The different clusters and therefore, the first step to consider is to eliminate algorithm will be faster but less some of the less efficient routes. For accurate. A larger percentile will of example, some of the routes that contain course have the inverse effect because less than one half-the total nodes in the there will be many more clusters with largest route starting with the same fewer nodes in them. The most efficient cluster, could be eliminated. Another choice would be to divided every node into possibility is to set up the set 1414 %M 6UTUOTqTqjed q9s aqq julod ST144 qV UT paqTjosap ST XT34EUI AITATIOR T T T 0 0 0 0 0 0 0 9 attl .UItjqTI06Te BUTUOTqTqjed qas aqq dn paeds Pug xTjq2m aqq asuaPuOD TTTm STqI 0 0 T T T 0 T 0 0 0 v apou e go pgaqSUT jaqsnTo e quesaidai ueo 0 0 0 0 T T T T 0 0 c A03 q0ea 'XT349W 9q4 AgTTdUITS 01 -paiTsap JT XT34VW AqTATIDe Ue OqUT paqpTsue3.4 aq mou ueo sainoi asaqj 0 0 0 0 0 0 T T T z uT umOqs eiv Pug paqejauab mou aje seqno3 alqTseaj aqq Is3e4snjo asaT41 qqTm T 0 0 0 0 0 0 0 0 T T (9T) oz, Z'Z T'T CT 6*0 T'Z 6*T VZ O*Z CT O*Z *(9T)uoiqenb9 UT UMOqS ST s3a-4snTo aqq Pug paqvTnolgo aq ueo (V) .194noi siqq qqTm '(T)aTqvl Pug (9)aan6f; UT UMOqS ST STqj -aqno3 6uT39qsnT3 atIq a4eiau96 ol sT daqs Ixau aT41 NOILVdVdSS HZISM0 SZIONSG saaqsnTo 9 6T* T OT* W EV ZT (GT) S = go jaqwnu s @T* OT W TT* 9V* TT z @T* s 6E' VC LS* OT sepou sjaqsnTo s *6E* z EV os* Ev* 6 (VT) go jaqwnu %d = go iaqwnu E 60* L W W L8* 9 E *v9* 8 ES* TO* ze, L (ET) E*Z = azTs 9qnoj xew 9 so* TT Z6* sz* st, 9 s ET* ZT T6* EV ov, s (ZT) %OV = %d z *OZ* 9 9p, EV 99' p T ET, E OV so* sz, E T *TV* v LZ' 86* vZ, z -6uTjaqsnjo ui aoueTeq poo6 e aATS v *LP* 6 EL' OT, Z6* T 0'4 %op 0-4 '49S ST (.%d) aTT-4uaojad A3e3-4Tq.7e z1flou ilia A X 3UON 9qI -S8TqeTJeA aSaqq ;0 4SIT 2 301 ,dzIsfIqD -30vds 'HOOD (ET Pug ZT) SUOT pvnba aaS -sapou a44 go u0T4Q30T aq4 q4Tm BuOT2 SaTqeT3eA TvIaAaS T 3`19VI auT;ap oq sT de4s 4sjT; aqj -e4noi 6uTlalsnTO aq4 6ut4ea3o UT pasn 'NOIIVUVdZS H3ISn73 S3,LON3G aq -4ou TTTM ainp9ooid BUTW93J aq4 , a3-enbs 4Tun e UT APQaITV aIV SaPOU aq4 a3UTS uaisrm ivNia s3iON3a .-aivnbs 4Tun e UT pa4nqTij9TP Ajwopue3 939 T43TqK 99POU 9ATOM4 sasn aTdwiexe STqI ,juos-aiia suoass zinou ssioNsa----- -wT4qTio6Tv aq4 ajei4suowap o4 paonpoid ST alduiexa TT2u's e 'UOTqOaS STqj UI X Z11CINVX3 'A aWT-4 6UT-4ndwoo go 4unouie T2104 a44 UT 96ueqo 3oulm 9 ATuO aNeul TTTm 1042'au96 94noi OTqTgeag SN4 go uoTITppv aq4 uaq4 'Apea3T2 wGTqoid 6uTuOTqTq3vd 4as a63eT e alpuvq ueo wa4sAs e ;I eqno3 aTqTseaj 8q4 q4Tm 40u Pug uIq4TI06Te 6uTuoT4TIaed qeS a44 qqTM SGTT U0Tj23aUa6 94no3 a3TIU9 aq4 uT q4moib go iapiO IsableT aql V 0.1 -4sit; pa3aPTSUOO eq plnoqs Aaq4 'SsaT ST Pa4TSTA aPOU 3ad 9 U119ij 8OUVISTP ;0 Pe9qJ9A0 8q4 M44 Pug Ua4;0 Iah, 'et ssaT awoq ujn493 KaT4-4 asnvoaq -4uaT6Tjja ajow aq oq pual saqno3 3efi3QT aqq 9OUTS -(g)a3n6T; uT umoqs s4lnsa3 eq4 qqTm 'Maq4 UT sapou ajow q4Tm sa4no3 8NTse@; MOTqoid pa4e3aueb aqq SaATOS wqqT3o6le 9q4 3ePTSu03 04 mOTqoad 6uTu0T4T4avd where VIII. REFERENCES the rows represent the clusters in the feasible route 1. R. E. MARSTEN and F. SHEPARDSON, "Exact Solution of Crew Scheduling the columns represent the feasible Problems Using the Set Partitioning routes themselves. Model: Recent Successful Applications", Networks, vol. 11, 165-177, (1981). FIGURE 8. 2. R. E. MARSTEN, "An Algorithm for Set Partitioning Problems", Management Science, Vol. 20, No. 5, (1974). to 3. R. S. GARFINKEL and G. L. NEMHAUSER, "Integer Programming", Wiley, 315-318,(1972). 4. R. S. GARFINKEL and G. L. NEMHAUSER, "Integer Programming", Wiley, (1972). b 5. J. J. BARTHOLDi, III and L. K. PLATZMAN, "Heuristics Based on the Spacefilling Curves for Combinational Problems on the Plane", PDRC Report Series 84-08, Vi. CONCLUSION Georgia Institute of Technology. 6. N. CHRISTOFIDES and S. EILON, This heuristic, which uses the "Algorithms for Large-scale Spacefilling Curve, introduces the Traveling Salesman Problems", Op. possibility of using set partitioning in Res. Quart. 23, 511-518, (1972). the general case of the vehicle routing problem. Up to now, set partitioning was used when the feasible routes were IX. BIBLIOGRAPHY obtainable in well defined applications. These routes either came pre-defined or were easily discernible. With this Kenneth S. Klesczewski heuristic, a random field of nodes can be visited with the feasible routes being Kenneth. S. Klesczewski is generated in a reasonable amount of time. currently working on simulation projects at the Coast Guard Research and Development Center, Avery Point, VII. ACKNOWLEDGMENT Connecticut, and is also an adjunct professor at the University of New Haven teaching statistics. He has an M.S. This research was done under degree in computer science from the contract with the University of New Haven University of New Haven and a B.S. degree Foundation and the United States Coast in Mathematics/Statistics from the Guard Research and Development Center. I University of Connecticut. His interests wish to give special thanks to Joseph A. include statistical models and Smith of the U.S.C.G. R.& D.C. without mathematical programming. whom this paper would not have been possible. b I @10 1416 RESULTS OF AN EXPERIMENT TO EXAMINE CERTAIN HUMAN FACTORS RELATING TO SEARCHES CONDUCTED WITH MARINE RADAR DUNCAN FINLAYSONi DONALD BRYANTi BYRON R. DAWEi A-J. ARMSTRONG* NORDCO LIMITED, P.O. BOX 8833, ST. JOHN'S, Nfld., CANADA A1B 3T2 *SEARCH AND RESCUE, CANADIA14 COAST GUARD, OTTAWA ABSTRACT As part of a series of Search and Rescue (SAR) training methods for electronic searches using experiments sponsored by Transport Canada, an marine radar. investigation of certain human factors relating to electronic searches using a standard marine radar was conducted using a navigation/radar 2.0 THE TCTI SOLARTRON NAVIGATION simulator system. Simulated search operations for SIMULATOR FACILITY weak radar targets partially obscured in a clutter field were carried out by experienced The system consists of a master control/monitor Canadian Coast Guard officers. The interrelated station and four individual bridge simulator factors of fatigue and motivation and the "area facilities (own ships). Each is equipped with a of concentration" within the detection envelope radar display, navigation equipment, were examined. The results have implications for communications gear, and helm control facilities. SAR planning and training. The equipment is controlled by proprietary software running on a DEC PDP-11/34 computer system. Each "vessel" can be navigated 1.0 INTRODUCTION independently or made to follow a prescribed route. A brief summary of system characteristics The importance of investigating certain human relevant to the experiment follows directly: factors affecting electronic searches using marine radar became evident in the initial search 1. Any specified search pattern and vessel speed and rescue detection experiment conducted by the can be simulated. Canadian Coast Guard (CCG) (Dawe et al., 1986). of specific concern were: 2. Up to 24 radar targets can be positioned in any desired location. 1. operator look direction selectivity on the radar plan position indicator (PPI); and, 3. Target strength and size may be varied in 2. the effects of fatigue and motivation on the steps allowing relatively weak targets to be ability of experienced CCG officers to detect simulated. Target characteristics, however, weak radar echoes. are not adjustable during an individual trial. Objective evidence and subjective observations suggested that radar operators tended to 4. Clutter simulation is not accurate but its concentrate on a relatively small sector of the intensity and range can be adjusted. Video PPI display centred on the heading marker and levels of the targets and clutter field can that the operators tended to become excessively be adjusted independently by the system fatigued after two hours on watch. To shed light operator during a simulation. on these issues in a short, cost effective manner, an experiment was planned using a 5. The simulated pulse length is menu adjustable Transport Canada Training Institute (TCTI) between 0.2 microseconds (us) and 3.0 us, and navigation/ radar simulator system. The specific antenna beamwidth is variable between 0.7 and objectives were to: 2.1 degrees to the 3 db point in steps of 0.17 degrees. 1. define quantitatively the area(s) of the PPI display that are viewed more frequently 6. A flat bed plotter is available to plot track and/or with greater concentration by and target positions at virtually any time experienced CCG officers; and, interval desired. 2. examine the time variability of detection success of weak targets in a clutter field. 3.0 THE EXPERIMENT The results of the experiment are expected to be used to improve search and rescue planning and For the purposes of this experiment each own ship operations and to provide guidance in SAR acted as an independent SAR vessel. Track CH2585-8/88/0000- 1417 $1 @1988 IEEE patterns were preset and all vessels maintained 1981). The actual detection envelope for a identical course and speed at all times. marine radar system is a complex function of target cross section area, and height above the The experiment was run over the five day period sea surface, sea state, 'and radar performance in July 1987. Two simulated searches of specifications, antenna height, and microwave approximately four hours duration were run each propagation conditions. For the purposes of this day. Eight CCG officers were made available for analysis the detection envelope has been defined the experiment and they were arbitrarily divided as the radar range selected by the operator. For into two groups of equal size. The individuals the most part, the radar operators elected to in each group participated as radar operators on use the 6 nautical mile range setting on the PPI the identical series of simulated search display. Since the theoretical radar range of patterns. targets very near the sea surface with an antenna height of 11 metres is approximately 7.5 nautical The CCG officers had either a Watchkeeping Mates miles, the commonly selected radar range was Certificate, an Ocean Navigator 2 Certification, appropriate. To examine the look direction or a First Mate Home Trade ticket. Sea time selectivity on the PPI display, the detection experience ranged from about I year to 25 years, data were binned by bearing, by range, and by with the median being very near three years. range and bearing. Actual and relative frequencies of detection were computed from the Eight parallel track or creeping line type search binned data. patterns, and two expanding square patterns were implemented. Twenty-four targets (the maximum InvestigatioIn of the fatigue/motivation factor possible) were distributed at random locations was reduced to an examination of the variability within the search area. The area was a function of target detection success in relation to time of the leg lengths and the vessel speed. The on search. For each operator and search, the actual number of detection opportunities varied ratio of the number of confirmed detections to since some targets became opportunities more than the total number of opportunities in consecutive once while others, normally because of their ten minute time intervals was determined. The location on the extreme edge of the search area, average number of the percentage of targets that never did become opportunities for detection. were detected for each time bin was computed to show the variability of detection performance An attempt was made to use ship parameters which with time. were, more or less, typical of CCG SAR vessels. Simulated search speeds, which were held constant during an individual search, varied between 10 5.0 RESULTS and 15 knots. A radar antenna height of 37 feet (about 11 metres) was selected. Table 1 shows a summary of detections by area. The highest percent detected was in the second The target "echos" were made as small as the quadrant; the lowest was in the third quadrant, system allowed. Video levels of both the targets even though there were approximately 10% more and the sea clutter representation were adjusted opportunities than in any of the other quadrants. to make the detection of the targets difficult. Comparing the port and starboard sides, the The clutter model employed is not indicative of highest number and percent detected was on the actual sea clutter, the pattern being symetrical starboard or right side of the PPI. The percent with the level decaying gradually with distance detected in the forward and aft semi-circles were from the centre of the PPI display to a minimum approximately equal. at 3 nautical miles. The simulated searches were conducted in deep water out of sight of land. With regard to detection range, Figure 1 shows This eliminated the requirement for navigational that more detections occurred in the 3 to 4 support. nautical mile (nmi) range than elsewhere. It is noted that the simulated clutter decreased The simulated searches were run with one CCG markedly after 3 nmi. However, a significant officer operating the radar unit in each own ship number were seen inside the 3 nmi range (area of for the length of the search; normally the greatest clutter) duration was approximately four hours. The operators were required to log the time, true An inverse relationship was noted between the bearing and range of each target sighting, and percent of targets detected and the nLanber of were asked to provide remarks whenever it was opportunities. The effect was removed by deemed appropriate. Since each operator had dividing the percentage of targets detected by identical opportunities for detection, no the number of opportunities. Figure 3 shows the communication regarding the experiment was results of all operators over all searches. permitted between the "vessels". There is an overall increase in detection success at the beginning (less than I hour) and towards the end of the searches (time on task of 3 to 3.5 4.02ATA REDUCTION AND ANALYSIS hours). The detection envelope is the moving area aro.und a search unit within which a target has a potential for being detected (Edwards et al., 1418 toward the end of the search. 6.0 DISCUSSION Another possibility is that the operator started The tendency to detect more targets in the second the search at maximum alertness, resulting in quadrant might be explained by the clockwise more detections, tedium and fatigue then rotation and the cueing effect of the heading followed, resulting in lower detections. marker. The time delay between the heading Finally, realizing that the search was coming to marker flash and the operators response to the an end the operator began to work more refreshed image may be such that full diligently. concentration may not occur until the second quadrant is being refreshed. By the time the The normalized time on search trends seem to operator had scanned the first and second suggest some residual effects after removal of quadrants and had started to look elsewhere the the "number of opportunities" bias. The overall third quadrant had been already refreshed, slight drop in performance level after about an phosphor decay had started and full hour or so does give some indication, however concentration was not achieved until the fourth small, that an operator should not be placed on quarter was being refreshed. Once the. heading radar watch for more than one hour. The slight marker flashes again the cycle repeats itself., A and temporary improvement around the 2 hour time greater concentration on the starboard side may mark correlates with the coffee break time and also be related to the regulation regarding risk indicates that a longer break period is necessary of collision in crossing situations wherein a for an operator's performance to return to the vessel which has another vessel on their own first hour level. The rise in performance in the starboard side shall keep out of the way of the latter period may be an artifact of the other vessel. fleagerness" effect discussed above. The significance of the peak at one hour is unknown. During the radar target detection experiment undertaken in Placentia Bay, Newfoundland (Dawe et al., 1986) it was found that all the 7.0.CONCLUSIONS detections occurred in the forward half of the radar displays and that the majority occurred The simulator experiments showed a number of within the + 45* of the ships heading. In the trends which have significance in terms of a SAR simulator experiment there were almost 11% more operations using marine radar. The look opportunities in the aft sector. Despite this, direction selectivity suggests that with current the percent detected in each sector is about radar systems, some investigtion is necessary in equal. This confirms the tendency to concentrate the need to utilize special search patterns to on the forward sector as indicated by the 1986 take into account this factor or some retraining experiment. In the simulator experiments the of radar operators to remove it. The use of operators knew they were not on a real vessel, raster displays might remove some of this look were not concerned about any potential hazards, direction bias. The tendency to detect more and were more concerned about detecting targets. targets at the edge of the clutter also suggests Had the situation been more realistic it is quite that some retraining might be necessary for SAR possible that they would have had an even greater operations. bias towards the forward.half. The significance of the inverse relationship The detection data suggested that more targets between the number of opportunities and percent were detected at the outer edge of the major detected is unknown in terms of its impact on a clutter zone. United States Coast Guard (USCG) SAR operation. The number of opportunities is personnel have noted that past experiments unrealistically large in terms of a real indicated that during visual searches, lookouts situation. tended to look along borders or edges (for example, along the side of a river, along the The time on search trend does indicate that to edge of a forest). This tendency might explain maximize radar search performance radar operators the phenomena seen here. The significant number should be relieved approximately every hour. A of targets seen inside the clutter zone is short coffee break is insufficient. The probably an artifact of the poor clutter significance of the other trends are unknown. simulation technique employed. Many of the trends indicated may have a lot to do No obvious explanation has been found for the with ergonomics and human physiology. Indepth inverse relationship indicated between the number investigations of these factors are clearly of opportunities and the percent detected. It beyond the scope of this experiment. might be, speculated that the operator thought after an, initial period of time he was seeing ACKNOWLEDGEMENTS false targets. (too many were detected), thus a reduction in the percent detected occurred in the This work was funded by the CCG and the mid-range after and initial higher percent Transportation development Centre. The authors detection. After a 2 to 3 hour period a "pattern wish to thank the staff of the navigation recognition effect" resulted in the operator facility at TCTI for their support and being able to more definitely recognize true co-operation. We also thank the-CCG officers who targets, giving a higher detection percentage volunteered for the experiment. 1419 REFERENCES Dawe, B.R., D.S. Bryant, and D.J. Finlayson, 1986. Search and Rescue Detection Experiment Placentia Bay, Nfld. Transport Canada Rpt. No. TP (8028(E). Edwards, N.C., S.R. Osmer, T.J. Mazour, and G.L. Hover, 1981. Factors Affecting Coast Guard SAR Unit Visual Detection Performance. Rpt. No. CG-D-09-82. U.S. Coast Guard Research and Development Center and Analysis and Technology, Inc. TOTAL TOTAL PERCENT DISPLAY SECTOR OPPORTUNITIES DETECTIONS DETECTED First Quadrant (0001-090') 3339 86 2.58 Second Quadrant (09cp -1800 ) 3672 124 3.38 Third Quadrant (18d'-2700) 3969 64 1.61 Fourth Quadrant (27CP-3600) 3567 88 2.47 Forward Sector 6906 174 2.52 Aft Sector 7641 188 2.46 Starboard Sector 7011 210 3.00 Port Sector 7536 152 2.00 TABLE 1 Summary of Total Number of opportunities, Total Numbe:: of Detections, and Percent Detected for Various Quadrants and Sectors of the Radar Plan Position Indicator Display 1420 NORMALIZED DETECTION RATIO NUMBER OF D p 9 9 p p p p p - 8 t: -- a a C3 m - - Q. a. 01 0 1. W 0 m cr> > A ;oi lb0 fA c 0 :m W n 4P- 0 0 RESULTS OF CANADIAN SHIPBORNE NIGHT SEARCH EXPERIMENTS DONALD BRYANT, BYRON R. DAWE, DUNCAN FINLAYSON, W. REYNOLDS7 M.J. LEWANDOWSKI NORDCO LIMITED, P.O. BOX 8833, ST. JOHNIS, NF, CANADA AlB 3T2 USCG R&D CENTER*1 GROTON, CONN. ABSTRACT metres. (m) , a beam of 16 m and a displacement of 3728 tonnes. It was equipped with a flume In the fall of 1987, a marine shipborne search stabilization system which acted to significantly and rescue detection experiment was undertaken dampen the natural roll of the vessel. off Nova Scotia employing the Canadian Coast Guard ship "Sir William Alexander". United Table 1 provides a summary of the major system States Coast Guard (USCG) visual search parameters pertinent to the two, night vision experiment methodology was employed. The targets goggles used. were 4 and 6-man life rafts. Night searches for these rafts with lights were carried out by Precise navigation information was essential to visual observers without visual aids. Night the experiment in order to determine ranges to searches were also carried out for unilluminated valid target detections and opportunities. liferafts using night vision goggles supplied by Fortunately the LORAN-C coverage in the area was the USCG. excellent. A differential hyperbolic LOkAN-C navigation system was utilized. A LORAN-C buoy, on loa 'n from the USCG, was moored in the centre 1.0 INTRODUCTION of the target area. Because the required accuracy (100 m) was available using the basic During the period from mid October to the last LORAN-C coverage, the LORAN-C buoy data was not week of November, 1987, a multifac Ieted ship based utilized. Be'arch and rescue detection experiment took place A Datawell non-directional wave rider buoy was on the Canso Bank off Nova Scotia, Canada. The primary objectives of the experiments were to deployed near the centre of the area., Onboard obtain visual sweep width information for processing, display and storage of wave data was Canadian Coast Guard (CCG) vessels under typical achieved through the use of a VHF telemetry link Canadian conditions; to add to the United States and a portable micro-computer. Coast Guard (USCG) data base on visual detection ,information for sea states 3 to 5, to determine Remote reading Weathermeasure Model 5121-B sweep widths for standard ship's radars against temperature/humidity sensors contained within unenhanced and radar reflector equipped life self-aspirating shields were mounted on either rafts; and to determine the sweep width side of the vessel. Measurements of the improvement that might be achieved using marine temperature and humidity on,the windward side of radars with more sophisticated signal processing. the vessel were made very hour. The targets used were commercial 4-man orange canopied life rafts and canopied 6-man yellow The ship's own meteorological instrumentation was military life rafts. In addition to the basic utilized by trained weather observers to measure, experiment visual night searches were undertaken on an hourly basis, temperature, humidity, wind for life rafts equipped with canopy lights and speed and direction, barometric pressure, and sea for unilluminated life rafts using night vision surface temperature. In addition, these goggles supplied by the USCG. Detection observers recorded sky conditions, prevailing experiments were also conducted utilizing targets visibility, precipitation, degree of equipped with search and rescue x-band radar whitecapping, and other pertinent data. transponders. This paper provides the results of the night searches for life rafts equipped with canopy lights and for the searches involving unilluminated life rafts using night vision 3.0 DATA COLLECTION AND METHODOLOGY goggles. It. was considered, to be very important that the data collection methodology be standardized and 2.0 THE EQUIPMENT 'compatible with that utilized by the USCG in similar experiments'. This would ensure that the The vessel used was the CCGS "Sir William results could be directly utilized by both Canada Alexander", a 1100 class Coast Guard buoy tender/ and the United States and indeed other countries icebreaker. The ship had an overall length of 85 undertaking research in this area. Thus the CH2585-8/88/0000- 1422 $1 @1986 IEEE methodology employed is essentially the same as 5.0 RESULTS that utilized by the USCG in their visual detection experiments. Night Vision Goggle Searches for unilluminated Approximately 20 single point target moorings Life Rafts were distributed over the experiment area. Targets were normally placed at 10 of these moorings. The experiment was broken into a The results are based on only 2 searches with a series of simulated searches averaging 4 to 5 total search time of 11.6 hours. hours in length. The targets were a mixture of 4 and 6-man life rafts. The factors tested for in the multivariate regression analysis were: After deployment of all the targets the ship was moved to a position well outside the radar and - lateral range visual horizon. A search pattern was derived. - wind speed During the search each detection was recorded - sea state (significant wave height) manually along with the appropriate data (range, - visibility bearing, time, system parameters, etc.). - precipitation cloud cover target type (4- or 6-man life raft) 4.0 DATA REDUCTION AND ANALYSIS METHODOLOGY The analysis showed that only lateral range had a The analysis methodology that was e .mployed significant effect on the probability of followed, in general, USCG procedures for visual detection M. detection experiments and involved a multivariate regression technique to compute a relationship Figure 1 shows the resultant POD versus lateral between various independent variables and. the range curve. The corresponding sweep width is probability of search object detection. In 1.36 nautical miles (nmi). ad'dition to lateral range, relevant independent regressors include attributes of the search object, and environmental conditions (for Night Visual Searches for Life Rafts with Lights example, significant wave height). Sweep width, the parameter used in search planning is determined by integrating the probability of The results are based on only 4 searches with a detection with respect to lateral range for total search time of 32.5 hours. Although more particular values of the significant independent searches did take place, it was discovered that variables. the canopy lights were not operating at full illumination all the time. Those searches for Prior to performing the analysis, a data which the illumination was low were subsequently reduction task was undertaken. A semi-automated eliminated from the analysis. The searches were procedure was developed to assist in Verifying conducted in the same manner as the day light target detections and for computing the lateral visual searches with the look-outs utilizing range or closest point of approach (CPA) of the binoculars. detection opportunities. After manually scrutinizing the detection data, the successfully The factors tested for in the multivariate detected opportunities were flagged in the regression analysis were: opportunities files. The data for individual searches were combined by search type forming the - lateral range following files: - wind speed - sea state.(significarit wave height) 1. Night vision goggles visual opportunity/ - visibility detection file, -. percent cloud cover 2. Night search for canopy lights opportunity/ - target type (4- or 6-man life raft). detection file. The analysis showed that lateral range was the only significant factor. Figure 2 shows the probability of detection versus lateral range The records also contained the observed curve. The derived sweep with is 2.2 nmi. environmental conditions at the time of detection Although this is only based on 4 searches the or, in the case of missed opportunities, at the spread in the data is low so that some confidence time of CPA. can be placed in this result. The multivariate regression technique was used to determine the statistically significant 6.0 DISCUSSION parameters. The variables were Ipreselected and checked for their significance at the 90 percent It is noted that in the analysis of both these confidence level in an interactive procedure. data sets it was seen that either wind speed or Those variables which did not contribute sigificant wave height could have been considered significantly were rejected from the analysis. significant parameters at slightly lower 1423 confidence levels, say 80% to 85%. This may be attributed to the fact that wind speed and significant wave height are loosely related to each other. It is felt that a larger data set Litton model m-912A/m-915A would be required to accurately determine their true significance. magnification 1.0 X No estimates of sweep width for night searches Field of View 40 degrees for life rafts equipped with standard lights have Resolution (limiting) .68 Lp/mR been reported previously. However the National Brightnetsgain (M-912A) 1000 Search and Rescue Manual (DND and CCG, 1985) does (m-915A) 1700 provide estimated sweep widths (in darkness) for similar devices. These are compared with the Litton Model M-927 NORDCO result in Table 2. Table 3 shows sweep widths derived by the USCG for white strobe life ring lights. An inspection of Tables 2 and 3 magnification 1.0 x + 5% suggests that the National Search and Rescue Field of View 40 de4-rees Manual estimates may be high while the USCG Resolution, on Axis (min) .68 Lp/MR results for life ring strobe lights are more Brightness gain 1200 comparable to those of the NORDCO results when the di f f er ing variable o f ship size, environmental condition, and light height are taken into account. Another interesting comparison can be made. For the prevailing condition encountered, the sweep TABLE 1 NIGHT VISION GOGGLES GENERAL SYSTEM width derived for this type of search is greater CHARACTERISTICS than the day light visual sweep width 0.2 - 1.5 nmi) now employed operationally. The searches involving the night vision goggles are believed to be the first of their kind ever undertaken. The small data set and the wide spread in the data suggests that a low level of confidence should be placed in the sweep width value obtained 0.3 nmi). Plus, the particular ESTIMATED SWEEP WIDTH night vision goggles used were meant to operate DETECTION AID (NAUrICAL MILES) with some sky light available. The 100% cloud cover was not optimum for their operation. Life Raft Canopy Light in Table 4 a summary of the environmental (From NORDOD Experiment) 2.2 conditions under which this data was collected is Electric Floating Lantern 1.0 provided. Hand Flashlight 3.0 Strobe LifeJacket Light 3.5 7.0 ACKNOWLEDGEMENTS This work was funded by the CCG and the Transportation Development Centre. The authors wish to thank the staff of the CCG, Maritime Region for their support and co-operation, especially Mr. Bernard Leonard and the officers Table 2 Comparison Between NORDCO Limited Derived Sweep width and crew of the CCGS "Sir William Alexander". We for Life Raft Canopy Lights and Estimates from the al so thank the USCG for its support and National Search and Rescue Manual for Similar Devices par tic ipat ion. 8.0 REFERENCES Department of National Defence and Canadian Coast Guard, 1985. National Search and Rescue Manual. Dept. of National Defence B-GA-209-001/FP-001, Canadian Coast Guard TP5421, Ottawa. 1424 TIME ON TASK (HOURS) SEARCH PLATFORM 1 3 WIND SPEED (KNOrS) 6 is 6 15 41 Foot Utility Boat 3.9 2.6 2.1 1.1 82/95 Foot Work Boat 3.9 2.6 2.1 1.1 Mean Experiment Environment Conditions Cloud Cover (Tenths): 0.6 Significant Wave Height (Feet): 2.4 TABLE 3 Sweep Width Values (NMI) for White Strobe Life Ring Lights Derived from USCG Experiments Nighttime Visual Searches (Lights On Rafts) VISIBILITY (run) WIND ( T,kt) HS (m) NOTES 1 15+ BCMG 5 SSE 20 VRG NWLY 8 1.0 [email protected] MAINLY WIND-WAVE 3 10 BCMG 15 SW-W 15 2.0 BCW. 1.4 24 12 BCMG 15 NNW 32-40 2.5 BcmG 2.3 WiND-WAVE ONLY 25 15 NWLY 25-Q 2.2 BCMG 1.9 WIND-WAVE ONLY Night Vision Goggles Searches VISIBILITY (nm) WIND ( T,kt) HS (m) NOTES 20 4-6 SW 25-30 VRG NW 25 1.8-2.0 WIND-WAVE ONLY 22 15 NW 10 BCMG 20 0.9-1.1 MAINLY SW SWELL TABLE 4 - ENVIRONMENTAL CONI)ITIONS 1425 z L) 0.4 x 0.2 0.0 1 1 ly X 0.0 1.0 Z.2 3.2 4.0 5.0 6.0 7.0 6.0 9.2 10.0 LATERAL RANGE C@ml) FIGURE I Night Vision Goggles Probability of Detection VS Lateral Range Curve 1.0 0.8 -x z 0 Li 0.6 H 0.4 m x 0.2 x 0.0 8.2 1.8 2.2 3.0 4.0 6.0 6.2 7.0 6.0 9.2 10.2 x Y LATERAL RANGE Cm@l) FIGURE 2 Night Search for Lights Probability of Detection VS Lateral Range Curve 1426 RESULTS OF A CANADIAN SHIPBORNE RADAR SEARCH AND RESCUE DETECTION EXPERIMENT BYRON R. DAWE, DUNCAN FINLAYSON, DONALD BRYANT NORDCO LIMITED P.O. BOX 8833, ST. JOHN"S, NEWFOUNDLAND AIB 3T2 ABSTRACT In the fall of 1987, a marine radar shipborne tender/icebreaker. The ship had an overall search and rescue detection experiment was length of 85 metres (m) , a beam of 16 m and a undertaken off Nova Scotia employing the Canadian displacement of 3728 tonnes. It was equipped Coast Guard ship "Sir William Alexander". United with a flume stabilization system which acted to States Coast Guard (USCG) visual search significantly dampen the natural roll of the experiment methodology was employed. The target vessel. A large A-Frame support for the ship's were 4 and 6-man life rafts, some with 2m crane was located just aft of the fo' c's le. Lunenburg lenses mounted on high flyer buoys at a The height of the top of the A-Frame was such height equal to that of the canopy top of 4-man that it was just below the height of the radar life rafts, and a small search and rescue radar antenna. transponder deployed on the ocean surface. The radar systems used were the Sperry 340, the The ship was equipped with 2 Sperry radar Sperry 4016 and the new Raytheon Pathfinder. ST systems: a conventional x-band system (Sperry radar. The number of searches and the target type 4016) and a Sperry 340 radar equipped with a varied for each radar. The sweep widths were Collision Avoidance , System. A Raytheon very narrow for all but the radar transponder. Pathfinder ST radar system was installed for the latter half of the experiment. Table I provides a summary of the major system parameters 1.0 INTRODUCTION pertinent to each radar. The radars are similar and it is felt that the only significant During the period from mid October to the last difference among them is the signal processing week of November, 1987, a multifaceted ship based utilized. search and rescue detection experiment took place on the Canso Bank off Nova Scotia, Canada. The The passive radar reflectors used were LENSREF primary objectives of the experiments were to (Lunenberg lens type) reflectors made by the obtain visual sweep width information for Maritime Facilities Co., Ltd. of Japan). The Canadian Coast Guard (CCG) vessels under typical reflectors have a minimum 2 square metre radar Canadian conditions; to add to the United States cross section in the horizontal plane and have a Coast Guard (USCG) data base on visual detection 250 wide vertical reflection pattern. information for sea states 3 to 5, to determine sweep widths for standard ship's radars against The radar transponders were Mitsubishi Electric unenhanced and radar reflector equipped life Corporation MELSWEEP Model H units. The r a f t s ;and to determine the sweep width transponder radiated omnidirectionally in the improvement that might be achieved using marine horizontal plane and up to 40 0 in the vertical radars with more sophisticated signal processing. plane. The effective radiated power was The targets used were commercial 4-man orange nominally 400 milliwatts. Once interrogated the canopied life rafts and canopied 6-man yellow unit responds with a signal that sweeps from 9.34 military life rafts. In addition to the basic to 9.48 GHZ. A train of radar "blips" are experiment visual night searches were undertaken transmitted extending outward from the for life rafts equipped with canopy lights and transponder's position. for unilluminated life rafts using night vision goggles supplied by the USCG. Detection Precise navigation information was essential to experiments were also conducted utilizing targets the experiment in order to determine ranges to equipped with search and rescue x-band radar valid target detections and opportunities. transponders. This paper provides the results of Fortunately the LORAN-C coverage in the area was the radar experiments. excellent. A differential hyperbolic LORAN-C navigation system was utilized. A LORAN-C buoy, on loan from the USCG, was moored in the centre 2.0 THE EQUIPMENT of the target area. Because the required accuracy (100 m) was available using the basic The vessel used was the CCGS "Sir William LORAN-C coverage, the LORAN-C buoy data was not Alexander", a 1100 class Coast Guard buoy utilized. CH2585-8/88/0000- 1427 $1 @1988 IEEE A Datawell non-directional wave rider buoy was 4.0 DATA REDUCTION AND ANALYSIS METHODOLOGY deployed near the centre of the area. Onboard processing, display and storage of wave data was The analysis methodology that was employed achieved through the use of a VHF telemetry link followed, in general, USCG procedures for visual and a portable micro-computer. detection experiments and involved a multivariate regression technique to compute a relationship Remote reading Weathermeasure Model 5121-B between various independent variables and the temperature/ hum id ity sensors contained within probability of search object detection. In self-aspirating shields were mounted on either addition to lateral range, relevant independent side of the vessel. These sensors were to be regressors include attributes of the search used to determine the presence or absence of object, and environmental conditions ( for microwave ducting and the subsequent derivation example, significant wave height) . Sweep width, of its effect on the sweep width. Unfortunately the parameter used in search planning is both the hunidity sensors failed shortly after determined by integrating the probability of installation. As a substitute, the humidity detection with respect to lateral range for derived from the ship's instrumentation was used. particular values of the significant independent The ship's own meteorological instrumentation variables. was utilized by trained weather observers to Prior to performing the analysis, a data measure, on an hourly basis, temperature, reduction task was undertaken. A semi- automated humidity, wind speed and direction, barometric procedure was developed to assist in verifying pressure, and sea surface temperature. In target detections and for computing the lateral addition, these observers recorded sky range or closest point of approach (CPA) of the conditions, prevailing visibility, precipitation, detection opportunities. After manually degree of whitecapping, and other pertinent data. scrutinizing the detection data, the successfully detected opportunities were flagged in the opportunities files. The data for individual searches were combined in individual files for 3.0 DATA COLLECTION AND METHODOLOGY each radar. It was considered to be very important that the The records also contained the observed data collection methodology be standardized and environmental conditions at the time of detection compatible with that utilized by the USCG in or, in the case of missed opportunities, at the similar experiments. This would ensure that the time of CPA. results could be directly utilized by both Canada and the United States and indeed other countries The records were sorted by lateral range then undertaking research in this area. Thus the binned using an iterative procedure to ensure methodology employed is essentially the same as that the equal size bins each contained that utilized by the USCG in their visual approximately the same number of records and that detection experiments. no bins were empty. Finally, the percentage of targets detected in each bin was calculated. Approximately 20 single point target moorings were distributed over the experiment area. For the passive radar targets a multivariate Targets were normally placed at 10 of these linear-regression technique was used to determine moorings. The experiment was broken into a the statistically significant parameters. The series of simulated searches averaging 5 to 6 variables are preselected and checked -for their hours in length. The targets were a mixture of: significance at the 90 percent confidence level in an interactive procedure. Those variables 1) empty life rafts; which do not contribute significantly are 2) life rafts with LENSREF radar reflectors rejected from the analysis. mounted inside (suspended from the canopy); 3) LENSREF reflectors mounted on spare buoys at Plots of the percentage of radar targets detected a height of 1.1 metres (simulating mounting versus lateral range, however, suggested that the at canopy height), and, choice of an exponential model normally used for 4) radar transponder deployed on the ocean visual detection analysis was not appropriate for surface. each of the radar detection data for all of the target types except the SAR transponder. Indeed, After deployment of all the targets the ship was it is clear that the detection process for these moved to a position well outside the radar and passive radar targets is quite different from visual horizon to any of the targets. A search that of visual detection. The radar operator has pattern was derived. Radar operators were to detect the target against a competing noise or deployed at the appropriate radars. Where clutter background. A number of detections often necessary screening was used to prevent the must occur before the operator can identify a operators from receiving visual clues. The "blip" on the screen as a target. Quite often a operator was changed every hour and given a one 11mental pattern recognition - like process" is hour rest. During the search each detection was required to identify the blip as a target. recorded manually along with the appropriate data If the target enters the zone of significant sea (range, bearing, time, system parameters, etc.). clutter, surrounding the ship in the near range it quite probably is lost again, and the 1428 probability of detection curve as a function of transponder results are not shown. The fewest range may actually decline in this area. Thus , detections occurred directly ahead where the intuitively, a monotonic exponential function is large A-frame was located. In both figures there probably not applicable for computation of the were more detections in the first and third probability of detection as a function of lateral quadrants and there were more detections on the range in this case. No other model has been starboard side. Overall there were more proposed to date, and it was decided that a cubic detections in the aft sections. spline curve fitting routine was the most appropriate to fit a curve to the percent detected data. The curve was extrapolated from DISCUSSION the first bin back to zero lateral range. The sweep widths for various target types using the General Issues three radar systems were then computed by numerically integrating the area under the The sweep width results provided for all but the empirically derived curves. transponder should be viewed in light of the methodology used to derive the sweep width Quite often the first detection of the values. The extrapolation of the curve to zero transponder occurred when the first response was lateral range is likely to lead to some error in received. Had the response not had a series of computation of sweep width. Further work will be radial delay lines associated with it the target required to develop an appropriate model for could no doubt have not been detected by the radar detection data. operator. This distinctive response would also no doubt enable easy detection in sea clutter and An examination of the plots of probability of no reduction in the probability of detection in detection versus lateral range indicates that a clutter. it was quite characteristic for the substantially reduced or zero probability notch radar operator to recognize the target at the will occur around zero lateral range. This, as first detection although it may have only been already noted, is probably due to sea clutter seen on one sweep. By the time the next single limiting detection and/or the A-Frame structure detection occurred (after as much as 5 to 10 full which might limit radar visibility for short radar scanner sweeps later) the operator was able lateral ranges. The question arises as to to obtain a range and bearing to the target. In whether the search track spacing should be this manner, with the transponder's distinctive sufficiently less than the derived sweep width to response, the human detection process is the same provide overlap in the gap. Further as that for visual detections. it was therefore investigation is obviously necessary to resolve considered appropriate to use the exponential this issue. For the purposes of the following curve and the standard visual analysis technique discussions the sweep width has been kept as is. to determine the probability of detection versus lateral range relationship and the resultant It was quite obvious during the experiment that sweep width for the transponder. one of the major problems in detecting the radar targets was that they were physically obscured a considerable portion of the time by large waves. This, then, is probably the primary reason for 5.0 RESULTS the reflector at canopy height giving the substantial improvement. The factors tested for each target type in the analysis for each were: lateral range, wind Radar operator motivation was clearly a problem. speed, sea state (significant wave height) and Each operator was relieved after 1 hour in order the presence of ducting. The analysis showed to obtain the best performance as indicated by that lateral range was the only significant the simulator experiment. Some operators liked factor. Table 2 summarizes the results, giving the Raytheon display while others preferred the derived sweep widths for the target types for standard PPI display. Some operators sometimes each radar. Table 3 provides a summary of the became pre-occupied with the Automtic Radar environmental conditions encountered. With the Plotting Aid (ARPA) functions on the Raytheon exception of the life rafts, for which all three system, while others were reluctant to adjust any radars were tested, and in which the standard of the displays to ensure optimum detection Sperry had a 0.0 sweep width, each pair gave per f ormanc e. essentially the same_,sweep width. Figures 1 to 4 show the POD versus lateral range curves for the Comparison Among Radars various radars and targets. As already noted with the exception of the transponder, these are Table 2 summarized the sweep widths derived. It based on splined fits. The fitted curves and is clear that for the weakest targets the more even the data points, in most cases, show a sophisticated radars outperformed the standard reduction in the near range. marine radar, although even their sweep widths were probably negligible. For the remainder of Figures 5 and 6 are polar diagrams divided into the targets there was no significant difference areas of equal size. The numbers indicate the between the Sperry 340 or the Raytheon radar and number of detections. The Sperry radar (both the the standard radar. This would initially suggest 340 CAS and the 4016) detections are shown that there was no significant difference between separately from those of the Raytheon radar. The the performance of the more advanced radars and 1429 that of the standard radar except in the case of second quadrant in the case of the Raytheon radar the weakest targets. Since these weak targets is likely to have been a direct consequence of were detected at the edge or within the zone of the shadow area created by the ship's mast. This significant sea clutter it would appear that the mast was behind and slightly to the right of the more advanced processing had its greatest benefit radar antenna, creating a shadow area covering within that zone. the 150* - 180* zone. Effect of Passive Reflector Enhancement The radar shadow areas (A-Frame and mast) noted 2 in the preceding paragraphs and the corresponding Table 2 suggests that placing 2 m passive radar reductions in the number of detections suggests reflectors within life rafts has little benefit that such factors will cause variations in the over not using a reflector at all. However, radar sweep width for individual ships. placing the reflector at canopy height shows a substantial improvement. In fact, the sweep Finally, no significant differences were found width derived is greater than that of the between the detection distribution of the normally used visual sweep width for comparable Raytheon raster display and the standard PPI conditions. displays. This suggests that display type does not affect operator look direction biases, an Radar Transponder indication that training is the primary factor. Both the radars tested gave about the same performance against the transponders. The sweep ACKNOWLEDGEMENTS width is about 6 to 7 times greater than that for a visual search under comparable conditions as This work was funded by the CCG and the specified by the National Search and Rescue Transportation Development Centre. The authors Manual. wish to thank the staff of the CCG, Maritime Region for their support and co-operation, especially Mr. Bernard Leonard and the officers The high performance of the radar transponder was and crew of the CCGS "Sir William Alexander". We only due to it being an active device. The also thank the USCG for its support and detection process, as described earlier was the participation. same as that of visual detection. As noted earlier one of the major problems observed in detecting targets on the radar was that they were physically obscured by waves. It is evident from REFERENCE the passive reflector results that had the transponder been placed at canopy height then a Finlayson, D., D. Bryant, B.R. Dawe, and A.J. considerably greater sweep width would have been Armstrong, 1988. Results of an Experiment to potentially obtained. Examine Certain Human Factors Relating to Searches Conducted with Marine Radar - Submitted Look Direction Selectivity to Oceans '88. Simulator experiment (Finlayson et al., 1988) results indicated certain biases as to where the radar operators looked. The small number of detections available from the experiment does make it difficult to make direct comparisons. However, some general comments can be made. The SY=K PARAML@ SPMRY SpMi RA@ Raytheon System was grouped separately from the other two radars because of its very different raster display. Peek @er W 50 50 25 Wa Guide Lose (dbl (Z ticieted, 2-Y) 6 5.5 1.0 As with the simulator experiment more targets Mtenna height were detected on the starboard side. The (-t-) 23.9 20.6 17.9 simulator results suggested a slight bias to the @idth Odi) V.rti1 25' 25' 25o forward section. The result here might be Hoti-L.1 0.8o 0.8o lo partially explained by the large A-frame P.1- 1,ength (US1 O@0710@2511@2 0.07/0.25/1.2 structure on the ship forward of the radars, @lae Repetition, which quite likely partially obstructed the view Frequency (HZ@ 4000/2000/500 3200/1600/53U 3000/2WO/lODG/750 Rec-i- -fee directly head for all three radars. This would Figure (di) 8.5 8.5 7.5 do have reduced detections in the forward section as Di,pl,y 16 inch PPf 16 inch PPI 23 inch ceeit- is evident from the polar diagrams which show P-eing CFM BinarY Pulse to pi@ overall minimal detections in the the 300 to 3300 - t__ sector. relation T@ I - Ne3or Cine-t-istics of @rlne Redar, The lowest number of detections overall occurred Wed in E.P.-ii,ent in the fourth quadrant, while the lowest number in the simulator experiment occured in the third quadrant. The slightly lower number in the 1430 SWEEP WIDTH (NMI) WIND SIGNIFICANT SPEED WAVE HEIGHT RADAR (kts) (M) RADAR RAFT RAFT WITH REFLECTOR ON TRANSPONDER REFLECTOR SPAR BUOY Sperry 4016 Lgt. - 43 0.6 -3.0 Sperry 340 8 - 43 0.9 -3.0 Sperry 4016 (Standard) 0.0 0.6 1.6 8.8 Raytheon Pathfinder Lgt. - 37 0.6 -2.5 Sperry 340 0.4 0.5 Not Tested Not Tested Radar Against Transponders 6 - 30 0.9 -2.0 Raytheon 0.3 Not Tested 1.8 8.7 Dominant Conditions 1 20 - 30 1.0 -2.0 TABLE 2 Stlmmary of Marine Radar Sweep Widths TABLE 3 Summary of Environmental Conditions Derived From Experiment Sperry 340 Raytheon Pathfinder @perry 4016 Am ne-red val@ Sperry 340 X 0 Are -red val@ 6.2 Ne 6.6 1.0 2.6 3.6 0.9 6.6 6.0 7.S 6.0 9.0 10.8 @.a 1.9 2.6 3@8 4ps spa LVERAL RANGE (nao) LAIERAL RANGE C.13 FIGURE 2 Derived Radar probability of Dete@tion - Lateral Range 11@ I Derived Radar Probb,l,ty of Data@tlon -- Lateral Range ag inat 4 and 6-an Life Raft3 vith 2 sq-e netre Relationahip against 4 and &--n Life Mft. Reflectors Hountd inside 1.0 Spa@@y 4016 9.4 ------- Raytheon Pathfinder X & 0 Are --d valuee 0.6 1.0 2.0 3.9 ..G 7.0 6.9 9.0 LATERAL RANGE (-i) FIGURE 3 Derived Radar Probability of Deb-tim versus Lateral Range against 2 square matte Reflectore motmted at 4-aan Life Raft Canopy Height 14311 1.8 @K x Rai Pathfinder ------ Spccy 4016 x x Measured val@ (::th red& a had ti.lly the same values) 9.4 9.2 O'S 0.9 1.0 ZmS 3mO 4 a 5.0 6@13 7.0 8.8 9.9 10.0 LAIEOAL RANOC (-) FIGUFM 4DMri-d Radar Probability of Detection -a- Lateral PAnge Relationahip Ageinet Radar Traapondm 0 0 330 30 330 30 300 60 2 27 s 27 z 2 2 2-- 12, 210 150 tic 150 is. 4 2 cang. rings 2 mi range rings -@4 1+7 Figure 5R@l&tLYS Target Detection Po3itions for the 5percy 4016 i,,,* 6R:ltZ.rarqet Detection P-t-a for the Raytheon and Radar sy- P thf@ ST Radar Systas Z 1432 RESLLTS OP A CAWTAN VL9LX EEARai AND REME DENECITCN D(@@ 9JVM FRIAY3M, BU;CN R. DAM, Dallam aWVT NaR[M LD= P.O. B3( 8833, ST. JCFN'S, Nfld., CXSAM AlB 312 ABSTRACT valid target detections and opportunities. A differential hyperbolic LORAN-C navigation In the fall of 1987, a marine shipborne search system was utilized. A LORAN-C buoy, on loan and rescue detection experiment was undertaken from the USCG, was moored in the centre of the off Nova Scotia employing the Canadian Coast target area. Because the required accuracy (100 Guard ship "Sir William Alexander". United m) was available using the basic LORAN-C States Coast Guard (USCG) visual search coverage, the LORAN-C buoy data was not utilized. experiment methodology was employed. The targets were 4 and 6-man life rafts. Searches for these A Datawell non-directional wave rider buoy was rafts were carried out by visual observers. deployed near the centre of the area. Onboard processing, display and storage of wave data was achieved through the use of a VHF telemetry link 1.0 INTRODUCTION and a portable micro-computer. During the period from mid October to the last Remote reading Weathermeasure Model 5121-B week of November, 1987, a multifaceted ship based temperature/humidity sensors contained within search and rescue detection experiment took place self-aspirating shields were mounted on either on the Canso Bank off Nova Scotia, Canada. The side of the vessel . Measurements of the primary objectives of the experiments were to temperature and hLunidity on the windward side of obtain visual sweep width information for the vessel were madeevery hour. Canadian Coast Guard (CCG) vessels under typical Canadian conditions; to add to the United States The ship's own meteorological instriffnentation was Coast Guard (USCG) data base on visual detection utilized by trained weather observers to measure, information for sea states 3 to 5, to determine on an hourly basis, temperature, humidity, wind sweep widths for standard ship's radars against speed and direction, barometric pressure, and sea unenhanced and radar reflector equipped life surface temperature. In add it ion , these rafts ; and to determine the swee p width observers recorded sky conditions, prevailing improvement that might be achieved using marine visibility, precipitation, degree of radars with more sophisticated signal processing. whitecapping, and other pertinent data. The targets used were commercial 4-man orange canopied life rafts and canopied 6-man yellow military life rafts. In addition to the basic experiment visual night searches were undertaken 3.0 DATA COLLECTION AND METHODOLOGY for life rafts equipped with canopy lights and for unilluminated life rafts using night vision it was considered to be very important that the goggles supplied by the USCG. Detection data collection methodology be standardized and experiments were also conducted utilizing targets compatible with that utilized by the USCG in equipped with search and rescue x-band radar similar experiments. This would ensure that the transponders. This paper provides the results of results could be directly utilized by both Canada the visual searches for life rafts. and the United States and indeed, other countries undertaking research in this area or using derived sweep widths in SAR missions. Thus the 2.0 THE EQUIPMENT methodology employed is essentially the same as that utilized by the USCG in their visual The vessel used was the CCGS "Sir William detection experiments. Alexander", a 1100 class Coast Guard buoy tender/ icebreaker. The ship had an overall length of 85 Approximately 20 single point target moorings metres (m) , a beam of 16 m and a displacement of were distributed over the experiment area. 3729 tonnes. It was equipped with a flume Targets were normally placed at 10 of these stabilization system which acted to significantly moorings. The experiment was broken into a damp the natural roll of the vessel. series of simulated searches averaging 4 to 5 hours in length. The targets were a mixture of 4 Precise navigation information was essential to and 6-man life rafts. the experiment in order to determine ranges to CH2585-8/88/0000-1433 $1 @1988 IEEE After deployment of all the targets the ship was Table I shows a comparison of Sweep width values moved to a position well outside the radar and obtained using this model with USCG sweep widths visual horizon. A search pattern was derived. for various values of the regression parameters. During the search each detection was recorded manually along with the appropriate data (range, bearing, time, system parameters, etc.). 6.0 DISCUSSION Table I shows newly derived sweep width values 4.0 DATA REDUCTION AND ANALYSIS METHODOLOGY developed by the USCG from experimental results which are to be incorporated in Canada's National The analysis methodology that was employed Search and Rescue Manual. Referring to Table 2 followed, in general, USCG procedures for visual it can be seen that the environmental conditions detection experiments and involved a multivariate for this experiment are generally closer to the regression technique to compute a relationship higher wind speed and sea state conditions listed between various independent variables and the in the table. it can be seen from Table I that probability of search object detection. In the NORDCO Limited derived sweep widths under all addition to lateral range, relevant independent conditions are approximately a factor of two regressors include attributes of the search greater than the values derived by the USCG for object, and environmental conditions (for smaller vessels. A great deal of confidence may example, significant wave height). Sweep width, be placed in this result as the maximum the parameter used in search planning is log-likelihood model used for analysis shows a determined by integrating the probability of near perfect fit has been obtained. The maj or detection with respect to lateral range for difference between the USCG experiments and this particular values of the significant independent experiment is the size, stability, and overall variables. seakindliness of the vessel. The CCG vessel was much larger than the USCG 90 ft. work boats and Prior to per forming the analysis, a data was equipped with an efficient flume reduction task was undertaken. A semi-automated stabilization system. The results indicate that procedure was developed to assist in verifying a more stable search platform greatly enhances target detections and for computing the lateral search detection performance. range or closest point of approach (CPA) of the detection opportunities. After manually An interesting result here is the significance of scrutinizing the detection data, the successfully the raft type parameter which was specified as detected opportunities were flagged in the either a 4 or a 6 to represent the number of men opportunities files. The oppor t unity/ detection each raft was designed to support. The larger data for individual searches were combined by 6-man life raft was constructed of yellow fabric search type; the file records also contained the while the 4-man life raft was international observed environmental conditions at the time of orange in colour. Further work needs to be detection or, in the case of missed conducted to determine the significance of the opportunities, at the time of CPA. size and colour of the search object. The multivariate regression technique was used to determine the statistically significant parameters. The variables were preselected and 7.0 ACKNOWLEDGEMENTS checked for their significance at the 90 percent confidence level in an interactive procedure. This work was funded by the CCG and the Those variables which did not contribute Transportation Development Centre. The authors significantly were rejected from the analysis. wish to thank the staff of the CCG, Maritime Region for their support and co-operation, especially Mr. Bernard Leonard and the officers 5.0 RESULTS and crew of the CCGS "Sir William Alexander". We also thank the USCG for its support and The factors tested for significance in the participation. multivariate regression analysis for the day light visual searches were: 8.0 REFERENCES lateral range wind speed Department of National Defence and Canadian Coast sea state (significant wave height) Guard, 1985. National Search and Rescue Manual. visibility Dept. of National Defence B-GA-209-OOI/FP-001, precipitation Canadian Coast Guard TP5421, Ottawa. cloud cover target type (4- or 6-man life raft) The analysis showed that the optimum regression model in the maximum log-likelihood sense consisted of lateral range, visibility, significant wave height, and raft type. 1434 SWEEP WIDTH VISIBILITY WINDS 15 KNOTS WINDS 25 KNOTS SOURCE (NAUTICAL MILES) SEA 2-@ FEET SEAS > 4 FEET 4-MAN RAFT 6-MAN RAFT 4-MAN RAFT 6-MAN RAFT 10 2.0 2.4. 1.0 1.2 USCG (90 Foot Work Boat) 15 2.3 2.8 1.2 1.4 10 3.7 5.1 2.5 3.8 NORDCO (280 Foot Icebreakers/ Buoy Tender) 15 5.4 7.0 4.1 5.5 NOTE: Significant Wave Height for NORDCO Derived Sweep Width was set at 3 Feet and 5 Feet, Respectively for this Comparison TABLE 1 Comparison of USCG Derived Sweep Widths (Nautical Miles) with those Derived from NORDCO Experiment for Similar Conditions VISIBILITY (NM) WIND T, kt) HS (m) NOTES 2 15+ NW 10 1.7-1.9 MAINLY SE SWELL 4 15 SWLY 10 1.0-1.3 MAINLY SE SWELL 5 8 INCR 15 NW 25-43 1.5 INCR 2.7-3.0 WIND-WAVE ONLY 7 15+ WNW 15-20 1.0 INCR 1.5-1.7 8 15+ SWLY 25-30 1.2 INCR 1.4 WIND-WAVE ONLY 9 15 BCMG 12 SWLY 20 1.4 WIND-WAVE ONLY 10 15+ SSE 20-28 1.1 INCR 1.4 11 15 1.4-1.6 WIND-WAVE ONLY 12 15+ NWLY 22-30 1.9-2.3 1.3-1.8m WIND WAVE + SELY SWELL 13 15+ LGT & VRBL 0.9-1.0 WSW SWELL ONLY 16 15+ SELY 18-28 0.9-1.1 WIND-WAVE ONLY 17 15+ BCMG 9 SELY 17-25 1.2 INCR 1.5 WIND-WAVE ONLY 18 2 BCMG 1/8 SSE 40 VRG SLY 20-25 2.3 INCR 3.0 WIND-WAVE ONLY 19 12 BCMG 5 SLY 25-30 BCMG SW15 1.6-1.8 21 12 INCR 15 NNwLy 6-12 1.2-1.4 MAINLY SKY SWELL 23 8 BCMG 5 NE 20-25 1.5 WIND-WAVE ONLY 26 15 NWLY 25-36 2.0-2.2 WIND-WAVE ONLY 29 15 wLy 15 BCMG SW 15-20 0.6 INCR 1.0 WIND-WAVE ONLY 30 1/8 INCR 15+ WSW 15 VRG WNW 15 2.2-2.4 MAINLY SSW SWELL 31 15 LGT & VRBL 1.8-2.1 MAINLY SSW SWELL 32 15+ SW 17-24 1.5-1.8 MAINLY SSW SWELL TABLE 2 Environmental Conditions 1435 A NEW COAST GUARD SEARCH TECHNIQUE Frank Replogle, Jr. U.S. Coast Guard Research and Development Center Groton, CT 06340 For three years the U.S. Coast Guard Research and Development Center has been pursuing a new airborne search technique. it shows promise of detecting liferafts with a greater than 95 percent probability at a search rate of 250 square nmi per hour. Historically the Coast Guard has used two search sensors, the eye in daytime for small objects and radar in day or nighttime for large objects. Search by aircraft has provided area coverage rates many times as great as search by boats. However, searching by eye in an aircraft produces high levels of stress because of the area coverage rates required. Searching for a liferaft from a jet aircraft might be compared to searching the area of a slide projection for an object of dimensions only 1/3001th of the width of the screen. In addition to possibly containing the target object, the screen is filled with wave clutter. If the aircraft is flying at 2000 ft at a speed of 250 knots, the searcher must search the full area every five seconds. Because of the difficulty of performing this search in the required time, the probability of detection is typically 60 to 75 percent for a single pass, and repeated passes must be carried out if this probability is to be increased. Ofttimes the availability of equipment and personnel does not permit this. Search for larger objects with radar is less taxing for the operator. Here the target "blip" may subtend 1/1001th of the screen dimension, and the time requirement for searching a frame is relaxed to 250 seconds. Evidently any improved technique for searching for small objects must utilize aircraft, must have automated detection -- with a high detection probability, and must operate day or night. At the R&D Center active and passive optical techniques have been considered. It was found that the amplitudes of wave clutter are drastically reduced in the Mid and Far Infrared wavelength region. In the Far IR wave clutter signals are typically small in comparison to target self emission signals out to nadir angles of 60 degrees. Thus in a passive system automating detection is reduced to thresholding of signals and determining if the object sensed possesses the characteristics of the target desired. Fortunately passive Far IR imaging apparatus has been developed for many applications. The apparatus nearest to the Coast Guard's search requirements is the Navy's high performance line scan IR reconnaissance equipment. Since this was developed for reconnaissance, readout is on film, which must be developed. CH2585-8/8810000- 1436 $1 @1988 IEEE However, the Coast Guard requirement is for real time imaging, which requires the addition of a real time TV presentation to this equipment. Since the number of scene pixels covered in a line scan is several times the number which can be presented in a TV display line, it has been necessary to sense the presence of a target automatically and then automatically to provide the operator with a zoomed TV presentation of the target. Thus the actions of the operator are reduced from performing a continuous taxing search to indicating occasionally if a target is the type desired. To test the utility of this approach, services of Rochester Institute of Technology to fly an Infrared scanner plane over targets on Lake Erie were contracted. The electrical signals from the scanner were recorded on magnetic tape. This was replayed at the laboratory into an apparatus of the type mentioned. When the signals met the amplitude and length criteria, an operator cueing ("target high lighting") bar was added to the video. Targets were a 41 ft Search and Rescue boat with two liferafts tethered to the stern. Figures 1 and 2 show a sample frame of a full field video image and a zoomed portion of this field. Current theoretical efforts include modelling probabilities of detection of small targets in different types of weather conditions. For this a model giving the temperature difference between a liferaft canopy and the sea under different atmospheric and time conditions is being developed. To this expected clutter and noise signals will be added to give probabilities of detection and false alarms. The current experimental effort includes having the Naval Air Development Center modify a high performance IR scanner to record on magnetic tape. The equipment will be flown in a P3 over Block Island Sound under a variety of weather conditions. Then the taped signals will be replayed into a laboratory imaging apparatus intended to simulate an aircraft search station. With this the effectiveness of the overall technique will be judged. If this is favorable, equipment of this type will likely be installed in Coast Guard search aircraft. 1437 FIGURE 1. Infrared Map Showing Search and Rescue Boat and Two Liferafts. FIGURE 2. Zoomed Portion of Figure 1 1438 ALASKA SAR FACILITY ARCHIVE AND OPERATIONS SYSTEM Robert W. Berwin Jet Propulsion Laboratory 4800 Oak Grove Drive Pasadena, CA 91108 ABSTRACT The Jet Propulsion Laboratory and the University of Table 1. Spacecraft and SAR Parameters Alaska Geophysical Institute have initiated a task to receive, record, process, and archive Synthetic CHARACTERISTIC E-ERS-1 J-ERS-l Aperature Radar (SAR) and optical images at the Fairbanks campus. Data will be acquired from the Orbit Altitude (km) 800 600 European Space Agency (ESA) E-ERS-1 and the Orbit Period (min) 90 85 National Space Development Agency of Japan (NASDA) J-ERS-1 spacecraft. Orbit Inclination (deg) 98 98 Downlink Data Rate (Mbps) 1 5 60 JPL is currently developing the Alaska SAR Facility Radar Transmit 5.3 1. (ASF), consisting of the Receiving Ground Station, Frequency (GHz) (C-band) (L-band) the SAR Processing System, and the Archive and Operations System. This paper focuses on the Swath Width (km) 80 75 Archiving and Operations System which catalogs and Mbps=Megabits/sec archives signal data and derived data products and provides an interactive product order function for users. 2. SYSTEM DESIGN In addition, a Mission Planning Subsystem provides the capability to predict satellite swath coverage of the spacecraft, manages data ' acquisition Fig. 1 shows the ASF functional block diagram. The requests, and produces an operations schedule for Receiving Ground Station (RGS) tracks the satel- the Receiving Ground Station and the SAR Processing lites based on predicted satellite locations and System. captures and records data specified by approved data acquisition requests and operational sched- ules. The SAR Processing System (SPS), which is Data products processed and archived at ASF will be based on an Advanced Digital SAR Processor available throughout the NASA research community, @eveloped at JPL [5][6], processes the data into as well as to non-NASA-researchers. images. These SAR products are then cataloged and archived by the AOS either on tape, magnetic disk, or 12" digital optical disk (DOD). 1. INTRODUCTION A Geophysical Processor System (CPS) is currently being designed and implemented at JPL and emphasis The National Aeronautics and Space Administration is on the SAR Ice Ocean portion (see note below). (NASA) has issued a Memorandum Of Understanding This processor will provide automated analysis of with the European Space Agency (ESA) [1] and the SAR imager acquired by the ASF. In particular, National Space Development Agency of Japan (NASDA) the GPS wi@l produce products for ice motion vector [2] to record and process Synthetic Aperature Radar maps and ice classification maps. These products (SAR) and optical data at the University of Alaska will be archived and distributed by the AOS. The Geophysical Institute (UAF GI) in Fairbanks. In GPS will be available to local UAF personnel and order to implement this memorandum, JPL is visiting science team members. currently developing the Alaska SAR Facility (ASF). An Interactive Ima e Analysis System (IIAS) is a Early in 1990 the European Space Agency will launch general purpose wortstation co-located with the ASF a Remote Earth Sensing Satellite, E-ERS-1 [31 and which will receive data products from the AOS. in 1992 NASDA will launch a Japanese Earth This system will support research investigations by Resources Satellite, J-ERS-1 [4). Characteristics local science team members and visit ing researchers of the spacecraft and SAR instrument parameters are at UAF. The IIAS is being developed by UAF. shown in Table 1. The optical sensor parameters for the J-ERS-1 spacecraft will be found in [7]. The Image Processor for Optical Data (IPOD) (7] [8) is a system which will be co-located with the ASF The ASF will operate 24 hours per day for at least to process data from the J-ERS-l optical sensor and 3 years beginning with the launch of E-ERS-1, with to provide for distribution of the resulting the goal of remaining operable for product ordering products. Only a small portion of the optical data up to 10 years after the end of the last satellite will be processed into final products at ASF. transmission. The ASF will receive 5 minutes of Users will be able to order these products through real time SAR data from E-ERS-1, 10 minutes of real the product order function of the AOS. time SAR and optical data from J-ERS-I, and up to 80 minutes of J-ERS-1 recorded SAR And optical dump for NASDA. CH2585-8/88/0000-1439 $1 @1988 IEEE GEOPHYSICAL ARCHIVE AND OPERATIONS PROCESSOR SAR PROCESSOR SYSTEM (SPS) SYSTEM(AOS)_ SYSTEM (G1PS) RECEIVING GROUND ----I - STATION (RGS) ALASKA SAR )ST ARCHIVE SAR PC _-*'IPROCESSOR@ P R OC 1E SS' 0 R AND ICE/OCEAN @C E @R CATALOG SUBSYSTEM IMAGE PROCESSOR SAR/OPS FOR OPTICAL DATA LAND (IPOD) OPS GIPOD ,UAF ------- GEOPHYSICAL : USER INSTITUTE I------- INTERACTIVE IMAGE ANALYSIS SYSTEM MISSION (IIAS) ALASKA PLANNING SUBSYSTEM Fig. 1 ASF Functional Block Diagram The AOS consists of an Archive and Catalog A catalog is the next level of information Subs stem (ACS) and a Mission Planning Subsystem granularity down from , the directory and is (MPS@ [9]. The ACS archives SAR data and products, physically located at each node. The user can gain manages information about the SAR data, products, access to the catalog from the directory. and optical data, and also provides means to order Typically, the catalog contains detailed informa- the SAR and optical products. The MPS predicts tion about whole * data sets specific to a swath coverage of the spacecraft, prepares data discipline, data center, and/or other facility. acquisition requests in a multi-mission environ- ment, and provides for multi-satellite conflict analysis. The MPS also provides operational The inventory provides information about data set schedules for the RCS to obtain data and for the granules usually used to identify or locate the SPS to plan its activities. granule(s) (see below), given the specification of the independent variable range(s). The AOS software architecture will be based on the concept of a catalog system. Conceptually, a A granule is a set of data samples. For example, a catalog system consists of directory, catalog, and granule may be the unit of data storage in a inventory levels. ph sical medium such as magnetic tape (one tape fiTe) or a paper chart. The directory is a level which provides the user with a brief description of what data are available The inventory is the main operational level of the at each node. These descriptions help to make an ACS. The inventory is being designed with a menu initial determination of the potential usefulness driven interface which presents the user with a of a data set for a user's application. Informa- series of menus, or screens, which prompts and tion on the location of more detailed descriptions guides the user through a menu network. These and/or the data set itself will be found in a menus allow the user to enter search criteria for directory. UAF G EO, INSTI 1440 selecting, assessing, and ordering data. Search 5. Geocoded lo-res image (+ photos): criteria include items such as 1) location and time Produced by rotating a lo-res image to of datatake, 2) image -id (a unique number given to grid north and projecting the image on the product when generated), and 3) site name a base map (usually Universal Trans- (assigned by a user for later reference). These verse Mercator or Polar Stereographic criteria allow the user to view a list of products projection) (both images and photos) which meet these input 6. Reprocessed full-res and lo-res images search parameters. (+photos): User can request reprocess- ing of signal data to full-res and From this product list, the user can then select lo-res images by specifying parameters one or more of these products and have descriptions different from standard processing. of each product displayed on the screen. These Useful for centering features and descriptions include the percentage of mutual highlighting certain characteristics. coverage of the image and the user's site area, and an earth coordinate vs. image pixel number display. At this time, the user can place an order for any The AOS is designed to archive 465 SAR and film of these products. products daily as a result of recording and processing data throughout the mission lifetimes of The design of the searching and ordering capabili- E-ERS-1 and J-ERS-1. This archivin will require storage facilities consisting of 29 Gigabytes on ties of optical data and GPS products is still 1211 DOD for lo-res, Geocoded full-res, and Geocoded preliminary. However, the procedures of the lo-res images, 3 Gigabytes on magnetic disk for the inventory are expected to operate in a similar catalog and inventory, 11,200 Gigabytes on magnetic manner to that of SAR products. tape cassetes for raw signal data, full-res, and complex images, plus a library for storing Note: Funding for implementation of IPOD and the approximately 112,000 photo packets. CIPOD (Geophysical IPOD) and SAR/OPS portions of the GPS is not approved at this time. 3.2 Optical Products Optical products are processed and archived by the 3. PRODUCTS IPOD and ordered through the inventory function of the AOS. Distribution of full-res, lo-res, Digital Terrain Maps, and Custom Produced Products is on 3.1 SAR Products CCT and 5.25" digital optical disk. Browse images are produced for any product and is routed through Standard products are produced by the SPS for all the AOS to the user's workstation. SAR signal data recorded by the RGS. Special products, are generated by the SPS and the AOS in response to user requests only. Standard, repro- The characteristics for optical products are shown cessed, and geocoded full-res photos are generated in Table 4 and are defined as follows (these from the corresponding image data by means of a definitions were taken from reference [7]): FIRE recorder. Lo-res photos are produced from a lo-ores image data file using a Lasertechnics camera. The lo-res photo products are negative and 1. Full-res image: an image including all positive transparencies and a positive print. correction data and raw video 2. Lo-res image: a subsampled, subbanded product providing browseable imales. The characteristics for these products are shown in This product will have a b ina'y c oud Tables 2 and 3 and are defined as follows: cover mask 3. Digital Terrain Map: an elevation Standard Products: image map derived from stereo pairs. These are produced at user-selected 1. Full-res image: 4 looks, <30 m spatial resolutions up to the source resolution, 12.5 m pixel spacing resolution of 18x24.2 meters 2. Lo-res image (+ photos) : derived from 4. Custom Produced Product: a variety of full-res image by 8x8 averaging of the customized products based on user pixels, approximately 60 looks, <24.0 m specified parameters resolution, 100m pixel spacing 5. Browse Product: compressed images of any image product residing in the IPOD Special (by request) Products: archive 1. Full-res photos 2. Computer compatible signal data: Converted raw signal data suitable for recording onto a Computer Compatible Tape for processing by users 3. Complex image data: an image in which the Doppler phase information at natural pixel spacing has been retained 4. Geocoded full-res image (+photos): Produced by rotating a full-res image to grid north and projecting the image on a base map (usually Universal Transverse Mercator or Polar Stereo- ,graphic projection) 1441 Table 2. SAR Image Product Characteristics SAR PRODUCT DAILY IMAGE PIXEL FRAME SIZE FILE SIZE QUANTITY RESOLUTION SPACING (KM) (MBYTES) (M) (M) Full-res image 25 (E) <30 12.5 10OX100 64 50 (J) Lo-res image 5 (E) <240 100 10OX100 1 10 (J) Geocoded full-res 1 <30 12.5 10OX100 128 image Geocoded lo-res image 10 <240 100 10OX100 128 Reprocessed full-res 15 <30 12.5 10OX100 64 image Reprocessed lo-res 15 <240 100 10OX100 1 image GPS full-res product 6 TBD TBD TBD TBD GPS lo-res product 60 TBD TBD TBD TBD Complex image 1 <10 NATU- 30X50 100 RAL M-E-ERS-1, (J)--J-ERS-l, TBD-to be determined Table 3. SAR Photo Product Characteristics 4. PRODUCT FORMATS AND MEDIA PHOTO DAILY QUANTITY MAP SCALE PRODUCT The SAR data will be distributed in the format Full-res 5 (E) 1:500,000 consistent with the Committee on Earth Observation 10 (J) Satellites (CEOS) [10) . This format organizes the data into 1) a Volume Directory File, 2) a Lo-res 25 (E) 1:500,000 SARLEADER file, and 3) a SARTRAILER file. 50 (J) Distribution of SAR and optical products will be on Ceocoded 1 1:500,000 Computer Compatible Tape (CCT) and 5.25" digital full-res optical disk. Geocoded lo-res 10 1:1,000,000 Reprocessed 15 1:500,000 full-res 5. FUTURE ENHANCEMENTS Reprocessed 15 1:500,000 lo-res I E-E-ERS-1, J=J-ERS-1 Browse Images: In addition to ordering data through the product order function, the user will be a le to request browse images. which are full-res and lo-res Table 4. Optical Image Product Characteristics images which have been compressed in size suitable for electronic transmission to the requestor's OPTICAL DAILY SIZE GEOGRAPHICAL remote work station. Once a compressed image has PRODUCT QUANTITY (MB) COVERAGE reached the user's workstation, it can subsequently be decoded, or uncompressed, and displayed in its Full-res 2 305 75kmx225km (approximately) original form. Lo-res 2 60 full swath Digital 3 7 75kmx75km Future Data Ordering: Terrain Map A user will be able to request data to be taken at some future time from a particular satellite and Custom 75 variable user sensor. These data acquisition requests, entered Produced selectable interactively from a user's workstation, will specify target location, observation period, Browse 150 65 variable datatake repeat interval, and whether the datatake Browse products are variable according to product is to be from an ascending or a descendin bit. wg or compressed Requests will be entered into a database ich the mission planner will use to merge them with Quantities based on 10 minutes of data previous requests and resolve multi-satellite acquisition/day processed conflicts. Updated data acquisition requests will be submitted to flight agencies for approval. The quantities represent one option considered in the Phase B study [7]. Exact amounts of optical image products are to be determined. 1442 6. CONCLUSIONS 8. REFERENCES The Archive and Operations System (AOS) is designed [1] "Memorandum of Understanding between NASA and to archive Synthetic Aperature Radar (SAR) data ESA on the Reception of E-ERS-1 Data at products produced and derived by an advanced Fairbanks", January 1986. digital SAR processor and to inventory data [2] "Memorandum of Understanding between NASA and products produced by the Image Processor for NASDA Concerning the Direct Reception of Optical Data. The AOS will also provide a data JERS-1 Data at the Fairbanks Station", management system whereby the scientific community January, 1988. can interactively search the inventory and order data products. Geophysical products pertaining to (3] Erich, U. and D.H. Gottschalk, "ERS-1 and Its ice motion vector maps and classified ice maps will Potential for Industrial Utilization", be available to users through the inventory and Earth-Oriented Applications of Space Technol- product order function as well. ogy, Vol. 4., No. 4, pp 211-219, 1984. [4] Y. Horikawa and H. Ohba, "Studies on Japan's Earth Resources Satellite-l", Earth-Oriented The Mission Planning Subsystem is designed to Applications of Space Technology, Vol. 5, No. provide users with ordering of data takes of 3, pp 185-192, 1985. interest from future fly-bys of the satellites over the Alaska station mask. The MPS also provides (5] "Computer Sciences and Data Systems", Vol.2, schedules of datatakes for ASF station personnel to p. 248, Proceedings of a Symposium held at plan antenna activities for receiving and recording the National Conference Center in Williams- incoming data and to assist the SPS operator in its bur Virginia, NASA Conference Publication activities. 245R: November 18-20, 1986 [6] Carande, R. and B. Charny, "Alaska SAR Processor", IGARSS'88, Edinburgh, Scotland, The ASF is an integrated system designed specifi- September 13-16, 1988. cally for receiving data, producing products, and distributing these products to the scientific [7] "J-ERS-1 Optical Sensor Image Processing community. The functions of the ASF represent a System Phase A Study Draft Report", Goddard major advancement from current systems - for Space Flight Center, December 1987. processing and distribution of data products. Data [8) "J-ERS Optical Sensor Image Processing System gathered within the Alaskan station mask will Conceptual Design Draft Report", Goddard support studies and scientific analyses . in Space Flight Center, December 1987. ocea@io rapl_@Iy of the open ocean and ice margins, glaciology, geologic processes of river and delta (9] J. Hilland, "ASF AOS FRD11, JPL D-4738, formation, and seasonal vegetation changes [11]. September 1987 (a JPL internal document) [10] Final Report on SAR Data Product Format Standards by the CEOS Working Group on SAR 7. ACKNOWLEDGEMENT Data Standards, November 1987. (11] Carsey, F. "Alaska SAR Facility Science The work described in this paper was carried out at Requirements", JPL Internal Document D-3668, the Jet Propulsion Laboratory of the California January 1988. Institute of Technology, and supported under a contract with the National Aeronautics and Space Administration. 1443 THE SEARCH FOR SOUTH AFRICAN AIRWAYS FLIGHT 295 Michael K. Kutzleb Steadfast Oceaneering, Inc. for various projects. For this task, the services of Steadfast Oceaneering, Inc., the Navy's prime contractor for underwater Steadfast Oceaneering, Inc. recently search and recovery operations, were performed a successful search operation required. for a South African Airways 747 Steadfast Oceaneering Inc., with aircraft that had crashed in the Indian Ocean. Using a special deep ocean sonar offices in Falls Church, VA and Fort system, the Steadfast search team Lauderdale, FL, is a leader in the field located the aircraft in less than 14 of underwater search and recovery, and hours of sonar search in water depths specializes in deepwater operations. of over 4400 meters. Steadfast is the Steadfast was the prime contractor for the prime contractor for search operations search for KAL Flight 007, the Air India for the U.S. Navy's Supervisor of 747, and the space shuttle CHALLENGER. Salvage office. The tiny island of Since its incorpration in August 1979, Mauritius served as the base of Steadfast has conducted over 100 other operations for mobilization and search tasks around the world for both deployment of all search assets. This military and civilian customers, in is believed to be the deepest success- addition to performing numerous routine ful search operation on record-deeper survey tasks in deep and shallow water. even than the TITANIC operation of Steadfast is a wholly owned subsidiary of 1986. Oceaneering International, one of the world's largest underwater services company providing diving, Remote Operated vehicles, and engineering services to both oilfield and government customers On 28 November.1987, South African. worldwide. Airways Flight 295 enroute from Taiwan to Mauritius disappeared into the Indian PRE-TASK PLANNING ocean shortly before its scheduled arrival. The Boeing 747 Combi aircraft, The first order of business was to which was carrying passengers as well as identify what assets were required and cargo, had reported smoke in the aircraft where to commence the search. Both the and had just completed an emergency Cockpit Voice Recorder (CVR), and the descent to 14,000 feet when communications Flight Data Recorder (FDR) on SAA Flight were lost. Floating debris was located 295 were equipped with Dukane 37 kHz the following day approximately 135 miles underwater beacons. These beacons are northeast of Mauritius. activated upon immersion in water, and are designed to survive an aircraft crash and Based upon the U.S. Navy's success in help pinpoint the location of the flight locating the pingers from the Air India reorders using underwater acoustic locator 747 in 1985, the South African Government equipment. With theoretical detection requested the U.S. Navy Supervisor of ranges for these beacons of up to two Salvage assistance. An agreement was miles, large areas can be covered fairly signed on 3 December 1987, after favorable quickly using suitable pinger locator determination under Section 607 (a) of the systems. Foreign Assistance act of 1961, and further legal determination that use of Based on this, it was decided that U.S. Navy assets for humanitarian purposes Phase I of the search would concentrate on would not violate the terms of the listening for the beacons, while Phase II comprehensive Anti-Apartheid Act of 1986. would involve the use of side scan sonar The SUPSALV office administers several either to map the debris if Phase I was civilian contracts for the provision of successful, or to continue the search if specialized search and salvage services the pingers were not located during Phase I. CH2585-8/88/0000- 1444 $1 @1988 IEEE PHASE I Two ocean-going salvage tugs from Pentow Marine in South Africa, the WOLRAAD The Supervisor of Salvage keeps two WOLTEMADE and the JOHN ROSS, were provided pinger locator systems in their inventory as platforms for the first phase of the for use in locating downed military search. A German research vessel that aircraft, and provided them for use on was operating in the area, the SONNE, was this task in reponse to an official tasked to conduct a bathymetric survey of request from the South African government. the search area using a SeaBeam survey Each system consists of a towed system. This information proved to be hydrophone, a cable, and a topside invaluable throughout both phases of the processing/display console. The search effort. The SONNE also had a video hydrophone and underwater electronics sled aboard capable of obtaining photos of housing are pressure rated to 10,000 psi, the seafloor in water depths of up to and could therefore be towed at a 6000 meters. The R/V AFRICANA, a South sufficient depth to ensure detection of African fisheries research vessel the beacons in the water depths in the outfitted with a hull-mounted acoustic search area, which ranged from 2500 to listening system, was also used during 4500 meters. Previous tests using these the Phase I pinger search. locator systems had shown a maximum An ARGO navigation chain, using detection range of approximately 3700 meters under ideal conditions. shore sites on Mauritius, Rodriguez, and Cocos I-gland, was set up and operated by A team of U. S. Navy and Steadfast Geoteam, a Norwegian survey company. The personnel, headed by Captain Bartholomew ARGO system could only be used during the of the SUPSALV office, departed for day, since ranges to two of the sites Mauritius on 4 December to meet with DCA exceeded the maximum night time range of and SAA personnel. Captain Bartholomew, the system. This resulted in the use of along with Tom Salmon and Bill Walker of GPS satellite positioning during the night the SUPSALV office, assisted the DCA after the ARGO signals were lost. With Executive Committee in the planning phase the two systems, accurate positioning was of the search. The organizational . possible approximately 18 hours per day. experience that the SUPSALV personnel gained in projects such as the Air India The WOLRAAD WOLTEMADE got underway and CHALLENGER search and recovery tasks on 11 December.,,with,,the JOHN ROgS*j.oining was of immense assistance in the the search the following day. A test preliminary phases of this effort. pinger was dropped near the area in 3600 meters of water, both to test the locator Mike Kutzleb, President of Steadfast systems used and to serve as a standard and in overall charge of the Steadfast to compare with the-sounds actually Project Team, together with Dr. Johan detected during the operation. The Strumpfer of the South African Institute AFRICANA arrived on scene to assist on 17 for Maritime Technology, collected all December. available loss data for analysis. Using metereological data and positions of Trial runs conducted on the test floating wreckage obtained by various pinger confirmed that the towed locator search aircraft over a two day period systems were indeed capable of detecting following the accident, debris set and the pingers in the prevailing water drift were calculated to derive a depths. A consistent detection envelope projected impact position. This position of one mile was observed, which resulted was then compared to other data such as in the selection of one mile spacing for the projected flight path to define a the search tracks. This gave an overlap preliminary search area for the Phase 1 100% to ensure that all areas were search. covered. The search was truly an international Phase I continued until 2 January effort. Equipment came in from around the 1988, when it was felt that the batteries world, and had to be cleared through in the pingers would have been exhausted. customs, transported across the island to Several possible sound sources were heard, Port Louis, unpacked, set up, and tested. but none of them were determined to be Pinger locator systems, navigation systems, definite pinger sounds. A total.of 1005 winches, and A-frames were needed for each square miles was covered by the three of the two vessels, and had to be ferried ships during this phase. to the ships at anchor since all berths in the port were full. Retrieval of floating debris by commercial and South African Naval vessels was still an ongoing effort. 1445 PHASE II and a take up reel capable of storing up to 14,000 meters of towcable. A Once Phase I was completed, it was hydraulically operated A-frame was used decided to demobilize all assets and plan for launch and recovery .of the towfish, for Phase II - the side scan sonar with an oceanographic sheave for search. All of the data used in the fairleading the towcable over the stern. preliminary analysis was reviewed and refined, and new bits of information The search teamgpt underway on 22 were added. During one of the SONNEIS January and arrived in the primary area video runs, newspapers and several pieces the following morning. High winds and of mail were observed on the bottom in rough seas from a nearby tropical one area. This area was along the depression prevented the sonar fish from calculated debris drift axis, coincided being launched for the next 36 hours, so with the flight path, and also fit in this time was used to lock in and well with the best estimates of the calibrate the GEOLOC navigation system. projected impact position of the By Sunday evening, 24 January, the weather aircraft. Based on this information, an had abated sufficiently to permit launch area of high probability, measuring four of the sonar fish, and the first line was by ten miles, was selected as the primary started just after midnight. Towing a area of interest for the Phase II sonar sonar fish 45 meters off the bottom in search. 4500 meters of water at the end of a 9000 meter wire is no small feat; thus the The search vessels used during the first search line was run in an area of pinger locator phase were not suitable relatively flat topography. This allowed as platforms for a deep sonar search due the operators to become familiar with to their inability to maintain the slow system control and response before search speeds needed and the lack of tackling the more difficult, central adequate deck space required for portion of the area where the aircraft installation of the deep ocean winch and wreckage was believed to be located. The the sonar A-frame and handling system. SeaBeam charts of this high probability, The OMEGA 801, a 65 meter supply type central area showed several peaks rising vessel with variable pitch propellors 600 to 1500 meters off the seafloor amidst and a large open deck, was chartered to generally rough terrain. serve as the platform for the sonar search. The 50 kHz sonar operating on a scale of 300 meters per side was used for the During the break between Phases I search. Lane spacing of 300 meters was and Phase II, it was decided to replace chosen to give a theoretical overlap of the ARGO navigation system with a system 100 percent, subject to steering errors, that would provide reliable navigation navigation system errorr and an around the clock. A GEOLOC system, anticipated offset in the towfish track owned and operated by CGG, a French relative to the search vessel. This track geophysical survey company, was offset was actually observed to vary from contracted to provide the high accuracy 0 to 200 meters during the course of the positioning required during the Phase II search. sonar search. GEOLOC, which is manufactured by Sercel, operates in the The second search line was run 2 mHz band using a spread spectrum through the center of the area, and a technique, and has a maximum range of promising contract was noted on the sonar approximately 1000 km. Radiated signal records. Subsequent runs over the next strengths are kept below the ambient two days showed a typical aircraft debris noise level to eliminate interference, pattern, with pieces spread out over a and the system is immune to the 1000 meter square section of the bottom. tropospheric and sky wave interferences News of the discovery was passed to the so common to other medium and long range DCA Executive Search Committee on 27 precise navigation systems. January. Sonar mapping continued for another week using shorter range scales The sonar system used for the search and the 100 kHz system for higher was built around the Klein Smartfish, a resolution and greater detail of the dual frequency (50 or 100 kHz) full ocean various pieces of debris. The largest depth towfish. The signals were piece in the debris field measured 40 multiplexed at the towfish and passed meters long by approximately 5 meters through a 9000 meter coaxial towcable to wide, with most pieces measuring less than the Klein topside processing, recording, 5 meters in overall size. and display equipment. The winch The water depth at the site where consisted of a traction unit/power pack 1446 the wreckage was located was 4450 meters. Rocky terrain existed within two miles east and west of the crash site ' with mountains rising 700 to 1500 meters f rom the -ocean f loor. At the site itself, the bottom was flat and featureless, which greatly assisted the search team in mapping the area and flying the towfish at the low altitudes required for the high resolution sonar imaging. Although the sonar portion of the search was almost immediately successful, it was by no means an easy task. Steadfast personnel had conducted numerous search tasks in water depths ranging from 2000 to 3000 meters, but had never searched in depths of 4500 meters. As a guage to measure the complexity of this task, consider that the deepest location of a lost object to date was the TITANIC discovery in 1985. The TITANIC, close to 300 meters in length, was located in less than 4000 meters of water, or over 500 meters shallower than the SAA 747. The entire fuselage of a 747 aircraft measures only 69 meters in overall length, and from previous experience it was predicted that the aircraft would be broken up into numerous small places and spread out over the bottom, thus making detection difficult at best. The wreckage was indeed broken up and spread out as predicted, but the experience of the sonar operators in interpreting the records, as well as the excellent quality of the records themselves, resulted in the successful completion of the task in spite of difficult operating conditions in water depths previously considered unreachable. 1447 PERFORMANCE OFCOAST GUARD MEDIUM RANGE SURVEILLANCE (MRS) AIRCRAFT RADARS IN SEARCH AND RESCUE (SAR) MISSIONS R. Q. Robe and D. F. Paskausky G.L. Hover U.S. Coast Guard Research and Development Center Analysis & Technology, Inc. Avery Point 190 Winthrop Blvd. Groton, Connecticut 06340-6096 New London, CT 06320-6223 ABSTRACT program, Coast Guard planners used visual search data developed during During June 1987, the U.S. Coast Guard World War II together with a limited conducted a series of tests to determine amount of detection data collected in the ability of two airborne radar 1955 (reference 1). Field experiments systems, mounted aboard a Coast Guard conducted before 1978 collected MRS aircraft (HU-25A), to detect small relatively few samples, ignored SAR-type objects. The two radar systems significant variables and did not which were evaluated were the AN/APS-127 account for missed targets. No reliable forward-looking airborne radar (FLAR) search guidance was available for and the AN/APS-131 side-looking airborne searches using electronic sensors prior radar (SLAR). to the current series of experiments. Realistic radar searches were conducted 2. EXPERIMENT DESCRIPTION to collect data using unalerted sensor operators and standard search patterns. The site used for the experiment was a 15- by 30-nautical mile area located in Analysis of the detection data confirmed the Atlantic Ocean off Ft. Pierce, that the probability of detecting these Florida. All SAR objects were located targets varied with significant wave in this area. Actual search tracks height, search altitude, and target size assigned to the aircraft for different as well as lateral range to the target. data collection objectives were not limited to this area. Allowance was 1. INTRODUCTION made for the aircraft to fly beyond the test area so that the aircraft would be During June 1987, the U.S. Coast Guard parallel to the major axis, wings level Research and Development Center (R&D prior to it entering the experiment Center), together with units of the CG area. For longer range data Seventh District and the Seventh collection/detection opportunities, the District CG Auxiliary, conducted a 4- flight track was parallel to, but week experiment to evaluate and measure seaward of the search area. the ability of the crew of a CG MRS aircraft (HU-25A) to detect SAR-type Three types of life rafts and ten objects using the aircraft's FLAR different boats were used as search (AN/APS-127) and SLAR (AN/APS-131) objects/radar targets during the sensors. The SAR-type objects used were experiment. The boats used provided a 6- to 10-person life rafts of commercial good variety of vessel types, 24 to 43 types and pleasure boats in the 24- to feet in length overall. This range of 43-foot range. The test was designed to vessel length was selected as collect data under as realistic a set of representative of the vessel size circumstances as possible. The expected to be encountered in a majority capabilities of the whole system of search and rescue cases. The vessels including radar operators, radars, were constructed of fiberglass except signal process ing/display, and aircraft for a 41-foot aluminum hull Coast Guard was included in the evaluation. utility boat. The boats and rafts were not outfitted with radar reflectors. Since early 1978 the R&D Center at Target boat radars were not operating Groton, Connecticut, has conducted at- during the experiment period. sea experiments designed to produce visual and electronic search guidance Target locations and aircraft positions for use in U.S. Coast Guard SAR planning were monitored using an automated and operations. Prior to this research microwave tracking system (NTS) CH2585-8/88/0000- 1448 $1 @1988 IEEE consisting of a Motorola Falcon 492 3. HU-25A RADAR SYSTEM DESCRIPTIONS system coupled with an HP350 computer. The HU-25A Guardian is a Falcon 20 jet This system, developed by the Coast aircraft specially modified to perform Guard R&D Center for the SAR Project, the medium-range surveillance missions provided target and search unit of the U.S. Coast Guard. These missions positioning and search track include SAR, law enforcement, fisheries reconstruction accurate to better than patrol, and marine environmental 0.1 nautical miles. The MTS master protection. A limited number of these station, for this experiment, was Guardian aircraft are equipped with both located in Ft. Pierce, Florida. Two the AN/APS-127 FLAR and the AN/ASD-6 fixed reference stations were located to AIREYE multisensor surveillance system. the north and south of the master The AIREYE system includes the AN/APS- station. These locations provided line- 131 SLAR. Both of the Guardian's of-sight tracking of searcher and target positions. airborne radars were evaluated during this experiment. The trackline of the HU-25A often took The AN/APS-127 FIAR is an X-band air-to- it beyond the range of the MTS. When surface search radar developed to detect this occurred, the position of the small targets in a sea clutter aircraft, as displayed on the onboard environment. The radar operates at a navigation computer, was used to peak power of 200 KW, scan rate of 720 reconstruct the track. These positions deg/sec, and a beam width of 5.0 degrees were recorded once per minute while horizontal and 6.5 degrees vertical. A outside MTS coverage area and were 7-inch plan position indicator (PPI) is recorded as a navigation tie-in two or used for a radar display . This PPI is three times on each search leg while designed primarily for operation in the within MTS coverage. During data search mode and was used for all FLAR reconstruction all aircraft positions data collection. The FIAR system were converted to the MTS coordinate contains special selectable features base. that may enhance system performance when Target and aircraft (when within MTS used correctly. Range scales are range) positions were recorded selectable from 5 to 160 nautical miles continuously by the computerized with the option of moving the display tracking system, displayed in real time origin from its normal centered position on the CRT and recorded on a to any location on the PPI. Only the microcomputer hard disk every 15 to 30 results on the 10 nmi ranges were used seconds. in this report. Recorded target and aircraft position The AN/APS-127 offers three distinct data were used to generate an 8- by 12- display modes: heading stabilized, inch hard copy plot of each search. ground stabilized, and north stabilized. Target locations were marked on these The heading-stabilized display provides plots with letters. The aircraft tracks a PPI presentation wherein targets and were plotted with a 11+11 to designate terrain move relative to the sweep every fifth position fix. A hard copy origin, which represents aircraft chronology of the search and target position. North stabilization aligns craft positions was also generated to the.display with magnetic north, and the accompany the plot of each search. degree marks around the scope represent Using these plots and the detection magnetic bearings from the aircraft. logs, accurate lateral range The ground-stabilized display provides a measurements and detection/miss PPI presentation that is an unchanging determinations could be made. A target view of the earth's surface as long as was considered an opportunity for the selected area remains within radar detection on any given search leg if the range. This stabilization mode provides aircraft passed it within the selected a greater signal gain than the other radar range scale distance. If a logged modes and has been determined to be the target detection could be correlated best mode for small target search with the target*s position, it was (reference 2). considered a detection. Otherwise, a Previous AN/APS-127 experiments were miss was recorded for the target on that performed in 1983 and 1984 using Coast particular search leg. These detections Guard HU-25A aircraft. The evaluations and misses, along with associated search of these experiments were found in parameters and environmental conditions, references 2 and 3. System parameter were compiled into computer data files settings that appear to optimize small for analysis. target detection performance were determined during these evaluations and used during this experiment. 1449 The AN/APS-131 SLAR is an X-band June 8, when wave heights were 3 to 5 surveillance and oil slick detection f eet and winds were 12 to 18 knots. system capable of operation in all During the remaining 7 days of data weather conditions, day or night. The collection, significant wave heights ran AN/APS-131 is one of f ive components 0.5 to 3 feet and winds were 0 to 11 that comprise the AN/ASD-6 AIREYE knots. system. The vertically polarized radar operates at a peak power of 200 KW with Range scale and search altitude were the a beam width of 0. 4 degrees in the only selectable parameters varied during horizontal and an elevation of -4 to -45 data collection so that statistically degrees vertically measured from the significant sample sizes could be horizontal. The SLAR produces an aerial generated. Other sensor parameters were map containing an imagery swath width of optimized by the radar operators as up to 160 nautical miles centered on the needed.- aircraft's ground track. This map is produced in a near-real time video The only potentially significant human format using the AIREYE multipurpose factor investigated for its influence on display (MPD) or on a permanent copy detection performance was time on task. dry-silver f ilm. Both display methods include annotations of critical flight 5. ANALYSIS APPROACH data, aircraft position, and target position (MPD only). AN/APS-131 imagery The primary performance measure can be recorded f rom the AIREYE MPD f or currently used by SAR mission post-operation viewing and processing. coordinators to plan searches is sweep width (W). Since this radar evaluation The SLAR radar antenna is designed to was intended to support improved Coast provide wide coverage in elevation and Guard SAR mission planning, W was chosen very narrow coverage in azimuth. The as the measure of radar search SLAR antenna is mounted on the aircraft performance to be developed during data so that it radiates nearly broadside to analysis. Sweep width is a single- the aircraft centerline. The imagery number summation of a more complex produced by this system includes a blind range/detection probability zone centered on the aircraft's ground relationship. Mathematically, track. The width of this blind zone is approximately twice the aircraft's C@ altitude. Sweep Widffi (W) =f00 P(x)dx An evaluation of a similar radar, the where AN/APS-135 SLAR, was performed in 1985 using an HC-130 aircraft. The AN/APS- x lateral range or closest 135 is nearly identical to the AN/APS- point of approach to 131 with the major difference being use targets of opportunity of a 16-foot vice an 8-foot antenna. A (see figure 1), and summary of this evaluation can be found in reference 4. Parameter settings P(X) probability of detection determined to optimize small target at lateral range x detection performance were chosen based upon this evaluation. Figure 2 shows a typical P(x) curve as a function of lateral range. In Figure 2, 4. RANGE OF PARAMETERS TESTED (x) is the lateral range of detection opportunities. The range of potentially significant parameters tested for each sensor Conceptually, sweep width is the includes controllable aircraft/sensor numerical value obtained by choosing a parameters (time on task, altitude, value of lateral range less than the range scale) , environmental parameters maximum detection distance for any given expected to influence radar performance sweep so that scattered targets that may (wind speed and wave height), and target be detected beyond the limits of W are type/size. equal in number to those that may be missed within those limits. Search speed was held constant at 250 knots indicated air speed (IAS) for all 6. ANALYSIS data collection. The influence of interactions among the Although wind speeds ranged from 0 to 18 primary search parameters of interest knots and wave heights ranged from 0.5 was investigated using a sophisticated to 5 f eet, it should be noted that the binary, multivariate regression analysis most severe of these conditions occurred- technique. The detection data were on the f irst day of data collection, combined into a set with all variables 14-50 was selected where A,B,C, and D are regression variables and x is lateral Target range. 7. TEST RESULTS A total of 904 valid sensor-target interactions were reconstructed from the experiment. Data quantities are categorized by sensor, and target type for the 10-nmi range scale in Table 1. Data quantities pertaining to the SLAR reflect elimination of targets occurring within the assumed 0.8-nmi blind zone to Lateral Range either side of the aircraft. Figure 1. Definition of Lateral Range Table 1. Number of Searcher/ Target Interactions 1.0 RADAR LIFE RAFT BOAT Targets not sighted SYSTEM TARGETS TARGETS 0.5- CL Targets sighted 10-nmi 10-nmi Range Range Observer Scale Scale 0.0 Lateral range (x) AN/APS-127 Maximum FLAR 224 412 lateral range of detection AN/APS-131 Figure 2. Relationship of Targets SLAR 83 185 Detected to Targets Not Detected of interest included. This data set was 7.1 FLAR Detection of Life Rafts divided into statistically significant Figures 3 and 4 depict the raw detection subgroups based on this multivariate data and least-squares fitted lateral analysis technique. Data groups which range curves for FLAR detection of life were not significantly different were rafts when the 10-nmi range scale was combined. For each data grouping a used. Significant wave height was the function which represents the only search variable in the data set probability of detection versus lateral that was found to exert significant range was established. Sweepwidth can influence on target detection be calculated by integrating this probability. Figure 3 provides the function. Inspection of the raw data lateral range curve for significant wave for many target/sensor/range scale heights less than 2 f eet, and Figure 4 combinations indicated that the provides the lateral range curve for 2- multivariate analysis model as to 3-foot significant wave heights. constrained by the need for a monotonic functional relationship, would not Inspection of the two lateral range adequately represent the observed radar curves indicates that detection detection performance as a function of lateral range. performance against these small targets degraded markedly with even a small In order to fit a lateral range curve to increase in sea return. It should be the detection data that exhibited noted that only a small data set was unimodal response, an appropriate collected in the higher sea conditions, fitting function had to be identified. resulting in a rather uncertain lateral The function range curve fit (the 90-percent confidence limits are extremely wide). P(X) A 7.2 FLAR Detection of Boats [(x-B)'+ Al Figures 5 to 7 depict the raw .detection _@ e Nted data and least-squares fitted lateral L/ rgb 7,t-ss'-grh 1451 .90-- Ran 9 scale: 10 nmi Ait,tugdes. 500 to 4500 ft 21/26 2:- .80-- Sig. wave hts: <2 ft .90- 16/19 Rations denote detections/total W.70-- opportunities; bars depict the 90% .80-- confidence limits on each ratio 25/32'' 2.60- W70 CL M .50 260 .2 18/41 CL r- .50 9/20 .2 .40 0 'D 18/51 230-- Z.40-- 10- 13/55 :6 Ran es ale: 10nmi it, gd 01 tu x 3/9, - .30.- A .500 ft Z.20 - Sig. wave hts: <2 ft 21 5/27 *6 Rations denote detectionsttotal .10- .20- opportunities; bars depict the 909% '1/16 2 confidence limits on each ratio 0 Cd .10 0 1 2 3 4 5 6 7 8 9 10 Ol Lateral Range (nmi) 0 1 2 3 4 5 7 9 10 Figure 3. APS-127 FLAR Detection of Lateral Range (nmi) Life Rafts (10-nmi range Figure 5. APS-127 FLAR Detection of 24- scale; seas <2 feet) to 43-foot Boats (10-nmi range scale; seas <2 feet; 500-foot,altitude) .90- Range scale: I Onmi Altitudes: 2500 to 4500 ft ZI .80-- Sil, wave hts: 2 to 3 ft Range scale: 10nml R ons denote detections/total .90-- Altitudes: 2500 to 5000 ft .70-- opportunities; bars depict the 90% Sig. wave hts: <2 ft confidence limits on each ratio 2@1 .80- Rations denote detedions/total 2.60-- opportunities; bars depict the 906/6 CL confidence limits on each ratio r- M.70-- .50-- .0 .2 2.60.. 22/40 Ii .40-- 0. 79 c.50- 32153 .30- - .0 `15 .40. 13131 :5 a' .20-- 1112 1/7 T - -- K21152 CL .30 .110- 0/2 V.20-- 2 00 1 2 3 4 5 6 7 8 9 10 .10 3/22 Lateral Range (nmi) 0 - - 0 1 2 3 4 5 6 7 8 9 10 Figure 4. APS-127 FLAR Detection of Lateral Range (nmi) Life Rafts (10-nmi range Figure 6. APS-127 FLAR Detection of 24- scale; seas 2 to 3 feet) to 43-foot Boats (10-nmi range scale; seas <2 feet; range curves for FLAR detection of 24-to 2500- to 5000-foot altitudes) 43-foot boats when the 10-rimi range scale was used. Significant wave height when the 10-nmi range scale was used. and search altitude were the search None of the search variables were f ound variables found to exert significant to exert a statistically significant influence on target detection influence on life raft detection probability. probability. This result is not surprising in light of the narrow range Figures 6 to 7 provide lateral range of significant wave height and search curves for significant wave heights less altitude represented in the data set. than 2 feet and 2 to 5 feet, respectively, when 2500- to 5000-foot The APS-131 SLAR was designed to be a search altitudes were used. Comparison high-altitude, wide-area surveillance indicates that a moderate increase in radar, thus, it was not tested at the sea return resulted in approximately a 500-foot altitude. Based on a 1985 test 30-percent reduction in boat target of the APS-135 SLAR using alerted detection probability near the middle of operators and an analog film display the 10-nmi lateral range curve. Near (reference 4), the 2500- to 4000-foot the inner and outer limits of the 10-nmi search altitude was expected to yield curve, target detection probability was much better life raft detection not affected by the increase in sea performance than is indicated by Figure return. 8. During the earlier experiment, detection probabilities of 60 to 80 7.3 SLAR Detection of Life Rafts percent were achieved with 10-person 1 @4 18/51 5/27 1/7 0/2 life rafts. The use of unalerted Figure 8 depicts the raw detection data operators, high operator workload and least-squares lateral range curve (including real-time target logging), fit for SLAR detection of life rafts and the introduction of the MPD video 1452 .90. Rftangdes ale: 10 nmi .90-- Rangescale: 10nmi tu s@ A . 2500 to 5000 ft Altitudes: 2500 to 5000 ft Sig. wave hts: 2 to 5 ft 80-- Sig. wave hts: 0.5 to 5 ft z, .80-- Rations denote detectionsAotal :8 , Rations denote detections/total opportunities; bars depict the 90% '0 .70-- opportunities; bars depict the 90% :6 .70-- confidence limits on each ratio -0 confidence limits on each ratio 19/30 0 60- b..60-- 2 rL 21/44 .50 10123 11/25 .2 .50-- 20/48 is .40 40-- 13/31 T 10126 a.30-- .a tZJ4/1 8 (D 7/34 0 20-- '.20-- LM * .10 10 0 - - - - - OT 0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10 Lateral Range (nmi) Lateral Range (nnni) Figure 7. APS-127 FLAR Detection of 24- Figure 9. APS-131 SLAR Detection of 24- to 43-foot Boats (10-nmi to 43-foot Boats (10-nmi range scale; seas 2 to range scale; all data) 5 feet; 2500- to 5000-foot altitude) 10-nmi scale does not accommodate the full range of SLAR capability to detect display vice a film display are 24- to 43-foot boats. potential factors in explaining this degradation in SLAR detection performance. Additional data collection 8. REFERENCES will be required to better identify the cause of the degraded detection 1 . Richardson, W.H., Empirical Sweep performance and to investigate the Width Analysis (Air to Surface), SIO effect of using a 500-foot SLAR search Reference 68-30, Visibility Lab, Scripps altitude for small targets. Institute of Oceanography, University of California, San Diego, California (October 1968). .90. 1 Range scale: 10nmi 2:..80" Altitudes: 2500 to 4500 ft 2. Ketchen, H.G., St. Martin, J., Sig. wave hts: 0.5 to 2 ft Hover, G.L., and Mazour, T.J., .70-- Rations denote detections/total Evaluation of U.S. Coast Guard Forward@ opportunities; bars depict the 90% Looking Airborne Radars, Report No. CG- 2 .60-- confidence limits on each ratio CL D-17-84, U.S. Coast Guard Research and .50-- Development Center and Analysis & .40- Technology, Inc., March 1984. 10/33 .30-- 5/18 3. Ketchen, H.G., Nash, L., and Hover, 4) .20'* 2/15 2 - G.L., Analysis of U.S. Coast Guard HU- 0.10-- 17 25A Visual and Radar Detection F- 0. Performance, Report No. CG-D-29-83, U.S. 0 1 2 3 4 5 6 7 8' 9 10 Coast Guard Research and Development Lateral Range (nnni) Center and Analysis & Technology, Inc., Figure 8. APS-131 SLAR Detection of June 1983. Life Rafts (10-nmi range scale; all data) 4. Robe, R.Q., Edwards, N.C. Jr., Murphy, D.L., Thayer, N., Hover, G.L., Kop, M.E., Evaluation of Surface Craft 7.4 SLAR Detection of Boats and Ice Target Detection Performance _j?y the AN/APS-135 Side-Looking Airborne Figure 9 depicts the raw detection data Radar (SLAR), Report No. CG-D-30-86, and lateral range curve fits for SLAR U.S. Coast Guard Research and detection of 24- to 43-foot boats when Development Center, International Ice the 10-nmi range scale was used. As Patrol, and Analysis & Technology, Inc., with the life raft targets, none of the December 1985. search variables evaluated in the regression analysis exerted a statistically significant' influence on 20148 7@4 boat detection probability. The most notable feature of the lateral range curve in Figure 9 is that target detection probability increases steadily with lateral range, implying that the 1453 SEA BASED AEROSTATS (SBA): EFFECTIVE SURVEILLANCE FOR MARITIME INTERDICTION James B. Brewster, LT., USCGR United States Coast Guard In any war, forces cannot be effectively used 3. Reduction of consumer demand - to until the enemy is located. In the war on effect whole system drugs, narcotics traffickers cannot be captured until they are spotted. The Sea Based Aerostat The question was where to effectively use US law (SBA) is efficient, effective and inexpensive in enforcement personnel. locating and causing the apprehension of During production, if narcotics can be traffickers. The SBA can: detected and destroyed, the supply will be Determine general vessel size. reduced However, this activity is beyond the Screen probable traffickers from capability of law enforcement agencies whose innocents mariners. jurisdiction is inside the United States. This Conserve cutter fuel and search time. must be left to the State Department, Drug Extend cutter on scene endurance by UNREP. Enforcement Agency (DEA, primarily an undercover Provide deterrence value. investigative agency) and the cooperation of production countries. SBAs are providing inexpensive operational During transportation large quantities can be intelligence that is making a difference in seized if the trafficker can be detected. maritime interdiction However, law enforcement agencies must seperate the trafficker from among legitimate commerce. This is laborious and difficult but the kind of work law enforcement personnel do daily. Given Mariners transiting the Caribbean Sea lately, proper detection and interdiction assets, have often spotted something similar to the results can be'seen immediately in this phase. Goodyear blimp. As they got closer they Distribution networks are usually destroyed discovered that the blimp was tethered to a ship through undercover work. While effective, these by a long cable. Frequently, the mariner would operations require long time frames and spot a United States Coast Guard Cutter and/or a historically catch people. The narcotics seized Coast Guard helicopter operating in the vicinity here is small compared to the transportation of this blimp. What the mariner saw was a Sea phase. Based Aerostat, the latest Coast Guard tool for combatting maritime narcotics trafficking. The South Florida Task Force recommended concentrating law enforcement activities against In 1982 narcotics trafficking into the United the transportation phase as the most effective States had reached alarming proportions. means t6 intercept and seize narcotics. This Responding to the public outcry for action, the was called the Border Interdiction Strategy. President set up the South Florida Task Force The agencies involved in implementing this under the supervision of Vice President Bush. strategy are: US Customs, US Immigration and The Task Force was directed to formulate a Naturalization Service, and the Coast Guard. comprehensive strategy to combat narcotics Customs was designated as the lead agency trafficking into the United States. The ashore, the Coast Guard was designated the lead strategy developed was based upon the three agency for Maritime Interdiction and the two major phases of narcotics trafficking: agencies would jointly share responsibility for Production, Transportation, and Distribution. air interdiction. It set up a three pronged attack to combat the The primary purpose of the Border Interdiction specific phases and involved: Strategy is to detect and intercept narcotics during shipment. This is the area in which the 1. Eradication/Elimination of the trafficker is most vulnerable. The trafficker narcotics supply - aimed at production cannot hide his movements. He must put his contraband on some form of conveyance to get .2. Interception and seizure of drug them to market. Most importantly, for drugs stockpiles - aimed at transportation grown or produced outside the United States, the and distribution trafficker cross the border zone before he can- 1454 united States Government work not protected by copyright get to market. In doing this, the trafficker The Coast Guard began to investigate options risks coming under the watchful eye of law that could provide greater, sustained radar enforcement personnel. The Border Interdiction coverage of the choke points. The requirement Strategy has proven to be a very effective way was for a radar with at least 60NM range. C- to intercept and seize large quantities of 130s, E-2C Hawkeyes and airships were considered narcotics. During fiscal year 1987, the DEA and rejected because the aircraft radar recorded 616 marijuana seizures each involving platforms were not as effective on slow moving over 200 pounds. A total 1018 tons was seized. targets like ships which required more extensive 82% of those seizures occurred in the border on station time than these platforms were interdiction zone during the transportation capable of. The Sea Based Aerostat was phase. This accounted for 81% of the total considered to have the best potential and was seized. In the same time period, the DEA selected. recorded 1613 cocaine seizures each involving a minimum of one kilogram. While only 39% of the The SBA is a radar surveillancev and command and seizures occurred in the border interdiction control platform. It consists of an Aerostat zone, it accounted for 64% of the quantity (an aerodynamically shaped balloon), a Radar, a seized. Command and Control Center, and a Platform Vessel. Narcotics traffickers must first be DETECTED. After being detected they must be IDENTIFIED. The AEROSTAT is an aerodynamic helium filled Once identified as a probable trafficker,, slaw balloon. It looks like a blimp and is tethered enforcement unit must be able to INTERCEPT the to the platform vessel. It has a diaphragm, trafficker. Only in this manner can a called a ballonet, that maintains constant trafficker be arrested and his contraband pressure inside the Aerostat in response to seized. The United States has manpower and changes in atmospheric pressure and altitude. equipment to intercept traffickers once they In this manner the Aerostat maintains it shape have been detected. The crucial step has always and provides lift. The radar is suspended been in detecting the trafficker. underneath the Aerostat. The helium filled aerostat can elevate the radar to a height of As lead agency for maritime interdiction, the, 2500'. The aerostat is 109FT long, has a Coast Guard analyzed the maritime aspects of diameter of 37FT, and holds 56,000 CU FT of narcotics transportation. Our concern was how helium. to use our fleet to efficiently and economically patrol the Caribbean Sea. The solution was The RADAR is suspended under the Aerostat and is obvious. All maritime traffic entering or protected by a fabric wind screen. It is a leaving the Caribbean Sea must transit through surface search radar that can spot small wooden an island pass. This concentrates or funnels vessels (30') up to 60NM away. Range is a all maritime traffic through an area of function of the square root of the altitude, manageable size. These are referred to,as times 1.25. Thus the range of theradar is "Choke Points". The Caribbean Sea has four increased by suspending it beneath the balloon. principle choke points: The Yucatan Strait, At an altitude of 250OFT the range of the radar Windward Passage, Mona Passage and the Sombrero horizon is 63NM. The aerostat is the primary Passage (called The Anegada). This phenomenon factor allowing the radar to see as far as it is advantageous for efficiently interdicting does. contraband. The choke points reduce area that a patrolling cutter has to cover and concentrates The COMMAND AND CONTROL CENTER is the operations traffic. This reduces search time, saves funds. center for the whole system. It consists of the and greatly increases the probability of radar repeater, radio communications gear, detecting a smuggler. Aerostat monitor display and Target Information Processing System (TIPS). All radar data is Patrolling choke points has proven to be a received, viewed, and analyzed in the Command successful concept, however, the major problem and Control Center. The data is then continues to be one of detection. Radars on our transmitted to operational commanders for largest cutters can only see 15-20NM. The action. largest choke point, the Yucatan Strait, is 114NM across. The cutters did not have radars The PLATFORM VESSEL is a 192' long offshore with sufficient range to see across the pass. supply vessel. It houses the crew, functions as the aerostat mooring platform and can remain on Smugglers could use planes to determine what station for 30 days. It also carries fuel, side of the Strait the cutter was on and then water, and food to extend the on station time of direct their trafficker to the opposite side. other interdiction cutters. If the cutter transited from side to side, hoping to stumble upon a vessel, their The Coast Guard stated its requirements for an unprogrammed movements might confuse the SBA that would properly operate in the open spotters or it just might waste fuel. ocean environment, and in all temperatures and Interception was mainly dependent on luck since conditions, from 55 degrees North to 55 degrees the cutter did not have the entire picture of South latitude with the following limitations: the traffic in the pass. 1455 AEROSTAT reality, he has given himself away. This type a. Altitude: 0-2500' feet above sea level. of intelligence further aids the Commanding b. Winds: Officer in selecting and pursuing contacts that Peak sustained: 50 knots relative. have the highest probability of being a Gusts: 70 knots relative. trafficker. c. Seas 15 feet. d. Be winchable in 50 knots relative wind. Deterrence is a value that is hard to measure. e. Be winchable at a minimum speed of 200 In the case of the SBA, its deterrence value is ft/min and stoppable at any point measurable. When on station, vessel traffic between 20 and 2500 feet. cannot get around the SBA. The only way to f. Use helium as the lighter-than-air gas. avoid detection by the SBA is to move through another choke point. SBA crews continue to RADAR observe a large drop in vessel traffic within a. Have a weather detection mode. five days of arriving on station. This appears b. Detect a small wooden vessel in sea state to represent a tactical response by traffickers three from 3-60NM. to switch to a choke point not covered by an c. Have a probability of detection of 50% aerostat or to wait and ship their narcotics per sweep. after the SBA leaves. It is an intangible d. Radiate and rotate continuously 27 days result and is therefore hardest to defend. each month. Analysts/program reviewers like to see results (eg do a cost benefit analysis of the value of When tested, the SBA was able to meet the narcotics seized against dollars spent). If the foregoing criteria and proved capable of SBA provides 100% deterrence, no narcotics are electronically seeing at one time all the seized. This creates the appearance of not traffic in the widest Caribbean passage, Yucatan being effective. In reality the SBA is being Strait. extremely effective. Another case did demonstrate and prove 100% deterrence. The SBA Coast Guard interdiction operations have become was deployed to enforce a closed fishing area. more efficient as a result of deploying the SBA. Fishermen commonly fished right through the When operating with an SBA a cutter commanding closed area realizing that there was no one to officer does not have to waste time in the catch them. The night the SBA arrived on detection phase of interdiction operations. The station the crew observed fishing traffic moving SBA detects traffic in a choke point and into and out of the closed area. During the produces a picture of the traffic. This picture next day, word got to the fishing community that is electronically sent to the cutter. The the Coast Guard was on duty. That night fishing commanding officer can then use this information traffic stayed clear of the closed area. The to produce a "plan of attack" to identify the SBA crew could see on the radar, fisherman contacts. If he has a helicopter, he can direct moving up to the edge of the closed area and his helicopter to fly over each contact. If a thenreversing course away from the closed area. contact should be identified as a probable trafficker, the commanding officer can intercept Since May 1984, Coast Guard SBAs provided the vessel. This maximizes the on station intelligence to cutters that was instrumental in cutter time in active interdiction. thirty-three seizures. In addition to providing detection services to Marijuana Cocaine Haitian Migration the cutter, the SBA crew can also make some 27 Seizures 2 Seizures 8 seizures determinations about the contacts. This helps 247,865 lbs 3010 lbs 792 Haitians the Commanding Officer screen out legitimate traffic from probable smugglers. With it's 60NM In addition to its primary role as an range, the SBA can observe a contact for a while interdiction resource, the SBA can also act as a and make some assumptions about it. SBA mother ship and service other cutters with fuel, operators can determine the contact's general water, and stores. These supplies are routinely size from the radar scope. By coupling general transfered to cutters on station eliminating the size with the observed trackline some contacts need for a cutter to make a port call for fuel can be determined to be commercial traffic and and water. This allows the cutter to become not worth trying to identify. This only more efficient in using its deployment time for enhances the efficiency with which the interdiction and has proven to be extremely cost Commanding Officer can pursue contacts. effective. The Coast Guard has only one aerostat on duty at this time. Four additional The SBA crew records and stores all radar data. aerostats are being procured by the Coast Guard. This data is later analyzed. From this data, These are expected to be in operations in the Coast Guard has been able to discern February of 1989. With five SBAs, the Coast trackline patterns that indicate a high Guard will be able to more effectively seal the probability that the vessel is a trafficker. passages to the Caribbean Sea. When such a trackline pattern is observed, the SBA immediately notifies the cutter and a direct intercept is plotted. What is humorous is the trafficker thinks he is avoiding detection. In 1456 INNOVATIVE TECHNOLOGY APPLIED TO MAXIMIZE A PORT'S LIFELINE: A CASE HISTORY FOR THE SEA LANES OF THE CHESAPEAKE BAY Douglas J. Evans Capt. Richard W. Owen Paul R. Farragut Evans-Hamilton, Inc. Assoc. of Maryland Pilots Maryland Port Administration 451 Hungerford Dr. 3720 Dillon Street The World Trade Center Rockville, MD 20850 Baltimore, MD 21224 Baltimore, MD 21202 Canal connects Chesapeake and Delaware ABSTRACT Bays and is the Port of Baltimore's "short out" to and from Philadelphia, New York A consortium of public and private marine and northern Europe. This sixteen nauti- interests have worked together to maximize cal mile waterway can save a large, new the available finite navigational resour- container ship up to seven hours of ces that serve the Port of Baltimore. The "steaming" time and approximately $12,000 port needed to make maximum use of the per voyage by use of the canal. The adage Chesapeake Bay ship channel and of the that "time is money" is particularly Chesapeake and Delaware Canal as its short appropriate for present container ship out to and from the Atlantic. Thus a services where, as a result of intense number of conventional and non-convention- competition, profit margins are thin. al improvements were undertaken to in-@ Large steamship companies have responded crease the vertical clearance (air gaps) by building new ships which carry fewer in the canal and the water depths of the crew members, more cargo and are faster channels. This paper deals with the new and more energy efficient. Even with such technological improvements which involve improvements, container steamship lines remote data sensing and real time trans- are constantly "rationalizingff ports of mission and the application of this infor- call which means basically that fewer mation to the maneuvering of the larger ports are served in order to maximize use ships entering the port. The net benefit of the large investment in vessel capital of the improved technology and increased costs and to diminish unnecessary transit shipping is national in scope since it time and related operating expenses. makes the export of U.S. commodities more Modern ship costs can run $35,000 - competitive in the world wide market. $50,000 per day so making effective use of the C&D Canal becomes of paramount impor- tance to the Port of Baltimore. In terms of economic impact, some eleven jobs are I. INTRODUCTION created for each 1,000 containers and some $57 are created for each container ton In order to keep the Port of Baltimore (1). Thus ports keenly compete for this more competitive, there has been a valuable cargo. concerted effort by many private, state and federal agencies to maximize use of In the March, 1988 issue of American Ship- the available shipping channels. Both Der magazine, Dr. John Hayes of NOAA notes conventional and non-conventional methods that the tidal gauges in the federal sys- are being used today to increase the size tem also allows for more efficient use of of ships entering the port. This paper large bulk carriers. He notes that while deals with some of the non-conventional the primary use of the 225 tidal gauges i n' methods.which have proven to be very cost the U.S. system in the past was to provide effective. The economicalg navigational data for oceanographic research and. hist- and operational considerations of these orical documentation, the new use of tidal new technologies are discussed as they datapermitting increased shiploads through apply to the Chesapeake Bay and Chesapeake use of this information has both economic, and Delaware (C&D) Canal, although they safety and therefore environmental bene- are equally applicable to all major fits. The economics of this situation are harbors. such that one additional foot of draft on some large Cape Class ships can equal 3600 A. Economic Considerations tons of additional coal handled. Through March of 1988, U.S. average coal prices The Chesapeake Bay Ship Channel is nearly equaled some $42 per ton (2) so this extra 150 miles long, requiring an average 10- foot provides an ability to move $151,200 hour transit from the mouth of the bay to worth of additional coal with little addi- Baltimore. On the other hand the C&D tional cost. CH2585-8/8810000-1457 $1 @1988 IEEE B. Favigational Considerations even greater extremes of water elevations. Of the nearly 200 miles of waterways in Wind tide elevations can be predicted on a the Chesapeake Bay serving oceangoing short-term basis (a few days), but the commerce and the Port of Baltimore appro- accuracy of this forecast depends on the ximately 100 miles are dredged channels. accuracy of the weather forecast and also These channels vary in size from the fully on the availability of a well-calibrated, restricted (i.e. a canal with walls) Ches- numerical-hydrodynamical model. This fore- apeake and Delaware Canal with a 450-foot cast system has not been developed for the width through a variety of channel config- Chesapeake Bay area to date. Fortunately, urations of various widths up to the 1000- due to the slow response of the Chesapeake foot width channel at the Bay entrance. and Delaware Bays to the strong wind vel- In addition, the C&D Canal is crossed by ocity, a lag time of 6-12 hours exists five bridges of heights varying from 136.4 before major water level changes occur and to 140.9 feet above mean high water (MHW). the change in water level can be antici- In order to maximize the use of the chan- pated before it occurs and monitored as it nel system in view of the increasing size develops. Since the wind tide is not of ships serving the port, a clear need easily predicted and since it can cause arose to develop a system which would considerable change in either the air gap provide water level data and vertical or the channel depths, it is a major con- bridge clearance data so that a rational tributor to the amount of underkeel or policy could be developed regarding water vertical clearance that the pilots require draft and air draft for the two different in navigating large ships in these water approaches to the Port of Baltimore. ways. This unknown variable could reduce the effective operating draft or bridge 1. Water Depths and Bottom Changes clearance for ships entering the Port of Baltimore. These are some of the natural The channel bottom, the water levels and variations affecting the water depths of the ships draft (and elevation) are all the Chesapeake Bay and C&D Canal. Ship variable in time and space. Only the loading and maneuvering can also alter the bridge elevations are fixed in elevation. "effective" water depth of these water- The Chesapeake Bay ship channel has a ways. guaranteed depth of 42 feet below mean low water (MLW) while the C&D Canal has a C. Ship Handling Considerations guaranteed depth of 35 feet (MLW). How- ever over a period of time, sediments will The establishment in the last century of be deposited in the channels. Consequent- conventions limiting overloading of ves- ly frequent hydrographic surveys and dred- sels led to the familiar Plimsol Mark or ging are required to monitor and maintain load line in order to protect againstlife the published depths. and property loss at sea. Today's increase in vessel size has out-distanced 2. Sea Surface Variability the ability of many ports to increase channel depths to accommodate fully loaded Of considerable more variability in both ships. hence large bulk carriers and time and space is the sea surface. The container ships rarely are able to load to astronomical spring tidal height ranges their Plimsol Marks, but seek guidance to from a height of 3.4 feet at Cape Henry load cargo to some maximum safe draft for (VA) and 6.0 feet at Reedy PT (DE), at channel transit. In order to determine the eastern end of the C&D Canal, to a low maximum safe draft, several factors must of 1.3 feet in Baltimore Harbor. The be considered in addition to the above water depth and the vertical clearance navigational considerations including, (air gap) can change as much as 6 to 7 vessel trim, vessel maneuvering character- feet in 12.5 hours at the Reedy Point istics, vessel roll and pitch in exposed Bridge during the period of spring tides. seaways, and the vessel hydrodynamic Fortunately the astronomical tides are phenomenon known as ship sinkage or squat. predictable and allowances can be made beforehand for these tidal height changes. Vessels transitting channels exposed to the open sea at the mouth of the Chesa- A' less frequent, but random change in peake Bay will experience to some degree water level occurs in response to passing pitch and roll (heel) and other forms of weather systems, which can result in eit- motion induced by long period swell. her a significant increase or decrease in water level depending on the wind speed A ship moving through the water also gen- and direction. At its maximum values, the erates large pressure differentials around wind tide exceeds the astronomical.tide by the hull which interact with the channel a factor of 2 or 3 and since it is banks and bottom and with other vessels in independent of the astronomical tide, it meeting and overtaking. These pressure can superimpose on the latter to produce changes also cause a ship to sink bodily 1458 in the water and, for a deeply laden ship the corresponding and the succeeding 24- with a large block coefficient, to trim by hour periods. The differences between the the bow, effectively causing an increase elevations are the result of the non-tidal in draft and reducing underkeel clearance. components, primarily the wind tide; thus, This phenomena has variously been called the pilot is forewarned of any anomalous sinkage or squat and is influenced by hull water level condition prior to entering form, water depth, ship speed and other the canail. The user can select either the factors. Investigator's analyses of squat air gap mode to provide the clearance provide another guide in determining draft beneath the bridge or the tide mode to limits and operational requirements. provide the water depth of the canal. Quality control and reliability are main- II. TECHNOLOGICAL DEVELOPMENTS tained several ways. The TAGMOS system is calibrated at each maintenance trip. The The advent of inexpensive microprocessors gauges are checked daily by phone for and simplified telecommunications systems operation, and the measured water level is has made rapid data transmission possible compared to the predicted tide level and with the finished data products available any anomaly verified. The two gauges at within minutes of data acquisition to the each location are compared weekly. The product's users. In turn, the user has newer system at Reedy Pt. has been up- responded by increasing his willingness to graded so that the field controller PC can accept deeper draft ships by using this be programmed by phone, thus permitting information to take advantage of favorable more rapid restoration of service. water level conditions and to adjust Finally, the redundant gauges provide accordingly to unfavorable conditions. backup in the event of a gauge failure. Some of the new technological developments are briefly discussed. The Chesapeake City TAGMOS has had less than 3% downtime in 24 months of operation A. Real-time Navigational Information with the lost data being mostly due to power surging from lightning and poor Two real-time navigational systems have commercial power. The system has been recently been installed in the Chesapeake proven to be reliable and accurate. Bay and C&D Canal shipping lanes. These two systems are NOAA's TIDES ABC and EHIts A typical TAGMOS graph of the predicted TAGMOS (Tide and Air Gap Monitoring Sys- and observed water levels is shown in tem) programs which provide similar types Figure 2 for a major winter storm. The of measurements, but the results are used graph shows an increase in water elevation for different purposes. In the C&D Canal of 2.3 feet above high water (air gap the lowest of the f ive bridges in Mary- decreased to 138.3 feet) associated with land is at Chesapeake City which has a an approaching storm. This wind tide 140.9-foot clearance above MHW and in stayed above 2.0 feet for several days. On Delaware at Reedy Pt near DelawareCity another occassion, the wind tide decreased which has a 136.4-foot clearance. In the to 3.8 feet below the predicted tide level spring of 1986, the Association of Mary- when the winds shifted to the northwest. land Pilots, through a grant from the Maryland Port Administration, contracted While the air gap is of interest to large Evans- H@Lmilt-on, Inc. to install a TAGMOS vessels using the north route to Balti- on the C&D Canal at Chesapeake City (MD). more, the water depth in the ship channel In June 1988 in cooperation with the in the Chesapeake Bay is of primary con- Pilot's Association for the Bay and River cern to those ships arriving from the Delaware, the TAGMOS system was expanded south. Consequently, in 1986 the National to Reedy Pt (DE) as shown in Figure 1. The Ocean Service (NOS) of NOAA installed two systems can be interrogated by a per- real-time water level monitoring gauges at sonal computer at any of the pilot's strategic locations in the Chesapeake Bay offices in Baltimore, Philadelphia, Lewes near the dredged ship channels as shown in (DE), Chesapeake City and Cape Henry (VA)., Figure 1. These gauges are part of NOS's larger program of "real-time" or "now" The TAGMOS system operation is straight tidal gauges which have been placed in forward. The water level is measured by operation in several major shipping ports. both a Leupold & Stevens float gauge and a The NOS system is described in more detail Sensotec pressure gauge, and the data are in a paper by Parker (3). A software automatically processed and stored in a package developed by NOS called TIDES-.ABC microprocessor for 24-hours. Upon inter- (4) is used to interrogate.the tidal rogation using a standard phone connect- gauges . and to process and di .splay the ion, the water level data of either gauge measured water levels and predicted tidal are transmitted to the pilott8 PC where elevations. the past 24-hours are graphically dis- played. The PC also computes and displays the predicted tidal height elevations for 1459 The TIDES-ABC and the TAGMOS systems have and ship speed through the water. The basically the same configuration of t'ide hydrodynamic forces generated by a moving gauges, field data micro-processors vessel, will vary approximately as the modems and the user's PC type graphic/' square of the shipfs speed, hence a modest tabular displays. However, the hardware speed reduction does much to reduce these components and the software programs are effects. If, for example a ship moving at indigenous to each system. The NOS gauges 13 knots reduces speed to 9 knots, the have also been installed in Delaware Bay magnitude of the hydrodynamic forces can at Lewes, Cape May and Philadelphia and be reduced by nearly 50% . Speed reduc- are used by the Delaware pilots bringing tions, however, are also accompanied by a ships into the C&D Canal. NOS maintains a certain degradation of maneuverability, comparable quality assurance program of especially in strong cross currents or daily checks and periodic calibrations of high winds. A minimum underkeel clearance their gauges. must also be maintained for proper steer- ing and control. A balance must be struck Figure 3 shows a typical NOS display of between navigational considerations and the predicted and observed water levels in maximizing vessel loads. By knowing water Baltimore Harbor following the passage of levels prior to and during time of pas- a strong storm off the east coast of the sage, proper assessment of these elements U.S. The water levels dropped over 3.5 can ensure safe transit. feet below the predicted tide level and 3 feet below MLW. Water elevation persisted below normal values for several days. III. RESULTS OF THE TECHNOLOGY B. Ship Maneuvering/Operations A. Improved Ship Operations Real-time tidal and air gap data is used Since introduction of the new technology in two important ways for improving ship providing both predicted and actual tides operations. By examining historical data, and air gap in real-time, maximum transit long range policy regarding draft can be drafts have been increased incrementally established. By examining immediate con- from 39.5 feet to 41 and in some cases 42 ditions an informed decision can be made feet. As pilots gained experience with as to the advisability of channel transit larger and deeper draft vessels and as at deep draft on a particular voyage. If they became convinced of the realiability conditions do not warrant departure from of the tide and air-gap acquisition sys- the berth or entrance from the sea vessels tem, the "productivity" of the channel are delayed until conditions improve. system was increased while maintaining Vessel owners, agents and operators have safe clearances. agreed to this policy and operational decisions are made based upon probability 1. Ride the Tide of delays, length of delays, economic benefits of increased draft and concern The astronomical tide wave enters the for safety. Chesapeake Bay and travels northward at an average speed of about 12 knots. With the B. Less Restrictive Navigational cooperation of shipowners and operators, Requirements deep laden, inbound vessel arrivals at the Bay entrance are timed so that northbound The use of NOAA's TIDES-ABC and EHI's transits can be made with the incoming TAGMOS has considerably reduced guesswork tide. If a ship maintains an average in channel navigation regarding underkeel speed of 12 knots it can arrive in the and vertical clearances. Excessive under- Baltimore approach channels, where keel clearances due to uncertainties in shallower channel sections exist, at high water levels have been reduced as a direct water. While this practice of transiting result of the new technology. The reluct- on a rising tide is not unusual in many ance of certain vessels to use the Chesa- ports, the new real-time tidal data pro- peake and Delaware Canal due to uncertain- vides immediate information on actual ties in bridge clearance have been conditions before and during transit of overcome by demonstrating the reliability the waterway. Since the tidal level in of the system to shipping officials. New the Chesapeake and especially Baltimore ship calls into the Port of Baltimore are Harbor is very sensitive to changes in in part the result of these new technolo- wind velocity and direction, this informa- gical innovations. tion is essential if vessel movements are to be made safely. IV. SUMMARY 2. Ship Maneuvering Technological innovations have contributed to deeper draft and taller structured Vessel maneuvering characteristics depend ships entering the Port of Baltimore, largely on water depth/ship draft ratio resulting in more economical shipping 1460 costs and increased attractiveness of the partnership of marine interests working port as a major transportation center. together at the federal, state and private These gains in navigational capabilitiees industry levels using both innovative and were achieved at a relatively s m a 11 conventio-nal techniques to exploit the capital cost which was less than the full potential of these waterways while income generated from one new container maintaining safety for the *equipment and ship arrival or one deeper loaded bulk personnel and respect for the environment. ship departure. While these gains have been achieved at a nominal cost, other agencies have been working in t h i s REFERENCES partnership of marine interests to improve ship operations at the port using more 1. Martin Associates, The Economic conventional techniques. The U. S., Army Impact of the Port of Baltimore, Maryland Corps of Engineers is in the process of Port Administration and the Greater Balti- constructing the new 50-foot Baltimore more Committee, January, 1988. Channel Project as well as maintaining and 2. Energy Information Administration, improving the C & D Canal. The U. S. Coast Weekly Coal Production, Week of June 18, Guard Aids to Navigation Section has work- 1988, Washington DC. ed with the other marine interests to 3. Parker, Bruce B, Application of provide improved navigational aids and to Real-Time Data Systems to the Forecasting relocate of channel buoys for safer ship of Water Levels and Circ,ulation, Proc. operations. Oceans 86, Vol 5, pp. 1401-1406, Sept. 23- 25, 1986. In summary, the enhancement of the ship- 4. U. S. Dept of Commerce, TIDES-ABC ping channels and canal into the port of Version 3.14, NOAA, National Ocean Survey, Baltimore is being achieved through a Rockville, MD 20852, Dec. 2, 1985. Figure I LOCATION OF CHANNELS, TIDAL AND AIRGAP MEASURING STATIONS CHESAPEAKE BAY-C&D CANAL C D CANAL BALTIMORE 2 NEW JERSEY (@_J 3 4 "AIN SOUTHERN 6 APPROACH CHANNELS 5 LEWES 0 ELAWARE 6 MARYLAND VIR61NIA 7 LEGEND: RAPPAHANNO A AIRGAP SYSTEMIREEDY PT BRIDGE 2. AIRGAP SYSTEM/ CHESAPEAKE SHOAL CHANNEL CITY BRIDGE 3. BALTIMORE TIDAL GAUGE 4. TOLCHESTER TIDAL GAUGE 5. ANNAPOLIS TIDAL GAUGE YORK SPIT 6@ LEWISETTA TIDAL GAUGE CH T. WINDMILL PT TIDAL GAUGE B. CHESAPEAKE BRIDGE/ TUNNEL TIDAL GAUGE CAPE HENRY CHANNEL 1461 ASSOC14TION OF MARYLAND PILOTS OBSERUED AND PREDICTED AIR GAP - CHESAPEAXE CITY BRIDGE, MD STARTIING DAY: 136 ................. ..........*------------------------------- * ------------------------- DEC 2 1986 137 ------- ----------------- ------------ 1* ----------------------- * --------*---------- PREDICTED :: A OBSERUED 138 ....... ...... ............. *,*,-**--*--*"**-,-* ....... ................ (FLOAT) ...... *...........I...... ............................................................. 13 9 BRIDGE ELEV: 140 ........... .........., ........................................................... A 148.9 FT NNW P .4 1. ....... ........... 141 ...... ..... ... . .. .............. If ......... .......... T" ---------- 'i ----- 142 ............. ........ . . . . . . . . . . .I. . . . . . . . . . . . . . . . . . . . . . . . . 143 .... .. T ................ 144 ....... .......... .......................... 07 11 15 19 22 3 03 Pr 1'1 15 19 23 0 3 07 EASTERN STANDARD TIME Figure 2. Example of a Typical TAGMOS PC Graphic Display. NOAA 2 H_R f i nn a I 00F, an A Times are EST DatUM is MLW 0 F e t -2 -31 11-3 preaictea 0 12 is b 12 is 0 b 12 is Fj ohser-ved ...... 11/20/87 11/21/87 11/22/87 Preliminarg Data: 8574680 Baltimore, Md. Figure 3. Example of a Typical TIDES-ABC PC Graphic Display. 1462 THE MULTI-AGENCY MDU ON PORT SECURITY: A MODEL FOR OONFLICT RESOLUTION NICHOLAS A. MARZIANI SYSTEMS PLANNING AND ANALYSIS, INC., FALLS CHURCH, VIRGINIA ABSTRACT security have introduced controversy and discussion in both interagency and Channels of communication and coordination intergovernmental relations. This controversy, between marine law enforcement and defense in turn, has generated difficulties in the agencies exhibit the same potential for port security planning and coordination misunderstanding, disruption, and inefficiency process. It appears that interagency that often exist between other coordination with regard to port security and multiple-interest marine entities. related mission areas has proceeded more Specifically, the diverse military, law effectively in the last decade than during any enforcement and commercial interests in port previous time in history; yet, important areas security as represented by the U.S. Navy, of coordination remain to be addressed and Coast Guard, Maritime Administration and other associated issues resolved. Indications are agencies offer the opportunity for a fruitful that these areas will be carefully fine tuned analysis of a model of conflict resolution in such a manner that they may serve as a applicable across a wide spectrum of marine model for interagency and perhaps policy concerns. It is the object of this intergovernmental coordination in the future. paper to demonstrate that the devolution of It is the central argument of this study policy formulation and implementation to the that the past and probable future success of local level where historical factors so allow such coordination has been and will be due to is, in fact, the best national level policy the establishment of local port readiness for marine interest group conflict resolution. committees (PRC's), the creation of a milestone in interagency understanding. The Coast Guard's assumption of the chairmanship of these committees is a logical outgrowth of The United States is critically dependent its long history of activity in the mission on the security of its waterborne channels and area and its concern for optimum cooperation marine terminals. and coordination. The effectiveness of local As one element in the coastal defense PRC's demonstrates the wisdom of allowing the community since its formal establishment in largely local nature of port security 1915 as "an armed service", the Coast Guard coordination concerns to control the manner in has been a military force-in-being as well as which they are addressed. It will be argued a maritime law enforcement agency. In this that the continued successful coordination of formation, port safety and security became port readiness and security interests and specific responsibilities of the Coast Guard, resolution of difficulties, as practiced in and the Coast Guard Captain of the Port (COTP) the PRC's, is predicated upon the devolution was charged with the safe passage of military of policy formulation and implementation to and non-military waterborne vessels and the local level. cargoes. By implication, this suggests that much of Most of the port security work of the Coast what falls under the rubric of marine policy Guard in wartime has contributed significant may be most efficiently implemented by expertise and resources which had been allowing the necessarily diverse interagency developed and accumulated during peacetime. It and .intergovernmental interests in the marine is because the port security mission has been environment to provide policy directions as executed in both peace and war that certain close to the specific area of contention as ambiguities in authority exist, and with them possible. Port security as a policy issue potential coordination problems. demonstrates the correctness of this Besides joint Navy - Coast Guard port hypothesis, and by extension, its security planning and operations, the Coast applicability to other areas of marine policy. Guard must deal with other military and None of this is to deny the appropriateness of civilian agencies in the execution of its port national level strategic planning where such security mission. The mission of port security efforts are logically required to initiate is a shared one and is critical to national local coordination bodies. It is only security. These complex aspects of port maintained that once macro-level steering CH2585-8/88/0000- 1463 $1 @1988 IEEE groups have performed their important organizing its Port and Environmental Safety function, they permit local entities to bring Program (PES) in may 1980. The new program their inherent efficiencies into the policy consisted of port safety, port security, and process. environmental protection. Questions of Coordination Naval Coordination Issues Early in the 19801s, prodded by renewed Port security coordination issues between interest in national defense preparedness, a the Navy and the Coast Guard are best number of officials in Congress and in the approached by first citing the institution of Administration, began to voice concern that Maritime Defense Zones (MDZ's) in 1984. previous initiatives to insure interagency Subsequent to the signing of an MDA between coordination in port security related matters the Coast Guard, Department of Transportation, were inadequate and ineffective. and Navy officials in 1984 that formally Typical of marine agency comments at that established the Maritime Defense Zones, a time is the following: number of MOU's, ad hoc committees, working MARAD and the Coast Guard appear to have groups, and the like on all planning levels, distinct areas of responsibility regarding active and reserve, have sprung up involving the smooth operation of ports in wartime. Coast Guard, Navy, and other agencies of the The duties of the Captain of the Port are federal govermient. Before considering these specifically spelled out in statute. How- initiatives, a brief excursus on potential ever, the Port Controller function only problems between the Navy and Coast Guard is exists during wartime, thus hampering in order. peacetime coordination and planning efforts whether proceeding from the orders of the between the TOTP and the Port Controller in President himself, the National Security each locale. Council, defense secretaries, or the Chief of (The Port Controller is the Maritiwe Naval Operations, the executive branch and its Administration's (MARAD) local agent during military instrumentalities have historically mobilization.) issued directives that have shaped large In considering the manner in which the portions of the mission known as port security diverse interests in port security have today. Recently, Congress has become much more addressed the coordination issue, insight will aggressive in designing the Coast Guard's role be gained in resolving similar difficulties in in port security, although traditionally less other areas of marine policy. active in policy formulation in the mission Historical Authorities area. Legal authorities for port security consist As with many of Congress' other efforts to of statutes, executive orders, and more be a more equal partner in policy development recently, international agreements. Statutory since the early 1970's, its initiatives in authority, as first developed during World war port security have obtained both a more solid I, was originally designed to meet imiEdiate basis for mission performance and the wartime and/or emergency threats. Executive potential for serious conflicts in the event orders issued in conjunction with these of national emergency or war mobilization threats continued the linking of direct between executive branch managers and field federal involvement in port security with officers whose responsibilities must include a emergency contingencies. This state of affairs proper regard for purely statutory duties. continued until the initiation of hostilities Such conflicts, left unaddressed or in Korea and U.S. - Soviet polarization during unrecognized, can only tend to produce the Cold War prompted Congress to adopt a potentially dangerous inefficiencies in landmark piece of legislation on 9 August coordination under crisis conditions. 1950, the Magnuson Act, 50 U.S.C. 191. The Without doubt, the most comprehensive Magnuson Act amended the Espionage Act of 1917 authority conferred specifically upon the U.S. (the first statute concerned with port coast Guard by Congress for port security and security) by permitting the President to safety was the Ports and Waterways Safety Act institute measures pursuant to port protection of 1972, as amended by the Port and Tanker without having to first issue a declaration of Safety Act of 1978 and subsequently. Bearing national emergency. close similarity to the regulations The decoupling of port security from promulgated by the Magnuson Act of 1950, the national emergency declaration was an provisions of implementing regulations grant important shift in the program's evolution. It specific, statutory authority to the Coast provided the basis for some standing authority Guard Captains of the Port, under the to perform port security duties, and led to Cormnandant and District ComTanders, to the legislation of the 1970's that granted regulate broad areas of vessel and waterfront permanent statutory status to the program. facility operations. Further development in port security The Role of the Captain of the Port (COTP) missions occurred as new concerns surfaced In the execution of the port security relating to environmental protection in the authorities granted to the Coast Guard, 60's and 70's. Ultimately the Coast Guard whether statutory or in time of mobilization recognized these related concerns by (and therefore also under the naval mDZ 1464 commands), it is the Coast GL1,11-(I Captai_n of- . . . under comitatus" provisions, the Port who bears the immedtato 1)OD cannot normally perform law enforce- responsibility for the mission area. Previous ment functions. Thus, DOD has no legal to the Ports and Waterways Safety Act oE 1972, basis to task the Coast Guard to perform responsibilities of the COTP had been codified law enforcement junctions as a military/ by executive orders, as well as by World War warfare mission. II legislation and naval assignments. And, Passage of the 1972 legislation did broaden DOD planning assumes the Coast Guard will COTP authorities; the 1978 legislation went continue to perform its regulatory and law further by defining the applicable marine enforcement missions 5as tasked by statute environment to include the 200-mile fishery and Executive Order. management zone and Outer Continental Shelf DOD planning intentions and COTP operational waters. Most importantly, however, both pieces realities may not be as harmonious under of legislation fully institutionalized, in crisis conditions as intended. effect, the COTP organization by making its Even in peacetime planning, the question as activities immune from rescission by executive to how the military harbor defense mission is authority. distinguishable from the statutory port Federal regulations which specifically security mission is problematic. The two implement provisions of the Ports and highly complementary missions were discussed Waterways Safety Act stipulate that the COTP's in a NAVGARD Port Security Working Group authority includes the management of vessel report; the following excerpts of that operation and anchoring when, among other discussion are illustrative of the potential conditions, "The District Ccmmander or Captain for real disruption of coordination between of the Port has determined that such order is Navy and Coast Guard concerns: justified in the interest of safety by reason A member of an armed force is generally of weather, visibility, sea conditions, free to use deadly force against an enemy temporary port congestion, other temporary while a law enforcement officer uses force hazardous ci@cumstances, or the condition of in a graduated manner against a suspected the vessel. 11 Inasmuch as a recent NAVGARD criminal . . . The Rules of Engagement for Working Group on Port Security report said, military action differ substantially from "The greatest vulnerability, even during the Use of Force policy for law enforce- mobilization, is damage from accid . . . ment. Although there are some clear , _fnts. Historical data bear this out" , it would distinctions between Harbor Defense and appear as if the COTP's statutory authority Port Security, there will also be an over- over vessel movement might even be greater in lap . . . The contrast between what Navy. war than during peace. units/personnel can do and what Coast COTP concerns are, in effect, being divided Guard units/personnel can do suggests that between MDZ-related naval planning functions the distinction between Harbor Defense and and statutory ones. Indeed, potential Port Security should reflect the statutory conflicts in both planning and operations authority of the armed force concerned might appear to be inevitable, with or without . . . Port Security, as a Coast Guard additional resources. specialized mission, mast include all law The Convergence of Statutory and Naval Warfare enforcement related activities and may Missions - Flow or Impact? include military activities. In practice The distinct but related responsibilities these distinctiogs will not be clear when involved in the performance of statutory and action is taken. naval warfare missions clearly converge, at According to a recent statement by a the operational/port level, in the person of reserve Coast Guard office@, "Coastal defense the Coast Guard Captain of the Port. Congress is a U.S. Navy mission." One's political expressed its firm conviction that the COrP sense as to whether executive or legislative ought to possess permanent, unequivocal agencies should prevail in port security authority to protect U.S. ports and waterways planning and one's perspective on the larger in order to assure safe and expedient passage question as to what really constitutes "peace" of waterborne cargoes. Various executive or "war" (and the proper way for a democratic branch agencies and the U.S. Navy have society to design and implement its public concurred, along with the Coast Guard, that policy in a world frought with crisis and the COTP office is the proper and logical danger) affects one's perception of the issues level to exercise important naval port and discussed here. one will either perceive the coastal defense missions. While opportunities foregoing as a call to denounce the statutory for interservice coordination are thereby basis for port security, or an opportunity to enhanced in the execution of the port security further delineate the wisdom in crafting a mission between the Coast Guard and the Navy, maritime policy dealing with potential crisis it is equally apparent that, depending on management on the basis of a firm other factors, increased opportunity for deliberational foundation. Indeed, whether confusion and conflict also exist. civil marine policy and military. naval policy While, converge and flow or impact depends ultimately 1465 upon deeper constitutional issues. 1983 and comprised of representatives from Whatever one's opinion, it is undoubtably each of six agencies responsible for various true that the greater the attempts to provide aspects of emergency marine transportation, explicit channels of communication and both civilian and military, finally developed coordination, the greater the probability of a memorandum, which was signed by them in success in achieving some measure of harmony. December 1984. This Multi-Agency MOU on Port The U.S. Coast Guard and U.S. Navy have begun Readiness was endorsed by the Departments of that process at a number of interagency 'befense and Transportation in January 1985. In levels. We turn now to other agencies who have the words of the Coast Guard Commandant, "For begun talking together about port security, the first time, responsibilities of key and the implications for marine policy federal agencies involved in port throughput conflict resolution in general. of strat ' cargo are presented in a single In the event of wartime mobilization, a document." number of military and civilian federal The signatories realized that merely agencies would be responsible for various documenting the roles and authorities of the aspects of the overall task of acquiring the participating agencies, as they were then necessary vessels and port facilities for understood, could not of itself insure their emergency shipping, as well as the management efficient and effective coordination in the of those vessels and the waterways they must stress of a crisis management contingency. For ply for the safe and efficient transfer of this reason, the MOU also mandated the military and economic cargoes between the creation of a Port Readiness Steering Group United States and its diffuse overseas forces which, in turn, established a Port Readiness and allies. State and local port and law Working Group (PRWG) "to develop guidance and enforcement/emergency personnel would also manage act . lish the MOU's need to be organized so as to support the qLons to accomp objective." The Port Readiness Working Group objectives of the national authorities. These recommended that the chairmanship of local include the Maritime Administration (MARAD) Port Readiness Committees, which were mandated and its National Shipping Authority (NSA), the by the multi-agency MOU for the purpose of Military Traffic Management Command (MEC) of coordination of local concerns, be assumed by the U. S. Army, the U. S. Army Corps of the Coast Guard Captains of the Port (COTP). Engineers, the Military Sealift Command (MSC), The basic reason for the selection of the the Naval Control of Shipping Organization Coast Guard COTP as the local committee (NCSORG) , the Federal Emergency Management chairman was simply but eloquently expressed Agency (FEMA), and even the Environmental by MIW in an intra-agency memorandum: Protection Agency (EPA). A COTP is permanently present in the . All the above organizations have been specified ports and provides a standard- assigned emergency/mobilization ized focal point for committee responsibilities by Executive Order 11490 - coordination . . . Although the COTP is without the benefit of that comprehensive and the chairman, the committee is an rational analysis that would have clearly organization of equals. The chairman's delineated the unambiguous domain of each. The role isi@o coordinate communications and object here will be to discover the degree to action. which these organizations have sought -to While emphasizing overall committee membership examine their roles during mobilization and equality, XD4C never forgot the Coast Guard's negotiate with the others a clear special capabilities, developed from nearly 15 understanding of the respective years of statutory responsibility for port responsibilities of each, with the objective security and local port awareness in general, of attaining efficient and effective use of as well as decades of port presence derived waterborne assets. Coordination activities in from executive department authority. local port readiness working groups will be With the establishment of the national featured as a reasonable cause for optimism in steering and working groups, and the local any future contingency requiring port-level port readiness committees, it would seem an cooperation. ideal forum has been developed wherein The Multi-Agency Memorandum of Underst@!n@ important players in the emergency marine on Port Readiness transportation field can address, and correct, Largely as a result of signing of an MTMC - some of the serious potential problems in Coast Guard memorandum, a number of other coordination that existed at the beginning of federal agencies with port readiness concerns the present decade. There is insufficient began to approach MTMC and the Coast Guard space to discuss all the issues the various with an eye toward participating in an federal signatories of the multi-agency MOU extended formal agreement. There was a clear have faced. need for a comprehensive, multi-agency While the ultimate judgement as to the understanding that would address the broad success of the local working groups may have sweep of issues that would likely arise during to wait that unsought, occasion for which they national mobilization. As a result, an ad hoc have prepared, it seems that much of value has port readiness .working group formed in June occurred in the few years in which the 1466 multi-agency MOU has been in existence. Summary and Conclusions It now remains to determine just how well the port security mission is being addressed by MOU participants and organizational perspectives and insights might assist policy and program analysts to craft the most effective and efficient approach to the mission of port security, and similar marine policy issues. Evaluation of the execution of the port security mission must proceed in two distinct, if related, contexts. First,the militayr character of the mission must be considered along with its special relationship to bothe the U.S. Navy and Coast Guard. This military aspect, in dialection tension with statutory responsibilites as a maritime law enforcement concern,provides one focus for evaluation. Second,the largely civilian character of the multi-agency MOU on Port Readiness and its non-naval aspect in emergency maritime transportation planning provides another perspective for analysis. These different contexts for involvement inport security will render diverse assessments as to how well interagency cooordination is being effected in the mission area.In its relationship to the U.S. Navy, the Coast Guard would have to fulfill two distinct mandates are a direct function of the constitutional division of power between the executive and legislative branches. Under its executive branch mandate, the Coast Guard must assume the chartacter of a naval service in its own right during national mobilization or emergency, Under the mandate imposed upon it by Congress, the Coast Guard must also maintain its status as a maritime law enforcement agency with statutory duties during both peacetime and mobilization. Even as Congress and the President frequently argue over various points of national policy, so do those agencies and organizations which view themselves as primarily executing either Congressional intent or Presidential operating authority. Recent events associated with the Iran/Contra affair demonstrate most vividly the strong potential for misunderstanding and fractured national policy that existes when exective branch operatives and stautory considerations run counter to one another.While both day-to-day operational expertise applicable to inshore naval missions and a clear law enforcement mission, the COTP must necessarily serve two masters during wartime. It is clearly an irregular boundary between general maritime and national defence/naval policy with which the COTP must contend and there papears to be little guidance available to the COTP in this regard from Congress or high exective officials. It must be acknowledged, on the other hand, that the Coast Guard's militayr relationship with the Navy has indisputably never been better. The formalization of that relationship within the context of the Maritime Defense Zone command structure, and the excellent forum for planning functions related to the MDZ's(as well as other military and non-military concerns) provided by the formation of local Port Readiness Committees augers well for the future. Ultimately, it will most likely be the low-level but poerful influence of the relationships forged in local PRC's that will provide the best opportunity for the Navy and Coast Guard to work out its potential conflicts rather than the high-level but often ineffective attempts to set things in order conducted for Washington. Examination of the difficulties and opportunities involved in the coordination of Naval-Coast Guard concerns inport security will now focus on the PRC's. analysis of the dynamics within the PRC's will proceed upon the insights of orhanizational theroy. the existence of "good faith" relationships within the context of a military command structure are also important, but the scope of the analysis must transcend such consideration as well as recognize the dynamic and fluid nature of port security as one policy issue within a total, synergistic government process. This "fluidity" is reflected in a situation in which information flows from the external enviroment (such as threat assessment data and the demands of public and special interest groups) to the federal agency as well as between the impacted marine agencies themselves. the public policy responsibilites requiring cognizance of rapidly changing conditions, of correctly interpreting those conditions, and finally of effectively responding to them impose special demands and insights on policymakers. a degree of mangerial flecibility has been developing which is inconsistent with the centralized model. According to organizational theortist Doanld P Warwick highly interdependent agencies whose decision making processes are centralized at a relatively high level of hhierarchy usually experience information " buck up" and subsequent inefficiency in the policy process. In addition, such interagency entities are generally resistant to necessary structural changes as external conditions shift in both nature and intensity, In the fluid world of port security, policy development and mission management cannot alfford the kind of inefficienciees inherent in this centralized model. Interestingly, Warwick discusses organizational development initiatives of the past quarter century designed to remedy the problems of centralized management that closely resemble the local PRC's philosophy and approach. From sensitivity training to organizational feedback and problem-solving sessions, Warwick describes those very elenvents that have been central features of Transportation) interpretations will the PRC's modus, operandi. While it is not prevail over those ?g a DoD (Department of possible to document the -degree to which Defense) component. organizational theory was or was not Indeed, Dr. Stephens indicated that MARAD port consciously employed by the framers of the representatives would require a lot of help local PRC's, it is significant such insights from other quarters., find thT@r reflection in these local In actuality, ai@?bther approach to the role entities. of the FPC may be more useful. According to SumTarizing the first part of our Mr. Joseph Corban, Military Liaison/Special evaluation of the port security mission, it Operations Officer with the State Port may be concluded that although potentially Security Office at the Port of Wilmington, serious problems in coordination do exist North Carolina, the appropriate model for the between the Navy and the Coast Guard, the pace Federal Port Controller is the Joint Chiefs of of efforts of the two services to understand Staff, whose directorship is likened to that each other's perspectives which have been of a coordinator, whose overall purpose is to ongoing since before the beginning of the achieve maximum efficiency of operation. 1980's (and especially since the establishment one danger which the agencies involved in of local PRC's) is likely to result in a port mobilization planning need to avoid is satisfactory mobilization relationship within the possible tendency to erect inefficient the next five to ten years, certainly before hierarchical structures in attempting to the end of the century. The MOU which attempts execute their intertwined responsibilities to integrate the naval, Coast Guard, and thereby negating the central benefit of the general interagency perspective and functional local PRC's. According to Warwick, public structure has been mutually declared to be a bureaucracies, especially those which need to "living document", and it would seem that operate in a multi-agency environment, tend to there is truly an organic character to the promulgate elaborate structures as a response relationships that , have been forged in to uncerta signals in the external connection with its preparation and execution. environment. !@ By continuing to design and implement its own The potential for inefficiency inherent in resolution of executive-statutory conflicts at a hierarchical environment has been discussed. the local level, local PRC's, chaired in the The remedy would appear to be the acquisition rain by Coast Guard officers, do appear to and maintenance of realistic threat offer the best opportunity for Navy - Coast assessments, response criterion, and Guard coordination requirements to find communication channels; in other words, the suitable satisfaction. external signals need to be well defined and Turning now to the non-naval aspect of the good working relationships must be maintained emergency maritime transportation planning in the local PRC's. Although the special efforts of the multi-agency MOU signatories, impetus provided by military command an extension of the foregoing analysis would structures does not exist between the naval be appropriate. Inasmuch as the relationship and non-naval multi-agency MOU co-signatories, of the non-naval signatories is, unlike that the forum provided by the local Port Readiness of the relationship to the naval command Committees can be a source of important structure, a largely negotiated affair among interaction and adjustment. equal but separate partners, the dynamics of . In the main, the interagency coordination coordination tend to produce more open-ended efforts in the execution of the port security results. This "indeterminate" quality of mission have improved markedly since the interagency linkage compels one to adopt a beginning of the decade. The outcome of more conservative prognosis for future interagency deliberation in the early 1980's coordination efforts than was the case with exhibits the wisdom both of perceiving mutual the Navy and the Coast Guard. needs at high levels of administration and one key interagency relationship which must then of allowing the resolution of those be maintained is that which concerns the issues at the regional and local level Maritime Administration. Comneenting on the wherever possible. Marine agencies must strive role of MARAD and its designated port to utilize the opportunities this shift in representative, the Federal Port Controller, policy formulation and management style Hugh W. Stephens wrote, affords them. . . . MARAD representatives believe their Thus far, interagency efforts to engender agency will assume a predominant role in significant cooperative relationships and coordinating federal agency interaction understandings have been most exemplary. with port authorities during the surge Though not without important difficulties yet phase of mobilization. MARAD may indeed to be resolved, the dynamics and trend of the assume this during the initial stages of interagency coordination process as mobilization, but, given the primacy of implemented in local PRC's will likely render military needs in general, and MTMC's a working model by which other responsibilities in particular, it is intergovernmental processes will be evaluated questionable that DoT (Department of in the future. By developing local fora for 1468 conflict resolution, with suitable national 1978, this executive level group, whose direction, as necessary, and identifying membership includes agencies as diverse relevant historic factors in the conduct of as the National Security Council and EPA, those fora, planners in port security strives to create an environment in which throughout government have designed a constructive activity can occur without management system worthy of consideration by the dampening effects of "superagency" those groups whose concerns include meshing oversight. military and non-military interests in marine 16) Hugh W. Stephens, "U.S. Ports and National policy. Defense Strategies", Defense Trans- NOTES portation Journal, December 1986, p. 67. 1) U.S. Department of Transportation, office 17) H. Brinton Milward and Hal G. Rainey, of the Secretary of Transportation, Coast commenting upon Warwick in "Public Guard Roles and missions, Staff working Organizations", in Organizational Papers, Washington, D.C., March 1982, Theory & Public Policy, edited by p. 504. Richard H. Hall and R rt E. Quinn, 2) 33 CFR 160.111(c). (Beverly Hills, CA: Sage Publications, 3) Port Security Working Group, "Report of 1983), p. 136. the NAVGARD Board", 25 June 1986, p.6. (Emphasis added). 4) U.S. Coast Guard, Port Security Branch, This paper is the distillation of "Operating Program Plan (OPP) - a master's thesis entitled "Inter- Emergency Preparedness Goal, Goal Number agency Coordination in the Execu- 1", 10 April 1987, p. 11. tion of the Coast Guard's Port 5) Ibid. Security Mission", prepared for the 6) Port Security Working Group, "Report of faculty of the University of Dela- the NAVGARD Board", 25 June 1986, pp. 2-3. ware's Marine Policy program, College 7) LCDR Edward A. Moritz, USCG(R), "Coastal of Marine Studies (1987). Defense", U.S. Naval Proceedings, June 1985r p. 48. 8) U.S. Department of Transportation, U.S. Coast Guard, Commandant Instruction 16601.5, dated 3 July 1985, Item No. 4.a.(2); also Lieutenant Mike Ditto, USCG, "Information Paper", 1 August 1984. 9) Commandant Instruction 16601.6, dated 9 February 1987, Item No. 3.a. 10) Charles A. Vickery, Brigadier General, USAFf MD,1C Vice Commander, Department of the Army, dated 18 March 1987, Item No. 4. (Emphasis added). 11) "There is a unique and diverse range of groups and institutions in the inter- governmental public policy arena whose claims, counter-claims and concerns are factored into any policy equations or strategies being considered by govern- mental or ocean industry corporate decision-makers." Edward W. Cannon in "National Ocean Policy 'Octagon"', Sea Technology, January 1985, p. 50. 12) Donald P. Warwick, A Theory of Public Bureaucracy, (Cambridge, MA: Harvard University Press, 1975), pp. 66-67, 192. 13) Ibid, p. 167. 14) Ibid, p. 196-7. 15) It is noteworthy that other areas of marine policy have experienced or are experiencing similar shifts to local and regional management schemes, perhaps the most important example being the estab- lishment of regional fisheries councils mandated by the Magnuson Act of 1975. An equally important trend in marine affairs is the increased level of discussion and coordination that occurs in the Ocean Principals Group. Founded by the USCG in 1469 MITIGATION PLANNING FOR PORT DEVELOPMENT Daniel J. Sheehy Susan F. Vik Versar, Inc. Aquabio, Inc. 9200 Rumsey Road P.O. Box 4130 Columbia, Maryland 21045 Annapolis, Maryland 21403 ABSTRACT These include construction of artificial wetlands, rehabilitation of existing degraded wetlands, Port expansions frequently result in adverse en- establishment of submerged aquatic vegetation (SAV) vironmental impacts that require mitigation to or kelp, excavation of existing shore sites to preserve living marine resources. A systems create open water, and construction of artificial analysis approach was used to select the location reefs. The decisions associated with deciding and design of an artificial reef that was part of whether or not mitigation is required, the per- the mitigation required for redevelopment of the formance requirements to be met, and the site, Port of Wilmington, Delaware. type, and scale of mitigation are complex inter- agency problems. Applying a systems approach to Multiple Attribute Decision Making (MADM) and de- support these decisions can serve to improve the cision risk analysis (via network simulation) were probability that a decision acceptable to the among the methods used to plan cost-effective and various agencies and interest groups can be reliable mitigation. This approach allowed de- achieved in a cost-effective and timely manner. cision makers to adjust attribute weighing factors and evaluate the uncertainty inherent in the ap- APPLYING SYSTEMS ANALYSIS TOOLS plication of this new technology. The results FOR DECISION SUPPORT suggest that these methods can be used to assist in resolving the complex problems common in miti- The purpose of applying a systems approach to miti- gation planning. gation planning is to clearly establish the assump- tions, criteria, and decision factors used to make judgments. Once this is accomplished, it is some- what easier to rationally debate the problem. This PORT MITIGATION REQUIREMENTS is particularly helpful when, as with most mitiga- tion projects, decisions are being made by groups Port redevelopments and expansions frequently of individuals from different agencies that reflect result in localized adverse environmental impacts different concerns and disciplines. that may require mitigation to preserve living estuarine or marine resources. In many areas, Systems analysis is a decision making process in however, mitigation options are both limited and which competing alternatives are analyzed for the expensive due to extensive land development and purpose of selecting the one that best satisfies multiple use conflicts with other on-the-water some predetermined set of requirements, in this activities. case, the requirement would be the mitigation of adverse impacts. Before applying the systems Port development often requires dredging and land analysis process to aid decision makers, however, reclamation to increase or maintain capacities or two basic questions must be answered: What re- adapt to new cost-effective shipping and transfer quirement, and "best" in what way? How these technologies. These actions may result in the central questions are answered is vital to the loss or degradation of wetland, reef, or shallow successful application of the analytic tools. water habitat that is critical to maintaining Objectives must be carefully selected, and suit- fish or shellfish resources and the vitality of able performance measures must be clearly speci- the local ecosystem. Concern over the loss of fied before the benefits of the systems approach this habitat and its associated resources has can be fully realized. resulted in increasing efforts to maintain the carrying capacity of our coastal areas. As a There are basically six elements to a systems consequence, mitigation for unavoidable adverse analysis, they include: impacts is now frequently required as a permit condition for port development activities. 0 The selection of useful objectives. 0 A listing of all feasible alternatives. A number of options exist for creating or aug- 0 An examination of expenditures associated menting wetland or shallow water habitats with the alternatives. impacted by proposed development activities. CH2585-8/88/0000-1470 $1 @1988 IEEE 0 An effectiveness scale for each attribute degree of attainment of the stated objective by 0 Estimates of effectiveness of various competing alternatives. Considered together, the alternatives effectiveness scale and effectiveness are referred 0 Criteria that functionally relate ex- to as the "measure of effectiveness" or "objective penditures and performance function." Each of these elements is significant to the de- The final step in the analysis process is the cri- cision making process; failure to fully address terion that is simply a statement about cost and them is often the cause of controversy and delays effectiveness that determines choice. Cost and in the environmental decision making process. effectiveness are functionally related. There- Since delays frequently result in cost increases, fore, an effort to maximize effectiveness and they may adversely affect project completion and, minimize cost is relatively meaningless. In most if fixed funding has been allocated for the cases, either cost or performance must be defined project, may reduce the scope or effectiveness of by thresholds or bands while the other is maxi- the final project. mized. The first step any analysis should be the selec- One advantage of applying a systems approach to tion of useful objectives. This should result in mitigation planning is that it serves to create a clear statement of the problem. The objectives an audit trail for the decision making process. must both satisfy the basic requirements and not Since many environmental decisions may ultimately conflict with organizational goals. This may may be reviewed by all interested parties, the public, appear to be a simple task. However, when more and possibly the courts, it is definitely advan- than one set of organizational goals are involved tageous to have applied a rational and open or a new conceptual objective is contemplated, approach to the problem that can be defended this task may require considerable discussion and logically and, if necessary, legally. The systems compromise to arrive at a mutually acceptable set approach is particularly useful for complex of objectives. problems involving new approaches and more than one organization. The second step is the listing of all feasible alternatives. A feasible alternative is one that DECISION SUPPORT METHODS satisfies all objectives within the limited resources that are available. The resources that A range of methods are available to systems ana- may be limiting factors include funding, material, lysts and others interested in solving real-world people, time, and information. For environmental environmental and natural resources problems. projects, information may be limited to a degree It is, however, essential to apply the correct that results in some measure of uncertainty with methods and to be aware of the requirements and respect to the ultimate decision. It is very im- assumptions inherent in their use. As with any portant to clearly state this uncertainty so that set of decision support tools, these methods must it can be accounted for in the decision making be carefully applied in a manner consistent with process. The analyst must define the real-world the decision maker's project objectives and situation as accurately as possible in order to institutional goals. Misapplied, such analytic determine which alternatives are valid. tools can create the illusion that a "ideal solution" has been found when, in fact, it has The third step is cost analysis and estimation not. wherein the expenditures required to fully imple- ment each of the alternatives must be assessed. Project Objectives An important point to remember here is that if any of the alternatives are developmental (not Although obtaining the decision maker's objectives fully operational, tested, and proven) the costs would appear to be a straight-forward problem, this as well as performance are bound to be uncertain is not always the case. Often objectives are not as well. clearly stated, or multiple and possibly conflict- ing objectives exist. This frequently occurs in The next step in the process is the development of mitigation projects that require concurrence of an effectiveness scale that indicates the degree to numerous federal and state agencies. To allow which each alternative achieves the objectives for the widest range of feasible alternatives, specified. This is a critical problem in analysis objectives should be functionally described in a and is one where errors are frequently made. Since manner that allows clear performance measures to effectiveness is rarely measured by a single attri- be established. bute, methods to combine readings from a multi- plicity of effectiveness scales are often required. Alternatives Reliance on simple parametric cost estimates is a very risky approach and should be avoided. Developing 4 list of feasible alternatives may re- quire the advice of specialists from a number of Determining where on the effectiveness scale each fields. For mitigation planning, what is feasible alternative lies is the next step in the analysis. depends on site-specific conditions. Frequently Effectiveness is a comparative measure of the the information needed to fully examine feasible 4471 alternatives is limited which introduced an combine the readings from each scale for each element or risk into the process. Developing a alternative. The question of commensurability of listing of alternatives acceptable to all parties these scales relates to whether of not they are can be facilitated by applying group decision directly comparable. If they can be converted to making tools designed to apply social choice theory comparable scales, the question of the relative im- to group expert judgment. These range from portance of each attribute to overall performance traditional "brainstorming" and Delphi methods to becomes the next problem. This is usually addres- methods using the Borda function or Successive sed by developing weighting factors that reflect Proportional Additive Numeration (SPAN) (MacKinnon, the decision maker's perception of importance. 1966). These methods can function to incorporate a range of group input to arrive at a consensus. The problem of uncertainty about the effectiveness of any developmental system or one that has not Cost been applied under the specific circumstances being considered is inherently probabilistic. This is Cost is also dependent on local conditions and especially true for new mitigation methods applied available information. The approach that should be in new areas. In incidences where the distribution used here is the concept of "life cycle cost," of each activity can be estimated, network simula- which incorporates the total cost of selecting an tion methods can be used to obtain a frequency alternative over its life expectancy. Alternatives distribution for the overall system performance an should be compared based on an economic analysis cost. that considers acquisition, maintenance, and re- placement costs as well as all other costs which Multiple Attribute Decision Making (MADM) methods may be associated with fully implementing that that aid in addressing the problems of importance, alternative over the period of performance required commensurability, and uncertainty have been for the project. Failure to consider the life applied to resource development and siting problems cycle costs often leads to decision errors. for some time. Some of these methods such as the Simply comparing initial acquisition costs does not the Technique for Order Preference by Similarity consider variations in functional life expectancy to Ideal Solution (TOPSIS) (Yoon and Hwang, 1985) and the future requirements to replace and/or or Hierarchical Additive Weighting Method (HAWN) maintain alternatives. (Saaty, 1978) can aid in reconciling disagreement among compromise solutions. These are robust Measure of Effectiveness techniques which permit decision makers to examine the significance of individual criteria, weighting, The development of a scale to measure each alter- or uncertainty factors. native's relative effectiveness is frequently the most difficult problem in analysis. In almost Although a detailed description of the application all cases, effectiveness is not measured by a of a systems analysis approach is beyond the scope single attribute and the problem of how to com- of this paper, our objective here is to present a bine the readings from a multiplicity of attribute framework that can aid decision makers to resolve effectiveness scales is difficult particularly the complex problems associated with mitigation when some amount of uncertainty is involved in and to illustrate briefly how this approach was the individual scores. The basic problems include: used in a case study. Our experience with recom- mending the application of a new technology for a 0 Multi-attribute effectiveness controversial project suggests that this approach 0 Measurability can help focus discussion on real issues and 0 Commensurability expedite the decision making process. It is 0 Importance particularly useful in interactive group situations 0 Sensitivity where it can be used to illustrate the consequences 0 Uncertainty of attribute weighting and uncertainty variations. Alternative effectiveness is typically a function CASE STUDY: THE PORT OF WILMINGTON, DELAWARE of many characteristics; this is certainly the case in mitigation performance. The problem arises from As part of a proposed project to maintain authoriz- the fact that the underlying relationship among ed channel depths in the Wilmington Harbor Federal these operating characteristics is not always Navigation channel adjacent to the Wilmington clear. Sensitivity analysis can be used to deter- Marine Terminal in order to accommodate current and mine which attributes are not worth considering in projected waterborne commerce levels at the Por terms of their relative impact on the decision. Wilmington, the construction of a disposal area Measurability is especially a problem for nonquan- at the Wilmington Harbor South Site was proposed. tifiable attributes that cannot be put on a car- dinal scale. The development of interval measure- This plan would also leave open opportunities for ment scales for these attributes using group deci- on-site port relocation and expansion to accommo- sion making methods can help address this problem. date projected growth. One of the primary areas of environmental concern associated with the If meaningful scales can be developed for each at- preferred plan was the filling of 326 acres that tribute, the problem then becomes one of how to comprised approximately 87 acres of uplands, 12 1472 acres of wetland vegetation, 85 acres of intertidal Define Fu -donal mudflat, and 142 acres of shallow water habitat. Obj..@'ives To address these losses, a plan for proposed mitigation to compensate for the losses of aquatic habitat was developed through interagency coordina- List Alternate tion (Corps of Engineers, 1985). This conceptual Sites mitigation plan called for the use of artificial reefs in the Delaware Bay. Screen List Alternate Sites Designs The application of artificial reefs for mitigation is a relatively new approach based on the introduc- Measurement of tion of designed and prefabricated artificial reef Effectiveness technology in the U.S. (Sheehy, 1979; Aquabio, 1980), and the results of testing and evaluation Criterion of this technology in Florida (Aquabio, 1981). Properly sited and designed artificial reefs have significant advantages over scrap material and Selection of Life Cy le rock reefs for mitigation applications (Sheehy and Sites Cos Vik, 1983; Sheehy, 1984). easurement The description of this case study will focus on of Effectiveness the application of designed prefabricated reefs to illustrate the application of a systems ap- proach to resolving some of the problems associated with the use of this relatively new technology for a nonstandard application. Some of the methods briefly described in the earlier section were applied to this project by Aquabio, Inc., which provided consulting services Figure 1. Reef site and design selection process to both the State of Delaware and the U.S. Army Corps of Engineers. The data base used to support held with the interagency group established to this study was developed by Aquabio and was based guide this effort. These discussions resulted in on extensive field and laboratory studies in the development of a set of objectives that focused Japan and Taiwan on prefabricated reefs and the mitigation on provision of compensatory habitat literature and field studies throughout the U.S. for a selected group of target species. This The specific approaches and methods are presented refinement permitted the application of existing here as examples tailored to the needs of this information on ranges, food habits, shelter, project and are not necessarily suitable to all spawning, and nursery requirements for each of mitigation projects. the species to be used to screen sites for con- sideration. Once sites were selected, data on Application of a Systems Approach to the Case Study the oceanographic and substrate conditions could be used to help select appropriate designs ("design Aquabio applied some of the approaches commonly to site"). used in systems analysis to address some of the multiattribute problems associated with selecting Discussions with the Division of Fish and Wildlife artificial reef sites and designs, and conducted indicated that, in addition to the habitat compen- a feasibility study to estimate costs. The flow of sation objective, enhancement of recreational events is illustrated in Figure 1. fishing opportunities was a secondary objective to be considered in the siting and design of the The project requirements were to select suitable reefs. This reflected the concern by fishermen reef sites and then select designs appropriate that continued development on the bay would for the site conditions and target species. adversely impact fishing opportunities and success. Therefore the major emphasis of the case study application of systems approaches was focused on The objectives were thus reduced to clear function- the measurement of effectiveness which includes al statements that had the concurrence of each of establishing an effectiveness scale, determining the participating agencies. Some compromises were effectiveness of each alternative, and then made due to questions about long-term performance; ranking the alternatives in terms of their overall however, a consensus was reached that enabled the preference. project to proceed. Defining Objectives. The initial objective of The basic objectives were to provide an artificial tigation/habitat compensation for losses identi- reef that would: fied in the impact statement was not sufficient to c t M establish functional performance criteria for the 0 Have an effective life expectancy of reefs. To resolve this problem, discussions were at least 50 years. 1473 0 Increase local carrying capacities for Effectiveness Scale. For each attribute identi- selected fish and shellfish. fied as important to reef siting or design perfor- 0 Be structurally stable and non-leaching. mance, a scale and weighting system was developed 0 Provide a effective surface area for by a team of experienced reef specialists. Where epibenthic communities. actual data was judged adequate (e.g. water depth 0 Be accessable to recreational fishermen. or reef height), it was used. Where some subjec- tive judgment was required to evaluate an attribute Alternative Identification. Alternative selection (e.g., effective surface areas/unit), expert reflected a two phase process. The first step was opinion was collected by using one of the group to identify potential sites that met the conditions decision making methods discussed earlier. All suitable for the target species, would provide attribute effectiveness scales were at least in- stable conditions, and would avoid conflicts with terval. In some cases, such as stability, a number other water use activities. A hierarchical of factors were combined to produce an overall screening method was used to select sites effi- stability scale. ciently. The exclusionary screening was based on established (e.g. shipping channels) or derived Effectiveness. Table I illustrates the type of (e.g. ice scour regions) exclusion zones. Next data used in the final MADM analysis for selecting more detailed information such as fish habitat reef sites. This type of decision matrix was used requirements or preferences and access areas for to rank the final sites in order of preference recreational fishermen was considered. based on weighting and uncertainty factors derived from case studies and expert opinion. A similar Feasible design alternatives were examined to de- approach was used with the selection of final reef termine if they complied with all the basic objec- designs (Table 2). In both cases, the number of tives. Initial screening eliminated developmental alternatives was reduced by initial screening. designs without performance data to support claims of stability, life expectancy, or effectiveness. The MADM methods applied permitted us to conduct a This eliminated certain types of reefs including sensitivity analysis on the rankings by varying scrap material units and units which were original- weighting, certainty, and individual score values. ly developed for other purposes (e.g., erosion This resulted in the selection of a set of sites control or submerged well head protection). The and designs that met all the criteria and had the feasible alternatives were a series of prefabri- lowest probability of failure. cated units which have been proven and tested. Criterion. Criterion that related cost to perfor- Cost Estimation. Due to the recent introduction mance were developed and used to aid in the final of prefabricated artificial reef technology in selection reef sites and designs. The cost the U.S., and the fact that only small-scale performance relationship was extended in time; projects have been completed here, cost estimates this affected the decision by indicating that were required as part of the feasibility study. some sites or designs exhibited a differential Since a number of the units under consideration reduction in performance over time. Final recom- were originally developed in Japan, it was neces- endations were based on the outcome decision risk sary to conduct aspects of the cost analysis in analysis conducted using network simulation models Japan where hundreds of such reefs have been which reflected potential uncertainties in cost, built for commercial fisheries development. A performance, and schedule. The combination of review of the current costs of components, labor MADM and Decision Risk Analysis methods resulted and material requirements, and logistic constraints in the recommendation of reefs that were proven that might affect deployment in Delaware Bay was and tested and had a high probability of success- conducted. Cost projections were made for those fully meeting the mitigation performance objec- units which appeared to have,the greatest potential tives. for mitigation applications in Delaware Bayi CONCLUSIONS The resulting data was used in conjunction with predictions of reef unit effective life expectancy A systems approach to mitigation planning provided at the proposed sites in Delaware Bay. These a rational framework for addressing the multiple predictions were based on wave data, substrate attribute problems associated with siting and de- conditions and current velocities at the proposed signing of artificial reefs. Results suggest that sites by using drag and lift coefficients for the decision errors can be avoided by clearly defining configuration most suitable for that site. objectives in terms of functional performance re- quirements, applying life-cycle cost estimates, and All costs associated with the acquisition, deploy- properly considering the relative importance and ment, and maintenance of a reef system were credi- uncertainty of the multiple attributes used to ted to that system. Since some of the reefs evaluate the alternatives. Risk analysis, using required different labor skills and equipment for network simulation methods, is useful in assessing erection and placement, this approach assured that the long-term reliability of mitigation options. what was being compared was the best estimate for the desired overall performance. 1474 Table 1. Decision matrix for final site ranking Weight Uncertainty Site A Site B Site C Site D Site E Site F Substrate type High 10 2.1 4.5 6.7 2.3 4.3 4.1 Proximity to ramps Low 0 4.5 3.2 4.2 2.1 4.7 5.6 Hydrographic High 35 14.6 6.3 5.2 4.5 6.0 7.1 Water quality Medium 15 5.6 7.9 6.5 6.8 7.6 8.3 User conflict pot. Medium 25 7.8 5.6 2.8 5.3 5.5 3.0 Biologic factors High 60 5.4 8.9 5.3 3.5 7.9 4.4 Score 2.2-7.1 4.0-8.4 2.7-7.6 2.8-7.6 3.2-7.7 2.4-7.2 Rank 6 1 3 3 2 5 Table 2. Decision matrix for final reef design ranking Weight Uncertainty Reef 1 Reef 2 Reef 3 Reef 4 Reef 5 Reef 6 Stability High 15 4.0 5.6 7.5 3.2 8.3 5.5 Life cycle cost Medium 35 7.0 8.2 5.2 8.9 3.7 5.7 Reliability High 25 4.0 10.0 6.2 7.5 8.4 4.3 Effective surface Medium 5 3.4 7.5 6.9 5.9 5.0 3.5 Fishability Low 10 7.8 4.3 6.7 3.4 5.7 3.6 Benthic fish Medium 45 4.0 6.9 6.5 4.3 8.0 6.7 Pelagic fish Medium 20 4.1 7.6 7.2 5.2 3.5 4.1 Shellfish Low 15 3.1 6.3 5.5 4.3 6.3 5.5 Design flexibility Low 0 3.1 8.9 7.8 3.2 5.2 4.2 Score 2.5-6.4 5.4-8.6 4.6-8.5 3.3-6.9 4.2-7.8 2.8-6.7 Rank 6 1 2 4 3 5 REFERENCES Sheehy, D.J. 1984. The Application of Designed Artificial Reefs In Coastal Mitigation/ Aquabio, Inc. 1980. Artificial Reefs as a Means Compensation and Fisheries Development Pro- of Marine Mitigation in Southern California. jects. In: Proceedings of the Ninth Annual Aquabio 80-TT-720: 75 pp. Conference of the Coastal Society: 331-338. Aquabio, Inc. 1981. Japanese Artificial Reef U.S. Army Corps of Engineers. 1985. Final En- Technology. D.J. Sheehy and S.F. Vik eds. vironmental Impact Statement. Wilmington Aquabio 81-TR-604: 392 pp. Harbor Federal Navigation Project. MacKinnon, W.J. 1969. Development of the SPAN Yoon, K. and C.L. Hwang. 1985. Manufacturing Technique for Making Decision, in Human Groups. Plant Location Analysis by Multiple Attri- Am. Beh. Sci. 9(9):9-19. bute Decision Making. Int. J. Prod. Res. 23(2):345-359. Satty, T.L. 1977. A Scaling Method for Priorities in Hierarchical Structure. J. Math. Psych. 15(3):234-281. Sheehy, D.J. 1979. Fisheries Development: Japan. Water Spectrum (Winter 1979-1980):1-9. Sheehy, D.J. and S.F. Vik. 1983. Recent Advances in Artificial Reef Technology. In. Pro- ceedings of Oceans '83 Vol. 11: 957-960. 1475 COAST GUARD RECREATIONAL BOATING PRODUCT ASSURANCE PROGRAM Alston Colihan Recreational Boating Product Assurance Branch U.S. Coast Guard Washington, D.C. ABSTRACT other things, the FBSA authorized the Coast Guard to establish national construction and performance standards applicable to The purpose of the Coast Guard Recreational Boating manufacturers of boats and associated equipment. Product Assurance program is to improve the safety of boats and associated equipment sold to the recreational Each regulation the Coast Guard has issued under the FBSA has boating public and thereby reducing injuries and deaths related to safety. That is, each was developed in response to a on the nations waterways. The program stems from demonstrated need. Section 5 of the FBSA specifically directs laws enacted as early as 1940. Initially these laws pre- that the regulations be stated in terms of performance. This scribed limited standards for motorboat equipment. gives the manufacturer the widest discretion in designing and Today's Recreational Boating Product Assurance pro- improving products, which still meet the required level of gram involves approximately 3000 companies manu- safety. Therefore, it is unlikely that the Coast Guard will ever facturing and importing more than 2.5 million units an- issue boating safety standards relating to the fit and size of com- nually. ponents. If accident statistics indicate aproblern in such an area, then a performance standard may be issued. By "performance" we mean specifying the load which a fitting must withstand, as In the 1960s, two laws affected noncommercial boats on the opposed to specifying the dimensions of that fitting orthe thick- navigable waters of the United States. The Motorboat Act of ness of the material from which the fitting is constructed. 1940 prescribed limited standards for motorboating equipment. The Federal Boating Act of 1958 was primarily a boat number- In establishing a need to develop and prescribe regulations and ingstatute. Both of these Acts applied to the boat operator only. standards, the Coast Guard: In many cases manufacturers would build boats in accordance (1) considers the need for and extent to which requirements with such requirements as backfire flame control or ventilation; imposed on manufacturers will contribute to boating safety; however, for manufacturers whochosenottodo so,only theper- (2) considers relevant available boating safety standards, sta- son who operated the boat could be cited for a violation of the tistics and data, including public and private research and law. development, testing and evaluation; (3) considers whether any proposed manufacturer require- Several considerations lent a sense of urgency to early passage ment is reasonable and appropriate for the particular type of boat of anew bill promoting boating safety. For one, annual fatalities or associated equipment for which it is prescribed; and in boating accidents were averaging fourper day andmany more (4) consults with the National Boating Safety Advisory Coun- boaters were involved in serious accidents which resulted in in- cil (NBSAC). juries and severe property damage. During the months of May through September of 1968 alone, the casualties were 195 per-, NATIONAL BOATING SAFETY ADVISORY COUNCIL sons killed in May, 169 killed in June, 199 killed in July, 158 killed in August, and 130 killed in September. nose were a lot After the Coast Guard is reasonably certain that a regulation is of casualties and only involved fatalities. needed, the matter is brought to the attention of the National Boating Safety Advisory Council, established under the author- The Coast Guard's present involvement with recreational boat- ity of section 33 of the FBSA, for their advice and concurrence. ing product safety began with the enactment of the Federal Boat The Council was established to further insure that all boat safety Safety Act (hereafter "The Act" or "FBSA") on August 11, regulations are needed and are reasonable, considering the 197 1, which became a part of the United States Code. The big- hazard which is being addressed by the regulation. NBSAC gest problem with previous boating legislation in the United consists of 21 members divided equally among State Boating States was that each requirement was part of the law passed by officials, representatives from the boating industry and repre- Congress. This meant that it took an act of Congress to change, sentatives of the boating public. The public members are usu- improve or add new requirements as the need arose. Among ally members of boat owners associations or have some other '1476 United States Government work not protected by copyright wide interest in boating, from the point of view of the consumer. effort by the Coast Guard to reduce the number of drowmings as- NBSAC normally meets twice per year. The members are re- sociated with recreational boating. Drowning� which ac- imbursed for their travel expenses. counted for over 90 percent of all boAng deaths, were most often the result of the sinking or capsizing of boats less than 20 Often NBSAC wishes to maintain. close liaison.with the Coast feet in length. Manufacturers of boats subject to the Safe Load- Guard during the development of a particular regulation, In ing and/or Safe Powering Standards must affix a U.S. Coast theseinstances, NBSAC appoints severalmembers to a subcom- Guard Maximum Capacities label. The Safe Loading Standard mittee or panel. These members attend frequent meetings and applies to monohull boats less than 20 feet in length, except sail- briefings with the Coast Guard project officers who are working boats, canoes, kayaks and inflatables. The Safe Powering Stan- on the regulation. They are invited to attenddemonstrations and dard applies to monobull boats less than 20 feet in length that boat testing sessions. In this way they are completely acquainted use one or more outboard motors for propulsion, except sail- with the solution of the problem as it is developed. boats, canoes, kayaks and inflatables. EARLY REGULATIONS AND STANDARDS The U.S. Coast Guard Maximum Capacities label must display Maximum Horsepower (if an outboard boat), the Maximum The first regulations and standards applicable to manufacturers Persons Capacity in pounds and the number of people and the were adopted from industry standards published by the Ameri- Maximum Weight Capacity (persons, motor and gear for out- can Boat and Yacht Council (ABYQ and the Boating Industry boards) (persons and gear for inboards, inboard-outdrives and Associations. The Coast Guard issued final rules in the Federal boats without mechanical propulsion). The manufacturer tests Register on August 4, 1972 covering Defect Notification and a boat and calculates the capacities to be displayed on the label Boat Identification for all boats and safety standards covering in accordance with the specific requirements in each of the Safe Loading, Safe Powering, and Flotation for boats less than standards. 20 feet in length. Later, standards covering Level Flotation, Electrical Systems, Fuel Systems, Ventilation and Start-In- SAFE LOADING STANDARD Gear Protection were published. The Safe Loading Standard and requirement for the display of Following are brief explanations of the present Coast Guard loading information provide a guide to selecting appropriate regulations and safety standards applicable to manufacturers weight capacities. A careless or inexperienced boat operator and importers of recreational boats: might crowd too many people or too much portable gear into a boat, thereby reducing its stability and freeboard and making it HULL IDENTIFICATION NUMBER more susceptible to capsizing and swamping. Capsizings fre- quently occur because of the simple fact that a small boat hull The Hull Identification Number regulations apply to all recrea- can physically accommodate a much greater load than the boat tional boats manufactured in or imported into the United States. can safely carry. Each Hull Identification Number (HIN) is a 12-character label which must be placed on the stem or transom of every boat hull SAFE POWERING STANDARD above the water line. The first three characters are a manufac- turer identification code assigned by the Coast Guard. The next The regulation governing horsepower capacity for outboard five characters in the HIN are the manufacturer's serial number. boats was taken directly from the voluntary industry standard The last four characters in the HIN indicate the date of certifi- used in the United States for a number of years. In this regula- cation and the model year. The HIN is used by the Coast Guard tion, horsepower capacity is a function of boat length and tran- and various Federal and State agencies to identify boats subject som width. The boat length in feet is multiplied by the transom to recall, to identify boats being registered, to identify boats width in feet to obtain a factor. The horsepower capacity al- involved in accidents and to trace lost, stolen or abandoned lowed for various factors is listed in a table in the regulations. boats. Generally, boats equipped with remote steering and a transom height of at least 20 inches qualify for larger horsepower DISPLAY OF CAPACITY INFORMATION capacities than boats which lack remote steering or a smaller transom height. A further reduction in horsepower is required The Display of Capacity Information Standard, Safe Loading if a boat has a flat bottom with a sharp square chine. Standard and Safe Powering Standard are part of a continuing 1477 A recent amendment to the Safe Powering Standard allows Faced with this situation, the Coast Guard began an analysis of manufacturers of certain small outboard runabouts less than 13 reports and statistical data pertaining to fatalities that resulted feet in length to use optional performance test methods as an al- from boats capsizing, flooding, sinking, or striking floating ob- ternative to the existing calculation method. Boats capable of jects. The results of the analysis showed that a large percentage less than 35 miles per hour must successfully complete a Quick of the fatalities investigated during the analysis period would Turn Test in which the boat operator performs a 180 degree of not have occurred had the boats involved in the accidents been the wheel at Itop speed. Boats capable of 35 miles per hour or equipped with "level flotation." In response to the findings of more must successfully complete both a Quick Turn Test and a the analysis, the level flotation regulations removed the major Safe Maneuvering Speed Test in which the boat is steered difficulties experienced with the basic flotation standard. through a test course with progressively sharper turns, depend- ing upon the boat's top speed. When the performance tests are LEVEL FLOTATION used, a boat's maximum horsepower capacity may not exceed 40 horsepower. Both performance tests were adopted from the The "level flotation" regulations are designed to greatly in- ABYC powering standard. crease both thevisibility and the survivability of individuals fol- lowing a boating accident by making the boat a safety platform FLOTATION STANDARD from which rescue can be effected, When the level flotation regulations were established, it was estimated they would save The Flotation Standard applies to monobull boats less than 20 approximately 210 lives annually. feet in length, except sailboats, canoes, kayaks and inflatables. The amount of flotation required and the attitude in which the The level flotation regulations apply to monohull outboard boat must float when swamped, depends upon whether the boat powered boats - the most common type of boat used on is powered by an inboard engine, an outboard motorgreaterthan America's waterways. These boats must be constructed with two horsepower, or an outbooard motor of two horsepower or enough properly placed flotation to cause them to float in an less or manual propulsion. Boats less than 20 feet in length approximately level attitude when submerged with occupants powered by inboards and stemdrives must be equipped with on board. This "level flotation" standard is substantially more "basic flotation." Outboard powered boats and manually pro- stringent than the "basic flotation" standard. pelled boats must be equipped with "level flotation." In general, outboard powered boats equipped with level flo- tation have these advantages: BASIC FLOTATION (1) They will float in a level attitude when swamped, thus pro- viding a platform on which the occupants can remain safe until The original "basic" flotation standard, effective in August rescued. 1973 (1974 model year); reflected the existing industry practice (2) Occupants can remain in the swamped boat with approxi- for emergency flotation. From 1968-70 there were 218 deaths, mately 50 percent of their bodies out of the water, thus lessen- 51 injuries, and over $1 million in property damage resulting ing the danger of hypothennia (subnormal body temperature re- from boats sinking. At the time, it was felt that a boat which sulting from exposure to cold water). would float when flooded gave its occupants a greater chance of (3) The powerhead of the motor, in most cases, will remain survival because it would be available to grasp until assistance out of the water when the boat is swamped. If the motor can be arrived. For search and rescue purposes, it was considered far started with a pull starter, the boat and its occupants may be able easier to find a boat than individuals in the water. to return safely to shore under the boat's own power. The basic flotation regulations require a boat to contain an FUEL AND ELECTRICAL SYSTEM STANDARDS amount of flotation that is determined from a complicated for- mula based on the estimated weights and densities of its differ- The Coast Guard Fuel and Electrical System Standards apply to ent components. Experience had shown however, that the use gasoline powered inboard boats and boats with permanently of basic flotation in outboard powered boats was not satisfac- installed gasoline powered auxiliary engines. The standards do tory. Outboard powered boats equipped with basic flotation had not apply to outboard engines, portable engines or portable fuel a tendency to capsize and float bow high when swamped. This tanks. Accident statistics on fires and explosions have shown made it difficult for survivors to hang onto and remain with the that inboard powered boats are 43 times more likely to have a boat and almost impossible for them to right the boat and climb fire or explosion than outboard powered boats. Even when out- back on board. board powered boats have a problem it is usually a fire, result- 1478 ing in fewer injuries, while the inboard boat typically explodes. report showed that a natural ventilation system is effective only when a boat is moving fast enough to force air through the ven- Both of these regulations were exposed to the broadest and most tilation ducts, or when the wind is at sufficient velocity and di- comprehensive discussion and analysis by a subcommittee of rection to blow through the ducts. The report concluded that the National Boating Safety Advisory Council. A Fuel and there is. a definite need for some type of forced ventilation on Electrical Panel was established by the Council to review the boats to remove flammable and explosive gases before the en- regulations. The panel included representatives from independ- gine is started. Boating accident statistics indicate that most ent testing laboratories, marine surveyors, boat builders, engine fires and explosions occur while the boat is dead in the water builders, and State Boating Law Administrators. The technical after fueling. A forced ventilation system assures that engine aspects of the proposed regulations were scrutinized by the compartments are properly ventilated before the engine is panel in meetings open to the public. The regulations were started. based on the existing American Boat and Yacht Council (ABYC) standards and on the standards of the National Fire Pro- Among the major features of the Coast Guard Ventilation Stan- tection Association (NFPA). dard are: (1) The term, "open to the atmosphere," is defined for the pur- Some of the major requirements in the Fuel System Standard poses of identif@ing compartments which do not require venti- are: lation: A compartment is open to the atmosphere if there is at (1) The use of terneplate (lead-tin alloy coated steel) for per- least 15 square inches of open area directly exposed to the at- manently installed fuel tanks is prohibited due to its susceptibil- mosphere for each cubic ibot of net compartment volume. ity to deterioration from exposure to salt water. (2) A powered ventilation system must be @ installed in any (2) Fuel tanks must be physically tested for shock and vibra- compartment (not open to the atmosphere) which contains aper- tion. manently installed gasoline engine with a cranking motor. (3) Tanks larger than 200 gallons must also be subjected to a (3) Minimum air flow capacities for blowers are specified. slosh test - designed to detect weaknesses in the fastening of (4) A natural ventilation system is req, ired for each compart- baffle plates inside the tank. ment which contains a permanently installed gasoline engine, (4) All components in the fuel systern must be capable of with- contains a permanently installed fuel tank and an electrical standing exposure to the flames of gasoline for 2-1/2 minutes. component that is not ignition protected, contains a fuel tank that (5) Fuel pumps of the diaphragm type, commonly used on vents into that compartment or contains a non-metallic fuel tank automotive and marine engines, are not permitted to be vented exceeding acceptable permeability. into the engine room. (5) Intake ducts must be above the normal accumulation of (6) The principle innovation in the Electrical System Standard bilge water and low enough in a compartment to draw fuel is a requirement that all components used within the engine vapors which are heavier than air. room must be externally ignition-proofed. This means that (6) A label must be displayed advising the boat operator to run these electrical components must be able to operate in an explo- the blower for at least 4 minutes prior to starting a gasoline en- sive atmosphere without igniting the vapors. gine. (7) The electrical wiring specifications are a combination of (7) Minimum internal cross-sectional areas for supply open- those published by the Society of Automotive Engineers and ings and supply ducts and exhaust openings and exhaust ducts Underwriter's Laboratories. in natural ventilation systems are specified. VENTILATION STANDARD START-IN-GEAR PROTECTION STANDARD The Coast Guard Ventilation Standard applies to boats having The Coast Guard Start-In-Gear Protection standard applies to gasoline engines for propulsion or electrical generation. Exist- outboard engines which are capable of developing a static thrust ing ventilation regulations required gasoline powered motor- of 115 pounds or more. A review of Coast Guard accident sta- boats or motor vessels to have "at least 2 ventilation ducts fitted tistics had found that when an outboard motor is started in gear with cowls or their equivalent, for the efficient removal of ex- it frequently produces a sudden movement of the boat which plosive or flammable gases from the bilges of every engine and causes its occupants to either fall inside the boat or be thrown fuel tank compartment." The effectiveness of this type of natu- overboard. The standard requires these motors and any controls ral ventilation method was questioned in a 1975 report based for such motors to be equipped with a device to prevent such upon information from two independent research projects. The accidents. 1479 MANUFACTURER CERTIFICATION can create the injury. It must occur substantially without wam- ing, which precludes normal wear on aproduct. Forexarnplethe mere fact that an engine fails to function is not necessarily a An important feature ofregulations issued under the FBSA is the safety related defect. VVMe the failure of an engine could allow regulation whichrequires Manufacturer Certification that aboat the boat to drift into danger, the assumption is that the operator complies with applicable regulations without the previous would normally have options available to cure such a condition, approval of the Coast Guard or the Government. Behind this including an anchorto stop the boat from drifting. If engine fail- policy is a presumption that reputable manufacturers will not ure were considered to be a safety related defect, then an empty risk the imposition of the penalties provided under the Act, nor fuel tank could also be considered a safety related defect. How- do they wish to expose themselves to civil liabilities arising from ever, if an engine fails because of a broken connecting rod or their failure to supply a legally complying product to the retail piston which breaks through the side of the engine block, it is customer. conceivable that flying engine parts could cause an injury. DEFECT NOTIFICATION Therefore, such a failure might be considered a safety related defect. The most revolutionary provision of the FBSA are the Defect ENFORCEMENT Notification Regulations which require manufacturers to notify all boat owners of any safety related defect in their boats, and The Coast Guardhas fourprincipal methods for detecting manu- repair these safety related defects at the manufacturer's "sole facturer or importer violations of the boating safety regulations: cost and expense". This requirement to pay for repair of defects (1) Coast Guard Marine Safety Office and Marine Inspection was the first such requirement in U. S. product and consumer Office personnel perform FactM Visit where they examine law. The law establishing safety regulations for automobiles the boats, with the dual purpose of detecting violations and edu- simply requires that the manufacturer notify the owner of the cating the boat manufacturer in the requirements of the regula- need for repair. However, in practice, most automobile manu- tion. These field personnel also recommend boats to be tested facturers have voluntarily made those repairs free of charge to in the laboratory to detect violations which cannot be deter- the customer. The FBSA definition of "manufacturer" includes minedvisually, such as the maximurnweight capacity displayed importers of foreign made boats and engines. It is impossible on a boat. for the Coast Guard to enforce regulations on boat builders who (2) The Coast Guard also purchases boats on the open market are outside the United States. Therefore the responsibility for for Conipliance Testing by independent testing laboratories meeting the requirements of the FB SA falls directly on the per- under contract to the Coast Guard. This method is particularly son who imports these marine products. appropriate for determining compliance with the flotation regu- Manufacturers of boats and engines are required to notify all lations and the safe loading regulations, wherein the boat must purchasers of defects which create a substantial risk of personal be immersed in a test tank of water. Boats are selected for test- injury to the public, as well as notifying customers of defects ing on the basis of suspected irregularities. This means a high arising from failure to comply with an applicable regulation. In percentage of the boats tested fail to pass one or more of the eithercase, the manufacturer is requiredtopay forthe cost of re- applicable regulations. In 1987 149 boats were purchased and pairing the defector correcting a failure to comply. Atthatlevel, tested for compliance with the Safe Loading and Flotation Stan- there is practically no difference between costs to repair a safety dards and 74 failed one or more of the tests. related defect or a failure to comply with a regulation. The sig- (3) Manufacturers and boat builders who discover that they nificant difference is that failure to comply with a regulation have failed to comply with boating safety regulations, usually might expose the manufacturer to the assessment of a penalty mport their violations voluntarily to the Coast Guard in order to under the terms of the FBSA. There is no specific penalty for reduce any possible penalty and in order to place themselves in a safety defect other than the cost of repairing that defect. abetterposition to defend civil liability law suitsthatmight arise from the violation. The FBSA states that the manufacturer should provide notifica (4) Manufacturers also Mort their coml2etitors'violations of tion for any defect, "which creates a substantial risk of person- boating safety regulations to the Coast Guard. Manufacturers alinjury to the public". The current policy of the Coast Guard who are posting legal weight limits on their products are natu- is to consider each case on its own merits. In general, a safety rally quite perturbed when they discover that competitors are related defect must be the primary cause of a condition which unlawfully posting higher weight capacities on their own com- parable products. 1480 PENALTIES The FBSA provides for penalties of as much as $2,000 for each violation of a regulation on each boat produced up to a maxi- mum of $100,000 for a series of related violations. These penalties would apply to the manufacturer in the United States or to the U.S. importer of a foreign made article. The Coast Guard normally considers the relative seriousness of a violation of a boating safety regulation when determining the amount of any penalty. Conscientious efforts by themanufac- turer to correct the defect quickly and to ensure that all purchas- ers of products containing the defect are notified are factors tending to mitigate the penalty assessed. We have found that most manufacturers are anxious to correct defects in the shortest possible time. They have cooperated with the Coast Guard to the extent that most penalties have been for a relatively small percentage of the maximum penalty which could have been assessed. The Coast Guard considers that its primary duty is to achieve safe boats. The penalty provisions of the FBSA are viewed only as an aid for achieving safe boats and associated equipment. We believe that excessively harsh and unsympa- thetic application of the penalties would only encourage manu- facturers to attempt to hide their problems instead@ of openly cooperating with the Coast Guard in the correction of those problems. 1481 DETERMINING HORSEPOWER LIMITS.ON RECREATIONAL BOATS Carol L. Hervey U.S. Coast Guard R&D Center Avery Point Groton, Connecticut 06340-6096 ABSTRACT and Yacht Council (ABYC). This organization was founded in 1954 and The Coast Guard regulates maximum charged with developing and implementing allowable horsepower on recreational safety standards relating to outboard boats up to 20 feet in length. recreational boating. A Hull Division Improved methods of determining the was formed which is responsible for a proper amount of horsepower have been variety of topics, one of which is proposed, including a new "dry land" powering. The Hull Performance formula based on lateral acceleration. Committee, formed in the early 170s, A test program was conducted which picked up on the powering subject begun evaluated the power-related performance in the 150s where the basis of the of 25 outboard boats of various designs. present length/width formula was From this information-it was determined developed and called the "Ralph Brown that lateral acceleration was not Curve" after its originator. Industry sufficient to set horsepower levels and standards making organizations, the that a fairly complex formula would be Outboard Boating Club (OBC) and the needed to adequately assess the Boating Industry Association (BIA), powering- related behavior of the wide worked with and amplified on the basic variety of hull designs currently formula., Compliance with the standards available to the consumer. set by the ABYC has been mandatory f or those manufacturers participating in the 1. INTRODUCTION organizations. However, participation in the organizations themselves was The issue of limiting horsepower on voluntary. Thus, prior to 1970, recreational outboard boats has been compliance with any standard was discussed and debated since the Coast strictly a matter of choice. Guard began regulating power in the early 1970s. opinions range from During the 1950's and 601s, when the questioning the need for such a data for the ABYC standard was regulation to a desire for more strict developed, there were primarily family, controls on the current limits. Since fishing and skiing boats. There was a the adoption of the Federal Boating significant performance gap before the Safety Act of 1971, in particular 33 CFR category of "race boat" was reached. It 183.53 (D) - Horsepower Capacity, the is important to note that the current Coast Guard has been mandated standard was developed at that time as a responsibility in this area. The guide for a family boat with a beginning current regulation, which simply relates or novice driver. Times have changed. maximum horsepower to principal boat With the advances in fiberglass dimensions, applies only to outboard construction making possible a wide motorboats 20 feet in length or less. variety of new hull designs, new It is the purpose of this paper to outboard designs and features, in describe the history behind the current particular power trim and high regulation, discuss some of the performance propellers, and power alternatives that have been availability quadrupled, the entire investigated, and describe the current range of recreational boat performance thinking on the subject of horsepower has broadened greatly. Step V bottom limits. hulls have been developed which offer a soft but fast ride but bring with them 2. BACKGROUND the instability known as "chine walking". The bass boat craze brings The Coast Guard regulationhas its roots high speeds in low-sided boats which in early work done by the American Boat often run in obstacle- infested waters. 1482 United States Government work not protected by copyright Gone is the gap between a family and a small boats 12 feet in length or less, high performance boat. remotely steered and designed for a maximum of 2 persons, cannot carry 3. CURRENT COAST GUARD REGULATIONS sufficient power to get them up on plane. Their power limit is now When the Federal Boating Safety Act was obtained by successful completion of the enacted, the Coast Guard borrowed the Avoidance Course. ABYC formula, H-26. The method of calculation is shown on Figure 1. In spite of the potential misuse, a "dry Basically, a factor is 'determined by land" formula is desirable for reasons multiplying the boat length by transom that it can be simple to use and width. This f actor is then used to requires no subjective interpretation. determine maximum horsepower according In 1982 the Coast Guard contracted with to the table in Figure 1. Not long Ideamatics, Inc., to try to develop a after this standard was adopted, more complete standard which would however, it was realized that it did not still be easy for the manufacturers apply to a variety of situations. Some to use. The Ideamatics effort involved manufacturers of high performance boats a relating powering, speed and turning began adding lightweight "beaks" to radius to a quantifiable variable, their bows for added length and wing- lateral acceleration. Their premise was like extensions to achieve the highest based on the statistical information possible formula power rating. which indicates that falls overboard are a major cause of recreational boating In order to research other methods of fatalities and that lateral acceleration determining horsepower limits, the Coast is a measure of the force needed to Guard contracted with Wyle Laboratories cause this. The Ideamatics formula uses to investigate the present standard, length, beam, weight and deadrise to test course methods and any alternate predict maximum attainable speed and proposals. The primary test course used turning radius. These predictions are for this study was the one developed a combined to calculate maximum lateral number of years previously and adopted acceleration. by the ABYC, the so-called Avoidance Course. It is shown on Figure 2. The formulas were developed by using a Wyle's final report, issued in March number of different boat hulls with 1974, presented results which showed 53* various engines, and measuring the of all outboard boats are underpowered physical parameters and the speeds, when horsepower capacity is restricted turning radii and lateral accelerations to formula horsepower as compared to obtained on the test course. Regression results from avoidance course tests. analysis was then used to f it the data This implies that the other half of the and develop the formulas. It was hoped outboard population is overpowered and that ultimately this formula could serve thus the current formula gives only an as a replacement for the current average value of horsepower capacity. standard. What, then, could offer a more accurate The Coast Guard Research and Development method for determining horsepower Center became involved in this work in limits? This question is not easily 1984. They were tasked with getting a answered. For one thing, there is no number of boat hulls representative of simple criteria to use to determine how various categories and testing them to much horsepower is appropriate. The see if the Ideamatics formula could be issue involves maintaining adequate used as a replacement. Field tests were maneuverability at high speeds without conducted at the Mercury Marine facility serious jarring of those in the boat. in Oshkosh, Wisconsin, during May of The current formula does not relate 1985 and 1986. In all 25 boats were power, speed and maneuverability. tested in the categories of jon, bass, Moreover a formula, such as the one semi-vee, deep-vee, trihull, tunnel currently in use, makes it possible for hull, cathedral hull and "sport" boats. a boat builder to put a boat on the market that has not undergone adequate, The boats were measured for beam, or in the worst case, any, water testing length, deadrise, transom height and yet is in full compliance with the Coast width, and for some hulls, complete Guard regulation. offsets were taken. Afterwards, the boats were run through the avoidance The current Coast Guard procedure does test course with their current maximum allow for exemptions to be granted in horsepower, a turn radius test was cases where a manufacturer feels his conducted to determine maximum radius, boat is being underpowered. For and lateral acceleration measurements instance, according to the present were taken. For some of the hulls, roll regulation, the "sport boats", which are angle, pitch, longitudinal and vertical ..1483 acceleration, heading, and rudder angle present time that further work towards were also measured. These additional an improved "dry land" formula will be quantities were measured so that if the done. However, if the interest in this Ideamatics formula did not prove valid, area is renewed, it is likely that an additional information would be investigation into the use of a test available to try another formulation. course approach will be pursued. An additional test used in this series is called the Quick Turn Test. This is used by the ABYC to determine maximum maneuvering speed for any remotely steered boat; inboard, outboard, and any length including those over 20 feet. As used by the ABYC, whenever a boat cannot remain under satisfactory control when put in a one-half steering wheel rotation snap turn at wide open throttle, a placard is af f ixed to the dashboard which states at what speed that turn can be negotiated. This was developed as an alternative to the Avoidance Course, since no marker buoys need be used and there is less driver influence. We used it to compare lateral accelerations from the Avoidance Course. After comparing the two years field data it became apparent that the Ideamatics formula was not sufficiently accurate in predicting the relationship between engine power and lateral acceleration. It was decided to suspend any further validation at that point. As mentioned previously, additional data was taken in the course of the test program to use in the event that the Ideamatics formula did not prove valid. Stevens Institute of Technology was tasked to examine the results of the field tests and to investigate the possibility of another formula based on easily measured boat parameters. They did make an preliminary formulation, however, much work remains to be done to determine its range of applicability and accuracy as a predictive tool. 4. FUTURE PLANS As this work was being finished, several things occurred which have delayed further development in the area of maximum horsepower determination for recreational powerboats. First is the recognition that an improved dry land formula which is both simple to use and applicable to the wide range of hull types available may not be achievable. At the same time the Coast Guard has experienced the budget cuts and reprioritization that so many government agencies have. Also, the area of overpowering is quite subjective and other areas of compliance, such as adequate flotation and the new alcohol and boating laws have refocused boating safety issues. It is unlikely at the 1484 TITLE 33 - Navigation and Navigable Waters CHAPTER I - COAST GUARD, DEPARTMENT OF TRANSPORTATION (CGD 73-250) Part 183 - BOATS AND ASSOCIATED EQUIPMENT 183.51 - Safe Powering 183.51 Applicability This subpart applies to monohull boats less than 20 feet in length, except sailboats, canoes, kayaks, and inflatable boats, that are designed or intended to use one or more outboard motors for propulsion. 183.53 Horsepower capacity The maximum horsepower marked on a boat must not exceed the horsepower capacity determined as follows: (a) Compute a factor by multiplying the boat length in feet by the maximum transom width in feet excluding handles, and other similar fittings, attachments, and extensions. If the boat does not have a full transom, the transom width is the broadest beam in the aftermost quarter length of the boat. (b) Locate horsepower capacity corresponding to the factor in Table 183.53. (c) If the horsepower capacity in Table 183.53 is not an even multiple of 5, it may be raised to the next even multiple of 5. (d) For f lat bottom hard chine boats with a factor of 52 or less, the horsepower capacity must be reduced by one horsepower capacity increment in Table 183.53. Table 183.53 Outboard Boat Horsepower Capacity COMPUTE: FACTOR = BOAT LENGTH X TRANSOM WIDTH If factor (nearest integer) is 0-35 36-39 40-42 43-45 46-52 Horsepower capacity is ........ 3 5 7@ 10 15 Note* For flat bottom hard chine boats, with factor of 52 or less, reduce one @apacity increment (e.g., 5 to 3) No remote steering, or less than 20" transom If factor is Remove steering For flat bottom For other boats over 52.5 and at least 2011 hard chine boats and the boat. transom height has ......... Horsepower capacity is (raise to (2X Factor)-90 (0.5X Factor)-15 (0.8X Factor)-25 nearest multiple of 5) ........ Figure 1. Current Standard 1485 Speed (MPH) Floating 35-37.5 Marker Buoys 37.5-42.5 42.5-50 20' ---------- - ------- 25 - ------------------ ------- @5 -75' 120' 20'--75' Figure 2. Avoidance Course 35 to 50 MPH 1486 REGULATION OF PASSENGER CARRYING SUBMERSIBLES L C D R Steve Johnson L C D R John Veentjer Commandant (G-MTH-4) U.S. Coast Guard 2100 Second St. SW Washington, D C 20593-0001 ABSTRACT dive to the submersible to provide entertainment such as feeding fish. At the conclusion of the dive the This paper is a history of commercial passenger passengers disembark to the ferry which has picked up carrying submersibles and the United States Coast Guard another load of passengers at the shore base. Ten or regulatory efforts to ensure their safe operation. A more dives may be completed during a day's operations description of a typical operation and vessel is given and an evening or night dive may also be offered if the with emphasis on safety features. Safety guidelines and submersible is equipped with suitable lighting. procedures for certification are described. The submersible pressure hulls are cylinders with hemispherical heads. Two hatches are located on top of the cylinder at the ends. overall hull lengths range from 50 to 65 feet. Acrylic viewports are fitted along both INTRODUCTION sides of the cylinder portion of the hull and large acrylic hemispherical viewports are fitted at one or both ends. There is a growing interest in operating The forward viewport is for the pilot and the after submersibles in the tourist sightseeing business. This viewport, if fitted, is for passenger viewing. An interest now centers in the Caribbean with operations in exostructure is affixed to the hull which includes the the Cayman Islands, Barbados, and the U.S. Virgin embarkation deck, skids, protective guards and fairing. Islands. Operations have also begun recently in Hawaii Attachments to the exostructure include hard and soft and the Mariana islands. The leader in this business has ballast tanks, trim tanks, propulsion units, jettison been Subaquatics of Vancouver, BC with two 28 weights, lights, liferails, and emergency locating passenger submersibles, ATLANTIS I AND la, in the devices. Power for propulsion is provided by Cayman Islands and Barbados and two 47 passenger rechargeable batteries. The interior is fitted with a sub m ersibles, A T L A N TIS III A N D IV, in St Thomas, USVI deck and seating arranged for passenger viewing out of and Hawaii respectively. ATLANTIS III was built in the viewports. The batteries are usually located in a Vancouver, BC under C oast G uard inspection. compartment below the deck. Ladders to the hatches ATLANTIS IV was built under Coast Guard inspection in are usually stowed up out of the way to make room for Washington State. Two other passenger sub m ersibles additional passenger seating. Pilot controls are located have co m e under U.S. jurisdiction recently. M A RIE A I, a in the forw ard he m ispherical head. Toile t facilities m ay 46 passenger submersible built in Finland, is now be installed depending on voyage length. operating in Saipan under Panamanian flag and the U.S. This new business is on the verge of expanding from control verification program for foreign passenger several operations to possibly dozens throughout the vessels operating within U.S jurisdiction. L 0 OKIN G United States. These tourist attractions have been very GLASS, built in Scotland, has recently been U.S. Coast popular and are potentially very lucrative for their Guard certificated for carriage of 48 passengers in St. owners. The Coast Guard, as the maritime safety Thomas. MA RIEA I and LOOKING GLASS were designed regulatory agency, has the responsibility for ensuring the by Fluid Energy Ltd. of Scotland. safety of the passengers and crew of these vessels. The basic operations of these attractions are similar. They are located in a strong tourist market with HIST 0 RY interesting dive sites. A typical excursion begins with a brief ferry ride from a convenient shoreside base. Considerable research and development has been At the dive site the passengers embark the conducted relative to the safe design, construction, and submersible for a 45 minute dive. Each passenger sits operation of small manned submersibles. These efforts with access. to a viewport on one side of the vessel. The involved the participation of the Navy, the Coast Guard, dive follows a predetermined standard course at 50 to the submersible industry, the classification societies and 200 foot depths. Commentary on the underwater other interest groups such as the Marine Technology attractions observable during the dive is given by the Society (M TS). pilot or attendant over a PA system. During the dive a The Navy has always been involved with the safety support vessel patrols the div e site m aintaining of military sub m ersibles. The loss of T H R E S H E R in 1963 communications and possibly visual contact with the intensified the Navy efforts and resulted in special sub m ersible. A scuba diver fro m the support vessel m ay safety programs. With the advent of the deep research 1487 United States Government work not protected by copyright vehicles such as TRIESTE and ALVIN, the Navy took introduced in both the House and Senate to obtain action to establish the safety of Naval personnel when authorization for regulation of non-military sub m ersibles embarked on manned noncombatant submersibles. The regardless of size, service, or nu m ber of passengers. military certification requirements were applied as In preparation for the legislative proposal, the Coast appropriate. Other requirements were applied as Guard developed a regulatory concept which would dictated by the submersible's specigh ed design and use. prescribe a reasonable standard of safety but not hinder ShCNAV Instruction 9290-1 was issued in 1966 development of submersible technology. The intention establishing the Navy safety policy and assigning was to produce performance standards for submersible responsibility for action to CNO. This responsibility vessel systems which would lead to low-level initial covered: the competency of operations; the planning of regulations for im ple m entation. Highly specific operations; and the materials adequacy of the regulations would not be promulgated until undersea sub m ersibles. technology passed beyond the developmental stage. The Naval Ship Systems Command (now NAVSEA) Indications at the time were that a slow and deliberate was delegated responsibility for development and effort would be possible. Research and development promulgation of the evaluation and certification efforts were initiated to determine some of the more criteria. The Navy published "Material Certification basic requirements for submersible regulations. I-i-gi on Procedures and C rite ria Manual for Manned, with the industry and standards organizations was Non-combatant Submersibles" in 1968. This manual was established to develop policy, codes, and guidelines for considered both broad and strict. The Navy certification submersibles. Submersibles requiring certification under program was open to innovation and new technology by existing laws and regulations were to be handled on an acceptance of new features which were demonstrated individual basis making extensive use of provisions for safe. determining "equivalent safety." The Navy's certification program is now published in Not long after the Coast Guard Underwater Safety N A V 14 A T P-9290, "Syste m s C ertffication Procedures and Project Team was formed, the priority of the project Criteria Manual for Deep Submergence Systems". This became questionable. The Bills in Congress had not certification program applies to all noncombatant passed and the anticipated demand for submersibles had sub m ersibles which are e m ployed carrying Navy not materialized. Although there had been steady personnel. The Navy still has a number of submersibles involve m ent in the advancement of sub m er-sible to which the certification program applies including technology, the industry itself had experienced the same ALVIN, the fimt research submersible certificated by decline as the marine industry in general. The use of the Navy. submersibles for other than industrial, experimental or In the mid-1960s, ABS was approached by industry research applications did not until recently appear representatives regarding the practicality of preparing probable. This being the case, Coast Guard efforts standards for the design and construction of commercial tapered off with the termination of the Underwater sub m ersibles. Because of the limited experience in Safety Project (USP) in the late 1970's. operation of co m m ercial sub m ersibles and because of the Non-mili-tary submersibles have now been used for unknown variety of their uses at that time, A BS felt that several decades; however, until 1984 had not been used specific Rules would be premature. Accordingly, a in any service for which the existing Coast Guard Guide for Classification of Manned Submersibles was inspection statutes and regulations would apply. prepared and published in 1968. The Guide was for use In 1984 Subaquatics first contacted the the Coast during the period of' accumulation of actual operating Guard about their plans to build and operate passenger experience. The Rules were written with the small carrying sub m ersibles. Their first two vessels, research, industrial, and experimental submersibles in ATLANTIS I and 11@ built from 1984 to 1986, were mind. designed to carry 28 passengers and two crewmen on In the decade to follow, there was a dynamic growth short voyages to a depth of 150 feet. These two vessels in the submersible industry. Builders, operators, and are now operating in Barbados and the Cayman Islands. ABS (the havy and Coast Guard as well) gained extensive With the success of these vessels, Subaquatics then experience relative to small submersibles, primarily approached the Coast Guard with a proposal for a 47 those for research, industrial, and experimental service. passenger submersible to be operated within U.S Consequently, ABS was able to publish "Rules for jurisdiction in the U.S. Virgin Islands. The Coast Guard Underwater Systems and Vehicles" in 1979. The Coast worked closely with Subaquatics to define the basic Guard participated in the development of these Rules. safety requirements and determine acceptable design M TS conducted three studies and published three and operational features necessary to ensure the safety sets of guidelines for submersible safety during the of passengers and crew at a level equivalent to that of a period from 1968 to 1979. The guidelines address design, small passenger vessel of similar capacity. This operations, personnel, maintenance, procedures, and submersible, ATLANTIS III, was certificated in July 1987 equipment. The Coast Guard participated in the studies and has been operating successf)iUy in St. Thomas, USVL and provided funding. The Coast Guard established the Underwater Safety COAST GUARD CERTIFICATION Project (USP) on July 15, 1968 in reaction to what appeared at the time to be a strong near-term need for The Coast Guard has taken a systems approach to Coast Guard regulation of underwater vehicles and certificating submersibles, evaluating the combined stations. At the time, there were about 50 civilian design, operation, dive site, and operator qualifications submersibles in existence in the U.S. In a decade of fro m the conceptual stages through the initial submersible operations, there had been only three major operations. After initial certification, the Coast Guard accidents resulting in the loss of one life. Nevertheless, monitors the operations and periodically inspects the to ensure at least a minimum standard of safety was vessel. Each operation is evaluated individually because maintained, the Coast Guard proposed legislation in bills of the developing nature of the business and absence of 1488 spe citic regulations for sub m ersibles. be coordinated with and accepted by the U.S. Coast The regulations that are the basis for passenger Guard if the vessel will be operatin in U.S. waters I sub m ersible certification are found in 46 CFR (SOLAS Chapter I, Regulation 4(b)@ Requests for Subchapter T-Small Passenger Vessels (less than 100 consideration of exemptions can be m ads to gross tons). Since these regulations were developed Commandant (G-MVI). If the Flag Administration issues primarily with surface craft in mind, many of the its own certificate of inspection, the Coast Guard needs requirements cannot be applied to or may otherwise be to know the standards and requirements that were inappropriate for submersibles. The Coast Guard's applied before accepting the vessel. Early identification approach to the novel design and unique operational and contact with the Flag Administration is necessary. hazards of passenger submersibles is to attain a level of The degree of Coast Guard inspection will depend on its safety at least equal to that required for a surface craft country of registry and the degree of inspection of similar size and service and otherwise minimize any performed by the Flag Ad ministration. inherent hazards of underwater operation. The failure of any single piece of equipment should not result in a D 0 C U MENTATIO N "fatal" or unsafe situation. This is ac cc m pLished in part through a co m bination of design require me nts, U.S. Customs Service has determined that vessels operational restrictions, and redundant systems. operating to carry passengers or cargo between U.S. Passenger submersibles over 100 gross tons would be ports or from one U.S. port to a point within the subject to 46 C F R Subchapter H-Passenger Vessels which territorial waters and back to the same port are is a more detailed and comprehensive set of regulations considered to be operating in coastwise trade. Vessels than Subchapter T; however, like Subchapter T it does engaging in coastwise trade must be documented under not address submersibles. No submersibles are likely to the laws of the United States with a coastwise license. be built to Subchapter H in the near future. Such vessels, in addition to being owned and operated by A s m all sub m ersible, carrying six or less passengers qualified citizens, must also have been built in the for hire, is an "uninspected vessel" as defined by 46 United States, and must never have been sold to a United States Code (USC) 2101 (42). Such vessels are non-citizen of the United States, registered under the not subject to Coast Guard inspection regulations but are laws of any foreign country, or (if greater than 500 gross subject to regulations for uninspected vessels, 46 CFR tons) rebuilt abroad. The foreign built and foreign flag Subchapter C-Uninspected Vessels, or recreational boats, passenger submerEibles currently in operation are an 33 CFR Subchapter S-Boating Safety. These regulations exception to these requirements because the U.S. Virgin do not address submersibles and do not provide for Islands and the Mariana Islands are not covered by the periodic inspection after the vessel has been coastwise laws. manufactured. Due to the potentially hazardous nature of operating a submersible vessel, the local Coast Guard COAST GUARD PLAN REVIEW Captain of the Port having jurisdiction over the operating area may make special operational Normally small vessel plan review for Coast Guard requirements under authority of 33 USC Chapter 25 - approval is done by the cognizant Officer in Charge, Ports and Waterways Safety Program, and 33 CFR part Marine Inspection (OCMI) who has the responsibility of 160 - Ports and Waterways Safety - General- Concerns inspectLng vessels within his jurisdiction. However, of the Captain of the Port include special operational because passenger submersibles are novel, detailed plan restrictions, navigational safety, and port security review is done by the Marine Safety Center (G-MSC), considerations. located at 2100 Second St. S.W., Washington, D.C. 20593-0100. Plan submittal procedures should, however, FOREIGN FLAG VESSELS be discussed with the OCMI as well. Detailed plan revie w will not be performed before jurisdiction Foreign flag vessels may operate in certain U.S. (evidence that Coast Guard inspection is required) has waters as discussed in the Documentation section below. been established and a contract has been awarded for A foreign flag submersible inspected and certificated by construction. Concept review prior to this is usually a country having inspection laws and standards similar to appropriate and can be accomplished by submitting a those of the U.S. will be inspected (control verification) proposal addressing the design) and operation of the to ensure that the condition of the vessel is as stated in vessel to Commandant (G- MTH , 2100 Second St. S.W., its certificates. A foreign country is considered to have Washington, D.C. 20593-0001. A written operations inspection laws and standards similar to those of the U.S. manual/safety plan giving details of the normal and if that country is a party to the International Convention emergency operational procedures is submitted to the for Safety of Life at Sea (SO L AS). Coast Guard early in the planning stage for conceptual Generally, the vessels curTent certificates of review and approval. inspection will be accepted as evidence of lawful Plans required to be reviewed by the Coast Guard inspection; however, the Coast Guard must be aware of are listed in 46 C F R Subehapter T, Section 177-05, and and in agreement with the design and inspection include the following: midship section, outboard and standards applied. If the Flag Administration issues inboard profiles, arrangement of decks, machinery and SOLAS Passenger Vessel Certificates to the vessel, as electrical installations, fuel tanks, piping systems, bull we would prefer, certain exemptions will be necessary. penetrations and shell connections. The SOLAS Regulations, like U.S. regulations, were In addition to the plans required in Subchapter T, written primarily with surface craft in mind. Thus, the following are required for detailed plan review of there are many regulations which cannot be applied to or sub m ersibles: may otherwise be inappropriate for sub mersibles. Other - Pressure hull construction, strength measures to attain an equivalent level of safety will be calculations, and tolerances; including windows, nec essary. Exe m ptions to the S 0 L A S R egulations m ust hatches, penetrations and attachments. 1489 Life Support system s/equip m ent. Guard's acceptance of the design of all the submersibles (1) Air filtration mentioned above. The ABS Rules provide a good base (2) Sensors and monitoring equipment for evaluating a submersible vessel. The vessel's (3) Oxygen system certification is also dependent on its operational plan (4) Emergency breathing and dive site which are evaluated along with the Fire Protection syste m s/equip m ent. submersible vessel. A submersible design certificated Bilge system. for operation in one location might not be acceptable in Ballast system plans and calculations. another location or for a different operation such as a W eight, stability and buoyancy data and longer dive because of' the close association of the vessel calculations, including emergency surfacing design and its operation. capability. A surface and submerged inclining will be required for which a proposed procedure Hull Structure is to be submitted. Intact and damage freeboard and limits of The hull structure should be designed to and will be heel/trLm calculations. reviewed to the standards of the American Society of Control Systems plans. Mechanical Engineers (ASME) Code for Pressure Vessels Quality control and testing procedures. for Human Occupancy (PVHO) or the standards of the M aterial identification. American Bureau of Shipping (ABS) Rules for Building and Classing Underwater Systems and Vehicles. Considering the unique operating parameters of a Proposals to use other established standards for submersible, there are certain areas where the Coast underwater vehicles may be submitted for consideration Guard will focus particular attention during plan review by Commandant (G-MTH). Plan submission must identify and inspection. So me examples are: the standard utilized and the pertinent design parameters such as maximum operating depth and - Escape and rescue from a submarine will be structural safety factor. difficult and hazardous, so the means for Since submersibles are more susceptible to rapid and returning to the surface in both the normal and catastrophic flooding than surface vessels, additional emergency modes will be reviewed. The requirements to those in 46 CFR Subchapter T may be submersible must be capable of returning to the necessary. The following are examples of features which surface in the event of failure of any system may need to be incorporated depending on the design and except the main pressure hull. Redundant whether or not an equivalent level of safety is achieved systems can make this possible. by the operational procedures addressed in the - Access to lifesaving equipment and means of operations manual/safety plan. exiting the submersible once on the surface may also be difficult. The Coast Guard - A protective cover over the exterior and requires that adequate freeboard and stability interior of windows to protect them from be available on the surface to permit the safe impact and chemical cleaning agents. debarkation of passengers under the worst - Windows, designed to the PVHO standards or expected surface conditions in the designated equivalent, are to be reasonably protected in operating area. case of underwater collision (during normal or - Review of calculations which demonstrate emergency operations) with natural formations adequate subdivision and stability to permit the or man-made obstructions in the operating area. vessel to surface in a timely manner, while - Quality control during construction (including maintaining an upright attitude after receiving 100 percent weld X-ray). damage which penetrates the hull or damage to any ballast tanks. Subdivision and Stability - Provisions for personnel protection, during the time it takes to surface and evacuate, from Subdivision is not a require m ent for vessels less than hazards such as smoke or toxic vapors in the 100 gross tons, less than 65 feet L 0 A measured over the event of a tire or from flooding in the case of weather deck, carrying no more than 49 passengers, and hull da m age. operating on other than ocean waters. Subdivision has - Differences between submerged and surface not been required for any of the submersibles mentioned operations including restricted visibility, above, but may be required for any submersible designed undetectability by surface craft, and to exceed any of the constraints listed here or for m odifications to the required e m ergency isolation of hull penetrations not adequately designed to instructions must be considered. prevent flooding. - Vital systems, such as those necessary for the Stability must be adequate for both surface and vessel to surface, to deploy lifesaving submerged conditions. While surfaced, the stability must equipment, or to debark personnel, must be be sufficient to safely withstand heeling moments caused shown to have an acceptable level of reliability, by wind, waves, and passenger movement. Hatch a manual override control, or redundancy. coamings must be high enough to prevent flooding by - The oxygen system will be subject to special waves, and the embarkation deck must be high enough to review due to the increased potential for fire. be reasonably dry for embarking passengers. The passenger submersibles built to date have been quite D ESlG N stable on the surface. The significant stability hazard while submerged is C o m pliance with the ABS rules and ABS the large trimming moment that can be generated by classification have been important factors in the Coast passenger movement. A submerged stability test is 1490 required to determine the vertical distance between the Lifesupport Systems centers of gravity and buoyancy in order to evaluate the sub m ersible's stability in loaded conditions. ABS lifesupport standards are considered by the Coast Guard to be adequate for passenger submersible Passenger A cco m m odations certification. Unless they carry more than 150 passengers, there Electrical Installation are no specific rules for structural fire protection. However, 46 CFR Subehapter T, Section 177-10-5 states The electrical system must meet the requirements that the "...general construction of the vessel shall be of 46 CPR Subchapter T and Subchapter J as such as to minimize fire hazards insofar as reasonable appropriate. The following are some of the specific and practicable." This means that a designer or builder items which must be considered and addressed in the should minimize the amount of flame, smoke, and toxic design; how ever, this list is not all inclusive. gas producing materials within the submersible to reduce Either marine shipboard cables orlow smoke cables the risk to passengers and crew. The fit and installation would be acceptable. The low smoke cables provide a of the materials must be to the satisfaction of the degree of safety surpassing that provided by the standard cognizant Officer in Charge, Marine Inspection. shipboard cables and their use is encouraged. Separation (e.g., a partition) to prevent passengers An emergency power source must be provided. It from interferring with operator performance is required must be independent of the main power source and must between the passenger and pi-lot sections. be sized to supply all connected loads for at least twice the duration of the voyage. At the end of this period the Lifesaving Equip m ent voltage of the battery must not be less than eighty eight per cent of the nominal battery voltage. Emergency Alternatives to on board stowage of required lighting must be automatically activated upon loss of the primary lifesaving equipment (e.g., liferafts) must be main power source. Emergency lights may have a means addressed. Liferaft stowage on board the support vessel to turn them off to conserve power, but such means has been accepted. should not be accessible to the passengers. Other Coast Guard approved life preservers will be emergency loads must be capable of being connected to required for everyone on board, including children. the emergency power source. There must be a means to Coast Guard approved inflatable life preservers are indicate a low charge on the e m ergency batte zies. available and may be used on submersibles. Life All batteries must be protected from salt water preservers must be stowed on board. contamination, yet remain accessible for regular servicing. Hydrogen gas generation must be monitored. Fire Protection Equipment Battery charging procedures and equipment must be addressed from the standpoint of hydrogen generation Required fire protection equipment is limited to and sources of ignition. portable non-toxic: extinguishers approved by the U.S. Coast Guard. Fixed Halon 1301 systems are typically INSPE C TIO N installed and are in addition to the required portable extinguishers. Fixed systems must be approved. Fire The inspection and certification requirements are detection is not required, but a detection system is addressed in 46 CPR Part 176. Application for usually installed. If instal-led a detection system must Inspection (CG-3752) should be submitted to the Officer have an approval from a nationally recognized testing in Charge, M arine Inspection (0 C MI) having laboratory as well as plan approval by the Marine Safety responsibility for the location at which the vessel is C enter. A fire pump and firemain system are not being built. Normally, inspection for certification is required. conducted only on U.S. flag vessels. Generally, inspection will begin when requested after there are Machinery Installation approved plans by which to conduct the inspection. The schedule of inspection is to be worked out with the 0 C ML Marine engineering systems are subject to the For a U.S. flag vessel built overseas, the ultimate requirements of 46 CPR Subchapter T and Subchapter F aim will be a degree of inspection during construction as appropriate. The following are certain specific items equivalent to that which would be attained if the vessel which must be considered and addressed in the design; were built in the U.S. The Coast Guard maintains only a however, this list is not all inclusive. limited number of inspectors overseas, hence overseas Pressure vessels (other than the main hull) which are inspections m ay be cc m plicated by delays in permanently installed on board the vessel are subject to communications and inspector availability. Additionally, 46 CPR 54.01-5. Portable pressure vessels for use on overseas inspections of U.S. vessels are subject to board, but serviced/refIlled ashore and remaining in rei m burse m ent. commerce, are considered ship's stores and must be Department of Transportation (DOT) approved (46 CPR MANNING 147-04). DOT cylinders which are modified in any way are no longer considered DOT approved and must be Since the Coast Guard does not presently have shown to be equivalent to the appropriate pressure vessel regulations which specifically address licensing and standards. manning of passenger carrying submersibles, a proposal Hydraulic and pneumatic systems must meet 46 must be submitted to Commandant (G-MVP), via the C FR 58-30. cognizant 0 C ML This proposal must address the level of Air conditioning syste ms must m set 46 C FR 58.20. training and qualification and number of personnel A system for de watering must be addressed. necessary for safe operation of the vessel. Each licensed 1491 operator will also be required to meet additional was proposed by the industry but never formalized. requirements and receive a special license endorsement Navy assistance when available wiD be utilized through for passenger submersibles. The number of unlicensed an agreement between the Coast Guard and the Navy. deckhands required, if. any, will be determined after The Coast Guard underwater SAR policy and guidance is evaluation of the manning proposal and operational published in the National Search and Rescue Manual. safety plan. A TLA N TIS IH is required to be manned by a licensed operator, a pilot trainee and a deckhand. FUTURE PLANS 0 PER A TIO NS The Coast Guard is presently working with the Transportation Syste m s Center (D epart m ent of Since a submersible is usually not as self sufficient Transportation), Cambridge, Massachusetts, in the as a surface vessel, special consideration will be given to development of a sytems safety analysis of a typical the overall syste m of operations, support and passenger submersible operation. This analysis will maintenance in view of the environment in which the provide the bads for a more systematic safety submersible will operate. Depending on the location of evaluation of future proposals. the analysis should be operation, certain conditions such as strong tidal co m pleted in early 1989. currents or hazards presented by other vessels or The Marine Board of the National Research Council underwater obstructions may be cause for certain has shown interest in this area and in cooperation with operating restrictions, additional design features, or the Coast Guard may conduct a study of the safety of possibly prohibiting operations altogether. passenger sub m ersible systems during 1989. Submersibles will be restricted to operations in The results of these two studies will be used by the waters no deeper than the designed and certified Coast Guard in the development of appropriate guidance maximum operatLng depth of the vessel. or regulations for passenger submersible builders, Certification will be for a particular operating operators and owners. If this emerging business location, and operations in other locations will not be continues to grow it will be necessary to have a uniform permitted without specific Coast Guard approval. All set of safety requirements in order to effectively aspects of the intended operations must be discussed guarantee the safety of the crews and the hundreds of with the cognizant OCKI and Captain of the Port thousands of passengers that will take the thrilling (COTP) for the proposed operating area early in the excursions these vessels can offer. planning stage. The written operational safety plan would be useful as a basis for that discussion. The operations manual/safety plan should address at least the following aspects of the operation. In the planning stages and during review, a draft m anual/plan is acceptable. Ultimately, the m anual/plan m ust be finalized and adhered to. Changes may be made subject to approval of the 0 0 MI and C 0 T P. - Support craft functions and capabilities including: sub m ersible shado wing, diver availability and capabilities, and emergency lift capability. - Normal operational procedures for: submerging and surfacing, surface operations, underwater operations (visibility, currents, communications, surface traffic, etc.), and ferrying and transferring passengers (vessels carrying more than six passengers to and from the submersible are also subject to Coast Guard inspection). - Emergency procedures for scenarios such as: inability to surface, loss of power, controllable leakage of hull, collision, evacuation out of and off the vessel. - Mooring and operational area proposal. - Dive site location. Search and Rescue (SAR) Also important to the safe design and operation of submersibles is a rescue capability should it ever be needed. Although the Coast Guard search and rescue responsibilities extend under, as well as on, the surface of the sea, there was, when the USP first formed, no facilities in existence designed for underwater rescue from non-military submersibles. A Mutual Assistance Rescue and Salvage Pian (MARSAP), whereby other submersible owners would provide all possible assistance, 1492 RECREATIONAL BOATING ACCIDENTS IN OCEAN WATERS Gary L. Traub United States Coast Guard office of Navigation Safety and Waterway Services Washington, DC 20593-0001 ABSTRACT number since the Coast Guard began keeping rec- ords in 1960. The U.S. Coast Guard's Boating Accident Report file contains reports of 6,860 boats involved in Regulations under the Federal Boat Safety Act Of 4,781 recreational boating accidents occurring 1971 have helped considerably in this reduction between 1969 and 1987 on the oceans, the Gulf of of fatalities. The regulations have covered Mexico, and Long Island Sound. These accidents electrical systems, engines, boat flotation, accounted for an average of 74 fatalities, 71 personal flotation devices (PFD), other safety reported injuries, and $2,380,000 in reported equipment, and safe operation. Despite the property damage annually for those nineteen growth in the number of participants and the years. Because of the considerable difference increase in congestion in popular areas, the between the environments of these waters and overall danger in recreational boating is being inland waters, differences are also expected in reduced each year. This reduction is a result the characteristics of recreational accidents. of the production of safer boats and life saving appliances and an increase in educational Fatalities on the oceans have dropped from 93 efforts and public promotion of boating safety. fatalities per year in the 1969 to 1973 period to 48 per year in the current five-year period. STATE RANKINGS The most frequent (45%) type of fatal accident is capsizing. Falling overboard is responsible For many years Florida and California have for another 17% per cent of the fatalities. reported the greatest total number of accidents occurring on all waters. Since both have long shorelines and are populous, they also have accounted for the most accidents occurring on INTRODUCTION the oceans (and the Gulf of Mexico). The top six states and the number of reported accidents since 1960 the Coast Guard has been collecting on the oceans from 1969 through 1987 are: and publishing statistics annually on recre- ational boating accidents. Data on reported STATE ACCIDENTS accidents occurring since 1969 are in a computer file that is updated as each year's statistics California 1,451 are published. Thi!@ paper analyzes those acci- Florida 1,266 dents that are coded as occurring on the oceans, New Jersey 348 which also include, for our purposes, Long New York 239 Island Sound and the Gulf of Mexico. Because Hawaii 185 the data base contains no indication of the Oregon 177 distance of the accident from shore, we do not know the number of these accidents which The list of the top states in the number of occurred on the open seas. fatalities occurring on the oceans shows the same two states at the top. ALL RECREATIONAL ACCIDENTS STATE FATALITIES Boating Statistics, U. S. Coast Guard publica- tion COMDTPUB P16754.1, shows that the number of California 422 fatalities in recreational boating has been Florida 323 decreasing for approximately fifteen years. Oregon 77 Since 1971 the fatality rate has decreased from New York 71 20.2 fatalities per hundred thousand boats to Washington 64 6.1 fatalities in 1987. In 1973 the peak number New Jersey 59 of 1,754 fatalities occurred. Last year 1,036 people died in boating accidents, the lowest 1493 United States Government work not protected by copyright The top two states reverse their order when the Among the twenty-eight causes of accidents which states are ranked by the number of reported are listed in Boating Statistics, the proportion injuries occurring on the oceans. of accidents on the ocean caused by failure of the steering, throttle, or other non-power STATE INJURIES equipment is the most significantly increased over that proportion found on all waters. Florida 421 Instead of four per cent of the reported boats California 372 having such a failure, 7.9% of the boats on the New Jersey 121 ocean had equipment failures. Speeding shows Oregon 65 the most significant decrease. Only 1.2% of New York 63 ocean operators caused their accidents by speed- Hawaii 49 ing, whereas more than four per cent of opera- tors in reported accidents on all waters caused The ranking of the states according to the them by speeding. amount of reported property damage in thousands of dollars follows. The figures are not In analyzing fatalities we find that inattention adjusted for inflation, so the ranking is or carelessness is blamed for proportionately weighted towards the more recent years. fewer fatalities on the oceans than on all waters. A wave or wake striking the boat is the STATE PROPERTY DAMAGE cause for proportionately more fatalities on the ($000) oceans. This finding agrees with the finding that fatalities when boats capsize, which is one California 14,877 of the results of being struck by a wave, are Florida 11,667 proportionately more frequent on the oceans. New Jersey 3,271 Hawaii 2,984 BOATS North Carolina 2,323 Oregon 1,341 Most boats involved in reported accidents on the ocean from 1969 to 1987 are open motorboats. CAUSES OF ACCIDENTS The median length of all boats involved in reported accidents on the ocean is 24 feet, but The Coast Guard's Boating Accident Report file the most common length is sixteen feet. For allows for three codes describing the sequence those boats involved in fatal accidents, the of events for each boat involved in an acci- median length is nineteen feet. The most common dent. This analysis of all accidents from 1969 length for boats involved in fatal accidents is through 1987 concerns only the first code. For also sixteen feet. On the oceans, as on all instance, many capsizings occur after the boat waters, fatalities are more likely on smaller has flooded. Such accidents are analyzed as boats. floodings. The median horsepower of boats reported to be As one would expect, accidents occurring on the involved in accidents on the ocean between 1969 ocean are different from the accidents occurring and 1987 is 120 for powered boats. A steady on lakes, rivers, or other bodies of water. rise over the years in the power used on boats Grounding is more frequent among accidents on has increased the median for accidents occurring the oceans than on inland waters. There are in 1987 to 140 horsepower. In fatal accidents proportionately fewer collisions reported with the median is 70 horsepower, again showing the other boats or with fixed objects since boats greater likelihood of fatalities on smaller have more room to maneuver. Reports of falls boats. In the last four years the median for overboard are proportionately less frequent, fatal accidents has fallen to 50 horsepower. No probably due to the limited use of small boats explanation for the decrease is apparent. on the ocean. People in the water are not reported to be struck by boats or propellers as ACCIDENT CHARACTERISTICS often because water skiers, swimmers, and divers are not as common far from the ocean beaches. For each item of information that is coded for These statements on the frequency of reported each accident, including the boat, operator, and accidents do not account for any differences in the environment, some of the characteristics the type or length of use of boats on the oceans which occur significantly more or significantly versus those on other waters. Fatalities from less frequently on the oceans than on all waters falling overboard, colliding with another boat, combined will be stated. Statistics for all and striking a fixed object are proportionately waters used in the statistical testing covered fewer on the oceans. only the year 1986, when there were 1,066 fatal- ities and 6,407 reported accidents. Most char- The Boating Accident Report file allows for acteristics do not change much from year to three causes of accidents to be coded for each year, so that the 1986 statistics are appropri- boat. This analysis concerns only the first ate for general comparisons. code. If more than one cause is blamed for the accident, they are coded in the order of their When the frequencies of occurrence of reported impact on the casualty. accidents are discussed, we must remember that 1494 accidents are rare events. Associated charac- rate of approximately five per cent per year. teristics do not necessarily reflect the same However, since the total number of fatalities on proportions as one might observe in recreational all waters is also trending down, there is no boating experience or through surveys because immediate reason to suspect that there is any- the probability of an accident occurring is the thing affecting ocean fatalities that is not product of risk and exposure. Knowledgeable affecting all recreational boating fatalities. people usually have an idea of the relative risks, but not about the overall exposure of Since the number of fatalities on recreational segments of the boat population under various boats on the oceans is declining, the usual conditions. Therefore, we can not be too spe- analysis will find many downward trends. In cific in analyzing accident statistics, but must order to produce more useful results, the sta- talk more generally. Another problem is that tistical calculations are performed on the ratio only a small fraction of accidents are believed of the number of fatalities having each charac- to be reported. Analysts must use their experi- teristic to the number of all ocean fatalities ence to interpret the results meaningfully. The year by year. Such an analysis will, instead of following paragraphs in this section consider finding that the number of capsizings is in a those characteristics of reported accidents downward trend, compare the capsizing fatalities which are more likely or less likely on the to the downward trend of the total number of oceans than on all waters combined. fatalities. In the case of capsizing, there is no downward trend in the relationship to all According to this analysis, boats are less ocean fatalities, even though capsizing fatali- likely to be maneuvered or used for water skiing ties have dropped from 37 per year in the 1969- on the oceans. Cruising while the passengers 1973 period to twenty in the current period. are fishing is more popular on the oceans. The proportion of cabin motorboats and auxiliary The most significant trend is a rise in fatali- sailboats is greater on the ocean and open ties in accidents caused by the loss of stabil- motorboats are less common. Boats on the ocean ity in strong current, weather, rapids, white are more likely to have inboard engines and they water, etc. It appears that this cause was not are longer. usually considered for ocean accidents before 1976, when there is a sharp break in the number Based on accident reports, operators of boats on of fatalities each year. This is possible the ocean are older, with a higher percentage because the different Boating Accident Report over 50 years old and a lower percentage under forms which the states print have limited room 26. more operators have more than 500 hours of to list causes. So the coder must choose from experience and fewer operators have operated our list of 48 cause codes depending on how the boats of the same type for less than 20 hours. events of the accident are understood. People in accidents on the ocean are less likely to be alone on the boat. They are also more The second most significant trend is a decline likely to have taken a safety course with the in the percentage of fatalities on wooden Coast Guard Auxiliary or the U.S. Power Squad- boats. This is a direct result of the popular- rons. Fewer operators have had no safety ity of fiberglass over wood for new boat con- instruction. When the boat has an accident, the struction. in 1969, 31% of the registered boats personal flotation devices are more likely to be were wooden. By 1987 this percentage has used. dropped to 4.6. In contrast, fiberglass boats increased from 33% of all those registered in Environmental conditions on the oceans are dif- 1969 to 52% today. ferent because rough and very rough conditions are more likely. Moderate winds are more com- Fatalities have not occurred on boats using mon. Water temperature from 50 to 59 degrees exactly 75 horsepower since 1978. In the ten Fahrenheit is overrepresented in ocean acci- preceding years 39 fatalities had occurred. In dents, while temperatures of 70 degrees and the eight years from 1969 to 1976, 38 of those above are underrepresented. fatalities occurred, for an average of almost five per year. The sharp contrast between this The proportion of accidents occurring in the average and no fatalities in the present time early evening on the oceans are less frequent may be the result of boat owners scrapping their than on all waters, but the early morning hours previously high-powered engines of 75 HP for from 6:30 to 10:30 account for more accidents. larger ones. In fact, there is a significant June and July, the usual peak months of recre- increasing trend of fatalities on boats powered ational boating activity, do not see as high an with exactly 150 HP. accident peak on the oceans. Rental boats account for a lower proportion of accidents on Fatalities occurring in accidents where the sea the ocean than on other waters. temperature is from 70 to 79 degrees Fahrenheit are increasing significantly. This trend, and TRENDS two less significant trends for temperatures in the 50's and 60's, is explained by a trend The total number of fatalities occurring on the toward fewer fatalities in accidents where the oceans, Long Island Sound, and the Gulf of Mex- sea temperature is unknown. The investigators ico is on a significant trend downwards at a 1495 of fatal accidents seem to be recording the tem- 1986. Six of these eleven accidents occurred at peratures more often now than they did years ago. some distance offshore. The five known dis- tances from land are a quarter to a half mile, The percentage of ocean accident fatalities 1900 yards, one and a half miles, two miles, and found to have been using Personal Flotation three miles. In three of the eleven accidents, Devices (PFD) is on a downward trend. There are the environment (fog, darkness, and 3 to 6 foot several explanations possible. one is that seas) was a factor in causing the accident. search and rescue is becoming more, effective, However, attentive operators would have avoided thereby saving those who do use flotation. all three collisions. Another seven were caused Another is that the PFD'S themselves have become by inattention or carelessness or not obeying more effective. The criticism, over the past the Rules of the Road. Five sailboats, (two 25 several years, of the effectiveness of the most foot sailboats hit each other) were hit while comfortable PFD's may be discouraging some boat- they were under sail. ers from using their PFD'si Another less likely reason is reduced use of the oceans by those who Capsizings of 16' Boats are likely to wear PFD'S. Of the twelve capsizings of sixteen foot boats Water entering the boat over the transom, gun- which I will discuss, three occurred in 1987, wale, or bow is declining as a cause of fatali- three occurred in 1986, and six occurred in ties on the ocean. This may be partly due to 1985. Two of the three in 1987 occurred while the ability of small boats built under the level small craft warnings were in effect. Seven of flotation regulation to remain level on the sur- the twelve occurred in surf or close to shore or face even when flooded. The regulation covers at the entrance to a harbor, where waves were at only boats under twenty feet in length. (Remem- least three feet. Two boats were kayaks on the ber that the median length of boats involved in Pacific ocean off the coast of California. One fatal accidents on the ocean is nineteen feet.) of the kayaks was three miles away, the other However, only one other cause, wave or wake was sighted one eighth of a mile from a harbor striking the boat, among those that should be before the victim was lost by the observer. affected by the same boat capability, also shows Four of the twelve accident reports cited the a significant downward trend. Deeper investiga- distance of the boat from shore. The distances tion into the accident files may determine if are 150 yards, 1/8 mile, 1/2 mile, and three there is a more probable reason. miles. None of the other eight reports indi- cated that the boats were far offshore, so they The number of ocean fatalities that have probably were pretty close. The motors of two occurred in fog has been very low since 1981, boats stopped, allowing the boats to drift into when there were 13. There have been only thir- the surf of three feet or more. Three of the teen more in the six following years. operators were trying to reach shore in bad weather and water conditions. Equipment failure is becoming a thing of the past in causing fatalities on the ocean. There Capsizings of 141 Boats has not been one recorded since 1984. In the period 1969 through 1976, an average of more With information on four of the nine capsizings than six fatalities per year were blamed on of fourteen foot boats occurring from 1985 equipment failure. through 1987, the distances from shore were six hundred yards, three hundred yards, and two There are other trends that are statistically boats just off a beach. Two catamarans capsized significant for the period from 1969 to 1987. in Hawaii, one just getting underway at a beach, However, those which were described are the most the other on trying to return to the beach. significant. CONCLUSION ACCIDENT CHARACTERIZATION Recreational boating accidents which occur on Some statements will be made about the most the oceans, the Gulf of Mexico, or Long Island common accidents, based on the length of the Sound are different in some of their character- boat involved and the type of accident, in order istics. However, no matter where people use to supply a little more detail about how their boats, one thing above all remains the accidents occur. Reports of the most recent same. Operators, and their passengers, could accidents in each group were read to find what avoid many accidents if they were more attentive the accidents had in common. In general, and careful. capsizings are responsible for the greater proportion of fatalities. Collisions with another boat are responsible for very few fatalities, but many injuries. Collisions of 25' Boats With Other Boats Seven collisions of a 25 foot boat with another boat occurred in 1987. Another four occurred in 1496 TRIBUTYLTIN AND WATER QUALITY: A QUESTION OF ENVIRONMENTAL SIGNIFICANCE Michael H. Salazar Michael A. Champ Naval Ocean Systems Center National Science Foundation San Diego, CA 92152 Washington, D.C. 20550 ABSTRACT time and resources, even the best environmental data has some scientific uncertainty. Given the The most critical scientific issues that have caus- accelerated time requirements of the regulatory ed strict regulation of tributyltin (TBT) antifoul- process, these data have even higher levels of ing paints were reviewed. It was determined that uncertainty. The environmental impact of new regul- the evidence cited by scientists and regulators as ations are also unknown. most significant is equivocal. Unreal expectations that science can provide definitive answers in a Most of the evidence considered by the regulatory short time have been established through increased process to be most significant comes from bivalve environmental awareness, and politics. The regula- molluscs. This has occurred for a variety of tory process has accelerated decisions using less- reasons. First, it is generally believed that mol- than-optimum scientific data with abnormally high luscs are more sensitive than other animal groups degrees of uncertainty. Research has been driven to to TBT, the primary toxic component of organotin develop quick-response tests to quantify effects antifouling paints. Second, many bivalves have a only and not explain why they occur. At parts-per- cosmopolitan distribution and are commonly main- million (mg/1) cause-and-effects levels with simple tained in the laboratory. Third, filter-feeding bi- contaminants, these regulatory approaches were cost valves may be more susceptible to TBT due to their effective, and conservative approximations. How- feeding strategy. Fourth, many bivalves have an ever, faced with effects in the parts-per- trillion economic importance in the commercial shellfish in- range (ng/1), these regulatory approaches appear to dustry. One significant point that should be made be inappropriate. They have perpetuated many inter- here is the subtle distinction between the environ- pretations of the available data and many unresolv- mental impact of TBT on the shellfish industry and ed questions of environmental significance. This the environmental impact of TBT on natural popula- paper reviews the scientific data utilized by regu- tions. However, it should be noted that effects on latory agencies to establish criteria for the use cultured shellfish do not necessarily demonstrate of TBT in antifoulant paints. similar ecological effects in a typical natural situation. This issue becomes particularly relevant when oysters are raised in the vicinity of marinas INTRODUCTION or high vessel densities. Even if TBT were not present, it is difficult to understand selecting Mariners have battled fouling organisms on boat culture areas immediately adjacent to marinas and bottoms since the first days of sail. Phoenicians boatyards with many other contaminant problems. nailed copper strips to their wooden hulls as along Furthermore, marinas which are usually located in ago as 300 B.C. Copper-based antifouling paints protected areas with low flushing rates, other were introduced in the early 1800's and have been contaminants, and high silt loads do not have opti- the predominant bottom coating for over 150 years. mum conditions for culturing filter-feeding bival- The first organotin-based paints were developed in ves. 1961 and were not used in significant quantities for at least 10 years (1, 2). Organotin antifouling Nevertheless, this paper will focus on the scienti- paints can provide protection for up to 7 years fic aspects of data that the regulatory process has compared with 2 years for copper-based paints. chosen as the most important evidence against or- However, the future use of tributyltin (TBT) com- ganotin antifouling paints. In Europe, the critical pounds in antifouling paints will be severely limi- evidence was associated with shell thickening in ted in many countries. There is little doubt that oysters (Crassostrea gigas and imposex in dog- unregulated use of TBT in antifouling paints has whelks (Nucella lapillus . In the U.S., the criti- generated environmentally unsafe levels in many cal evidence was associated with three laboratory marinas. However, there are many unresolved ques- studies that reportedly demonstrated unacceptable tions of environmental effects in areas removed effects on growth and development in oysters (L._ from marinas. Where TBT concentrations are low, gigas, Ostrea edulis) and clams (Mercenaria problems of interpretation are due, at least in mercenaria . All of this evidence is based on only part, to a highly accelerated collection and analy- four species, a similar number of laboratory tests sis of scientifically questionable organotin data and field observations, generally unsupported by used for regulatory purposes. Assuming unlimited chemical measurements. In general, the laboratory 1497 United States Government work not protected by copyright studies utilized questionable methodology and field test conditions. It should be mentioned that cor- studies were little more than casual observations. roborating field measurements only establish The collection of this type of data was apparently another correlation with shell thickening. This does not disprove the field correlation with sus- driven by the short-term, crisis type needs of the pended sediment documented by Key et al. (10). regulatory process. Regulatory offices in the U.S. Further, the Waldock and Thain laboratory study is do not generally fund research, but only solicit questionable due to abnormally small sample size data and information from users, product manufact- (10 animals/replicate), artificial seawater car- urers and other interested parties. In effect the rier solvents, artificial algal diets, stati'c con- information requirements of the regulatory process ditions, and no statistical analysis. Additionally, monopolize many research resources in an attempt to all comparisons were made relative to a seawater get the information needed for policy and decision- control rather than comparing suspended sediment making. The result with TBT appears to be an abun- treatments with a clean suspended sediment control. dance of scientific information that is not effec- tive in predicting environmental effects. The regu- Due to the different methods and procedures used to latory process is forced to be even more conserva- study the effects of organotin compounds in many tive because of the abnormally high scientific studies like this, the National Oceanic and Atmos- uncertainty in the data. pheric Administration in the U. S. sponsored an SHELL THICKENING IN C. GIGAS Interagency Workshop on Aquatic Monitoring and Analysis of Organotin Compounds in 1986 to review The Earcpean Experience and recommend standardized procedures for field studies (12). By 1976 French and U.K. scientists had observed an Thain and Waldock (13) suggest that TBT is at least abnormal thickening in the shells of Pacific oy- a contributory factor and probably a major factor sters, C. gigas (3) The first difficulty in inter- in the failure of 0. edulis to reproduce naturally preting these field observations of shell thic- in the Crouch estuary. Thain also points out that kening is that this species is not native to the natural population fluctuates widely and has Europe. It had been introduced to France in 1968 been decimated by harsh winters, over fishing and and to the U.K. in 1972 (4, 5), about the same time disease. The last major collapse followed the harsh TBT antifouling paints were becoming more widely winter of 1962-1963 (14). It is well known that the used. There was very little baseline data on chance of repopulation diminishes when stocks are "normal" growth of this exotic (non-native) species low. before the anomalies were reportedly associated with TBT. The second difficulty was that existing analytical methods were insufficient to support the After the natural failure of the natural population early French work with appropriate chemical measur- of 0. 'edulis, a number of species were introduced, ements. Field studies were generally not conducted unsuccessfully. These include Crassostrea angulata by commonly accepted procedures (6). This environ- Crassostrea virginica Ostrea lutaria and Ostrea mental work appears to have been conducted in es- chilensis (4, 15, 16). This should not be surpris- sentially scientific isolation. Most of the evi- ing since Mann (15) has concluded that "The intro- dence was largely correlative and therefore the duction of a exotic (non-native) species can be a interpretation was subjective (1). Supporters of highly unpredictable and potentially harmful act, the 1982 French ban point to the marked improvement with results ranging from total failure to repro- in spatfall and reduced shell thickening in duce to practically uncontrolled proliferation of Arcachon Bay but fail to discuss that LaRochelle the species in the new environment". After over 50 Bay showed little improvement after the ban (7). years of culture of C. _gigas on the Pacific coast Alzieu (3) does report some measurements of heavy of North America, its future as a self-sustaining metals in Arcachon Bay but suggests no correlation species in that region is still uncertain. with shell thickening. It is interesting to note that the concentrations of copper in Arcachon Bay It also seems too coincidental that the decrease in are an order of magnitude above TBT levels and are spat from Japan and British Columbia was concurrent within the range reported to reduce growth (8) and with the first reports of shell thickening in cause shell deformities (9) in mussels. France and the U.K. Although hatcheries provide a large number of progeny from a small number of In the U.K. , thickening was first attributed to broodstock and a source of spat without natural suspended sediment by Key et al. (10). Waldock and breeding, there is a significant potential problem. Thain (11), conducted a laboratory study to deter- Mann (15, 16) has suggested that the impact of mine whether this shell thickening phenomenon was exotic species on the environment may not be evi- induced by suspended sediment or TBT. They found dent for many years. In the context of TBT and that "clean" sediment did not cause shell thicken- simultaneous introduction of C. gigas, it is likely ing. TBT alone and sediment contaminated with TBT that the natural effects of The environment on the both induced thickening. As with the Key et al. oyster may not be evident for many years. C. gigas study, this only shows a correlation, and does not has a tendency toward homozygosity and, without prove a cause-and-effect relationship in the field. clear genetic lines to promote vigorous hetero- zygotes, the progeny may be more susceptible to any Other contaminants in the sediment were not measur- environmental perturbations. It should be remem- ed. If sorbed on sediment surfaces they could have bered that the main reason for introducingg.. - Rigas been a contributing factor to the observed shell was the failure of a natural population from natur- thickening in addition to the stressful laboratory al causes. 1498 The U.S. Experience a serious problem for the oyster industry because meats are sold out of the shell. The thickened The concern created by the European experience has shell was regarded as a trivial problem. They also had a profound effect on expediting the U.S. regu- cite papers that report shell chambering in C. latory process. Without waiting for the national gigas from Japan before the introduction of TBT regulation, several states have already placed paints. Similar early accounts of shell thickening strict limits on the use of TBT. The results from that were apparently unrelated to TBT have been TBT use in France and the U.K. most likely caused reported in British Columbia (21). Chambering in 0. EPA to be more conservative, lowering their TBT edulis was reported as early as 1916 (22, 23). criteria below the calculated values. Data were European biologists generally agreed that cham- used from three tests that have significant limita- bering was caused by body shrinkage and deposition tions and should not be used for environmental of partitions to occupy the unused space (24, 25, prediction. EPA has acknowledged that they did not 26). Although TBT was not measured in the trans- properly scrutinize methodology or results (7) due plant experiments, Okoshi et al. (20) doubt whether to regulatory deadlines and the inadequacy of the a relation exists between shell thickening and TBT data. It is interesting to note the coincidence in antifouling paints. They do believe that there is a the literature cited for significant effects at 20 significant genetic and environmental component ng/l TBT and the original 20 ng/l TBT EQT (Environ- responsible for the observed thickening response. mental Quality Target) in the U. K. Again, commer- cially important species are emphasized and the Since C. gigas does not reproduce naturally in the environmental significance in the typical natural U.K. and Japan has not exported seed since 1978, situation is only implied. all laboratory and field test animals come from hatchery operations that produce animals with un- Shell thickening studies with C. gigas in the U.S. certain genetic backgrounds. It seems likely that have been driven by the European experience and a inbreeding has occurred over the years and the need for organotin field data. This shell thic- resulting oyster may be even more prone to thicken- kening index has been used by Wolniakowski et al. ing than the original Japanese Miyagi race. In (17) in Oregon, and Stephenson et al. (18, 19) in numerous studies of natural and hatchery popula- California. Its validity as a reliable bioindicator tions, heterozygote deficiencies have been reported in any environment has never been scrutinized. It for over two dozen bivalves species including is Important to recognize how pressures of the Mytilus edulis C. gigas and 0. edulis (27, 28). regulatory process have limited and focused science This tendency toward homozygosity has been cor- in such a narrow direction. The reported specif- related with low growth rates and high mortality icity of the shell thickening index is also its (29). Hatchery reared animals, particularly C. weakness. gigas tend to have a genetic as well as an en- vironmental liability. First, offspring tend to The first International Organotin Symposium was lack hybrid vigor due to homozygous tendencies. held in Washington, D.C. in 1986. One obvious ques- Second, hatchery animals are conditioned to grow tion asked was; Is there anything other than TBT well in the hatchery and not necessarily in nature. that can cause shell thickening? European scient- Natural stocks generally grow faster, exhibit ists reported that nothing they had tested induced higher survival and are less susceptible to en- shell thickening in C. gigas None of these results vironmental changes than their hatchery cohorts have ever been published however, so there has been (30). no peer or regulatory review. At the 1987 Interna- tional Organotin Symposium in Halifax, Nova Scotia, Although this work on shell thickening in C. gigas a panel discussion addressed two similar questions. from Japan was not reported in the international (1) Is there anything else that might cause shell literature until 1987, the actual research was thickening? (2) Why have there been no reports of completed in 1984. Since they had predicted poten- shell thickening from Japan where C. gigas is a tially catastrophic effects on the oyster industry, native species? European scientists present at the it seems odd that European scientists did not con- meeting dismissed these concerns even though TBT tact Japanese scientists. The possibility of hat- use in Japan is widespread. chery deficiencies was apparently disregarded, even though oyster inbreeding is well known in the sci- entific community. The Japanese Connection MOLLUSCAN GROWTH EFFECTS In 1987, the first reference to shell thickening in C. gigas from Japan appeared in an international 1uposex in Nucella lapillus journal. Okoshi et al. (20) reported that there were distinct differences in propensity for shell Another reportedly TBT-specific bioindicator that thickening in two local races of the Japanese oy- has influenced the bans in Europe and the TBT ad- ster C. gigas The Miyagi race has a genetic ten- visory in the U.S. (7) is imposex in dogwhelks CN. dency for shell thickening. This race was primarily lapillus , i.e., the imposition of male sexual exported to Europe and the U. S. (15, 16). Contrary characteristics on the female of the species. It is to what had been reported in the U.K., Okoshi et unclear whether TBT is the only chemical that can al. (20) found no correlation between chambering induce this phenomenon, particularly when this and either shell height or flesh weight. Further- tendency toward hermaphroditism is quite common in more, they did not consider shell thickening to be gastropods. Severe imposex can apparently affect 1499 the population but the degree of imposex and the conservative lowering of the advisory in itself TBT concentration that induces these effects suggests that the regulatory process is flawed if remains unclear. The laboratory studies have extra- limited data are available. For purposes of this polated effects from TBT concentrations beyond discussion, it will be assumed that the calculation analytical capabilities and reported corroboration of the Final Acute Value was valid and only the by casual field observations that is highly specul- reduction of the calculated value will be scrutin- ative (31, 32). Further, in similar studies on a ized. closely related species, Smith (33, 34, 35, 36) reports that other chemicals like organoleads might The major problems with larval and juvenile bivalve induce the same phenomenon in Nassarius obsoletus bioassays are unrealistic test conditions (static and that imposex does not necessarily have any renewal, artificial seawater, unialgal feeding, significant impact on the population. The other carrier solvents), unme 'asured test concentrations important factor to consider is the genetic propen- and extreme variability in aliquots used to repre- sity for imposex. Just like inbreeding and result- sent survival. Additionally, TBT adsorption to ing homozygosity in C. gigas dogwhelks have a particles could make them less bioavailable for (42). The very strong tendency to inbreed. N. lapillus part- bivalve larvae as previously suggested icularly, spend their entire livJ-in a very limit- unrealistic test conditions might represent what ed area and there are no pelagic larvae. There is Carpenter and Huggett (43) have referred to as very little genetic exchange. A recent study on "biogeochemical quackery" in assuming that experi- Nucella lamellosa has shown the greatest popula- mental conditions are similar to those in the tion subdivisions reported for a marine gastropod field. Unmeasured test concentrations are unaccept- (37). This generally supports the theory that these able when testing in the parts-per- trillion range. particular gastropods with limited larval and adult The EPA calculated the advisory concentration based dispersal have genetically fragmented populations that would be very susceptible to environmental on a pre-determined mathematical formula and then changes. lowered it based on "other considerations." Several steps in this scenario are flawed and suggest that The two most significant findings that led to bans unresolved questions of understanding environmental significance are being perpetuated by the regula- in France and the U.K. could be attributed to po- tory process. The derivation of the Final Acute tentially large genetic deficiencies in two mol- Value based on the four most sensitive species luscan species, C. gigas and N. lapillus One is a tested for saltwater will not be discussed here. No non-native species that is @_ultured for consump - bivalves were used in the calculation even though tion. The other is naturally occurring but appears the advisory was lowered because of reported ef- to be at an evolutionary standstill due to its fects on bivalves. It will be assumed that the inability to exchange and distribute genetic infor- acute value, approximately 500 ng/l is correct. It mation. If the philosophy of regulatory criteria is should be remembered that this value is a function to protect 95% of the species, dogwhelks and oy- of the most sensitive species tested to date. The sters with potentially large genetic limitations do next critical step is determining the acute/chronic not seem like appropriately representative species. ratio. Under the EPA guidelines, acceptable studies The Japanese paper (20) has stressed the importance for determining this value are only available for of genetics and environment in explaining observed two freshwater species and one saltwater species. thickening effects. The importance of genetics It is important to note that although the saltwater (stock) and environment (site) has also been stres- mysid is the most sensitive species used in the sed in explaining mortality and growth in a number calculation, the acute/chronic ratio is signifi- of bivalve species including M. edulis 0. edulis cantly higher for the less sensitive species. This C. virizinica (38, 39, 40, 41). Additionally, the ratio is more a function of procedures than the data from questionable laboratory studies and actual effects of tributyltin. The Daphnia tests casual field observations with minimal chemical were conducted for only two days. According to the measurements precludes establishing a clear cause- EPA guidelines (44), this is the maximum time they and-effect relationship. could be kept alive without feeding. This gives an artificially high acute value. If the animals were Regulatory Procedures exposed for a longer period of time, e.g., 96 hours, the acute value and thus, the ratio would be The majority of laboratory testing in the U.K. has significantly lower. The animals were fed during been on the oyster C. gigas Although this work was chronic tests and a more reasonable estimate of intended to corroborate field observations in projected effects in nature was obtained. It should Europe, the results are subject to a variety of be noted that the chronic values for mysids and interpretations and unresolved questions of en- Daphnia are quite similar. It might be expected vironmental significance. Nevertheless, these and that the acute values would be similar if the similar studies were used by EPA to lower their Daphnia were fed and the test conducted for 96 organotin advisory criteria calculated by a precise hours as with the mysids. formula from 47 ng/l to 10 ng/l TBT. More impor- tantly, EPA chose to regulate the use of TBT anti- The EPA also suggested that the freshwater numbers fouling paints by inference of unacceptable effects can be used to calculate the acute/chronic ratio on larval and juvenile clams and oysters. Reduc- for the saltwater advisory concentration. However, tions in growth and development were reported near the most sensitive and the most resistant of fresh- 20 ng/1 TBT in laboratory studies. These studies water organisms differ by a factor of 20 as compar- violate EPA's own criteria for acceptable data, but ed to a factor of 463 for saltwater organisms. Ad- were used for proposed national regulation. The ditionally, the calculated advisory concentration T500 for freshwater is only half of the calculated value reported at 25 ng/l TBT, these concentrations were for saltwater. not measured as they were below their limits of detection. This violates another of EPA's criteria Interestingly, over 5 years ago, preliminary static for accepting data, i.e., measured test concentra- tests with mysids provided a 96-hour LC-50 estimate tions (44). In addition there was extreme variabil- of approximately 500 ng/l TBT. At the time, this ity in estimated survival from aliquots with diffe- was one of the most sensitive species tested. In rent results on different sampling days. More than preparing the Navy's Environmental Assessment (45) half of the solvent controls has significantly that number was divided by a safety factor of 10 to different survival than the seawater control. Test estimate 50 ng/l as a safe level. Although a gross validity is questioned because control survival was estimate, it is amazing that 5 years later with only 3% at the end of the test. Further, larvae sophisticated mathematical manipulation of signifi- from adults "conditioned" on a diet of corn starch cantly more data from a variety of species, the and algae were probably under stress passed on from regulatory process would produce a number that is the adults (53, 54, 55). virtually identical (47 ng/1). This also suggests that only a few key species have a large impact on The second test that was used to lower the U.S. the criteria and that greater care should be exer- advisory was conducted by Thain and Waldock (56) on cised in the interpretation of those bioassays in 0. edulis, another species not native to North particular. America. The results could be rejected on the basis of non-native species and unmeasured test concen- Larval and Juvenile Bivalve Growth trations. These tests were conducted under static conditions, using artificial seawater, artificial However, the regulatory process suggests that there algal diets, and solvent controls, all of which are unacceptable effects on commercially important either separately or in combination could signifi- bivalves at concentrations less than this calculat- cantly affect the results. There was no replication ed value. EPA cites reductions in growth and rate and no statistical analysis. The authors suggest of development in three different species. Pre- that significant reductions in growth occurred at viously White and Champ (46) have discussed the 60 ng/l TBT. EPA selected 20 ng/l TBT as a critical shortcomings of bioassays in general for predict- concentration because growth was slightly lower ing environmental impact and their questionable use than the controls. In a conservative regulatory by the regulatory process. Cunningham (47) has scenario it seems reasonable that any statistically enumerated essential factors in the design and significant reduction of growth rates In nature conduct of experiments using bivalves. The environ- could be considered environmentally significant. mental significance and interpretation of organotin Any reduction in laboratory growth, statistically bioassays have been discussed with an emphasis on significant or not, should be interpreted with understanding bioavailability (42). The three bio- extreme caution, In this case, no statistics were assays cited by EPA are subject to the same pro- used. blems noted in other bioassays, but the test condi- tions are so extreme that they warrant further The only study used to lower the advisory in which scrutiny and interpretation. test concentrations were measured at levels of reported effects was in the most recent study con- All three tests were conducted under static renewal ducted by Laughlin et al. (57). Observed effects at conditions, which are more realistic than static, 25 ng/l TBT were considered significant by EPA even but not representative of the environment. All though the authors reported significant effects at three tests used an approximation of seawater which 50 ng/l TBT. Solvent carriers were used which may was either artificial seawater, and/or seawater reduce survival and increase bioavailability. This that had been filtered and sterilized in addition study suffers from many of the same problems of the to stock algal cultures as food. These test condi- other tests, partly due to difficulties in main- tions are so far from natural conditions it is taining larvae in the laboratory. Static conditions difficult to believe that animals in nature would must be used because there are no techniques avail- respond similarly. The use of carrier solvents and able to maintain them under flow through condi- filtered seawater may have increased bioavailabil- tions. The algal species used as food were much ity by concentrating TBT on food particles. different than field populations. In addition, this was an uncontrolled experiment without replication Bivalves could be more sensitive to TBT because that defies statistical analysis. If replicate they filter and retain bacteria, algae and sediment containers had been counted at each sampling time, as food with the associated TBT. Laboratory studies a statistical variance could be associated with the have shown that TBT rapidly binds to suspended results. Another problem was the utilization of sediment particles, algal cells and bacterial cells rarefied data. That is, only animals that grew were (48, 49, 50, 51), and that suspended particulates included in the growth rate computations. This enhance TBT bioaccumulation in mussels (42, 48) and biases the results in favor of finding a difference oysters (42, 11). between test concentrations. It does not account for a natural distribution where some animals grow The EPA regulatory Protocol has suggested that only slowly but survive to grow when conditions improve. native species be used in predicting environmental Larvae were obtained from an experimental hatchery effects (7). The unpublished Springborn Bionomics with potentially the same inherent problems prev- study (52) violates this criterion by using C. iously discussed. gigas larvae. The validity of the experimental design should be scrutinized. Although effects were 1501 As with the European experience, EPA did not scru- Davies et al. (68) have shown that TBT accumulation tinize the three studies it used to lower the ad- by oysters in the field is generally higher than in visory (44). Particularly, EPA did not consider the laboratory and it takes longer to reach equili- problems in methodology or analysis. Further, the brium. This field study also casts doubt on the potential genetic and environmental deficiency of validity of laboratory studies which were used to hatchery-reared animals and their offspring were lower the advisory values. Thain and Waldock (56) not considered. reported equilibrium after only 14 days where Davies et al. (68) report equilibrium in C. Rigas ANALYTICAL LIMITATIONS AND BIOAVAILABILITY after 16 weeks. Part of these differences could be attributed to stressed animals in the laboratory. There has been a rapid improvement in development Stress in the laboratory can be caused by main- of analytical techniques to measure the various tenance under unnatural conditions (53, 54, 55). species of butyltins and progress in identifying Stress in the field could be attributed to TBT, the species that are most toxic. Bioavailability other contaminants or physical/chemical factors has yet to be quantified in any study. How can (70). environmental effects be predicted when there still is not a definitive method for reliably measuring Kiorboe et al. (71) suggested that growth rates in the bioavailability of TBT in nature? Partitioning optimum laboratory studies do not approach growth experiments have reported TBT associated with part- rates in the field primarily because M.,edulis iculates from 3 to 70 % in nature (58, 59). This derives additional nutrition from suspended parti- demonstrates a high degree of natural variability culates. They may also accumulate additional TBT and the inability of the detector to distinguish from these suspended particulates. between truly dissolved TBT and TBT associated with paint dust. Since these measurements were made in Mysids exposed to TBT survived better with bottom different types of estuaries, at different times of sediment than without sediment (72) while the pre- the year, with different levels of suspended solids sence of suspended sediment appears to enhance the and only by differentiating a small fraction of deleterious effects of TBT on oyster spat (11). particles, the environmental significance is un- Henderson suggests that TBT did not bind to phyto- clear. To predict environmental effects, this rela- plankton, but two species of coral had significant- tionship must be more clearly defined and it cannot ly higher mortality when exposed to TBT with phyto- be measured by chemical measurements alone. It is plankton than without (73). Laughlin (48) found still not clear if TBT measured by detectors is that phytoplankton enhanced TBT accumulation in equivalent to TBT available to organisms and how mussels. Waldock and Thain (11) showed that TBT bioavailability varies by species. with sediment particles reduced growth more in C. Rigas than TBT alone. These three bioassays demon- Laughlin et al. (48) have shown that lab values are strate that TBT binds to bottom sediment and that not reliable measures of the bioaccumulation pro- sediment and feeding type affect bioavailability. cess for mussels and have suggested that accumula- tion under field conditions is necessary for ac- curate predictions of environmental effects. This has been confirmed in a recent field study (60). FIELD STUDIES Previous work by others on heavy metal accumulation by mussels has shown that bioaccumulation is af- Given the high degree of scientific uncertainty in fected by age, sex, season, tidal height, and an- the laboratory studies previously discussed, it is imal condition (61, 47). It is well established not reassuring that field studies and field that organic complexes reduce the toxicity of monitoring of TBT have an even higher level of metals to phytoplankton and zooplankton. These same uncertainty. Environmental complexity and vari- complexes may enhance the bioavailability and toxi- ability have precluded establishing a clear cause- city of metals to filter-feeding bivalves (62, 63, aDd-effect relationship between measured TBT con- 64, 65). Organic complexes may also increase the centrations and observed effects in the field. This bioavailability of TBT for bivalves and reduce the caused Stebbing to conclude as late as 1985 that bioavailability of TBT for phytoplankton and zoo- "While the weight of evidence strongly implicates plankton. organotins, strictly speaking the case remains inconclusive." Stephenson (19) reported significant The early laboratory work suggests bivalve growth reductions in bivalve growth along an increasing rates are directly affected by accumulated TBT (14 ' TBT gradient but other conditions along that gra- 66, 11). Recent work (60) shows that very low dient were not documented and the control site growth rates are associated with high TBT concen- could not reasonably be called a true control. trations in mussel tissues and that below 1.5 ug Further, few chemical measurements were made to TBT/g tissue there is no clear correlation between establish a true gradient. TBT in mussel tissues and mussel growth rates. Salazar and Salazar (60) have questioned the These shortcomings were addressed at Oceans 187 utility of extrapolating environmental effects from (70) and Stephenson acknowledged that these methods tissue levels, extrapolating seawater TBT concen- could not be used as previously suggested. Similar trations from tissue levels, or using field data criticisms could be made of the European experience to verify laboratory data (13, 59, 67, 68, 69, 32, in the field. Now the validity of the shell 66). thickening index has also been questioned as inap- propriate. The failure to bring this information to the attention of the scientific and regulatory 1502 communities has resulted in potential overestimates In the U.S., results from studies on C. gigas and of TBT effects in nature and a regulatory response N. lapillus could be rejected as non-native species which is likely overly conservative. and results from the three larval and juvenile studies on C. gigas M. mercenaria and 0. edulis A number of authors have discussed the questionable rejected for not adhering to commonly accepted value of single chemical measurements at a point in guidelines for toxicant testing. In Europe, the space or time but there are few meaningful measure- question of environmental importance was over- ments available for TBT on a real-time basis. This whelmed by the significance of the oyster industry. is absolutely necessary for field experimentation Everywhere, scientists and regulators were pressur- and environmental prediction. Otherwise, the "mind- ed for quick results. If not, advisories might not less monitoring" and "biogeochemical quackery" have been lowered in both the U.K. and U.S. and TBT discussed by Carpenter and Huggett (43) are ap- might not have been so severely restricted. In proached. short, the existing regulatory process may not be appropriate for evaluating environmental effects in This is particularly true of chemical measurements the low parts-per-trillion ranges. Regulatory agen- reported from the U.K. Most samples were taken at cies should consider holding an international re- slack water low tide-which can overestimate actual view of the protocols for regulating toxicants with concentrations. The number of measurements that ng/l level effects. Perhaps GESAMP or NAS/NRC exceed the arbitrary EQT of 20 ng/l has also been could coordinate such an effort on an international reported and emphasized. Order of magnitude differ- basis. ences in TBT concentrations have been reported to vary with tidal flow (74) and in weekly survey It seems that at least some of the problem in per- samples (60). Further, that relatively short time petuating these unresolved questions of environmen- of slack water low tide represents a small fraction tal significance can be attributed to advocacy of the time that animals are exposed to the highest instead of science, and conducting most of the concentrations. It has yet to be demonstrated that early work in virtual scientific isolation without animals respond to mean TBT concentrations over the peer review. Some of the problem can also be attri- entire tidal cycle or just the extreme highs. Any buted to the regulatory process accelerating decis- valid environmental assessment should include means ions using less- than-optimum scientific data with as well as extremes with some integration of expo- abnormally high degrees of uncertainty. This has sure over time that is commensurate with the natur- perpetuated many interpretations of the available al temporal and spatial variability of the eco- data, many unresolved questions of environmental system in question. It has also been shown that significance and many overly conservative regula- surface samples are generally much higher in TBT tions. The remainder of the problem can be attri- concentration than deeper samples (75, 60), so it buted to the uncertainty of environmental science. hardly seems representative to use one water sample For TBT, this uncertainty appears much higher than taken near the surface at slack water low tide to acceptable, much higher than it could have been. represent the entire tidal cycle and all water within the basin. ACKNOWLEDGMENTS SUMMARY Opinions expressed in this paper are those of the authors and may not necessarily reflect those of Studies have reported that TBT exposure produce the their respective organizations or U.S. government following effects: (1) shell thickening in C. policy. gigas (2) imposex in N. lapillus (3) growth re- ductions in larval C. gigas and M. mercenaria as well as juvenile 0. edulis. Data from a limited REFERENCES number of laboratory tests and field observations were used to regulate TBT antifouling paints in 1. Stebbing, A. R. D. 1985. Organotins and water Europe and the U.S. A careful review of the sci- quality - some lessons to be learned. Mar. Poll. entific methodologies used in these studies shows Bull. 16(10) 383-390. that these critical bivalve test results remain equivocal and that a true cause-and-ef feet rela- 2. Champ, M. A. and F. L. Lowenstein. 1987. TBT: tionship in nature remains unproven. The data may The dilemma of high-technology antifouling paints. also have been biased by emphasizing reported ef- Oceanus 30(3):69-77. fects on cultured, non-native species with possible genetic deficiencies. It has been acknowledged that 3. Alzieu, C. 1986. TBT detrimental effects on the information reported for most studies and time oyster culture in France - Evolution since anti- constraints did not allow a thorough evaluation of fouling paint regulation. In: Proceedings, Oceans the methodology or the results. Most of the labora- 1986 Conference, Washington, D. C., 23-25 Sept tory tests did not follow commonly accepted proce- 1986, Organotin Symposium, Vol. 4, pp. 1130- 1134. dures. Most of the field studies report mostly anecdotal information without measuring other fac- 4. Walne, P. R. and M. M. Helm. 1979. Introduc- tors that could have affected molluscan growth. tion of Crassostrea izigas into the United Kingdom. Furthermore, baseline studies to clearly demon- In: R. Mann (Ed.), Exotic Species in Mariculture, strate growth effects or existing environmental MIT Press, Cambridge, Massachusetts. Proceedings of conditions before the introduction of both TBT and a Symposium of Exotic Species in Mariculture: Case gigas were not conducted. Histories of the Japanese Oyster Crassostrea.gigas, Woods Role, 1978. pp. 83-105. 1503 5. Maurin, C. and J. LeDantec. 1979. The culture of 17. Wolniakowski, K. U. , M. D. Stephenson and G. Crassostrea izigas in France. In: R. Mann (Ed.), S. Ichikawa. 1987. Tributyltin concentrations and Exotic Species in Mariculture, MIT Press, Pacific oyster deformations in Coos Bay, Oregon. Cambridge, Massachusetts. Proceedings of a In: Proceedings, Oceans 1987, Halifax, Nova Symposium of Exotic Species in Mariculture: Case Scotia, Canada, September 28 October 1, 1987, Histories of the Japanese Oyster Crassostrea gigas Organotin Symposium, Vol. 4:1438-1442. Woods Hole, 1978. pp. 106-122. 18. Stephenson, M. D. , D. R. Smith, J. Goetzl, G. 6. Popham, J. D. and J. M. D'Auria. 1983. Combin- Ichikawa and M. Martin. 1986. Growth abnor- ed effect of body size, season, and location on malities in mussels and oysters from areas with trace element levels in mussels (Mytilus edulis high levels of tributyltin in San Diego Bay. In: Archs. Envir. Contam. Toxic. 12:1-14. Proceedings, Oceans 1986 Conference, Washington D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 7. Environmental Protection Agency. 1987. Tri- 4, pp. 1246-1251. butyltin Technical Support Document, Position Docu- ment 2/3. Office of Pesticides Programs, Office 19. Smith, D. R., M. D. Stephenson, J. Goetzl, G. of Pesticides and Toxic Substances, U. S. Environ- Ichikawa and M. Martin. 1987. The use of trans- mental Protection Agency, October 1987. planted juvenile oysters to monitor the toxic ef- fects of tributyltin in California waters. In: 8. Manley, A. R., L. D. Gruffydd and P. C. Almada- Proceedings, Oceans 1987 Conference, Halifax, Nova Villela. 1984. The effect of copper and zinc on Scotia, Canada, 28 Sept I Oct 1987, Organotin Sym- the shell growth of Mytilus edulis measured by a posium. Vol. 4, pp. 1511-1516. laser diffraction technique. J. Mar. Biol. Assoc. U K. 64:417-427. 20. Okoshi, K. , K. Mori and T. Nomura. 1987@. Characteristics of shell chamber formation between 9. Sunila, I. and R. Lindstrom. 1985. Survival, the two local races in the Japanese oyster, Cras- growth and'shell deformities of copper- and sostrea Rigas Aquaculture 67:313-320. cadmium-exposed mussels (Mytilus edulis L.) in brackish water. Estuar. Coastal. Shelf Sci. 21. Quayle, D. B. 1969. Pacific oyster culture in 21:555-565. British Columbia. Fish. Res. Bd- Can Bull. 169. 193 10. Key, D. R. S. Nunny, P. E. Davidson, and M. A. PP. Leonard. 1976. Abnormal shell growth in the 22. Houlbert, C. and C. Galaine. 1916. Sur le Pacific oyster (Crassostrea gigas : Some pre- chambrage des huitres et sur 11infection possible liminary results from experiments undertaken in des chanbres par le fait d1une Annelide tubicole 1975. ICES Paper CM 1976/K:11, International parasite de la coquille. Comptes Rendus Heb- Council for the Exploration of the Sea, Copenhagen. domadaires des Seances de l'Academie des Sciences, Vol. 162:54-56. 11. Waldock, M. J. and J. E. Thain. 1983. Shell thickening in Crassostrea izigas: organotin 23. Houlbert, C. and C. Galaine. 1916. Sur les antifouling or sediment induced? Mar. Poll. Bull. causes du chambrage et sur l1entretien raisonne des 14(11):411-415. bancs d1huitres naturels. Comptes Rendus Heb- domadaires des Seances de I'Academie des 12. Landy, R. B. , S. T. Holm, W. G. Conner. 1986. Sciences, Vol. 162:301-304. Intragency workshop on aquatic monitoring and analysis for organotin compounds. NOAA/National 24. Korringa, P. 1951. On the nature and func- Ocean Service National Marine Pollution Program tion of "chalky" deposits in the shell of Ostrea Office, July 1986. edulis Linneaus. Proceedings of the California Academy of Sciences Series 4, Vol. 27(5):133-158. 13, Thain, J. E. and M. J. Waldock. 1986. The impact of tributyl tin (TBT) antifouling paints on 25. Orton, J. H. and C. Amirthalingam. 1927. molluscan fisheries. Wat. Sci. Tech. 18:193-202. Notes on shell- depositions in oysters. J. Mar. Biol. Assoc. U. K. 14:935-953. 14. Thain, J. E. 1986. Toxicity of TBT to bi- valves: Effects on reproduction, growth and sur- 26. Worsnop, E. and J. H. Orton. 1923. The cause of vival. In: Proceedings, Oceans 1986 Conference, chambering in oysters and other lamellibranchs. Washington, D. C., 23-25 Sept 1986, Organotin Sym- Nature 3(2775):14-15. posium, Vol. 4, pp. 1306-1313. 27. Zouros, E. and D. W. Foltz. 1984. Possible 15. Mann, R. 1979. Exotic species in aquacul- explanations of heterozygote deficiency in bivalve ture. Oceanus 22(l):29-35. molluscs. Malacologia 25(2):583-591. 16. Mann, R. 1983. The role of introduced bi- 28. Fujio, Y. 1982. A correlation of hetero- valve mollusc species in mariculture. J. World zygosity with growth rate in the Pacific oyster, Maricul. Soc. 14:546-559. Crassostrea gijzas Tohoku J. Agri. Res. 33(2):66- 75. 1,%4 29. Gaffney, P. M. and T. M. Scott. 1984. Genetic 42. Salazar, M. H. 1986. Environmental signi- heterozygosity and production traits in natural and ficance and interpretation of organotin blo- hatchery populations of bivalves. Aquaculture assays. In: Proceedings, Oceans 1986 42:289-302. Conference, Washington D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, pp. 1240-1245. 30. Wilkins, N. P. 1976. Genic variability in marine bivalvia: implications and applications in 43. Carpenter, J. H. and R. J. Huggett. 1984. molluscan mariculture. In: G. Persoone and E. Meaningful chemical measurements in the marine Jaspers (Eds.), Proceedings, 10th European environment transition metals. In: H. H. White Symposium on Marine Biology, Ostend, Belgium, Sept. (Ed.), Concepts in Marine Pollution Measurements. 17-23, 1975. Vol. 1:549-563. Maryland Sea Grant Publication, University of Maryland, College Park, pp. 379-403. 31. Bryan, G. W. , P. E. Gibbs, L. G. Hummerstone and G. R. Burt. 1986. The decline of the gastro- 44. Environmental Protection Agency. 1987. Am- pod Nucella lapillus around south- west England: bient aquatic life water quality advisories for evidence for the effect of tributyltin from anti- tributyltin.. U. S. Environmental Protection fouling paints. J. Mar. Biol. Assoc. U. K. 66:611- Agency, Office of Research and Development, 640. Environmental Research Laboratories, Duluth, Minnesota, Narragansett, Rhode Island. 32. Bryanp G. W.t P. E. Gibbs, G. R. Burt and L. G. Hummerstone. 1987. The effects of tributyltin 45. United States Naval Sea Systems Command. 1984. (TBT) accumulation on adult dog-whelks, Nucella Environmental Assessment for Fleetwide use of or- lapillus long-term field and laboratory experi- ganotin antifouling paint. U. S. Navy, Code 56YP, ments. J. Mar. Biol. Assoc. U. K. 67(3):525- Washington D. C. 544. 46. White, H. H. and M. A. Champ. 1983. The 33. Smith, B. S. 1980. The estuarine mud snail, great bioassay hoax, and alternatives. In: R. A. Nassarius obsoletus: Abnormalities in the re- Conway and W. P. Gulledge (Eds.), Hazardous and productive system. J. Moll. Stud. 46:247-256. Industrial Solid Waste Testing: Second Symposium. ASTM STP 805. American Society for Testing and 34. Smith, B. S. 1981. Reproductive anomalies in Materials. pp. 299-312. Stenoglossan snails related to pollution f rom 47. Cunningham, P. A. 1979. The use of bivalve marinas. J. Applied Toxic. l(l):15-21. molluscs in heavy metal pollution research. In: W. 35. Smith, B. S. 1981. Male characteristics on B. Vernberg, A. Calabrese, F. P. Thurberg, F. J. female mud snails caused by antifouling bottom Vernberg (Eds.), Marine Pollution: Functional paints. J. Applied Toxic. l(l):22-25. Responses. Academic Press, New York, pp. 183-221. 36. Smith, B. S. 1981. Tributyltin compounds 48. Laughlin, R. B. , Jr. , W. French and H. E. induce male characteristics on female mud snails Guard. 1986. Accumulation of bis(tributyltin) Nassarius obsoletus = Ilvanassa obsoleta. J. oxide by the marine mussel Mytilus edulis. Applied Toxic. 1(3):141-144. Environ. Sci. Tech. 20:884-890. 37. Grant, W. S. and F. M. Utter. 1988. Genetic 49. Yamada, J., K. Tatsuguchi and T. Watanabe. heterogeneity on different geographic scales in 1978. Uptake of tripropyltin chloride by Nucella lamellosa (Prosobranchia, Thaididae). Escherichia coli. Agric. Biol. Chem. 42:1867-1870. Malacologia 28(1-2):275-287. 50. Blair, W. R., G. J. Olson, F. E. Brinckman and 38. Dickie, L. M., P.R. Boudreau and K. R. W. P. Iverson. 1982. Accumulation and fate of tri- Freeman. 1984. Inf luences of stock and site on n-butyltin cation in estuarine bacteria. Microb. growth and mortality in the blue mussel (Mytilus Ecol. 8:241-251. edulis . Can. J. Fish. Aquat. Sci. 41:134-140. 51. Dooley, C. A. and V. Homer. 1983. Organotin 39. Mallet, A. L. and L. E. Haley. 1983. Growth compounds in the marine environment: Uptake and rate and survival in pure population matings and sorption behavior. Naval Ocean Systems Center crosses of the oyster Crassostrea virizinica Can. Technical Report #917. J. Fish. Aquat. Sci. 40:948-954. 52. Springborn Bionomics, Inc. 1984. Chronic 40. Lannan, J. E. 1980. Broodstock management of toxicity of tributyltin fluoride to the Pacific Crassostrea gigas I. Genetic and environmental oyster (Crassostrea gigas Thunberg). Report Number variation in survival in the larval rearing system. BF-84-8-66, Project Number 1462.8000, Springborn Aquaculture 21:323-336. Bionomics, Inc., Pensacola, Florida, August 1984. 41. Newkirk, G. F. and L. E. Haley. 1983. Selec- 53. Bayne, B. L. and R. J. Thompson. 1970. Some tion for growth rate in the European oyster, Ostrea physiological consequences of keeping Mytilus edulis: response of second generation groups. edulis in the laboratory. Helgo. Meeres. 20:526- Aquaculture 33:149-155. 552. 15%5 54. Bayne, B. L., P. A. Gabbott and J. Widdows. 67. Waldock, M. J. 1986. TBT in UK estuaries, 1975. Some effects of stress in the adult on the 1982-86. Evaluation of the environmental problem. eggs and larvae of Mytilus edulis L. J. Mar. Biol. In: Proceedings, oceans 1986 Conference, Wash- Assoc. U. K. 55:675-689. ington, D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, pp. 1324-1330. 55. Bayne, B. L. , D. L. Holland, M. N. Moore, D. M. Lowe and J. Widdows. 1978. Further studies on 68. Davies, I. M. , J. C. McKie and J. D. Paul. the effects of stress in the adult on the eggs of 1986. Accumulation of tin and tributyltin from Mytilus edulis-. J. Mar. Biol. Assoc. U. K. anti-fouling paint by cultivated scallops (Pecten 58:825-841. maximus and Pacific oysters (Crassostrea gigas Aquaculture 55(2):103-114. 56. Thain, J. E. and M. J. Waldock. 1985. The growth of bivalve spat exposed to organotin 69. Thain, J. E., M. J. Waldock and M. Helm. 1986. leachates from antifouling paints. ICES Paper CM The effect of tri-butyl-tin on the reproduction 1985/E:28, International Council for the Ex- of the oyster Ostrea edulis. ICES Paper CM ploration of the Sea, Copenhagen. 1986/E:14, International Council for the Explora- tion of the Sea, Copenhagen. 57. Laughlin, R. B., Jr., P. Pendoley and R. G. Gustafson. 1987. Sublethal effects of tributyltin 70. Salazar, M. H. and S. M. Salazar. 1987. TBT on the hard shell clam, Mercenaria mercenaria. In: effects on juvenile mussel growth. In: Pro- Proceedings, Oceans 1987 Conference, Halifax, Nova ceedings, Oceans 1987 Conference, Halifax, Nova Scotia, Canada, 28 Sept 1 Oct 1987, Organotin Sym- Scotia, Canada, 28 Sept I Oct 1987, Organotin Sym- posium, Vol. 4, pp. 1494-1498. posium. Vol. 4 pp. 1504-1510. 58. Valkirs, A. 0. , P. F. Seligman, and R. F. Lee. 71. Kiorboe, T. F. Mohlenberg and 0. Nohr. 1981. 1986. Butyltin partitioning in marine waters and Effect of suspended bottom material on growth and sediments. In: Proceedings, Oceans 1986 Con- energetics in Mytilus edulis. Mar. Biol. 61:283- ference, Washington D. C., 23-25 Sept 1986, Organo- 288. tin Symposium, Vol. 4, pp. 1165-1170. 72. Salazar, M. H. and S. M. Salazar. 1985. The 59. Waldock, M. J., J. E. Thain and M. E. Waite. effects of sediment on the survival of mysids 1987. The distribution and potential toxic effects exposed to organotins. Proceedings, 11th U.S. - of TBT in UK estuaries during 1986. Appl. Japan Experts Meeting on Management of Sediments Organomet. Chem. 1:287-301. Containing Toxic Substances, November 4-6, 1985, 60. Salazar, M. H. and S. M. Salazar. 1988. Tri- Seattle, Washington. pp. 176-192. butyltin and mussel growth in San Diego Bay. Pro- 73. Henderson, R. S. 1985. Effects of tri- ceedings, Oceans 188, This Conference. butyltin antifouling paint leachates on Pearl Harbor organisms. Naval Ocean Systems Center 61. Phillips, D. J. H. 1980. Qualitative Aquatic Technical Report #1079. Biological Indicators. Applied Science Publishers, Ltd., London. 488 pp. 74. 'Clavell, C., P. F. Seligman, P. M. Stang. 1986. Automated analysis of organotin compounds: a 62. George, S. G. and T. L. Coombs. 1977. The method for monitoring butyltins in the marine en- effects of chelating agents on the uptake and ac- vironment. In: Proceedings, Oceans 1986 Con- cumulation of cadmium by Mytilus edulis. Mar. ference, Washington D. C., 23-25 Sept 1986, Organo- Biol. 39:261-268. tin Symposium, Vol. 4, pp. 1152-1154. 63. Harrison, F. L. 1979. Effect of the physi- 75. Seligman, P. F. , A. 0. Valkirs and R. F. Lee. cochemical form of trace metals on their accumula- 1986. Degradation of tributyltin in marine and tion by bivalve molluscs. Am. Chem. Soc., Symp. estuarine waters. In: Proceedings, Oceans 1986 Ser. 93:611-634. Conference, Washington D. C., 23-25 Sept 1986, Organotin Symposium, Vol. 4, pp. 1189-1195. 64. Luoma, S. N. 1983. Bioavailability of trace metals to aquatic organisms a review. Science Tot. Environ. 28:1-22. 65. Martincic, D., H. W. Nurnberg and M. Branica. 1986. Bioaccumulation of heavy metals by bivalves from Limski Kanal (North Adriatic Sea). II. Copper distribution between oysters, Ostrea edulis and ambient water. Mar. Chem. 18:299-319. 66. Thain, J. E. , M. J. Waldock and M. E. Waite. 1987. Toxicity and degradation studies of tri- butyltin (TBT) and dibutyltin (DBT) in the aquatic environment. In: Proceedings, Oceans 1987 Confer- ence, Halifax, Nova Scotia, Canada, 28 Sept - 1 Oct 1987, Organotin Symposium, Vol. 4, pp. 1398- 1402. 1506 REDUCING PLASTIC POLLUTION IN THE MARINE ENVIRONMENT: THE U.S. COAST GUARD AND IMPLEMENTATION OF ANNEX V OF MARPOL 73/78 Lieutenant Commander Joel R. Whitehead, USCG Port Safety and Security Division U.S. Coast Guard Headquarters Washington, D.C. 20593 ABSTRACT year history of the U.S. Coast Guard. Yet this new task is closely aligned to other Coast Guard This paper explores the U.S. Coast Guard's ef- mission areas, for preservation of the marine forts to reduce ocean pollution by plastics and environment assures continuity of the very other ship-generated refuse. Implementation of medium on which all mariners rely. It was not Annex V of the International Convention for the until after World War 11, when the Coast Guard Prevention of Pollution by Ships, 1973 (MARPOL inherited the duties of the Bureau of Marine 73/78) will outlaw dumping of plastics at sea Inspection and Navigation, did it begin to enforce from vessels of signatory nations. It will also a panoply of arcane marine pollution laws. severely restrict vessels from discharging other Among these were the Refuse Act of 1899 and the types of garbage at sea and will require nations Oil Pollution Acts of 1924 and 1961. These laws which are party to the Convention to provide were almost impossible to enforce: either they adequate reception facilities for ships' refuse. authorized only meager fines which did not deter This paper discusses the problems caused by ship- violators or they were filled with loopholes. By generated garbage and traces the Coast Guard's the late 1960's, worldwide attention was fixed on role in reducing marine pollution through the need for effective marine pollution controls MARPOL 73/78 and particularly through Annex after events such as the catastrophic oil spill V following the grounding of the tankship Torrey Canyon. The growing importance of marine I. THE COAST GUARD AND MARINE environmental protection within the Coast Guard ENVIRONMENTAL PROTECTION was formally recognized in 1971 with the creation of the Office of Marine Environment and Marine environmental protection is a Systems. relatively new n-dssion in the almost two hundred United States Government work not protected by copyright 1507 II. MARPOL 73/78: THE INTERNATIONAL grounds. Only two days later, the Torrey FRAMEWORK Canyon's sister ship, Sansinena, explodedather berth in Los Angeles. These domestic tragedies The Torrey Canyon disaster, along with prompted the U.S government to threaten several other major environmental debacles, unilateral enforcement of strict safety standards convinced the Coast Guard leadership that an on all ships entering U.S. waters. Faced with enforceable international maritime pollution this threat, the maritime nations of the world treaty was essential to protect our coastal re- agreed to convene an international Conference on sources. The U.S. Coast Guard firmly believed Tanker Safety and Pollution Prevention in London that an international approach was required, in October 1977. The resulting agreement, which because domestic law could not effectively control the United States ratified, was the Protocol of the great number of foreign flag ships after they 1978 and it amended the 1973 Convention. left U.S. ports and ventured into our coastal MARPOL 73/78 was born. waters. By the early 1970's, almost 95% of MARPOL 73/78 addresses the problem of petroleum products entering the United States operational pollution from ships on a global scale were transported in foreign bottoms. Domestic and consists of five Annexes, each of which is political pressures, combined with a growing designed to combat a specific type of marine awareness on the part of the European shipping pollution. Annexes I and II address shipboard nations, finally led to acceptance of the In- pollution by oil and noxious liquid substances ternational Convention for the Prevention of (bulk liquid chemicals) and were the first to be Pollution by Ships, 1973 (also known as enforced internationally. Annexes 111, IV and V MARPOL, for Marine Pollution). are optional to Party states, and address ship- Although the 1973 Convention was a wa- generated pollution by packaged hazardous tershed event in international maritime co- materials, sewage and garbage. By terms of the operation, the United States believed the 1973 Convention, these annexes only enter into force Convention's ship construction standards were too one year from the date on which at least fifteen lax. Many other maritime nations agreed: by nations (which must cumulatively represent fifty 1976, only three others had accepted the 1973 per cent of world shipping tonnage) have ratified Convention. The Coast Guard's vision of a strong them. and effective international convention seemed far As an international legal instrument, from reality until a series of fifteen major marine MARPOL 73/78 is acknowledged as being very disasters struck within a three month period. In successful. Unlike its ineffective predecessors, December 1976, the tanker Argo Merchant MARPOL took into account the economics of foundered off the coast of Massachusetts and international shipping. The Convention had threatened the abundant George's Bank fishing some unusual, yet effective, terms for signatory 1508 nations. Nations which ratified the Convention agreed to enforce its provisions on all ships III. PLASTICS: ENVIRONMENTAL AND which entered their waters. This removed any ECONOMIC IMPACTS economic advantage to those shipping nations which failed to ratify the Convention and As early as the 1973 Convention, it was discou raged them from switching to flags of recognized that garbage from ships, especially convenience to avoid compliance. MAR- plastics and synthetic fishing nets, posed a POL 73/78 also required Party states to ensure multitude of problems. The unregulated dumping that shoreside reception facilities for certain of plastics at sea menaced marine life, was a MARPOL pollutants were available at ports and safety hazard to mariners and was an affront to terminals. This assured ship operators that our aesthetic values. Although the disposal of there would be a site to dispose of these wastes ship-generated garbage at sea has been the norm once the ship arrived in port and also served as for commercial and military vessels for an incentive for compliance. millennia, this practice has had severe Through the late 1970's, technical experts environmental and economic effects. A 1975 study. from the Coast Guard led U.S. delegations to the by the National Academy of Sciences indicated Marine Environmental Protection Committee of that 6.4 million metric tons of refuse were dis- the International Maritime Organization in charged annually into the oceans from London. At this specialized body of the United commercial merchant ships, naval vessels, Nations, the U.S. Coast Guard negotiated the fishing vessels, oil rigs and offshore platforms. ship construction standards and other technical About one million metric tons of this was details for the MARPOL annexes. In 1980, after estimated to be plastic debris. Over fourteen these details had been agreed upon million U.S recreational boats also compound the .internationally, the United States government problem.by depositing garbage into the inland formally enacted MARPOL 73/78 into domestic waters of the United States. The accumulation of law with passage of the "Act to Prevent Pollution plastics in the marine environment is also from Ships, 1980" (33 U.S.C. 1901 et By exacerbated by their persistency: non- 1987, Annexes I and 11 were in force in- biodegradable plastic. "six-pack" beverage yokes ternationally and had significantly reduced are believed to have a life at sea of several pollution by ships from oil and bulk liquid hundred years. chemicals. With the first two Annexes of The central environmental concerns sur- MARPOL73/78 in place, the stage was set for rounding the discharge of plastics at sea are implementing Annex V, which targeted op- entanglement and ingestion. Entanglement by erational pollution of the oceans by plastics and various marine species in synthetic fishing nets other ship-generated refuse. and monofilament fishing lines severely reduces 1509 their mobility to avoid predators and inhibits have attributed these to fouling hooks and lines their ability to feed. Juvenile sea creatures can on discarded or lost gill nets. According to one become entangled in small packing bands or six- small passenger boat operator association in New pack beverage yokes, which constrict as they England, approximately $50,000 worth of hooks, grow, resulting in strangulation, infected lesions synthetic lines and lures and one million dollars or starvation. Likewise, ingestion of persistent in operating expenses are lost every year in debris is a serious problem for all types of marine dealing with monofilament gill nets in the Gulf life. Turtles often ingest plastic sheeting of Maine alone. Lost fishing nets and traps are materials and bags mistaking them for jellyfish, also believed to deplete stocks of commercially one of their favorite delicacies. Birds ingest valuable species of fish, lobster and crabs. Recent small plastic resin pellets which block digestive research indicates that abandoned or lost gillnets tracts, impair the absorption of nutrients and can remaining in the ocean can continue to entangle cause bacterial infections. When plastics are fish for months, indiscriminately killing fish continually ingested, the eventual outcome is and other economically significant marine life. often death. The impact of persistent marine debris on the While data detailing the effects of persis- economies of coastal communities is also tent marine debris are becoming more available noteworthy. The Padre Island National for certain species, the aggregate effect of Seashore Park in Texas spends over $10,000 plastics on marine life is not fully understood. annually on beach cleaning efforts, primarily Populations of the Northern Fur Seal on the concentrating on a one half-mile stretch of beach Pribilof Islands and the Hawaiian Monk Seal frequented by the public. The beaches of Long have precipitously declined during the same Island, New York were closed to swimming in June timeframe in which the use of synthetic fishing 1976, as were many beaches between Virginia and nets became prevalent. Many scientists feel Massachusetts in the summers of 1987 and 1988, confident that the ubiquity of plastics in the when unusual amounts of floating debris washed marine environment is directly related to these as hore. The total economic impact on the tourist and other unexplained population declines. industry (including loss of revenues to restaurants, Ship-generated garbage also causes adverse pier fishing, bait and tackle shops and in economic effects through the depletion of fishery reduction of beach attendance) has been resources, vessel damage and degradation of our estimated in the hundreds of millions of dollars beaches. One of the most expensive economic im- annually. pacts is on the commercial fishing industry, both in terms of the loss of fishing gear and the depletion of fishery stocks. Fishermen have reported substantial losses of fishing gear and 1510 IV. POLITICAL PRESSURES AND A ternational agreement, and that Annex V of SOLUTION: ANNEX V OF MARPOL 73/78 MARPOL 73/78 was the best mechanism to accomplish this. Kime testified before Congress While environmental groups had been several times over the summer and fall of 1987. pressing the federal government for a legislative He persuaded them to apply the discharge answer to the hazards of entanglement and prohibitions of Annex V to the internal waters of ingestion of plastics, researchers around the the U.S. as well, thus providing an effective world convened at the 1984 "Workshop on the enforcement tool to replace the Refuse Act of Fate and Impact of Marine Debris." They 1899. The U.S. Senate gave its advise and presented research which confirmed the damage consent to ratify Annex V in November 1987 and being done to the marine environment and President Reagan signed the "Marine Plastic identified new areas for future research. Pollution Research and Control Act of 1987" into Within the United States, economic concerns law on December 29, 1987. With U.S. ratifica- prompted several powerful states to join in the tion of Annex V, the Convention's entry into force call for legislative action. Texas, New Jersey, requirements were fulfilled, and as a consequence, and other coastal states were fearful at the Annex V would be in effect throughout the world prospect of declining tourist revenues because of on December 31, 1988. unprecedented accumulations of beach litter. By the spring of 1987, over thirty U.S. Senators were V. ANNEX V OF MARPOL 73/78 so aroused that they wrote to the President of the United States outlining the severity of the Annex V of MARPOL 73/78 is designed to marine debris problem and asked him to take reduce the introduction of ship-generated garbage action. A federal Interagency Task Force on into the marine environment. It particularly aims Persistent Marine Debris was directed to study at preventing the discharge of plastics, including the dilemma and to report back to the President's synthetic fishing nets, into the marine Domestic Policy Council. Shortly afterwards, environment. Congressman Studds of Massachusetts introduced Annex V has three mandates. First, it legislation to prevent plastics from being prohibits discharge of any plastic materials at discharged into the ocean. Rear Admiral J. sea by ships. Second, it limits the locations at William Kime, Chief of the Coast Guard's sea where other types of non-plastic garbage can Office of Marine Safety, Security and be discharged. The Annex accomplishes this by Environmental Protection, strongly supported the setting minimum distances from land where legislation. He convinced Studds and other certain types of garbage can be discharged. For lawmakers that the only hope of an effective instance, large floatable items such as dunnage, solution to this problem was through an in- packing and lining materials may only be 1511 discharged outside twenty-five miles from land. VI. THE FUTURE Similarly, other refuse, such as crockery, food wastes, paper products, and bottles may only be just as the oil pollution prevention laws of discharged outside twelve miles from land, unless the early 1970's enforced a radical change with they have been "comminuted" or ground so that past practice, so will implementation of Annex V they can pass through a mesh screen with open- of MARPOL 73/78. Effective implementation of ings no greater than twenty-five millimeters the Annex will take several years before (about one inch). In this case, they may be dis- measurable results can be detected and will charged outside three miles from nearest land. require the cooperative efforts of government, Strict limitations are also imposed on oil rigs and mariners, private citizens, port and terminal platforms: they can only discharge food wastes operators and shipping related industries. when outside twelve miles from nearest land. Moreover, significant efforts will have to be "Special Areas" are also established in which undertaken in the areas of shipboard engineering, only comminuted food wastes may be discharged enforcement and education. Ships undergoing outside of twelve miles. Special Areas are construction and some existing ships will have to waters which must receive extra attention be fitted with compactors, incinerators, grinders because of peculiar oceanographic conditions or or other equipment to enable them to meet the because of their present polluted state. There are discharge requirements of Annex V. Cooperative now five Special Areas under Annex V: the enforcement mechanisms must also Mediterranean Sea, Black Sea, Baltic Sea, be put into place which can make the most ef- Persian Gulf and Red Sea. Because of the vast fective use of the combined resources of various .amount of ship and oil industry related garbage federal, state and local law enforcement agencies. accumulating in the Gulf of Mexico, the United Finally, a series of educational campaigns States Coast Guard is considering the designation must be targeted at commercial fishermen, of the Gulf as a new Special Area. merchant seamen and recreational boaters. The Finally, the Annex requires that party na- concern and strong sense of partnership displayed tions ensure that ports and terminals provide to date by a multitude of marine interests reception facilities for ships' garbage, so that strongly suggest that solving this serious envi- plastics and other garbage can be brought ashore ronmental dilemma can become a reality in the and that ships will not be delayed in ports. As near future. provided by the Act to Prevent Pollution from Ships, the U.S. Coast Guard is empowered to deny entry to a ship if the port or terminal it is bound for does not provide the required reception facilities. 1512 THE PRODUCTION OF MENHADEN SURIMI Anthony P. Bimbo Director of Applied Development Zapata Haynie Corporation Post Office Box 175 Reedville, VA 22539 USA INTRODUCTION The world's catch of fish and shellfish reached 92.2 million metric tons (mmt) in 1986 with about 30% or 28 mmt processed into fish meal and oil. Non-traditional sources of low cost high quality fish protein are needed to meet both the consumer and institutional markets worldwide. There is strong commercial interest in many countries to use industrial fish and under-utilized species for human consumption. Part of the basis for this interest is a growing movement world-wide, especially in developing countries, to limit the amount of fish that can be used in the production of fish meal and oil even though, at the present time, few commercially viable products for human consumption can be made from them. And finally, new uses for these under-utilized species may be vital to survival of the fish meal industry. The value of fish meal is in.direct relationship to soybean prices. Wide fluctuations in the price of commodities over the last few years has led-to financial problems in the industry. In fact, because of low prices, several countries now burn as much as 40% of their fish oil as fuel. THE MENHADEN FISHERY Various species of fish, found along the Atlantic, Gulf and Pacific Northwest/ Alaskan coasts of the U.S. may be used as raw material for surimi. Alaskan pollock has been the major resource used for surimi by the Japanese because of its abundance, low price and physical properties. Gulf croaker, Atlantic and Gulf menhaden, Atlantic mackerel, Pacific whiting and rod hake are among other species which offer possibilities. Of all of these, menhaden representing around 40% of the total U.S. landings offers not only unique possibilities for further development but also some major challenges to its utilization as food. Commercial landings by U.S. fishermen at ports in the 50 states were 3.1 mmt in 1987. This was composed of 1.8 mmt of edible fish and shellfish and 1.3 mmt of fish for reduction to meal and oil. Menhaden represented 92% (1.2 mmt) of the fish used for reduction. An 8 year history of the U.S. menhaden catch appears in Table 1. Menhaden are small, oily, herring-like fish similar in appearance to the alewife and shad. They are dark blue to brown in color with silver along the sides. They are distinguished from other fish of the herring family by the presence of a dark shoulder spot on both sides of the body. Menhaden are migratory fish. They appear in dense schools in the open waters of large bays and along the shores of the Atlantic and Gulf coasts of the U.S. The menhaden are pelagic and feed primarily on plankton which they convert to CH2585-8/88/0000- 1513 $1 @19.88 IEEE energy and store as fat. DEMONSTRATION PLANT During what has been described as the Age of Engineered Seafoods, 1975-1986, U.S. imports of surimi products rose from 5.9 to over 100 million pounds. The period 1985 to 1987 has seen a slower increase in consumption of these surimi based analogs, but now we are beginning to see interest from other segments of the food industry. Table 2 shows the phenomenal growth in the consumption of surimi products in the U.S. for the period 1975-1987. The period since 1985 is now being described as the Age of Engineered Foods. With the continued promotion of the health benefits of seafood, research is currently underway to engineer new food concepts produced from basic building blocks. Interest in omega 3 fatty acids and surimi from red fleshed, oily species as a possible source of these fatty acids has sparked new interest from the red meat industry as well. Table 3 shows the unique omega 3 fatty acid composition of both a high fat and low fat surimi produced from Atlantic menhaden. And so economic stresses on the menhaden fishing industry, increased market demands for traditional food fish and emphasis on full utilization of our fishery resources have directed attention to the higher value utilization of the menhaden resource. With the possibility for new market opportunities for surimi and fish mince, counterbalanced by the unique challenges associated with red fleshed species, the U.S. menhaden industry requested technical .assistance from both academic and government groups. In response, the Congress of the U.S. designated increased FY1985 funding for menhaden research and development in the conversion of whole menhaden into surimi and minced forms. The NMFS in cooperation with industry was directed to use the funds to achieve the following objectives: 1. determine the economic and technical feasibility of producing menhaden surimi, 2. produce developmental quantities of test menhaden surimi material for formulation and evaluation of end products by public and private sector food scientists and, 3. determine onboard and inplant procedures necessary to convert menhaden to human food products. On February 20, 1986, Zapata Haynie Corp., the largest processor of menhaden in the U.S., signed a contract with the National Marine Fisheries Service (NMFS) to conduct a two-year, $2 million pilot project designed to determine the most efficient way of making surimi from menhaden. Under the terms of the contract, the NMFS provided $1.4 million, with Zapata Haynie providing the additional $600,000. Since that time, Zapata Haynie's contribution has increased well beyond that initial committment. Although the original contract ended in February, 1988, an extension was arranged to permit continuation of the project during the 1988 fishing season and further evaluate improvements involving harvesting and process optimization. CONTRACT Under the terms of the contract, each party has specific responsibilites: 1514 Zapata Haynie must provide: 1. An appropriate physical site, 2. An appropriate food grade building, 3. Refrigerated holding facilities, 4. A plate freezer, 5. Frozen storage facilities, 6. Utilities, 7. Fresh food grade menhaden, 8. Disposal of all processing wastes. U.S. Government Funds Are Used For: 1. Purchasing, installing and maintaining fish handling and processing equipment, 2. Plant operating expenses, 3. Collection of data, 4. Storage and distribution of test products, 5. Preparation of reports, 6. Public access/technology transfer, 7. Overhead costs. The contract calls for a plant capable of producing at least I ton of finished surimi per 8 hour day. The contract spells out the steps needed to produce surimi, however it recognizes that the processing requirements to make a quality surimi from menhaden are not well defined and calls for a plant design that includes sufficient flexibility to allow the evaluation of a number of different processing options through experimental runs. It was expected that the processing options would lead to an alternate method of producing surimi from menhaden. Based on contract requirements, Zapata Haynie initially set up the traditional Japanese surimi production process as the baseline method for producing surimi from menhaden, this process is outlined in Figure 1. The steps involve removal of the head, tail and viscera followed by removal of the skin and large bones. The meat fraction is then washed 3 times until the majority of the fat and water soluble proteins are removed. Finally, the meat is refined through a I mm screen to remove fine pin bones, scales and some of the dark meat. The refined meat is then pressed to obtain the desired moisture content and finally mixed with cryoprotectants before freezing. Waste is generated at almost every step of the process. A fish meal plant or some process for treatment of the solid and liquid waste is essential to the success of such a plant. The plant was designed with enough flexibility to allow design changes for improvement of the product and process through a series of experimental runs. These runs were designed to determine the technical feasibility of the processing options, provide data on preliminary cost estimates and provide products and by-products for evaluation. Product quality and yields were major factors for the development of operating procedures for subsequent production runs. OVERVIEW OF THE PROCESS WITH MENHADEN The surimi process is essentially a series of processing steps designed to isolate the undenatured muscle protein of the fish from the other components of the flesh. How well these steps perform and at what cost in yield will determine the eventual economic and technical feasibility of producing surimi from red fleshed species such as menhaden. 1515 In general terms, protein will increase as we proceed through the process as more non-protein wastes are removed, but will drop in the surimi block because of the addition of 8% cryoprotectants, which appear as carbohydrate material. The fat drops as we proceed through the process eventually leveling off at 0.5-1.0% in the final surimi block. In addition to the isolation of muscle protein, the process removes pigment, blood, and some dark flesh. Color is measured on the Hunter Scale. The L value stands for lightness (white-black), A for redness (red-green), and B for yellowness (yellow-blue). The raw mince is very red, and the process tends to increase the lightness (L value) and decrease the red color (A value), but so far we have had very little effect on the yellow color (B value). In fact over several hundred observations, the B value (yellowness) has deviated very little regardless of process changes. Yellowness may be species specific. Because of large variations in fish size we experienced wide fluctuations in yield, in spite of on line fillet machine adjustments. Fish sorting before filleting should improve yield significantly. Yields can also be affected both positively and negatively by process equipment changes. During the first year, the technical program of experimental work was broken down into two major tasks. TASK I: HARVESTING AND PRE-PROCESS HANDLING. Bulk handling methods for food grade menhaden were developed. Menhaden vessels with some modifications were utilized. In addition to the menhaden vessels, arrangements were made with local trap net fishermen so that an alternate source of locally caught menhaden were available. In most instances, these fish are less than 6 hours old. Alternate methods of harvest included the small local snapper rigs or single vessel purse seiners. These normally fish in shallow waters and are available as an alternate source of fish. Since these vessels do not have refrigeration systems, chilled sea water and slush ice systems were tested. Since the plant is set back away from the water, fish were unloaded into insulated containers and trucked to the surimi plant. After delivery, the fish containers were drained, refilled with fresh refrigerated water and held in the plant's refrigerated storage facility. A major question was how long menhaden could be held in refrigerated water on shore without adversely affecting the surimi quality. Results from the 1987 fishing season indicated that in order for fish quality to be acceptable for next day processing to surimi 1. chilling must begin immediately after harvest and bring the fish temperature down to 50'F. within 4 hours 2. chilling must then continue to approximately 34'F. and held and 3. if the initial chilling takes longer than 4 hours, then the fish will appear good but will be unacceptable for surimi within 12 hours. 1516 TASK II. PROCESSING Processing covers the range from sorting through final storage of the finished surimi. CUTTING PROCESS Cutting is the first major processing step in the production of surimi. The cutting step is critical to the process, since it has a lasting effect on the economics of the overall process. Edible meat lost in this step cannot be recovered as surimi at the end of the process. Three different types of cuts were evaluated; heading and gutting, filleting, and scalping (lateral cut along the top of the backbone). In each case, cutting was done with commercial equipment, except that in the case of the scalp cut, it was done manually. The scalp cut simulated separation of red and white meat and should be a good predictor of what skinning might do. Table 4 shows the composition of red and white meat from 3 different batches of fish. Fat is significantly higher in the red meat while moisture is lower. We made several runs of scalped fish. Results indicated that a significant improvement in the Hunter A and B values could be obtained if we could mechanically separate white and dark meat. It was not possible to obtain a good headed and gutted fish with our cutting machine, so this option was dropped. MINCING The mincing process (deboning) is designed to separate fish flesh from bones, skin and scales. Different sized screens in the deboner and different belt tensions have an effect on yield and quality of the mince. Different cuts of fish will behave differently in the deboner, and will also give different yields depending upon cut and screen size. Our Bibun deboner is equipped with a 5 mm screen. We looked at several variations in the mincing operation. With a lighter belt tension color improved slightly, protein and fat were significantly improved but unfortunately, surimi yields were significantly lower. And while orienting the fillet meat side down lowered the A and B values, it also lowered the L value. With the plant configuration, orienting the fillet meat side down was not possible however this could be done under full scale commercial conditions. WASHING Washing of the mince is another critical step in the production of surimi. Washing removes water soluble proteins and other undesireable constituents from the mince. The initial process called for a batch wash system: 3 cycles, 5 to 1 ratio water to fish with a dewatering screen at the end. We looked at ratios of 3, 5 and 10 to 1 and 4 wash steps at a 3 to I ratio. There was very little change in color between 5:1 and 10:1 but both were improved over the 3:1 ratio. Four batch wash steps at 3:1 were similar to 3 steps at a 5:1 ratio. 1517 In late July 1987 we began evaluating the Alfa Laval inline wash system. This consists of a pump, a variable length of stainless steel pipe made up into loops with inline static mixers, and an Alfa Laval NX 414 decanter. Deboned, refined, unwashed mince is mixed with water and pumped through the wash loops to the decanter where the water and solids are separated. The wet solids are then pressed before mixing with cryoprotectants. While there doesn't appear to be much of an improvement in the quality of the surimi, yields are improved, water usage is cut and there is a definite savings in labor. REFINING Zapata Haynie decided on a refining step instead of the traditional straining step for removal of pin bones, scales, dark meat and skin particles. The refiner handles wet meat, so less heat is generated and bench scale research at North Carolina State Univ. indicated that it might be possible to separate red meat from white meat. We evaluated several different machines as refiners. The Bibun refiner was difficult to control, required additional water and did not have sufficient capacity to meet the plant's needs. We evaluated a Beehive, Yield Master and a strainer as substitutes for the Bibun. The Beehive gave the best combination of results, had a greater capacity and could handle a dryer product. We also looked at placement of the refiner in the process either before or after the wash cycle and found that before the wash cycle gave the best operational results. At this point, pin bones and scales were completely removed and this helped with the rest of the process. We were not successful in separating red meat from white meat in the refiner. DEHYDRATION OR PRESSING Since the purpose of the dehydration step is to r'emove water and maintain a product at 75-80% moisture, dehydration studies consisted of achieving this moisture level under the various test conditions. We also evaluated single and double pressing. There was very little difference in product color between the two methods however double pressing did improve protein and lower fat. MIXING In addition to the normal addition of cryoprotectants i.e. sugar, sorbitol and sodium tripolyphosphate, we evaluated different levels and types of cryoprotectants. Because of the unique challenges that are associated with surimi from menhaden, Zapata formed a technical advisory committee consisting of researchers from several universities, government and industry technical staff, as well as Zapata technical staff. This committee has, among other things, advised the surimi plant on experiments, test runs and needed data. In addition, several of the members of this committee have supplemental research grants designed to complement the data generated by the surimi 1518 plant. These efforts show a unique cooperation between private industry, the Federal Government, the Sea Grant College Program and the Fisheries Development Foundations. 1986 AND 1987 FISHING SEASONS Plant construction was completed on October 6, 1986. A full scale production run was made on November 19, 1986. The run demonstrated the 1 ton per day capacity of the plant, as defined in the contract. The 1986 fis hing season ended on November 21. The 1986 winter work was done with frozen fillets and provided us with the opportunity to test the technical structure of the program as well as the operation of-the various pieces of equipment. Production in the plant actually began in May 1987. A series of bench scale experiments defined the initial benchmark for menhaden surimi. In early June 1987 we made our first series of extended production runs.-These were followed by a series of experimental runs to determine the optimum operating parameters for deboning as well as washing. We also experimented with skinning machines during 1987 but were unable to find a satisfactory machine for handling the menhaden. Since separation of the red meat will be critical to producing a lighter colored, lower fat product, we continue to explore this type of equipment. The project progressed rapidly during the 1987 fishing season but as with most projects of this type, it raised many questions which formed the experimental basis for the 1988 fishing season. 1. A modified process utilizing the in-line wash system with pre-refining although defined still must demonstrate a product of consistent quality over the wide range of fish composition available during the season. A series of continuous production runs are needed. 2. Based on these runs a set of possible specifications must be developed for menhaden surimi. These specifications must be reproducible in the plant. 3. Fat in surimi can either be an advantage or disadvantage. If the block can be stabilized then menhaden surimi will be able to offer not only funtionality but also a source of omega-3 fatty acids. 4. If the residual fat in the surimi presents stability problems, then additional research will be needed to release the fat from the meat since it appears to be bound to the proteins. 5. The relationship of the 3 colorvalues for surimi must be evaluated. Higher L values seem to be associated with higher A and B values. Skinning may offer a solution to this problem. 6. Finally, the yield must be optimized without adversely affecting quality. 1988 FISHING SEASON 1519 Based on the 1987 experimental results a modified process for the production of surimi from menhaden was developed. A flow diagram of this new modified process appears in Figure 2. The 1988 fishing season began in May of this year. A number of modifications were made in the plant to improve fish handling and processing. On the harvesting side, we have modified a menhaden vessel to deliver food grade Atlantic menhaden to the surimi plant. The surimi plant has been modified to handle the bulk unloading of menhaden from the vessels directly into 5500 gallon stainless steel insulated tankwagons. These vehicles are then used to transport the fish in slush ice to the plant and to act as storage containers for supplying fish to the plant. The tank wagons are loaded from the vessel and unloaded at the plant with wet fish pumps. The pumps operate on a pulsed cycle, vacuum unloading the fish in a mass of water to a sealed tank and then pumping the fish and water under pressure to the tankwagon. The fish enter the plant pass over a dewatering screen and onto a weigh belt. The only other modifications to the original process was the inline washing with a pre-refining step. The only mechanical sorter available for trial was a Patco 3500. It separated fish according to body thickness at the widest point, just behind the gills. While a favorable correlation exists between fish size and this cross dimension, the machine was not capable of handling the volume needed by the process. The menhaden had a tendency to wedge in the channels, blocking the machine and restricting fish flow. We are still looking for a sorting machine. We recently purchased a Trio Skinning machine and are evaluating different deboner screen sizes to see if the refining step can be dropped. We anticipate further modifications to the process after the 1988 fishing season. Research on menhaden surimi is also underway at a number of Universities in the U.S.. North Carolina State Univ., which did much of the original bench scale work on menhaden surimi along with researchers at the NMFS Laboratory in Charleston, SC, is currently investigating the use of a flavor mask developed by the Ogawa Company to extend the use of menhaden surimi in analog products. They are also evaluating the presence of a protease that affects the functionality of surimi in certain foods. In cooperation with the National Fisheries Institute they have helped standardize the methods of analysis to be used for testing surimi. Researchers at Cornell Univ. are evaluating surimi and intermediate minces in a number of products and have evaluated several anti-oxidants that might be used to stabilize the fat. Researchers at the Univ. of Massachusetts Seafood Lab in Gloucester, Mass. are evaluating the basic lipid chemistry of menhaden surimi. Virginia Tech. Univ. scientists are evaluating a number of convenience and fast food type products that can use menhaden surimi. Univ. of Georgia scientists have produced pasta type products with menhaden. Rhode Island Univ. researchers are evaluating process modifications and the North Carolina State Univ. Seafood Laboratory is evaluating ultrafiltration as a means of recovering valuable proteins from the waste waters. 1520 In addition to the University research, a number of food companies both in. the analog business and in other areas of the food industry have been evaluating the product for some time. The recent precedent setting approval of surimi for use in a cured meat blend by the U.S. Department of Agriculture will open up a new arena of opportunity for surimi producers. The first product approved was for "Spicey Bites" breaded pork nuggets with 15% surimi produced by the Alaska Fisheries Development Foundation. A second product "Southwest-Style Golden Morsels" developed by Manning, Batson and Associates of Phoenix, Arizona was also approved. According to the company, the product is low in fat, adequate for either commercial market or restricted diet programs and is less expensive to manufacture than similar all-meat products. They have also successfully used menhaden surimi in some of the test formulations. In the future we will see snack foods, drinks, luncheon meats, hot dogs and many other foods which will utilize the unique functional properties of surimi. The Age of Engineered Foods is now. BIBLIOGRAPHY 1. Sonu, Sunee C. "Surimi", NOAA Technical Memorandum, NMFS, January 1986, NOAA-TM-NMFS-SWR-013, 2. Anonymous, Surimi: The Explosive Blended Seafood Market, Cos Cob, Connecticut, 06807-0398, Seafood Management Corporation, 1984. 3. Suzuki, Taneiko, Fish and Krill Protein, Processing Technology, London, Applied Science Publishers, 1981. 4. Dreosti, G.M., and T. J. Billy, "Potential Food Products From Industrial Fish", IAFMM Technical Report No. 1, January 1983. 5. Anonymous, "The Surimi Explosion", Seafood Business, Vol 5, No 3, May/June 1986. 6. Anonymous, "Surimi Sensation", Seafood International, March 1986, pg 25. 7. Vondruska, John, "Blue Crab Markets and Analog Products", Paper presented to the National Blue Crab Industry Association's Annual meeting, Feb. 26-28, 1986. 8. Vondruska, John, "Market Trends and Outlook for Surimi-Based Foods", Paper presented at the International Symposium on Engineered Seafoods, Seattle, Washington, Nov. 19-21, 1985. 9. Lanier, Tyre C., Roy E. Martin, and Anthony P. Bimbo, "Nutritional Implications of Increased Consumption of Engineered Seafoods", Food Technology 42(5); 162, 1988. 10. Vondruska, John, W. S. Otwell, and R. E. Martin, "Seafood Consumption, 1521 Availability and Quality", Food Technology 42(5); 168, 1988. 11. Bimbo, Anthony P. "U.S. Menhaden Surimi Production: First Interim Report", Paper presented at the 26th Annual Meeting of the International Association of Fish Meal Manufacturers, October 8, 1986. 12. "Proceedings of the International Symposium on Engineered Seafood Including Surimi", Edited by Roy E. Martin, National Fisheries Institute, Washington, D.C., November 19-21, 1985. 13. Anonymous, "Alfa Laval Decanter-Based Surimi Production", Bulletin PB 40984E/8610, Alfa Laval Fish and Meat Engineering A/S. 14. Andres, Cal, "Surimi Poised For Next Step Forward", Food Processing, August 1987, pg. 28. 15. Martin, Roy E., "Surimi Products A Review", Paper presented at the 194th National Meeting of the American Chemical Society, September 2, 1987. 16. Bimbo, Anthony P., "U.S. Menhaden Surimi Production: Second Interim Report", Paper presented at the 27th Annual Meeting of the International Association of Fish Meal Manufacturers, October 7, 1987. 17. Bimbo, Anthony P., et. al. "U.S. Menhaden Surimi Production: A Progress Report", Paper presented at the National Technical Conference: Fatty Fish Utilization: Upgrading from Food to Feed, Raleigh, NC, December 10, 1987. 18. Moore, Raymond E., et. al. "Harvesting and Pre-Process Handling of Food Grade Menhaden- An Interim Report", Paper presented at the National Technical Conference: Fatty Fish Utilization: Upgrading from Food to Feed, Raleigh, NC, December 10, 1987. 19. Martin, Roy E., "Surimi", Proceedings of the IFT Short Course on Seafood Technology- Preparing for Future Opportunities, New Orleans, LA, June 19, 1988. 20. Bimbo, Anthony P., "Fish Oils", Proceedings of the IFT Short Course on Seafood Technology- Preparing for Future Opportunities, New Orleans, LA, June 19, 1988. 1522 TABLE 1. US AVAILABILITY OF FISH, SHELLFISH AND MENHADEN, 1980-1987, IN 1000 METRIC TONS. ----------------------------------------------------------------------------------------------- FISH AND ATLANTIC GULF TOTAL SHELLFISH MENHADEN MENHADEN MENHADEN ----------------------------------------------------------------------------------------------- 1980 2940 430 702 1132 1981 2711 402 553 955 1982 2888 400 854 1255 1983 2921 420 924 1344 CIO 1984 2920 329 983 1311 1985 2839 359 884 1243 1986 2736 256 829 1085 1987 3128 323 907 1230 ----------------------------------------------------------------------------------------------- SOURCE: FISHERIES OF THE UNITED STATES, 1980-1987. TABLE 2. ESTIMATED US SUPPLY OF SURIMI PRODUCTS ------------------------------------------------ YEAR CONSUMPTION ------------------------------------------------ MILLION POUNDS COOKED EDIBLE WEIGHT ------------------------------------------------ 1975 5.6 1976 4.7 1977 5.8 1978 4.8 1979 4.9 1980 6.1 1981 9.1 1982 20.6 1983 39.9 1984 69 1985 90.6 1986 99 1987 110.2 ------------------------------------------------ SOURCE: VONDRUSKA, J. SYMPOSIUM ON MARKETS FOR SEAFOOD AND AQUACULTURAL PRODUCTS, 1987.. 1524 TABLE 3. N-3 FATTY ACID COMPOSITION OF MENHADEN SURIMI. ---------------------------------------------------------------- FATTY ACID LOW RANGE HIGH RANGE ---------------------------------------------------------------- EXPRESSED IN MG PER 100 GRAMS OF RAW EDIBLE PRODUCT. ---------------------------------------------------------------- 16:3N-3 3.5 11.4 16:4N-3 5.1 16.4 18:3N-3 6.1 19.6 18:4N-3 18.8 60.6 20:3N-3 0.93 3 20:4N-3 8.74 28.2 20:5N-3 57.85 186.6 21:5N-3 3.41 11 22:5N-3 14.14 45.6 22:6N-3 106.08 342.2 ---------------------------------------------------------------- SOURCE: ZAPATA HAYNIE CORP. INTERNAL DATA. 1525 TABLE 4. PROXIMATE COMPOSITION OF MENHADEN RED AND WHITE MEAT. --------------------------------------------------------------- PROTEIN FAT ASH MOISTURE --------------------------------------------------------------- BATCH A RED MEAT 16.43 18.47 0.77 63.90 BATCH A WHITE MEAT 17.97 10.56 1.30 70.05 BATCH B RED MEAT 18.12 7.14 1.79 72.31 BATCH B WHITE MEAT 18.70 3.46 1.62 76.68 BATCH C RED MEAT 16.82 15.27 1.48 65.83 BATCH C WHITE MEAT 18.80 8.43 1.42 72.06 --------------------------------------------------------------- SOURCE: ZAPATA HAYNIE CORP. INTERNAL DATA. FIGURE I ORIGINAL PROCESS WHOLE REFRIGERATED FISH FILLETING --------------- WASTE DEBONING ---------------- WASTE 3 STEP WASHING ---------- WASTE DEWATERING -------------- WASTE REFINING ---------------- WASTE DEHYDRATION ------------- WASTE MIXING EXTRUSION FREEZING STORAGE 1527 FIGURE 2 MODIFIED PROCESS WHOLE REFRIGERATED FISH FILLETING --------------- WASTE DEBONING ---------------- WASTE REFINING ---------------- WASTE INLINE WASHING DECANTING --------------- WASTE DEHYDRATION ------------- WASTE MIXING EXTRUSION FREEZING STORAGE 1528 AN EXPERIMENTAL ESTUARINE SALINITY GRADIENT Eric Klos Marine Ecosystems Research Laboratory Graduate School of Oceanography, University of Rhode Island, Narragansett, RI 02882-1197 Abstract investigation of chronic, low-level pollution effects. For the last decade ' we3have operated 14 The Marine Ecosystem Research laboratory large marine mesocosms (13 m ) to model a well- (MERL) is a mesocosm facility at the University mixed, high salinity (30 */..) coastal system. of Rhode Island'i Graduate School of Oceano- Mid-Narragansett Bay, R.I. has been the natural graphy. A 560 m building serves as a support reference station. These systems, however, are laboratory and operations center for large ex- capable of being adapted to simulate other natu- perimental ecosystems. Fourteen cylindrical ral situations. Recent interest in the behavior tanks 1.8 m in diameter and 5.5 m in depth are of low salinity estuaries has led us to develop arranged along an access deck outside and adja- a salinity gradient model suitable for biogeo- cent to the building. The facility has labor- chemical experiments. atories for biology, organic, trace metal and We adapted 4 mesocosm tanks to create a sal- nutrient chemistry, and radioisotope tracer inity gradient. River water was introduced at studies. 3 one end of the system and seawater (30 */.o) was The mesocosms (Figure 1) contain 13 m of introduced at the other. Dilute salinities of Narragansett Bay water with associated pelagic 5 -/oo, 10 -/o@ and 30 */.. were maintained community. Benthic communities from Narragan- between these two extremes. Air driven dis- sett Bay are removed, as undisturbed as possi- placement pumps that are non-destructive to ble, with a box core, loaded in trays and placed plankton were designed to transfer water "up" in the bottom of each mesocosm. Physical para- and "down" the gradient and calibrated to main- meters such as mixing, water turnover and temp- tain a 27 day turnover time in each tank. . In erature can be scaled to simulate natural con- this design the same equipment allows a wide ditions. MERL has typically operated these range of operating conditions. parameters to simulate the "average" conditions of mid-Narragansett Bay. The water column is INTRODUCTION stirred to simulate wind and tidal mixing (Nixon et al., 1980). Mixing is by a vertically rotat- Marine mesocosms are more than research ing plunger, 60 cm dia. with a 60 cm travel, run aquaria; they are experimental ecosystems at 5 rpm, 2 h on : 4 h off. Seawater from created to behave as analogs of the larger nat- Narragansett Bay is pumped to a head tank by ural systems from which they are derived. By diaphragm pumps which are nondestructive to design they are interactive systems that contain plankton. The head tank periodically (120 feedback networks found within ecosystems and 1/tank, 4 times a day) delivers seawater by are thus self-sustaining and ecologically dur- gravity feed through a manifold to the mesocosms able. Mesocosms serve as unique environmental providing the tanks with a 27-day turnover time; research tools and are an important bridge the average turnover time of Narragansett Bay between field and laboratory studies. They are (Pilson, 1985). Temperature within the tanks is- cost effective when compared to the cost of controlled and maintained at the average daily obtaining a similar data set in the field. Of ambient bay temperature with heat exhangers more significance, they can be operated at a placed in each tank. An IBM-PC and compatable level of experimental control and replication digital/analog input/output interface device is that is not possible in the field. Field ex- used to control many of these operations. periments in the marine environment are confused Over the last decade numerous experiments by advection and the difficulty in locating have been completed at MERL with this "average" valid control sites. Bioassay type laboratory model of Narragansasett Bay. Long-term experi- experiments while useful in their ability to ments of a duration greater than a year have efficiently project the response of a species to been conducted to observe ecosystems under the a perturbant, are not capable of predicting a influence of hydrocarbons, enhanced nutrients, whole system effect. Mesocosms can be operated nutrient ratios, sewage sludge and effluent, and for relatively long periods of time over sea- polluted sediments. Many short-term experiments sonal and annual cycles and because of this have been run either within the context of long- feature ecosystem responses such as adaptation, term experiments or as separate studies. These and succession of communities can be observed. experiments have been diverse and include radio- This capacity has made mesocosms ideal for isotope studies and specific organic, hydro- CH2585-8/88/0000. 1529 $1 @1988 IEEE tained as in other experiments and the tanks would be cleaned on a regular schedule. The only deviations from our normal operations was the introduction of river water with seawater to C produce dilute salinities and our final goal of A connecting tanks together so that water could move "up" and "down" the-gradient. Four condi- CONTROL tions were selected: freshwater (0 */..), 5 A SEAWATER TURNOVER O/Oo and 10 0/00 low salinities in the middle WATER COLU@N 120 L/FEEO and seawater (30 */00). Tank turnover was kept 480 L/OAY the same as previous experiments - 27 days. The Pawcatuck River in western RI (Figure 3) was selected as the reference. The estuarine a MIXING part of the river is about 8 km long - from 2 @ ON - 4 @ OFF 5 RPIA Westerly to Watch Hill - and opens into Little Narragansett Bay to the south. Salinities in the estuary vary primarily in relation to the C TEMPERATURE amount of freshwater input but typically there MIXING HEAT EXCHANGER ONIOFF is a clear salinity gradient in the estuary from I 1C OF BAY TEMPERATURE entirely fresh at Westerly to around 30 O/G. at Watch Hill. There are two sewage treatment plants along the river about 2 km south of Wes- terly on opposite banks serving a modest popu- Figure 1. A MERL Mesocom lation of around 10,000. Although this input represents a measurable nutrient enrichment carbon and metal spikes. On several occasions (Nowicki, unpublished data) there is little natural conditions other than Narragansett Bay industrial discharge and water quality is satis- have been investigated. In one experiment sev- factory (Sisson, RI Dept. Environmental Manage- eral tanks were stratified to simulate an off- ment, personal communication). shore ecosystem with a strong pycnocline (Fox and Kester, 1985). Recent interest in ESTUARY low-salinity estuaries led us to attempt construction of an experimental estuarine RIVER OPEN salinity gradient making use of our existing SEA mesocosms. This work was done in the latter F1W part of 1987 as a feasibility study and initial examination of contained ecosystems at low 0 5 salinity. 10i%. Design Concept and Goal 30 A simplified and conceptual view of an estuary was generated initially to set design ANIN r, 1141"5111 P P criteria for the experimental salinity gradient mesocosm (Figure 2a). This view is not unlike those represented in many textbooks. Seawater enters the estuary at the bottom by tidal forc- ing and is mixed and diluted with freshwater EXIT entering from a river at the opposite end. There is a net flow of freshwater out of the estuary observed as lower salinity. Mixing 1XIIII. -EN within the estuary is both horizontal and verti- cal and the rate of mixing is a function of 8ENTH0j$ tidal currents and wind shear. 0 6 10 30 96 This view of an estuary was used to define an overall design goal making use of available tanks, equipment and resources (Figure 2b). We Figure 2. decided that this initial experiment would be (a) Conceptual view of an estuary. vertically well mixed - the gradient would be (b) Experimental estuarine salinity gradient. strictly horizontal - and mixing would be the Freshwater and seawater are introduced at either same magnitude as in most other MERL experi- end. Mixing is a f tion of tides and wind in ments. A benthic community would be included. the estuary, and transfer pumps and mechanical Tank temperature would be monitored and main- mixing in the mesocom. __ e - 1530 The Pawcatuck River was examined on several of laminar flow maintained 3in the feed system occasions to locate suitable sediment sites and (Reynolds No. Ca. 2 x 10 . A4though fully measure salinity and other water quality vari- turbulent flow (Reynolds No. > 10 ) may be more ables. (A larger unpublished salinity data set desirable for flow control, we do not feel this over I year was made available to us by Sisson, is good for plankton. RI-DEM.) Sediments in the estuary are most 'ly fine silt-sand. Low salinity (3-20 0/..) and high salinity (22-30 O/oo) stations (Figure 3) MA were selected as the sediment sources for the experimental benthic communities. WESTERLY METHODS Experiment Initiation In early September, 1987, four sediment trays were loaded by the standard MERL method with sediments from the Pawcatuck River - 3 from the low and 1 from the high salinity station., These were transported to MERL and each lowered to the DEPTH CONTOURB bottom of an empty tank. Seawater and river IN METERS water were added to each tank in the proportion required to bring each tank to its set salinity. Mixers and heat exchangers were placed in the tanks and water feed systems were turned on. PAWCATUCK RIVER After a one week "equilibration" period routine ESTUARY monitoring and sampling commenced. Mesocosm Design Mesocosm design goal was accomplished in two phases. In the first phase tanks were operated ------ independently. Freshwater and seawater end members were introduced into each tank at the proper rate to maintain set salinity. Our in- tent was to first observe the behavior of low salinity ecosystems run under the same protocol as other MERL experiments with salinity as the LITTLE NARRAGANSETT 171* 5o'. only variable. This allowed a more straight- SAY WATCH HILL forward analysis of their condition. In the second phase, four independent tank systems were converted into a single estuarine "system" with four compartments. Freshwater and seawater were introduced only at the ends of the model and Figure 3. The Pawcaturk River estuary. transfer pumps were installed to move tank water up" and "down" the gradient. The seawater feed manifold in use at MERL is River water was delivered to us periodically constructed of PVC and designed to have an equal about every 2 or 3 weeks - via tank truck path from the head tank inflow to the outlet for (E.H. Benn & Sons) in ca. 20 m 3 volume shipments each-tank. This feature, and equal length lines and was pumped into two mesocosm tanks set aside to each tank provide an even distribution of for use as a freshwater reservoir. plankton to the tanks at equal flow rates. When Phase I design required a new method of con- not in the process of feeding tanks the manifold trolling the feed into the tanks; both fresh- and feed lines are drained to prevent freezing water and seawater. Experience at MERL has and fouling. In previous experiments all 1 tanks taught us that from a single source equal flow received the same volume input (480 1 d ) and rates among tanks is the best way of insuring an only I control point was necessary. This design equal split of plankton. (Different flow rates was expanded for this experiment with installa- may transport different plankton concentrations tion of 2.5 cm dia. spring return pneumatic (per unit volume) from a single source.) There actuator valves (Chemtrol, mfg.) to allow indi- is subtlety to this problem: just having equal vidual on:off control for each of the salinity flow rates to tanks Is insufficient. The split- gradient tanks. This particular valve is inter- ting manifold must be constructed with each path changeable with the standard manual ball valves geometrically equal (Vargo, unpublished data). used with other tanks and has the same pressure This situation is in part due to the condition drop. 1531 Actuation with air at 80 psig turns the pneu- our own design. A prototype was fabricated and matic valve against a spring. When pressure is tested. Concentrated zooplankton samples dis- removed the spring returns the valve to its charged'through this pump were indistinguishable initial state. Valves were configured to be from control samples (Sullivan, personal commun- normally closed. Air pressure was supplied from ication). Other criteria were satisfied and it the MERL compressor (joy, mfg. : 20 hp., 80 was relatively inexpensive to fabricate in our SCFM) and was directed either to the pneumatic shop. valve or vented with .64 cm dia. 3-way solenoid The submersible transfer pump works by dis- valves (ASCO, mfg.). The action of the solenoid placing water flooded into the pump chamber with valve was controlled with a relay that received air. PVC check valves direct the flow of water its signal from a computer driven digital output out of the pump and into the next tank. A 2 s module. on : 2 s off cycle of air was supplied to each The freshwater manifold incorporated the same pump with 3-way solenoid valves of the same type general design principles as the seawater sys- used with the seawater manifold. Five pumps tem. One Important difference: 1.3 dia. 2-way were independently controlled to attain the solenoid valves (normally closed) were used circulation shown in Figure 2b. Each tank ex- directly to control water flow. This was done, cept the 0 0/.. treatment received water from in part, to compare the performance of solenoid two sources. vs. pneumatic valves in the same application. Determination of flow rate proportions that As with seawater flow, freshwater valves were maintain salinity and supply a 27 d turnover directed by computer. time for each tank is more elaborate for an The total amount of water fed to each tank to interconnected system than a single compartment provide a 27 d turnover time is 480 1 per day, model. It was necessary to have an overflow or 120 1 each feed cycle. An equal amount of from each tank. The volume fed to a tank from tank water drains through an overflow pipe. The one source during each feed was calculated as amount of freshwater and seawater required each follows: feed to maintain set salinities can be deter- mined by a simple proportioning of known and Se S2 1 selected salinity times the total flow rate: J, X 120 1 feed (2) S 2 JS = Se x 120 1 feed-1 (1) where: a il=Volume flow rate, of first source where: Se=Tank Salinity, 0/.o J = volume flow rate of seawater S2=Salinity of second sourcet 1/.0 s Se= Tank salinity, */oo S1=Salinity of first source, Ss= Salinity of seawater, 30 0/00 Flo rates from transfer pumps averaged 10.2 1 min Seawater and fres water end members Flow rates through th seawater system aver- q aged 16.5 1 min- 1 tank- f . Flow rates through averaged 16.5 and 20.2 1 min- respectively. A the fr.1shwater system were slightly lower: 14.5 new schedule for timing feed was programmed. - 1 All feeds were added simultaneously 4 times a 1 min tank . These flow rates were trans day at the same time as before. lated into a feed system "ON" schedule by cal- culating the time required to supply the neces- Analytic Methods sary volume to each tank each feed cycle. Tanks were fed 4 times each day; at 0300, 0900, 1500 Salinity was monitored daily with a refracto- and 2100. Both seawater and freshwater were meter with a calibrated precision of + .5 O/oo. added at the same time. Samples were taken after feeds and duilng mixing Phase II - the interconnection of 4 tanks to ensure that they were representative. Sev- into a single estuarine system - involved the eral times a Beckman induction salinometer was use of pumps to transfer water between tanks. used to look for vertical salinity structure In this phase salinity within a tank would be during and following feeds. This salinometer the result of input from adjacent tanks. Sea- had a precision of + .05 */o. and was addition- water and freshwater end members were added only ally useful as a c5eck of refractometer accur- at the end of the experiment (Figure 2b). acy. On two occasions high precision (+ .005 Several pump designs were examined for their ability to transfer zooplankton undamaged. measurements with an Autosal were made to Other desirable qualities were that the pump be study small scale vertical structure during feed non-contaminating, self-priming, draw minimal with transfer pumps. Bay salinity was measured power and require minimal connecting plumbing. routinely and at times, over a tidal cycle. We concluded that our best choice was a pump of Flow rates were measured volumetrically about 1532 every week. Measurement prceision was + .05 1 communities were impoverished (a condition min Flow rates among tanks through the observed in nature). Less macrofauna. may in freshwate i and seawater manifolds varied up to + part explain the greater abundance of zoo- 1 1 min f rom the mean. Flow rates throufh plankton in the low treatments. The greater transfer pumps varied up to + .5 1 min . filter feeding community in the high salinity Timing of individual feeds allow@_d us to correct tank may have removed phytoplankton at the for individual differences. Total system preci- expense of zooplankton. The ratio of dissolved sion was estimated to be 5 to 8%. inorganic nitrogen:phosphorus (DIN/DIP) sug- Ecosystem behavior was monitored by other gested phosphorus limitation of phytoplankton in members of the MERL staff. The following para- the 0 O/oop 5 0/000 and 10 0/00 tanks and meters were measured weekly in the tanks as part nitrogen limitation in the 30 */0. tank (Suttle of the ongoing MERL mesocosm sampling program: and Harrison, 1988). While the mechanism of 1. Phytoplankton biomass (as chlorophyll) this shift in limiting factors is not fully 2. Zooplankton abundance understood, it appears in nature and was present 3. Total system metabolism in the mesocosm. 4. Dissolved inorganic nitrogen, phosphorus and silicate Table 1. Oandition of the experimental ecosystem 5. Particulate nitrogen and carbon 6. Dissolved and particulate iron. Methods for these analyses are described in 0*/o. 5o/.. 10*/o. 30o/o. Lambert and Oviatt, 1986. RESULTS Production (02 M-2d7l) 1.0 1.0 1.5 1.6 Salinity varied + 1 */., in the mesocosms Respiration (g02m_2d_1) 0.5 0.3 0.6 1.2 over the duration oCthe experiment (102 days). P/R 2.0 3.3 2.5 1.3 What changes in salinity did occur were gradual over a period of days or weeks. Such drift was Zooplankton (no. 1-1) 17.0 40.0 22.5 4.0 corrected with minor adjustments in the feed schedule and the system allowed to drift back to its set salinity. Because tank salinity was a DIN/DIP 165.9 161.8 53.4 3.2 function of flow rate such adjustments (at least theoretically) increased the precision of turn- over time. Differences in flow rate between actual and DISCUSSION that used to set schedules accounted for this drift. The pneumatic ball valves were superior For more than a decade mesocosms have been to solenoid valves in controlling water flow. developed for a number of marine environments. The solenoid valves were diaphram. seated and The substantial diversity of mesocosm designs easily trapped matter flowing through the valve mirrors the different underlying concepts from on the seat when closed thus allowing sane leak- which these designs were generated. Banse age. Ball valves, because they rotate across (1982), Grice and Reeve (1982), Pilson and Nixon the seat when closing, tend to wipe off any (1980), Steele (1979), Davies and Gamble (1979) accumulated matter. and others have reviewed and compared some of Vertical salinity profiles taken during both the larger and more prominent of these systems. phases of the experiment indicated that there At present there are at least a dozen major were virtually no gradients within a tank. types gf design. They range in volume up to Samples taken within 30 cm of an input during 1300 m . Some of these systems have been de- feed revealed differences from tank salinity ployed at sea as large plastic enclosures < .5 */.0 but these were rapidly mixed out fol- attempting to trap the local water community as lowing a feed. The transfer pumps did not in the CEPEX experiments (Menzel and Case, apparently entrain newly added water and over- 1977). Living models of such diverse ecosystems flows were always whole tank salinity. as coral reefs (Adey, 1982) and estuaries with strong salinity gradients (Cooper and Copeland, Ecosystem Behavior 1973) have been constructed. Many marine meso- cosms have only isolated part of the water Data obtained from routine tank monitoring column, but some like MERL have coupled their suggested that in a general way the experimental systems to the benthos. All of these efforts estuary behaved like the natural one. Table 1 have had to come to terms with a fundamental characterizes the ecosystem in the different problem: as you isolate a piece of the eco- treatments (Oviatt, unpublished data). Divers system from nature you are removing it from its observed that the benthic community in each physical environment and must provide for that treatment was viable and similar in appearance with mechanical systems. to the collection site throughout the experi- Perhaps in some situations containment alone ment. The 0 '/oop 5 O/oop and 10 0/,, benthic is sufficient. Light, flushing, turbulence and 1533 temperature, however, are important forcing and less expensive than traditional electro- functions in the marine environment that require mechanical controls. Following an initial in- consideration. Scaling of physical forcing vestment in this equipment, the cost and effort functions to mimic nature is an important design of radically changing the operation of a process concept. If physical parameters are scaled to like a MERL mesocosm can be minimized. relative magnitudes, the biological components There are many types of estuaries: low sal- should behave naturally (Perez et al., 1977; inity (relatively high river input) or high Oviatt et al., 1977). The technology for con- salinity (relatively low river input) and well structing and operating mesocosms is not sophis- mixed or vertically stratified is one simple ticated; the challenge is how to apply it. The classification scheme. In this experiment we behavior of experimental ecosystems is linked to have modelled one possibility. We recognize physical design in much the same way that that in nature there is awesome complexity and natural systems are coupled with their physical variability both within and between estuaries in environment. Mechanical structure and function everything from sediment type to the prevalence are managed more or less to the researchers of Internal waves. This should not discourage ,specification. mesocosm. development - it should serve as an Design can be changed, scale can be adjusted, invitation to explore. and a range of conditions may be created that simulate different natural conditions. For ACKNOWLEDGEMENTS example, varying the speed of mechanical stirrers has been shown to influence plankton All members of the MERL staff contributed in community structure and function in tanks many ways. Candace Oviatt, Associate Director (Oviatt, 1981; Donaghay and Klos, 1985). of MERL, managed the project and encouraged this An experimental estuarine salinity gradient paper. Edwin Requintina produced many of the was established by blending freshwater and sea- new devices and helped keep the experiment run- water in different proportions. An ecosystem ning. Stephen Kelly assisted with control pro- not unlike the reference estuary was maintained gramming and nutrient analysis. Barbara Nowicki throughout the experiment. The results have was in charge of the routine mesocosm monitoring demonstrated the feasability of operating meso- and was supported by Laura Weber, Cynthia Heil, cosms to these specifications for longer periods Lynn Beatty, Barbara Sullivan, Viva Banzon, J. of time and for more complex experimentation, Douglas Cullen and Aimee Keller. Special thanks for example adding nutrients to a parallel sys- to Richard Sisson, RI-DEM, and John Hall, Frank tem. Conversion to a vertically stratified Hall Boat Yard, for introducing us to the Paw- salinity gradient is a future goal and will catuck River. Research presented in this paper incorporate methods applied previously to main- was supported by EPA Grant No. CR812487 and tain strata in single tank systems. Managing funds from the Mellon Foundation. flow between compartments will become more complex however. Tanks linked together to produce a multi- REFERENCES compartment mesocosm offer unique research opportunities. The compartments are interactive Adey, W.H. 1982. The microcosm: A new tool for at rates that can be controlled. Estuarine box reef research. Coral Reefs 1:193-201. models which describe replacement time and transport through a non-homogeneous environment Banse, Karl. 1982. 2. Experimental marine can be applied directly to an experiment. it ecosystem enclosures in historical perspect- has not escaped our notice that by allowing tank ive. In: Marine Mesocosms, G.D. Grice and overflows when the tanks were connected compli- M.R. FEe-eve (eds). Springer Verlag, pp. 11-24. cates calculation of mass balances and other factors. We allowed overflows so that each tank Cooper, D.C. and B.J. Copeland. 1973. had an equal turnover time at its selected sal- Responses of continuous-series estuarine inity. A decision to have water exit only the microecosystems to point source input vari- high salinity end of the experiment would have ations. Ecol. Monogr. 43:213-236. resulted in each of the tanks turning over at different rates. We chose the former option for Davies, J.M. and J.C. Gamble. 1979. this experiment for comparison with past work. Experiments with large enclosed ecosystems. It is not clear at this point which option is Phil. Trans. R. Soc. Lond. B. 286, 523-544. more desirable or if different types of investi- gation are best operated differently. Donaghay, P.L. and E. Klos. 1985. Physical, With the advent of microcomputers and compa- chemical and biological responses to simu- tible interface devices, process control has lated wind and tidal mixing in experimental changed dramatically. Electro-mechanical hard- marine ecosystems. Mar. Ecol. Prog. Ser. 26: ware must have the same power wiring no matter 35-45. how it is controlled, but with use of a com- puter, control is by written program. This approach has the potential to be more flexible 1534 Fox, M.F. and D.R. Kester. 1985. Fate of ocean- -dumped acid iron waste in a stratified microcosm. In: Duedell, I.W., B.H. Ketchum, P.K. Park, a@_n_d D.R. Kester, (eds.), Wastes in the Ocean, Vol. 5, Wiley-Interscience, N.Y., pp. 171-185. Grice, G. and M. Reeve. 1982. Introduction and description of experimental ecosystems. In: G. Grice and M. Reeve (eds.), Marine Meso- cosms; Biological and Chemical Research in Experimental Ecosystems, Springer-Verlag, NY, pp. 1-9. Lambert, C.E., and C.A. Oviatt. 1986. Manual of biological and geochemical techniques in coastal areas. MERL Series, Report No. 1, Second Edition, The University of Rhode Island, Kingston, RI. Menzel. D.W. and J. Case. 1977. Concept and design: controlled ecosystem pollution exper@- iment. Bull. Mar. Sci. 27(l):1-7. Nixon, S.W., D. Alonso, M.E.Q. Pilson, and B.A. Buckley. 1980. Turbulent mixing in aquatic microcosms. In: John P. Giesy, Jr. (ed.), Microcosms in7_Ecological Research, U.S. Tech- nical Information Center, U.S. DOE Symposium Series 52 (CONF-781101), pp. 818-849. Oviatt, C.A. 1981. Effects of different mixing schedules on phytoplankton, zooplankton and nutrients in marine microcosms. Mar. Ecol. Prog. Ser. 4:57-67. Oviat t, C.A., K.T. Perez and S.W. Nixon. 1977. Multivariate analysis of experimental marine ecosystems. Helgolander wiss. Meeresunters 30:30-46. Perez, K.T., G.M. Morrison, N.F. Lakie, C.A. Oviatt, S.W. Nixon, B.A. Buckley and J.F. Heltshe. 1977. The importance of physical and biotic scaling to the experimental simu- lation of a coastal marine ecosystem. Helog- lander wiss. Meeresunters 30:144-162. Pilson, M.E.Q. 1985. On the residence time of water in Narragansett Bay. Estuaries 8:2-14. Pilson, M.E.Q. and S.W. Nixon. 1980. Marine microcosms in ecological research. In: John P. Giesy, Jr. (ed.), Microcosms in Ecological Research, U.S. Technical Informa- tion Center USDOE. Symposium Series 52 (CONF-781101), pp. 724-741. Suttle, C.A. and P.J. Harrison. 1988. Amonium and phosphate uptake rates, N:P supply ratios, and evidence for N and P limitation in some oligotrophic lakes. Limnol. Oceanogr. 33(l): 186-202. 1535 DEPURATION OF OYSTERS IN A CLOSED RECIRCULATING SYSTEM Fox, J. M. and A. L. Chauvin Water Management Inc. 1920 Shrewsbury Road Metairie, Louisiana 70004 ABSTRACT The present methodology for determining end point quality in depurated oyster tissue is the A pilot scale depuration facility for Crassostrea recognized procedure for fecal coliforms (6). virginica has been established in _i@ -heavily- This methodology is quite slow (96 hours), and industrialized suburb of New Orleans. This requires either wet storage or refrigerated facility consists of five 10,000 liter holding storage of product until the product has been tanks with a capacity of 200 bushels per 48 determined to be in conformance with State hour period. Because of its location, tanks are guidelines. Current Louisiana regulations prepared with well water and synthetic sea salts regarding water quality and end point criteria to achieve the proper salinity. Effluent water for tissue are shown in Table 1. is circulated through a series of treatment steps and eventually contacted with a form of Currently there is only one commercial-scale, advanced photoxidation (APO) in order to certified, oyster depuration plant operating in eradicate pathogenic bacteria and reduce the United States. This paper presents the accumulation of toxic metabolites. various technologies used to achieve end point criteria for successful depuration of Louisiana oysters in a semi-closed, recirculating system located in an industrial section of a large city 1. INTRODUCTION (New Orleans, Louisiana). It also shows results from a typical run in which compliance Depuration, also known as controlled purifica- with regulations for end-point criteria were tion, is a process whereby the number of met. pathogenic organisms that may be present in shellfish harvested from polluted (restricted) 2. ADVANCED PHOTO-OXIDATION TREATMENT waters is reduced to levels considered safe for human consumption. Depuration of oysters is accomplished by addition of an oxidizing gas known as Photozone Research on depuration began in the late nine- (TM) to the water. The process by which teenth century and has continued over the oxygen is converted to a greater activity level years with basic and applied research on is known as advanced photo-oxidation (APO). bacterial assimilation/ removal and physiological This is accomplished by passing either ambient effects of a wide variety of physical, chemical air or oxygen through a PVC housing or quartz and environmental water quality parameters (1). sleeve, containing a patented ultraviolet lamp Recent developments in environmental pollution with a spectral emission of 185 nM light. Gas and dwindling resources have accelerated this stream analysis by mass spectrometer shows research and caused all government regulatory that of the 1.8% oxygen converted to Photozone agencies to consider requiring depuration of all shellfish, possibly within the next five years. Table 1. Depuration Standards Purification of oyster tissue via the depuration Parameter Concentration process is a function of acclimation of the animal to environmental conditions existing in dissolved oxygen greater than 50% saturation the depuration tank (temperature, salinity; (2), coliform bacteria less than 1. 8 (3), (4). It is directly associated with (mpn/100 ml) pumping activity of the oyster and is typically salinity (ppt) 20% source manifested by shell activity in conjunction with temperature (OC) 10 25 release of feces and pseudofeces into the water. pH @ 7.0 8.4 Monitoring of pumping activity is usually accom- turbidity (JTU*) less than 20 plished via soluble and insoluble nitrogen, mean fbcal coliform less than 20 dissolved oxygen, and turbidity sampling at (mpn/100 g tissue) various points in the system. Various mech- mean fecal coliforms less than 70 anisms have been developed to directly monitor for upper 10% shell activity (5); but this measurement cannot be conclusively correlated to production of feces. * Verified for individual facilities CH2585-8/86/0000-1536 $1 @1988 IEEE MURE 4. (TM), 65% is negatively-charged ozone, 14.7% OYSMR DEPUMMON PLWT FLOW SCHEMAMC hydroxal radical, 6.3% hydrogen dioxide, 5.9% hydrogen peroxide, 4.4% atomic oxygen, and less than 0.1% nitrogen oxides. This gas carries an oxidation potential of 2.60 volts DE@7M TA" compared to classical ozone (96.4% ozone, 3.3% nitrogen oxides) at 2.07 volts. Because of its P long half-life in water (approximately two hours,) it can be easily recirculated. Photozone (TM) is usually injected into water via an air compressor or liquid oxygen dis- WAM penser. The actual mechanism by which it is injected varies; by venturi injector, air diffusion tubing, micropore airstones, static U@ 1@ @ G-) mixer devices or oxygen solubilizers. . Two -------- types of gas generators are used on the Water . ........... Management depuration facility: a gas-phase generator consisting of a metal or PVC cell T" t 5, .3 T@ 180 or 200 nM lamp or a A containing less than q9TZ'. . 1. _-@- T gas-liquid generator that utilizes both UV L-------- treatment and Photozone (TM) treatment. The latter generator utilizes a quartz sleeve to separate gas and liquid phases. Photozone (TM) is created by passing the source gas between the lamp and the quartz sleeve. It is PWOO then injected into a hydraulic line entering the cell and passing between the steel jacket of the L i cell and the outer surface of the quartz sleeve. 12.0 d.) 3. DESCRIPTION OF SYSTEM Because of the location of the facility, the depuration process requires a recirculating hydraulic flow utilizing Activated Oxygen for maximum of three trays wide by seven trays water treatment. A flow schematic of the long exists. This allows for a minimum of three process is shown in Figure 1. Other de- inches clearance on all sides of each tray (FDA, puration plants currently use flow-through 1987). systems and UV for disinfection. The closed loop design optimizes water use, provides water 5. DEPVRATION PROCESS DESCRIPTION quality surpassing process water standards and year-round operation in a controlled The depuration process begins with purified environment. water entering each tank via a 211 PVC manifold situated at the base of the tank. Holes drilled This facility has a depuration capacity of 200 in the pipe assure a uniform flow of water is bushels every forty-eight hours. Both the dispersed into all areas of the tank. Together, National Shellfish Sanitation Program regulations all depuration tanks operate as one process and Louisiana State regulations require a flow system. The tank size, tray configuration, rate of 3.785 liters per bushel per minute, water circulation rate and flow pattern are which provides 151 liters per minute to each identical for each tank in the treatment system. tank. There are five depuration tanks in the system, each holding 40 bushels of oysters and The water treatment process begins with the 10.2 metric tons of artificial seawater. The water from each depuration tank overflowing design criteria for the plant are shown in Table 2. Table 2. Design Criteria 4. SHELLSTOCK PRE-TREATMENT Parameter Criterion Before oysters are placed in the depuration Oyster Capacity/Tank 40 bushels tanks, they are culled to remove heavy ac- Oyster Capacity Total 200 bushels cumulations of fouling organisms and dead Tank Circulation Rate 151 lpm/tank oysters or oysters in broken or cracked shells Total Circulation Rate 757 lpm/tank and then washed to remove mud and detritus. Depuration Rate 3.785 Ipm/bushel* After the pre-treatment process, the oysters Depuration Time 48 hours are arranged in trays with layers no more than Temperature 200C 3 inches thick. The trays are stacked four Salinity 20 ppt trays high in the depuration tanks resting on pallets made of pvc to allow bottom circulation. The trays are configured in each tank so that a 1 Bushel 120 to 133 oysters 2/3 sack 1537 into a top skimmer drain with simultaneous in Figures 2a and 2b, respectively for one run removal via a bottom drain. This water is during the summer of 1988. pumped into a lamella tube separation chamber for settling of larger solids. From there it As can be seen in Figure 2a, there is an flows into a second chamber containing a increase in tank total coliform bacteria at the parallel arrangement of protein fractionation 24-hour sampling point, followed by a decrease columns which assist in removal of suspended to less than 1.8 mpn/100 ml at the 48-hour solids (bacteria), dissolved solids, and soluble sampling point. Samples taken in the recircu- nitrogen. Aeration in the skimmers is provided lation line past the UV port show only a slight at a flow rate of 0.5 cfm per column diffusor. increase at the 24-hour sampling point that Foam from the protein fractionators is concen- decreases to less than 1.8 mpn at 48 hours. trated down into a sump, chlorinated after each run, and wasted into the local sewer system. Results from fecal coliform sampling of oyster All water that passes into subsequent sections tissue (Figure 2b) show a high 0-hour level of the facility must pass through this protein (92,000 mpn/100g), followed by a decrease to fractionation column. Columns are designed to less than 18 mpn/100g at both the 24-hour and process a hydraulic flow of 64 Ipm. 48-hour sampling points. Water from the protein fractionators is pumped into the first of two 15,000 liter Photozone (TM) Figure 2a. Depuration Results treatment tanks through a 90 elbow at the base (total coliform bacteria) of the tank wall oriented to create clockwise circulation. The circular flow pattern assures 70 total coliforms (Mpn/100MI) that there is no short-circuiting in the treat- 60 24 HIRS ment process. Photozone (TM) for this contact tank is produced by four gas-phase generators 60-.--- driven by a small 0.5 hp compressor. Photo- 40-- zone (TM) from the generators is added to the 30------ water by a series of connected concentric rings 20 1 8.- HIRS of 1.27 cm weighted diffusor tubing. This 10 treatment also serves to aerate the water and _J provide sufficient oxygen for the oysters. 0 10 20 30 40 50 Additional treatment is introduced into this depuration hours contact tank via a recirculating venturi (Run 6/18 - 8/20) arrangement from the second treatment tank. Oyster tank - uV port Water then flows into a second 15,000 liter Lot 8118 treatment tank from a 4 inch lateral standpipe. This tank receives the same type of Photozone (TM) treatment as the first contact tank. There is a refrigeration coil in this tank used to maintain the process water temperature at 200C. The treatment water is then pumped from the contact tanks through a parallel arrangement of two gas-water generator units Figure 2b. Depuration Results and 1 parallel UV sterilizer unit. These units have a hydraulic capacity of 151 lpm (APO) and (fecal coliform bacteria) 302 Ipm, respectively. A small bypass of 151 fecal coliforms (mpn/100g) (Thousands) lpm does not receive the previous UV treatment 100 and is mixed with treated water and returned 80 via a manifold distribution system to the depuration holding tanks. Flow into these 60 tanks is regulated by hydraulic flow meters/ valve arrangements (one per tank). 40 6. DEPURATION RESULTS 20 24 HIRS __48. HIRS 01 Typically, a depuration run will last 48-72 0 10 20 30 40 60 hours. Monitoring of physical and chemical depuratIon hours water parameters takes place at 0, 24 and 48 @Run 8/18 - 6/20) hours. Tissue bacteriological samples are taken oyster tissue from each oyster depuration tank at the same time. Water bacteria samples (total coliforms, Lot 8118 fecal coliforms) are taken from a port immedi- ately prior to water recirculating back into the depuration tanks) at similar times. Typical total coliform levels in recirculating water and fecal coliform levels in oyster tissue are shown 1538 Sampling of insoluble nitrogen as protein What must be emphasized in developing or particulate nitrogen (PP-N) from protein evaluating oyster depuration systems is the fractionation unit effluent (condensed foam) ability to sequentially decrease both particulate showed concentrations of 3.9 mg/liter, 156.3 and soluble materials as either bacteria or mg/liter, and 10.5 mg/liter at 0-, 24-, and substrates for bacterial growth by very simple 48-hour sampling points. This data is.shown in and cost effective means prior to contact with Figure 3. UV sterilization and/or ozone oxidation equip- ment. Strict maintenance procedures allowing for cleaning of system build-up or biofouling of tanks, lines, UV lamps and other equipment Figure 3. Fractionation Effluent must be incorporated into operational schedules. (protein particulate nitrogen) The ability to depurate oysters in any system will inevitably be a function of shell and tissue PP-N (mg/liter) bacterial status for 0-hour oysters, oyster 160 biomass, the maintenance of physical conditions 160 conducive to depuration within holding tanks, 140 system processing efficiency, and treatment 120 time. The development of more-rapid and 100 8 ___40 MRA accurate methodologies for determination of end 0 point tissue quality in conjunction with statis- ao 40 1------ tically verifiable levels of sampling could assist 20 in determining the status of individual lots of oysters prior to removal from the depuration 36 40 60 0 10 20 system. This would ultimately result in depuration hours increased depuration. capacity and decreased (Run 8/1,B-8/20) storage time of product awaiting verification of tank water foam eftiuant end point quality prior to market release. Lot 6118 REFERENCES I . Richards, G. P. Microbial purification of shellfish: A review of depuration and relaying. , Journal of Food Protection. 1988. Vol. 51:0): 218-247. 2. Banoub, S., N. J. Blake and G. E. Roderick. Final Report: Studies concerning the cleansing mechanisms of non-01 Vibrio cholerae and V. vulnificus in fl-o:R d a shellfish. DOCTNOAA Award No. NA 83-GA-H-0007. February, 1985. 63 pgs. 3. Rowse, A. J. and G. H. Fleet. Effects of 7. DISCUSSION water temperature and salinity on elimina- tion of Salmonella charity and Escherichia As can be seen from the previous results, there coli froni -sydney rock oysters (crassostrea is a general decrease in tissue fecal coliform commercialis). Appl. Environ. MicriTb_iol. concentration over the 48-hour depuration T9-84.Vol. 48: 1061-1063. period. The decrease to compliance level, as seen in most successful depuration trials, is 4. Fufari, S. A. Shellfish purification: A usually achieved within 24 hours post-stocking review of current technology. FAO Tech. of tanks. The increase in total coliforms in Conf. on Aquaculture, Publ. FIR: AQ/Conf/ tank water can probably be explained by either R.11. Kyoto, Japan. 1976. 16 pgs. release of bacteria (both fecal and total) from oysters as part of the depuration process or by 5. Soniat, T. A microcomputer based shell release from sediment associated with the shell activity monitor for oysters. Proceedings of the oyster. As is seen in the results, of the National Shellfish Association Con- protein particulate nitrogen concentration ference, June 27 - June 30, 1988, New increased in fractionation column effluent in Orleans, Louisiana. 1988. correspondence with a similar increase in total coliform bacteria in tank water. Apparently, 6. American Public Health Association. protein fractionation units can serve a useful Laboratory Procedures for the Examination function in reducing total coliform bacteria in of Sea Water and Shellfish. 5th edition. recirculating systems. Because of the relatively Chapter 3. Procedures for the Bacterio- low cost of construction of such columns, their logical Examination of Water and Shellfish. use could mean a substantial reduction in 1985: 37-63. overall capital outlay for UV or oxidation equipment. 1539 ON RAY TRAJECTORIES AND PATHTIMFS FOR ACOUSTIC PROPAGATION IN A MEDIUM WITH VELOCITY GRADIENTS W.J. Vetter Faculty of Engineering and Applied Science Memorial University of Newfoundland St. John's, Nfld. AlB 3X5 ABSTRACT velocity distribution. This calls, in principle, for sound transmission experiments between many Sound, in addition to its traditional role sources and receivers, and for finely gridded for communication and sensing in the ocean ocean cells with associated cellwise constant environment, serves now also for tomographic sound velocities, for the purpose of ray tracing determination of the spatial distribution of some and mathematical inversing of the traveltime medium parameters. This paper contributes new information. Such many source/receiver probing expressions and insights for paths and pathtimes being as yet not practically feasible, it is of rays in volume cells of the medium for which appropriate to ponder the implications of medium the sound velocity can be characterized in terms heterogeneity on signal travel paths and travel of a point velocity and a constant velocity times, and to contemplate the use of larger volume gradient. Such a parametrization permits the use cells, say with constant gradient velocity models of relatively large cells in the segmentation of (ramp velocity models). These will be concerns to the medium, this having the consequence that the be addressed in this paper. dimensional complexity of the tomographic formulation and inversion is significantly reduced. In the case of a small number of 2. HETEROGENEITY AND VELOCITY AGGREGATIONS segmentwise ramping vertical velocities, including velocity reversals, the profile parametrization In OAT it is the discerned propagation time enables the computational determination of arrival of the sound signal from a source to a receiver point targeted (APT) or eigenrays, the rays from a which encapsulates information about raypath given source point to a specified arrival point. geometry and about the local propagation velocity at every Incremental of the paths. For the ray model idealization the important aggregation 1. INTRODUCTION variables are pathlength 1, pathtime t, and a velocity-pathlength product v P1, which can be Important questions posed by climatologists represented as line integrals or partitions and oceanographers point to the need for sensing thereof as follows [41: and mapping of ocean current and temperature fields and their dynamics, on various spatial and Y f dC = f v(C)dt(C) = vTt temporal scales. A promising approach for such C C large scale sensing has been pioneered by Munk and fc, + f + Wunsch [1-31, ocean acoustic tomography (OAT), C2 whereby traveltime information of sound signals passing through the medium is processed to yield estimates of the spatial sound propagation t= f dt(C) = f dC distribution at the time of probing. Temperatures C C =v vT can be inferred from the sound velocity = f + f + (2) information, whereas the time differential of Cl C2 oppositely directed sound signals gives information about currents, gyres, or eddies. A 2(C)dt(C) = v2t key component of the OAT method is the tracing of vpl fC v(OdC fCv G rays, transmitter to receiver through the f f (3) heterogeneous mediumq including multipaths due to C1 C2 + vPvTt bottom and surface reflections and due to velocity gradient reversals (deep sound channel). It is seen that the above variables for the raypath traversed define time-average velocity The ocean, as a sound transmitting medium, is VT = 1/t, path-average velocity inherently heterogeneous, and the objective of OAT vp = (vPy)/y' and a geometric mean velocity is to retrieve detail of the spatial sound CH2585-8/88/0000- 1540 $1 @1988 IEEE 0 1/2 1/2 rays through a medium cell with a velocity vG [vP/t] [vPvT] in seismic usage distribution which can be modelled by a ramp this latter is known as the RMS or root-mean- velocity profile. For tractability the squared velocity. It is only for a homogeneous relationships are given for the simplest case, for medium and attendant straight line raypaths that velocity ramping by a constant gradient g in the the above three velocities have the same numerical vertical direction, and for the x vs. z plane h > 1.0 is origin positioned at the starting point of a ray. value. The dimensionless ratio vP/vT = As indicated in Figure 3, variables and parameters a quantitative measure of the velocity pertaining to the ray commencement point are heterogeneity encountered along the path [4). It subscripted by a, those pertaining to the ray has attributes somewhat like a variance; the arrival point by 8, those along the ray locus may larger the value above the base value of unity, be unsubscripted or they may be qualified by (z) the larger is the velocity variation. or (x,z), as appropriate. We have thus at the The line integrals may be arbitrarily starting point(xa=ozaolalea); at the terminal partitioned, say corresponding to the boundaries point (xp,zp,vp,Gp); along the raypath of a medium segmentation into cells with which the (x,z,v(z),O(x,z)). Coordinate axes for z and x raypath interacts. Clearly, any medium are directed downward and to the right heterogeneity within a cell will give rise again respectively; all angles are measured counter- to numerically different time- and path-average clockwise from the z-axis, but geometrically velocities for the pathsegment through the cell. identical angles may also be used to simplify Most important however, is that the components in representations. Snell's law holds at the end the partitions of the variables t, vp I in points and at all points along a ray, thus the equations (1) to (3) may be arbitrarily permuted equivalent ray parameter expressions without affecting variable values, or without Q x z = sinl)X,z) coso x,z) 1iXj1 v(, affecting numerical values of vT, vp, vG, or h. For further clarification of the effects of sina = sinp = pa (5) medium velocity heterogeneity, we consider a va vp medium segment suf f iciently small so that it , is only heterogeneous in one dimension, say in the Along an incremental ray segment dC, path vertical [4]. This is then a Snell's medium, i.e. geometrical and time incrementals are related by Snell's law applies for raypaths. Figures l(a) and (b) respectively display a velocity profile dC = v(z)dt and an associated raypath for some given ray parameter value (the solid curves). A partition dx = sinO(x,z)dC (6) of this profile and the Snell's law associated raypath at z, say, and a reordering of profile dz = cos(x,zq)dC and path segments (the broken curves) will not affect any aggregation variables or velocities The gradient of the linearly ramping velocity along the raypath or for the profile which relate in the vertical direction is g, so that the to the total transitions, provided that rays are profile function takes the form monotonic. Figure 2 displays a strongly vertically heterogeneous profile. By extension of v(z) = Va + gz for z < +- za = 0 (7) the situation in Figure 1, this profile may be finely partitioned, and the partition parts With (5) and (7), incremental relationships reordered to generate the monotonically increasing (6) can be integrated to give explicit expressions profile as shown without affecting any of the for raypath points and time as profile or path aggregation variable values or velocity and heterogeneity parameters [4]. The Z(P,O,) = _1 (sinO(x,z) - sin%) (8) pg canonical profile is in fact the cummulative form of the velocity distribution function for the X(P,(q)a) = -I(cosqea- cosqe(x,z) (9) pg profile. The ramp trend in the canonical (sin0(x,z)(l + cose (monotonic) profile suggests that a ramp velocity profile can be an appropriate approximator for medium segments which have inmogeneity of = 1 In (tanWx,zq/2) (10b) consequence. 9 tanu,/Z) Summing sin2 O(x,z) and cos2 O(x,z) from (8) 3. RAMP VELOCITY PROFILE and (9), one obtains the well-known circular raypath locus, its coordinates of the centre and The ramp velocity profile model is relevant radius being transparent, as and important as a segmentwise approximator for 2 2 2 some characteristic deep ocean sound speed (x VQ + 8(z + Va v. -8) profiles like the SOFAR channel, in addition to g ta2nqlu0-8) 9 2F12n08Ta a its role as the simplest of approximators for modeling medium heterogeneity. In this section we Some rearrangement of (11) leads to an summarize and extend somewhat the basic expression which gives the ray departure angle relationships which have relevance for tracing of 1541 for an arrival point targeted ray or APT (21) ray at as (12) It can be frequently convienent to use ray Likewise expression (12) and the related endpoint-specific parameters, as complements to expression for 1/tan with. at a the parameter pair, namely a profile or raypoint , take forms vertical ray path-average velocity vPN and a ramp profile contrast factor q. The defining relationships and various other parameters relating to these are then as follows [4]: (22) (13) (ray point velocity) (23) (14) We have introduced here decompositions of the ray departure angle and ray arrival (profile or vertical path-average velocity) angle being the angle of the chord and a being the angle between the chord and the tangent to the raypath at the endpoints (15) and , as shown in Figure 3. Angles a and are linked through the simple but (ramp contrast factor) important relationship (24) (16) The consequence of these decompositions is that the straight line chord can facilitate (vertical ray time-average velocity) thinking and analytical relationships pertaining note q = to the attendant circular arc raypath. For the ramp velocity model being considered, (17) a ray departing at at an angle 0 will eventually reach its depth extremum at Relevant relationships linking these turning ray angle and coordinate values include the following: (profile or vertical ray heterogeenance) (25) The arrival point targeted ray parameter and cumulative variable expressions can then be (26) obtained as follows: (18) (27) (ray parameter; note q = 4. RAY TRACING AND APT-RAYS (19) From the above, it is seen that the ramp velocity model gives rise to simple and tractable raypath related relationships. These can serve for the ray tracing task through a conjunction of medium cells, where profile parameters for each known from ray direction at exit from the adjacent cell. Relationship (23) among others,in conjunction with geometry expressions for cell (20) boundaries can serve for obtaining directly the 1542 ray exit point from the cell. Appropriate Figure - 4 exemplifies up-starting soundpaths translations and rotations of coordinate axes in in a shallow channel of 200 m, over horizontal the 3-D space enable the above results to be range of 2000 m, vSURFACE = 1480 and g = .1 adapted for ray tracing through 3-D cells in the (m/sec)/ sec . Figure 5 exemplifies rays in the general 3-D medium. In the ocean environment, deep channel, over a horizontal range of 100 km, ramp velocity models would apply for relatively with v THERMOCLINE @ 1475 at 1000 m, gUp = .04, large medium cells, making these then attractive = .01, ZTRANS= 950 m, zREC = 1100 m. for OAT by virtue of the small count of parameters gDOWN to be determined from the applicable perturbation equations. 5. REFERENCES For a single ramp velocity profile seginent, [1) W. Munk and C. Wunsch (1979): Ocean Acoustic relationships which have been detailed unable the Tomography - A Scheme for Large Scale direct determination of arrival point targeted or Monitoring; Deep-Sea Research, 26A, pp. APT rays (also variously referred to as eigen- rays) ., plus their ancillary detail of raylength, 123-161. raytime, etc. For two or three ramp segments and [21 Munk, W. and C. Wunsch, (1982): Observing horizontal reflecting boundaries, applicable the Ocean in the 1990's; Philosophical expressions can be combined with boundary, and Transactions of the Royal Society of London, interface constraints to give low order Vol. A 307, pp. 439-464. polynominal equations for the x-coordinates of ray segment transitions. The root solutions of these [31 Ocean Tomography Group, The, (1982): A equations allow then again construction of the Demonstration of Ocean Acoustic Tomography; APT-rays and segmentwise evaluation of their Nature, Vol. 299, 9 September 1982, pp. ,raytimes. Simulation results demonstrate 121-125. feasibility and computational efficiency for direct and multipaths in shallow channels and in [41 W.J. Vetter (1987): Vertical Heterogeneity deep sound channels from velocity gradient and Noveout in the One-Dimensional Medium; reversals. Geophysical Prospecting, 35, pp. 700-717. 1543 ZI I(Z) X'O x ZO- za -@Y ZY Z16 -Zr 0Y (Z) a;, Y zY-- OY\ ZY-- Z.8 z zV (b) Fig. 1: Raytime and raylength invariance for Z'8 profile and trajectory segment permutation. z (a) Profile, (b) trajectory. A f(V Fig. 2: Canonical Profile. (a) Profile segment and canonical profile, (b) velocity distribution. V (b) G R R - G min 0'@ P, OdL -x (0,0) (0.0) 0( velocity gradient is I G oil Z -ZO-MX - - - - - - - - - - - - - - - m-lan 5 P2(XI3 Z(O (0, z 0) z 1,z (X TURN Z TURN (a) (b) -Y Y@ G @P' Fig. 3: (a) Ramp velocity profile, (b) raypath geometry Iw A -Too R too \V\ I \1 V X X@ A A@ ---@VV X N z 0 -1000- 0 A/ @L V V \V-/ D (b) /000 2000 Fig. 4: APT rays in single ramp segment shallow sound channel (a) profile, (b) arrival point targeted rays 0 - V 09 ../Soo vw -2000 25,00 too kift Fig. 5: APT rays in 2 ramp segment deep sound channel (a) profile, (b) arrival point targeted rays ol So VC, 1545 THE DEPENDENCE OF THE MICROWAVE RADAR CROSS SECTION ON OCEAN SURFACE VARIABLES DURING THE FASINEX EXPERIMENT David E. Weissman Dept. of Engineering Hofstra University Hempstead, New York 11550 ABSTRACT 1. INTRODUCTION A measurement program, to develop The Frontal Air-Sea Interaction Experiment more precise relationships between the (FASINEX) was a cooperative program to microwave (Ku-band) radar cross section investigate the role of horizontal and the ocean surface variables that variability in air-sea interaction. it affect the roughness and reflectivity, is took place during the Winter-Spring of in progress. An extensive data set was 1985-86. The region selected for study obtained during the Frontal Air-Sea was an oceanic front in the subtropical Interaction Experiment (FASINEX) during convergence zone southwest of Bermuda. the winter of 1986 in the Atlantic south The operations involved numerous agencies of Bermuda. of interest in this study are and research organizations [Stage and the airborne Scatterometer measurements Weller, 19851. observations and that were coincident with surface wind measurements were made from buoys, ships, speed and stress measurements from aircraft and spacecraft. The scientific aircraft and ships, and the directional objectives spanned a wide range of surface wave spectrum. The airborne radar subjects and included: the role of data set spans several different flights atmospheric forcing in maintaining the (typically at an altitude of 3000 feet) subtropical front; changes in surface during which there were a variety of roughness, stress and drag coefficients different atmospheric and ocean across the front; and mean marine conditions. This also includes a variety atmospheric boundary-layer structure in of radar parameters such as incidence the area [Stage and Weller, 1986]. angle of the radar beam, the polarization and a circular sweep of look direction The goal of this study is to improve the relative tothe wind vector. The goal of this research is to explain the properties understanding of the relationships between of the Ku-band radar cross section of the microwave radar backscatter from the ocean ocean (as a function of incidence angle, and basic geophysical parameters of azimuth angle and wind speed) in terms of interest to oceanography and meteorology, the physical forces and variables, such as such as surface winds and stress. The surface stress and currents, long wave uniqueness of this experiment is that for tilting and hydrodynamic modulation of the the first time it is possible to short Bragg waves. Current theoretical quantitatively examine the combined effect models based on Bragg scattering are being of ocean surface winds and stress, air-sea tested. In situations where the Bragg temperature, sea surface wave spectrum, scattering theory does not agree with the and total mean square slope on the measured radar cross section (magnitude measured radar cross section as a function and azimuth dependence), revisions of the incidence and azimuth angles, because of theory are proposed. the near simultaneous manner in which these measurements were conducted. CH2585-8/88/0000- JW $1 @1988 IEEE Empirical [Wentz, 1984 ] and physically (dual linear polarizations, 3.5 deg. beam based model functions [Fung and Lee, 19821 width) was rotated by a gimbal system; have had their successes and shortcomings every 4 seconds the antenna was slewed by when compared to satellite and aircraft 10 degrees for a total azimuth range of scatterometer data. Newer RCS model 330 deg. This rotation can be completed functions [Plant, 1986; Donelan & Pierson, in about 130 seconds. Therefore local (13 1987] based on more detailed physical km regions) and short duration estimates effects and related experimental results of the physical variables that affect the have appeared recently. They merit RCS can be studied. The accuracy, in the application and evaluation to determine estimate of the radar cross section is their ability to account for the observed believed be 0.5 dB@or better. Features in RCS data characteristics. The FASINEX this data such as upwind/downwind data set is the best available collection asymmetry and spacing between maxima and for providing atmospheric and sea between minima are of interest as well the measurements to be used with the variations measured across the front. theoretical model functions for RCS Figure 1 is diagram indicating the a predictions, for comparison with the data. typical 3-aircaft flight formation. The C-130 carrying the Scatterometer is the lead aircraft, followed by the NCAR Electra conducting gust probe measurements of the mean and fluctuating wind that 2. BACKROUND yield stress measurements and u . The last aircraft is the NASA Wallops P-3. This study involves the analysis of the onboard is the Surface Contour Radar that NASA-JPL Scatterometer (AMSCAT) data measures the two-dimensional spectrum of obtained from 10 measurement flights the long wave field. It also carries the during FASINEX. A total of approximately Goddard ROWS short pulse radar that 30 hours of data were collected, for a measures the total mean-square slope (and wide range of sea and environmental at higher altitudes the directional wave conditions. The radar incidence angle spectrum). Fig. 2 shows the general ranged from 0 to 60 deg., and for each structure of the flight lines on Feb. 18. case the azimuthal scan was close to 360 The numbers next to each line indicate the deg. (within a time duration such that time sequence in which they occured. The the aircraft traveled 12 km along a sea surface temperature front running east straight flight path). Winds encountered and west is indicated by the solid line. over this experiment period ranged from 2 to 20 m/s. The supporting data on sea and atmospheric conditions will be obtained from several sources. other aircraft that often flew in the same formation with the C-130 were the NCAR Electra (wind and wind stress, temperatures, and wind NASA fluctuations), the Navy-NRL P-3 (with NASA-P-.3- C13.OB winds, stresses and radar backscatter AMSCAT measurements), and the NASA P-3 (with the surface contour radar for wave sprectrum SCR measurments)[Friehe and Williams,19881. NCAR ELECTRA 1000M Ships and buoys also provided surface 400m q_qST.PR0BE measurements. 35m S E.A-S. U. R. FA.C E On each day the C-130B aircraft that carried the AMSCAT Scatterometer followed a planned flight pattern that usually consisted of 80 to 100 km straight Figure 1 The aircraft relative positions and segments that crossed the frontal boundary altitudes of the three different aircraft are several times. The altitude was 3600 shown. feet. The variation in azimuth angle was performed by a conical scan rotation of the antenna mounted on the underside of the fuselage of the C-130B. The antenna IWO_ @@'C@T@ R SC@ NCAR E@TR A !400- LECM qUST @RO BE 35 1547 Detailed comparisons between theory and experiment will continue to be conducted over a wide range of wind and sea 28.5- T 5 conditions. Specific tests will be made +E to separate experimental situations where SEA SURFACE only one important quantity changes over a TEMPERATURE FRONT flight line, such as: winds (and/or 14 stress), the long wave spectrum (from WI.ND: surface contour radar data) or the total mean square slope (from the ROWS radar). 2&0 - 0+ -9 OVS FROM170* Comparisons between the theoretical and (23.40C1 6 measured changes should give an excellent test of some of the specific terms in the 7 theoretical equations. 27.51 1 70.5 wo 69.5 WEST LONGITUDE Figure 2. The flight pattern of the airplanes on Feb. 18 is shown, the arrows and numbers in each FIB-THEORY EXP. V-POL .035 - - - - - - - - - - - - - - - -- - -- - - segment indicate the flight directions and the relative sequence. The curved line indicates the .230 - I L L - -,L. -J...... L - -k- - j - - 1. - - sea surface temperature front. A wind arrow for conditions south of the front (at a particular .025 E-, time) is given. The "Ell and "Ou are the locations of the ships Endeavor and Oceanus. 0 .020 4- T n n --T U) 3. DATA ANALYSIS 4_ This study is providing an evaluation of -Ij the relative and absolute accuracy of these RCS models over a wide range of radar angles and environmental conditions. a. Mae In addition, whenever possible, errors in A the theoretical predictions will be STEP-RNGLE examined with the purpose of identifying those aspects of the theory that might be Figure 3. RCS data (V-Pol), on a linear scale, modified to yield better results. For from the last cycle of Run 2 on Feb. 18 (south of example: the FASINEX data & theory the front) is given with a theoretical calculation comparisons shows that Plant's model using Plant's model modified to include two other provides good agreement with the measured swell wave components. V-Pol RCS results (see Fig. 3), but does not fully account for the 4 dB change in RCS ocean the ocean front (see Table 1 - sequence of antenna circles for Run 2, from I to 7). in addition, the measured H-Pol RCS is noticeably higher (3 dB) than Wave slope information is an important the theory for incidence angles above 40 part of all the theoretical models of RCS. deg. (for the 50 and 60 deg. data; see Mean square wave slope measurements were Fig. 4). This latter situation (a slower made with a Ku-band broad beam radar decline in the measured RCS with incidence altimeter [Jackson, et. al. 1987]. This angle) has been seen in other studies instrument measures the total mean square [Donelan & Pierson, 1987; Fung & Lee, slope (averaged over all azimuth angles). 19821. Plant's theory contains a modulation transfer function term that is based on measurements of the hydrodynamic interaction; this term is important for RCS estimation. 1548 rr 2 0 rl- - t-h 10 1050) 0) M M 0@ @- 0 500 ;V rn CL :g tY @-' M V IC 000MFh M M r@ 0 OD ft (D0t-h F@ r- 1,- 0ct, P. 0 or Is U) H. l- ri ze a) I-h 0, rr " rt- - m A)zm(D 0 otl ED :3 (0 rl- rt' (D COMF- (D H. p) -(D ED 0, t-h (D (D F@8r- La ft Mr, 0)0rt-z"C:cA)0F, "Is (a a tr U) rr rr H. to t-h"0rt0cr Is (D Im Iwo F@ M rl-0En (D0H- H- 0@C (D rr tr kO (D H- V40ct 0:c C: mpo5:3 :3 0)5fD (D 000Di cr tr (D 0) Di (D [a (D (D rt, (D 'I (D<(D CL ODZ(D 0) <rn rn -0Zn, 0, > rn (D :30rr :3 (D rr X rt* rt rt' (D rt' CI) rr rt-0)- 0)0C: J-- ctW 0 CA) :3 0) (D tY0::r (n (a (D a :V4 :; :5 H- 13 PUC:5 H- 0M0)0WtV Z0 (D l< 0) 0) 9)) rt- M @-,00,,CZ) (D M rt rt- M Mpi (a (n l< F@ 00 (D013 P. 1.4m 0Cr ID m 0rr @3 xrr0Cril (nCm CL C3 0.40In rt* (D PI H- EO D, tr) ::r Ai rt (D F@ Ntr pi0CD 91 (D Fj (r-IN pi"(I (a rr (DMrt (D0E:I (D M0) t3 (D :3 CD 0) 0) CY)0>A) ::% :y0rr m0) rt, ct F' C, tj @-N 0@ Z0- tomrr A) PI (D rr (D ti P-? 81---"73 ... PQ m(n P_ 3c @r @I0F- @- zi ::r rt 0)000ell - - Jr: CLIM 0) @-' Wrt rl, F,- (D PI I-h rj) k< (T 0) -CL :5 Z% rLmw@0 U) (DC0H- Z0 PC 1-4 zE tr :j ri00rn rr 0) La -4 CA ru :j U:) pam0(D 010rr 3C0 010) .17' cnW0) (D ct C= w(1) @- (D001 H. ri) Di (D rt P-h 0, IC (D<@r rt, U) CL tr lee rt tl' to"H- rn :c A) rt*M0)01-3 ::rm00 z cl 000 0 @0 r_ 51 C) 0)<:Y En 01 :C tr (D En50)0 PC (1) ::1 (D5rt' MW<MM () I-h (D 0 @(D0Is rn m@- 0 (D fti --- V ::% (1) rn @u H- (a Ea I- @d SD (D (D 0) rn ;-v rr 1< rt (D CoM(D :C ft (D (j) %0 - to I- P.P0 co :c @_ m::r0OD C: r- T..; w @0 CD CDm99 0 pi 0) DiMQW Ln -tr Hm L 0rtm tj to U)00)<m ct ct mMcr 0 M [-- (D0co (D 0H::s @_ F.- @_ m0xm m0to mm 4ti, r_mto Di m am:3 & o":j al cr t@r cr F@ .0 (D rt- COMt:r M - ED 0 tO (D (D rr 0:y 1< rt*,X0380 "m m:5 000 . . . ra m En 0) m01-5% I I I Hto 01 (n0:C H- :30::Y IV rl' 0 (D5F" (a --r- T--I 0P)0 . I 7IT- :y 0) ri P. @- 0) ka 0- rt, 0) M::l rr 0) (D rr 0) Ct ct- P) -A. _L --- : P) H- rr W < t:y' A) 1- CT ri<Cr 0A)M0rr @r Z%mI-- F@ : : : :3 1< m (D rt. Fu :3 @J (D P) (D - MM Q,000m "I U ct rr SI) rr M A) a] C; 0 - 1- Ell (D ft @_j En ID En Di HQftj ::1- 0- 0 (D 00-0) " @- (D 0) 0) :V4 a @- (a 0 0) L- :3 OD L-000F- I.- m m to III) m In SD 0 10m@_w m 0rr :c "@d (I , 0@v CL 19 0 UQ 0 00 F" ca :3 :5wIn 0 @, ::1 v@J. 0 to151 t@ or" " mr. pm co LO"z w ct m@ " m 0 "1 rr P_ 0 mm 0 l< M Ln M0) @- :r H- In CL "H_ ::r :3zmr l" CD rt 0) A)Mrt* En 00to lee u C> mto MM-M M a ;@ I -":cmrt, rr*:3 P. Ism%." T mmH. 1-m0tr rr rno co00 ca :4 :1 cr ai " mrn0mDi"rrWA) ":cm Ef 91 C+ ISO -4b-:L-- (a (D rr F1 (1) Z :X 0) V. ::$ rr Z@ &- ::l A) Ln (D fa (D F-I P. -3X En co to :I M ID<"0) IP7, ZA "Ii rr MMI.' = CD00 in' rn :Cmcot (D (OD) t0h rr 0:r @, ::r 0) pol C+m w zr (D (D (D @.,. w020_IV 0 CL (D H- 00 (t 0,c0mCL rt (D pi ca0rr t-h ftJ 0) Ct (D :3 M=3c0D, to0iM (D013 M (1)pq:3 H- F, ::r rt Lo @c LQ 14 :3 (D I- =100 13 :3 NJ 1- LQMri Qj (D 0) U) Ct, P, rr 11) rr a0) M 0fj. =1 I.... :c to Fi. Mi to @r cf, 01 rr (D 0F@ 0) Ur CL rr ri- rr 0ma) C+ P) 240 ----- r.!. - C2 rt zr 0) (D (D :3ZLD IQ (DCH0:3' H. 4zE0rt (D 0, N ::r H- rr Ib En ft rr W OD IVrr rrMIC W 0(D0(+M 0) LQ P- tj R, 00 00) P- (D 0) rt Zr_ pi 0i (+ :K :5 @r rr Ab rt (D I-h H- -9@- -0H. 9)) 1`1 U) Qj I:D 2GO' @rclmo 0)0H- 0, 0) ai H. 0 z t mm ""0 0"s" L) mm cC: I-h M r-0(t M 0co (D0 F ft Qj ti @I a ft-0Mrt a) :r (DmM H. @Cl0-4 C+ rq 290 -------- tr P) < (D H- 0=@-- a F3 ZI :3- @- (1) @_, 0(a 0) (D c+ CD Cr m = I%Q z 00ct00 t:r ft rr m 0"0) rt @r cr C+0P. .5 CD Cli PV COLO m@-- @' H- 0rt (D0:C0:3 10 OD rr 1- 0 Z, 0) (D ct'0 : 1 001--- 0, (D Z ED ct r_ SD (D CD MH. 0:4 (D ct (D P)NmM ct F" Al rr rt "-J (D =1 Eg Qj V(1) rr0M(D M 'Y- H- rt @- M00IVMtY W(I' Cli F@ ID La0Xa a20 r_ mN = m H. (D Di Fl-0CiM(aM(D " La (D (D M 0) 0 " u:l co a La 00"11-ca H. lb LQ00) C+mQS 5I= (D A) &< cr <-10 (D ":c CDm (D kQ H- 0:3* 0 F@ CD VI ti0 340 w -ai 5rr @if 0) 0) ri ct LQ En ft @-, (D rr w. (D (D FA- m Im m(DZrr<I- PI rr 0(D 1-1 PMZ :30 ct LQ H. ctM0 %b (D rr H.00m La cy0 rr 0 cn 0 M44@ 0@" 0rn :71 M00510rt ZI H- 10 =s, 0' @- 0t:l (D 36 F- (D M0- I-h M Z -(D (a EQ @@ 0 LQ EO M LQ - (D M Di Is 0to E43 'SIGO' - COMPARISON BETWEEN MEASUREMENTS AND THEORY -141-- 40 ft- FEB. 18 (50 DEG. INCIDENCE) F4 R7-V-POL-LO MTF(-2.5)THY&DRTA RUN 2 ---------- SM IM MEAS-(V-POL) IHEORY RCS IL*-(Cm/s) .w 1 28.6 0.013 29 ----------- 28.5 0.020 29 ------ .04 L 1--J 2 28.4 0.015 3 -------------- 28.3 0.023 3 4 28.2 0.028 02 5 0.029 28.1 0.028 35 .ell .41 6 28.0 0.030 4 ---- j 7 27.9 0.029 27.8 0.028 29 RZIM-RNGLE RUN 4 MEAS.(H-POT-) Figure 6. Comparison of V-Pol RCS with theory, 1 27.8 0.0073 from Feb. 24 data. Incidence angle is 40 deg. 2 28.0 0.0090 3 28.1 0.0080 .9 4 28.2 6.010 0.0038 30 R -- F4 RG H-POL L0-MTF(-5)THY&DATA @a,t- t@s, Nmum 5 28.4 0.0087 0.0030-0.0033 24 .025 6 28.5 0.0055 28.6 0.0028 24 ---------- Table 1: Radar cross section measurements (at > .013 up-wind maximum) for individual scatterometbe azimuth cycles (labeled Sub# 1 to #7 for Run 2 and .. . . . .. #1 to #6 for Run 4) are listed and compared with U) .010 - - - - - - - theoretical estimates based on long square slope and u* wave mean 14 The important conclusion here, consistent with other theoretical comparisons with measurements [Fung and Lee, 1982; Wentz, et. al., 1984], is that for the larger AZIM-ANGLE incidence angles ( > 40 deg) the measured H-Pol RCS is higher than the theory. In -igure 7. Comparison of H-Pol RCS with theor order to explain this discrepancy it is y proposed to re-evaluate one of the f* om Feb. 24 data. Incidence angle is 40 deg. .important theoretical terms that is common to all of the theoretical models. This is the term in the RCS model that is referred Measurements of the height PDF have shown to as the "quasi-specular term" that departures from Gaussian statistics at arises from the scalar approximation to higher winds and slopes [Huang, et. al. the Kirchoff (Physical optics) method 19831. This is the cause of the well [Ulaby, Moore and Fung, 1982, Chapt. 121. known electromagnetic bias effect that The usual assumption has been that the sea must be compensated for on satellite surface height and slopes are independent altimeter measurements. Furthermore, the (the latter being much less than 1) and measurements by Tang,& Shemdin show that have Gaussian statistics [Ulaby, et. al., slope magnitude is correlated with the 1982; Durden, 1986]. The RCS formula for long wave height, contrary to the usual this term is a Fourier transform of a assumption in deriving the radar cross Gaussian joint probability density section from the Kirchoff integral. It is function and thus has a Gaussian form possible that departures of the actual which decays very rapidly with incidence surface from the Gaussian PDF could have angle. There is ample evidence that an appreciable impact on the H-Pol RCS at Gaussian height and slope statistics is the higher incidence angles where the only an approximation [Cox & Munk, 1954; Tang & Shemdin, 1983; Wu,. 19751. 1550 Bragg scattering term is relatively small FEB. I B-RUN 6 V-POL (SUB 7) (35 dB below the nadir RCSJ. The V-Pol backscatter is usually 4 to 6 dB higher than the H-Pol.so it would not be affected by this additional contribution to the RCS Li 92 as much as the H-Pol will be. Recent H-Pol X-band radar measurements in the NRL T wavetank also find this same discrepancy at high wind speeds. Measurement at 20 deg. incidence angle on Cc W Feb. 18 show interesting features. In Fig. 8 & 9 (V-Pol and H-Pol respectively) _J .2. there are very little directional effects in these azimuth scans taken south of the frontal boundary. These show no maximum in the radar cross section looking upwind. AZIMUTH ANGLE However other scans along the same flight line do show a small up/cross ratio of Figure 9. Comparison of V-Pol RCS with theory, about 2 or less. The Bragg theory (dotted from Feb. 18, south of front. 20 deg. incidence. curves in Fig. 8 & 9) gives a good estim- ate of the average RCS, but not of the up/crosswind ratios. The possible,role of "physical optics" scattering will be exam- 4. DISCUSSION ined for this case. This project is advancing the understanding of the relationship of the It is intended to continue theoretical and wind stress on the ocean, atmospheric numerical investigations of the turbulence, the long wave two-dimensional consequences of non-Gaussian surface spatial spectrum and the mean square slope properties, on the "quasi-specular" term to the average radar cross section as a in the RCS at higher incidence angles. If function of polarization, the incidence this effect can be proved and explained it angle and azimuthal angle. Future could have important ramifications for applications of Scatterometer data, future Scatterometer measurements and whether from space or aircraft will other radar techniques of ocean remote greatly benefit from a better physical sensing. understanding of these relationships. The continuation of the analysis of The FASINEX data from the airborne Scatterometer data from FASINEX will also Scatterometers and the related atmospheric proceed to explore the wide range of and ocean measurements are the incidence angles and wind speeds "state-of-the-art" in terms of their encountered during this program, and carry accuracy, time and spatial coincidence, out more comparisons with the theory under and.range of conditions. The issue of the these varying conditions. higher measured H-Pol backscatter (at the higher incidence angles) has been a puzzle for some time [Donelan & Pierson, 1987]. FEB.18-RUN 8H-POL (SUB 1) 20 DEG. This study offers some new opportunities and ideas to pursue better answers to the questions about 1) the higher levels of H-Pol backscatter at the higher incidence angles ( > 40 deg.) and 2) the effect of the long wave field (and the modulation transfer function) on the RCS function, at various wind speeds. For example, it has 44 been suggested (Lyzenga, et. al., 1983) Li V_ that the coherent scattering from wedges _J and surface elements with small radii of curvature may make an appreciable contibution to the total backscatter at the higher incidence angles. This study . . . . . . . suggests an alternate view; that the incoherent scattering should be AZIMUTH ANGLE re-examined, but with attention to the character of the statistics and probability density functions of the ocean Figure 8. Comparison of H-Pol RCS with theory, surface wave heights and slopes. Then it from Feb. 18, south of front. 20 deg. incidence. 1551 will determined how these can affect the Jackson, F.C., W.T. Walton, B.A. Walter, Kirchoff integral and the quasi-specular and C.Y. Peng, Sea-surface mean square RCS theoretical term. slope from Ku-band backscatter data, Submitted to J. Geophys. Res., 1987 The 20 deg. data for Feb. 18 and other days will also be very useful in compar- Keller, M., Naval Research Laboratory ing theory based on Bragg scattering vs. (private communication). quasi-specular (physical optics) scatter- ing. Lyzenga, D.R., A.L. Maffett and R.A. Shuchman, The contribution of wedge scattering to the radar cross section of 5. ACKNOWLEDGEMENT the ocean surface, IEEE Trans. Geosci. Remote Sensing, Vol. GE-21, No. 4, pp 502-505, 1983 The authors appreciate the contributions of the following people to the progress of MasuIko, H., K. Okamoto, M. Shimada, and this work: Ed Walsh, who provided us with S. Niwa, Measurements of the Microwave wave spectrum results from the Surface Backscattering Signature of the Ocean Contour Radar; Fred Jackson, who gave us Surface Using X-band and Ka-band Airborne his wave slope measurements from the Scatterometers, J. Geophys. Res., Vol. Ku-band broad beam altimeter; and Bill 91, pp. 13,065-13,083, 1986 Plant, who shared his theoretical model function program to permit efficient and Plant, W.J., A Two-Scale model of Short accurate calculations of the radar cross Wind-Generated Waves and Scatterometry, J. section. Scott Shafer of JPL provided Geophys. Res., Vol. 91, No. C9, pp. easy access to the Scatterometer data. We 10,735- 10,749, 1986 are pleased to acknowledge the support provided by the NASA HQ Physical Schroeder, L,C., P.R. Schaffner, J.L. Oceanography Program, Dr. James Richman, Mitchell, W.L. Jones, AAFE RADSCAT 13.9 Manager, to Hofstra University and the Jet GHz Mesaurements and Analysis: Wind-Speed Propulsion Laboratory. Signature of the ocean, IEEE J. Oceanic Engg, Vol. OE-10, No. 4, pp 346-357 No. This is FASINEX Contribution No. 66 Cll, pp 13,063-13,083, 1985 6. REFERENCES Stage, S.A., and R.A. Weller, The Frontal Air-Sea Interaction Experiment (FASINEX); Cox, C.S. and W.H. Munk, Statistics of Part I: Backround and Scientific sea surface derived from the sun glitter, objectives, Bull. of Amer. Met. Soc., J. Mar. Res., Vol. 13, pp 198-227, 1954 Vol. 66, No. 12, Dec. 1985 Donelan, M.A. and W.J. Pierson, Radar stage, S.A., and R.A. Weller, The Frontal scattering and equilibrium ranges in wind Air-Sea Interaction Experiment (FASINEX); generated waves with applications to Part IIz Experiment Plan, Bull. of Amer. scatterometry, J. Geophys. Res., Vol. Met. Soc., Vol. 67, No. 1, Jan. 1986 82, No. C5, pp. 4971-5030, 1987 Tang, S. and O.H. Shemdin, Measurement .Durden, S., Microwave scattering from the of high frequency waves using a wave ocean surface, Center for Radar Astronomy, follower, J. Geophys. Res., Vol. 88, Stanford University, Sci. Rept. No. No. C14, pp 9832-9840, 1983 D901-1986-1, June 1986 Ulaby, F.T., R.K. Moore, and A.R. Fung, Friehe, C.A., and E.R. Williams, Aircraft Microwave remote sensing: active and measurements in Fasinex, Proceedings of passive, Vol. II, Addison-Wesley 1982 the 7th Conference on Ocean-Atmosphere Interaction, Amer. Meteor. Soc., Walsh, E.J., D.W. Hancock III, D.E. Anaheim, CA., pp 38-40, Feb. 1-5, 1988 Hines, R.N. Swift and J.F. Scott, Directional wave spectra measured with the Fung, A.K., and K.K. Lee, A surface contour radar, J. Phys. Semi-empirical sea-spectrum model for Oceanog., Vol. 15, No. 5, pp 566-591, scattering coefficient estimation, IEEE J. 1985 Oceanic Engg., Vol. OE-7, No. 4, pp. 166-176, 1982 Walshi E.J... D.W. Hancock III, D.E. Hines, R.N. Swift and J.F. Scott, Huang, N.E., S.R. Long, C.C. Tung, Y. Surface contour radar observations of the Yuan, L.F. Bliven, A non-Gaussian directional wave spectrum during FASINEX, statistical model for surface elevation of Proceedings of the 7th Conference on non-linear random wave fields, J. ocean-Atmosphere Interaction, Amer. Geophys. Res., Vol. 88, No. C12, pp Meteor. Soc., Anaheim, CA., pp 103-106, 7597-7606, 1983 Feb. 1-5, 1988 1552 Wentz, F.J., S. Peterherych and L.A. Thomas, A Model Function for ocean Radar Cross Section at 14.6 GHz, J. Geophys. Res., Vol. 89, No. C3, pp 3689-3704, 1984 Wu, J.' Correlation of micro and macroscopic structures of wind waves, Proceedings of IEEE OCEANS 175, pp 321-326 OONA: AN INTELLIGENT DSS FOR THE U.S. COAST GUARD Hemant Bhargava Steven 0. Kimbrough University of Pennsylvania Department of Decision Sciences/6366 Philadelphia, PA 19104-6366 (215) 898-5133 Abstract: This paper describes Oona, an intelligent de- vessel cost models (CASHWHARS). Significantly, SAR cision support system (DSS) developed at the University and ELT are Fortran models that were written well be- of Pennsylvania for the U.S. Coast Guard, in conjunction fore Oona was envisioned, and CASHWHARS is-like with the Coast Guard R. & D. Center, Groton, CT. As Oona-written in Prolog. Thus, Oona has proved able to a DSS, Oona is a software tool that integrates data and integrate disparate models into a single coherent system, models into a coherent system, thereby permitting more in spite of the fact that some of these models were never effective and economical use of application data and mod- intended to be used this way. els. Data and models for support of vessel acquisition Besides her ability to integrate models from different decisions are the primary application models currently sources, Oona has a number of distinctive features. We available in Oona. These data and models are useful for shall now discuss several of these. cost analysis, performance analysis, and for forecasting First, Oona is more than a particular DSS, she is costs and requirements. Although Oona was built with what we call a DSS shell [41, a software system provid- extensive use of artificial intelligence programming tech- ing DSS features and functionality in a ge-neric, reusable niques, Oona is, in many ways, traditional in concept for way. A particular DSS consists of applica-tion-specific a DSS- We envision substantial feature enhancements for data and models, plus software for manipulating, exercis- Oona, which enhancements are facilitated by our hav- ing, exploring and reporting on the data and models. In ing built Oona using artificial intelligence programming our design for Oona we have managed to make this latter techniques. software generic, so that the specific data and models may be removed and replaced with other, very different, data and models, and the feature set for manipulating, exercis- 1.0 INTRODUCTION ing, exploring and reporting on them will remain in place. By carefully developing and reusing the shell software in Oona is a prototype DSS. She is written mainly in Pro- Oona, we anticipate that enormous benefits will be real- log, runs in the VAX VMS environment, and addresses ized, including: lower incremental costs for building new VT240 terminals. Oona was developed for the purpose particular DSSs; decreased user training costs and user of supporting analyses of competing vessels by the Coast resistance to computer systems due to user familiarity Guard's Office of Acquisition. The purpose of this pa- with the generic shell; and lower software maintenance per is to describe and discuss Oona briefly, to review the costs due to reuse of debugged code. main, general lessons we have learned from building and Second, Oona is written in Prolog, employing pro- experimenting with her, and to discuss our plans for fu- gramnling techniques normally associated with expert sys- ture releases of Oona, and other DSSs, in light of these tems. A typical, basic expert system consists of a domain- lessons. specific knowledge base that holds the system's knowledge about the application at hand, and a domain-independent, (generic) inference engine for extracting information from the knowledge base. Because the inference engine and other system routines (e.g., for screen management) may be. used for more than one expert system application, 2.0 OONA they are often referred to collectively as an expert system As in any DSS, the primary job of Oona is to facilitate shell. Oona works as a DSS shell precisely because her the exploring and exercising of data and models. The par- software architecture resembles that for an expert sys- ticular data and models that Oona currently works with tem, although it is much more elaborate. Oona in fact include vessel performance models (SAR and ELT) and consists of a high-level inference engine (the meta-level CH2585-8/88/0000-1554 $1 @1988 IEEE control ), several inference engines devoted to particu- u(shipi) = k(1)*u(1,sar(ship1)) + lar tasks (e.g., for differentiating symbolic expressions), k(2)*u(2,elt(ship1)) + and several distinct knowledge bases, some of which are generic and some of which contain information about the k(3)*u(3,cost(ship1)) data and models in this specific DSS. The job of the meta-level control is to task work out to the other infer- Note that the model is presented symbolically to the user. ence engines as appropriate (a technique pioneered in the At this point, generic commands are available for explor- symbolic mathematics program, PRESS [6]). New data ing, and for obtaining information on, any symbol in the and models may be added simply by declaring them in model. The user simply issues an explore command and the appropriate knowledge bases; the previously-existing types the symbol of interest and the system retrieves and inference engines are then automatically available to op- reports on the knowledge it has of the symbol. These and erate upon the new knowledge in the system. Further ' related commands axe part of Oona's shell and will work this software architecture allows both standard procedu- on any symbol that is part of a particular DSS. ral models (e.g., SAR and ELT) and deductive models (as 3.0 LESSONS FROM OONA in expert systems) to reside comfortably in a common system. Our purpose in this section is briefly to discuss the main Third, Oona contains an elementary, but powerful, lessons we have gleaned from Oona. These shall serve to symbolic mathematical modeling language. Models ex- motivate the discussion that follows. pressed in this language may be evaluated and manipu- (1) The use of symbolic programming proved instru- lated by a variety of the built-in inference engines. mental in delivering the generic features of the shell part Fourth, like its prototype predecessors [3], Oona of Oona. (2) Expert systems programming techniques makes essential use of a multiattribute utility model in also proved key to providing a rich feature set in a mod- order to integrate the cost and performance models math- ular, expandable way. We want to emphasize that Oona ematically. Significantly, the multiattribute utility model is not an expert system. Instead, she has been written is expressed internally in Oona's symbolic modeling lan- making extensive use of declarative knowledge bases and guage. domain-independent inference engines. The expertise en- Fifth, Oona exploits her symbolic, declarative-style coded into Oona is expertise about how to manipulate programming to deliver a hypertext user interface with models, how to put together a report, and so forth. None many of the links between information items (nodes) gen- of this knowledge had to be extracted from an expert erated automatically by the system. Once a model (or and engineered.. Instead, it is mostly common, public, data item) is declared in an appropriate knowledge base, and-most significantly-static knowledge. (It is static generic commands are available for displaying the model because, for example, the rules of differentiation are fixed (or datum) in symbolic form and for further exploration and will never change. Such stasis yields a tremendous of the symbols constituting a particular model. As noted maintenance benefit and helps promote the reuse of coded above, three models-SAR, ELT, and CASHWHARS- knowledge, precisely what the shell concept was intended axe integrated mathematically by a multiattribute utility to achieve.) model expressed in the system's modeling language. At (3) We consider our experiment here with command- one level, a user might see, for example, the utility model driven-rather than mouse-based point and click-hy- expressing the overall value of a given ship: per-text to be a success. Of course, a much better dis- play system could be built, given a hardware and soft- ware environment for building user interfaces that is more sophisticated than a VT240 terminal (e.g. Macintosh 1555 [2]). The essential point, however, is that the design con- lesson is that our exploratory development of a declara- cept motivating hypertext-variously described as dis- tive-style model management system has worked extreme-1 cretionary display of information (DDI), knowledge on ly well and merits significant additional development. We demand (KOD), and what you want when you want it now turn to a more general discussion of model manage- (WYWWYWI)-has been shown, we find, to be immense- ment in DSS. ly useful and valuable. In the future, no DSS should be without it. 4.0 MMS DESIGN CONCEPTS (4) The fourth lesson we draw from our experience The purpose of this section is to describe briefly some with Oona has to do with a significant conceptual short- of the central model management system (MMS) design coming we discovered in operating the system. In spite concepts implemented and/or planned for Oona, as well of the innovat 'ive technical approaches used to implement as other DSSs we are building for the Coast Guard. Oona and certain of the resulting features made possi- ble, Oona remains in concept a traditional DSS; she is 4.1 PROCESS-0111ENTED DSS a friendly environment for exploring data and exercising models interactively. All that is well and good, but more Early on in the development of the field of DSS, it was is needed and what that is may be briefly described as found that DSSs in general and DSS features in particu- follows. lar can be categorized as either data-oriented or model- Arriving at a decision in a rational and considered oriented [Alter 1977]. However correct this may be as manner amounts, we believe, essentially to developing a description of actual practice, it is our finding that a and comparing competing arguments (or reasons) for al- third orientation needs to be provided for in DSS. As- ternative courses of action. This is called the argumenta- suming the argumentation theory, DSSs should be ex- tion theory of decision support [5]. This point of view is plicitly oriented towards the process of coming to a de- reflected nicely, we believe, in federal procedures guiding cision, and that process-tempered by case-specific re- major decisions, for example in the A109 circular (Office quirements, e.g., the A109 process-is basically one of de- of Management and Budget) that outlines procedures for veloping and evaluating arguments for competing courses major systems acquisitions. The conceptual shortcoming of action. We see the requirement, then, for argumenta- we find in (the current version of) Oona and in traditional tion- or process-oriented DSS and we are actively design- DSSs is that they focus on the trees-on exercising sets ing and developing implementations. of data and models-and fail to help the user see the Our process-oriented concept, with regard to pro- forest-i.e., the arguments for competing courses of ac- viding DSS shell support for the Office of Acquisition, tion, which need to be constructed and evaluated. Put exploits and enhances concepts from the field of project another way, a major systems acquisition is a very large, management and is, briefly, this. The Office of Acquisi- lengthy project.. Making particular cost or performance tion (in conformance with the OMB A109 circular) has or configuration models available to an analyst or deci- carefully articulated what it wants done during a major sion maker is useful and to be welcomed, but that leaves systenLs acquisition project. Phases, key decision points, unsupported the critical task of managing the process of and so forth are identified and sequenced. Our concept arriving at a decision. Consequently, we are developing a is to model-in a declarative, knowledge-based system broader DSS concept and plan to incorporate it into fu- fashion-this process, both as it is set out by policy and ture versions of Oona and other DSSs we are developing as it actually unfolds in a particular project. This process for the Office of Acquisition [2]. In �4.1, below, we briefly model, a kind of deductive model, would then be available describe our expanded DSS concept. to the DSS shell, and hence to the DSS users, who could (5) Oona has a rudimentary, but powerful, model inquire as to project status, discover, e.g., what decisions management system (MMS), which provides the basis for were made and by whom, add information as the project very many of her interesting and useful features. Our fifth goes forward, and develop project reports in the DSS. 1556 These reports would then have full access to the sorts .sions based on these analyses, and seek justification for of models and data traditionally associated with a DSS. their decisions. To support these users,,.we define the In our current design, we distinguish between process- following feature categories for a model management sys- level entities-e.g. processes that report on how the de- tem. cision process is doing-and subject-level entities g. 1. Construct. New models need to be added to the data and models having to do with the particular subject model base by model builders. The system can support matter of the acquisition project, such as helicopters or this task by reporting on existing models, and providing types of cutters. The DSS is arranged so that the user a facility to input models and solution techniques in a has the feel of being, at any given moment, "in" a certain very natural form (particularly for mathematical expres- step of the acquisition process, Given the user's location sions). Models are expressed symbolically in a model dec- in the process model, certain process-level functions and laration language, analogous in some ways to a database certain subject-level functions will be available, and the data definition language. The model definition language user can then proceed with the business of contributing is declarative so that, for example, builders need only to the progress of the project. In sum, the goal is for declare the mathematical form of a function instead of the DSS to serve as a central respository of project infor- specifying a procedure to compute it. mation, which information is available to the DSS to be reasoned about for the sake of controlling and compre- 2. Formulate. The system will assist users in formulat- hending what is going on-in the project. Space does not ing models to solve their problems. This has two parts, permit a full discussion of the potential benefits of such identifying the model(s) to be used, and obtaining a sym- a concept, but we strongly believe that it can be instru- bolic formulation for the model. The first task includes mental in contributi'ng to a solution of the institutional identifying the problem type or context, determining a set memory problem for lengthy decision processes. of appropriate models, and selecting from this set. The 4.2 FUNCTIONS OF AN MMS second involves paxtial instantiation of the chosen mod- els by identifying model parameters and variables for the Oona is meant to be an integrated modeling system that particular problem, and obtaining a symbolic form for handles various modeling techniques (mathematical mod- the model. In the example in �2 (expression (1)), the els, deductive models, decision analysis models, and data first step involves determining that the overall value of a models), for solving problems in several domains (e.g. fi- ship is calculated by using a utility model, and the second nancial analysis, vessel acquisition, capital budgeting), step involves identifying the parameters for the model to across different phases in the modeling life cycle (e.g. obtain the form of expression (1), in �2. formulating models, solving them, interpreting their re- 3. Instantiate. Once a model has been obtained in a sults), and usable by users with varying degrees of domain symbolic form, it can be used to solve different problems knowledge and computer skills. In developing a feature of the same type by instantiating it (fully or partially) set for model management, we begin by defining the dif- to problem-specific data. Thus we distinguish between ferent user roles that such a system must support, and a model and a model instance. The system will use the which of the modeling activities they perform. symbolic knowledge about models to assist users in cre- Broadly speaking, we see three user roles: 1) model ating model instances. For the ship model, this requires builders, who construct the models and express the do- instantiation of the kW's and cost and performance pa- main knowledge that enables the MMS to perform its rameters, which may themselves be obtained from other functions in that domain, 2) model users and analysts, models. who exercise, the models, solve or analyze real-world prob-1 lems by providing the data, and create reports based on 4. Exercise. Once models have been formulated, they their analyses, and 3) browsers, or decision makers, who may be exercised or executed to obtain solutions. This browse through various reports and models, make deci- includes evaluating a model (e.g. drawing a set of con- 1557 clusions, or computing model outputs), performing sym- if the input is valid (e.g. checking that the dimensions bolic manipulations (e.g. rewriting in a different form, are consistent). or differentiating a model), and executing queries on a model. Depending on circumstances, and the kind of 9. Maintain. Models and data need to be stored and execution, model paxameters will need to be fully instan- maintained. In the case of data, scenarios (or cases) may tiated, partially instantiated, or completely instantiated, need to be altered (in which case links to outputs need to and the system will use knowledge about the relevant be revised), deleted, and created. Models need to be cre- models to manage this task. Models may be solved by ated, edited, deleted. Further, for both data and models, internal solvers (as in the utility model) or by external additional (often meta-level) information will need to be solvers (e.g. SAR and ELT). added, altered, or deleted. Model maintenance involves a tradeoff between regenerating every value (i.e. wasteful 5. Analyze. Modeling is generally an iterative activ- computations), and storing results (which requires set- ity, since problem characteristics, problem data, etc. can ting appropriate links so that the effects of changes are change, and/or not be known with certainty. The system automatically managed, and the values are always up- should assist in answering questions about these changes dated). by supporting ad hoc post-execution analysis, which can 4.3 SOFTWARE DESIGN be analytical (e.g. computing derivatives to understand the implications of changes in some parameters) or nu- merical. A useful feature would be the ability to gener- The application of modeling techniques to solve decision ate a list of all the models and parameters affected by problems involves a multi-step progression from the gen- a change in some value, so that users are aware of the eral to the specific. For example, the system might infer implications of the change. from a user's description of a problem that it requires minimization of shipment costs subject to satisfying de- 6. Explain. An important requirement of a good DSS mand and supply constraints, and may select to formu- is that it be able to explain its results, so that users can late it as a linear program. The final form will be a fully form arguments to support decisions made using these instantiated problem which was obtained (by instantiat- results. This involves the ability to explain the mod- ing index sets and coefficients) from a general product els themselves, as well as the algorithms used, and how shipment problem which is a special case (with specific parameter values were obtained. For example, the justi- forms for constraints and objective function) of a lin- fication for selecting a particular ship may be that it had ear program. For each step in this process, the system the highest utility, which was computed using the model has inference engines to perform the required inferences of �2, and so on. (interacting with the user, if necessary), and there are 7. Describe. At any stage in the modeling life cycle, meta-interpreters controlling the application of inference different kinds of users will need information about the engines (since in general, there is no pre-determined pro- cedure for using the inference engines). The inferences models they are working with; so that they may under- are performed on knowledge resident in various knowl- stand the model better, or determine whether or not it is edge bases. This knowledge is obtained from users either appropriate for their problem. The system should be able explicitly through statements in a (model) declaration to generate such descriptions from its knowledge about language, or through interactions between the system and the models. users. 8. Validate. At various stages in the process of mod- The system performs each of the functions in P-2 by eling, the system may need to get inputs from the user. using inference rules that manipulate knowledge about These might be data values for parameters (e.g. k(l)), models, obtained through a declaration language, and or the choice of a model to obtain values for some param- stored in knowledge bases. For example, the same equa- eter. In either case, the system can perform checks to see tional form of a model would be used in describing it, 1558 executing it, and explaining its results. In general, the system will have several inference engines that use a com- bolic Programming Envirorunent," Proceedings of thel bination of inference rules that operate on a set of knowl- Nineth International Conference on Information Sys-1 edge bases. Hence there is a many-to-many relationship tems, 1988. between inference engines and knowledge bases, which is [3] Ximbrough, Steven 0., et al. "A Decision Support depicted in the system schematic design, Figure 1. System for Evaluation of Advanced Marine Vehi- Figure 1 cles," Transactions of DSS-86, Sixth International Conference on Decision Support Systems, Jane Fe- /-_U`ser_-\ Meta-level dorowics, ed., Washington, D.C. (April 21-4, 1986) USER Interface Control 15-26. [4] Kimbrough, Steven 0., "On Shells for Decision Sup- port Systems," University of Pennsylvania, Depart- ment of Decision Sciences working paper, 1986. [5] Kimbrough, Steven 0., "The Argumentation The- ory for Decision Support," University of Pennsylva- IE, 1E 2 ............ IEN nia, Department of Decision Sciences working paper, 1987. [6] Silver, Bernard, Meta-Level Inference, North-Holland, Amsterdam, 1986. Kb I Kb 2 p Kb IE Inference Engine Kb: Knowledge base P : Process REFERENCES [1] Alter, S. "A Taxonomy of Decision Support Sys- tems," Sloan Management Review, 19, no. I (Fall 1977) 39-56. [2] Bhargava, Hemant, Michael Bieber, and Steven 0. Kimbrough, "Oona, Max and the WYWWYWI Prin- ciple: Hypertext and Model Management in a Sym- 1559 THE ZENO ALLIANCE NETWOR& A DUAL-LOOP FIBER OPTIC INSTRUMENTATION NETWORK FOR SHIPS M[ichael Reynolds, Roger Hendershot, Mark Jungck, and Brian Reid Coastal Climate Company, Seattle WA. 98104 ABSTRACT The ZENO ALLIANCE Network is a dual-fiber, fiber optic ring connecting any number of intelligent sensor in- terface (ISI) units to a central controller. Network devices transtrult three-layer messages simultaneously in opposite directions around the loop, and each device has the ability to accept a message from either direction. In this way the loop is immune to breakage. Each ISI unit responds to its own address or to 'global' commands. ISIs are capable of extensive data compression and operate on the network at 9600 baud. Thus collisions are rare, easily detectable, and A resolvable. A recent ZAN installation is described. Your 18b col- lect statistics of wind, air temperature, relative humidity, barometric pressure, and solar radiation from 3 different locations on the ship. The data are collected in a M[1- crovax 2000 with a sophisticated data base management system (DBMS). Ship information is collected concurrently. 1. INTRODUCTION Figure 1: Photo of the sensors and intelligent interface ZENO ALLIANCE Network (ZAN) is a local area net- asseinblies on the foremast of the RIV ALLIANCE. The work (LAN) made up of ZENO data acquisition devices. fiber optic and power cables are protected in flexible coated Fiber optic inter-connections assure immunity from electro- metal conduit. Photo courtesy of Dr. Peter Minnett magnetic interference and wide bandwidth data transmis- siou over long distances. Our first complete ship installation was a satellite ground meteorology verification system for ZAN devices use a peer-to-peer communication protocol the new NATO research vessel the RIV ALLIANCE. ZAN called Carrier Sense Multiple Access wia Collision Detec- is a next-generation shipboard data system that empha- tion (CSMA/CD) which is similar to Ethernet. Devices aim flexibility and expandability. Figure I shows the final monitor both inputs for transrnissions and switch to the installation of the meteorological instruinentation on the active line. A ZAN message is comprised of an attention foremast of ALLIANCE. This paper will discuss the basic character (#), an address, a return address, the message concepts of ZAN and present details of the installation. string, a checksum, and an end of transmission ch ter. ZAN consists of a number of Intelligent Sensor Inter- The Central Computer Facility (CCF) is responsible for in- faces (ISlo) and a Central Computer F461ity (CCF) con- terrogating each ISI and collecting the various data. riected with a dual, fiber optic loop (Figure 2). Each ISI The CCF operates a comprehensive Data Base Man- operates its own set of instrumentation with its own par- agement System (DBMS) which maintains and operates ticular software package with common communication pro- the network. The DBMS software retrieves and archives tocol. An earlier loop protocol which has been success- data from the ISIs as a background procedure. In the fore. ful in the oceanographic community is SAIL (Serial ASCII ground, the user can use the DBMS to examine data, create Instrumentation Loop) which was originally developed for output reports, and run processing or plotting program. shipboard applications. The DBMS allows one to create and schedule customized CH2585-8/88/0000-1560 $1 @1988 IEEE LOCATION I LOCATION 2 I.LOCATION N C-TYPE In-A-out I--)- -B-in: i-<- LfR out F 181 181 IS1 COM-A COM-B 1 2 N N-TYPE Manual Data Data Archival Entry CCF "Scheme7 R T rts, Blue 'L Definition Dal: Configurations, Maintenance Activity Grapnics COM Figure 2: Sketch of a ZAN system. The dual fiber, fiber Figure 3. The fiber optic interface box and the direction optic cable connects the CCF to all ISIs in a loop config- of signals in the fiber optic cables. The orientation of the uration. The sensors at remote sites are processed by the cable corresponds to the direction of the light in the orange ISIs using individually tailored software. (A) fiber. sampling schemes. When the time for a particular scheme configurations. The loop configuration is the most reliable comes, the CCF automatically reconfigures all ISIs and be- configuration as it allows any device to talk to any other gins collecting the new data. All data are marked according device and assures that the Controller can communicate to the current sampling configuration. with any device, even when the loop is broken or a node Devices in a ZAN system communicate over dual-fiber, device has failed. fiber optic cables. Each fiber optic serial port on an 181 is in LOOP effect a half-duplex port. Figure 3 shows two different types of fiber interfaces and how signals travel on the LAN. The direction of the light in the fibers is in opposite directions. The forward direction is called the A fiber and the reverse direction is called the B fiber. Actually there is no real N N distinction in the two but for practical purposes it is useful to distinguish between them. Signals can arrive at the ISI on either the A or B fibers. An ISI can have one or two UART buffers. In the former case it will collect characters from only one input but will CCF switch between buffers automatically. 9 it is monitoring the A input when a transmission comes on B, it will switch DISTRIBUTED with no loss of data and begin monitoring that input. Sig- nals received at any input will be re-transmitted down the N N corresponding fiber. When an ISI transmits an internally generated message it will transmit over both A and B fibers simultaneously. N N The Contmiler is the part of the CCF that communi- cates and manages the LAN. The Controller uses a special type of fiber optic interface box (FOIB) to communicate C onto the LAN (Type-C in Figure 3). The fiber optic in- terface box is a dual RS232-to-fiber optic converter. Each CCF R8232 port is a full duplex port that transmits onto one fiber and receives from the other. Thus the Controller can frigure 4: Two different fiber optic networks possible with communicate in either direction. In the present configum- ZAN hardware. The loop is the most reliable because it tion, all signals stop at the Controller interface box. will operate even if the loop is completely severed at a spot. Many different network configurations are possible with The distributed configuration is easier to install but not as ZAN hardware. Figure 4 shows examples of star and loop failsafe. VN 1561 Section 2 presents details of a ZAN network and de- ment from the addressed device. scribes the various ZAN hardware available from Coastal A message is any exchange of information from one de- Climate Company. Section 3 describes the installation of vice to another. A message can be one of two types: a this equipment on the R/V ALLIANCE. Finally, Section 4 request or a reply. Messages will have one of two formats: is a brief description the future of the ZAN and planned formatted messages or unformatted messages. Dialog be- new hardware. tween two devices begins with a formatted request. Two devices can agree to go into an unformatted or open state 2. BASIC CONCEPTS by a formatted exchange. A formatted message consists of six partr. an attention 2.1 ZAN Protocol character, an address, a return address, the message string, a checksum, and ending character. ZAN is designed for the interconnection of oceanographic The '#' (ASCII decimal 35) is the attention character and meteorological instrumentation on a single dual-loop for all elements of the ZAN. When any device receives #, LAN although other applications are in no way ruled out it begins monitoring the incoming string. Simpler devices by this emphasis. It is an extension of the earlier SAIL must suspend sampling to do this but multi-tasking devices protocol (EEEE,19ss). Primary differences between the two can collect data and monitor the LAN simultaneously. protocols are- (a) Messages are bi-directional with forward A device addreas will be 4 numeric characters. We rec- and return addresses; (b) Messages are terminated with a ommend that device addresses and device serial numbers be checksum and ending character, and (c) Baud rates can be the same as this facilitates configuration control and con- up to 19,200 baud. figuration archival. The ISI checks each numeric character As in any serial communication there are two states, the and compares to its address. The moment the incoming ad- mark and the space. A mark or logical one is defined as the drew disagrees with the 181 programmed address, it returns presence of fight in the cable sufficient to drive the receiver to waiting for #. circuits to a positive output. A space or logical zero is the It the first character after # is one of the letters A-Z, case of zero light. The duty cycle for messages is low and then all ISIs will check the messige to we if it is one of the majority of time there will be no right. This is different their programmed global commando. Global commands are than standard fiber optic convention but conserves power given by the Controller. They are interpreted by all de- for battery powered devices. Special circuitry in the ISI vices. Since all devices obey the global command they do interface distinguishes between true communication signal not usually issue a reply, Global commands can be used to and random light from an exposed fiber. set the clock, change baud rates, or other such communal All communication within the ZAN system is asynchronous operations. Two global commands are: ASCH and will use the 256 character IBM standard charac- ter set including control characters. Each character has 10 BDbbbb Sets the baud rate to bbbb. bits consisting a start bit, 8 data bits (no parity), and one TMyymddhhmss Sets the device clock to year, month, stop bit. No parity bit is required. The use of an end-of- day, hour, minute, and second speci- m age checlmum provides ample checking of the message. fied. If any device uses or requires parity, it is up to the commu- The return address, the four digit address of the sending nicating equipment to deal with that requirement. device, is the next 4 characters in the message string after In loop operation, the Controller communicates through the device address. This will be the address for any reply two full duplex serial ports, COMI and COAR. One port messages. A responding ISI will invert the device addrew transmits on the A fiber and receives from the B fiber. The and the return address in its reply. other port now the reverse order. Signals transmitted from The message string is any string of ch ten that will one port travel directly through each ISI around the loop be interpreted by the receiving device in a particular way. and are received at the other port. The Controller can The string can be a simple command or it can be an exten- assess loop health and pinpoint any faults using a logical sive data output with embedded control characters. The 'process of elimination. If the loop is broken or if a star attention character (#) and the ending character (-D) are configuration is used, the Controller treats each network as not allowed to be in any message string. a separate full duplex communication port. The checkmm is two decimal numbers which are com- ZAN protocol permits any device on the LAN to ex- puted by adding the ASCII value of all characters following change data with any other device. This uncontrolled us- and not including the attention character. Only the last age can lead to conflict when two devices transmit at the two digits of the sum are transmitted. same time. High baud rates and short messages are recom- mended to reduce the probability of a conflict . The use of Checkoum a message checksum. will reduce the probability of an error. (ASCII characters) mod 100 The Controller can check its own message and re-transmit Finally the last character in the entire message string Jf it is bad. If a loop break is encountered, the Controller is the end-of-transmis8ion character. In ZAN, the ending can switch to the other loop. The ISI cannot check its own character is the control-D, written 'D, whose ASCII value ni age; instead it waits a nominal time for acknowledge- is 3. 1562 When the address@ checksum, and ending character all SENSORS TEnUNAL/SHUTOFF agree, the addressed device interprets the command string 23 5 and replies with the appropriate responses. The addressed device must respond within two seconds and the mean time CPU TRANSIENT PROTECTION to respond should be less than one second. The response A/D 25 time should be as short as possible to reduce the time the RECULATOV loop is committed. All responses to a formatted command INTERFACE menage will be formatted reply messages. In the case of global comman4 the global command, checksuni@ and end- ing character mug agree for the device to respond. CONFRTIM A null message is a command message with no message string, that is there is an attention character, address, re- 10 to tum address, checksum, and end character only. On receipt FIBER FIBER of a null message the ISI will enter into an open state where OPTIC OPTIC XCVR XCVR communication is not formatted and straight serial commu- 'A' IBI nication is possible. As an example, if the address of an 181 _J is 1111 and the address of the sending device is 0000, then _J the null message would be: K11100008WD LOOP LOOP AC POWER On entering the open state the ISI sends a formatted reply OUT IN Of. #000011110PENOWD. Figure 5: Block diagram of the Intelligent Sensor Interface. In the open state, the two devices will be able to com- municate on the LAN in a half- or full-duplex mode depend- and high-voltage shunt. Finally a 2.1 A-hr internal bat- ing on the connections to the LAN. ISI software provides tery protects the ISI from line surges and fluctuations and a menu oriented operating system that allows the operator maintains operation in -the case of external power failure. to change software parameters, test seasor operation, read The fiber optic receivers and transmitters are made by status or down-load data. From the open state it is possi- Hewlett Pac1card Co. The transmitter (HFBR-1404) con- ble to re-load software or change code. Since ISIs are often tains a planar 820 am GaAs emitter and the receiver (HP placed in un-sheltered locations, this is a major advantage. HFBR-2402) incorporates a monolithic photo-IC which con- The open state ends if. (a) a command is given from tains a photodetector and DC amplifier. All fiber optic con- the command device; (b) a # occurs on the LAN; or (c) a nections are SMA connectors and the fiber optic cable used BREAK occurs. The BREAK is a pulse of light with a width is 50/125 ism glass fiber. between 120 msec and 150 msec. It causes a complete LAN The fiber optical power budget is as follows: The optical reset. transmission power into the cable is rated at - 17.6 dBm. The receiver requires at least - 11.2 dBm to operate reli- 2.2 Intelligent Sensor Interfwes ably. Thus about 6.3 dBm surplus light is available. At -40 OC temperature this figure is degraded by approximately 2 A block diagram of the major components of a Coastal dB. Climate Company ISI are shown in Figure 5. The major The fiber optic connectors degrade the signal by about components are the two fiber optic interface circuits, tran- 0.5 dB each and feed-thru connectors degrade the signal by sient relief and power regulation circuit, digital interface as much as 1 dB. The attenuation in the So pm glass fiber is circuit, analog-to-digital (A/D) circuit, CPU, and power less than 5 dB/Km. Our longest run will less than 0.2 Kin. supply. Thus we expect a surplus light power of 3.3 dBm typically The ISI can be powered in three ways. An external 9-18 and 1.3 dBm in the coldest conditions. volt DC power source can be connected to terminals in the The fiber optic connectors are soldered directly to the sensor input connector. Typical current drain is less than fiber optic interface printed circuit boards and protrude 75 mA including fiber optic communication. Standard 106@ through the wall of the 181 housing. 0-rings seal the holes 115 volt AC power, 50-60 Hz, can be supplied via a separate and provide mechanical support to the connectors. AC power connector. The power fine has 'hot', 'neutral', ISIs can accept 3 different types of data signals: ana- and ground terminations. Each ISI has power protection log voltages, frequency variable sine or square waves, and capable of withstanding 2500 V Wpoecoud voltage tram- RS232 serial signals. Resident software in the 181 deter- sients and thus will withstand all expected power surges. mines how the data signals are sampled and converted to The equipment should be safe if operated with the ample final numbers. Data are stored as fully calibrated floating standard power although if protected power is available it point numbers. In this way complicated data manipulations 'o should be used. The transient protection circuit combines at the CCF are reduced and overall flexibility is enhanced. capacitors and protection diodes as a combined RF filter Two identical frequency circuits are provided. Each is 1563 designed to read the AC signal from an R.M. Young wind nal temperature. monitor. The wind monitor output is a sinusoidal voltage A setup is a particular assignment of values for the spe- whose frequency is proportional to wind speed aud whose cific parameters which controlsainpling and averaging in a amplitude varies from about 10 mV at near-stall. speeds to particular ISI (parameters 1-5, see below). The user defines about I V at high winds. When other types of inputs are one or more setups for all ISIs in the DBMS. Setups can be required, the input buffer circuitry is easily modified. given names such as 'one-minutel or 'on-station'. A scheme There are eight analog inputs. The number can be ex- is a particular group of setups for one or more of the IS19 panded if necessary as the A/D circuit boards are otackable defined in the ZAN DBMS. The scheme can be defined in and addressable. The A/D circuit uses an ADC1205 12-bit the DBMS by a uniquename. One or more schemes can be A/D integrated circuit. A frei&runuing oscillator provides a defined. conversion clock to the circuit. Reference voltage is gener- A DBMS table allows the user to schedule when dif- ated by a precision reference and voltage-follower amplifier. ferent schemes are to be initialized. This feature is one of The analog input signals are buffered in low-paw filters and the most powerful aspects of the ZAN DBMS. It is possible multiplexed under CPU control. to partition a complete cruise for satellite overpasses, in- There are two serial RS232 inputs: COMI and COM2. tercomparison times, and less intensive periods in advance. COMI is used by the LAN and optionally by the termi- All current parameter values are collected from each ISI nal. COM2 is available in the sensor connector and can be whenever data are collected. used for any variety of inputs. One example is to operate Certain of the parameters have specific functions in the a digital barometer such as the Paroscientific Digiquartz Coastal Climate ZAN: Parameters Pi-P5 are used exclu- barometer. The COM2 port has four signal lines: trans- sively for the sampling schedule. Parameter P23 contains mit line; receive line; Data Terminal Ready line (DTR); the ISI four-digit address. Parameters P29 and P30 contain and ground. The DTR line is a bi-directional 0-5 V high- the date and time of the initiation of the present scheme. impedance digital i/o signal line. The CCF Controller tranders A-Ph and Pn-& to each ISI as part of a scheme initialization. Each time the controller 2.3 ISI Software collects data from an ISI it also reads back the entire pa- rameter list and compares the date-time to the date-time of The above descriptions of ZAN protocol and the ISI the current scheme. If they do not agree, then the data will hardware provide a nucleus of an operational network. The not be logged into the data base. It is felt that this crow specific software in the ISIs determine how they will be checking is essential to insure a maximally un-corrupted able to function as an integrated system. ISIs are capable data set. of running distinct sets of software. In fact any intelligent Another important concept in the ISI-DBMS interac- device such as a PC can operate as an ISI as long as it tion is the concept of fields. The ISI stores computed statis- follows the few simple guidelines. tics (samples, means, standard deviations) in memory as ISI commands follow a menu oriented syntax. There individual ASCII lines. Each line begins with the date and are four menus: main, data, parameter, and tesL They are time of the Ime, followed by a string of data separated by designated by P, D>, P>, and T>, respectively. The menu commas. The line is terminated by a carriage return and domain is entered via the main menu. If you are in a par- Due feed. The comma separators define data fields. Each ticular menu and want a command from that menu, simply field for each ISI is defined in the DBMS. As the data are enter the request syntax. If you want a command from an- collected they are stored in the DBMS measurement ta- other menu you must first transfer to the correct menu by ble according to their definition in the DBMS. All data are typing the menu letter. Commands can be concatenated stored in association with the scheme, parameters, time, into one long command string but menu jumps must be and ZAN setup that was current at that time. made correctly. Communication between a terminal and the ISI is possi- Fundamental to the operation of ISI software is the con- ble over the fiber optic cables or with the terminal connec- cept of parameters. Each ISI maintains a parameter array of tion. To speak directly to an ISI over the LAN simply plug numbers (Currently there are 30 32-Nit numbers possible.) a terminal or computer into the fiber optic interface box. that can be used to control timing, sampling options, or Either the A or the B loop can be used. A terminal can data conversions in the ISI. They can act as flags signalling be programmed to send formatted commands to individual the presence of certain sensors. Parameters are designated IS19 or it can mud a null message and open up a particular by the symbol A, i = 30. ISI for unformatted communication. With a terminal you Parameters are read and written in the parameter menu. will be able to: (a) set calibration coefficients; (b) set ISI Parameters are altered in RAM memory with the syntax addresses; (c) sample all sensors for calibration or check- Pa/m where n is the parameternumber and z is the desired out; (d) set the clock; (e) set sampling duration, interval, value. Parameters can be stored permanently in the CPU and time or; (f) re-program. the 181 partially or completely. EEPROM using the command E. Eaampk: The message string P1/20ETI changes param- 2.4 Central Computer F"ility eter 1 to the value of 20, stoma all current parameters in EEPROM, then moves to the test menu and requests inter. The complete computer system is called the Central 1564 Computer Facility (CCF) and the Controller is the part information in a relational data base so that any data can that maintains communication with the ZAN devices. The be related to other data through defmed pointers. The Controller can be a piece of hardware in a distributed pro- DBMS provides a quickly assimilated hierarchy of menus cessor system or it can be a background procedure in a mul- with which one can configure the network, schedule schemes titasking computer. The CCF described here uses the latter and VMS procedures, and print out reports of past data. approach with a MicroVax 2000 computer and peripherals The window/menu technique though somewhat slower to (Figure 6). Our CCF software can be implemented in AT, manipulate then direct commands, provides an easy sys- IBM mini, or HP environments as well as Digital Equip- tem for occasional users. ment Corporation's VMS. For more sophisticated reporting and data rnanipula- The software in the CCF is a combination of contin- tion, the advanced user must become familiar with the uously running system programs and prov-Ams which are FOCUS database command stractum With this he will called by others on a predetermined schedule. Figure 7 have immense processing capability including most statis- shows the interaction of the different components of the tical procedures. Alternatively, the user can write out se- software. Rectangles represent batch jobs that run concur- lected data to ASCII archival files on tape or hard disk and rently and interact with the DBMS. Tilting parallelograms process them with individual software packages. represent processes that are initiated, run to completion, Great thought and care was given to DBMS architecture and terminate. These processes are usually called as VMS with the following guidelines: (4) The DBMS must allow jobs by the VMS Controller. Disk files are represented by complete flexibility for the addition of diverse instrumen- the cylinders. tation and sampling characteristics. As an example, occa- The Data Base Management System (DBMS) is the sional data from an expendable bathythermograph should heart of the ZAN CCF. Of several options we have chosen be as easily entered as data from regularly sampled air tem- the FOCUS software package from Information Builders perature sensors. (b) The DBMS should provide complete Incorporated for the nucleus of the DBMS. FOCUS stores configuration control and archival. The ZAN configuration at any moment must be part, of the DBMS and all data in the DBMS must be associated with the ZAN configuration at the time it was measured. This includes hardware serial numbers, parameter settings, calibration coefficients, and, of course, time. (c) The DBMS must provide for the casual Solar t Radiation user and the nou-sophisticaWl user. We decided on a menu oriented multi-table DBMS which leads the user through all 15 steps of the system operation. If this is frustrating to an J Isi J ISI ISI J-] a A 8 A @t @ @t @t Barometer Debug ZAN cent Orr Orr DIST S TERN JB BOX out: Log F11 C-z'!-'@o ut Ir FOREMAST JB R ISI J Da a Ethernet End Node WC ov SHIP ims Dec Server V Wofr VMS FOCUS Contra[,- IMS Doris Collector Nnter 0woff F _P1-_tt-_] L 3FW03/ M3 AC POWER Figure 7: Block diagram of the Central Computer Facility software. Rectangles are continuously running programs; slanted parallelograms are called procedures; cylinders rep- Figure 6: Sketch of ZAN installation on the NATO Re- resent disk files. Solid lines are lines of control and hashed search ship RIV ALLLANCE. fines represent data flow. @[email protected] tt- 1565 expert user, they may work directly with FOCUS command access to the network. The user can communicate directly language. However, the scientist who comes on board once in unformatted text or he can send and receive formatted in six months or the ship technician win find the menu op- messages. Debug allows direct contact with each device on eration with its checks and cross-checks infinitely usdul. the network. There are actually two controllers in the VMS system; Hourly coded meteorological surface observations are the ZAN Controller which up to now we have called the prepared by a sophisticated melding of all available data. Controller, and the VMS Controller. The ZAN Controller The code conforms to the World Meteorological Organiza- communicates messages onto and receives responses from tion (WMO) synoptic code for ships at sea. The software the ISIs on the LAN. Messages for the LAN are stored in that does this is called SFCOBS. SFCOBS uses DBMS ap- the schaste fde. The schedule file is a text list of times plication routines to create 5 data files for the past ten and activities for the next several houm The schedule file minutes of data from the database. Files 1-3 are meteoro- is routinely updated by a VMS procedure called in the VMS logical measurements from three different locations on the schedule file. In this way the process is self-perpetuating. ship. Three differeut sets of instrumentation can be defined An activity might be a command menage for a particular with a DBMS table. Thus there can be a 'port-, 'starboard', ISI and it might be a global command for all ISIs. Data and 'stem' location of instruments. The DBMS allows the received by the Controller are stored in the LAN data file. operator the flexibility to specify each sensor and also to The ZAN Controller codes out-going messages with ad- designate for what relative wind directions each sensor is dresses, checksums, and staA/stop characters. It decodes valid. (i.e. Ignore a sensor when it is leeward of an obsta- incoming messageL The Controller handles collision errors, cle.) File 4 is the IMS data collected during the past ten LAN faults, and other management functions. minutes and File 5 is information from the SFCOBS setup Routinely a VMS procedure loads the data into the data table of the DBMS. base. When the capture mode is on, all LAN communica- The process of melding the four time series into one is tions are also stored in the Capture file. If a received re- involved. AD time series are filtered and resampled onto a sponse does not check out, the Controller will repeat the common one-w%ond sample period. All valid relative wind request. The operator can select how many times the Con- directions are reduced into a single direction for each second troller will try to get a good menage. Currently, with one and these directions are used as criterion with the DBMS repeat try a 99% data recovery is achieved. If the LAN table to compress the three time series to a single 'best' is broken, the Controller will send and receive on different relative time series. Next, the ship speed and directions and loops and notify the operator of the problem. the relative winds are used to compute a true wind vector The VMS Controller operates in a manner siM*1 to time series. Finally, the ten-miuute second-by-second final the ZAN Controller and allows system batch jobs to be time series of all meteorological variables are averaged and scheduled and run via the DBMS. A pnxedure is a VMS job the WM0 weather code is produced. that is initiated and runs to completion. Procedures can be scheduled in the DBMS. In this case they will be entered 3. R/V ALLIANCE INSTALLATION into the VMS schedule file and then are called by the VMS Controller. They can be run on request by a call directly We have recently installed a ZAN system on the research from the DBMS. Procedures may be command language vessel ALLIANCE The ship is operated by the Undersea film, executable files, or a combination. Users can write Research Centre of the Supreme Allied Commander At- and execute their own procedures using the DBMS com- lantic (SACLANT), located in La Spezia Italy. ALLIANCE maads table. Thus plotting, report writing, and archival is 93 m long, displaces 2450 tonues, has a sustained speed activities can be run routinely by the system. The VMS of 16.3 kta, has a crew of 10 officers and 17 crew, and sup- Controller reads the VMS schedule file for time and ac- ports a scientific staff of 23 persons. Figure 6 is a diagram tivity. At the appropriate times, the Controller launches showing the ISI locations and sensors. VMS batch jobs or processes. Jobs can run periodically For ALLIANCE, ZAN is sub-divided into seven major (1-minute, 5-minute,...,hourly, 3-hourly,. daily), 01, M sub-assemblies: (1) Foremast sensor mast; (2) Foremast ISI quest, or 1-time only. plate for starboard sensors; (3) Foremast ISI plate for the We define the ship's Information Management System port sensors; (4) Stem sensor mast; (5) Stern ISI plate; (6) as a single computer port which provides all current ship op,- Lab ISI plate (a spare system); and (7) Central Computer erational information: position, ship speed, heading, gyro Facility. information, and engine status. The IMS collector software Figure 1 is a photo showing the foremast mast assembly periodically reads data from the IMS and formats them for and 181 plate& The foremast provides an excellent platform the DBMS as though they came from an ISI on the LAN. for measuring surface layer conditions in uncontaminated An ISI could perform this task in the case when an IMS is air over a broad forward sector (relative wind directions not present. from 230* to 1201. Instruments at the top of the foremast The Debug software allows the operator direct control of will be appraKimately 16 m above the sea surface. the LAN from the CCF monitor. The Debug program first The foremast sensor mast assembly includes the mast turns off the ZAN Controller and then allows the user direct structure, the sensor mounting hardware, and the sensor 1566 asseniblies. The foremast mast is mounted on the foremast pedestal. It holds the port and starboard meteorological sensors and the radiation sensors. A completely redun- dant set of instrumentation is installed on the foremast for maximum reliabiflity. The short-wave downward radiation measurement is not redundant and is measured with the starboard ISL WIND MONITOR Each 'plate' assembly is a heavy aluminum plate on which are mounted the ISL sensor junction box, and any additional item for that particular location. The LAB ISI is located in the ship laboratory. At this time there are no sensors attached to it. It will serve as a spare ISL Sensors can be attached to it over time and the data will be collected into the data base. ,The wind monitor and temperature-relative humidity HousING sensors are combined into a single interchangeable mod- tF FILTER ule mlled the WEATHERPAK. Coastal Climate Company manufactures WEATHERPAK in a vaXiety of configura, FIMI tions of intelligence, memory, and sensors. The model used SHIELD in ZAN combines an R.M. Young wind monitor and a mod- ified Rotronk tempemtureRR sensor into a single module UCK DiXNNECT lie with transient protection and radio frequency filtering on BASE !4. all signal and sensor power connections. Figure 8 is an exploded view of WEATHERPAK. All components of the -SOLAR system use sealed connectors and can be removed or re- RADIATION pLued with little trouble. SCREEN Air temperature and relative humidity (RH) are mea- sured with the Rotrouics MP-100 hygromer sensor. The air temperature is measured by two different sensors. A plat- inum RTD sensor is used by the RH circuit for temperature Figure 8: Exploded view sketch of the WEATHERPAK compensation but while the RTD has excellent long-term sensor package. The wind monitor plugs directly into a con- stability, the linearization circuitry used by Rotronics is nector with o-ring seals. The temperature-RH sensors are only accamte to about �0.8*C. For better temperature ac- covered with a teflon particle filter and the entire electronic curacy, Coastal Climate Company has modified the MP-100 package is shielded with a metal cover. With the Kamlok by adding a thermistor bridge circuit to the MP-100. The clamping base the unit is ea@ily removed or replaced. thermistor is the Yellow Springs Inc. (YSI) bridge Model 44212, interchangeable to 0.2 -C . Complete electromagnetic shielding is provided with the WEATHERPAK The Rotronic T-RH sensor is mounted tion coefficients for the barometer are stored in the sensor inside a metal tube and then inside the radiation screen. EEPROK and it can be altered with a computer connec- All sensor signals and power lines are protected from tran- tion. In all barometers the most typical error will be an sieuts and LM by circuitry including resistor-capacitor fil- offset; the unit will track pressure changes correctly, but ters and ferrite toroids. Lightening and corona transients will be offset by a constant amount. We have provided an are shunted by transient protection Itransorb' diodes.@ ISI parameter to correct for oftet error in the barometer. The WEATHERPAK mounts to a special clamping base Thus it is easy to correct the barometer during a cruise or which incorporates the unique 'Kamlok' fitting. The Kam- in inclement conditions. Jok fitting was originally designed for high pressure hoses in, Solar insolation is measured at the foremast starboard fire fighting and tanker industries. Its cam-shaped locking location. The black-and-white pyronometer has a conver- arms press the mating insert down onto an internal gas- Sion gain of 10 1&VoltS/Wm-2 and in fall Sunlight the sensor ket making a water-tight seal. We have added an internal output will be about 10 millivolts. Preamplification by a cylinder with additional o-ring seat to the Kamlok fitting. factor of 200 is required to convert the sensor output to the The internal cylinder supports a 10-pin connector that con- nominal 2 V input required by the ISL A very low noise in- nects the WEATHBRPAK to the ISL AD electrical wiring strumentation amplifier, MAX420, Is Installed in the fore- is entirely enclosed in metal tubing, either in side the mast mast mast just under the radiation sensor. By amplify- or in flexible coakd metal conduit. ing as close as possible to the sensor and keeping electrical WINI 1i Barometric pressure is measured with a Paroscientific wires inside the mast and conduit we have eliminated any intelligmi pressure sensor at the stem location. All calibm- detectable noise in the data. 1567 4.0 CONCLUSIONS AND FUTURE DEVELOPMENTS ACKNOWLEDGEMENTS The ZENO ALLIANCE Network was developed so that The ZAN system concept has been on these author's a network of ZENO data collection platforms could be cou- minds for some time but it required an act of faith by the nected together in a peer-to-peer local area network. We scientists of the SACLANT Undersea Research Centre to use fiber optic connections because it provides noise-free bring it to reality. In particular we appreciate the help reliable operation athigh baud rates over long distances. and effort by Dr. Peter Munett, the chief scientist for The installation on the RIV ALLLANCE met or ex- the project. The chief engineer Mr. Frederico, De Strobel ceeded all expectations. We routed the fiber optic cables and his technicians Roberto della Maggiora and Antonio in two days. The fiber optic connectors were installed after d'Augustino provided much help in the installation. the cables were pulled through the ship wireways. They were made in place; on the deck or the lab. Ship's techni- REFBRENCES cians learned to make the connectors in a couple of hours. Each connector was assembled in less than 20 minutes and Atallah, Deif N. (1988) Peer-to-peer protocol facilitates real- the failure rate was about one in ten. time communications. fflectronie Design New, Aug 18, After the entire network was confipred and operating 179-186. we added the fourth ISI in the laboratory. This involved only making the fiber optic connections and editing the IEEE Standard ANSI/IEEE Std 997-1985. DBN5 configuration tables. The entire process was com- pleted in less than thirty minutes. The data have been completely free of ship induced noise. Even though the foremast mast is placed in front of two ship radars and one end of the ship's HF radio an- tenna connects just below it, we have not seen the slightest electromagnetic contamination. Any one who is familiar with ship instrumentation installations will recognize the threat these noise sources pose. We are just beginning to collect long data sets with this installation. During a test cruise the redundant in- struments on the foremast tracked perfectly. The tempera- tures were within 0.1 *C of each other and the RH sensors were within 3%. Vector averages of wind speed were within 0.1 ma-1 and directions within 2 degrees. The stern in- struments tracked ahnost as well but for obvious reasons differed somewhat. With any new development, there are enhancements that will be developed in the near future. Coastal Climate Company is actively pursuing the following improvements: � Expendable bathythermograph (XBT) reader ISI con- nects to an XBT launcher and logs each profile for entry to the DBMS. � ISI interface to an infrared detecting sea surface tem- perature sensor. � ISI display device. This unit will connect into the network and collect data for display. The display will be able to interrogate any ISI or the CCF for data to display. The display will be a plasma display with touch screen control for operator convenience. It will be water resistant and can be placed in unsheltered locations. We believe that ZAN is an ideal extension to the SAIL principle. It is time that all research ships begin using a common data management in an organized effort to up- grade the quality and the availability of ship data. ZAN is a system which will provide this needed next step. 1568 ENVIRONMENTAL CONDITIONS IN NEW YORK BIGHT, JULY-AUGUST, 1987 Robert E. Dennis, Richard P. Stumpf, and Martin C. Predoehl Assessment and Information Services Center National Environmental Satellite, Data, and Information Service National Oceanic and Atmospheric Administration conditions. Several stations experienced ABSTRACT drought conditions during this period. The exceptions were Kennedy Airport at the apex of During the summer of 1987 the coastal area the New York Bight, which had near-normal or of New Jersey experienced short periods of excess rains, and Atlantic City and Baltimore, local and regional anomalies in coastal and which received at least 25 percent more than offshore characteristics of water and flow normal amounts of precipitation during July. during the period of late July and early August. Intervals of warmer-than-normal 3. WINDS surface waters occurred, associated with slightly higher-than-normal salinity. Short Wind data for the area are available in periods (up to ten days) of wind direction several forms. Climatological analyses provide blowing from the northeast and onshore may monthly means of temperature and winds averaged have created conditions which would favor over large areas. The National Climatic Data onshore transport. Large-scale systematic Center (1987) publishes Local Climatological abnormality of the environment in the New Data giving daily values as well as summaries Jersey coastal region is not evident. for each month for each NWS station. The predicted wind field from the NWS Limited-area Fine Mesh model archived at NCDC provides still another estimate of the surface wind values for 1. INTRODUCTION the region, particularly over the ocean. To describe the. environmental situation Observed Winds along the mid-Atlantic coast during the months Long-term average surface winds in this preceding events of interest in August 1987, we region flow from the south and slightly offshore utilized selected standard information collected during this period of the year (Williams and by NOAA through three of its Line Offices: Godshall, 1977; Bishop, 1980). South and observed and modeled winds from National Weather southwest winds will produce offshore movement Service (NWS), water temperature and salinity of surface waters along the New Jersey coast. from National Ocean Service (NOS), and satellite Williams and Godshall give only July data in sea surface , temperatures from National this summary form. These authors have rated the Environmental Satellite, Data, and Information summer constancy of the winds near 40 percent, Service (NESDIS). which is relatively steady compared to the spring or fail periods and is slightly less 2. AIR TEMPERATURE AND PRECIPITATION constant than the values found for winter. Bishop displays both July and August summaries During June and July 1987, coastal stations in wind roses, whose mean values agree well with from Bridgeport, CT, to Cape Hatteras, NC, Williams and Godshall. experienced warmer-than-normal air temperatures. In August this regime changed with stations NWS observations from Atlantic City, NJ, north of Atlantic City, NJ, recording below- show the winds, as indicated by the 'fastest normal temperatures. In the middle region. mile', predominantly onshore during July 1987, including Maryland and Delaware temperatures then southward alongshore during early and were near normal, and to the south temperatures middle August 1987 (Figure la). The remainder remained above normal for August. Boston of August winds were highly variable, trending recorded lower-than-normal average monthly generally shoreward. These "fastest mile" winds' temperatures for the entire year until September probably include a strong measure of sea breeze. 1987, a pattern possibly related to the presence of a cold pool of water east and southeast of the New England states. A simple analysis cannot easily separate the wind induced by the general circulation from the From June through August the area effects of sea breeze generated by the experienced predominantly drier-than-normal differential heating and cooling of land and 1569 United States Government work not protected by copyright ocean. During this period the land was 4. SURFACE TEMPERATURE AND SALINITY experiencing warmer-than-normal air temperatures and lower-than-normal precipitation. Such The NOS maintains observations of surface conditions may have contributed to a stronger- salinity and temperature at selected coastal than-normal evening onshore breeze, in turn stations. The data are published periodically creating a 'fastest mile' predominantly in the series Surface Water Temoerature and reflecting the sea breeze, not the general Density. Atlantic Coast, North and South America circulation. Studies have shown the perils of (Department of Commerce, 1968). The NOS has using land-based wind measurements over water updated these averages for station data (Overland and Gemmill, 1977). Other detailed requested individually. In the mid-Atlantic wind velocity measurements were not available Bight area stations at Montauk, NY, Ventnor, NJ, for the study. One must exercise extreme and Chesapeake Bay Bridge-Tunnel, VA, are caution in any attempt to apply these Atlantic sufficiently open to coastal oceanic influence City winds to infer movement of surface waters offshore. to provide information on events in the mid- Atlantic Bight. Modeled Winds The daily salinity for summer 1987 at both The NWS' Limited-area Fine-Mesh Model II Montauk, NY, and Chesapeake Bay Bridge-Tunnel, (LFM) forecasts winds for grid points over North VA, followed the normal seasonal march of America at eleven levels (Gerrity, 1977). salinity during this period. The values for Boundary layer winds in this archive are roughly Ventnor, NJ, however, did not follow the the lowest 50 millibars of the atmosphere and seasonal march and remained above normal include the effects of friction over the water. throughout the' summer. of the three coastal Values are chosen at six hours after model stations only Ventnor recorded monthly water initialization, in order that the winds and the temperatures consistently above normal during atmospheric pressure systems are in dynamic July and August. equilibrium and spurious transient surface wind patterns generated from the widely-spaced 5. REGIONAL OCEAN TEMPERATURE SUMMARY observations used for initial conditions will have damped out (Reeves and Pytlowany, 1985). Regional sea surface temperature data are summarized by NWS from ships' observations, The dataset comprises the most recent ten buoys, and satellite data, and are published years of modeled winds. Since wind observations monthly in the Oceanographic Monthly Summaxy over water are sparse, these modeled winds are (National Weather Service, 1987). Off the New the best available estimate of what the Jersey coast sea surface temperatures ranged instantaneous wind field was over the water, from near normal in June to 20C above normal in provided modeled data can be related to observed July and August, 1987. No large water data. Hess and Pytlowany (1987) studied the temperature anomaly was present in the New York statistical relationship between the modeled Bight, though water far offshore was below winds and observed winds for two stations in normal through much of the summer, and water .Chesapeake Bay. Modeled variables explained temperatures far south off the South Carolina between 50 and 71 percent of the variance, coast in August were higher-than-normal. hence, these winds represent reasonable estimates of the wind fields for this region. During August 1987, average surface temperatures over the region were mixed, with The mean monthly wind field from the LFM large areas above-normal, large areas below- model compares favorably with the climatology of normal, and much of the intervening area near- Williams and Godshall. No consistent anomaly normal. Along the coast, summer warming appears for the summer of 1987, nor even for a continued quite strongly in the Carolinas, where single month. However, examination of the daily water temperatures reached more than 40C above values of the modeled winds reveal patterns of a normal. week to ten days duration which created surface flow quite different from normal monthly South of Nova Scotia and southeast of New averages. England was a large mass of water nearly 40C The LFM wind displacement plotted as two below normal centered near 410N, 660W. The percent of the wind speed shows two periods when depth of this cool pool is not available without winds were northeasterly, from 05-08 August and additional shipboard measurements. This pool again from 12-16 August (Figure lb). Northeast persisted from July into September. Warming in winds normally will create conditions of onshore the Gulf of Maine continued, however, and in surface drift. Coincidentally garbage washed September the warmed surface waters extended onto the New Jersey shore during each of these throughout the Gulf and south of New England, periods. Climatological data for the region effectively separating the continent from the suggest consistent northeast winds during the cold pool offshore. late summer is unusual. 1570 7. SURFACE CURRENTS Hess, Kurt W., and Peter J. Pytlowany. Seasonal mean surface currents in the region Predicted Winds for Chesapeake Bay from LFM are fairly constant in the nearshore region and Observational Data. NOAA Technical Memo- within 30 kilometers of the coast (Bishop, randum NESDIS AISC 11, 1987. 1990). Ship current measurements and surface Huang, Norden. A simple model of ocean surface drifter computations indicate a mean southward drift current, in The Physical Behavior of Oil surface drift in the nearshore region during the in the Marine Environment. MIT Press, 1979. summer months. The rates are on the order of 10 cm/sec. Farther offshore the drift is highly variable due to both shifting wind structure and National Weather Service. Oceanographic Monthly movement of eddies and Gulf Stream meanders Summary. National Oceanic and Atmospheric through the area. Administration, U.S. Department of Commerce, 1987. The NWS Limited-area Fine-Mesh Model (LFM) output for the wind field are applied with the National Climatic Data Center. Local Climatolo- equations by N. Huang (1979) to approximate gical Data, Atlantic City, New Jersey. surface drift and integrated mass transport. National Oceanic and Atmospheric Administra- Transport was shoreward over the entire Mid- tion, U.S. Department of Commerce, 1987, Atlantic Bight for the period of 09-15 August 1987, a pattern which is unusual at this time of Overland, J. E., and W. H. Gemmill. Prediction year. of marine winds of the New York Bight. Monthly Weather Review, 105(8), August 1977. 8. CONCLUSIONS Reeves, Robert W., and Peter J. Pytlowany. During summer of 1987, coastal New Je rsey Comparison of Boundary Layer Winds from NWS experienced short periods of local and regional LFM Output and Instrumented Buoys, NOAA anomalies in coastal and offshore water Technical Memorandum NESDIS AISC 2, 1985. characteristics and flow during the period of late July and early August. Intervals of U.S. Department of Commerce. Surface Water warmer-than-normal surface waters occurred, Temperature and Density, Atlantic Coast, North associated with slightly higher-than-normal and South America, C&CS Publication 31-1, salinities. A large-scale systematic Third Edition. abnormality of the ocean environment in the New Jersey coastal region is not evident, although Williams, Robert G., and Fredric A. Godshall. surface water temperature anomalies occurred far Summarization and Interpretation of Historical offshore in the ocean southeast of Cape Cod Physical Oceanographic and Meteorological (cold water anomaly) and well south off the Information for the Mid-Atlantic Region. coast of South Carolina (warm water anomaly). National Oceanic and Atmospheric Administra- The analysis suggests short periods (up to tion, U.S. Department of Commerce, 1977. ten days) of northeast winds may have created conditions which localized the appearance of floatables and garbage on the beach. Any debris floating offshore may have drifted onto the beaches when the wind-driven surface currents flowed onshore during two short periods of early August. Surface drift computations indicate the garba e could have reached shore between the 6th and X of August and again between the loth and 15th of August, from areas just offshore to as far as 50 km offshore. 9. REFERENCES Bishop, Joseph M. A Climatological Oil Spill Planning Guide, No. 1, The New York Bight. National Oceanic and Atmospheric Administra- tion, 1980. Environmental Science Services Administration, Coast & Geodetic Survey, U.S. Department of Commerce, 1968. Gerrity, J. F. The LFM model--1976: a documen- tation. NOAA Technical Memorandum NWS NMC 60, 1977. 1571 50 25 8 AUG 18 AUG 0 29 JUL Pi 20 JUL -26 -50 -76 -76 -60 -26 0 25 60 KILOMETERS (a) 80 1 1 so 40 30 -10 1@ 30 W 79 W 20 4 AUG t 0 0 20 JUL 30 JUL 9 AUG -to 25 JUL 19 AUG 14 A 6 -20 -10 0 10 20 30 40 50 80 KILOMETERS (b) Figure 1. Progressive wind displacement diagrams for a) Fastest mile at Atlantic City, NJ, July-August 1997; and b) LR4 winds at grid point located at 39*07'42" N, 72*22'49"W, July August 1987. 4 AU @@29 JUL 3 JUL 9 AUG 19 AU US 014 fA 25 < JUL 1572 DEVELOPMENT OF DEEP WATER TECHNOLOGY AS IT RELATES TO FUTURE SALVAGE Andre Galerne International Underwater Contractors, Inc. During World War II, the British used an old decrepit ship, the " Niagara" , to transport gold from Singapore to At a time when technology in deep water Australia. They felt the Japanese would research has made tremendous progress, never fire on an old ship. They were the author will review the options wrong, it was sunk and this led to a available in deep water archeology and salvage in waters beyond diver depth by-a treasure salvage in terms of survey, remarkable salvor, Sir Williams, t C) identification and recovery. retrieve a multimillion dollar cargo, This paper will cover not only diving but using an observation bell. also various other equipment including in Russian waters in depths of 850', tethered and/or untethered ROVs, using new technology and improved diver manned submersibles, and support vessels. capability , a salvage was made with great success, with an alleged profit of $80,000,000. Normally, the salvage process is an orderly progression of steps in which a salvor (1) acquires or performs research on a targetv (2) attempts to locate the The development of deep water technology target, (3) performs an initial and the imp'rovement of the surface verification, (4) attempts to establish support ship will undoubtedly thrust the rights, (5) launches a full-scale salvage USA toward a new salvage era. effort, (6) brings the cargo to market, (7) sells and distributes the proceeds. 16 more recent yeai-s, the most publicized salvage was performed by Mel Fisher in Research into potential targets will take very shallow water, maximum 60'. He many forms. .it may include (a) spent 17 years of his life researching, obtaining certain information concerning locating and doing the salvage. It was the circumstances of the casualty, its a very long and tedious job. approximate location and environmental conditions at the time of the loss, (b) Numerous individuals and groups have obtaining a cargo manifest, and (c) spent millions for this same type of determining the possibility of salvage research with some success. In fact, based on technical considerations, an there's been some very special but evaluation of the cargo's marketability, unpublished salvage work during the last projections of the economic return on the quarter century. salvage, a determination of the liklihood of a previous successful salvage attempt, After World War I, a Royal Navy ship and, finally, the possibility of "Laurentic" which was sunk by a mine near establishing rights to the salvaged Greenland, led to a salvage of very cargo. little publicity, bringing up 14 tons of gold worth $550,000,000 at current value Once a wreck's "file" is in place, the. in the form of 3,211 ingots, leaving only location phase may begin. Estimated 25 ingots behind. location offset by known wind, current, and tidal conditions can be marked off to In the 30's, a substantial amount of gold approximate the likely boundaries of the was salvaged from a SLInken ship, the ship's initial resting place on the ocean "Egypt.,, by Sorina, a private Italian floor. Using state-of-the-art remote company. The salvage yielded 5 1/2 tons sensing tools, such as side scan sonar, of gold and 42 tons of silver, worth a magnetometer, and sub-bottom profiler, total of $78,000,000 at current value. one can develop a hard copy image of the CH2585-8/88/0000-11573 $1 @1988 IEEE area in an effort to identify the target. Many legends also muddied the water (pun These surveying techniques are tied to not intended). For instance, legends of geodetic positioning references and a race of giants in South America deliver coordinates of potential targets reporting people to be 10 feet tall were to further investigate. common. No trace has been found of this race of giants. And legends of sunken The products of the above survey are used ships may, in reality, have been ships for an initial verification process in that had been mutinized and their cargo which the targets are investigated stolen. through the use of diving techniques, remotely operated vehicles (ROVs), and But we do know, in fact, that many manned submersible intervention depending important cargoes are on the bottom of on the specific circumatances. This the sea. The new age of electronics will operation is to identify the vessel give us the possibility to locate these and then to verify the cargo's manifest treasures. in both the qualitative and quantitative aspects with emphasis on completing; (1) Thousands of c 'argo ships sank during confirmation of cargo, (2) initial sight World War Il in the Atlantic. If you survey to establish optimum salvage plan, search for one, you will invariably find and (3) obtaining a sample of cargo or two or three others. Many have no artifacts to assist in establishing commercial value. Nevertheless, people rights and value. have always been interested in the adventure of the treasure hunt. Legal work will begin to establish rights and possible claimants (if any) to the Being in the diving business since 1946, salvaged cargo. This is sometimes a I must say that I am amazed at the lengthy process depending on the wreck's gigantic progress made i.n this industry. geographic location, reported cargo and The hard hat was king of the road for recent publicity. nearly a century but was slowly replaced by the invention of SCUBA. The first Upon the recovery of the cargo, it must patent talking about a self contained be evaluated from the aspect of its underwater breathing apparatus was by a archeological worth. Some cargoes must Mr. Denerouz in the late 19th century. be treated to avoid spoilage, or restored, depending an its contents. The In 1933, Commandant Yves Le Prieur cargo must be stored and valued while a developed a SCUBA system with a full face distribution network is established. mask and free flow system. The real breakthrough was the Cousteau-Gagnan Successful salvage depends a great deal system developed in 1942. This equipment upon the experience of the salvor and was a true revolution. Suddenly, man resources that are available to it. became free to swim and explore underwater, I've just narrated the ideal salvage procedure for a modern ship. But it's Today, divers using sophisticated not that simple. For years, I've been mixutures go to 1500 ft and hope to go working on developing a technique to deeper on hydrogen/oxygen mixutres which retrieve cargo in deep water. A salvor are currently being tested by Comex. trying to perform this type of task faces huge problems. The first one, but not Not only did the diver make tremendous the least, is to locate the wreck and progress but the appearance of the manned determine its cargo. If the wreck dates submersible brought our industry to back to the 16th, 17th or 18th century, another threshhold. However, it did get navigation being very imprecise and off to a rather, slow start. We had to communications being non existent, many wait for Perry to develop a relatively ships disappeared without a trace and small exploratory SUbmers ib I e. One of only vague information is found in the first was the PC38 which was used to old archives. If, after a long struggle locate the sunken atomic bomb off Spain. in difficult-to-read archives, one is I subsequently purchased the PC3B and successful in finding a cargo of used it on different jobs. It was really interest, there remains the task of a crude piece of equipment, and, in trying to find its location. Remember fact, I recently donated it to a marine that at that time maps were often a museum. national secret and the published ones were often misleading. In fact, if one From that first sub to the highly reads old books, some navigators reported sophisticated Woods Hole "Alvin" and sighting islands that we now know are non Ifremer's "Nautile"@ submersibles capable existent, of diving to 20,000 ft now are 1574 being surpassed by two new Russian this areav you will often find the cause submersibles built in Finland, and Japan to be the lack of a stable work platform. is preparing one to dive to 21,000 ft. These subs are capable of exploring 97% From another technological standpoint, I of the bottom of the sea. The remaining also can foresee the development of a new 3% to 35@000 ft is not really worth the generation of equipment able to locate excess cost needed for development of the and pinpoint wrecks with valuable cargo. necessary technology. However, a recent A highly sophisticated underwater sle ,d panel for NOAA recommended the was instrumental in showing Woods Hole construction of a 10,000 meter Dr. Ballard and Ifremer the first submersible for the 90's. pictures of the lost "Titanic". Manned submersibles are often limited to Many, many wrecks are buried and this new 1,000 meters. Very few have been built technology will make the difference in to go deeper and often they have been one being able to find them. And if you of a kind. An exception was the Canadian consider that many old wrecks are made 9f company, Hyco, which built a series of wood with bronze cannons which offer very Pisces. These subs were very useful in little magnetic deviationg You will developing deep water technology. For understand the magnitude of the needed five years, my Pisces 6 was on the technology. Something must be developed drilling ship "Discoverer Seven Seas" to find precious metal. There is some roaming the world making dives to 6,000 hope that the satelite can be of great ft. It was in the forefront of deep help. This is not a far fetched idea. water oil work development. Although, the development of this The ROV made a modest debut. The eyeball technology is not an easy problem to RCV 225 became widely used. Now there solve, there is a more difficult obstacle are so many different types that Frank t o overcome and that is the legal Busby, the man who tries to track and aspect. publish new developments in manned submersibles and ROVs, has a hard time After a wreck has been located, then the keeping up with the new equipment. legal maze begins. Maritime laws are very complex and not too well defined. From their original capability of an One can be in international waters and eyeball view, ROVs have progressed to still face the wrath of various groups; .50me very sophisticated equipment, like i.e. the French going back to the AT&T's "Scarab" capable of cutting and "Titanic" instigated protestation in the burying cable, digging trenches, and U S congress. Yet Peter Gimble diving on making surveys. These are becoming the the "Andrea Doria" stirred no protest. backbone of deep water research and (Our company supported his first movie an work. this vessel.) If a wreck is close to Other hybrid ROVs, like the Mantis, also shore within the 3 mile limit in state have the dual capability of being used controlled waters, permission is needed. with a pilot and/or as an ROV. Some can More red tape. There is now in the works perform these tasks simultaneously. a tentative law to extend the 3 mile limit to 12 miles. This will increase the power of the states. All this underwater technology has also been helped by the development of a new Salvors recognize and respect that class of ships - the SWATH - small water archeological finds must be handled with area twin hull. The SWATH made a very great care, but all underwater research slow debut into this market. I purchased can come to a halt with all these the former "Duplus", renamed "Twin restrictions. If this comes to pass, an Drill". Like the semi submersible entire segment of our past history will drilling ship, as long as the wave does be unknown to us forever. But this is not reach its flatatiOTI, thR ship is very another story. stable. There are only a few SWATHs in existence today. The Japanese have one -"Kascio". The U S Navy has three now under construction. I believe that in the 1990s, all research vessels will be of a SWATH design, providing a More stable work platform and more comfortable quarters. Many salvors underestimate the importance of the surface support equipment. In fact, if you look at the accident rate in 1575 ART AND TECHNOLOGY ON 20th-CENTURY VESSELS Carol Olsen Olsen and Associates - Art for Marine Architecture Part of the $14 billion Americans spent last year on pleasure boating was for hull carvings and graphics. New tech- nology now revolutionizes the ways in which vessels are personalized, posing new competition for the few remaining j"t - American shipcarvers. Based on a photographic survey from many ports, this paper shows how the American spirit is diversely expressed on a broad range of vessels, including military craft, commercial fishing and passenger I vessels, and pleasure craft. Contrast is also made with commercial vessels from other nations, which show some of the finest innovations in modern hull Some of the social, ornamentation. political, economic, ideologic, and artistic influences on vessel decoration are also considered. New materials and techniques have created better opportunities to decorate military, cargo, passenger, fishing, and pleasure vessels than ever before in the 5,000 year history of seafaring. Among the most innovative designs in recent years are the 25,000-piece glass mosaics on the bows of some Fred. Olsen ships in Norway (Fig. 1). The strength of these artworks was first tested for six Fig. 1. In 1961 M.S. BONNARD carries a months by placing samples of the glass and mosaic bow shield, made of 25,000 pieces adhesives at the waterline of an Olsen of glass secured to a curved steel Line ship sailing in frigid northern plate which is welded to the ship. waters, and success with those experiments Depicting the Impressionist led to commissioning the full bow work. painter Pierre Bonnard, the shimmering glass conveys the effect of an The Fred. Olsen Lines has commissioned Impressionistic play of light. (Photo: over 50 bow ornaments for their ships Fred. Olsen & Co.) since 1936; most have been bronze sculptures by the best artists available (Fig. 2). It appears that the immense added recognition of these ships because of their decorations has off-set the financial investment for the artwork. CH2585-8/8810000.1576 $1 @1988 IEEE In fact, throughout history, nations have W1_.11r_:,@_._,_17_-- =T,7, 77- IN often employed their best artists as naval sculptors. one of America's best examples is the bronze Victory on the U.S.S. L OLYMPIA, put on the bow after Commodore (later Admiral) George Dewey's victorious return from the Spanish-American War (Fig. 3 other particularly noteworthy American efforts include works by William Rush, who created figureheads for America's first Navy vessels in the late 18th-century; his work on early 19th-century American merchant ships caused artists in foreign ports to study and sketch his work.[1] Also, an extraordinary eagle, weighing 3200 pounds and with nearly a 20-foot wing span, was carved in 1881 by John Haley, Bellamy for the U.S.S. LANCASTER and is one of the masterpiece marine sculptures produced in our country. It is now on display at The Mariners' Museum in Newport News. Also the beautifully graceful eagle on the bow of SHENANDOAH, a passenger vessel sailing from New England, is one of A the best at sea today. In the 17th-century, court artists such as the famous Pierre Puget were partially responsible for decorating ships.[2) Jean-Baptiste Colbert, Finance Minister to Louis XIV, said, "Nothing can be more impressive, nor serve more to exalt the majesty of the King, than that his ships should be more beautifully decorated than any others at sea." (3] Fig. 2. Bronze sculpture by Gunnar Janson on bow of M.S. BOREALIS, 1948. (Photo: The importance of the appearance of Fred. Olsen & Co.) nations' ships.is still significant. When vessels gather in parades such as operation Sail 76, operation Sail 86, and for the 1988 Australian anniversary, Harman developed a logo for Destroyer photographs published in books and Squadron 23, based on the Indian character articles around the world not . only Little Beaver from the Red Rider comic document for posterity the quality of strip.[4] Thereafter this became known as naval architecture, but also the level of the Little Beaver Squadron. The design art considered acceptable at the bow. was painted on the side of the bridge wing on the CHARLES AUSBURNE, and Harman's It should be a matter of national pride original artwork is in the Navy Yard for the best artwork possible to be on museum in Washington, D.C. United States ships such as the U.S. Coast Guard's EAGLE which so frequently A very animated look is achieved by the represents our great nation in ports painted shark's teeth and eyes on the around the world. submarine U.S.S. TORSK, docked in the Baltimore, Maryland harbor. According to one of the most interesting Coast Guard Henry Lingenfelder, Director of the TORSK figureheads was on the cutter BEAR. now on display to the public, this echoes Before becoming a Coast Guard ship, this the shark mouth design used on the Flying was the vessel on which naval officer and Tigers P-40 fighter planes, which were polar explorer Admiral Richard E. Byrd flown by the American Volunteer Group in sailed, and her original polar bear 1941 and 1942, under the command of Claire figurehead is in The Mariners' Museum L. Chennault. TORSK has carried this collection at Newport News, VA. design since at least 1970.(5] In some cases, it has been appropriate for Interestingly, this popular motif is also military vessels to have more playful seen on a recently developed low-cost ROV decorations. For example, at Admiral (6], as well as a Dutch-built Arleigh Burke's request, cartoonist Fred passenger-carrying hydrofoil vessel.[71 1577 77,,@' '. 0,;77 TsE@@ .47 W- often one sees painted graphics. In 7@ , , - C Seattle's Fishermen's Terminal they vary OW for example, from images of Pacific Northwest art to jumping dolphins to the logo of the Seattle Seahawks football team. Images that people choose for hull decorations usually have important meanings to them. For example, oyster fisherman George Carmines of Poquoson, 4 Virginia, had a modest little portrait of his wife's grandfather painted on the boat that he named for the fine old man, Capt -14 A& died, and it shows him puffing away on a Henry. The artwork was added after Henry 'irk cigar, one of the vices he loved but was Fig.3 Classical bronze figurehead on told to avoid for health reasons. Captain Cruiser U.S.S. OLYMPIA, turn of 20th Carmines' voice softens to admit that some century. Photo in National Maritime of their favorite times together were long Museum, San Francisco. hours spent smoking and talking about the sea and those with whom they share it. Surprising to some perhaps, issues such as Tribute is often paid to heroes through water pollution have influenced boat boat decorations. For example, the decorations. For example, to help raise Seattle, Washington commercial fishing funds to clean up the Hudson River in New vessel GJOA is named for the exploration York, the day-sailing passenger vessel vessel upon which the Norwegian explorer CLEARWATER hosts concerts by singers such Roald Amundsen made the first voyage as Pete Seeger. One of the volunteer crew through the Northwest Passage and around members, Pete Lintene, carved one of his the northern Canadian coast. On the bow favorite birds, a Canada goose, as a is painted a Viking; the owner himself is figurehead, which he describes as Ila Scandinavian. symbol of a free, healthy environment."[8] And in Virginia, sport fisherman Curtis Elsewhere, commercial fishermen and their Reeves considers Francis Marion one of the families create some of the most colorful great American heroes; his boat carries decorations to be seen anywhere, for the Marion's nickname SWAMP FOX, earned not blessing of fleets held along United only for his ability to successfuly engage States shores. The work involved is 'British troops, but particularly for sometimes monumental, taking weeks to avoiding them by hiding in the South prepare large numbers of elaborate paper Carolina swamps. A cartoon graphic on the flowers, papier mache seahorses, cardboard transom shows a fox riding a fish, a cartoon characters, and other imaginative combination used merely because the owner displays. Galveston shrimper Johnny is also one of the millions of Americans Patane has covered his boat with Mickey who also greatly enjoys fishing. Mouse and other cartoon characters in recent years, and Joel and Cheryl Coward Forbes magazine recently reported that in Kemah, Texas took first prize in 1984 approximately $14 billion was spent on for a dragon figurehead that breathed pleasure boating in 1987. Part of that smoke; it was rigged to a C02 cartridge. revenue covered paintings (Figs. 4, 5), sculpture (Fig. 6) and vinyl-cut names and graphics for boats. On Stonington, Connecticut lobster boats, there have been excellent paper sculpture At least a dozen shipcarvers can still be renderings of Mother Goose subjects such found at work in the United States, as the Old Lady Who Lived in a Shoe, and producing hand-carved nameboards and other Humpty Dumpty Who Sat on a Wall. Fleet work in the long-standing tradition of blessing decorations have also lightly this craft. Also, as in the past, some nodded toward national interests; in 1983, are itinerant, while others prosper in one Galveston, Texas fishing family established locations; both do work other created a Shrimp Shuttle in the image of than shipcarving for their full income.(9] the Space Shuttle. The success of those who still survive in When the paper decorations come off, the slim shipcarving trade is challenged a permanent carvings may be found on some little more by the approximately 5-year fishing boats, such as the Chesapeake Bay old revolution in computer-aided graphics skipjacks which have trailboards showing for boat lettering and hulls. Today drawn American flags, eagles, and cannons. More artwork can very easily and quickly be 1578 computer copied and duplicated . in States who now practice in 48-foot long, high-performance vinyl. This means that 2000-pound teak boats which are complete custom and stock graphics are speedily with carved dragon heads and tails. available at reasonable cost, and Spring-time competitions between teams boatowners usually can install the work from New York, Virginia, Pennsylvania, themselves with very little effort.[10] Illinois, and elsewhere are held in Philadelphia and the winning team goes to The'range of boat decorations chosen by Hong Kong to compete in the Interhational Americans is fascinating to consider. A races in June. At both the American and few examples are the carved wood portraits Far East events, the competition is begun of twin baby daughters, Sarah and Molly with a colorful ceremony conducted by a Morfew, on MEKA II in North Carolina. on Taoist priest, to open the eyes of the the Florida-based yacht named LUCKY DUCK brightly painted wooden dragon figureheads belonging to Mr. and Mrs. Philip Ross, and bring good luck to the Dragon Boat there is a painted duck head as a family races.(11] v, A Id e 'j, 41-44-11- ,- 44,W Fig. 4. Custom painted graphics adorn many American pleasure boats. Custom and stock vinyl graphics are also increasingly popular because of the economy and ease with which they are prepared and applied. (Photo: C.Olsen) X; reminder of their children's favorite story, MAKE WAY FOR DUCKLINGS. Carved trailboards on the boat of Wendell Crosby of Beverly Farms, Massachusetts shows a Fig. 5. Decorative paintings on hulls special ivy motif; it had been one of the are a 5,000-year old tradition. This decorations at his daughter's wedding. work on the bow of T.T. BORGHOLM, 1959, Just for fun, Wilson Downs of Texas copied is by Tore Heramb. (Photo: Fred. a cartoon frog from a cereal box for his Olsen & Co.) boat GRIBBET, and the training vessel PILGRIM in California carries an image of Richard Henry Dana holding the text of TWO Art symbols express an attitude toward YEARS BEFORE THE MAST, the 19th-century life and the ideals one strives to attain. book written to acquaint the public with Malvina Hoffman wrote, in part, "The study the hardships of the common seaman. of art is the study of life... 11[12], and there are many reasons for decorations on Additionally, as Americans embrace the ships. They help in ship identification; maritime traditions of other countries, figurehead descriptions were, in fact, some of those decorations become part of part of official 19th-century custom house our heritage. Just within the last few records.[13] They increase the value of years, the International Dragon Boat Races a ship (14], and give added prestige to in Hong Kong have greatly captured the nations, businesses (Fig. 7), and interest of athletes throughout the United particularly to individuals represented. 1579 Perhaps more than anything else T, 'PT 7W W@rli oZ'W-' 9 -1 31 figureheads have been associated with good luck, both on ships and off. In fact, at the United States Naval Academy in Annapolis, Maryland, a replica of the 2 U.S.S. DELAWARE's portrait figurehead of A V the Indian leader Tamanend has long been called the God of 2.5 - the passing grade at the Academy; a new system has now A44 J P, changed the standard to 3.0. Before Z, exams, some cadets still pitch coins up to the Indian figurehead, considering it good luck if the coin stays up on the figure, rather than dropping off.[15) K N Ship decorations have also served as gestures of good will. In one instance, the 6100-ton German vessel HENRIETTE SCHULTE gave its approximately 6-foot long and 4-foot wide bow emblem bearing a city emblem to the City of Newport News, Virginia, in appreciation for repeatedly excellent harbor facilities and Z1, assistance. [16] one story of friendship though is Fig. 6. Bronze or carved wood figureheads particularly remarkable.[17] In 1891, the can be designed for all bow shapes. DICTATOR was coming up from Florida, loaded with lumber. Hit by one storm after another, she finally sank off of Virginia Beach, Virginia. The local people grieved that the captain's young wife and son were among those lost, and when the ship's wooden figurehead of a questions of marine management and woman washed ashore, it was set up to look policies and sea technology and out over the sight as a memorial. safety,[18] there is one area of the ocean's history that is always uplifting But decades later, due to more storms and to consider. The ways in which people the ravages of tourists and time, little have decorated ships is fascinating, remained of the original carving. Then an showing national pride, personal heroes, amazing thing happened. A whole new allusions to literature, a sense of humor, .generation of people at Virginia Beach and more, and new technologies suggest still felt it was important to have a -that there will be continued and fuller memorial, and arrangements were made for chapters in art,for marine architecture to Ornulf Bast of Norway to create a new more fully express the seafaring spirit of bronze figure to replace the original our people. figurehead. The bronze was erected not only in Virginia Beach, Virginia, but also For those who are working with today's in Moss, Norway, the homeport of DICTATOR. ships and who have the opportunity and Today the two figures are said to look out interest to develop some type of hull across the ocean to each other. decoration, whether a painted or vinyl Additionally, Moss and Virginia Beach are graphic, or a wood, bronze, or glass officially sister cities, and even have ornament, please feel welcome to contact had a student exchange program. Carol Olsen of Olsen and Associates, in Washington, D.C., a business that The story of ship decoration throughout specializes in art for marine history is immensely rich. if one architecture. This business was imagines that for about 5,000 years on the established so that the marine art world's lakes, rivers, and oceans, @raditions of the past, coupled with maritime cultures have each been innovations of the present and future, decorating vessels differently and in ways will be readily available to the maritime that express something of their social, community. Naval artwork is part of our political, economic, technological, living maritime legacy and it can be ideological and artistic uniqueness, then awe-inspiring, as one author observed: the scope of the subject begins to suggest itself. And when a ship with sails outspread Came up the bay like a bird awing Thus, as the world's scientists and Inanimate the figurehead leaders struggle today with critical Seemed yet a living, breathing thing. 1580 77 "4 alp, Fig.7. Company logos enhance the bows of some ships. Z, NOTES (1] H. Marceau, WILLIAM RUSH, 1756-1833 [13] See, for example, the Custom House THE FIRST NATIVE AMERICAN SCULPTOR records at the Newport Historical (Philadelphia 1937). Society, Rhode Island. [2] P. Norton, SHIPS' FIGUREHEADS (New [14] W. Baker, A MARITIME HISTORY OF BATH, York, 1976) 60-62. MAINE AND THE KENNEBEC RIVER REGION (3] Maurice Michael and Martin Grass, (Bath, 1973). trans.,THE SWEDISH WARSHIP WASA, (15] The Indian's quiver is solid, so it n.d. is not possible for the pennies to (4] Mr. John Riley, Ships History Branch, be pitched into it as some sources Navy Yard, Washington, D.C., have described. personal communication. [16] The Mariners' Museum, Curatorial [5] H. Lingenfelder, Sept.1988, personal records, OB9. communication. (17] Every March the Ladies Auxiliary to [6] R. Frank Busby, "Low Cost ROVs-Some the volunteer and paid Virginia Early Returns," SEA TECHNOLOGY Beach Fire Department commemorates Feb.1987, p.11. @he DICTATOR with a ceremony that (7] THE MARINE PHOTOGRAPHY OF PETER includes placing a wreath in the BARLOW (New York, 1972, Motor ocean. Also see William Foss, THE Boating and Sailing Books), p.106. NORWEGIAN LADY AND THE WRECK OF THE DICTATOR (Norfolk, 1970). [8] All information in this essay [18] "The Dirty Seas," TIME magazine, concerning modern boat decorations August 1, 1988, 44-50. is based on personal communication within the last decade with the boatowner and/or the artist. (9] In addition to shipcarvers already working, many academically trained SELECTED BIBLIOGRAPHY in addition to artists are available to do naval sources shown above. sculpture and would like to receive these commissions. Brewingtonf M.V.f SHIPCARVERS OF NORTH (10] The equipment most frequently used AMERICA (1962, Barre, Massachusetts) for boat lettering and graphics is Laughton, L.,G.C.f OLD SHIP FIGURE-HEADS by Gerber Scientific Products, Inc. AND STERNS (1925, London) [11] The U..S Dragon Boat Association is Pennsylvania Academy of the Fine Arts, currently headed by Mr. Jack Seitz WILLIAM RUSH AMERICAN SCULPTOR (1982, of Ambler, Pennsylvania. Pennsylvania) [12] M. Hoffman, SCULPTURE INSIDE AND OUT Pinckney, P.A., AMERICAN FIGUREHEADS AND (New York, 1939). THEIR CARVERS (1940, New York) 1581 MARITIME TRAINING AND OCEAN EDUCATION Samuel Teel Associate Professor Maine Maritime Academy, Castine, Maine INTRODUCTION ground in the practical skills needed to competently perform the jobs at sea. The second An intrinsic part of the future aspect of the education, and is undeniably the ocean and its what I referred to as part resources. The impact of the of the transition, has to do ocean environment on the with _-an ocean oriented civilizations of the world is education. This term implies taught to young students from an intentionally broad field the very start of their of study. The personnel who education. have been practically trained need, now, to go beyond Maritime education does exist wheelhouse and join more fully in the United States. The. in the scientific, political, education process in this field economical, and managerial has for many years followed the aspects of the changing marine industry. The key word being and ocean environment. used here is "followed". Could the old cliche apply in this Companies, organizations, case? Have we found ourselves societies, regulatory authori- being followers rather than ties and individuals who, in leaders? some way, are connected to the ocean environment are all I believe that the time has troubled by a common factor. come for educators to modify That is, where do you find techniques and adjust the. qualified personnel to work in curriculums of the maritime. and around the extreme education process. The environment of the world's academicians need to join with oceans? I suggest that the industry and design the students need to be taught the education of the next genera- practical skills of seafaring tion rather than waiting until and then provided with a the future is upon us and then selected series of classes and trying to catch up as best we topics that will enhance the can but never actually doing students' potential. By doing it. so the graduate becomes an asset, both to himself and to OCEAN EDUCATION the industry in general. Their horizons are broadened Maritime education, as it is to enable them to see the being applied and taught at multitude of factors affecting present, is in a transitional the international maritime state. Undoubtedly the first environment. requirement of the maritime training is to provide the student with a strong back CH2585-8/88/0000- 1582 $1 @1988 IEEE CURRICULUM age allotment of the various courses being taught to the At the Maritime Academy in undergraduates. Note that the Castine, Maine, we have Nautical Science courses undertaken the task of reflect a 50% contribution to broadening our curriculum and the entire curriculum. These going beyond the skill and are the true professional- task-based training that has license courses students take seemingly ruled curriculum during their matriculation. development in the past. They include topics ranging from Terrestrial and Celestial Once we targeted our goal, we listened to various sectors of Navigation to Vessel Stability the ocean oriented community, and Cargo Handling Procedures. one being the shipping compan- ies and associated services, Marine Transportation is shown and the other being maritime as having a 26% share of the and management colleges and curriculum. These courses universities. We observed and represent our effort to round researched the development of out the undergraduates' transportation oriented studies in ocean education. curriculums and, by combining Naturally, we have slanted our these two facets, were able, courses to reflect a transpor- during the early nineteen tation-directed series of eighties, to develop and classes, and each successive implement a marine transporta- course builds on its predeces- tion track within the standing sor. navigation and nautical science curriculum base. We start teaching Marine Transportation basics at the A priority was to maintain a beginning of the freshman strong practical-based curricu- semester in an introductory lum. This stems back to the level course. It is a current mission and objective of the overview of the maritime college which is, in essence, industry, what it consists of, to carry on Maine's heritage of terms, definitions, and the the sea. This happens to be basic concepts prevailing in the mission of our college but the maritime world. This I firmly believe that it needs course prepares the student to be the mission and directive for a detailed study of each force of all maritime education of the listed topics in in this country. In the coming greater detail during the spring we will be graduating college career. the first of these students into the industry. Each Maritime Shipping Finance is a individual has an expansive study of decisions and actions practical background in ship that affect the value of a operations, and also carries firm. Current and long-term with him or her a broad assets, liabilities, financ- understanding of the various ing, and capital structure managerial aspects of the decisions are considered, industry in its entirety. along with a study of finan- cial management in the Figure (1) depicts the percent- maritime environment. A 1583 SCIENCE AND MARINE HUMANITIES T. TRANSPORTATION 24 % 26 % t ZL NAUTICAL SCIENCE 50 % f igure 1 separate course has been review liability, risk, developed to study the theory protection and indemnity, and and case analysis of maritime salvage. This course is economic issues. Topics in complimented by an admiralty this class include vessel and international law course acquisition and disposal, which introduces the student chartering, and pricing along to international conventions, with the study of the govern- national legislation, and ment and international general maritime law. regulatory organizations and policies impacting the In addition to the topics and maritime community. course I have listed, the students will complete a An ever important facet of the series of traditional shipping industry is the undergraduate management management of human resources courses covering the basic and labor relations. In this management techniques common course the students analyze the to all industry, plus the more functions performed in organi- specialized aspects of zations which facilitate the maritime marketing management, effective use of people to seaport management, and achieve organizational goals. shipyard management. Maritime insurance and salvage We also offer a course in is an economic consideration of Managerial Leadership in which growing importance for the students identify the manage- maritime world. Our students ment skills commonly believed 15M, to belong to recognized leaders ADVANTAGES and exploring the issues concerning the same. It also I firmly believe that the should be emphasized that the graduate of Maine Maritime structure of a maritime college Academy, or of any other is regimented and, by its very college which teaches this nature, these skills are type of curriculum, is what is learned and practiced as needed to lead us in. the students live and work within coming developmental years as the college community. international advancements are made in all the various The remaining 24% of our disciplines that influence the curriculum provides the ocean environment. undergraduate with a solid Many graduates will find mathematics and science employment within a few months background, accompanied by a after graduation from institu- series of courses in litera- tions such as ours. There are ture, composition, and many opportunities for these oceanography, to name a few. young people. They will staff It is important to note that the vessels of the world, even within the science and operate merchant vessels, humanities departments an research vessels, and explora- effort is constantly being made tory vessels. to address the needs of the maritime industry. An example Some will target other of this effort is revealed in specific areas of ocean our composition course which related employment. Such a emphasizes technical writing as decision would likely lead to it pertains to the maritime postgraduate work. These industries. We additionally young men and women thus find offer a mathematics course themselves in a very desirable designed for business manage- position. They have already ment. Topics addressed include been exposed to a number of optimization techniques, ocean oriented subjects which probability, and decision- will undoubtedly help them in making theory. making a career decision. They additionally possess the The result of this effort is a practical seafaring background graduate who is cross-trained which will constantly serve in a variety of ocean related them in their coming years. subjects; one who is specif- ically trained in the operation Consider, as an example, an and management of the merchant individual who, after gradua- marine industries, solidly tion from a maritime college, educated in navigation and goes on to a graduate degree related subjects and, and then to a doctorate. At additional possesses an some time in the future he interest and exposure to the finds himself preparing to ocean environment from a enter the world of deep ocean firsthand, multi-perspective remote-sensing research. Part viewpoint. of the research application will be conducted from a three-man submergible to be launched from a ship platform 1585 900 miles at sea. On the day being spoken of for the planned day of the launch, first time? Technology will weather conditions are come, but it will be the predicted to deteriorate making application of the technology recovery difficult and that makes the difference. A dangerous. Decisions must be source from which to draw the made. Our individual has individuals that will be experienced these conditions needed in the environmental before. Remembering the days conditions of the ocean, and before graduate school, he to provide leadership in the assesses the situation correct- coming century can be found in ly, uses his expertise, and the various academic institu- continues with a successful tions that exist today. launch and recovery of the submergible craft. The SUMMARY illustration is melodramatic but the point is that the I suggest three guidelines as individual's prior practical we move in this direction: experience and education served him well. I. Academic institutions need the flexibility to adjust All sectors of the maritime their curriculums as world will benefit from technology advances; not in individuals who possess the such a way that makes them qualities described in the followers of industry, but previous paragraphs. Whether rather to be leaders of one is profit motivated, a technology. private business, or a government agency, all will be II. Financial support of the best served by an individual world's ocean-oriented with the practical skills of colleges and universities from seafaring combined with the both the public and the complementing knowledge of a private sector is a primary specific area of study. requisite for advancement in realizing and then utilizing CONCLUSION the potential of the world's oceans. ocean education is the key. Cross-training at the under- III. Conferences such as graduate level should be OCEANS 88 help us to chart the supported by all the various best course as we move into sectors of the maritime and the coming century. ocean community. Identification of the issues at this preliminary stage will Changes are upon us. We, as a determine the success of ocean society, will turn to the ocean education, research and as a source of energy, development, and their impact minerals, and sustenance. Who on world society. is going to lead us into this forthcoming era? Who will oversee the application of the current technology and who will work to prepare for advancements that are only this 1586 THE NEXT GENERATION WATER LEVEL MEASUREMENT SYSTEM:- THE NEXT STEP IN REAL-TIME DATA FOR NAVIGATION Daniel W. Farreill ABSTRACT The Sutron Corporation was recently awarded a contract to develop the Next Generation Water Level Measurement System (NGWLMS) for the National Ocean Service (NOS), an agency of the. National Oceanic and Atmospheric Administration (NOAA). The NGWLMS system will integrate existing sea level and Great Lakcs measurement networks and modernize them with state-of-the-art sensors, satellite transmission, and centralized data receiving/processing. The NGWLMS consists of five functional subsystems: 0 sensor/measurement subsystems; 0 data collection and recording subsystem; 0 data transmission subsystem; 0 data processing and analysis subsystem; and 0 data and information dissemination subsystem. The integration effort will include near real-time capability and links with other NOAA centers and systems for ocean, weather, and climate services. The NGWLMS was designed to upgrade the operation and maintenance of the National Water Level Observation Network (NWLON), a network of approximately 225 permanent stations and approximately 150 temporary stations. The NWLON was established to aid in measuring, collecting, analyzing and disseminating water level data and related products. The technology used to support the NWLON is aging and largely obsolete; therefore, NOS determined to upgrade it with a more modernized system. The NGWLMS system provides a major upgrade of the existing system through complete end-to-cnd replacement with technological advances made during the past decade. NGWLMS will apply state-of-the-art technological capabilities to upgrade and augment the existing system's capabilities. NGWLMS uses acoustic ranging and silicon strain gauge pressure sensors and micro p rocesso r- based data collection and recording subsystems designed to improve data qual 'ity as well as to collect data quality assurance parameters. A self -calibrating acoustic sensor that takes 181 samples in three minutes (centered at six-minute intervals) will be used. All data quality assurance parameters will be storcd every six minutes. Such multiple sampling will greatly increase the reliability of each of the six- minute measured values. Moreover, because the acoustic sensor will be lcvclcd directly, an observer will no longer be necessary. The NGWLMS will relay data, every 1. Vice President, The Sutron Corporation, 2190 Fox Mill Road, Herndon, VA 22071 1587 three hours, from a data collection platform (DCP) to an OMA central computer facility in Rockville, Maryland, through NOAA's Geostationary Operational Environmental Satellite (GOES). The new capabilities of the NGWLMS will include: 0 acquisition of near real-time watcr-level data from the system and full automated data processing; 0 accommodations by the DCP of as many as 11 ancillary sensors for observations including air and water temperatures, water density, conductivity, current speed, current direction, wind speed and direction, and barometric pressure; 0 capability of real-time water-level data for navigation; 0 capability of line-of-site radio links to ships supporting real-time hydrography operations and harbor traffic control; 0 automated acquisition of data and dissemination of products through computer network links; 0 acquisition of ancillary meteorological and oceanographic data for use in real-time by the National Weather Service's Storm Surge and Tsunami Warning Systems; and 0 application to other new NOAA systems to determine absolute changes in sea level by correlating with the Global Positioning System (GPS) and Very Long Baseline Interferometric measurements (VLBI) for climate and long-term sea level applications. These new capabilities will center around shortened response times to user requests, for example: 0 a few minutes for hydrographic and other navigation applications (present response time, 30-60 days); 0 a few hours for automatic detection and preliminary diagnosis of malfunctioning field equipment (present response time, 60 days); 0 1-71 days response to requests from general users for basic data (present response time, 90 days); and 0 7-30 days beyond the data c-ollection time required for tidal data computations (present response time, 3-6 months). Implementation of the NGWLMS is already underway. The first 20 prototype f icid units, provided by The Sutron Corporation, were installed last year and have been under evaluation since June 1986. The locations of these prototypes were selected to include a wide range of environmental and tidal conditions. Evaluations to date indicate satisfactory performance and an extremely successful rate of data return. An additional 50 field units are being installed in 1988. The installation of the central computer facility in Rockville, Maryland, is also underway. Completion of the entire NGWLMS is expected within about three years. 1588 Electronic Chart Display Information Systems: Operational, Policy and Legal Issues Peter W. Mushkat and Cynthia Lamson International Institute for Transportation and Ocean Policy Studies. Halifax, N. S., Canada, B3H 3J5 Unlike paper charts, electronic charts are tronic chart display. SINADS will also be able to dynamic information systems which can be incorporate positioning systems such as LCRAN-C. manipulated to support specific user require- OSL is also developing a passive precise radar ments and, by incorporating other electronic positioning system called NAVFIX to replace the navigation equipment, provide real-time navi- active microwave transponders. gation. What does and ECDIS do? An BCDIS provides real- The potential exists to enhance marine safety time navigation for a vessel. It gives the master through adoption and use of electronic chart an accurate, contemporaneous picture of a vessel's display and information systems (ECDIS). position (through the use of a positioning system However, how navigators access, interpret and such as SATNAV or LORAN or. soon, GPS) on an up-to- act upon the information will determine whe- date chart (which, potentially, will be kept cur- ther or not ECDIS is a blessing or a menace. rent through automatic updating) and reflects the real circumstances of the vessel by integrating This paper highlights issues related to ECDIS radar returns, including ARPA. and itsusers. It reports on the first Cana- dian seminar devoted to ECDIS, organized by 2. THE TBCHNOLOGY IITOPS on behalf of the Marine Advisory Board and the Canadian Hydrographic Service, held The technology in terms of hardware and software in June, 1988 at the Marine Institute in St. for an ECDIS is here now. Several companies inter- John's, Newfoundland. nationally, in addition t? Canada's OSL, are mark- eting operational DMIS. Each system employs its own software to manipulate the chart data and navi- 1. INTRODUCTION AND HISTORY gating information to produce its display. Already we have seen several generations of this-type of At a relatively early stage in their development, system come and go. As has been previously noted, Canada has attained a great deal of experience in OSL is presently developing its third generation electronic charts (ECs) and electronic display and system. The refinements in these systems are gene- information systems (ECDIS). The reasons for this rally designed to make them more accessible. user- are varied. In some respects they were fortuitous: friendly. safer. and more independent. the right people coming together at the right time. In others certain applications peculiar to Canada Apart from the purely technical advances - new favour the development of new navigation systems, microchips, etc. - the primary concerns in this such as the number of areas which are difficult to are ergonomic. What are the most effective consols navigate using traditional means, e.g., the Arctic. ("knobology"); what are the most effective colour displays (colour perception). The result is that Canada hosts one of the few companies in the world. Offshore Systems Ltd. The other aspect of technology important to an (OSL), which has a viable electronic navigation ECDIS is positioning. A breakthrough in this area display system operating in a commercial context on is expected when Global Positioning System (GPS5) the Marine Atlantic ferry, the MV Caribou. %1L comes on stream. It has been delayed by the hold- developed its first system. a Precise Navigatinq up in deploying the necessary satellites due-to the System (PNS), which used microwave transponders to "Challenger" space shuttle disaster. With the get ships in and out of narrowly defined channels advent of GPS it will be possible for civilian in the drctic.(Note) Its second-generation system, vessels to obtain a position within � 100 m and, a Precise Integrated Navigation System .(PINS) which with differential GPS within � 2 m. The ability to incorporated an electronic colour-filled chart position a vessel accurately is critical. display, is used on the Caribou and 13 other ves- sels. OSL is currently developing a Shipboard In- tegrated Navigation and Display System (SINADS) which will add a radar overlay to an improved elec- CH2585-8/88/0000-1589 $1 @1988 IEEE 3. INFORMATION the courts will have very little to go by in deci- ding the status of these systems. Internationally, Having developed the hardware to display the EC and the SOLAS Convention (Safety of Life at Sea) re- the software to manipulate the data, what is then quires that vessels carry paper charts; the Regula- needed is the "grist for the mill". In Canada, the tions passed pursuant to the Canada Shipping @ct Canadian Hydrographic Service (CHS) is now proces- require that vessels carry and use paper charts. sing its data digitally but the end product is still a paper chart. The CHS is responsible for As the courts decide, so the insurance industry 993 charts. The information for each chart is will adhere. collected on field sheets (there may be up to 20 field sheets in one chart) plus information gather- It was noted at the seminar, however, that it was ed from plans of wharves, dredging surveys, engin- unlikely that Marine Atlantic, which operates the eering plans, hydro and cable plans, etc. All of Cari , would have been able to obtain coverage this information has to be considered in developing on that vessel operating under those conditions and electronic chart information base (ECIB) which without the PINS. will be used to create charts, both electronic or paper. One aspect of this technology which the insurance industry has already picked-up is the "black box". George Macdonald, Manager of chart production for As in aircraft, some ships are already being fitted the CHS, commented at the recent seminar that, from with a device which will record their activities his region, at the current rate of 12 charts per and serve as a after-the-fact record in the event year it would be 2003 before the 202 charts for of a disaster. that region where digitalized and 2013 before the 702 field sheets used to create those charts were Premiums and litigation are only one legal concern. in digital form. ECs and ECDIS have introduced a new player to the Internationally, the International Maritime Organi- scene, the ECDIS manufacturer. Before liability zation (IMO) and the International Hydrographic was divided between the HOs and the vessel. Now the Organization (IHO) have jointly formed the manufacturer of the ECDIS exercises control over Harmonization Group on EMIS (HGE) to formulate the information as well (through the hardware, standa.rds/regulations concerning ECs and ECDIS. software and, if they digitize the chart, the daL- The HGE has stated that the national hydrographic ta). offices (HOs) should continue to be charged with the production of official charts. The reason for One final legal issue to be mentioned is copyright. this is two fold; first, to ensure accurate safe In Canada courts have accepted that software is a charts; and. second, to facilitate exchange of suitable subject of copyright. It has yet to be information through the implementation of uniform determined either in the courts or by legislation standards in this area. However. as has been not- whether charts or databases may be copyright. As ed. there are ECDIS on the market. Where are they Capt. Mukherjee noted at the MAB/CHS Seminar, "It getting their data? is well known that the objective of copyright con- trol is to ensure safe distribution and use of Most manufacturers are taking charts produced by charts. The same objective can be secured by other the FrDs and digitizing them. This fact identifies means" (eg., regulation or licensing). one of the critical aspects of ECs and ECDIS. 5. TIE 'WILD CARD' To date, one reason users have been generally re- luctant to consider ECDIS in their operations is a There is one user group who has already embraced lack of relevant ECs. To the potential user this the EMIS and that is the recreational boater. It limits its utility. Those are using an ECDIS have has been reckoned there are 45,000 vessels world- mostly been involved in situations, like the wide which would come under the IMO-IHO regulations CARIBOU. where they are employing it in a restric- when they are adopted, in the US alone there are in ted situation (access to a difficult harbour often the vicinity of 16,000,000 registered pleasure obscured by fog) so they do not require in exten- craft. sive BC and where the benefits far outweigh the costs. If this is a case of "seeing is believing". The potential of this market speaks for itself. until users are convinced of the value of ECDIS Recreational ECDIS are already available for less there will be little demand (and consequent resour- that $10,000. and with the general reduction in ces available) for the HOs to produce the data prices for micro-computer technology the probabil- necessary. Despite this. HOs, and the CHS in pw- ity is that these will be half that price in the ticular, are going ahead on the belief the demand foreseeable future. These people are less con- will be forthcoming and, judging from the reaction strained by budget or the need to justify the pur- of Marine Atlantic axid other present users, they chase of such a system in economic terms. will be proven correct in their prediction. Pleasure boaters are not regulated as stringently 4. CHARTING AND THE LAW as commercial vessels. To this is added the consideration as to the legal This is a market which, generally. is not as skil- status of ECs and ECDIS. Until the IMO/IHO through led or as practised in navigation as the commercial the HGE agree on their recommendations/standards. practitioners. This is both good and bad. On the 1590 positive side, an ECDIS may compensate for a lack it adds aLnother dimension to their repertoire of of experience or knowledge; on the negative side, navigational skills. it may promote an undue confidence in the system. According to experienced EC users, reliability is The consequence of the ECDIS market developing in the single most important benefit derived from the this are is that standards and systems will be acquisition of electronic navigation technology. market driven and regulators will have to accom- Electronics facilitate precision navigation and modate a multiplicity of established systems which reduce the margin of error associated with manuad may not be the best all round. calculations and observations. When navigators gain confidence in the accuracy of information Ultimately the need for information will be the provided, they are better able to concentrate on same whether it originates with commercial or re- manoeuvring operations. In turn, the ability to creational users. concentrate relieves some of the stress and fatigue induced by uncertainty. 6. OPERATIONAL AND TRAINING ISSUES Mariners are also cognizant that lack of understan- Too often the human dimension of technological ding or over-confidence in system capabilities may change is neglected in the rush to gain a perceived result in excessive risk taking. Whereas equipment operational advantage and many aspects relating to manufacturers bear responsibilities to inform users the application and use of innovative systems are about system limitations, ultimate responsibility either neglected or left to chance, EMIS depends for ship safety rests with the Master. Therefore, on the capacity of users to access and correctly discussions about data inputs and system capacities interpret information. Given the costs associated are integral components of BCDIS training exerci- with navigational errors, it is imperative to iden- ses. tify user needs in tandem with the development of BMIS hardware and software. In 1987, the Canadian Coast Guard issued a Ship Safety Bulletin advising mariners that existing From an operational perspective, ECDIS are revo- BCDIS should be used with caution and are not to be lutionary navigational aids enabling mariners to regarded as an acceptable equivalent to convention- access and manipulate relevant information for al charts: "The safe use of these displays for different purposes simultaneously. For example, navigation purposes depends on a number of factors dual display capability allows navigators to moni- including the existence of an adequate data base, tor the ship's route on a real-time basis and to the accuracy of digital information, the availabil- examine options for advance routing. In addition ' ity of satisfactory means of updating that base and most systems have the capacity to store historical the subsequent provision of safe and c?mprehensive navigational data which can be retrieved for both methods of data selection for display." planning and documentation purposes. To date, questions about training mariners to use Canadian mariners who have used ECs are generally EMIS safely and responsibly have been overshadowed enthusiastic about the potential of ECDIS to en- by debates about data acquiFition. specifications hance marine safety, particularly in close man- and system administration. The 1988 Marine oeuvring situations and during icebreaking. Al-- Advisory Board seminar provided a unique opportuni- though ECs are regarded as being a significant ty for mariners. hydrographers, commercial opera- addition to the kit--bag of navigational aids. most tors and regulatory agency representatives to as- mariners do not foresee that they are likely to sess their collective experience with EMIS and to become a single, all-purpose navigational tool. identify future training needs, Navigators with Radar, for example, provides essential information hands-on experience argued that competence in about vessel traffic. Participants at the 1968 "knobology" was achievable in a relatively short seminar in St. John's were of the opinion that timeframe, that is, given a good introduction to an radar overlays should be incorporated into ECDIS on-board system and access to a reference manual design. (backed up by a system "help" menu). competent personnel should be capable of operating an ECDIS. Mariners are often described as being extremely However. a majority of these users argued that conservative. sceptical and slow to innovate. The there are two levels of training: Level 1 training expression. "if it ain't broke don't fix it," des- involves the study of system principles and focuses cribes the attitude of the seafaring community on knowledge. comprehension and judgement. Level 2 towards change. However, the introduction of ECDIS training concentrates on enhancing operational technology aboard Canadian ships has not encounter- skills. Responsibility for Level 1 training rests ed significant opposition. probably because naviga-- with schools and professional nautical training tors do not perceive the technology as a threat to institutes. while Level 2 training is probably best job security or professional competence. Unlike provided by equipment manufacturers. technologies which transfer functions originally performed by human operators to machine, ECDIS do Instructors from several marine training institutes not eliminate the requirement for experienced and expressed frustration with the current system of properly qualified navigation personnel aboard training and testing marine personnel. Because ship. Canadian Coast Guard experience with ECDIS curriculum changes are linked to regulatory devel- in the Laurentian region indicates that mariners opment, nautical education tends to lag behind are receptive to this innovative technology because technology and commercial practice. Thus, individ- uals may earn qualifying certificates that do not. 1591 in fact, adequately prepare personnel for working there was agreement that all the information in the dynamic environment of commercial shipping. contained on the paper chart should be acces- Although most seminar participants shared the view sible to the mariner. The critical question that it was premature to (add ECDIS to nautical is achieving a proper balance between a mini- curricula immediately, there was consensus that mum amount of information which should be instruction in precise positioning systems was displayed at all times and the mariner's probably overdue. Of course, when international ability to avoid "clutter" by selecting from standards for EMIS are accepted by national requ- the total information available those items laLtory agencies, training institutes will then be necessary to create a meaningful display. compelled to instruct students in the use of such Thought should be given either to allowing systems. mariners to create their own charts by provi- ding information/database and allowing them Possibly the most vehemently argued point by marin- to build up a composite display based on what er participants at the MAB seminar was the need to they need to know in a given situation, or a involve users at all stages of system development. minimum data set (default setting) should be Although there is no evidence that mariners have provided with the ability to add or delete been deliberately excluded from EMIS R & D activ- information as needed. ity nationally, the Canadian Hydrographic Service and manufacturers have assumed leading roles in Mariners present stressed the importance of system design, testing and standaa-ds-setting. An good, user-friendly design. Perceptual, as important outcome of the seminar was the follow-up well as ergonomic, considerations were men- action taken by the Canadian Coast Guard Super- tioned. intendent of Navigation Safety, Mr. Tom Brooks, who distributed copies of the draft ECDIS performance The focus should be changed from concentra- standards prepared by the IMO/IHO Harmonization ting on display standards to clatabase/inform- Group (NAV-HGE) arf solicited comments from all ation standards. interested parties. Thus, one tangible result of the MAB/CHS Seminar was the expansion of the ECDIS Speakers supported the use of EC test beds as ,.constituency" in Canada. a means of improving technology and. more importantly, gaining practical experience. 7. THE MAB/CHS) SEMINAR The use of test beds, or simulators, in Nautical Training Institutes was also endor- The following points about the introduction of sed. EMIS in Canada emerged in the final session of the MAB/CHS seminar: As an ancillary issue. resulting from the CHS-' statement that sufficient furKIs were riot * It was felt that uniform standards for elec- available to proceed with the speedy digitiz- tronic navigation displays should be develop- ation of data, there was some discussion of ed internationally in order to ensure safety the concept of "user pay". In this context and ease of access to information, which the recreational user was identified as pos- would enhance commercial viability. However, sibly an extremely lucrative market. there was general agreement that more hands- on experience is needed before standards are Although electronic navigation display sys- ..cast in stone". tems would be of great value in commercial shipping, from a user perspective the largest * The rate at which electronic navigation dis- market would be recreational boaters who are plays will develop is not clear given the generally not as stringently regulated as the lack of international standards and the un- commercial sector. certainty as to when GPS will be available and with what accuracy. There was some discussion about the infra- structure required to support electronic Another factor hampering the development of navigation, including: digital data from the electronic navigation displays at this staqe Hydrographic Service; the ways and means to is a lack of usei demand which would generate update navigational information; and differ- the government resources riecessany to provide ential GPS broadcast from the Canadian Coast the chart information needed to operate the Guard. systems. One reason users are relLICtant to install systems is because there is no gov- It would be useful to try to foster a con- ernment chart information with which to oper- tinuing process of user input into the devel- ate systems. opment of electronic navigation displays. In the same area it may be useful to consider There is a need to establish a clear- voca- separately the issue of training/retraining, bulary in relation to this field. 8. CONCLUSION There is a need to realign thinking from a mere electronic representation of a paper Ultimately, the successful introduction of ECDIS in chart to a true concept of equivalency, ie. a Canada will depend on whether it answer-s a need of system that will provide information for the marine community. There is no question that it safe, accurate navigation. At the same time has proven itself extremely capable in certain 1592 specific situations and, no doubt, will continue to do so. The question is will it find a more general acceptance. The hardware and the software required to prove the benefits of ECDIS to the user are now available. They have been relatively easy to develop. It is more difficult (i.e., costly because of the labour involved) to produce the data required to operate these systems. Commercial interests are holding back because of the lack of certainty in the areas of standards and liability. Rvort-atlorkai @v_-;ers are sA.@_n as a poterr- tially profitable way of stimulating demand for this service which will benefit all sectors of the marine community. The development of ECDIS technology offers an un- paralleled opportunity for cooperative activity among different sectors of the marine community. The gaps between hydrographers and mariners, for example, can be bridged if conscientious efforts are taken to involve all interested parties from the earliest stages of design through to decisions about system application and administration. TO date, mariners have not resisted the introduction of ECDIS because they perceive ECDIS as a tool to enhance precision of navigation. However. the experience and concerns of the user community must be incorporated into the development process to promote acceptance of this new technology and maxi- mize the potential benefits associated with ECDIS. Questions about trainirj(@ and certification should be addressed prior to widespread shipboard applica- tion of ECDIS. Too often training follows regula- tion but, given the rapid advances in ECDIS tech-- nology. the need to anticipate training need---,, is paramount and should not be permitted to lag behind legislative/regulatory development. Certainly it is the CHS's contention, based on their experience with ECDIS. that when the demand comes it will be big. 9. REFERENCES 1. eg. "Geonav" is manufactured by Navonics in Italy; CTiartNav 20/20 by Laser Plot Inc. in the US; Disc Navigation by Disc Navigation Sales A/S in Norway; and NavPlus by Cordrafix Inc. in the US. ?. C.R.C. 1978. c. 1415. 3. Canadian Coast Guard, 1987 Ship Survey Bulletin. No. 2/87. 29 January. 4. Kerr, A.J., R.M. Eaton. and N.M. Anderson. 1986. "The Electronic CTiart: Present Status and Future Problems". Journal of Navigation. Vol. 39. pp. 24- 31. 5. IMO/IHO Harmonization Group (NAV-HGE). 1988. "Performance Standards for Electronic Chart Display System (first draft). Hambux-g. FRG. 1593 A Computer Navigation System using Kalman Filter Smoothing Dr. Everett F. Carter and Joe Levkowicz University of Rhode Island Graduate School of Oceanography South Perry Road Narragansett, RI 02882 - 1197 The device command and device query ABSTRACT processes work in concert to extract and process data from their device. The command process implements any device The University of Rhode Islands specific command protocols and data Graduate School of Oceanography and the translations between the host and URI/GSO Technical Service Department are navigation device. It is designed to building a general shipboard data allow several processes to issue commands collection system for the R/V Endeavor. to the device without interfering with The collection system is implemented an a each other. DEC MicroVax II running VMS. The device query process is a client This paper describes the computer of the device command process. It knows software comprising the navigation data all about the device specific commands collection portion of the overall system. and the data contained therein, but does not have to worry about the other device The navigation software collects, specifics. Its job is to do those things logo and processes information from a required to collect navigation position diverse set of electronic navigation and'quality data, convert the data into a aide. standard format, and send the "standardized" and raw data to the A "smoothed beat guess" set of logger/router. navigation parameters is continuously produced using a Kalman filter system. The grabber process simply catches the ejected data, tacks an a "standardized* translation, and BendB it and the raw data to the logger/router. 1. DRIVERS 2. LOGGER / ROUTER The device specific portions acquire The logger/router logo all incoming data from the navigation devices, convert data to a daily file, distributes new it into a wcommon format", and hand it data to other processes, and keeps track off to the router/logger. They are of currently active devices. designed to allow multiple processes to access interactive navigation devices. There are two classes of clients of the logger/router; the "producers" who Navigation devices come in two send data for logging and/or flavors; those that require bidirectional distribution, and the "consumers" who interaction with the host, and those want to examine the data. Generally, the that passively "dump" data. For the device specific processes are the former type, the device specific portion producers and user processes are the- consists of a device command process and consumers. A process may be both device query process; the latter type is producer and consumer; the kalman filter implemented with a simple device grabber process consumes data from devices like a process. user process and produces navigation data like a device. CH2585-8188/0000- 1594 $1 @1988 IEEE The device clients of the 4. REMOTE ACCESS logger/router hand it some device specific information when they *connect" Remote access to the device command to it. This data is available to other servers and logger/router is available clients who may wish to examine data from through TCP/IP and DECNET. Special the device. server processes running on the host field network requests, routing them The 'consumer" services offered by through the appropriate server and back the logger/router range from a log file to the requesting node. to continuous delivery of the latest data from selected devices. The device information service of the logger/router 5. UTILITIES acts as the "yellow pageen of the active 'producers". Included in this There are several utilities that use information is the "producers" unique the navigation services. The time server name. A consumer receives data from a updates the system time using the GPS producer by sending the producers name to clock time. The navigation plotter the logger/router. It may request one produces plots of current or previous fix fix or sign up to receive all fixes as data. they come in. There can be many real- time consumers of a single producers data. 6. SOFTWARE FRAMEWORK The log file and data from the To make a system of this type logger/router is in ascii. maintainable and flexible the software has been designed and written in "layers'. The following is a description 3. KALMAN FILTER of these layers and how they are used. The Kalman filter is the heart of For starters, all device 10 is the whole system. It takes the output device independent. All io is done via a from all navigation devices and produces set of standard calls. This helps remove as "best guess" smoothed output. device and operating system dependence from the navigation system. A device can Each navigation device in the system be reconfigure to run on a different io is assign a "weight' that allows the system without making any code changes. filter system to make allowances for the accuracy of the device. These weights Navigation device dependence is are currently chosen by the informed user pretty well confined to the navigation of the navigation system. When a new device specific drivers. Adding a new reading is entered the filter system, navigation device to the system requires makes a prediction (based on previous a driver specific for that device and readings) of navigation parameters and some configuration changes to the Kalman combines the new reading with its filter startup file. predicted position according to the weight of the navigation device and the All the processing involved with weight of the predicted position. This making the logical connections between produces a "maximum likelihood" set of "server" and "client" processes exists in current navigation parameters. a template form. This means that writing a new device command process, device The filter system is designed to grabber process, or device query process allow the user to change its operating requires filling in the software blanks. parameters an the fly. It uses an ascii configuration file to describe device A fully implemented set of commands input format and the processed navigation is available for all supported output. interactive devices. Direct command access to a device is as easy as calling The filter system can process an init routine and then calling the navigation data either in real-time or in specific command routines. a batch mode since it obtains all time data from the navigation message it is Interprocess communication is done processing. Output from the system is via mailboxes. The operating system produced whenever it receives a special dependent calls are confined to a mailbox "output data" message. module for portability purposes. There are many ways to implement interproce8s communication; mailboxes were chosen because they seemed to offer the greatest ease of portability. 1595 Error handling is done by a yet another process that receives error messages from the other processes. When things go wrong or when a routine wants to log some event, it send a message to the error logger. The logger records the time, date, process name, id, error, image name, source file, software version, source code error line, compile date, and an auxiliary error message in a file. Depending on the severity of the error and upon the desires of the reporting process, it remains silent, flashes the panic light, catches fire ... etc. All error messages must be "registered" with the error logger. This means at least a minimum effort by the unruly programmer to "document" the error types and messages in the program. it also leaves a trail of what went wrong, when, and where. This makes it easier to debug over the horn ... no weird messages that print cryptic messages and then scroll into the void. 7. CONCLUSION The result is a flexible, easy to use system that runs with a minimum of operator intervention - and provides continuous navigation data. 1596 Process Interconnection in the Havication System ................... ....... Date File :::User 9393::: ..................... ..................... v .................. .............. ........... .................... .................. HHHetwork Inter face::::::< g' t: e r:::::::::: H:Xalman Filter'::::' A A ':::::Time Server!"U: :::'::Devi ce Query:::::: Process Process A A v v ............................. .................... .............. ......................... 'HiDevice Commandifl: :'::::Device Grabberfl':: Process Process ............ F.............. ....... .............I............. ........................... :;:HP89393iVe:H:: Device in fl: D e v i c e ................................. zi. ................. .................. .1597 AWAKENING THE CONSCIOUSNESS OF THE BOATING PUBLIC REGARDING POLLUTION, INTOXICATION, AND COMMON SENSE SAFETY ON THE NATIONS WATERWAYS F. Hunt Anderson United States Coast Guard Boating Education Branch Washington, D.C. 20593-0001 ABSTRACT speculate ontheenvironmental damage of recreational Possibly the least regulated of all forms of transporta- boating. We can only make an educated guess, based tion occurs on ournations rivers, lakes, bays andoceans on experience with small microcosms, as to these in the form of recreational boating. The amount of effects and try to forecast what will happen if nothing environmental damage and relative safety of these is done while substantial growth continues. waters is left largely to the individual boater, and yet, the relative damage which can be done to the planet, Anyone who remembers their basic grade school earth animal life and fellow humans is staggering. This science class recalls the importance of the water cycle paper deals with the problem of changing public atti- and how delicate the balance maintained by Mother tude to a more responsible perspective. These prob- Nature. Being candid, we are literally playing on the lems include lack of funds, motivation, and direction. lifeline of mankind, water. The impact has not ap- peared in truly catastrophic terms and is not defined The roles of government, associations, federations, simply as a recreational, commercial or residential industry, retailers, media, educators, and law enforce- problem. In some areas we are witnessing more severe ment are considered in this uncharted process of effects than others. Are these examples of more change. Those involvedwith education are often over- general conditions thatwe can expect as ourre-creating whelmed by the vast number of issues they are tasked on the water increases? One of the primary reasons to address with little or no support material to help many .of us are here at this meeting is because of our them. Looking at specific programs undertaken by concern for the ecology of our planet and the delicate national nonprofit organizations, industry and balances that we know exist, Unfortunately there are government,one can see a very plausible model which many others who do notknow of these concerns and use can be expanded by cooperative efforts of various and consume our nations waterways with little or no regard for these issues. groups. The objectives are basically simple and are centered around educating the public through short There seems to be little comprehension of the fact that messages, involvement and interaction to not only be our garbage outlives us. Soda cans, beer cans, bottles, aware of potential problems but empowered to inter- potato chip bags, plastic bags and wrappers all live vene in preventing them and, thus, awakening the many times longer than the average human being. Itis consciousness of the boating public. all out of sight out of mind, until a prop becomes hopelessly entangled, marine animals ingest it thinking UNTRODUCTION its food, animals become caught in it or a storm washes it ashore. The aggregate effect of this garbage accumu- The purpose ofthispaper is to explore the current status lation may leave an unfortunate legacy for our of one of the nation's most unregulated transportation children's children, moving closer and closer to the activities and to assess some options relative to dealing -headwaters of our rivers and covering the bottoms of with negative aspects of it. Each year over 1,000 our estuaries. people die in recreational boating accidents which is over 90% of the water transportation fatalities. When COMMON SENSE AND RESPONSIBILITY we get right down to analyzing the data, we really do not know precisely how manypeoplehave beeninjured So far we have been lucky. The majority of boaters or otherwise been involved in some sort of boating exhibit common sense and common courtesy as they accident. Likewise, we have little more than the use the nations waterways. They respect the environ- garbage and debris on the shore lines to help us ment and the safety of their fellow boaters. We may United States Government work not protected by copyright 1598 however not be so lucky in the future. We take for foster this sense of responsibility. Unfortunately, en- granted that people will use common sense. Many, forcement will continue to be needed to remove those however, don't know what it is, and we cannot assume who persist in operating vessels in an unsafe manner to that everyone has it. I will define common sense as: satisfy their ownego. However, the majority ofboaters Having the knowledge to do what is best for the safety will be headed for the waterways in search of fun and and common good of others as well as oneself We relaxation but not knowing enough to be truly respon- would assume that the father who took his family and sible boaters. Who will be the conscience of mankind friends to view the fireworks this past July Fourth and who will influence boaters to adopt behaviors would have had the common sense to know that it is which will embrace good "common sense?" unsafe to put 15 people in a 20 foot boat to go for a cruise on a busy harbor. We might also assume that THE MODEL operators of the two high speed boats which passed close to this overloaded craft should have had the What is heartening is that an embryonic model exists common sense to know that their wakes could have for implementing this "common sense approach." In caused an accident. some areas of the country it is more widely developed We expect boaters to exhibit a great deal of common thanothers. It is amodel that has within it the flexibility sense. Unlike the automotive market place, the pro- to address specific needs of geographic areas andpopu- spective boat buyer is often asked to piece together the lations. The generating of a common sense base means boating system that he or she wants for the intended the instilling of an awareness that it is needed, then a areas and activities the boat is,to be used. This may program of information to support it with constant mean buying a hull and propulsion system separately, reminders through role modeling and endorsement of or adding on consoles, instrume .ntation, and a whole the need to be responsible and use it. This goes beyond host of aftermarket items which may effect the per- just offering classroom training or providing public formance of the vessel depending upon placement service messages. It means weaving it into the cultural adjustment etc. Because a boat is an expensive item, fabric which includes entertainment, social activities, many will buy a basic boat and add to it each year as advertising, and folklore. With primarily volunteer they get theinoney. The industry only assumes respon- horse power the boating organizations and the national sibility for the specialized components it makes. The boating federations have sought to do all of this. It is dealer is often overwhelmed by the quantity and vari-@ time consuming long term work fraught with many ety of items that he is being asked sell and may or may frustrating experiences. The intention is there but the not know the best choice for the intended purpose. funding and the person powerto support it are woefully When these choices are made they are most often lacking. driven by concern for performance and not concern for safety or the environment. The boater is expected to INFORMATION - TBE KEY ELEMENT use his orhervessel in waters appropriate forthe vessel, yet many operators will "push the envelope" of the For years people have realized that education and in- vessel in pursuit of more challenging activity in formation dissemination are key elements to prevent- rougherwater. We expect boaters will go out prepared ing accidents andpromoting safe behaviors, How often to meet sudden changes in the environment such as have we heard the accident victim say "I didn't wind, rain, currents, tides and changes in temperature. know..." Various nonprofit, national, state, local, and How often have we beard the story of the fisherman commercial groups have sought to educate the public who goes out and at the end of the day can't get the in one way or another in safe boating practices. The motor started and spends several hours orihe night efforts of these organizations vary in scope in different drifting farther away from the intendeddestination? In geographic areas largely a factor of the number of addition to exhibiting common sense and knowing personnel, particularly volunteers, involved with the their craft we expect boaters to be good and courteous program. On a national level these organizations have seaman, which includes being concerned for the im- banded together in the forums of the National Safe pact of their boats and related activities to the environ- Boating Council, the National Association of Boating ment. Law Administrators and the National Water Safety Congress, to exchange ideas and generate some com- The intent here is to show that the operator not only has mon programs such as the National S afe B oating Week to have common sense but an innate sense of respon- campaign. These national federations are run on the sibility as well. No amount of enforcement is going to dedication of leaders who donate time apart from their 1599 regular jobs to make these organizations function. for boaters. At this point some would opt for a bureau- While the efforts of these organizations have grown cratic solution which would legally mandate boating phenomenally in the past 3 years, it has not kept pace education through legislation at the state level. VAiile with the increase in size of the boating public. The this is one option that has the advantage of guarantee- National Safe Boating Week program, for example, ing funding for a program, it still does not assure that has grown by a factor of at least ten, but it is estimated the boater actually has the information he or she needs that this effort is reaching only twenty percent of the to be operating the particular boat that is being used. boating population. The boating industry has verbally been in support of ALTERNATIVES educating the public. Jeff Napier, president of the Na- Like boating, learning to play a musical instrument is tional Marine Manufacturers Association, stated "The a difficult task requiring knowledge of musical signs, teaching of boat operation techniques and safety note values, key signatures, rhythms, etc. Yamaha measures should be as common place in our nation's Music Corporation developed a pre-instrumental pro- school systems as is the teaching of ground strokes for grain, with the aid of psychologists, educators and mu- tennis... Boating has an enviable safety record: of the sicians that gives a student many of the basics before 70 million people who go boating every year, only ever touching an instrument and enhancing the ability 6,000 are involved in accidents resulting in 1,036 of the individual to master it more rapidly. Because fatalities or a fatality rate of 6.1 per 100,000 boats.... Yamaha had a rigorous instructor training program, Despite the small number of fatalities or the dropping complete media and material support, and a certifica- fatality rate, our industry recognizes that by seizing the tion process, the program was assimilated into a num- initiative we can better prepare our future boat owners ber of school systems. With some adaptation, a similar for more enjoyable and safer hours on the water." model could be developed for a boating education While statements made by industry representatives program. reflect an awareness that growth may create problems in the future and an honest desire to become involved, Another approach was taken by Kaiser Permanente to there is a hesitancy to go all out and create complete educate youth on certain health hazards. Small acting boating education or information programs to support companies were hired to perform informative plays de- their endorsements. There are the factors of time, picting the various hazards in an entertaining way. The money and the differing levels of commitment of the companies traveled from school to school putting on 1,500 manufacturers, but in this author's opinion, the assembly programs geared at different age levels. problem is related to specifically what to do and how Similarly, a team of experts could be assembled to to get it into the system. develop dramatic productions on boating. It is a given that no boating safety education course or With the computer age comes possibilities that are information program will adequately cover all areas truly innovative. Computer assisted training modules boat handling, first aid, seamanship, navigation, regu- have been successfully developed for other subjects. lations and environmental impact. There is always Why not boating? Video games teaching navigation as more to learn, some of which is dependent upon the well as safety information would seem a natural. type of boat and waters that are used. The Minimum Education Guidelines established by the National Automobile seat belt campaigns and anti drug cam- Association of Boating Law Administrators does offer paigns offer yet another avenue of approach. The some help in narrowing the field ofpossible topics, but media blitz is an unaffordable task for most nonprofit it would still be a monumental task to come up with a and government agencies. As apart of advertising and program that all manufactures would support. In public service campaigns industry couldpromote vari- addition to this problem, each educational system has ous aspects of boating safety. Alcohol is as significant its own standards and associated review and selection aproblem onthe water as itis on the highway. Several process. How many school systems would make the breweries have been forthright enough to publish bro- same choice as the industry? chures on the subject but more is possible. In fairness to the industry, it should be pointed out that One element of alcohol awareness programs is inter- this problem exists for all the organizations who have vention. Too often when people see an unsafe behavior developed or are creating public education programs or a problem related to intoxication they just watch. Iwo Most people do not know how to effectively intervene. provide leading information, a motivation to pursue Parents will snatch children from the jaws of death but more indepth knowledge can be generated. Govern- will often not know how to prevent it from happening ment and industry will need to provide leadership, again. Programs have been developed to address the some funding and coordination to facilitate the devel- communications skills that are needed to effectively opment ofnew programs andkeep duplication ofeffort intervene but they are not well known and are not well to a minimum. It is a strategy that presumes that funded. everyone will accept more responsibility rather than waiting for infamous accidents to color our approach There are a range of possibilities for industry involve- and produce costly legal mandates. The decision ment. Some might also be exercised by other organi- makers of government, industry and boating related zations with the sophistication to develop a funding organizations will have to develop a common commit- base. The bottom line seems to be the development of ment so we can begin to weave responsible boating into a product. Those organizations that are conducting our cultural fabric. We have an opportunity to effect education and information programs successfuily are significant behavioral change to preserve the environ- doing so because they have the support materials and mental integrity and the safety of our waterways for personnel for training instructors as well as the mate- future generations without imposing legal hoops and rials to conduct the program. wickets we may not have the ability to enforce. It is not necessary for industry or any other group to go back to square one and start with a whole new program. There are many organizations which have programs in various forms and levels of proficiency. Some are kept more current than others. A few have gained the accep- tance of school systems. By gathering representatives from some of the larger member organizations of the National Safe Boating Council it may be possible to provide support for existing programs and allow for a quantum leap in the evolution of these programs and broader use. THE OPPORTUNITY AND THE CHALLENGE The exciting element is that boating and water sports are perceived as fun. The growth in the marine market is just one indicator. Advertisements, movies and TV shows are often staged on the water because of this perception. Learning how to deal with the water environment also offers the opportunity to present standard english, mathematics and science programs with a practical application and appeal. It offers opportunities for the development of communications and teambuilding skills. The issue is we are not exploiting a golden opportunity. It is a question of nurturing the "common sense base" which is a part of many current programs andproviding the funding so they may flourish. It means influencing the media; the movie makers, TV producers, story writers, advertisers and textbook writers to take a responsible approach when depicting the water envi- ronment. By providing role models that reinforce safe and responsible practices, a subliminal message can be sent that may far exceed the effectiveness of any training program. By creating short messages which 1601 GEOLOGY AND HYDROLOGY OF REEFS AND BANKS OFFSHORE TEXAS AND LOUISIANA Richard Rezak and David W. McGrail Texas A&M University College Station, TX From 1967 to 1983 the Department of Oceanography at Texas A&M University was involved in research on reefs and banks off Texas and Louisiana. Funding was originally provided by Texas A&M University, University of Texas Flower Garden Ocean Research Center, and NOAA/Sea Grant College Program. Beginning in 1974, funding was provided by the U.S.D.O.I. Bureau-of Land Management (now Minerals Management Service). Studies were conducted on the geologic structure, sediment distribution, biol'ogical zonation, and water and sediment dynamics at over 30 reefs and banks. Geological studies utilized precision navigation, high resolu- tion subbottom profilers, and side-scan sonar to produce accurate bathymetric and structural maps. Grab samplers were used to obtain sediment samples. Direct observations of the seafloor and the sampling of rock outcrops were accomplished by use of the DRV DIAPHUS. Video cassette recordings of the seafloor were made on all submersible dives. Hydrologic data was obtained by long term moorings of current meters and temperature sensors. In addition, profiles of temperature, salinity, horizontal current velocities, and transmissivity were obtained on seasonal cruises. -The results of the research show an intimate relationship between geologic, biologic, and hydrologic processes and each has a direct influence on the other two. 1602 LAUNCH AND RETRIEVAL OF A 1000 TON BARGE-SHAPED VESSEL FROM A CONVENTIONAL TANKER Chandru (Jack) Kalro Diversified technologies, Inc. Alexandria, Virginia This paper describes the successful application of ship salvage technology .to solve the problem of transferring a 1000 ton Single Anchor Leg Mooring (SALM) Base from the deck of a conventional tanker to the water and back to the deck. Hydraulic Linear winches, long popular with salvors, are used to haul in or pay out the SALM Base on Nylatronlined skid beams at the rate of 1.5 fpm. The skid beams have hinged sections which open out during the launch and retrieval operations., The skid beams slope at an angle of 12 degrees and the entire tanker itself is heeled over to about 12 degrees during launch and retrieval operations. 1603 DEVELOPMENT OF EXPERIMENTAL ECOSYSTEMS FOR THE STUDY OF COASTAL LAGOONS S. W. Nixon and S. W. Granger University of Rhode Island Narragansett, RI Much of the world coastline is characterized by shallow, macro- phyte-dominated lagoons that increasingly are subjected to pollu- tant inputs. A major concern on the R.I. coast is eutrophication possibly caused by nitrate enriched groundwater entering such lagoons from increasing numbers of individual sewage systems or leach fields. In order to examine this @roblem, we have designed and constructed ten r1plicate lagoon mesoco5ms, each 2.3 m x 1.8 m x 1.1 m deep (4.1 m surface area, 4.55 m volume). The bottom of each mesocosm is covered partially (45%) by unvegetated, organic-rich sediments, 45% by lower organic sediments with eel- grass, and 10% by clean sand. This ratio is similar to that of natural R.I. lagoons. Sediments were carefully collected from the field (to about 30 cm deep) to preserve their vertical struc- ture. Water for the systems is collected regularly from a natu- ral lagoon, trucked to the mesocosm site, and stored in large tanks. The mesocosms are run as open systems with flushing rates of 5% per day. Water circulation is provided by large transparent plastic paddle wheels. Several species of small fish, crabs and shrimp, as well as macroalgae have been added. After an initial study of replication among the ten systems, three different levels of daily nitrate additions will be started to observe the responses of plankton, seagrasses, macroalgae, and animal growth rates. An automatic oxygen sampling system allows frequent meas- urement of total system and water column metabolism. 1604 IMPROVING SEARCH SUCCESS; REAL-TIME DATA COLLECTION AND TRANSFER TO USER D.F. Paskausky W. Reynolds R. Gaines R.Q. Robe U.S. Coast Guard R&D Center Avery Point Groton, CT 06340-6096 ABSTRACT To improve the probability of detection of the target in search and rescue (SAR), the USCG R&D Center is conducting research to determine sweep widths for search platforms and sensors such as Night Vision Goggles and new radars (including side and forward- looking and synthetic aperture radars). Research in measurement of currents in real time to hindcast and nowcast target movement includes development of Loran-C position transmitting buoys, satellite image analysis to initialize a current field in a search area and a search planning test bed utilizing desktop computers to analyze and transmit this information to the rescue command center. This SAR planning system is to be included as an ocean module in a ports and waterways management information system (PAWMIS). PAWMIS utilizes a microcomputer-based automated information system which will provide information required for emergency response, port security, vessel traffic management, aids to navigation management and permit review by means of computer-generated site-specific charts, databases and video response-planning images. The modules of PAWMIS are designed for use by Marine Safety Officers, Captains -of the Port, Operations Centers and others. It will permit access to mainframe databases as necessary. 1605 NEW APPROACHES FOR INDICES/MONITORING MICROBIAL PATHOGENS IN SEAFOOD R. R. Colwell University of Maryland College Park, MD Assessment of the public health safety of seafood requires microbiological moni- toring, either for the presence of potentially pathogenic microorganisms or of indicator species. Culture methods used for-microbiological analyses are, histori- cally, recognized to be limited in accuracy and reliability. Most probable number (MPN) of coliforms, fecal coliforms and related enteric bacteria yields an esti- mate of the number of such bacteria in a sample but does not offer precision or accuracy. Furthermore, the coliform group has been used to indicate the probable presence of human pathogens, e.g., Salmonella, Shigella, Hepatitis virus, etc. The inadequacy of the coliform MPN test has been decried widely in the literature. However, until recently, suitable alternative methods have not been available. Recent discovery of the viable, but nonculturable phenomenon, characteristic of many Gram negative bacteria and nonspore-forming Gram positive bacteria (Roszak and Colwell, 1987) and first demonstrated in Vibrio cholerae, poses a new challenge to the standard methods approach. The phenomenon, briefly stated, is a viable stage, whereby cells can be demonstrated to be endogenously respiring, responsive to substrates, but not culturable by ordinary microbiological methods. Such cells can be cultured when passaged through animals, e.g., the rabbit illeal loop, in the case of Vibrio cholerae. The viable but nonculturable stage has also been docu- mented for Campylobacter and Legionella. New methods have been devised to detect bacteria in the viable but nonculturable stage. Research is needed to determine the occurrence of.these forms in seafood and to assess their public health signifi- cance. 1606 INFORMATION IN A CHANGING MANAGEMENT CONTEXT: THE OCEAN DUMPING EXAMPLE R. H. Bi-;rroughs University of Rhode Island Kingston, RI Policy toward ocean dumping as expressed in legislation, regulation, and judicial interpretation has shifted. In addition multimedium management and assimilative capacity, concepts introduced after passage of the law, have further Affected perceptions about appropriate policy. As a result the information relevant to measuring success in program implementation has been redefined. Policy relevant information in reports for the Congress has been reviewed. Initially, policy was directed toward the magnitude of the materials dumped in the ocean, but units of and techniques for the measurement of amounts are not consistent. The magnitudes of other anthropogenic sources are ignored. Nonetheless, ocean du'mped dredged material and industrial/chemical wastes show a decline while sewage sludge dumping is increasing. Since passage of the law, policy has shifted from a proposed elimination of sewage sludges to restricting only those sludges that might unreasonably degrade the ocean. After these refinements, agency success may be measured in avoidance of unreasonable degradation or through the other considerations introduced by multimedium management or assimilative capacity. The changes redefine policy relevant information and place greater emphasis on composition, fates, and effects of materials proposed for the ocean. 1607 A BIOTECHNOLOGICAL SYSTEM FOR THE UTILIZATION OF WASTE PRODUCTS OF THE SEAFOOD AND CHEESE MANUFACTURING INDUSTRIES John P. Zikakis University of Delaware Newark, DE The U. S. Shellfish and Cheese Manufacturing Industries are producing huge amounts of wastes and they have serious problems with their proper disposal. The value of tNe nutrients lost by dLmping thZ unused byproducts is estimated to be in excess of $500 million annually. k1though the cost Eor cleaning the envirorment is difficult to assess in terms of dollars, a conservative figure may be well over one billion dollars. The main reason for the underutilization of whey is that it contains a high quantity of lactose. Nearly 80 percent of the world's population is lactose intolerant, an incident which is similar for most species of animals. Since lactase is not an inducible enzyme in most animals, the main problem is how to increase the ability of an individual to digest larger amounts of lactose in the diet. Using chitinous products in diets containing high amounts of whey, we were able to overccme the lactose intolerance in econcmically ijuportant animal species. The enzymatic breakdown of chitin (in the shellfish processings) in the gastrointestinal tract, releases bioactive growth- prcmot ing canpounds (such as Beta-N-acetyl-D-glucosamine glycosides and oligosaccharides) which stimulate the growth of beneficial lactase-producing bifidobacteria in the gut. The manipulation of these bacteria in the gut supplied indirectly the needed lactase and enabled us to feed broiler chickens up to 8 times more whey than it is possible without the chitinous supplement. These and other nutritional as well as enzymological studies will be presented and discussed. 1608 EPA's RESPONSE TO THE FLOTABLES INCIDENTS OF THE SUMMER OF 1987 P. Molinari Office of Marine and Estuarine Protection U.S. EPA Summary of EPA actions to identify sources of flotables problems in New York Bight. How EPA plans to use its various authorities (MPRSA, CWA and RCRA) to reduce future flotable problems. How EPA plans to work with state/local governments and other federal agencies to handle future flot- able incidents that occur. 1609 LOW COST ROVS FOR SCIENCE Lance L. Stewart and Peter J. Auster The University of Connecticut Groton, CT NOAA's National Undersea Research Program at the University of Connecticut, Avery Point, is the national center for test and evaluation of low cost ROVs for scientific applications. LCROVs were viewed as easily transportable and deployable undersea platforms that could deliver specially designed sampling devices and cameras to depths of 330 meters. Since 1984, we have evaluated and developed procedures and tools for utilizing ROV technology to address a wide variety of sampling tasks. Operations in 1988 include a dedicated ROV science program supporting 21 research missions covering a multitude of scientific tasks. Examples of research missions include: herring egg bed studies in the northern Gulf of Maine; juvenile finfish and crustacean habitat definition on the southern New England shelf; hypoxia effects in Long Island Sound; hydrothermal activity in Yellowstone Lake; macrozooplankton migration and sediment transport in the U.S. Great Lakes; deep sound scattering layer definition in the Gulf of Alaska; and the ecology of the deep bdsins--of@-the-East African Rift Lakes system. 1610 THE HYDROCARBON POTENTIAL OF THE FEDERAL OCS, OFFSHORE NORTHERN CALIFORNIA James M. Galloway and Michael R, Brickey Department of the Interior Los Angeles, CA A series of four sedimentary depositional basins, extending from Cape Mendocino to Point Conception, underlie much of the Federal OCS and State Tidelands. These basins, the Point Arena Basin, the Bodega-La Honda Basin, the Ano Nuevo Basin, and the Santa Maria Basin, are closely related in their stratigraphic and structural framework. The first exploration efforts conce-ntrated in the onshore portions of the basins, most notably in the Santa Maria and La Honda Basins. Offshore exploration began in the Federal OCS following the 1963 Lease Sale. Early inconclusive results were followed up by the recent successes in the off- shore Santa Maria Basin. Natural oil seeps and exposed oil-bearing strata are present at many lo- cations along the coastline, and suggest that hydrocarbon-rich rocks also occur in the offshore basins. In the 1960's, the principal exploration tar- gets were sandstones of Pliocene age. Currently, exploration emphasis is placed on older Miocene-aged rocks, especially the Monterey Formation. The Monterey Formation, and its equivalents, are prolific source rocks. Its arguable quality as an economic reservoir rock onshore suggested to many geologists that it would not be a viable exploration target offshore. How- ever, discoveries on Lease Sale 48 and 53 tracts in Mioce-ne rocks of the Santa Maria basin suggest that Miocene rocks in the other offshore basin may also contain hydrocarbon reservoirs. 1611 WILSON ROCK FIELD: A CASE HISTORY Scott Sorenson, Cn-ris ALonzo and Maher Ibrahim U. S. Department of the Interior Los Angeles, CA Wilson Rock Field is located in the Santa Barbara Channel, approximately 20 miles southwest of Gavicta, California. The blocks containing the field were leased during OCS Lease Sale 48 in June 1979 for bonus bids totaling $32,922,144. The discovery well was spudded in July 1983 from a semisubmers- ible in 1,547' of water. Sediments ranging in age from Pleistocene to Cretaceous were penetrated. Drill stem tests produced 43 API gravity oil and gas at cumulative rates of approximately 600 BOPD and 700 MCFPD. The Wilson Rock Unit was formed in August 1984 to facilitate development of the field. Constraints on development included water depth, distance from shore, estimated recoverable resources and declining oil prices. Analysis of the available data indicates that these factors prohibited development under the prevailing economic conditions, resulting in the termination of the leases in June 1987. The Wilson Rock Field is an ideal exampla of recoverable resources that remain undeveloped due to technological considerations and changing economic conditions. 1612 OUTER CONTINENTAL SHELF FISHERIES AND RESOURCES IN THE GULF OF MEXICO B. G. Thompson National Marine Fisheries Service Department of Commerce About 44 percent of the total Gulf commercial landings came from the Outer Continental Shelf (OCS) in 1986. OCS landings in this region are dominated by shrimp trawling and menhaden purse seining operations. In 1986, 134.7 million pounds of shrimp were taken in the OCS, almost 45 percent of the entire Gulf shrimp catch. The shrimp fisheries off Texas harvested 79.2 million pounds, followed by Louisiana with 22.8 million pounds. The Louis- iana and Texas shrimp fleet is comprised of approximately 3,400 vessels and 4,000 boats employing 16,000 full- and part-time fishermen. Menhaden landings of 821.3 million pounds, all from off Louisiana, dominated the commercial OCS finfish harvest in 1986 volume. Snappers, groupers, and king and Spanish mackerel were the dominant finfish catch from the Florida OCS; snappers and flounders in Alabama; snappers in Mississippi; and snap- pers, groupers, and tuna species dominated the Texas OCS finfish harvest. The.total estimated 1986 recreational harvest from the Gulf OCS was 27.9 million pounds, about 33 percent of the Gulf-wide catch. Catches of spotted seatrout, Spanish mackerel, snappers and groups dominated the harvest. In total, almost 4 million recreational anglers made 17.9 million trips in the Gulf in 1986 Since 1984, the National Marine Fisheries Service has obtained data on the association of recreational fishing with oil and gas structures. In the Gulf OCS off Louisiana, up to 70 percent of all recreational fishing trips took place within 200 yards of a structure, although there was no significant difference in the number of fish caught per trip compared to fishing away from structures. 1613 AN INEXPENSIVE MOBILE SELF-CONTAINED HABITAT SYSTEM FOR MARINE RESEARCH Robert I. Wicklund Caribbean Marine Research Center Riviera Beach, Florida A mobile, self-contained habitat system has.been developed for marine research by Perry Offshore, Inc. and the Caribbean Marine Research Center. The habitat is a two-person ambient system measuring 9ft x 7ft x 612ft with full-size flat windows on three sides. Initially, power has been provided by,a battery pod set on the bottom allowing the habitat to be co mpletely free of surface support. The batteries provide 60 amps at 24 volts to power a 002 scrubber, internal lights, exterior lights and auxiliary outlets for instrumentation. A solid polymer fuel cell producing 40 amps at 36 volts is in the final stages of development as an alternate source of power. The habitat system is highly mobile, designed to be launched and retrieved from the deck of a vessel in less than two hours. Decompression can be' conducted in a deck chamber, while the habitat is being moved from one location to another. 1614 CUMULATIVE ENVIRONMENTAL EFFECTS OF THE DEPARTMENT OF THE INTERIOR'S OFFSHORE OIL AND GAS PROGRAM: 1987 REPORT TO CONGRESS Jeffrey P. Zippin Department of the Interior Reston, VA As part of the Minerals Management Service's (MKS's) responsibilities for administering oil gas leasing and development activities on the Outer Continental Shelf (OCS) and in the Exclusive Economic Zone, the MMS prepares an annual report on the cumulative effects of the program on the environment. As the first cumulative effects report, the 1987 report traces the effects,of offshore oil and gas activities since the program's inceptiou in 1954. Studies sponsored by the MMS and effects observed and reported in the literature indicate that acute effects of oil and gas development activities to marine biota and habitats are both highly localized and short lived. Follow-up investigations on recovery of marine and nearshore habitats following oil spills have revealed a remarkable resilience by many species and rapid recovery of affected areas. No chronic, low level, long term effects attributable to offshore oil and gas activities have been noted, although the MMS continues to monitor for these effects through the Environmental Studies Program. 1615 LEGAL AND POLICY ISSUES AT STAKE IN THE CURRENT 5-YEAR PROGRAM Emmett Turner Paul R. Stang U. S. Department of the Interior Washington, DC The current 5-Year OCS Oil and Gas Program was approved by the Secretary of the Interior in July 1987 and immediately litigation was filed against it, asserting that the program was deficient and in violation of the OCS Lands Act in several respects. The ruling of the U.S. court of Appeals of the District of Columbia is expected in fall 1989. If the Court has ruled by the oceans 88 Conference, the paper will focus the Court opinion and its implications. If the Court has not ruled, the paper will outline some of the issues and the key-arguments (in laymens language) of the Federal Government and the petitioners in the case. Also, the paper will address evolving policy issues confronting the OCS leasing program with emphasis on the Department of Interior's efforts in conflict resolution. 1616 A SHALLOW WATER SONAR PERFORMANCE PREDICTION SYSTEM Arnold Novick Mission Sciences Corp. Cormack, NY Standard Navy sonar system performance prediction models have typically focused on deep water appli-cations. Mission Sci,--nces' has developed a family of ray based shallow water predi-c-,ion models which form a physically realistic, cqpsistent and robuL->t system for the prediction of system signal, noise and reverberation levels. The embedded propagation loss model, SMART, has tested well against several difficult ocean data sets. This paper describes the shallow water prediction system including comparisons with measurement and other prediction models. 1617 A VCR BASED DIGITAL DATA RECORDER FOR UNDERWATER ACOUSTICS MULTIPATH MEASUREMENTS J. Matthew Tattersall, Joseph A. Mingrone and Peter C. King Naval Underwater Systems Center New London, CT A family of multichannel underwater acoustic data acquisition systems which are deployed to the ocean floor and utilize Video Cassette Recorders as a digital data storage medium have been developed. One of these systems, the Acoustic Transient Recording Buoy (ATRB), was used recently as part of an-ocean boundary-interaction acoustic measurement program. Explosive SUS cha"es were used as a source of transient high energy wide band acoustic signals and the direct, surface, and bottom-interacting multipath arrivals from the detonation of. these charges were recorded by the bottom moored ATRB-system. Acoustic data obtained from the first two deployments of ATRB are of very high quality with wider bandwidth and dynamic range than data sets acquired during earlier experiments using analog recording methods. The ATRB recording system is described in this paper and some of the acoustic data collected during the measurement program are presented. Funding for the development of the ATRB system was provided by the ONR AEAS Program (K. Dial, program manager). 1618 AN INTRODUCTION TO THE PHYSICS OF UNDERWATER SOUND AND THEIR APPLICATION TO PASSIVE-ANTI-SUBMARINE WARFARE Douglas K. Roderick Texas A & M University Galveston, Texas The importance and interaction of the physical aspects of underwater sound to the United States' passive anti-submarine warfare effort has been recog- nized and well documented. The United States involvement in the undersea warfare effort is linked directly to the massive Soviet build-up of its subsurface force. We have recognized the need for a thorough understanding of the basic principles of underwater sound as the primary phenomena gov- erning our application of passive anti-submarine warfare. This material displays the principle basics of underwater acoustics and its association with the undersea warfare effort. In addition, this paper presents some highlights of the present state- of-the-art applications in passive anti-submarine warfare and gives some possible future directions that this field might pursue. 1619 REALTIME SIDE SCAN SONAR TARGET ANALYSIS Ray Gandy and Steven Paulet SEAQUEST Niantic, CT. Sonar records require analysis by experienced personnel who can identify potential targets from the surrounding natural features and other seabed debris. The data required in an anomaly analysis is generally obtained from a topographical representation of the seafloor using manual measuring techniques, i.e. rulers, calipers, etc. Consequently the anomaly analysis is routinely performed at a location remof@from the actual sonar scanning location and often occurs several days after an area of interest has been mapped. Furthermore, the manual measurement procedure and computation of the relative location of an identified target is often very inaccurate. Subsequent attempts to observe and/or retrieve the object associated with the identified anomaly at the computed latitude and longitude can be very time consuming and possibly unsuccessful. It is the intent of this paper to present and discuss a revolutionary method of analyzing Side Scan Sonar records. The Instant Target Analysis Computer (INTAC) System enhances target analysis in two ways. The first speeds up the analysis procedure by avoiding tedious manual measurements of a sonar contact. The second enhancement is the ability of INTAC to determine the position in latitude and longitude of a sonar target. This position can then be used to 'guide submersibles or divers to the exact location of the target for investigation or recovery purposes. This instant target analysis is performed realtime, thus avoiding costly project delays. 1620 HARDWARE/SOFTWARE PACKAGE FOR CTD DATA ACQUISITION WITH A NEIL BROWN CTD David Nebert, Howard Saklad, Gil Mimken Institute of Marine Science University of Alaska Fairbanks, Alaska 99775-1080 INTRODUCTION ASCII configuration files that sit outside of the compiled program. The Institute of Marine Science Once the software was tested and (IMS), University of Alaska, has proven, new deck hardware was built built a low cost hardware/software that eliminated the need for the CTD system to replace an aging Neil Neil Brown deck unit. The complete Brown CTD deck unit and 9 track system is comprised of an IBM-XT/AT tape drive. The new IMS CTD Deck or clone, a rack mounted p9wer System includes a commercial power supply and a compact, 8"x8"x2 I /2It supply, a small hardware package case containing the CTD deck and microcomputer software which hardware. ,provides enhanced features such as data acquisition concurrent with real-time computer screen plotting THE SYSTEM of the cast. The package demonstrates that a versatile, Major concerns driving the design of reliable, computer based system can the system are that it be simple to provide low cost redundancy and operate, reliable as far as data serve most CTD data collection acquisition is concerned , and easy needs. The software and hardware to maintain. While development schematics will be provided at no costs were somewhat higher than Cost to interested institutions. originally anticipated, all the above goals have been met. BACKGROUND The software package has been designed so that it will lead an The policy at the Institute has operator through a series of menus. been to maintain redundancy for sea It has taken only few casts for a going CTD components for the sake person who has only minimal of reliability. Since we had a knowledge of CTD cast operations to MARK III Neil Brown deck unit, it become competent with the system. was decided that to add a new model The software has been constructed so would not only be expensive, but that only authorized responses are would leave us with two different accepted; others are ignored (with a model deck systems. beep) . This makes the system difficult to look up. Critical The first step was to build operations (such as halting data electronics to interface the Neil collection) require a specific, Brown deck unit with an RS-232 simultaneous two-key response. The computer interface. The software capabilities of the system include package was designed to meet our cast preparation, data collection to various needs but has been kept hard disk while viewing real-time modular and flexible by the use of data, dumping plots or data to a 1621 printer, collecting calibration The hardware consists of two information, and plotting or components; the signal demodulator listing previously collected data. and a power supply. Thedemodulator There is an option to select a has been designed to be minimal with series of stations and then have emphasis on simplicity, placing most them plotted or . listed of the system sophistication in automatically. Up to three software. This was done to allow parameters can be simultaneously easy enhancement without having to plotted on the screen; the user can return to the electronics shop for alter both plotting scales and soldering iron and parts, as well as parameters to suit changing needs. to ensure maintainability in the (A detailed list of features for f ield. The design for the the IMS CTD Deck System Software demodulator uses simple circuitry are available on request.) and modern componentry. Functionally, it replaces the Neil The software is written in Brown MARK III deck unit. The Microsoft QuickBASIC 4.0 and is compact circuit design has proven documented within the program. highly stable, requiring minimal Both compiled and source codes are adjustment. The cost of fabricating available. The software does not of the demodulator can be under $200 include Microsoft QuickBASIC, but plus labor. Schematics will be the compiled version operates as a provided at no cost to those who standalone. Code modification will wish to build their own signal require Microsoft QuickBASIC. demodulator. Evaluation of many of the system A Hewlett Packard 6443B power supply features (exclusive of data ($1300) was chosen to drive our CTD collection) are possible by using system, providing sufficient power the software without the hardware for add-on devices such as a or CTD connected; test CTD data fluorometer. It is a constant will be supplied. voltage/constant current supply with meters which aid in system Raw data are recorded to hard disk diagnostics. We feel that these during data collection. One-meter added capabilities justify the averaged data are computed (ASCII) additional cost of a more after the cast is terminated, and sophisticated power supply. A less can be used to speed subsequent elegant power supply will meet the shipboard analysis. An alterable basic needs of the system for under batch file is run after each cast $800. that allows raw or one-meter averaged data to be copied to It's possible to use the software floppy, streaming tape or other module with a Neil Brown MARK III preferred means of storage for deck unit with no modifications to redundant data security. Because the NB deck unit except a level of the low cost, exclusive of the shifting adapter to match the TTY MS-DOS computer ($200 plus labor output to RS-232. for signal demodulator and $800 for a power supply) it is possible to obtain affordable redundancy with COMPUTER REQUIREMENTS this system. In additioni since the computer (or two computers, if An IBM-XT/AT or clone is needed, you wish complete redundancy) is with two floppies or one floppy and not dedicated, it may be used for a a hard drive, one serial port f or variety of other data processing data input and a CGA or EGA video tasks. adapter. The software was 1622 originally developed with Microsoft ACKNOWLEDGMENTS QuickBASIC versions 2.1 and 3 which do not support Hercules Monochrome This project was financed primarily graphics or VGA resolution. Only with funding from National Science CGA is currently supported. Since Foundation grant OCE 8619385. QuickBASIC 4.0 does support Supplemental funding was provided by Hercules and VGA standards, our the Institute of Marine Science. We plans include an upgrade to support also wish to thank the many users of these standards. A parallel port the system, David Leech in is needed for screen dumps or for particular, who have provided the listing data. The preferred feedback necessary for us to build a configuration is an AT operating at better system. 8 MHz (or above) with a hard drive and one or more floppy drives. If many casts are normally taken per cruise, it may be desirable to add a Bernoulli drive or a demountable hard drive (or optical disk?) for transporting data from the ship. ADDITIONAL PROCESSING OF DATA There presently is no data processing capability beyond that described above. We expect to build limited plotting capabilities to expand the simple screen dump and listings already provided. This might include such things as parameter/depth, T-S and cross- sections plots. There is no commitment on the part of the Institute of Marine Science to provide processing capabilities beyond those present in the original software package. Applications programs can be developed to interface with both One-meter averaged files and raw data files (one-meter averaged files are presently ASCII; raw data are binary). Users are free to develop their own applications programs in any language they wish since applications do not interact directly with the acquisition program, but rather with MS-DOS data files. 1623 OUTER CONTINENTAL SHELF FISHERIES AND RESOURCES IN THE NORTHEAST REGION Darryl Christensen National Marine Fisheries Service Department of Commerce About 39 percent (517.7 million pounds) of the total commercial landings from the Northeast Region were harvested from the Outer Continental Shelf (OCS) in 1986. These landings had a total value of $367.4 million, of which Massachusetts contributed over 48 percent. The predominant landings in the Region were Atlantic cod, pollock, flounder, surf clams, sea scal- lops and ocean quahogs, which comprised 57 percent of the OCS harvest by weight and 69 percent by dollars. Sea scallops were the most valuable fishery at $86.1 million, while flounder comprised the greatest quantity at 73.9 million pounds. The commercial fishing fleet is comprised of approximately 4,604 vessels and 45,748 boats employing 80,450 fU117 and part-time fishermen. The total estimated 1986 recreational harvest from the Northeast Region OCS was 58.4 million pounds, about 38 percent of the total marine recreational landings. Catches of black sea bass, bluefish, Atlantic mackerel, weakfish, and scup were the most popular, composing 42-percent of the harvest. In total, almost 4 million marine recreational anglers made over 28 million trips in the Northeast during 1986. 1624 OCEAN SENSING CAPABILITIES ON LANDSAT 6 Matthew R. Willard Earth Observation Satellite Company Lanham, MD The Earth Observation Satellite Company (EOSAT), in conjunction with NASA and the U.S. Navy, will fly a Sea-Wide-Field-Of-View Sensor (SeaWiFS) on Landsat 6. SeaWiFS builds on the herizage of I NASA:s Coastal Zone Color Scanner (CZCS) and NOAA's Advanced Very High Resolution Radiometer (AVHRR), and will provide real-time ocean color and 1hermal data in the early 1990s. Some of the potential uses of SeaWiFS data by operational and commercial users include: locating fish populations, optimizing ship routing, and assisting in design of offshore oil platforms. SeaWIFS data will also* be@.:.used by the oceans research community to support global research projects involving: quantative evaluf1tion of the ocean's role in the global carbon cycle and other major bio- geochernical cycles; determination of the magnitude and variability of annual primary production by marine phytoplankton on a global scale; and study of marine optical properties and upwelling patterns on a regional and global basis. In sum, the SeaWIFS mission will provide an unique oceans data set in the early 19190s essential to a number of global oceans research missions as well as being essential to improving the cost- effectiveness of a wide range of ocean industry users. 1625 AN INEXPENSIVE INTERACTIVE PROCESSING SYSTEM FOR NOAA SATELLITE IMAGE W. B. Campbell and M. L. Weaks NOAA/NESDIS Camp Springs, MD A PC based image processing system has been produced to enhance the image analysis functions within NOAA and to facilitate high quality product creation and rapid dissemination. It uses basic commercially available components and NIOAA writ- ten software. This system allows for-display of mapped, gridded, full 11-bit resolution image iriform@tion from LAC, GAC, or GOES images. The dynamic color enhancement utilities allow for easy interpretation of ocean thermal structure details because each area of an image can be custom enhanced witb as many or as few color intervals as necessary and annotated UL, separable non-destructive overlay planes for future use or direct distribution. The software is menu driven and quite simple to operate. 1626 CONSEQUENCES OF THE ABANDONED SHIPWRECK ACT: THE NORTH CAROLINA EXAMPLE Richard W. Lawrence Underwater Archaeology Kure Beach, NC One of the uncertainties surrounding the "Abandoned Shipwreck Act", should it be passed, is how the various states would manage their shipwreck resources. Perhaps the best way to answer this question is to look at state with existing submerged cultural resource programs. This paper will examine North Carolina's underwater archaeology program, which has been active for over twenty yea=rs, and how the Underwater Archaeology Unit manages the state's shipwrecks, and deals with such issues as diver access to shipwrecks, recovery and salvage permits, and treatment of recovered artifacts. Particular attention will be given to the contribution that sport divers and volunteers have made in discovering and preserving our maritime heritage. 1627 LONG-TERM ABRASION AND CORROSION DAMAGE TO THE HAWAII DEEP WATER POWER CABLE J. Larsen-Basse, B. E. Liebert, K. M. Htun and A. Tadjvar National Science Foundation Washington, DC A 300 kV deepsea power cable is being designed and tested for possible deployment between the islands of Oahu and Hawaii. Part of the deep ocean section of the route includes regions with steep lava ledges washed by strong tidal currents. It is anticipated that sections of cable could become suspended between ledges and be moved back and forth by tidal drag forces. Then abrasion and abrasion-corrosion become potential failure mechanisms. In this papex,--"we,-evaluate-li4l-.erature data and corrosion test results for both armor wire samples and prototype cable specimens exposed in both surface and deep ocean Hawaiian waters. The results are combined with abrasion and friction data fo.r,armbr wire-lava rock combinations to develop an estimate of the expected worst-case damage in an 30-year design life. According to this estimate the greatest damage will occur for a cable catenary which, being sufficiently short to avoid tension fracture (<60 m) or long-term fatigue of the lead sheath (<40 m), has a contact load of 100 kg at mid-point, a corresponding friction-limited excursion of 0.38 m and a 30-year distance of travel over the rock of 33 km. Here, the expected wear is-7 mm and general corrosion is 2-mm, leaving about 3 mm before the rock face comes in contact with the lead sheath. On this basis, it is concluded that the cable most probably will survive damage due to abrasion-corrosion. 1628 UPWELLING MONITORING OFF WESTERN SAHARA 2 Clemente-Colon and J. Zaitzeff NOAA/NESDIS Washington, DC Upwelling off the northwestern coast of Africa occurs year around mainly as a response to prevailing northerly winds which parallel the coastline. Bottom topography and non-local large scale atmospheric events have also been cited as important upwelling factors in this region, one of the most important commercial fishing grounds in the world. A database of sea surface temperatures (SST) has beldn produced for the Canary Islands waters ant the African coast between 21N and 30N to study the ocurrence, satellite detectability, and temporal variability, of upwelling activity in the region. Thermal infrared data from the advanced very.high resolution radiometer (AVHRR) aboard the NOAA 9 satellite was used to produce atmospherically corrected SST maps.. A particularly important fishing ground for the spanish fishing fleet based in the Canary Is. lies between 24N and 25 N off the coast of Western Sahara. This area has been selected for a detailed study of SST variability. A time series of STT over a 1/2 degree latitude by 1/2 degree longitude area was produced for a one year period, summer 1986 to summer 1987, utilizing the available cloud-free data. The data shows that while SST's are generally cooler during the winter and spring months, relatively strong upwelling events occur troughout the entire year. on the other hand, decreased upwelling activity is observed twice in the year during early October and mid-March. 1629 OPTIUM TECHNIQUES FOR TRACKING PLUMES IN THE OCEAN: A CASE STUDY OF SLUDGE PLUME DISPERSION AT THE 106-MILE SITE S. E. McDowell, C. S. Albro, W. R. Trulli, W. G. Steinhauer and F. G. Csulak Battelle Ocean Sciences Duxbury, MA With increased public concern for pollution of the marine environment, ocean disposal of manmade wastes has become a topic of growing interest. Under contract to the U.S. Environmental Protection Agency, Battelle Ocean Sciences is conducting an oceanographic monitoring program at the 106-Mile Deepwater Municipal Sludge Site to determine the fate and effects of sewage sludge dumped at the site. A series of cruises.has been conducted to, collect physical oceanographic and chemi.cal data from within sludge plumes to determine rates of plume dilution and the effect on water quality within.and around the site. The success of the recent studies has been due to an integration of physical and chemical sampling techniques Which allows 1) tracking of sludge plumes for periods up to 1 day, 2) collection 'of hydrographic and current data to resolve plume advection and dilution, and 3) directed water sampling within the plumes for analyses of trace metals and organic compounds. The optimum measurement tool for tracking sludge plumes is an in situ transmissometer interfaced to a CTD profliler and Loran-C navigation system to provide real-time, 3-dimensional information on plume turbidity. Results .from these field studies will be presented, and the use of this measurement capability for other pollution monitoring programs will be discussed. 1630 A MONITORING PLAN FOR DISPOSAL OF SEWAGE SLUDGE AT THE 106- NILE SITE C. E. Werme, P. D. Boehm, W. G. Steinhauer and F. G. Csulak Battelle Ocean Sciences Duxbury, MA The 106-Mile Deepwater Municipal Sludge Site,(106-Mile Site), located approximately 120 nmi southeast of the Ambrose Light, New York, beyond the edge of the continental shelf, is currently designated for the disposal of sewage sludge originating from-19 sewage treatment plants in the New York metropolitan area. A monitoring program for the 106-Mile Site has been designed which will generate data to be used by site managers to make decisions about site redesignation or dedesignation, and continuation, termination, or modification of dumping permits. The monitoring plan for the program describes the regulatory basis of monitoring, and how site and waste characteristics have been used to predict possible impacts of sludge disposal and to formulate null hypotheses that these predictions suggest. The impacts of sludge disposal have been organized into an implementation framework of tiers. The tiered approach organizes the null hypotheses into a hierarchy, whereby data collected in each tier are required as the foundation for the design and extent of monitoring activities in the next tier. The four tiers included in the monitoring program are (1) Sludge Characteristics and Disposal Operations, (2) Nearfield Fate and Short-term Effects, (3) Farfield Fate, and (4) Long-Term Effects. 1631 MONITORING WATER QUALITY CHARACTERISTICS DURING DISPOSAL OF SEWAGE SLUDGE AT THE 106-MILE SITE C, D, Hunt, W. G. Steinhauer, C. E. Werme, P. D. Boehm, and F. G. Csulak Battelle Ocean Sciences Duxbury, MA Regulation of sewage sludge disposal at the 106-Mile Deepwater Municipal Sludge Site is contingent upon achieving Environmental Protection Agency saltwater quality criteria (WQC) or limiting permissible concentrations (LPC) within four hours of disposal. These criteria must be met in less time if the plume is transported across the site boundary in less than four hours. Verification of whether WQC are being met requires sample collection from within the plume. Collecting these samples presents several difficulties, including positioning of the sampling platform within the plume, location of the maximum concentrations of the plume at depth, and collecting the samples from within this maximum. Combining in situ sensors, -I T_ including CTD and transmissometry, with realtime shipboard disp ay of data, and high-volume pumping systems has provided the necessary direction and control over sample depths to successfully sample sewage plumes for up to 8 hours following disposal. This capability allowed collection of water-samples for several parameters useful for tracing the dispersion and dilution of the sludges in addition to determining concentrations relative to marine WQC. The results from samples collected in several plumes during September 1987, indicate dumping rates currently used at the 106-Mile Site may be affected if permits to dump at the site are issued. 1632 HYPOXIA IN LONG ISLAND SOUND (LIS), SUMMER OF 1987 Barbara L. Welsh Marine Sciences Department University of Connecticut Avery Point, Groton, Connecticut The seasonal depletion of oxygen in western LIS in summer, 1987, reached the lowest levels heretofore documented for the area. Deficits in waters below the pycnocline began in the westernmost basin in mid-Jun6, and, hypoxic conditions (< 3ppm) spread eastward to encompass most of the western half of the Sound by mid-August. The area of deepwater affected in 1987 was similar to that in 1986. In 1987, however, a dinoflagellate bloom developed in mid-July, aggravating the deepwater situation to the point of causing anoxia in the mainstem of the system, linking the mass of hypoxic deepwater to hypoxic water in the adjoining embayments, and causing unusually severe kills of fish and invertebrates. The widespread hypoxia in.1986 is believed to have resulted from the eutrophying effects of nutrient additions from sewage treatment plants (STP). The extreme conditions brought on by the dinoflagellate bloom in 1987 are believed to have resulted from a coupling of unusually stable meteorological con- ditions to regular STP additions plus additional STP failures. 1633 EPIDEMIOLOGY OF BOTrILENOSE DOLPHIN DISEASE --U.S. ATLANTIC COAST, 1987-1988 Joseph Geraci University of Guelph Guelpj, Ontario, Canada This paper describes the history of the epidemiologic investigation into the bottlenose dolphin disease which ranged from northern.New'Jersey to Central Florida in 1987-1988. A simultaneous die-off of dolphins occurred in New Jersey.and Virginia in July 1987. An interagency'team was formed which examined animals for biotoxins, pollutants, viruses and bacteria in an extensive patho.bi.ology invesElgation. Results of the several lines of investigation are presentedand an epidemiologic synthesis is discussed. 1634 DRIFT CURRENT MEASUREMENTS FROM A SUBMARINE Peter J. Hendricks Naval Underwater Systems Center Newport, RI A simple, accurate technique for measuring ocean currents from the set and drift of a submarine is described. The drift current is the vector difference between two velocity vectors: the inertial velocity determined from navigation data and the relative velocity, which is calculated from the electromagnetic speed log and the gyrocompass heading. Drift-currents may be calculated routinely and displayed for submarines underway with a modest amount of computation or calculated from data recorded on tape. An example of submarine drift current calculations is presented from a multi- ship exercise in the Sargasso Sea, including a cold-core, Gulf Stream ring. These data illustrate the capability of the technique for resolving the complex current structure of the region and demonstrate that the current estimates from different ships are consistent to within a few tenths of a knot or less. 1635 IMPLICATIONS OF USING A WIDE SWATHE SOUNDING SYSTEM Dr. Roger L. Cloet Bathymetrics Bath BA2 3DW England There is an ever growing evidence that surveys of the seabed obtained by means of echo sounding does no longer meet the stringent demands for greater positive identification of navigational and seabed engineering hazards and for a quality assurance of the data. Although sidscan sonar has been able to assist the surveyor over the last couple of decades, it has required an interpretative skill, not always available to the busy operator, and has proved somewhat subjective as a consequence. Interferometric swathe sounding has enabled the generation of depth data in a uniform density pattern across the survey area thereby removing the orientation bias inherent in echo sounding surveying. The widths of the swathes achieved are such that it is operationally possible to ensure sufficient overlaps, and thus duplication of data, so that objective quality control procedures can be applied. Using the procedures has resulted in evidence that hitherto sometimes unsuspected inadequate per- formances of complementary instrumentation, or insufficiently precise environmental data, can be monitored, quantified and, if need be, corrected. The data rates are greatly augmented thereby improving the resolving capacity of the survey in terms of identification and delineation of seabed features. The other side of the coin is that the high data rates require the application of a strict discipline in data base management while also providing opportunities to supply the customer with much more detailed environmental information wherever he wants, or needs, it. The proposition is that several versions of the data base should be available, each being tailored to the specific requirements of major customer type groups. In this way data will be used for fully if selectively enhancing the cost benefit derived from the original survey operation. 1636 DEVELOPMENT OF SEPARATOR TRAWL TECHNOLOGY Philip H. Averill Maine Department of Marine Resources Augusta, ME A project to develop a shrimp separator trawl (SST) has resulted in a number of viable designs that catch shrimp without catching or retaining unwanted fish species. A body of knowledge on net development strategies as well as on fish reactions to various net configurations has also been assembled. All designs tested as possible SST configurations will be described. Also explained will be the evolution of the various techniques used to comparitively evaluate the wide variety of designs tested. Experience with "trouser trawls" of various designs will be emphasized. 1637 RECENT WAVE ENERGY RESEARCH IN SWEDEN Lennart Claeson Technocean AB Gothenburg, Sweden Wave energy research has been carried out in Sweden since 1976. About $6 000 000 have ,been spent - one half by a governmental research programme and one half by Swedish industry. The main aim of governmental program is to @evaluate if wave energy could contribute to the future Swedish energy supply - especially in the light of the decision that the nuclear power shall be settled before 2010. There are three industrial projects: - The IPS-buoy (Interproject Service AB). A prototype was built in 1979 and tests were .carried out at sea in 1980 and 1981. In 1985 a smaller buoy, for navigation purposes was tested. - The hose pump concept (Celsius Industries AB). Sea trials have been carried out in 1983 and 1984. A small light-buoy was also tested in 1985. Calculated cost for large scale power plants is less than 4 @/kWh. - The Wave Rotor is a new promising concept of which only small models have been tested yet. 1638 THE SSV CORWITH CRAMER: SEA EDUCATION ASSOCIATION'S NEW SAILING RESEARCH VESSEL R. Jude Wilber, Charles E. Lea and Susan E. HuMphris Sea Education Association Woods Hole, MA With the launching of the SSV Corwith Cramer the Sea Education Association of Woods Hole, MA now operates two sailing vessels which serve as research platforms for Sea Semester. This program, using the 125' schooner Westward, has provided practical, at- sea experience in oceanography to over 2000 undergraduates over the past 17 years. The Cramer, which is the first vessel built to meet U.S. Coast Guard regulations for Sailing School Vessels, is modeled closely after Westward. A variety of design modifications and equipment chancres greatly improve her performance as a research vessel. Cramer's greater length (135 feet) and bean (26 feet) serve to increase speed, stability and working deck area. The dry lab amidships is nearly twice the size of Westward's and designed specifically for the type of research projects conducted during Sea Semester. The vessel's on-board research equipment consists of standard oceanographic sampling gear such as plankton nets, trawls, cores, dredges, and Nisken bottles. Sampling gear is deployed amidship over either side of the vessel using a 9hp variable speed winch carrying 500 m of 1/4" hydrowire. In addition, the Cramer carries a variety of modern sampling and analytic gear including a 3.5 and 12 kHz PDR, digital recording BT and CTD salinometer, fluorometer, spectrophotometer and four micro-computers for data recovery, processing and storage. With its new design and equipment the SSV Corwith Cramer provides SEA with an economical, stable and well-equipped research platform which will continue to provide students with a practical introduction to oceanography for many years to come. 1639 HAZARDOUS MATERIALS IN MARINE SALVAGE OPERATIONS J. Kenneth Edgar Diversified Technologies Alexandria, VA Hazardous materials are on board ships in abundance. In addition, materials that are normally benign may become hazardous when exposed to heat, air, water or undergo a chemical reaction. materials long held as being safe have been found-to be hazardous or toxic; for example, cellulube is carcinogenic. Similarly, statutes and regula- tions, handling, reporting, transporting and disposal requirements are dvnamic and of tremendous impact on any operation involving hazardous/ toxic materials. The problems of managing hazardous materials and hazardous situations have grown in scope and complexity. Recent salvage cases worldwide have demonstrated this. Most salvors have very little'knowledge, an,"' even less experience,-with operations involving haz'ardous materials. Trai-ning is almost non-existent, and the available documentation was.not written towardis the salvor. Similarly, the salvor normally does not have the ezuiTment readily available to react to a hazardous situation. Any hazardous material operation may very likely tie to a pollution or firefighting operation or both. Other issues such as generator designation, liability, disposal, etc. can add enormous.complexity, risks and expense to an operation. This paper addresses the issues a salvor must confront when faced with an oneration that involves hazardous or toxic materials. 1640 THE FUTURE OF THE TOURIST SUBMARINE INDUSTRY W. Burton Hamner University of Washington Seattle, WA The market for underwater sightseeing aboard submersibles specifically designed for it has been well established, and competition is underway. Sites for operations are being developed worldwidi-. Eventually the number of people going on submarine rides in the coastal oceans of the world could exceed 40,000 per day. Because they are becoming mass transport vehicles for recreation, tourist submersibles will suffer the accidents that befall such operations eventually. Unlike other kinds of transportation, an accident with loss 'of life could totally discourage the market. The future of tourist submersibles will likely be strongly affected by the position and cohesiveness of the industry when an Accident occurs. An outline is presented of a predictive approach that utilizes ihree separate but similar models from the literature. They are a technology assessment, an industry competition analysis, and a behavioral stimulus- response model of one particarly important regulatory agency, the U.S. Coast Guard. The models are used as overlays on a complex picture to help identify significant features and their interactions. The most apparent scenarios for the future of the tourist submersible industry are described. 1641 HYDRODYNAMIC AND MASS TRANSPORT MODELI'NG OF NAVY HARBORS J. J. ZAGEL, R. T. KILGORE AND S. M. STEIN GKY and Associates Springfield, VA Hydrodynamic water quality modeling of eight U.S. Navy harbLors is being performed to acertain the effects of organotins on harbor life, Organotins are the essential constituents in a fairly new-family of powerful anti-fouling nautical paints that have proven very effective against biofouling when applied to seaborne vessels, and have thus resulted in significant associated fuel and maintenance cost savings. DYNHYD (hydrodynamics) and TOXIWASP Cmass transport) a6r_e being used to model the hydro- dynamics and the transport and fate of these organotins within the following harbors: Mayport - Jacksonville, FL; Pearl Harbor, HI; Puget Sound, WA; Charleston, SC; Long Beach, CA; Philadelphia, PA; Portsmouth, NH; and Alameda, CA. For each harbor a detailed link-node network is established,, hydrologic and hydraulic data is collected, calibration and verification of the DYNHYD model, and dispersion calibration of the TOXIWASP model is performed. DYNHYD modeling characterizes the movement of water in the area of interest under alternative climatic conditions. TOXIWASP modeling determines parameter disperston throughout the water and sediment layers of the system. The modeli,ng is intended as a support to the design and implementation of monitoring efforts and as a general support tool for the Navy's water quality program. 1642 SCHEDULING PATROLS USING A HYBRID INTEGER PROGRAMMING/RULE-BASED SYSTEM APPROACH S. F. Roehrig University of Pennsylvania Philadelphia, PA Formal mathenatical programming algorithms have been developed for many scheduling problems of interest to the maritime community, but computational COMDlexity often restricts their use to disappointingly small problems. An interactive system which uses a rule base and advice from the scheduler to.formulate constraints for an integer progran will be discussed. The interaction results in fixing the values of many decision variables, thus considerably reducing the size of the decision space, and consequently, the-computational burden. As a sample problem, the task of scheduling U.S. Coast Guard Atlantic Area cutters is addressed. Some 30 ships are involved, with varying patrol lengths, maintenance and training requirements, and capabilities. The system illustrates the coupling of symbolic computing techniques with a high-level math programming language (GAMS). 1643 SHIPBOARD TECHNICIAN PROGRAM OF THE NATIONAL SCIENCE FOUNDATION H. Lawrence Clark National Science Foundation Washington, DC The NSF Shipboard Technician Program of the Ocean Sciences Division pro- vides support for basic technical support services on research vessels. Technician support grants are provided to.institutions specifically for the support of NSF-sponsored research projects scheduled on those research vessels operated by the institutions, and are concurrent with an award for Ship Operations. . The Shipboard Technician Program has two components: an at-sea com- ponent and an ashore component. The at-sea component consists of instruct- ing scientific personnel in the use of shared shipboard equipment; mainten- ance, repair, and calibration of scientific instrumentation; instruction and supervision of deck operations; and providing logistics assistance to on-coming and off-going scientific parties. On-shore activities include the maintenance, repair, and calibration of shared use seagoing equipment and assisting scientific personnel in preparing for field work on the vessel. Shipboard technicians funded by this Program have broad responsi- bilities for providing the coordination and technical assistance needed for the successful completion of research projects at sea. 1644 CHRONIC EXPOSURE EFFECTS OF TRIBUTYLTIN ON PEARL HARBOR ORGANISMS R. Scott Henderson Naval Ocean Systems C-nter Kailua, Hawaii Several species of harbor organisms were exposed in a chronic, flowthrough experiment to environmentally relevant tributyltin (TBT) concentrations in the range of about 10 to 100 ng/liter. Effects were seen only at the highest mean level of exposure (100 ng/liter) and included depressed condition indices of adult oysters (Crassostrea virginica) and reduced settlement of two endemic oyster species. The settlement and growth of 15 common fouling organisms and survival of anchovy baitf ish and portunid crabs were unaf f ected by all TBT exposures. Oysters, anchovies and crabs were sampled periodically for future analysis of TBT tissue burdens. The results of this experiment reaffirm observations from previous site-specific flowthrough tests in Pearl Harbor that show that chronic exposures to TBT levels of about 80 to 100 ng/liter and higher cause significant deleterious effects on sensitive marine epifauna. 1645 CONTROLLED DEPRESSOR TOWED SENSOR PLATFORM THE U.S. NAVY'S Mk28 SEARCH SYSTEM Michael Higgins and Roger Whyte Eastport International, Inc. Upper Marlboro, MD Eastport International, under contract to and direction of the U.S.'Navy's Explosive ordinance Disposal Technical Center, has developed, tested and delivered to the Navy a towed search system based on entirely new techniques and technology. The Mk 28, Mod 0 Search System, offers the Navy unequaled shallow water capabilities in locating objects using marine magnetrometers and, eventually, side scan sonar. The development and testing of high performance naval search systems is the subj ect of this paper. The Mk 28 is a fully integrated system composed of: an actively controlled towed depressor fish, a small tow winch and cable, surface control consoles, an integrated navigation system, and the search sensor and its processing equipment. The towfish has remotely controlled depressor wings which permit the fish and sensors to be flown at either a constant depth or a constant altitude. The fish also incorporates an acoustic positioning beacon and the ability to support side scan sonar. This paper will address the developme nt and testing of the innovative features of the system, mainly: the controllable depressor wing, control algorithms.(with hydrody- namic analysis), the integrated navigation system, and the sensor integration. Substantial use of the Navy's technical sea te6ting'data will be used to sup .port the paper. 1646 DEVELOPMENT AND TESTING OF A HEAVY-DUTY WORK ROV FOR 10,000 FOOT SERVICE Michael Higgins, B.ill,Lawson and Bill Field Eastport International, Inc. Upper Marlboro, MD EASTPORT INTERNATIONAL has upgraded the depth capability of the entire.GEMINI ROV system for 10 '000 feet bringing the GEMINI*into the vanguard of deep-ocean commercial ROV systems. The GEMINI development used designs at the forefront of today's sub- mersible technology, including: a fiber optic umbilical cable, dual spatially- correspondent heavyduty manipulators, and an a-ir-transportable aluminum handling system. The development and testing, due to be.complete in mid 1988 is expected to be com- pletely successful. This paper addresses and describes the engineering development and systems integration. Specific areas of discussion include: telemetry system design and integration for this great depth, cable design, power system engineering, vehicle sensor and manipulators/tools specification and integration, and handling system design. Sea testing is also addressed with a focus on operational problems and solutions when working in 10,000 foot depths. Deep ROV system technology has many. Iuseful applications in supporting offshore oil. and gas exploration, drilling, and production, and cable and pipeline support. Development of a 10,000 foot submersible proves the capability exists to support commer.cial operations.to this extreme depth. 1647 DESIGN AND APPLICATIONS OF SEATRAC, AN INTEGRATED NAVIGATION AND DATA MANAGEMENT SYSTEM Ray Gandy and Steven Paulet Structured Technology Corporation Niantic, CT Typically, position data processing equipment is manufactured to interface with particular positioning sensors or devices. In addition, the data processing and its overall effectiveness can be somewhat limited in particular applications. The need for the integration of navigation equipment along with enhanced data management capabilities has been evident for some time in the offshore industry. TITis need has been recognized in developing the state-of-the-art integrated navigation and data management system known as, SEATRAC. SEATRAC combines two primary functions incorporating the latest in microprocessor technology. The first is to integrate the signals from a variety of offshore positioning devices and optional digital data. The second function is to manage the collected data for visual display and permanent storage. The interface versatility and increased data processing capabilities make SEATRAC an affordable integrated system for offshore use as a shipboard display, plotting and recording system. Mission operations can be more readily evaluated with multicolor simultaneous display of ship and submersible position and track history. This results in instant recognition of prior and future positions. The operating area video presentation can be changed from an overall perspective to a close-up view wi.th the push of a button. Beacons, anchors, pingers, markers, etc., can be visually displayed as desired. Position fix colors even change when the automatic dead reckoning position mode is activated. This paper details the design and applications of SEATRAC as a high-tech, low-cost alternative in providing integrated navigation and data management. 1648 THE MEASUREMENT OF PRECIPITATION AT NATIONAL DATA BUOY CENTER STATION Eduardo D. Michelena National Data 13uoy Center NSTL, MS An operational capability to measure precipitation at shore stations and deployed buoys is being developed at the National Data Buoy Center (N--DBC). Measurement accuracy and sensor reliability are of prime importance in this program. The design and field evaluation of several precipitation sensors is described. Included are sensors specially developed and built for. N.-DBC, as well as some coftercially available instruments. NDBC's experience with tipping-bucket rain gauges, including laboratory calibration and use on shore stations and ocean platforms, is presented. The installation on a large discus buoy of an optical precipitation-rate sensor and a new design of rain-accumulation-type gauge is explained. Inter-comparison data from both land and at-sea test installations of several precipitation gauges are evaluated. 1649 MINI-DRIFTER TEST DEPLOYMENT DATA GULF OF MEXICO, SPRING 1988 Ralph R. Miller and Raymond Canada Computer Sciences Corporation Moorestown, NJ A dozen mini-drifter meteorological buoys were tested in field conditions in the Gulf of Mexico,in the spring of this year. This paper presents a summary of the data in terms of data quality, sensor stability, and sensor accuracy. The buoy data are compared with NWS data. Conclusions are presented reflecting our assessment of the state of development of the mini-drifter meteorological buoy. 1650 A SURVEY OF RADIONAVIGATION SYSTEM USERS Adeste F. E. Fuentes U. S. Coast Guard Washington, DC In these times where cast-effectiveness is being emphasized in all government programs, the Coast Guard is reviewing the status of federally-operated radio- navigation (RA) Systems to Support a recommendation to the Department of Trans- portation on the optimal mix of R.A systems. To conduct a thorough. review, the Coast Guard believes that direct input from the users of these systems (marine and terrestrial (land/offshore]) must play'-an integral part. However, there is a lack of available valid baseline data on current RA system users. This neces- sitated the sponsorship by the Coast Guard Headquarters Radionavigation Division of a comprehensive nationwide RA system user survey to accomplish the following: a) receive feedback on each system's effectiveness; b) determine each system's extent and method of use; c) determine the impact on the user of each system's discontinuance; d) determine the impact upon these systems of changing user needs and technology advancements; and e) provide an estimate of and categorize the marine user population. This paper will provide a description of the survey (including antecedent data collection efforts - estimation of the respondent universe, sampling methodology, copy of each questionnaire type, etc.) , analysis of the results, and publication plans. 1651 EXPERIMENTAL MANIPULATIONS OF DRAINAGE IN A GEORGIA SALTMARSH: LESSONS LEARNED Alice G. Chalmers University of Georgia Marine Institute Sapelo island, GA 31327 ABSTRACT movement in combination with plant and microbial activity (8, 7). Although the heterogeneity in In two separate studies of factors productivity in the two areas has been correlated controlling primary productivity of Spartina with a number of features (e.g., water flow, free alterniflora, subsurface water movement was sulfide, interstitial salinity, available Fe2+), increased in stands of intermediate and short there has been no clear demonstration of the factor Spartina. The objective of the manipulation was to or combination of factors which control simulate conditions found in creekbank marsh, where productivity of S. alterniflora (1). productivity was 2-4 times higher than in the Soil drainage has been suggested by several experimental areas. The small-scale pilot study investigators (9, 7, 15) to be an important factor had resulted in a two-fold increase in aboveground controlling productivity of S. alterniflora. biomass in one growing season and in shifts in Patterns of heterogeneity in grass height, standing belowground chemical conditions toward those crop and productivity have also been correlated typical of creekbank marsh. In the large-scale with interstitial salinity (10), iron availability experiment, belowground chemistry changed as (12), nitrogen availability (8) and sediment redox expected, but although plant biomass increased at (7, 9), all of which are affected by drainage. all sites during the first growing season, it Here I report the results of two studies, declined steadily during the next three years. In similar in design but different in scale, designed addition to those contradictory results, to determine the effects of increased drainage on substantial site-to-site variation was observed in Spartina productivity. The results of the first, the large-scale study. These results demonstrate small-scale study have been reported in detail by the need for caution in making broad Wiegert et al. (15) and King et al. (7), and are generalizations about salt marshes and their only summarized here. response to disturbance, and the value of long-term The site for the small-scale study, conducted studies in determining the effects of disturbance. in 1980-81, was in a stand of intermediate height S. alterniflora on the southern end of Sapelo Island, GA. For the large-scale study, conducted in 1983-86, four study sites were selected in short INTRODUCTION S. alterniflora marshes, two in the Airport Marsh (AP 1 and AP 2), one of the most studied marshes on Spartina alterniflora Loisel., which dominates Sapelo Island (12), one in a relatively new marsh the intertidal marshes along much of the East and at Raccoon Bluff (RBl) on the eastern side of the Gulf Coasts of the United States, displays a island and the fourth near the north end (NE) of striking heterogeneity in growth form which the island. corresponds to a major physical differentiation in At each site three plots were marked out and habitat. The highly productive "tall Spartina" is designated as undisturbed control (UDC), disturbed invariably found growing on and immediately behind control (DC) and drained (Dr). Plot size in the the levees bordering the meandering tidal creeks pilot study was 5 m X 5 m and 10 m X 10 m in the which intersect the marsh, while "short 'Spartina" large-scale study. For the large-scale study occupies the remainder of the marsh area, usually drainage systems were installed in the drained and called the high marsh (see 13 for a detailed disturbed control plots, but the outflow pipes in description of these zones). While short Spartina the disturbed control plots were capped so that no exhibits considerable variability in productivity drainage would take place. For the small-scale and height, nowhere does it equal that of the grass study, trenches were dug in the DC plot to simulate growing in the creek bank-levee zone. Interstitial the disturbance of installing a drainage system, water moves through the levee and creekbank but were refilled without burying any pipes. The sediment relatively rapidly under the influence of design of the drainage systems is reported in the hydraulic forces created by the rising and detail in Chalmers et al. (ms. submitted to falling tides (4), while horizontal sediment water Ecological Monographs). movement in the high marsh is virtually nonexistent During late September or early October of each (11). Thus productivity is correlated with the year, aboveground biomass was clipped from 0.25 m2 amount of sediment water movement. quadrats in each plot at each site. The culms were There are a number of differences in the washed, stripped of dead leaves, sorted into 25 cm chemistry of the sediments in these two zones which height classes and dried to constant weight at result from the difference in interstitial water 60*C. Dead material was also washed and dried. ICH2585-8/88/0000- 1652 si @1988 IEEE Belowground chemistry was sampled once during _tlease of protons (3). At each site the increased the small-scale study and at 3-4 month intervals acidity was greatest at depths of 10 - 18 cm, near throughout the later study. Acrylic dialysis' the bottom of the root zone. Although we do not samplers with Gelman Versapor-200 acrylic copolymer know how deep air penetrated into the sediment, we membranes (pore size 0.2pm) were used to obtain do know that the depth of the water table at low interstitial water samples (5). One sampler was tide was deeper than that at which the maximum placed in each plot and allowed to equilibrate for decrease in pH occurred, and might assume that some two to three weeks. Salinity, pH, redox potential air penetrated to the depth of maximum acidity. and the concentration of sulfide were measured at 4 Apparently the peak of increased acidity was due to cm depth intervals. Details of the chemical oxidation of reduced sulfur compounds which were analyses are reported in Chalmers et al. (ms. accumulated at that depth. Sulfate/chloride ratios submitted). and differences between pyrite concentrations in drained and undisturbed control plots also support RESULTS AND DISCUSSION this hypothesis (Chalmers, in prep.). At all four sites, sulfide concentrations were In the pilot project experimentally increased lower in the drained plots than in the undisturbed drainage caused, in one growing season, a two-fold marsh, although the differences at Raccoon Bluff increase in end-of-the-year biomass in a stand of were small. In the undisturbed plots intermediate height S. alterniflora (15). Average concentrations were low at the surface and culm height also increased. Sulfide concentrations increased with depth. were lower and soluble iron concentrations were Aboveground biomass increased in the Dr plots higher in the experimental plot than in the control at each site during the first year of increased area (7). Interstitial salinity was also lower in drainage, although the increases were small at all the drained plot than in the undisturbed control sites but AP 1. During the remainder of the study, area. All of these results were consistent with a however, aboveground biomass in the Dr plots shift of conditions toward those found in creekbank declined at every site, while that in the UDC plots marsh. remained relatively constant. There was little In the large-scale study, interstitial change in aboveground biomass in the DC plots salinity was lower in the drained and DC plots than during the first year of the study except at AP 1, in the undisturbed controls. The average reduction where it increased slightly. At AP I it continued in salinity over the upper 24 cm (root/rhizome to increase during the next two years, but zone) ranged from 1.7 ppt at the North End to 7 ppt decreased substantially at the AP 2 and NE sites. at the Airport 2 site. The salinity decrease in At three of the four study sites, aboveground the large-scale study was not as large as expected biomass in the Dr and DC plots was lower at the end based on the results of the pilot project. The of the study than it had been before the drainage reduction in salinity caused by increased systems were installed. Mean plant height infiltration of tidal water may have been increased at the two Airport sites as a result of counteracted in the later study by increased the increased drainage, but changed only slightly evapotranspiration in the higher elevation short at the other two sites. Spartina marshes. Drainage was clearly increased by the Increased movement of water through the experimental manipulation. In both studies, we sediment raised the redox potential at each site. could easily see the differences in the marsh At Raccoon Bluff the experimental manipulation had surface caused by drainage; in the undisturbed the least effect; redox potential was higher in the marsh, water covered the surface at low tide as -a undisturbed control there than at any other site thin film and as pools filling small depressions, and nearly the same as in the Dr and draining DC but not in the drained plots. Water ran out of the plots at the other three sites. The depth profiles outflow pipes at a slow but steady rate throughout of redox show steep gradients in the upper 8 - 20 the low-tide period of exposure of the ends of the cm of the undisturbed marsh at each of the sites. pipes. In fact, during some summer neap tide In contrast, redox increased just below the surface periods, the surface sediment at the NE site at in each of the Dr plots. times became so dry that cracks and fissures There was also a marked increase in the developed and it took on a powdery appearance. acidity of the interstitial water at the two Similar but less severe drying was occasionally Airport Marsh sites. The effect was less observed at Raccoon Bluff, but not at either of the pronounced at Raccoon Bluff and the North End, but Airport sites or at the pilot project site. Such was nonetheless evident in the depth profiles. The extremes of dryness could undoubtedly cause a greatest reduction in pH occurred between 10 and 18 degree of water stress for Spartina that would cm deep in the sediment at each site, which counterbalance or outweigh the beneficial effects corresponds to the zone with the largest biomass of of lower salinity and more oxidized root zone * roots and rhizomes. Values as low as 2.9 were Although the frequency of inundation was the same measured in the drained plots at the two Airport at all sites, duration varied from site to site sites; data from January and July 1986 show that pH because of differences in elevation. Thus, there in the original drained plots is returning to were small but possibly significant differences levels similar to those in the undisturbed plots. from site to site, even in the undisturbed control The marked decrease in pH observed in the plots, in the length of time the sediment surface drained plots during the later study indicates that was wet or covered with water and in natural rates increased oxidation was an important drainage of drainage. effect. The lower pH's are consistent with Although there were initial increases in oxidation of reduced sulfur compounds and the aboveground biomass in the Dr plots at each site in 1653 the large-scale study, the decline in productivity alterniflora. But it is also clear that similar during succeeding years demonstrates that drainage perturbations in apparently similar sites can alone does not determine the productivity of S. produce quite different effects. There is to date alterniflora. Clearly, a number of factors no evidence that prolonged increased drainage has interactively control plant growth in the salt been detrimental at either the AP I or the Raccoon marsh, but the pilot project which preceded this Bluff site. At Airport I productivity in the study suggested that the amount of drainage drained plots continues to be higher than in the determined how favorable for growth the interaction undisturbed control, due perhaps to lower natural of controlling factors would be. Simply increasing drainage and more frequent and/or prolonged tidal the drainage doubled the aboveground biomass in one inundation. growing season. Why then were the results of the The ability of S. alterniflora to alter its later study so different? own belowground environment (6), especially insofar The first experimental drainage was done in as it controls movements of the water table and the intermediate height S. alterniflora, a type with potential for air entry into the sediment (2) could higher initial rates of drainage than those of be of great importance, but the age of the marsh short-form grass. Our selection of four short-form and its sediment type or the source of its Spartina sites for the present study was made sediments also are important in determining deliberately in the expectation that what was good belowground chemistry. Because of the large site- for partially-drained marsh would be even better to-site differences that can occur even in a small for stagnant (interstitially) marsh. In geographical area, broad generalizations about retrospect, this was clearly a naive expectation. typical high marsh or typical low marsh sediment Adjacent to the laboratory building at Sapelo chemistry should be approached with caution. The Island is a large area of marsh which was, many aboveground appearance of a S. alterniflora marsh, years ago, drained in the expectation of using it in particular the height of its grass, is not as cultivatable land. When resultant soil necessarily an indication of a single specific conditions prevented its use, it was returned to aspect of the belowground chemistry. its original intertidal condition. The outcome of Drainage alone does not control primary this "experiment" should have warned or enlightened productivity in S. alterniflora marshes and can, in US. fact, be detrimental in some circumstances. When During the course of this study we found that drainage rates are high, water stress, which is there are substantial differences within the generally considered to be a function of salinity marshes adjacent to Sapelo Island in the natural in salt marshes, can become a more serious problem rates of drainage. Of the four sites used for the than interstitial salinity alone would indicate it present study, the site with the largest amount of to be. On and near creek banks, where drainage is natural drainage was Raccoon Bluff, where there was naturally high, tidal period and hydrological almost no plant response to increased drainage. dynamics of the marsh combine to prevent the The Airport 1 site had the smallest amount of excessively dry conditions that increase water natural drainage, yet was the only site where stress in plants. In those circumstances it is aboveground biomass in the drained plot was higher sediment water movement, or flushing of at the end of the study than prior to installation interstitial water, rather than drainage per se of the drainage systems. Had the study been that creates conditions so favorable for S. limited to these two sites we might have concluded alterniflora growth. Similarly, at AP 1, where that some degree of increased drainage could be elevation was low and sediment drying was not so beneficial to growth in stands of short Spartina. great as at other sites, the increase in sediment The Airport 2 site most closely resembled the water movement, rather than the increase in site of the earlier project visually, yet drainage in the drained plots, was the cause of aboveground biomass in the drained plot at AP 2 increased productivity. increased only slightly (12%) during the first The dramatic difference in the conclusions growing season, and decreased by 75% over the that would have been drawn about the effects of course of the experiment. At the North End site, increased drainage if the large-scale experiment which had more natural drainage than either of the had ended at the end of the first post-drainage two Airport sites, biomass declined by more than growing season (1984) and those drawn after three 80% during the study. The sustained increase in growing seasons demonstrates the need for long-term drainage was clearly detrimental to plant growth at studies, particularly when large-scale disturbances these sites. are part of the experimental design. The site-to- There was less change in sediment chemistry at site differences in response to increased drainage Raccoon Bluff than at the other sites, but water demonstrate the dangers of generalizing about marsh stress should have been most severe there, even response to disturbance or alteration. Although under natural conditions. Its relatively high all S. alterniflora marshes may be similar in elevation results in high natural drainage and less appearance, there can be substantial differences, frequent and less prolonged tidal inundation than even within a small geographical area, in controls at either Airport site, and its sandy sediment on important ecosystem-level processes and holds less water than those with high clay content susceptibility to disturbance. found at the other three sites. The lack of response to increased drainage, either positive or ACKNOWLEDGEMENTS negative, suggests that the plants at Raccoon Bluff were already adapted to severe water stress. This research was supported by NSF Grant No. It is clear that prolonged increased drainage BSR-8304928 to Alice G. Chalmers, William J. Wiebe can have an adverse effect on growth of S. and Richard G. Wiegert. This is Publication No. 1654 621 from the University of Georgia Marine Institute. REFERENCES 1. Chalmers, A. G. Soil dynamics and the productivity of Spartina alterniflora. In: Estuarine Comparisons, V. S. Kennedy, (ed.), Academic Press. 1982. pp. 231-242. 2. Dacey, J. W. H., and B. L. Howes. Water uptake by roots controls water table movement and sediment oxidation in short Spartina marsh. Science. 1984. 24: 487-489. 3. Giblin, A. E., and R. W. Howarth. Porewater evidence for a dynamic sedimentary iron cycle in salt marshes. Limnology and Oceanography. 1984. 29: 47-63. 4. Harvey, J. W., P. F. Harvey, and W. E. Odum. Geomorphological control of subsurface hydrology in the creekbank zone of tidal marshes. Estuarine, Coastal and Shelf Science. 1987. 25: 677-691. 5. Hesslein, R. H. In situ sampler for close- interval pore water studies. Limnol. Oceanogr. 1976. 21: 912 - 914. 6. Howes, B. L., R. W. Howarth, J. M. Teal, and I. Valiela. Oxidation- reduction potentials in a salt marsh: Spatial patterns and interactions with primary production. Limnol. Oceanogr. 1981. 26: 350-360. 7. King, G. M., M. J. Klug, R. G. Wiegert, and A. G. Chalmers. Relation of soil water movement and sulfide concentration to Spartina alterniflora production in a Georgia salt marsh. Science 1982. 218: 61-63. 8. Mendelssohn, I. A. Nitrogen metabolism in the height forms of Spartina alterniflora in North Carolina. Ecology. 1979. 60: 574- 584. 9. Mendelssohn, I. A., K. L. McKee, and W. H. Patrick, Jr. Oxygen deficiency in Spartina alterniflora roots: metabolic adaptation to anoxia. Science. 1981. 214: 439-441. 10. Nestler, J. Interstitial salinity as a cause of ecophenic variation in Spartina alterniflora. Estuarine and Coastal Marine Science. 1977. 5: 707-714. 11. Nestler, J. A preliminary study of the sediment hydrology of a Georgia salt marsh using rhodamine WT as a tracer. Southeastern Geol. 1977. 18: 265-271. 12. Nixon, S. W., and C. A. Oviatt. Analysis of local variation in the standing crop of Spartina alterniflora. Botanica Marina. 1973. 26:103-109. 13. Pomeroy, L. R., W. M. Darley, E. L. Dunn, J. L. Gallagher, E. B. Haines, and D. M. Whitney. Primary production. In: The Ecology of a Salt Marsh, L. R. Pomeroy and R. G. Wiegert, (eds.), Springer-Verlag, New York. 1981. pp. 39-67. 14. Pomeroy, L. R., and R. G. Wiegert. The Ecology of a Salt Marsh. Springer-Verlag, New York. 1981. 271 pages. 15. Wiegert, R. G., A. G. Chalmers, and P. F. Randerson. Productivity gradients in salt marshes: the response of Spartina alterniflora to experimentally manipulated soil water movement. Oikos. 1983. 41: 1-6. 1655 BUTYLTIN RELEASES TO HARBOR WATER FROM SHIP PAINTING IN A DRY DOCK by C.M. Adema, W.M. Thomas, Jr., and S.R. Mangum David Taylor Research Center Ship Materials Engineering Department Annapolis, MD 21402-5067 ABSTRACT they are used predominantly by the Navy. Therefore, the Navy can control and monitor TBT use at these sites. The The results of a study to determine the release of presence of unknown and uncontrolled TBT mass loadings butyltins into harbor waters from shipyard painting of the from non-Navy sources would make it difficult to accurately underwater hull of USS Badger (FF 1071) are presented. The relate TBT concentrations resulting from Navy sources. An study was part of a Congressionally mandated program to implementation plan for the entire Pearl Harbor phase of the conduct a two-harbor case study to ascertain the fate and two-harbor case study describes the joint DTNSRDC/NOSC effects of organotin in marine waters resulting from the effort.2 application and use of organotin antifouling paints. Base line surveys of organotin concentrations in Pearl Implementation of the program involves development of a Harbor were conducted in 1984.3 The surveys show low to mathematical model of Pearl Harbor, documenting the undetectable levels of butyltin compounds present in water shipyard activities involved in the application of the paint, (5 ng/L TBT) and low to,undetectable levels of extractable tin monitoring the concentration of organotin in the harbor in sediment (<50 ng/g) and oyster tissue samples (<400 ng/g). waters following undocking of the ship, and studying the Conversely, Honolulu Harbor, with heavy commercial and environmental fate and toxicity of organotin in Pearl Harbor private use, showed significant impact from tributyltin inputs waters. averaging 97 ng/L TBT in water and elevated levels of The shipyard took extraordinary steps to contain and extractable tin in tissues and sediment. collect the paint overspray in the drydock. We found that This paper presents the results from sampling of drydock more than 99% of the paint overspray was collected from the and harbor waters during and after the painting and undock- drydock floor before the ship was undocked. The remaining ing of the first ship painted at Pearl Harbor Naval Shipyard, overspray and the initial release of tributyltin from the freshly USS Badger (FF 1071). painted hull resulted in a release of approximately 15 grams There are two pathways by which butyltin compounds of tributyltin from the drydock. The release did not cause a associated wth antifouling paints can enter the estuarine measurable increase in the concentration of tributyltin in environment. The first pathway, the release of butyltin adjacent harbor waters over the background concentration of compounds from the underwater hull of the vessel, is most 15 to 20 ng/L. The release of butyltins into the marine significant when the vessel is docked or anchored and environment from shipyard activities can be reduced to represents a continuous source of release. The second, the negligible levels, but practical means to achieve that reduction release of butyltin compounds associated with the painting within the constraints of shipyard operating procedures are and cleaning of underwater hulls, represents an intermittent required. discrete release at the point of boatyard or shipyard activities. In the past 3 years, a great deal of effort has been INTRODUCTION devoted to describing the first pathway. Concentrations of butyltin compounds in estuaries and, in particular, in pleasure The U.S. Navy has been authorized by Congress to craft marinas have been determined. The rate at which conduct a two-harbor case study involving the application of butyltin compounds are released from antifouling coatings has organotin antifouling paint to Navy ships.1 This study will been measured, and attempts have been made to relate the provide data for the Environmental Protection Agency's concentrations in the marinas to the rate at which the butyltin (EPA) Special Review and for evaluating the effects of the compounds are released from the coatings. However, no Navy's use of tributyltin (TBT) paints. Organotin-based known attempts have been made to describe the second antifouling paints have been applied to only 13 active Navy pathway, either from small boatyards Or from large shipyards. ships for testing purposes. Three ships were painted at Pearl This paper describes the first attempt to measure the release Harbor Naval Shipyard during Fiscal Year 1987 under the of butyltins into adjoining harbor waters during the painting two-harbor case study. and undocking of a ship. Pearl Habor and Mayport, Florida, the other harbor in this study, were selected as the sites for the study because CH2585-8/88/oooo- 1656 $1 @1988 IEEE ................................. ............ ....... ...... .......................................................... ...............................*............... PEARL HARBOR, Hl::*'*""'*"*"''* .......... .................... .. ......... ............... ............. ... ....... ..... .. ................ . .................... ...... . ........ ............... I........................................ ................. ... ...................... ... . ........ Waipahu ...... Alea ....................... .. ....... ...... .......................... .............. ...... ........... ........... .. ................. ....... ...... . ....... ............ Middle ............ ..... East Loch Loch .... ..... ........ ........... ........ WAI 10 15 .......... West j;EM@@ULA Loch ch .......... ...... ..... 9 ...... . SA 9B ............ ....... ......... .................. 7 7 6 ............ HAR PEARI .......... ............. ... ... VAL B ASE ......... .......... ............. ....... RYDOCK 2 ............ ...... ... ........... ....... .... ............. ............. .......... ..... ...... ............ .......... ........ ............. .I ................... ...... ........... ............... ...... ............ .... ...... ...... ...... ........ ........... ........... .......... . ... . . ....... .... BASE NAUTICAL HICKAM AIR FORCE ............. ...... .. .. ......... .... .... ......... ...@'::::X 0 ............... .............. ... .................................. ..... ............. .............. ............... ..... ..... ...... ......... ... ........ .......... ... .............. 0 2 ............. ........... .......................... ............. KILOMETERS .................................. ...... .................... ...................... ...................................... ...... .... .. ...... .......................... ..... ........ Ewa Beach ...................... ........... ............ ....... .MWEIZ PACIFIC OCEAN Fig. 1. Map of Pearl Harbor. DE ATERING 92 3 --O@IRAINAGE SUMPS OPENINGS INTERMEDIATE 0 5HARBOR CAISSON DRAINAGE 4 OPENINGS Fig. 2. Drydock water sample locations; schematic of drydock drainage system. 16,57 METHODS AND MATERIALS Drydock Cleanup Site Description Three coats of paint were applied to the ship over a 3-day period; then protective materials containing the over- The ship was painted at the Pearl Harbor Naval Shipyard spray in the drydock were collected. First, the masking of the in one of its four graving drydocks* (see Figure 1). This bilge and keel dock blocks and the staging was removed and drydock has an intermediate caisson which divides the then, the Herculite on the drydock floor was cut into 4- to drydock into two separate sections which are independent of 5-ft (1.2- to 1.5-m) strips, rolled, and placed in 55-gallon each other. The section of the drydock used was drums for disposal. Next, the drainage trenches and tunnels approximately 500 ft long by 140 ft wide (153 x 43 m). The were cleaned manually using narrow shovels. Finally, the drydock floor has 4-in.-wide by 2-in.-deep (10.2- x 5.1-cm) entire drydock floor was vacuum-swept using industrial-sized lateral drainage trenches on 6-ft (1.8-m) centers, which are manually operated vacuum sweepers. designed to transport precipitation and infiltration water to The drydock preparations and cleanup described above, one of the longitudinal drains located on each side of the although successful, were costly; other engineering approaches drydock (see Figure 2). There are eight openings to each of must be evaluated to minimize the release of paint overspray the two longitudinal drains. The longitudinal drains discharge into the harbor when the ship is undocked. into one of two sumps which empty into a common pump well from which the drydock water is pumped into the Drydock Sampling harbor. USS Badger (FF 1071) with approximately 22,500 ft2 Sampling in and about the drydock was designed to (2090 m2) of underwater surface area was painted. It is 438 ft provide a complete picture of the release of butyltins during long (134 m) and has a 46.8-ft beam (14 rn). The ship was painting and undocking of the ship. We took samples to approximately centered in the drydock with the stern 20 ft quantify the amount of antifouling paint which fell to the (6 m) from the intermediate caisson and the bow 40 ft (12 m) drydock floor during the painting process, the amount that from the seaward caisson. The ship rested on concrete dock was discharged to the sanitary sewer system during painting blocks afthe center line and bilge keels about 10 ft (3 m) and cleanup, and the amount released to the harbor whilethe above the drydock floor. The water line was 15 ft (5 m) ship was being undocked. below the top of the drydock wall. Paint Overspray Dry4ock *Preparations ABC-2, an organotin antifouling paint manufactured by The shipyard took unusual and substantial efforts to DeVoe-Reynolds Corp., was applied to the ship using airless contain and collect the overspray in the drydock in order to spray equipment in three coats of 5-mils (5 mils = 0.005 in. minimize the amount of overspray that could enter the = 0.13 mm) dry film thickness, for a total antifouling coating harbor. Four of the openings to each of the longitudinal of 15 mils. A 4-mil-thick polyethylene sheet was wrapped drains were covered completely during painting. The other around 8- x 8- x 1/8-in. (20- x 20- X 0.3-cm) aluminum four were fitted with 1/4-in. mesh steel screens lined with panels to measure the overspray from the painting; 22 of cheese cloth to filter particulates from the water. The sump these panels were placed on the drydock floor and six on the which collected the water from the longitudinal drains was top of the drydock wall. The panels were placed on the isolated from the pump well, and two diaphragm pumps were drydock floor prior to the start of painting and picked up at used to pump the water from the sump to the sanitary sewer. the completion of painting each of the 3 days to prevent The other sump was covered and used to discharge ship's redistribution of the overspray by wind or rain. After the first cooling water through the normal pump well. The portion of and second coats of paint, a circular piece of the black the drydock floor directly under the ship and extending out polyethylene with a surface area of 1.25 CM2 was excised from 20 ft (6 m) toward the drydock walls (representing six of the panels in a transect across the dock. A 1.25-CM2 approximately 50% of the drydock floor) was covered with circular piece was excised from each of the panels after the Herculitej a phenolformaldehyde reinforced plastic cloth. It last coat of paint was applied. is very durable and will not tear if mechanical equipment runs over it during painting operations. The bilge and keel blocks Drydock Discharge During Painting and Cleanup were covered completely also. The staging, caissons, painting baskets, and ship brows were masked so that shipyard Infiltration or leakage into the drydock and precipitation personnel would not inadvertently contact an organotin are pumped continuously from the drydock. We sampled this painted surface.# discharge before the painting began to establish the back- ground level of butyltins. This discharge was diverted to the sanitary sewer while the ship was being painted and while the dryclock was being cleaned. Daily samples of the water A graving dryclock is a large hole in the shoreline into which the ship is discharged to the sewer were collected from the time that floated. The harbor end of the drydock is sealed with a caisson and then the water is pumped from the drydock so that repairs can be made to the painting began until the ship was undocked. These samples underwater hull of the ship. Alternatively, repairs can be made to the ship were collected in I-L polycarbonate containers by submerging in a floating drydock which lifts the ship out of the water. the container in the sump water. A daily discharge of f Herculite is a trade name of Herculite Products, Inc., 1107 Broadway, approximately 60,000 gallons (227 m3) was estimated for the New York, NY 10010. # To our knowledge, no one has established whether these personnel safety 10-day period. precautions were necessary, i.e., whether skin contact with a dry organotin coating is harmful. 1658 Drydock Release During Ship Undocking Triplicate water samples were collected 1/2 m below the surface and I to 2 m above the bottom, using a 1.5-L Teflon The drydock is flooded with water from the adjacent Kemmerer sampler, and were transferred to I-L polycarbonate harbor to facilitate undocking of the ship. The caisson is bottles. Water quality measurements were taken at each removed when the drydock has been filled with water, and the station prior to water sample collection. Parameters we ship is floated out of the drydock. Then, the caisson is measured included salinity, pH, temperature, and dissolved replaced, and the drydock is pumped out in preparation for oxygen to determine the extent of stratification. Additional the next ship. We collected samples from the drydock at the samples were collected at several stations to be filtered and locations shown in Figure 2 after the drydock was flooded were used to evaluate the partitioning of TBT to suspended before the caisson was removed, and again after the ship was particulates in the water column. Split samples were taken at removed and the caisson was replaced. Approximately 23 hr approximately 10016 of the stations for quality control elapsed between the two samplings in this particular instance, purposes. All samples were chilled in ice chests during because high winds prevented the caisson from being replaced sampling and placed in a freezer within 4 hr of collection. immediately after the ship was undocked. The samples were collected at three depths (0.5, 6, and 12 m) using a 1.5-L ANALYTICAL METHODS Kemmerer sampler with Teflon* wetted surfaces. Triplicate samples were taken at each depth at Location 2 to estimate Water sample variability. A second set of triplicate samples was taken at each depth at Location 2 and filtered within 4 hr. All samples were collected in 500-mL or I-L poly- through nominal 0.45-,um glass fiber filters using a poly- carbonate bottles and frozen at - 20*C within 4 hr of carbonate filtering apparatus to minimize butyltin losses to collection. The samples remained frozen until they were surfaces. The apparatus was rinsed with methanol between thawed for analysis. Seawater samples preserved in this samples. The filter and filtrate were retained for butyltin manner have been shown to be stable up to 3 months.4 The analysis to determine if the butyltins were associated with samples were thawed and analyzed by gas chromatographic particulates or were dissolved. A third set of triplicate samples separation followed by flame photometric detection (GC- was drawn at each depth at Location 2 to analyze for total FPD).5 Sample preparation consisted of taking a 200-mL suspended solids. A composite sample of the harbor water aliquot of seawater, spiking it with a dipropyltin dichloride adjacent to the drydock was collected before the drydock was internal standard, adding dichloromethane and sodium flooded to determine the background concentration. All tetraborate III, and shaking the sample for 10 minutes. The butyltin water samples and filters were placed in a freezer dichloromethane layer is removed, another 4 mL of within 4 hr after sample collection or filtration. dichloromethane is added, and the extraction is repeated. This The drydock water was sampled again at the pump well sample preparation converts the organotin compounds to the while the drydock was being pumped out, which occurred more volatile and hydrophobic organotin hydrides and isolates over a I 1/2-hr period immediately following the second them from the seawater matrix. The combined dichloro- sampling of the flooded drydock. Samples were collected at methane layers are evaporated to less than 100 1AL, of which 15-minute intervals. Triplicate samples were taken at the 15- 5 ;AL are injected into a gas chromatograph equipped with a and 75-minute sampling points. Simultaneously, samples also tin-sensitive flame photometric detector. The chromatograph were taken and filtered for suspended solids and butyltins, as separates the organotins and quantifies them by peak area described above. comparisons with standards prepared in the same manner. A detection limit for tributyltin of 2 ng/L can be achieved. Harbor Sampling Filters Harbor water was sampled three times; twice before and once after the ship was undocked. The former was to provide Nominal 0.45-iAm glass fiber filters6 were used to filter background organotin concentrations which may be attributed samples for butyltin analysis. We analyzed six dibutyltin and to ships and pleasure craft located in Pearl Harbor before the tributyltin standards, three at 100 ng/L and three at ship was painted. Sampling stations were selected to 1000 ng/L, filtered through these filters, and found that less correspond to stations used for Navy base line organotin than 5016 of either butyltin species is adsorbed onto the filter. surveys.3 Six sampling stations concentrated in the main and Filters were frozen at -20'C within 4 hr of collection. south channels of Pearl Harbor were selected for the survey. First, the thawed filters were Soxhlet-extracted for 6 hr with We collected the background samples at high slack tide the methanol. Then, we added 10 mL of the methanol to 200 mL day before the scheduled undocking of the ship. Mechanical of water and analyzed the solution in the same manner as the problems with the ship delayed the scheduled undocking, so water samples. The detection limit for tributyltin was 5 ng. we took a second set of background samples at low slack water 5 days later. Four tidal cycles (2 days) after undocking, Overspray we took samples at the six stations to describe the plume of TBT that resulted from the undocking of the ship. Sampling Pain overspray samples were prepared for analysis by was delayed I day because the drydock was not pumped out dissolving the paint with dichloromethane. Toluene was added until the day after undocking of the ship. and the dichloromethane was allowed to evaporate. The samples were analyzed for total tin by graphite furnace atomic absorption spectrophotometry. Three of the samples were dissolved in dichloromethane and speciated to determine the Teflon is a trade name of E.I. DuPont de Nemours & Co., 1007 Mar ket types of organotin in the paint. The results are. shown in St., Wilmington, DE 19898. Table 1. 1659 Table 1. Distribution of butyltin in paint overspray. Paint Overspray Sample No. Monobutyltin/Tributyltin Dibutyitin/Tributyltin 8 0.17 0.26 9 U.Ub U.28 11 0.06 0.14 Average 0.10 L u.23 Suspended Solids relatively windy conditions that existed at the time of painting, more than 9007o of the overspray was collected on Samples for suspended solids were taken in 1-L poly- dropcloths placed under the ship. carbonate bottles. The samples were analyzed within 8 hr of collection by filtering the entire contents of the I-L poly- carbonate sample bottle through a 0.45-;Am glass fiber filter in Drydock Discharge During Painting and Cleanup accordance with "Standard Methods.-6 Table 2 shows the c,; -gentration of butyltin in the Quality Control drydock discharge. The tL:4,@ii mass of TBT discharged to the sanitary sewer was approximately 20.5 g. An average daily Several quality control measures were taken to ensure drydock discharge of 60,000 gallons (227 M3) was diluted in that the results obtained were accurate. Chromatagraphically the average daily sewage treatment plant flow of 5.64 million pure quality control standards prepared by the National gallons (21,300 M3) so that the average TBT concentration Bureau of Standards were analyzed daily to ensure that the entering the sewage treatment plant was approximately GC-FPD was operating within � 25 076 of the historical 80 ng/L. The results clearly show the effects of cleaning the tributyltin value for these samples. Twenty samples were split drydock in that the concentration of butyltins in the drydock and analyzed by personnel from the National Bureau of discharge were reduced by approximately an order of Standards (NBS) and this Center. There was no significant magnitude after cleanup was complete. The results also difference in the results. Split samples that contained more indicate that an insignificant amount of overspray is washed than 20 ng/L were within 6507o of each other with a mean from the drydock by transient rain showers; rain occurred at difference of 22076. Split samples that contained less than four different times during the painting and cleanup of the 20 ng/L were within 5 ng/L with two exceptions. drydock, and the discharge concentration showed no In addition, we conducted replicate analyses on 26 significant difference on those days. samples. Replicate analyses that contained more than 20 ng/L Drydock Release During Ship Undocking were, within 26016 of each other with a mean difference of 1507o. Replicate analyses that contained less than 20 ng/L were within 7 ng/L. The flooded drydock samples show the concentration of Nineteen samples were spiked with 50 ng/L of tributyltin. butyltins resulting from paint overspray remaining after the The recovery ranged from 66076 to 15007o, with an average cleanup and from the initial release of butyltins from the recovery of 102.507o. Similarly, when 13 samples were spiked underwater hull of the ship. The organotin release rate of the with 50 ng of dibutyltin, the recovery ranged from 6507o to Devoe-Reynolds ABC-2 paint was measured using the method 13807o, with an average recovery of 90.607o. specified by the U.S. Environmental Protection Agency (EPA).8 The results are shown in Table 3. RESULTS AND DISCUSSION Weused the sum of the initial release (0 to 10 minutes) and the 10-minute to 3-hr release rate to calculate the mass of Paint Overspray TBT released from the underwater hull prior to sampling and estimated that 3.1 g of TBT would be released, resulting in a Approximately 68 kg of TBT were applied to the ship. concentration of 34 ng/L in the 90,000 M3 of water in the Figure 3 shows the distribution of solvent-extractable drydock. Table 4 shows the initial concentration of butyltins organotin on the drydock floor. Analysis of the paint in the drydock prior to ship undocking. overspray showed a dibutyltin to tributyltin ratio of 0.23. The The average TBT concentration in the drydock was total mass of organotin on the drydock floor after painting of 180 ng/L. Of the 180 ng/L, 17 ng/L was contributed as the ship was approximately 5.4 kg or 8016 of the TBT applied background by the harbor water and approximately 34 ng/L to the underwater hull. Interestingly, this is within the 5076 to was contributed by the initial release from the ship hull. 9% range reported by Aderna and Schatzberg.7 Therefore, approximately 130 ng/L or 11.7 g of TBT The concentration of TBT at the harbor end of the remained from the paint overspray, which represents a drydock indicates,that only very small amounts of TBT could cleanup of 99.8%. have entered the adjacent harbor waters as a result of paint Three samples out of 21 contained a particle of paint. overspray due to the predominant wind direction, as shown in This was evident in the elevated concentration and in the Figure 4. These results also indicate that even with the chromatogram for the sample (Figure 4). Those samples 1660 2.2 1:1 79 11 MONOBLITYLTIN TRIHYDRIDE 11 INTERNAL STANDARD 200. DIBLITYLTINDIHYDRIDE 12 TRIBLITYLTIN HYDRIDE 22 3.0 PANEL DAMAGED PEAK ASSOCIATED 0 WITH PAINT PARTICLE 0.54 Fig. 4, Typical chromatogram of sample containing paint 6.8 51 particle. 33 1.9 had a DBT/TBT ratio that was closer to the 0.23 ratio found in the paint overspray samples. The high concentration of DEIT in the samples which did not contain a paint particle Cl 0 may be because DBT is being released from the coating more 24 190 rapidly than the TEIT, which causes a high DBT/TBT ratio. Table 5 shows a comparison of filtered and unfiltered C3 0 samples with the associated suspended solids concentrations 3.0 5.8 from Drydock Station 2. 6.5 5.0 The filtered samples indicate that approximately 1601o of 19 37 the dibutyltin and 4701o of the tributyltin is associated with . suspended solids even when the sample does not contain paint 0 13 particles. The sum of filter and filtrate is about equal to the 0.34 250 V 0.20 results of whole water samples. This indicates that the 0.86 94 0.57 analytical method allows recovery of the butyltins from the 2.5 220 1.1 suspended solids. Cl Drydock water samples collected after the caisson was 38 replaced show approximately 24076 of the TBT that was present in the drydock before the caisson was removed 22 hr before, and only 23 ng/L higher than the harbor background 0 0 of 17 ng/L shown in Table 4; see Table 6. 7.8 120 Note that it is unusual for the caisson to be left out of 130 the drydock for this long. Typically, the caisson is replaced 460 NOTE: WHEN immediately after the ship is undocked, but in this case high ID 0 THREE winds prevented its replacement in the drydock until the 21 40 NUMBERS following day. 'ARE SHOWN, THEY REPRE- Concentrations of TBT in the drydock averaged 40 ng/L 100 SENT THE after the caisson was replaced. The average DBT/TBT ratio OVERSPRAY was 0.6 compared to 2.2 in samples taken before the caisson AFTER THE was removed. Filtered samples indicate that 76016 of the TBT 1ST, 2ND, rs 20' 3RO COATS and 83% of the DEIT is associated with particulates; see OF PAINT. Table 7. However, the sum of filter and filtrate is 5 to 6 times higher than the results from whole water samples. This 134.4 3RD COAT ONLY indicates that the butyltins are irreversibly attached to the particulate within I day and the higher absorption of dibutyltin would account for the lower DBT/TBT ratio of the Fig. 3. Distribution of dichloromethane extractable tin samples shown in Table 6. (Mg TBT/CM2). The results of drydock water samples collected at the pump while the water was being pumped out are shown in which contain paint particles were not included in the Table 8. The higher concentrations of butyltins at the calculation of average concentrations because the paint beginning and end of the pumpout appear to correlate with particle disproportionately affects the bulk water the higher suspended solids concentrations found at these concentration. The environmental effects of TBT are times during the pumpout. However, we could not verify this dependent upon its bioavailability; i.e., it must be dissolved in correlation statistically. As with the second set of flooded @ 0 200 8 69 the water column. The samples containing paint particles also drydock samples, the combined TBT from the filter and 1661 Table 2. Butyltin concentration in drydock fluids discharged to the sewage treatment plant (ng/L). Drydock - Butyltin Go centration [email protected] Day Operation Dibutyltin Tributyltin Conditions 1 Prior to painting 17, 14 15, 3U 3 4 6 Rain 4 Painting 11,000 16,000 11 to 12 mph 4,200 750 average winds from the 5 2,700 6,000 northeast 3,000 8,500 6 Cleanup 2,200 13,000 Rain 5,600; 5,000 17,000; 6,000 7 1,500 8,000 Rain 8 5,600 12,500 9 2,600 24,000 10 1,600 3,100 11 4,200 6,200 Rain 12 2,500 2,450 13 2,600 b,000 14 After cleanup 1,400 200 15 8UU 1,25U 16 4,200 170 17 1 Drydock flooded Table 3. Release rate of organotin by EPA Method. Time After Paint was First Exposed to Water Release Rate (Ug TBT/cm2/day) During which Measurement was Made Mean* Standard Deviation* 0 to 10 min 2.83 0.34 10 min to 3 hr 1.01 0.22 4 to 6 hr 0.83 0.18 24 to 30 hr .0.78 0.06 2 days** 1.11 0.05 3 days 1.39 0.18 7 days 1.75 0.08 15 days 2.79 0.05 24 days 1.80 0.05 28 days 1.66 0.07 31 days 2.36 0.09 56 days 1.96 0.04 77 days 1.89 0.09 93 days 2.53 0.03 128 days 1.71 0.16 *Mean and standard dk-jiation are based on the three cylinders painted with the same paint as required by the EPA Method. **The measurement is taken over a 6-hr period on the days shown. 1662 Table 4. Butyltin concentration of drydock water before the cassion was removed. Butyltin Dibutyltin/ Depth Concentration (ng/L) Tributyltin Location (m) Dibutyltin Tributyltin Ratio Harbor 12.0 48 17 2.8 Drydock Station 1 0.5 1500 2500** 0.6 6.0 220 120 1.8 12.0 140 84 1.7 Drydock Station 2* 0.5 470 140 3.4 6.0 280 160 1.8 12.0 310 160 1.9 Drydock Station 3 0.5 1000 400 2.5 6.0 730 320 2.3 12.0 290 220 1.3 Drydock Station 4 0.5 120 63 1.9 6.0 390 150 12.0 350 140 2.5 Drydock Average 0.5 530 200 2.7 6.0 410 190 2.2 270 150 L.8 Drydock Average, 390 180 2.2 Ail Depths I I *Drydock Station 2 samples were taken in triplicate. "Sample contained paint particle and was excluded from average. Table 5. Butyltin concentrations of filtered and unfiltered samples with associated suspended solids concentrations at Drydock Station 2 before the caisson was removed. Butyltin Concentration (ng/L) Suspended Filtered Solids Depth Water Filter I Total Unfiltered Concentration (m) DBT TBT DBT TBT DBT TBT DBT TBT (mg/L) 0.5 600 3100* 58 47 170 65 200 38 42 110 690 2600** 550 210 50 60 550 210 Average 375 124 50 72 425 196 360 . 138 24 0.6 365 53 35 190 280 54 200 60 36 110 290 340 170 41 46 37 265 88 Average 245 51 39 112 284 163 278 161 22 12.0 29 30 47 42 120 100 155 170 10 21 300 85 140 56 90 80 500 295 Average 108 85 49 1 48 157 133 307 160 22 Average, 243 87 46 F7 7 289TI64 31-51 153 All Depth4 Sample contained paint particle and was excluded from average. 1663 Table 6. Butyltin concentrations of the drydock water after the caisson was replaced. Butyltin Dibutyltin/ Depth Con-centra ion (ng/L) Tributyltin Location (m) Dibutyltin Tributyltin Ratio Drydock Station 1 0.5 20 3u 0.7 6.0 12 19 0.6 12.0 8 28 0.3 Drydock Station 2* 0.5 19 24 0.8 6.0 10 24 0.4 12.0 13 20 0.7 Drydock Station 3 0.5 13 5 - 6.0 40 38 1.1 12.0 9 32 0.3 Drydock Station 4 0.5 80 85 0.8 6.0 50 50 1.0 12.0 24 120 0.2 Drydock Average 0.5 33 36 0.9 6.0 28 33 0.8 12.0 14 50 0.3 Drydock Average, 25 40 0.6 Ail Depths *Urydock Station 2 samples were taken in triplicate. Table 7. Butyltin concentrations of filtered and unfiltered samples with associated suspended solids concentrations at Drydock Station 2 after the caisson was replaced. Butyltin Concentration (ng/L) Suspended Filtered Solids Depth Water Filter Total Unfiltered Concentration (m) DBT TBT DBT TBT DBT TBT DBT TBT (mg/L) 0.5 28 20* 350 200 18 26 5 8 730 3100* 10 20 10 16 16 35 29 25 Average 14 15 183 117 197 -132 19 24 18 0.0 32 50 97 95 12 35 15 19 39 99 9 18 21 20 47 82 LO 20 Average 23 30 61 92 84 122 10 24 17 12.0 770 300* 130 82 20 35 45 70 14 7 11 21 10 25 110 160 8 4 Averag )b 48 85 1 83 1 113 1 131 13 -)0 14 Average t22 31 109 97 131 128 14 23 Ail Dep@h *Sample contained paint particle and was excluded from average. 1664 Table 8. Butyltin concentration in drydock water during pumpout. Time from Butyltin Concentration (ng/L) Suspended Start of Filtered Solids pumpout Water Filter Total Unfiltered Concentration (min) DBT TBT DBT TBT DBT TBT DBT TBT (mg/L) 15 70 120 110 140 1000 800 17 140 300 160 520 82 225 29 34 335 62 too 13 116 27 Average 81 251 110 250 191 501 48 170 24 30 70 90 11 5 52 38 21 45 21 35 NO ND* 31 29 19 60 38 45 7080 1270* 57 39 17 75 140 120 1760 4600* bo 50 57 400 148 160 110 950 500U* 48 384 153 470 310 540 5000* 31 Average 308 140 315 210 623 350 - - 45 85 570 1 280 1 720 1 1400 1500 510 54 Table 9. Butyltin concentration prior to and following the undocking of the ship (ng/L) standard deviation). 5 Days Before I Day B fore 2 Days After Station DBT TBT DBT TBT DBT TBT 5 Surface 14 15 3 8 13 13 Bottom 4 5 12 8 5 4 7 Surface 18 18 19 22 29 22 Bottom 7 6* 2 4 16 28 73 Surface 2 2 5 12 14 14* Bottom 6 4 5 21 22 17* 9A Surface 7 5 - - 8 5 Bottom 12 13 6 1 18 34* 9B Surface 4 6 9 38 14 11 Bottom 9 3 4 12 6 6 15 Surface 7 6 2 2 42 24 Bottom 11 10 3 24 23 20 Average Surface 11+8 9+9 8+6 16+13 20+23 15+11 Bottom 8+6 7+8 5+3 12+8 11+14 17+20 Average A-11 Stations 9+7 7+7 6+5 14+Ll 17+19 16+16 *One sample apparently contained a paint particle. filtrate was about four times greater than the amount of TBT Harbor Water in whole water samples. A comparison of the amount of butyltin present in the flooded drydock to that present during Samples of harbor water were taken twice before the ship pumpout is not possible because the pumping rate varied and was undocked and once 2 days after the ship was undocked. was not recorded. However, the observed lower water All harbor water samples were taken in triplicate. The results concentration after replacement of the caisson apparently is are shown in Table 9. due to two factors: mixing with the water and settling of Harbor water samples from Stations 5, 7, and 713 were suspended solids to the drydock floor. filtered. The results for samples collected 5 days before ship 1665 undocking and I day after ship undocking are shown in between the filtered and whole water analyses. The Table 10. Table I I compares the two sets of samples shown in discrepancy appears related to the butyltin collected on the Table 10. filters associated with the suspended solids in the water The stations where samples were collected represent the column. A partition coefficient of approximately 70,000 was portion of Pearl Harbor adjacent to the naval station and calculated using the average water, filter, and suspended shipyard. Although the results shown in Table 9 exhibit solids concentrations. This is significantly higher than the considerable variability, the average concentration of TBT in 3,000 to 4,000 reported by Valkirs9 and Slesinger.10 that portion of the harbor after the undocking of the ship is Additional work would be required to identify the cause of not statistically different than before the undocking. The this difference, but it may be due to a difference in the nature results shown in Table I I reflect the same phenomenon seen of the suspended solids in Pearl Harbor or possibly a in the flooded drydock samples taken after the caisson was difference in the recovery efficiency of the analytical replaced (Table 7); specifically, there is some discrepancy techniques. Table 10. Butyltin concentrations in filtered and unfiltered samples and corresponding suspended solids concentrations in harbor water samples. Butyltin Concentration (ng L) Suspended Filtered Solids Water Filter Total Unfiltered Concentration Station DBT TBT DBT I TBT DBT I TBT I DBT I TIJT (mg/L) Before Undocking of the Ship 5 Surface 7 25 8 110 15 135 14 15 9 Bottom 6 42 17 92 23 134 4 5 25 7 Surface 10 10 260 2100* 5 10 6 9 150 84 33 33 14 10 11 28 17 11 Average 10 10 140 56 150 66 18 18 15 Bottom 4 4 11 25 6 7 8 12 11 28 8 4 3 2 - - 110 210 Average 5 6 11 27 Ib 33 42 74 17 7B Surface 2 10 4200 76UO* 2 2 20 Bottom 2 5 4 11 6 4 15 After Undocking of the Ship 5 Surface 6 4 84 60 90 64 46 13 19 Bottom 10 10 17 42 27 52 5 4 20 7 Surface 8 10 110 26 118 36 32 24 15 25 10 13 25 38 42 33 16 35 20 47 36 82 12 10 Average 13 23 47 29 60 52 29 22 20 Bottom 33 37 24 22 57 59 6 65 4 12 140 14 144 26 10 lu 12 12 51 14 63 26 16 8 Average 16 20 72 17 88 37 11 28 21 7B Surface 12 28 14 40 34 12 56 16 Bottom 1 120 1 200 1 51 1 150 1 171 1 350 1 18 1 17*1 15 *Sample contained a paint particle and was excluded from the average. L 1666 Table 11. Comparison of filtered samples collected before and after undocking of the ship. Ratio of Total Tributyltin in Average % on Filter Filtered Sample Versus Dibutyltin Tributyltin Whole Water Sample Before Undocking 72 79 5 - 6* After Undocking 69 56 2** *Excluding bottom at Station 7. **Excluding bottom at Station 7B. CONCLUSIONS REFERENCES The release of butyltins into the marine environment 1. House of Representatives Appropriation Committee from shipyard drydock activities can be controlled and Conference Report 99-1005 (15 Oct 1986). reduced to a very low level, as low as 15 g during the 2. Seligman, P.F., J. Grovhoug, and C.M. Adema, undocking of a ship. This release resulted in an unmeasurable "Implementation Plan for Pearl Harbor Phase of Navy increase of tributyltin in the vicinity of the shipyard 2 days Organotin Two-Harbor Case Study," NOSC/DTNSRDC (Dee after undocking of the ship. Therefore, if extraordinary 1986). precautions are taken, the contribution to the marine 3. Grovhoug, J.G., et al, "Baseline Measurements of environment from shipyard activities can be minimized, and Butyltin in U.S. Harbors and Estuaries," Proceedings of the the environmental effect of organotin resulting from these Organotin Symposium of the Oceans '86 Conference, activities can be negligible. The techniques used in this case to Washington, DC (23-25 Sep 1986). minimize the release of TBT were those that were ava i ilable; 4. Valkirs, A.0., P.F. Seligman, G.D. Olson, F.E. they are not practical for routine use. Additional work is Brinkman, C.L. Matthias, and J.M. Bellama, "Di- and required to identify the practical means by which the release Tributyltin Species in Marine and Estuarine Waters. Inter- of butyltins can be minimized within the constraints of laboratory Comparison of Two Ultratrace Analytical Methods shipyard operating procedures. Employing Hydride Generation and Atomic Absorption or Flame Photometric Detection," Analyst, Vol. 112, pp. 17-21 ACKNOWLEDGMENTS (1987). 5. Matthias, C.L., G.D. Olson, F.E. Brinkman, and We gratefully acknowledge the assistance of Mr. Gordon J.M. Bellama, "A Comprehensive Method for Determination Smith for his assistance in the collection of samples. We also of Aquatic Butyltin and Butylmethyltin Species at Ultratrace appreciate the cooperation and assistance provided by the Levels Using Simultaneous Hydridization/Extraction with Gas Pearl Harbor Naval Shipyard during the sampling program. Chromatography-Flame Photometric Detection," Environ, Sci. Technol., Vol. 20, No. 6, pp. 609-615 (1986). LIST OF ABBREVIATIONS 6. American Public Health Association, American Water Works Association, and Water Pollution Control 0C Degrees Celsius Federation, Standard Methods for the Examination of Water cm Centimeters and Wastewater 209D, 15th Ed. (1980). CM2 Square centimeters 7. Aderna, C.M., and P. Schatzberg, "Organotin DBT Dibutyltin Antifouling Paints and the Environment - Drydock Phase," EPA Environmental Protection Agency Naval Engineers Journal (May 1984). ft3 Cubic feet 8. U.S. Environmental Protection Agency, "Data Call- m Meters In Notice for Data on Tributyltins Used in Paint M3 Cubic meters Antifoulants," Washington, DC (29 July 1986). GC-FPD Gas chromatographic separation followed by 9. Valkirs, A.0., P.F. Seligman, and R.F. Lee, flame photometric detection "Butyltin Partitioning in Marine Waters and Sediments," mm Millimeters Proceedings of the Organotin Symposium of the Oceans '86 9 Grams Conference, Washington, DC (23-25 Sep 1986). kg Kilograms 10. Slesinger, A.E., "The Distribution of Organotins in L Liters the Workplace and the Environment," 23rd Annual Marine P9 Micrograms and Offshore Coatings Conference, Baltimore, MD (11-13 IAL Microliters May 1983). Pm Micrometers mg/L Milligrams per liter mL Milliliters min Minutes ng/L Nanograms per liter NBS National Bureau of Standards NOSC Naval Ocean Systems Center TBT Tributyltin 1667 ACCUMULATION AND FATE OF TRIBUTYLTIN SPECIES IN MICROBIAL BIOFILMS W. R. Blair, G. J. Olson, T. K. Trout, K. L. Jewett and F. E. Brinckman Polymers Division National Bureau of Standards Gaithersburg, Maryland 20899 ABSTRACT microbial biofilms on TBT-painted panels affected the release rate of the TBT into surrounding Microbial biofilms, composed of microorganisms solutions. Laughlin and co-workers (7) found originally obtained from estuarine harbor-exposed that microbial biofilms growing on TBT painted commercial organotin-painted panels, accumulated panels accumulated TBT. They suggested the tributyltin (TBT) spiked into estuarine water. microbial biofilm may contribute to paint Algal or bacterial-dominated biofilms were grown performance by reducing organotin loss, in effect by varying nutrient conditions. Both types of acting as a secondary controlled release medium, microbial communities accumulated TBT in excess as well as having protective functions toward of 300 ng/mg biomass (dry weight basis) from non-target marine life. The extent of estuarine water solutions containing 50 pg/L TBT, bioaccumulation of TBT in microbial biofilms and corresponding to bioconcentration factors of over its possible transformations remain uncertain, 7000. No degradation of TBT to dibutyltin (DBT) because-it is difficult to measure and speciate species was detected, either in the biofilm organotins within the biofilm matrix. The material or in the surrounding solution. Thus, problem is further complicated by the difficulty the microbial biofilms concentrated but did not of removing biofilm from painted surfaces without degrade TBT. These results suggest that also removing some of the paint coating below. microbial films on TBT-painted structures in marine environments concentrate TBT at materials Our earlier studies showed that estuarine surfaces, thereby acting as "capacitors". Such heterotrophic bacteria can rapidly accumulate TBT effects might make practical the use of paints from solution without TBT molecular with extremely low TBT release rates delivering transformation (8). The purpose of the current antifouling service with reduced environmental research was to determine the extent of TBT hazard. accumulation and transformation in microbial 1. INTRODUCTION biofilms. Methodology developed for speciation of TBT in waters and sediments (9,10) was applied Organotin-based antifouling coatings are in to the analysis of butyltins in the biofilm world-wide use to prevent attachment of marine matrix. We grew biofilms on glass slides and life to ship hulls, saving time, fuel, and exposed them to TBT in solution. While this does maintenance costs (1). The U.S. Navy has an not precisely mimic the leaching of TBT from a interest in painting its fleet with advanced painted surface, it does offer a reliable system controlled release (CR) polymer coatings which in which we could be certain that none of the TBT provide antifouling protection for six years or observed in the biofilm had been physically longer (2). However, tributyltin (TBT), the most removed from paint. We assumed that films widely used organotin compound for CR antifouling capable of accumulating TBT leached from paint activity, is extremely toxic to some non-target would also be able to accumulate TBT from marine life (3,4). A number of countries surrounding free solutions, if grown on a surface including France and the United Kingdom, and more not painted with TBT. In addition, we recently individual states and the Environmental investigated using organotin-fluorescent ligand Protection Agency in the United States, have combinations and epifluorescence microscopy to imposed or are considering restrictions on the directly detect organotins on biofilms (11,12). use of these materials on recreational vessels We did not find a fluorescent probe for TBT but (5,6). monobutyltin in a microbial biofilm was detected using epifluorescence microscopy. The actual mechanism by which TBT repels fouling organisms is unclear. Microbial biofilms rapidly 2. EXPERIMENTAL develop on TBT-painted hull surfaces, but higher forms of life (i.e. barnacles) are effectively 2.1 Biofilm Growth repelled (7). The role of the microbial biofilm in concentrating or affecting the release rate All biofilm experiments employed a mixed culture and form of the TBT toxicant has received very of estuarine microorganisms originally obtained little attention. Loeb and colleagues from the surface of a steel panel painted with a (unpublished data) found that the presence of 1668 United States Government work not protected by copyright commercial ablative copper and TBT-containing BIOFILM GROWTH TANK antifouling paint (International Hysol BFA-254 copolymer). The panel had been immersed in the ,-Plexiglass Cover Severn River at Annapolis, MD for 6.5 months at an average salinity of approximately 8-10 parts per thousand (ppt) (J. Mihm, David Taylor Research Center, Annapolis, personal communication). The panel displayed some green growth on the surface, suggesting the presence of Glass Slide t- Pyrex Tank algae. Upon removal from the river, the panel was gently rinsed with 0.5 L of Chesapeake Bay ------------ water (salinity 10 ppt), made sterile by autoclaving at 1210C for 15 min. The panel was then gently scraped with a sterile teflon spatula over a 10 x 10 cm area, and the scraped material Stirring Bar was washed into a sterile polycarbonate bottle with 100 mL of sterile Chesapeake Bay water. The Magnetic Stirrer wash water was immediately transported to the laboratory, where 30 ml, of the suspension was inoculated into a conical flask (250 mQ Figure 1. Apparatus used to grow microbial containing 170 mL of sterilized Bay water having biofilms on glass slides. 10 mM ammonium chloride and 2 mM dibasic potassium phosphate as nutrients. This flask was placed under fluorescent lamps (intensity 350-400 candles), depending on the experiment. foot candles) at 220C. Another 30 mL of wash water was inoculated into a flask containing 170 2.2 Tributyltin Uptake on Biofilms ml. of water containing 0.03% yeast extract and 0.03% polypeptone (Difco). These samples were After several weeks, glass slides containing incubated a-t 281C on a gyratory shaker at 200 biofilms were removed from the growth tank and r.p.m. (New Brunswick model G76). After one week immersed in smaller growth cylinders of pyrex of incubation, the contents of both flasks were glass containing sterile Bay water amended with mixed and centrifuged (6000xg, 20 min). The 50 Mg/L TBT (as the cation). Additionally, two supernatant was poured off and the pellet groups of control slides were included, which containing cells was resuspended in 70 mL of consisted of clean glass slides placed into TBT sterile Bay water. Sterile glycerol (7.7 ml) was solutions and biofilm slides placed into sterile added, the cell suspension was continuously Bay water minus the TBT spike. Small teflon stir stirred to mix the contents, and 1.0 mL aliquots bars were added to the cylinders and the were removed and added to each of 60 sterile 1.5 solutions were held in the dark or under mL plastic screw-capped vials designed for fluorescent lamps at room temperature. After cryogenic sample preservation. The vials were incubation, the slides were dipped twice into Bay transported to the American Type Culture water free of TBT and scraped with a clean razor Collection (ATCC, Rockville, MD), where they were blade. A small volume of water was used to rinse frozen and stored under liquid nitrogen. The the scraped material into 15-mL glass centrifuge cell suspensions contained 2.0 x 108 cells/mL tubes. Scrapings from the duplicate or triplicate (Petroff-Hauser counting chamber) and 4.3 x 10' biofilm slides incubated with TBT were pooled in colony forming units/mL when spread on to one centrifuge tube. Aliquots were removed to peptone-yeast extract (0.03% concentration of separate tubes for spiking with butyltin each) agar and incubated at 28*C for one week. compounds (method of additions quantitation) and for dry weight determinations. Microbial biofilms were grown on glass microscope slides (2.5 x 7.5 cm) immersed vertically in a 2.3 Butyltin Extraction and Analysis small glass tank (Figure 1) containing 500 ml, of sterile Chesapeake Bay water amended with peptone The procedure is a modification of a method for and yeast extract (0.05% of each), or ammonium extraction of TBT from sediment (10). All chloride and dibasic potassium phosphate (10 and concentrations of organotins are listed on a 2 mM, respectively) . The Bay water had been weight basis in terms of the cations. collected in a 50 L polyethylene carboy from 1.0 Approximately 200 pl, of the pooled biofilm m depth at the mouth of the Severn River. The suspension was added to a 150-mL round bottom sample was passed through a 35 pm filter during flask along with an internal standard consisting collection. It was stored in the dark at room of 50 ul, of tripropyltin chloride (1.5 pg/uL in temperature in the laboratory, and aliquots were methanol). Concentrated HC1 (0.5 mQ followed by removed and autoclaved as needed for biofilm methanol (25 mQ was added and the mixture was experiments. The inoculum was prepared from refluxed for 30 min at 80-85'C. The sample was frozen cells in storage at ATCC. The cells were cooled to room temperature and was extracted quickly thawed in warm water (301C) and added to twice with 5-ml, aliquots of cyclohexane in a 125- 1 le the biofilm growth tank. A sterile teflon stir ml, separatory funnel with shaking (10 min) on a bar was added to the tank and the tank was placed wrist action shaker (Burrell). The combined on a magnetic stirrer in the dark or under cyclohexane layers were evaporated to about 2 mL fluorescent lamps (intensity 350-400 foot under a gentle stream of air and then shaken for 1669 45 min with 1.0 mL of 4% (w/v) aqueous NaBH4. A TABLE 1 small aliquot of the cyclohexane layer was injected into a gas chromatograph equipped with a Bioaccumulation of TBT by Microbial Biofilms flame photometric detector operated in a tin- selective mode as described previously for detection of DBT and TBT (9,10). Incubation TBT Soln. TBT Accum. Bioconc. Cone. Cone. Factor 2.4 Epifluorescence Imaging (dry wt) Spectrophotometric detection of monobutyltin Dark, 97 hr 50 jug/L 353 pg/g 7060 adsorbed to the biofilm was successfully demonstrated by epifluorescence microscopy (11). Monobutyltin, which has a more intense Light, 168 hr 50 Ag/L 373 pg/g 7460 fluorescence emission than tributyltin when complexed with flavonol (11), was chosen for this experiment to demonstrate that the concept of visual imaging of adsorbed organotins on the material. Control slides (clean slides exposed to biofilm was practical. Glass slides covered with TBT, or biofilm slides minus TBT) gave no a well developed biofilm growth were exposed to a butyltin peaks in the chromatograms. solution of monobutyltin in ethanol (4.5 x 10-4 M concentration) for approximately 1 hour. The Another experiment was run using biofilms that slides were then rinsed with ethanol and exposed had been grown under fluorescent lamps. These to an ethanol solution of flavonol (1.4 x 10-4 M biofilms contained a large proportion of algae, concentration) for 15 min. Control slides, as evidenced by their green color and large either exposed to monobutyltin but not to numbers of red autofluorescing cells visible flavonol, or exposed to flavonol but not to under epifluorescence microscopy using 365 rim monobutyltin, were also prepared. excitation (owing to chlorophyll in the cells). These films were incubated under fluorescent Following the above treatment, the slides were lamps for 7 days and analyzed for TBT and DBT as examined by epifluorescence microscopy. The UV excitation frequency was set at 366 nm with a 28 nm bandpass. A barrier filter was used to BU3Sn eliminate any radiation at wavelengths over 400 nm. Pr3Sn ? The following compounds were tested for their abilities to chelate to butyltin species and generate luminescent emission: 3-hydroxyflavone, BU4Sn morin, apigenin, galangin, emodin, flavone, Biofilm chrysin, 4',5,7-trihydroxyflavone. All of these 97 hours compounds were purchased from a commercial source (Aldrich Chemical Co., Milwaukee, WI). 3. RESULTS AND DISCUSSION Preliminary experiments showed that microbial biofilms accumulated 331-784 ng TBT/mg (dry weight) on exposure of films to Chesapeake Bay water containing 0.344 mg TBT/L for 28 hr. Subsequent experiments were performed at lower (50 Mg/L) TBT concentrations and comparably substantial quantities of TBT were also bound to + spikes biofilms in these experiments. The first DBT 345 ng experiment involved biofilms grown in the dark, BU2Sn then immersed in the spiked Bay water and held in TBT 516 ng the dark. After four days incubation the pooled biofilm sample contained 353 ng TBT/mg (dry weight; Table 1, Figure 2). This represents a bioconcentration factor of 7060. Each glass slide contained an average of 0.43 mg of biofilm material (dry weight) over an area of 25 cm2. The concentration of TBT in the film was determined by standard additions of di- and 0 5 10 min tributyltin chlorides to aliquots of the pooled biofilm sample. These were then carried through Figure 2. Chromatograms showing the the entire extraction procedure. Samples of the bioaccumulation of TBT in microbial biofilms solution of Chesapeake Bay water were analyzed after 97 hours incubation (top). Spikes of periodically and showed no biodegradation of TBT tributyltin and dibutyltin species into the to DBT, nor was any DBT detected in the biofilm biofilm matrix are shown at bottom. 1670 above. This biofilm material contained 373 ng TABLE 2 TBT/mg (dry weight; Table 1). This corresponds to a bioconcentration factor of 7460. Again, Detection by EMI of Monobutyltin Bioaccumulation there was no evidence of light- or dark-induced degradation of TBT to DBT either in the biofilm SLIDE EMISSION MAXIMUM RELATIVE INTENSITY material or in solution during the course of the experiment. No cells- 435 8.58 0.25 no tin Tetrabutyltin was detected in biofilms run in the no ligand first (dark incubated) experiment (Figure 2). This compound was detected as a contaminant in Biofilm- 435 8.54 0.27 reagent TBT used as a spike and was evidently no tin bioaccumulated. This peak was not seen in the no ligand light incubated experiments where a chromatographically purified source of TBT was Biofilm- 436 9.03 0.28 employed which did not contain tetrabutyltin. with BuSn3+ The concentration of TBT used in these no ligand bioaccumulation studies (50 Ag/L) greatly exceeds Biofilm- 458 7.85 0.19 concentrations of TBT reported in bulk water in with ligand aquatic environments. However, bioaccumulation no BuSn3+ of TBT in more dilute samples is also likely, especially since biofilms accumulated nearly as Biofilm- 452 30.39 0.87 much TBT from solution at 50 Mg/L as at 0.344 with BuSn3+ mg/L (see above). In addition, it is likely that with ligand the concentration of TBT in the interstitial water of a microbial biofilm on an organotin- flavonol had significantly greater fluorescence painted ship hull is considerably higher than in surrounding open waters. We therefore do not intensity than any of the control slides. Visual believe 50 Mg TBT/L to be an unrealistic examination of this slide under the concentration to illustrate potential microbial epifluorescence microscope revealed a dark blue- biofilm effects. black background with localized areas of bright blue fluorescence, characteristic of the The TBT bioaccumulation results are similar to monobutyltin-flavonol complex. These images our previous findings with isolated strains of suggest that the monobutyltin moiety is not estuarine bacteria (8). In that study we found homogeneously retained in the biofilm, but that the organisms accumulated TBT to 3.7 to 7.7 additional studies will be required to determine Ag/mg (dry weight; bioconcentration factors in whether the organotin is adsorbed chiefly by the hundreds). There was no measurable viable cells or by exocellular adhesive polymers. degradation of TBT to DBT. Laughlin et al. (7) found that a microbial biofilm on a TBT painted Only 3-hydroxyflavone, morin and galangin ligands panel contained 20-60 pg/mg (wet weight), form complexes which emit sufficiently for substantially greater than the concentration we consideration as "fluorogenic ligands" for tin observed in our biofilms. This is not species (data not shown). In each of these surprising, given the quite different cases, Bu3Sn+ had a fluorescence of only about experimental systems between that study and the 1/100 of that of the other tin species, perhaps present investigation. Laughlin et al. employed due to the UV degradation of TBT to more TBT painted panels immersed for four months in fluorescent species (12). It therefore seems San Francisco Bay and removed biofilm by lightly very unlikely that any of these ligands could be scraping the film from the painted surface. a sensitive, selective reagent for the detection However, both studies show that microbial of tributyltin species in mixtures of organotins. biofilms bioaccumulate TBT to a significant However, it seems reasonable to employ the EMI extent. technique to survey biofilm communities for the presence of TBT degradation compounds. In Previous studies have shown that TBT is degraded conjunction with electron microscopy with element to DBT in the environment, with half-lives on the specific energy dispersive microanalysis, order of 1-2 weeks (13,14). In some cases, questions could be answered concerning the microalgae have been implicated in the location of organotin species in the biofilm biodegradation process (14). However, the (cell mass or exocellular polymer). current experiments show that biodegradation of TBT did not occur in the microbial biofilm It is interesting to note that reproducible composed of organisms originally obtained from a results indicate that the relative intensity of panel painted with TBT-containing paint. We emissions of either morin or 3-hydroxyflavone believe that our past and current results complexes of butyltin species in EtOH follow the indicate biodegradation is not a common mechanism trend: of TBT resistance in estuarine bacteria. . BuSn3+ > Bu2Sn2+ > Sn 4+ > Bu,Sn+ Table 2 summarizes results employing EMI to This trend differs from the relative intensities detect monobutyltin in microbial biofilms. The reported for butyltin compounds in aqueous biofilm slide exposed to both monobutyltin and solution (10). The EtOH/H20 ratio (v/v) greatly 1671 affects the relative emission intensities of 2. Schatzberg, P., in Proc. Internat. Organotin these samples. There may be a solubility or Symp., Oceans 87', Sept. 29 to October 1, 1987, miscibility problem in the case of organotin- Halifax, Nova Scotia, IEEE, Piscataway, NJ, pp. flavanoid complexes in a highly aqueous 1324-1333 (1987). environment. This may hinder analyses of "in- vivo" samples which require an aqueous 3. Champ, M.A. and Lowenstein, F.L. Oceanus, environment. The best use of the EMI may be to 30, 69-77 (1987). detect tin species on a solid surface, after chromatography has been performed (i.e. paper 4. Maguire, R.J. Appl. Organometallic Chem. 1, chromatography, TLC). 475-498 (1987). 4. SUMMARY AND CONCLUSIONS 5. Champ, M.A. and Pugh, W.L. in Proc. Internat. Organotin Symp., Oceans 87', Sept. 29 to October Biofilms containing microorganisms obtained from 1, 1987, Halifax, Nova Scotia, IEEE, Piscataway, organotin painted panels accumulated relatively NJ, pp. 1296-1308 (1987). large amounts of TBT from solution. There was no detectable biodegradation of TBT to DBT in the 6. Champ, M.A. and Bell, D.F. Research Needs films or in the surrounding solutions. The films Concerning Organotin Compounds Used in may thus influence the action of organotin Antifouling Paints in Coastal Environments, coatings. Microbial biofilms may concentrate TBT Report to NOAA/National Marine Pollution Office, on the surface of painted ship hulls. In this Rockville, MD, March 1988, 133 pp. manner it is conceivable that the TBT charged microbial biofilms act as the primary repellent 7. Laughlin, R.B. Jr., Cobet, A.B. and Guard, to settling larvae of hard fouling organisms. If H.E. in Proc. Internat. Conf. Marine so, then organotin paints with extremely low Biodeterioration, in press. release rates may still provide antifouling protection by virtue of a TBT enriched surface 8. Blair, W.R., Olson, G.J., Brinckman, F.E. an@ biofilm, and may release very little TBT to Iverson, W.P. Microbial Ecol. 8, 241-251 (1982). surrounding waters. Such paints would minimize environmental impacts of the use of 9. Matthias, C.L. Bellama, J.M., Olson G.J. and TBT-containing antifouling paints. Additional Brinckman, F.E. Environ. Sci. Technol. 20, 609- experiments involving exposure of TBT enriched 615 (1986). biofilms to macrofouling; organisms are warranted. Epifluorescence microscope imaging of 10. Matthias, C.L., Olson, G.J., Brinckman, F.E. monobutyltin in microbial biofilms was and Bellama, J.M. Internat. J. Environ. Anal. demonstrated. The technique may be useful for Chem., submitted. the detection of such TBT degradation products. Direct detection of TBT in solution or in biomass 11. Blair, W.R., Parks, E.J., Olson, G.J. and will require the development of a ligand which Brinckman, F.E. J. Chromatogr. 410, 383-394 specifically chelates to TBT, yielding a complex (1987). which fluoresces with reasonable quantum efficiencies. Selected commercially available 12. Trout, T.K., Olson, G.J., Brinckman, F.E., materials were found to be unsatisfactory for Bellama, J.M. and Faltynek, R.M., submitted. these requirements. 13. Seligman, P.F., Valkirs, A.0. and Lee, R.F. 5. ACKNOWLEDGMENTS Environ. Sci. Technol. 20, 1229-1235 (1986). We thank Paul Schatzberg, David Taylor Research 14. Olson, G.J. and Brinckman, F.E. in Proc. Center (DTRC), Annapolis, MD, for his Organotin Symp., Oceans 86', September 23-25, suggestions, ideas and support in undertaking Washington, DC, vol. 4, IEEE, Piscataway, NJ, pp. this project. We are grateful to George Loeb and 1196-1201 (1986). James Mihm of DTRC for helpful discussions and for providing the initial biofilm samples. We especially thank Lieselotte Schebek and M.O. Andreae of the Max Planck Institute, Mainz, for assistance in conducting the algal biofilm experiments. Portions of this work will be included in the doctoral dissertation of TKT in partial fulfillment of the requirement of the Graduate School of the University of Maryland. 6. REFERENCES 1. Ludgate, J.W. Jr., in Proc. Internat. Organotin Symp., Oceans 871, Sept. 29 to October 1, 1987, Halifax, Nova Scotia, IEEE, Piscataway, NJ, pp. 1309-1313 (1987). 1672 UPDATE OF OCCURRENCE RATES FOR ACCIDENTAL OIL SPILLS ON THE U.S. OUTER CONTINENTAL SHELF Cheryl McMahon Anderson and Robert P. LaBelle U.S. Department of Interior, Minerals Management Service ABSTRACT addresses the likelihood of oil spills occurring The Minerals Management Service estimates the during the production and transportation of likelihood of oil spills of 1,000 barrels or offshore oil. A realistic, objective methodology greater occurring in association with the for estimating oil spill occurrence rates is needed production and transportation of offshore oil on for this application of the oil spill risk the U.S. Outer Continental Shelf (OCS). The analysis. estimation process uses a spill rate constant, based on historical accidents, expressed in terms SPILL RATE DEFINITION of number of spills per billion barrels of oil produced or transported. The mean spill occurrence This paper presents the results of analyses of oil estimate is obtained by multiplying the rate spill occurrence rates (hereafter referred to constant by the volume of oil projected to be simply as "spill rates") for U.S. OCS platforms and handled. The mean number of spills is then applied pipelines and worldwide spills from tankers. Spill to a Poisson process to further estimate the rates are expressed in terms of number of spills probability of one or more spills occurring in a per billion barrels of oil handled. only spills of given production period. The calculated occurrence 1,000 barrels and greater are addressed, since the rates of 0.60 spills per billion barrels produced OSRA is used primarily to estimate contacts over on U.S. OCS platforms and 0.67 spills per billion days, not hours, without consideration of explicit barrels transported in U.S. OCS pipelines represent spreading or weathering of oil. Consequently, only a decline of 40 and 58 percent, respectively, since those spills that are large enough to travel long last evaluated in 1983. Spill occurrence rates for distances on the ocean surface and that could worldwide tanker transport remain unchanged, since persist for several days or longer are appropriate last evaluated in 1983, at 0.90 for "at-sea" spills for simulation by this model. Another and 0.40 for "in-port" spills. consideration is that a large spill is unlikely to go unnoticed; therefore, reporting records should be more reliable. The 1,000-barrel cutoff meets INTRODUCTION the above requirements. In addition, the estimation of the size of an oil spill is generally The Minerals Management Service (MMS) of the an inexact measurement, which is often rounded up U.S. Department of the Interior (DOI) conducts oil or down by onsite observers to the nearest 500 or and gas leasing on the U.S. Outer Continental Shelf 1,000 gallons, barrels, or metric tons. Therefore, (OCS) and regulates activities on tracts that have spills near 1,000 barrels are likely to be reported been leased. The MMS normally prepares an as 1,000 barrels. Large spills account for less Environmental Impact Statement (EIS) for each than 1 percent of all spill events on the U.S. OCS; offshore lease sale. Within each EIS, MMS must almost all smaller spills (98 percent) are less evaluate the potential risks of oil spills than 10 barrels in size. These smaller spills do occurring and contacting environmentally sensitive not lend themselves to trajectory modeling. Any resources. further reference to "spills" in this paper should be assumed to be "spills of 1,000 barrels or The Oil Spill Risk Analysis (OSRA) model was greater" unless otherwise specified. developed in 1975 by DOI as a tool to assist in the evaluation of offshore oil-spill risks POISSON DISTRIBUTION FOR (Smith et al., 1982). This model is used primarily ESTIMATION OF SPILL OCCURRENCE to provide environmental impact analysts with probabilistic estimates of oil-spill occurrence and Efforts to estimate the probability of a spill contact with biologic and economic resources occurring are part of MMIS's assessment of potential located on the U.S. OCS. The trajectory portion of environmental impacts. In the past, the occurrence the model uses seasonal surface current and local of spills has been modeled as a Poisson process wind conditions to simulate many trajectories from (Smith et al., 1982). Using the updated database, potential offshore spill sites. Results of the we have reexamined the assumptions necessary to use trajectory runs are then used to estimate the this distribution and have determined that they are statistics of potential contact. The OSRA also still applicable. CH2585-8/88/0000- 1673 $1 @1988 IEEE Since spill occurrences do meet the criteria for a The spill rate, A, is expressed in the form of Poisson process, we use the following equations in equation 1c as the number of spills per billion our estimation of spill rates. The estimated barrels of oil handled. In the case of platforms volume of oil handled is the exposure variable. and pipelines, MMS addresses the U.S. OCS Smith et al. (1982), using Bayesian inference experience exclusively. The advantages of this techniques, presented a derivation of this process approach are that the rates will better reflect the assuming the probability of n spills over some magnitude of spill occurrence under U.S. regu- future exposure t is assumed to occur at random lation and operational controls and that the with some frequency specified by equation 1a: individual spill and production records are readily P[n spills over future exposure t] = (,t n _Xt accessible to MMS. The disadvantages include the ) e limitations involved in analyzing a small number of n! observations. This precludes cross-sectional analysis of possible rate variations that may exist where X is the true rate of spill occurrence per between the northern Pacific or Alaska waters unit exposure. The predicted probability takes the versus the Gulf of Mexico and Southern California, form of a negative binomial distribution: where the majority of the U.S. OCS experience has (lb) P(n) = (n+v-l)!tn-rv occurred. The rates should not be adjusted based on the intuition that more hostile environments are n!(V-1)(t+T) n+v riskier. The more hostile environments have more stringent engineering and procedural regulations, where T is-past exposure and v is the number of which may offset or even reduce the risk of a spill spills observed in the past. The negative binomial occurring relative to historic OCS experience. is then shown to converge over time to the Poisson with X estimated using equation 1c (Smith et al., 1982): RATES FOR SPILLS OF 1,000 BARRELS OR GREATER (1c) X = VIT We separately addressed three sources of oil spills: platforms, pipelines, and tankers. Only SELECTION OF EXPOSURE VARIABLE spills of the size of 1,000 barrels or greater were rigorously analyzed. Platform and pipeline spills Two basic criteria for selecting an exposure were analyzed based upon U.S. OCS experience from variable are that the exposure should be simple to 1964 through 1987. Tanker spills were based on measure, and it should be a quantity which is worldwide data from 1974 through 1985. In each predictable. Volume of oil handled is the chosen analysis, each spill event record was examined and exposure variable primarily for these reasons. verified to the extent possible. Each spill was Past production is well documented, as shown in classified based on its applicability to the figure 1. Volume of oil handled makes the analysis in terms of size, product spilled, and calculation of the spill rate simple--the ratio of spill source. the number of historic spills to volume of oil handled--in the form of equation 1c. Volume is There are only 11 and 8 spill observations, a Iso predictable. Estimates of volume are prepared respectively, for platform and pipeline spills; see by the Resource Evaluation Program within MMS, tables 1 and 2. Nonparametric tests were applied whose function and expertise is.in the assessment to determine whether these observations were random of oil resources using comprehensive geological and and independent. In both cases, the volume of oil geophysical databases and related models. In handled between spills appeared to be nonrandom and addition, almost all other exposure variables that increased over time. This indicated that the spill could be used--number of platforms, number of rate, based on volume of oil handled, had declined tanker trips, etc.--are currently estimated by MMS over time. as a function of oil handled. Table 1 Figure 1 Oil spills of 1,000 barrels or greater from U.S. OCS production of crude oil and condensate, platforms on the U.S. OCS, 1964-1987. 1964-1987. Date Area Size Cause 500 (barrels) condensate 01/08-09/64 Eugene Island 2,559 Collision, fire. 400 crude oil 10/03-19/64 Ship Shoal/Eug. lsl. 11,869 Hurricane, blowout.* 07/19-26/65 Ship Shoal 1,688-- Blowout. 0 300 01/28-02/06/69 Santa Barbara 77,000 Blowout. 03/16-19/69 Ship Shoal Collision. blowout. 2,500 X.: 02/10-03/31/70 Main Pass 30,000 Fire. X@ x MM 12/l/70-4/17/71 South Timbalier 53,000 Blowout, fire. 0 . .. ... ... ... .... ... 'i*@ ix'A ',x 01/09/73 West Delta 9,935 Storage tank rupture IGO 01/26/73 South Pelto 7,000 Storage barge leak. X. xx X 0 1/23/79 Main Pass 1,500- Collision with rig. 10 11/14/80 High Island 1,456 Tank overflow. 1'?JF._ -1 .414- .0 8�0 year Sources: MMS 88-0010, MMS 88-0011. 6 platforms. includes 1,589 BBL storage oil. (Santa Barbara estimates vary. Condensate. Sources: MMS 88-0010, 1988a and 1988b. 10,000-77,000 BBLS.) ... Diesel. 1674 Table 2 Figure 3 Oil spills of 1,000 barrels or greater Percent of production and pipeline spills in the from pipelines on the U.S. OCS, 1964-1987. record disregarding previous 1.0 BBBL intervals. percent total production percent total pipeline spills Date Area Size Cause 100 80 60 40 20 0 0 20 40 60 80 100 (barrels) too 7.5 too 10/15-27/6'e West Delta 160,638 Anchor dragging. 871 6.5 75 03/12/68 South Timbalier 6,000 Anchor dragging. 73 5.5 163 02/11-16/6E Main Pass 7,532 Cause unknown. 60 4.5 38 (0.67 spill@/BBSQ 05/12/73 West Delta 5,000 Internal corrosion. 471 3.5 13 04/18/74 Eugene Island 19,833 Anchor dragging. 33F_ 2.5 13 09/11/74 Main Pass 3,500 Hurricane. 20 [::: 1.5 0 12/18/76 Eugene Island 4,000 Trawler dragging. 7[ 0.5 0 total production record (BBBL) 12/11/81 South Pass 5,100 Anchor dragging. I I I 1 100% includes entire U.S. OCS record, 1964-87. Source: MUS 88-0011. Production include. crude oil Subsequent percents represent distribution as and condensate. the oldest L.0 BBBL intervals are disregarded. To properly examine the data, uniform intervals the percentages of the record included as the based on volume of production, rather than time, oldest data are dropped in increments of 0.5 and were selected. Interval lengths were calculated to 1.0 billion barrel production intervals for maximize the percentage of intervals with a single platforms and pipelines, respectively. The shaded spill occurrence. This minimized misleading area indicates the percentage of the record observations of zero spill intervals created by included in our final spill rate selections. For using production intervals of such short length example, figure 2 indicates that the selected 0.60 that it was highly unlikely that a spill would be platform spill rate is based on the most recent observed. Historical spill rates were calculated 67 percent of the production record, which includes cumulatively from the first available interval only 27 percent of the total number of historical forward through 1987, dropping the previous spills. This means that by dropping the record for interval's record. The'spill rate was selected by the first 2.5 billion barrels of production, indentifying the point in the database where major 73 percent of the spills occurred during this early decreases in the spill rate are no longer apparent production and are discounted. Likewise, figure 3 while trying to maximize the amount of the indicates that the 0.67 pipeline spill rate is historical record used. based upon the most recent 60 percent of the OCS production record, which contains only 38 percent Spill rates of 0.60 and 0.67 spills per billion of the historical spills. barrels were selected for U.S. OCS platforms and pipelines, respectively. These rates were selected The database of worldwide tanker spills provided because they reflect the observed decline compared substantially more observations than the U.S. OCS to the intervals based on the longest records, and pipeline and platform data. Rates for tanker they are based on the most recent 60 percent or spills of 1,000 barrels or greater were calculated more of the historical record. Figures 2 and 3 directly from the data and were relatively constant illustrate the comparison of the historical over the entire time,period from 1974 through 1985. production record against the spill record, showing Spill rates for spills occurring "at-sea" were Figure 2 calculated separately from those occurring Percent of production and platform spills in the "in-port" based on an exposure of 107.8 billion record disregarding previous 0.5 BBBL intervals. barrels transported. These spill rates were 0.9 and 0.4 per billion barrels, which were based on percent total production percent total platform spills 97 "at-sea" and 43 "in-port" spills, respectively. 100 80 60 40 20 0 0 20 40 60 80 100 lbo 1 7.5 - - 160 93 1 7.0 173 SPILL RATES FOR SPILLS OF 10,000 BARRELS .1, 1 6.5 64 OR GREATER .0 6.0 45 73r- 5.5 36 Rates for spills of 10,000 barrels or greater were 87 5.0 27 (0.60 spR1s/BB8L) 60 4.5 18 calculated by applying the percentage of historical 53 4.0 18 spills equal to or greater than 10,000 barrels 47 3.5 =18 40 3.0 =18 against the 1,000 barrel or greater spill rate. 33 2.5 =] 9 This resulted in a rate of 0.24 spills per billion 27 2.0 0 20 1.5 0 barrels of oil produced and/or handled for 13 = 1.0 0 platforms and 0.17 for pipelines. The 10,000 7@ 0.5 0 0 0.0 0 barrel and greater spill rates are 0.55 for "at-sea" tanker spills and 0.16 for "in-port" total production record (131313L) tanker spills. 100% includes entire U.S. OCS record, 1964-07. Production includes crude on Subsequent percents represent distribution as and condensate. the oldest 0.5 1111131. intervals are disregarded. 1675 CONCLUSIONS REFERENCES The updated platform and pipeline spill occurrence Lanfear, K.J., and Amstutz, D.E., 1983. A rates are significantly reduced from those of Reexamination of Occurrence Rates for Stewart (1975, 1976) and Lanfear and Amstutz Accidental Oil Spills on the U.S. Outer (1983). This may be due to improvements in Continental Shelf: Oil Spill Conference, technology and stricter safety regulation over 1983, Proceedings, 5 pp. time, combined with overall gains in experience in Smith, R.A., Slack, J.R., Wyant, T., and offshore oil production. A comparison of how the Lanfear, K.J., 1982. The Oil Spill Risk rates have changed since 1983 is shown in figure 4. Analysis Model of the U.S. Geological The primary reason for these reductions is the low Survey: U.S.G.S. Professional Paper 1227, incidence of spill occurrence since 1980. 40 pp. Stewart, R.J., 1975. Oil Spillage Associated Figure 4 with the Development of Offshore Petroleum OSRA Model oil spill rates, 1988 revision Resources. Report to Organization for compared with previous rates. Economic Cooperation and Development, 49 pp. Stewart, R.J., 1976. A Survey and Critical 5 Review of U.S. Oil Spill Data Resources, M Stewart, with Application to the Tanker/Pipeline 1975, 1976 Controversy. Report to the U.S. Interior 0 4 SLanfear & Amstutz, 3.87 Department: Cambridge, MA, Martingale, Inc., 4. 1983 75 pp. 0 @@Anderson & LaBelle, M U.S. Department of the Interior, Minerals DC@3 August 1988 Q) 3 Management Service, 1988. Federal Offshore Statistics: 1986: OCS Report No. MMS 88-0010, 2.30 95 pp. 2 1.87 U.S. Department of the Interior, Minerals 1.60 Management Service, 1988. Accidents 0 1.30 1.30 Associated with Oil & Gas Operations: ............ 1.00 OCS Report No. MMS 88-0011, 264 pp. 0.67 ... ........... 0.60 U.S. Department of the Interior, Minerals Q) ............ Management Service, 1988a. Communication ............ ........... ........... 0 ........... - ML with MMS Pacific Region for 1985-86 OCS OCS platforms OCS pipelines worldwide tankers crude oil and condensate production figures. spill source U.S. Department of the Interior, Minerals (spills of 1,000 barrels or greater) Management Service, 1988b. Communication with MMS Gulf of Mexico Region for 1985-86 crude oil and condensate production figures. 1676 MYTH AND MANAGEMENT--The Shipwreck Management Act Jack Fullmer, Legislative Committee, N.J.C.D.C. Congress recently passed the Abandoned Shipwreck Act and this has caused a great deal of concern in the sport diving community. As legislative Committee Chairman of the N.J. Council of Diving Clubs, I was asked to formulate a position for our Council and took an active Part in trying to change the wording of the bill. Eventually this effort led to testimony before the Senate Subcommittee on Public Lands, National Parks, and Forest. The changed wording of the Senate Bill eventually became 'Law. The purpose of this paper is to examine some common myths about shipwrecks and shipwreck management, to look at some of the wording in the Shipwreck Manaqement Bill, and to show how some of the positive concepts in the law can be used at the state level to build a framework that will both protect historic shipwrecks and fully utilize this multipurpose resource. The rise of sport diving in the second half of the century along with advances in marine electronics have changed the role that shipwrecks play in our society. At one time shipwrecks were looked upon only as objects of salvage. It was difficult to find them and difficult to dive on them. Modern SCUBA gear and advanced electronics have changed all that. But the myths and stereotypes that arose during that era when shipwrecks were looked at as treasure trove are still around us along with a few new myths generated by the media. Perhaps the most common myth is that shipwrecks are mainly important for salvage and, more recently, for underwater archaeology. Certainly most of the T.V. and news coverage on shipwrecks Portrays this role. Yet treasure wrecks Probably constitute less than 1% of the total resource, while historic shipwrecks are estimated to constitute about 5%. A historic shipwreck is a wreck that could meet National Register criteria. The Predominate importance of almost all shipwrecks in marine waters is for fishing and sport diving value, or what is often referred to as reef value. For example, in New Jersey 70% of all sport diving is done on shipwrecks. Party and charter fishing boats fish the wrecks extensively, for the wrecks are concentration or focal points for bottom fish, lobsters, and other marine life in an otherwise barren seafloor. Every day during the season a whole fleet of fishing and dive boats visit the wrecks off the Jersey coast. Another myth that may have instigated the Shipwreck Management Bill is the belief that salvage is the most serious threat to shipwrecks. Yet the traditional cause for the destruction of shipwrecks in our rivers, harbors, and bays is dredging and large scale construction projects. Trawlers and draggers account for much of the, damage in the oceans. It is interesting to note that this sort of damage is accidental, and if the location of the wreck were well known, the damage could be avoided. CH2585-8/88/0000. 1677 $1 @1988 IEEE Natural causes such as storms, hurricanes, worms, and corrosion eventually doom all shipwrecks in the marine environment, some at faster rates than others. One Revolutionary War wreck had laid on the bottom of the Mullica River in south Jersey in a fair state of preservation. But within the last 10 years a 30 foot deep channel has meandered into it causing the wreck to tip over into the channel at a precarious angle. Large timbers are now falling off the wreck and are being washed away by the current. All shipwrecks should be viewed as transient resources. It is important to locate those historic shipwrecks that have entered periods of accelerated deterioration so that archaeological work can be done before it is too late. Salvage as a factor in the destruc- tion of shipwrecks is rather minor. Another misconception that crops up in the Shipwreck Management Act is the belief that an embedded wreck is probably an old or historic shipwreck. The simple fact is that whereas most historic shipwrecks are embedded, so are most non-historic shipwrecks. Almost all wrecks are embedded in the normal sense of the word, as shifting sands and storms will embed a wreck in as short a period of time as a couple of years. Yet another misconception is that 'states will make good managers of shipwrecks because they have archaeologists on staff that are interested in shipwrecks. In New Jersey not one dime of Historic Preservation Funds or discretionary grant funds has ever been used to fund an underwater archaeology project. Nor is there a single person in the agency that is designated for historic preservation in N.J. that dives or has ever seen a shipwreck underwater. on the other side of the coin the N 'J. Dept. of Marine Fisheries has undertaken an ambitious program to increase the number of wrecks (artificial reefs) with little available funding, a program that will benefit both fishermen and sportdivers, and replace a transient resource. The fact is that some states have taken an active interest in shipwrecks while others have shown little or no interest. This can even vary between departments within a state whose jurisdic- tions both involve shipwrecks. Another interesting point is that states must approve dredging projects in rivers, harbors, and bays, 'and have inadvertently been responsible for much of the destruction of historic shipwrecks in this country. sometimes the location of historic research informa- tion is known to sport divers but this information is not known to state officials that issue permits for dredging or other large scale projects. A similar situation took placein 1983 in N.J. when a memorandum of agreement was reached on a proposed bridge across'the Crosswicks Creek near Bordentown; all of the state and federal officials had overlooked the fact that. an entire Revolutionary War fleet of merchant vessels and privateers (22 vessels) had been sunk by the British in the immediate area of that proposed bridge, and it took some sport divers and amateur archaeologists to bring that point home. 1678 Another prevalent assumption is that state and federal agencies find most shipwrecks. Nothing could be further from the truth. Sport divers, dive boat captains, and charter and party boat captains, who often ask sPort divers to check them out, find almost all wrecks, including historic shipwrecks. State officials are almost totally dependent on these sources, or on publications, usually by sport divers, for information about the locations of shipwrecks. Strangely enough the logical assumption that archaeologists and historians are responsible for most historic research about ship- wrecks is also correct. Most historic research on shipwrecks is done by sport divers, who also write numerous local histories about shipwrecks in books and for newspapers, and magazines. This is also true for mapping and drawing of shipwrecks, and much of what passes for underwater archaeology in this country is handled advoca- tionally by interested sport divers. The last myth that needs some comment is the idea that shipwrecks are fragile resources. This fallacy even occurs in the House report language on the Shipwreck Management Act. ShilDwrecks may be fragile in the sense that pollution could affect their ecosystem, but common sense should tell anyone that a wreck that can survive the violence of numerous hurricanes is not going to be affected by a sport diver swimming over it. MANAGEMENT The previous discussion sugge sts certain logical principles that should be involved in any shipwreck management program. The first and very obvious point i s that shipwrecks should be treated as a total resource (multipurpose) and not just as an historic.resource, which is. actually a very minor aspect of the resource. Treating shipwrecks as a multipurpose resource means that you take a close look at how they are actually utilized in our society, what purposes they serve, and who are the people who utilize them. You consider the large numbers of fishermen and sport divers that visit them every year. You consider the lobster fisherman that sets his traps near a shipwreck, but tries to avoid entangling them in the wreck. You consider the trawler captains that are trying to avoid hitting them. You also consider the very tiny number of underwater archaeologists and's.alvors that make a living from ship- wrecks. The Shipwreck Management Act does recognize some of these different uses of shipwrecks. In Sections Four and Five of the Act, it mentions habitat value, tourism, recreational exploration, under- water parks, and historic research and salvage. Strangely absent is the almost total lack of reference to recreational fishermen and commercial fishing interests, which is one of the main uses of the resource. 1679 Another thing that suggests that the Act was not intended to take a total resource Point of view is its jurisdiction. The categories of shipwrecks transferred to states (National Register vessels and those vessels embedded in submerged lands of a state so that tools of excavation are required ... ) would indicate that the law is oriented to historic and/or digging up vessels that could be historic. The impression one gets is that the law was first aimed at historic shiowrecks, and then the legislators, after hearing testimony, decided that it might be important to mention other uses as well. The concern snort divers have repeatedly expressed is that almost all shipwrecks are embedded in the normal meaning of that word. Yet states are given the mistaken impression that there is some special importance to an embedded wreck. Tools of excavation probably refers to power dredging equipment, and the act of removing substantial amounts of sediment from the bottom would also be of environmental concern. Yet that is not made clear anywhere else in the Bill. In hindsight, the Act should have given jurisdiction to all abandoned shipwrecks in state waters, or to just historic shipwrecks, and not become involved in the embedd'ed mystery. I have enclosed a letter from a Congressman that states his belief that Admiralty Courts still have jurisdiction over some wrecks in state waters. The second Principle of management, which follows logically from the first, is that state management of shipwreck resources should not be solely vested in historic preservation offices, but should reflect the interests of sport divers, fishermen, salvors and other groups that utilize the resource. The idea of manage- ment by the utilizers of the resource is not a new idea. it occurs in the N.J. Marine Fisheries Council, in the N.J. Boat Regulation Commission, and in other such regulatory agencies. It is particularly important in a resource that is only seen and utilized by a small segment of the population. For example, 99% of the people that actually dive and can see shipwrecks fall into the sport diver category. If this group were to be excluded from its manacrement, a state would have very little information and ,input about the resource and, in addition, it would have no handle on the feelings or concerns of that segment. Furthermore, the people that are most likely to understand the issues and problems relating to shipwrecks are those that are familar with shipwrecks. By involving the utilizers of the resource, a-state is less likely to:run into a situation where the managers of the resource have never seen a shipwreck, a situation that is bound to foster myth and ill conceived regulations. T@e type of management form could be a commission which would accurately reflect the number and involvement of these different groups with shipwrecks. Thus it might have a large portion of fishermen and sport divers as well as a salvor, archaeologist, and a few representatives of state agencies involved with shipwrecks. 1680 The Shipwreck Management Act suggests that sport divers, fishermen, archaeologists, salvors and other interests be involved in the management of shipwrecks at the state and federal level in Section Five, and this was a key aspect of the testimony of the N.J.C.D.C. before the S enate. Unfortunately it occurs in a section of the Act that asks the National Park Service to draw up guidelines. The N.J.C.D.C. suggested to the National Park Service that they appoint sport divers and fishermen to serve on the Committee to draw up these guidelines. This idea was refused by the Park Service, and apparently the only persons that will serve on this committee will be historic preservation professionals who will listen to testimony at hearings. The question is how sport divers and fishermen are supposed to manage shipwrecks at the federal and state level if they are not allowed to serve on the advisory committee. Where else at the Federal level could they serve in a management capacity. In Section Five, the word "partnership" is written in conjunc- tion with the idea of management. The testimony of the N.J. Council of Diving Clubs intended partnership to have a wider meaning. Partnership refers to a cooperative effort between sport divers, fishermen, archaeologists, salvors, and (inferred),state agencies to not only manage shipwrecks, but to work together to fully realize the potential of this resource. For historic shipwrecks, partner- ship means that sport divers and state officials work together to identify historic shipwrecks, especially those that may be threatened by natural or other factors. It means cooperation between sport divers and archaeologists to map and study these wrecks It means sharing information about the location of historic ship- wrecks with trawler captains so that they can avoid them. For our shipwreck reefs, partnership means sport divers working with the marine fisheries agency to monitor fish kills, to conduct fish surveys and water sampling, and to help out with other scientific endeavors where the numbers of sport divers can do what the small numbers of scientists cannot do. Partnership means fishermen and sport divers working with the State to replenish our shipwrecks through artificial reef programs. Partnership means sharing the location of shipwrecks so that they become a truly public resource, not hidden as secret loran numbers known only to a few. Partnership is not helped by the attitude of some archaeologists and ".historic preservation professionals" that tries to limit access to historic shipwrecks. This elitist approach (that only a select few with degrees should be permitted on shipwrecks) ignores the statistics that most historic shipwrecks are not found by archaeolog- ists, are not initially explored by them, are not mapped by them, and that most historic research is not done by them. Yet this elitist Philosophy, which contradicts the voted on policy of Congress, is reflected in the House Language Report. 1681 Since there are very few underwater archaeologists in this country and few states employ any, the direction that historic preservation is going to take for shipwrecks is not a profess- ional archaeological approach, but is going to stress manning, National Register nominations, and historic research by para- professionals. Several surveys have already been conducted in New Jersey in this manner, and have been highly successful. A state might consider hiring an underwater archaeologist to help lead these sport diver volunteers. Partnership, therefore, is based on guaranteed access. Not only is this the voted policy of Congress, but common sense would indicate that no snort diver would report the location of a historic shipwreck if a state had a policy to then deny access to it. Access is also important because historic shipwrecks, unlike land sites, may also be important sport diving and fishing reefs. New Jersey has many old shipwrecks that are important sport diving wrecks. Even for important archaeological sites, states have alternatives to denial of access. Report language is both Houses suggest that states hide the location of.historic shipwrecks by issuing vague descriptions of their location. This foolish policy would not only increase the risk of trawlers hitting them, but also ignores the fact that it is the diving public that normally finds them. This public would probably return the favor by hiding the location from the state, and this is not a partnership relationship. Finally states should stress education and innovative designs over fear of nenalities when trying to protect historic shipwrecks. In a paper by Alan B. Albright of the Institute of Archaeology and Anthropology, University of South Carolina called the Law and the Amatuer in Resource Management, Mr. Albright stresses that coopera- tion is far more effective than confrontation and threats of law enforcement. He cites the cooperation that his state has received from sport divers and the numerous finds that are archaeologically important that have been reported by sport divers. He mentions that enforcement is always time consuming, expensive, and tends to establish an adversary relationship. Voluntary compliance, a product of education, understanding, and compromise, is less expensive, self motivating, and establishes cooperation as the norm. Maryland recently passed a law that calls for an educational program for the training of interested members of the public (sport divers) in the identification and registration of submerged archaeological property. New Jersey is considering draft legislation that will include advocational mapping of shipwrecks. Sport Divers are already mapping historic shipwrecks in the Mullica River, and Amateur involvement is a very large part of archaeological efforts in the Crosswicks Creek. Organizations such as the Atlantic Alliance with notable archaeologists such as Duncan Mathewson are 1682 sponsoring underwater archaeological training programs for sport divers. Consideration is being given to inserting positive incentives such as an annual awards program for the best mapping and nomination of historic shipwrecks in the N.J. Shipwreck Management Bill. 'I could go on and on about what the Shipwreck Management Act did and did not do, and what principles of management offer the most promise for the full utilization of this multipurpose resource. The fact is that the national law allows for a great deal of latitude for states to shape their own management pro- grams. , The sport diving community needs to meet the challenge of overcoming the myths and stereotypes that cloud this resource. It must defeat elitist ideas that try to hide or deny access to historic shipwrecks. Since 99% of the people that actually dive on and see shipwrecks are sport divers, the sport diving community has a unique opportunity to take a lead role in the partnership that will manage this important resource. This should be a positive role that stresses sport diver involvement in scientific and archaeological programs, and a cooperative effort with fishermen, archaeologists, and state agencies to map, explore, and replenish this multipurpose resource. The N.J. Council of Diving Clubs believes that shipwrecks are a public resource and should be open to the public. We believe that shipwrecks are not the exclusive monolopy of archaeologists, salvors, or anyone else, but that openness aftd cooperative effort is what will best serve the development and preservation of the resource. State progEams should encourage innovative designs to reach this goal. 1683 THE CONTINUING NEED FOR ACCURATE POSITIONING IN NAVAL TACTICS John L. Hammer, III Wayne R. Hoyle Oubit North America, Inc. ABSTRACT of determining one's place on the earth to a Modern naval warfare has, over the years, specific tolerance. resulted in the relegation of accurate position- ing and navigation to a second class role. The While the two words mean the same thing to modem naval officer worries about positioning many people, there are subtle differences. Fun- usually only when concerned with navigation. damentally, the word 'positioning' implies a During the battle he most often conducts his more accurate knowledge of one's position than operations in a relative motion sense using rela- is required in navigation. The act of navigation tive ranges and bearings. The once widely- is more dynamic in that it includes movement respected title of navigator has been replaced by from one place to another. Positioning reflects a that of the Tactical Action Officer. Despite these greater interest in one's position as accurately lamentable trends, accurate positioning remains fixed on the surface of the earth. It is very often as important today as it ever was. The trends used in association with activities requiring close toward over-the-horizon targeting/battles and tolerances such as in surveying or the offshore strict Emission Controls (EMCON), coupled oil industry. Positioning is necessary where the with the more traditional tactics in Anti-Sub- mariner wishes to accurately mark a place for marine Warfare (ASW) and Mine Warfare, con- return or some future action. tinue to require good geographic positioning as DECLINE IN NAVIGATIONAL AND POSITIONING a matter of paramount importance. EXPERIENCE INTRODUCTION The term 'Navigator' within the U.S. Navy has Naval warfare, as with all other operations at lost some of its luster over the years because of sea, depends heavily on the positions of one's two factors. One, improved techniques and own vessel and those of other contacts in the equipment have made navigation less difficult to area. While much activity is conducted in the practice but at the same time more accurate. relative frame of reference using bearings and Secondly, the evolution of modern warfare has ranges, it is still necessary to have accurate and shifted emphasis from true positioning to a rela- current knowledge of one's position on the sur- tive frame of reference. A navigator is generally face of the earth. less interested in precision positioning because his ship's motion does not allow him time to take In this discussion on determining one's location full advantage of multiple or more complex we need to review two fundamental definitions. positioning strategies. The first is for navigation, which is the art or Less than a hundred years ago, the title science of conducting one's self from one place Navigator was one which inspired admiration. to another. The second is for positioning, the act Now navigation is something that must be done to keep out of trouble. It's still possible to ruin CH2586-8/88/0000- 1684 $1 @1988 IEEE one's reputation through bad navigation. For Navy ships to keep a hard-copy (paper) this reason, navigation is maintained as a neces- geographic track plot. sary but not altogether desirable skill. Unfor- RELATIVE AND GEOGRAPHICAL POSITIONING tunately, the imperatives of today's fast-moving tactical and battle situations have diluted the One need only look at any tactical radar screen emphasis placed on traditional navigation skills. today to get an appreciation for how operation- al tactics have developed after World War 11. Before that, they were always played out on plot- The modern naval officer must concentrate ting tables using geographic references. Virtual- more on threat indicators and less on finding his ly every navigational and tactical radar displays position. This lamentable state of affairs appears targets in the Plan Position Indicator (PPI) for- in both the lack of strong and comprehensive mat. With PPI, the radiating ship is positioned navigation training and in the lower priority in the center. Contacts are located on the screen given its actual practice. along radials, at some true or relative bearing, extending from the radiating vessel's blip. Some The evolution of modem electronic navigation radars allow radar operators to displace the systems has led to the design of relatively simple- origin of the cursor to an off-center position to to-use equipment with direct reading displays. aid in the visualization of the tactical situation. One needs only take latitude and longitude Computations may be made within the equip- readings from the equipment and plot them. ment to solve for the maneuvering board solu- This may be well and good in the open sea where tions of targets' courses and speeds thereby the hazards of running into a shoal are less un- reducing relative data to true motion. Similarly, likely. However, in restricted waters, or where bearings and ranges from one target to another an accurate position is required, these systems can be determined by moving the cursor origin are not always as accurate as manufacturers to a desired blip and rotating the cursor to a would like users to think. Potential problems third. range from what we call range holes, to system failures to poor pattern geometry. Navigation in Modem tactical operations are normally con- well-travelled waters, is generally easy because ducted within a relative motion framework be- provision has been made for complete and ac- cause most weapons systems deal in range and curate electronic navigation signal coverage. bearing data elements. When contact is made, However, in less frequently travelled waters, it the target is generally identified by its range and becomes more problematic. It is into these bearing from the detector. Ile resulting evolu- waters that our navies must sail and fight. How tion in tactical thinking, therefore, has con- many of you, ten years ago, would have thought centrated almost solely on the relative motion that there would be major actions in such out-of- solution. This is a dangerous trend because, with the-way places as the Falklands and Grenada? the flood of information now available from all In order to position oneself well, one must pay our sensors, there is a higher probability of full attention to all the variables. losing the 'big picture.' That's why, when you see a picture of the Combat Information Center There are still many operations personnel who of an Aegis ship, you see a bank of very expen- find it difficult to adjust to a digital environment sive screens. Unfortunately most ships cannot because they feel they've lost touch with the real carry such an expensive suite of equipment and world. They find it difficult at times to interpret must make do either with small and very the information presented. It's not surprising limitecd screens or with the more traditional and that we still find the requirement aboard U.S. underdeveloped plotting tables. Nevertheless, one of the best ways for keeping a (NTDS) or Link 11 or 14 Systems. These systems clear picture of what is going on is to plot the ac- pass disposition information on a number of tion on a table. This effectively results in conver- contacts automatically between computers. One sion of relative motion into a true motion picture weak link in these systems is the frequent in- on a geographic plot. The resultant plot then ability of data generating ships to position them- directly relates the action to the real world. selves accurately. Any errors in their positions are passed through the system and multiplied by The fact that such solutions are still available, any other errors present at the receiving vessel. dramatically points to a definitive need for tac- tical operations officers to get their target infor- If every ship and aircraft was properly posi- mation on a true or geographic display. tioned, the problems could be significantly Movement is then unambiguously portraye. reduced. In a recent exercise, the Link system Mariners feel at home working tactical solutions was observed at one instant to have three con- in true motion where action is depicted in real tact symbols portraying the same vessel in three world terms. It significantly reduces the com- different places on the screen. Two of these posi- plexity and allows for faster decision-making. tions were more than five miles distant from each other. The force was in radar silence and at The major advantage is that one gets a true ap- darken ship. The positions provided the only in- preciation of the interaction of all the par- formation by which the ships maintained their ticipants in an action without having to work out stations in the main body and screens. ne ship a relative motion solution in one's head. This is from which we made these observations was car- particularly advantageous when there are multi- rying a navigation integration system which ple contacts to contend with both true and rela- combined three navigation system inputs with tive data flows being received. an automated tactical plotting table. We were LARGE AREA BATTLES able to obtain the finest positioning information of any ship in the force. Since we knew the loca- Modern weapons systems have longer and tion of our ship to within a hundred meters, we longer stand-off ranges. The result is that tac- were easily able to see the failings in other ships ticians have had to consider ever extended not possessing good navigation or position infor- detection horizons. The only good method for mation. adequately referring target positions to scat- tered elements of a task force is on some sort of A common tactic during periods of EMCON grid or geographic coordinate system. This is be- silence is to send a radar aircraft aloft which cause not all vesssels can support complex, large radiates from a location far removed from its and expensive detection and tracking sensor task force. One of the greatest sources of error suites, or their sensors are limited by EMCON in the Link system arose from the fact that the restraints. Ships must rely on other sources of in- helicopter had taken its relative position from formation to learn the whereabouts of the the carrier which was in turn not well-posi- enemy. As with most detection systems, the tioned. The result was that all the contacts il- detection is made in the relative mode but con- luminated by the helo, which was the only radar verted into geographic positions for transmis- in the EMCON-shrouded force, were in as much sion to, and plotting by, other elements. error as that of its guide, the carrier. She was, in- NTDS AND THE LINK SYSTEMS terestingly, the element which. had appeared in Much of these data communications are per- triplicate (i.e., in error) on the Link screens of the force earlier in the exercise. formed through the Navy Tactical Data System 1686 When enemy targets are detected by such a mis- plotted with the positions of other participating positioned helo's radar, then the error is ships and aircraft. The plot is most critical for the propagated throughout the force. Any ships Attack Coordinator who must be able to see the needing to plot those targets would be in error entire situation and call in his forces to maxi- as would any weapons which were to be fired. mum effect. During a similar exercise transit, the Link con- Most modem plotting tables still require the trol ship role was given to one of the screening concerted efforts of several persons to keep the escorts. It was easy to see that the destroyer, plot current. The ship's position is projected by which was also the screen commander, was soon a light source from below and a person must overloaded trying to run the ASW problem mark its position manually at fixed intervals. against three submarines whilst trying to main- Similarly, all other contacts and targets must be tain the tactical data base. When the going got similarly tracked. The result is that many per- hot, it was obvious which effort was left to lan- sons must lean over the table while hand plot- guish. We observers noted an early breakdown ting. Meanwhile, the Attack Coordinator must in the navigation function as disorientation be- look over everyone's shoulders to get a clear pic- came the rule on the Link system. We believe ture of the unfolding attack. Not only is this this occured because the navigation function is method manpower-intensive, but it is prone to not considered critical when compared with the human error in the plotting..Very little develop- priorities of the battle. This might possibly be ment has taken place in this area. considered acceptable in battle there is no need for this to happen. There are systems available Few equipment manufacturers have attempted which will automatically track the geographic to automate this function. While there are a plot for low cost and minimal effort. Positioning number of systems attempting to consolidate all information was of paramount importance to this information on a geographic plot, most other ships in the force when in EMCON. equipment tends to be very expensive and com- A7TACK SCENARIOS plex. Additionally, many operators tend to feel alienated from a strictly digital solution. The Attack scenarios are most often played out on plotting table still has an intimacy that is not geographic plots because of the clarity of their found from viewing a video screen. presentations. This is particularly true in situa- MINE WARFARE tions where there is a 'time-late' or datum to deal with. The solution to such problems is best The need for good geographical positioning is found on- a geographic frame of reference be- even more critical in the area of mine warfare. cause the "lost" contact can be fixed to a position It is of paramount importance that mine warfare on the earth where it was last detected. In this vessels, aircraft, Q-Routes and objects on the way its projected movement can be easily seabed be accurately positioned. Each of the portrayed. Similarly, it is far easier to set up elements in the mine warfare equation must be search and attack patterns on a geographic grid fixed on a common reference frame on which all than it is to work in relative motion. This fact be- players may track and display objects of interest. comes even more true when there are multiple The most common grids are those developed vessels and helos involved in the action. around geographic coordinate systems, whether they be in Latitude/Longitude or Universal Traditional Anti-Submarine Warfare (ASW) at- Transverse Mercator. Without a common tack scenarios are traditionally played on a tac- reference frame, the ability to avoid or re-ac- tical plotting table. The datum or target is quire the contacts is significantly reduced. 1687 Precise positioning is of even greater impor- survey run line may be regarded as accurately tance in this critical area of naval operations. positioned. There is very little margin for error. Many of the larger equipment manufacturers are quick to Side scan sonar traces are good reconnaissance provide very large and technologically advanced tools but there. are a number of variables which suites of equipment by which they claim to solve are often not taken into account when accurate many of the mineman's problems. We've found positions are desired. A number of products on that a far better and less expensive approach is the market and in development project the to increase the precision in solving the naviga- paper side scan trace onto a video screen. This tion/positioning challenges. This is a relatively should allow the operator to rapidly classify and low cost solution with leveraged benefits sur- "position" objects detected. Many of these sys- passing those of the 'more and larger' equip- tems, however, fail to address such problems as ment syndrome. a display (waterfall) scroll rate tied to ship's Q-ROUTES speed. This is common on paper recorders but causes distortion on video, leaving doubt as to Mine warfare begins with the complete survey where the object really is. Other interesting fea- of Q-Routes, i.e., the intended routes through tures touted are 'freeze framing' and 'zooming,' restricted waters down which vessels may steam. which allow the operator to stop the scrolling to This concept limits the amount of area needing better view a contact or blow up its scale. Both to be continually searched. The objective is to these functions serve a useful purpose but condition a channel by accurately surveying it obscure the cascading display currently being and then resurveying it frequently to ensure that collected. no new mine-like objects have appeared. When the initial surveys are undertaken, all objects of Positioning objects detected by side scan sonar interest on the bottom are accurately positioned is possible, but there are a number of variables and identified. These become the reference ob- which must be addressed. The principal con- jects against which future contacts will be com- sideration is the position of the towfish at the pared. It is important that all objects are located time of each ping. Ibis is no small problem. Just in order to reduce future requirements for clas- knowing the position of the towing ves- sification of 'new' contacts. sel/aircraft is not good enough when obtaining positions for use in mine warfare. One of the Q-Routes should be regularly "conditioned"' so other papers at this conference on the Bathyscan that, should war break out, the number of new system (in the Acoustics - Side Scan Session) contacts needing to be classified would be provides an interesting and accurate approach. limited. Positioning during these conditioning MINE HUNTING AND NEUTRALJZATION surveys must be as accurate as possible to provide accurate positions for future com- Return to a previously-detected mine-like ob- parison with objects previously detected and ject for neutralization is an extension of the classified. problem discussed above. It's no good just having the objects accurately positioned and Much of this kind of survey work is done with the charted. The mine hunter must be able to aid of side scan sonars. These allow operators to navigate back to the object. Without proper sweep a swath of bottom between the survey run navigation, useless time is wasted looking for the lines. There are inherent inaccuracies in the side object to be neutralized and the risk is always scan sonar principle which must be considered present that the wrong object will be destroyed. before the positioning of objects far from the Therefore, as much attention must be paid to 1688 positioning the hunter and disposal ves- the very best solutions for positioning themsel- sels/aircraft as in the initial survey. ves and the objects they're looking for. The point being made here is that positioning is CONCLUSION the key to good mine warfare tactics from begin- We've tried to point out that the evolution of ning to end. Mine warfare operators cannot be modem naval warfare has drawn us away from content to be haphazard in their approach to the careful and assiduous application of naviga- navigation. Ile best solution is to use more than tion and positioning information. 'Me ease of one positioning system and mathematically obtaining a 'fix' and the imperatives of tactical evaluate the inputs to reduce the effors. We are threats have relegated this aspect to a lesser role. not even sure that the GPS will be the final solu- Nevertheless, the importance of accurate tion to all these problems, though it is supposed positioning, whether it be for navigation or some to give very accurate positions. Mine warfare warfare application cannot be dismissed. Only personnel cannot afford to be cavalier in their with good positions may naval forces optimize approach to positioning. their power. Recently we heard a story about some mine for- Without good positions, tracking the battle is ces in the Gulf. They had come up with the tradi- made difficult. The lack of a plot makes tional "Cocked Hat" problem where three lines decision-making difficult. Too much effort is put of position resulted not in a point but in triangle. into making large systems which do everything, Their solution was to turn off one of the three with too little attention being paid to the key ele- stations in order to reduce any ambiguity. With ment, the positioning of the force. Neither can two lines of position, they had no doubt where we let the speed of the battle detract from main- they were and they had a point fix. We have no taining our positions because they are what we knowledge what criteria they used in selecting need to deconflict our scenarios in the end. the deselected station, but it points to the fact that they really needed more than three lines of position to get a statistical evaluation of their positioning systems. There are too many lives and High Value Targets riding on their efforts. There is an ever greater need to provide for the integration of a number of positioning systems in order to come up with a statistically weighted answer. If nothing else, you will know how bad your position is. There is an old mappers' axiom that says, "The next battle will be always be fought at the junc- tion of at least two map sheet lines." The exten- sion of this rule in mine warfare is that the next mine will be seen at the outer limits of your radiolocation chain pattern. Minemen will al- ways be on the fringes; out where the coverage is weak and support is weak. That's why they need systems which can be counted upon to give 1689 ECONOMIC IMPACTS FROM COAL EXPORTS: THROUGH THE PORT OF BALTIMORE AND THE PORT OF NORFOLK Morris Va. Clark, Jr., Dennis P. Robinson, and Lloyd G. Antlel U.S. Army Engineer Institute for Water Resources Fort Belvoir, Virginia 22060-5586 ABSTRACT One should remember, however, that an impact assessment constitutes only one of the many sets of socioeconomic inputs to the political decision- Paper presented at the OCEANS '88 Conference and making process by which scarce Federal and state Exposition on 'A Partnership of Marine Interest' resources, such as the dredging of harbors, are co-sponsored by The Marine Technology Society and distributed. One aspect, for example, of deter- The Institute of Electrical and Electronics mining whether a port should be dredged at a given Engineers- -Ocean Engineering Society, in coopera- point in time is to show with geographic and tion with the Port of Baltimore, 31 October - 2 sectoral specificity how movements of commodities November 1988, Baltimore, Maryland. Paper sum- through that port contribute to regional and marizes the results of using a multiregional national economies. input-output methodology, developed by economists at the U.S. Army Engineer Institute for Water Re- This impact assessment makes use of three major sources, to conduct an economic impact analysis of assumptions: (1) economic impacts of shipping one the export shipment of coal through the Port of million tons of export coal through each port, Baltimore and the Port of Norfolk, respectively. respectively, can be analyzed to provide the Paper is based on research and other work con- impacts attributable to the first unit million ton ducted by the authors in their pursuit of the use shipment as well as successive unit million-ton of regional science and economic structural shipments; (2) economic impacts of shipping coal analysis in the planning and development of through each port can be derived by multiplying federal navigation projects. An earlier version the actual many millions of tons of export coal of this paper was delivered at the 27th Annual handled by each port in the year 1986 by the Meeting of the Southern Regional Science Associ- economic impacts of the unit million-ton shipment ation, 14-16 April 1988, Morgantown, WV, This of export coal; and (3) coal exported through paper does not necessarily reflect the official either port will come from mines in the 8-state policies and deliberations of the U.S. Army Corps coal shed defined below. of Engineers. At this juncture, let us view our input data in a time perspective.2 An average 1984 mine mouth 1. INTRODUCTION price of $34.70 per ton for coal was derived, as discussed below, from executing the MRIO models with 1980 data in 1977 prices and then expressing This paper exhibits the efficacy of using of a the 1980 results in 1984 prices. Then 1986 multiregional input-output (MRIO) methodology to waterborne traffic data (Waterborne Commerce of assess the economic impacts attributable to the the United States, Calendar Year 1986: Department exports of a commodity through one port or of the Army, May 1988, p. 74 and p. 83) for each another. In this instance, the commodity is coal port were reduced to million-ton units and and the ports are Baltimore, Maryland, and multiplied by the unit million-ton impact data for Norfolk, Virginia. The economic impacts are ex- each port in order to derive the total economic pressed in terms of output, value added, income impacts for each port in the year 1986, the most (wages and salaries), and employment (person- recent year for which export coal traffic data are years). All results are expressed in 1984 prices. available. Navigation planning in the U.S. Army Corps of Engineers is characteristically conducted in 2. METHODOLOGY conformity with the Corps navigation planning requirement (as stipulated in the Principles and Guidelines for Water and Related Land Resources The MRIO methodology was selected because it Planning, published by the U.S. Water Resources provides explicit information on the trade among Council) that commodity movements be identified states, and on exports and imports. The source and traced. Thus, the economists at the U.S. Army data, used in the model, is the 1977 Jack Faucett Engineer Institute for Water Resources used a Accounts for the United States which were created multiregional input-output (MRIO) methodology to on a state-by-state basis. These accounts provide identify, trace, and provide an assessment of the the latest consistent estimates of production, economic impacts of the export coal movements consumption, and trade among the 50 states. The through the ports of Baltimore and Norfolk. 1977 Jack Faucett Accounts were derived from data 1690 United States Government work not protected by copyright in the 1977 Economic Censuses pertaining to consumption or capital formation production, consumption, and trade among the industrial sectors within each state. The 1977 ExRorts - sales to foreign customers Jack Faucett Accounts are also consistent with the 1977 National Input-Output Accounts compiled by In the transactions matrix, the sum of each row the U.S. Bureau of Economic Analysis. (interindustry and final demand) gives total gross output. The sum of each column (industry transac- The twenty five sectors listed below are utilized tions and value added) gives total gross outlay. to provide the requisite industry detail. In balance, total outlays equal total outputs, and the model provides a complete description of an 1. Agriculture economy. A particular value of the use of in- 2. Forestry & Commercial 'Fishing put-output models is the industry detail they 3. Other Mining (Non-coal) provide (as well as regional detail in multi- 4. Coal Mining regional models) and the capability to estimate 5. Crude Petroleum & Natural Cas direct and indirect economic impacts due to a 6. Contract Construction change in final demand (which may include ex- 7. Food & Kindred Products ports). 8. Apparel & Textiles 9. Lumber & Other Wood Products In matrix notation, the impact model is as 10. Paper & Printed Products follows: 11. Chemicals 12. Plastics, Rubber & Leather AX - (I-TA)-lTAY 13. Petroleum Refining where: X - Total Output, 14. Stone, Clay & Class Products I - Identity matrix, 15. Iron/Steel Fabricated/Structural Products T - Trade coefficient matrix, 16. Other Metallic Products A - Technical matrix, and 17. Non-Elect. Machinery & Equipment Y - Final demand 18. Electrical Machinery & Equipment 19. Transportation Equipment The multiregional input-output methodology used in 20. All Other Manufacturing this analysis is designed to evaluate the economic 21. Transportation & Communications impact of one million tons of coal exported 22. Utilities through Baltimore Port or Norfolk Port based on: 23. Wholesale & Retail Trade 24. F.I.R.E. & Services * Direct Sales by coal and 25. Government Enterprises transportation industries * Total Output of all industries The model is configured in a way that identifies * Value Added by all industries each of the 8 states which contribute most of the * Employment export coal shipped through Norfolk Port and Baltimore Port, and the remaining 42 states are The maj or inputs to the model, expressed in 1977 aggregated into a single region, giving a total of prices for coal shipments by truck and rail to the 9 regions which are defined below: Norfolk Port or Baltimore Port from each coal producing state in the 9-state region, are: Region Number State Name * Coal sales 1 Maryland * Associated transportation costs 2 Virginia which include port costs, 3 West Virginia handling costs and vessel 4 Pennsylvania services 5 Kentucky 6 Ohio Although the model provides detail on 9 regions 7 Tennessee and 25 industrial sectors, the following discus- 8 Illinois sion simplifies the presentation by emphasizing 9 Rest of U.S.A. the 5 major states that ship coal through each port, and the remaining 45 states, depending on Multiregional input output models replicate the the port in question. The following sections transactions among sectors and regions. There are summarize the results of the impact analysis. major categories of transactions data: Interindus try Transactions - sales by 3. FINAL DEMAND CHANCES one sector to another in the intermediate stages of production A shipment of coal increases the demand for goods Value Added - payments made for wages, and services throughout the United States, espe- management and to capital cially and primarily in the coal producing regions. The MRIO methodology is used to calculate Imports - purchases from foreign sources the effects of producing and shipping a unit used for production of goods and services or for million tons of export coal through the Ports of consumption Baltimore and Norfolk, respectively, by appropri- ately making changes in the final demand for the Final Demand - sales by each sector for coal sector and the transportation sector. Coal 1691 export tonnage for each port for the year 1986 ton of coal. Thus, the total 1980 port cost (the most current export data available) expressed is$4' 53 per ton, which is $5.58 per ton in 1984 as million-ton units (24.177 million tons for pric's.5 Norfolk Port and 6.177 million tons for Baltimore Port) was multiplied by the corresponding unit Average transportation costs for both ports add million-ton impact factors for each port. slightly more than $25.90 per ton to the $34.70 average mine mouth value per ton resulting in a According to the most recent data available from total loaded-on-ship value of approximately $60.60 the Norfolk and Baltimore Engineer Districts, the per ton of export coal in 1984 prices. The total regional distribution of the coal sales through U.S. transportation and handling costs for coal each port, in 1984 prices by originating state, is exported through Norfolk Port or Baltimore Port, shown in Table 1.3 are apportioned as follows: Table 1 Rail 75% Port 18% REGIONAL DISTRIBUTION OF THE SOURCES OF THE Truck 7% EXPORT SHIPMENT OF ONE MILLION TONS OF COAL Total 100% OTHER ALL of course, the total costs per ton of coal to PORT ND WV PA REGIONS foreign customers will increase by the costs of the ocean voyage, unloading, and transhipment. BALTO. 34.0% 38.5% 26.1% 1.4% 100.0% OTHER ALL 4. OUTPUT IMPACTS PORT VA WV KY REGIONS NORFOLK 37.0% 43.1% 17.1% 2.8% 100.0% The overall impact on output of shipping through Baltimore Port is $113.5 million per million tons According to the same reference sources cited of coal and $769.2 million for 6.778 million tons above, the regional distribution of the mine mouth of coal. For Norfolk Port, the overall change in prices per ton of export coal, in 1984 prices by total output is $115.9 million per million tons of originating state, is as shown in Table 2. coals and $2,802.6 million for 24.177 million tons of coal. Generally, trucks haul coal to rail loading facilities and railroads transport the coal to Total industrial output impacts exceed the amount each port. Ports add handling costs for transfer- of direct sales to export coal customers by about ring the coal to ships, fuel costs, and the cost 2 times. This multiplier effect is due mainly to of providing other services to the ships, to the the indirect impact on other industries of the average price of coal per ton. Rail costs are purchases by the coal industry and the transporta- distributed according to the distance traveled in tion industry. The multiregional input-output each state and handling costs are allocated to the model solves directly the overall equilibrium port sector in each port state. impact of the sequence of rounds of interindustry impacts. For the year 1980, it was estimated that truck conveyor and rail haul to the rail loading tipples Economic impacts accruing to states other than cost $2 per ton and rail transportation costs to Maryland, Virginia, West Virginia, Kentucky, and each port range from $14 to $15 per ton.4 Port Pennsylvania (the states primarily involved in handling costs are estimated to be $0.75 per ton. production and transportation of coal) are about The 1980 costs of fuel and supplies, provided by 25% or more of the total increase in output attri- the port to coal-carrying ships, average $3.78 per butable to export of coal through either port. Table 2 REGIONAL DISTRIBUTION OF NINE MOUTH PRICES PER TON AND MINE MOUTH VALUE OF ONE MILLION TONS OF EXPORT COAL Mine Mouth Prices Per Ton Throukh Nine Mouth Value Of A Million Tons Throu (1984 Prices) (Millions of 1984 Dollars) PORT OF OTHER ALL PORT OF OTHER ALL BALTO, MD WV PA REGIONS REGIONS BALTO. MD WV PA REGIONS REGIONS $30.43 $37.29 $32.43 $36.09 $33.67 $10.35 $14.36 $8.47 $0.51 $33.67 PORT OF OTHER ALL PORT OF OTHER ALL NORFOLK VA VV KY REGIONS REGIONS NORFO VA WV KY REGIONS REGIONS $36.76 $37.12 $30.71 $31.25 $35.73 $13.60 $16.00 $5.25 $0.88 $35.73 1692 Table 3 REGIONAL DISTRIBUTION OF THE TOTAL MINE-TO-LOADED-ON-SHIP TRANSPORTATION COST AND VALUE OF ONE KILLION TONS OF EXPORT COAL (Millions of 1984 Dollars) Total Mine-To-Loaded-On-Ship Transportation Cost Of A Million Ton Export Shipment PORT OF OTHER ALL PORT OF OTHER ALL BALT0. MD WV PA REGIONS REGIONS NORFOIK VA WV KY REGIONS REGIONS Rail 12.39 4.78 0.90 0.12 18.19 Rail 12.13 4.72 0.87 0.12 17.85 Truck 0.79 0.89 0.64 0.03 2.35 Truck 0.85 0.99 0.40 0.07 2.31 Port 5.58 NONE NONE NONE 5.58 Port 5.58 NONE NONE NONE 5.58 Total 18.76 5.68 1.54 0.15 26.12 Total 18.57 5.72 1.27 0.19 25.74 71.8% 21.7% 5.9t 0.6% 100.0% 72.1% 22.2% 4.9% 0.7% 100.0% Total Mine-To-Loaded-On-Ship Value Of A Million Ton Export Shipmen PORT OF OTHER ALL PORT OF OTHER ALL BALTO. MD WV PA REGIONS REGIONS NORFOIK VA REGIONS REGIONS M14 Val. 10.35 14.36 8.47 0.51 33.67 MK Val. 13.60 16.00 5.25 0.88 35.73 Transp. 18.76 5.68 1.54 0.15 26.12 Transp. 18.57 5.72 1.27 0.19 25.74 Total 29.11 20.03 10.00 0.66 59.80 Total 32.17 21.71 6.52 1.06 61.47 48.7% 33.5% 16.7% 1.1% 100.0% 52.3% 35.3% 10.6% 1.7% 100.0% Depending upon the economic (structural) composi- data, the employment impacts are 26, 323 person- tion of each state, indirect impacts vary, since years for export shipment of 24.177 million tons different sectors have varying multipliers. of coal through Norfolk Port and 7,228 person- years for export shipment of 6.778 million tons of coal through Baltimore Port. 5. VALUE ADDED IMPACTS The 23-job difference in the employment totals (about 2% of either total) for the unit million- Value added is a fundamental indicator of the ton shipment, is well within the margin of error benefits attributable to economic impacts. Value for this type of analysis. Such differences in added represents returns (payments) to employees, total employment, if actual, would be due to the managers, and capital investments due to increased differences in the output -per- employee estimates output. Additions to value-added total $63.14 for the transportation and manufacturing sectors million per year per million tons of coal exported in the impact states. Since we do not know the through Norfolk port and $62.96 million when actual routes of the coal from the mines to the exported through Baltimore port. The correspond- ports, however, we also do not know the actual ing value added totals for the 1986 traffic are aggregation of employment impacts that are $1,526.6 million for 24.177 millions tons through attributable to the transportation sectors of the Norfolk Port and $426.8 million for 6.778 million impact states along the routes of the component tons through Baltimore Port. The major impact shipments. Consequently, the lack of detail in states (Maryland, Virginia, West Virginia, and the transportation input data necessitates our Pennsylvania) receive about 80 percent of the viewing the 23-job difference as being too small value added for shipment through either port. to give either port an actual employment advan- tage. We suggest that the actual employment impact stemming from the difference in the total 6. EMPLOYMENT IMPACTS employment generated by a unit million-ton shipment through each port is likely to be much smaller. Export coal provides jobs in coal mining, trans- portation, and other industries. Estimates of the full time equivalent job impacts are shown in the 7. IMPACTS BY INDUSTRY SECTOR following table. In the 9 regions, based on the $60.60 average loaded-on-ship coal value per ton in 1984 prices, the million-ton shipment of export Where, one may inquire, besides the direct impacts coal through the Norfolk port will generate 1,089 on coal mining and transportation sectors, do the jobs, and through the Baltimore port will generate impacts accrue? For shipment through Norfolk 1,066 jobs. In terms of the 1986 export traffic Port, about 35 percent of the total employment 1693 Table 4 REGIONAL DISTRIBUTION OF FINAL DEMAND CHANCE IMPACTS LARGEP THAR THREE PERCENT STEMMING FROM EXPORT COAL SHIPMENTS THROUGH BALTIMORE PORT AND NORFOLK, PORT (USA Total Impacts In 1984 Prices And Person-Years) REGION OUTPUT VAL ADD INCOME EMPLOYMENT REGION OUTPUT VAL ADD INC EMPLOYMENT One Million Tons Through Baltimore Port One Million Tons Through Norfolk Port --------- $Millions -------- Person-yrs --------- $Millions -------- Person-yrs MD 39.01 23.80 13.12 443.3 VA 43.42 24.76 14.87 476.5 WV 28.94 16.64 9.97 257.7 WV 31.21 17.92 10.69 274.5 PA 17.06 8.85 5.17 142.8 KY 10.13 5.67 2.99 84.0 REST/SHED 6.74 3.21 1.71 62.4 REST/SHED 8.95 4.15 2.37 85.4 REST/USA 21.72 10.46 3.55 160.2 REST/USA 22.21 10.64 .3.72 168.4 Total 113.48 62.96 33.51 1,066.4 Total 115.92 63.14 34.64 1,088.8 --------- Percentage Analysis --------- --------- Percentage Analysis -------- MD 34.4% 37.8% 39.2% 41.6% VA 37.5% 39.2% 42.9% 43.8% WV 25.5% 26.4% 29.7% 24.2% WV 26.9% 28.4% 30.9% 25.2% PA 15.0% 14.1% 15.4% 13.4% KY 8.7% 9.0% 8.6% 7.7% REST/SHED 5.9% 5.1% 5.1% 5.8% REST/SHED 7.7% 6.6% 6.8% 7.8% REST/USA 19.1% 16.6% 10.6% 15.0% REST/USA 19.2% 16.8% 10.7% 15.5% Total 100.0% 100.0% 100.0% 100.01 Total 100.0% 100.0% 100.0% 100.0% 6.778 Million Tons Through Baltimore Po 24.177 Million Tons Through Norfolk Port --------- $Millions -------- Person-yrs --------- $Millions -------- Person-yrs MD 264.44 161.32 88.96 443.3 VA 1,049.70 598.60 359.42 11,520.4 WV 196.19 112.82 67.55 257.7 WV 754.70 433.24 258.45 6,637.0 PA 115.64 60.01 35.01 142.8 KY 244.88 137.19 72.30 2,030.2 REST/SHED 45.71 21.74 11.59 0.0 REST/SHED 216.37 100.42 57.26 2,063.8 REST/USA 147.23 70.89 24.03 160.2 REST/USA 536.95 257.18 90.02 4,072.0 Total 769.21 426.77 227.15 1,066.4 Total 2,802.60 1,526.63 837.46 26,323.3 impacts are in sectors other than coal mining and impacts attributable to shipment through either transportation, about 31 percent of total value port. Non-coal-producing states receive over 20 added impacts are in the other sectors, and about percent of the output.impacts. 35 percent of total output impacts are in the other sectors. Table 5 provides greater detail on Generally, we recommend that one should be total output shares by sector and port. cautious in extending these preliminary. results. Although our knowledge of the sources, quantities, and prices of the coal shipped through each port 8. FINDINGS appears quite good, we must emphasize that we have virtually no information concerning the actual routes of the coal from the mines to the ports. Exporting coal through the Ports of Baltimore and To the extent, from a theoretical perspective, Norfolk generates direct impacts in the 5 regions that the data onthe coal shipments and associated which are major coal producing states and indirect transportation costs used in this analysis reflect impacts in all regions. Differences in the the actual tonnages and associated transportation indirect impacts are due to trade and production costs of specific component shipments from source- patterns. The largest impacts are induced in the region to port, we find it appropriate to attri- states which produce coal and provide transporta- bute the attendant impact differences to the tion and port handling of cargo. Overall impacts, differences in the output-per-employee estimates.. however, are significant in other regions of the United States. The port states receive about 35 Calculation of employment totals, the natural percent or more of all of the impacts, while West basis of the difference in employment totals, is Virginia receives more than 25 percent of all dependent in part upon the output-,.per-employee 1694 Table 5 SECTORAL DISTRIBUTION OF FINAL DEMAND CHANGE IMPACTS LARGER THAN THREE PERCENT STEMMING FROM EXPORT COAL SHIPMENTS THROUGH BALTIMORE PORT AND NORFOLK PORT (USA Total Impacts In 1984 Prices And Person-Years) INDUSTRY OUT VAL ADD INCOME EMPIDYMENT INDUSTRY OUTPUT VAL AD INCOME EMPLOYMENT One Million Tons Through Baltimore Port One Million Tons Through Norfolk Port -------- $Million --------- Person-yrs --------- $Millions -------- Person-yrs Coal Min'g 40.17 23.01 12.34 268.8 . Coal Min'g 42.19 23.61 12.96 282.3 Crude Pet. 5.96 4.48 0.50 20.1 Crude Pet. 5.70 4.28 0.48 19.2 Pet.Refin. 6.31 0.87 0.23 6.4 Pet.Refin. 5.96 0.81 0.20 6.0 Trans/Com. 32.94 20.56 13.24 430.2 Trans/Com. 32.32 19.67 13.49 422.1 FIRE/Serv. 8.16 5.42 2.14 138.5 FIRE/Serv. 8.42 5.56 2.14 142.8 All Other 19.94 5.05 202.5 All Other 21.32 9.21 5.36 216.3 Total 113.48 62.96 33.51 1,066.4 Total 115.92 63.14 34.64 1,088.8 -------- Percentage Analysis -------- --------- Percentage Analysis -------- Coal Hin'g 35.4% 36.5% 36.8% 25.2% Coal Min'g 36.4% 37.4% 37.4% 25.9% Crude Pet. 5.3% 7.1% 1.5% 1.9% Crude Pet. 4.9% 6.8% 1.4% 1.8% Pet.Refin. 5.6% 1.4% 0.7% 0.6% Pet.Refin. 5.1% 1.3% 0.6% 0.6% Trans/Com. 29.'0% 32.7% 39.5% 40.3% Trans/Com. 27.9% 31.2% 39.0% 38.8% FIRE/Serv. 7.2% 8.6% 6.4% 13.0% FIRE/Serv. 7.3% 8.8% 6.2% 13.1% All Other 17.6% 13.7% 15.1% 19.0% All Other 18.4% 14.6% 15.5% 19.9% Total 100.0% 100.0% 100.0% 100.0% Total 100.0% 100.0% 100.0% 100.0% 6.778 Million Tons Through Baltimore Port 24.177 Million Tons Through Norfolk Port -------- $Millions -------- Person-yrs --------- $Millions -------- Person-yrs Coal Hin'g 272.28 155.94 83.66 1,821.6 Coal'Min'g 1,020.17 570.81 313.33 6,825.1 Crude Pet. 40.41 30.39 3.41 136.2 Crude Pet. 137.72 103.60 11.62 464.2 Pet.Refiu. 42.76 5.90 1.55 43.2 Pet.Refin. 144.09 19.53 4.84 145.5 Trans/Com. 223.27 139" '38 89.77 2,916.0 Trans/Com. 781.50 475.58 326.26 10,206.1 FIRE/Serv. 55.33 36.72 14.51 938.4 FIRE/Serv. 203.60 134.40 51.83 3,453.0 All Other 135.16 58.43 34.24 1,372.5 All Other 515.52 222.71 129.58 5,229.4 Total 769.21 426.77 227.15 7,228.0 Total 2,802.60 1,526.63 837.46 26,323.3 estimates in the sectors of the impacted regions. of access to detailed transportation input data Actual employment totals would also be based on also necessitates our reiterating our opinion that detail actual transportation data that show the the 23-job difference (for the unit million-ton actual routes of coal shipments from mines to basis or the 1986 actual tonnage basis) is too ports. Data on actual routes of component ship- small to be significant. ments would be used to aggregate actual trans- portation costs and derive actual transportation Even without the benefit of detailed transports- cost totals. tion data, however, it is apparent that the MRIO methodology can be used to readily quantify (at The significance of this lack of detailed trans- least in relative orders of magnitude when detail portation data is apparent in the example in which transportation data are unavailable) the inter- the export of a million tons of coal through dependence (expressed as economic impacts) of All Norfolk Port generates an estimated total employ- states. in, the U.S. economy when a bulk commodity ment that is 23 full-time jobs larger than the is exported through an American port. estimated total employment stemming from the export of a million tons coal through Baltimore Port. Additionally, please remember that the 23- ENDNOTES job difference is approximately two percent of the total employment impact of a million-ton coal 'Economists in the Navigation Analysis Division shipment through either port; therefore, it is an at the U.S. Army Engineer Institute for Water Re- estimate that is well within the margin of error sources, Fort Belvoir, Virginia 22060-5586; for this type of analysis. Consequently, our lack telephone: (202) 355-2240. 1695 2Due to a paucity of data and time ' technolo- 3Booz Allen and Hamilton, Transporting Export gies from different years were consciously Coal from Appalachia, a report prepared for the aggregated to derive results that are primarily U.S. Department of Energy (November 1982). useful as indicators of relative orders of magnitude. When detailed data on actual trans- 41bid., November 1982. portation costs, for specific shipments of export coal from mine to port, become available, results 5Siberman and Yochums, The Economic ImRact of which are more definitive will be produced. Virginia's Ports on the Commonwealth. 1696 Interpretation of SEASAT Radar Altimeter Returns from Ali Overflight of Ice in the Beaufort Sea L.S. FEDOR Wave Propagation Laboratory, Environmental Research Laboratories, National Oceanic and Atmospheric Administration, Boulder, CO 80303 E.J. WALSH t Wallops Flight Facility, Goddard Space Flight Center, National Aeronautics and Space Administration, Wallops Island, VA 23337 ABSTRACT SEASAT radar altimeter data are examined for rn the effects of sea ice on the returns during an over- flight of the Beaufort Sea. Waveforin parameters <8 and their statistics are combined to form a param- > eter sensitive to the presence of sea ice. Variations 0 in the value of this ice parameter are compared with ice charts obtained from the Canadian Atmo- 2 spheric Environment Service. Particular attention 08 Z. is paid to the sensitivity of the radar altimeter to the open-ocean sea ice boundary. -@2 -@4 -@6 -8 0 a 1,6 @4 32 GATE NUMBER INTRODUCTION Figure 1. Modeled ra.dar altimeter waveform re- The satellite-borne microwave radar altimeter is spouse to ocean waves. SEASAT parameters were an all-weather high-resoluLion instrument. Esti- used for input. The significant wave height used was mates of surface roughness and variations in surface 2 rn. topography are determinable from the returned- pulse waveforms. Figure i illustrates the expected waveforin from the ocean surface as determined by rough-surface scattering, based on specular point tp is a function of the radar characteristics and the theory [Barrick, 1972]. In the figure, gate Dumber 0 surface height roughness but not the altitude of the defines the rnean surface position; for ocean returns, instrument above the surface. The time constant of this is approximately at the half-power point of the the trailing edge-decay of the waveform, t,, is the two- waveform. Gate numbers less than 0 are indicative way time of flight difference between the suborbital of ranges closer to the radar and therefore above the point and the edge of the effective scattering region: mean surface level. Gate numbers greater than 0 indicate ranges farther from the radar, and signals in 2H these gates come from off-nadir reflections (and also t' = c C2 (2) from below the mean surface level in the presence of waves). The rise-time parameter of the leading edge, tP1 is a function of rms elevation (height roughness). where H is the effective height of the instrument This time constant is defined to be above the surface, 1/V).2 - [(8ln2) /lfi2l + [1/82], 0, B is the half-power beamwidth of the radar antenna, 2 and s is the rms surface slope. For most ocean - X., + 2h2 tp = c returns a > @bB so that 0, - V)B/vlr8lii2 and t, is nearly constant. For the SEASAT altimeter, 0, - 8 when s < 0.003 radians. Since s - 0.07 for an where c is the speed of light, 'X@ = cT-/[4.(ln2 )1121, 'r open-ocean local wind of 1 mls [Fedor and Brown, is the transmitted (compressed ,) pulsewidth, and h is 1982] the surface winds would have to be virtually the rins surface elevation; for SEASAT X. - 28 cm. nonexistent before the altimeter waveform would exhibit a response in t, significantly smaller than t On assignment at the NOAA/ERL Wave Propa- that determined solely by the beamwidth. gation Laboratory The functional form of the received power from a CH2585-8/88/0000-1697 $1 @1988 IEEE rough surface, to first order, is given as [Fedor et al., 7- 6- 19791 A2 2 4- 112 JR.(0)J P"(t) = I( ' [21r X H F (t) (3) H 4 3- where K contains the radar constants including the transmitted power and antenna gain; A, is the radar -4 !-3 -2 -1 0 3 4 wavelength; H, and s have been defined previ- CROSS-TRACK DISTANCE (km) Ously; IRR(O)l is the rough-surface Fresnel reflection coefficient; and 2t J7(t) [1 + erf exp(- tp t' is the waveform function with a value of 2 at its maximum. The parameters tp and t, are described earlier, and erf (x) is the error function with argument X. The surface effect on the magnitude of the received power is primarily influenced by the average PULSE-LIMITED backscatter cross section per unit area at normal FOOTPRINT incidence, o-O = (JRR(0)J')1,3'- We concentrate our discussion on the surface effects of sea ice on o, tp@ and t_ ALTIMETER RETURNS from SEA ICE Earlier studies note that radar returns from sea ice are characterized by narrow waveforms and by greater signal strength than returns from ocean waves Figure 2. Top: Schematic of a radar altimeter [Brown, 1982; Dwyer and Godin, 1980). From the pulse advancing toward a smooth surface. The pulse previous discussion we note that the waveforms may (blackened) is just beginning to reach the surface. be made more narrow relative to ocean returns by The vertical scale is magnified by a factor of 400 reducing the slope roughness of the surface. In the relative to the cross-track scale, so the incidence extreme, assume that the surface is perfectly smooth. angles are distorted. Bottom: Plan view of the area Then the power received at the radar will have the illuminated by the pulse. form P,r = K A,@,@ [-7r H2 I R@ (0) 1'] (4) indicated above. H4 Only those portions of the pulse that are incident on facets of the surface oriented normal to the direction where JR,(O)I' is the Fresnel reflection coefficient of of propagation are reflected back to the radar. This the smooth surface, and all other parameters are as occurs only at the nadir point for a perfectly smooth defined previously. Thus, the signal is returned to the surface. Any off-nadir returns require that the radar as if reflected from a smooth sphere of radius surface be tilted toward the radar. If we define M H, the shape of the returned pulse being the same as as the ratio of power received from a smooth surface the transmitted pulse. In contrast to ocean returns, to the power received from an ocean surface at the note that the position on the waveform for locating waveform maximum, then the range to the surface occurs at the maximum of the return rather than near the half-power point. (5) Figure 2 illustrates the interaction of the radar pulse with a smooth surface. The top of the figure shows the vertical plane, with sixteen positions of the where spherically expanding pulse indicated, as it advances toward the surface. The position at which the pulse 010 H JR@, (0)12 just begins to interact with the surface is blackened. 27rI/2 X. The vertical scale is magnified by a factor of 400 with respect to the cross-track scale, and the apparent off-nadir angles are correspondingly distorted. The Using SEASAT values, we find H 800 kni and bottom of the figure shows a plan view of the area X. - 28 cm, and the coefficient multiplying JRI(0)J' 9 open ocean illuminated by the pulse for two of the pulse positions is 5 dB. For the the Fresnel reflection 1698 70 J I 160 -155 -150 -145 -140 -135 -130 -125 -120 Longitude (*) Figure 3. SEASAT 3-day repeat suborbital tracks for the Beaufort Sea. The tracks are limited to less than 72'N by the orbit inclination angle. coefficient is approxi ately -2.1 dB. If we use 0.005 41t, and through the relative standard deviation 'm as the value for s@ (for surface winds - 1 m/s)the S'/P. The value for o,'.. in decibels is used in the average backscatter cross section per unit area is computation. about 21 dB. Thus, even for a poor reflector such as first-year sea ice, which has a Fresnel reflection SEASAT DATA OVER SEA ICE coefficient of - 11 dB [Brown, 1982], a smooth surface During the last month of its lifetime SEASAT will greatly enhance the returned power; that is, was in a 3-day repeat-orbit inode. In this period, M - 27 dB. from the end of September to the middle of October To explain the appearance of the waveform shape 1978, we obtained several data sets over the Beaufort and its magnitude the radar returns from sea ice Sea. Figure 3 shows the Beaufort Sea area and indicate that there must be a high probability of the SEASAT suborbital tracks for this repeat-orbit flat surfaces at the satellite 'subpoint [Brown, 1982]. period. Since the radar altimeter is a nadir-looking Off-nadir returns would result if portions of the instrument, the tracks also are indicative of the surface were rough and/or if there were flat surfaces altimeter coverage of the Beaufort Sea. Although tilted toward the radar. Using parameters that are this results in large areas where there is no data estimated routinely during processing of over-ocean coverage, the radar altimeter can still provide a cost- radar returns, we define an ice parameter _T, that is effective means for long-term monitoring of sea ice sensitive to the non-ocean-surface response expected dynamics by providing detailed information over the from a surface with sea ice present. The quantity I same set of ground tracks every few days. is given as We selected SEASAT radar altimeter data from orbit number 1439 obtained on 5 October 1978 to tp S, illustrate the effects of sea ice on the radar returns. _T 100.1--01 (6) tI P max In Figure 4, we show a schematic of the ice conditions on this day that was derived from ice charts compiled by the Canadian Atmospheric Environment Service. where Sp is the standard deviation of five wave- The region marked as new ice, extending from Banks form values, beginning with the waveform maxi- Island in the east to approximately Point Barrow in mum, about their mean value P; o,'.x is the average the west, is no more than a week old. The ice chart backscatter cross section estimated at the waveform compiled in the previous week, dated 28 September maximum; t, and tp are as defined earlier; and the (not shown), indicates that this whole region is open coefficient 100 is a scaling factor. This quantity -1 water. Superimposed on the ice chart is the satellite is responsive to flat surfaces both through the ratio suborbital track. The middle part of the track is over 1699 750 700 -1600 -1550 -1500 -1450 -1400 -1350 -1300 -1250 -1200 050CT OOPEN r----ISTRIPS& NEW NEW Y. L.L.J. [M MM-y- I F-IM-Y. 1978 WATER PATCHES LL@JICE ICE, M NEWICE ICE Figure 4. Schematic of the Canadian Atmospheric Environment Service ice chart for 5 October 1978. Solid diagonal hatching is new ice with some niulti.year ice present. Cross-hatching is multiyear ice with sorne new ice present. Superimposed on the chart is the SEASAT suborbital track for orbit 1439. BEAUFORT SEA-ORBIT 1439 DATE 278 .1500 06 4000 3. -5000 A? 17 .0 NC Figure 5. Three-dimensional radar altimeter waveform plot for orbit 1439. Time (in seconds after the beginning time of 1528:41.58 GMT) increases from right to left. Radar range increases toward the upper right hand corner of the figure. Below the waveforms is a reference plot of amoax - 1700 BEAUFORT SEA, REGION C - ORBIT #1439 DATE=278 TIME= 15:30:33.32 140 WAVEFORM FILTERING CY IV NIV 1500 "60 5@ IV @ 2 1000 3: 500 Z,? CD 0 (SeC01111.) cc 4N Figure 6. Expanded view of the three-dimensional waveform plot for the data set Iq marked region C in Figure 5. this new ice region, whereas the parts of the track Other features in Figure 5 can be seen more clearly over the Amundsen Gulf and the Chukchi Sea are in the expansion of the area labeled region C (Figure mostly open water. 6). There are numerous regular patterns of apparent signals that appear above the mean surface level (at The presence of seaice in the radar altimeter signals lower gates) that decrease in radar range as a function is clearly demonstrated in Figure 5, which shows a of time. These signals are a result of the sampling three-dimensional plot of the altimeter wavefori-ris filter gate response to the strong sharp signal (John L. as a function of time. The time axis is labeled MacArthur, The Johns Hopkins University, Applied in seconds after the beginning time of 1528:41.58 Physics Laboratory, private communication) and are GMT, with time increasing from lower right to upper not a physical phenomenon. However, although there left. Each waveform used in the plot represents 100 is a feature occurring at 1530:51 GMT where smooth pulses averaged on board the satellite. The waveform ice is not present (no strong main return), there is a sampling time is approximately every 0.1 s. The axis signal beginning some 17 gates above the surface (8.0 labeled Gate represents the position of the waveforni in). The along track extent of this feature is about sampling gates; the center of the axis denotes the 0.8 s (5.28 kni). We believe that this feature is a large mean surface level. Increasing range from the radar ice island sensed by the altimeter. The ice island follows the axis toward the upper right side of the rr-Q, northwest of Banks Island, was observed by the figure. The power in the gates is plotted on the SEASAT Synthetic Aperature Radar on orbit 1438 vertical axis. As mentioned previously, the first (W. J. Campbell, United States Coast and Geodetic and last minutes of data are open-ocean returns * Survey, private communication). Other features in The waveforms in these regions are as shown in this mixture of sea ice and open water extending a Figure 1. Their amplitudes are relatively small few gates above mean surface level are inferred to because of the vertical scale that was necessary to be multiyear floes. The features at 1530:45.57 and accommodate the over-ice returns dominating the 1530:48.02 are data dropouts. center of the figure. Each waveforin is normalized to have equal area, to prevent overwhelming of the ICE PARAMETER over-ocean waveforrns by the over-sea-ice waveforms. This feature is demonstrated more clearly in the plot A test of the proposed ice parameter's response of o-__ along the bottom of the figure. The over to sea ice is shown in Figure 7. For reference, the ice returns are at the 40 dB level whereas the over average backscatter cross section is plotted above the water returns are at the 10 dB level. The waveform satellite suborbital track and IT is plotted below the shape about the maximum for the over ice returns track. The maximum value for I on the plot is 2000. is much as we described previously for a smooth The over-open-water values are less than 10. surface. Still, the off-nadir returns visible behind the We refer to the right-hand portion of the figure, maxima are riot negligible. The reduced levels, to where sea ice is first encountered. We note that o-L. around 25-30 dB, at 1531:35-00 GMT are probably increases in value before there is a corresponding due to the presence of multiyear floes (see Figure 4). response in I. Comparisonofo'."- with the waveform 1701 ORBIT #1439' 75 Radar Backscatter Cross-Section at 'Pulse Maximum(dB) 701 _J Ice Pararn ete -1 0 500 1000 1500 2000 -160, -155 -150 -145 M -135 -130 -125 -120 Longitude (") Figure 7. Calculated ice parameter I for orbit 1439 (plotted below the track). a'... is plotted above the track for reference. plot in Figure 5 shows a similar behavior; that is, to 155.5'W the altimeter parameters are again in- the waveform shape is similar to a normal open- dicative of ice. Now, however, the character of IT water response. Furthermore, the response of I is different, without large excursions, but having a agrees well with the ice chart in Figure 4. Either, more rapid variation along track; is also reduced the surface near the ice edge is calm relative to in magnitude. These characteristics are all indicative the open ocean earlier in the track, or there are an of rougher surface conditions but with a considerable insufficient flat surfaces oriented toward the radar to number of flat surfaces. In contrast to results in the compensate for a water-dominated mixture. From ice chart, the altimeter responses west of 155.5'W this point until the track crosses 138*W longitude, indicate that the ice did not extend down to the the surface is dominated by very flat structure. This satellite track. is indicated by the large values of I as well as the high values in Next there is an open- water CONCLUrarqGRFMARKS area clearly marked by both altimeter paranieters The ice parameter 1, as defined here,provides a and the waveforms. The satellite suborbital track is reasonable estimator of the presence of sea ice in well away from the open-water region indicated on the returned radar altimeter signals. It is empiri- the ice chart. We believe that the altimeter correctly 4csees" open water and that this difference from the cally based on the waveform shape and magnitude ice chart is indicative of the dynamic conditions of expected of returns from a linear combination of the region. The ice charts are composited from a smooth and rough surfaces. It is expected that, as series of aircraft reconnaissance flights during the the ice undergoes stresses and rafting, surface tilting week prior to the given date. These aircraft flights will enhance or degrade the returns. Hummocking are similar to the the 3-day repeat-tracks shown in will further reduce the number of reflecting facets as Figure 3, but more restricted in extent and duration. the ice grows older even though weathering smooths Since the flight track data is than interpolated in the rough edges. We have shown data suspected of order to complete the charts it is reasonable to indicating multiyear ice; here the signal signature is expect that their accuracy would not be sufficent sufficiently different from the expected open ocean for our purposes. Other ground truth that would response to give a significant value for 1. The major limitation of the discussion is lack of high-quality be available is Landsat imagery but that was not ground-truth. The ice charts are valuable but lack pursued at this time. the unique one-to-one evidence that is needed. Beyond this open-water region is evidence of the multiyear floes discussed previously. From 144'W 1702 Acknowledgments. We thank Konrad Steffen and Edgeworth Westwater for reading the manuscript and for their helpful com- ments. One of us (LSF) received partial support from the Office of Naval Research under contract #NR 307-536/3-22-85. REFERENCES Barrick, D.E., Remote sensing of sea state by radar, Chapter 12, Remote Sensing of the Troposphere, edited by V.E. Derr, NOAA Wave Propagation Laboratory, Boulder, Colo., 1972. Brown, G.S., A theory for near-normal incidence microwave scatter- ing from first-year sea ice, Radio Sci., 17, 233-243, 1982. Dwyer, R.E., and R.H. Godin, Determining sea-ice boundaries and ice roughness using GEOS-3 altimeter data, NASA Contract Rep. CR-156862, 1980. Fedor, L.S., and G.S. Brown, Waveheight and wind speed measure- ments from the SEASAT radar altimeter, J. Geophys. Res., 87, 3254-3260, 1982. Fedor, L.S., T.W. Godbey, J.F.R. Gower, R. Guptil, G.S. Hayne, C.L. Rufenach, and E.J. Walsh, Satellite altimeter measurements of sea state-An algorithm comparison, J. Geophys. Res., 84, 3991-4002, 1979. 1703 Airborne Pulse-Linlited Radar Altimeter Return Waveforin Characteristics Over Ice in the Beaufort Sea L.S. FEDOR Wave Propagation Laboratory, Environmental Research Laboratories, National Oceanic and Atmospheric Administration, Boulder, CO 80303 G.S. HAYNE AND E.J. WALSH t Wallops Flight Facility, Goddard Space Flight Center, National Aeronautics and Space Administration, Wallops Island, VA 23337 ABSTRACT Pulse-limited radar data taken in March 1978 with the 13.9 GIlz AAFE altimeter from 1500 in altitude over ice in the Beaufort Sea are registered to high quality photography. The variations of 61 BEAUFORT SEA the radar return waveform shape and signal level are correlated with the variation of the ice type determined from photography. :D 70 L - E-4 0. INTRODUCTION Diring mid-March 1978, the NASA C-130 aircraft Z 4 171V b I was deployed to Eielson Air Force Base in Fairbanks, 01 Alaska, to make a series of flights over ice in the Beau- MACKENZIE I fortSea. The radar altimeter data analyzed here were BAY obtained northeast of Mackenzie Bay on March 14th 69 L L - - - in the vicinity of 69.9'N, 134.2'W (Fig. 1). The data were taken with a 13.9 GHz radar altimeter developed under the NASA Advanced Applications 136 135 134 133 132 Flight Experiments (AAFE) Program [Hughes Air- craft Company, 19761. This airborne radar was WEST LONGITUDE (0) built as a forerunner of the SEASAT radar altimeter Figure 1. Region (indicated by the form6e cross) and utilized the same pulse compression technique. of the AAFE altimeter data acquisition March 14, It has tile same pulse compression ratio (1000 : 1) 1978. as the SEASAT althneter and approximately the same range resolution (0.417 in versus 0.469 in for SEASAT). One significant difference was that the intervals equal to tile pulse duration as it spherically AAFE altimeter has only 24 range gates instead of expands within the antenna pattern. the 60 that SEASAT had. Tile bottom of Fig. 2 shows a plan view of the area illuminated oil the flat surface for two of the AAFE ALTIMETER RETURN WAVEFORMS pulse positions. As the pulse advances from the darkened position to the next one, the illuminated Iii the two flight lines analyzed in this paper, the area increases from zero to its maxinlimi value and aircraft altitude deviated only about I ni froni its thereafter remains constant. As the pulse continues mean value (1524 in and 1566 in). The top of Fig. to advance, the illuminated area becomes an annulus, 2 shows the position in space occupied by a radar which increases in radius but narrows in width, the pulse (darkened region) transmitted from 1545 in two trends compensate to maintain the illuminated altitude just as it begins interacting with a flat sea area equal to that of the pulse-limited circle. surface. Also indicated are the next 12 positions that If the range to the mean surface is in its nominal the transmitted pulse sequentially occupies at tijrne position, centered in the 24 range gates of the AAFE altimeter, tilen only the last 12 gates (indicated in t On assignment at the NOAA/ERL Wave Propa- Fig. 2) would probe a flat surface. It, is apparent fron't gation Laboratory Fig. 2 that under these circumstances the maximum CH2585-8/88/0000-1704 $1 @1988 IEEE 6- constant altitude, antenna-pointing variation during Ji 5- the flight lines being analyzed would essentially be caused by aircraft roll attitude variation. Analysis E-4 4- of the photography indicates that the roll attitude 0 variation was typically within �'1'. That means Chat H 3- the return signal front a surface whose scattering is 2- independent of incidence angle would be down to about 65% of the maximuin by the 12th gate past the nadir return (3.3' off-iiadir). But we sliall see that when the surface is highly specular, the entire return -150 11-100 -50 1 0 1 50 100 )1 150 signal can rise and fall within two adjacent gates. 11 CROSS-TRACK DISTANCE (in) DATA ANALYSIS lligh-quality photography was taken simitltane- ously with the radar data, using a 9-inch format Zeiss cailiera with a 6-ilich lens. Figures 4, 5, arid 6 each show two noricontiguous segments of approx- imately 30-s duration where the photography has been registered to the variation of the radar signal level and the return waveforms. In each segment, time increases from left to right and the average return waveforms are at the top, normalized to their respective peaks. Below the waveforms is the vari- ation (in dB) of the signal level at the peak of the return. The AAFE altimeter was not absolutely cal- PULSE-LIMITED,)i ibrated and the indicated power values are relative. FOOTPRINT Below the signal level variation is the photography of the swath interrogated by the radar. The AAFE altimeter pre-averages the retiirn wave- forms and outputs average returns at about a 10-1-1z rate. The pulse repetition frequency (PRF) used for the data under analysis was 476 Hz, and each output Figure 2. Schematic illustration of the interaction waveform was the average of 52 returns. In Figures of radar altimeter pulses with a smooth surface. The 4, 5, and 6, each waveform shown is a nonoverlapping tipper part of the figure shows the position in space average of 10 output waveforms spanning a time of a pulse (darkened region) transmitted from art interval of 1.09 s and representing 520 transmitted altitude of 1545 in altitude just as it reaches the pulses. The standard deviation of the fluctuations in surface. The lower part of the figure is a plan view of the area illuminated on the surface for two of the pulse positions. OFF-NADIR ANTENNA POINTING width of the swath interrogated on the surface would 0 be approximately 250 in. On the other hand, a highly 60 specular surface would return a signal to the radar only from the nadir point, and the effective area > contributing to the return would be much smaller in M 40 diameter. The monostatic horn antenna used by the AAFE -1 1,,20 .C@ 100 altimeter had a 3 dB beamwidth of .15', so the @2 two-way beamwidth was 10.6' and signals from 5.3' a4 off-riadir would be reduced by half. Figure 3 shows Z C@ what the return waveform would look like if the AAFE altimeter were at 1545 in altitude over a flat, diffusely scattering surface. The four waveforms essentially show the variation with off-iiadir pointing -8 0 8 16 24 of the antenna gain pattern as a function of range GATE NUMBER to the surface. The curves past gate 12 are shown dashed as a reminder that the AAFE altimeter Figure 3. Modeled returned waveforms for the would generally have only 12 of its gates probing the AAFE altimeter at 1545 m altitude. The four surface if the return front nadir were centered in its waveforms show the effect of varying the off-nadir 24 gates. Since the pitch attitude of the aircraft pointing angle of the antenna over a surface whose would be very stable as the C-130 maintained a scattering is independent of incidence angle. ;0 1705 X ID"", @@ J@ vllj@ Il" rs@ @q, p r" rp-, IS Y f@.'l f ll, X "Ik "A""..""'A"'O" N""!"O 0111"'If 16 4Z Q KX, 2W Al", E W", @,l, fl, r'q ZZ I;Z, QRT'1@5 iai 3,4,`;"'I@PEK (dB) 4LEVEL ............... Lt K 4 2;@@ $P' MMM,;! I F@' N2 ........... Figure 4. The response of the AAFE altimeter to ice returns. Each of the two noncontigluous data segments shown displays the returned wavefornis at the top, the. AC4C power in the middle, and the corresponding photography at the bottom. these average returns due to Rayleigh fading would (A,C) of homogeneous first year white ice (thicker be only about 5% of the mean value at each point on than 30 C111; refer to [Steffen, 1986] generally exhibit the waveform. both the sharp peak at nadir indicative of specular The aircraft ground speed was approximately 76 scattering and the slow decay following the antenna m/s and the length of each of the segments in Figitres pattern (Fig. 3) in the later gates, which indicates 4, 5 and 6 was about 2.3 km. The cross-track width of diffuse scattering. Tile waveforms from the area the photographic images is 250 in, the diameter of the of lowest backscattered power in the shear zone region nominally interrogated by the altimeter (Fig. (B) had a signal level of approximately 3 dB and 2). Letters are used to identify specific features. The showed virtually no decay in tile plateau region. The registration of the radar data to the photography absence of a plateau falloff seems to suggest that was made as close as possible but should not be the backscattered signal may actually have increased considered exact. off-iiadir, compensating for tile falloff of the antenna In the top segment of Fig. 4, the aircraft traversed a gain (Fig. 3) to maintain tile return signal level shear zone that was probably caused by the Beaufort constant. gyre. The width of the zone was about 625 in The bottom segment of Fig. 4 shows that the re- and the AAFE altirneter indicated that the mean turned power could reach levels about 20 dB higher elevation within this rubble field was approximately than the surrounding areas when the radar encourl- 1 ni higher than the surrounding area. The radar tered gray (10-15 ciii thickness) and gray-white ice return peak power in this region was 10-15 d13 (15-30 cm thickness) either at a refrozen lead (D) lower than the surrounding areas, and the return or in a matrix of white ice flows (H). Notice that waveforms indicate that the surface was scattering tile response of the radar to the refrozen lead (D) is diffusely. The waveforms in the surrounding area strong even though the width of tile lead is smaller 1706 J N A t PEAK S I GNAL. LEVEL (dB) L 10@ J JR, .. . ..... . ... M P Figure 5. Two additional noncontiguous data segments in the same format as Fig. 4. than the altimeter ptilse-limited footprint (Fig. 2). Because of the rapid cross-track variation of the ice On the other hall([, the radar signal evidences sharp structure occurring in the gray ice and light nilas at decreases when it encounters compression ridges (0), it is. difficult to assess what is causing the signal (E,F,G) formed by the buckling, bending, or local to be LO dB higher than in the more uniform region crushing of colliding ice features. of light nilas at (P), which exhibits some rafting. The top segment of Fig. 5 shows the radar response The region (S) of dark nilas (less than 5 cm to regions of gray ice (10-15 cin thick) and light ililas thickness) oil the left side of the top segment of Fig. (5-10 cm in thickness). The uniform regions of 6 has a typical signal level of 32 dB as well as the gray ice (1,K) produced a highly specular signal highest value seen in the data set (37 dB just before of about 20 dB. In these regions the return signal the first-year ice is encountered at the right of the rose and fell within a few range gates, and there area). Tile waveforms are also the most specular was relatively little energy in the later gates. The of any of the data shown. The gray ice on the signal froin the light nilas (J) was almost 30 dB. right side of the top gegilient (T) produced a signal The radar signal fell abruptly when the first-year ice level of about 25 dB, and the narrow bands of dark was encountered (L) and thereafter maintained all nilas (U,V) in the matrix of flows beside it have average level of about 9 dB. The return waveforms backscattered signal levels of about 30 dB. in this region generally exhibited both specular and The bottom segment in Fig. 6 demonstrates just diffuse scattering components. ]low small a break in the first-year ice can cause a The bottom of Fig. 5 indicates that localized areas significant response in the radar signal level. The of old rafting (M,R) can produce signal levels in the 5- very small fracture at (W) is only about 3 in wide, dB range. It also shows that visually similar smooth yet it produces a peak response of 18 dB, about 10 areas of first-year ice could produce significantly dB higher than the surrounding area. The narrow different average signal levels, for example, 20 dB area of (lark Dilas at (X) produces a signal level just at (N) and 10 dB at (Q). The recent finger rafting as high as the broad area at (S) because part, of it is at in evidence just past (M) on the lower half of the nadir. But the light nilas and gray ice at (Z) produce photography suggests that, the first-year ice on the no response becallse they do not quite extend to the left side of this segment may be thinner than the ice radar nadir point. As a result, the signal level at (Z) on the right side. is 15 dB lower than at (X). As is the case throughout, 1707 @Rl 'MT U`,'@'W s @7 pt fy. "P, R", PEAK SIGIiAL LEVEL (CIB) 12 S T V@ @Y 1, k 1 1, r,;J Y Z W X Figure 6. Two additional noncontiguous data segments in the same format as Fig. 4. the radar signal level is depressed to about 5 d13 output waveforms. The major tick marks along whenever a region of significant ridging such as (Y) this axis are every ten waveforms and represent the is encountered. averaging interval utilized in Figs. 4, 5, and 6. The The ten-output-waveform averages displayed in axis labeled GATE is the radar range in the return Figs. - 4, 5, and 6 give a very good impression of waveform (0.41-7-ni interval). The display has 60 the composition of the backscattered signal as far gate positions along this axis because the prograrn as diffuse and specular scatter is concerned. But used to generate these three-dimensional plots was because they are averages over 1.09 s,. the aircraft orginally developed to analyze SEASAT altimeter has moved approximately 83 in and the response to data. The AAFE altimeter waveforn-is occupy the some very localized phenomena would be smoothed central 24 gates in the display. over. Figs. 7 and 8 present every output waveform The return waveforms are plotted twice (Figs. 7-8). for four short segments of the data already discussed Below the photography, range increases toward the so the altimeter response at a 0.109-s rate can be upper left. In plots of the same waveforms above the observed. Since the aircraft moved only 8.3 in during photography range decreases toward the upper left. this interval, there would lie very little sn-toothing of This allows one to view the backs of the waveforms, the response, even for highly specular returns. Even which are hidden in the bottom plot. It is similar to at this highest output rate of the AAFE altimeter, holding a mirror behind a t] iree- dirn ensio nal ruodel the returns are the average of 52 return pulses. The to look at the rear face. As in the case of Figs. standard deviation caused by Rayleigh fading in the 4 through 6, the width of the photography is the radar signal would be only 14% of the mean value at 250 in cross-track distance nominally interrogated each poiijt on the return waveform. by the 12 range gates of the AAFE altimeter past I-'he top segment of Fig. 7 shows the region near the range to the nadir point (Fig. 2). This segment (B) in Fig. 4. The waveforms are plotted to provide demonstrates that the off-nadir scattering in the last a three-dimensional effect, The vertical axis is linear AAFE altimeter gate decreased by only a factor of 2 in returned power to give an imniediate impression over the shear zone relative to the region outside the of the relative strength of the returns. The axis zone. However, the near nadir scattering was greatly el parallel to the aircraft ground track can be thought reduced. of as either the distance or the time interval between The bottorn segment of Fig. 7 shows the region 1708 Figure 7. Three-dimensional time plots of wave- forms output by the altimeter at 0.109-s intervals. Below the corresponding photography the waveforms are plotted with range increasing to the upper left. The same waveforms are plotted above the photogra- phy with range decreasing to the upper left to present a view of the back of the waveforms otherwise hidden in the lower plot. These wavefornis are from the regions near (B) in Fig. 4 (Top) and (M) in Fig. 5 (Bottom). "A A ""'?'@ ..... ... . . . ..... . .... . near (M) in Fig. 5. Note that the peak wave r value in this segment is 400, 16 times the power o peak return in the top segment. The returns ha e much more specular appearance here, although the is still a noticeable amount of energy on the rear fac of even the high-amplitude returns near the end o the segment. inost). The aircraft ground speed was the same The top segment of Fig. 8 encompasses the (N)- at the noininal value of the aircraft airspeed, and the photography over the 1-min interval surrounding through-(R) region of Fig. 5. The peak value is this segment indicated that the aircraft had virtually 2,000, 5 times higher than the peak of the bottorn no drift angle. This combination indicates that the segment of Fig. 7. The returns are highly specular wind speed at the aircraft altitude was negligible, and and.show aln-tost no indication of off-nadir scattering. suggests that it was also negligible at the stirface. The bottom segment of Fig. , 8 shows the region The structure apparent in the photography indicates near (S) in Fig. 6. It contains the highest amplitude that the highly specular region contains dark nilas return waveform seen in the data set (5,000). The (up to 5 cin thick), but the possibility that there was high- amplit tide returns in this segment generally also some very smooth open water cannot be ruled rise and fall within two range gates (three at the out. 1709 5f, R $ Figure 8. Three-dimensional time plots as in Fig. 7, but from the regions near (P) in Fig. 5 (Top) and (S) in Fig. 6 (Bottom). CONCLUSIONS The AAFE altimeter has demonstrated that the Acknowledgments. We thank Konrad Steffen and Edgeworth return waveform shape and signal level of an airborne Westwater for reading and commenting on the manuscript. A spe- pulse- limited altimeter at 13.9 G1lz respond to sea cial thanks to Konrad Steffen for his assistance in interpreting the ice type. The signal level responded dramatically to photography. One of us (LSF) received partial support from the even a very small fracture in the ice as long as it Office of Naval Research under contract #NR 307-536/3-22-85. occurred directly at the altimeter nadir point. Shear zones and regions of significant compression ridging RFFURENCES consistently produced low signal levels. The return Hughes Aircraft Company, Ground Systems Group, "Final Report of wavefornis frequently evidenced the characteristics the Advanced Application Flight Experiment Breadboard Pulse of both specular and diffuse scattering, and there Compression Radar Altimeter Program," NASA CR-141411, Au- was an indication that the power backscattered at gust 1976. Steffen, K., "Atlas of the Sea Ice Types, Deformation Processes, and 3* off-nadir in a shear zone was actually somewhat Openings in the Ice," North Water Project, Ziircher higher than that from nadir. Geographijehe Schriflen, Heft 20, Ziirich 1986. 1710 SEA GRANT FACES OCEANS OF PLASTIC Xanthippe Augerot Washington Sea Grant 3716 Brooklyn Avenue N.E. Seattle, Washington 98105 ABSTRACT The National Sea Grant College Program The specific impacts of plastic debris on (NSGCP) is well-suited to serve as a partner fish, turtles, marine mammals, and birds in marine plastic debris education and have not been quantified, nor have the problem-solving due to its unique economic impacts on fishermen and coastal character. A non-regulatory, university- communities. Information from beach based organization, it can reach segments clean-ups and research tell us that the of the population which national campaigns sources and composition of plastic trash cannot. Sea Grant's greatest strengths are are very different around the country. the ability to nurture innovative Fishing vessels are the largest source in the educational efforts and hand them on to North Pacific, while along the Gulf of local groups, its experienced corps of Mexico and the north Atlantic seaboard, marine advisory agents, and a built-in the major culprits are commercial vessel national network. Marine debris is not a traffic and non-point shore-based sources, key area on Sea Grant's research agenda, respectively. with the possible exception of aiding applied research into the usefulness and MARPOL Annex V and the Marine Plastic effects of alternative materials in the Pollution Research and Control Act of 1987 marine environment. (MPPRCA), which prohibit the disposal of plastics at sea, will begin to address the overall problem by minimizing vessel INTRODUCTION sources of plastic trash. However, as recognized in the Interagency Task Force The advantages of plastic products are well Report on Persistent Marine Debris, which known. They are durable, inexpensive, builds significantly on a specially designed and versatile. It is almost inconceivable Sea Grant workshopl and working papers, that any marine enterprise could survive enforcement at sea is difficult and costly; in the modern world without the benefits therefore, the best approach to reducing of this ubiquitous material. plastic trash impacts is education to change public behavior.2 But plastic has a darker side in the marine environment. Six-pack yokes, strapping bands, and lost fishing gear, when dropped The National Sea Grant College Program has into the ocean, may disable fishing vessels been aware of the issue since 1984, when and entangle, starve or maim wildlife. On the Pacific Sea Grant College Program the beaches, plastic trash has the same (PSGCP, composed of Sea Grant college potential for harming wildlife and is, in programs in states bordering the Pacific addition an aesthetic eyesore that may Ocean) helped support an international harm a local tourist industry. workshop in Honolulu on the fate and CH2585-8188/0000- 1711 $1 @1988 IEEE effect of marine debris. More specifically, researchers and the users of the marine PSGCP was responsible for publishing the environment. They serve both as advisors proceedings, which have become the basic to marine industries and as educators reference document on this subject.3 promoting environmental stewardship. Texas and Oregon Sea Grants became involved in specific local efforts to tackle Not affiliated with any regulatory agency the problem, and as awareness of the issue or single constitutency, advisory agents spread, other Sea Grant programs actively seek to remain neutral and incorporated the concern about plastic provide information upon which the public debris into their ongoing education and can make better decisions. The agents also advisory efforts.4 Most of these efforts work closely with Sea Grant's extensive have been undertaken cooperatively with Communications network, which provides a the NMFS Marine Entanglement Research broad publications and outreach capability, Program (MERP), state and local considerably extending the educational governments and private organizations effort. Their unique position enables Sea such as the Center for Environmental Grant to extend educational programs to Education (CEE). marine user groups not easily reached by national campaigns. WHAT IS SEA GRANT? Several Sea Grant programs have The National Sea Grant College Program is a developed expertise regarding this issue, cooperative research, education and which they try to share nationally advisory network supported by funds from throughout the network. The first and the National Oceanic and Atmospheric most concerted marine debris education Administration (NOAA) together with projects began at the Texas and Oregon Sea matching public and private funds from Grant programs, and both demonstrate the the participating states. It comprises 30 slow-growing but potent networking effect programs involving more than 300 common to Sea Grant nationally. colleges, universities, and affiliated institutions in the coastal and Great Lakes The Newport, Oregon marine debris states and territories. Created by an act of disposal project, funded by NMFS, Congress in 1966 and modeled in part after developed from contact between Oregon the Land Grant College system, the mission Sea Grant and MERP. At MERP's request, of the Sea Grant Program is to foster the Oregon Sea Grant wrote a proposal for wise use, development, and conservation of funding, and helped to find and train the marine resources. Sea Grant programs use manager who so successfully carried out three tools to carry out their mission - the project. The Port of Newport, MERP, research, education and and the marine Oregon Sea Grant and the fishing advisory service. community succeeded in providing more accessible waste handling facilities and a recycling program which actually lowered WHY SEA GRANT? port waste handling costs, while simultaneously involving the entire Sea Grant's marine plastic debris education community and creating a new attitude of program has been carried out primarily by stewardship toward the ocean.5 marine advisory services. Patterned after the Land Grant system's Extension Service, In Texas, the successful and nationally the marine advisory service is a network visible Adopt-A-Beach program was of marine agents and specialists who work developed by the state based on marine with many varied marine activities, such advisory agent Charles Moss's active beach as seafood processing, commercial and adoption program in Brazoria county. One recreational fisheries, port management, of Sea Grant's strengths is that it is able to aquaculture, tourism, and waterfront nurture new ideas and fund innovative development. Housed at Sea Grant programs which can then be taken over by universities, the field agents and specialists community activists. These programs tend serve as a bridge between the academic 1712 to be more successful in the long run, since activity has had similar impacts. Members they are locally "owned" and controlled. of our staff have been actively involved with the Washington State Marine Plastic Most of the 30 state Sea Grant programs Debris Task Force, convened by state are now incorporating marine debris Public Lands Commissioner Brian Boyle in educational materials in their in their January of 1988. Task Force members advisory and education efforts. They have (including representatives of state written newsletter articles and issued agencies, community groups, legislative press releases informing the public how to staff, educational institutions, public and make a difference. Several programs have private sector organizations) developed a developed posters warning about the report for the Commissioner, outlining how hazards of marine debris and have the state should go about increasing public cooperated with state agencies to coin sensitivity to the marine debris issue, catchy slogans as well as to run public enforcing the MPPRCA and mitigating service announcements in the media. plastic debris problems in a coordinated and cost-effective manner. Washington THE EDUCATION CAMPAIGN Sea Grant may play a larger role in the state's response to this issue if it is The list of major Sea Grant publications on designated the state's clearinghouse for marine plastic debris is short: the Texas marine plastic debris information.10 Shores 6 summer 1987 issue on marine plastic trash, the Oceans of Plastic Other Sea Grant activities include a port workshop proceedings,7 Plastic in the debris disposal project in Bellingham, Ocean:What are we doing to clean it Up? 8 Washington, similar to the MERP-funded and a condensed version of the project in Newport, Oregon, but different in Interagency Task Force Report .9 However, a number of ways: the Sea Grant agents they are distributed widely and have a are acting in an advisory capacity to the slow but profound effect as network Port of Bellingham, whereas in Newport builders, helping people in small the project manager was a port employee; communities to share information and the fishing fleet comprises primarily minimize duplication of efforts. Some of seiners and gillnetters, not trawlers; and these publications are being used and there was more prior knowledge and reused, either in full or in part, with interest among fishing community leaders adaptations for specific local audiences. in Newport than in Bellingham. Workshops are another form of strategic Gear type makes a significant difference. talent-sharing. The Oceans of Plastic Whereas the Newport fleet of trawlers fish workshop in February, 1988, brought in Alaska as well as near home, and often together experts to discuss aspects of make extended fishing trips, the fisheries-related marine debris problems. Washington fleet of seiners and gillnetters Aside from producing a report about the are primarily overnigbters or day importance of educational efforts vis-a-vis fishermen. They do not need to do repairs increasing enforcement mechanisms, the at sea, and their gear is too cumbersome to workshop generated many personal ties do so. Wildlife entanglement problems are and several spin-off effects. Marine debris not so visible or acute here as in Alaska, disposal projects will be established at two and fishermen have an antagonistic Texas ports in the coming year (with relationship with harbor seals due to their Saltonstall-Kennedy funds), and the state poaching of catch. For all of these reasons, of Alaska has increased its activity with formulating an effective educational regard to its small and isolated fishing campaign is more complex in Bellingham, ports. Topics for future workshops might as the fishermen perceive themselves to be include alternative fishing gear less of a problem and have little self- construction, or small port and terminal motivation to change. Funded partially by marine debris reception facilities. a Puget Sound Water Quality Authority Another recent Washington Sea Grant (PSWQA) public involvement and 1713 education grant, this project is part of a T11E BROADER CONTEXT larger effort to educate the citizens of Puget Sound about. water quality The marine plastic debris issue is strategic management. in that it provides an easy opening to discuss larger and, in the long term, more Sea Grant is also helping to produce significant issues, such as deteriorating educational videos about marine plastic coastal and estuarine water quality and the debris. Alaska Sea Grant has cooperated national solid waste disposal crisis. The with Saltwater Productions to produce a public relates immediately to images of fur MERP-funded video, "Trashing the Oceans", seals entangled in trawl gear, or to sea for general audiences. They plan to turtles choking on plastic bags. Attention collaborate again in the future to produce a may then be drawn to the larger issues video targeted at Alaska fishermen. inherent in any discussion of marine plastic debris. In this sense, marine plastic Some Sea Grant programs are very debris is a key piece of any advisory involved with local and state beach clean- educational program that attempts to ups. For example, Georgia, North Carolina, promote the wise use and conservation of and Louisiana are compiling databases on our natural resources. the nature and origins of the plastic debris found during beach sweeps. This information should help focus educational A RESEARCH ROLE? programs more closely on the primary sources of the trash. It is also useful for The Interagency Task Force recommended legislative testimony when legislators several broad areas for future research: attempt to make new policies regarding fish and wildlife impacts; economic marine debris.11 impacts; improved knowledge of the contribution of sea-based versus shore- Although the port project in Newport, based debris sources; and cooperation with Oregon is no longer receiving NMFS the fishing and gear manufacturing funding, the port is continuing the debris industries to reduce gear loss, improve disposal and recycling program and Oregon gear recovery, and develop recycling Sea Grant has picked up the education and opportunities for used gear. The Task advisory role. The educational materials Force also recommended research on developed during the course of the project degradability for specific plastic products. are available through Oregon Sea Grant, These research subjects are by and large and they continue to publicize the efforts best addressed as augmentations to of individual fishermen who haul their ongoing work by NOAA, the U.S. Fish and trash to shore. In future educational Wildlife Service, state and local materials, Oregon Sea Grant will target governments, and the plastic industry charterboat operators and recreational itself. Very specialized academic research fishermen. organizations have evolved which deal with the more technical research issues of The Saltonstall-Kennedy (S-K) grant plastic recycling and degradability.13 program is funding another outgrowth of Sea Grant may have a role to play if the the Newport project. Former project plastics or fishing gear industries are able manager Fran Recht, based at the Pacific to develop alternative materials or gear Marine Fisheries Commission, will be configurations. Sea Grant could then forge serving as roving advisor to small ports on the link between users and industry to test the West coast as they come into compliance with the Coast Guard's new applications in the environment, while regulations implementing the MPPRCA. relating this work to existing resource Her expertise will be shared by a larger management research. audience both in this manner, and through a guide published by NMEFS.12 1714 CONCLUSION A @For a good description of the initial Sea Grant debris programs and ongoing efforts The major activities that Sea Grant will throughout the Sea Grant network, see emphasize over the next few years with Guthrie, D., "Castoff Plastic Debris," Oceanus respect to marine plastic debris are to: 31(3):pp 29-36. a) Continue networking, among Sea 5For more information about the Newport Grant programs, user groups, and Marine Refuse Disposal Project, see Recht, other interested bodies. F., Report on a port-based project to reduce b) Target public education with marine debris. NOAA-NWAFC Processed publications, workshops and problem Report 88-13, July 1988:75 pp. solving activities. 6Texas Sea Grant College Program. Marine c) Identify problem areas where Litter. Texas Shores Summer 1987:28pp. additional policy development is 7AIaska Sea Grant. Oceans of Plastic; needed. Workshop Proceedings (forthcoming). 8Augerot, X. Plastic in the ocean; what are No single organization can carry the entire we doing to clean it up? WSG-AS-88-6, marine plastic debris educational effort. 1988:8pp. As evidenced by ongoing projects, many 91nteragency Task Force on Persistent different organizations contribute funds, Marine Debris. Summary of Report to ideas and expertise. By working together, Congress. Alaska Sea Grant (forthcoming). projects that start small are able to reach 1OWashington State Marine Plastic Debris much broader audiences and have a Task Force. Marine Plastic Debris Action greater effect. We must continue to Plan for Washington State. October expand the educational network-locally, 1988:65 pp. regionally and nationally-in order to IlDearborn, R.K., Director, Alaska Sea increase our reach yet further. And we Grant, Statement regarding marine plastic must also continue to emphasize the debris, on behalf of the Sea Grant marine plastic debris problem in the Association and Liffman,M.M., Assistant context of broader water quality, resource Director, Louisiana Sea Grant, Testimony management and environmental concerns. concerning marine plastic debris, before the National Ocean Policy Study, U.S. Senate I would like to thank Mr. Ron Dearborn, Committee on Commerce, Science, and Director, Alaska Sea Grant for the thoughts Transportation, July 29, 1988. and ideas he shared with me in the 12Recht, F.. Dealing with Annex V - a re rence guide for ports. NMFS MERP preparation of this paper. . fe 1988. 13Michigan State University's School of RENCES Packaging, Rutgers Center for Plastics Recycling Research and the Stevens 10ceans of Plastic; A Workshop on Institute of Technology's Polymer Fisheries Generated Marine Debris and Processing Institute, for example. Derelict Fishing Gear, Portland, Oregon, February 9-11, 1988. 2Report of the Interagency Task Force on Persistent Marine Debris, printed by the Department of Commerce, National Oceanic and Atmospheric Administration, 1988, p. 141. 3Shomura, R.S. and H.O. Yoshida (ed.s), Proceedings of the Workshop on the Fate and Effect of Marine Debris. NOAA TM- NMFS-SVvTC-54 and UNIHI-SEAGRANT-CR- 85-04, 1985:580 pp. 1715 ANALYTICAL CHEMISTRY OF BUTYLTINS K.W. Michael Siu, James W. McLaren, Paulette S. Maxwell, Graeme J. Gardner, Shier S. Berman The common analytical technique for butyltins involves extracting the species as halides, derivatizing to hydrides or tetraalkyltins, separating by using gas chrom- atography, and quantitating with flame photometry or atomic absorption spectrometry. To arrive at certified concen- trations of butyltins in standard reference materials (SRMs), it is, however, imperative to have results from at least two independent analytical techniques. In order to satisfy this requirement, analytical techniques have been developed for the determination of butyltins in a harbour sediment SRM, PACS-1. These are: (1) Gas chromatography - flame photometric detection (GC-FPD) of butyltins as chlorides, (2) High performance liquid chromatography - inductively coupled plasma mass spectrometry (HPLC-ICPMS), and (3) Ion spray mass spectrometry / mass spectrometry (ISMSIMS). In GC-FPD, butyltins are extracted from the sediment as chlorides, and separated and quantitated with gas chromatography as such. Two extractants may be used: toluene/iso-butyl acetate/tropolone and hexane/iso- butyl acetate. Extracted butyltins are separated by cation exchange in HPLC-ICPMS using citrate buffers and determined on-line with ICPMS. ISMSIMS is only applicable to tributyltin (TBT) determination. The TBT ion is evaporated from the extract by ion spraying, separated from matrix ions by the first MS stage (no chromatography is necessary) , and quantified by using the second MS stage with selected reaction monitoring. 1716 United States Government work not protected by copyright 'The U.S. Coast Guard: A Prototype for National and International Ocean Policy Duplerwntation Presented by Edward-W. Cannon Chief, Governmental Affairs Staff U.S. Coast Guard and Executive Director Oceans '88 Conference and Exposition Baltimore, Maryland October 31, 1988 p of Marine Interests A Partnershl A Partnership of Marine Interests 4t9c -0 C Figure Figure 2 INTRODUCTION and THE RELEVANCE OF THE OCEANS 188 THEME In the tandem roles as Chief, Governmental Affairs Staff, U.S. Coast Guard, and the Executive Director of the Oceans '88 Conference and Exposition, a managerial responsibility I was honored to undertake, there are many policy, programmatic and operational issues and perspectives I could convey with respect to the U.S. Coast Guard; to what has been called 'ocean policy'; or to their combined relevance to Oceans '88 and its theme of consensus construction within the most comprehensive paraneters which might yet be postulated for a major policy and technical forum. 1717 United States Government work not protected by copyright There is an intended correlation between the theme of the Oceans ?88 Conference and Exposition, 'A Partnership of Marine Interests' and the thesis that the U.S. Coast Guard represents a unique prototype for the cost-effective implementation of both national and U.S. foreign policy 'goals. There is an additional relationship between the overall conference theme and this paper's later recommendation that the U. S. Coast Guard, as well as other Federal agencies, can be more responsive to their multiple constituencies, and reciprocally assure a better stake in Federal budgetary allocations, if these agencies can adopt what I call the 'External Affairs' model. Indeed both the logos precisely designed for Oceans '88 are depicted to convey these themes; i.e " the Marine Partnership, and the notion of consensus involving all interested players in the oceans policy/ocean technology ballgame; the reality that relevant structures are helpful to comprehend and pursue individual and joint, Federal interagency and intergovernmental strategies; and the necessity to assure more harmonious conflict use resolution. The themes are also crafted to assist in rallying the support necessary to attract national attention to the national, inland marine, coastal, oceanic and related atmospheric needs of this nation. In commercial chatter, it's called institutional advertising for long-term, market retention. If you want attention, utilize the most sophisticated, creative and professional ideas which can be marshalled to support your cause. The single droplet of water illustrated in Figure #1, and suspended in the water column radiating photosynthesis, suggests the sc ient if ic-atmo spheric integration. The mosaic of ocean uses within the droplet signifies both harmony and diversity, peaceful conflict resolution, and the premise that any one ocean use or concern is, in an Aristotolian sense, as significant as the others for their respective constituencies. The 'octagon' logo illustrated in Figure #2 is designed as a more permanent symbol to represent all the categories of the participants in Oceans '88, as well as all of the participants involved in any modern nation-state's policy formulation process. Those participants include what I call the '8 Estates of Government': Congress; the Executive Branch; the Judiciary; the Trade Media, historically called the 4th Estate of Government; a combine of Industry and Interest Groups as the Fifth Estate, an aggregate of Academic Institutions and Professional (or Disciplinary-Oriented) Institutions as the Sixth Estate)* State and Local Governments as the Seventh Estate; and the International Community as the 'Eighth Estate' involved in any individual, nation-states' policy formulation. In a larger public policy context, it can be predicated that a local government initiative such as California's Proposition #13, bilateral agreements such as the recent US-Canada Trade Agreement, or the role of the media, the 'Fourth Estate' in public diplomacy, can have profound consequences on national ocean policy formulation. The private sector, public and special interest groups, academic entities, state and local governments and indeed international representatives, all lobby before the Congress; and, the general and trade media print and electronic, in turn report upon those interactions and influence both the 'inside the Beltway' and 'outside the Beltway' citizens, voters and decision-makers. Unless a federal agency, as well as Governmental Affairs components of corporations, national associations and other institutions, comprehend this precise networking, as well as the prescriptions in the U.S. Code regulating lobbying activity, their efforts at advancing from the $reactive mode' to the 'proactive mode' can only be marginal; and, in some cases their obsolete organizational styles can be counterproductive. Public policy strategies cannot be optimized without rational and constituent-responsive structures. 1718 THE U.S. COAST GUARD AS A PROTOTYPE So much for a primer on the symbols of Oceans '83; now, to their symbiosis with the U.S. Coast Guard itself, and how it has been utilized, and can further be refined, as a remarkable prototype for the National interest. The Coast Guard is a multi-faceted model in which its delegated missions require the agency to be an active participant in a plethora of interagency.9 intergovernmental and international programs, events, task forces, conferences and proceedings, ranging from the National Narcotics Border Interdiction System to the UN/IMO, to the Federal Ocean Principals Group, to the American Oceanic Organization, to EXPO '86 in Vancouver, to the Marine Technology Society and the Chairmanship of Oceans 188. My window or personal perspective of the U.S. Coast Guard is as a unique, and traditional, yet dynamic, Federal agency and Armed Service, implementing U.S. domestic law as well as international covenants, in the inland waterways, coastal zones, and high seas as well as in a record of visits to other national ports of call. It is an agency which not only presents a unique 'peace and defense model' similar in texture perhaps to NATO's Committee on the Challenges of Modern Society, but one which is committed to both national needs and international interchange ... one which, as I hope to describe, has operated quietly yet efficiently across the intergovernmental spectrum, across the 'Eight Estates of Government' ... as will be elaborated. Thus, the U.S. Coast Guard: 'A Prototype for National and International Ocean Policy Tnplementation'. Thus the utility of the 'Partnership of Marine Interests' and 'Octagon' logos to fit the needs of both the Coast Guard and Oceans '88. The U.S. Coast Guard has been continually involved in the process of charting both contemporary and emerging ocean-related needs since its origin as the Revenue Cutter Service in 1790 when Secretary of the Treasury Hamilton petitioned the Congress for a fleet of cutters to assure import revenues for the infant Nation. The agency was established in name in 1915 when it was merged with the Life Saving Service. It became part of the newly f ormed Department of Transportation in 1967. Its current and 18th Commandant is Admiral Paul A. Yost who assumed command in May 1986. The U.S. Cost Guard, is the nation's historic maritime law enforcement agency and has a major operational presence, with a myraid of functions in the coastal, offshore and high seas as well as inland waterway areas. Many of those missions and activities are familiar to many of the attendees at Oceans '88 and the readers of its formal Proceedings. Thus, they will only be briefly enumerated and described. The Coast Guard is an agency which maintains traditional safety-at-sea and national defense functions, yet continually evolves in response to national and international needs. Much of the direction of its current missions, programs and activities emanate from the last 16 years of both Legislatiave and Executive initiatives which have included the following: � National Environmental Policy Act (1970) � Water Quality Improvement Act (1970) � Port and Waterways Safety Act (1970) � Coastal Zone Management Act (1970) � Federal Boat Safety Act (1971) � Marine Protection, Research, and Sanctuaries Act (1972) � Amendments to the Oil Pollution Act (1973) � Fisheries Conservation and Management Act (1976) o Executive Initiative on Oil Pollution (1973) 1719 � Executive Initiative on Drug Law Enforcement (1973) � Clean Water Act of 1977 (Amendments to Federal Water Pollution Control Act) � Presidential Order on Interdiction of Illegal Migrants (1981) � Vice Presidential Task Force on South Florida Crime (1982) � National Narcotics Border Interdiction System (1983) � Presidential Proclamation for a 200-mile Exclusive Economic Zone (1983) � Maritime Defense Zone enactment (1984) ... and the � Anti-Drug Abuse Act (1986) The synergistic effect has been the expansion of some traditional missions and entry into new areas such as Marine Environmental Protection and working closely with the states and jurisdictions in a Joint education and enforcement effort for its Recreational, Boating Safety program. Innovations in development of the. 'Pollution Strike Force' concept, have become internationally known. These newer responsibilities, combined with its historic roles, have led the service to structure, as modern management systems require, a statement of strategic objectives; which includes the responsibilities: To minimize the loss of life, personal injuries, and property damage on, over, and under the seas and waters subject to U.S. jurisdiction. 'Help Others - Help Yourself' is one recruiting message which attracts young men and women to this humanitarian role; To facilitate waterborne activity in support of national economic, scientific, defense, and social needs; To maintain an armed force prepared to carry out specific naval or military tasks in time of war or emergency; the Coast Guard, is, as another recruiting ad conveys, 'An Armed Service and More;' To assure safe and secure ports, waterways, and shoreside facilities; To enforce federal laws and international agreements on and under those waters subject to U.S. jurisdiction and on and under high seas where authorized; To maintain or improve the quality of the marine environment; and To cooperate with other governmental agencies - Federal, state and local - so as to assure efficient use of public resources. This latter responsibility, as well as Executive mandates such as the Presidential Task Force on Private Sector Initiatives, established to link Federal with intergovernmental resources in the pursuit of the National interest, provide the rationale for the Coast Guard Chairmanship of Oceans '88. The Executive initiative for an Exclusive Economic Zone (Presidential Proclamation #5050 of March 10, 1983) represented an expansion of U.S. national resource management over a spacial area of some 3.4 million miles, estimated to be over one and one-half the total land mass in the United States. And, although this act essentially recognized both the judicial framework and legality of ongoing, offshore activities evolved as the result of customary international law, and has in the short term primarily required major research attention by both NOAA and the Department of the Interior, the U.S. Coast Guard must also consider in its planning whatever private sector, public sector, interest group or academic forecasts will be projected for offshore resource development within that zone The agency must factor these inputs into what I call a 'Long-Term Perspective' so 'event management' can be related cohesively to 'Trend Management'; so reaction can be reconciled with proaction to assure that the proper level of resources will be allocated; and to also guarantee that the Coast Guard's multiple constituencies are adequateky and cost-effectively-served. 1720 However, for the 'foreseeable future, a major portion of the Coast Guard's resources must be committed to the national effort to reduce the supply side of the narcotics equation through sea-air interdiction. Yet, in the current mix of -priorities, three mission areas will remain paramount: Maritime Law Enforcement, Maritime Safety, and Defense Readiness. As activity is generated from increased hydrocarbon expoloration and extraction, ocean transportation, fisheries development, offshore mining, specific Arctic offshore and onshore development, recreational boating, disposal of oil and other hazardous substances, such as plastic harnesses and toxic medical waste, and the possible scenarios for waterborne terrorism, the Coast Guard's role and mission base must be expanded. Its statutory responsibilities have evolved in terms of inland marine and coastal as well as high seas or 'trans-oceanic' functions beyond the jurisdictional 'shores of the United States ... and they extend beyond the economic factoring to include general law enforcement, safety-at-sea, environmental protection and national defense activities. The specific, current programs as derived from law and Executive direction, currently include: Search and Rescue ... in which the agency must maintain a 24-hour, 'real time' operational infrastructure of small boats, cutters, aircraft, communications facilities and Operations Cent -ers ... and in which under international agreement, it provides the search and rescue coordination for 21 million square miles of ocean ... that's already a sizeable segment of both the EEZ's seawater and the high seas to supervise. Unquestionably, as resource limitations are imposed upon the public sector, another significant element or question which has arisen is the issue of what degree of current governmental activity can be transferred, at reduced cost and without loss in the threshold of public safety, to the private sector. In addition, to its ongoing extensive Office of Management and Budget A-76 reviews, the U.S. Coast Guard issued a classification of non-emergency towing policy in June 1983. It is proceeding with a new system in which assistance cases, which are clearly of a non-emergency nature, are transferred to the private sector via local lists of available towers and salvors. However, the infrastructure for response to both inshore and offshore emergency cases must still be maintained; and, as both distant offshore and Arctic activity expand .. given the time, weather and expense variables, it remains highly uncertain what degree of private sector enterprise would be available for those locales. There is a threshold of public response which must be maintained. The strategy is for public and private sectors to work together and more creatively in the national and foreign policy interest. Dearch and Rescue, the most publicly known mission of the Coast Guard, ranges from removal of hazards to navigation to flood relief in tranquil waters, to firefighting, to response to offshore boaters (72% of the life saving missions involve recreational boats and almost 95% of all rescue missions take place within 20 miles of shore), to high seas rescues, to helo/medevac. operations ... all requiring skilled seamanship and exposure to mariner hazards. And passenger vessels are included as well. Search and Rescue is an international obligation. In I-larch 1987 the Coast Guard rescued the merchant seaman of the Soviet disabled vessel, the Komsomolets Kirgizii, foundering in the Atlantic. In July 1986, three Coast Guard cutters (Farallon, Mau, Cape Gull) and three Dolphin choppers helped extinguish fires and airlifted passengers when an explosion occurred on the cruise ship SS E24ERALD SEA offshore a resort island in the Bahamas. From 1979 to 1984, the Coast Guard annually averaged 6000 lives saved and over a billion dollars in property loss prevention. 1721 In a more systemized format for international safety-at-sea, the Coast Guard operates the AMVER program, or Automated Mutual Assistance Vessel Rescue System, from its Third Coast Guard District Headquarters on Governors Island in New York. Participating flag merchant vessels send both sail plans and periodic position reports to the AMVER computer. In an emergency, a recognized Rescue Center of any nation may obtain a computer-predicted list of ships in the vicinity of an emergency. An AMVER mock-up has been displayed at the Oceans '88 Exhibit. Enforcement of Laws and Treaties ... In the year 1990, the U. S. Coast Guard will celebrate its bi-centennial of service as the primary Federal agency for maritime law enforcement ... which includes not just the enforcement of those domestic laws within the 200-mile Fisheries Conservation and Management Act, but other domestic laws and international agreements extending to interdiction of drugs, other controlled substances and contraband, waterborne movements of illegal migrants, and a variant of laws relating to protection of endangered species, marine mammals and marine sanctuaries. In this effort the service participates in a unique, intergovernmental coordination within the National Narcotics Border Interdiction System, comprising the Coast Guard, Customs Service, Drug Enforcement Administration, U.S. Navy and at times up to 27 different, local, state, and Federal agencies, depending on the specific mission. The effort is part of a multi-faceted, national program reaching from crop substitution to crop eradication ... to seizure at or prior to exportation ... to land, air and sea interdiction ... to education and elimination of drugs in the workplace and school ... to continuing drug research ... to stricter penalties for drug abuse ... to increased international cooperation as well as cooperation between the public and private sectors to reduce consumption. The U.S. Coast Guard has primary responsibility for the at-sea enforcement of over 20 different fishery management plans and associated enforcement regulations which regulate domestic and foreign fishing in the U.S. EEZ pursuant to the Magnuson Fisheries Conservation and Management Act. In addition, the U.S. 'Coast Guard is tasked with monitoring compliance with endangered species and marine sanctuary regulations, as well as certain international fishing agreements in waters beyond the limits of the EEZ,. Fisheries enforcement is also a close interagency effort; concurrent jurisdiction is shared with the National Oceanic and Atmospheric Administration of the U.S. Department of Commerce for the enforcement of the Magnuson Act provisions. The fourth significant program area is Port and Environmental Safety and Security. In this decentralized effort, under the Ports and Waterways Safety Act and earlier legislation, Coast Guard officers acting as Captains of the Port (COTP's) in 47 port cities, are authorized to enforce rules and regulations governing the safety and security of ports and the anchorage and movement of vessels in the U.S. waters. Port safety and security functions include supervising cargo transfer operations, both storage and stowage, conducting harbor patrols and waterfront facility inspections, establishing security zones as required, and the control of vessel movements. In the reality of potential for terrorism, this role is also evolving as evidenced by the large fleet of small patrol boats on duty in 1984 at Olympic Village along Long Beach, California. Merchant Marine Safety ... in which the U.S. Coast Guard assures the safety of e nation's ports and inspected merchant vessel fleet. Vessel safety standards are established for construction, maintenance and repair, machinery, electrical systems, lifesaving equipment and the manning of vessels. .1722 In addition the service must also review and approve plans f or vessel construction, repair and alteration; establish merchant marine qualification standards and investigate maritime casualties and incidents involving merchant marine personnel. To assist the Coast Guard in meeting these responsibilities, memoranda of understanding are now in effect with the American Bureau of Shipping, the National Cargo Bureau, Underwriters Laboratories and other related certification organizations. International obligations for merchant marine safety, the related environmental safeguards, and the allied responses to waterborne terrorism are standardized through the United Nations' Intergovernmental Maritime Organization in London. Ice Operations ... in which the U.S. Coast Guard maintains the sole Federal capability for icebreaking in both domestic waters and polar regions. In its domestic ice operations, certainly the increased focus on Alaskan/Arctic resource development will continue to increase in importance. In another interagency effort, the agency provides assistance to ships and assists the U.S. Army Corps of Engineers in the prevention, mitigation or relief of flooding caused by ice accumulation in waterways. This opening involves ice-blocked waters in the Northwestern United States, the Great Lakes - St. Lawrence Seaway System, the Upper Mississippi River Basin and Sub-arctic Alaska. In addition to maintenance of a domestic fleet of buoy tenders and tugs with icebreaking capabilities, the U.S. Coast *Guard operates polar icebreakers to support resupply for the Department of Defense in the Arctic and to support the National Science Foundation and other Federal interests for resupply and scientific investigation in the Antarctic. The icebreakers must meet the personnel and equipment standards to meet those continuing treaty obligations. Yet, the icebreaker fleet has been greatly reduced; and, as this writer submits, only a coordinated, 'External Affairs' approach can help alleviate this Coast Guard and national shortfall. Marine Environmental Response ... The U.S. Coast Guard is responsible for enforcing the Federal Water Pollution Control Act and related legislation on the navigable waters of the United States and tributaries to ensure that public health and welfare are protected. Plastic harnesses and hazardous medical waste are two emergent problems in which the agency is involved in cooperative interagency activities. Pollution monitoring and response is again another Federal interagency effort in which the U.S. Coast Guard shares on-scene coordinator responsibilities with the Environmental Protection Agency, depending on whether the spill has occurred in the inland waterways or coastal/of f shore areas. And, the agency views the Marine Environmental Protection responsibility in a systematic fashion extending from preventive measures in the operation of vessel traffic systems and anchorages, the enforcement of laws governing weaterfront facility safety, and the stowage of explosives, oil and hazardous cargo on board vessels ... to the remedial responsibilities for responding to oil spills and the release of other harmful substances in the inland marine, coastal and offshore area. Closely related to the Ports and Waterways, Safety and Environmental Protection responsibilities is a designated Bridge Administration program. The Coast Guard assumed responsibility for administering bridges over navigable U.S. waters in 1967, shortly after the Department of Transportation was f ormed. This encompasses approximately 18,000 bridges in the U.S. from foot bridges to the Verrazano. The Coast Guard is charged with ensuring safe and unobstructed navigation through or under bridges, while meeting the needs of other modes of transportation, such as car and truck traffic on the bridges. This program may soon be transferred to the U.S. Corps of Engineers; and, hopefully, more resources can then be directed to coastal and other bluewater concerns. 1723 Recreational Boati2g__Saf-ety- -. In 1980, 81.8% nf the population lived in Amrica's coastal states and the cl--trrent projections indicate an increasinq movement of population concentrations to the Coastal Zone cities, and clusters of beaches, bays and harbors in the megalopolitan corridors. Concomitant with this trend is a rapid growth in recreational boating, now a national maritime pastime with about 70 million boaters (operators, passengers, skiers, hunters and fishermen) and over 15 million recreational vessels. To meet this national recreational and economic activity, the U.S. Coast Guard administers a comprehensive program of assuring manufacturers' compliance with standards, and facilitates safe operation and enforcement through dual Recreational Boating educational programs. These tasks are pursued through the 40,000 member Coast Guard Auxiliary in the form of formal public education courses, courtesy boat examinations, and regatta patrols, as well as a $30 million financial grant system to the states and jurisdictions for carrying out the significant education and enforcement efforts. Given the size of the constituency, this is a significant, intergovernmental responsibility in which the Coast Guard's role is largely managerial and less asset-intensive. However, the boating population remains the Coast Guard's largest, most identifiable constituency. Defense Readiness ... is a fundamental underlying mission and the U.S. Coast Guard is the Nation's oldest continuous seagoing service with a record of participation in every defense encounter since the birth of the nation. Federal law requires that the U.S. Coast Guard be at all times a military service and a branch of the Armed Forces within the Department of Transportation. The only exceptions are in time of war or as directed by the President when it operates as part of the U.S. Navy. In 1984, following a Memorandum of Understanding between the Secretary of Transportation and the Secretary of the Navy, Coast Guard Atlantic Area and Pacific Area Commanders were designated by the U.S. Navy as Maritime Defense Zone Commanders for their respective coastal area. They are tasked by CINCLANTFLT and CINCPACFLT to exercise command and control over DOD and Coast Guard forces supporting maritime defense. Coast Guard Area Commanders assume additional responsibilities to plan, direct and conduct defensive operations in the safety of ports, harbors and navigable waters. In addition, they will support other commands and agencies in anti-submarine warfare, mine countermeasures, protection of shipping, intelligence, civil defense and related matters. To support and complement the Defense Readiness mission, the U.S. Coast Guard maintains a 13,000 member Reserve Program, including men and women, to provide qualified individuals and trained units for active duty in time of war or national emergency. In addition to its role in national defense, the Reserve augments the active service in the performance of peacetime missions during domestic emergencies and during routine and peak operations. Research and Development - In support of all Coast Guard operational programs, a modest but mission-specific, engineering, research and development program has been deigned to apply state-of-the-art technology to the agency's needs. The Coast Guard's R&D support tasks, like its operational program, are characterized by both an intergovernmental and international perspective. Interagency cooperation increases the pool of expertise, and reduces redundancy and costs in research and development. There has been a significant measure of US-Canadian cooperation in this support program. In addition to its R&D effort, the U.S. Coast Guard maintains a separate Marine Science Activities program, which is entirely applied oceanography in support of Coast Guard missions. It includes the conduct of the International Ice Patrol, begun in 1914 following the TITANIC disaster, to detect and chart iceberg movement into shipping lanes; operational support for NOAA's data buoy project; early marine and coastal weather observation and reporting from 124 Coast Guard shore stations and 81 cutters; and several other Federal 1724 interagency projects such as ocean sounding. The Coast Guard has been involved in SEASAT, the ocean-observing satellites, participating in the operational requirements for the Alaskan Synthetic Operative Radar (SAR) Facility for sea ice and iceberg reconnaissance. Operational ocean products from future satellites will be used to assist in predicting surface object drift in the ocean for Search and Rescue, International Ice Patrol, Marine Environmental Protection and Response, and Law Enforcement. Infrastructure: Budget and Personnel: To carry out the foregoing missions the U.S. Coast Guard received an estimated $2.5 billion in its PY88 budget, including some 1.8 billion in operating expenses. Its manpower level is just over 38,000 military with an officer corps of 5,000. It has a civilian complement of about 5,000 and is assisted part time by some 40,000 Coast Guard Auxiliarists. Women are significant in the Coast Guard; the Academy was the first to accept women cadets and the service was the first in assigning women as CO's of armed vessels. It now has approximately 190 female officers and over 2,300 women enlisted personnel. International Participation: In addition to the international involvement previously cited, the U.S. Coast Guard is a major participant in international affairs, both in the furtherance of U.S. foreign policy as well as in the interest of the international ocean community at-large. The Coast Guard assists other nation's marine agencies under the provision of over 50 international instruments, exclusive of an additional number of international fisheries agreements. The Coast Guard has assisted these other maritime agencies in training foreign nationals within the U.S. from over 60 nations ranging from Argentina to Zaire. This training includes, but is not limited to, cooperation with the Department of Defense's Military Assistance Program and training in accordance with LORAN agreements between the U.S. and other nations. It has provided mobile training teams to nations such as Jordan, Saudi Arabia and Haiti and all in the varied training areas of search and rescue, merchant marine safety, port security, law enforcement and aids to navigation. In addition to those forms of cooperation identifiable as 'training assistance,' "operational" forms of assistance have also been provided, such as dispatching oil pollution strike forces to the Straits of Malacca and Magellan and elsewhere. Whether one acknowledges the marine scientific and operational functions of conducting the International Ice Patrols, providing operational and technical direction to LORAN stations overseas (under various treaties and agreements), comprehensive participation in major UN/IMO subcommittees, or representation within the special inter-American Port and Harbor conferences and the UNCLOS itself, the U.S. Coast Guard's international, socio-technical assistance is extremely broad. The international involvement is also performed within a healthy interagency arrangement with EPA, NOAA, the National Science Foundation, the Department of State and other Exectuive components. However, the Federal interagency and international participation must, at some point, be more cohesively and consistenly organized within the Coast Guard itself and in a policy-integ rated manner, if the agency's fullest measure of 'External Affairs' benefit is to be achieved. In July of 1986, a Coast Guard flight crew f lew a C-130 to the Peoples Republic of China to demonstrate its Search and Rescue capabilities to senior military and government officials. The PRC's SAR interest has been motivated by the start of offshore drilling operation for oil in the South China Sea and some severe accidents they have experienced in ferrying people from the mainland to the rigs.- The Coast Guard Captain who headed the project was born in Beijing and is fluent in Chinese. 1725 On the more diplomatic and mini-Olympian f ront, the Coast Guard was a major participant in the SAR week, Marine Commerce Week, and 'Sea the Future' Symposium of EXPO the '86 World's Fair in Vancouver, B.C. The participation in sporting events with the other nations' Coast Guard forces, and the visits of helicopters and the cutters MORGENTHAU and BOUTWELL, were designed to accommodate that fair's theme of 'Han in Motion ... Man in Touch' with an operational presence. Yet, these aforementioned activities are only a few of the intergovernmental and international cooperative projects in which the U.S. Coast Guard is 'Involved. THE U.S. COAST GUARD AND NATIONAL OCEAN POLICY The Coast Guard is an 'Armed Service and More.' And, to accommodate all of these responsibilities, it must employ a multi-mission system of management which is designed to effect the optimum utility of whatever level of physical, financial and personnel resources is available. Also, to perform the multiple range oil missions, the vessels and aircraft must be so equipped with the necessary hardware and state-of-the-art technology; and its personnel must have a corresponding measure of expertise and versatility In their training. So, it seems that the sum total and key elements to this management challenge, at least for the remainder of this millennium, may be 'productivity, and professionalism.' Those twin goals and attributes will apply to both the public sector and private sector. 'Productivity' will become a significant byword and basic barometer of performance. And productivity can be enhanced in a critical path network when interagency and intergovernmental partnerships are maximized. The Coast Guard's programs reach to the full range of those parameters enveloping in what is circumscribed within the Exclusive Economic Zone; and, in which its mission frame of reference must include policy considerations for maritime safety, transportation, maritime law enforcement, national defense, environmental protection, and of course all other functional policies inherent in its responsibilities for the overall facilitation of waterborne transportation. Those policy considerations will more and more involve interagency, intergovernmental and international coordination as budgetary limitations are imposed; and, national and international oceanic wants and needs are multiplied. To assess these newer dimensions of national ocean policy, the Coast Guard must, like its motto, be 'always prepared' ; and such preparation will necessitate fine tuning Its key, policy-oriented, and multi-programmatic staffs so they are postured for the future. Because the U.S. Coast Guard has been a respondent to many initiatives and issues addressed in the name of 'National Ocean Policy,' it had to respond accordingly. For example, in responses to various inquiries and initiatives regarding the extent of interagency cooperation and coordinat4on, it submitted a straw definition to the National Advisory Committee on Oceans and Atmosphere (1971-1985) i.e., it stated generally that "ocean policy is a reflection of energy policy, transportation policy, fisheries development and other forms of economic policy, foreign policy, defense policy, fiscal policy, law policy." That definition, like the theme of a 'Partnership of Marine Interests' designed for Oceans '88, and the octagon logo of intergovernmental players, was fashioned to guarantee apertures for participation by 411 concerned parties in the national ocean policy formulation process. The agency prototype is based on promoting the highest degrees of interagency and intergovernmental consensus. 1726 .In 1978, to provide a model for interagency cooperation, the U.S. Coast Guard established the 'Ocean Principals Group' as an ad hoc forum within the Executive Branch - to discuss marine interagency issues which have become mutual commonality and concern. That forum still meets for quarterly luncheon discussions and includes as members 'Principals' from the Office of Science and Technology Policy, the National Security Council, the Coast Guard, NOAA, the Navy, MARAD, the Army Corps of Engineers, the U.S. Geological Survey, the Minerals Management Service, the Environmental Protection Agency, NASA, the National Science Foundation, the Energy Department and State Department. The Coast Guard, as founder, remains as Executive Secretary; however, the chair is rotated quarterly with the meetings. It works because it has a comprehensive roster of all the Federal interests, is objective, and is one high-level arena and captive audience where emerging interagency problems can be headed off or mitigated; and where each agency's concerns can be introduced - and, it takes place within a very informal setting. Because the U.S. Coast Guard implements such a variety of inland marine, coastal and trans-oceanic, national and international activites - with an extensive multi-mission infracture of over 200 helicopers and fixed wing aircraft, over 240 large vessels, over 2,000 small boats and over 1,200 shore stations, and shares interagency concurrent jurisdiction for many of its statutory responsibilities, it interacts closely with all the players in the intergovernmental arena. The Ocean Principals is informal; there are other informal arrangements. To assure appropriate networking with other agencies and the private sector, it has at last count, over 80 memoranda of understanding with entities ranging from the American Bureau of Shipping to the Veterans Administration, from the states of Texas and Florida to Environment Canada and other nations. To participate formally with industry and education in the intergovernmental arena, the U.S. Coast Guard has both chaired and directed the marketing and press operations for the biennial, Washington-based "Oceans" conferences. Those conferences have been identified as germane Federal participation under the Presidential Task Force. on Private Sector Initiatives, and heralded a lasting 'Industry-Government-Education' partnership. The theme has ranged from 'Partners in Progress', to 'Designs for the Future', to 'Science, Engineering and Adventure,' to this year's theme of 'A Partnership of Marine interests.' Returning again to the nature of ocean policy and the notion of a unique interagency and international perspective, it is my view that national government policy formulation is evolving into a newer, 'state-of-the-art' process. We can thank the French for recognizing the press as the Fourth estate of government. In recent years the public policy pundits have discerned the interest groups, or 'gatekeepers,' as tantamount to the Fifth Estate. Now, I think we might recognize the roles of the private sector and public and special interest groups in the national public policy process as the Fifth Estate; the roles of think-tank et. al. educational groups as the Sixth Estate; the states and localities as the Seventh Estate; and the international community of interests as the Eighth Estate. We might indeed call the configuration an 'Ocean Policy Octagon.' The array is again depicted in Figure #2. Whatever value in such modeling, the bottom line is that institutionalized cooperation is far better than the episodes of intransigent confrontation which can occur if all concerned interests are not consulted.. 1727 A POSSIBLE DIRECTION FOR THE COAST GUARD IN INTERNATIONAL AFFAIRS Given such an expanded perspective within the Coast Guard, and given the origination and nurturing of a 'Partnership of Marine Interests,' a major initiative is suggested to posture the Coast Guard as a more viable instrument of foreign policy. It is stated as follows: given the myriad of national missions with which the agency has been tasked; given the experience of working with many mechanisms for national interagency and intergovernmental and international cooperation; given the extensive involvement of the U.S. Coast Guard in the area of international marine technical assistance, as manifested in cooperative projects with a myriad of nations; and given its unique 'white ship' receptivity in international participation in SAR agreements and other multilateral events (EXPO '86, etc.); ... how might the Coast Guard's utility as a significant instrument of American foreign policy be more properly recognized and utilized in the furtherance of overall foreign policy and defense objectives? Can the military/bluewater mission base be further enhanced by greater technical and operational assistance to coastal nations, particularly developing states now acceding to various zones of coastal control, and in return gain support for U.S. seapower and other national and allied objectives? Is there a future national defense role where an expanded Coast Guard can provide such 'transfer of expertise' and 'transfer of technology' in close cooperation with private sector firms such as TRACOR Industries, now involved in the design of a master coastal surveillance system for the Government of Indonesia? Might this also be a variation of the original mandate of the Presidential Task Force on Private Sector Initiatives? Might it provide some of the agency's share of privatization overtures and offset the displacement or dilution of other needed resources? To reinforce this national and international initiative, perhaps the Coast Guard could also, as a follow-on to its EXPO '86 participation, take the leadership for establishing an 'International Association of Coast Guards?' THE CONCEPT OF EXTERNAL AFFAIRS FOR THE U.S. COAST GUARD During the remainder of this millennium, the United states, as well as other seafaring nations, will advance their focus on the coastal regions and oceans in order to satisfy emerging economic and recreational. needs, meet national defense requirements, and concurrently plan for the significant socio-environmental concerns associated with accelerated ocean resource and ocean space development. Oceans 188 has been offered as a major milestone in this coastal and offshore evolution. The U.S. Coast Guard can also be a major milestone in the continuing national ocean policy evolution, but its internal organization must match the external world with far greater compatability. Given the complexity of the managerial challenges ahead, and the reality that budgetary resources must be utilized prudently, my perspective is that a new 'Partnership of Marine Interests,' integrating the diverse industrial interests, Federal interagency particpants, the state governments, the range of comprehensive intergovernmental 'groups and the relevant ' international interests should be pursued by the Coast Guard as well as the ocean community at-large. A more precise integration of external activities,must take place @within the Coast Guard if it is to match its internal organization with the challenges and opportunities presented in the external ocean policy arena. 1728 If the Coast Guard is to come of age and develop any carefully considered, proactive strategies to survive in an increasingly austere, national budgetary environment, it should establish: first, an 'External Affairs Secretariat;' and later when oportune, an txternal Affairs stall7f directed by a flag officer. Such an organizational component should integrate its Congressional Affairs Staff, Governmental Affairs. Staff (the integrated Federal interagency and ('within the Beltway' groups), International Affairs Staff and Public Affairs Staff (Media Relations, internal communications and 'Beyond the Beltway' community affairs functions), so that their respective directors can discuss external interrelationships more cohesively. Such a senior staff is further necessary: to relate internal management to the reality of how policy is formulated and budgets are orchestrated in the external world; to provide a continuum of critical analysis and insights on 'Within the Beltway' interest group initiatives and needs; to provide the integrated feedback and necessary recommendations for the Commandant's decision-making; to design responses and innovations to relate short-term, reactive 'Event Management' to longer term, proactive 'Trend Management' or indeed 'Survival Management in Strategic Planning;' to articulate a more consistent voice and posture with key Legislative, interagency and intergovernmental groupss; to chart the future mission base of the Coast Guard in closer communion with its critical support groups; and, enhance its resource base to implement, and take fullest advantage of, the evolution of inland marine, coastal and trans-oceanic needs. The Coast Guard can be a leader and designer of the marine policy agenda, help draft new mission areas, and concurrently enhance its status in the budgetary allocation area. To paraphrase Franklin Delano Roosevelt, it must generate "the will to prosper as well as to survive." CONCLUSION I close my paper with the notion that the more we meet, within international fora such as Oceans '88 for professional and purposeful exchanges and the associated camaraderie, we will recognize that the short-term decisions we make for immediacy, visibility and independent needs, sooner or later can and must be reconciled with the longer-term decisions for imagination vision and interdependent needs. Federalist James Madison may have observed that man is inclined toward factions, but fellow Bostonian John Adams expressed more profoundly that, "if we don't all hang together surely, we will all hang separately." . . . . . . . . . . . . . This article is an elaboration of my presentation before the 'Sea The Futuret symposium of the EXPO '86 World's Fair, Vancouver! B.C., September 14, 1986. The views offered are solely my own and do not necessarily reflect any current official position of the United States, the U.S. Coast Guard/Department of Transportation or other federal entities. I do not suggest however, that the government would have any occasion to disfavor the loyal and presumably positive, accurate and - I trust - helpful thoughts I have conveyed for my colleagues in that public policy domain we call the ocean community." -- Edward W. Cannon 1729 THE UNSPOKEN YET VITAL PARTNERSHIP BETWEEN THE UNITED STATES COAST GUARD RESERVE AND THE PUBLIC SECTOR AT LARGE Petty Officer Second Class Louise A. Berney, USCGR There is a component of each branch of our armed forces known as the reserve. The intention here is to focus upon the United States Coast Guard Reserve. The words "reserve" or "reservistil , within the USCG, are in many ways misnomers. The USCG is the singlemost counterpart of all the branches of the armed forces whose missions dictate constant contact with the public *. This is also true of a Coast Guard Reservist in an IDT (weekend drill) status. However, let's carry that several steps further. The interaction amongst USCG reservists and the public does not end at 1700 on a sunday at the close of an IDT weekend. A reservist does not "reserve" his USCG - public partnership activities to two days per month. It operates on a daily basis in an "unofficial" capacity throughout the remainder of the month. The members of the USCGR bring their Coast Guard know-how to the public daily. The premise is that, in this, there is an exchange of ideas, expertise and understanding between the USCG and the public that goes beyond the reservists' official military capacity. This is the unspoken partnership. It does exist. It is beneficial to both the USCG and the public. It is not a funded function yet has positive econimic ramifications as well as positive public relations benefits. It does deserve to be brought to the forefront. The intricacies and varieties of the exchanges between reservist and public have been observed for nearly five years. They have been scrutinized, questioned, ruminated upon, dissected and-even participated in by this author. This has included speaking with over.250 reservists from every state and Puerto Rico. Perhaps this may not statistically be considered a large cross section out of approximately 12,000 members of the USCGR. However, direct conversation and detached observation has provided candid, unsolicited input. Of the 250 plus reservists, their civilian occupations included science, academia, law enforcement, firefighting, indusrty, technology, naval archictecture and engineering, environmental consulting, port operations, diving, government and recreation. There is a two way street that has developed between the reservists' USCG role and their civilian (public) roles. Each of these 250 reservists were examples of mutual benefit to the USCG and the public. It evolves naturally with the reservist serving as the catalyst and enabler for the partnership pathway. The following examples will serve as a representative sampling. No statistics have been included since this good will and exchange of expertise cannot be quantified, nor for the intended purpose of this expose should it be statistically manipulated. For purposes of this expose, public is defined as all academic, industrial, governmental, scientific, maritime/marine, persons and business entities outside of the USCG and USCGR.) 1730 United States Government work not protected by copyright A reservist on a civilian business trip stops by a USCG district office to visit. This sparks a dialogue amongst USCG personnel and the reservist. It so happens, coincidentally, that there has been a significant pollution incident at a petrochemical plant that handles the same products as the plant the reservist manages. The outcome of this "unofficial" dialogue resulted in the USCG personnel gaining new insights into the realities of daily plant operations and of non-intentional pollution incidents that occur at facilities who are typically in compliance with regulations. This occurred in such a manner where none of the parties in the dialogue had to act defensively. On the other hand, the reservist gained new insights as the the position and responsibilities of the USCG in such a matter. He could then use this to foster ongoing cooperation and understanding between plants such as his and the USCG. The perspectives gained by both entities were "unofficial", yet valuable. A group of reservists have been invited to visit a Coast Guard cutter. In the engine room, one reservist noticed a USCG junior officer intently working on a problem with an engine component. After several moments of observation, the reservist engaged in a casual conversation with the officer. Soon after, they were approaching the technical problem together. The reservist happened to be a naval engineer/architect who designed engine rooms such as the one on the cutter. The officer learned valuable "tips" on the quirks of the system. The reservist then returned to his company to relate to them the temperment of the engine system in the field under actual daily operation. Another partnership develops. Several local police officers (reservists) patrol a major port area which is also under the jurisdiction of the local USCG Marine Safety Office. The police officers and MSO personnel-are found discussing access to the port facilities. The police officers (reservists) know the most rapid access from the shoreside. The MSO personnel know every little nook and cranny of the port facilities and waterside access. This conversation resulted in the MSO personnel becoming aware of-alternate and more rapid access that they had no opportunity to previously explore. The police officer (reservists) learned of the more minute details of the port area and thereby could be more effective to the public. "No way" shouted a reservist in a class being attended by 29 persons (14 active duty USCG and 15 reservists) and being taught by a combination of regulars and reservists. The outspoken reservist was a vessel agent. The discussion involved the economic realities of the shipping industry versus the Codes of Federal Regulation overseen by the USCG. Reasonable solutions and new perspectives considering both economic reality and regulatory obligation develop. The active duty personnel will be more aware of this in future inspections. The reservists in their civilian roles will be cognizant.of regulatory responsibility. In the field, both entities will use these insights to accomodate needs and responsibilities. The instructors can then pass this on to future classes, thus enhancing the learning process. A company president approaches an employee (a reservist), posing a question about planned bridge construction on a major highway project. The bridge will cross navigable waters. The reservist has patrolled these waters during drill weekends. She then calls USCG Captain of the Port for the particulars of a permit process. COTP personnel explain the criteria to the reservist and do so easliy since the reservsist is familiar with the area due to her USCGR patrols and knowledge of USCG terminology and regulation processes and understands the USCG terminology and basic policies On the other hand, the.COTP personnel query her to explain more on how a major thoroughfare's @bridge crossings come into being from the earliest stages to completion. The engineers at the reservist's consulting firm are confident of the information from the USCG that she passes to them since they are aware of her,dual perspective in the matter. Partnership once again develops. These are but a few examples cited to show such "unofficial" exchanges between the USCG and the public occur on a daily basis. These are partnerships far more common and beneficial than we realize. The dual perspective (reservist as civilian and reservist as Coast Guardsman) lends itself well to fostering true understanding since exchanges are casual (without the defensiveness that develops in many official exchanges) and are candid. Best of all, it is of great economic benefit. These parnerships require no purchase orders and no budgeting and save time. The economic benefits are inestimable. The unspoken and unofficial role of the United States Coast Guard Reservist is truly that of educator, arbitrator and facilitator. There is a dual perspective and combined knowledge (USCG and public) that reservists possess due to their duality of roles. They bring this to both the United States Coast Guard and to the public. They are not "officially" ordered or obligated to do so, nor are they paid to do so. It simply develops and is for the benefit of all concerned. Herein lies a true partnership of marine interests. 1732 DATE DUE 36668141061178