[From the U.S. Government Printing Office, www.gpo.gov]






                                    Science for Solutions


            NOAA COASTAL OCEAN PROGRAM
            Decision Analysis Series No. 9



                     ATMOSPHERIC NUTRIENT INPUT TO
                                  COASTAL AREAS

                        RED UCING THE UNCER TA IN TIES




                                                          Richard A. Valigura
                                                       i ji
                                                          Winston T. Luke
                                                          Richard S. Artz
                                                          Bruce B. Hicks


                                                           NOAA AIR RESOURCES
                                                               LABORATORY




                                               '-wr7@77,f













      TD427                             JUNE 1996
      .N87
      A8 '        I             U.S. DEPARTMENT OF COMMERCE
      1'9 9 6
                          National Oceanic and Atmospheric Administration
                                      Coastal Ocean Office










                                          DECISION ANALYSIS SERIES

                                        The Decision Analysis Series has been
                                        established by NOAA's Coastal Ocean
                                       Program (COP) to present documents for
                                       coastal resource decision makers which
                                         contain analytical treatments of major
                                       issues or topics. The issues, topics, and
                                       principal investigators have been selected
                                           through an extensive peer review
                                       process. To learn more about the COP or
                                          the Decision Analysis Series, please
                                                             write:


                                                             NOAA
                                                  Coastal Ocean Office
                                                 1315 East West Highway
                                                 Silver Spring, MD 20910

                                                  phone: 301-713-3338
                                                  f ax:      301-713-4044












                Cover photo: Convective coastal storms are an efficient mechanism
                for the transfer of atmospheric nitrogen to a given water body.
                Soluble nitrogen compounds in the atmosphere are dissolved into
                precipitation during storm formation as well as by failing rain. (Photo
                by Glenn Rolph, NOAA Air Resources Laboratory, 1992)









                                             Science for Solutions



                                                                                            p3'D
             NOAA COASTAL OCEAN PROGRAM
             Decision Analysis Series No. 9

                                                                                              EhWraT OF


                        ATMOSPHERIC -NUTRIENT INPUT TO
                                          COASTAL AREAS


                                                        -U
                            REDUCING THE NCERTAINTIES



                                            Richard A. Valigura
                                              Winston T. Luke
                                               Richard S. Artz
                                               Bruce B. Hicks


                                     NOAA AIR RESOURCES LABORATORY











                                                 JUNE 1996





                                      U.S. DEPARTMENT OF COMMERCE
                                              Michael Kantor, Secretary
                               National Oceanic and Atmospheric Administration
   Z                                       D. James Baker, Under Secretary
                                              Coastal Ocean Office
                                               Donald Scavia, Director




















                  This publication should be cited as:

                  Valigura, Richard A., Winston T. Luke, Richard S. Artz, and Bruce B. Hicks. 1996. Atmospheric
                  Nutrient Input to Coastal Areas--Reducing the Uncertainties. NOAA Coastal Ocean Program
                  Decision Analysis Series No. 9. NOAA Coastal Ocean Office, Silver Spring, MD. 24 pp. + 4
                  appendices.










                                     This publication does not constitute an endorsement of any commercial product or
                                     intend to be an opinion beyond scientific or other results obtained by the National
                                     Oceanic and Atmospheric Administration (NOAA). No reference shall be made to
                                     NOAA, or this publication furnished by NOAA, in any advertising or sales
                                     promotion which would indicate or imply that NOAA recommends or endorses any
                                     proprietary product mentioned herein, or which has as its purpose an interest to
                                     cause directly or indirectly the advertised product to be used or purchased because
                                     of this publication.







               Note to Readers

               Atmospheric Nutrient Input to Coastal Areas--Reducing the Uncertainties by Richard A.
               Valigura, Winston T. Luke, Richard S. Artz, and Bruce B. Hicks of NOAA's Air Resources
               Laboratory is a synthesis of the findings and the management implications of research sponsored
               by the NOAA Coastal Ocean Program as part of its nutrient enhanced productivity activities.
               The atmospheric deposition of nutrients to estuaries, particularly along the U.S. East Coast, is
               being increasingly recognized as a significant contributor to eutrophication with its concomitant
               adverse impacts on living marine resources. This study by the Air Resources Laboratory is a
               seminal investigation of the subject particularly relating to the Chesapeake Bay.

               The NOAA Coastal Ocean Program (COP) provides a focal point through which the agency,
               together with other organizations with responsibilities for the coastal environment and its
               resources, can make significant strides toward finding solutions to critical problems. By working
               together toward these solutions, we can ensure the sustainability of these coastal resources and
               allow for compatible economic development that will enhance the well-being of the Nation now
               and in future generations. The goals of the program parallel those of the NOAA Strategic Plan.

               A specific objective of COP is to provide the highest quality scientific information to coastal
               managers in time for critical decision making and in a format useful for these decisions. To help
               achieve this, COP inaugurated a program of developing documents that would synthesize
               information on issues that were of high priority to coastal managers. A thrce-step process was
               used to develop such documents: 1) to compile a list of critical topics in the coastal ocean
               through a survey of coastal resource managers and to prioritize and select those suitable for the
               document series through the use of a panel of multidisciplinary technical experts; 2) to solicit
               proposals to do research on these topics and select principal investigators through a rigorous
               peer-review process; and 3) to develop peer-reviewed documents based on the winning
               proposals. Seven topics were selected in the initial round, and COP is planning a second round.
               Additionally, the series is being expanded to include the synthesis of findings from other COP-
               funded research. A list of titles in print appears on the inside back cover.

               As with all of its products, COP is very interested in ascertaining the utility of the Decision
               Analysis Series particularly in regard to its application to the management decision process.
               Therefore, we encourage you to write, fax, call, or E-mail us with your comments. Please be
               assured that we will appreciate these comments, either positive or negative, and that they will
               help us direct our future efforts. Our address and telephone and fax numbers are on the inside
               front cover. My Internet address is [email protected].





                                                            Donald Scavia
                                                            Director
                                                            NOAA Coastal Ocean Program









              ACKNOWLEDGMENTS


              Numerous individuals and agencies contributed their time and/or resources toward the
              completion of this project. Special thaks go to the following individuals who provided
              unpublished datasets used in this report: Bill Matuszeski and Lewis Linker (EPA Chesapeake
              Bay Program Office); Paul Stacey (Connecticut Dept. of Environmental Protection); Holly
              Greening (Tampa Bay National Estuary Program); Mark Alderson (Sarasota Bay National
              Estuary Program); Bill Keene (University of Virginia); Joe Scudlark (University of Delaware);
              Kevin Civerolo, Bruce Dodridge, and Russ Dickerson (University of Maryland Meteorology
              Dept.); Russ Brinsfield, Ken Staver, and Mary Catherine Morrisey (University of Maryland's
              Wye River Education Center); Bill Boicourt (University of Maryrland's Horn Point
              Environmental Laboratory) ; Bob McMillan, Tylden Meyers, Jeff Mcqueen and Barbara Stunder
              (NOAA Air Resources Laboratory).

              The literature sysnthesis was facilitated by the Chesapeake Bay Research Consortium. Thanks is
              due to co-authors Laura McConnell, Joe Scudlark, and Joel Baker.

              The Mt. Washington Workshop was facilitated by the Chesapeake Bay Program's Scientific and
              Technical Advisory Committee. Special thanks must go to the participants at the Workshop,
              many of whom contributed to the text: Viney Aneja, Carmen Aquillar, Joel Baker, Richard
              Batuik, John Benedict, Brandon Bonanno, Owen Bricker, Mark Bundy, Al Cimorelli, Dave
              Correll, Greg Cutter, Robin Dennis, Russ Dickerson, Rebecca Dickhut, Bruce Dodderidge, Jim
              Galloway, Rieke Hood, Carolyn Hunsaker, Betsy LaRoe, Lewis Linker, Melissa McCollough,
              Laura McConnel, Jeff McQueen, Paul Miller, Ed Myers, John Ondov, Jess Parker, Eileen
              Rowan, Bill Ryan, Dan Salkovitz, Richard Scheffe, Terry Schemm, Joseph Scudlark, Sidney
              Steele, Robert Summers, Eric Van DeVerg, and Marv Wesely.

              The Airlie Workshop was facilitated by the Alliance for Chesapeake Bay and the Chesapeake
              Bay Research Consortium. Thanks are due to all who participated in the Shared Resource
              Workshop, and especially to all who contributed material that found its way into the present text
              - Mark Alderson, Suzanne Aucella, Joel Baker, Richard Batiuk, Glen Besa, Rona Birnbaum,
              Karl Blankenship, Donald Boesch, Walter Boynton, Peg Brady, Dail Brown, Diane Brown, John
              Calder, Mark Castro, James Collier, Chris Deacutis, Peter de Fur, Robin Dennis, Ellen Parr
              Doering, Fred Durham, Charles Ehler, Kelly Eisenman, Fran Flanigan, David Foerter, Elizabeth
              Gillelan, Normand Goulet, Holly Greening, Sonia Hamel, Patricia Harrington, Sheila Holman,
              Rieke Hood, Carolyn Hughes, Bill Keene, George Keller, Margaret Kerchner, Erica Laich, Lewis
              Linker, Jim Lynch, Gail MacKiernan, Tom Maslany, William Matuszeski, Paul Miller, Brian
              Morton, Kent Mountford, Thomas Noel, Marria O'Malley Walsh, Karen Rice, Barry Rochelle,
              Thomas Rogers, Eileen Rowan, Dan Salkovitz, Joseph Scudlark, Moira Schoen, Bill Sharp, John
              Sherwell, Deborah Shprentz, Paul Stacey, Ronald Stanley, Nancy Summers, Robert Talbot,
              Robert Thomann, Julie Thomas, Darryl Tyler, John Wickham, and Derek Winstanley.




                                                             v









               FOREWORD


               The overall goal of Atmospheric Nutrient Input to Coastal Areas (ANICA) Program was to
               develop methods for assessing the importance of atmospheric nutrient input, using the
               Chesapeake Bay as a first target of contemporary importance (as evidenced by specific mention of
               atmospheric deposition to the Bay, among other water bodies, in the Clean Air Act Amendments
               of 1990).

               As an atmospheric component of the nutrient research program of NOANs Coastal Ocean
               Program, the long-term objectives of the research program were:

                0     To determine the wet and dry deposition of nitrogen to Chesapeake Bay and other East
                      Coast estuarine areas selected for intensive study

                 ï¿½    To develop a strategy for assessing the dry and wet deposition affecting other coastal
                      watersheds in the Northeastern United States and Maritime Canada


                ï¿½     To apply the models that are developed or modified in this program to describe and
                      predict present and future atmospheric deposition scenarios for catchment areas impacted
                      by nitrogen deposition

                ï¿½     To link the findings from ANICA with the ecological components of NOAA's Coastal
                      Ocean Program and the Clean Air Act's Great Waters Program objectives























                                                            vii









                      TABLE OF CONTENTS


                      LIST OF FIGURES AND TABLES                                         ................................................................................ xi


                      LIST OF TERMS AND ACRONYMS                                            ........................................................................... xiii


                      EXECUTIVE SUMMARY                                    ............................................................................................... xv


                      CHAPTER 1: INTRODUCTION                                         .....................................................................................I
                         Nitrogen in the Environment                    ......I ............................................................................................I
                         Atmospheric Nitrogen                  .............................................................................................................I
                         Defining the Problem                  .............................................................................................................4


                      CHAPTER 2: SCIENTIFIC FRAMEWORK                                                    ................................................................7
                         Mass Balance Paradigm                     .........................................................................................................7
                         Sources and Emissions                   ............................................................................................................7
                                  Reactive Nitrogen: Nitrogen Oxides                       ...............................................................................9
                                  Reduced Nitrogen: Ammonia and Ammonium                                  ...............................................................9
                                  Organic Nitrogen               .......................................................................................................... 10
                         Deposition Flux Estimates                     .................................................................................................... I I
                                  Wet Deposition               ............................ ............................................................................... 11
                                  Dry Deposition               ........................... ............................................................................... 12
                         Loadings to Chesapeake Bay                        ............... ............................................................................... 13
                                  Direct Loadings              .........................................;................................................................. 13
                                  Indirect Loadings                ....................................................................................................... 13
                                  Total Loadings Estimates                   .............I............................................................................... 14
                         Loadings Intercomparison Between Coastal Waters                                  ............................................................. 16


                      CHAPTER 3: REMAINING ISSUES                                             .......................................................................... 19
                         The Literature Synthesis                    ......................................................................................I .............. 19
                         The Mt. Washington Workshop                         ............................................................................................. 20
                         The Airlie Workshop                  ............................................................................................................. 22

                      APPENDICES:
                         A. LIST OF ANICA TASKS                                     ................................................................................... A-1


                         B. SPECIFIC STUDIES                             ............................................................................................... B-1
                                  A Regression Approach for Estimating Precipitation Chemistry                                     ...............................  B-1
                                  Wet Deposition Over Chesapeake Bay                            ...................................................................... B-4
                                  Wet Deposition of Organic Nitrogen                         ......................................................................... B-6


                                                                                            ix










                             Dry Deposition Inferential Method (DDI" Measurements at Wye, Md                                . .................... B-8
                             Atmospheric Nitrogen Speciation Measurements                         ....................................................... B-1 I
                             Intercomparison ofFilterpack and Denuder Techniquesfor HN03Measurements                                        .... B-16
                             Estimating A ir- Water Transfer ofHN03                      .................................................................. B-19
                             Operational RAMS 15 Km Forecasts over Chesapeake Bay                                ..................................... B-26


                      C. ANICA ASSOCIATED SCIENTISTS                                          ............................................................. C-1


                      D. BIBLIOGRAPHY                           .................................................................................................. D-1








































                                                                                    x








                 LIST OF FIGURES AND TABLES

                 Figures

                 Figure 1. 1      Atmospheric input of nitrogen to coastal areas

                 Figure 2, 1      A simplified schematic of the coastal nitrogen cycle

                 Figure 2.2       A simplified schematic of the terrestrial nitrogen cycle

                 Figure B. 1.     Spatial distribution of the wet deposition of inorganic nitrogen across the
                                  Chesapeake Bay watershed

                 Figure B.2.      A representation of the sequence of mechanisms influencing the deposition from
                                  the atmosphere of trace gases and diff-using particles

                 Figure B. 3.     The 48-hour back trajectories from Wye, Md. corresponding to periods of low
                                  measured NO,, concentrations

                 Figure BA.       The 48-hour back trajectories from Wye, Md. corresponding to periods of high
                                  measured NO,, concentrations

                 Figure B. 5.     The frequency distribution for 18 months (excluding winter) of estimated
                                  deposition velocities in Chesapeake Bay and the distribution for I week in
                                  December 1993


                 Figure B.6.      Ten-minute deposition velocities calculated for December 1-8, 1993

                 Figure B.7.      Ten-minute Monin-Obukhov dimensionless stability estimates versus measured
                                  windspeed, calculated from the Chesapeake Bay Observing System data for April
                                  1994


                 Figure B. 8.     Horizontal (vectors) and vertical (contours) windspeed simulations performed over
                                  the northern Chesapeake Bay employing the finest nested grid spacings (AX=2.5
                                  km)

                 Figure B.9.      Operational 15 km RAMS simulations of northern Chesapeake Bay








                                                                      xi










                 Tables

                 Table 2.1        Budget of global NO. emissions by source (Teragrams [I Tg=I          012 g] N yr-')

                 Table 2.2        Budget of global NH3emissions by source (Tg N yr-')

                 Table 2.3        Watershed retention values (in % of nitrogen loading) used in Bay loading studies
                                  to date


                 Table 2.4        Proportion of total nitrogen loading to coastal waters attributed to atmospheric
                                  input

                 Table B. 1.      Regression coefficients of the selected models (p-values)

                 Table B.2.       Testing statistics for total nitrogen (N03-+NH4') deposition estimates (in
                                  percent)

                 Table B. 3.      Basic03, CO, daytime NO, N02, NOY, and No,/NOY statistics for data collected at
                                  Wye, Md. between September 2-28, 1993

                 Table BA         Mean and maximum observed NO. mixing ratios for "Clean" and "Dirty" days
                                  measured at Wye, Md. between September 2-28, 1993

                 Table B. 5.      Comparison between deposition (g HN03         m=2 ) estimates (Vd * [HN03]) derived
                                  using different averaging schemes and the time series average

                 Table B.6.       Summary of surface data available for RAMS

                 Table B. 7.      RAMS operational configuration

















                                                                     xii









                LIST OF TERMS AND ACRONYMS


                CO              Carbon monoxide                    DH            Heat transfer coefficient
                Fd              Dry deposition Flux                Gg            I Gigagram. = 10' g
                HCL             Hydrochloric acid                  HN03          Nitric acid
                Mg              I Megagram = 10' g                 N2            Diatomic nitrogen
                Norg            Organic nitrogen                   NT            Total nitrogen
                N20             Nitrous oxide                      NH3 Ammonia
                NH4'            Ammonium                           (NH4)2SO4 Ammonium sulfate
                NO              Nitrogen oxide                     N02 Nitrogen dioxide
                NOx             Sum of NO and N02                  NOy           Total reactive nitrogen
                N03-            Nitrate                            N03NH4        Ammonium nitrate
                p-N03           Particulate nitrate                PAN Peroxyacetyl nitrate
                03              Ozone                              Ra            Aerodynamic resistance
                Rb              Quasi laminar resistance           R.            Canopy resistance
                RT              Total air-surface resistance       SLM           Standard liters per minute
                S02             Sulfur dioxide                     UZ            Windspeed at height z
                Vd              Deposition velocity                V9            Gravitational settling velocity
                Z/L             Monin-Obukhov coefficient


                AEOLOS          Atmospheric Exchange Over Lakes and Oceans Study
                AEROCE          Air/Ocean Chemistry Experiment
                AIRMoN          Atmospheric Integrated Research Monitoring Network
                CAAA            Clean Air Act Amendments of 1990
                CAD             Citric Acid Denuder
                CBOS            Chesapeake Bay Observing System
                CL              Ozone Chemiluminecence technique for measuring NO
                DDIM            Dry Deposition Inferential Method
                HY-SPLIT        Hybrid Single Particle Lagrangian Integrated Trajectory Model
                GFC             Gas Filter Correlation
                GPCP            Global Precipitation Chemistry Project
                1C              Ion Chromatography Analysis
                NADP            National Atmospheric Deposition Program
                NAPAP           National Acid Precipitation Assessment Program
                NCDC            National Climatic Data Center
                NCEP            National Centers for Environmental Prediciton
                NDIR            Non-Dispersive Infrared Adsorption
                PMT             Photomultiplier Tube
                RADM            Regional Acid Deposition Model
                RAMS            Regional Atmospheric Modeling System
                UV              Ultraviolet
                WATOX           Western Atlantic Ocean Experiment



                                                                   xiii









                EXECUTIVE SUMMARY

                A significant fraction of the total nitrogen entering coastal and estuarine ecosystems along the
                eastern U.S. coast arises from atmospheric deposition; however, the exact role of atmospherically
                derived nitrogen in the decline of the health of coastal, estuarine, and inland waters is still
                uncertain. From the perspective of coastal ecosystem eutropl-kation, nitrogen compounds from
                the air, along with nitrogen from sewage, industrial effluent, and fertilizers, become a source of
                nutrients to the receiving ecosystem. Eutrophication, however, is only one of the detrimental
                impacts of the emission of nitrogen containing compounds to the atmosphere. Other adverse
                effects include the production of tropospheric ozone, acid deposition, and decreased visibility
                (photochemical smog).

                Assessments of the coastal eutrophication problem indicate that the atmospheric deposition
                loading is most important in the region extending from Albemarle/Pamlico Sounds to the Gulf of
                Maine; however, these assessments are based on model outputs supported by a meager amount of
                actual data. The data shortage is severe. The National Research Council specifically mentions the
                atmospheric role in its recent publication for the Committee on Envirom-nental and Natural
                Resources, Prioritiesfor Coastal Ecosystem Science (1994). It states that, "Problems associated
                with changes in the quantity and quality of inputs to coastal environments from runoff and
                atmospheric deposition are particularly important [to coastal ecosystem integrity]. These include
                nutrient loading from agriculture and fossil fuel combustion, habitat losses from eutrophication,
                widespread contamination by toxic materials, changes in riverborne sediment, and alteration of
                coastal hydrodynamics."

                Much of the initial understanding of the atmospheric deposition problem derived from work of
                NOAA's Air Resources Laboratory (ARL). During the 1980s, under the auspices of the National
                Acid Precipitation Assessment Program (NAPAP), ARL conducted a major study (the Western
                Atlantic Ocean Experiment - WATOX) of the fate of air pollutants carried by the wind beyond
                the east coast of the continental U.S. This study revealed that almost all of these pollutants are
                deposited to the ocean, with greatest deposition rates occurring in the nearshore region. About
                3 0% of the total U.S. air emissions of nitrogen are deposited to the Atlantic, again with heaviest
                deposition to the near-shore region. Further impetus was added to the atmospheric transport
                issue by the passage of 1990 Clean Air Amendments which require that:

                       Yhe Administrator [of the Environmental Protection Agency], in cooperation with
                       the Under Secretary of Commercefor Oceans andAtmospheres, shall conduct a
                       program to identify and assess the extent of atmospheric deposition of hazardous
                       airpollutants (and in the discretion of the Administrator, other airpollutants) to
                       the Great Lakes, the Chesapeake Bay, Lake Champlain and coastal waters.

                To address the need for an objective methodology to assess the importance of the atmospheric
                input to coastal regions, the ARL/Atmospheric Nutrient Input To Coastal Areas (ANICA)
                program was developed through the NOAA Coastal Ocean Program. During the four-year


                                                                 Xv










                lifetime of ANICA, methods for assessing the role of atmospheric nitrogen loadings to coastal
                areas were developed, and other stakeholders were alerted to the importance of the atmospheric
                deposition issue.

                Much of the progress under the ANICA program was made by cooperative effort with the
                Chesapeake Bay Program through alliance with the NOAA Chesapeake Bay Office. Through this
                alliance, ARL researchers were invited to assume the chairmanship of the multi-agency Air
                Quality Coordination Group (AQCG) of the Chesapeake Bay Program. Subsequently, the AQCG
                hosted two major atmospheric loadings workshops. The first was a scientific workshop at
                Baltimore, Md., in 1994 (the "Mt. Washington Workshop") which succeeded in defining and
                prioritizing immediate research needs to reduce existing uncertainty surrounding the atmospheric
                loadings issue. Six research areas were identified which refelected the need to:

                       1) conduct integrated (intensive and coordinated) monitoring at specified locations
                       2) work to improve atmospheric models
                       3) work to improve biogeochemical watershed models
                       4) improve emissions inventories
                       5) conduct process-oriented measurements to extend spatial representativeness
                       6) develop an extensive network of less intensive measurements.

                Through its leadership role in the AQCG, the success of the Mt. Washington Workshop
                strategically positioned ARL at the head of coastal nutrient/atmo spheric loadings research and
                assessment in the U.S. This position was strengthened in areas to the north and south of the
                Chesapeake Bay through a the second workshop of local, state, and federal policy makers
                conducted near Warrenton, Va., in 1995 (the "Airlie Workshop"). The Airlie Workshop
                concluded that there is need for:


                       1) a better understanding of how all atmospheric nitrogen species affect coastal
                       ecosystems and the related policy options
                       2) a cross-media approach to the atmospheric deposition and loadings problem
                       3) a coalition of interested parties extending from the north to the south of the potentially
                       affected eastern coast of the continental U.S., including both terrestrial and biological
                       interests as equals.

                Participants of the Airlie Workshop noted that the work needed is essentially multimedia,
                requiring attention by a consortium of workers rather than separate attention by specialists
                operating independently

                The ANICA program was designed as a targeted research program designed specifically to
                answer two particular questions: To what extent is the perceived problem due to deposition
                from the atmosphere and how can this understanding be extrapolated to other circumstances?
                After several years of work on the Chesapeake Bay, there is now a strong recognition of the
                importance of the atmosphere among the scientific community which deal with Atlantic coastal


                                                                xvi









               ecosystems. This is widely seen as an area of NOAA leadership, providing crucial guidance to
               EPA and the states to assure that regulatory controls on industry, agriculture, and waste
               treatment are considered in proper context with air quality controls. ANICA scientists have
               been exporting the lessons that have been learned. In fact, ANICA has been slowly expanding
               its horizons, with recent activity in all of the Great Lakes, Pamlico-Albemarle Sounds, the
               Gulf of Maine, and Tampa Bay.

               There have been approximately 35 studies around the world which addressed at least one aspect
               of atmospheric loadings, the majority of which were published since 1990. However, the
               measurement and modeling techniques used varies considerably between individual studies,
               making intercompari sons difficult. Though published estimates of the relative contribution from
               the atmosphere fall in an apparently constrained range (10-45%), the actual amount of
               atmospheric loadings vary widely depending, primarily, on the size of the waterbody and its
               watershed (as elucidated in this report). The uncertainties of the studies to date make it
               imperative that a better understanding be obtained of the processes that transport and deposit
               nitrogen to estuaries and coastal zones, and that the scope of this understanding be extended to
               areas along both the East Coast and the Gulf of Mexico which remain to be investigated. The
               most recent estimate of a 27% contribution of atmospheric deposition to total nitrogen loadings
               to the Chesapeake Bay (Chesapeake Bay Program, 1996), falls within the range reported for other
               estuaries (10-45%). Projections for the coming decades estimate that the atmosphere will become
               a more significant source of nitrogen loadings to coastal areas with anticipated increases in
               population and land development resulting in more mobile and new power plant emissions.

               The results and recommendations of the two Workshops listed above, and the underlying science
               behind atmospheric nutrient loadings to coastal areas are substantiated in the documentation that
               follows. This report will: a) review the underlying framework needed for understanding
               atmospheric loadings issues, b) present the specific research conducted by ARL and associated
               scientists under the ANICA background/measurement/modeling framework, c) summarize the
               current state of the science, and d) suggest the remaining research and management needs.




















                                                            xvii








               Chapter I
               INTRODUCTION

               Nitrogen in the Environment

               Though the availability of nitrogen normally limits biological productivity in coastal waters,
               overabundance of nitrogen is of concern in areas which have developed nutrient enrichment
               problems (i.e., eutrophication). In addition to increasing productivity, nutrient enrichment
               generally alters the normal ratios of nitrogen to phosphorus and to other elements such as silicon.
               This alteration may induce changes in phytoplankton community structure. Species which
               normally occur in low abundances may be favored, and, in some cases, toxic and/or noxious algal
               blooms may result. For the New England coast in particular, the number of red and brown tides
               and shellfish problems from nuisance and toxic plankton blooms have increased over the past two
               decades. Furthermore, in coastal areas with poor or stratified circulation patterns (e.g.,
               Chesapeake Bay, Long Island Sound) the "overproduction" of algae tends to sink to the bottom
               and decay, using all (anoxia) or most (hypoxia) of the available oxygen in the process, causing
               loss of habitat. In extreme cases, the increase in suspended matter due to overproduction
               decreases light infiltration, in turn causing a loss of submerged aquatic vegetation.

               There is some anecdotal evidence that nitrogen loadings have been an issue for longer than is
               generally appreciated. Paleoecological studies have revealed that there is a threshold up to which
               ecosystems can tolerate stress without obvious adverse consequences. Once this threshold is
               passed, however, the system may no longer be able return to its original equilibrium state. What
               is seen now could be a manifestation of a nitrogen problem that is just starting to be observable.
               In some cases (i.e., Chesapeake Bay, Long Island Sound), the overall nutrient loads have already
               tipped the balance and are producing obvious negative effects. The decline has been evident in
               seagrass and in fisheries productivity. Other estuaries may not have yet reached their balance
               points. Here, the emphasis will be on the role of atmospheric nitrogen compounds.

                       Note that thefollowing discussion uses a standardized distinction of the terms
                       "deposition" and "loadings. " DgPosition is the flux of nitrogenftom the
                       atmosphere to whatever surface is beneath. Indirect           is the portion of
                       nitrogen deposited onto the terrestrial watershed which is transmitted to the water
                       body itse4f, defined in terms offlow to the tidal waters. Deposition to the water
                       surface itse@f constitutes a direct loyding.

               Atmospheric Nitrogen

               The exact role of deposition of atmospheric nitrogen compounds as contributors to the decline of
               the health of coastal, estuarine, and inland waters is still uncertain. From the perspective of
               coastal ecosystem eutrophication, nitrogen compounds from the air, along with nitrogen from
               sewage, industrial effluent, and fertilizers, become a source of nutrients to the receiving ecosystem
               (Figure 1. 1). Eutrophication, however, is only one of the detrimental impacts of the emission of
               nitrogen containing compounds to the atmosphere. These compounds are also

















                                                                      Oak
                                                                                                    Aw'.. 4!                                                                                                          &3
                                                                                                                                                                                                                            :77-
                                                                                       1@11






                                                                NITROGEN                                    DAY                                WET                                                   CRY
                                                                 OXIDES                                  DEPOSITION                        DEPOSITION                                            DEPOSITIO14
                                                                                                             OF                                Of
                                                                                                          NITROGEN                           M17RATE                                                 Of
                                                               INOUStR@Al                                  OXIDES                              AND                                               AMMONIUM
                                                                 PLANTS                                                                    AMMONIUM                                                 IONS







                                                                                                                                                                         AMMONIA

                                                                                  NITROGEN
                                                                                  OXIDES
                                                                                                                                                                       AGRICULTUR
                                                                                                                                                 LEACHING
                                         miTRDGEw
                                                                                                                                                   tfwM
                                                                                  SEWAGE
                                           0X*fS
                                                                                  PLANTS                   4@,                                    FARMS
                                                                                                            41@


                                       URBANIZATION
                                                                                                                                  RIVER
                                          TRAFFIC                                                                               RUN OFF


                                                                                             SEWAGE
                                                                                             RUN-OFF                                                                 J-4

                                                                                                                                                             'A
                                                                                                                                                              41


                                                                                _4
                                                                                        ff
                                                                                                                                  g1W111                                                                             LEGEND

                                                                                                                                                                                                                     ATWSPHERIC
                                                                                                                                                                                                                     SVJRCES
                                                                                                                                                                                                                     ATMO&PHERIC
                                                                                                                                                                                                                     DEPM1704
                                                                                                                                                                                                                     TERRESTRIAL
                                                                                                                       OXYGIEN                                                                                       SWKES
                                                                                                                     DEPLETION                                                                                       RIVER(wf
                                                                                                                                                                                                                    It
                                              AL.
                                                                                                                                                                                                                OR



                                                                                                                                                                            @ONIA
                                                                                  m                                                                                         JLT@
                                                                                                                                                                               LIRE














                    Adopted Imm Acid4utm. June mo                                                                  KILLS SIENTRIC                                                                            0       SWWAS
                    Illustration: vailult vem"Ita                                                                       LIFf                                                                                 <>      RESULYS



                                                                                  Figure 1. 1 Atmospheric input of nitrogen to coastal areas









                significant contributors to decreased visibility (photochemical smog), acid deposition, and the
                production of tropospheric ozone.

                Atmospheric nitrogen compounds are ubiquitous. Diatomic nitrogen (ND is the dominant
                component of the atmosphere, and is comparatively unreactive. However, nitrogen is also
                present in the atmosphere in other chemical forms, some of which are relatively reactive. These
                fate of these nitrogenous compounds are the central topic for the present report.

                Most atmospheric nitrogen compounds (other than N2 and nitrous oxide [N201) fall into two
                categories: reactive nitrogen (composed primarily of nitrogen oxides), and reduced nitrogen
                (typically dominated by ammonia [NH3D. There are also organic nitrogen species that are
                typically referred to as a subset of reactive nitrogen. These chemical species arise in the
                atmosphere from the interaction between nitrogen oxides and biogenic or anthropogenic
                hydrocarbons. Though available evidence suggests that organic nitrogen originates from both
                anthropogenic and natural sources, speciation of organic nitrogen in the atmosphere is very
                poorly understood.

                The dominant source of the reactive oxides of nitrogen present in air over North America is high-
                temperature combustion (industry, power plants, automobiles, etc.). Nitric oxide (NO) generated
                by combustion (some also derives from natural biological processes) reacts quickly in the lower
                atmosphere generating nitrogen dioxide (NO2), which is then the dominant nitrogen oxide in the
                lower atmosphere in polluted regions. N02 Slowly deposits to the underlying surface, but a large
                proportion of it remains in the atmosphere where it is subjected to further chemical reactions.
                One of these reactions generates nitric acid vapor (HN03) which is easily and quickly deposited.
                Other reactions generate ozone; the ozone issue is therefore intimately related to the NOX
                (defined as NO + NOD question.

                The different atmospheric nitrogen compounds deposit differently, and once deposited they vary
                greatly in their ability to move through a watershed to affect the water body it contains. Specific
                influencing factors, in addition to the chemical form in which the nitrogen compound is
                deposited - reactive, reduced, or organic nitrogen species - vary widely based on proximity to
                sources, receiving waters, transformations in the atmosphere as well as on the land surface and
                transfer through the watershed via surface runoff (Paerl, 1996). Current estimates are that
                nitrogen oxides are the largest contributor to atmospheric nitrogen loads to coastal waters of
                eastern North America (40-60%), with ammonia (20-40%) and organic nitrogen (about 20%)
                also contributing significant amounts on an annual basis.










                                                                3









                Defining the Problem

                The initial impetus to investigate the importance of the atmosphere as a transport mechanism for
                nitrogen to estuarine areas was the 1988 Environmental Defense Fund (EDF) report (Fisher et al.,
                1988). Based on one year of measurements (1984), and assuming relationships between wet and
                dry deposition that are likely to be conservative, the authors estimated that one-third of the
                inorganic nitrogen entering the Chesapeake Bay comes from the atmosphere. There have been
                several subsequent efforts to quantify the atmospheric nitrogen loadings to the Bay. However,
                the assumptions used in these studies overshadow the results; the error associated with these
                estimates is typically quoted as being a factor of two or sometimes as much as a factor of about
                three. Further impetus was added to the atmospheric transport issue by the passage of 1990
                Clean Air Amendments that require:

                        The Administrator [of the Environmental Protection Agency], in cooperation with
                        the Under Secretary of Commercefor Oceans andAtmosphere, shall conduct a
                        program to identify and assess the extent of atmospheric deposition of hazardous
                        air pollutants (and in the discretion of the Administrator, other air pollutants) to
                        the Great Lakes, the Chesapeake Bay, Lake Champlain and coastal waters.

                In FY 1991 the NOAA Coastal Ocean Program initiated the Atmospheric Nutrient Input to
                Coastal Areas (ANICA) program. The ANICA program was proposed and managed through the
                NOAA Air Resources Laboratory (ARL). Through participation in the multiagency National
                Acid Precipitation Assessment Program, ARL scientists gained experience in creating data bases,
                techniques, and models usefW for evaluating the atmospheric sources of nitrogen to coastal
                waters. The long-term goal of ANICA was to:

                        Develop methodsfor assessing the importance of this atmospheric input, using the
                        Chesapeake Bay as afirst target of contemporary importance.

                ANICA was first proposed as a program to address scientific uncertainties involved in
                assessments of the kind put forward by the initial EDF report. When ANICA commenced, only a
                minimal data base existed with which to test computer models, and little experimental capability
                to obtain additional data. However, there was already a large, multi-state and federal
                organizational research effort addressing the Chesapeake Bay problem, the Chesapeake Bay
                Program. Most of attention was being directed at monitoring the flow of nutrients into the Bay
                via streams and rivers, monitoring changes in the Bay water quality and in its biology, and
                developing models to describe the interaction between such nutrient inputs and the recycling of
                nutrients within the Chesapeake Bay ecosystem. The ANICA program was designed to provide
                precisely the information that has previously been missing - direct experimental quantifications of
                areal deposition rates of atmospheric nutrients - and to introduce this new information into the
                models being developed.

                        ANICA was a targeted research program designed specifically to answer two particular


                                                                  4









                      questions: To what extent is the perceivedprohlem due to depositionftom the
                      atmosphere and how can this understanding be extrapolated to other circumstances?

               After several years of work on the Chesapeake Bay, there is now a strong recognition of the
               importance of the atmosphere among the scientific community which deal with Atlantic coastal
               ecosystems. The specialization of ARL is widely recognized as being crucial to this science. This
               is all widely seen as an area of NOAA leadership, providing crucial guidance to EPA and the
               states to assure that regulatory controls on industry, agriculture, and waste treatment are
               considered in proper context with air quality controls. ANICA scientists have been exporting the
               lessons that have been learned. In fact, ANICA has been slowly expanding its horizons, with
               recent activity in all of the Great Lakes, Pamlico-Albemarle Sounds, the Gulf of Maine, and
               Tampa Bay.

               In FY 1991 ANICA was implemented as a series of 23 specific tasks (listed in Appendix A).
               These tasks were divided into three general areas of research: background development,
               measurement, and model and data evaluation. At the end of FY 1995, nine (primarily background
               oriented) tasks were completed or implemented and six of the measurement and modeling phase
               tasks were initiated. Much of this progress was made through cooperative efforts with the
               Chesapeake Bay Program through the NOAA Chesapeake Bay Office. This report will: a)
               review the underlying framework needed for understanding atmospheric loadings issues, b)
               present the specific research conducted by ARL and associated scientists under the ANICA
               background/measurement/modeling framework, c) summarize the current state of the science, and
               d) suggest the remaining research and management needs.























                                                              5








               Chapter 2
               SCIENTIFIC FRAMEWORK


               Mass Balance Paradigm

               Biogeochemical cycling of nitrogen and contaminants on local, regional, and global scales is a
               complex system of emissions, transformations, dispersion, and deposition. This system is most
               complex in coastal environments, such as Chesapeake Bay, where landscape processes affect
               coastal processes and vice versa. Figure 2.1 is a simplified schematic that shows an idealized
               picture of a coastal ecosystem divided into four basic compartments or reservoirs, and the
               transfers among these compartments. This view provides a convenient basis for the current mass
               balance paradigm that attempts to quantify the transfers (loadings) between designated system
               components. To date, most emphasis has been placed on quantifying and identifying direct,
               well-defined emissions to each compartment (i.e., emissions from smokestacks, outfalls).
               However, the terrestrial watershed and the riverine and estuarine processes affect the transport of
               these emissions to the Bay both directly and indirectly.

               Atmospheric deposition includes wet deposition (through rain and snow) and dry deposition (as
               gases and aerosols). In each case, it also is necessary to consider whether the deposition is
               deposition to the water surface (i.e., direct deposition) or deposition to the watershed, with
               subsequent transport to the receiving stream (i.e., indirect deposition). Determining the
               magnitude of deposition is necessary but not always sufficient for addressing management issues.
               To deal efficiently with such issues requires an understanding of both the sources of atmospheric
               nitrogen (e.g., long-range versus short-range transport, emission source type, and associated
               meteorological conditions) and the relative magnitude of the atmospheric depositional loadings
               compared to that of all other sources of nitrogen surface waters.

               Sources and Emissions


               Most atmospheric nitrogen compounds (excluding N2 and N201 which are relatively inert in the
               lower atmosphere) fall into t W@o categories: reactive nitrogen, sometimes referred to as oxides of
               nitrogen or odd nitrogen, and reduced nitrogen (typically dominated by ammonia [NH3])' Some
               organic nitrogen species arise in the atmosphere from the interaction between nitrogen oxides
               and biogenic or anthropogenic hydrocarbons, and are thus typically referred to as a subset of
               reactive nitrogen. The relative portions of the different fonns nitrogen can take--i.e., nitrate,
               nitrite, ammonia, dissolved organic nitrogen--vary wideiy based on proximity of sources to
               receptors, receiving waters, and atmospheric transformations as well as on the land surface and
               transfer through the watershed via surface runoff (Paerl, 1996). Current estimates are that
               reactive nitrogen is the largest contributor to atmospheric nitrogen loads to coastal waters (40-
               60%), with ammonia (20-40%) and organic nitrogen (0-20%) also contributing significant
               amounts.























                                                                         internal
                                                                          Cycling         Anthropogenic

                                                                                             Sources



                       Anthropogenic                                  Atmosphere
                          Sources
                                                          Transfer                       Transfer


                                                                                                     Terrestrial          internal
                            In er al       Tidal Waters                    Transfer                  W2tershed            Cycling
                            C cling

                                                           Transfer                       Transfer
                                                                          Surface                                     Anthropogenic
                                                                         2nd Ground
                                                                           Watem                                         Sources

                                               Anthropogenic

                                                   Sources
                                                                             internal
                                                                             Cycling











                     Figure 2. 1. A simplified schematic of the coastal nitrogen cyle.

                                                                              8









              Reactive Nitrogen: Nitrogen Oxides

              Reactive nitrogen compounds (primarily oxides of nitrogen) are emitted to the atmosphere
              through both natural and anthropogenic pathways, overwhelmingly (95%) as nitric oxide, or NO.
              Natural sources of NO include emission from so.,ls and generation by lightning; dominant
              anthropogenic sources include emission from automobiles, power plants, and biomass burning.
              On a global basis, anthropogenic and natural sources d reactive nitrogen are approximately equal
              in strength (Table 2. 1). In North America, and especially in northeastern North America, the
              overwhelming majority of nitrogen oxides is of anthropogenic origin.

              Table 2.1. Budget of global NO. emissions by source (Teragrams [I Tg = 10" g] N yr-1).
              Adapted from Watson et al., 1992.
              r-sources of NO,, = NO + NO,                                     Tg N yr-1

               Fossil fuel combustion                                             14-28

               Lightning                                                          2-20
               Microbial activity in soils                                        5-20

               Biomass burning                                                    3-13

               Oxidation of ammonia                                               0-10

               Oceans                                                             < I

               Stratospheric input                                                <1

              Once in the atmosphere, NO is transformed via a multitude of secondary reactions to form higher
              oxides of nitrogen. Of the reactive nitrogen components, FIN03 and particulate nitrate (p-NO3-)
              are believed to dominate nitrogen deposition (Huebert and Robert, 1985; Huebert et al., 1988;
              Meyers et al., 1989). However, deposition of N02 also may be significant in near-urban
              environments where concentrations are high; its spatial and temporal extent is an important
              question being addressed. Peroxy acetyl nitrate (PAN) is a compound which arises from the
              reaction of N02 with the photo-oxidation products of both natural and anthropogenic
              hydrocarbons. PAN is efficiently produced in polluted urban air due to the reactive nature of
              many anthropogenic hydrocarbons. Few studies of PAN deposition have been conducted.
              Smullen et al. (1982) views PAN deposition as insignificant; however, Singh (1987) states that it
              may be important near polluted urban and semi-rural locations.

              Reduced Nitrogen: Ammonia and Ammonium

              Ammonia is emitted into the atmosphere through both natural and anthropogenic pathways.
              Natural sources of NH3 include microbial decomposition of organic nitrogen compounds in soils
              and ocean waters and volatilization from animal and human wastes. Anthropogenic sources


                                                            9









                include the manufacture and release of commercial and organic fertilizers during and after
                application and fossil fuel combustion. Human activities such as manure management and
                biomass burning exacerbate emissions from otherwise natural processes. Two estimates of the
                global emission budget are presented in Table 2.2. The uncertainties apparent in these estimates
                are related to the intrinsically local nature of ammonia emissions, which make regional estimates
                highly difficult to construct.

                Ammonia is a highly reactive compound and has a short residence time in the atmosphere. It is
                primarily emitted at ground level and quickly deposits to the area near its source unless it reacts
                with other gaseous chemicals (e.g., S02, HN03) and converted to ammonium (NH4) aerosol
                (Langland, 1992; Asman, 1994). Ammonium can be transferred regionally as an ammonium salt
                (e.g., ammonium nitrate [N03NH4D and is the primary contributor to ammonium concentrations
                measured in precipitation.

                Table 2.2. Budget of global NH3emissions by source (Tg N yr-'). Adapted from Schlesinger
                and Hartley, 1992 (numbers in parenthesis indicate range); Dentener and Crutzen, 1994.

                 Source                   Schlesinger and Hartley, 1992        Dentener and Crutzen,
                                                                                         1994

                 Domestic. animals                 32.0 (24-40)                          22.0

                 Sea surface                        13.0 (8-18)                          7.0
                 Undisturbed soils                  10.0 (6-45)

                 Wild animals                                                            2.5

                 Vegetation                                                              5.1

                 Fertilizers                        9.0 (5-10)                           6.4

                 Biomass burning                     5.0 (1-9)                           2.0

                 Human                                  4.0

                 Coal combustion                        2.0

                 Automobiles                            0.2

                  Total                            75.2 (50-128)                         45.0


                Organic Nitrogen

                To date, studies of atmospheric nitrogen deposition have almost exclusively addressed diss olved
                inorganic nitrogen (reactive and reduced nitrogen). Because of the paucity of reliable
                measurements, the historical variability in analytical techniques and results, and the current lack
                of suitable and uniform analytical measurement techniques, only limited work has been reported


                                                               10









                on the deposition of organic nitrogen. In fact, only wet deposition of organic nitrogen has been
                addressed.


                This dearth of information is becoming widely recognized by the scientific community and is
                receiving increased attention. Recent reports (Gorzelska et al., 1992; Milne and Zika, 1993;
                Cornell et al., 1995; Scudlark et al., 1995) suggest that organic nitrogen is a significant fraction
                of the total nitrogen measured in precipitation. Various estimates for the relative flux of organic
                versus total N via wet deposition range from <1 0% to >60%. These recent data suggest that the
                contribution of the unresolved organic fraction may significantly augment the atmospheric
                deposition of nitrogen to coastal waters. However, in addition to the lack of dry deposition data,
                there remain many conceptual questions related to source identification and bioavailability of
                deposition organic nitrogen.

                Current analytical techniques are unable to speciate specific organic N compounds measured in
                precipitation. Therefore, it is difficult to determine relative contributions of biogenic versus
                anthropogenic sources. Cornell et al. (1995) speculated from their open ocean studies that the
                higher concentrations in continental rains, the need for a relatively large sea-surface fractionation
                to sustain a marine source, and the isotope results all suggest that a continental source is more
                likely. They also suggest that industrial combustion sources could contribute to dissolved
                organic nitrogen formation via reactions of soot with NO,, and NH3.

                Deposition Flux Estimates

                The process by which atmospheric nitrogen is transferred to terrestrial and water surfaces is
                generally termed atmospheric deposition. Deposition is divided into two categories: wet and dry.
                Wet deposition involves the incorporation of gaseous or particulate nitrogen into cloud/rain
                formations (including fog), and the subsequent deposition of this water onto an underlying
                surface. Dry deposition involves the exchange of gaseous and particulate nitrogen between the
                atmosphere and a surface either through settling or by impact deposition.

                Wet Deposition

                With the exception of organic nitrogen compounds, the analytical techniques for major nitrogen
                species in precipitation are well established and can be considered fairly routine. Most of the
                current debate over the appropriate approach for measurement of wet nitrogen deposition is
                centered around field sampling protocols (i.e., the process by which the sample is collected and
                transported to the analytical laboratory). The question of precipitation sample collection
                frequency remains a topic of active debate. Many precipitation sampling networks have adopted
                a one-week sampling protocol as a reasonable compromise between high data quality and
                reasonable operational costs. Shorter collection intervals are generally not required to quantify
                trends in the wet deposition of stable chemical compounds; ecosystems do not usually respond
                rapidly to changes in deposition. On the other hand, longer-term sampling (e.g., monthly) is
                usually avoided because of difficulty in maintaining adequate quality control over precipitation









                samples which remain in the field for long periods of time. Overall, weekly sampling provides
                data which are quite adequate for long-term trend detection and ecosystem input assessment, but
                which are of limited use in coupling with meteorological models, and are often useless for
                process-oriented studies. In addition, while many ions in precipitation are relatively stable for
                periods of weeks or months, nitrogen compounds are more chemically labile and so require more
                frequent sample collection and analysis intervals to ensure sample stability and, by extension, a
                proper assessment of their environinental impacts. Thus, as the questions posed to the research
                community become more difficult, the demand for daily sampling is increasing.

                Wet deposition is chaotic in nature, which makes estimating short-term patterns difficult.
                However, review of nitrogen deposition data shows that deposition at one site is much the same
                as at a neighboring site when long-term averages are evaluated. This fact allows meaningful
                areal (isopleth) maps of wet deposition and nitrogen chemistry to be constructed using long-term
                data.


                Dry Deposition

                When considering total atmospheric deposition of nitrogen to a given watershed, the largest
                uncertainties are associated with the ability to estimate the spatial distribution of dry deposition.
                Enough is known about the processes that control dry deposition to permit deposition to be
                estimated on a point-by-point basis. However, in contrast to wet deposition, long-term averaging
                does not reduce differences between sites; it clarifies the differences. This fact makes
                understanding large-scale dry deposition patterns difficult. Thus, it is not possible to extract
                meaningful site-specific dry deposition data from large-area, time-averaged data without detailed
                consideration of the site in question.

                The mechanisms that control dry deposition are tied to biological and land surface features that
                are highly variable from one specific location to another. The problems with estimating dry
                deposition are further complicated by the fact that the capability for long-term, continuous
                monitoring is limited by both scientific and economic factors. In an effort to circumvent these
                problems, investigators have used mathematical models to estimate the distribution of dry
                deposition. However, even detailed, site-specific models used to estimate dry deposition from
                field measurements at a site under investigation are limited in their ability. Comparative
                measurement and modeling studies have shown that site-specific models - at good sites and in
                selected conditions - perform well a little over half the time. The reasons for the inadequacies
                are known to be associated with hard-to-quantify surface characteristics (e.g., leaf wetness,
                moisture conditions). These issues are not new, and they have been the subject of considerable
                debate. Efforts to resolve these problems have been made and continue to be made, but a
                universally acceptable spatial dry deposition model is still far distant.

                The primary assumption made by published loading studies is that dry deposition is equal to wet
                deposition. This assumption is necessary because there is no independent measurement basis for
                estimating areal dry deposition rates. This same lack of data also means that there is no existing


                                                                 12









               evidence that this estimate is grossly misleading. It would seem obvious, considering the vastly
               dissimilar atmospheric and surface mechanisms that cause dry and wet deposition, that equality
               between them would be an extremely unlikely and quite fortuitous finding, in practice.

               Loadings to Chesapeake Bay

               Direct Loadings

               Studies have explored the idea that atmospheric deposition may contribute a significant
               proportion of phytoplankton nitrogen demands in coastal areas (Paerl, 1985; Paerl, 1988; Paerl et
               al., 1990). Fogel and Paerl (199 1), for example, have estimated that 20-5 0% of annual new
               nitrogen demands for Pamlico-Albemarle Sound may be supplied by direct atmospheric
               deposition to the water surface (wet and dry). Furthermore, there have been two recent, and
               opposing, papers published on the effects of wet-deposited nitrogen effects on phytoplankton off
               Bermuda (Owens et al., 1992; Michaels et al., 1993). It is not currently known if the over-land
               measurements are representative of over-water deposition, but it is suspected that this is not the
               case. To investigate this question, a daily precipitation chemistry site was established on Smith
               Island, Md., in late 1995. This site will provide the first time series measurements of over-water
               wet deposition along the East Coast. Estimates of wet deposition to the Chesapeake Bay surface
               range from 3.45 to 4.2 Gg N03-N yr-1 (I Gigagram = I * 10'g) (Fisher and Oppenhierner, 1991;
               Hinga et al., 1991; Tyler, 1988).

               Although the air/surface exchange of nitrogen has been estimated for most nitrogen species over
               open ocean (Galloway, 1985; Duce et al., 1991), these rates may not apply to coastal situations in
               that coastal areas involve different meteorological processes (NRC, 1992).
               Through the use of instrumented Chesapeake Bay Observing System (CBOS) buoys, owned by
               the University of Maryland, estimates of nitrogen (HN03, N02, NH4) dry deposition rates to the
               Bay surface have been developed (Valigura, 1995). These estimates corroborate those given by
               other investigators to some extent, but still cover a wide range of values, from 745 Mg y-' (I
               Megagrain = I * I O'g) to 2.24 Gg y-. From this data set, calculations were performed to
               determine the effect of atmospheric dry deposition on phytoplankton dynamics. This analysis
               demonstrated that dry-deposited nitrogen may provide 10% of the annual "new nitrogen"
               demands in Chesapeake Bay, and that individual events could supply up to 75% of the new
               demands for periods of several days (Malone, 1992; Owens et al., 1992).

               Indirect Loadings

               The greatest uncertainty in the quantification of loadings to coastal areas such as Chesape    ake Bay
               is how much is transferred through the terrestrial watershed to the surface waters, and how much
               is subsequently transported downstream to the Bay. Nitrogen activity within a given watershed
               depends on the amount of soil nitrogen, historical acidic deposition, physical characteristics of
               the soil, site rainfall and temperature characteristics, the elevation and slope of the land, and the
               type and age of the vegetative cover (see Figure 2.2). These characteristics vary at all scales,


                                                                  13





















                                                                           Oxidation,
                                                                           Fixation by lightning


                                      N    NITROGEN OXIDES                                      NH 3 NITROGEN OXIDES

                                              Industrial                                                 tinospheric           Volitalization
                                              fixation
                                                               .9         CROPREMOVAL                  deposition
                                              (fertilizer)                                      I
                                                                                                     IF
                                                               nn

                                   0                                                                                   SOIL

                                   ;4
                                                                                Ingestion        ANIMAL
                                                              PLANT N                            N


                                                                                                          Death and wastes







                                                 Assitnilation



                                          NO'      Nitrification     NH  +       Arnmonification         ORGANIC N
                                              3                          4                                                                EROSION


                                                                                                                                        --IN- RUNOFF
                                                PARENT MATERIAL
                                                                                               MI          N                                 SUB-
                                                                                                                                          SURFACE
                                                                                                                                            FLOW






                                     LEACHING







                      Figure 2.2. A simplified schematic of the terrestrial nitrogen cycle,


                                                                                      14








               making it difficult to determine the general fate of atmospherically deposited nitrogen over any
               area of significant size. However, using these criteria, a classification scheme for forested sites
               has been developed to evaluate a site's potential to retain/leach nitrogen (Melillo et al., 1989;
               Johnson and Lindberg, 1992). For example, sites in the Chesapeake Bay watershed generally fall
               into the zero-to-low leaching classifications (Hunsaker et al., 1994).

               A central difficulty in all discussion of watershed retention relates to the use of average values in
               assessments. Watershed retention assumptions used in the Chesapeake Bay studies (see above)
               are presented in Table 2.3. These estimates vary over several orders of magnitude, primarily
               because of the lack of adequate evaluation data. Assembling an adequate understanding of long-
               term behavior when the processes involved are ftmdamentally episodic is one of the major
               challenges, and represents an area in which contemporary

               Table 2.3. Watershed retention values (in % of nitrogen loading) used in Bay loading
               studies to date (numbers in parenthesis indicate range tested).

                 Land Use                 Tyler, 1988              Fisher and               Hinga et al., 1991
                                                              Op enheimer, 1991

                 Forest                   95.2-100.0            80.0 (51.0-100.0)            80.0 (25.0-95.0)

                 Pasture                  93.7-99.96            70.0 (51.0-90.0)             80.0 (25.0-95.0)

                 Crop and                 76.0-99.97                  70.0                    0.0 (45.0-75.0)
                 Resi ential__@           62.0-95.3              35.0 (0.0-70.0)            25.0 (10.0-50

               models are sorely deficient. It is apparent from measurements that the majority of the
               atmospheric wet deposition occurs during a few episodes (Dana and Slinn, 1988; Fowler and
               Cape, 1984), such that the wet-deposited nitrogen (as well as previously dry-deposited nitrogen)
               is deposited directly to or can flow quickly into the surface waters without intermediate reduction
               in concentration.


               Total Loading Estimates

               The role of atmospheric transport in providing an important path for nitrogen to estuarine areas
               was publicized in the EDF report (Fisher et al., 1988). Based on one year of measurements
               (1984), and assuming relationships between wet and dry deposition that are likely to be
               conservative, the authors estimated that one-third of the inorganic nitrogen entering the
               Chesapeake Bay comes from the atmosphere. There have been several subsequent efforts to
               quantify the atmospheric nitrogen loadings to Chesapeake Bay (Fisher and Oppenhiemer, 1991;
               Hinga et aL, 1991; Tyler, 1988). The approach taken in these studies can be divided into two
               components: estimating wet and dry deposition and estimating watershed retention. However,
               the assumptions used in these studies overshadow the results; the error associated with these
               estimates is typically quoted as being a factor of 2 or sometimes as much as a factor of about
               three. The most recent and best estimate of atmospheric nitrogen loads (i.e., 27% of the annual
                      I

                      d































                                                                 15








                load) to the Bay were developed through the Chesapeake Bay Program (CBP; Chesapeake Bay
                Program, 1996). This estimate falls within the range reported for other estuaries (10-45%), see
                below. To estimate wet deposition, the CBP combines output from a regression model developed
                from NADP weekly and daily precipitation chemistry measurements with data from the high
                density NOAA rainfall network. This approach yields daily estimates of rainfall to 74 sub-basins
                of the Chesapeake Bay watershed. Dry deposition is assumed to be equal to wet deposition for
                over-land areas and 44% of wet for over-water areas. Indirect atmospheric loadings from the
                terrestrial watershed are estimated with the CBP Watershed Model, a bulk-parameter model. In
                the coming decades, it is predicted that the atmosphere will become a more significant source of
                nitrogen loadings to the Chesapeake Bay with anticipated increases in population and land
                development resulting in more mobile and new power plant emissions (Fisher et al., 1988;
                Pechan, 1991).


                Loadings Intercomparison Between Coastal Waters

                There have been approximately 40 studies around the world which addressed at least one aspect
                of atmospheric loadings, the majority of which were published since 1990. However, the
                measurement and modeling techniques used vary considerably between individual studies,
                making intercomparisons difficult. Table 2.4 presents a summary of selected studies performed
                along the U.S. East Coast which are comparable in broad terms. There were two criteria for
                selection based not on scientific merit but on the approach and information content. These
                criteria required that the results were either:

                  - published in a credible peer reviewed journal
                  - advocated by a major management organization (e.g., an EPA National Estuary Program)

                These studies can be divided into two groups: those which considered both direct and indirect
                nitrogen loads and those which considered only direct loads. Table 2.4 confirms the common
                belief that the amount of atmospheric nitrogen input is related to the size of a waterbody and its
                watershed.


                It is interesting to note that although the percentages listed in Table 2.4 have an apparently
                constrained range (10-45%), the numerical estimates of atmospheric loadings vary widely. This
                point highlights the danger of using percentages as the basis for large scale management
                decisions. Many areas along both the East Coast and the Gulf of Mexico remain to be
                investigated. The uncertainties of the studies to date make it imperative that a better
                understanding be obtained of the processes that transport and deposit nitrogen to estuaries and
                coastal zones.









                                                                 16






                     Table 2.4 Current estimates of nitrogen loading (Millions of Kg) to selected coastal waters attributed to direct and indirect atmospheric input.
                                                              Surface area      Surface area       Deposition         Deposition onto        Atm. load           Total atm.      Total load from       % load        Ref.
                                                              watershed         of tidal           onto the           the tidal waters       delivered from      Load            all sources           from the
                                                              (km')             waters (kM2)       watershed          (i.e., direct load)    the watershed                       (incl. Atm.)          atm.
                                                                                                                                             (i.e, indirect
                                                                                                                                             load)            I

                        Narragansett Bay (R.I.)               4708              328                4.2                .3                     .3                  .6              5                     12            1
                        Delaware Bay                          36905             1846               53                 3                      5                   8               54                    15            2
                        Long Island Sound                     43481             4820               45                 5                      7                   12              60                    20            3
                        Albemarle-Pamlico Sounds              59197             7754               -39                3.3                    6.7                 10              23                    44            4
                        Chesapeake Bay                        165886            11400              175                16                     29                  45              170                   27            5
                        New York Bight                        50107             38900              69                 54                     8-                  62              164                   38


                        Rhode River (Md.)                     33                4.9                -                  .005                   -                   .005            .012                  40          -6
                        Waquiot Bay (Ma.)                     -70               -8                 .062               -                      .0065               .0065           .022                  29            7
                        Flanders Bay (N.Y.)                   83                39                 -                  .027                   -                   .027            .36                   7-            8
                        Delaware Inland Bays                  800               83                                    .28                    -                   .28             1.3                   21            9
                        Sarasota Bay (Fl.)                    524               135                                   .16                    -                   .16             .6                    26            10
                        Patuxent River (Md.)                  2393              137                                   .22                    -                   .22             12.6                  13            11
                        Newport River Coastal Waters          340               225-1600           -                  .095-.68               -                   .095-.68        .27-.85               35-80         4
                        (N.C.)

                        Narragansett Bay (R.I.)               4708              328                -                  .4                     -                   .4              9                     4             12
                        Choptank River (Md.)                  1779              361                -                  .57                    -                   .57             1.54                  37            11
                        Guadalupe Estuary (Tx.)               -                 551                -                  .31                    -                   .31             4.2-15.9              2-8           13
                        Potomac River (Md.)                   29940             1210               -                  1.9                    -                   1.9             35.5                  5             11
                        Tampa Bay (Fl.)                       6216              1031.              -                  1.1                    -                   1.1             3.8                   28            1@4
                        Massachusetts Bays                 I  -                 3700           -   -                  1.6-6                  -                   1.6-6           22-30                 5-27          15
                     1. 11inga et al., 1991; 2. Scudlark and Church, 1992; 3. Long Island Sound Study, 1996; 4.Paerl and Fogel, 1994-1 5. Chesapeake Bay Program, 1996; 6. Correll and Ford, 1982-1 7. Valiela et al.,
                     1996; 8. Peconic Bay NERI 9. Delaware Inland Bays NEP; 10. Sarasota Bay NEP; 11. Boynton et al., 1995; 12. Nixon et al., 1995; 13. Brock et al., 1995; 14. Tampa Bay NEP; 15.
                     Massachusetts Bays NEP.









                Chapter 3
                REMAINING ISSUES

                As the ANICA Program progressed it became clear that there is a large need for information
                dissemination regarding atmospheric nitrogen issues. ANICA scientist were directly involved in
                three such informational projects. In early 1994, a literature synthesis entitled, "Atmospheric
                Deposition of Nitrogen and Contaminants to the Chesapeake Bay and Its Watershed" (Valigura et
                al., 1995) was completed for and published by the Chesapeake Bay Program Scientific and
                Technical Advisory Committee. The second effort was a scientific workshop (referred to as the
                Mt. Washington Workshop) entitled, "Atmospheric Loadings to Coastal Areas - Resolving
                Existing Uncertainties," and was held in June 1994 in Baltimore, Maryland. The third effort was a
                policy-oriented workshop (referred to as the Airlie House Workshop) entitled, "Airsheds and
                Watersheds - The Role of Atmospheric Nitrogen Deposition," and was held in October 1995 in
                Warrenton, Virginia.

                The Literature Synthesis

                The literature synthesis concluded with the following recommendations for steps to reduce
                uncertainties associated with prediction of atmospheric loadings. Depending on chemical species,
                the current uncertainties in estimates of atmospheric loading of nitrogen and contaminants to
                Chesapeake Bay and its watershed range from ï¿½ 20% to orders of magnitude. However, if ways
                are to be found to reduce such atmospheric deposition, it is essential that the processes be better
                understood and that the quality and reliability of the estimates be improved. Accordingly, it is
                recommended that the following steps be initiated. These steps are not comprehensive or specific
                suggestions, but one list of key areas of uncertainty which will serve as discussion points in future
                deliberations. They are:

                 ï¿½     Conduct monitoring and research experiments focused on improving measurements and
                       modeling techniques to further understand and quantify the emission cycles of the key
                       chemical species

                 ï¿½     Develop and perform nitrogen speciation experiments including on organic nitrogen and
                       ammonia compounds; subsequently, conduct intensive studies of the dry deposition rate of
                       nitrogen compounds from air crossing the watershed zone of the Chesapeake Bay region

                 ï¿½     Investigate the effect of localized contaminant deposition in both urban and near-urban
                       environments; specifically, develop estimates of surface-water loadings attributable to
                       urban runoff and investigate the temporal and spatial distribution of NOY deposition

                 ï¿½     Establish integrated monitoring sites of atmospheric emission and deposition; initial focus
                       should be on the concurrent measurement of the chemistry, intensity, and duration of
                       precipitation and streamflow events











                  ï¿½     Establish over-water precipitation chemistry sites and compare the results with those from
                        land-based precipitation chemistry sites

                  ï¿½     Establish common data bases to process and store data that can be used as input to models
                        and/or to test specific predictions of the models; this effort should include emission
                        inventories (anthropogenic and natural), wet and dry deposition rates at specific locations
                        to be covered by the model, spatial distributions of soil and atmospheric moisture,
                        chemical concentration distributions, and relevant meteorological supporting data

                  ï¿½     Clarify the role of urban areas as a source of atmospheric contaminants to surface waters',
                        conduct research on sampling methods for small particle deposition and source attribution
                        for organic contaminants

                  ï¿½     Investigate the bioavailability of materials deposited from the atmosphere; conduct
                        exposure studies to learn how chemical speciation influences exposure

                The Mt. Washington Workshop: Developing Science Priorities

                To focus attention on the problem, and to take a first step towards ordering scientific programs to
                reduce existing uncertainties in the quickest manner, a meeting of active scientists from different
                contributing disciplines was conducted at Mt. Washington, Maryland, on June 29-3 0, 1994. The
                workshop was sponsored by the Scientific and Technical Advisory Committee and the Air Quality
                Coordination Group of the Chesapeake Bay Program (CBP). The gathering included
                representation from the Great Lakes research community and federal and state agencies.

                The challenge given to the workshop was simple -- to construct a prioritized listing of practical
                studies that would make the greatest impact on reducing the current uncertainty in estimates of
                the contribution of atmospheric deposition to declining aquatic ecosystem health. The workshop
                was constructed to produce such an ordering through meetings of interacting working groups and
                concluding with extensive plenary discussion of the working group conclusions.

                The listing that resulted is summarized below and substantiated in the documentation that follows.
                It was concluded, however, that scientific investigations that are already under way are making
                considerable progress; in essence, any new efforts should be arranged to build on existing
                programs rather than risk new starts that compete with older ones. It was also concluded that
                there is a general need for improvement in the scientific programs already ongoing for all
                measurement methods, models, and evaluation of pollution reduction strategies.
                The emphasis of the workshop discussion was on all nitrogen species, toxic chemicals, trace
                metals, precipitation chemistry, airborne aerosols, and supporting meteorological investigation. In
                every one of these cases the general call for a new focus applies, although with different weights
                according to the particular emphasis. The workshop put priorities in the following order:



                                                                  20









               Priority I -- Conduct intensive, coordinated integrated monitoring at special locations
                      within the watershed, with wet deposition, dry deposition, and local catchment area
                      characterizations. It was concluded that the single most limiting factor in assessing the
                      adequacy of current models is the lack of quality data on actual deposition within the
                      target watershed. Until an integrated monitoring station is operational, there will be no
                      comprehensive data set for evaluating model performance.

               Priority 2 -- Work to improve existing atmospheric models. In brief, there are many limitations
                      of current models, especially including their limited grid size (smaller grid cells are
                      desired) and their inability to handle orographic and chemical factors that are likely to be
                      of critical importance.

               The above top priorities reflect the workshop's recognition that current models are likely to he
               misleading, but that the extent of any errors cannot be judged because the key observations of
               deposition are not yet made.

               Priority 3 -- Improve biogeochemical watershed models. The workshop recognized the
                      important role of watershed chemical retention and emphasized the need for close linkages
                      with the appropriate expert scientific community.

               Priority 4 -- Improve emissions inventories and projections. It was noted that assessments of
                      atmospheric deposition are necessarily at the mercy of emissions estimates and that such
                      estimates are currently highly imperfect both in the adequacy of reporting requirements
                      and the spatial resolution used to report the emission values.

               Priority 5 -- Conduct process-oriented measurements to extend vertical and spatial
                      meteorological and chemical concentration coverage and to quantify
                      representativeness. The models that are now needed for assessment purposes need more
                      advanced input data than do the simpler models used in early assessments. As time
                      progresses and as these models evolve further, input data requirements will increase. It
                      was concluded that measurement programs to provide the data required by the models
                      should be established.


               Priority 6 -- Establish an extensive array of less intensive measurements. This item follows
                      from Priority 1. in essence, a nested network is envisioned with a small number of
                      Priority I intensive stations supporting a denser array of simple stations designed to
                      provide improved spatial resolution for some selected variables.

               The workshop considered the needs for uncertainty reduction in two distinct contexts:

                0     to build confidence by separate attention to individual parts of the source-emission-
                      dispersion-deposition-delivery-effects process



                                                              21









                  0     to test the accuracy of overall understanding and model predictions

                In addition, it was pointed out that a quality assurance program to consolidate the existing
                network of wet and dry deposition (for nitrogen) sites is currently not in place. Therefore,
                deposition must be monitored, keeping ecological considerations in mind.

                The meeting noted that current assessments are almost entirely based on large-grid model outputs
                without the benefit of actual deposition observations to the coastal areas that are thought to be
                affected. Consequently, the top research priority was associated with the need for one (or
                preferably several) Integrated Research Sites where actual deposition measurements could be
                made at locations where supporting ecological data are collected. Other priorities reflected the
                current state of science and the relative importance of different areas of uncertainty. In general,
                each participant was encouraged to think in terms of the science at the highest possible level
                leaving personal agendas and agency perspectives behind.

                        In general, it was concluded that work should continue on the development of
                        atmospheric models which produce output offiner-grid resolutionfor the airshed
                        affecting the Chesapeake Bay. Ancillary work shouldfocus on studying the influence of
                        key urban areas on the atmospheric loading to the watershed and tidal waters and to
                        de ve lop spatially-de tailed we t/all and dryfall atmospheric-deposi tion input data for the
                        watershed, tributaries, and coastal waters off the Bay mouth. Coordination of this
                        activity appears to be an appropriate role for the CBP Air Quality Coordination Group.

                Finally, it was noted that no part of the workshop deliberations should be considered relevant only
                to the Chesapeake Bay. It was specifically concluded that the ordering listed above would be
                equally appropriate for other affected coastal areas.

                The Airlie House Workshop: Identifying Implicationsfor Management

                A workshop was conducted at Airlie House Conference Center, Warrenton, Virginia, in October
                1995, where leading scientists and key policy and regulatory officials assembled to explore
                mechanisms by which air and water pollution control programs can work together to help protect
                coastal ecosystems. The focus of the workshop was on atmospheric nitrogen compounds, but
                many of the conclusions would apply equally well to other pollutants occurring in the air, such as
                toxic chemicals, trace metals, and persistent organic compounds. In all such instances, the
                atmosphere constitutes a resource that is shared among many different coastal jurisdictions and
                between the air and water regulatory communities.

                Scientific uncertainty has historically been an obstacle to proactive management response. The
                Airlie House Workshop reported here benefitted from the ability to draw upon scientific
                conclusions and recommendations about the coastal atmospheric deposition problem that were
                developed following the Mt. Washington Workshop. Given this information, participants
                concluded that scientific uncertainty has been reduced to the extent that some new or modified


                                                                 22









                regulations and controls can be justifiably implemented.

                Atmospheric deposition of nitrogen species is recognized by all East Coast estuarine programs as
                either a significant contributor to estuarine eutrophication or a mechanism of possible concern.
                The region from which the atmospheric nitrogen pollution arises is much larger than the water
                surface that is potentially affected, and even much larger than the watershed that drains into it.
                The extent of "airsheds" are now starting to be recognized. The Chesapeake Bay airshed,
                depending on the definition used, is up to 600,000 square miles    (106 km) in area, extending
                upwind of and bordering the water body itself Emissions from an airshed of this magnitude affect
                more than a single estuary. For example, the emissions from the Chesapeake Bay airshed may
                affect the entire coastline such as from the Carolinas to New York. Thus, airsheds constitute an
                important "shared resource" that must be recognized. Reductions in emissions in airsheds benefit
                many downwind ecosystems, and assessments of the benefit of such reductions must take all
                benefitting water bodies into account, not just one single ecosystem that is especially favored.

                It was noted that some reductions in nitrogen emissions have already been made as a result of the
                controls mandated by the Clean Air Act Amendments (CAAA) of 1990. However, participants
                expressed concern that political pressures may not permit the full emissions reductions proposed
                by the CAAA to be realized. Water quality scenarios need to be weighed carefully, since some
                assume full implementation of the CAAA controls whereas others disregard them. Current
                assessments of projected water quality (and of ecosystem viability) do not take the additional
                controls proposed by the Ozone Transport Comn-fission into account. It should be noted that the
                CAAA, and therefore the present document, concentrates on the deposition of the products of
                emissions of nitric oxide (NO) into the atmosphere - essentially nitrates (N03-). Very little is said
                about the role of products of ammonia emission - essentially ammonium compounds - which are
                of considerable importance but about which relatively little is known. The role of organic
                nitrogen species is also not emphasized; these are poorly understood but are known to be strongly
                influenced by biological sources that cannot be regulated.

                The following five recommendations summarize the conclusions drawn by the participants of the
                Shared Resources Workshop.

                  ï¿½     Efforts to resolve scientific uncertainties associated with the quantification of atmospheric
                        deposition and the resulting loadings should be continued. The priorities identified at the
                        1994 Mt. Washington workshop serve as a useful reference for planning future work.
                        Future research should also focus on quantifying atmospheric nitrogen fluxes to the
                        coastal ocean and characterizing the biochemical cycle of organic nitrogen through the
                        Chesapeake Bay watershed.

                  ï¿½     Although there is uncertainty in many areas, enough is known to determine a general
                        direction for action. Managers and regulators should move forward and not wait for all of
                        the uncertainties to be resolved. Accuracy is possibly most needed when weighing costs
                        of controls versus benefits.



                                                                 23










                 ï¿½    A set of basic information for use in explaining the cause for concern about atmospheric
                      deposition and waterbody effects to the public, politicians, regulators, etc., should be
                      generated. It was considered likely that a single set of basic material could be used as the
                      core of issue-related material addressing current understanding about emissions,
                      atmospheric deposition loadings by watershed and water body, areas of greatest
                      uncertainty, etc. This would promote cooperation and coordination across the
                      organizations involved and avoid sending mixed messages.

                 ï¿½    A cross-media approach to quantifying atmospheric deposition and resulting loadings
                      needs to be developed. Greater cooperation across issues, estuaries and bays, scientific
                      disciplines, and governmental units is essential. Barriers to greater cooperation should be
                      identified and eliminated.


                 ï¿½    In order to assure that such coordination continues, a future meeting of the present kind
                      (but with representation from an enlarged group of organizations) should be held in about
                      a year's time.

               The workshop concluded that there is need for: a) a better understanding of how all atmospheric
               nitrogen species affect coastal ecosystems and of the related policy options, b) a cross-media
               approach to the atmospheric deposition and loadings problem, and c) a coalition of interested
               parties extending from the north to the south of the potentially affected East Coast of the
               continental U.S., including both terrestrial and biological aspects as equals. Theworkthatis
               needed is essentially multi-media, requiring attention by a consortium of workers rather than
               separate attention by specialists operating independently.

                      The workshop resulted in a clear and loud callfor more cooperation across different
                      issues, estuaries and hays, scientific disciplines, and state andjederal agencies.
                      Outreach to state andjederal agencies, non-government organizations, indust?Y, and the
                      puhlic at large, is critically needed

               Even in these days of calls for cooperation and integration, the close association required in the
               present context calls for attention by teams of workers that has classically been difficult to arrange
               and maintain, In the present case, there seems to be no option; the importance of the issue and its
               environmental sphere of influence combine to make it more important than ever before to build a
               working alliance of researchers and regulators addressing the atmospheric deposition issue. The
               role of atmospheric deposition is known to be important in many areas of concern, the
               atmosphere does indeed constitute a genuine shared resource, and a coordinated attack on the
               problem is recognized as an essential element rather than a desired goal. The Airlie House
               workshop was a start along a road to more extensive cooperation specifically intended to
               lead to more accurate assessments of atmospheric deposition to all East Coast estuaries and
               to arrange for the results to be presented in an optimal format for assimilation by policy
               and regulatory processes.



                                                              24








                Appendix A
                LIST OF ANICA TASKS

                TASK 3. 1. 1: Bring together data from various wet deposition networks, examine quality
                        assurance, and assemble appropriate information into a single data set for further analysis.

                TASK 3.1.2: Assess deposition of organic nitrate using specially-equipped monitors at a single
                        site, and later using a number of sites as determined on the basis of the initial exploration.

                TASK 3.1.3: Sort deposition data according to event precipitation rates and amounts, and
                        reported in a probabilistic manner as well as in terms of long-term average deposition
                        rates.


                TASK 3.1.4: Set up new sampling sites. Sites will be sought that provide precipitation data
                        representative of the central portions of the Chesapeake Bay, and precipitation records
                        will be compared against values interpolated from terrestrial isopleths. If no existing sites
                        that are suitable can be located, then special sites will be set up.

                TASK 3.1.5: Central-bay sites will be sought for data on wet deposition on an event basis. If no
                        existing sites can be located, then special sites will be set up to test the hypothesis that wet
                        deposition to the water surface of the Chesapeake Bay can be estimated adequately by
                        interpolating terrestrial wet deposition isopleth maps.

                TASK 3.2. 1: Measure HN03 concentrations routinely at the Wye River site.

                TASK 3.2.2: Conduct intensive intercompari sons against annular denuder methodologies to
                        evaluate the quality of HN03 data reported by the NOAA filterpack techniques.

                TASK 3.2.3: Conduct intensive air chemistry investigations of other airborne nitrogen
                        compounds, especially N02.

                TASK 3.2.4: Assess the spatial variability of air chemistry across the Chesapeake Bay watershed
                        by use of Regional Acid Deposition Model or some better modeling capability
                        benchmarked against the Wye River data.

                TASK 3.2.5: Initiate Dry Deposition Inferential Method operation at the Wye River site and
                        commence archiving deposition velocities for HN03 and N02-

                TASK 3.2.6: Obtain satellite imagery and derive land-use categorized depiction of the
                        Chesapeake Bay catchment area.

                TASK 3.2.7: Produce modeled wind fields and related meteorological data for the Chesapeake
                        Bay catchment area.

                                                                  A-1










               TASK 3.2.8: Produce areal representations of deposition velocities for the entire catchment area.

               TASK 3.2.9: Benchmark the DDIM predictions for the Wye River sites for both N02and 14N03.

               TASK 3.2. 10: Instrument water buoys to monitor the stability regime over the water surface.
                       Use the data obtained to compute deposition velocities for N02and HN03appropriate for
                       the water surface itself.


               TASK 3.2.11: Derive probabilistic representations of deposition velocity regimes in order to
                       express water surface deposition of N02and HN03 (primarily) as a function of month and

                       season.


               TASK 3.2.12: Combine terrestrial deposition velocity fields (from Task 3.2.8) with air
                       concentration fields (Task 3.2.4) to provide monthly and seasonal quantifications of the
                       dry deposition of nitrogen species to the terrestrial watershed.

               TASK 3.2.13: Combine aquatic deposition velocity fields (from Task 3.2.11) with air
                       concentration fields (Task 3.2.4) to provide monthly and seasonal quantifications of the
                       dry deposition of nitrogen species to the water surface of the Bay.

               TASK 3.2.14: Conduct intensive field studies as required to evaluate the accuracy and adequacy
                       of these quantifications.

               TASK 3.3. 1: Assemble a data base giving weekly, monthly, and seasonal values representing:
                       a) dry deposition to the water body, b) wet deposition to the water body, c) dry deposition
                       to the surrounding watershed, d) wet deposition to the surrounding watershed,
                       e) precipitation quantity, f) riverine nutrient input, and g) representative Bay water
                       concentrations of related nutrients.


               TASK 3.3.2: Conduct multiple lag correlation analyses of the assembled data so as to deduce
                       statistical relationships among the variables as a function of season.

               TASK 3.4. 1: Conduct aircraft assessments of the horizontal fluxes of nitrogen species in air
                       flowing across the Chesapeake Bay watershed as a ftinction of altitude and season.

               TASK 3.4.2: Utilize the ANICA data set to improve regional deposition models for future use in
                       assessing atmospheric nutrient deposition to vulnerable ecosystems.









                                                               A-2








               Appendix B
               ESTIMATING DEPOSITION: SPECIFIC STUDIES


                               A Regression Approachfor Estimating Precipitation Chemistry

               Background

               The modeling subcommittee of the Chesapeake Bay Program was interested in an immediate
               improvement to the input of atmospheric nitrogen from N03- and NH4' into the Bay Watershed
               model via precipitation (i.e., wet deposition). Data from 15 National Atmospheric Deposition
               Program (NADP) sites within and around the Chesapeake Bay watershed were used to investigate
               statistical relationships between the amount of precipitation and the concentration of nitrogen
               species in the rainfall. Using these statistical relationships and NOAA precipitation records,
               improved methods for estimating wet deposition nitrogen loading to the Chesapeake Bay were
               developed. Although statistical methods are not complete solutions to the problem both are an
               improvement over older approaches used by the Bay Program.


               Method


               Two datasets were used to develop the proposed methods; one from the NADP and another from
               the MAP3S precipitation chemistry network. The primary network collecting precipitation
               chemistry data is coordinated by the NADP. The NADP was initiated in 1978 and is still in
               operation, consisting of some 193 sites funded on a site-to-site basis by a variety of organizations.
               These sites have identical siting and sampling criteria, and all data are analyzed in the same
               laboratory and using the same methodology. Precipitation samples are collected for one week
               with wet-only samplers and then sent to the Illinois State Water Survey for analysis. The
               proposed methods were either developed from or directly use the dataset collected by the NADP
               during 1984-1992 from the 15 NADP sites in and nearest to the Bay watershed.

               The MAP3S network was begun in 1976 and collected data until 1990 from nine sites along the
               Northeast United States. The MAP3 S sites, like the NADP sites, had coordinated siting,
               sampling, and analytical criteria to allow intercompari sons between sites. However, precipitation
               samples were collected on a event basis using wet-only samplers. The regression method was
               tested using datasets collected from 1978-1986 by three A4AP3S sites located within the
               watershed.


               The regression method is based on the basic logarithmic relationship between amount of
               precipitation (mm ha-') and the N03- and NH,' concentrations in the precipitation (mg mnf'). A
               relationship was developed using weekly data collected over an eight year period at the 15 NADP
               sites, Due to the weekly sampling protocol of NADP, the data were quality controlled by
               selecting those data where the precipitation event occurred only on the last day of the weekly
               sample. These data can be interpreted as daily samples, analyzed by the NADP network. Using
               this criterion, 265 samples were obtained from the approximately 5020 samples collected from the

                                                            B-1









                   NADP sites, and these samples were used to develop the regression model. T                        he coefficients of
                   the final regression model are shown in Table B. 1. The final model was chosen from a list of
                   variables which included log of precipitation, month, month', sine of month, cosine of month,
                   latitude, and longitude. This model was then tested (estimated vs. measured) on 8 years of event
                   based data (approximately 1800 samples) collected independently at three MAP3S sites in and
                   around the watershed (Table B.2). Estimates of annual total nitrogen (N03+ NH4) deposition
                   were within 20% of that measured. On an individual event basis, estimates generally fell within
                   a factor of two.


                   Table B.1. Regression coefficients of the selected models (p values).
                                                     I       ln(ppt)             month              month' -T Latitude                   R'

                      N03          -1.289 (.0811)        -.3852 (.0000)                         -.0037 (.0000)          .0744           .41
                                                                                                                       (.0001)

                                   -1.226 (.0000)        -.3549 (00LO)       .3966 (.0000)      -.0337 (.0000)                          .31

                   Table B.2. Testing statistics for total nitrogen (N03-+ NH4') deposition estimates (in
                   percent).

                      Site Location & Network             Mean Error         Mean Error in      Error Variance       Maximum/Minimum
                                                            in Annual            Event              in Event             Event Errors
                                                            Estimates          Estimates           Estimates


                      State College, Pa.                        -19                 5                   77                 998/-80
                      MAP3S

                      Charlottesville, Va. MAP3S                -18                 21                  95                 649/-94

                      Lewes, Del.           MAP3S               -2                  44                 339                 973/-81

                      15 Sites             NADP                 -19                 17                  65                 490/-


                   Implementation

                   The Watershed Model used by the Chesapeake Bay Program is a bulk parameter model which
                   divides the Chesapeake Bay watershed into 70 segments related to small-and-intermediate-scale
                   watersheds. Each of the watershed segments has an associated hourly precipitation data record.
                   To develop wet deposition estimates for the entire watershed, the regression method was
                   employed using the precipitation data records. The result was a daily wet deposition data set for
                   each segment. Figure B. 1 shows the average annual deposition estimates by watershed segment.








                                                                                B-2










                  Dissolved laorpnic N
                  Wet Deposition (Kgs/Ha)


                                 0 - 5A


                                  4.4-4.9


                                 4
                                     -4.4
                                  .0


                                  3.6-4.0


                                  3.1-3.6





















                   Figure B. i. Spatial distribution of the wet deposition of inorganic nitrogen across the Chesapeake
                   Bay watershed.


                                                                             B-3









                                             Wet Deposition Over Chesapeake Bay

                Background

                The Mt. Washington Workshop listed as a top priority the need to provide actual data against
                which to compare model predictions and assessments. Wet-fall deposition of nitrogen has
                traditionally been estimated by creating isopleths of deposition between over-land precipitation
                chemistry sites. It is currently suspected that over-land wet deposition data do not adequately
                represent direct wet deposition rates to large waterbodies, primarily due to the complex
                meteorology in coastal areas such as Chesapeake Bay. The dearth of data concerning wet
                deposition of nitrogen to the Chesapeake Bay surface waters prompted ANICA scientists to
                propose the establishment of an overwater precipitation chemistry site. The site was established
                in September 1995. The resulting data set will provide a first opportunity to evaluate the
                effectiveness of over-land based sampling for estimating over-water deposition to large
                waterbodies, and to test model outputs against actual data obtained over the target area. The
                project will provide measurements of wet deposition of nitrogen species to the waters of the
                Chesapeake Bay using a wet-only collector erected on Smith Island in the lower Bay. Samples
                will be collected and analyzed using the National Atmospheric Deposition Program's (NADP)
                Atmospheric Integrated Research Monitoring Network (NADP/AIRMoN) field and laboratory
                protocols.

                Wet deposition is the result of scavenging of airborne particles and trace gases by clouds, fog,
                condensation, and by the collection of other particles and gases by hydrometeors as they fall from
                the clouds in which they were formed. These processes are sometimes referred to as "in-cloud"
                and "sub-cloud" scavenging, or alternatively as "rainout" and "washout" respectively. In general,
                scavenging of nitrogen by precipitation processes is highly efficient. Concentrations in collected
                precipitation samples are easily measured, although rapid chemical reactions occurring after
                samples are collected can give rise to misleading concentration data if analysis is not performed
                soon after collection, or if adequate steps are not quickly taken to chemically preserve sample
                integrity. The most widely accepted method for precipitation sampling is the wet-only collector,
                which is designed to remain covered until precipitation is occurring and returns to covered
                condition after rainfall. To quantify trends in the wet deposition of stable chemical compounds,
                short-term deposition data are generally not required; ecosystems do not usually respond rapidly
                to changes in deposition. However, long-term (e.g. monthly) sampling is usually avoided
                because of difficulty in maintaining adequate quality control over samples that remain in the field
                for long periods of time. Many networks have adopted a one-week sampling protocol as a
                reasonable compromise between high sample quality and reasonable operational costs. Weekly
                sampling provides data that are quite adequate for long-term trend detection and for assessing
                ecosystem inputs for stable ions, but are limited in their utility for coupling with meteorological
                models, and are often useless for process-oriented studies. In addition, while many of the
                heretofore studied ions are relatively stable for periods of weeks or months, problems such as
                nitrogen loading to estuaries and other coastal ecosystems require even shorter period collections
                as well as special handling to adequately ensure sample stability. As the questions posed to the


                                                               B-4









               research community become more difficult, the demand for daily sampling is increasing.

               With the exception of organic nitrogen compounds, the analytical techniques for major nitrogen
               species in precipitation are well established and can be considered fairly routine. Most of the
               current debate over the appropriate approach for measurement of wet nitrogen deposition is
               centered around field-sampling protocols (i.e., the process by which the sample is collected and
               transported to the analytical laboratory).

               Objectives

               Field protocols for the major networks around the world have been developed over 40 years of
               experimentation and debate. In the United States, the national network is the NADP. As of
               1993, NADP consisted of 193 weekly sampling sites, and seven daily sampling sites (affiliated
               with the NOAA AIRMoN Program). The NADP weekly and AIRMoN sites have identical siting
               criteria for precipitation chemistry measurements, and all data are analyzed in the same
               laboratory using similar methodologies. The one major difference is that the weekly NADP
               samples are filtered at the laboratory to stabilize sample chemistry; AIRMoN samples are
               stabilized through continuous refrigeration beginning immediately after collection. In general,
               the site operator will visit the site on a daily basis. By evaluating the collector function record on
               each visit, the operator will check for precipitation during the previous 24 hours, excessive
               bucket lid openings other than during precipitation, and if more than one week has passed
               without a bucket change. If one of these events has occurred the site operator will change the
               bucket. If a sample has been collected, field chemistry (as appropriate) will be done, and the
               sample will be prepared and shipped to the central NADP analytical laboratory (the Illinois State
               Water Survey). The sample will be analyzed using NADP/AIRMoN guidelines. Quality
               assurance/quality control will be performed before the,data is entered into the NADP/AIRMoN
               electronic database housed at Colorado State University.

               The goal is to produce two years of data on wet deposition of central nitrogen compounds
               (nitrates, nitrites, and ammonium) to the waters of the Chesapeake Bay, with daily time
               resolution.



















                                                             B-5









                                               Wet Deposition of Organic Nitrogen

                Background

                Wet-deposition fluxes of inorganic N (primarily N03- and NH4') to coastal eastern North America
                are well characterized and the corresponding dry-deposition fluxes, though substantially less
                certain, are reasonably constrained. However, the deposition fluxes of organic nitrogen (N.@)
                compounds are very uncertain. The lack of reliable measurement techniques and the associated
                dearth of field data are responsible for the poor state of current understanding. Various estimates
                for the relative flux of organic versus total N via wet deposition range from <10% to >60%. The
                four types of methods currently being evaluated for atmospheric investigations of N.rg are those
                commonly used to analyze total nitrogen (NT) in seawater. These methods include: persulfate wet
                chemical oxidation (most widely used), UV photo-oxidation, high-temperature oxidation, and
                high-temperature catalytic oxidation. In all cases, Norg is determined by subtraction from NTthe
                sum of the ammonium, nitrate, and nitrite concentrations. Recent findings (Suzuki, 1993) suggest
                that the agreement between methods is reasonably good. The pyro-chen-tiluminescent method
                currently shows the most promise due to advantages of rapid and automated analysis, and small
                sample (100 pl) requirements. Other techniques are labor intensive and require approximately
                100 ml of sample -- potentially a very serious constraint when studying small-volume precipitation
                events. With support from ANICA and the EPA Great Waters Program, the University of
                Virginia (UVa) has developed a pyro-chemiluminescent technique to reliably measure total N and
                organic N in dilute aqueous solutions, and is currently investigating sampling artifacts and
                assessing sample stability.

                Analytical Method

                Wet-only precipitation is sampled using an Aerochern Metrics collector. As indicated above,
                sampling techniques and sample stability assessments are currently under investigation; specific
                sampling and handling procedures may be refined as a result of these investigations. Preliminary
                results suggest that organic N in precipitation is very unstable; to generate representative data, it
                may be necessary to subsample precipitation events in stainless steel containers and to preserve
                samples immediately against microbial degradation.

                Samples are analyzed for NTusing a modified ANTEK model 7000 analyzer. All N compounds
                are converted to NO via high-temperature combustion, and NO is subsequently quantified via
                chemiluminescence. Over the past several years, this approach has been used with increasing
                frequency to analyze atmospheric and marine samples. As described below, however, potentially
                serious problems may compromise resulting data. Few data generated using this approach have
                been published in the peer-reviewed literature.

                Initial studies by UVa revealed that data generated using a range of "standard" conditions
                (recommended for application by the manufacturer and by other users) and an off-the-shelf
                instrument (configured for application by the manufacturer) were subject to large, significant


                                                               B-6









               positive and negative artifacts; all compounds were not recovered with equal efficiency. The
               direction and magnitude of bias varied as a function of the chemical composition of the test
               solution and the corresponding method of calibration. "Standard" operating conditions also
               resulted in poor reproducibility and low precision. Because the organic fraction is calculated by
               difference, reliable results require highly precise and unbiased analytical data; performance using
               "standard" configurations and procedures was unacceptable.

               UVa subsequently modified the instrument and the analytical procedure and are now able to
               generate unbiased and highly precise results at expected concentrations for a representative range
               of test compounds including NH4', N03-, urea (CH4N20), and ethylamine(CAN).

               Proposed Research

               UVa has applied to several funding sources for funds to conduct a subsequent and more
               comprehensive analysis (i.e., over space and time) of NT, inorganic, and (by difference) N.,g in wet
               deposition to eastern North America and to the western North Atlantic Ocean using this new and
               well-characterized technique. NH4' and N03-will be measured by automated colorimetry and ion
               chromatography, respectively, using standard procedures for the Global Precipitation Chemistry
               Project (GPCP) and the Air/Ocean Chemistry Experiment (AEROCE); resulting inorganic data
               are precise to approximately +/-4% and unbiased. The above approach would comprise the most
               reliable and precise method currently available for measuring the wet-deposition flux of NTand
               N.,g.

                       Ae goal of this investigation willprovide thefirst comprehensive assessment of the
                       atmospheric deposition of N,,rg in eastern North America and the northwestern Atlantic
                       Ocean thereby providing critical information concerning atmospheric N inputs to the
                       Chesapeake Bay and other coastal waters in polluted regions.



















                                                              B-7










                           Dry Deposition Inferential Method (DDIM) Measurements at Wye, Md

               Background

               As of 1992, the NOAA Dry Deposition Inferential Measurement (DDIM) network was composed
               of I I dry deposition sites. In December 1992, a DDIM site was established at Wye, Md.,
               constituting the second NOAA/DDIM site in the Chesapeake Bay watershed, the first being at
               Leading Ridge, Pa. Both Chesapeake Bay sites are collocated with NADP weekly and
               NADP/AIRMoN stations, thereby allowing estimates of total nitrate nitrogen deposition. The
               DDIM network was designed to develop methods of deposition estimation at a three core sites,
               and then expand these methods throughout the DDIM network. Concentrations of sulphur
               dioxide, particulate sulfate, particulate nitrate, and nitric acid are measured via heated inlet
               filterpacks. The filterpack samples are exposed for one week with an air flow of three liters per
               minute. Meteorological variables (e.g., wind speed and direction, standard deviation in wind
               direction, global radiation, air temperature and humidity, surface wetness, precipitation) are also
               measured at each site, thereby allowing dry deposition to be estimated via the inferential
               technique. These techniques were developed over several years of intensive research at the core
               network prior to network initialization. However, the core sites are not representative of the
               coastal environment, and it was not clear, a priori, if the DDIM method would work in a coastal

               areas.


               The DDIM Method


               This method estimates dry deposition using air concentration data and calculating deposition rate
               based on measurement of atmospheric and surface factors or by consideration of land-use type
               (i.e., forest, cropland, etc.). For nitrogen compounds, this method is applicable only to those
               species that can be considered to be depositing at all times -- primarily HN03, but also particulate
               nitrate (p-N03-; Meyers et al., 1989). These chemicals allow surface deposition fluxes to be
               treated as essentially uni-directional (toward the surface), and thereby inferred from atmospheric
               concentration data, provided that the characteristics of the situation in question can be formulated
               in terms of properties that can be measured. In this technique, the transfer coefficient is termed a
               deposition velocity (Vd). The value of Vd is computed from field observations and used to
               estimate the dry deposition rate (Fd) from air concentration measurements C:

                                                          I'd =PVdC'

               Estimation of Vd involves a balanced consideration of atmospheric and particle physics,
               chemistry, and biology (see Figure B.2). The deposition velocity is normally considered as an
               overall "conductance" of the pollutant to the surface and, consequently, the inverse Of Vd is
               described as the total resistance to pollutant transfer, RT. This basic "resistance analogy"
               commonly specifies four major resistances:




                                                              B-8























                                                                                                      Airbourne Source



                                                                                   Large Particles                             Gases


                                           Aerodynamic                          Settling
                                            Factors
                                                                               Turbulence                                          Turbulence
                                                                               -- I
                                                                           Therinophores!;
                                           Near-Surface                              I
                                             Phorelic                      Electrophoresis
                                              Effects                                I
                                                                           DiffusiopF
                                                                                   I
                                                                            Stephan 1-71 w                                       Stephan     ow
                                                                                      ,01rIc E S a

                                           Quasi-Laminar                       Impaction
                                                                                     I
                                               Layer                           Interception
                                              Factors
                                                                        Brownian Diffusion                 - - - Brownian Diffusion








                                                                                     1                                                     1
                                                                               Orientation                Stoniata                   Wetness
                                            Surface                                  I                          I                          I
                                           Properties                          Flexibility                Waxiness                  Chemistry
                                                                                     I                          I                          I
                                                                               Smoothness                 Vestiture                 Emissions
                                                                                     I
                                                                               Motion                     Exudates






                                                                                                          Recepto




                      Figure B.2.            A representation of the sequence of mechanisms influencing the deposition from
                                             the atmosphere of trace gases (right side) and diffusing particles (left). Not all
                                             processes are identified. (adapted from Ilicks, 1986).
                                                                                                                 s



















                                                                                            B-9









                     an aerodynamic resistance, k that is a function only of atmospheric properties, such as
                     atmospheric turbulence and stratification;
                     a near surface "quasi-lan-@inar" resistance, Rb, that relates to the diffusion of pollutants
                     across the near-surface layers where molecular and Brownian properties are important;
                     a surface canopy, or residual, resistance, R,, that expresses the consequences of the
                     chemical, morphological, and biological processes influencing pollution adsorption or
                     "capture" by the surface itself,
                     gravitational settling, V, as modified by particle growth due to humidity interactions, is
                     used to incorporate particle dynamics into R.., and Rb(note that Vg is zero for gases).

              It is important to note that these resistances are conceptually independent so that they may be
              considered in series, thereby allowing the calculationOfVd:

                                        Vd @ [(R. + Rb+ R, + RaRb Vd-1 + Vg]

              There are three major areas of concern with this approach: i) the accuracy of the concentration
              data, ii) the accuracy of the deposition velocity, and iii) the applicability of the deposition velocity
              concept in the experimental circumstance. The first two issues have been and are currently being
              addressed by a variety of air chemistry and research programs; current estimates of the error
              associated with them is about ï¿½ 25%. The third concern points to the conceptual uni-
              directionalityof Vdas a potentially severe limitation, important to both modeling and measuring
              programs. The Vdconcept is derived from relationships developed to describe turbulent transport
              to surfaces that are effectively horizontal and uniform. When considering non-uniform landscape
              patterns (i.e., checkerboard forest and croplands), edge effects will be of considerable importance.
              The mechanisms involved will be largely advective, once again making the problem three
              dimensional. This is an over-riding problem in that a one-dimensional description is used by
              almost all existing deposition models. It is especially a problem in the coastal zone where the
              majority of the landscape is broken and varied.


              Results


              Data from the Wye site have been collected, archived and analyzed from December 1992 through
              December 1994. These analyses revealed deposition rates which were similar to those reported
              from other DDIM stations in the region, and that dry deposition of nitrate (HN03/P-NO3)was
              approximately 46% of total nitrate deposition. However, further analysis indicated that the filter
              pack (chemical concentration) data may be suspect due to artifacts associated with high ammonia
              and sea salt concentrations over Maryland's Eastern Shore. Additional studies are planned to
              quantify errors associated with filterpack measurements in coastal areas.









                                                         B-10









                                       Atmospheric Nitrogen Speciation Measurements

               Background

               Reactive nitrogen compounds are emitted to the atmosphere through both natural and
               anthropogenic pathways, overwhelmingly (95%) as NO. Nitrogen oxide is highly reactive in the
               atmosphere and is transformed into a variety of higher oxides of nitrogen. Understanding these
               transformations and their by-products is crucial to the understanding of tropospheric ozone
               production, air quality, and atmospheric nitrogen effects on coastal ecosystems. Through
               ANICA, NOAA/ARL developed the ability to perform research quality measurements of nitrogen
               oxides and their associated chemicals (i.e.,03 and CO). Furthermore, several speciation
               experiments were performed, two of which are relevant to Chesapeake Bay. The first was a
               cooperative effort with the University of Maryland's Meteorology Department in September 1993.
               The second experiment, in August 1995, consisted of shipbourrie nitrogen speciation
               investigations as part of the Atmospheric Exchange Over Lakes and Oceans (AEOLOS)
               Experiment in Chesapeake Bay. The University of Maryland performed a experiment
               investigation into NO,,/NO, speciation from September 2-28, 1993 at Wye, Md., making
               simultaneous measurements of NO, NOx, NOY,03- Concurrently, NOAA/ARL's Hybrid Single
               Particle Langrangian lntegrated Trajectory (HY-SPLIT) model was used to run 24 hour "back
               trajectories in an attempt to determine the geographic history of the air-mass being sampled.

               Analytical Method

               The accurate determination of NOY speciation is an intricate and complex undertaking which
               necessitates the use of several experimental techniques. NO, NO,, and NOY may all be measured
               with a single instrument, a standard ozone chenffluminescence NO detector (Luke and Valigura,
               1996). The ozone chemiluminescence (CL) technique is, for all practical purposes, the only
               widely used method of measuring atmospheric levels of nitric oxide. The technique is a
               continuous measurement method based upon the detection of photons released through the
               reaction of ozone (0,) with NO. Ambient air is drawn into a reaction vessel at a controlled flow
               rate by a mechanical vacuum pump and mass flow controller. Ozone is generated within the
               instrument as a reagent by passing a flow of pure, dry oxygen through a high-voltage electrode; it
               is then directed into the reaction vessel where it reacts with NO in the sample flow to form N02.
               Approximately 10% of the N02 "ormed in an electronically excited state (N02*) (e.g., Fontijn et
               al., 1970), a fraction of which relaxes back to the ground state by emitting a photon. The
               broadband emission spectrum ranges in wavelength from about 600-3000 mn, and the photons are
               detected by a red-sensitive photomultiplier tube (PMT). Because thermal emission and
               amplification of electrons by the PMT can overwhelm detection of the low-energy photons, it is
               necessary to cool the PMT with dry ice or thermoelectric junctions to minimize background noise.
               As long as ozone is in great excess in the reaction chamber, the amount of light detected at the
               PMT is proportional to the mixing ratio of NO in the sample air stream and the instrument
               pumping speed. NOx (NOx = NO + N02)is measured by first passing the air through a photolytic
               conversion cell, where a fraction of the ambient N02 'Sphotolyzed to NO and subsequently


                                                              B-11









               detected. NOY is measured by first passing the ambient sample air through eithera gold or
               molybdenum catalyst which efficiently reduces all NO, species to NO. The detector provides
               real-time, continuous output of NO, NO, and NOY concentrations.

               Given the strong photochemical links between NOY and 03, ozone concentration was measured
               using continuous UV photometric technique. Carbon monoxide was also be measured using a
               continuous technique, gas filter correlation (GFC), non-dispersive infrared (NDIR) absorption.
               Carbon monoxide is one of several compounds which participate in the photochemical oxidation
               of NOX and production of ozone. It is also an excellent tracer of incomplete combustion and can
               provide information on the severity of pollution events.


               The HY-SPLIT Model


               Grouping of trajectories into common spatial clusters is a method to classify different
               meteorological situations associated with pollutant transport from different source regions. These
               regimes can then be identified with specific periods of time, such as those associated with an air
               sample or deposition measurement.

               As with many models using gridded meteorological data fields, the trajectory calculations (HY-
               SPLIT model; Draxler, 1992) follow a geometric approach. The trajectories' 3 -dimensional
               motion is computed from the u,v, and (o (dp/dt) component winds output and archived every two
               hours from NOAA!s National Centers for Environmental Prediction's Nested Grid Model (NGM).
               The time series consists of twice-daily consecutive model forecast fields from +2 h after
               initialization to +12 h. Archives of the 2-hour fields are available since 1991 from the National
               Climatic Data Center (NCDC; reference TD-6140). The fields are given on a 180 kni polar
               sterographic grid at 10 model sigma levels from 0.982 to 0.434. There are about 4 levels within
               the boundary layer.


               Results


               The fraction of NOY which is most reactive is NO,,. Air influenced by fresh NO,, emissions will by
               characterized by high NO,,/NOY ratio. Table B.3 shows that the mean ratio for the project was
               about 0.65. During the day this ratio varied from a broad morning maximum of 0.80 to an
               afternoon minimum of <0.50. Polluted air arriving at the site mixes downward to the surface
               when the sun rises and instability occurs, but photochemistry is still weak. Later in the day,
               photochemistry dominates and NO, is converted to higher oxides of nitrogen (e.g., HN03). In the
               late afternoon to early evening, FIN 03 deposition combined with NO emissions from the soil lead
               to an increased ratio later in the day.

               Results from the HY-SPLIT model back trajectories gave some insight into the origin of
               pollutants arriving to the Wye site. Figures B.3 and BA show the outputs from six model runs;
               three low NOx ("clean") days and three high NOx ("dirty") days, respectively. Note that the air
               highest in NOx came from Pennsylvania or the Baltimore/Washington region, while low-NOx air



                                                             B-12

















                                                                                                                                                                                                                                                                                                                                                                                               ............













                                                                                                                                                                                                         . . . .  .........
                                                                                                                                                                         . . .     . . . . . . . . . ..                                                          .. . . . . . . . .. . .
                                                                                                                  ............
                                                                                                                                                                                                      ....................                                  .........
                                                                                                                                                          ..........                                                                  %





                                                                                                                                                                                                          .................




                                                                                                                                                                                                                                       d

                                                                                               ...............
                                                                                                                     ...........             ........
                                                                                                                                                                     ...............             ...........



                                                                                                                            .......                 ...
                                                                                                                                                                       ...... .....








                                                                                    W"y   .                          ..........












                                                            Figure B.3.                                             The 48-hour back trajectories from Wye, Md. corresponding to periods of low
                                                                                                                   measured NO, concentrations.


                                                                                                                                                                                                                                      13- 13






































                                                                                                                                                                             ..........    ...

                                                                                                                                                                                                         t
                                                                                                                                              .............

                                                                                                                                          .... .........
                                                                                                                                                                                            .............
                                                                                                                          ..........
                                                                                      ..........   .........                               ................            ... ...           : ..............
                                                                                                                    ......                                                                               Ilk








                                                                 ............
                                                                              ...................
                                                                                                                   ..................................
                                                              ................                    ......
                                                                                                             .............



















                                           Figure BA                                 The 48-hour back trajectories from Wye, Md. corresponding to periods of high
                                                                                     measured NO. concentrations.


                                                                                                                                                                          B-14









               originated from the more rural south. Table BA contains the mean and observed maximum NO,,
               mixing ratios during each 24-hour trajectory run.

               Table B.3. Basic031CO, daytime NO, N02, NOY, and No,,/NOY statistics for data collected
               at Wye, Md. between September 2-28, 1993.
                                        1 CO ppb I Nodavppb I NO, ppb                NO, ppb          N

                 Mean            28          211         0.56          4.20            7.67             0.65

                 Median          26          196           1           3.05            5.24             0.65

                 Minimum          1     1     95         -0            0.52            1.07             0.26

                 Maximum         76          628         14.78        20.68           48.66             - 1

                 M d 67%      1 11-45      142-274 1 0.03-0.76       1.21-7.39       2.97-12.14      0.40-0.84
                 Mid 90%         3-59      117-334 1 0.02-2.14       0.90-11.78      2.01-20.81      0.33-0_95_J

               During the trajectory for September 11, the lowest CO/NO,, and CO/NOY ratios were observed.
               The trajectory came from the industrial midwest. Relatively low CO mixing ratios coupled with
               high NO_, mixing ratios suggest emissions from coal-burning power plants. Concurrent highS02
               mixing ratios would further suggest that power plant emissions influence this particular
               trajectory.

               Table BA Mean and maximum observed NOx mixing ratios for "Clean" and "Dirty" days
               measured at Wye, Md. between September 2-28,1993. -

                         Day             Mean Concentration (Ppb)         Maximum Concentration (ppb)

                                               "Clean" Days

                         9/13                      1.47                            2.58

                         9/26                      2.01                            3.69

                         9/27          11          2.03                            3.28

                                               "Dirty" Days

                         9/11                      7.49                            16.38

                         9/12                      6.03                            10.55
               @L-9/22                J1           7.49                            20.89


                                                              B-15










                       Intercomparison of Filterpack and Denuder Techniquesfor HN03Measurement

                Background

                Despite the importance of HN03 tooverall N deposition, there are remarkably few suitable
                techniques for its measurement. Real-time instruments typically employ spectroscopic methods of
                detection, and are often complex, expensive, and power consuming. Integrated, or batch
                techniques (e.g., filterpacks) have been widely used due to their simplicity and low cost.
                However, the time resolution of these techniques is limited. Furthermore, artifact formation
                arising from collection or volatilization of HN03frorn ammonium nitrate aerosols is a perennial
                worry in filterpack methods, and can pose a problem whenever long sampling times are used. In
                coastal areas the adsorption of HN03on coarse-mode seasalt aerosols can carry the nitrate flux in
                large particles; therefore, traditional gas-phase filterpack BN03measurements may be inadequate
                in coastal environments.


                The perceived need to evaluate the filterpack technique in coastal areas was reinforced by the
                results from the Wye, Md., Dry Deposition Inferential Measurement (DDIM) station which
                indicated that it may not be possible to use traditional inferential techniques for estimating dry
                deposition in the near-coastal environment. The DDIM station pushes the filterpack methodology
                to its theoretical limit by using low flow rates and a weekly sampling protocol. This is especially
                true in areas with high NH3and/or seasalt concentrations (both a problem on Maryland's Eastern
                Shore). In these areas the chemical equilibrium between HN03/NH3/NH4No3on the initial
                particle filter can influence the amount of nitrogen trapped by the secondary nylon filter (see
                below). Through ANICA and the Great Waters Program denuders and support equipment have
                been acquired for a filterpack/denuder intercomparison.

                Filter Pack


                The most widely used method for routine HN03 concentration and flux measurements is the nylon
                filterpack (FP) technique (e.g., Goldan et al., 1983). In this method, two filters are typically
                employed; a Teflon prefilter is typically used to trap particles (NHN03, (NH,)2SO4, etc.) and pass
                HN031which is quantitatively scrubbed by reaction with basic -NH2active sites on the nylon
                filter. Both filters are extracted after collection and N03- is usually analyzed by ion
                chromatography. Fast air flow rates and filter pore sizes of I @Im are typically used, and the open-
                faced, inlet-less filterpack assembly is faced directly into the wind to avoid surface losses of
                HN03. The nylon filter technique showed no interferences in laboratory tests (Goldan et al.,
                1983) and has been successfully used in a variety of gradient measurements of HN03flux (e.g.,
                Huebert and Robert, 1985; Huebert et al., 1988). Side-by-side deployment of the Teflon/nylon
                filterpacks demonstrated a precision of ca 1%, more than adequate for the successful
                measurement of HN03gradients over Illinois grassland (Huebert et al., 1988). Provided that
                sample collection intervals are kept short, however, the problem with artifact formation from the
                collection and volatilization of ammonium nitrate aerosols can be minimized (e.g., Spicer et al.,
                1982; Huebert et al., 1988).



                                                               B-16











                Denuders


                Denuder measurements Of NH3 are based upon the concept of selective scavenging by passing the
                sample airstream through an acid-coated tube (simple denuder) or concentric tubes (annular
                denuder). Gas-phase ammonia rapidly diffluses to the walls, is absorbed, and is extracted after
                sampling for analysis by ion chromatography (IC), colorimetry, or other methods. Ammonium-
                containing aerosols and vapor phase HN03 pass unimpeded through the denuder and are collected
                on Teflon and nylon backup filters (Luke and Valigura, 1996).

                The denuder method employed for this study will utilize a citric acid-coated denuder (CAD),
                followed by a traditional filter pack assembly, with extraction and analysis of ammonia (as NH,')
                by IC. An inverted quartz tube, 60-70 cm long, is suspended vertically from a sampling tower.
                Air is drawn through the denuder at a flow rate of about 2- 10 standard liters per minute (SLM)
                using a small pump and electronic mass flow controller. Gas-phase ammonia in the sample
                airstream rapidly diffuses to the walls of the denuder and is absorbed by the citric acid coating.
                After sampling, a small aliquot of deionized water or 5 mM HCL is used to rinse the walls of the
                tube and extract the ammonia in the form of an ammonium salt. The backup filter is also
                extracted. Ammonium concentrations in the tube and filter extracts are quantified by ion
                chromatography, referenced to independent calibration standards. The concentration in air may
                be calculated from the measured (NH4'), air sampling rate, and length of the sampling interval.
                Although free of readily identifiable interferences, the CAD/IC method may suffer from artifacts
                under some conditions. Turbulent particle impaction at the inlet and walls of the denuder, as well
                as particle settling, can artificially increase measured (NH,). These artifacts can be virtually
                eliminated, however, by maintaining larninar flow through the tube and positioning the denuder
                vertically (Ferm, 1979). Potentially severe artifacts may arise from the evaporation of ammonium
                nitrate within the denuder as gas phase N113 is depleted at the denuder walls. The magnitude of
                this error will depend strongly upon temperature and residence time of the sample air within the
                denuder. While this error is difficult to quantify, Langford et al. (1989) estimated its magnitude
                under conditions of high aerosol loading using the measured NH4N03 evaporation rates of
                Richardson and Hightower (1987). At denuder residence times of I s (typical residence times for
                the research proposed here is less than 0.5 s) and temperatures of 25-35 'C, Langford et al,
                (1992) estimated that particle evaporation will introduce approximately 6% error in measured
                (NH3) at high aerosol concentrations ((NH4N03)/(NH3) z 70). At the lower aerosol loadings
                typical of the rural and remote troposphere, artifact formation from aerosol evaporation should be
                small. Exceptions may occur at locations near strong local NH3 sources (such as agricultural
                areas) where aerosol loadings may be high.

                Previous Intercomparisons

                A multitude of HN03 intercomparison campaigns have been conducted over the last 10- 15 years,
                and results have not always been consistent. In general, the various chemical techniques have
                shown substantial agreement at HN03 concentrations greater than a few ppbv. At lower
                concentrations the results begin to diverge. In a 1979 study in Claremont, Ca. (Spicer et al.,


                                                              B-17









                1982), a variety of filter pack and denuder methods were compared with the NOY conversion
                difference and FTIR techniques (Luke and Valigura, 1996). At HN03concentrations of 4-20
                ppbv, good agreement was found among most of the methods. Hering et al. (1988) compared the
                results of HNO, measurements using several integrated (denuder, thermal denuder, FP) and
                spectroscopic methods. At HN03 concentrations of several ppbv agreement was generally good,
                but scatter increased below 3 ppbv. Overall agreement among all methods was a factor of two.
                The demonstrated precision of the denuder method was poor, and the denuder method yielded
                consistently lower results than the spectroscopic methods used. Artifact formation arising from
                ammonium nitrate volatilization in filterpacks exposed for long periods of time resulted in
                overprediction of HN03by FP techniques. However, for short sampling intervals the FP method
                agreed well with the spectroscopic techniques. In areas where the majority of nitrate is present as
                gas-phase nitric acid, intercomparison campaigns suggest that agreement between filterpack and
                denuder measurements of HN03are generally good, owing to the lack of artifact formation from
                volatilization of ammonium nitrate aerosols (e.g., Benner et al., 1991).

                Measurement Strategies

                The first priority is to deploy the CAD system alongside the DDIM filterpack and conduct
                calibration, collection efficiency, and artifact formation tests to judge the suitability of the CAD
                technique for longer term measurements. Collection efficiency of the denuder system will be
                determined by spiking the ambient airflow with a quantified source of N113, obtained either from a
                calibration gas mixture or a permeation tube. Measured efficiencies will be compared to those
                calculated from the Gorn-dey-Kennedy relationship (e.g., Ferm, 1979; Bollinger et al., 1983).
                Side-by-side CAD/filterpacks will be deployed in parallel to assess the representativeness of daily
                integrated samples by comparing results from a single CAD/filter sample collected for 24 hours
                with those from 8 (6) denuders sampled for 3 (4) hours each. Day/night effects could similarly be
                studied and may be important considerations when sampling in a humid environment where
                condensation may be a problem.

                The secondary priority is to evaluate the importance of NH4N03artifact formation in high
                concentration areas such as Chesapeake Bay. Collection efficiency and "breakthrough" tests will
                be performed by connecting denuders in series and comparing the amounts of NH4' collected on
                front and back tubes. Artifact formation from particle impaction and/or evaporation may be
                explored by comparing results from parallel samplers, where a Teflon prefilter is placed at the
                inlet of one denuder and the outlet of the other. In a parallel effort, base-impregnated denuders
                for HN03will be deployed alongside the CAD denuders, and will be used in conjunction with
                filterpack measurements to assess the importance of aerosol NH4N03fbrmation at the sampling
                sites and its effects upon long-term filterpack measurements of HN03carried out routinely at the
                Wye Institute. Analytical recovery of the extraction and IC analysis will also be evaluated.








                                                                 B-18









                                           Estimating A ir- Water Transfer of HN03

               Background

               When considering the atmosphere as a pathway for nitrogen input, it is necessary to consider
               whether the nitrogen is transferred directly to the water surface, or indirectly to the watershed,
               with subsequent transport to the receiving waters. Several ongoing research programs seek to
               estimate the amount of nitrogen entering the coastal waters via precipitation and indirect
               deposition; however, rigorous studies of directly deposited nitrogen are few. Air-water exchange
               has been estimated for most nitrogen species over open ocean (Galloway, 1985; Duce et al.,
               1991), however, these rates may not apply to coastal situations because coastal areas experience
               meteorological processes not found over the open ocean (NRC, 1992).

               To date, the primary obstacle to estimating coastal air-water nitrogen exchange has been the lack
               of near-surface, over-water coastal meteorological data. These data are needed to generate
               improved computer models, which will then be able to simulate existing small-scale coastal
               conditions. In response to this need, there has been an increased deployment of measurement
               buoys along the East Coast of the United States. One such network of buoys is the Chesapeake
               Bay Observing System (CBOS) owned and operated by the University of Maryland's Horn Point
               Laboratory. Through ANICA, the instrumentation aboard a CBOS buoy was augmented to
               include relative humidity and water temperature measurements. The results of this project are
               published in the Journal of Geophysical Research (Valigura, 1995). The two primary objectives
               were to: i) develop and evaluate an iterative bulk exchange model to estimate air-water exchange
               of heat, water and momentum from buoy data, and ii) use the model outputs to estimate air-water
               transfer rates of nitric acid (HN03).

               The Iterative Bulk Exchange Model

               Because of a high affinity for water and relatively high ambient concentrations in coastal areas,
               HN03is considered to be the primary nitrogen species of interest for deposition directly to water
               surfaces. Given its affinity for water, HN03 transfer can be considered uni-directional (i.e.,
               downwards). Furthermore, for transfer to water, the quasi-boundary layer resistance (Rb) is small
               (Garratt and I-Ecks, 1973; Kanemasu et al., 1979) relative to the aerodynamic resistance (R.,@03)
               even in very light wind conditions. Consequently, the arguments here will focus on R, HN03alone,
               and the resistance form of the general flux (FHN03)equation can be written

                                      FHN03 = IHN0310-[HN031, / R@,HN03 = Vd IHNO31,

               where the surface concentration, [HN0311, is taken to be zero and the deposition velocity, Vd (M
               s-'), is the inverse of the aerodynamic resistance Ra,@03 (S M-')-

               Air-water exchange of sensible heat is similarly regulated by aerodynamic resistance (R.,@,
               allowing for the assumption that Ra,@03 is equivalent to R;,,H. Using this assumption, it was


                                                             B- 19









               therefore possible to use the heat transfer coefficient, estimated with general bulk-transfer
               equations, to estimate the deposition velocity of HN03

                                                             _' = D.u.
                                                    Vd =Ra,H          (2)

               where DHis the dimensionless heat transfer coefficient, see below, and u. is the measured
               windspeed (m s-') at height (z).

               Model Evaluation Results


               Three days of eddy correlation measurements of heat, moisture and momentum fluxes were
               collected on tower, boat, and airplane platforms from June 16-20, 1990 near NOAA's Looe Key
               National Marine Sanctuary in Florida (Crawford et al., 1993). During the first day, data
               collection was taken in shallow gulf waters (10 m), 500 m upwind of shore. During the next two
               days, data collection was performed 13.5 km offshore, 100 m outside the barrier reef in 25 m of
               water. Mean meteorological variables (wind speed, air temperature, and relative humidity) were
               recorded from all platforms, with water surface temperature measured by infrared temperature
               sensor from the boat, for the duration of the experiment. These mean data were incorporated into
               the bulk exchange model and the resulting output was compared against the eddy correlation data
               collected from the airplane (Crawford et al., 1993).

               Model versus measurement intercomparisons of friction velocity, latent heat flux, and sensible
               heat flux were quite favorable. Average differences between measured and modeled friction
               velocities and latent heat fluxes were small, :@ I cm and ï¿½ 10 W m-' respectively. Sensible heat
               fluxes were very small and intercomparisons were inconclusive. This was due to an apparent
               disconnect between eddy correlation data measured on the airplane and local water- surface data
               collected by boat. The disconnect can be attributed to the inherently local nature of the surface
               temperature measurements and the fact that eddy correlation integrates surface characteristics for
               some considerable distance upwind. The noise in the water surface temperature measurements
               was directly translated into noise in the model output, thereby resulting in the measurement
               versus model differences. Further comparisons are being conducted as data becomes available.

               Experimental Set-Up

               In late March 1992, the CBOS buoy was anchored in the north Chesapeake Bay, off Howell
               Point, Maryland (39.36N, 76.1'W) for 3 periods between April 1992 and July 1994 (excluding
               the winter months) totaling 18 months. The CBOS buoy is owned and operated by the Horn
               Point Laboratory, Center for Environmental and Estuarine Studies, a member of the University of
               Maryland system. The buoy is of the wave-follower variety, and allowed data to be telemetered
               to a central computer system for remote access. The cross-section of the Bay narrows to 4.5 krn
               in the area of the buoy, making the fetch 4.5 km or greater in the geographical window from 2 10'
               to 50', and I km or less in the window from 5 V to 2090. Water depth is variable across this
               section of Chesapeake Bay (average depth 5 m), but the buoy is anchored in the deep channel


                                                             B-20









                where the depth of water is 12.5 in.

                Deposition Velocity Estimation Results

                Deposition velocities were calculated for the approximately 25,000 10-minute periods that
                comprise the CBOS dataset to date. The overall frequency distribution is presented in Figure
                B.5. When viewed on shorter time scales, the distribution begins to change, as is shown for the
                first week in December 1993 (Figure B.5). The potential danger of using average/general
                deposition velocities in short term analysis becomes apparent when the actual time series of Vd is
                reviewed for the same week in December 1993, Figure B.6. Meyers and Yuen (1987) found that
                concurrent, high resolution Vd and concentration data improved estimates of 03 deposition, but
                did not improve estimates of S02 deposition. The primary reason for the difference between the
                two chemicals was that the measured variability in 03 concentration was significantly correlated
                with the corresponding measured variability in Vd (i.e, concentrations were proportional to
                deposition rates), and concentrations of S02 Were not correlated with corresponding Vd (i.e.,
                concentrations were not proportional to deposition rate). It is unknown if 14NO, concentrations
                ([HN031) over Chesapeake Bay are correlated with Vd ([HN031 was not measured during this
                study). To illustrate the potential errors associated with using average or time series Vd, a simple
                matrix analysis was performed using the December 1993 time series, Table B.5. Deposition was
                estimated using three different Vd regimes (the actual times series Vd, the time series average Vd,

                Table 13.5. Comparison between deposition (g HN03 m-') estimates (Vd * [HN031) derived
                using different averaging schemes and the time series average$ estimate for the first two
                weeks of December 1993 (% differences are shown in parenthesis).

                                                      Low-High             High-Low              Average
                                                     1.2-2.7 ppb          2.7-1.2 ppb            1.95_2pb

                  Time Series Vd                    .485( 4.5)            .443(-4.5)            .464( 0.0)
                  Time Series Average Vd             .429(-7.5)           .499( 7.5)            .464:
                  (.0049 ms-')                                                               1                    1

                  1993 Annual Average Vd             -567(22.0)           661 (42.0)            614(32.0)
                  (.0065 ms-')                 IF                         -                     -                 @1

                and the 1993 average Vd), and three different HN03 concentrations representative of [HN031
                commonly measured within the Bay region. The concentrations were i) switched from a constant
                low (1.2 ppb) to a constant high (2.7 ppb) after the front, ii) switched from a constant high to a
                constant low after the front, and iii) maintained at the average (1.95 ppb). The deposition
                estimate derived from using the time series average Vd and average [HN031 was used as the
                reference value for intercomparisons, Table B. 5. The analysis shows that if Vd and [HN03] are





                                                                B-21






















                  0.30




                  0.25  - ---------- - ----- Z-a'




                  0.20    ---



              C2
                  0.15  . ...................... ... ... ... ... ...




                  0.10




                  0.05




                  0.00
                      0.0          0.5          1.0         1.5         2.0          2.5         3.0
                                                  Deposition Velocity (cm s-1)
                          ------- ---------






















             Figure B.5.   The frequency distribution for 18 months (excluding winter) of estimated
                           deposition velocities in Chesapeake Bay (solid line) and the distribution for I week
                           in December 1993 (dotted line).


                                                      B-22























                      2.5




                      2.0




                      1.5




                      1.0




                      0.5




                      0.0
                            1          2           3           4           5           6           7         8
                                                                   Date









               Figure B.6.    Ten-n-tinute deposition velocities calculated for December 1-8, 1993.


                                                             B-23









                not correlated, the error in deposition estimates is primarily driven by errors in estimating mean
                values of Vdand [HN031, If they are correlated, determining the source the associated errors is
                more complex. If [HN031 andVdare correlated, adequate estimation of deposition can only be
                obtained by concurrent measurements of [HN03] andVd. These analyses demonstrate the need
                to determine if variability in HN03 concentration is significantly correlated with the
                corresponding variability in transfer rate to Chesapeake Bay.

                TheVddistribution (Figure B.5) is likely to be conservative for two reasons. The first and most
                important reason is that the winter months are not accounted for because the buoy is removed
                due to ice. During the winter, air-water temperature differences are likely to be at a maximum
                (with water being consistently warmer than air), causing transfer rates to be at their peak.
                Therefore, any distribution that excludes this period is likely to be conservative. Another reason
                to believe that these estimates are conservative is that assuining equivalent transfer rates for
                HN03and heat does not adequately account for scavenging of 14N03by aerosol water droplets
                and particles, which tend to increase deposition rates.

                There are theoretical limitations with this approach as well: lack of homogeneous conditions in
                the coastal zone and the inadequacy of similarity theory to describe turbulent conditions
                measured. Because the northern Chesapeake Bay is narrow, the local landscape tends to affect
                the meteorology. These effects make fetch assumptions unreasonable, thereby making it difficult
                to assume that the "local" buoy measurements are representative of any sizeable area. Estimating
                large-scale deposition patterns will require deployment of buoys at a range of different sites with
                concurrent mesoscale modeling efforts. The second theoretical consideration concerns the
                Monin-Obukhov similarity theory upon which the bulk transfer equations are based. This theory
                has been evaluated, and is considered valid, over a certain range (-I @. z/L :@ 1) of meteorological
                conditions. Beyond this range, the atmosphere is either extremely stratified so that the flow is
                effectively disconnected from the surface or so unstable that it is under free convection
                conditions. During the two years of data collection, there were periods where model outputs
                showed that conditions were too stable/unstable to be considered in the "normal" Monin-
                Obukhov frame. These periods were closely related to low wind speeds, as shown in Figure B.7
                for the month of April 1994.

                The approach used in this study has been shown to be applicable toSo2and should be applicable
                to other hygroscopic chemicals such as ammonia. To improve upon this technique, further eddy
                correlation projects must be performed to evaluate/modify the bulk transfer equation assumptions
                under low wind conditions. In future investigations, concurrent evaluation of HN03
                concentrations will allow for the quantification of the actual differences between the time-series
                and the single-deposition velocity approaches.









                                                              B-24






















                         15





                         10





                          5




                          0                                                                    11 '00-OW0000000 so




                         -5





                       -10
                             0                 2                 4                 6                8                 10
                                                                  Wind Speed (m/s)




                Figure B. 7.    Ten-n-tinute Monin-Obukhov dimensionless stability estimates versus measured
                                windspeed, calculated from the Chesapeake Bay Observing System data for April
                                1994. Solid lines denote outer ranges of applicability for Monin-Obukhov
                                similarity theory.


                                                                  B-25










                                 Operational RAMS 15 Kin Forecasts over Chesapeake Bay

               Background

               The Regional Atmospheric Modelling System (RAMS) version 3a, developed at Colorado State
               University, has been configured to simulate mesoscale atmospheric circulations on both
               IBM/6000 workstations at the Air Resources Lab (ARL). The software developed at ARL allows
               the user to setup and initialize RAMS anywhere in the world using the ARL packed
               meteorological fields and global land and water surface data sets. The ARL modeling system is
               described in detail by McQueen et al. (1994, 1995).

               RAMS Description


               Grid Structure
               RAMS utilizes an Arakawa-C grid stagger of the thermodynamic and momentum variables to
               reduce finite differencing error. An oblique stereographic horizontal grid coordinate can be
               specified. This varies from the normal polar stereographic, coordinates in that the grid is true at
               the center of the RAMS domain with the "pole" position defined at that center as well. A two-
               way interactive multiple nested grid scheme exists so that scale interactions can be incorporated.
               This nested grid approach also allows for a finer mesh to resolve local-scale circulations in the
               area of interest and a coarser mesh outside this area. Simulations performed over the northern
               Chesapeake Bay employed the finest nested grid spacings to date (,&X=2.5 kin), Figure B.8. A
               15 krn finest mesh grid spacing is used to produce operational 24-36 hour forecasts. Grid
               domains of the finest mesh have ranged from 50 to 600 kni depending on the grid spacing. The
               model can be run in 1, 2 or 3 dimensions.


               Initialization
               The model can be initialized from spatially inhomogeneous meteorological observations and,
               therefore, simulations are not limited to synoptically undisturbed cases. A program to ingest
               ARL packed National Centers for Environmental Prediction's (NCEP) model gridded fields into
               RAMS was developed. More than one input NCEP dataset can be used by combining these
               fields to create a complete input through the troposphere. Alternatively, NCEP model gridded
               data and surface and rawinsonde sites can be objectively analyzed to isentropic surfaces before
               being interpolated to the model grid using the RAMS ISAN package. Objective analysis follows
               the widely used Barnes approach. Currently, the model defaults to using NCEP model data on a
               2.5 degree latitude-longitude horizontal grid and mandatory pressure levels.

               The user can choose constant or spatially varying surfaces for RAMS ingest. Spatially varying
               surface variables such as soil moisture, soil and vegetation type, canopy temperature and water
               content, terrain height, land roughness, land percentage and water surface temperature can be
               ingested into RAMS on the model grid. Table B.6 summarizes the available ground surface data
               sets and their coverage. Over the U.S., surface -data exists at a higher (30" latitude-longitude grid
               or about I kin resolution) than available globally (10' latitude-longitude grid or about 18 krn



                                                            B-26












                                  RAMS Fire grid doma@r (2KM)

                                                                                                                 rr
                                                                                                                 CD
                                                                                                                 cc

                                                                                                                 __j

                                                                                                                 V)
                                                                                                                 w
                                                                                                                 0
                                                                                                                 Ir

                                                                                                                 CD
                                                                                                                 cr)






              C,


              D                                                                                                  CD



           CD











                                                                                                                 LU
           U
           C-5                                                                                                   0-
                                                                                                                 U)
           0                                                                                                     0
           01
           rr
                                                                                                                 <
           LLJ
                                                                                                                 0
           LLJ                                                                                                   Z


                                                                                                                 L)


                                                                                                                 <
                                                                                                                 LU
                                                                                                                 L)





                                                                                                                 CD




            -25.00 -20.00 -15.00 -10.00 -5.00                 0       5.00     10.00    15.00    20.00     25.00





               Figure B. 8.   Horizontal (vectors) and vertical (contours) windspeed simulations performed over
                              the northern Chesapeake Bay employing the finest nested grid spacings (AX=2.5
                                             A




























                              km). Note the high vertical motions on the western shore of the Bay indicating the
                              beginnings of a "Bay Breeze" (Triangle indicates position of CBOS buoy).


                                                            B-27









                 resolution). Soil moisture is predicted by first computing an Antecedent Precipitation Index
                 (API) derived from the last few months of observed or model gridded precipitation archived at
                 ARL.


                 Table 13.6. Sum ary of surface data available for RAMS.
                   Type                      ge I Resolution                 Range              Update

                   Topography        Global          10' lat/lon grid        lom
                   11                U.S.            30" lat/lon grid        lom
                   SST               Global          V lat/lon               . 1 K              weekly

                   soil type         Global          I degree                14 categories
                   soil moisture     Global          NMC forecasts           0-1                daily
                   1111              U.S.            observed precip         V1                 daily
                   Land use          global          I degree                167 H-S
                                                                             categories

                                     U.S.            I km                    167 H-S
                                                                             cateRories

                   Roughness         global          I degree


                 Operational Capabilities

                 RAMS has been used operationally for air quality dispersion forecasts. At ARL, RAMS is run in
                 both a forecast and hindcast mode. Table B.7 summarizes the current operational configuration.

                 Table B.7. RAMS operational configuration.

                   Workstation         Mode                  Location              Domain         CPU/wall      AX
                                                                                                I time        I

                   IBM 6000/530        12 H hindcast         Mid Atlantic/         2400 km'       .33            80 km
                   (ARL RISCI)         w/ FDDA               Ches. Bay             680 kM2                       20 km

                   IBM 6000/560        24-36 H               Mid Atlantic/         2400 kM2       .16            60 km
                   (ARL RISC2)         forecast              Ches. Bay             680 kM2                       15 km
                                                                                                                     ni

                 RAMS is normally run over the Chesapeake Bay and Mid Atlantic region when it is not required
                 for an emergency, exercises, or to support air quality experiments. The RAMS domain can easily
                 be moved through the Real time Emergency Application and Display sYstem (READY; Rolph


                                                                     B-28









               et al., 1993; Draxler et al., 1993) by simply specifying the center latitude and longitude of the
               coarse grid through the READY menu. On RISC2, RAMS is configured to produced 24-36
               hour forecasts twice-per-day after the latest ETA model fields are available. On RISCI, RAMS
               is configured in a hindcast mode while also assimilating four dimensionally surface wind data
               normally available every hour. An example of  'operational RAMS predicted winds and
               temperatures on the fine 15 km domain are shown in Figure B.9. RAMS outputs are available in
               real-time on a one week rotating archive on the ARL work stations.
















































                                                            B-29












            FN      RAMS Operational Fine Grid (15 km) Archive

                                                                                                     czz






                                                                                                     Ui
                                                                                                     U


                                                                                                     0
                                                                                                     Lo
                                                                                                     Li
                                                                                                     cr





                                                            All

                                                                                                     L)


                                                                z                                    LLI


                                                                                                     cl)
         0
         C-D


         CD
         0                                                                                           <c
         LLJ


                      "WHU,                                                                          <




                                                                                                     <
                                                                                                     LIJ





                                                                                                     C)
                                                                                          wwi z
           36.00  40.00 44.00   48.00 52.00   56.00  60.00 64.00   68.00  72.00 76.00   80.00 84.00
                                                        ML
















            Figure B.9.   Operational 15 krn RAMS simulations of northern Chesapeake Bay.



                                                     B-30









               Appendix C
               ANICA ASSOCIATED SCIENTISTS


               UniversiU ofDelaware
                      Thomas Church - College of Marine Studies: Member of the Chesapeake Bay
                              Atmospheric Deposition Study (CBADS) group within the Chesapeake Bay
                              Program. Specializing in wet deposition of nutrients and heavy metals.
                      Jin Wu - Director, Air-Sea Interaction Laboratory: Investigating physical interactions
                              during low wind speed conditions.

               UniversiU ofMar-yland
                      Joel Baker - Chesapeake Biological Laboratory: Head of the CBADS group. Specializing
                              in wet deposition of organic compounds, including nutrients.
                      William Boicourt - Horn Point Environmental Laboratory: Head of the Chesapeake Bay
                              Observational System (CBOS) buoy project. Specializing in the physical
                              hydrodynamics of the Chesapeake Bay.
                      Russell Brinsfield - Director, Wye Research Center: Investigating nitrogen cycling
                              through cropped watersheds within the Chesapeake Bay area.
                      Russell Dickerson - Department of Meteorology: Specialist in the measurement of
                              atmospheric nitrogen species.
                      John Ondov - Department of Chemistry, College Park: Member of the CBADS group
                              within the Chesapeake Bay Program. Specializing in particle chemistry and
                              transfer.


               UniversiU of Virginia
                      James Galloway - Professor, Department of Environmental Sciences: Investigating wet
                              deposition of organic nitrogen to the coastal ocean.
                      William Keene - Department of Environmental Sciences: Investigating wet deposition of
                      organic nitrogen to the coastal ocean. Specializing in atmospheric chemistry.

               UniversiU offorth Carolina
                      Hans Paerl - Professor, Institute of Marine Sciences: Specializing in nitrogen-production
                              interactions in coastal waters.


               Smithsonian Environmental Research Center
                      Dave Correll - Director: Specialist in forest watershed nutrient dynamics.
                      Jess Parker - Forest Ecologist: Specialist in forest canopy Structure dynamics.
                      Donald Weller - Forest Ecologist: Specialist in forest watershed dynamics modeling.







                                                            C-1








               Appendix D
               BIBLIOGRAPHY


               Asman, W.A.H. 1994. Emission and deposition of ammonia and ammonium. Nova Acta
                       Leopoldina. 70: 263-297.

               Benner, C.L., D. J. Eatough, N.L. Eatough, and P. BhardwaJa. 1991. Comparison of annular
                       denuder and filter pack collection of HN03(g), HN02(g), S02(g), and particulate-phase
                       nitrate, nitrite, and sulfate in the south-west desert. Atmos. Environ. 25: 1537-1545.

               Bollinger, M.J., R.E. Sievers, D.W. Fahey, and F.C. Felisenfeld. 1983. Conversion of nitrogen
                       dioxide, nitric acid, and n-propyl nitrate to nitric oxide by gold-catalyzed reduction with
                       carbon monoxide. Anal. Chem. 55: 1980-1986.


               Boyton, W.R., J.H. Garber, R. Summers, and W.M. Kemp. 1995. Inputs, transformations, and
                       transport of nitrogen and phosphorus in Chesapeake Bay and selected tributaries.
                       Estuaries 18:285-314.


               Brock, D.A., J. Matsumoto, and N. Boyd. 1995. A nitrogen budget for the Guadalupe Estuary,
                       Texas, over a range of freshwater inflow. Contrib. In Mar. Sci, Submitted.

               Cornell, S., A. Rendell, and R. Jickells. 1995. Atmospheric inputs of dissolved organic nitrogen
                       to the oceans. Nature. 376: 243-246.


               Correll, D.L., and D. Ford. 1982. Comparison of precipitation and land runoff as sources of
                       estuarine nitrogen. Estuarine, Coastal and Shelf Sci. 15:45-56.

               Crawford, T.L., R.T. McMillan, T.P. Meyers, and B.B. Hicks. 1993. The spatial and temporal
                       variability of heat, mass, and momentum air-sea exchange in a coastal environment. J.
                       Geophys. Res. 98: 12869-12880.

               Dana, M. T., and W. G. N. Slinn. 1988. Acidic deposition and episode statistics from the
                       MAP3S network database. Atmos. Environ. 22:1469-74.


               Dennis, R. 1994. Review of RADM modeling results. Presented to the Modeling Subcommittee
                       of the Chesapeake Bay Program, I February, Annapolis, Md.

               Dentener, F. J., and P. J. Crutzen. 1994. A three-dimensional model of the global ammonia
                       cycle. J. Atmos. Chem. In press.

               Draxler, R.R. 1992. Hybrid Single-Particle Lagrangian Integrated Trajectories (HY-SPLIT):
                       Version 3.0 - Users Guide and Model Description. NOAA Technical Memorandum ERL
                       ARL- 195. Silver Spring, Md.: National Oceanic and Atmospheric Administration. 40 pp.

                                                             D-1









                Draxler, R.R., G.D. Rolph, J.T. McQueen, J.L. Heffier, and B.J.B. Stunder. 1993. Capabilities of
                        the NOAA Washington Regional Specialized Meteorological Center for atmospheric
                        transport model products for environmental emergency response. International Workshop
                        on Users' Requirements for the Provision of Atmospheric Transport Model Products for
                        Environmental Emergency Response, Montreal, Canada. 17 pp.

                Duce, R. A., P. S. Liss, J. T. Merril, E. L. Atlas, P. Buat-Menard, B. B. Hicks, J. M. Miller, J. M.
                        Prospero, R. Arimoto, T. M. Church, W. Ellis, J. N. Galloway, L. Hansen, T. D. Jickells,
                        A. H. Knap, K. H. Reinhardt, B. Schneider, A. Soudine, J. J. Tokos, S. Tsunogai, R.
                        Wollast, and M. Zhou. 1991. The atmospheric input of trace species to the world ocean.
                        Global Biogeochernical Cycles 5:193-259.

                Ferm, M. 1979. Method for the determination of atmospheric ammonia. Atmos. Environ. 13:
                        1385-1393.


                Fisher, D.C., J. Ceraso, T. Matthew, and M. Oppenheimer. 1988. Polluted coastal waters: the
                        role of acid rain. New York: Environmental Defense Fund.


                Fisher, D.C., and M. Oppenheimer. 199 1. Atmospheric nitrogen deposition and the Chesapeake
                        Bay estuary. Ambio 23:102-208.

                Fogel, M. L., and H. M. Paerl. 1991. Nitrogen isotope tracers of atmospheric deposition in
                        coastal waters off North Carolina. Carnegie Institute Geophysical Laboratory. pp. 147-
                        154.


                Fontijn, A., A. J. Sabadell, and R. J. Ronco. 1970. Homogeneous chemiluminescent
                        measurement of nitric oxide with ozone: implications for continuous selective monitoring
                        of gaseous air pollutants. Anal. Chem. 42: 575-579.

                Fowler, D., and J.N. Cape. 1984. On the episodic nature of wet deposited sulfate and acidity.
                        Atmos. Environ. 18:1859-1866.


                Galloway, J.N. 1985. The deposition of sulfur and nitrogen from the remote atmosphere. In:
                        Galloway, J.N., R.J. Charlson, M.O. Andreae, and H. Radke, eds. Biogeochernical
                        cycling of sulfur and nitrogen in the remote atmosphere. Dordrecht, Netherlands: D.
                        Reidel Publishing Co. pp. 27-53.

                Garratt, J.R., and B.B. Hicks. 1973. Momentum, heat, and water vapor transfer to and from
                        natural and artificial surfaces. Quart. J.R. Met. Soc. 99: 680-687.

                Goldan, P.D., et al. 1983. Calibration and tests of the filter-collection method for measuring
                        clean-air, ambient levels of nitric acid. Atrnos. Environ. 17: 1355-1364.




                                                                D-2









               Gorzelska, K., J.N. Galloway, K. Watterson, and W.C. Keene. 1992. Water-soluble primary
                      amine compounds in rural continental precipitation. Atmos. Environ. 26(A): 1005-1018.

               Hering, S.V., Lawson, D.R., Allegrini, I., Febo, A., Perrino, C., Poosanzini, M., Sickles, J.E.,
                      Anlauf, K.G., Wiebe, A., Appel, B.R., John, W., Ondo, J., Wall, S., Braman, R.S.,
                      Sutton, R., Cass, G.R., Solomon,, P.A., Eatough, D.J., Eatough, N.L., Ellis, E.C.,
                      Grosgean, D., Hicks, B.B., Womack, J.D., Horrocks, J., Knapp, K.T., Ellestad, T.G.,
                      Paur, R.J., Mitchell, W.J., Pleasant, M., Peake, E., Maclean, A., Pierson, W.R.,
                      Brachaczek, W., Schiff, H.I., Mackay, G.I., Spicer, C.W., Stedman, D.H., Winer, A.M.,
                      Biermann, H.W., and Tauzon, E.C. 1988. The nitric acid shootout: field comparison of
                      measurement methods. Atmos. Environ 22:1519-1539.


               Hinga, K.R., A.A. Keller, and C.A. Oviatt. 1991. Atmospheric deposition and nitrogen inputs to
                      coastal waters. Ambio 20:256-260.


               Huebert, B.J., and C.H. Robert. 1985. The dry deposition of nitric acid to grass. G. Geophys. Res.
                      :2085-2090.


               Huebert, B.J., W.T. Luke, A.C. Delany, and R.A. Brost. 1988. Measurements of concentrations
                      and dry surface fluxes of atmospheric nitrates in the presence of ammonia. J. Geophys.
                      Res. 93:7127-7136.


               Hunsaker, C.T., P.J. Mulholland, and C.T. Garten. 1994. Proceedings from workshop on
                      nitrogen dynamics in forested watershed for the Chesapeake Bay watershed model.
                      Unpublished.

               Johnson, D.W., and S.E. Lindberg, eds. 1992. Atmospheric deposition and forest nutrient
                      cycling. Ecological Series 9 1. New York: Springer-Verlag.

               Kanemasu, E.T, M.L. Wesely, B.B. Hicks, and J.M. Heilman. 1979. Techniques for calculating
                      energy and mass fluxes. In: Barfield, B.J., and J.F. Gerber (eds.) Modification of the
                      Aerial Environment of Plants. ASAE Monograph, pp. 157-182.

               Koch, S.E.. 1993. Proc. of a colloquium. and workshop on mesoscale coupled modeling.
                      February 22-25, Calverton, Md. NASA Conference Pub. 3217, 112 pp.

               Langford, A.O., P.D. Goldan, and F.C. Fehsenfeld. 1989. A molybdenum oxide annular denuder
                      system for gas phase ambient ammonia measurements. J. Atmos. Chem. 8: 359-376.

               Langland, M. 1992. Atmospheric deposition of ammonia from open manure-storage lagoons in
                      south-central Pennsylvania. Environ. Prof. 14:28-37.




                                                            D-3









              Luke, W.T., and R.A. Valigura. 1996. Methodologies to estimate the air-surface exchange of
                      atmospheric nitrogen compounds. In: J. Baker (ed.) Atmospheric deposition of
                      Contaminants to the Great Lakes and Coastal Waters. Denver, Col.: Society of
                      Environmental Toxicology and Chemistry.

              Malone, T.C. 1992. Effects of water column processes on dissolved oxygen, nutrients,
                      phytoplankton and zooplankton. In: Smith, D.E, M. Leffler, and G. Mackiernan (eds.)
                      Oxygen dynamics in the Chesapeake Bay. College Park, Md.: Maryland Sea Grant, 234
                      PP.

              McQueen, J.T., R.R. Draxler, and G.D. Rolph. 1994. Examining the possibility of real-time
                      mesoscale model forecasts in regions of moderately complex terrain. Preprints of the
                      I Oth Conference of Numerical Weather Prediction. Portland, Ore.: American
                      Meteorological Society, pp. 412-414.

              McQueen, J.T., R.R Draxler, and G.D. Rolph. 1995. Influence of grid size and terrain resolution
                      on wind filed predictions from an operational mesoscale model. J. Appl. Meteor.
                      34:2166-2181.


              Melillo, J.M., P.A. Steudler, J.D. Aber, and R.D. Bowden. 1989. Atmospheric deposition and
                      nutrient cycling. In: Andreae, M.O., and D.S. Schimel (eds.) Exchange of trace gases
                      between terrestrial ecosystems and the atmosphere. Report of Dahlem workshop on
                      exchange of trace gases between terrestrial ecosystems and the atmosphere, Berlin, 1989.
                      Great Britain: Wiley-Interscience Publication.

              Meyers, T.P, B.J. Huebert, and B.B. Hicks. 1989. HN03deposition to a deciduous forest.
                      Boundary-Layer Meteorol. 49:395-410.

              Meyers, T.P., and T.S. Yuen. 1987. An assessment of averaging strategies associated with
                      day/night sampling of dry-deposition fluxesof S02andO3. J. Geophys. Res. 92: 6705-
                      6712.


              Michaels, A.F., D.A. Siegel, R. Johnson, A.H. Knap, and J.N. Galloway. 1993. Episodic inputs
                      of atmospheric nitrogen to the Sargasso Sea: contributions to new production and
                      phytoplankton blooms. Global Biogeochem. Cycles 7:339-351.

              Milne, P.J., and R.G. Zika. 1993. Amino acid nitrogen in atmospheric aerosols: occurrence,
                      sources, and photochemical modification. J. Atmos. Chem. 16: 361-398.

              National Research Council Panel on Coastal Meteorology. 1992. Coastal meteorology: A review
                      of the state of the science. Washington D.C.: National Academy Press, 99 pp.




                                                            D-4









                Nixon, S.W., S.L. Granger, and B.L. Nowicki. 1995. An assessment of the annual mass balance
                       balance of carbon, nitrogen, and phosphorous in Narragansett Bay. Biogeochern 31:15-
                       61.


                Owens, N.J.P., J.N. Galloway, and R.A. Duce. 1992. Episodic atmospheric nitrogen input to
                       oligotrophic oceans. Nature 357:397-399.

                Paerl, H.W. 1985. Enhancement of marine primary production by nitrogen enriched acid rain.
                       Nature 316:747-749.


                Paerl, H.W. 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters.
                       Limnol. Oceanogr. 33:823-847.

                Paerl, H.W. 1996. Atmospheric nitrogen deposition in coastal waters. In: J.E. Baker (ed).
                       Atmospheric Deposition of Contaminants to the Great Lakes and Coastal Waters.
                       Denver, Co.: Society of Environmental Toxicology and Chemistry.

                Paerl, H.W., J. Rudek, and M.A. Mallin. 1990. Stimulation of phytoplankton production in
                       coastal waters by natural rainfall inputs: Nutritional and trophic implications. Marine
                       Biol. 107:247-254.


                Paerl, H.W., and M.L. Fogel. 1994. Isotopic characterization of atmospheric nitrogen inputs as
                       sources of enhanced primary production in coastal Atlantic Ocean waters. Mar. Biol.
                       119:635-645.


                Pechan, E.H. 199 1. Background documentationof S02and NO,, forecasts. Springfield, Va.:
                       E.H. Pechan and Associates. U.S. Envirom-nental Protection Agency Contract 68-
                       D00120.


                Richardson, C.B., and R.L. Hightower. 1987. Evaporation of ammonium nitrate particles. Atmos.
                       Environ. 21: 971-975.


                Rolph, G.D., J.T. McQueen, and R.R. Draxler. 1993. Real-time environmental applications and
                       display system. American Nuclear Society meeting on environmental transport and
                       dosimetry. Charleston, S.C., September 1-3.

                Schlesinger, W.H., and A.E. Hartley. 1992. A global budget for atmospheric NH3. Biogeochem.
                       15:191-211.


                Scudlark, J.R., and T.M.Church. 1993. Atmospheric input of nitrogen to Delaware Bay. Estuaries
                       16:747-759.






                                                            D-5










               Scudlark, F.R., K.M. Russell, J.A. Montag, J.R. Maben, W.C. Keene, T.M. Church, and J.N.
                      Galloway. 1995. Dissolved organic nitrogen in precipitation: collection, analysis, and
                      atmospheric flux. National Atmospheric Deposition Program, Technical Committee
                      Meeting, Toronto, Ontario.

               Singh, H.B. 1987. Reactive nitrogen in the troposphere. Environ. Sci. Technol. 21: 320-327.

               Smullen, J.I., J.L. Taft, and J. Macknis. 1982. Nutrient and sediment loads to the tidal
                      Chesapeake Bay system. In: Chesapeake Bay Program technical studies: a technical
                      synthesis. E.G. Macalaster, D.A. Barker, and M. Kasper (eds.) U.S. Environmental
                      Protection Agency, Washington, D.C., pp. 147-162.

               Spicer, C.W., J.E. Howes, Jr., T.A. Bishop, L.H. Arnold, and R.K. Stevens. 1982. Nitric acid
                      measurement methods: an intercomparison. Atmos. Environ. 16: 1487-1500.

               Suzuki, Y. 1993. On the measurement of DOC and DON in seawater. Marine Chemistry 4 1:
                      287-288.


               Tyler, M. 1988. Contributions of atmospheric nitrate deposition to nitrate loading in the
                      Chesapeake Bay. Columbia, Md.: VERSAR Inc. Maryland Department of Natural
                      Resources Report RP1052.

               Valiela, L, G. Collins, J. Kremer, K. Lajtha, M. Geist, B. Seely, J. Brawley, and C.H. Sham.
                      1996. Nitrogen loading from coastal watersheds to recieving waters: review of methods
                      and calculation of loading to Waquoit Bay. Ecol. Appl., in press.

               Valigura, R.A., J.E. Baker, J. Scudlark, and L.L. McConnell. 1995. Atmospheric deposition of
                      nitrogen and contaminants to the Chesapeake Bay and is waershed. In: Perspectives on
                      the Chesapeake Bay: avances in etuarine scences. Annapolis, Md.: Chesapeake Research
                      Consortium.


               Valigura, R.A., 1995. An iterative bulk exchange model for estimating air-water transfer of
                      HN03' J. Geophys. Res. 100:26,045-26,050.

               Watson, R.T., L.G. Meira, Filho, E. Sanhueza, and A. Janetos. 1992. Greenhouse gases: sources
                      and sinks. In: Houghton, J.T., B.A. Callander, and S.K. Varney (eds.) Climate change
                      1992: the supplementary report to the IPCC scientific assessment. Great Britain:
                      Cambridge University Press. 198 pp.








                                                            D-6










                                             OTHER TITLES IN THE
                                        DECISION ANALYSIS SERIES



             No. 1. Able, Kenneth W. and Susan C. Kaiser. 1994. Synthesis of Summer Flounder Habitat
             Parameters.


             No. 2. Matthews, Geoffrey A. and Thomas J. Minello. 1994. Technology and Success in
             Restoration, Creation, and Enhancement of Spartina alterniflora Marshes in the United States. 2
             vols.


             No. 3. Collins, Elaine V., Maureen Woods, Isobel C. Sheifer and Janice Beattie. 1994.
             Bibliography of Synthesis Documents on Selected Coastal Ocean Topics.

             No. 4. Hinga, Kenneth R., Heeseon Jeon and Nodlle F. Lewis. 1995. Marine Eutrophication
             Review.


             No. 5. Lipton, Douglas W., Katharine F. Wellman, Isobel C. Sheifer and Rodney F. Weiher.
             1995. Economic Valuation of Natural Resources: A Handbook for Coastal Resource
             Policymakers.

             No. 6. Vestal, Barbara, Alison Rieser et al. 1995. Methodologies and Mechanisms for
             Management of Cumulative Coastal Environmental Impacts. Part I -- Synthesis, with Annotated
             Bibliography; Part 11 -- Development and Application of a Cumulative Impacts Assessment
             Protocol.


             No. 7. Murphy, Michael L. 1995. Forestry Impacts on Freshwater Habitat of Anadromous
             Salmonids in the Pacific Northwest and Alaska--Requirements for Protection and Restoration.

             No. 8. Kier (William M.) Associates. 1995. Watershed Restoration--A Guide for Citizen
             Involvement in California.


















































                                                              ATMO


                                                                         ,a

                                                      0
                                                                           @h
                                                                       C'









































                                                      3 6668 00004 9074