[From the U.S. Government Printing Office, www.gpo.gov]
Aill . . . . . . . . . . low 7.w JA 14, COASTA L ZONE INFORMATION CENTER ik@ Amp- Gc 11.2 .025 1988 V1 41, A '-C@ ir At OCEAN SYS""l""l:"f)' N 0 AA?l"0,A R I S-jcl '014 y i4szll' 777 -,-V. 4.@ 91 '40 t. Aii ville Ilk -4 jr Awl COASTAL ZONE INFORMATION CENTER T s e tv, GC Ree 11.2 .025 1 D-On esol 1988 Pred c V1 W, i4 Ar low Ocean System Studies .e. i/ NOAA/OAR Research Strategy I The Ocean System - Prediction and Resources Produced by the University Corporation for Atmospheric Research based on a series of NOAA Symposia and Meetings Boulder, Colorado December 1988 V) rl< Table of Contents Foreword ................................................................................................................................. Editor's Preface .......................................................................................................................3 1. Ocean Circulation and Global Climate Change .........................................................5 11. Ridgeflux: Hydrothermal Venting on a Global Scale ............................................. 25 111. Fisheries Oceanography .............................................................................................. 37 IV. Sea Ice and Arctic Ecosystems .................................................................................... 51 V. Estuarine Systems ......................................................................................................... 59 VI. Marine Chemistry, Biology and Climate .................................................................. 71 Editors: Shirley Fiske, National Sea Grant Office; Victor Omelczenko, National Sea Grant Office; John Ball, OAR. Editorial Assistant: David Schlezinger, National Sea Grant Office. Administrative Assistant: Sue Borda, National Sea Grant Office. Type composition: Barbara Mericle, National Center for Atmospheric Research (NCAR). Art director: Justin Kitsutaka, NCAR. Cover design: Gaylynn Potemkin, NCAR. Publication design and layout: Barbara Mericle, NCAR. Illustration and produc- tion: Lee Fortier, Wil Garcia, Barbara Mericle, Gaylynn Potemkin, Mike Shibao, NCAR. Foreword The responsibility of the Office of Oceanic and Ocean Circulation and Global Climate Change Atmospheric Research (OAR) is to formulate and Ridgeflux: Hydrothermal Venting on a Global Scale execute purposeful, comprehensive research pro- Fisheries Oceanography grams that provide technological and scientific prin- Sea Ice and Arctic Ecosystems ciples on which to base improvements of NOAA's Estuarine Systems services and products. As such, OAR must look ahead Marine Chemistry, Biology and Climate to opportunities that will enhance the Nation's strength and provide the knowledge base upon which Future reports will address other issues. to wisely pursue these opportunities. It must also The purpose of these reports is to communicate to 'de early warnings of problems our Nation will others OAR's view of opportunities for NOAA in prov. face and the knowledge upon which cost-effective critical areas of research for which we have a respon- solutions can be based. Accordingly, OAR must offer sibility. It is my hope that the statements will stimu- continuity that transcends the changing political cli- late comments from a varied audience, and that the mate because many of the research programs require research community can use them to identify oppor- time scales of longer duration than that of the politi- tunities for cooperation with NOAA. cal process. At the same time, it cannot be isolated from changes in priorities that are communicated by our Nation's people through the political process. To accomplish this, an effective dialogue is required. It is in the above context that I am-initiating an Joseph 0. Fletcher episodic series of publications on contemporary is- Assistant Administrator sues that OAR is addressing in concert with other Oceanic and Atmospheric Research NOAA elements, starting with the following subjects NOAA that are addressed in this report: Preface Editors, Ocean System Studies I: Prediction and Resources is dress it, and coordination with other agencies or the first of an episodic series of reports identifying universities in developing NOAA's unique contribu- NOAA/OAR research opportunities and strategies. tion. The contents address a number of issues global in The papers are statements of strategy and are not scope and vital to mankind. The problems are greater meant to be budget initiatives or program imple- in magnitude than any one nation's capability to mentation plans. They focus on strategic research attack them. Given the wide nature of NOAA's char- needed to answer pressing scientific problems for ter, this document articulates salient research areas resource management and for an understanding of where NOAA can make a unique contribution, and the global environment. where public and private issues intersect NOAA's In most cases, the research statements are the re- research capability. sult of a series of working groups and/or workshops As a strategic statement, this document highlights on the topic. They usually involve the input of NOAA the issues to which OAR is committed in the broad line organizations, other agencies and universities. categories listed in the Table of Contents. It reflects The following individuals participated in the devel- the direction we want to go, regardless of the vicissi- opment of these papers: tudes of the budget process. The modulation of the effort may change but not the direction. It is not an exclusive commitment to these areas of research at L Ocean Circulation and Global Climate Change the expense of others. Rather than paint an entire Based on Global Climate Change Workshop re- landscape, this document settles on a handful of high port written by Robert Molinari, AOML, using writ- points on the research horizon at this time. ten contributions or private communications from The papers cover a diverse set of research agen- Kirk Bryan, GFDL; John Diamante, CCAR; J. 0. das, research methods and goals. In order to provide Fletcher, OAR; Richard Gammon, PMEL; Dave a semblance of organization, each chapter is pre- Goodrich, OCAR; J. Mike Hall, OCAR; Don Hansen, ceded by an abstract summarizing the research prob- ACIML; Ed Harrison, PMEL; Stan Hayes, PMEL; Joe lems and opportunity, the essential elements of the Huang, CCAR; Richard Legeckis, NESDIS; Ken research strategy, thequestion of timeliness (the "Why Mooney, OCAR; George Philander, GFDL; Ed Sara- now?" question), NOAA's role, and the benefits to be chik, University of Washington; Jorge Sarmiento, derived from investment in the research. The text of Princeton University; Uwe Radok, CIRES; William each chapter reflects the style of a number of differ- Woodward, NOS. ent work groups. You may expect divergences in Drawn from current research of: Kirk Bryan, S. style, tone and approach. Rather than force uniform Manabe, M.J. Spelman, R.J. Stouffer-GFDL. consistency and editorial prerogative, we have sim- ply tried to insure that each paper provides neces- IL Ridgeflux sary background information on the scientific issue Eddie Bernard, PMEL; Steve Hammond, PMEL; and the sense of urgency surrounding the research, Dick Feely, PMEL; Ed Baker, PMEL; Bob Embley, details of the research questions and strategy to ad- PMEL; Chris Fox, PMEL; Gary Massoth, PMEL; Peter 4 Editor's Preface Rona, AOML; Terry Nelsen, AOML; Steve Piotrowicz, AOML: Atlantic Oceanographic and Meteorological AOML; Dave Duane, NURP; Bill Lavelle, PMEL; Laboratory OAR/NOAA. Glenn Cannon, PMEL. CIRES: Cooperative Institute for Research in Envi- III. Fisheries Oceanography ronmental Science, OAR/NOAA. Gene Fritz, National Sea Grant Office; Jim Schu- maker, PMEL; Peter Ortner, PMEL; Tom Fontaine, GFDL: Geophysical Fluid Dynamics Laboratory, GLERL; Kenneth Sherman, NMFS; Donald Hoss, OAR/NOAA. NMFS; Gary Stauffer, NMFS; Reuben Lasker, NMFS; Robert Francis, University of Washington; Larry GLERL: Great Lakes Environmental Research Labo- Crowder, North Carolina State University; James ratory, OAR/NOAA. Kitchell, University of Wisconsin. NESDIS: National Environmental Satellite & Data X. Sea Ice and Arctic Ecosystems Information Service, NOAA. Jim Overland, PMEL; Carol Pease, PMEL; Knut Aagaard, PMEL; Roger Barry, CIRES/ University of NMFS: National Marine Fisheries Service, NOAA. Colorado; Don Cavaheri, Goddard Space Flight Cen- ter; Gary Wall, Navy /NOAA Joint Ice Center; Norbert NOAA: National Oceanic and Atmospheric Admini- Untersteiner, University of Washington; Vera Alex- stration. ander, University of Alaska. NOS: National Ocean Service, NOAA. V. Estuarine Systems Brian Edie, GLERL; Larry Swanson, SUNY/Stony- NURP: National Undersea Research Program, OAR/ brook; Garry Mayer, National Sea Grant Office; Don NOAA. Atwood, AOML; Herb Curl, PMEL. Based on Coastal Marine Research Plan PDP, Larry OAR: Office of Oceanic and Atmospheric Research, Swanson, SUNY/Stonybrook; Carry Mayer, National NOAA. Sea Grant Office; Harold Stanford, NOS; Bud Cross, NMFS; Jim Thomas, NMFS. OCAR: Office of Climatic and Atmospheric Research, OAR/NOAA. VI. Marine Chemistry, Biology and Climate Written by: Bill Graham, National Sea Grant Of- PMEL: Pacific Marine Environmental Laboratory, fice. OAR/NOAA. Drawn from research by: Steve Piotrowicz, AOML; George Harvey, AOML; Peter Ortner, AOML; Tim SUNY: State University of New York. Bates, PMEL; Richard Gammon, PMEL. Using material produced or provided by: Don Atwood, AOML; Alex Pszemny, AOML; Steve Piotrowicz, AOML; Tim Bates, PMEL; Richard Gam- mon, PMEL 40 Chapter I Ocean ange Circulation and Global Climate Change 4_ -A@ 4 M", OW Proble and M @0 -Researck -p un -0 q 44 t ,por @en For several years concern@ has' be 'Ah k4ka"il-k-01A," "climatigsyst h' h WW em can e c, aracterized as a global the earth's global environment 'May be! cha"itg,ing"'in _'heaiengim having tw ,o working fluids (the oceans and 4J-t-- rin -a" m os h@W 1' `@ -kls@h4tnlainlyfrom theequa- ways to which-we can- not- easily-,adjust',tize 11,16h p ere) t ranspor trend in atmospheric carbon dioxide and- the known -rop '. olai regions-The research radiative effi@ts an rrcent.-- @pakisllvi beab -,to'u-,tid@r,@--tith'd"andpt@dict.'the,behavior le, data su tingsignificantch nge5i theeafth,'i@bzone--4_ clkllhte)'@f the'sysiem-im-0 scales. gges it In ,@@ent time egy. Aayer are evidence ofM@-@b A rf,ll reseaktl @stYdh Js-directea at resolving the ie 1@iidi&A wer n or o ed"i der t the global climate.- an ts-n _Jonger@qpqssi aMie@, Meg" zR t eb vior of the system that pe, It is clear that mj icipartt,,@- oal.. irsi, h eha 4 qeds, 'u-ii r,-gto6d@is@defi*ned@f@rom,currentobserva-,, I @o J "h -e q -';q - to b, in the global enp h s 41 -a,, the scaioewh-e@r'e it aff-e'ds-the, @ecpqstructipqs'qf its g1lo -envrro"ent,-@,-,,1,, 'i,anif post behavior. Then, h d ng,-25 -,diagnostic studies of the Je tl,@vu grgo n i n&Y jsw@ ways that we do not Pity e@ .6h - Tati6h�'iqnd,through--numericaI simulations of the activity, coupled with_the: natural paritibitityJAi,th. I sei ',cou- and-_ ns. An- @71:esp p oatm, ere ocea @ M global climate 109 . V _ ,, . Pie _o@p swers,to@crilicatq, f@tiq, s- -opided- by focused research for the ft"re. Current, and.,anticipate4,'cMi@g@s@in,,tho-@ Y @pi wz -Podu, @-:40sqvv 6 ersiahdingof the system and' global environmenf "ll'i, ce@@ignip@ ie6h@64@1 f"refiO'b tfi-u@hd models ultimately pro- social and political- 16de& k4ilied" repare& ns cooperation with@other nations; *ust@be-p @vide Ihe -,ba8is P-piedictio of the-future behavior of address. @@the @s @t "roa NOAA has, the -respon5ibility,;io@uWe'--rs-reia' hd4hW_WikX-`@,,__ Jas must be' addressed. These are: timately forecast interannu 1_j4ecadq,(qn4,1qjr -be, refined and- verified; niila @ ?I @@a trvu tio -,models'@must, climate change. On these' time scales, the,,,transport obserbvititnuz' systems musthe designed and deployed storage and exchange 6 -fiki -@ndsub-@ jor, 6cean"dimate m onitaring; and new in situ technol- fhiat bythesuir surface waters of the world oceans and the -mass, mo--i, nitoring must be developed. mentum and energy exchqrges,,ke;W_ theoqqq engine o t t gI bal he A ab u he o at the atmosphere, are of crucial importance - @eZ in, iii& that need to'be answered are: ing climate change. @,?U jVt @4,,@[email protected] -heat @ transport -,of the- oceans (how q,- Ahom -is mu d from where to where)? `hbw is1he tnafisportperformed (what water in asses 'i's t 7ke:invotpec andhqw@ hesy@tem-forced)? M aij !t,007@_nq=@ ILL_ 6 Ocean Circulation and Global Climate Change ------ 7-1 .4- * what is the temporal and spatial variability of the My Now? i behavi .or2 * how do changing ocean conditions influence the, @-ilfov The world, oceans play a central role in climate atmosphere and what ocean features are most im- change, and the problems presented by global climate portant in this respect? change -are very real. Recent scientific, advances and, These questions lead to increasingly more specific 1'-,,,'planned technological improvements fe.g., super- A questions which the research strategy mustaddress. .43&t- computers and satellite systems) now make it possible A specific research question of importance in take a truly global look at the earth, including the All interannual climate variability concerns the effect on iMI, world oceans, as a system and engage, for the first time, the phenomena associated with El Nino and the South- in a national (and international) scientific program to -b t ern Oscilation (ENSO) of water mass exchange's e ween 'i! understand and predict changes fboth natural and man- the Indian and Pacific Oceans. On interdecadal and @Z',T made) in the global environment. longer time scales, some important research questions concern: the role of North Atlantic deep water formation, 'Iv@-@ Benefits and exchanges -with the northern- branches,of, the -@,M subtropical gyre, in regulating the thermohaline The earth may face climate and environmental circulation, and --im2k changes on time scales of importance to living genera- t, -V av the identification of feedback control mechanisms 0@,Y tions, their civilizations, and future progeny. These in the Atlantic thermohaline circulation and- their tXklr changes may be due to natural variability, or may be role in possible major, and relatively sudden,, C,,li7-,,, human-7inducled,. such as the effects of increasing C02 mate changes taking place -on -the century time.`@" levelsand changes in ozone levels. Reqardless of the' AN - scale. 36 source of change, the ability of scientists to predict the On both interiinnual and interdecidal -time sca direction and magnitude of change will allow nations-- some important research questions relate to: 4: and international institutions to be better prepared to understanding the dynamics of the ocea,nl accommodate to these changes, or to otherwise mitigate SAI -'atmosphere fluxes -,of heat -moisture,and-mom n@,@ their. consequences. The research strategy @laid out for. I e tum, particularly 'With respect toareas controlling ocean_ circulation is directed toward providing fore- I the tinie-variability of water mass formation, casters and modelers with refined ocean models havinq the relationship of changes of heat content-, of the':;1.-'Z predictive capability commensurate with the existing 1 equatorial- zones and changes in the position of the 11-:4,," atmospheric, models and to support development of Intertropical Convergence Zone (ITCZ) and their. -,-- coupled models of the two fluid systems. relationship to climate variability in higher lati- W, tudes, particularly in the Atlantic basin. properly pose the sequence of questions that needs to be answered. In their last conversation, Alice B. Toklas asked Ger- trude Stein, "What are the answers, Gertie?" Gertrude Stein's reply was, "But, what are the questions, Alice?" Research is directed toward a goal, most often posed in terms of a question, or questions, to be answered. In particular, there is usually a logical The traditional definition of climate is the long- sequence, or hierarchal order, of questions leading to term behavior of the atmosphere, described in terms the goal or purpose. The most general question at the of averages of the basic parameters that characterize top of the hierarchy must set the proper direction for the weather, such as temperature, pressure, wind a research strategy designed to achieve the goal. velocity, and precipitation. Climate is then distinct Therefore, the first order of business in formulating a from the climate system, which consists of those ele- well focused ocean climate research strategy is to ments of the earth system that determine climate. Ocean System Studies 1 These are, at a n-dnimum, the total atmosphere, the the climate system provides a self-consistent meta- oceans, the land surface and the cryosphere (Fig- phor for formulating critical questions and a research ure 1). If our ultimate goal is prediction on climate strategy to answer them. In order to predict changes time scales, then we must concern ourselves with in climate, we need to answer the question: simulating and predicting the interactive processes What is the temporal and spatial variability of of the climate system. the forcing fields driving the two working fluids The essential nature of the climate system illus- of the global heat engine and what is the vari- trated in Figure I is a "global heat engine" involving ability of the response of these fluids in space two principal working fluids, the atmosphere and and time? oceans. The oceans and the atmosphere redistribute This constitutes the most general question which heat between the "sources" and the "sinks." The the research strategy must address. major source areas are located mainly in the equa- torial and tropical regions, while the major sinks are Figure 1: The Climate System as a Heat Engine. The dynamic and located principally in polar areas. This description of thermodynamic coupling of the earth's atmosphere, oceans, cryosphere and land surface provide the global heat transport of the climate system. N STRATOSPHERE"I"':w, POLAR TROPOPAUSE V V mob M RIU0 p@ MA 01- @_g 2 101k A U, 4 2P NN. ANTARCTIC ICE CAP 8 Ocean Circulation and Global Climate Change The ultimate driving forces of the climate system's Consequently, even more so than with the weather heat engine are the incoming solar radiation, the prediction problem, we must depend to a great ex- outward radiation flux (the reflected albedc, field tent on development of sophisticated computer and planetary radiation), the gravitational and rota- models that can be used to perform various numeri- tional (Coriolis) forces. In terms of the oceans, the cal experiments (hypothesis testing, sensitivity analy- main fields associated with the forcing of oceanic sis, simulation, diagnosis and predictions). Data on component of the heat engine act at the interface the past behavior of the climate system will also be between the oceans and the atmosphere. These fields necessary. Back beyond about two centuries in time, are the distribution of wind stresses, the fluxes of these data must be derived from indirect sources of sensible and latent heat, and mass fluxes. The mass information, such as lake levels and ice cores. These and latent heat fluxes are mainly concerned with reconstructed time series will serve to define the moisture transport at the ocean surface. The heat flux behavior that we seek to understand. The retrospec- also is closely related to sea surface temperature (SST) tive data will be used, along with current data from distributions. The main fields associated with the monitoring systems, in the model development (veri- forcing of the atmosphere are the same surface fluxes fication, calibration, and assimilation) and applica- and SST field, but the corresponding quantities for tions (simulations and predictions) portions of the the land and cryosphere surfaces also are involved in NOAA program. Understanding gained from well- driving the atmospheric circulation. Hence, defining formulated research efforts into the fundamental and understanding the spatial and temporal vari- oceanic processes and circulation patterns will serve ability of these ocean surface fields must be an im- to advance the development of the circulation mod- portant concern of the research strategy. els. Diagnostic efforts and simulations using the Starting with the general question above, a logical advanced models provide more sophisticated insights sequence of questions for ocean climate research fol- into the climate system, refine the research questions lows with: and so forth (Figure 2). � What quantity of heat is transported by the global Calibrated and verified models that can explain oceans and how is that transport distributed the past behavior of the climate system, including globally? 11 global climate changes" in prehistoric and even his- � How are the different oceanic water masses in- toric times (e.g., the Little Ice Age) will eventually volved in the specifics of the global heat trans- provide the basis for reliable predictions and assess- port? ments of future changes in the world's climate sys- � What is the temporal variability of the oceanic tem. An early step in the process will involve appli- heat transport? cation of surface wind stress fields reconstructed from At the same level of generality, the next questions about a century of historic observations to simulate relate the role of the oceans back to the atmosphere in the global ocean circulation. The corresponding his- the overall global heat transport problem: toric record of SST (Sea Surface Temperature) obser- � How do changes in the oceanic heat transport vations, including the position of oceanic fronts, will influence the atmosphere? be used along with the sparser records of land-bas'ed � What are the most influential features of the observations to simulate the past circulation patterns oceanic heat transport system affecting the atmo- of the atmosphere. Limited historical records of sub- spheric variability? surface ocean observations are available for valida- Sequences of increasingly specific questions fol- tion. low from the above, and these are posed in the subse- On climate time scales, most of the true external quent sections. forcing functions are now variable: they must be In order to predict climate change, we must deal explicitly dealt with in the prediction problem. Also, with the complexity of the earth as a system, where the land, ocean and cryosphere are fully coupled to significant changes may take place gradually over the atmosphere in both a dynamic (mechanical en- centuries or relatively dramatically over a time pe- ergy) and thermodynamic (thermal energy) sense riod encompassing the lives of a few human genera- (Figure 1). If we look at the climate prediction prob- tions. We can not afford to wait while we accumulate lem from the side of the atmosphere, what had been adequate time series of basic geophysical variables, essentially fixed ocean, land and cryosphere bound- nor can we afford to undertake what inadvertently ary conditions for the weather prediction problem mdght become irreversible in situ "experiments" on become time-variable as we address the longer cli- the climate system itself. mate time scales. For example, the sea surface tem- Ocean System Studies 1 9 NUMERICAL CLIMATE MODEL SIMULATION ment) programs brought early recognition that a cli- AND PREDICTION mate program would need to focus on numerical modeling as a basic means, and prediction of climate change on different time scales as a basic end. Fur- (a) Numerical,., Climate Model thermore, the understanding of the roles of the oceans' transport of momentum and heat (including inter- actions with the cryosphere) were seen as being deci- @trn..Ph.,,C ai ra sive for the development and application of climate .'e'al models. In the seminal GARP document (D66s, et al., Circulatiom p icq jqq@, Model 1975), it was emphasized: "On seasonal, annual and :2 decadal time scales, climate models must take into Air-Ocean account an interactive upper ocean and sea ice... On Boundary Interactions long time scales (e.g., 100 to 1000 years), considera- tion must be given also to changes in the deep ocean @N NumericalZ 'Ocean and the variations of the continental ice sheets... The 8P, Gerieral role of the oceans is believed to be a dominant one on Circulatic climatic time scales." lkp Model, el@ 'roc, _6 _,r9c, U Figure 2: Application of Ocean Circulation Models and Ocean Data in Two decades ago, E. Lorenz (1968) speculated that Climate Simulation and Prediction. Ocean data (circulation, thermo- the large scale behavior of the climate system may dynamic state, surface fluxes) provide (a) initiallboundary values for exhibit nonlinear instability under certain circum- ocean circulation models coupled with atmosphere General Circulation stances, particularly for its long term behavior. Evi- Models (GCMs) and models for cryospheric and terrestrial inter- dence for this kind of behavior in the global climate actions, (b) identification and removal of systematic errors in historical system is the apparent abruptness of the onset of at data for use in the climate model calibration and verification process, least three major cooling episodes or glaciations over and as statistical constraints (in data assimilation mode) for (c) simulations and (d) predictions of climate changes. the last 150,000 years. The evidence suggests that a substantial portion of the cooling effect may have been experienced within the interval of about a cen- peratures shift, significantly with the seasons, as does tury (Lamb, 1982). There also is evidence that less the land surface albedo and sea ice extent. dramatic examples of sudden transitions in the cli- On time scales beyond the seasonal, the dissipa- mate system have taken place within the last 150 tive processes cause the atmosphere to "lose mem- years. Fletcher et al. (1979 a,b) noticed sudden drops ory" of the initial conditions, so that the boundary in the strength of the Southern hemisphere Wester- conditions dominate the behavior of the atmosphere lies in 1870,1903 and 1917. on long time-scales. However, the natural response The conventional concept for climate change sees times of the oceans, cryosphere and solid land are long-term change taking place continuously and much longer than those of the atmosphere. Conse- gradually as a quasi-linear process. Many adherents quently, while the atmospheric side of the climate of the greenhouse warming hypothesis accept this system is dominated by boundary conditions on cli- concept implicitly and expect change to take place as mate time scales, the initial conditions for the oceans a gradual, more or less irreversible trend. The evi- and cryosphere will still play a major role in the dence for a gradual global warming since the begin- prediction problem for the total climate system. ning of the industrial revolution in the mid-eighteenth Hence, observations of the present state of the oceans century is usually taken to support this gradualistic are essential for model initialization. concept of climate change. Oceanic relationships to the boundary condition However, Broecker (1987 a, b) has presented a dominated features of the atmospheric system are challenge to this conventional view of climate change. @ 'm S he G9e n P ,a Ci@ , .1-t Model certainly candidates for focused research. Experience He points out the evidence for a number of very with the GARP (Global Atmospheric Research Pro- rapid transitions, mostly from and occasionally to, gram) and GATE (GARP Atlantic Tropical Experi- cold periods (including full blown Ice Ages) and 10 Ocean Circulation and Global Climate Change cites evidence that dramatic changes in atmospheric Basic observations concerning the role of the carbon dioxide content may have been involved in thermohaline circulation in long-term climate change such climate changes. He also concludes that the suggest that high latitudes are regions of critical pro- evidence suggests a key role for some mechanism in cesses (including ice processes) and that salinity the North Atlantic involving a shut-down of the changes are critical, if not decisive, for the operation Atlantic portion of the density/ salinity-driven deep (or shut-down) of the thermohaline circulation sys- c@irculation (thermohaline circulation, Figure 3) trans- tem (Bryan et al., 1988). Hence, changes in salinity porting the excess cold waters of the Atlantic into the (along with temperature) are the key observables Pacific. Recently, Manabe and Stouffer (1988) have that must be mapped and monitored. Other impor- demonstrated numerically, using a coupled ocean- tant observables are related to surface fluxes, inter- atmosphere simulation model, the existence of a bi- actions with major wind fields, and sea ice inter- stable equilibrium condition for the climate system, actions. at least within the approximations and limitations of the model. Such a bistable state involves two modes of the climate heat engine having markedly different statis- tical averages of the principal variables (i.e., climate). Each mode is stable for a wide range of boundary values and forcing conditions, but both states are The field of global sea surface temperatures is simultaneously possible for a relatively narrow range generally recognized as a key climatological vari- of the boundary and forcing conditions. The climate able, as it is the major boundary condition related to system can make relatively rapid transitions between the forcing of the atmospheric circulation that is also the two modes of the bistable state, under the condi- reflective of ocean processes. Analyses of the avail- tions where both modes can exist. The climate sys- tem is actually unstable for that narrow range of con- ditions and markedly increased variability over a Figure 3: The Great Ocean Conveyor Belt (Broeker, 1987b). A range of frequencies can be expected. simplified view of the thermohaline circulation system of the global oceans. &lamb Sea-to-Air Heat Transfer J%Ib A C 0 PACIR N AM Ocean Systeni Studies I able SST records (see for example Folland et al., 1986, relative strength of the interdecadal signal is stronger and Parker, 1987) identify three prominent modes of in the Atlantic Ocean and Atlantic basin land areas temporal variability, other than the annual signal. than it is in the Pacific, at least outside of the Tropics These are: an interannual signal representative of the (Cayan, 1987). The most prominent interdecadal cli- El Nino related phenomena and having a mainly mate signals relate to the Saharan/Sahel drought in tropical ocean focus, a signal reflecting interdecadal North Africa, and the Mid-Westem drought cycle of fluctuations, and a long term warming trend. It is the United States, while the drought pattern in North- tempting to identify the long term trend as being a eastern Brazil near the equator does not exhibit a secular change due to greenhouse warming, but given relative prominence of the interclecadal component the limited duration of the available record it can of the signal over the interannual (Lamb et al., 1986). only be characterized as representative of century The relative strength of the interdecadal climate sig- scale or longer fluctuations. nal compared to the interannual oscillations tends to increase as we move North and South away from the Equator in both oceans. The relative strengths of the interannual and interdecadal climate signals over the globe can be de- scribed as a stronger interannual effect modulated The strongest, most coherent, interannual climate by a weaker interclecadal signal for the Tropical Pa- signals known are in the Equatorial Pacific and are cific basin as a whole. To some degree this holds true categorized as ENSO (EI-Nino-Southern Oscillation) for the Equatorial Atlantic. A stronger interdecadal phenomena. The ENSO manifestations include: the signal modulated by a weaker interannual effect tends see-saw oscillation (Southern Oscillation) in atmo- to be more characteristic in the higher latitudes. spheric pressure as recorded in Darwin, Australia in the West, and Tahiti in the Eastern Equatorial Pacific,- El Nino, the strong thermal anomaly in the Eastern Tropical Pacific, off the Peruvian and Ecuadorian Coasts; and the general pattern of tropical Pacific temperature and sea level anomalies and propaga- A possible sequence of global ocean circulation tion of trans-Pacific Kelvin and Rossby waves. There programs, prioritized as to allocation of resources, seems to be some degree of correspondence in the and designed to make rapid progress in understand- Equatorial Atlantic, but whatever signal exists there ing of climate change on the interannual through does not seem to be as strong or coherent. An ex- interdecadal /century time scales, might start with ample of a relatively pronounced interannual signal the interannual signal. It mostly involves the ocean in the Equatorial Atlantic is the upwelling in the Gulf surface, atmosphere and some major surface current of Guinea in the Eastern Tropical Atlantic. systems. Measurements in the lower atmosphere and The climates of the major land areas of the Pacific within the ocean surface layers are the most cost basin (Australia, New Zealand, Coastal China, Peru, effective. The Equatorial Pacific, with the best de- and California) also tend to show a strong inter- fined and localized effects of this kind, would appear annual signal as the dominant effect at climate fre- to be the best place to start, subsequently moving quencies lower than the seasonal/annual band. Gen- outward to the Tropics, then the middle latitudes in erally, the interannual signal in the Pacific tends to the Pacific or the Equatorial Atlantic. The neighbor- become weaker and less coherent at higher latitudes, ing Indian Ocean, which behaves sympathetically but remains pronounced nevertheless. with the Equatorial Pacific, would also receive early In the Atlantic, the signal strength and coherence attention. The Atlantic equatorial and tropical zones of the interannual signal are also diminished as we could follow next if not given attention earlier, and move away from the Equator, and diminished over- higher latitude areas of the Pacific and Atlantic would all compared to the Pacific manifestations. In the probably be last. Northern Atlantic, the interannual oscillation fre- At some point, the study of the interdecadal signal quently is not as pronounced as the decadal/ and the deeper ocean, with the greater associated interdecadal climate signals (Cayan, 1987). costs and technical difficulties, would get underway. The decadal/interdecadal signal shows almost an Again, the best place to start would also be where the opposite character from the interannual signal with interdecadal signal trends to be clearest and strong- respect to its geographic distribution. Generally, the est and that would be in the North Atlantic, probably 12 Ocean Circulation and Global Climate Change following to the South Atlantic, the South and North Pacific. This broad strategic approach to programmatic prioritization and sequencing is, in essence, what NOAA has been following since the inception of the The programmatic basis of the NOAA research OACIS (Ocean-Atmosphere Climate Interaction strategy in ocean-climate, as articulated in the 1979 Study) program in 1984, and in certain respects for plan (Fletcher, et al., 1979), led to a series of budget- the last ten years since about 1979. OACIS, which ary initiatives for specific program elements over the was the predecessor of the TOGA (Tropical Ocean next few years. Out of this planning and budgetary and Global Atmosphere) Program, and TOGA itself, process, emerged four new core programs (see Fig- set the strategy firn-dy in the mode described, for the ure 4) which were to form the foundation of the interannual process by 1985 (Figure 4). The STACS NOAA programmatic strategy in ocean-climate re- (Subtropical Atlantic Climate Studies) Program set search for the next decade. the pattern for the decadal problem with its Atlantic All four core programs have supported the devel- emphasis in 1982. opment of basic means (technology and techniques) The prioritization described above would be flex- required to conduct subsequent programs. All ex- ible. Certain investigations might be conducted in cept the Comprehensive Ocean-Atmosphere Data Set geographical areas out of the suggested sequence. (COADS) effort, have focused upon specific regional Any critical features of the ocean surface fields in- problems and basic processes that are representative volved in the forcing of the global climate heat en- of important aspects of the general ocean-climate gine would be candidates for special priority. Ex- system critical to the development of understanding amples include regions of high wind stress in the of climate change. COADS is fundamentally global Southern Indian and South Atlantic Oceans, the Ant- in concept and approach, but with regional resolu- arctic Convergence Zone and major oceanic fronts in tion. Together, the four programs cover the entire the North Western Atlantic and North Western Pa- frequency band in the climate system identified as cific Oceans. the concern of NOAA's long-term effort (seasons through centuries) and represent the essential first Figure 4 Evolution of NOAA Ocean Climate Programs. A summary steps toward realizing the overall predictive goals. view of NOAA1OAR ocean climate programs since 1975 with meteoro- logical antecedents since 1963, and projected programmatic strategy through the first decade after the year 2000. CLIMATE SYSTEM TIME-SCALES (1 ON YEARS) @IAVP - loterannual Variability Program 3PBV-'Paleo-Bi6IogkaI Indicators' 4 N'-_@ -2 BONE%Y CPF_ C@OoPerative Program Efforti 1963 @V W@ (with WOCE, GSP, GQFS) "510VP --Interdec@atvariabiljty Program eather --meather-, Time-S(,-,ales GATE 6GOVAC'- Global Ocean variabittiyand Climate 1974 Weather future thru ,Sea' n6al sohai/A & k@, Viri Time-Scates NO Climate N=O Annual thru - , A' TOG] interannuat EP7s OACIS% 9>@@ ZiAVIf 29 Time-Scales 1978 1984 ,,7--", 185 1 1991-19 5 0- GOVA 777@ C6 STACS E, Decadal 1982 GSP, GOFS) Ihru 1990's >2000 )nte TTO ABVP1 rder-adal 5 ID P 9 Time-Scales 1989/1990 >1 95 1985 JR7 I All N=2 @X Interdecadal th u r -PB71\ /50 A D 7ZsF,__ A Centuries 'R Time-scales _N,9@901199,1 N=3 1988 '' , t 1975 1980 1985 1990 1995 2000 PROGRAM START-UP DATES (Year) Ocean System Studies 1 13 of NOAA's chemical tracer work in the mid- 1980's. -1 V@r A f1Q The chloroflourocarbon tracer studies had the addi- tional advantage of being complementary with other Two of the core programs, Equatorial Pacific global chemistry and carbon dioxide flux work sup- Ocean-Climate Studies (EPOCS) and STACS, are the porting NOAA's ocean climatic mission, i.e., the earliest and are officially funded as ongoing base ocean's role in the greenhouse problem. The work programs to this day. In 1978/79, NOAA initiated has progressed at a low funding level, but has pro- the EPOCS program to study the causes of the large ceeded more or less continuously, supporting other scale sea surface anomalies in the equatorial Pacific NOAA climate research programs. and their effects on the atmosphere. The Equatorial and Tropical SST distribution is believed to be a major factor in the forcing of the atmospheric compo- a a Ila nent of the global climate heat engine. Because of the large interannual signal in sea surface temperature in The fourth effort, the Comprehensive Ocean- the eastern equatorial Pacific, initial studies were di- Atmosphere Data Set (COADS), is also internally rected at this region, funded and has continued at a steady low level of Oceanic heat flux is generally believed to be an effort for nearly a decade, but unlike the other three, important process involved in interdecadal climate does not involve an actual field measurement pro- variability. However, long-term monitoring of the gram. COADS represents a systematic attempt to use sub-surface ocean processes on interdecadal time past data sets relevant to ocean climate on time scales scales presents'a formidable challenge. In 1982, NOAA beyond a decade or two. Since 1854, ships of many began the STACS program for studying the transport countries have been taking regular observations of of equatorial heat to higher latitudes in the Atlantic local weather, sea surface temperature, and many and related processes. The initial emphasis of STACS other characteristics near the boundary between the was directed at the western boundary currents of the ocean and the atmosphere. In later years, fixed re- North Atlantic Subtropical Gyre, in particular, the search vessels, buoys and other devices have contrib- Florida Current and Gulf Stream, which have been uted similar marine reports. The collection of surface shown to be a major component in the wind-driven data spanning the. global oceans, from the mid- modes of Atlantic heat flux and the subtropical gyre. nineteenth century to date, is the historical ocean- Given the interdecadal and interannual implications atmosphere record utilized in the COADS effort. To- of the program, it is shown in Figure 4 as based upon gether, these four programs form the four legs upon both of these frequency bands. which the NOAA ocean-climate strategy stands. Mr A third effort, Transient Tracers in the Oceans (TTO), has been intermittently funded as an internal rammatic e As not ent effort supporting other programs, and has been di- in Figure 4 started earlier than the others, and will rected at a sequence of regional problems having assume growing importance in the future. This is the much broader implications for climate. The 1979 ocean circulation /climate dynamics modeling devel- NOAA strategy document noted that plans were opments that have been ongoing as base efforts in underway for one study, "Transient Tracers in the the Geophysical Fluid Dynamics Laboratory (GFDL) Mh Mset o Mprograommatic Meorts Rnot i eRntiie A fift Ocean," using "radio-carbon and tritium tracers," to since the mid 1960's. These model developments and study "residence times, transport mechanics and numerical experiments also cover the entire climate vertical mixing." However, under the subsequent frequency band of concern. Like COADS, they do not budgetary constraints, the tritium work was judged involve field work, but are involved in development to be too expensive for NOAA's limited resources, so of essential techniques needed to achieve the ulti- that chloroflourocarbons (freons) became the focus mate predictive goals of the program. 14 Ocean Circulation and Global Climate Change ity involves a broadened focus for TOGA to include the, as yet poorly understood, "global teleconnec- tions," relating global interannual oceanic and atmo- spheric phenomena. The full ENSO cycle, rather than just the warm episodes, and the intraseasonal mode Since the inception of EPOCS in 1978, NOAA's of variability also are included in the new focus. ocean-climate research strategy for interannual time The limits of the tropics, by usual definition, are scales has been executed through a sequence of spe- confined by the � 30 degrees latitude band, which cific programs: EPOCS, OACIS, and TOGA (see Fig- accounts for about one half of the Earth's surface. ure 4). EPOCS concentrated on the equatorial ocean However, as a practical matter the ocean concerns of between about 20 degrees N and 10 degrees S lati- TOGA are approximately bounded by the � 40 de- tude, with the major initial emphasis on the domi- grees latitude band, providing a beginning to exten- nant equatorial Pacific Ocean phenomena. In 1983, sion of the monitoring system into the mid-latitudes. starting with the EPOCS scientific plan, the National The extension of the monitoring system is essential Academy of Sciences (NAS) produced a national plan for understanding the nature of the teleconnections. for a program called OACIS thai was directed to- The massive Pacific warm episode of 1982-83 was ward the problem of inteTannual climate variability. associated with a clearly defined pattern of inter- NOAA initiated its part of OACIS in 1984, building ocean teleconnections around the equatorial belt. around the continuing EPOCS effort. When the World These evolved over a period of about three years to Climate Research Program (WCRP) adopted the strat- include a strong and well defined global signal in egy of OACIS, it was developed as a full interna- rainfall, atmospheric circulation and SST. New analy- tional program called TOGA. In 1985, NOAA sub- ses of the historical data have shown this to be a sumed the EPOCS and OACIS efforts under the recurrent mode of interannual global variability in international program plan (see Figure 4). the tropical ocean and atmosphere. The tropical Atlantic also has a mode of inter- annual variability that has many similarities with the MMMMMM Pacific Ocean manifestation of ENSO. The tropical Atlantic measurement program under TOGA during The program title, TOGA, defines the essence of the next several years aims at providing data to check the main concern: the role of the dominant inter- the hypothesis that changes in the position of the annual processes of the tropical ocean and its Intertropical Convergence Zone (ITCZ) relate to interactions with the global atmosphere producing changes in the heat content of the equatorial zone. interannual climate variability. These efforts, which will probably extend beyond After extensive review and debate, the TOGA the TOGA time-line, are also candidates for TOGA Program proceeded with the broad strategic concept interactions with the World Ocean Circulation Ex- outlined in Section 2.0, continuing to concentrate periment (WOCE) Core 3 Project that is concerned resources allocated to investigations in the interannual with ocean gyre dynamics. climate research area, first for investigations of low latitude Pacific Ocean processes, then moving out- ward to higher latitudes (Tropics and Subtropics) in the Pacific, the Indian Ocean, then the tropical Atlan- tic, etc. During the initial implementation stage of TOGA, The 1990's will be a period of intense interagency the major concern was the strong equatorial Pacific and international research in the oceans, involving at warmings which occur at irregular intervals of about least two major NSF-led programs: WOCE, con- 2-7 years. However, now both diagnostic studies and cerned with basin scale ocean circulation, and the theoretical work have more clearly delineated the Global Ocean Flux Study (GOFS), concerned with global tropical linkages associated with ENSO vari- surface fluxes in the global ocean. These programs ability. Recent work also indicates an important role offer important opportunities for cooperative efforts of the large-amplitude, coherent atmospheric fluc- with NOAA programs. With leveraged interagency tuations on the intraseasonal timescale (30-60 day efforts, achievement of the predictive goals of the waves). As a result, the current description and NOAA program becomes feasible. Therefore, it is understanding of tropical ocean-atmosphere variabil- important that NOAA have its main programmatic Ocean System Studies 1 15 thrusts on both interannual and interdecadal ocean- sible role of ocean processes in these areas in African climate variability in place by the early 1990's. Fig- and Brazilian drought, and the effect on ENSO of ure 4 shows the existence of these Cooperative Pro- water mass exchange between the Indian and Pacific gram Efforts (CFE) for the decade of the 1990's. Oceans. The problem of ocean-atmosphere "tele- STACS, concerned with deriving climate indices connections", particularly involving mid-and-high for western boundary current phenomena, is shown latitudes will probably remain a conundrum, so long in Figure 4 as, more or less, operating off-line by as the observational framework remains confined to itself before interacting with the interannual and the traditional TOGA � 40 degrees latitude band, interdecadal ocean-climate program elements in the and mainly in the Pacific/Indian Ocean Basins. In early 1990's. By that time, STACS will have accumu- addition to expansion of observational networks, lated about a decade of time series needed to address cooperation between program elements focusing on the cross-effects between the interdecadal and inter- both interannual and interdecadal variability (Fig- annual time scales, particularly in the sub-tropics, in ure 4) will be essential for progress in understanding time to interact with the new elements designed to the true causal mechanisms for the teleconnections. address these questions in the early 1990's. TOGA probably can not answer all of the impor- tant research questions concerning the ocean's role in interannual climate variability within existing levels of resources and within a single decade. A follow-on program will need to continue investigations into the Meridional circulation involves the overturning more complex issues of higher latitude responses of the ocean. It brings into play both the surface and and the global ocean that may have been started intermediate layers of the ocean and the deep and under TOGA. bottom layers. Hence, the aspects of the ocean circu- The follow-on program is tentatively named lation that relate to interannual and interdecadal phe- Interannual Variability Program (IAVP) (Figure 4). nomena must be connected through the merictional In actuality, it may be an expansion and extension of circulation. Recent model and observational results TOGA. The broader based program must build upon suggest that we need to have a better understanding both EPOCS and STACS, rather than mainly EPOCS of the meridional heat transport from the tropics and as has been the initial situation with TOGA. Also, its variability. For instance, some model simulations this program will need to interact with a NOAA of the ocean indicate that the time scale of El Nino interdecadal ocean variability program element hav- events is determined by, the heat loss from the trop- ing an Atlantic Ocean focus (see Figure 4). Activities ics during a warm episode and, the time required to discussed below could be initiated, either as part of a replenish the western Pacific warrn pool. new interannual variability program, or under the Meridional heat transport is crucial in both the later stages of the present TOGA effort. cooling and warming processes of the global climate The EPOCS, TOGA and the STACS programs need heat engine. In the Atlantic, unlike the Pacific, there to be enhanced by observations in new areas, while is considerable northward heat flux across the equa- continuing observations in selected areas already tor, related to the thermohaline circulation in that explored on an operational monitoring basis. The basin. Variability in this cross-equatorial flux will area of EPOCS and TOGA (or post-TOGA) research have important implications to heat flux further north. should be expanded poleward to include interactions In both basins, processes on the western boundary of the equatorial/tropical circulation systems and apparently have a large influence on meridional heat subtropical gyre circulation in both the Atlanti@ and flux. A better understanding of the interaction be- Pacific, and the area of STACS concern should be tween the equatorial circulation and the subtropical increased equatorward for the same reason. The same gyres, with particular emphasis on the western observational framework applied in the earlier stages boundaries, is required. of STACS and TOGA should be used in the enhance- The meridional heat flux away from th e equato r in NIEMEN W@W_ -1 ment: two to three years of intensive observing peri- the tropics and the cross equatorial heat/chemical ods should be directed at identifying important in- tracer fluxes are key components of the global de- dices and the methodology to monitor the indices. scription of ocean circulation required by global Priority also must be given to full expansion of change studies. Measurement of these transports near monitoring and research effort into the Atlantic and the equator requires that special consideration be Indian Ocean basins, in order to examine the pos- given to the likely importance of boundary condi- 16 Ocean Circulation and Global Climate Change tions (particularly in the Atlantic), abyssal circula- tion, and the failure of the geostrophic approxima- tion at the equator. However, coupling a deep circulation study to the upper ocean TOGA (or post- TOGA) program, as indicated by the various inter- annual and interdecadal program interactions in Fig- W ure 4, and utilizing the satellite altimeter and wind F611owing the over-all programmatic strategy out- fields available during WOCE will permit a coordi- lined in Section 2.0, the next emphasis should be on nated attack on this problem. the decadal, interdecadal and longer time scale ocean The meridional velocities generally are small processes including possible mechanisms of poten- compared to the prominent zonal currents. A combi- tially rapid climate change. The important role that nation of direct and indirect techniques are going to the Atlantic Ocean Basin appears to play in and the be required to estimate these flows. The program thermohaline circulation and the importance of the will require: interdecadal signal in the climate of neighboring land � Zonal transects at tropical latitudes (�10') to areas of the North Atlantic make it high priority as measure density and chemical tracer distribu- an early focus of research into the interclecadal ocean tions. These sections should be repeated after variability problem. However, when we look at the about five years, with modifications based upon implied scope of the program, involving both the what has been learned in the intervening period Southern and Northern Atlantic Oceans, including of time. polar and near-polar regions, as well as the equato- � Meridional cross equatorial transects 00'N to rial ocean area (with emphasis now including deep, 10'S) of tracers in the western, central and east- water processes), the task remains great. Further- ern regions. Enhancement of the TOGA/Post- more, NOAA lacks experience in some of these areas, TOGA upper ocean thermal field measurement such as the Antarctic/Southem Ocean. program by including salinity measurements. One practical solution is to create a highly lever- Determination of surface currents with surface aged NOAA program, strongly cooperative with drifters and satellite altimetry; constant level other (mostly non-NOAA) programs scheduled to drifters may also be needed in very high lati- get underway in the early 1990's. Since the NOAA tudes. objectives are, for the most part, not the same as � Determination of surface wind fields and wind those of the other programs, the NOAA effort can stress using satellite based scatterometer instru- not simply depend passively upon the non-NOAA ments and island meteorological stations in those efforts. The other non-NOAA programs are more of broad areas of the oceans where the thermohal- a pure research nature, in which understanding of ine circulation is believed to be at the surface or processes and overall dynamics is the main goal, within the mixed layer. although guided by climate concerns to some de- � Integration of the above measurements with the gree. However, NOAA's efforts must be directed at GFDL high resolution model in order to esti- development of predictive capability for the climate mate mericlional transports and refine the over- system on longer time-scales, and establishing the all picture. essential base-line observational networks needed � Exploration of eddy variability in tropical re- for the diagnostic and predictive applications. To do gions and assessment of the importance of eddy this, the NOAA program must also be long-term. heat flux. Long-term monitoring would then follow the ex- ploratory studies. Other aspects of the research effort directed toward questions related to the meridional circulation and its relationship to interdecadal and interannual research elements are discussed in Sec- In many ways, the Atlantic Basin Variability tion 4.2. (ABVP) effort identified in Figure 4 is an analogy of TOGA, but there are essential differences. TOGA is an intense ten-year effort, built upon a strong on- going base funded NOAA effort in the Pacific Ocean System Studies 1 17 (EPOCS) which is directed to core questions. The In both the Atlantic and Pacific basins, processes ABVP effort will build upon STACS to some degree, on the western boundaries apparently have a large but must provide continuing funding for the other influence on meridional heat flux. A better under- core effort (TTO), that has never been formally standing of the interaction between the equatorial funded. The ocean tracer work must now also be circulation and the subtropical gyres, with particular applied on a long-term basis in the Atlantic (it has emphasis on the western boundaries, is clearly re- been ongoing in the Pacific), specifically for the inter- quired. decadal problem. The vertical pathways between the surface and The ABVP must form a permanent core program deeper layers of the ocean must be modelled cor- continuing through to an expanded global ocean rectly in order to simulate the effects of atmospheric interdecadal variability focused research effort some- carbon dioxide on climate and natural climate vari- time later in the 1990's. Tentatively, this expanded ability on longer time scales. The most valuable data focus is shown as a new program element, the Inter- sets for verifying that the vertical pathways have decadal Variability Program (IDVP), in Figure 4. been included in ocean circulation models correctly In the context of understanding the mean circula- are measurements of transient tracers, such as the tion, interclecadal changes, and the meridional trans- freons, tritium and bomb-produced carbon-14. For ports of heat and gases, it is clear that studies related this reason the transient tracer program in WOCE is to global climate change must include the intermedi- of particular importance for the continued develop- ate and deep circulations and their interactions. ment of NOAA models. Modeling has been very successful in simulating the The freon measurements taken by NOAA in its upper equatorial ocean in the TOGA Program, but own TTO program in the North Pacific are a unique the deep water circulation has received less atten- resource in this respect, and this program should tion. It is known, however, that the meridional over- serve as the basis for similar observations in the At- turning and deep flows across the equator play a lantic and in particular, those high latitude areas very important role in the global heat balance. In the (North and South in the Atlantic Basin) where deep Atlantic, the mean annual heat flux is northward water renewal occurs. Activities in the Antarctic and across the equator, a dramatic contrast to the sym- high latitudes of the North Atlantic, associated with metrical poleward pattern in the Pacific. Decadal deep water formations and the thermohaline circula- changes in this flux have been suggested as a mecha- tion, will involve extensive NOAA participation in nism for forcing atmospheric circulation changes. other programs on an international basis. Major regional climate questions such as the source Another component of the early Atlantic variabil- of rainfall variability in the Sahel, northeast Brazil ity activities will be initiation of an Atlantic Volun- and the U.S. Great Plains may be intimately tied to teer Observing Ship (VOS) program, similar in sys- ocean variability in the Atlantic basin. tem design to the VOS program which has been More specifically, recent paleoclimatic studies have invaluable in the support of TOGA Pacific activities. substantiated the variability of North Atlantic ther- The proposed Atlantic VOS will provide improved mohaline circulation as a major factor in the initia- upper ocean thermal fields for atmosphere-ocean tion of warming and cooling events observed in the interaction studies. climate record. This is supported by coupled ocean- It should be noted that deep XBT units are now atmosphere model results that suggest that the global available, but are considerably more costly than the climate system may be characterized by two stable standard ones now widely used. Hence, the deep states that differ primarily by the presence or ab- application units will probably need to be concen- sence of the global circulation cell associated with trated in the very high latitude and equatorial areas North Atlantic deep water formation. The geographic emphasized here, and if applied elsewhere, will constraint of the areas of deep water formation being probably need to be "coarsely" interspersed with in the Atlantic suggest this region as an area of early standard XBT units. Observations and monitoring emphasis. strategies are made cost-effective by applying results In the Atlantic, the considerable northward heat of modeling diagnostic studies of the systems being flux across the equator may be related to the thermo- analyzed. haline circulation in that basin. Therefore, variability in this cross-equatorial flux will have important implications to heat flux further north. 18 Ocean Circulation and Global Climate Change First, in the Atlantic, the objective is to derive Separation of the interannual and interdecadal indices for such cross-equatorial, western boundary variability effects is only a convenient artifact of the features as the Deep Western Boundary Current (a programmatic strategy constrained by limited re- major component of the thermohaline circulation of sources. Clean separation of physical processes and the North Atlantic) and surface currents. Modeling circulation based on characteristic frequency is only and observational studies suggest that both of these approximate. Consequently, a later research focus, features play an important role in meridional heat perhaps an independent program element, ultimately flux. During the intensive observing period, direct may be needed to address the full range of questions current, water mass and water mass age observa- concerning the combined interannual through inter- tions need to be taken. To trace water masses, an decadal climate variabilities on a true "global ocean- indication of cross-equatorial flow, nutrients (e.g., global atmosphere" basis. This is tentatively desig- silicate and phosphate) coupled with oxygen need to nated in Figure 4 as Global Ocean Variability and be used. To study water mass ages, helium/tritium Climate (GOVAQ for startup in the 1995-2000 time ratios for short time scales (months to years) and frame. However, certain critical questions will re- halocarbons for longer time scales (decades) need to quire coordination of interannual and interclecadal be applied. The latter approach has recently been research efforts prior to that time. utilized to trace Deep Western Boundary Current The internal NOAA ocean-circulation related pro- waters across the equator. grams that will support the ABVP effort are STACS, Later on in the Pacific, the strategic emphasis which already has been dealing with long-term mea- should include intermediate water mass formation. s.urements of gyre variability in the Sub-tropical At- In contrast to the Atlantic, there is apparently no lantic, EPOCS and TOGA/Post-TOGA (Section 3.0), deep water formation in the North Pacific. Thus, by which must deal with the equatorial Atlantic and the process of elimination, mericlional heat flux is proba- mid- latitude connections on interannual time scales bly strongly related to circulation of intermediate that are strongly coupled with the interclecadal ef- waters. Process studies in the critical formation re- fects of gyre dynamics and interactions in the Atlan- gions of these waters, and large-scale mapping of tic tropical areas. Any TOGA/Post-TOGA Atlantic their distribution using tracers, are required. In the variability work will need to be well coordinated approach described here, the heat flux away from the with the ABVP effort. A possible example of such a equator and the cross equatorial heat /chemical tracer coordinated TOGA-ABVP project might concern the fluxes are key components of the global description branching of the Atlantic Equatorial Current at the of ocean circulation required by global climate change geographical "nose" area of Brazil, between 5 de- studies. Aspects of research concerned with the grees and 8 degrees S latitude. North or South shifts mericlional circulation have been discussed in Sec- in the axis of the branching current system will affect tion 3.3. the transport of warm South Atlantic water into the Caribbean and the Gulf Stream (Lamb, 1972) and consequently is related to both interannual and inter- decadal climate variability in the Atlantic basin. There also remains a possibility of an international pro- gram effort of some kind in this area, that would A major part of the problem confronting human allow broader based cooperative efforts. Again the institutions in making use of climate forecasts.is that NOAA concerns would be directed toward the long- global average conditions have little meaning locally. term and development of predictive capabilities. Generally, a local or regional manifestation of a global change is significantly more extreme than the global average. Our insights into the regional implications of global change are not well developed. The impe- tus for this line of research is the long-term (ten years or longer) ecological consequences and the near-term human impacts (less than ten years) resulting from Ocean System Studies 1 19 large and long-lived climate fluctuations. The com- even the Pacific Basin outside of the Tropics. Obser- plexity of the climate system, and the volume and vatiohal and modeling studies of mechanisms affect- variety of data that are needed, require a strategy ing large-scale, ocean-atmosphere variability (focus- directed initially at a selected subset of these critical ing on the Atlantic Basin, North and South America, regional problems. Europe, and Africa) at time scales on the order of Present research reveals regional problems that decades, but including the interannual modulation, might yield to concerted study, based in part on need to be undertaken as soon as possible. This in- observational evidence linking ocean surface param- cludes critical questions such as the wider role of eters to regional variations, and in part on evidence ENSO events. from experimentation with atmospheric General Cir- culation Models (GCM's). Sea surface temperature (SST) anomaly patterns have been correlated with U.S. weather patterns on interseasonal to interannual time scales. For instance, El Nino SST anomaly pat- terns have been used with some success to predict atmospheric conditions over portions of the U.S. Mid- A variety of national and international programs latitude SST anomaly distributions in the Pacific have having common interests, or complimentary concerns, been used in a sin-dlar mode. In the Atlantic, tropical with the NOAA research objectives on interdecadal SST distributions have been used to predict Brazilian time scales are in various stages of planning and and African rainfall patterns-. At higher latitudes, development. Most are scheduled to get underway SST distributions have been correlated with Euro- during the decade of the 1990's, offering important pean rainfall. Much of the atmospheric variability opportunities for cooperative efforts. occurs in the form of the teleconnection patterns ob- served in pressure fields at various heights. Although the evidence is weaker, other studies indicate changes in the atmospheric circulation pat- terns and corresponding changes in regional weather The international WOCE program under the World patterns on decadal time scales. These changes are Climate Program (WCP), with the national WOCE known to be related to changes in the strength of the program led by the NSF, will be the main external global atmospheric circulation, which in turn is proba- programs of concern to the ABVP. The national pro- bly related to variability in oceanic thermal forcing of gram is mainly concerned with the description of the the atmosphere. For instance, winter season air tem- global ocean circulation and processes on the decadal/ perature, precipitation and synoptic weather system interdecadal scale, which are designated as part of frequency over the contiguous U.S., and drought in the Core 1 Project under the international scientific the Great Plains have been shown to vary on decadal plan for WOCE. The Core 1 large scale observational and longer time scales in phase with large scale atmo- network and hydrographic program will involve the spheric circulation patterns over the U.S. These pat- Atlantic (as well as other oceans) and should provide terns, in turn, depend on Pacific SST patterns. opportunities for cooperative efforts with NOAA. The number, severity and genesis of extratropical However, more important opportunities for the ABVP storms along the Atlantic coast of the U.S. has in- effort may be with the Core 2 Project (The Southern creased from the 1920's to the 1960's, with post-1960's Ocean Project) and with Core 3 (The Gyre Dynamics distributions sirrdlar to pre-1920's distributions. Simi- Experiment). lar long-term variability has also been observed in The international WOCE Core 2 Project concerns other regional climate features such as the Sahel correspond very closely with the ABVP concerns. drought. A concentrated effort is now warranted to The Core 2 Project (Gordon et al., 1987) has desig- increase our understanding of the role of large scale nated a "spoked wheel" plan composed of nineteen atmosphere-ocean interactions in regional climate deep vertical sections between Antarctica and the 30 variability, in order to develop regional prediction degree S latitude zonal line. Three of these lines are capability. designated by the Core 2 Project as "choke points" As already noted, the issue of interclecadal vari- with lines for repeated monitoring and deployment ability cannot be completely disassociated from the of bottom pressure sensors and long-term coherent interannual oscillations, particularly in the mid- current meter arrays. The Drake Passage, alone, would latitudes and higher zones in the Atlantic Basin, or have a permanent monitoring system consisting of 20 Ocean Circulation and Global Climate Change bottom pressure gauges, closely spaced moored ar- rays (intensive transport array) of current meters, temperature and conductivity sensors. Ship based CTD profiles to the bottom and XBT casts would be The Arctic Ocean Sciences Board WAS, 1977) has taken on all lines. Oceanic tracer samples are also proposed a Greenland Sea Project (GSP) which has a proposed in these areas. number of international participants, including cer- The three "choke point" lines are clear candidates tain science funding or policy levels of governments, for a possible cooperative effort by NOAA, as is the other than the United States. Given the similar inter- ocean tracer work. ests between the proposed GSP and the Atlantic vari- An E-M cable system is not mentioned as part of ability strategy outlined here, NOAA would be a the long-term Drake Passage monitoring system, so logical candidate to undertake the U.S. agency role. that STACS and the ABVP may be able to contribute The project's region of study is consistent with the such a system in a joint effort. It should also be noted identified area for NOAA's North Atlantic climate that W. Emery ran hydrographic surveys essentially concerns. along some of the same lines about a decade ago as The project is concerned with water mass produc- part of the FGGE Program, so that a baseline, in part, tion, sea ice variability and its relationship to climate, already exists. NOAA could support repeats of these and atmospheric exchanges driving the system. Geo- lines on a regular, long-term basis as part of the chemical tracer work, modeling, deployment of ABVP Program and contribution to WOCE. moored arrays of current meters, tracked surface The Gyre Dynamics Experiment part of the inter- drifters, pressure gauges are all proposed. However, national WOCE MOCE Core 3,1987) appears to be E-M cable efforts are not explicitly proposed. The focusing on the North Atlantic, perhaps with some passages between Iceland and Greenland and be- South Atlantic work and a possible deep circulation tween Spitsbergen and Greenland offer potential lo- experiment in the Brazil Basin. The concerns appear cations for such deployments, although the other to be highly compatible with the ABVP strategy. sides of Greenland and Spitzbergen facing Northern There will be a component of the Core 3 effort con- Europe pose more design questions and implemen- cerned with meridional circulation and deep convec- tation problems. Again, STACS technology, or ad- tion driven by diabatic processes as involved in both vanced developments of that technology (see Section deep water formation in the sub-polar areas and warm 5.0), may be applicable here, as would NOAA tran- 18 degree water formation in the sub-tropics. sient tracer work and related carbon dioxide flux and Another Core 3 concern involves the sub-polar ocean process research. frontal exchanges (at the boundary of the sub-tropical and sub-polar gyres). Tracer work is also proposed along with hydrographic surveys, velocity floats and moored systems. The NOAA concerns in ABVP would seem to stress gyre interactions in both the The NSF-led Global Ocean Flux Study (GOFS) equatorial/ tropical area and high latitudes in both project (Brewer, 1986) is concerned with a variety of southern and northern hemispheres. The surface surface processes and fluxes, including those involv- fronts at the boundaries of the subpolar gyres and ing carbon dioxide. The NOAA program must be subtropical gyres in both the Northern and Southern concerned with surface fluxes of fresh water, latent hemisphere are singled out for special attention. These and sensible heat, and carbon dioxide, so that mutu- interactions are associated in the South Atlantic with ally advantageous inter-relations between the Atlan- the important convergence zone near the Antarctic tic component of GOFS and the NOAA program Circumpolar Current. Hence, cooperative opportu- seem both feasible and desirable. In this respect, the nities with Core 3 efforts appear promising. Also, NOAA effort may be able to complement GOFS by STACS-type cable systems are not explicitly listed concentrating on the ocean margins and the shore- for Core 3, so that this could provide a unique NOAA ward edges of the ocean boundary currents (both contribution to any cooperative effort. Also, the area western and eastern). of WOCE concerns cuts off at the 60 degree N lati- tude line, whereas NOAA interests will go to still higher latitudes. NOAA's contribution to a coopera- tive effort could involve determination of the North- em boundary conditions for the WOCE efforts. Ocean System Studies 1 21 rate. It may be possible to calibrate fish scales and other paleo-biological evidence in the varves by parison with written historical fish population com records and the whole sequence of overlapping di- rect and indirect ocean records, beginning with There is evidence of a long-term trend in the time COADS (Sharp and De Vries, 1988). In principal at series of some climatic variables over the past cen- least, these can be extended back for several millen- tury or so. However, the earliest studies by Fletcher, nia. Such a paleo-biological component of a compre- Radok and Slutz (1979) indicated that climate vari- hensive NOAA Climate and Global Change Program ability on the decadal to centuries time scale is domi- needs to be developed (Figure 4). nated by only a few abrupt adjustments of the circu- The strategic approach involves the application of lation regime, rather than by gradual change. The these long time series with computer simulation evidence of the greatly expanded COADS data set models. If sudden changes are indeed characteristic now reinforces that conclusion. However, questions of climate change on interdecadal and longer time have lingered about possible systematic errors due scales, there will be a particular problem in simulat- to changes in measurement technology and prac- ing the actual transition processes. Typically, a non- tices, and the transition from sailing vessels to steam linear system undergoing such a transition, reorgan- powered ships. Every effort has been made over the izes itself so that boundary layer processes and other years to identify and remove sources of systematic sub-grid effects become critical, at least during the error, but it is difficult to prove that no systematic transition. Unfortunately, lack of resolution of the error sources remain when one cannot repeat the models leads to parameterizations that are usually observations under controlled conditions. based upon assumptions dependent upon the exist- Consequently, efforts are continuing to compare ing state of the system, which may be incompatible the COADS data with indirect historical and paleo- with transitional conditions and the future state. climatic evidence. These are mostly from land areas, An atmospheric GCM, or ocean circulation model, and range from the historical record of Nile River is simply a discretization of the known equations of flood levels at Cairo, lake sediments in Africa, and motion of the atmosphere (or ocean), with physical ice core data. The records generally show changes of processes treated explicitly when possible (radiation, an abrupt nature correlated with the transitions indi- orography, large scale precipitation, etc.), and param- cated in the COADS series. One implication of the eterized when necessary (cumulus convection, small observed sudden transitions is that we need much scale mixing, etc.). The mix between explicitly com- longer retrospective time-series for numerical model puted and parameterized processes changes as the development and verification purposes. Longer resolution increases. For example, if the resolution oceanic retrospective time-series are preferred over could be refined to one kilometer, it would no longer land series, as being more directly indicative of past be necessary to parameterize cumulus clouds be- ocean circulation patterns. cause they would be explicitly resolved. Since model Since we have mostly reached the historical limits resolution is purely a function of the capability of of scientific observations with COADS, indirect computer technology, we can expect models to im- sources will be required. Since these indirect sources prove as advances in computer technology allow are inherently very "noisy" data that are difficult to higher resolution. This progress will be slow, since interpret, redundancy of many data types will be every doubling of resolution involves 16 times more needed to produce statistically based reliability in computation. Doubling of the resolution in each of their interpretation. Ship logs may be one source, but the two horizontal spatial dimensions, multiplied by there is another. Man has been fishing on the ocean the mandatory reduction in the temporal step size, surface for as long as he has been sailing upon it, and and assuming that the number of vertical layers in fish catch and whaling records indicating changes in the model is also doubled, accounts for the factor of species populations and their patterns of behavior 16. exist going back centuries. These records overlap A national effort is being made to develop super- with the COADS data in some of the same areas since computers, and climate modeling is considered to be the nineteenth century. These can also be correlated the prime nondefense application of these machines. with studies of mud layers (varves) in areas where NOAA will need to update its computing facilities as sedimentation rates are laid down ten or more times these new generation machines become available. as rapidly as the world average ocean sedimentation This will require a more frequent updating of GFDL 22 Ocean Circulation and Global Climate Change computers from the present ten-year cycle to an ac- telephone cables might yield direct evidence of heat celerated five- year cycle or less as dictated by the flux variations if long period temperature and other actual advances in computer systems, if we are to possible noise of the power stations can be accurately achieve important modeling capabilities, such as monitored. simulation of transition regimes between climate The point electromagnetic measurements, devel- states. oped and successfully used in deep ocean magne- totelluric studies, may also prove to be valuable be- cause of the inherent spatial smoothing (vertically and horizontally) of the electromagnetic signals and MMINNEM the lower costs of deployment compared to installing cables. An array of bottom magnetic and electric. It is imperative that NOAA continue to develop, measurements should be tested as an alternative to on a more systematic basis, technology for monitor- cross stream cable voltages. Such arrays will be use- ing climatically important oceanographic processes. ful in regions such as the Arctic where it is not fea- Both shipboard and in situ instrumentation are re- sible to install long cables. quired. In particular, an acoustic Doppler velocity The vertical electric field is useful as a measure of profiler which can be deployed on ships of opportu- the magnetic east-west transport, but will not pro- nity is needed to provide global distributions of vide any horizontal spatial averaging since the verti- surface velocities, an important variable for climate cal electric currents are small due to the effects of the model validation. Similarly, accurate meteorological insulating atmosphere. Such measurements would packages, including sensors to determine surface be particularly useful in the equatorial regions where energy fluxes, are required for installation on ships- the flow is predominantly east-west. of-opportunity. Finally, accurately towed electric field measure- A "scientific cable" is necessary for use in loca- ments will provide the important regional survey of tions other than straits and passages where land-to- the ocean currents; namely, its extent and spatial land connections are possible. The cable should be variations. This information will be needed to design designed to provide data on the internal structure of the minimum cable length and the maximum spac- the ocean, rather than merely total transport, as does ing allowed between the sea floor point electro- the present cable system. A possible array would magnetic recorders. consist of suites of instruments connected by cable and having remotely programmable sensor packages located at selected positions on the seafloor. Other electromagnetic methods for measuring large-scale variations in ocean transport are promis- With the development of a predictive capability ing. The electromagnetic methods that are ideal can- for climate changes taking place over a range of time didates for measurements of ocean transport vari- scales as a goal, the concept of the climate system as a ations are: (1) voltage measurements using passive "heat engine" with two working fluids (the atmo- or active telephone submarine cables that span ocean sphere and the oceans) provides an effective meta- currents, (2) bottom horizontal electric field mea- phor for formulating the critical questions that must surements, (3) surface and bottom vertical electric be answered. The questions can be structured into a field measurements, (4) bottom horizontal magnetic hierarchy going from the most general to the very field measurements, and (5) accurately towed elec- specific. At the top of the hierarchy, the most general tric field measurements. question concerns the temporal and spatial variabil- Cross stream cable voltage measurements yield ity of both the effective forcing fields and the vari- accurate and continuous real time measurements of ability of the response of the two working fluids on the transport variations in the Florida Current. The different characteristic time scales. development of inexpensive cable laying techniques The research strategy is designed to answer the therefore should be started for continuously moni- important questions. At the most general level of toring the transport variations of other important approach, the strategy seeks to recover retrospective ocean currents. The use of existing submarine tele- information on the past behavior of the climate sys- phone cables should also continue to be explored as tem and to use this information, along with data these measurements will be the least expensive of from present monitoring systems, for calibration, any of the electromagnetic methods. Trans-oceanic initial conditions and boundary conditions for com- Ocean System Studies 1 23 puter simulation models of the climate system. The computer simulation models involve the coupling of the atmospheric and oceanic general circulations and the interactions with the land areas of the earth and BrewerP. G., et al., 1986, "The Global Ocean Flux Study cryosphere. The simulation models must be sequen- (GOFS): Status of the U.S. GOFS Program," EOS, 67, tially refined in an iterative process from the infor- 827-837. mation gained through focused process research, di- Broecker, W.S., 1987 a, "Unpleasant Surprises In The agnostic studies and vigorous comparisons of simu- Greenhouse," Nature, 328, pp. 123-126. lation results with known past climati(f variability patterns and current data from monitoring systems. Broecker, W. S., 1987 b, "The Biggest Chill," Natural His- Predictions emerge from this process by numerically tory, 96 pp. 74-83. carrying forward in time the projections of the cali- brated and validated models. Bryan, K., Manabe, S., and Spelman, M. J., 1988, Inter- Knowledge of the characteristics of the climate hemispheric Asymmetry in the Transient Response of signal on different time scales and in the difficult A Coupled Ocean-Atmosphere Model to A Carbon Di- regions of the world oceans allow ordering of the oxide Forcing, NOAA/GFDL, Princeton University, research priorities and structuring of the research Princeton, New Jersey, in press. into programmatic components. A plan involving Cayan, D. R., 1987, "Low Frequency Variability of SST and the development of a sequence of programmatic steps Its Relationship to Atmospheric Forcing," Proceedings is summarized in Figure 4.0. While the program units of the Twelfth Annual Climate Diagnostics Workshop, shown are not intended to be definitive at this time, U. S. Department of Commerce, NOAA, NWS, pp 167- the questions that they are designed to answer are 178. critical and the programmatic structure supporting the research strategy that is formally adopted will D66s, B. R., et al., 1975, "The Physical Basis of Climate and need to provide those answers. Climate Modeling," GARP Publication Series No. 16, A specific research issue of importance that con- ICSU/WMO. cerns the role of the oceans in interannual climate variability is the effect on ENSO of water mass ex- EPOCS, 1982, EPOC's: Equatorial Pacific Ocean Climate changes between the Indian and Pacific Oceans. On Studies, U.S. Department of Commerce, NOAA, ERL, interdecadal and longer time scales, important re- Boulder, Colorado. search concerns include: (1) the role of North Atlan- Fletcher, J. 0., et al., 1979, An Ocean Research Plan, U.S. tic deep water formation, and exchanges with the Department of Commerce, NOAA, ERL, Boulder, Colo- northern branches of the subtropical gyre, in regulat- rado. ing the thermohaline circulation, and (2) the identifi- cation of feedback control mechanisms in the Atlan- Fletcher, J. 0., Radok, U., and Slutz, R., 1979, "Climate tic thermohaline circulation and their role in possible Signals of the Antarctic Ocean," presented to: Sympo- major, and relatively sudden, climate changes taking sium on Progress in Antarctic Meteorology, XXIIth place on the century time scale. On both interannual General Assembly of IUGG, Canberra, Australia. and interdecadal time scales, some of the important Folland, C. K., Parker, D. E., Ward, M. N., and A. W. research questions relate to: (1) understanding the Colman, 1986, "Sahel Rainfall, Northern Hemisphere dynamics of the ocean /atmosphere fluxes of heat, Circulation Anomalies and Worldwide Sea Tempera- moisture and momentum, particularly with respect ture Changes,"Proceedings of the Pontifical Academy to areas controlling the time- variability of water of Sciences Study Week, Vatican, 23-27, September 1986, mass formation, (2) the relationship of changes of also U. K. Meteorological Office Long-Range Forecast- heat content of the equatorial zones and changes in ing and Climate Research Series, Report No. LRFC 7A, the position of the ITCZ and their relationship to Sept. 1986, Amended July 1987. climate variability in higher latitudes, particularly in the Atlantic basin. Gordon, A. L., et al., 1987, The Southern Ocean, WOCE Other questions, some more specific and some Core Project 2 Planning Meeting, Bremerhaven 20-23, more general, have also been posed. Still, other ques- May 1986, WCP/ICSU/SCOR. tions will emerge from the learning process of the GSP, 1987, Greenland Sea Project: An International Plan of research itself. the Arctic Ocean Sciences Board, National Academy Press, Washington, D.C. 24 Ocean Circulation and Global Climate Change Lamb, H. H., 1982, Climate History and the Modern World, Parker, D. E., 1987, "The Sensitivity of Estimates of Global Methuen, London and New York 387 pp. and Hemispheric Marine Temperatures to Limitations in Geographical Coverage," U. K. Meteorological Of- Lamb, H. H., 1972, Climate: Present, Past and Future-Volume fice Long-Range Forecasting and Climate Research I Fundamentals and Climate Now, Methuen, London, 1972, Series, Report No. LRFC 12, April 1987. 613 pp. Sharp, G. and T. J. DeVries, 1988, Paleocology Workshop, Lamb, P.J., Peppler, R.A., and S. Hastenrath, 1986, Inter- NOAA, NSF and U. of S. Carolina, S. Carolina. annual Variability in the Tropical Atlantic," Nature. 322, pp. 238-240. STACS, 1982, STACS, Subtropical Atlantic Climate Stud- ies, Program Development Plan, U. S. Department of Lorenz, E. N., 1968, "Climatic Determinism," Meteorologi- Commerce, NOAA, ERL, Boulder, Colorado. cal Monographs, vol. 8, No. 30, pp 1-3. WCP-WOCE, 1987, "The Gyre Dynamics Experiment, Manabe, S. and Stouffer, R. J., 1988, Two Stable Equilibria WOCE Core Project 3 Planning Meeting, London, 2-5 Of A Coupled Ocean-Atmosphere Model, NOAA/ September 1986, WCP/ICSU/SCOR. GFDL, Princeton University, Princeton, New Jersey, in press. Chapter II RIDGEFLUX Hydrothermal Venting on a Global Scale ive Eke-cut @_Su mima Y Problem and Opportunity ocean'-lbasins would allowNOAA to take a major step 4 -orw 4 1 d, in-, quantifying theglobal- significance of sub ar Ocean basins can, no lonzer be -considered to 'be 'Mart ne hydrothermal venting. @nz@'cMtr'a'l@o-'p'er'a-t'i'ng- yothesis that OAR tan te merely passive st by subaerial erosional processes. Indeed., ocean Y'basins, is the f -ollowing: oor ina worl wide '@drot er @ I venting plays a major role by virtue of their containin@g@lhe- U S-eafl h in con- spreading center system, are, now known to hqstqcfive, 'trolling chemical, budgets of the world ocean. processes which are having si@niftcard, *4- o-it"he" "Testing this@ hypothesis-requires work in four princi- US 4 -esi ch e@rts' 'will help determine, in I chemical, heat and mass@-biidgets ofied water 'J 't as-,, @pal-_areas.-Th rese'ar plate tectonics revolutionized earilisciencei in the-(66s,,, @,orde@ to@p 'reqiet,theeffectsof hydrothermal venting on dynamic processesnow, Iknown6to,@occur@,worldtaideate@t@;S-'Y,t,4g.@@c-hemistry@,ofthaocean@@nvironment. eading centers, in pdrticiilar@hydrofh&mal - .6 rc-@At seafloor spr SJU'l."esfength ofhydrothermal emissions inte venting processes-, are- becoming, an-anaI6�@usf, relationship Qcusfttr grated over, ridg es eg- m e n t s a nd, th ei r -to underlying geologic structure and processes marine sciencefor the@90s. Hydrother'mdl,.ven-t--@i-@fi','g.'@4"ffe@"is-"th-e-@ f@'c'hem-ical'' '-71 :1 rt of:,conservative and noncon- egion @, ra,,nsp compositions 4 the -ocean and underlying sedimettq serivalive'hydrotherinal emissions, including through long-term' iin"pu@_(of' -numerous, chemi6alf eie@-f horoiheirm- --W-e@ m-ission-16ss rates -from the water ments and dissalved-gasses incIudinksiIica;iroii;pkos-, 'c6lumn,'and,- - phorous; and CO,. The global'oceanic @budgets-vf silica' e-Temporal variability'ofhydrothermal emissions at 5 and phosphorous, for, example,are especially4mportant -time scales, of 10@' to >10 years. because they are micronutrients which play-major-roies@ These-processes will be studied at distances@ranging in the biogeochemical cycles of [email protected].,Hy4rotherm,,ql-,,,,, lroir,m t eve e- eis to ral hundred kilometers-from active e 0 a hydrothermal sources. Far-field (>100km) studies will venting is also directly responsibl for tIie:ev,iiutio4qf globally distributed major and heretofore- totally un-,,, focuson theregional extent, age,a,-nd evolution of chemi- known vent fauna -ecosystem. calanonialies. Mid-field (I km,to 100km) scale investi- gafio'ns-,,will,focus on-the axis of seafloor spreading d theirsegmentation. Near field (< I m to I centers an Research Strategy knO i@ vestigations, will concentrate on detailed geologic analyst's a nd mapping of targeted vent fluids, and tem- I demonstrai ion-bf`h An unequivoca t i, dche@mical studies of vent emissions. rydrptherlm in- @pora an fluence on the- re-g@ional@chemistr-y-vf,,majorzipertions 26 RIDGEFLUX-Hydrothermal Venting on a Global Scale ARr Why Now? J 1-ioli )br-'quantitatively understanding and predictin A g qtsof hydrothermal processes. Seafloor hydrothermal venting, which, was diSCOV7., ered in the late '70's,' i� now known', to be- a @gl6balll-,,P phenomenon. The full range 6i seafloor sp'r_eading_@cen--_1'A-P Why NOAA ? ter processes and the magnitude,of theireffects, through@ out the world ocean are, however, stilVin 'a stage of,--.- V,- resea q,, NOAA's ' rch mandate includes theresponsibil- discovery. GrowinR nuinberibf are n 'investika-ho"n's,' clearly ity to u der@tdnd and assess global chemical processes d that contin lly alter the composition of the oceans, sliqwin t t these vrocesses are, not isoldie" d' but'are g ha is Jif@_ distributed throughout' the world, bc-Mii andii" is@@ a @ @6'- 004- their ments, -and their The remoteness'of the clear thqt1hCse.prmesses have persisted fuqqqmei*l , .@! 7r spreqdin centers requires exten- deep pcqgii @qaflo, contributions to ocean chemical budgets for hundreds-, sive and very sophisticated research platformsand in- of, millions of years. Recent results 6, t, , live iANY strumentation. NOAA has, in addition to its research ugvs correlations between major plate motion changes '@n- 'A@ responsibilities, unique capabilities for undertaking the creases in sMftoor hydrofhermal acti ity,(with accom- -361: necessary field work to understand ridgecrest processes panying changes in chemical flux rates to the ocean),- including highly specialized research vessels, manned and long-terinv1imate change. NOAA, in collaboration and remote . submergence vehicles, high-resolution with other government and academic investigators, is 1`1M.,, bathymetric and photographic systems, and prototype studying regional chemical,ocean6graphie@eff,@ctls-iii the;%W vent chemical monitoring and -vent fluid. sampling sys- North Paciftc which may be the result of hydrothermdl, -%11@ tems C ivit'' "Id e e@flo6isp@,Eiading@'ent@r m act y taking p ac' along the s systems off theWest -coast of,the'U.-S,: Most-recently, 0- 0- Be@i@ftis NOAA has discovered oiefitirelk newtype of-hydr thermalactivity., in, 1,98@6 'and again-Jii, 1987 NOA-A'131 researchers detected and studiedlarge episodic bur@ts of 7 Among the benefits of NOA hydrothermal research A hydrothermal activity occurring over- the- Northern 'AN,"14 are, (1) new numerical models for chemical fluxes in the 1 Pdcificseafloorspreiidin@c@nt&,which,itia-,gitigl@@@v@-nt "Co oceans, (2) acquisition of critical data for establishing models, cfor @ circulation-, of heavand chemicals -in the ntained- quantities, lof heat, and, inassrequivalent to-AM 46=, " -long continuous- hydroihermal iput-of entiie_iqll ocean, -and 0) development of deeps a logy which year ou , 1,1, e techno "AM ridge segments. These and other discoveries, compel-J '.1', will make it feasible to addiess the global,scope of the its'un""ite-nationa y ro N044 -to' continue, and expand, id h d thermal impact. found, more on less commonly, along the entire 65,000 km length of the worldwide seafloor spread- ing center system. Subducting margins may also play an important The discovery of warm and hot springs at the role in affecting the regional, if not global, oceano- Galapagos (Corliss et al.,1979) and East Pacific Rise graphic budgets of dissolved elemental species and (Rise Project Group, 1980) spreading centers con- gasses, notably methane. Recently Kulm et al., (1986) firmed the existence of mineral rich hydrothermal have discovered extensive areas where sedimentary fluids emanating from seafloor vents. Associated with prisms along the subducting margin of the U.S. west the vents were massive polymetallic sulfides and a coast are being compressed and faulted with an ac- diverse and unique animal community with an ecol- companying release of large quantities of methane, ogy completely dependent upon chemoautotrophic iron, manganese, ammonia, barium, and other nutri- bacteria. In succeeding years, a growing number of ents. In comparison with hydrothermal vents, these investigations at other spreading centers in the Pa- "cold" vents, too, are replete with abundant, but cific and, most recently, the Atlantic Ocean, have biologically distinct, exotic fauna. Subduction zones documented the existence of many other venting are nearly equivalent in length to spreading centers. systems. It now appears that hot springs may be In terms of the global extent of such environments, Ocean System Studies 1 27 the mass flux from these regions may be shown to Both estimates indicate that hydrothermal circula- rival the seafloor spreading centers and other vol- tion plays a major role in the distribution and con- canic regions within the ocean basins. centration of many elements within the global ocean. Since these estimates were made, active venting has been observed and some measurements made in the Atlantic Ocean along the mid-Atlantic Ridge (Rona, et al.,1986; Ne1sen et al., 1986/87). As a consequence, those earlier calculations, which were only able to Prior to the discovery of seafloor springs, the ocean assume other sources of venting, now have added basins were considered passive sinks for materials credence. The recent discovery of episodic releases of weathered from continents and transported into the massive amounts of heat and mass at the seafloor of oceans by rivers and wind. Discovery of the mid- the Juan de Fuca Ridge (Baker and Massoth, 1986a) depth 5('He) anomaly in the Pacific (Clarke et al., suggests that short duration episodic events may be 1969) was an early indication that the ocean basins at least as important to the oceanic heat and mass themselves were, at least for some conservative ele- budgets as the steady state inputs from the hot mental species, active sources of a magnitude that springs. significantly affected the global ocean environment. Hydrothermal circulation also has a global effect Subsequent sampling of vent fluids (Edmond et al., on the heat budget of the deep ocean. It is now 1979 a,b; Edmond, 1981) and observations of basin estimated that hot springs and warm water vents scale plumes of 'He originating on the East Pacific account for 30 percent of the cooling of newly formed Rise (Lupton and Craig,1981; Riser, 1985) have con- oceanic crust and 20 percent of the earth's total heat firmed the global impact of hydrothermal emissions loss (Sclater et al., 1980). The local and regional effects (Figure 1). [Note: The 'He /4 He isotope anomalies in of the input of volcanic heat will vary according to natural waters are defined as the percentage "delta" the rates at which this heat is supplied to the oceans. values relative to the atmospheric 'He /4 He ratio, Stommel (1982) has proposed that the heat flux over 8(3 He) = (R/R atm-')*1001 where R=3 He/4 He. A 8(3 He) value of approximately 70 indicates an essentially pure source of 3He from the earth's mantle.] Figure I Contours of 6('He) in section view over the East Pacific Most hydrothermal constituents, however, are not Rise at 15'S (after Lupton and Craig, 1981). [Note: The lkfelkfe conservative in seawater. Their contributions to the isotope anomalies in natural waters are defined as the percentage oceans are detern-tined by a process of normalization "delta" values relative to the atmospheric 'HePHe ratio, 8ffie) = (R1R_-1)*100, where R = 3HePHe. A 8(111e) value of approximately with the assumed conservative hydrothermal sup- 70 or higher indicates an essentially pure source of 'Hefrom the plies of 'He and heat. For example, Edmond et al. earth's mantle.] (1979a,b) and Edmond (1981) used hot spring data from the Galapagos spreading center and 21'N on 8(1 He) (%) the East Pacific Rise to estimate global fluxes of both 7 6 5 4 3 2 major and minor elements from submarine hot 0 1 1 1 1 1 springs. By assuming that the majority of chemical 1 5 fluxes were from 350'C hot springs and that the 10 8(3He) versus heat and other elemental heat ratios 1- 7---+-15 1 --------- r-20 were uniform between vent systems, they estimated I I 1 25 that Li, Rb, Ca, Si, and Ba are released in amounts I I 1 13,0 comparable to or greater than the terrestrial input to _ff 2- T Z@51 1 @ 11 @4J the oceans. In addition, the Mn input from hydro- A thermal sources is sufficient to account for its entire authigenic inventory in deep sea sediments. On the 0 3 other hand, Mg and SO 4are removed as percolating 20 1 seawater interacts with basalt, so spreading centers 25 are the major sinks for these species. The more so- 4 phisticated approach of Thompson (1983), which considers low temperature as well as high tempera- 1000 0 10 0 km ture venting, has refined but not substantially altered 5 1 the original flux estimates of Edmond et al. (1979a,b). 130 120 110 100 90 WEST LONGITUDE 28 RIDGEFLUX-Hydrothermal Venting on a Global Scale a basin long active ridge is sufficient to alter mid- Richardson, 1985). The probable causes for these water circulation patterns. Joyce et al. (1986) have changes are that, as a result of tectonic rearrange- described a regional heat anomaly in the bottom ments of seafloor spreading centers during the Eo- water of the northeast Pacific Ocean that can be used cene (and other periods) the input of hydrothermal as a unique tracer of the relative age of the bottom CO 2and other constituents increased to a level where water. oceanic concentrations were several times higher than the present. The concomitant release of C02 to the atmosphere was of sufficient magnitude to effect some of the global warming trends that were known to have occurred during those periods. The implication for the present day oceans and atmosphere is that Several geological factors appear to play a major hydrothermal effluents contribute to the overall bal- role in determining the location and vigor of along- ance of the carbon cycle in the oceans which, in turn, axis hydrothermal venting. These include the volcano- ultimately controls natural variations of C02 in the tectonic stage of evolution of the rift axis, the rate atmosphere. (continuous or episodic) at which volcanism and/or The association of spatial and temporal hydro- rifting occurs at the spreading axis, the proximity to thermal variability is apparent at several scales. Long subcrustal melting anomalies, the extent and depth cores from the DSDP Program have been used to of faulting, and the thickness of sediment cover. On a quantify low-frequency (>106 years) changes in hydro local scale, fracture pattern, bathymetric relief, and thermal activity caused by large-scale tectonic reor- surface lava type may also influence the location and ganizations in ridge structure (Owen and Rea,1985). extent of venting. The relative importance of low- Standard sediment cores have the unique potential temperature or diffuse venting, either within or out- for resolving hydrothermal flux variability on the side the rift axis, is presently unknown although the order of 101 years within the last one million years. heat and mass flux from such sources may be large Leinen (1984), for example, has found factor of three (Morton and Sleep,1985). changes in the accumulation rate of hydrothermal Continued observations along spreading centers, emissions from the Juan de Fuca Rid e in sediments 9 however, have demonstrated that venting has a dif- of Brunhes age (<750,000 years). Detailed studies of ferent spatial and temporal variability than origi- hydrothermal emissions in sediments cored from the nally anticipated. While the association,of major vent ridge systems may thus allow the determination of fields and long wavelength along-axis bathymetric the history of hydrothermal processes on regional highs remains valid, detailed surveying of individ- scales and their impact on paleo-environments within ual ridge segments (Baker and Massoth, 1986a; their respective ocean basins. On an even finer scale, Macdonald et al., 1986) has revealed more extensive the recent observation of an apparently brief but along-axis hydrothermal activity, implying that some intense release of a large quantity of hydrothermal ridges or ridge segments may be considered line effluent from the southern Juan de Fuca Ridge-equal sources of hydrothermal emissions to the surround- to about four years' production of a large vent field ing ocean. on the same ridge segment-indicates that high- , For chemical constituents of hydrothermal fluids frequency episodic processes may contribute sub- that behave nonconservatively and rapidly precipi- stantially to the hydrothermal flux of material into tate in seawater, such as Mn and Fe, the impacts are the water column (Baker and Massoth, 1986a,b). These more localized and are preserved as concentration observations demonstrate that determination of the anomalies in the sediments (Lyle, 1976; Lyle et al., ridge integrated hydrothermal source strength will 1986; Metz and Trefrey, 1985). In such cases, the ultimately require determination of the spatial and sedimentary record can provide a temporal history temporal variability of venting over a broad range of of hydrothermal venting. Recent studies of Deep Sea scales. Drilling Project cores indicate that substantial in- creases in the flux of hydrothermal materials to the oceans occurred during the Cenozoic (Lyle et al., 1986; Rea and Leinen, 1986). Some of these changes in the rate of hydrothermal input may have resulted Some data sets suggest an intriguing connection insignificant changes in the earth's carbon budget among basin wide anomalies in the distribution of and climate (Owen and Rea, 1985; Kasting and heat, mass, and hydrothermal activity along a ridge Ocean System Studies 1 29 system. Reid (1982) attributes the tongue of rela- tremely important because they can be used to deter- tively warm water extending 6000 km westward from mine the integrated effects of hydrothermal venting the East Pacific Rise between 5'S and 20'S (Fig. 2) to on time scales of decades to hundreds of years. hydrothermally emitted heat. Riser (1985) noted that To deepen our understanding of the chen-dcal this plume also shows elevated 8('He) concentra- composition and evolution of a regional hydrother- tions and determined that the plume resulted from mal plume requires a determination of the plume's advection. by large-scale abyssal flow and eddy dif- fusion processes. In the North Pacific, a basin-wide transect at 47N reveals aubiquitous silicate maxi- Figure 2 Potential temperature K) along the a3 = 41.50 isopyc"al in mum centered at 2000 m, the depth of the Juan de the depth range between 2500-3500 meters (after Reid,'1982). Fuca Ridge crest (L. Talley, pers. comm., Fig. 3). The 160 100-fold elevation of silicate in hydrothermal fluids _7 from the Juan de Fuca Ridge (Von Damm. and Bis- hoff, in press) suggests that this maximum may at -10 least represent a residue of long-term hydrothermal - 1.3 venting into the mid-depth waters of the northeast Pacific. [Note: The global mass balance of Si in the oceans is particularly important because it is a micro- 0 nutrient and, therefore, plays a major role inthe bio- 7:@ 1.45 geochemical cycles of the sea.] Unequivocal estab- lishment of a hydrothermal origin of the silicate 1.4 maximum will require corroborating evidence from other conservative and nonconservative hydrother- J. mal tracers. NOAA has already directed preliminary 110 efforts At mapping large-scale features of hydrother- mal emissions. Baker et al. 0 985) identified a concen- 60 tration maximum in particulate hydrothermal Fe ex- tending westward from the southern Juan de Fuca 160 Ridge for at least 100 krn at ridge crest depths. Chemi- cal and heat anomalies in the water column are ex- Figure 3 Distribution of dissolved silicate (S'03 ymollf) from an August, 1985, cruise along 47'N latitude. Station density shown by tick marks along the top (after Talley, Joyce, and Swift, in prep.). 471N TRANSECT AUGUST 1985 -60 120 1000 140 - 182.5 160 170 2000 170 170 801 1080 175 in 1, 175 180 800 7.5 80 UJ 175 X= 3000 @07 177'5 W @160 1 .5 U) W cc 170 a. 4000 160 5000 Pill Juan de Fuca Ridge LONGITUDE 160*E 167*56'E 1751E 178120'W 170130'W 162*40'W 155055'W 148*03'W 140110'W 1320501W 124159'W STATION 038 045 058 063 070 077 083 090 097 104 115 30 RIDGEFLUX-Hydrothermal Venting on a Global Scale origin at individual vent fields on various segments Juan de Fuca seafloor spreading system and it effects of the ridge crest. Sampling the integrated plume on the chemistry of the Northeast Pacific Ocean. The above each of three vent fields on different segments VENTS Program involves the collaborative efforts of of the Juan de Fuca Ridge has shown each to be other governmental agencies and universities bor- chemically distinct (Massoth et al., 1985). Chemical dering the northeast Pacific and has discovered diversity of lavas within a segment, interpreted as an numerous active vent sites along major segments of indication of small-scale mantle heterogeneity, has the Juan cle Fuca Ridge as well as the first major site been revealed by detailed dredging of the East Pa- of venting in the Atlantic. cific Rise (Langmuir et al., 1986). Intra-segment chemi- In order to better understand the linked processes cal diversity of plumes, arising from two distinct of hydrothermal venting, chemical composition of venting processes, i.e., "normal" chimneys in vent oceans and atmosphere, and climate change, NOAA fields with a life span of tens or hundreds of years, proposes an expanded program of observations and and episodic venting with a fife span of days or measurements along several ridge crests. The weeks, has been recently observed on the Juan de RIDGEFLUX Program will emphasie the quantitative Fuca Ridge (Baker and Massoth, 1986a). determination of global oceanic chemical effects of venting Changes of similar magnitude may also occur over occurring over time scales extending from nearly instan- shorter time periods, and the techniques to quantify taneous to geological. Over short time intervals, higher frequency hydrothermal changes are in pre- processes will be monitored to discern the high- liminary stages of development. In situ vent moni- frequency content of the hydrothermal variation sig- tors will provide information on the temporal stabil- nal in order to be able to effectively design a long- ity of the concentration of key chemical species in the term sampling and measuring strategy. Over longer undiluted fluid of individual vents. These sensors time scales, the basin-wide dispersal of hydrother- will also replace anecdotal information on the birth mal emissions from major portions of the global ridge and death of individual vents with continuous and crest will be quantified. Over the longest timescales, quantitative measurements. Even more challenging research will determine the historical causesand ef- is the detection and study of brief and episodic, but fects of venting preserved in the stratigraphic record. potentially very large, events of volcanic and associ- Integrating results at those time scales (and the corre- ated hydrothermal activity. sponding space scales) should provide a basis for predicting the effects of venting on global changes to the oceans and atmosphere. An initial step in quantifying the global impact of MEMO= hydrothen-nal venting on ocean chemistry will be to examine carefully select active venting locations Despite the growing awareness and measurements within the global ocean basin (Fig. 4). Building on the of hydrothermal activity, its quantitative influence present NOAA/PMEL VENTS Program which is on the character and composition of the global ocean seeking to undertake such a mission in the northeast and atmosphere is not yet known. With what is pres- Pacific on the Gorda/Juan de Fuca/Explorer ridge ently known, however, it is reasonable to hypothe- system, RIDGEFLUX investigations will expand this size that seafloor venting does have a significant quantitative approach to include other major venting effect upon the chemical composition of the global regimes such as those recently found along the Mid- oceans, and perhaps even the atmosphere. Atlantic ridge (Rona et al., 1986). The Indian Ocean Among NOAA's research missions is the respon- venting system is under investigation by French and sibility to continually assess the state of the ocean English scientists. An unequivocal demonstration of from several perspectives. In addition to research hydrothermal influence on the regional chemistry of focusing on physical and biological processes, major portions of ocean basins would allow NOAA NOAA's mission also mandates an evaluation of to make a major step forward in quantifying the global chen-dcal processes that continually alter the global significance of submarine hydrothermal vent- composition of the oceans, their sediments, and their ing. life. The central operating hypothesis of the NOAA's base hydrothermal research program RIDGEFLUX Program is thus: VENTS has been since 1984 principally directed to- Hydrothermal venting plays a major role in ward achieving a quantitative and predictive assess- controlling the chemical budgets of the world ment of hydrothermal venting occurring along the ocean. Ocean System Studies 1 31 Testing this hypothesis requires work in four prin- cipal areas: �Ridge-integrated source strength of hydrother- mal emissions and its relation to underlying geologic structure and processes, Much of the needed equipment for the field work � Regional transport of conservative and noncon- required for RIDGEFLUX are available from NCIAA servative hydrothermal emissions, facilities. These are, for example, Class I research � Hydrothermal emission loss rates from the water vessels, deep submergence vehicles (manned and column, and, unmanned), swath mapping systems (Sea Beam), � Temporal variability of venting at scales of 10-4 water sampling systems (SLEUTH), corers, grab to 10-1 years. samplers, deep-tow camera systems. Many other The goal of the NOAA RIDGEFLUX Program is to systems are available for lease. determine, in order to predict, the effects of hydro- Systems for making long-term in situ observations thermal venting on the chemistry of the ocean envi- and measurements of processes at, or near active ronment. Research will emphasize measuring, or hydrothermal vents are only in initial stages of de- otherwise detern-dning, the magnitude and composi- velopment. Under sponsorship of the University of tion of the present and historical vent flux integrated Washington, however, a prototype system of a long- over major ridge systems. Such a task requires a term ocean bottom observatory (LOBO) was designed, large-scale observational measurement and model- deployed, and recovered. Having succeeded with ing program, supported by specific process studies, the initial experiment, efforts are now underway to to quantify the rates of input and loss of hydrother- more completely determine what kinds of in situ mal constituents such as heat, trace metals, SiO 2' He, seafloor experiments will be required for the long and CO 2at specific ridge segments. term study of active ventfields. 'Piate tN@ undary, 0 600 Uticertain 'fv C R TiIJ,4 0%4 N L 7 Aleutian Trench -.7, 400 P 'P' LA 7 San Andreas PACIFIC Fault IL P L A T E P A C I F I C CAROLINE PLATE 00 COCOS PLATE k RI LATE law P L A T E Andesii., ntains 'I Trench 0"Q)T H P@L NAZCA IND PLATE A US T R A14 A'N nga Trench R I CA N P L A T E Mid A la 400 East Pacific e Rise P L A T E so e-e- @AN TA R C T I C A N T A R C T I C P L A T E 600 1 9 - PLATE @c'OTJAPLATE 1800 1200 600 00 600 1200 1800 Figure 4Global map of plate boundaries. Hot-water venting is concentrated in volcanic environments, principally along seafloor spreading centers (//). Other hot-water venting occurs in association with mid-plate, or hot-spot, volcanism and in volcanically heated backarc basins. Cold-water venting is associated with the subducting edges of plates (A), primarily adjacent to continental margins. 32 RIDGEFLUX-Hydrothermal Venting on a Global Scale The LOBO concept is important to RIDGEFLUX within advecting plumes. The development of such a goals and therefore is a cornerstone to experimental unique tracer will be of major importance for the .Plans. The NOAA/PMEL VENTS program is pres- establishment of dispersal rates for hydrothermal ently supporting development of a prototype chemi- emissions near active spreading centers. cal monitoring system which was successfully tested In order to address the global consequences of during the 1987 field program. ridge crest venting, research to establish relation- At present, the LOBO is viewed as basically a ships between the magnitude of hydrothermal emis- stationary system, although limited movement of sion signal and geological setting, as well as studies some sensors using a manned submersible is planned. of the regional transport, loss rates, and temporal A new concept being explored jointly with National variability of hydrothermal en-dssion products, will .Bureau of Standards (NBS) and OAR (NURP and be conducted at distances ranging from meters to SQ is the development of an unmanned ' untethered several hundred kilometers from active hydrother- ROV incorporating artificial intelligence. A number mal sources. The research tasks outlined below are of scenarios for such a device are being considered. referenced to three distance scales: "Near field" (<I m Coordinated budget initiatives are being planned. to 1 km); "Mid-Field" G km to 100 km); and "Far- Field" (> 100 km). In terms of the scale of features and processes being studies (e.g., vents, vent fields, ridge now segments, entire ridges, and, at the largest scale, ma- jor ocean basins), this categorization is useful but The NOAA/PMEL VENTS program, as outlined somewhat arbitrary since the effects of hydrothermal above, focuses on hydrothermal effects on the North venting are not rigidly bounded. Effects- of hydro- Pacific basin and will serve as a foundation for the thermal venting observed at these spatial scales per- RIDGEFLUX Program and its global objectives. The tain to the variability of hydrothermal venting which VENTS program has shown that initial observational occurs over periods of time ranging from minutes to results should be followed by process oriented ex- millions of years. Variations in hydrothermal activity periments designed to obtain quantitative limits and which occur over very short as well as very long time time series data on which predictive models can be scales will be studied in an effort to converge on developed. Consistent with this rationale, the princi- understanding hydrothermal processes which take pal thrust of RIDGEFLUX Program research is to place over the time interval of primary interest, i.e., quantitatively understand the effects of submarine interannual to,centuries, and therefore contribute to hydrothermal venting from ridge systems on global our ability to predict change in the global ocean. ocean chemistry. Program objectives will be aug- mented by collaborative and cooperative research with other elements of NOAA, other agencies, and academic institutions with complementary programs. Determination of the ridge-integrated hydrother- Regional chemical oceanographic anomalies, mal signals necessary for addressing the RIDGEFLUX which appear to originate from spreading center hypothesis, begins with a systematic inventory of the hydrothermal sources on the Gorda/Juan de Fuca/ chemical and heat anomalies of the various vent field Explorer ridge system, the East Pacific Rise, and, by plumes along the ridge segments chosen for study reasonable assumption, from the MAR and other and a characterization of the gross geologic features ridge systems as well, extend throughout large areas of those segments. Mapping of the plumes is particu- of surrounding ocean basins. The regional extent, larly useful because the anomalies can be used to age, and evolution of these anomalies will be deter- determine the integrated effect of hydrothermal vent- mined by means of: ing on the water column over time scales that expand e Acoustic and other mapping techniques applied as the spatial scale of the observations expand. A to ridge systems in ocean basins selected for promising new technique for determining the rate of study to determine anomalies indicative of vent- dispersal of hydrothermal plumes has been proposed ing, by Lupton and Craig (1981). They indicate that, by * Water column chemistry surveys designed to determining the scavenging rate of hydrothermal Mn reveal the regional extent of hydrothermal prod- in the water column, the variable Mn/3He relation- ucts and effects including heat, trace metals, S'021 ship may be used to provide an "internal clock" He, and CO 2' Ocean System Studies 1 33 Sediment coring to determine the type, quan- Determination of rates and mechanisms for pre- tity, and chronology of regional hydrothermal cipitation (including biologically-mediated contributions to the sedimentary record. mechanisms) and dissolution of plume constitu- N Determination of regional ocean currents to ents through field and laboratory studies, permit calculation of the heat and particulate e Sediment coring to determine the duration, type, fluxes within the hydrothermal plume, and amount of hydrothermal activity associated Synoptic CTD/transn-tissometer surveys to de- with specific vent fields and ridge segments, termine the geographic extent as well as the heat e Numerical modeling to synthesize field obser- and particle loading in the hydrothermal plume. vations into flux estimates. * Correlation of regional tectonic history of the spreading center segments with stratigraphically preserved hydrothermal chronology. The axis of seafloor spreading centers is divided into morphologically discrete segments which ap- Ram, parently result from large-scale processes controlling magma supply to the crust (Francheteau and Ballard, Results of near field investigations of short-term 1983). Ridge segmentation is especially well devel- hydrothermal events are required to establish the oped along the Mid-Atlantic Ridge and the Gorda/ appropriate time and space scales for longer term Juan de Fuca/Explorer ridge system. Adjacent seg- sampling and measuring strategies. In order to ments range in length from about 50 km to 100 krn. achieve the intended results from the activities listed These segments are bounded by structural disconti- below, experiments will be conducted on the seafloor nuities which often appear to mark boundaries be- within vent sites for extended periods of time. Dur- tween regions of contrasting volcanic, tectonic, and ing the next five years, the use of submersibles (and/ hydrothermal activity. Mid-field scale investigations or remotely operated vehicles) as well as surface that focus on regional contributions of hydrothermal ships will be required. Many mid-field-scale experi- emissions identified with discrete vent fields include: ments will extend to the near-field as well; other � Acoustic, magnetic, and electrical mapping; near-field experiments include: geodetic surveys, supplemented by photogra- * Detailed geological mapping of target vent fields, phy; thermal surveys (sediment heat flow and e Mineralogical, petrographic, and radiometric water column temperature); and sampling, to analyses of hydrothermal precipitates and host determine geological and geophysical factors rock to determine the age and chemical history which affect the location and size of active vents of venting on the scale of years to decades, as well as their flow rates and the duration of *Temporal studies of vent behavior including active venting, variations in temperature, flow rates, rates of 9 Remote acoustic surveys augmented by photo- accumulation and dissolution of hydrothermal graphic, transmissometer, and thermal surveys precipitates, effluent chemistry, and correlation (sediment heat flow and water column tempera- of these variations with geological controls, ture) to identify and map the regional extent of Collection of undiluted vent fluids from selected active hydrothermal activity, sites to determine undiluted, representative vent � Determination of regional ocean currents to fluid chemistries which can then be compared permit calculation of the flux of heat and par- between different vents, vent fields, segments, ticles carried within the hydrothermal plume. and ridges. � Synoptic CTD/transn-dssometer surveys to de- Resolution of seawater/rock interactions at a termine the geographic extent as well as the heat range of extant temperatures using field and and particle loading in the hydrothermal-plume, laboratory studies, � Discriminating the total hydrothermal source Sampling of dissolved and particulate constitu- between discrete high-temperature and low- ents within the buoyant plume at dilutions temperature sources, roughly between 1 and 1000 to establish the � Chemical sampling and laboratory modeling to chemical evolutionary links between the undi- determine the chemical evolution and reaction luted and diluted plume. rates of particulate and dissolved constituents Incorporation of the quantitative results from the of the plume, above work will constitute the basis for, and even- 34 RIDGEFLUX-Hydrothermal Venting on a Global Scale tual verification of, physical and chemical models for Canada Geological Survey and Canadian universities hydrothermal plume evolution. Physical models will focused on hydrothermal processes in the northeast include descriptions of the injection and dispersal of Pacific. dissolved and particulate components of the plume Under established guidelines, MMS has organ- based on geological setting, ocean currents, momen- ized a Gorda Ridge Task Force whose responsibility tum and buoyancy fluxes, and entrainment of the is to consider the possibilities for, and ramifications hydrothermal plume into surrounding water. Chemi- of, leasing portions of the Gorda for mineral explora- cal models will describe changing chemical compo- tion and -mining. NOAA has a formal relationship nents and reaction rates within the plume. These with the MMS Gorda Ridge Task Force and several results will provide a quantitative look at the mecha- NOAA scientists have and are actively participating nisms and effects of submarine hydrothermal vent- in collaborative Gorda Ridge research. ing on the chemical evolution of significant portions France continues to be very active both in the of major ocean basins and will provide the frame- Atlantic and the Pacific. France has an ongoing and work for evaluating cause and effect links between vigorous hydrothermal research program. Planning hydrothermally produced variations in regional ocean is now underway to have US/France ridgecrest col- chemistry and long term climate variations. laboration in the Atlantic through the US/France Bilateral Agreement. Activities of RIDGEFLUX will be coordinated and In the 1960's, plate tectonics became a unifying linked to those of several other bureaus in NOAA, theme and powerful hypothesis for testing the proc- other agencies here and abroad, as well as the uni- esses which govern the dynamic interaction of the versity community, and, where possible, private global system of crustal plates. By the 1990's, seafloor industry. processes will become an analogous unifying research Within NOAA, NOS will provide ship support a's theme which will serve to resolve longstanding ques- research platforms and for aerial surveys. NURP will tions concerning, for example, ocean chemistry bud- provide submersible and ROV support necessary to gets and long-term climate trends. Based on its ongo- accomplish program goals. NSGCP will support re- ing leadership research position in hydrothermal search which complements and expands upon, research, its unique facilities and organizational ele- RIDGEFLUX. For example, Sea Grant is currently ments and broad environmental mission, NOAA will supporting research in new instruments and con- play a key role in advancing ourunderstanding of cepts for long term and large scale measurements of venting and its connotations for predicting ocean processes at ridge crests; venting associated with hot climate. spots; and cold temperature venting associated with subduction zones and the US/France Bilateral: pre- liminary discussions are underway for possible joint missions on the MAR, a probable field site for ele- ments of RIDGEFLUX Baker, E.T., J.W. Lavelle, and G.J. Massoth (1985). Hydro- Interagency coordination: NSF continues to fund thermal particle plumes over the southern Juan cle Fuca academic research proposals focused on ridgecrest ridge, Nature 316, 342-344. and hydrothermal processes. Recently the NAS/NRC Baker, E.T., and G.J. Massoth, (1986a). The along-strike sponsored a national workshop to develop an NSF distribution of hydrothermal activity on a spreading program. RIDGE, which if supported, will result in a segment of the Juan de Fuca Ridge, EOS 67,1027. long term academic ridgecrest research program. RIDGE has been specifically planned to integrate Baker, E.T.,.and G.J. Massoth (1986b). Hydrothermal plume NOAA research which has the unique role in long measurements: A regional perspective, Science 23, term determination of oceanographic hydrothermal 980-982. effects. Numerous academic researchers are involved in ongoing NOAA hydrothermal investigations. Clarke, W.B., M.A. Beg, and H. Craig, (1969). Excess helium- There also is a strong collaborative relationship be- 3 in the sea: evidence for terrestrial primordial helium, tween NOAA, the USGS and researchers from the Earth Planet, Sci. Lett. 6, 213-220. Ocean System Studies 1 35 Corliss, J.B., J. Dymond, L.I. Gordon, J.M. Edmond, R.P. Macdonald, K.C., R.M. Haymon, and L.J. Perram (1986). von Herzen, R.D. Ballard, K. Green, D. Williams, A, Thirteen new hydrothermal vent sites found on the Bainbridge, K. Crane, and T.H. van Andel, (1979). Sub- East Pacific Rise, 20-21'S, EOS 67, 1232. marine thermal springs on the Galapagos Ridge, Science 203, 1073-1083. Massoth, G.J., E.T. Baker, J.P. Cowen, K.K. Roe, and P.Y. Appriou, (1985). The geochernical diversity of Juan de Edmond, J.M., (1981). Hydrothermal activity at mid-ocean Fuca Ridge hydrothermal plumes, EOS 66,929. Ridge axes, Nature 290, 87-88. Metz, S., and J.H. Trefry (1985). Geochernistry of metal- Edmond, J.M., C. Measures, R.E. McDuff, L.H. Chan, R. liferous sediment from the TAB hydrothermal field, Collier, B. Grant, L.I. Gordon and J.B. Corliss (1979a). Mid-Atlantic Ridge, EOS, Trans. AG U, v. 66., 936 pp. Ridge crest hydrothermal activity and the balances of the major elements in the ocean: The Galapagos data, Morton, J.L. and N.H. Sleep, (1985). A mid-ocean ridge Earth Planet. Sci. Lett. 46,148. thermal model: constraints on the volume of axial hydro thermal heat flux. J. Geophys. Res. 90, 11345-11353. Edmond, J.M., C. Measures, B. Mangum, B. Grant, F.R. Sclater, R. Collier, A. Hudson, L.I. Gordon and J.B. Nelsen, T.A., G.P. Klinkhammer, J.H. Trefry, and R.P. Corliss (1979b). On the formation of metal-rich deposits Trocine (1986/87). Real-time observation of dispersed at Ridge Crests, Earth Planet. Sci. Lett. 46, 19-30. hydrothermal plumes using nephelometry: examples from the Mid-Atlantic Ridge, Earth and Planet. Sci. Lett. Francheteau, J. and R.D. Ballard, (1983). The East Pacific 81, 245-752. Rise near 21'N, 13'N, and 20*S: inferences for along- strike variability of axial processes of the Mid-Ocean Ridge, Earth Planet. Sci. Lett. 64, 93-116. Nelson, T.A., G. Klinkhammer, and J. Trefry (1985). Real- time observation and tracking of hydrothermal plumes Joyce, T.E., Warren, B.A., and L.D. Talley (1986). The geo- on the Mid-Atlantic Ridge, EOS, Trans. AGU, v. 66, 936 thermal heating of the abyssal subarctic Pacific Ocean, PP. Deep Sea Research 33, 1003-1015. Owen, R.M. and D.K. Rea, (1985). Sea-floor hydrothermal Kasting, F., and S.M. Richardson (1985). Seafloor hydro- activity links climate to tectonics: The Eocene carbon thermal activity and spreading rates: the Eocene carbon dioxide greenhouse, Science 227,166-169. dioxide greenhouse revisited, Geochim. Cosmochim. Acta 49, 2541-2544. Rea, D.K. and Leinen, M. (1986). Neogene controls on hydro thermal activity and paleoceanography of the southeast Kulm, L.D., E. Suess, J.C. Moore, B. Carson, B.T. Lewis, Pacific Ocean, In: Leinen, M., and Rea, D.K. et al., Init. S.D. Ritger, D.C. Kadko, T.M. Thornburg, R.W. Embley, Rept. DSDP, Vol. 92: Washington, D.C. (U.S. Govern- W.D. Rugh, G.J. Massoth, M.G. Langseth, G.R. Cochrane, ment Printing Office), 597-617. and R.L. Scamman (1986). Oregon subduction zone: venting, fauna, and carbonates. Science 231, 561-566. Reid, J.L. (1982). Evidence of an effect of heat flux from the East Pacific Rise upon the characteristics of the mid- Langmuir, C.H., J.F. Bender, and R. Batiza (1986). Petro- depth waters, Geophys. Res. Lett. 9, 381-384. logical and tectonic segmentation of the East Pacific Rise, 5'30'S-14'30'N, Nature 322,422-429. Rise Project Group (1980). East Pacific Rise: hot springs and geophysical experiments, Science 207,1421-1444. Leinen, M., (1984). Rates of hydrothermal sedimentation ontheJuandeFuca Ridge, EOS 65,1112. Riser, S.C., (1985). Maintainence and driving of helium-3 and thermal plumes in the abyssal South Pacific, EOS Lupton, J.E. and H. Craig (1981). A major helium-3 source 66,930. at15'SontheEast Pacific Rise, Science 214,13-18. Rona, P.A., G. Klinkhammer, T.A. Nelsen, J.H. Trefry, and Lyle, M.W., Owen, R.W. and Leinen, M. (1986). History of H. Elderfield (1986). Black smokers, massive sulfides, hydrothermal sedimentation at the East Pacific Rise, and vent biota at the Mid-Atlantic Ridge, Nature 321, 33- 19'S, In: Leinen, M., Rea, D.K. e t al., Init. Repts. DSDP, 37. 92: Washington, D.C., (U.S. Government Printing Of- fice), 585-596. Sclater, J.G., C. Jaupart, and D. Galson, (1980). The heat flow through oceanic and continental crust and the heat Lyle, M. (1976). Estimation of hydrothermal manganese loss of the earth, Rev. Geophys. Space Phys. 18, 269-311. input to the oceans, Geology 4, 733-736. 36 RIDGEFLUX-Hydrothermal Venting on a Global Scale Stommel, H. (1982). Is the South Pacific helium-3 plume dynamically active? Earth and Planet. Sci. Lett. Gl, G3- G7. Thompson, G. (1983). Hydrothermal fluxes in the ocean, In: Chemical Oceanography, Vol. 8, J.P. Riley and G. Skir- row (eds) Academic Press, Condon., 272,337. Von Damm, K.L., and J.L. Bischoff (in press) Chemistry of hydrothermal solutions from the southern Juan de Fuca ridge, 1. Geophys. Res. Chapter III Fisheries Oceanography Executive Summary Problem and Opportunity is intiniatelij linked to the physical dynainics; of the oceans and large-scale climatic 'fluctuations. Fisheries Tire fisheries resources of tire Un ited States are, with scientists throughout tire world have identified recruit- a'feu, notable exceptions, products of coastal ocean waters inent fisheries oceanography as tire most' important and adjacent estuaries. Viese resources support the topic in fisheries science, and the key to accurate predic- Nation's nia jor cominercial and inarine recreational tions. fishing industries. However,fish populations and cont- A research prograin on fisheries recruitment Prot inunities fluctuate, sometimes substantiallY,from year- only has the potentialfior improving fisheries manage- to-year and to even greater extents on longer intervals inent responsibilities, but also for improving man's (interdecadally). This variability and lack of knowledge understanding of the linkages between the physical qj its causes reduce the @ffe'ctiveness of resource environment and tire ocean's productivity. management and cause economic dislocations so char- Recruitment ,fisheries oceanography needs to address acteristic to fisheries. the following list of questions: * Miat is the role of a biotic environmental vari .- ation in controlling survival of eggs, larvae, and Research Strategy juveniles? a Mat is the role of 'food availability in controlling A rnajor goal of fisheries science is the accurate survival of larvae and juveniles? prediction of tire status of species 'for both short-terin 9 What is tire role of invertebrate and vertebrate (inter and intra-annual) and long-terin (interdecadal) predation in controlling survival of eggs, larvae, ti?ne periods. Most of the existing forecasting inethods and juveniles? are known to be inadequate, particularly in regard to e Kqzat is the role of phYsical factors in transporting long tinie scales and radical population change. Short- eggs or larvae to appropriate juvenile habitats or terin predictions tire needed to better set harvesting in controlling predator-prey interactions? levels arid conditions (timing, place, etc.). Long-terin e How do the above factors interact to affect total predictions are essential in identifying the boorns, col- survival to recruitment? lapses, and changes in comintini(v structure that cause e Does the relative importance of various survival the niaJor dislocations in fisheries. Recruitment is the mechanisins change within or among years? process by which young fish (or shellfish) are added to 7here is general agreement aniong researchers that the adult or fished stock, and is, therefore, the essential the prograin should focus equally on short- and loizg- process on which tire continuity of a fishery depends. to-in time Scales, seek to deterinine factors proinoting Evidence to date suggests that recruitment variability survival rather than accounting various sources of 38 Fisheries Oceanography mortality, and f6cus on compqring recruitment proc- Benefits esses in a limited number of tatget coastal ocean envi,NklAt ronments (upwelling' nks V!@ As a, result of the activities in the research strategy, frpnts,@ cont4", Pl,,Aelfffia - '@%%A - _U - I and large lakes). _V it should be possible to identifyand efficiently monitor the environmental-factors that are the best predictors of 11 stock abundances. Improved understanding of the causes WhyNow? of population size and variability will enable the conse- quences of man's activities in coastal and marine envi- Since the initiation of the Exclusive Econamic-Zone-,'01' ronments to be evaluated. Such research should also and its concomitant fisheries management responsibili-' lead to the Prediction of major climate and oceano- ties, fisheries oceanography has become a high priority. graphic events since it has recently been demonstrated The first requisite of effective fisheries management, a'@A%,@,- that movement, distribution, and variation in recruit- way to accurately forecast stock size, is@ not available. `45i- - ment of marine organisms are early indicators of such Existing forecasting Methods have been criticized for, events. When we gain an understanding of the factors bein inaccurate, unrealistic, oversimplified- and that control the size offish and shellfish populations, it 9 i ased towards over-representing the fftts, Offishing,', 4-AN will be possible to determine how the -actions of man All, affect these populations. 4mll dynamics on both of these time scales and so assist management, the fishing industry, and the public. Recruitment is the process by which young fish A major goal of fisheries research is to provide a (or shellfish) are added to the adult or fished stock. scientific basis to develop the capability to manage Research in fisheries during the past 5 to15 years fisheries for maximum yield in inherently variable provides substantial evidence that major fluctuations physical, biological, and sociological environments. in populations of fish and shellfish are often caused Where possible, prediction of both interannual and by variable survival rates among early stages of these long-term variability in resource production is highly organisms. Survival prior to recruitment has g@ner- desirable. Even in cases when prediction is not at- ally been identified as the single most important tainable, fisheries scientists can assist management process influencing the dynamics of commercially by partitioning the variation in resource production important fish and shellfish populations (Steele et al., into understandable components (e.g., that variabil- 1980). Simply stated, recruitment variability, which ity caused by the physical dynamics of the ocean, is an outcome of variable larval and juvenile mortal- biotic interactions, and large-scale climatic fluctua- ity, is presently considered the central problem of tions). In order to satisfy short-term management fisheries science. demands, it is important to understand inter-annual Although elements of fisheries recruitment proc- variability. In some fisheries this may be accomplished esses have been studied for many years, only during by simply indexing pre-recruit year class abundances the past 5 years has the topic been identified within through direct stock assessment methodologies. But NOAA as a high priority area of research. In part, this technique is generally unreliable, still, most sig- this is related to NOAA's fisheries management re- nificant changes in fisheries production seem to oc- sponsibilities in the Exclusive Economic Zone (EEZ). cur on a much longer time scale (e.g., interdecadal). The first requisite of effective fisheries management These are the blooms, collapses, and major changes is the ability to effectively estimate or forecast stock in community structure and species composition that size. Many existing forecasting methods are inade- cause significant dislocations in fisheries (e.g., sardine- quate particularly in regard to long time scales and anchovy in eastern boundary current ecosystems, radical population change. A major criticism of most gadoid-flatfish-crustacean dynamics in continental current approaches is that they are inherently unre- shelf ecosystems). The goal of recruitment research is alistic, eliminating complexity by over-integrating to improve our understanding of fish population the effects of many perturbing factors and/or over- Ocean System Studies 1 39 emphasizing the effects offishing. For example, in most fishery yield models, recruitment is represented as a pure density-dependent deterministic function of the size of the spawning stock, and fishing itself is One of the best known hypotheses has been re- the major controlling factor instock dynamics. Fur- ferred to as the starvation hypothesis. This hypothe- ther, current fishery management theory is based on sis is based on the observation that the average den- assumptions of steady state or equilibrium processes. sity of food items for larval fish in the sea is low Empirical evidence simply does not bear this out compared to their requirements for growth and sur- (Francis et al., 1987). vival. Further, small marine fish larvae are extremely If our marine fisheries resources are to be wisely vulnerable to these low food levels; many small lar- managed, we need to establish an integrated na- vae have only a few days from first possible feeding tional program of fundamental recruitment research. to their point of no return (irreversible starvation). In addition to its practical management applications, When oceanic conditions are stable, aggregations of this research has intrinsic scientific merit. Recruit- food apparently supply sufficient energy for growth. ment research requires integrating the biotic system But turbulence due to storms and other physical vari- with the abiotic (ocean and atmospheric) environ- ability can disrupt these patches leading to reduced ment. The importance of these linkages differs among growth rates of larvae, and in the extreme case, star- species and ecosystems; however, particularly in the vation. First feeding larvae seem to be particularly context of global change it is essential to our under- vulnerable to starvation, but the entire larval stage standing of the ocean system and its dynamics that seems to depend upon ample food to maintain high these processes be studied. growth rates. A second hypothesis, the predation hypothesis, is based on the observation that eggs and larval fishes are consumed by a wide variety of invertebrate and vertebrate predators. Copepods, salps, jellyfish, chae- tognaths and a variety of other invertebrates have A number of hypotheses on the mechanisms that been shown to eat eggs and yolk sac larvae in the control recruitment of larval and juvenile fishes have laboratory. Fishes, including adult conspecifics, ac- been suggested. These mechanisms include both abi- tively consume larvae and juveniles in laboratory otic factors such as changes in temperature, salinity, experiments. Unfortunately, few of these experiments or local circulation, and biotic factors such as food have examined the importance of predation on lar- availability or the presence of predators. While each vae when alternate prey are available. Field evidence of these factors could act alone and affect larval and of predators occupying the same water masses as juvenile survival, most likely it is their interactions eggs and larvae is available as is some diet data to that control year class strength. In the past, much suggest that fishes, in particular, do consume eggs research has focused on elaborating evidence to sup- and larvae in the field. A major problem with the port one of these mechanisms (e.g., starvation, preda- field diet data are that eggs and larvae are extremely tion, or advection). If in fact they interact, these mecha- ephemeral in the stomachs of predators, so the ef- nisms cannot be considered as independent, alter- fects of predation are difficult to document from nate, or competing hypotheses. These factors are in- field samples. tegrated in the growth dynamics of larvae and the A third hypothesis states that advection processes size or growth rate-dependence of predation and and other factors controlling larval transport can have other direct sources of mortality. We need to address marked effects on larval survival by transporting more clearly how these potential mechanisms inter- larval towards or away from appropriate habitats. act and how we can evaluate these interactions. Re- When environmental conditions favor transport to cent studies at the Northeast Fisheries Center have appropriate "nursery areas," growth and survival been directed toward determining the effects of in- may well be high. Physical oceanographic processes teractions of temperature, prey abundance, and related to larval transport are potentially very impor- predation on haddock re -cruitment (Laurence 1985, tant to recruitment success. Lough 1985, and Cohen et al., 1985). A fourth hypothesis, developed at the Northeast Fisheries Center, argues that recruitment is deter- mined largely by predation on post-larval fish by other fish (Cohen and Grosslein 1982). 40 Fisheries Oceanography on the causes of high mortality rates among young stages of fish and shellfish must continue, future analyses might more profitably focus on the unique One problem with traditional thinking is that these features of survivors rather than on estimating mor- mechanisms are not truly "alternatives." Both envi- tality rates or explaining how the larvae may have ronmental factors (e.g., temperature, salinity) and died. the availability of appropriate food affect larval fish Studies of recruitment in fishes must consider the bioenergetics and growth. In larval and juvenile fishes, effects of microscale and mesoscale environmental all surplus power (energy per time, Ware 1982) is variation on the abiotic and biotic factors which to- allocated to growth, so when environmental factors gether influence growth dynamics of the larvae. They or food vary, larval growth rates will vary. The star- must also obtain information on the density and size vation hypothesis only explains the extreme case structure of the predator fields to which the larvae when the energy balance remains negative long are exposed. Last they cannot neglect detailed physi- enough for the fish to die. cal oceanographic, meteorologic, and climatic infor- Predation on young fish is also influenced by both mation to address the issue of larval and juvenile nutrition and growth. Firstly, there is evidence to transport, and to explore the fundamental causes of show that starving larvae or larvae receiving lower long-term ecosystem change. rations swim more slowly and are more vulnerable to predators. Secondly, predation is often size de- pendent. If growth rates in larvae are reduced, they are exposed to gape-limited predators for a longer period of time (Werner and Gilliam 1984). One can also imagine situations where the converse is true depending on the predator field. The major questions of relevance to recruitment in Virtually no evidence is available to support the fish and shellfish address the effects of changes in hypotheses that 'advection per se is a major source of abiotic or biotic factors on variations in spawning mortality. Eggs and larvae may be advected to areas activity and on the survival of eggs, larvae, and juve- in which conditions are'less suitable for normal de- niles. How does environmental variation influence velopment or growth. The causes of mortality are spawning activity of pelagic marine fishes? What identical to those attributed to the starvation hy- factors control survival of eggs, larvae, and juvenile potheses. In short, factors that influence larval growth fishes? Much recruitment research has been based on dynamics cannot be considered to be independent, the notion that adult stock size has little influence because they are integrated in the growth bioener- upon recruitment except when stocks are extremely getics of the larvae. Because risk of predation de- low. Intraseasonal spawning activity in adults might pends on prey size, predation cannot be considered well be responsive to environmental variation; this as a separate factor. Larval transport cannot so read- issue cannot be neglected. ily be integrated into the growth dynamics frame- The following list of related questions concerns work. Larvae and juveniles which are transported to factors considered central to recruitment dynamics an inappropriate habitat may have less food or higher of fishes. The list is neither exhaustive nor mutually energetic costs or a higher risk of predation, but the exclusive. critical issue is that they failed to reach a critical 0 What is the role of abiotic environmental vari- habitat required to successfully complete their life ation in controlling survival of eggs, larvae, and history. juveniles? In order to progress more rapidly toward under- - What is the role of food availability in control- standing recruitment, mechanisms which were for- ling survival of larvae and juveniles? merly considered to be the bases of alternative hy- * What is the role of invertebrate and vertebrate potheses must be integrated. Fish which recruit to predation in controlling survival of eggs, larvae, the fishery are the survivors of various interacting and juveniles? mortality mechanisms. One can invoke a variety of * What is the role of physical factors in transport- mechanisms to explain why 99+ percent of the eggs ing eggs or larvae to appropriate juvenile habi- and larvae die or one can focus on the unique charac- tats or in controlling predator-prey interactions? teristics of survivors which allowed them to succeed - How do the above factors interact to affect total where so many others had failed. Although research survival to recruitment? Ocean System Studies 1 41 9 Does the relative importance of various survival 1-0 1 1 1 1 1 1 1 1 183. 1 mechanisms change within or among years? 82. Duration of a Long-Lived The most effective means to answer these ques- d A-Type Cold Water Mass 810 N tions is to emphasize the interactions among the vari- s 350 - ous potential recruitment mechanisms and to focus 80 0 on characteristics of the survivors to the age at which 300 - 0 70 @4 year-class strength is established. U) st 0 0 0 250 60 0 @O ca 200 0 2. 5 Cn 0 0 0 - 40 150 Z -1? 0 CD Recruitment dynamics can be examined at a vari- 0 0 00 0 - 0 30 ety of spatial and temporal scales. Appropriate scal- 2 100 0 .0 0 cz ing requires a clear understanding of relevant physi- w o: 0 0 20 P 09. 00 0 0 cal oceanographic processes which 1) create and LL 50 .0 0 0 4% 10 so *. 0 cob 8 maintain a favorable habitat to be occupied by eggs, 0 so 01@ , liiisiis'@11r I I !'P - larvae, and juvenile fish and shellfish during the L) 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 recruitment period; 2) influence egg or larval trans- Figure I Large-scale variations in catch of three species of sardine port to anappropriate nursery habitat; 3) serve as (Far Eastern, Californian and Chilean). migration routes for juveniles to appropriate nursery areas; or 4) attract or concentrate predators. This suggests that one appropriate time scale over which to explore recruitment mechanisms is short- term (e.g., intra-annual). Studies of intra-annual dynamics and larval survival mechanisms provide insights as to which portion of the spawned cohort accounts for the bulk of the survivors rather than simply estimating the time course of mortality. The suggested focus is on how and why survivors differ 0 from the average fish spawned and whether survi- vors possess any unique characteristics or co-occur Cat with critical oceanographic conditions which corre- late with their success. Characteristics of the survi- vors provide useful information to interpret how the Temp interacting constraints on successful recruitment behave and allow us to screen among the potentially important recruitment mechanisms that determine 20-29 30-39 40-49 50-59 60-69 70-79 80-83 which individuals survive from a spawned cohort. Climatic fluctuations and their subsequent oce- DECADE anic effects are becoming recognized as important Figure 2 Ten year mean Bristol Bay winter (Nov.-Mar.) air tempera- phenomena affecting major long-term shifts in ma- ture anomalies and Bristol Bay sockeye catch anomalies (standardized rine fisheries production. Figure 1 shows an 80-year (obs-meanlstd)). time series of variations in estimated abundance of three stocks of sardine, one each in the Northwest Pacific, Southeast Pacific, and Northeast Pacific. This figure suggests a relationship between climate-ocean CA oi vx. (over relatively long periods of time) process, and the status of fish populations that reside within the subtropical transition zones of the Pacific Ocean ba- sin. Figure 2 shows evidence of a direct correlation between interdecadal climate fluctuations and fish production-in this case, Bristol Bay (Alaska) sockeye 42 Fisheries Oceanography salmon. Similar relationships have been shown for lated. Concurrently, new methods of studying them other commercially important Bering Seastocks. have been perfected. In particular, we have become Debates have been raised'for years over the relative especially knowledgeable about the seasonality of impacts of exploitation and environment on major spawning, egg production, spawning frequency, bio- shifts in fisheries (e.g., California sardine, Bering Sea mass fluctuation, and early development of a num- king crab, North Sea gadoids). It seems clear that ber of important fishes (e.g., the clupeiform major fishery resource collapses often occur at times species-the anchovy, the sardine, the menhaden, the of intense fishing. However, causality does not nec- herring, and the gadoids). Studies of the early life essarily follow. history of fishes have advanced greatly in recent Therefore, major questions that need to be asked years, specific hypotheses on the respective roles of with respect to long-term (interdecadal) shifts in fish- fish eggs, larvae, and juveniles in variable recruit- ery production include: ment and population fluctuation can now be exam- �How are large-scale atmospheric and oceanic ined in the field and in thelaboratory. Advances have processes linked to major changes in fish com- similarly been made in biological and physical ocean- munity structure? ography and these can and should be applied to �Does the exploitation process itself affect major fisheries problems. Last, emphasis on recruitment changes in fish community structure and under research within NOAA is consistent with the na- what circumstances? tional and international focus on the biological impli- � Are community shifts primarily due to responses cations of global weather and climate change. of individual populations to environmental change or are more subtle food web interactions important? The answers to these long-term (interdecadal) questions most likely will require a more broad-based Various NOAA researchers have been integrally approach to examining marine ecosystem dynamics involved in developing the scientific background than the answers to the short-term (interannual) upon which much current and,proposed recruitment questions. Studies of long-term dynamics will focus research is based. Studies relevant to recruitment on such things as community structure, changes in dynamics have been conducted under the auspices spatial and temporal dynamics, food web structure of NMFS, OAR, (including both ERL, Sea Grant, and and how they are affected by both global weather/ NURP (National Undersea Research Program)) and circulation processes and resource harvesting strate- other NOAA agencies. These studies provide a dem- gies. Of utmost importance, however, is that these onstrated commitment to cooperation among fisher- studies be driven by specific hypotheses rather than ies biologists, oceanographers, meteorologists, and an intuitive longing for a full and complete ecosys- ecologists in NOAA to address the complexities of tem characterization. The latter approach is both in- recruitment dynamics. Approaches have included herently and practically infeasible except in very re- field surveys of eggs and larvae and larval food in stricted circumstances. concert with detailed physical oceanographic re- It is unclear how the processes and mechanisms search, laboratory experimentation to explore partic- critical to intra-annual recruitment dynamics are ular mechanisms and mathematical modeling. One coupled to those determining long-term shifts in fish of the major ongoing NOAA efforts in recruitment community composition. Understanding the recruit- fisheries oceanography is the Fisheries Oceanography ment mechanisms relevant to both of these time scales Coordinated Investigations (FOCI). This is a joint is, however, critical to successful fisheries manage- ERL/NMFS project to study larval recruitment into ment. the walleye pollock fishery of the western Gulf of Alaska. Critical aspects of the physical environment under investigation are the temporal and spatial behavior of wind-driven currents, the Alaska Coastal Current and slope /shelf exchange. Biological research Fishery scientists have never been in a better posi- focuses on the major sources of mortality of eggs, tion to study the factors affecting recruitment than larvae, and juveniles and how these are influenced they are today. In the last 25 years, an enormous fund by oceanographic conditions, egg quality, larval and of information on the biology of commercially and juvenile food resources and predator interactions. recreationally important fish species has accumu- The study employs advanced technology in conduct- Ocean System Studies 1 43 ing detailed biological and physical field sampling as the so called "estuarine dependent" species, the trans- well as historical satellite imagery to relate physical port of larvae and juvenile stages from offshore to environmental variation to recruitment success. "nursery areas" is critically important to recruitment An earlier ERL/NMFS cooperative project, con- of these species. Brewer et al., 1984, were among the ducted between 1978 and 1982 was an investigation first scientists to show in field studies that plankton of the relation between the physical features of the organisms are potentially formidable predators on Mississippi River Plume and the feeding growth and young fish. Recent studies on survival of juvenile survival of larvae of several economically important salmon (Pearcy, personal communication) provided species. This project focused on the biological impor- information suggesting that salmon recruitment was tance of the Mississippi River front rather than larval intimately related to predator concentrations at the transpo IA and currents. mouth of spawning rivers. Most Sea Grant investiga- The NMFS Georges Bank Ecosystem study con- tions have emphasized experimental biology with tains a recruitment research program on the limited emphasis on oceanographic conditions or groundfish community of the Bank. Findings to date long-term field efforts. indicate that most fish biomass produced is consumed by the fish themselves (i.e., predator-controlled com- munity). Also noted, is that fish biomass remains constant but species composition can change drasti- cally. Combined laboratory and field experiments Pioneering work by Methot (1983), an NMFS re- provided evidence which suggests that interannual searcher, used otolith analysis to examine the birth- recruitment variability in cod results from variable date frequency distribution of northern anchovy lar- predation rates on juveniles. The longer term species vae caught in the CalCOFI surveys to that of survi- shifts noted also appear to be related to predation in vors caught in the fishery at age 1+. He found that concert with undefined environmental conditions that harvests were drawn disproportionately from only a seem to confer a competitive advantage to one or portion of the spawning season. By comparing the more species. intra-annual patterns of recruitment success, one can A long-term joint state and NMFS research project identify periods of disproportionate mortality or on the California Current Upwelling System (Cal- exceptional survival. These periods can then be cor- COFI) has led to a proposed comparative project on related with short-term abiotic or biotic events, which the four major Eastern Boundary Current may influence survival, to more tightly link cause Systems-Peru, Canary, Benguela, and California-in and effect. This approach has also been successfully collaboration with the Intergovernmental Oceano- applied to examine the effects of abiotic and biotic graphic Commission (Sardine Anchovy Recruitment factors on growth and relative survival of larval and Program-SARP). This project seeks to develop im- juvenile American shad (Crecco and Savoy 1985) and portant linkages between the physical features of to relate survival and recruitment success of Lake these systems and recruitment dynamics. The project Michigan bloater to their growth dynamics (Rice et uses the most advanced techniques to examine the al., 1987). The power of this approach is that it fo- details of intra-annual recruitment mechanisms cuses on the unique characteristics of survivors and (Methot, 1983). One of the strengths of this and other suggests hypotheses regarding specific recruitment NOAA programs (e.g., the Georges Bank Ecosystem) mechanisms which can subsequently be tested either is the combination of field sampling, laboratory ex- experimentally or via designed observations. Over perimentation and mathematical modeling employed. the past decade, our capacity to physically sample To date, Sea Grant-sponsored academic research the ocean environment has been radically enhanced. has focused on various individual processes integral Fundamental understanding of upper ocean dynam- to recruitment success-trophodynamics of larvalfish ics requires current measurements. Acoustic cloppler and shellfish, transport of larvae to appropriate nur- based methods for doing this have been recently sery areas and predator-prey relationships among emphasized. Because these permit real-time charac- early life stages. Rice et al., 1987, was able to demon- terization of the entire upper water column, the data strate that growth rate, which integrates the physical are particularly suitable for interdisciplinary and nutritional environment, is strongly correlated biological /physical sampling studies. to survival among fishes. Research on the recruit- Conductivity/ temperature/ depth (CTD) technol- ment of blue crab (McConaugha 1983) and spot and ogy has also rapidly advanced and CTD's are now croaker (Petrafesa et al., 1986) have shown that among commonly incorporated into biological oceanographic 44 Fisheries Oceanography samplers. Instrumented buoys are becoming com- mon and can yield both hydrographic and meteoro- logical data over time/space scales impossible to achieve with shipboard sampling (e.g., see Weller Since the implementation of the Magnuson Fish- and Davis 1980). The combination of satellite al- eries Conservation and Management Act of 1977, the timetry, advanced high-resolution radiometer federal government has had a major role in the as- (AVHRR) and scatterometry imply that meteorologi- sessment and management of the fishery resources cal and physical coupling may be observed in near of the EEZ. NOAA, as the designated fisheries agency real time over global scales. New methods of biologi- and the nation's lead civilian ocean science agency, cal oceanographic sampling (reviewed in Ortner et not only is obligated but is uniquely qualified to al., 1981 and Dicky 1988) employing acoustics, bio- pursue a research program to increase understand- optics, satellites, telemetry and pattern recognition ing of recruitment. provide a basis for developing new and powerful NOAA's diverse capabilities in fisheries science tools for quantitatively sampling larval and juvenile and management, biological oceanography, physical marine fish andtheir food resources. Both in situ and oceanography, meteorology, and computer model- remote sensing have inherent advantages and disad- ing provide a pool of expertise capable of addressing vantages (e.g., see Esias 1981) but there is no doubt major recruitment problems in a comprehensive satellites are fast becoming indispensable tools in the manner. field of fisheries oceanography (Anon 1981). A criti- NOAA's organization infrastructure including cal attribute of the newest biological oceanographic mission-oriented components (NMFS), funda- samplers is that they make it possible to obtain bio- mental research components (ERL), academic logical data on time and space scales comparable to granting programs (Sea Grant), and support available physical data. Combining this with real- groups (NOS) make NOAA the logical agency time feedback will be essential to addressing physical- for implementing such a broad-based and com- biological coupling. Such feedback is also essential to prehensive program. progress in utilizing artificial intelligence techniques NOAA has a national network of marine fisher- to control sampling. These advances are required to ies and oceanographic research facilities that adequately sample comparatively rare organisms like includes-NOAA ERL and NMFS laboratories, juvenile fish. Sea Grant programs and several NOAA/ academic cooperative institutes. NOAA scientists have excellent relationships with scientists in academia and state agencies. These relationships promote collaborations A growing recognition of the urgency of describ- among scientists interested in recruitment in ing and understanding the central role of the ocean NOAA, the academic community, and the states. in global change and the multiple effects of the ocean The partnerships between NOAA and acade- upon our society have stimulated interagency strate- mia can be used efficiently to respond to com- gic planning. This is reflected in the recently@ released plementary needs for both long-term and short- U.S. Global Ocean Science Program (GOSP). One of term recruitment research. the principal complements, and one for which a ma- NOAA is sensitive to resource management jor role for NOAA is forecast, is Global Ecosystems needs. In partnership with the states, NOAA is and Productivity Processes. Even in the short-term, it responsible for managing the living marine re- is felt that NOAA can and ought to specifically direct sources of the EEZ. effort toward support of Fisheries Oceanography. Given NOAA's organizational character, this role is both appropriate and realistic. Failure to exploit this opportunity will result in a loss of research momen- tum, and could deprive society of information needed The NOAA recruitment research program will to wisely manage marine fisheries into the 21st cen- focus on the mechanisms that control the survival of tury. larval and juvenile fish and shellfish both on the short-term (intra-annual) and long-term (interdeca- dal). The ultimate goal is to develop an understand- ing of the recruitment process directed toward achiev- 11 CAT SID: 06090 OL Beginning of record displayed. OLUC ti "OCEAN SYSTEM STUDIES NOAA OAR RESEARCH STRAT... Record 1 of 1 HELD BY NO@ - 34 OTHER HOLDINGS OCLC: 19133167 Rec stat: c Entered: 19890210 Replaced: 19950507 Used: 19970103 Type: a ELvl: I Srce: d Audn: Ctrl: Lang: eng BLvl: m Form: Conf: 0 Biog: MRec: Ctry: mdu Cont: b GPub: f Fict: 0 Indx: 0 Desc: a Ills: ab Fest: 0 DtSt: m Dates: 1988,1989 1 040 OLA *c OLA *d OCL *d AAA *d AGL I 2 070 0 GC11.2.025 *b 1988 1 3 072 0 M400 I 4 090 GC11.2 *b .025 1988 1 5 090 *b T 6 049 NO@M 11 7 245 00 Ocean system studies : *b NOAA/OAR research strategy / *c produced by the University Corporation for Atmospheric Research. I 8 260 [Rockville, Md. : *b National Oceanic and Atmospheric Administration, office of oceans and Atmospheric Research, *c 1988-1989] 1 0 9 300 2 v. : *b col. ill.1 col. maps ; *c 28 cm. I 0 10 504 Includes bibliographical references. I @ 11 500 "August 1988"--Vol. 1 "November 1989"--Vol. 2. 11 ptov Ocean System Studies 1 45 ing the capability of predicting fish productivity shown to differ in some way (birth dates, growth (quality and quantity) in the context of abiotic and rates, appearance of low ration checks on the otolith biotic change. c.f. Rice et al., 1987) from the average fish which did N The research program will focus on comparing not survive. These differences reflect the interaction recruitment processes of a limited number of species of the various sources of mortalityand the patterns in a limited number of target marine ecosystems should suggest the factors which allowed particular (upwelling systems, riverine and pelagic fronts, shelf larvae to be successful (e.g., first fed when food was systems, and large lakes) which offers an alternative seasonally most abundant and/or available, grew to previous studies that have attempted comprehen- exceptionally rapidly in the first week of life, hatched sive examinations of single marine ecosystems. This during a period favorable for transport to an appro- comparative approach will lead to a more generic priate nursery habitat, hatched after important understanding of recruitment processes (Bakun 1985). predators on larvae moved to another habitat). Specifically, the approach: The characteristics associated with successful re- � Systematizes fragmentary information and in- cruits should allow the researchers to reject some sights in order to foster useful generalities con- recruitment mechanisms and to focus further research cerning physical, biological, and human impacts in that system. Tighter recruitment hypotheses can on fisheries production, then be proposed which can be tested in laboratory � overcomes problems associated with the com- or field mesocosm. scale experiments or via specific plexity and heterogeneity of marine ecosystems field observations. by studying only specific system attributes and parameters, a" � reduces the overall effort needed by focusing on similarities among ecosystems instead of the uniqueness of individual systems, As was mentioned earlier, the most significant � improves the chance of detection of specious changes in fisheries production seem to occur rela- relationships, tively infrequently, and are the changes that cause � increases confidence in weak relationships, and, major dislocations in fisheries. Because they occur on � is most informative in situations not amenable such a long time scale (e.g., interdecadal), they have to controlled experiments. proven to be most difficult to study (Figure 3). Figure 3 Annual catch (total and by species) for Georges Bank (after Hennemuth, 1979). The most promising of the approaches is to inves- GEORGES BANK tigate intra-annual recruitment mechanisms by documenting the unique characteristics of the survi- 106 vors relative to the average fish (Methot 1983, Crecco and Savoy 1985, Rice et al., 1987). This requires field Herrin Total sampling to collect the species of interest at different AOther Finfish Mackerel, stages of development throughout the spawning Silver Hake season. In order to track a cohort, detailed under- 105 V'V AA standing of fish movements relative to the physical 0 ly \1% Other circulation is absolutely required. Given this physi Flounder cal constraint, one can use otolith analysis to com- _J < pare the birth-date frequency distribution and daily D Z growth trajectory of the survivors at any point in the Z od < 104 YT FL life history to fish sampled earlier. Redfish'* Fled Hake The null hypothesis is that the survivors are drawn Poll@@k@i 'I at random from the spawning distribution. In prac- Haddock tice, if the null hypothesis cannot be rejected, then there is no point to look for particular mechanisms Squid contributing to exceptional mortality or survival. If 103 F the null hypothesis is rejected, the survivors are 1955 1960 1965 1970 1975 1980 46 Fisheries Oceanography Any serious study of long-term changes in re- of each fish species throughout its ontogeny. A sec- source production requires careful analyses of rec- ond way of examining food webs is to weight food ords of the past. These changes have been hypothe- web connections by rates of energy or material flows. sized to be strongly influenced by changes in oceanic This is the classical "ecosystem approach" and often variables which in turn are often significantly af- focuses more on materials or energy than the qualita- fected by fluctuations in the atmosphere (Wooster tive characteristics of food web components (such as 1983). Therefore, a combination of biological, oce- what species occupies a particular niche). In terms of anic, and climatic records of the past must be studied fisheries, which species is abundant as the major and linked. The work of Koslow et al. (1987) and piscivore or planktivore, for example, is often of great Hollowed et al. (1987) use fishery records of the past management importance. A third way of looking at to explore long-term resource dynamics. Boehlert and food webs is to examine food web linkages and look Yoklavich (1986) use otoliths of long-lived species to at the strength of species interactions or food web develop long-term growth records. A critical factor connections (Paine 1980). In this approach, the em- in any analyses attempting to link long-term atmo- phasis is on the "strong interactions" in the food web sphere and ocean environmental variability with fish- rather than on all possible interactions. The data from ery resource production is the linking of time and a number of food webs suggest that the number of space scales upon which climatic changes occur to strong interactors is much lower than the total num- those upon which fish population and community ber of interactions. If the strong interactors can be changes occur. The work of Smith (1978),and Bakun identified, then community changeovers become (1983) address this issue and suggest ways that these more predictable. In many aquatic systems, pis- problems might be scientifically addressed. These civorous fishes are relatively unimportant from the basic scientific questions are multifaceted: statistical, perspective of total food web connections or energy biological, and physical. If the recruitment problem flow, but in aquatic systems, they have been shown is to be significantly addressed, major breakthroughs to have a profound impact on food web structure in the study of long-term environmental-biological and function (Kerfoot and Sih 1987, Carpenter 1988). linkages must be made. Top piscivores are thus often strong interactors. To Long-term changes in the relative abundance of date, the only way to prove whether a particular stocks within a fish community could not only relate species is a strong or weak interactor is to conduct to long-term climatic variability but to species inter- controlled experiments. This is difficult to do in the actions. The key question is whether community pelagic, but environmental variation and exploita- changes are due to independent population responses tion are more-or-less constantly performing experi- to environmental change (autecology) or whether ments that fisheries biologists may be able to take some of the changes relate to species interactions advantage of if they pursue hypothesis-oriented field- (synecology). Improved understanding of the mecha- studies. nisms behind recruitment dynamics of particular If we are to understand the processes that control species are very useful if species interactions are un- long-term recruitment dynamics and community important. But, mechanisms that are important to change, we will need to understand both climate- single species on an intra-annual (short-term) basis population recruitment relationships and species may or may not be the same mechanisms responsible interactions. This is a formidable task that may re- to long-term changes in a population's abundance. quire different methods than those that have proven Even if species interactions are important, environ- useful in understanding intraseasonal dynamics. mental factors may contribute to the shifts. If envi- ronmental changes alter the abundance of particular species, the whole community may then shift due to interactions (e.g., Paine 1980) between this species and the others in the community. Continued progress toward understanding recruit- Recent progress in our understanding of food webs ment will then depend upon the development of suggests that there area number of ways to examine conceptual models of the recruitment process and any particular web and the interactions among its the explicit use of the comparative approach upon components. The first and oldest view is the connec- representative fisheries in diverse oceanographic tance web, which focuses on all the trophic connec- systems. Though systems differ dramatically in their tions among food web components. This way of view- basic oceanography and species composition, the ing webs requires a detailed understanding of diets same general set of processes control recruitment. By Ocean System Studies 1 47 comparing within and among oceanographic/ ences on recruitment dynamics in major U.S. fisher- fisheries systems and across species of larvae, we ies. Oceanic frontal zones (e.g., Gulf Stream) can in- may be able to derive some generalizations regard- fluence the spatial distribution of spawning, as well ing the interaction of mechanisms controlling the re- as larval stages, food resources and predators of pe- cruitment process. lagic fish species. In addition, these fronts may pro- vide the hydrodynamic mechanism that affects suc- cessful transport of pre-recruits to their nursery areas. Estuarine fronts (e.g., Mississippi and Columbia River plumes) have major impacts on both economically The four major Eastern Boundary Current sys- important species with a life history strategy wherein tems, the California, Peru, Canary, and Benguela spawning is offshore but nursery areas are in estuar- Currents are characterized by very similar environ- ies (e.g., menhaden, spot, redfish, summer and white mental changes, e.g., intense seasonal upwelling. flounder) and salmonid fish that spend critical juve- These regions contain very similar, recurrent groups nile life history stages near frontal boundaries. In of marine fishes and other organisms. This is particu- these cases, the links between physical and biological larly striking when one considers the species compo- processes are likely to be both especially significant sition of the fish fauna; each Current is dominated by and amenable to analysis. a sardine, an anchovy, a hake, a jack mackerel, a scombroid mackerel, and a squid. Many of the major fisheries in the world are located in these regions presumably because of enhanced availability of food to fish and shellfish larvae, juveniles, and adult fishes Continental shelf areas contain some of the most due to stimulation of primary and secondary pro- productive waters on the globe. Within these, some duction. of the world's largest and most valuable groundfish Upwelling is also a characteristic of water move- and invertebrate fisheries take place. Two distinct ment along the shore at headlands. The Kuroshio types of these systems are shallow shelves directly Current impinging on Japan is an excellent example associated with a continental land mass (e.g., E. Ber- of a current which produces this effect. Sardines, ing Sea shelf) and shallow banks totally surrounded anchovies, and other commercial fish and inverte- by ocean water masses (e.g., Georges Bank). Produc- brates common to upwelling systems also abound tion in the former tends to be associated with physi- there. cal conditions which allow for transport of eggs and larvae to appropriate habitats. In the latter, produc- . .. ......... ...... tion seems to be more dependent on stable local gyres which prevent dispersal of eggs and larvae and on enhanced lower trophic level productivity associ- Fronts, defined as zones of discontinuity that sepa- ated with enhanced nutrient supply. rate water masses, are receiving increasing attention from fishery biologists. Because these features pro- vide horizontal structure they affect the spatial dis- tribution of organisms of fishery importance as well as the distribution of food organisms and predators. Although small with respect to tonnage, many A growing body of information indicates that fishes, locally important fisheries are located in lakes. This their larval forms, and their zooplanktonic food and is particularly true in the less developed nations of predators may be aggregated along frontal zones. As Africa, South and Central American, and Asia. The a result of this aggregation, it is likely that the factors Great Lakes system can serve as a model for lake most influential to recruitment success operate along fisheries as well as for the more complex marine frontal zones and may affect the greatest proportion environment. The lakes are large enough that spatial of an annual year class. Aggregationis affected by sampling problems are identical to those encoun- frontal convergence, whereas frontal divergence could tered in the oceans, yet in fact they do have definite disperse organisms. Fronts are often dynamic, and boundaries and their present fisheries management their movement can displace or disrupt existing dis- practices reduce some of the uncertainties that com- tributions of fishes, fish larvae, and zooplankton. monly confound studies of open oceansysterns. Lake Both oceanic and estuarine fronts have major influ- Michigan fisheries, for example, are totally managed 48 Fisheries Oceanography at the top carnivore level, which gives scientists bet- special types of data and aid in the development and ter control of the system under study. evaluation of sampling devices. The participation of From an experimental and modeling viewpoint, the National Environmental Satellite, Data, and In- the Great Lakes may be particularly attractive for formation Service will be critical to track movements investigating the dynamics of fisheries because the of large water masses, oceanographic and atmo- fish have life cycles common to most lake species spheric frontal systems. To insure that appropriate (e.g., large eggs, relatively low fecundity, deposit locales are sampled at appropriate times, it will be spawning). The low number of species and associ- necessary to coordinate scheduling of NOAA and ated interactions as well as the limited number of UNOLS research vessels. Special emphasis must be physical and chemical factors that must be consid- placed on developing techniques for acquiring bio- ered are advantageous for research on the lake sys- logical, physical, and chemical information during tem. The principal forage fish species, the alewife, is extreme weather conditions. a marine fish that has successfully adapted to fresh- water. Alewife populations have been highly vari- able, showing dramatic changes since the time of their introduction. Empirical data indicate that marked swings in alewife abundance may bestrongly Anon. 1981. Oceanography from space. Oceanus 24(3), Fall influenced by harsh winters that severely limit re- 1981. cruitment. Bakun, A. 1983. Part III-Report of the working group on environmental studies and monitoring. In: G. D. Sharp and J. Sirke (eds.), Reports of the expert consultation to examine changes in abundance and species composi- tion of nerific fish resources. San Jose, Costa Rica, April 1983. FAO Fish., Rep. 291(1):41-54. Although recruitment research offers the promise of information that has great utility for fisheries and Bakun, A. 1985. Comparative studies and the recruitment habitat management, it is. a high-risk research topic. problems: searching for generalizations. CaICOFI Rep. As such, there are no guarantees for success. The 26:30-40.Boehlert, G.W. and M.M. Yoklavich 1986. Long- nature of the problem necessitates that a long-term term cycles of growth in Sebastes: Extracting informa- financial and logistic commitment (5-10 years) be tion from otoliths. Pro. Int. Rockfish Symp., Alaska Sea made. Recruitment studies have to extend for a Grant. number of generations of the species of interest. Ef- Brewer, G.D., G.S. Kleppel, and M. Dempsey 1984. Appar- forts will focus on organisms with short life-spans ent predation on ichthyoplankton by zooplankton and and/or species for which there is an historical data fishes in nearshore waters of Southern California. Mar. base or some historical index of abundance (e.g., fish Bio. 80:17-28. scales in sediments). Similarly, critical field experi- ments may have to await the occurrence of changes Carpenter, S.R. (ed) 1988. Complex Interactions in Lake Com- in the physical environment that are not predictable munities. Springer-Verlag, Berlin (in press). and occur at irregular intervals. Currently most field research on fisheries recruit- Crecco, V.A. and T.F. Savoy 1985. Effects of biotic and ment conducted by NOAA elements particularly abiotic factors on growth and relative survival of young NMFS is "piggybacked" on existing survey and as- American shad, Alosa spaidissima, in the Connecticut sessment programs. University research also is lim- River. Can. J. Fish. Aquat. Sci. 42:1640-1648. ited by the availability of vessel time and appropriate Dickey, T.D. 1988. Recent advances and future directions laboratory facilities. If significant progress is to be in multi- disciplinary in situ measurement systems. In: made, then the additional logistical support required Toward a theory of biological-physical interactions in will have to be provided. An effective fisheries re- the world ocean, B.J. Rothschild (ed). Reidel, Dordrecht cruitment program will make major demands on the Netherlands (in press). many NOAA agencies. The Environmental Research Laboratories must contribute essential expertise in Esias, W.E. 1981. Remote sensing in biological meteorology and both physical and biological ocean- oceanography. Oceanus 24:32-38. ography. The NOAA Undersea Research Program will be called upon to facilitate in the acquisition of Ocean System Studies 1 49 Fiedler, P.C., G.B. Smith and R.M. Laurs 1984. Fisheries Rice, J.R., L.B. Crowder and M.E. Holey 1987. Exploring applications of satellite data in the Eastern North Pacific. mechanisms regulating larval survival in Lake Michi- Mar. Fish. Review 46:1-13. gan bloater: A recruitment analysis based on charac- teristics of individual larvae. Trans. Amer. Fish Soc. 116 Francis, R.C., S.A. Adlerstein and A. Hollowed 1987. Pa- (in press). cific hake stock assessment: steady-state models ap- plied to the assessment of a dynamic population. Paper Shannon, L.V. (ed) 1985. South African ocean color and presented at IRIS-INPFC Symposium on Recruitment upwelling experiment. Sea Fisheries Research Institute. and Errors in Stock Assessment Models. Vancouver, Capetown, South Africa.Sharp, G.D. 1987. Climate and B.C., October 1987. fisheries: causes and effects in energy flows among coastal and offshore areas. Hennemuth, R.C. 1979. Man as predator. In: Contemporary Quantitative Ecology and Related Economics., pp. 507-532. Smith, P.E. 1978. Biological effects of ocean variability: Ed. by G.P. Pahl and M. Rosenzelg. Int. Coop. Publ. time and space scales of biological response. Rapp. P-V. House, Fairland, MD. Reun. Cors. Int. Explor. Mer. 173:117-127. Hollowed, A.B., K.M. Bailey and W.S. Wooster 1987. Pat- Steele, J., C. Clark, P. Larkin, R. Lasker, R. May, B. terns in recruitment of marine fishes in the northest Rothschild, E. Ursin, J. Walsh, and W. Wooster 1980. Pacific Ocean. Biol. Oceanog. 5:99-131. Fisheries Ecology: Some Constraints that Impede Our Understanding. Ocean Science Board. National Acad- Kerfoot, W.C. and A. Sih (eds.) 1987. Predation: Direct and emy of Science, Washington, D.C. Indirect Impacts on Aquatic Communities. University Press of New England, Hanover, New Hampshire. Ware, D.M. 1982. Power and the evolutionary fitness of teleosts. Can. J. Fish Aquat. Sci. 39:3-13. Koslow, J.A., K.R. Thompson and W. Silvert 1987. Recruit- ment to the northwest Atlantic cod (Gadus morhua) and Weller, R. and R.E. Davis 1980. A vector measuring current haddock (Melanogrammus aeylejinus) stocks: influence of meter, Deep-Sea Res. 27:565-582. stock size and climate. Can J. Fish. Aquat. Sci. 44:26-39. Werner, E.E. and J.F. Gilliam 1984. The ontogenetic niche McConaugha, J.R. 1987. Blue crab larval distribution and and species interactions in size-structured populations. recruitment. Coastal Oceanography and Climatology News Ann. Rev. Ecol. Syst. 15:393-425. 5(4):43-44. Wooster, W. (ed) 1983. From year to year: interannual Methot, R.D., Jr. 1983. Seasonal variation in survival of variability of the environment and fisheries of the Gulf larval Engraulis mordax estimated from the age distribu- of Alaska and the Univ. of Washington, Seattle, 208 p. tion of juveniles. U.S. National Marine Fisheries Service Fish. But. 81:741-750. Mulligan, T.J. 1986. Stock identification based on elemental composition of otoliths. Int. Cors. Exp. Scos., C.M. 1986/ M:19. Ortner, P.B., R.E. Pieper and D.L. Mackas 1982. Advances in zoolplankton sampling. In: Fish Ecology III, B.J. Rothschild and C.G.H. Rooth (eds), pp. 355-379, Uni- versity of Miami Tech. Rep. No. 82008, Miami, Florida. Paine, R.T. 1980. Food webs, linkage interaction strength and community infrastructure. J. An im. Ecol. 49:667-685. Pietrafesa, L.J., G.S. Janowitz, J.M. Miller, E.B. Noble, S.W. Ross and S.P. Eppsily 1986. Abiotic factors influencing the spatial and temporal variability of juvenile fish in Pamlico Sound, North Carolina, p. 314-353, In: D.P. Wolfe, Ed., Estuarine Variability, Academic Press, NY, NY. Chapter IV The Role of Sea Ice in Controlling Arctic Ecosystems 7 i I I E M -@Y v", @ Problem and Opportuni 4 conseauen, tl better allocate hu n and y@ to ma Sea ice exqts an,'infly -th entiql,cqntrqf,q -;,:, J qc c T @iP@ marine ecosystem. ts extent, orination,ah fl die t'are_' critical factors forithe a a pi6ducfibiz npyii prim ry y 'searc,, 'Shelf. Sea ice acts as a platform vn 'Ri h _S of the Bering Sea I qtegy habitat for mammals@-and,@a substrate for,thati-ftyplaht- _P0, , TeS that form the base of the food chain svpportih' b' e' ed, earj* will',test g,, 0, r the hypothesis that -Pro shellfish, fish and maniinals@'The maximu-m ice 'extent and sea- rong-, spri n8_ 'O'Od year-to@year lizes the sea and allows'a st pulse" :ft ice- retreat@ account r the major f the Bering production, making- the AinericaifAfatici 6' 6f@.`the',O-@ qki@tproductivity 0 dnl Chukchi Sed8. Elements'in this study include:. richest commercial fisheries in, the world.' Thexetreat eas is-_@ _ti6ri!qfexistt ng, to ogica the ice edge throujlr'ihe-Biring@and Chukch -i'S'"" b"I "' I and physical equal to @ertih@inkan`qreaftoin te:i@s .to Wa,ta sets'and'initidii6n of aWhistorical study us- 6urfNins 'to --the Mississippi, -trig satellite -and other remote sensing border and the Rocky M data, River. The Northern Bering Sea-has ntly-beenshow B p rece@ i hys;:ical"coceanographic sampling ry prod h riod, to have the world's highest primii uction rates. a'16 ng te ice e @ze over an' eig t year pe s of, energy flow Yet the critical pathway ,,,'_E_,,xamihati6rt otthe,dyhamics of primary produc- ion, -nut rient re ling -and fluxes at the iceedge, to bottom organisms and then to commer'cial,fisher@iei,,' t" -' __ _' Wye and mammals are not well -understood. I NOAA, with its research capabilities' a e ite ini- rticu e 's 6eitiddiflux- of pa lat material, qgery and environmental data archiv ,.! @,fis4eries,@ Examination of therelatiort ice,edge, primary Of mandates is well configured to address this important, --,prodoctiontooccUrrencea-ndti'mingofzooplank- problem. jv&ahd larval fisheg and'crustaceans, Understanding the role and dyizamiesof this -search Process Re- on variabil'ity'of-high-latitude weather c 0, 91 an d' hti,' i-oh)ce'eiient, -location, and timin' then will enable NOAA'to'lJr6iiide, the tools to pe rm closer management andJ6, Make - Tfo y4y ce,ahd oceanogra hic processes in P more prudent fisheries allocation decisions.' The le 't6astalareasby use of in situ and remote sensing timethat thisapproach yields @willl also,enllqble@ cqm-niar,7@@,_@ tl@chniques,- cial fishermen, seafood #roces@dis, and e reevan, 52 The Role of Sea Ice in Controlling Arctic Systems Synthesis of field studies into models and exami- ploitation, and competition. The answers to the impor- nation of fish and mammal population dynamics tant questions of production of groundfish, crabs, and in the light of the expanded data base on weather, salmon, and of the timing of salmon harvests, lie in sea ice and oceanography, understanding the dynamics of the Alaskan continen- Reactivation of the Bering Sea ice model developed tal shelves. earlier, its extension to tire Chukchi Sea, and its In his statement on United States Arctic policy rise to design sampling strategies, (April 14, 1983) the President emphasized that the Measurements of currents, nutrient dynamics, United States has unique and critical interests in the biological productivity and particulate flux in the Arctic region. In light of the region's strategic impor- vicinity of Bering Strait during the ice melt-back. tance, the Administration feels that the Arctic war- rants priority attention by this country. A review of national issues and priorities for the Arctic Research Why Now? and Policy Act has been completed by the Polar Re- search Board of the National Research Councill and the Dramatic, largely unexplained changes are occur- U.S. Arctic Research Commission. Based on these re- ring in the Arctic system: ports, both the Interagency Arctic Research Policy The king crab and Tanner crab harvests have Committee (IARPC) and the State of Alaska estab- plunged since 1979 with serious economic impact. lished an implementation plan to address the structure Other commercial species such as pollock are being of ecosystems of the major arctic shevles (Bering Sea, heavily exploited. Beaufort Sea, Chukchi Sea) thorough integrated pro- The northern fur seal population is declining rap- grains with a strong physical oceanographic and idly (4-8% per year). All large baleen whale popu- weather/climate component. They specifically identi- lations have been severely reduced and mail not be fied the biological production and food web dependen- recovering. The walrus population has become cies in relation to physical features such as ice edges, very large and is in danger of crashing. Sea otter polynas and hydrographic structures (i.e. fronts) as populations are increasing and competing with priority areas for research.The Arctic and Antarctic fishermen for shellfish. are less well known than any other area of comparable Dramatic increases in marine growth and sur- size. Most Arctic-rim countries, particularly the Soviet vival have been documented for salmon popula- Union, possess Arctic technologies far more advanced tions of the Bering Sea. Recent indications of an than those currently available in the United States. end to the boom in production of the past ten years Sponsorship of currently neglected research in basic have very serious economic implications for both science is a necessary andproper functionof the federal commercial and subsistence fisheries. government to fulfill national objectives in Arctic re- search. Benefits Polar Research Board, 1985 National Issues and Research Priorities in the Arctic National Research Council, Washington, DC.123 pp U S Arctic Research Commission 1986: National Needs and Arctic The ocean's productivity defines the limits on all Research: A Framework for Action, Los Angeles, CA. 27 pp marine biological systems. The work here will provide arctic marine ecosystem researchers with tire basic in- Polar Research Board. 1985. [bid, p 44 put to estimate parameters of biomass, recruitment, ex- Ocean System Studies 1 53 plays an important role in determining the timing of the spring bloom. Increased stratification from ice melt allows the development of an intense phyto- The coastal aboriginal peoples of Arctic America plankton bloom at the ice edge as soon as active melt- developed cultures based on exploitation of the ma- back begins. Consequently, the growth season is ini- rine mammals and fish of the Bering, Chukchi and tiated earlier than would be possible in the absence Beaufort Seas. The same resources were the basis for of sea ice. For this reason, variations in the southerly exploration and settlement of Russian America and extent of sea ice in winter have major ecological con- later, the interests of distant water fishermen and sequences. The retreat of the ice edge fertilizes a whalers from the United States provided an impor- region equivalent to Texas to the Canadian border tant incentive for the purchase of Alaska. From the and the Rocky Mountains to the Mississippi River of beginning the Arctic environment has been the major the continental U.S. (Figures I and 2). limiting factor in exploitation of these resources. Our The influence of year-to-year variations in sea ice knowledge of the role of ice in the ecosystem is criti- extent and retreat is an important question if the cal for understanding and exploiting biological pro- impacts on the biological system are so extreme. A ductivity. We now have some understanding of re- significant climatic change occurred during the span gional variations in primary production in the south- of the previous interdisciplinary Bering Sea ice edge eastern Bering Sea and ideas of the physical/chemi- work (1975,1976,1977, OCSEAP/NOAA; 1982,1983, cal factors which produce these variations. The an- 1987, NSF/Ocean Sciences; 1983, 1985, 1987 ONR/ nual primary production cycle of most of the Bering Arctic Programs). The three years during the late Sea shelf is dominated by a spring pulse. Spring 1970's were cold, whereas during the warm years blooms occur at the onset of ocean stratification, with 1982 and 1983 the ice only reached the latitude of St. their duration highly dependent on storms to replen- Matthews Island. The warm period has extended ish the surface layer with nutrients. Nutrient enhance- through 1987. The decade of the 80's has been the ment by wind mixing from deeper layers is respon- sible for from 10% to 50% of the yearly spring bloom total nitrate uptake, depending on year. Ice cover Figure I The biologically productive seasonal ice zone of the Bering and Chukchi Seas is roughly equivalent in area to the major grain producing regions of the American farm belt. Average Ice Edge Retreat Average Ice Edge Advance 54 The Role of Sea Ice in Controlling Arctic Systems warmest decade measured to date. The first three rial sinks through the water column. Food chain rela- years were cold with the ice reaching across the Ber- tionships linking the ice edge and spring bloom to ing Sea Shelf, whereas in the warm years 1982 and the other ecosystem components are not well known, 1983 the ice only reached the latitude of St. Matthew but it seems reasonable that the rich benthos on the Island. The chlorophyll content of the water at the ice shelf and its top level consumers, shellfish, walrus, edge and the primary production were significantly whales and fishes, are in part dependent on this lower during the warm years than during the cold efficient use of early-season solar radiation. In par- years. When the ice edge is closer to the shelf break, ticular, the benthic-mammal food link maybe excep- the higher nutrient concentration below the surface tionally important in the American Arctic, and these layer increases nutrient supply through ice-edge large organisms provide a degree of biological stabil- upwelling. Furthermore, the earlier water column ity. Their role in nutrient recycling is not known. stratification associated with ice melt prolongs the Spring bloom phytoplankton may also be particu- total bloom period. The total annual primary pro- larly important in feeding juvenile fishes and crusta- duction in the outer shelf domain is probably in- ceans. creased significantly during a cold year. Therefore, For crab, the period from egg hatch to settlement we state the hypothesis for the ICE program as fol- of larvae is likely to be the major determinant of year- lows: Interannual variation of maximum ice extent class strength. The area of hatching must be more and seasonal retreat account for the major year-to- clearly delineated using intense survey efforts. year variability in the biological productivity in the Oceanographic and sea ice information is important Bering and Chukchi Seas. To clarify the effects of the in determining advection of larvae and subsequent position of the ice edge, it will be necessary to follow settlement. Because considerable spawning and sub- the ice northward during its retreat and examine sequent larval dispersion occur near the ice edge, ice variations in ice-edge phytoplankton productivity as edge phenomena may be important to this stage of the ice retreat passes over various parts of the shelf. crab larvae survival. This is particularly true for crabs A large proportion of the ice edge production in the northern Bering Sea and Norton Sound. probably reaches the benthic community, since the grazing community in the water is small and not very active. As the bloom progresses, organic mate- Figure 2 Melting sea ice stabilizes the water column at the ice edge promoting primary biological production as the ice retreats in the spring. Sun Wind Fres P Potar ro G --l 0 v HI n4v I io ia, Ocean System Studies 1 55 The seasonal variability of the extent of ice cover YEARS o MOST ICE EXTENT o (a) and of its residual melt water is governed by the 57/58 65 OCT-FE.B 65 60/61 4 storm climatology. Figure 3 shows the composite 63/64 8 75/76 storm tracks for the 5 heaviest (A) and 5 lightest (B) 76/77 60. 7 3 1 years during 1958-1982. During the heaviest ice years 60* the tracks were shifted southward along the Aleu- 5 tian Islands and eastwards into the Gulf of Alaska. 43 This gives rise to more north and northeast winds 55o 2 3 57 54 55* which move the ice farther south, increasing the ex- 46 tent of ice cover. During the lightest ice years (Fig. 42 1 54 62 6 3B) more storms move north across the western part 51* 5 '41o - 65 75 51* of the Bering Sea. The result is a greater incidence of 171 oE 157o south and southwest winds on the shelf, compacting 180 170oW 160o the ice cover and closing the ice-growing leads. Fig- ure 3C shows the maximum extent of ice cover in the YEARS 65. LEAST ICE EXTENT 65* (b) Southern Bering Sea. In an average year about one- 58/59 OCT-FEB 65/66 2 half of the domain is covered. In heavy years ice 66/67 23 29 surrounds the Pribilof Islands and covers the entire 77/78 78f79 21 25 23, continental shelf. Interannual variability in seasonal 60' 60' sea-ice extent in the Bering Sea is controlled by vari- ation in storm-track position related to large-scale 52 46 42 49 differences in the general weather circulation. 55*6 1 4 46 42 38 55* 39 2 @Z' IF, cb - cz> 2 6 67 51* 5 45 41 38 51, The research is based on a program of field meas- 171*E 180 170*W 160o 157o urements, historical analysis, and modeling. Physi- cal process studies address the movement of ice by 65' o M variable ocean currents, the relative importance of 65 local radiation versus heat advection by currents to V A melt during ice retreat, and the link between sea- sonal and interannual atmospheric variability and 60o 60' ice extent. Biological process studies address the in- fluence of the ice edge on primary productivity and Light Ice r the efficiency of transfer of energy from the surface Extent Ice Extent Average to bottom living organisms. Remote sensing provides 55 Section ea 55 a means of longer term monitoring. Modeling activities will synthesize the understand- ing of causal mechanisms between sea ice and the 51* 51o regional biology determined from the field studies 17loE 160o 157* with the historical atmospheric time series to test the 180, 170oW ICE hypothesis and corollaries: Figure 3 Storm track counts by 2' latitude v 4 longitude over October to Februaryfor the five heaviest (a) and five lightest (b) ice years in the period 1957 to 1980. The arrows suggest core pathways of the preponderant storms. Panel (c) shows the corresponding extents Of the ice cover. (Overland and Pease, 1982: Monthly Weather Rev.) 56 The Role of Sea Ice in Controlling Arctic Systems able to differentiate the relative contribution to melt- ing of the local radiation budget versus northward advection of heat by ocean currents. Studying the ice pack will require measurements over five years, since the interannual variability of the weather conditions To extend our knowledge of sea ice behavior and and the severity of the resulting ice conditions is biological consequences to the northern Bering Sea, large. The variability of the barotropic current com- Bering Strait, Chukchi and nearshore Beaufort Seas, ponent through the Bering Strait also has large vari- all regions of intense environmental, commercial, and ations from year-to-year which may affect the ice strategic interest, three important physical processes drift patterns. must be considered. First, the transport of ice by The coupling between the atmosphere and sea ice ocean currents must be understood. Net ocean trans- on seasonal and interannual time scales will be ad- port through Bering Strait is toward the north and dressed by detailed analysis of the North Pacific over a year averages about 30 cm / s. On shorter time- Oscillation (NPO) from historical atmospheric data scales, meteorological forcing results in stronger sets in the context of the northern hemispheric gen- events (current speeds up to 125 cm/s) which can eral circulation. The first element is to determine the reverse the flow over time periods of two to seven persistence of the NPO as a basis for statistical pre- days. This current system varies remarkably in diction. Such approaches have not been entirely suc- strength and direction and can cause the ice to move cessful at mid-latitudes, but, since the duration of in the opposite direction as the local wind, but is NPO events is often greater than 10 days and the poorly understood. Second, the role of heat advec- anomalies are large, such approaches should be at- tion by this current in the spring and summer months tempted. Statistical methods now exist but they must versus the local vertical heat flux and radiation bal- be based upon more complete causal hypotheses. ance in the melt-back of the ice pack is not known. The data base which exists for the study are the sea Third, there is no realistic theory describing the inter- level pressure, 500-1000 mb thickness (which corre- annual variations of ice cover and their relation to late with air temperature) and 500-mb steering level atmospheric circulation. Understanding of the sea- wind fields beginning in 1947 as well as long time sonal cycle of arctic weather can be improved by series of weather observations from Arctic stations basic research on the relation of high-latitude atmo- and sea ice extent fields derived from satellite obser- spheric circulation to forcing by the land and sea ice vations continuously since 1969 and sporadic before distribution and by lower latitude circulation. that time. The first measurement component of the ice pro- A second feature is to systematically assess the gram is an array of satellite-position drifting ice buoys possibility of positive feedback of heat between atmo- deployed in key areas along the coast. These mea- sphere, land, and ice surfaces for the sub-Arctic. It is surements will be used to evaluate the extent of shore known that the land/sea distribution and coastal effects from the coast on the ice velocity, to test open orography provide forcing to the long-wave atmo- pack constitutive laws, and to create a nearshore spheric circulation pattern. Of particular concern to constitutive law, if necessary. A second set of mea- the seasonal time scale is the feedback between the surements will be conventional current meters and seesaw relation of Bering Sea and the Sea of Okhotsk, pressure gauges to address year-to-year and seasonal and the Alaska and Siberian landmasses. The ther- variations in ocean transport. A third measurement mal mass of Alaska and Siberia are substantial and component will estimate thermodynamic variables. ground temperatures can increase by several ten's of Sensible and latent heat fluxes from the atmosphere degrees over a week's time. These temperatures rein- to the ice and sensible heat flux from the ocean to the force the long wave atmospheric weather patterns ice can be computed by careful temperature mea- ability to maintain the existing storm track pattern. surements in the boundary layers. High resolution Sea ice acts as an extension of the continent and thermistors will be added to the anemometer and effectively acts as an insulator between the air and current meter masts at the drifting buoy sites for this the relatively warm sea temperatures. When ice is purpose. In addition, a broad-band radiometer will retreating in the Bering Sea due to southerly winds, it be attached to each station for estimates of the solar is advancing in the Sea of Okhot@k. It is not known and long-wave radiation incident on the upper sur- whether this relative ice motion tends to reinforce face of the ice. This measurement program will be the existing weather pattern. Ocean System Studies 1 57 HISTORICAL KING CRAB CATCH Eastern Bering Pea, 1953-84 70 1 60 The renewable resources of the American Arctic, TOTAL 0 ---- U.S. I specifically fish and shellfish, represent one of the .......... Japan Nation's greatest commercial assets. Yet there are 50 U.S.S.R. dramatic, unexplained fluctuations in the resources. The recent catches of king crab in the Bering Sea are 0 shown in Figure 4. This was an international fishery -r- 40 - t shared with Japan and the USSR until the United U) z States developed the capacity to capture the full har- 0 30 - vest in 1981. The catch grew rapidly from 1958 until 0 EE 1963, reaching a peak of 28,000 mt, then declined W from 1964 until 1971 to a low of 9,000 mt. It rapidly 2 20 - increased from 1972 until 1979 with a peak harvest of 65,000 mt, and has since declined to less than 5,000 mt in 1985. Such large year to year variability in 10 - stocks, both natural and that due to fishing, has cre- ated major difficulties for both industry and manage- 0 ment. 1954 58 62 66 70 74 78 82 86 The Bering Sea is a vitally important rearing and YEAR feeding ground for a majority of the salmon stocks Figure 4 Recent catches of king crab in the Bering Sea. which serve as cornerstones for a large number of commercial and subsistence activities in Alaska and other areas of the Pacific Rim. Very dramatic recent increases in commercial production (Figure 5) from 21.9 million salmon in 1974 to 144.6 million in 1985 Figure 5 Recent explosive increase in commercial salmon production from 1974 to 1985. ALASKA SALMON CATCHES (Millions) 1887-1986 160 1 1 1 1 1 1 1 1 1 Peak 1985 140 - 120 - 100 - 80 - 60 - 40 - mom= 20 - 0 1887 1897 1907 1917 1927 1937 1947 1957 1967 1977 YEAR 58 The Role of Sea Ice in Controlling Arctic Systems have illustrated the importance of the marine envi- electromagnetic spectrum and is, therefore, an ideal ronment in determining the well-being of both parameter to be measured remotely from space. subsistence and commercial fisheries. The first promising instrument is the passive mi- The biological sampling strategy is to make stan- crowave imager, SSM/l, flown on the DMSP satellite dard nutrient dynamics, productivity and vertical in 1987. The bands selected for passive microwave flux measurements in the vicinity of the melt-water sensing are largely unaffected by clouds and require stabilized region of the ice edge and contrast these few atmospheric corrections at high latitudes. At measurements with adjacent open-ocean measure- present a single reading is incapable of being inter- ments. These tasks will follow the retreat of the ice preted beyond the statement that ice is present or edge through the Bering and Chukchi Seas. The sec- absent. With the new instrument, it should be pos- ond part of the sampling strategy is to determine sible to use a set of multi-spectral, multi-polarization benthic (bottom) biological activity to determine the readings to distinguish various sea ice properties for efficiency of transfer of energy from primary produc- Alaskan waters. tion to bottom fauna. The shallow Bering and Chukchi The most powerful instrument, combining high Seas may be unique in their efficiency of transferring spatial resolution with virtual independence of atmo- energy from the ice edge enhanced primary produc- spheric effects, especially in the dry polar atmosphere, tivity to higher trophic level species via intermediate is the synthetic aperture radar (SAR). The European bottom communities. Sampling will be oriented on Space Agency plans to launch a satellite (ERS-1) with primarily south-north transects and will occur dur- SAR in 1990. A SAR receiving station for ERS-1 is ing spring ice retreat. Measurements will be required planned for Fairbanks, Alaska which will be made over a five year period to provide representative available for dissemination SAR data for research, sampling and consider year-to-year variability. This and will provide the all-season, all-weather data base strategy will allow determination of the variations in necessary for sea ice processes research on scales as the importance of ice by biological regimes over the small as 25 m. Analysis of SAR imagery will provide entire north-south area, To date, ice edge biological baseline information for verification of the sea ice work has been restricted to its southwest parts. and climatology models proposed for the Alaska In parallel with field experiments, efforts to model region. Biological studies can benefit from synoptic regional ice/ ocean circulation, heat budget, and bio- coverage with color imaging for chlorophyll, a mea- logical system will be undertaken. Initially modeling sure of phytoplankton biomass. Aircraft-based capa- will concentrate on regional ocean circulation, ice bility would be of greatest use, to provide coverage drift, and stabilization of the mixed layer by ice melt. in communities with less surface-based sampling As results from the field measurement programs and from a research vessel. atmospheric studies become available, systems stud- ies can be formulated to address the causal links between interannual atmospheric variability and year-to-year changes in the biological conununities of the Bering and Chukchi Seas. The hypothesis to be tested by the ICE program is These measurements will be used by the Alaska that the interannual variability of the location of Department of Fish and Game to test the hypothesis maximum ice extent and seasonal ice retreat is the that the times of arrival of maturing salmon at the primary cause of year-to-year variation in the bio- fixed geographic reference frames sampled by com- logical productivity of the waters of the American mercial fisheries are driven by physical factors re- Arctic. This program will implement the highest pri- lated to ice edge location. ority research recommendation of the U.S. Arctic Research Commission as mandated by the Arctic Research and Policy Act of 1984. The program is based on a ten-year program of physical and biologi- cal field measurements, study of historical variabil ity of arctic weather and its influence on ice motion, and modeling. Modeling activities will synthesize In situ measurements of sea ice and biological the understanding of causal mechanisms between processes are cumbersome, expensive, and limited sea ice and the regional biology determined from the with respect to aerial coverage and spatial and tem- field studies with the historical atmospheric time poral resolution. Fortunately, the presence or absence series to test the ICE hypothesis. of ice on the ocean affects virtually all regions of the t gg", Chapter V 74, g,@ ... ...... ...... Estuarine Systems: Productivity and Environmental Change ty@@to,@pridict,,re@ponses,of,estuari*ne Problem wid ea A 40' n it @aetions an atural' events, n 4vtr- nine ,their -assoetafei 'Vim"I neh -ective, teaMjzies to man- Estuaries and -,,, _c6ds@ta nd dr em fii@Coff -S luable and M va -yet I-v Ine u r 4 14-A environment. They provide, the. biolvgjc@ql@f *d, waters'. -4- for our,productive coastal j,.,eswarr@ 'a' @t, two-thirds 'of the estuarine-dependent. Fu qV ind-j 'ulatiq@ turep _ic, i@,[urtc -st @'Of, estu and the growth is forecasted f&b';r`eoas:t@iI'ii' t- '-.inqWeas@ Mg arine processes yar @n Ter an @j 2- such that, by'-1990`,75%c,'_',bfIh,' -LIS"' W, n-', n "'O@O tftese@ systems " in creases e@l I, ;%VOM qkip. jected to -live within 50 miles@@f aim regarding dy the are alr@a 's @iijji' c @seq ,e f it tibie. @hageni&tit strategies. 'nice sh`o@t'O@_ OAR will !in- p n heavy use. Evide thdtlhi hea@hW`@ 'ourW lenient- -a s prog-ria-m" fi@cused 6' y jken@44-@W__&',@@Am;7 @_. @@), " - , " - " D 4@ hing. @egraw i6h'*AvWii@4i iion s'esiuallkliel's", I ity mental quali a' 'r angrQve forests, 1 TO@ 4, f6 r in of fis h e r ies Ids s e's and @ t It rea fs@ t, to -"Ply b h c. hea I i e@" i4tie' d -, nd" -er I flIats the,wid co umn. Unlike many - ?q ti@ h_'@" hich empha-' safety have become wide,,, ixygen, epe@ s@ t7c, programs w bay waters -have resulted--in', 'specific pollutants, xamme, - ftd re the factors levels of toxic contaminaint@'have`been-r_p dikkel,", 'i,@@'e` is-lapproach-witt-e-J '-"-a compa ments and benthic organisnisft uaries. om urban?est ,-'cc@ii,troll.ihglpr6du@tivity in a limited number of estuar- There is widespread disappearance9f subinft& un nj,, esear the co t R ch will 'on Y in coastal-'- " ' _' ' vegetation; theie'iicorit" ued kiidf wetldhdi@'--"cus,` variation, in freshwater inflow on and there arediamatic declines -in eiMa=,r'in'@@"'d'ep-'e"'nd@en't,@@---@"&@'E@-'-,'..'@ -I-,`V'he4&Ot @qf duct n species including oyst@rs-;clarhs,@t@liped,bag@,@ii-h'4t,@had. W,` ical,'@s nicture'and biological pro ivity i P 7-7 -' Recognizing the urgentneedfor-actitm @io maihtAin-@' and enhance the ecql6gica-l y Z4V,@#eet--q_f,nutrie_hts and nutrient cycling on estuaries, the US Co _d' ngress hd_s:ur'ge, Aidevd@ prodactivit 3;,The,. relationship, of-habdat structure to habitat of management plans:-fgr these, -cr,itical-,,areas,.@-@0,AR@-,, el di the cumulative eftts of @,In u research can ngot -de -1 (tial"for, -vet;,, 60 Estuarine Systems: Productivity and Environmental Change g .4. The'relationship' nc ion resources cross geo between the nature,and Ju t political boundaries. As the Mition@s w- V ing of estuarine food ebs and habital lead civilian ocean science agency, NOAA is uniquely ity. V qualified to assist in this role by conducting a program Q' @@q to increase understanding of estuarine systems, and by Q%_ a providing information necess ry to &edict and assess, Why Now? changes in these sy ems. NOAA's div rse c pabi7ities Aff st e a atm6sphei4c sc- ce, ocean systems'dy amic' ien n s, blo g, the Clean Water Act and the,411 , logical processes, resource p rotection, and coastal zone Recent legislation (e. Water uality Act of M7) an achims-by-Federal"..,1@1, management provide a pool of expertise capable of ad- V-P ajor-@estuarine prob n a,com" prehensive `Mv'dressihOn ...fldministrators@have Legislators, environmental @managers, the media and'114@' manner. the general public are being made increasingly aware of,4 NOAA has a national network of estuarinelcoastal estilarthe systems, their importance to coastal environ facilities that includes 27 NOAA laboratories, 17 ments, and the potential harm, that ma come'to theselk_,'@i Estuarine Research Reserves, Sea Grant Programs, y systems as a result of short-sighted or inadequate man-@-011_1 and several, NOAAlacademic cooperative insti@ 1._-aggivent.. This.heightened-awareness coincides, with.thei6jr. -,,itutes.-_ I maturation of many subdisciplines within -,enpiron- NOAA employs or funds many of the Nation's is' dinkestuarine scientists, some of whom- mental science. In particular: outstan -Improved computer technology and techniques, 011 serve in NOAA's research laboratories and others andenhan@ed modeling ca ed pab il iiiesin a,ble%;f-o'r' " th llfxkg@ who are associat with NOAA through existing @firsVtime -desc tion9ft yari, 'namics-withibsk"'. ,@',-,`_progrdms_ rip some realism. NOAA scientists have excellent relationships with New instruments andfacilities, afida better under---.,.'. sci .entists in academia. These relationships have i V M standing' of ihe'time Ji-aines of estuarine processes, formed the bases for partnerships among, estuar-, ine interests in NOAA, the academic community, can now be applied byscientists to-achieve, major-, new insight into estuarine processes. T; and the states. These partnerships can be used'to o- rafii'df NOAA has ih,'ptace a, c re pkog -e, stuarine ,1 -@iovidi expertise to re''spond to research- and as- Ir efforts and is- positioned 'to undertake a fpcused com--,M needs. sessinent- 'd Ofe'st,utaries. ,Ad itio rtn rships,between NOAA'an'd. academia-,. re, einsive'stu y Vq, e' p h I - I - 37he, pa forNOAA; its researchers and their academic collengul, e5.7'.-K can be used efficiently and effectively ifirough'the rd tC6fte rogranitorespon d- th ' i M-- , upon t 's1reng s. - I-Sela-G'', m""W" res.of)'@nds,on,site-spe'cifi@='problems@are,@ii,%'k@ Expenditu to complementary needs r long-term estuarme@ fo seen as a poor, investment for developing a nati6nal',"flu, res-ea rch and monitoringas well as for short-term approach to estuarine management by a research rather i process-oriented studies. is resource managent than a resource m aln-dgement agency. Meeded -how- ari@ NOAA' sensitive to ent needs. studies on the basic functionime of esiuari .he lidbitats "Af In partnership with the states, NOAA is'respon- and relationships to overall productivity around the-l! sible for managing living marine resources and country. the coastal zone. Additionally, NOAA advises NOAA is poised to' initiate "Mu'it'idise@i"p@ii'na'r-y''a-nd."--o"@' other agencies (e.g., U.S. Army Corps of,Engi- multi-institutionalresearch to addr6s cr estuar- _.t, neers and Environmental Protection Agency) on ine issues. Failure to exploit this opportunity -will re- qV@i living mari .ne resource s, tinder numerous legisla- suit in a loss,of research momentum, and-c6uld deprive iltv tive directives. society of information needed to NOAA has a great -deal of experience in predict- ies into the 21st century., ing and forecasfing-experience that should be di- _I- "EM'. rected to estuaries. n en NOAA's role and experience i , vironmental pre- iction and'forecasting will form the basis@ for tying' Why NOAA? A d theseres6urcesa' responsibilit* together. nd ies L The Federal govern-m- ent.has a'role in estuarine go ,v- ernance; especially where estuaries- ana -,uses of th :e,Fede er e,51 Ocean System Studies 1 61 IJI Ben"fits-, o Improve our ability, t sustain and increase long- '-wa -,w, _plj'hji@ term econom iEreturns fiom estuaries through rec-, Scient@fic findin --res urpf 'extraction,,, transportation, and ,,gsftom our understandingof estitariheprod@@'ti'@it-'y--aiid-ther@by - waste.disposdt. provide the critical knoWleqge,need8dto_:'I iwis ne'z@4nozvledge can be used by resource man- Develop appropriate@:,tvdter-@,,[email protected],@ .agement@a ncies to develop cost-effective regulations, gies, -.@.,,Jm qw,,permitting, and reduce litigation. The ulti- Develop cost-eflective clontrb stidtegies,@ behefit"wilt be'enhanced'economic development *Maintain and restore eriticalest'u'@,ril@e'@6@it@t@,,';,-@",-@ 4hr u -the wise, use and *ahagment of estuaries. Restore and enhance@ estu"a rine-depehdem@, fisher-w -------------- 9 hypoxia -related fish kills have increased in fre- quency and severity, as has the incidence of Estuaries and their associated waters are a valu- disease in estuarine finfishes and shellfish; and able, yet vulnerable component of the world's oceans. *landings of many economically-important Although the estuarine/ coastal complex comprises estuarine-dependent fishery stocks show long- less than one percent of the marine environment, it is term declines. by far the most productive. Spartina salt marshes, for Although excellent research on estuaries is being example, produce ten tons of organic material per conducted: acre per year, compared to four tons per acre per e ongoing estuarine research often is short-term, year produced by fertile hay fields. Estuaries provide reactive and oriented toward local problems; the food, shelter, and spawning grounds for over 70 and percent of the Nation's commercial fisheries land- e basic research often is fragmented and accom- ings by weight and over 66 percent of these fisheries plished on small scales. by value ($5.5 billion in contribution to the GNP in There is need for a national program of funda- 1986). Seven of the ten most valuable commercial mental research on estuaries that will lead to a quan- fisheries-Gulf shrimp, sockeye salmon, menhaden, titative understanding of these systems, and ulti- pink salmon, oyster, South Atlantic shrimp, and blue mately to the ability to predict the implications of crab-depend on estuaries to survive. The estuarine/ changes to estuarine environments. To be fully suc- coastal system is equally critical to sustaining recrea- cessful, such a program must include provisions for tional fishing: an estimated 17 million sport fisher- interpreting findings as they become available and in men in the Nation generate expenditures of over $7.5 terms that are relevant to managers. An integrated billion annually. program of fundamental estuarine research is re- Estuaries also are important to society as avenues quired if estuarine resources are to be managed for transportation, sites for industrial development, wisely. areas for the disposal or treatment of wastes, and environments for recreation. These and other con- ECEREEM11111111111 M flicting uses may be taking their toll on estuarine 0 ME on N I Run Em OEM T"W" ecosystems. Several trends suggest there is cause for concern over the well being of estuaries: Research undertaken as part of the estuarine re- � the demographic shift toward the coastal zone is search program will focus on estuarine productiv.;ty2 placing increased demand on estuaries for trans- and factors that control secondary productivity within portation, waste disposal, urban development, specific estuarine habitats. The ultimate goal of work recreation, and natural resources; � nationally, the loss of estuarine wetlands has 'Hypoxia refers to a level of oxygen dissolved in water that is stressful to organis s; generally, the level is defined as less than 2 milliliters oxygen/ increased from about 20,000 acres per year to liter wamter. more than 30,000 acres per year, over the last ten 'Estuarine productivity can be considered in two principal forms: primary productivity and secondary productivity. Primary productivity refers to the years; creation of organic compounds (fixation of carbon into organic forms) by plants, while secondary productivity is the growth of animals that consume plant materials, either directly or indirectly. 62 Estuarine Systems: Productivity and Environmental Change is to develop the capability to predict estuarine pro- e amenability to a range of experimental designs ductivity (quality and quantity) in the face of envi- and scales; and ronmental change. * appropriate scales for management activities. The research program will focus initially on com- The Spartina marsh and seagrass bed habitats will paring productivity in a limited number of target be the initial foci of the program. Once the program estuarine habitats. Investigation of estuarine produc- is well underway, other habitats will be considered tivity, focused on target habitats, offers an alterna- for study. Following are general description of each tive to previous studies that have attempted compre- of thefive habitat types selected. hensive examinations of single estuaries, or investi- gations that have focused on specific estuarine wastes or pollutants. The comparative approach will im- prove generic understanding of estuarine processes. The more tightly focused comparative habitat Spartina is the dominant plant type of coastal salt approach will make estuarine studies more tractable marshes of the United States. On the East and Gulf and allow comparisons of estuarine productivity in Coasts, Spartina dominates the extensive coastal similar and in different habitats over space and time. marsh areas that extend from Maine to Georgia, and Specifically, the approach: along portions of the Gulf of Mexico. On the West �overcomes problems associated with the com- Coast, Spartina marshes occur in the Southern Cali- plexity and heterogeneity of large estuarine sys- fornia and San Francisco Bay areas. Salt marshes are tems, by limiting the number and types of habi- believed to serve as nurseries for juvenile fishes and tats and species to be evaluated; invertebrates, contribute shelter and nutrition for an �reduces overall effort needed by focus on simi- array of fishes that migrate with the tides, act as larities among estuaries instead of uniqueness coastal stabilizers from storms, and play a role in of individual embayments; and land accretion through the trapping of sediments �enables application of process-oriented findings and the formation of peat from decomposed marsh within similar habitat types around the country. plants. In effect, such sediment trapping and peat Given the program focus, there is a high probabil- formation will, in the very long term, cause these ity that efforts will result in substantive "break- habitats to filled completely, thereby bringing about throughs" in understanding of estuarine productiv- their natural demise. ity and, thereby, will provide critical information Spartina species exhibit high rates of primary pro- necessary for effective estuarine management ductivity and contribute to secondary productivity nationwide. primarily through the cletrital foodweb-. Historically, The United States Fish and Wildlife Service (F&WS) Spartina marshes have been dredged for navigation has developed a classification system of wetlands and cleared, drained and filled for development. The and deepwater habitats. This F&WS system was de- potential long-term losses associated with destruc- signed by wetlands scientists, with input from re- tion of this habitat include reduction of fisheries pro- source users of the private and public sectors. The ductivity, destabilization of shoreline areas, and re- classification system provides a flexible framework cluction of overall productivity of these and adjacent for delineating habitats, based on interactions of both habitats. physical and community structure criteria-including salinity regimes, flooding periodicity, substrate type, vegetative life form, and species dominance type. From this classification system, five target habitats have been selected for study: Spartina marshes, Seagrass beds constitute one of the world's most seagrass beds, mangrove forests, unvegetated flats, conspicuous and common coastal habitat types, and and water column systems. These habitat types were contribute a large portion of the total primary pro- selected on the basis of pragmatic criteria, including: ductivity of estuarine ecosystems. These plants are e aerial extent in the Nation's coastal zone; dominant in many areas of the Pacific Northwest, the * present and projected rates of loss and degrada- Bering Sea, the entire coastline from North Carolina tion; to Texas (except South Carolina and Georgia), and * contributions to living marine resources, through Puerto Rico, the U.S. Virgin Islands, and Hawaii. both primary and secondary productivity. 'Food webs refer to the matrix of plants and animals through which energy and materials flow. Ocean System Studies 1 63 Rhizomes and upright leaves of seagrasses allow the trapping and holding of sediments in coastal bays, @Nt@ thereby increasing protection from storm abrasion. Seagrasses form the basis for an abundant and Unvegetated flats are those portions of the bottom diverse assemblage of plants and animals. They ex- of sounds, lagoons, estuaries, and river mouths that hibit high rates of primary productivity, producing lie between high and low water and lack macro- large amounts of organic matter. This organic matter scopic plants. These flats are common in many estu- serves as food for grazing animals such as fishes, arine areas of the U.S. Virgin Islands, and Hawaii. birds, turtles, and manatees, and, as detritus, sup- Unvegetated flats frequently occur adjacent to habi- ports a complex community that includes bacteria, tat types including mangroves, seagrasses, and benthic algae and encrusting invertebrates. Little is Spartina marshes. While an intertidal flat appears known about the natural variability of the functional barren of flora, a host of organisms may be found, relationships of seagrasses or about the trophic link- including micro- and macroalgae, and benthic inver- ages within and between these and adjacent habitats. tebrates (such as oysters, clams, mussels, and fiddler Recent quantification of seagrass habitat losses sug- crabs) that utilize this habitat as feeding grounds gests a linkage with reduced fish productivity and during periods of high tide. Unvegetated flats are an general degradation of estuarine quality. integral part of the estuarine regime, providing food and substrate, especially for burrowing organisms. Mangrove forests don-dnate 75 percent of the world's tropical and subtropical coastline, develop- Estuarine waters exhibit regions of strong hori- ing in low-lying areas where fresh water is supplied zontal and vertical density gradients. These regions to the coast by rivers or terrestrial run-off. Four spe- generally form at the boundaries between water cies occur in Texas, Louisiana, Florida, Puerto Rico, masses of differing temperature and/or salinity. U.S. Virgin Islands, and Hawaii. Mangroves have Horizontal regions of rapid density change are called thickly entangled aerial prop roots that contribute to frontal zones. These may occur both within an estu- shoreline stability by reducing tidal influences and ary and as plumes extending from the mouth of an dissipating wave energy. While mangrove habitats estuary. Vertical regions of rapid density change are produce large amounts of organic matter, their ma- termed pycnoclines. In temperate waters, pycnoclines jor contributions to the functioning of estuarine sys- generally form in spring/summer when the density tems include: the deposition of nutrient-laden mud of surface water is reduced as the result of freshwater and silt; the provision of sites for attachment, breed- runoff and seasonal warming. This buoyant plume ing and nursery areas; and shelter for crustaceans/ overrides colder, more saline water at depth. mollusks and vertebrates. The decomposition of Conditions within estuarine water column habi- leaves maintains a detrital food web believed to sup- tats reflect dynamic processes occurring within and port an extremely diverse community. between the strata, between the water and the over- Realization of the ecological value of mangrove lying atmosphere, and between the water and bot- assemblages is only recent. On a global scale, man- tom sediments. Pycnoclines and frontal zones are of grove habitats are being cleared at alarming rates to special interest because they concentrate particulate provide charcoal and timber, and to create space for materials such as suspended sediments and associ- aquaculture ponds, marinas, housing, and industrial ated contaminants, bacteria, phytoplankton and zoo- development. While mangroves form only a small plankton. Unlike the previous four habitat types dis- part of the estuaries of the United States, they are cussed, the nature of the water column habitat is important ecologically where they occur. Once de- variable over relatively short time scales (e.g., n-dn- stroyed, they recover slowly. utes to hours), as well as over more commonly exam- ined diurnal/tidal, seasonal and annual regimes. Variations are determined by the physics of estuar- ine processes. The functioning of water column habi- tats is of special interest in estuaries that suffer from nutrient over-enrichment. 64 Estuarine Systems: Productivity and Environmental Change pp MENEM= Su orting hypotheses:, Changes in the vol'- ume and/or timing of freshwater and sea water In order to improve abilities to describe and pre- inflows-into estuaries alter the physical structure dict changes in estuarine productivity that result from and. biological- processes within estuarine habitats. natural factors and human-related activities, it is necessary to improve substantially our understand- Estuarine circulation and inflows of freshwater ing of estuarine ecosystem functioning. This can be and sea water exert major influences on estuarine accomplished by testing the following central hy- habitats. Data indicate that changes in the volume pothesis: and/or timing of freshwater and sea water flows into estuaries affect biological processes and biological Central Hypotheses:- Changes in estuarine pro- productivity. Modifications of the timing and flux of ductivity are fImctions@ of hu @hih--indiicedl' aid freshwater and sea water into estuaries are known to natural'alteration5 upon habitat Structure and food affect the structure, location and intensity of density web dynamic- s. fronts. Climatic changes (e.g., seasonal, interannual, decadal) modify the extent and timing of annual Figure I provides a general indication of the rela- freshets. Environmental warming could contribute tionships between habitat substructure4 and func- to prolonged density stratification and increased tion, food web structure and function, and their in- occurrences of hypoxia. Episodic events, such as fluences upon estuarine productivity. For research storms, can create physical changes in estuarine purposes, the central hypothesis has been case into morphology that otherwise would take decades or four supporting hypotheses that deal with specific more to occur. ecosystem functions or structural components. These Redistribution and allocation of freshwater re- four provide focus for specific research thrusts that sources are major problems facing the Nation's estu- are critical in quantifying estuarine productivity and aries. Changes in land use patterns, such as removal man's impact on it. Figure 1 shows conceptual link- of ground cover around estuaries and along upstream ages among these research thrusts. Important research tributaries inevitably affect the quantity, quality, and questions and strategies that help quantify the link- timing of freshwater inflows. Similarly, removal of ages and provide understanding for the supporting freshwater from riverine systems, for irrigation and hypotheses have been identified. Addressing the municipal water supplies, alters salinity distributions supporting hypotheses by specific habitat type will and dilution capabilities of estuaries. Human-induced provide major new insight into human-related modifications of inlets and channels also may alter changes that impact estuarine productivity. Problem the movement of sea water into estuarine areas. At- areas and proposed research strategies for address- tendant with such changes are variations in frontal ing the supporting hypotheses are described in the structure that can affect transport of suspended sedi- text that follows. ments, nutrients and contaminants in the water col- A major problem area not addressed in this pro- umn. Recruitment of many fish and shellfish stocks gram is contan-driants. Both NOAA and EPA have is closely linked to hydrodynamics of estuaries and already established sizable research programs on associated coastal waters, as are the composition and contaminants. The result of this estuarine research well being of estuarine plant communities. Water program will be integrated with the results of the existing contaminant programs through synthesis Figure 1 Relationships among central and supporting hypotheses. efforts. Estuarine Productivity 'It is important to note that habitat type and habitat structure are quite different. Habitat type is a generic term that refers to the dominant compo- nent or ecological unit of an area. For example, Spartina marsh is a recog- nizable habitat type because it is dominated by a single plant species. Habitat Food Web, Other major habitat types includ idows, mangroves, interti- Modification, Modification dal flats, etc. Habitat structure, hoivev@r, refers to the biological, chemical Jj5@Fol: M flc and physical components that comprise a habitat type, and that contribute to and regulate the primary and secondary productivity of the habitat. Fresh Water Nutrient' Inflow Enrichment Ocean System Studies 1 65 management, thus, is integral to effective estuarine management. Among the important questions that Su"p, porting hypotheses: Natural and human- require resolution with regard to this supporting iehibid perturbations alter the transfer of energy hypothesis are: and, !'&'ad io,chang6 in theseconclary productivity � How do changes in freshwater and saltwater of estuarine- habitats,., inflows affect productivity of estuarine habitats? � What are the effects of alterations in freshwater High levels of estuarine finfish and shellfish (sec- and saltwater inflows on the structure of the ondary) productivity appear to be associated with water column, sediment distribution, and dis- high levels of estuarine plant (primary) productivity. solved and particulate loads of nutrients and The latter may be as much as two to five times that of carbon? lakes, streams and waters of the continental shelf. � What is the relative importance of episodic However, the relationship between primary and sec- events, such as storms and floods, in controlling ondary productivity is not understood well enough estuarine productivity? to develop informed management decisions concern- ------- ing fishery and habitat management. Recent declines Supp-orti-ng-h-y-pot-hes-i's-:'-H--u-ma-n,@r@tated,nutr-i- in a number of economically important fisheries have ent loadings to estuarine waters, adver Ise affect,,! been attributed to overfishing, loss or modification IF estuarine habitat productivity, of nursery habitat, or natural cycles. Unfortunately, there is little information to indicate how environ- Nutrient additions are increasing in many estuar- mental management decisions have altered food webs ies because of increased use of inorganic fertilizers and, thus, have impacted productivity in desired and the conversion of wetlands to urban and agricul- species. Previous research has been largely qualita- tural uses. Unfortunately, at present, there is insuffi- tive or descriptive. Emphasis must now be placed on cient understanding of the multiple relationships understanding the dynamics of trophic interactions. between nutrient inputs, nutrient recycling, and eco- Improved understanding of pathways that lead to system productivity to aid substantially in the man- secondary productivity is needed. With this informa- agement of estuaries. Associated problems that have tion, resource managers will be better able to oversee been identified include shifts in the productivity, estuarine-dependent living marine resources that composition and abundance of phytoplankton, and might otherwise be threatened by overfishing and/ reductions in the abundance and extent of desirable or habitat degradation. rooted aquatic plants. Such changes often are associ- The following are among the issues to be addressed ated with reduced dissolved oxygen and reduced with regard to estuarine trophic dynamics: light penetration (which further affects the stability * What are the food web linkages between plant of estuarine habitats). productivity and the productivity of living ma- These ecological changes can be detrimental to the rine resources within estuarine habitats? production of estuarine-dependent organisms 9 What factors determine the relative importance through shifts in food resources and changes in the of food web linkages? What are the implications availability of suitable habitat. At present, it is not of shifts among the pathways to secondary pro- possible to predict in a quantitative fashion the con- ductivity? sequences of such changes to estuarine productivity. e How are primary and secondary productivity in The following are among issues to be addressed with estuarine habitats partitioned among loss to hu- regard to nutrient loading, productivity, and trophic man harvest, loss to neighboring habitats (estu- structure: arine and coastal), or retention and cycling within * What is the relative importance of external nu- the habitat? trient sources and internal nutrient cycling? - How do primary and secondary productivity -,,--,,Supporting hypothesis. Habitat modification vary as functions of nutrient level? JisfOpts,'the physical structure of estuarine envi- * What changes in community composition occur rortinents, loading toreduced capatity for biologi- in conjunction with increased levels of nutri- cal pro uctivitv. ents? 66 Estuarine Systems: Productivity and Environmental Change A variety of estuarine habitats is required for spawning, growth, and survival of estuarine- dependent organisms. The productivity or yield of many species of living marine resources has declined Development of the estuarine research program during the past 25 years, coincident with deteriora- will follow a series of steps outlined in Figure 2. tion and physical loss of estuarine habitats-due pri- important components are: marily to human activities. Correlations of habitat 9 Models' of interrelationships among factors deterioration and declines in productivity can be mediating productivity within target habitats drawn, but there is a lack of well documented cause will be developed or adapted from existing in- and effect relationships between habitat deteriora- formation and analyses. The models will form a tion and loss with loss in productivity of living ma- framework for studying each habitat type, and rine resources. The most critical limiting habitat will provide a conceptual structure for planning factors that affect fisheries productivity include: experiments and testing hypotheses. acreage and configuration of habitat type; preclator- - the nature and complexity of the models will prey relationships; and habitat use by resource spe- depend upon the level of understanding of cies at different life stages. To date, studies have each habitat type; and established what habitat types appear important to - models may range from qualitative, descrip- living marine resources, but the specifics of how these tive formulations to quantitative or semi- habitats are used or which aspects of habitats make predictive constructs. them valuable for growth and production of living - models will be revised as new information is marine resources are poorly known. Without this generated. information, it is not possible to predict quantitative Research on each habitat type will consist of the cumulative effects of habitat modification on pro- following approach: ductivity of living marine resources. In addition, lack - local, multidisciplinary, site-specific field and of fundamental knowledge of the value of different laboratory experiments; habitat types to resource productivity also precludes - mesocosm6 studies to integrate findings from realistic assessment of remedial mitigative action. local habitat/ productivity studies and to test The latter must be judged not only in terms of re- broader hypotheses on habitat functioning; placement of physical habitat, but in terms of re- - large-scale field manipulations (experiments) placement of functions and processes (e.g., utiliza- of estuarine habitats, where feasible and ap- tion by fishes and shellfish) similar to those in unper- propriate; turbed habitats. - follow-up field evaluation studies. The following are examples of issues to be ad- Several iterations of the steps may be required dressed with regard to estuarine habitats: as knowledge about the functioning of target � What are the relative values of estuarine habi- habitats increases and the sophistication of ac- tats to the productivity of living marine re- companying models improves. sources? In general, multi-year studies will be designed to � What are the cumulative effects of habitat loss quantify natural variability and to allow comparison and alteration on populations of living marine of findings for different habitats within estuaries, as resources? well as findings for similar habitats among estuaries. � Are mitigation and enhancement methods vi- Undoubtedly, much of the variability observed for able approaches to ameliorating habitat loss? similar habitats may be related to differences among Does habitat produced as a result of mitigation parent estuaries. In addition, ecological interaction serve the same functions as the undisturbed among target habitats and adjacent habitats will be habitat? considered. Investigations will lead to quantitative understand- 5Ecological models may vary from relatively simple verbal or mathematical ing of relationships between physical and biological d ' tions of the functioning of estuarine habitat types toguantitative factors that affect estuarine productivity. Results will oescrWations. Models wilt form the basis for interpretin di I fcirrnu ef an pre icting be available at all stages of the effort and will perrnit the response of estuarine habitattypes to human-induc and natural en- vironmental alterations. environmental planners and managers to consider 'Mesocosms are artificial environments, often in the form of large aquaria, in which estuarine habitat Tpes may be simulated. Plants and animals are more thoroughly the impacts of alterations to estuar- included, insofar as possib) C. Mesocosm studies enable scientists to vary one or more factors (e.g., temperature, salinity, hydrology) under con- ine ecosystems. r trolled conditions, and, by so doing, test hypotheses refar ing the regula- i tion of habitat productivity. Sin-dlar habitats may be re ated to differences among parent estuaries. Ocean System Studies 1 67 energy. Models of individual habitat types will be formulated early in the investigation; they will be- come more complex and increasingly quantitative as Two important aspects of synthesis for the estuar- research progresses. The goal is to develop quantitative ine science program are: (a) the use of models, at the understandings of habitat types that will permit predic- habitat level, to direct research and integrate find- tions of changes in productivity that result from natural ings; and (b) the development of projections, on na- or human-induced changes in environmental conditions. tional and regional scales, of the impact of human Figure 2 illustrates the progression of research and activities on productivity, through modification and accompanying development of increasingly complex destruction of habitats and food webs. Associated models. with the latter will be efforts to evaluate the accuracy Laboratory and small-scale field studies will be and precision of projections, and to recommend ap- required to refine linkages among elements of the propriate follow-up actions. Models will be used to models. As models become more substantial and the structure and guide the direction of research. At least ability to describe habitat functioning quantitatively one model will be developed for each combination of increases, experimental studies may become more supporting hypothesis and habitat type under study. elaborate. In all cases, however, progress will be made The complexity of the models will reflect the levels of through an iterative process that evaluates findings understanding of the processes being studied. In their and reexan-dnes the accuracy of models in light of most rudimentary forms, models may be conceptual new data. Research at several levels of complexity frameworks for organizing information on processes may be carried out simultaneously. that regulate estuarine productivity. More commonly, The progression of research will culminate in an initial models may take the form of box models or observational program for evaluating the accuracy material budgets such as for carbon, nutrients, or Figure 2 Approaches for testing supportinghyp0theses by habitat. CHOICE OF ESTUARINE HABITATS FOR STUDY Mangrove Unvegetated Density Flat Interface . . ....... K71@@k`@@t-@' @M es u a fb --dfi, Z@ @k ,N Interpretations for ion,p environmental managers, legislators, and the public A L Re-evaluation of hypotheses and models I Large-scale field experiments/manipulations I Re-evaluation of hypotheses and models: refinement of predictive capabilities Comparisons Long-term observations among estuaries ("monitoring") 68 Estuarine Systems: Productivity and Environmental Change of predictions of habitat-associated productivity re- *What is the role of the physical structure in sulting from natural or human-induced changes. determining productivity of a seagrass bed? How Evaluation of predictions is needed to ensure the plastic are rooted aquatics and benthic algae in accuracy of answers to fundamental resource man- filling this role? agement questions such as: *What are the time-scales, rates and magnitudes �What are the relationships between changes in of natural and man-made changes in fresh- and land use and estuarine productivity; saltwater inflow? How do these changes affect �What are the quantitative relationships between secondary production? estuarine primary productivity and productiv- *To what degree is the carbon produced in ity at higher trophic levels; seagrass beds utilized within these habitats? Are �What is the relative importance of major habitat significant fractions sequestered within or trans- types to fish and shellfish productivity; ported from seagrass beds? �What are the implications to fish and shellfish Water column habitat: productivity of alterations in the quality or quan- What is the relative importance of "new" vs. tity of estuarine habitats; and "recycled" nutrients in water column systems? �Can secondary productivity be protected or How do changes in the chemical form and rates improved by managing for specific levels of pri- of external nutrient inputs affect internal nutri- mary productivity? ent cycling? What temporal and spatial lags in- Emphasis will be placed on providing useful fluence the expression of these inputs? information-in useful forms-to managers. The infor- Have man-induced changes in the number and mation will be provided by supporting hypotheses composition of estuarine benthic or pelagic filter- and by habitat. Special care will be taken to ensure feeding faunas affected overall secondary pro- that adequate attention is devoted to synthesis of duction? What is the significance of greatly en- data and information, as well as to research. hanced microbial and microplankton popula- tions in estuarine water column habitats? What is the role of density surfaces (pycnoclines and fronts) in mediating water column secon- dary production? Priority research will focus first on two of the Studies undertaken to answer these questions will habitats described earlier, seagrass beds and water include measurement of: column systems, in a cross-section of estuaries around -time variability of freshwater inflow, including the country. These habitats are extremely important measurements made during storms; because they: *nutrients; �are critical to estuarine transport and productiv- *extent and nature of change of habitats in areas ity; selected for study; and �are essential to the completion of early life his- *linkages of habitats to productivity, as functions tory stages of economically-important species; of time. �occur in most coastal states; and Modeling of efforts will begin immediately to en- �are increasingly threatened by continued urbani- sure that important pieces of needed information are zation and industrialization of the coastal zone. identified early in the program to guide field studies. Research projects will address the four supporting Research will address the nature of habitat changes hypotheses using the approach presented in Figure 2. as well as the rate at which such changes take place. Early efforts will consist primarily of: evaluating ex- This is important for modeling and for controlling isting information; developing conceptual models and mesocosm. investigations. Other important consid- working hypotheses; and initiating field, laboratory, erations. in design efforts relate to comparability of and mesocosm research. A list of potential research studies, particularly measures of productivity. issues for each habitat is provided below. The estuarine research strategy will be imple- Seagrass habitat: mented by NOAA's line organizations and OAR will What are the consequential sources of fixed car- play a major role in conceptualizing and conducting bon in seagrass habitats? In what forms does the research. carbon occur and how is it used? How efficient NOAA's Estuarine Programs Office will coordi- and plastic are the paths linking primary pro- nate the efforts, including distributing funds, devel- duction and production at higher trophic levels? oping a data management plan, and overseeing the Ocean System Studies 1 69 integration and synthesis of information collected by individual studies. Program research and synthesis efforts will be conducted through partnerships of NOAA components and academic institutions. The The proposed estuarine research strategy offers objective will also be to develop strong institution- an opportunity to enhance understanding of proc- to-institution relationships. The private sector also esses that regulate and influence the magnitude and will be encouraged to participate in these partner- nature of estuarine productivity. Emphasis is placed ships. Competitive proposals for research and syn- on the study of target estuarine habitats (eg., Spartina thesis efforts will be solicited. After peer review and communities, seagrass meadows, mangrove stands, evaluation, selected studies will be initiated. A Re- unvegetated flats, and density interfaces). This ap- view Committee will be established, under NOAA's proach complements more traditional studies that Chief Scientist, to provide advice on program efforts. center on comprehensive investigations of single es- The Committee will review the program plan, target tuaries or that address the sources, fates, and effects habitat types, and oversee the peer-review process. of estuarine pollutants. While these sorts of studies The bulk of funding for the first and second year is are well-suited for basin-specific problems or. local to be used for field, laboratory, and mesocosm. stud- environmental management issues, they have not ies, with little funding allocated to administration or provided the quantitative information needed to pre- coordination. The Estuarine Programs Office will dict impacts of man's activities on estuarine produc- coordinate the effort and the line organizations will tivity. implement the program. Consequently, no additional Comparative studies of estuarine habitats form administrative or managerial personnel will be re- the core of the proposed investigations. Estuaries quired. thus will be examined from a different perspective The relative involvement of NOAA line organiza- that will provide considerable insight into processes tions will depend on the focus of efforts and results that are common to estuaries around the country. of the competitive process used to allocate funds. Ultimately, this approach also will provide the frame- The major NOAA participants will be the Office of work for interpreting and integrating findings from Oceanic and Atmospheric Research, the National case-history" studies of estuarine issues. Marine Fisheries Service, and the National Ocean In summary, the estuarine research strategy will Service, each in partnerships with members of the provide a generic understanding of estuarine pro- academic community. The relative balance between ductivity and provide critical knowledge needed to: internal and external research will depend upon the * predict the effects of human activities on estuar- results of the competitive process. ine productivity; Data management will be largely a function of 0 restore and maintain critical habitats; NOAA's National Environmental Satellite, Data, and e restore and enhance fisheries; Information Service. A management system will be - develop appropriate water management strate- established and maintained for the estuarine research gies; and program, in order to provide a central and accessible - develop cost-effective pollution control strate- automated repository for program data and informa- gies. tion. To accomplish this, it will be necessary to de- This new knowledge and its use in management velop a data management plan that includes guid- efforts will result in cost-effective regulation, im- ance on data submission consistent with those of proved permitting, and reduced litigation. The ulti- NOAA and other federal agencies. mate benefit will be enhanced economic develop- ment through wise use and management of estuar- ies. Chapter V1 Marine Chemistry, Biology and Climate Executive Summary Problem and Opportunity climate is now a possibility that must be explored if global climate changes are to be accurately predicted. For many years the industrial nations of the world These new advances onlw compound a problem that ha ve l7een aII0U'ing large atnounts o 'f carbon dioxide -was recognized wears ago. Carboti dioxide is in con- and other "greenhouse oasses" into the atmosphere. tinuous flux, being added to the atmosphere through Recent studies suggest that ozone and the marine sul- processes such as coinbustion of fossil '6tels, chemical fur cwcle as Well as carbon dioxide n7ay have doininant weathering of rocks, and deconiposition of organisms roles in global cliniate change. Unfortunately the ulti- etc. and being fixed, or removed 'froin the atmosphere Inate consequence, global clitnate chattge, can only be through forination of carbonate rocks, incorporation approxiinated. Because of the potentiallil severe into organi-sins, and other processes. Present budget constluences this lack of predictive precision could be models of CO., art, not adequate; in the end, the models disastrous. al-ways shou? a carbon de 'ficit. The largest CO, "sink" Ozone in the atmosphere 'functions as an iinportant appears to be the oceans. Widt, the general areas where oxidant, a precursor to highlij reactive radicals, and as a CO, is absorbed or released by the oceans is known, s(gmficant absorber o 'f ultraviolet and infrared radia- sellsonal exchange cycles, rates, and the transport rates tion. The oceans are n0711 thought to be not only a sink of CO, to the deep -waters throughadvection or biologi- for oZone, but possib(il a storage reservoir (or ozone and cal transport are still poorly understood. The ocean a source of ozone precursors. Because of the absorption niargins are one olf the most biologically productive properties of ozone, pathways fioi- production and de- global areas and therefore importatit fior fixation of struction of this molecule need to lie understood more atmospheric carbon. The ocean nzargi .ns may account OflIV if @Oobal climate changes are to be predicted accu- for,;ome of the deficits seen. Carbon fixed bV inarine or- C rately. The evidence to date irnplicates the ocean sur 'face gant .sms i.n the coastal margins becomes part o 'f the'flux and the rnarine boundatij layer as major contributors to o'f particulates that gradually travel to the continental these processes. Ael 'f to be buried in the sediments. Anthropogenic nu- In yet another way, the oceans may play a major role trient runoff may further augment fixation of CO, in in the U101111latiOl? of world climate. Veiy recent data coastal margins, thus increasing their importance. suggest a link between ocean productivity and climate, 4 fi'A mediated by the marine sulffir Lycle. A 'i1toplankton are a major source of volatile sulfur in the niarine tropo- Research Strategies sphere. Preliminary evidence suggests that fluxes o 'f these compounds are correlated -with precursors to cloud 011111 7VII611 the global chynate models accurately in- fiormations. This niechanism of biological regulation of clude all the dominant uariables can a complete 72 Marine Chemistry, Biology and Climate ,understanding, and hen rticles through sampling re- global climate change be in areas. standing the research in r in the global climate ques- els in each of the areas th Although recentdata imply nitude of ocean sinks and Ozone: The niag s u es a connection with volatile marine sulfur no definite of ozone are thebasis of the maj6rquestions that need to,13,15, causative relationship connects the two phenomena. be addressed -for a more complete, understanding of thej,@@ Research efforts to determine the existence of a causa- contributions of the- ozone f4cet of @elim dfie-, 'research'!U!"f tive relationship and the magnitude of that relationship Research to be conducted,should include, -4H if it exists must, include: Tm n rt, and dis tribidionpf ozone iin ir"opo"s'p here, 41@ o Measurement of the concentrations of relevant I To including,themarine boundary jay@T,,, sulfur,compounds and cloud condensation nuclei, The role @focean, productivity as a source of ozimeASPET plankton speciation and productivity, cloud a]- precursors in, the I'marine-boundary [aye 'r-; bedb, and other important oceanographic and cli- @oneandre-, Th p , mate variables, e- pssi@ility thatthe ocean stores -o,.,, leases it under certain meteo@logic'al, conditions,,11 erm Det' ination. of the physical, chemical, and1bio- thus becoming an ozone source. Av logical variables that control planktonic prod-uc- Carbon dioxide: Weneed to understand the flux of Al"', tibn of-these sulfur compounds, I HP carbon dioxide between the sea and air-particularly as ' ;:@ Identification and quantification of the reactions @40 a function @of oceanic *gion'dndseason, and 6ver'time, IN, rel tin:g these compounds with the formation of as a result of past increases in atmospheric CO , This 1 cloud condensation nuclei, 2' Production of models linking changes in ocean w I help us predic't@ 'change in the future, To fully 4 derstand thes flvxes,w need.the.followin productivity, global albedo, and climate change. gjines.o('@@�@i research: Essential questions of causative polarity and sys- eIn - igati6 tem,stability can, only, be answered through the vest -npffli@,inz@asio@,6fanth@ropogmic,,Iri7c-@;S p ers through the North Pacific thermocline at peri .- J,#, testing of such models. odic intervalsj, Establishment of a,_@eiies of oceanic site's for the V"Y IVOW? ongoing seasonal measurements pf, tTaqr,tran_,@.dlit sient fluxes in.key areas. of the North Pacific from_-il" 'the tropics to the. pdlar-regio hs,- I Sin'ce'the -start of the industrial' revolution, atmo Extensive study of a North Pacific CO sink re- - spheric levels 9f carbon dioxide h' e be n in av e creasing i gion. due to the burning offossil fuels. As the concentrations vi 11 have increas d'"A Additionat important au stions ing to car- of C02- 1 1 -gasses e e _pertain and-other "greenhouse bon flux concentrate on determining the relative im- in the atmosphere, a global warming is predicted. Esti- portance of carbon fixation and subsequent burial of-_01t', mates of CO@ increases suggest that atmospheric levels that -carbon in the regulation of tropospheric levels of are increasing at an accelerating rate A g obal lem- carbon dioxide and the ossible role of anthroppgqnic perofur increase of only one degree Celsius could have -,p 11 nutrient enhancement in accounting for the CO, defi:- profound effects on the world climate. Atmospheric cits encountered, in present carbon dioxide budget"AiM, levels of "greenhouse gasses" are rapidly approaching 1 - "k' models. To answer these questions thefallowing linesaf IW@ 30, the concentration sufficient to cause this increase. With -tKY new avenues research need to be pursued: of research now opened by the elucidation Examination, of existin models toWete'rmihe the of possibly significant climate modulators suchas ozone 9 1-A knowledge needed to, increas e'- their-' cability, [,!-Qt@ and sulfur, the probability of predictin g potential de7 using-r-eseardh i@sulfs'ihe model's predictions and,g1t struction 4s greatly enhanced.. --adequa _' can-be t_es_t_ed,,--' Identification of temporal trends ofjhevariationlq@-A in ocean circulation usinglsynbptic; remotely, Benefits sensed data,, Peter in n the rtan Sful, will m - @i @ g, mipp I .,.,tl,,pr se@, involpedin- 4", The results of these investigqtions, if succes CO fixation,,particle development@, settling, -and @j@im, allow, rediction- of global climate change cau d b p se y V Ocean System Studies 1 73 disturbances in the biogeochemical flu.Yes in- theworld be significantly iff@ctedby even slight global warming. ceans. In additi IoIn t -e to e-ojel'- 11, ih@ change 0 1@ h4mdtf@ induced-,hi-, ese s in, global climate can be predictable, creases in co 2 concentrations,can he +etter@ assesse&,@,,,, then @ preventative measures or informed preparations ma World food production, supplond demandfqr enefgy-;1, maybe defominimize thepotentia ly drastic effects. all ecosystems, water resources,,and human health can, u2W- Wit' �,ali MENNEN= by the photolysis of ozone with water. In addition, ozone is a significant absorber of ultraviolet and in- In the last decade of the twentieth century, a para- frared radiation. The earliest view of tropospheric mount concern will be man-induced changes to our ozone assumed stratospheric injection and surface global climate. Research aimed at understanding, destruction with no intervening chemistry. A great detecting, and predicting the causes and the conse- many observations support these basic tenets of quences of climate change will engage much of stratospheric injection and surface destruction. NOAA's attention and resources. Opportunities ex- However, it has now been demonstrated 'that a ist for NOAA's marine chemists and biologists, both great deal of chemical production and destruction of in the Environmental Research Laboratories and in ozone occurs in the troposphere, the rates of which NOAA's extramural programs, to contribute through sometime exceed the estimated transport fluxes. A their research to the world's efforts to cope with combination of transport theory and chemical theory climate change. predicts ozone production in the upper troposphere Three areas of research are likely to engage and chemical destruction in the lower troposphere. NOAA's marine scientists as they address climate Geophysical Fluid Dynamics Laboratory (GFDL) sci- questions. The role that oceans play in the ozone entists have refined the transport model with a chem- cycle is poorly understood, yet understanding this istry approach in an effort to understand the role of role is essential if intelligent decisions are to be made transport in determining ozone distributions in the on the regulation of ozone-destroying compounds. troposphere. In their model, the sources of ozone are Within the carbon cycle, a good portion of man's exchanged across the tropopause with chemical pro- fossil fuel emissions cannot be accounted for in either duction occurring only in the upper troposphere. the atmosphere or in the oceans as inorganic carbon. The results of their model significantly disagree with A better understanding of the circulation of the oceans some observations. There is an apparent need for an and the degree to which the carbon cycle is affected additional 0 3 loss mechanism (in excess of the loss by ocean biology is needed if knowledge of the cycle mechanisms already in the model) "corresponding is to be complete. Finally, recent research suggests to an equivalent desposition velocity of around 0.1 the marine sulfur sources may contribute significantly cm/s" in the boundary layer of the tropical and sub- to climate modulation by controlling open ocean tropical Pacific. The model also has high gradients of cloud formation. The mechanisms that underlie this 03 mixing ratios and low surface 0 3levels over land. phenomenon need to be understood if the impor- In many cases the landsurface is a source of 03 and tance of marine sulfur emissions in regulating global not a sink, whereas the ocean surface is still assumed climate is to be assessed. to be a sink. Current research by AOML scientists has focused on determining the magnitude of the oceanic sink for ozone. During a recent research cruise in the tropical Atlantic, AOML scientists observed an atmospheric ozone increase corresponding to an increase in ma rine biogenic non-methane hydrocarbons. Because of Ozone has three important functions in the atmo- the reactivity of these hydrocarbons, their source sphere. It is an important oxidant and the precursor region was probably within a few hundred kilome- for highly reactive radicals, particularly hydroxyl ters. Within that spatial envelope, an increase in bio- which is formed by the reaction of O(M), produced logical productivity was observed. The data suggest ,m 7n i g ctable, 74 Marine Chemistry, Biology and Climate a possible, but untested, cause and effect relationship between the biological productivity of the oceans ENININEMENNEENIM and the formation of ozone precursors. Production of ozone in the marine boundary layer To determine the answers to the first two research is most likely to be through a complex series of pho- questions, the magnitude of oceanic sinks and sources tolytic and free radical reactions occurring in surface of ozone, research requirements are the following: seawater and/or in lower portions of the marine To quantify the transfer rate of ozone from the boundary layer, involving biogenic hydrocarbons free troposphere to the marine boundary layer ranging from one to four carbons in size. Photo- by studying ozone fluxes at the free troposphere/ chemical oxidation of these materials results not only boundary layer interface and on the downward in production of carbon monoxide (CO) and ozone, side of convective circulation cells, to determine but also aldehydes, ketones, organic and, peroxyocyl the scale over which this transfer rate is appli- nitrates, and aerosols. Methane is three orders of cable, and to investigate whether turbulent proc- magnitude more abundant than the other hydro- esses and/or convective activity play a signifi- carbons in the atmosphere, but it is one to three cant role in this transfer. orders of magnitude less reactive than other hydro- * To study under what conditions advective proc- carbons with OH radicals. esses dominate the distribution of a chen-dcal In addition to ozone precursors, such as methane species like ozone in the boundary layer. and the non-methane hydrocarbons, the generally * To compare the strictly chemical destruction rate accepted mechanisms for ozone production require of ozone with the difference between the photo- the presence of NO. or organic nitrates in sufficient chemical production and photochemical/chemi- concentration. Measurements of NO x in the remote cal loss rate that occurs during the day. Pacific indicate that the concentrations there are not There are two ways to arrive at an understanding high enough to support ozone production. However, of the magnitude of the oceanic sink. The first is to measurements of NO mixing ratios in air masses on experimentally isolate a column of air from the sea the Irish Coast and at Bermuda indicate that there surface to the top of boundary layer (over the tropi- should be sufficient NO xto catalyze ozone produc- cal ocean that is assumed to be the trade wind inver- tion over the North Atlantic. sion) and then measure the change in ozone in that column. The second way is to compute a columnar budget, i.e., quantify the fluxes as well as the steady NNEENNOMMENNEM state concentration. Neither of these is totally experi- mentally feasible at this time. The budget approach ErMEMINSEEMENSM is the one that can be best addressed at this time. This process can potentially be investigated OAR marine biogeochemical research on the ozone through shipboard observations because it appears cycle is an integral part of the Radiatively Important to occur below 30 m in altitude. Ship observations, Trace Species (RITS) program of NOAA. The central however, cannot evaluate the other, supposedly larger marine question to be answered is what is the magnitude source of ozone in the boundary layer, namely the of the oceanic sink for ozone? free troposphere. However, aircraft such as the heav- Key questions that must be answered if this cen- ily instrumented NOAA WD-P3 can, because of sen- tral question is to be resolved include: sors investigating vertical winds and turbulence, al- �What are the spatial and temporal distributions low investigations of turbulent and convective trans- of ozone sources in the marine troposhere, in- fer of ozone from the free troposphere to the bound- cluding the boundary layer? ary layer. � What role does the marine production of low A comprehensive research program to determine molecular weight hydrocarbons play in the pro- the role of marine-derived low molecular weight duction of ozone in the marine boundary layer? hydrocarbons in the ozone cycle will require: � What is the source of the nitrogen compounds e The measurement of low molecular weight required for ozone production in the marine hydrocarbons in the marine boundary layer and boundary layer? the water column in selected oceanic regions. � Can the ocean store ozone as ozonide and re- 9 Determination of the pathways and mechanisms lease ozone under appropriate meteorlogical con- by which low molecular weight hydrocarbons ditions? Ocean System Studies 1 75 are produced and transformed in the marine sink for the missing carbon appears to be the oceans. environment. Present estimates based on various box-diffusion This research program will require a close work- models indicate that anywhere from 25 to 45% of the ing relationship between physical, chemical, and fossil-fuel-derived carbon resides in the oceans. The biological oceanographers. A comprehensive pro- major causes of the present uncertainty in these esti- gram would consist of: mates are the insufficient knowledge of physical and � A field program of research cruises that will biogeochernical processes involving carbon in the undertake transects; across regions of high and oceans and inadequate treatment of these processes low biological productivity. Upwelling zones in ocean transport models. Improved accuracy in the and the high latitude regions of the Atlantic and estimates of the oceanic carbon uptake will undoubt- Pacific oceans at their peak times of productiv- edly rely on an improved understanding of essential ity are of particular interest. A detailed prograrn physical and biogeochernical processes involving of chemical analysis and biological productivity carbon and the development of ocean carbon models measurements will be undertaken in the water based upon more realistic general circulation mod- column and the lower boundary layer as appro- els. priate. �A laboratory-based biological sources prograrn that will be continued and expanded. Biochemi- cal and biological experiments will be under- taken to determine the biological, chemical, and Carbon in surface w5ters is fixed by photosyn- physical factors which control the production thetic processes to form organic particulates which and fluxes of low molecular weight hydrocar- sink into deeper waters. By far the largest fraction of bons from the water column to the boundary this carbon is remineralized in the deep waters or on layer. the seafloor. Estimates of the magnitude of this flux In current research at AOML, ozone is continually to deep water range from 2 to 6 gigatons of carbon recovered at low levels when sea water samples are per year. In addition to the large uncertainty in the stripped with inert gases. It has been hypothesized marine biological carbon flux, the geographic distri- that the source of this ozone is the decomposition of bution of the flux is still poorly understood. The ozonide present in the water samples, and that the contributions of the central gyres, the polar regions, ocean may be serving as a storage site for ozone in and the ocean margins have yet to be resolved. the form of ozonide. If confirmed by additional stud- Of particular interest is the role played by the ies, this phenomenon may represent an important ocean margins (the regions between the coast line new variable in the marine ozone cycle. and the ocean boundary currents) in the carbon cycle. Necessary research activities needed to evaluate The margins are among the most productive areas of the significance of these preliminary laboratory ob- the ocean and are the regions where the stimulation servations include: of biological productivity by man's activities is most �Development of analytical techniques and meth- likely to have occurred. ods for the direct measurement of ozonides in Shelf regions represent the areas where very high sea water. rates of oceanic carbon fixation occur. As carbon fixa- �Retrospective analysis of GMCC surface ozone tion is dependent on the supply of nutrients, this and meteorlogical data to assess the importance results from transport of "new" nitrogen supplies (as of the oceans as a direct source of ozone. opposed to nitrogen that is being recycled on a time scale of days) to the margin euphotic zone by various upwelling phenomena, onshore advection of deeper nitrogen-rich water, and from continental runoff. Measurements of carbon metabolism, production and Estimates of the global carbon budget indicate exchange along food webs indicate that large frac- that up to 47% of the fossil-fuel derived carbon can- tions of organic matter produced on continental not be ascribed to increases in the atmosphere. While shelves is exported to continental slopes and to the some of the debate about the missing carbon is re- deep ocean. This transport occurs as fine particle flux lated to uncertainties in the estimate of the amount of which is in turn modulated through processing of CO 2released from terrestrial land biosphere, the major the fixed carbon into larger particles, i.e., grazing of 76 Marine Chemistry, Biology and Climate phytoplankton by zooplankton and the concomitant missing carbon in unbalanced chemical and biologi- metabolism and "packaging" into fecal pellets cal budgets for global carbon dioxide, i.e., 0.7 to 1.6 particles and/or aggregates. gigatons of carbon per year. Recent evidence indicates that this transport may be significantly enhanced through filaments of high pigment (carbon rich coastal waters which extend into open ocean currents, e.g., the California Current, the Gulf Loop Intrusion, or by mesoscale eddies such as warm core eddies formed by the Gulf Stream along the Eastern shelf of North America). However, Two key research questions to be addressed if the a significant portion of shelf carbon may be "trapped" carbon cycle is to be fully understood: on continental slopes. This is especially possible for 9 Is carbon fixation in the ecosystems of the ocean that portion of carbon fixed via new nitrogen from margins and subsequent burial in shelf sedi- continental runoff since this carbon would be fixed ments an important factor in the regulation of on the inner shelf and require a longer transport time global tropospheric levels of carbon dioxide? time to the shelf edge, thus giving time for grazing a Does the additional carbon fixed due to anthro- and other processes to increase particle sizes and pogenic nutrient inputs account for the "miss- form aggregates that would settle on the slope. ing" carbon in the unbalanced (chemical and Supplies of essential nutrients have increased biological) carbon dioxide budgets? markedly in coastal runoff since the advent of the industrial revolution. This anthropogenically driven nutrient enhancement may well be responsible for enhanced productivity in the ocean margins and in- creased export of carbon to continental slopes. This transport and subsequent burial and diagenesis in A multidisciplinary OAR research program will slope deposition centers could represent "missing" be developed to identify and quantify the important carbon in global carbon dioxide budgets. As an ex- physical, chemical, and biological processes so as to ample of the enhancement of coastal ecosystem nu- allow their inclusion in models of global ocean flux trients, an increased nitrogen transport in the Missis- and climate. sippi River system during the period 1937 to 1980 Through enhanced monitoring of nutrient flux to has been documented. There is also evidence of in- the coastal ocean and coastal productivity (both in creased coastal productivity resulting from this flux situ and via remote sensing), much can be done to as shown by the documentation of seasonal hypoxia provide an adequate data base, in space and time, to associated with high chlorophyll levels down cur- allow assessment of potential carbon export from rent from the Mississippi Outflow. Similar enhance- ocean margins. However, specific, intense studies of ments of coastal productivity may be occurring in carbon fixation and the subsequent food web proc- areas such as the New York Bight, the Chesapeake esses which package and metabolize this carbon must Bay and flow plume, and the Rhine (as evidenced by also be conducted synoptically with physical studies high pigment levels in the southern North Sea), etc. of transport, sediment dynamics, and benthic proc- However, the lack of adequate spatial data on deca- esses. This will allow assessment of the amount and dal time scales for any such systems prevents quanti- rate of transport as well as of the eventual fate of the fication of the extent of enhanced carbon fixation or fixed carbon. It is also reasonable that shelf depos- export. It is clear that no similar enhancement of tion centers of particulate carbon should be evalu- productivity has occurred in the open ocean and that ated in terms of their existence and extent and whether analyses of abiotic storage of carbon dioxide (vertical or not such storage is temporary (steady state) with mixing, polar sinking) cannot account for all of the an eventual transport to deep ocean sediments, or carbon dioxide en-dtted by fossil fuel burning. An- permanent. The following types of studies are neces- thropogenic nitrogen loading to the coastal ocean sary to supply our understanding of the processing would allow export of 1.5 gigatons deposition cen- of carbon in the ocean margins. The proposed pro- ters of carbon per year, via fine grained sediments, gram can be considered to have four components. from the shelf to deposition centers on the upper A modeling component which will serve to guide slopes. If half of this amount represents increased the other program components toward a predic- loading since the industrial revolution it accounts for tive. capability with an ultimate product of Ocean Systein Studies 1 77 enhanced global climate models in which the ing the approach of our atmosphere to radiative equi- ocean modulation of tropospheric C02 is prop- librium with its changing trace gas composition. Thus, erly parameterized. the role of the ocean is crucial in setting the warming A time series component which will generate rate, even though the ultimate equilibrium tempera- data on temporal trends in the productivity of ture of the global system will be determined by the ocean margins and nutrient inputs to nearshore atmosphere. waters which is reasonably synoptic with data While atmospheric C02 time-series began during on variations in ocean circulation. Circulation the International Geophysical Year (IGY), and atmo- data coming from other global research efforts, spheric measurements of trends and distributions of e.g., WOCE. other greenhouse gases have been initiated during An experimental component conducted in se- the past decade, corresponding time-series and dis- lected ocean margin areas to elucidate the im- tributions of CO 2and other chmate-linked trace gases portant processes involved in ocean margin C02 in the sea are very sparse or nonexistent. Excepting fixation such as particle development, settling, some pioneering work on tritium in the North Atlan- and transport so that these processes can be tic, ocean model interpretations of chemical tracer quantitatively included in the models. fields in the thermocline are completely lacking, A remote sensing component which will allow though such studies clearly offer the best hope of a real application of the results of the other three defining the timeframe for the predicted global warm- components and generalization of these results ing. The transient is occurring now, so the need for to global models. the measurement of the evolving chemical tracer fields Interagency and international aspects: An OAR in the upper ocean tnd their use in improved ocean- Ocean Margin Global Ocean Flux Study will be a atmosphere models is an opportunity that should be integral part of the U.S. Global Ocean Flux Study. grasped. The ocean basin portion of GOFS is being organized Fossil-fuel derived CO 2 enters the ocean surface and sponsored by the National Science Foundation via gas exchange processes and is ventilated down- Ocean Margin GOFS; this would also be an impor- ward by n-dxing and thermohaline circulation. Deep tant part of the international J-GOFS program, or- convective mixing of the water column takes place in ganized to bring together the resources and abilities winter in the polar and subpolar regions which pro- of the international community in addressing the vide transport pathways for advection of heat, CO, role of ocean biology in the carbon dioxide cycle and and other tracers into intermediate and deep water climate regulation. masses. The newly formed water masses spread lat- erally along isopycnal surfaces. By coupling high precision and accurate CO, measurements with chemical tracers, such as chlorofluorocarbons, C and 14 3H, it will be possible to improve our understanding of the role of intermediate and deep water mass While the potential for increasing atmospheric formation and related processes in C02 uptake, and levels of CO 2from combustion of fossil fuels to cause for constraining future general circulation models. a global warming has been recognized for some time, The seasonal variations of carbon-nutrient rela- only very recently has the comparable climatic sig- tionships and tracer chen-dstry in the oceans are poorly nificance of other greenhouse gases been realized. known. Recent studies by PMEL scientists indicate The new NOAA initiative RITS specifically addresses that wintertime gas exchange and mixing processes only the atmospheric measurement of the global play a major role in the transfer of CO, and chloro- trends and distributions of these other climate-linked fluorocarbons in the subpolar regions of the North gases. Very recent work suggests that the limiting Pacific. Corresponding information for other oceanic uncertainty in predicting the warming scenario for regions is very sparse. However it is known that our planet during the next decades to century may intra- and inter-annual variations (i.e., El Nino) can not be in the uncertain growth rates of the green- be very large, especially near the equatorial zone house gases in the atmosphere, but rather in the very where the oceans are a major source to the atmo- poorly understood circulation processes in the upper sphere. With the development of general ocean cir- ocean, where the heating transient is being tempo- culation models which include C02 uptake and re- rarily stored. This ocean thermal lag is greatly delay- lease, there will be a major requirement to include 78 Marine Chemistry, Biology and Climate seasonal variations of CO 2and other tracers to verify shots will be taken at five-year intervals. The model simulations of seasonal changes in CO 2 sources survey will be conducted with such a spatial and sinks. coverage that all the major features of the CO 2 Air-sea exchange of C02 across the air-sea inter- distribution are adequately contourable and that face is probably the major rate limiting step for C02 the resulting chemical snapshot serves as a ref- uptake in the oceans. The strong and non-linear de- erence state against which future changes in pendence of gas exchange on wind speed suggest circulation and/or chemical composition might that, if the physical parameters influencing gas ex- be reliably detected. change are well characterized, gas exchange rates The establishment of ocean carbon time-series may be predicted with good accuracy from satellite sites in the North Pacific Ocean to follow sea- scatterometers. Experimental studies of C02 and other sonal and interannual variablility in CO, fluxes trace gas exchange rates under varying open-ocean will be established. These sites will cover the windspeed cinditions are needed to provide a more polar, semitropical gyre, and upwelling regions realistic estimate of regional gas exchange rates. of the North Pacific, thus including waters which are major sources and sinks of C02' An international, focused study of the C02/ freon sink region in the northwest Pacific off Kamchatka in cooperaton with WOCE. In addi- tion to the above, a modeling component will be required to guide the optimal sampling of the Major questions that need to be answered are: marine carbon system and thereafter to inter- �What is the flux of carbon dioxide between the pret the resulting measured CO 2 distributions in sea and the air as a function of oceanic region terms of present and future C02 uptake capacity and seasons? of the sea. � In what way has the natural distribution of re- Interagency and international aspects: An OAR gional CO 2sources and sinks already been al- tracer field study will be closely linked to WOCE, the tered by the 25% increase in atmospheric CO, World Ocean Circulation Experiment, to the Global during the past one-hundred fifty years? Ocean Circulation Experiment,to the Global Ocean �How will the present regional distribution of Flux Study, and to evolving programs such as Global air-sea CO 2fluxes change in the next century in Tropospheric Chemistry. Much of the fieldwork will response to much higher atmospheric levels and be conducted in phase with WOCE. climatically induced ocean/ atmospheric C02 levels and climatically induced ocean/atmo- spheric circulation changes? NONE= The Gaia hypothesis, as proposed by Lovelock To address the research questions listed above and Margulis in the early 1970's, states that life has will require both a field program of experimental regulated and stabilized the environment of the earth, measurements and a modeling program. Advances keeping it within narrow bounds that allows the in analytical methodology, especially in the preci- continuation of life. This hypothesis considers the sion of sea water inorganic carbon measurements, lower atmosphere as part of life itself, and argues now hold out the possibility that the invasion of that plant and animal populations regulate certain fossil fuel CO 2 will be detectable over the coming aspects of this atmosphere, including temperature years. The measurement of C02 and dissolved inor- and chemical composition. ganic carbon, along with the freon tracers, is then an This hypothesis is controversial and has not won essential part of the observational program. The field widespread acceptance, partially due to the inability program will consist of these major activities: to test examples raised in its support. Now, however, Chemical snapshots of the invasion of anthro- in the marine sulfur cycle, there appears to be a pogenic climate gases (C021 freon) throughout degree of climate regulation which is consistent with the North Pacific thermocline from the Aleu- the Gaia hypothesis and is also testable. The year tians to the equatorial upwelling zone. The snap- 1987 saw significant amount of evidence published Ocean System Studies 1 79 in support of the marine sulfur cycle providing a mechanism for a link between oceanic productivity and climate. Among the facts and calculations which have appeared in the recent literature are the follow- ing: 77EM � The major source of volatile sulfur in the remote The key research question that must be addressed marine troposphere is from the production of is what, if any, role does the volatile marine sulfur dimethylsulfide (DMS) by marine phytoplank- cycle play in the regulation of global climate? Impor- ton. tant subsidiary questions include: � There may be a possible direct correlation, in the e What are the controls on distribution of DMS remote marine atmosphere, between the rate of producing plankton species? DMS flux and the number of cloud condensa- - What are the important factors that control DMS tion nuclei. emission? � There is an apparent direct correlation between * What is the relationship between atmospheric the amount of daily solar radiation and the rate DMS and CCN in the remote marine atmo- of DMS flux from the ocean surface. sphere? � There is only a very weak correlation between e What impact does changes in CCN in remote conventional measures of primary productivity marine regions have on cloud cover and global and DMS flux rates. albeclo? From the above information, a new hypothesis * What role is played by anthropogenic sulfur has been postulated: that the natural biogeochen-dcal emissions? cycle of sulfur in the marine environment constitutes a negative feedback loop that may modulate global @7 warming. The hypothesis essentially states that the major source of cloud condensation nuclei (CCN) over the oceans is non-sea salt sulfate aerosol. This The proposed program will have the following aerosol is an oxidation product of planktonically elements: produced DMS. Because the reflectance (albedo) of A field component which will, through ship clouds (and thus, the earth's radiation budget) is cruises and aircraft overflights, measure key sensitive to CCN density, biological regulation of sulfur species and CCN concentrations in the climate is possible through the effects of variations in marine troposphere, plankton speciation and solar insulation at the sea surface and through the productivity, cloud albedo, and other important physical oceanographic forcing functions acting on oceanographic and climate variables. Measure- the planktonic production of DMS. ment will be made on a seasonal basis from the There is a very poor understanding at present of tropics to the polar regions to define the impact the hypothesized marine sulfur cycle. While nega- of the volatile marine sulfur cycle on climate. tive feedback has been postulated as operating within A laboratory component to determine the physi- the current global climate regime, evidence from cal, chemical, and biological variables which 20,000-year old antarctic ice suggests a positive feed- control DMS production by marine phytoplank- back loop may have operated during the past ice age. ton. This ice contained a higher than average number of * A laboratory component to identify and quan- particles derived from DMS, with possibly concomi- tify the reactions which relate DMS to CCN. tant higher numbers of CCN, higher cloud albedo, - A modeling component to link changes in ocean and a resultant drop in temperature. It is not at all productivity, global albedo, and climate change. clear what the impact of global warming by C02 Essential questions of the polarity of feedback emissions will be on the volatile marine sulfur cycle and system stability can only be answered and subsequent marine cloud formation. through the development and testing of predic- The volatile marine sulfur cycle, then, may play an tive models. important, but as yet, poorly defined role in climate regulation. Understanding global climate requires that the volatile marine sulfur cycle and its climate regulatory role be thoroughly studied. so Marine Chemistry, Biology and Climate We have a much poorer understanding of the chemical interactions between ocean and atmosphere. The tools to develop such an understanding are avail- If we are to have any hope of reasonabaly predict- able or are being rapidly developed at the present ing the directions, rates, and consequences of global time. There is a read-y opportunity for NOAA, through climate change, we must understand the critical in- the programs descr ibed above, to take the lead in teractions of the oceans and the atmosphere. Physical addressing major research problems in understand- oceanographers have been conducting extensive re- ing the influences of the carbon, ozone, and sulfur search programs on this subject in recent years and, cycles and how these cycles impact, in turn, affect as a consequence, are making major progress in our global climate. The programs described above understanding such phenomena as the periodic El are unique but are important portions of multiagency, Ninos. Circulation and climate modelers, utilizing international efforts in global marine chemistry as these research results, are beginning to develop they relate to global cycling and change. They are models with realistic chances of simulating these appropriate for NOAA to undertake and consistant events as first steps towards predictions of climate with NOAA's role and mission. variation and change. lop Ive AL AW 1w. 4S, V -4- A fs. 36 68 14101 2205