5.1 Review of the Status of the Science Plan
The US JGOFS Long Range Plan (US JGOFS, 1990) laid out the scientific justification for process studies. The overall goal of the process study component of JGOFS is to provide insight into key biogeochemical ocean processes, leading to mechanistic understanding which can be expressed in equations and coded in a hierarchical selection of models. A detailed inventory of process study objectives and elements is given in the International JGOFS Science Plan (SCOR, 1990). Biogeochemical processes themselves are generic and common to all the major biogeochemical provinces of the global ocean, but the relative importance, and regulation of processes differs among basins and provinces. In the Long Range Plan, 13 process studies in 11 oceanic provinces were proposed. Following the completion and assessment of the Pilot Study in 1989, and a further assessment of resources available for JGOFS, the list of 13 major studies has been reduced to a group of 4 high priority process studies. Provision has also been made for an unspecified number of smaller, focused, "miniprocess" studies. The four major regional process studies are described following a brief description of the Pilot Study.
5.2 Description of a Process Study
Process Studies fill an observational gap not covered explicitly addressed by Large Scale Surveys and Time Series operations. They are conceived to address fundamental problems of carbon and associated fluxes in strategically-chosen ocean basins on scales at once more focused in time than surveys and wider ranging spatially than time series stations. The scientific designs of different studies will take somewhat different forms depending on the specific objectives and logistic demands in each region, but the generic model of a centrally-coordinated, international collaborative program of observations and synthesis will be followed in each case.
The model for Process Studies in JGOFS is the Pilot Study, or North Atlantic Bloom Experiment (NABE), which took place in 1989-91. NABE was coordinated by the JGOFS Scientific Steering Committee through a Pilot Study Coordinating Committee made up of the national coordinators of the participating nations. The study itself was international, multidisciplinary and broadly collaborative. Research vessels from five nations (Canada, The Federal Republic of Germany, The Netherlands, The United Kingdom and The United States), a NASA aircraft and satellites for remote sensing, and moorings were deployed for this study of the spring phytoplankton bloom (Ducklow and Harris, 1992). In the USA, NABE consisted essentially of 17 projects with 25 principal investigators, at a cost of less than $10M. The results of NABE were reported at a Scientific Symposium held at the National Academy of Sciences in November 1990, and a volume of scientific papers was published in 1992 (Ducklow and Harris, 1992).
The first full scale process study was the Equatorial Pacific Process Study (EqPac), which completed its fieldwork in early 1993. Its composition, size and cost are detailed below in Section 5.6.1.
5.3 Guidance and Planning
Since the initiation of the NABE Pilot Study, JGOFS has set up a number of mechanisms to ensure efficient and proper planning, execution and synthesis of process studies. The various groups and activities are described briefly below, and the individual elements for each study are given in the sections on each study.
The US JGOFS SSC has ultimate responsibility for the design, implementation, execution and synthesis of each Process Study. The composition (PI's) and size (cost) of the studies are determined by the review and budget process in the funding agencies. While most of the tactical and scientific decisions regarding the operational phases of the studies will be made by the planning and coordinating groups and even by individual scientists, the SSC, through overall guidance of the planning process should decide the priority questions, form planning and coordinating committees, coordinate observational elements and set the timetables for each study. The SSC has taken several important steps toward implementing the process studies: i) initiated the planning process for the Equatorial Pacific, Arabian Sea, Southern Ocean and North Atlantic studies; and ii) requested informal proposals for additional process studies in different ocean provinces.
Following the precedent established during the Pilot Study, Regional Planning Groups or Process Study Coordinating Committees have been formed to define the scientific rationale and plan and guide each study. Each groups is composed of scientists with particular expertise in the geographic region, chosen to provide disciplinary balance, with a Chair or co-Chairs recognized by the SSC. The composition of each of the planning groups is listed in the appropriate section below. The planning groups are responsible, among other charges, for convening a series of planning meetings as defined below.
The Leaders of the Regional Planning Groups are members of the extended Executive Committee of US JGOFS. The "Exec-plus" is be responsible for coordinating and guiding the process studies, establishing a liaison between the SSC and the regional groups, and among the groups, and organizing synthesis and modeling activities for the process study component of JGOFS.
Much of the planning activity for Process Studies is carried out in a series of meetings following the Pilot Study model. The time frame and sequence of the meetings and the implementation of the field study defines a time line for implementation of each process study which lasts approximately 4-6 years. Time lines for each process study are included in the individual sections below.
Timeline for implementation of a major process study.
5.4 Resource Spectrum
Process studies require a large and diverse array of observing systems and analytical resources to meet their goals. The major types of resources are listed here, and some of the details of their deployment are given in the following sections on the individual studies. Process studies are neither as extensive in time as time series, nor in space as surveys. Observational resources need to be chosen to facilitate extrapolation and generalization of process study results. Mooring data can place the discrete observations from individual cruises in a temporal context. Aircraft observations will do the same for the spatial context. Satellite data, after being carefully compared to the multidisciplinary cruise data and mooring data, will be used to extrapolate to interannual time-scales and basin scale space scales. All of these data combined with the models gives us a synthesis of present-day response to climate variability that can be used to study past and future characteristics of the carbon cycle of the basin, and to apply the insight to other regions of the world.
Research vessels. Cruise data will provide the basis for gaining a mechanistic understanding of how various interdependent processes control the carbon and nitrogen cycles in study regions. JGOFS process studies are focused primarily on open ocean problems. Following the Pilot Study model, a minimum of ca. 500 days of shiptime is required on ca. 5 vessels per year of internationally-coordinated field study. The US JGOFS Program alone requires at least one dedicated large research vessel for a period of about one year. To accommodate the scientists and technicians required to address the Core Measurements (see below), large vessels with 40--50 day endurance capable of housing 20--40 scientists are needed. Smaller (coastal) research vessels may also be needed for ancillary investigations, mooring deployment/recovery, etc.
Satellites. In addition to their primary role in data collection for the Global Synthesis (Chapter 3), remote sensing satellites are needed for real-time guidance of sampling operations and interpretation of shipboard operations. Satellite data, after careful comparison with multidisciplinary cruise data and mooring data, will be used to extrapolate to interannual time scales. Thus provision of ground stations or shipboard receiving and data processing capabilities during field operations is essential for JGOFS. The satellite-borne sensors of greatest utility are ocean color instruments, sea surface temperature radiometers and altimeters (e.g., Robinson et al., 1991).
Aircraft. Satellite observations are subject to interference from clouds. Remote sensing aircraft equipped with LIDAR's and radiometers for SST and ocean color provide high resolution viewing of process study areas and are less vulnerable to cloud cover interference. The NASA P-3 equipped with LIDAR was used in NABE and its successor was flown in EQPAC.
Moorings. Ships can make observations of limited temporal continuity and satellites only view the ocean surface. For more comprehensive temporal coverage, depth-dependent observations and relative independence from weather, a variety of moorings have been and will be used in JGOFS. Moorings will provide data on the variation of properties both during process measurements and at times when process measurements are not being made. The daily process observations can be extrapolated to wide areas and to monthly time scales using the mooring data. These include LOTUS-type and other upper ocean physical/meteorological moorings and bio-optical moorings (Dickey, 1991) in addition to deep ocean sediment trap moorings (Deuser, 1986). Chemical moorings capable of sensing dissolved gases and nutrients are under development and will be used in later process studies.
5.5 Core Measurements
One of the most important aspects of the JGOFS observational strategy is the development of protocols for a list of Core Measurements. A preliminary version of these protocols was developed for the Pilot Study (JGOFS Report 6). More detailed sets of protocol descriptions were developed for the Time Series sites (described in the time series data reports and at their Web sites). A record of the methods (core and additional) employed in EQPAC is also available (JGOFS, 1995). A comprehensive set of protocols for the International JGOFS Program, written in a uniform format, has been published (JGOFS, 1995). The list of core measurements was developed in consultation with modelers and addresses the variables and rate processes required to provide adequate descriptions of biogeochemical processes and their hydrographic background for development of biogeochemical process models. An important goal of JGOFS is to build the capability to perform each measurement to JGOFS standards following the accepted protocol, on each process study cruise. The list of JGOFS Core Measurements includes the following parameters which should be measured in most phases of each process study:
The results of the process studies form a large and diverse array of products, including data sets generated by individual PI's and technical service units, research papers, and new models. Proper synthesis of these products requires a planned sequence of activities to ensure optimal use by the program and future scientists. Process study synthesis begins with interaction between the process study coordinators, PI's and data managers to facilitate data submission and access. Several scientific meetings and at least one large data workshop are usually needed to familiarize project PI's with the initial and individual results of component projects in a process study. Other smaller, more focused workshops devoted to special topics may also be required. Following these data presentations, a large synthesis workshop is convened to promote interaction among PI's and initiation of collaborative and interdisciplinary interpretation of data. The data and synthesis workshops for EQPAC were held 1 and 2 years following the completion of the fieldwork, and each lasted for 5-10 days and a university or conference center.
5.7 The Process Studies
5.7.1 Equatorial Pacific Process Study (EQPAC)
Time Frame: (Field Studies): 1991-1993
Planning Meetings, Reports and Workshops:
5.7.2 The Arabian Sea Process Study
The Arabian Sea Process Study is currently being implemented. Proposals were submitted to NSF for review in February, 1993, and decisions were made approximately one year later. A more detailed description of the Arabian Sea Process Study is given in the U.S. JGOFS Arabian Sea Implementation Plan
(Smith et al. , 1992).
Rationale: At the present time, it is unclear whether the northwestern Indian Ocean (Arabian Sea) is a sink for atmospheric carbon dioxide via its high rates of primary productivity and large concentrations of sedimentary carbon, or a source via outgassing of carbon dioxide brought to the surface during upwelling. The unique properties of the Arabian Sea can be used to expand our general understanding of the carbon cycle, productivity, and vertical flux of particulate material and biogeochemical transformations in the sea. Its principal unique feature is the regular oscillation of high rates of primary production and generally oligotrophic conditions under relatively constant levels of illumination. The oscillations in productivity and biomass that in high latitudes are forced by temporal variations in solar irradiation are here of a similar magnitude, but are forced by monsoonal atmospheric conditions which, via surface pressure fields and baroclinic adjustments, affect mixed-layer development and nutrient supply. The Arabian Sea experiences extremes in atmospheric forcing that lead to the greatest seasonal variability observed in any ocean basin. The wide range of climatic variability in the Arabian Sea makes it an excellent place in the present-day ocean to look clearly at past climates and possible future climates.
Time Frame: late 1994 - early 1996.
Objectives: The objectives have been broadly summarized in the U.S. JGOFS Planning Report Number 13 (Smith et al. , 1991). In that report, each objective is broken down into specific questions for the general categories of primary productivity and carbon/nitrogen cycling, heterotrophic processes, water column geochemistry, and benthic fluxes/paleoceanography. The questions are:
Scope of the Study: The U.S. JGOFS Process Study in the Arabian Sea will take place from September 1994 through January 1996. Other U.S. research programs in the region, specifically U.S. WOCE and the U.S. Office of Naval Research's (ONR's) Forced Upper Ocean Dynamics Program, will be conducted in parallel with U.S. JGOFS and coordinated with JGOFS. The U.S. JGOFS investigation alone will require 10 months of ship-time in 1994-1996. The schedule follows. Figure 2 shows the locations of U.S. JGOFS moorings and stations for process observations. A cruise track has been devised that includes six long stations, nine intermediate stations, and twelve hydrographic stations (Figure 2).
During pelagic process cruises, each of six long sampling sites will be occupied for a period of two days each for the purpose of establishing diel and variability in all relevant processes. These six sites are located in the area of very positive wind stress curl (2), in the area of very negative wind stress curl (1), in slightly positive curl (1), in slightly negative curl (1), and in the suboxic area near India. During benthic/paleoceanographic cruises, coring and other sampling are focused on the sites where sediment traps are moored (Figure 2) on the continental shelf and on the continental slope.
Mooring instrumentation for biological and chemical time-series will be emphasized in the Arabian Sea Expedition. Observations that will be made from moorings include temperature, salinity, scalar irradiance, particle concentration, chlorophyll, sedimentation, meteorology, and zooplankton biomass.
It is anticipated that the scatterometer, AVHRR, and altimeter satellites will be fully operational during all of the investigation of the Arabian Sea. SeaWiFS may be operational during the final months of the expedition. Aircraft overflights for surface pigments will be done during July, a month in which the historical CZCS records indicate haze or dust often rendered the CZCS records useless. Attempting to quantify the haze and dust of the region by satellite should be a high priority investigation.
Models for the region exist at Florida State University, Nova University, Princeton University and the Naval Research Laboratory. Efforts to incorporate U.S. JGOFS biological and chemical variables into these models were made prior to 1994,with the result that the cruise track was modified on the basis of model results. Development of these models for the Arabian Sea should continue, as a guide to planning field work, and in the future to synthesize the large data base anticipated from the ship and mooring observation programs.
Finally, knowledge of meteorological forcing is essential to the successful completion of the investigation. Meteorological instrumentation will be included in the mooring program and all research vessels intending to participate in the investigation must be equipped with complete and up-to-date meteorological capability.
Observations: This suite of observations is considered the fundamental data set that should be acquired during pelagic process cruises in the Arabian Sea.
Meteorological Observations Hydrographic Observations (depth-resolved) wind speed conductivity wind direction temperature humidity oxygen sea-surface temperature nutrients air temperature pigments barometric pressure fluorescence precipitation POC, PN and DOC solar radiation CO2 system properties longwave radiation particles (transmissometry) dust fall scalar irradiance Pelagic Standing Stocks Pelagic Productivity and Grazing
phytoplankton carbon mesozooplankton nitrogen microzooplankton oxygen bacteria mesoplankton and microplankton
On benthic process cruises, meteorological and hydrographic observations above should be made in addition to the following sediment and productivity measurements.
Sediment Composition Benthic Productivity and Standing Stocks carbon content foraminiferan production and biomass nitrogen content nutrients in pore water grain size mineralogy stable oxygen isotope abundance stable carbon isotope abundance
Mooring Program: One surface mooring contains 13 instrument packages. Instrument packages that measure fluorescence, beam attenuation, photosynthetically active radiation and oxygen are located at 10m, 20m, 35m, and 50m. The acoustic sensor for zooplankton biomass is at 40m. Conductivity and temperature sensors are at 8m, 13m, 18m, 48m, and 55m. Current meters are at 5m, 10m, 15m, 20m, 30m, 35m, 45m, and 50m. The meteorological instrumentation is on a discus buoy at the surface.
The four subsurface moorings contain either an upward-looking acoustic Doppler current profiler (ADCP) at 200 m or a profiling current meter and CTD. One subsurface profiling mooring contains a zooplankton biomass sensor at roughly 200m. The two upward looking ADCPs are also zooplankton biomass sensors.
Five separate subsurface moorings contain sediment traps (see Figure 2).
International Planning and Potential Collaborations with Other Programs: International planning for the Arabian Sea has been in progress since January 1991. The international planning group includes representatives from Pakistan, India, Oman, Kenya, Germany (chair), The Netherlands, United Kingdom, United States, and France. Of these, only France plans no field operations in the Arabian Sea. The Netherlands' Program commenced in 1992 and ends in 1993, and in a sense is a pilot study. It is collaborative with Kenya and Pakistan. In 1992 the U.S. and Pakistan began a four-year observational program in the northern Arabian Sea; this program is collaborative with the U.S. India has formed a JGOFS Committee, and the Indian Program will concentrate on the west coast of India and the central Arabian Sea in the 1994-1996 time-frame. The U.S. JGOFS planning group for the Arabian Sea has been approached by Oman for cooperation in setting up a JGOFS Time-Series Station on one of the coastal islands off Oman. Oman will be collaborating in the British, German, and U.S. JGOFS programs in the Arabian Sea. The Netherlands, the Intergovernmental Oceanographic Commission (IOC), with support from Germany, and the IOC in cooperation with the U.S. Office of Naval Research conducted training courses for regional scientists, with JGOFS methods/protocols being part of some of these courses. There is also an effort under way by the IOC to organize an exchange of scientists for data analysis. If the countries of the region each established and operated a JGOFS Time-Series Station in their territorial waters, the long-term scientific benefit would be substantial.
In an interagency meeting concerning research plans for the Indian Ocean, the 1994-1996 time-frame, with a focus on 1995, was identified by U.S. JGOFS for its Arabian Sea Process Study, by ONR for its Forced Upper Ocean Dynamics Program, and by WOCE for its World Hydrographic Program (WHP). The U.S. JGOFS sampling plan (Figure 2) has taken the U.S. WOCE WHP Plan into account and has identified some stations that overlap the U.S. WOCE WHP Plan; some lines in Figure 1 are part of the U.S. WOCE I7N line. Because U.S. JGOFS and U.S. WOCE are in the northern Indian Ocean at the same time, some of the U.S. JGOFS observations in the Arabian Sea are repeat hydrographic lines useful to U.S. WOCE. Additionally, the moorings for studying air-sea interaction and mixed layer dynamics which are part of the ONR Forced Upper Ocean Dynamics Program are necessary to understand the physical forcing which is part of U.S. JGOFS' goals, and to understand seasonal and possibly interannual variability of the region. Both the U.S. JGOFS Arabian Sea Process Study and the ONR Forced Upper Ocean Dynamics Program would like to turn their moorings around at the end of the field program and leave them in the Arabian Sea for an additional year. There have been preliminary discussions about this possibility, which clearly increases the understanding we can gain from the Arabian Sea
Implementation Schedule: This schedule contains plans of both the U.S. JGOFS Arabian Sea Process Study and the ONR Forced Upper Ocean Dynamics Program since these two investigations will work cooperatively on moorings and will coordinate use of a single research vessel. ONR's program is identified as NRL/SeaSoar.
Planning Meetings, Reports and Workshops:
A post-field work synthesis workshop and an international symposium to be held in the Netherlands are planned.
5.7.3 The Southern Ocean Process Study
The planning effort for U.S. JGOFS studies in the Southern Ocean began in earnest in 1990 (U.S. JGOFS Report 16), and has proceeded in parallel with the international planning effort (SCOR/JGOFS Report 10, 1992). The implementation summary presented here is taken from the U.S. JGOFS Southern Ocean Science Plan (U.S. JGOFS Report 17). A formal Implementation Plan was published in May 1995, U.S. JGOFS, Southern Ocean Implementation Plan (May 1995).
Rationale: The Southern Ocean, defined for the purposes of this study as the region south of, and including, the Subtropical Convergence, covers nearly 20% of the global ocean area. The Antarctic Circumpolar Current (ACC) has the largest volume flux of any major ocean current ( 130 Sverdrups). It is the only continuous circumglobal current, without beginning or end, and it is responsible for mixing of the deep waters of the other major oceans. Most of the ventilation of deep-sea water masses takes place in the Southern Ocean; in other words, deep water masses exchange gaseous components, including CO2 with the atmosphere. Furthermore, most deep waters derive their physical, chemical, and biological characteristics in the regions of the Southern Ocean where isopycnals outcrop at the sea surface and where mixing, cooling, and sea ice formation produce new water masses which sink into the ocean interior and renew the intermediate and deep waters of the world's oceans.
A unique feature of the Southern Ocean is the extensive regular seasonal advance and retreat of sea ice, oscillating between a maximum coverage of 20 106 km2 and a minimum of 4 106 km2. This surface feature, too, can be thought of as a frontal system, one that migrates north and south many hundreds of km annually. Biological productivity of surface waters is strongly influenced by the presence, and melting, of sea ice. Ice-edge productivity supports an abundance of life at higher trophic levels including mammals and birds as well as zooplankton and fish.
Fluxes of carbon in the Southern Ocean are large and play an important role in the global carbon cycle, yet the magnitudes of these fluxes remain poorly constrained. Our view of the air-sea exchange of CO2 in the Southern Ocean has undergone a substantial challenge during the past few years. Recent efforts at modeling the interhemispheric gradient of atmospheric CO2 have suggested that there is no net oceanic uptake of CO2 in the Southern Hemisphere (Keeling et al., 1989a and b; Tans et al. , 1990). If, as suggested, the net air-sea flux of CO2 in the southern hemisphere is close to nil, then it must reflect the compensating effects of large fluxes of CO2 into, and out of, the ocean in different regions. Takahashi et al. (1986) estimated that the largest CO2 flux into the sea, on a global basis, occurs in the zone between 40S and 50S. A counterbalancing net flux of CO2 out of high latitude waters in the Southern Ocean is expected to exist, the magnitude of this upwelling-induced efflux of CO2 is not well constrained by measurements of surface-water pCO2.
Global warming is likely to perturb circulation, ventilation, and biogeochemical processes in the Southern Ocean and these, in turn, represent potentially significant feedbacks into the nature of global change. At present, we know too little to predict the role of the Southern Ocean in global change, or the response of biogeochemical cycles in the Southern Ocean to anticipated warming. By successfully conducting process studies in the Southern Ocean, and then incorporating the results into ongoing efforts to construct coupled physical-biogeochemical models, we can better determine the present role of the Southern Ocean in the global carbon cycle, and improve our capability to predict the likely response of the region to anticipated global change.
Time Frame: Ship-of-opportunity and other pilot studies may begin during the austral summer of 1994-95. The major U.S. field program in the Southern Ocean will take place in 1995-96 and 1996-97. An announcement of opportunity for the main U.S. JGOFS Southern Ocean studies was issued in late 1993, with proposals due in mid-to-late 1994.
Objectives: The broadly-defined goals of the international JGOFS program have been articulated in a number of planning documents. These goals can be refined into more specific research objectives pertaining to the Southern Ocean:
Much of the variability of biogeochemical processes in the Southern Ocean can be ascribed to the different characteristics of the various zones described above. Therefore, in order to determine the magnitude and variability of carbon fluxes, and to understand the factors which regulate these fluxes, it is necessary that a JGOFS process study examine each of the zones. After considering the distinguishing characteristics of the diverse regions of the Southern Ocean, planning groups at both the national and international level have identified four general zones that represent the minimum number of subsystems having distinct biogeochemical properties that must be examined to provide a comprehensive assessment of carbon fluxes in the region. Physical, chemical and biological features within each zone have been thought to be sufficiently similar around the Southern Ocean that each zone can be considered as a continuous entity. The validity of that assumption needs to be tested, however, by comparing the results of the various national JGOFS process studies in different sectors of the Southern Ocean. The four representative zones include:
Program Elements: Like the Equatorial Pacific and Arabian Sea Process Studies, the strategy for studying the Southern Ocean includes:
JGOFS commenced with a Pilot Study in the North Atlantic, and its final Process Study will revisit the basin after 1998. The planning process for future work in the North Atlantic began with formation of an international planning group in 1992, and an international workshop in April, 1993. US JGOFS held a preliminary planning meeting in March, 1993, and a larger workshop in Bermuda in 1994 (US JGOFS Report #X). Scientific objectives for the US JGOFS North Atlantic Process Study were identified at that Workshop. The North Atlantic Process Study will provide US JGOFS with its only major opportunity to revisit a region of earlier study and test basin-specific hypotheses arising from analysis of earlier cruise results and other observations. Observations from the Bermuda Atlantic Time Series (BATS) as well as the North Atlantic Bloom Experiment ( NABE) were used to define priorities for future work.
There is ongoing research by European JGOFS programs in the eastern North Atlantic, and several nations are planning major studies for 1995-97. The major US campaign in the region will be in the period 1998-99, following the Southern Ocean Process Study.
Most analyses of interhemispheric gradients in atmospheric CO2, and assessments of CO2 sources indicate that the major global sinks for anthropogenic carbon are in the Northern Hemisphere. but, as an indication of our lack of understanding, is the sink terrestrial or oceanic? The North Atlantic has been identified as a potential major sink for anthropogenic carbon dioxide in part due to the siginificant amount of deep and intermediate water formation, yet current model estimates of the size of the North Atlantic CO2 uptake vary widely (eg, Sarmiento and Sundquist, 1992; Sundquist, 1993), attempts to constrain the anthropogenic flux are further complicated by the possibility of a net natural, background flux into the north atlantic (Broecker and Peng, 1992). the uptake of atmospheric CO2 can be estimated either directly, using seasonal maps of surface pCO2 and gas exchange rates, or indirectly, by computing the net advective divergence of CO2 for the basin. both approaches are at present plagued by a paucity of data on the appropriate scales. only a handful of meridional CO2 fluxes are currently available, all of which lack information on seasonal variability and the flux of doc. for surface pCO2, field work indicates a large amount of spatial variability (Watson et al., 1991) and a seasonal variabilty (Michaels et al., 1994; Keeling 1993) far greater than that in the atmosphere from which the long-term basin-scale averages need to be extracted.
Several recent lines of research have raised critical questions regarding the cycling and fluxes of carbon in the north atlantic. The JGOFS NABE studies revealed an unambiguous role of the biological pump in the seasonal drawdown of CO2 during the spring bloom, but quantitative budgeting of the relative importance of biological vs. physical processes in governing the seasonal CO2 cycle remains to be accomplished. The NABE sampling, unfortunately, did not observe the initial winter condition which precedes, and sets the stage for the bloom process. Studies of a local carbon budget constructed from the first five years of observations at the Bermuda Atlantic Time Series station indicate a greater than factor of 2 uncertainty in the balance between the observed seasonal drawdown in DIC and estimates of specific removal mechanisms (Michaels et al., 1994). More comprehensive sampling of the key properties and processes of the ocean carbon system are needed to improve our mechanistic and phenomenological understanding of the physical and biological pumps over an entire ocean basin. A recent analysis by Sarmiento (1995) suggests that CO2 uptake in the North Atlantic might be low with the DIC outflow to the south balanced by DIC/DOC supply through the Bering Strait and Arctic, rather than from atmospheric uptake. Narrowing the large uncertainties in the carbon budget for the North Atlantic basin is a critical task for JGOFS to address: without an effort to resolve our understanding of CO2 fluxes in this best studied and still poorly understood basin, JGOFS would not be complete.
Based on reviews of the NABE results and the considerations noted above, the US and international planning groups agreed that it was appropriate and timely for the next North Atlantic Process Study to focus on studies aimed at an improved answer to the question:
"What is the size of the CO2 sink in the North Atlantic Ocean?"
Answering this question not only means placing improved bounds on a new estimate, but improving our understanding of the processes contributing to the CO2 sink in the North Atlantic. The current estimates for the total oceanic uptake range from 0 to over 2 GtC yr-1, with the best ocean models giving 2 +/- 0.8 GtC yr-1 (Sarmiento and Sundquist, 1992). Achieving a satisfactory improvement will require an integrated program of time series, survey and process observations, remote sensing and modeling, as described below. The uncertainties in the carbon budget at basin and local scales suggest that an improved estimate requires improved understanding of carbon fluxes and transports, not merely more finely resolved data to estimate CO2 exchanges. Approaches to improved understanding are described below.
Finally, the final JGOFS Process Study will serve as a pilot study for future monitoring activities in the Global Ocean Observing System.
The US planning workshops identified the following as principal components for future work in the North Atlantic:
1. Constraining basin-scale carbon fluxes. The study would by necessity be seasonal and basin-scale (i.e. of larger spatial scale than NABE). It would emphasize novel techniques (biogeochemically instrumented buoys and/or drifters, underway systems, data assimilation) and would in many ways be a pilot study for future monitoring activities. The second North Atlantic study would incorporate elements from both a traditional process study and the expanded global survey (pCO2, Chl, nutrients) from the JGOFS science and implementation plans.
In a bit more detail, the study would involve extrapolating the seasonal pCO2 coverage for the basin using correlations among surface pCO2, hydrography, mixed layer depth, nutrients and would most likely rely heavily on data assimilation of field and satellite (wind speed, temperature, ocean color) measurements. Field work would be used to fill in critical gaps (many of them during the winter) of both surface pCO2 and its correlations with other variables and will emphasize "cheap" approaches (e.g. instrumented buoys, VOS) where possible.
2. Meridional flux estimates. A second thrust of the program would involve better estimates of the meridional fluxes of key biogeochemical variables (TCO2, NO3, PO4, DOC, DON, O2) from which the net divergence of carbon and nutrients for the North Atlantic could be calculated. The meridional fluxes would be estimated using the same approach as Brewer et al. (198?) but with seasonal resolution and improved constraints on the total flow field (from ADCP's pr IES') inverse modeling. These observations and sections will require a dedicated ship for a period of one year. The ship will alternate between regional pCO2 surveys and meridional sections, with seasonal coverage for each activity.
3. Control volume experiment near Bermuda. In order to close the Bermuda carbon budget, a better accounting of carbon removal by sinking particles (POC) and water motions and mixing (DOC, DIC) are required. Existing modeling and data assimilation tools will be applied in concert with observations from SeaWiFS, aircraft, biogeochemical moorings, routine time series cruises, and special process study cruises to balance the observed decline in surface concentrations of CO2 with specific process estimates of carbon export via sinking, DOC transport, etc. Fieldwork for a properly designed control volume experiment could be completed in a single field season, once moorings were in place and models were up and running. Some key core measurements are not made routinely, if at all, in BATS (e.g., grazing, 15N assimilation) and would need to be added for the process study.
The following are the highest priority measurements for each study. A more focused planning effort may add or subtract measurements from these lists:
1. Bermuda experiment: meteorology, CTD, oxygen, CO2 system properties, nutrients, particles (transmissometry and CHN analysis), DOC, pigments and optics. Particle flux from radiochemistry and floating traps. Optics, pCO2, oxygenand nutrients from moorings and drifters. Aircraft surveys of ocean color. New production, grazing rates and DOC utilization.
2. pCO2 and meridional carbon surveys: meteorology, CTD, oxygen, CO2 system properties, nutrients, particles (transmissometry and CHN analysis), DOC, pigments and optics. Underway CO2, nutrients, oxygen, pigments. DIC and DOC. SeaSoar mapping of upper ocean physical, chemical and optical fields.
Provisional implementation schedule:
The Bermuda Control Volume experiment could be carried out with a temporary enhancement of support for the time series study. Another control volume experiment could be carried out in 1986-7, concurrently with work by west European JGOFS nations in the Irminger Sea.
Elements of the zonal sections may be accomplished in collaboration with the WOCE Atlantic Program in 1996--1997. The pCO2 survey and meridional surveys might be carried out in conjunction with DOE- and NOAA sponsored work and WOCE operationsin the same time period. Large scale NSF support for coordinated and intensive process study observations is unavailable until after 1998.