1.1.1 Purpose of Document
JGOFS grew out of a workshop held by the National Academy of Sciences in Woods Hole in 1984 with broad community participation, which resulted in publication of the document Global Ocean Flux Study. A Steering Group was established as a result of this activity whose goal was to establish a viable scientific effort within the U.S. to examine carbon flux in the ocean and its role in the global carbon system.
An early focus of these efforts was to organize a domestic program focused on carbon flux issues, the Global Ocean Flux Study, initiate work on international analogs, and to develop a consensus in the U.S. community concerning a scientific agenda for such studies. In less than two years, an international program, the Joint Global Ocean Flux Study, endorsed by the Scientific Committee on Ocean Research, and a linked domestic program supported by U.S. federal agencies were in place. Early on it was decided that the ability to accomplish multi- and interdisciplinary research in biogeochemistry needed to be demonstrated, so a pilot process study, the North Atlantic Bloom Experiment, was organized. Simultaneously, a science plan (U.S. Joint Global Ocean Flux Study Long Range Plan) for the enterprise was under development and published in May, 1990.
The science plan is far ranging, in fact it goes beyond looking at the role of carbon in the ocean system. The science plan was written to provide an accurate description of the state of the science and the issues associated with biogeochemical cycling in the ocean, circa its writing. The multitude of issues discussed in the science plan are not easy to resolve. In many cases they demand new approaches to science. These approaches encompass not only the sociology of different disciplines working together, such as physics, chemistry, biology, geology and meteorology, but also need for major advances in ocean observations, conceptual and numerical models of the ocean and our nation's ability to manage the results of such activities.
The present document is the initial version of an evolving Mid-Program Implementation Strategy for the U.S. JGOFS program. It was initially written after the pilot study, the first large-scale process study in the Equatorial Pacific, implementation of the Global CO2 survey, and successful operation of two time-series stations for five years. U.S. JGOFS has elected to follow a variant of `rapid prototyping' in its implementation planning. Rapid prototyping as used by U.S. JGOFS has involved going to the field with early process studies such as the North Atlantic Bloom Experiment to demonstrate multi- and interdisciplinary observing and early synthesis capabilities, elucidate near-term implementation issues, and then, once a workable approach is demonstrated, to codify the overall programmatic implementation based on the early results. While this approach has given clear guidance on process studies and time-series observations, and the large-scale observations effort as part of the WOCE Hydrographic Program (WHP) has given us indicators for survey work, we have had no similar large scale pilot activities in modeling, data management, synthesis. Thus, the reader should view the current version of this plan as reflecting an advanced understanding of currently viable approaches to in-situ studies and slightly more speculative approaches to the other parts of the JGOFS spectrum. Future editions of this plan will address improvements in our understanding in these other parts of the JGOFS program, as well as address the changing fiscal realities.
The goal of the U.S. JGOFS Mid-Life Strategy is to provide a map to guide the program from mid-life to the attainment of its stated goals. In a more general sense, this plan can be viewed as describing planned field efforts through the end of the decade and analysis and synthesis activities continuing through 2003. It is the program's intent that by the end of the millennium it can provide new insights into the oceanic carbon cycle in the context of interactions between biogeochemical, ecological, and physical processes in the ocean. The resulting synthesis of the JGOFS data and process studies' interpretations should advance us towards our stated operational goal:
"To assess more accurately, and understand better the processes controlling, regional to global and seasonal to interannual fluxes of carbon between the atmosphere and ocean interior, and their sensitivity to climate changes" (JGOFS Science Plan).
Within this broad framework stated in the science plan, we have five specific objectives:
1.2 U.S. JGOFS Structure
U.S. JGOFS has developed a five-fold approach to address its implementation. It can be considered as five cooperating, complementary activities. Broadly defined, these efforts are: Large Scale Observations, Time-Series activities, Process Studies, Modeling, and Data Management. For extensive background and a thorough description of each of these areas, the reader is asked to review sections III through VII in the U.S. JGOFS Long Range Plan (May 1990). The following gives summaries of each area.
Large Scale Surveys. Large scale studies of biogeochemical variables in the ocean are sparse. Extant views describe an ocean which is heterogeneous on many scales with the greatest part of the variance at low spatial and temporal frequencies. Unfortunately these large time and space scales are currently the most poorly observed in the ocean. The rationale for the large scale survey component is to provide a composite, basin to global scale, biogeochemical view of the ocean surface, mid-depth and deep waters. During the JGOFS implementation phase specifically we plan to provide seasonal resolution of key biogeochemical parameters on regional and basin scales and to provide a consistent global description of surface pigment, primary production, CO2, and export fluxes and transformations.
Time-series. Prior to JGOFS there were no long term biogeochemical time-series in the oceans. At the mid-life of JGOFS, we now have two five-year records (Hawaii and Bermuda) and the beginnings of a multi-year high latitude record (Kerguelen Island). Resource limitations dictate a limited number of such time-series sites. Such limitations also suggest that future time-series efforts will utilize autonomous observing systems rather than human/ship mediated observations. The objective of the time-series effort is to provide well-sampled seasonal resolution of biogeochemical variability at a limited number of ocean observatories, provide support and background measurements for process-oriented research, as well as test and validate observations for biogeochemical models.
Process Studies. Early on in JGOFS it became clear that many of the links between key biogeochemical parameters were not well understood. The objective of the process studies component is to target key process links in our current models of the oceanic biogeochemical system and enhance our causal understanding of the processes. The goal of process oriented studies is to provide a mechanistic understanding of ocean processes in sufficient detail to predict and simulate biogeochemical fluxes at representative sites in the ocean.
Modeling. Modeling represent the synthesis of our process understanding as well as an approach for testing our current understanding of various biogeochemical cycles. With sufficient development, models can be used to examine sampled strategies for process studies as well as large scale observations. In the near term, models also suggest possible linkages where improved understanding will provide the greatest advantage. U.S. JGOFS views models in all these ways, however, our objective is to provide a useful synthesis of our understanding which can be used for diagnosis of the current ocean role in the carbon system, as well as for future forecasts of the ocean state.
Data Management. Historically, data management of biogeochemical parameters has not had high priority. This has led to minimal catalogs of past data, little exchange of data holdings, and a general lack of confidence in biogeochemical datasets. U.S. JGOFS, by necessity, believes a critical component of U.S. JGOFS implementation strategy is an agile, cost effective, workable data management scheme which facilitates ready sharing of data models, analysis and other knowledge. The objective of the U.S. JGOFS data management activity is to provide a living dataset which is readily accessible during and after U.S. JGOFS and provides sufficient data documentation to permit intercomparison and quality control.
1.3 Specific Goals
U.S. JGOFS, after completion of the Long Range Plan, started a three year sequence of discussions on an implementation plan. Five working groups, composed of U.S. JGOFS Scientific Steering Committee (SSC) members, were established to develop prospectuses in each of the five streams of activity. The prospectuses were then discussed by the whole Steering Committee, changes suggested, and final versions prepared. Following this activity the draft plans were reviewed by an outside panel. At the conclusion of this process, consensus was reached on seven goals for the program.
The operational goals of the U.S. JGOFS implementation plan are to provide:
Currently we believe these goals should be obtainable by the year 2003 and would be a major step towards the previously stated objective.
A rational approach to implementing a large scale program requires a ranked list of priorities. U.S. JGOFS, in concert with the discussions concerning implementation goals, has also developed a set of implementation priorities. These priorities are based on our current understanding of the available resources, activities in each implementation area, needs in each implementation area, and the aforesaid goals. In a linked set of Steering Committee meetings, community input was used to develop a set of priorities for the program. The following are the priorities established by this process:
U.S. JGOFS accords top priority to the creation and implementation of an integrated system of data management analysis, synthesis and numerical modeling which provides the interface between the work at sea, in laboratories and the finished scientific product that we wish to leave as a JGOFS legacy.
A. Technology Implementation
It is clear that the two existing time-series stations, HOT and BATS, cannot alone provide enough information "To determine the response of the ocean carbon system to physical and chemical forcing from subseasonal events to decadal changes." Additional time-series experiments or observations or research being conducted by International JGOFS will provide valuable data, but there remain several crucial regions where no time-series study is underway. These include sites of deep convection, characterized by remote location, extreme seasonal and probably interannual variability and hostile environmental conditions. Time-series data for future major process study sites, including the Arabian Sea and the Southern Ocean, are also a high priority.
The U.S. JGOFS supports only the establishment of autonomous time-series observation sites in the future, because these appear to be the most cost effective approach. The SSC assigns the highest priority to deploying moorings incorporating proven sensors to measure the light field, biomass based on pigment signals, dissolved oxygen concentrations, conductivity and temperature, and current speed and direction, at time-series and process study sites.
Further technology development is needed for the autonomous measurement of pCO2, TCO2, alkalinity, pH, nutrients, primary productivity, and zooplankton abundance; however, these are being adequately funded outside the JGOFS program. The SSC assigns a lower priority to testing newly available sensors at accessible locations, such as the HOT and BATS sites, and to incorporating sensors which operate successfully into process study moorings.
B. Moderate-scale Process Studies
Because JGOFS will have the resources to conduct large-scale process studies in no more than 4 oceanic regions before the year 2000, it is clear that the large-scale process studies alone are not sufficient for the objective "To characterize the present geographical distribution of key biogeochemical properties and rate processes pertinent to the oceanic carbon system." The SSC recognizes that moderate-scale process studies, each requiring $2.5M in total funds could contribute greatly to our understanding of the global distribution of biogeochemical properties. These include but need not be limited to: the role of mesoscale ocean eddies; studies of nutrient sources and demand in the oligotrophic gyres; and follow-up studies of the spring phytoplankton bloom in the North Atlantic. The scientific background for these and other studies is presented in the U.S. JGOFS Long Range Plan. The SSC encourages further discussion on additional candidates for smaller, process-oriented studies.
The U.S. JGOFS program is operating in a budget constrained environment. Resources for the program come from the National Science Foundation (NSF), National Oceanic and Atmospheric Administration (NOAA), National Aeronautics and Space Administration (NASA), Office of Naval Research (ONR), and the Department of Energy (DOE). Currently these organizations provide approximately $15M annually for support of the spectrum of U.S. JGOFS activities. These resources are split between process studies ($9M), global surveys ($3M), time-series ($2M), data management ($0.5M), and modeling ($0.5M).
The most significant change implicit in this version of program implementation is a progressive investment shift toward synthesis and modeling over the next eight years. This plan assumes that modeling, synthesis and data management will be supported with increasing investment starting in FY1996 (+$1M) which will plateau in an investment of $8M per year by FY2001. Early investment would be targeted at currently available data sets while later investment would focus on integration of satellite observations (SeaWiFS, ADEOS, EOS/MODIS and EOS/COLOR) and coupled numerical models.
As stated in the Introduction, we expect this plan to be the first of several iterations towards definition of a programmatic structure and implementation which will achieve the stated goals of the program by 2003. We believe that U.S. JGOFS has already made significant progress towards determining and understanding specific processes during spring blooms and in the equatorial ocean which control the time varying fluxes of carbon in the ocean, as well as its exchange with the atmosphere. As this plan will make clear in the following pages new or enhanced effort in a number of our programmatic areas is needed to optimize the use of current data holdings and process insights as well as to answer questions in other biogeochemical regimes of the ocean system. These include development and application of a data management system and of both small and large-scale models coupling biogeochemical processes and ocean physics; implementation of technologies, including autonomous moorings and satellite remote sensing, which can collect long time-series of data over a wider geographic area than is possible using ships; and execution of moderate scale process studies to address significant gaps in global understanding of the oceanic carbon cycle.
During its approximately ten-year observational lifetime, U.S. JGOFS will provide a foundation for determining the response of the ocean carbon system to physical and chemical forcing over time scales ranging up to the sub-decadal. Thus, it is important that JGOFS encourage activities that will lead to continuing progress over the next 10-20 years. Key areas include the implementation of autonomous sensors at sites where intensive ship-based biogeochemical and physical data are available for verification and interpretation of the mooring data. JGOFS can play a leading role in establishing the groundwork for the biogeochemical and bio-optical measurements of the proposed Global Ocean Observing System. Another important activity is ground-truthing, leading to improved interpretation of satellite observations of ocean biogeochemical properties during the SeaWiFS mission. JGOFS will clearly demonstrate the value of satellite-based observations or inferences of biogeochemical properties and should use its accomplishments to support continued satellite observations such as the Earth Observing System (EOS). JGOFS will be the principal model for programs which will support and exploit the next generation of earth-observing sensors.