Preface

The following document is a report of a Workshop held in Bermuda in April, 1994 to consider the U.S. JGOFS contribution to future JGOFS research in the North Atlantic. Two particular themes emerged from this meeting and subsequent considerations by the U.S. JGOFS Steering Committee:

1. a Control Volume Experiment (CVE) focussing on the carbon budget near the Bermuda Atlantic Time Series (BATS) site was discussed in some detail.

2. a larger-scale, regional study aimed at constraining meridional carbon fluxes, to be carried out in concert with the WOCE-JGOFS CO2 Survey, was also considered.

This report mainly concerns the former study. The regional survey was to have been the focus of a subsequent, major U.S. JGOFS Workshop. However, due to the expectation of insufficient resource availability in later years, and the growing realization that resources need to be devoted to Synthesis activities, U.S. JGOFS has recently curtailed planning of that study. As the following report shows, a CVE could be performed in coordination with BATS operations, and would enhance the scientific value of BATS and of JGOFS' understanding of local, 1- and 3-dimensional carbon budgets. This report is being placed on the U.S. JGOFS Web Pages to elicit further discussion of scientific issues and priorities for JGOFS as it begins to consider its accomplishments and future needs. The final report will be published in the U.S. JGOFS Planning Report Series following a suitable period of public discussion.

Hugh Ducklow

28 November, 1995

1 Findings and Recommendations

1. Local, vertical (1-dimensional) carbon budgets cannot be constructed with better than factor of 2-3 accuracy at the best-studied sites (e.g., Bermuda). The uncertainty is caused by some unspecified combination of conceptual and observational inadequacy, but there are few data for the design of future studies addressing this shortcoming.

2. Current estimates of carbon uptake in the North Atlantic are based on coarsely gridded (5-degree boxes, seasonally-resolved) data. Observations in much of the basin are still sparse.

3. Control volume experiments (CVE) provide a powerful approach to better understanding and quantification of carbon flux processes, but require heavy investment in moorings, shiptime and PI support.

4. A powerful array of modeling techniques now exist to enable design and implementation of CVE's optimized for addressing questions about carbon flux.

5. CVE's probably require prior time series and survey information in order to optimize site selection and timing, and allow proper interpretation of the results.

6. At the present time, Bermuda has the richest archive of observational and modeling data, but flux signals there are low relative to advection.

7. Higher latitude sites (e.g., the Irminger Sea south and east of Greenland) have stronger signals and possibly less advective interference, but almost no data for proper design of future studies.

8. A properly designed and implemented CVE would contribute to narrowing the uncertainties in carbon budgets at time series sites, and appears to be a logical extension of time series operations. The CVE model is ideal for answering focused questions about processes. But CVE's alone cannot serve as strawman models for the entire basin.

9. U.S. JGOFS should contribute toward time series and survey operations now being planned by European JGOFS investigators. In addition, a limited and focused selection of specific process studies (e.g., zooplankton grazing and egestion, DOC cycling, export processes) should be undertaken to sharpen questions and improve models.