On 8 July, 2000, a special workshop was held at the Woods Hole Oceanographic Institution (WHOI) to address the issue of the importance of continental margins in global ocean carbon and nutrient budgets and cycles. The workshop was attended by members of the SMP Continental Margins Working Group (Table 1) and also included Ken Brink (WHOI) and Larry Atkinson (ODU), who were invited because of their experience with shelf environments. The workshop was followed by three separate meetings of the SMP Continental Margins Working Group, held during the SMP PI meeting at WHOI.

The question of the importance of margins is relevant because the U.S. JGOFS program focused on developing a carbon budget based on deep ocean observations and models since its inception in the early 1980's. As we enter the twenty-first century, numerical models used by SMP investigators do not have the spatial resolution that may be required to accurately assess the contribution of continental margins to these global budgets, and there is increasing evidence that processes along margins cannot be ignored.

Workshop findings:

Studies that have attempted to estimate global production based on historical field observations invariably conclude that continental margins (shelves, slopes, rise) comprise somewhere between 25 and 50% of the global ocean net production (Berger et al.,1989; Koblenz-Mishke et al., 1970; Walsh, 1988, 1991). Global assessments based on ocean color satellite data also show a band of high pigment concentrations and primary productivity around global margins (Behrenfeld and Falkowski, 1997), beyond the area that could be expected to be "contaminated" by bottom-reflected light. Clearly, the optical "contamination" due to chromophoric dissolved organic matter (CDOM) needs to be accounted for in these images, but all such estimates show a four- to five-fold increase in production in a steep gradient going from open ocean waters to continental margins.

There is also increasing evidence that a large percentage of the particulate organic carbon generated by primary production on margins is exported to the ocean "bottom". Specifically, estimates of vertical organic particle flux and ocean bottom respiration rates suggest that the export ratio (organic carbon particle flux / primary productivity) along continental margins is perhaps 20-fold higher than in the open ocean (Walsh, 1988; Jahnke, 1996; Berner, 1992). A conservative estimate based on the results of Jahnke (1996) is that 44% of total global organic C flux reaches the bottom of the slope and rise. Since Jahnke examined only areas deeper than 1000 m, there is still the question of how much is deposited in the shallower slopes and shelves. Further, there is an unresolved question as to the magnitude of carbon exported from shelves in dissolved organic form.

The eyebrow-raising question that rises from such results is  "Did JGOFS miss one-half of the global ocean carbon flux signal to the deep ocean?"

As this question was raised during the plenary of the SMP PI meeting, the response from modelers was that Global Carbon Models (GCMs) generally account for the known oceanic tracer distribution (for example bomb-derived ¹⁴C) and that they do not point to any large unaccounted for source of carbon to the deep ocean. If current oceanic models indeed are able to balance the carbon budget, then the question may be rephrased to ask whether GCMs "misrepresent" the carbon flux processes to account for the effect of margins. Alternatively, may GCMs have been "tuned" all too well to reproduce observed patterns? The answer will not be easy to find, however, because comparisons of results from various GCMs conducted under the OCMIP effort really show wide discrepancy in tracer distributions.

The Working Group grappled with the question of whether there should be a conceptual or operational definition for margins. The conclusion was that this would depend on oceanographic discipline and on the application. A definition could be based on a specific depth or on an arbitrary distance beyond a certain bathymetric contour. A dynamic definition could also be based on time-varying patterns observed in ocean color and infrared satellite imagery, as areas where blooms can be observed directly associated with upwelling or other shelf areas. Other definitions may be process-driven, such as where a marked change in isopycnals is observed, or where marked horizontal physical, biological, chemical, or geological gradients occur. Conceivably, a definition could use several of these elements. The group felt that an operational definition for modeling purposes is not critical at this time.

At this stage, while we have some understanding of the causes of high primary production along margins, we don't fully understand the scales of variability and processes leading to nutrient supply in these diverse environments. We also know that current global carbon models are grossly under-resolved spatially and in forcing functions like the wind. Paradigms and models developed for the deep ocean do not apply to margins.

The major conclusion of the workshop was that margins should no longer be dismissed as areas whose impact is limited to carbon recycling. There are unresolved questions of food web control of the export ratio and carbon burial, of dissolved organic carbon input, production and export, of processes of nutrient supply (wind-driven, boundary current, eddy upwelling, atmospheric deposition, river), and of nutrient sinks like denitrification.

Recommendations:

The SMP Continental Margin Working Group compiled a list of recommendations for the SMP, the JGOFS Steering Committee, and relevant government agencies:

1) Quantify the uncertainty introduced by margins in global carbon (and nutrient) cycling and budget models. Specifically, it would be useful to estimate the magnitude of the error in global assessments derived by the JGOFS Synthesis and Modeling program by omission of margins.

2) Explicitly include continental margins into SMP planning and into any new global carbon/nutrient cycle initiative.

3) Margins should be incorporated into global carbon cycling models. This includes a study of whether there are inconsistencies in continental margin data, oceanic tracer distribution, and global models. Can experiments be done in a global model to elucidate the role of continental margins? For example, if all of the continental margin particulate organic carbon is put into a GCM, what would be the resulting signal? In the process, it will be important to identify which margin processes are either improperly parameterized, or are missing, in GCMs. This may require developing a testbed program for coupled open ocean-margins models.

There are a variety of exciting recent developments with respect to modeling of the interactions between the deep ocean and continental margins. These new initiatives take two basic forms: a) high resolution regional models of the continental shelf which extend outward to the deep sea, and b) deep-ocean models that reach inward toward the continental shelf. Examples of the former include various NOPP projects focused on the west coast (see NOPP home page links to projects led by McWilliams et al., Allen et al., Paduan et al.). Examples of the latter can be found in Dale Haidvogel's spectral element model (http://marine.rutgers.edu/po/). Future work on the ocean carbon cycle would benefit greatly from linkage with both approaches for modeling deep-ocean/continental shelf interactions.

4) An ambitious but realistic activity by JGOFS SMP is needed at the international level, to take advantage of the interest on coastal zones and margins by the international community (e.g. Land Ocean Interactions in the Coastal Zone/LOICZ).

5) Future carbon cycle programs will need a list of scientific questions that require answers. Specifically, SMP can start identifying which questions are most relevant. For example:

What is the role of cross-shelf particulate organic transport?
What is the role of shelf/slope sources of iron?
What is the role of sedimentary transport or burial of carbon?
Should shelves be regarded as temporary storage or permanent burial (over say, a 100 y time scale)?
What are the effects of shelf-bound processes like denitrification?
How can the growing coastal observing systems support global assessments conducted by the SMP?

Summary:

In summary, there is a question of whether JGOFS missed a significant portion of the global ocean carbon flux signal to the ocean by omission or inaccurate parameterization of continental margin processes. The source of this error (the margins) needs to be at the center of planning for any future program focusing on global carbon and nutrient assessments. SMP can facilitate addressing this issue by identifying key scientific questions associated with margins, and encouraging funding to address such questions in future programs.

Aknowledgements:

Support for the workshop was provided by the U.S. JGOFS Planning Office and Ken Buesseler, Executive Scientist - U.S. JGOFS Steering Committee. We especially thank Mary Zawoysky (WHOI) for her assistance in making the workshop possible.

References:

Behrenfeld, M. J. and P. G. Falkowski (1997) Photosynthetic rates derived from satellite-based chlorophyll concentration. Limnol. & Ocean. 42: 1-20.

Berner, R. A.  (1992) Comments on the role of marine sediment burial as a repository for anthropogenic CO2. Global Biogeochem. Cycles. 6(1). 1

Berger W.H. (1989) Appendix: Global maps of ocean productivity. In Productivity of the Ocean, Present and Past (eds. Berger W.H., Smetacek W. G.), pp. 429-455. John Wiley & Sons, New York.

Jahnke, R. A. (1996) The global ocean flux of particulate organic carbon: Areal distribution and magnitude. Global Biogeochem. Cycles. (10:1), 71-88.

Koblenz-Mishke, O. J., V. V. Volkovinsky, and J. G. Kabanova (1970) Plankton primary production of the world ocean. In: W. S. Wooster (ed.). Scientific Exploration of the Southern Pacific. National Academy of Science, Washington, DC. 183-193.

Walsh, J. J. (1988) On the nature of continental shelves. Academic Press, San Diego. pp. 1-520.

Walsh, J. J. (1991) Importance of continental margins in the marine biogeochemical cycling of carbon and nitrogen. Nature. 350. 53-55.
 

Table 1. U.S. JGOFS SMP Continental Margin Working Group Members and Workshop Participants*.

Larry Atkinson*	Atkinson@ccpo.odu.edu	Old Dominion University
Ken Brink*	Kbrink@whoi.edu	Woods Hole Oceanographic Institution
Anand Gnanadesikan	gnana@splash.princeton.edu	Princeton University
Rick Jahnke*	Rick@skio.peachnet.edu	Skidaway Institute of Oceanography
Joanie Kleypas	kleypas@ncar.ucar.edu	Nat. Center for Atmospheric Research
Frank Muller-Karger*, Chair	carib@marine.usf.edu	University of South Florida
Dennis McGillicuddy	dmcgillicuddy@whoi.edu	Woods Hole Oceanographic Institution
Don Redalje*, Co-chair	Donald.redalje@usm.edu	University of Southern Mississippi
Peter Verity*	Peter@skio.peachnet.edu	Skidaway Institute of Oceanography
Jim Yoder*	Yoder@emu.gso.uri.edu	University of Rhode Island