The Role of Eddies in Basin Scale Biogeochemical Budgets of the North Atlantic

Coupled physical--biogeochemical models provide a useful framework for testing our ideas about the underlying physical, biological and chemical controls on elemental cycling in the ocean. Up to now, efforts to do so on basin scales have produced mixed results. The seminal work of Sarmiento et al (1993) and Fasham et al (1993) demonstrated that the large scale distributions of chlorophyll in the surface waters of the North Atlantic could be simulated with a reasonable degree of accuracy using a simple planktonic ecosystem embedded in a coarse resolution general circulation model. While the general characteristics of these large scale patterns were consistent with observations, detailed comparisons with time series data at specific sites revealed that the underlying flux balances maintaining these distributions were not as satisfactory. In particular, the nutrient budget in the Sargasso Sea was dominated by horizontal advection. This result is quite different from recent regional modeling studies (McGillicuddy 1995; McGillicuddy and Robinson, 1997) which have suggested that upwelling in the interiors of cyclonic mesoscale eddies is the dominant mode of nutrient supply to the euphotic zone in the open ocean.

During the last several years, the World Ocean Circulation Experiment Community Modeling Effort (CME) has made significant progress in the development of more realistic general circulation models. Areas of improvement particularly relevant to biogeochemical processes include (1) aspects of the time-mean flow, (2) treatment of the surface boundary layer, and (3) the resolution of mesoscale eddies. This next generation of GCMs represents a substantial step forward in the physical bases on which coupled biogeochemical simulations can be built. It is proposed herein to construct the first basin scale eddy-resolving biogeochemical simulations of the North Atlantic to examine the impact of mesoscale processes on large scale elemental budgets. A suite of experiments is planned which are based on 1 degree (coarse), 1/3 degree (eddy-permitting) and 1/6 degree (eddy-resolving) CME simulations. The coarse resolution runs will be used to benchmark the Sarmiento and Fasham results, and serve as a control run for the higher resolution calculations. Although the eddy-permitting case will most likely not be entirely sufficient for a comprehensive study of mesoscale processes, it does represent a meaningful intermediate step that will provide useful information without incurring the substantial computational cost of the 1/6 degree runs. This will set the stage for a relatively small number of fully eddy-resolving runs which have only recently become feasible from a computational point of view. Understanding gained from these experiments will be used to guide the development of parameterizations for mesoscale biogeochemical processes to permit their inclusion in larger scale (global) models.