Linking ocean biogeochemical cycles and ecosystem structure and function: results of the complex SWAMCO-4 model

We present results obtained with SWAMCO-4, a complex model of the marine planktonic system calculating C, N, P, Si, Fe cycling within the upper ocean, the export production and the exchange of CO 2 between the ocean and atmosphere. The model, constrained by physical, chemical and biological (grazing, lysis) controls, explicitly details the dynamics of four relevant phytoplankton functional groups with respect to C, N, P, Si, Fe cycling and climate change. Those are diatoms, pico/nano phytoplankton, coccolithophorids, and Phaeocystis spp. whose growth regulation by light, temperature and nutrients has been obtained based on a comprehensive analysis of literature reviews on these taxonomic groups. The performance of SWAMCO-4 is first evaluated in a 1D physical frame throughout its cross application in provinces with contrasted key species dominance, export production, CO 2 air-sea fluxes and where biogeochemical time-series data are available for model initialisation and comparison of results. These are: (i) the ice-free Southern Ocean Time Series station KERFIX (50°40S, 68°E) for the period 1993-1994 (diatom-dominated); (ii) the sea-ice associated Ross Sea domain (Station S; 76°S, 180°W) of the Antarctic Environment and Southern Ocean Process Study AESOPS in 1996-1997 (Phaeocystis-dominated); and (iii) the North Atlantic Bloom Experiment NABE (60°N, 20°W) in 1991 (coccolithophorids). We then explore and compare the ocean response to increased atmospheric CO 2 by running SWAMCO-4 at the different locations over the last decade. Results show that at all tested latitudes the prescribed increase of atmospheric CO 2 enhances the carbon uptake by the ocean. However, the amplitude of the predicted atmospheric CO 2 sinks displays large regional and interannual variations due to the actual meteorological forcing that drives the local hydrodynamics. This is particularly true in the marginal ice zone of the Ross Sea (AESOPS) where the magnitude of the predicted annual CO 2 sink is positively related to the length of the surface ocean ice-cover period which determines the iron surface concentration at the time of ice melting. © 2004 Elsevier B.V. All rights reserved.