Coupling of coastal zone color scanner data to a physical‐biological model of the southeastern U.S. continental shelf ecosystem: 3. Nutrient and phytoplankton fluxes and CZCS data assimilation

Abstract

Phytoplankton and nutrient fluxes across the 40‐ and 75‐m isobaths at a nominal depth of 17 m were calculated from the distributions obtained with a physical‐biological model of the southeastern U.S. continental shelf ecosystem. These flux estimates represent integrated values over a 200‐km region of the southeastern U.S. continental shelf and cover the period of April 1980. The phytoplankton and nutrient fluxes showed considerable variability with time; however, the fluxes across the 40‐m isobath were primarily onshore while the fluxes across the 75‐m isobath were primarily offshore. The fluctuations in the nutrient flux were associated with the passage of Gulf Stream frontal eddy events. The effect of frontal eddy events on the phytoplankton fluxes was less apparent. Comparison of the time average of the two fluxes over one month with similar estimates for the fluxes at 37–45 m (Hofmann, 1988) showed a clear pattern of strong up welling between the 40‐ and 75‐m isobaths and an associated increase in phytoplankton concentration in the upper water column. In general, the pattern of the flux estimates was insensitive to the values chosen for the biological model parameters. The chlorophyll distributions obtained from nine Coastal Zone Color Scanner (CZCS) images from April 1980 were assimilated into the physical‐biological model to improve the phytoplankton flux estimates. Three techniques were tested for CZCS data assimilation into physical‐biological models, all of which gave stable model results. In general, the accuracy of the simulated chlorophyll fields was improved by assimilation of the CZCS data; however, the simulated fields rapidly converged to the fields obtained with no CZCS data assimilation after a few days. The time change of the phytoplankton carbon fluxes from the model with CZCS data assimilation were very similar to the fluxes obtained from the model with no CZCS data assimilation. However, the magnitude of the phytoplankton fluxes was less than the one obtained with no data assimilation because of the overestimation of the chlorophyll concentrations by the original model. The assimilation of CZCS chlorophyll into the physical‐biological model significantly improved the predictive capability of the model for phytoplankton distributions and associated phytoplankton fluxes. However, it is difficult to determine if the accuracy of other components of the model, such as the nutrient fields, is also improved by the assimilation of CZCS data.