Parameter estimates of a zero-dimensional ecosystem model applying the adjoint method

Assimilation experiments with data from the Bermuda Atlantic Time-series St udy (BATS, 1989-1993) were performed with a simple mixed-layer ecosystem mo del of dissolved inorganic nitrogen (N), phytoplankton (P) and herbivorous zooplankton (H). Our aim is to optimize the biological model parameters, su ch that the misfits between model results and observations are minimized. T he utilized assimilation method is the variational adjoint technique, start ing from a wide range of first-parameter guesses. A twin experiment display ed two kinds of solutions, when Gaussian noise was added to the model-gener ated data. The expected solution refers to the global minimum of the misfit model-data function, whereas the other solution is biologically implausibl e and is associated with a local minimum. Experiments with real data showed either bottom-up or top-down controlled ecosystem dynamics, depending on t he deep nutrient availability. To confine the solutions, an additional cons traint on zooplankton biomass was added to the optimization procedure. This inclusion did not produce optimal model results that were consistent with observations. The modelled zooplankton biomass still exceeded the observati ons. From the model-data discrepancies systematic model errors could be det ermined, in particular when the chlorophyll concentration started to declin e before primary production reached its maximum. A direct comparision of me asured C-14-production data with modelled phytoplankton production rates is inadequate at BATS, at least when a constant carbon to nitrogen C:N ratio is assumed for data assimilation. (C) 2001 Elsevier Science Ltd. All rights reserved.