Deriving four parameters from patchy observations of ocean color for testing a plankton ecosystem model

The success of ocean-color observations from space has raised interests in the application of this technology to ecosystem models. We investigated the time series of SeaWiFS data near the Azores during the first half of 1998, with the intention of testing a plankton ecosystem model. Results show that the data are very patchy, due to mesoscale variability and cloud mask gaps. The general approach of taking averages of these patchy observations introduces a bias and provides only limited information. Based on our knowledge and experience with spring blooms for characteristic subdivisions of the world's oceans, we propose a new approach: apply a four-parameter Gaussian curve fit to the gappy time series at each grid, and then extract for four parameters. These are the background chlorophyll, the increase in chlorophyll at the height of the bloom, the timing of the maximum, and a width parameter related to the duration of the bloom. Histograms of these parameters can then be compared to similar measures derived from an ecosystem model. This provides an effective way to test the model and compare its bloom dynamics with that of the satellite observations.