Assimilation of coastal and open sea biogeochemical data to improve phytoplankton simulation in the Mediterranean Sea

The high spatial and temporal variability of biogeochemical features, induced by local dynamics and terrestrialand atmospheric inputs in shelf seas, are challenging issues for the implementation of data assimilation in theseareas. The objective of this study is to integrate satellite ocean-colour observations with a coupled physicalbiogeochemicalmodel in order to improve the spatiotemporal descriptions of chlorophyll and other biogeochemicalvariables in the Mediterranean shelf seas. We adopted a specifically developed three-dimensionalvariational data assimilation scheme for the assimilation of satellite chlorophyll data. The non-homogeneousvertical component, the non-uniform and direction-dependent horizontal component of the background errorcovariance are the key features of the upgraded three-dimensional variational data assimilation scheme for shelfseas. The application of the new assimilation scheme significantly improves chlorophyll estimates in shelf seas,particularly in the representation of their spatial and temporal variability. Based on these results, we provide anestimate of the annual primary production of the Mediterranean basin.