Application of the ROMS Incremental Strong 4D-Variational data assimilation in the California Current System

Gregoire Broquet (UCSC), Christopher A. Edwards (UCSC), Andrew M. Moore (UCSC), Brian S. Powell (University of Hawaii), Milena Veneziani (UCSC), James D. Doyle (NRL), Hernan G. Arrango (Rutgers), Javier Zavala-Garay (Rutgers)

The Incremental Strong constraint 4D Variational (IS4DVAR) algorithm of ROMS is used to study the impact of data assimilation on a realistic, high resolution model of the California Current System. The model is forced with regional COAMPS atmospheric data and with ECCO data at the open boundaries. Climatological fields and both satellite-derived surface and in situ observations are assimilated to significantly improve many characteristics of the circulation dynamics. The parameterization of the background error statistics are shown to be particularly critical to providing a compatible and consistent use of these different observations. Additionally, the use of ROMS-IS4DVAR to adjust surface forcing provides sensible corrections to the wind stress and heat flux improving the results even further.