Application of 4Dvar Data Assimilation for Reanalysis and Short Range Forecast: Circulation of the Chukchi and Bering Seas

We present two sets of results of the variational data assimilation applied for the reanalysis and short-range forecast of the currents in the Chukchi and Bering seas the model used for the reanalysis is designed specifically for the efficient variational assimilation of long-term observations in ocean regions strongly governed by flow through open boundaries and by atmospheric fluxes. The experiments prove that this model can be also applied for the short-range forecast of the ocean currents. In both cases the preparation of the proper first guess solution is important. The data assimilation results demonstrate the ability of the model to reproduce realistic mesoscale variability of the circulation in the regions. The circulation in the Chukchi Sea is reconstructed from various sources of observations, including 2.5 months of velocity, temperature and salinity data from moorings and CTD observations in autumn 1990. Assimilation of mooring velocities allows us to estimate volume, heat and salt transports in the Chukchi Sea and to reproduce periodical reverse of the East Siberian Current, which is an important feature of the Chukchi Sea circulation. The hindcast and forecast of the circulation in the Bering Sea is based on the assimilation of the absolute SSH altimetry, which was initially prepared from the Topex/Poseidon anomalies by the reconstruction of the climatological ocean state. Preliminary results reveal that the assimilation of the SSH data improves prediction of drifter trajectories and transport variations through the Bering Strait.