Volume, heat and salt transport in the North-Eastern Bering Sea during 2007-2010 derived through the 4dvar data assimilation of in-situ and satellite observations

The rich collection of BEST-BSIERP observations and other sources of data provide an excellent opportunity for synthesis through modeling and data assimilation to improve our understanding of changes in physical forcings of the Bering ecosystem in response to climate change. Assimilating data of different origins, which may be sparse in space and time, is difficult using simple algorithms (traditional optimal interpolation, correlation analysis etc.).
The 4Dvar approach is effective for performing spatiotemporal interpolation of sparse data via interpolation (covariance) functions with scales based on ocean dynamics (Bennett, 2002).