A Fully Conserved Adjustment Scheme for Ocean Data Assimilation Systems

Ocean (T, S) data assimilation (optimal interpolation, Kalman filter, and 3D-Var), conducted in the three dimensional physical space, is a generalized average of modeled and observed data. Due to high nonlinearity of the equation of the state of the sea water and non-uniform vertical distribution of the observational profile data, false static instability may be generated. Such a false instability occurs in various reanalyzed ocean (T, S) data using optimal interpolation, Kalman filtering, and 3D-Var. In this paper, we present a fully conserved adjustment scheme for eliminating the false static instability. This method is on the base of conservation of heat, salt, and static stability. Comparison to the existing methods (i.e., convective adjustment and minimal adjustment) shows great improvement. Various examples are provided.