Simulating large-scale circulation in seas and oceans

The problem of simulating ocean general circulation is considered using an approach developed by G.I. Marchuk. Most attention is focused on problems associated with the application of multi-component-splitting and adjoint-equation methods and with the parameterization of turbulent-exchange processes. The statement of the problem, an algorithm of its solution, and the results of the four-dimensional temperature and salinity data assimilation in a model of World Ocean hydrodynamics are discussed. This algorithm yields qualitatively reliable results, and its main features are modularity, usability, and the possibility of using efficient implicit schemes.