Application of the Fokker Planck Equation to Data Assimilation in a Dynamic Ocean Circulation Model

Resumo: The most popular data assimilation scheme based on the Kalman Filter approach is considered. This scheme uses the Fokker Planck equation to calculate the time evolution of the covariance matrix of the error, where the error is the difference between the numerical model and observations. The present method does not require linearity of the model operator and is computationally efficient. Some theoretical aspects are discussed. Results of applications of the method in conjunction with the ocean circulation model MOM_2 and PIRATA in situ observational data are presented.