NUMERICAL EXPERIMENTS ON METEOROLOGICAL DATA ASSIMILATION USING A SUBOPTIMAL KALMAN FILTER

E. G. Klimova

A serious problem encountered in the application of a Kalman filter algorithm to modern prediction models is a high order of the forecast error covariance matrices used in this algorithm. One of the approaches to the solution of this problem is the use of simplified models for calculating forecast error covariances. A brief overview of the author's works on the study of applicability of the Kalman filter theory to the problem of meteorological data assimilation. Basic principles of constructing simplified models for calculating forecast error covariance matrices are described, different versions of the suboptimal Kalman filter are proposed, and results of numerical experiments on data assimulation are discussed.

Joomla templates by a4joomla