State, Parameter and Noise Estimation for a Coupled Ocean-Atmosphere Model

D. Kondrashov (1), C. Sun (2) and M. Ghil (1,3)

(1) University of California, Los Angeles, USA
(2) NASA Goddard Space Flight Center, USA
(3) Ecole Normale Supérieure, Paris, FRANCE

We extend previous work on applying Extended Kalman Filter for both model state and parameter estimation in an intermediate, nonlinear, coupled ocean-atmosphere model with synthetic data sets. Current work includes adaptive estimation of model-error covariance and assimilating real observational data. The coupled model consists of an upper-ocean model of the Tropical Pacific and a steady-state atmospheric response to the sea surface temperature. The model errors are assumed to be mainly in the atmosphere, i.e. in the wind stress. Preliminary results show that assimilation of observed SSTs allows to track changes in dynamical coupling parameter with the strength of ENSO events, and leads to a better state estimation.

Last Update: March 25, 2005