Collaborative Research: CMG: Ensemble Data Assimilation Based on Control Theory

Florida State University, Tallahassee FL

Investigators

Abstract

This project involves research on data assimilation methods for atmospheric and oceanic forecasting. The focus of the work is to develop methods for incorporating non-Gaussian error statistics into the Ensemble Kalman Filter, overcoming the need for one of the assumptions on which widely used operational methods are based. The approach involves the use of nonlinear optimization methods taken from control theory. Once the method has been developed, the performance of the new approach will be compared with that of existing methods. The new scheme will be used to assess the impact non-Gaussian error statistics on forecasts. If successful, the work will provide a method that may improve forecasts of weather and ocean forecasts under certain conditions, particularly extreme conditions.