The impact of using reconditioned correlated observation error covariance matrices in the Met Office 1D-Var system
Recent developments in numerical weather prediction have led to the use of correlated observation error covariance (OEC) information in data assimilation systems. Ill-conditioned OEC matrices may cause problems with the convergence of a variational data assimilation procedure, meaning that reconditioning methods are used to improve their conditioning.
We apply the ‘ridge regression’ reconditioning method to assimilate Infrared Atmospheric Sounding Interferometer (IASI) observations in the Met Office 1D-Var system. This is the first systematic investigation of the impact of reconditioning parameter choice on convergence of a 1D-Var routine. 1D-Var is used for quality control, and to estimate key variables that are not analysed by the main 4D-Var data assimilation system. We find that the current (uncorrelated) OEC matrix requires more iterations to reach convergence than any choice of correlated OEC matrix, suggesting that using a correlated OEC matrix in the 1D-Var routine would have computational benefits.
However, the impact on retrieved variables is less clear, with a small number of large differences between retrievals for the control and correlated choices of OEC. The improved convergence associated with correlated OEC matrices would allow for more efficient quality control procedures, which is likely to reduce the impact of such large retrieval differences.
Ms. Jemima M. Tabeart University of Reading & National Centre for Earth Observation