Comparison of the forecast background errors for nested grids in COAMPS
Bogumil JAKUBIAK
Warsaw University, Poland
Abstract
In this study an estimation of the background error covariances will be presented for selected nested grids of the mesoscale NWP system. In data assimilation schemes the background error covariances determine the filtering and the propagation of the information included in observations. Formalism of the method applied follows the Kalman filter approach in that sense, that whole algorithm is devided into a forecast and analysis step. In the forecast step the expression for the background error evolution is used, in the analysis step the equation that transform the background and observation errors into analysis errors is used. The model used is the COAMPS system of the US Navy. The set-up consists of four computational grids: 37x37 points with spatial step of 54km, 55x55 points with 18km step, 85x85 points with 6km step and 115x115 points with 2km spatial step. All four grids has 35 vertical leveles with variable vertical resolution. Model uses lateral boundary conditions from global NOGAPS fields. Presentation will concentrate on spatial distribution of forecast error variances and estimation of some unknown parameters of the error covariances.