Investigation of statistical structure of temperature short-range forecast errors in the atmospheric boundary layer for the purpose of objective analysis

The procedure is described of the estimation of statistical structure of short-range forecast errors of temperature field in the atmospheric boundary layer for the purpose of objective analysis. The numerical experiments on the estimation of forecast error covariances were carried out for the WRF (Weather Research and Forecast, NCEP, USA) model. The comparative experiments were implemented on the estimation of the influence of stability type in the atmospheric boundary layer on the variability of vertical and three-dimensional covariances. It is demonstrated that the variance and radius of correlation of temperature field forecast errors along the vertical as well as the variability of three-dimensional covariance functions in the atmospheric boundary layer differ considerably at different stability types. The results are cited of the numerical experiments on the estimation of the WRF model temperature field forecast error covariances in the atmospheric boundary layer for summer and winter periods.