Assimilation of ERBE Data with a Nonlinear Programming Technique to Improve Cloud-Cover Diagnosis

Abstract

A method is developed to assimilate satellite radiation data for the purpose of improving fractional cloud cover diagnosis within a numerical weather prediction (NWP) model. The method employs a nonlinear programming technique to find a set of feasible parameters for the diagnosis such that the difference between the observed and model produced outgoing longwave radiation (OLR) is minimized. The theoretical basis and methodology of the method are provided. The method has been applied in two forecast experiments using a NWP model. The results from the winter case demonstrated that the root-mean-square (RMS) difference between the observed and forecast OLR can be reduced by 50% when the radiatively optimized cloud diagnosis is used, and the remaining RMS error is within the background noise. The optimized diagnosis also reduced the RMS error in the summer experiment, but the reduction is inadequate, possibly because of the inability of the current cloud diagnosis in handling convective activity. The optimization procedure is both stable and sensitive. The largest impact of the optimized cloud diagnosis is that on the forecast of surface temperature, while the impact on forecast of other model variables is insignificant, partly due to the model's simplified cloud process, and partly due to the limited model integration time compared to the time scale of radiative forcing. Several ways in which the method can be used are discussed.