Dynamical-statistical methods of meso-scale precipitation forecasting over mountainous terrain

Summary Two diagnostic fine-mesh models for the determination of the vertical wind component over mountainous terrain have been developed. Both models rely on large-scale grid-point data. The first is based on the equivalent-barotropic scheme and uses 500 mbar heights as input data; the second model is built on the omega-equation and makes use of four input levels. Large-scale terms are interpolated from the large-scale grid to a fine-mesh grid with tenfold smaller mesh-width, while orographic terms are computed in the fine-mesh grid only. The slope of the model-topography and frictional effects are taken into consideration. Humidity and thermal convection are ignored. From computed vertical velocities and observed precipitations, an approximate prediction of precipitation is inferred by a regression technique. In the majority of cases, the baroclinic model gave slightly better results than the barotropic model.