The goal of our research is to improve the skill of Quantitative Precipitation Forecasting (QPF) in complex terrain. The Sierra Nevada mountain range (average crest of 2 km and a half width of 100 km) is selected for the study. The precipitation prediction is hindered by many factors in atmospheric modeling such as model resolution, topography, complexity in representation of microphysical processes, and inaccuracy of water substance measurements in the atmosphere and its use in model initialization. Present numerical prediction of precipitation shows that there is a problem of over and under prediction of precipitation amounts at the windward and leeward mountain slopes, respectively. Recent studies have shown that considerable attention has to be given to the representation of microphysical processes, if QPF is to be improved, especially in the regions of complex terrain.

In this study, we focus on the sensitivity of QPF in the Sierra Nevada on the representation of microphysical processes in a mesoscale model. The mesoscale model is MM5, non-hydrostatic, limited area model with multiple nesting capability. A number of heavy winter orographic precipitation cases from the Sierra Co-operative Pilot Project (SCPP) in the 1980's have been selected in this study. In addition to the SCPP cases, the flood event of Dec 1996-Jan 1997 will also be discussed.

High resolution simulations with the finest horizontal resolution of 4.5 km and 29 levels have been carried out for the cases using a number of existing microphysical schemes in MM5. The skill of MM5 model in predicting orographic wintertime precipitation in the Sierra Nevada using the existing microphysical schemes will be discussed based on the validation against the available data from SCPP observational network and the data from the archives of the Western Regional Climate Center at the Desert Research Institute.