Quantitative Precipitation Forecasting of Wintertime Storms in the Sierra Nevada: Sensitivity to the Microphysical Parameterization and Horizontal Resolution

Abstract

The skill of a mesoscale model in predicting orographic precipitation during high-impact precipitation events in the Sierra Nevada, and the sensitivity of that skill to the choice of the microphysical parameterization and horizontal resolution, are examined. The fifth-generation Pennsylvania State University?National Center for Atmospheric Research (PSU?NCAR) Mesoscale Model (MM5) and four bulk microphysical parameterization schemes examined are the Dudhia ice scheme, and the Schultz, GSFC, and Reisner2 mixed-phase schemes. The verification dataset consists of ground precipitation measurements from a selected number of wintertime heavy precipitation events documented during the Sierra Cooperative Pilot Project in the 1980s. At high horizontal resolutions, the predicted spatial precipitation patterns on the upwind Sierra Nevada slopes were found to have filamentary structure, with precipitation amounts over the transverse upwind ridges exceeding severalfold those over the nearby deep river valleys. The verification results show that all four tested bulk microphysical schemes in MM5 produce overprediction of precipitation on both the windward and lee slopes of the Sierra Nevada. The examined accuracy measures indicate that the Reisner2 scheme displays the best overall performance on both sides of the mountain range. The examined statistical skill scores on the other hand reveal that, regardless of the microphysical scheme used, the skill of the MM5 model in predicting the observed spatial distribution of the Sierra Nevada orographic precipitation is fairly low, that this skill is not improved by increasing the horizontal resolution of the model simulations, and that on average the quantitative precipitation forecasting (QPF) skill is better on the windward than on the lee side. Furthermore, a significance test shows that differences in skill scores obtained with the four microphysical schemes are not statistically significant.