The 13–14 December 2001 IMPROVE-2 Event. Part II: Comparisons of MM5 Model Simulations of Clouds and Precipitation with Observations

Abstract

This paper compares airborne in situ observations of cloud microphysical parameters with the fifth-generation Pennsylvania State University?National Center for Atmospheric Research (PSU?NCAR) Mesoscale Model (MM5) simulations, using the Reisner-2 bulk microphysical parameterization, for a heavy precipitation event over the Oregon Cascades on 13?14 December 2001. The MM5 correctly replicated the extent of the snow field and the growth of snow particles by vapor deposition measured along aircraft flight tracks between altitudes of 4.9 and 6 km, but overpredicted the mass concentrations of snow. The model produced a broader number distribution of snow particles than observed, overpredicting the number of moderate-to-large-sized snow particles and underpredicting the number of small particles observed along the aircraft flight track. Over the mountain crest, the model overpredicted depositional growth of snow and mass concentrations of snow, but underpredicted the amount of cloud liquid water and conversion of snow to graupel. The misclassification of graupel as snow and excessive amounts of snow resulted in the model overpredicting precipitation on the lee slopes and in localized areas along the foothills of the Cascades. The model overpredicted cloud liquid water over the lower windward slopes and foothills, where accretion of cloud liquid water by rain was the primary precipitation-producing mechanism.