Sensitivity of Summer Precipitation Simulation to Microphysics Parameterization Over Eastern China: Convection-Permitting Regional Climate Simulation

With a six-year (2009-2014) summer climate simulation using the Weather Research and Forecasting model at convection-permitting resolution (4-km grid spacing), the effects of microphysics parameterization (MP) schemes on precipitation characteristics are investigated in this study. The convection-permitting simulations employ three popular MP schemes, namely, Lin (single-moment bulk MP), Weather Research and Forecasting Single-Moment 5-class (one-moment and mixed-phased MP), and Thompson (two-moment and mixed-phase MP) scheme. By evaluating the simulations against the CMORPH, rain gauge (Station), and ERA-Interim data, it is found that the convection-permitting model reproduce well the summer precipitation amount and the associated large-scale atmospheric circulations, which are insensitive to the choice of MP schemes. The simulations with three MP schemes overestimate the precipitation amount, especially over the Yangtze-Huaihe River Valley. The overestimations may be due to the systematic biases, and cannot be significantly reduced by using different MP schemes. Moreover, all simulations capture well the major features of precipitation diurnal variations and their transition characteristics, but they significantly overestimate the precipitation frequency while underestimate the precipitation intensity. The analysis on the microphysical hydrometeors shows that the model-simulated precipitation amount is considerably affected by the vertical profiles of solid hydrometeors, especially the snow and graupel particles. The Thompson scheme creates more snow particles and less graupel than the other schemes, while produces the least precipitation amount that best matches the CMORPH.