2013年“菲特”台风暴雨数值模拟中微物理方案的对比试验

利用TRMM（热带测雨雷达）搭载的TMI（微波成像仪）反演廓线资料，分析“菲特”台风登陆前、后云团内部水凝物的分布种类，依此选择WRF区域中尺度模式下符合条件的6个云微物理过程参数化方案（Lin，WSM6，Godgce，WDM6，Morrison以及Thompson方案），模拟2013年10月6~8日的台风过程.从降水落区、强度，水凝物及风场垂直分布，台风路径及强度等方面对预报性能进行对比，结果表明，选用的6个云微物理方案都较好地模拟了浙江暴雨的范围和强度.结合Ts评分，降水量级越大，模拟效果对云微物理方案选择越敏感，其中，Lin方案效果最佳，尤其对极端降水的模拟，其次为WSM6、WDM6及Thompson方案，Morrison和Godgce方案相对较差.结合水凝物平均值廓线分布发现，除WDM6方案外，其他方案对暖雨过程的模拟基本一致，而对冰相过程的模拟6个方案差别较大；同时，各方案对风分量的模拟结果较水凝物廓线差别小，说明对于动力因素模拟不敏感.另外，6个方案对于台风强度的模拟整体偏弱，相较之下，Lin方案较好地模拟了强度变化趋势.

The effects of different microphysical schemes in WRF on the rainstorm of typhoon Fitow in 2013

Based on the measurements by TRMM Microwave Imager (TMI), we select six cloud microphysics parameterization (MP) schemes(schemes lin, wsm6, godgce, wdm6, morrison, and thompson) in the weather research and forecasting (WRF) model to simulate a typhoon case occurred during October 6-8, 2013. The applicability of these MP schemes to the simulation of the typhoon Fitow is studied accounting for the rainfall area and intensity of precipitation, vertical distributions of averaged hydrometeors and wind and the track and intense, forecast performances. It shows that the simulations results of all selected MP schemes t agree with the observation data of the rainfall area and intensity. According to the analysis results of threat score (Ts), the more actual precipitation, the more sensitive the choice of MP schemes, it seems that the lin scheme has the best performance among the six especially for extreme precipitation while the scheme of morrison and godgce have the worst performance. Referring to the vertical distributions of averaged hydrometeors and wind, we found that the simulation results of all MP schemes (except wdm6) on warm rain processes are basically the same, but a wide range of differences on ice phase process. Meanwhile, all MP schemes are less sensitive to dynamic simulation. Besides, our studies show that all these MP schemes do not have major impact on the track and their effect on the simulated intensity, is underestimated. Overall, lin scheme provides the best tendency forecast of the strength.