利用数值预报系统Weather Research Forecast Model（WRF）与三维变分同化系统Data Assimilation（WRF-DA）， 通过控制方案（Con）、 NOAA-19方案（MHS）和FY-3C方案（MWHS-2）， 研究了FY-3C搭载的Micro-Wave Humidity Sounder 2（MWHS-2）和NOAA-19（National Oceanic and Atmospheric Administration-19）的Microwave Humidity Sound‐er（MHS）微波湿度计资料同化了雅鲁藏布江大峡谷暴雨模拟预报的影响。结果表明： 利用WRF-3DVAR（Three Dimensional Variation）同化MHS与MWHS-2微波辐射资料的模拟， 改善了降水的落区位置， 但MWHS-2试验降水落区更偏北； 同化使得水汽场的落区明显改善， 但相较于落区的改善， 其对强降水量级的改善作用较小。同化增强了700 hPa南北风分量， 加大了研究区域水汽的输送强度， 有利于水汽聚集。同化也改善了温度场， 如700~400 hPa层形成具有不稳定性的垂直温度场结构， 有利于降水产生和发展。总之， MHS试验的模拟结果优于MWHS-2， 主要体现在风场、 温和湿度场。此外， MWHS-2试验的24 h预报均方根误差变化较稳定， 说明该数据更有利于中后期的模拟。

This study uses the Weather Research Forecast Model （WRF） numerical forecast system and the Three-Dimensional Variational Data Assimilation （WRF-DA） system to investigate the impact of assimilating data from the Micro-Wave Humidity Sounder 2 （MWHS-2） onboard FY-3C and the Microwave Humidity Sounder （MHS） from NOAA-19 （National Oceanic and Atmospheric Administration-19） on the simulation and prediction of heavy rainfall events in the Yarlung Zangbo Grand Canyon.Three assimilation schemes are compared： the control （Con） scheme， the NOAA-19 scheme （MHS） and the FY-3C scheme （MWHS-2）.The results indicate that assimilation of MHS and MWHS-2 microwave radiance data using WRF-3DVAR （Three-Dimensional Variation） improves the simulation performance compared to the Con experiment.It improves the accuracy of the precipitation location， although the MWHS-2 experiment shows a northern bias in the precipitation area.Satellite data assimilation significantly improves the moisture field， but its effect on heavy rain intensity is less pronounced than its effect on precipitation area improvement.Data assimilation enhances the 700 hPa meridional wind component， leading to increased moisture transport within the study area.With respect to temperature， the assimilation of satellite microwave moisture data has a moderately positive effect， which forming an unstable vertical temperature structure in the 700~400 hPa layer， conducive to the generation and development of precipitation.Overall， the simulation results of the MHS experiment outperform those of MWHS-2， especially in the wind field， temperature and humidity fields.In addition， the root mean square error changes in the 24-hour forecast of the MWHS-2 experiment are relatively stable， indicating that MWHS-2 satellite data are more advantageous for medium to long-term simulation studies.