Abstract: CMA-MESO model is utilized to investigate the impact of assimilating multi-source polar-orbiting satellite microwave radiance data on forecast of "23.7" severe rainstorm event in North China. This event is primarily associated the remnants of Typhoon Doksuri (2305) and is influenced by weather systems, including the subtropical high and Typhoon Khanun (2306). The study utilizes CMA-MESO assimilation system to assimilate radiance data from microwave temperature sounding channels, including ATMS on NOAA-20 and SNPP, AMSU-A on Metop-C, MWTS-2 on FY-3D, and MWTS-3 on FY-3E. The process of preparing satellite microwave radiance data for assimilation involves several steps, including cloud detection, quality control, and bias correction. The cloud detection algorithm utilizes the absolute value of observation minus background (OMB) for 52.8 GHz channel to identify and exclude cloudy observations. Bias correction is implemented using the static bias correction method developed by Harris and Kelly, which addresses both scan-dependent bias and air mass bias. Two experiments are conducted: A control experiment (CTRL), which utilizes only conventional observations (surface, radiosonde, and aircraft observations), and a multi-source microwave temperature sounding channel data assimilation experiment (MTS), which assimilates both conventional observations and the aforementioned satellite data. Results demonstrate that the accuracy of the model's initial conditions is improved through the assimilation of radiance data from multiple microwave temperature sounding channels. Errors in initial temperature fields are minimized, and the representation of weather systems is enhanced through data assimilation. MTS experiment demonstrates enhanced precision in temperature, humidity, and wind forecasts, particularly in the lower and middle troposphere. This improvement subsequently leads to better skills in predicting precipitation. In particular, MTS experiment demonstrates enhancements in 24 h accumulated precipitation forecasts, particularly for heavy rainfall events. There is a notable improvement in critical success index (CSI) for the extreme heavy rainfall threshold, which has increased by 14%. Physical mechanisms that enhance precipitation forecasts are further elucidated through diagnostic analysis. The integration of radiance data from multiple microwave temperature sounding channels provides a more accurate representation of the dynamic, thermal, and moisture conditions in regions prone to heavy rainfall. MTS experiment simulates increased upward motion intensity, enhanced low-level convergence, and upper-level divergence, along with a more favorable distribution of moisture and instability energy. These factors collectively enhance the accuracy of precipitation forecasts.