Abstract:The raindrop spectrum shape parameter, which has significant spatio temporal distribution features, is set as a constant in the Double moment Bulk Microphysics Scheme (DBMS) using Gamma distribution function. Based on Milbrandt 2 mon (MY) DBMS, four Meiyu precipitation cases are simulated coupled with four empirical relationships between shape parameter (μr) and slope parameter of raindrops which are concluded from the observations of raindrop distribution. The analysis results suggest that μr has some influences on precipitation. Adopting the diagnostic formulas of μr is positively helpful for improving systematic biases of Meiyu rainfall and shows certain correction ability to the localization of rainfall distribution. The good response of heavy precipitation to μr reflects the clear tendency of improvement. Calculated by the empirical formula, μr increases generally. In the mid and lower troposphere, μr decreases with the increase of rainfall intensity. The decline in raindrop water content and the increased raindrop mass weighted average terminal velocity, which is directly related to μr, are the direct causes for the changes of precipitation. The difference in improving degree between the light and heavy rainfalls mainly results from the different response features of their variable fields. The varying trend of warm cloud particle features of severe rainfalls is consistent with that of light rainfalls, but there are considerable differences in their varying degrees. While it rains lightly, the response characteristics of physical fields to μr present similar varying trends and some linear features. However, in the case of heavy precipitation, the cloud microphysical process and ambient condition experience complex interactions with each other and no significant laws can be followed.