江淮梅雨对流结构复杂，双线偏振多普勒雷达能够对降雨系统内三维微物理结构进行高时空分辨率观测，所得偏振参量表征云、降水粒子宏、微观特征，有利于加深了解江淮梅雨对流结构。然而偏振参量同化进数值模式的研究工作在国内尚处于一个起步阶段。本研究利用南京信息工程大学C波段全相参双线偏振多普勒雷达获取梅雨期降水资料，开展双线偏振雷达偏振参量的集合-变分混合同化对江淮梅雨期对流降水预报影响的研究。构建及优化双线偏振雷达差分反射率因子、差分传播相位常数偏振量的观测算子，明确适用于我国双线偏振雷达资料混合同化技术的最优参数及方案，通过建立包含冰相态水凝物控制变量的同化算法来进一步改善云微物理初始场的分析，最终本项目将建立适用于我国实际双线偏振雷达偏振参量的集合-变分混合同化系统，为改进我国江淮地区梅雨期对流降水预报水平提供基础。

The convective structures of Meiyu events are complicated. Dual-polarimetric radar could observe the three-dimensional convective microphysics structures of Meiyu with polarimetric variables that characterize the cloud and precipitation microphysical particles. However, the assimilation of dual-pol radar observations in numerical models is still in the prototype stage to better analyze the macro and micro characteristics of the structure of Meiyu. In this project, the polarimetric variables as differential reflectivity and the specific differential phases from the C-band dual-polarimetric radar located at Nanjing University of Information Science and Technology will be assimilated for Meiyu cases using 3DEnVar method for heavy Meiyu events. The optimized parameters of 3DEnVar for assimilating the polarimetric variables will be obtained through case studies after the observation operator constructed and optimized. The microphysics structure is expected to be improved by exploring the schemes of adding ice phase hydrometer variables as additional control variables. Finally the project will construct the data assimilation system that works with the 3DEnVar method, improving the level of the precipitation forecast for the Meiyu season in Yangtze-Huaihe River basin.