采用WRF中尺度模式及其三维变分同化系统WRF-3DVAR，对2014年5月24—25日发生在萍乡、宜春地区一次大暴雨天气过程进行了雷达资料直接同化敏感性试验，并对同化结果和模式降水预报进行了对比分析。结果表明：1）雷达资料的同化能有效改善初始场中风场、湿度场和热力场的量值大小及空间分布，使其包含更多的中尺度系统信息。其中，雷达反射率因子的同化对大气湿度场和温度场的影响较大，而雷达径向速度的同化作用主要是调整大气风场。2）短时降水预报方面，雷达反射率因子的同化对0—12 h短时降水预报的量值影响较大，而径向速度的同化对降水预报的落区影响较大，两者同时的同化对降水预报效果较好。3）雷达资料的同化对较长时间（6—12 h）的降水预报仍有较好效果。%Based on the mesoscale model WRF and its three dime nsional variational assimilation system WRF-3DVAR , some sensitivity experiments were carried out to analysis the improvement of model precipitation forecast by directly assimilating Doppler radar reflectivity and radial velocity data to a heavy rainfall case occurred in Pingxiang and Yichun areas on May 24 , 2014. The results showed that:1) assimilation of radar data can effectively improve the magnitude and spatial distribution of the initial field in terms of atmospheric wind, humidity and thermal fields due to more mesoscale information included. Assimilation of radar reflectivity had a greater impact on the atmospheric humidity and temperature fields , while assimilation of radar radial velocity majorly adjusted the atmospheric wind field. 2) for the short-time (0—12 h) precipitation forecast, assimilation of radar reflectivity played an important role in the precipitation amount forecast, while assimilating radar radial velocity influenced on the distribution of the rainfall. And the accumulated precipitation was forecasted with better rainfall pattern when radar reflectivity and radial velocity data were assimilated into the model at the same time. 3) The rainfall forecast of 6—12 h can be improved by assimilating radar data.
展开▼
