多普勒天气雷达反射率资料的三维变分同化试验

Study on Doppler Weather Radar Reflectivity Data Assimilation in WRF 3D-Var

基于美国新一代数值预报模式WRF(The Weather Research and Forecasting)及其3DVAR(Three-Dimensional Variational)同化系统,采用3DVAR直接同化雷达反射率资料,对2010年10月8日发生在云南省的暴雨过程进行三维变分同化和数值模拟试验。考察了采用不同水平分辨率和垂直层次的雷达反射率进入同化系统对暴雨预报带来的影响。结果表明:同化雷达资料相对无同化任何资料,雨带位置南移,更接近实况降水。同化不同水平分辨率的雷达反射率资料,其中水平分辨率为0. 1度的反射率资料同化后,模拟的降水相对其他方案更接近实况。垂直方向上选取3500m和6000m高度的资料,对整个降水带位置和降水强度影响较大。在此个例中选取垂直层次在1000m,2000m,3000m,3500m,4000m,5000m,7000m,8000m的雷达反射率资料,模式模拟的降水更接近实况。

Based on the Weather Research and Forecast model(WRF)and three-dimensional data assimilation system(WRF 3D-var),assimilating radar reflectivity factor directly,on October 8,2010 occurred in the process of rainstorm in Yunnan province,3D-var and numerical simulation experiment was carried out.Results show that the assimilation radar data without any data assimilation,relatively stretching south location,relatively close to precipitation.Different horizontal resolution radar reflectivity data assimilation,the horizontal resolution of 0.1 degrees after the reflectivity of data assimilation,simulated rainfall relative to other solutions closer to observation.Vertically to choose those of 3500 m and 6000 m height,greater influence on the location of the precipitation and precipitation intensity.In this case a"vertical"level within the selected in the 1000m,2000m,3000m,3500m,4000m,5000m,7000m,8000m of radar reflectivity data,model simulated precipitation closer to observation.