基于雷达资料同化的飑线过程数值模拟试验研究

利用中尺度WRF模式及其3DVAR同化系统对2014年3月30日发生在我国华南地区的一次飑线过程展开多普勒天气雷达资料的同化效果试验研究。首先对雷达资料进行去地物杂波、退速度模糊等预处理,后设计了基于不同雷达观测量的同化试验及同化频次的敏感性试验。结果表明:直接循环同化雷达径向风资料和雷达反射率因子能够增加数值模式中的中小尺度信息,提供可靠的水汽分布;不同的同化频次对同化结果影响显著,每12 min同化间隔的结果略优于30 min、60 min同化间隔;同化雷达反射率因子和径向风资料分别对模式的总水场和风场有显著调整,联合同化雷达反射率因子和径向风资料比单独同化反射率因子或径向风更能改善飑线垂直结构配置,促使地面冷池和雷暴高压配合,进一步改善模式对大雨和暴雨量级降水预报效果。

Numerical simulation of a squall line process based on radar data assimilation

The mesoscale Weather Research and Forecasting(WRF) model and its 3DVAR assimilation system were used to conduct assimilation experiment of Doppler radar data on a squall line event occurred in South China on March 30, 2014. Firstly, the radar data was preprocessed by removing the ground clutter and dealiasing folded radial velocity, and then sensitivity tests were designed to assimilate different radar observations on different assimilation frequency. Results show that the direct circulation assimilation of radial velocity or radar reflectivity factor can add mesoscale and microscale information to the numerical model and provide reliable water vapor distribution; assimilation at different frequency has significant effect on the simulation result. Every 12-minute interval radar data assimilation shows better result than that at every 30-minute or every 60-minute interval. Assimilating radar reflectivity factor or radial velocity can adjust the total water and wind fields, and assimilating both two can improve the simulation of the vertical structure of the squall line more, and reflect the surface cold pool and thunderstorm high, further enhances the precipitation forecasting effect on heavy rain and rainstorm.