高级搜索

雷达反演的多尺度风场在临近预报中的应用研究

A study of the multi-scale winds from radar retrieval for nowcasting

  • 摘要: 基于多尺度IVAP(MIVAP,the Multi-scale Integrating Velocity Azimuth Process)方法反演风场进行临近外推的预报方法原理可以概括为:首先采用大尺度的分析单元通过IVAP(Integrating Velocity Azimuth Process)方法得到反演风场作为引导气流,表示雷暴的系统性移动,再通过小尺度的分析单元来反演风场,作为雷暴内部移动信息的表征,最后将不同尺度的风场进行合成得到外推的运动矢量场以进行临近预报。相比传统的多尺度TREC(MTREC,the Multi-scale Tracking Radar Echoes by Correlation)方法依赖于追踪回波过去的变化,需要相邻一个或者几个时次的雷达回波来跟踪雷达回波的移动,使用MIVAP方法外推预报只需要最新时刻的雷达反射率因子和径向速度。通过个例分析和统计分析,并且与MTREC方法进行对比试验,结果表明:MIVAP方法反演风场用于临近外推预报具有可行性,MIVAP方法具有实际应用的意义;MIVAP方法得到的外推运动矢量场方向要比MTREC方法方向总体整体偏右,且外推的平均速度要比MTREC方法快;MIVAP方法在30 min以内前几个时次的预报效果略低于MTREC方法,但是30 min以后的预报评分明显优于MTREC方法,MIVAP方法的检验评分整体要优于MTREC方法。

     

    Abstract: A new method to obtain multiple scale wind retrievals is proposed by using the Multi-scale Integrating Velocity Azimuth Process (MIVAP) method with different scale analysis sectors for nowcasting. The retrieved winds from the IVAP method with the large-scale analysis sectors are considered as the steering winds involving the systematic movement, while the small-scale internal motion vectors are estimated by the IVAP method with small scale analysis sectors. Then the retrieved winds with different scales are synthesized as the motion vectors for extrapolation nowcasting. Compared to the traditional Multi-scale Tracking Radar Echoes by Correlation (MTREC) methods, which rely on tracking radar echoes of a few time slices, the MIVAP method only needs the radar echo and radial velocity observations at the latest time. Comparison experiments based on a series of convective rainfall cases demonstrate that it is practical to apply the retrieved winds from the MIVAP method for nowcasting. The statistical verification tests show that extrapolated motion from the MIVAP method appears at the right of that from the MTREC method, and the speed of the former is faster than the latter. Nowcasting using the MTREC method performs better than that applying the MIVAP method in the few periods of the first 30 minutes, while the MIVAP method achieves better evaluation scores than the MTREC method after 30 minutes of forecasting periods, so the MIVAP method has a better performance than the MTREC on the whole.

     

/

返回文章
返回