基于雷达回波最大重叠率的一种临近降水预报方法

The Nowcasting Method of Precipitation Based on the Maximum Overlap Rate of Radar Echo

基于天气雷达的临近降水预报是雷达短时预报业务系统的重要组成部分,本文在回顾雷达临近降水预报技术发展的基础上,充分考虑新一代天气雷达高时间、空间分辨率的优点,提出两个相邻时刻雷达回波具有最大重叠率的移向、移速是与降水系统整体的移向、移速密切相关的论点,再用此移向、移速对当前雷达回波进行平移,最后采用九点平均的回波生消模型进行订正,形成逐次外推预报,利用时间权重合成法对每6分钟的逐次外推预报合成1小时的临近降水预报场。文中使用合肥、厦门和南昌雷达站的数据进行计算、预报和检验,结果表明:两个时刻的最大重叠率移向、移速可以反映降水系统整体的移动特征,移向、移速的计算结果和熟练预报员的主观预报相一致;利用生消模型对平移后的降水回波进行订正可以反映实际降水中心和分布情况;采用时间权重合成法,对每6分钟的降水预报合成的1小时的降水分布场和实际的降水分布场也具有较好的一致性。

The nowcasting method of precipitation based on the maximum overlap rate is an important part of shortterm forecast radar system. On the basis of recalling the development of the radar nowcasting technology of precipitation, considering the merits of a new generation of weather radar with high spatial and temporal resolution, this paper proposes that within a short time interval, the moving direction and speed of the two adjacent radar echos with the greatest overlap rate are also the moving direction and speed of the precipitation system. The current radar echoes are translated by calculating the moving direction and speed, and are revised by use of nine-point average change amounts of echo dissipation model to form extrapolation forecasting; by use of the time weights synthesis method, the extrapolation forecasting every six minutes is synthesized to form the precipitation nowcasting field. In order to test the above-mentioned methods, the data at Hefei, Nanchang, Xiamen radar stations are used to carry out the calculation and prediction. The results show that： the moving direction and speed with the maximum overlap rate at two moments may reflect the overall movement characteristics of precipitation system, the calculation result is consistent with the subjective forecasting of forecasters and moreover it is objective, fine, and stable; using the growth and decay model to revise the translated echo can reflect the actual precipitation change; the 1-hour precipitation distribution which is synthesized by using precipitation forecasting every six minutes with the time weights synthesis method is in good agreement with the actual one.