基于X波段双线偏振天气雷达的雷暴云粒子识别

An approach to the hydrometeors classification for thunderclouds based on the X-band dual-polarization Doppler weather radar

为了进一步认识雷暴云中影响起电的主要水成物类型,联合714XDP天气雷达观测到的偏振参量(Z_H、Z_(DR)、K_(DP)和p_(HV)),在利用模糊逻辑方法的基础上引入温度(T),对中国西北地区夏季雷暴云中的水成物类型进行识别,将水成物分为11种。参考S和C波段及少量的X波段偏振雷达的粒子识别研究,得到11种粒子对应的各偏振参量阈值。分析了2007年7月24日的一次观测过程几个时刻的偏振参量、地面电场、放电频数及识别结果,并与利用三维雷暴云动力电耦合数值模式模拟出来的粒子分布结果做了对比。结果表明,选取的偏振参量阈值基本合理,采用的模糊逻辑方法能有效、合理地识别出雷暴云内水成物的相态和分布,而且利用K_(DP)可以大致判断云内电荷区。

In order to understand the main hydrometeor types affecting the electrification within a thunderstorm,a fuzzy logic algorithm for the classification of hydrometeor based on Xˉband dualˉpolarization Doppler weather radar measurements is deˉ scribed.Four radar parameters,namely,horizontal reflectivity （Z H ）,differential reflectivity （Z DR ）,specific differential phase （K DP ）and correlation coefficient （ρHV ）,have been used in the algorithm.Hydrometeor is classified into 11 types which have closed relationships with the nonˉinductive electrification mechanism.The thresholds of polarizatic parameters for the 11 hydroˉ meteor types are obtained by refering to the research results of the hydrometeor classification using Sˉ,Cˉand Xˉband dualˉpoˉ larization weather radars.In addition,environment temperature is added to restrict the altitude range of every hydrometeor type.The hydrometeors of the two convection processes taking place on 24 July 2007 in northwestern China are identified based on this algorithm.Furthermore,the characteristics of the radar parameters,ground electric field and flash frequency are anaˉ lyzed preliminarily.Compared with the 3ˉD thunderstorm electrical model,the results show that the hydrometeor types could be identified effectively.It implies the fuzzy sets of polarimetric parameters and temperature ranges determined are basically valid.The charge regions could be inferred roughly based on the characteristics of K DP .All of these results are helpful for furˉ ther understanding of the electrication mechanism within a thunderstorm.