基于雷达回波拼图资料的风暴单体和中尺度对流系统识别、跟踪及预报技术

以中国气象科学研究院灾害天气国家重点实验室的区域雷达组网三维数字产品作为数据输入,在雷达基数据的SCIT(The Strom Cell Identification and Tracking)算法基础上,完成了三维格点风暴单体识别、追踪和预报,用Davis发展的客观诊断评估方法识别雷达拼图资料中的中尺度对流系统,实现了雷达数据的中尺度对流系统识别、跟踪和预报,并利用这两种方法对多个强天气过程进行风暴和中尺度对流系统识别、跟踪及预报.在单雷达区域内用原SCIT和修改后的SCIT算法做了风暴单体定量识别检验.结果表明,(1)修改后的SCIT算法能够实现三维风暴的自动识别、跟踪和预报,在单雷达区域内与原算法识别风暴数量大体相当,中尺度对流系统识别方法能够实现中尺度对流系统的自动识别,并完成跟踪和预报；(2)SCIT算法预报误差较小,中尺度对流系统算法预报误差相对较大,它们的预报误差随时间延长而增大.

A new techniques for strom cell and mesoscale convective systems identification, tracking and nowcasting based on the radar mosaic data

The SCIT (The Strom Cell Identification and Tracking) based on the radar mosaic 3D data developed by State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences is presented for automatic identification,tracking and nowcasting of storms. A technique based on the MODE (Method for Object based Diagnostic Evaluation) developped by Davis is presented for automatic identification tracking and nowcasting of MCSs (Mesoscale Convective Systems). The performance of the detection and nowcasting are evaluated via applying it to several convective cases. The results show that the new SCIT could achieve automatic detection, tracking and nowcasting. It is considered that the revised algorithm shows a similar capacity in identifying storms to the original one for a single radar area. The forecast position of a strom is better than MCSs.