基于随机森林模型的小时强降水订正分析  

Hourly Precipitation Forecasting Correction Based on Random Forest Model

作　　者：王彬雁 王佳津 曹萍萍 杨康权 陈朝平 肖递祥 WANG Binyan;WANG Jiajin;CAO Pingping;YANG Kangquan;CHEN Chaoping;XIAO Dixiang(Sichuan Provincial Meteorological Observatory/Heavy Rain and Drought-Flood Disasters in Plateau and Basin Key Laboratory of Sichuan Province,Chengdu 610072,China)

机构地区：[1]四川省气象台/高原与盆地暴雨旱涝灾害四川省重点实验室,成都610072

出　　处：《高原山地气象研究》2023年第3期91-98,共8页Plateau and Mountain Meteorology Research

基　　金：四川省科技计划重点研发项目(2022YFS0542);高原与盆地暴雨旱涝灾害四川省重点实验室科技发展基金项目(SCQXKJYJXMS202214);中国气象局创新发展专项(CXFZ2021J027);四川省气象局智能网格预报创新团队;中国气象局西南区域气象中心成渝双城经济圈气象服务技术创新团队;中国气象局复盘总结专项(FPZJ2023-12);国家重点研发计划项目(2021YFC3000900)。

摘　　要：基于2020年7—8月西南区域高分辨率模式输出的物理量因子,以及四川省区域自动气象站逐小时降水资料,在分区基础上,采用随机森林模型进行小时降水订正,检验该方法在本地的可用性。结果表明:(1)建模前,对全省进行分区有利于改进大量级小时强降水预报。(2)随机森林模型得到的小时强降水分布整体趋势与原模式一致,对模式未预报出的小时强降水也有一定反应,且实况未出现小时强降水的区域订正后量级有所减小,但仍存在大量级降水分布不均以及空报较多的情况。Based on the physical factors output by SWC-WARMS and hourly precipitation of automatic stations in Sichuan province from June to August 2020,the random forest model was used to forecast hourly precipitation for the rainfall process,and TS score was conducted to understand the local availability of this method.The results show that:(1)Partition of the province is conducive to the improvement of hourly heavy precipitation.(2)The overall trend of hourly heavy precipitation forecast obtained by the random forest model is consistent with that of the original mode,which has a certain response to the hourly heavy rainfall not predicted by the model,and the precipitation magnitude of the regions without actual heavy rainfall has decreased after correction,but there are a lot of uneven distribution of precipitation and many false predictions.

关 键 词：随机森林 短时降水预报 TS检验 西南区域模式 

分 类 号：P456.7[天文地球—大气科学及气象学]