GPM卫星降水产品在“7·21”河南极端暴雨过程中的误差评估

Assessment on error of GPM satellite-based precipitation products during“7·21”extreme rainstorm in Henan

为评估卫星定量降水估测(Quantitative Precipitation Estimation,QPE)产品在气象预测以及灾害预警监测的能力,研究针对台风“烟花”给河南省带来的极端降水过程,以地面雨量站观测数据为参考,采用相关系数(CC)、相对偏差(RB)、均方根误差(RMSE)、分数标准误差(FSE)、探测率(POD)、误报率(FAR)以及临界成功指数(CSI)这7种精度评价指标和统计方法分析和评估全球降水测量计划(Global Precipitation Measurement,GPM)的多卫星融合降水产品中的IMERG(Integrated Multi-satellitE Retrievals for GPM)Version 06中的准实时产品IMERG_ER和近实时产品IMERG_LR,以及全球卫星降水制图(Global Satellite Mapping of Precipitation,GSMaP)Version 07中的准实时产品GSMaP_NRT和标准产品GSMaP_MVK。与雨量站观测结果直接对比,IMERG_ER、GSMaP_NRT和GSMaP_MVK在不同程度上低估降水量,RB值分别为-7.22%、-10.37%和-16.26%,表现为明显低估了强降水中心的雨量。结果表明:(1)IMERG_LR卫星降水产品总体表现最优,对极端降水事件的监测有一定的探测潜力;(2)四种卫星降水产品存在明显提前预估峰值的现象;(3)各个卫星产品在降水量小时高估,在降水量大时低估。

With global climate change,the frequency and intensity of extreme precipitation events are gradually increasing on global and regional scales.Impacted by Typhoon“In-Fa”in 2021,Henan Province encountered an extreme heavy precipitation storm.The object of paper is to evaluate the capabilities of satellite quantitative precipitation estimation(QPE)products in meteorological prediction and the alarming and monitoring of disaster through the“7·21”extreme rainstorm in Henan.In this research,gauge observations are used as a reference to evaluate the performance of version 6.0 Integrated Multi-satellitE Retrievals for Global Precipitation Measurement(IMERG)products and version 7.0 Global Satellite Mapping of Precipitation(GSMaP)products.Seven skill indexes and statistical methods are used to assess the accuracy of these two kinds of satellite products.These scores include Correlation Coefficient(CC),Relative Bias(RB),Root-Mean-Squared Error(RMSE),and Fractional Standard Error(FSE),Probability of Detection(POD),False Alarm Ration(FAR)and Critical Success Index(CSI).Compared with the gauge observations,IMERG_ER,GSMaP_NRT and GSMaP_MVK underestimate the rainfall,with RB values of-7.22%,-10.37%and-16.26%,indicating that they significantly underestimate the precipitation in the center of heavy rainfall.The results show that:(1)IMERG_LR performs better than other three satellite precipitation products,and it has promising potential to detect extreme precipitation events;(2)The four satellite precipitation products estimate precipitation demonstrate the precipitation peaks in advance of 1-2 hours.(3)Those satellite-based QPE products generally overestimate precipitation when the precipitation rate is small,and show underestimation when the rainfall volume is large.