一种基于静止卫星观测数据的实时降水识别方法研究  

Research on a Real-time Precipitation Recognition Method based on Geostationary Satellite Observation Data

作　　者：崔梦圆 姬大彬 贾立 郑超磊 蒋卫国 CUI Mengyuan;JI Dabin;JIA Li;ZHENG Chaolei;JIANG Weiguo(Key Laboratory of Remote Sensing and Digital Earth,Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100101,China;University of Chinese Academy of Sciences,Beijing 100049,China;Department of Geographic Sciences,Beijing Normal University,Beijing 100875,China)

机构地区：[1]中国科学院空天信息创新研究院遥感与数字地球重点实验室(中国科学院),北京100101 [2]中国科学院大学,北京100049 [3]北京师范大学地理科学学部,北京100875

出　　处：《遥感技术与应用》2024年第4期1000-1012,共13页Remote Sensing Technology and Application

基　　金：风云卫星应用先行计划(FY-APP-ZX-2022.0210);国家自然科学基金项目(42171039)共同资助。

摘　　要：极端降水的准确、实时监测对于提高洪涝灾害的预报具有重要意义,然而当前基于静止卫星降水产品普遍存在降水识别精度较低的问题,严重影响了其在洪涝灾害预警中的应用。基于Himawari-8静止气象卫星观测的红外波段亮温数据、ERA5再分析大气廓线数据以及地面雨量计观测数据,通过实时、动态地建立基于随机森林的降水识别模型,开发了一套适用于静止气象卫星的实时降水识别方法。该方法一方面通过引入实时的地面雨量计小时降水数据,用于实时训练降水识别模型,解决了静态训练的机器学习模型降水识别精度随时间衰减的问题;另一方面通过加入与降水的形成和发展密切相关的大气廓线数据,有效提高了基于静止气象卫星红外数据识别降水的精度。采用中国内地地区2157个地面雨量计站点的小时降水观测数据进行验证,本研究所提出的降水识别算法在小时尺度上的击中率(POD)为0.73,虚警率(FAR)为0.49,成功系数(CSI)为0.43,各指标表现情况均优于实时降水产品GSMaP_NOW和FY4A官方降水估计实时产品QPE。Accurate and real-time monitoring of extreme precipitation is of great significance for improving flood forecasting,however,the current geostationary satellite-based precipitation products are generally character⁃ized by low precipitation recognition accuracy,which seriously affects their application in flood warning.Based on the infrared band brightness temperature data observed by Himawari-8 geostationary meteorological satel⁃lite,ERA5 reanalysis atmospheric profile data and ground rain gauge observation data,this study developed a set of real-time precipitation recognition method suitable for geostationary meteorological satellites by establish⁃ing a real-time and dynamic precipitation recognition model based on random forest.On the one hand,this method solves the problem that the precipitation recognition accuracy of the static training machine learning mod⁃els decays with time by introducing real-time hourly precipitation data of the surface rain gauge for real-time training of the precipitation recognition model.On the other hand,it effectively improves the accuracy of precipi⁃tation recognition based on the infrared data of the geostationary meteorological satellite by adding the atmo⁃spheric environmental condition data closely related to the formation and development of precipitation.The hour⁃ly precipitation observation data of 2157 surface rain gauge stations in Chinese Mainland are used for verifica⁃tion.The proposed precipitation recognition algorithm has a POD of 0.73,a FAR of 0.49,and a CSI of 0.43 on an hourly scale.All indicators are better than the real-time precipitation product GSMaP_NOW and the official real-time precipitation estimation product QPE of FY4A.

关 键 词：降水遥感 降水识别 静止气象卫星 红外亮温观测 ERA5 

分 类 号：P407[天文地球—大气科学及气象学]