Super typhoon Rammasun successively slammed into coastal area of south China for three times during 12pm July 18th, 2014 to 12pm July 19th, 2014, bringing extreme storm and causing severe floods.Against this event, five multi-satellite precipitation estimates based on two kinds of mainstream satellite retrieval algorithm(Global Satellite Mapping of Precipitation, GSMaP; Integrated Multi-satellitE Retrievals for GPM, IMERG)were evaluated and intercompared with independent gauge observation networks which are governed by China Meteorological Administration(CMA)by using a variety of statistical indices(NB, CC, RMSE, POD, FAR and CSI)in hourly scale.The assessments indicate that:All multi-satellite precipitation products had an obvious underestimation on rainfall amount during the extreme event, but in some way, the bias were reduced after applying calibration algorithm(-6.8%for GSMaP_ GAUGE and -0.5%for IMERG_ CAL);Errors mainly come from rainfall which over 100 mm as FAR index showing an upward trend while POD and CSI indices beginning to reveal a downward trend in this scope;Moreover,the calibrated products lost pre-cipitation variability characteristic seized at a finer scale due to the use of lower temporal -spatial resolu-tion gauge analysis in calibration algorithm.%针对2014年7月18日12时至2014年7月19日12时"威马逊"台风先后3次在中国华南沿海地区(海南、广东、广西)登陆所带来的暴雨过程,以中国气象局提供的地面观测站点数据作为参考,使用标准化偏差(NB)、相关系数(CC)、均方根误差(RMSE)、命中率(POD)、误报率(FAR)和关键成功率(CSI)等评价指标,对基于2种不同国际主流卫星降水反演算法GSMaP和IMERG的5套高分辨率遥感降水产品进行了小时尺度上的精度验证与分析.结果表明:遥感降水产品在暴雨事件中均表现出不同程度的低估,但地面校正算法较好地修正了遥感降水产品与地面观测数据间的整体偏差[GSMaP_ GAUGE(-6.8%), IME-RG_ CAL(-0.5%)];遥感降水产品的误差主要来自大于100 mm 的强降水,降水量超过100 mm时,5套产品的误报率(FAR)整体呈现上升趋势,命中率(POD)和关键成功率(CSI)呈现下降趋势;由于校正算法的数据来源是低时空分辨率的地面观测数据,校正后的遥感降水产品也在一定程度上丢失了高分辨率下捕捉到的降水变异特征.
展开▼
