资源环境科技发展态势分析平台

The objective of this study is to comprehensively evaluate the Integrated Multi-satellite Retrievals for Global Precipitation Measurement Mission (IMERG), especially on heavy and extreme rainfall events, by using 144 rain gauge observations at daily/hourly scale from April 1, 2014 to December 31, 2017 over the Sichuan Province, China. Terrain in the study area is very complex which is composed of mountains, hills, plain basins and plateaus. The evaluation was conducted based on general assessments, graphical comparisons, precipitation detection capability, probability density function, extreme indicators and one specific extreme precipitation event.

The results show that: For the general analysis, IMERG_F tends to overestimate rainfall events especially light rainfall events, which is contrary to IMERG_E and IMERG_L. The precipitation detection ability of IMERG is influenced by the rainfall type and topography. For the extreme indicator analysis, IMERG has a better estimation of the rainfall intensity than the precipitation detection ability. For one specific extreme rainfall event, all the three IMERG products could capture the rainfall center and basic spatial changes. But the amount of rainfall has a bigger uncertainty especially at the beginning and the end of a rainfall. The Final Run product has a better performance, but the difference is insignificant. In addition, when the interval time of two heavy rainfall events is small, IMERG may regard them as one event. These findings provide valuable feedbacks to satellite heavy rainfall studies in complex terrain region regarding the development of the IMERG algorithm and data use.