Evaluation of the TRMM 3B42 and GPM IMERG products for extreme precipitation analysis over China

摘要

Accurate estimation of extreme precipitation is vital for the prediction of hydrologic extremes and flood risk management. Recent satellite-based precipitation products provide important alternative sources of data for such applications, yet their quality and applicability with respect to extreme precipitation have not been studied sufficiently. In this study, the performances of the Tropical Rainfall Measuring Mission (TRMM) 3B42 and Global Precipitation Measurement Integrated Multi-satellitE Retrievals (GPM IMERG) data in extreme precipitation estimation were evaluated over China. Both annual maximum precipitation and extreme rainfall events exceeding the 90th percentile were examined and compared with gauge measurements for the periods of 2000-2017 and 2014-2017. It was found that: (1) both satellite products captured the spatial pattern of extreme precipitation well over China with an overall underestimation for extreme rainfall rate and an overestimation for annual total extreme precipitation volume; (2) TRMM 3842 data had limited ability to detect extreme rainfall events, while GPM IMERG performed slightly better; (3) both products produced good estimation of extreme precipitation with short-medium recurrence intervals, but exhibited consistent underestimation at all return periods; (4) GPM IMERG outperformed TRMM 3B42 for nearly all evaluation metrics when compared over the same time period; (5) the performances were better in south and east China with humid monsoon climate, than in arid west China with high altitude, indicating a significant influence of topography and climate. Our results indicated high potential of satellite products to represent the spatial pattern, overall volume and probability characteristics of extreme precipitation over China, and revealed the general superiority of GPM IMERG to TRMM 3B42. Meanwhile, more studies are still needed to validate data in regions with complex topography and dry climate, and further improve the retrieval algorithm to better support disaster risk reduction and other hydrological applications, especially in areas with a sparse gauge network.