Evaluation of the validation of TRMM data over the region of Qilianshan mountain in Northwest China

Abstract

The Qilianshan Mountain area is very important for water resources and ecosystem safety of the Northwest China and Qinghai-Tibet Plateau. Satellite remote sensing is the best way to estimate precipitation over this region in the future due to the complex terrain and sparse of ground weather stations. The primary goal of this research is to evaluate the Tropical Rainfall Measuring Mission (TRMM) 3B43 rainfall products during 2008 ~ 2017 over the region, by using the gridded precipitation data and routine ground-based observation data from National Meteorological Information Center (NMIC) of China, combining with the Land Use and Land Cover (LULC, MCD12Q1) and topographic data (SRTM). Results show that accuracy of TRMM precipitation has changed a lot except in winter (arid season). Correlation coefficient of TRMM precipitation against the ground-based observations varies from 0.33 to 0.67, indicating that TRMM product is applicable over the Qilianshan mountain area. Seasonal variation of the relative error is mainly in the northeast and southwest areas. The TRMM rain products are greatly affected by topography, and its overestimations are basically distributed in the valley or trough areas. According to analysis of the land use classification, accuracy of the TRMM precipitation is obviously impacted by the sparse vegetation, evergreen broad-leaved forest and city area.