Search

The distribution of precipitation in alpine regions with cold climates, such as Upper Reach of Shule River Basin (URSRB), northeastern edge of the Tibetan Plateau, is notoriously difficult to estimate from conventional measurements. Nine precipitation products, including CMFD, CMORPH, ERA-Interim, GPCC, GPCP V2.3, JRA-55, TRMM 3B43, HAR, and CAPD, were evaluated by six alpine ground-based measurement stations from 2001 to 2013 on yearly, seasonal, and monthly scales, respectively. The precipitation gradients of the nine precipitation products in rainy period were calculated by linear regression and then compared with recently observations during 2013–2016 from numerous total precipitation gauges. CMFD has the best overall performance in both yearly, seasonal, and monthly precipitation because it is primarily based on the merged precipitation product from surface measurements and satellite remote sensing data, followed by GPCC and TRMM 3B43. It is also found that CMORPH shows the worst overall performance due to only derived from remote sensing data. Nine precipitation products all showed obvious precipitation gradients. The absolute precipitation gradient of GPCC exhibited the greatest similarity to observations, followed by JRA-55 and CAPD. Although HAR obviously overestimated precipitation, it had an accurate pattern of monthly distribution and relative precipitation gradients in alpine areas, indicating that it has some advantages for applying on Tibetan Plateau. The evaluation suggested that the uncertainty in estimated alpine precipitation has largely decreased in some precipitation products, which will help to comprehensive understanding the climate change and its impact on hydrology of northeastern basins of the Tibetan Plateau. [ABSTRACT FROM AUTHOR]