The spatial downscaling of TRMM precipitation data for the middle part of the Chinese Tianshan Mountains

Abstract

Remote sensing data makes great contribution to complements the inadequacy of the instruments in the region, especially at the high elevation and with complex underlying surfaces. In the paper, downscaling models based on the relationships among the precipitation, elevation, and vegetation are used. The multiple linear regression models are established with Tropical Rainfall Measuring Mission (TRMM) 3B43 dataset, the Digital Elevation Model (DEM) from the Shuttle Radar Topography Mission (SRTM), and Normalized Difference Vegetation Index (NDVI) at a high various spatial scales. The spatial resolution of TRMM 3B43 precipitation fields were downscaled from 0.25° to 1 km during 2014–2016 in the middle part of Chinese Tianshan Mountains. The results of downscaling show better agreements with the ground measurements. The downscaling results reduce the underestimation for TRMM precipitation at Bayanbulak and Balun, and it also reduces the overestimation for TRMM at Korla. However, at Yining, the differences value between the measured data and downscaling results are increased in 2014 and 2015, whereas precision of downscaling is increasing in 2016. In general, spatial downscaling can improve the precision of TRMM precipitation data. Therefore, spatial downscaling method is a feasible approach for studying the precipitation of Chinese Tianshan Mountains; it will provide more precise precipitation data sources.