Improvement of the multi-source weighted-ensemble precipitation dataset and application in the arid area of Tianshan Mountains, central Asia

• Published in: Advances in space research Vol. 72; no. 2; pp. 327 - 348
• Main Authors: Li, Xiaoqian, Li, Xiaolong, He, Xinlin, Du, Yongjun, Yu, Jiawen, Yang, Guang, Li, Jiangyue, Li, Dongbo, Xu, Wenhe
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 15.07.2023
• Subjects: Arid area; Data fusion; MSWEP; Precipitation; Spatial heterogeneity; MSWEP; Arid area; Spatial heterogeneity; Data fusion; Precipitation
• ISSN: 0273-1177; 1879-1948
• DOI: 10.1016/j.asr.2023.03.008

With high resolution and wide coverage, satellite precipitation products like the Multi-Source Weighted Ensemble Precipitation (MSWEP) could support hydrological research in arid regions, where the spatial heterogeneity of precipitation is high, but rain gauges are sparse, but these products are inevitably subject to errors. In this study, we integrated measured precipitation and topographic data, and improved MSWEP by interpolation and an optimal regression model. The Tianshan Mountain area was taken as a typical arid area, and spatial stratified heterogeneity of precipitation was analyzed. The results show that: (1) The optimal regression model combined the advantages of site interpolation and MSWEP and improved the underestimation of precipitation in mountainous areas; (2) The precipitation decreased from West to East, and from North to South, and the precipitation in West and East subregions increased linearly with the elevation. The maximum precipitation height zone in the North is located at 4300 m. There are two maximum precipitation height zones in the South; (3) The accuracy of spatial interpolation of station data was limited by the uneven distribution of the stations in the East subregion, and the inaccuracy of MSWEP was due to the land cover and topography of the North and South subregions. The improved MSWEP (IMMSWEP) not only resolved the problem of lack of rainfall stations in arid area, but also resulted in an inversion of regional spatial precipitation differentiation with high resolution and high precision, providing a reference for meteorological researchers.