关键词: 降水量, 时空变化, 线性倾向估计, Mann-Kendall趋势检验法, 相关系数法, 数据适用性, 精度评估, 青藏高原

Abstract: The Qinghai-Tibet Plateau, renowned as the “water tower of Asia” and the “third pole,” exerts a profound influence on water security and the livelihoods of people in China and neighboring countries. This study examines the changes in water resources on the Qinghai-Tibet Plateau and their influence. To carry out this analysis, remote sensing and reanalysis data from 103 meteorological stations surrounding the plateau, as well as eight types of satellite remote sensing data from domestic and international sources, are utilized. Linear trend estimation, Mann-Kendall trend tests, and correlation coefficient methods are employed to investigate the trend of precipitation change on the Qinghai-Tibet Plateau from 1980 to 2019, as well as the suitability of each dataset. The applicability of these eight satellite remote sensing and reanalysis data is assessed by using correlation coefficients (R), relative errors (BIAS), and root mean square errors (RMSE). The findings are as follows: 1) the eight precipitation datasets can generally depict the spatial distribution of precipitation on the Qinghai-Tibet Plateau, although significant differences in accuracy exist. Among them, Yang Kun’s CFMD dataset exhibits the highest quality, with a relative error of 13.64%. 2) Various meteorological data reveal a significant increase in overall precipitation over the past four decades on the Qinghai Tibet Plateau, covering an area of 12.8%-69.82%. However, arid regions with complex topography or low precipitation display higher relative errors. Additionally, the scarcity of observation stations greatly affects dataset quality. 3) During the past four decades, approximately 66% of the Qinghai-Tibetan Plateau witnessed an upward trend in precipitation, particularly in the central and northern regions. Conversely, the Yarlung Zangbo River, Nujiang River, Lancang River, and the lower reaches in the source region of Yangtze River in the southeastern part of the plateau experienced a significant downward trend. 4) Among the plateau’s watersheds, the source area of the Yellow River observed the most rapid increase in precipitation, at a rate of approximately 5 mm/year. Conversely, the Nujiang River Basin and the lower reaches of the Yarlung Zangbo River Basin experienced the fastest decrease, exceeding 9 mm/year.

Key words: precipitation, temporal and spatial variation, linear trend estimation, Mann-Kendall trend tests method, correlation coefficient methods, data applicability, accuracy evaluation, Qinghai-Tibet Plateau