基于MSWEP降水产品的新疆干旱时空特征分析

Meteorological stations in Xinjiang are sparse and unevenly distributed, resulting in drought monitoring based on in-situ observations in Xinjiang which are insufficient in spatial representativeness. Remotesensed precipitation products have the advantages of wide coverage, high spatial resolution, and timeliness compared to data from stations. Thus, remote-sensed precipitation products are important for drought monitoring in Xinjiang with scarce in-situ observations. Multi-Source Weighted-Ensemble Precipitation (MSWEP) has been applied in various applications worldwide. With the foundation of global climate change, Xinjiang has experienced an increasing trend in the frequency of droughts and the variability of precipitation. This paper evaluates the accuracy of the MSWEP products using data from 106 observatories. Based on the Standardized Precipitation Index (SPI), MSWEP was adopted to study the temporal variations in dry/wet conditions, identify drought events, and study drought characteristics in Xinjiang from 1980 to 2021. The main results were as follows: (1) MSWEP was highly correlated with data from in-situ stations (>0.8), which supports its application in drought monitoring. (2) Xinjiang had become more humid in the studied period. (3) Thirteen severe drought events were identified since 1980. Among them, the one from 1985 to 1987 was the most severe, and the one from May-October, 2009, was the most intense. (4) Drought events have various features in duration, intensity, and severity. Some droughts were intense with a short duration, while others were long and more severe. Overall, MSWEP has a high potential for drought monitoring, especially for regions where ground-based observatories are scarce. Based on MSWEP, Xinjiang was found to experience a wetting trend and frequent drought events with different characteristics.