北疆地处内陆干旱半干旱区，降水的多少对其影响十分巨大。尤其在全球变暖背景下，北疆地区的降水异常增多。本研究利用1979-2017年北疆40个站点冬季降水资料、ECMWF ERA-Interim和NCEP/NCAR再分析资料，基于HYSPLIT v4.9模式讨论北疆4个区域型暴雪日的水汽输送特征及可能影响机制。研究发现：（1）各区域型水汽以西边界输入为主，但西天山型有少量水汽从北边界输入，天山型有部分水汽从高层南边界输入；（2）北部型、西部型和西天山型的水汽通道均位于中纬西风带，但具体位置有所差异。北部型以地中海和黑海水汽输送为主，贡献率为58.8%；西部型以里海西南侧水汽输送为主，贡献率为70.8%；西天山型以黑海和里海东南侧水汽输送为主，总贡献率为72.9%；天山型的水汽源地主要位于印度、伊朗附近，贡献率为64.2%；（3）各区域型位势高度的距平南北均呈现“正-负”分布，西东均呈现“负-正-负”分布，但中心强度、位置及范围有所不同，这种差异导致影响区域的不同。
China’s Northern Xinjiang is located in a typical inland arid and semi-arid area，over which anomalous precipitation plays an main role. In recent years, anomalous precipitation events have increased with rising global temperatures. In this study, the inter-decadal variability of the wintertime precipitation in Northern Xinjiang have been investigated by using the observational rainfall data of 40 stations、ECMWF ERA-Interim and NCEP/NCAR reanalysis dataset for 1979-2017. This study discusses water vapor transport characteristics of wintertime snowstorm days and possible mechanisms in four regions based on the HYSPLIT v4.9 model. Our main results are as follows: (1) The west boundary input of water vapor is mainly in each region, but a small amount of water vapor is imported from the northern boundary in the western TianShan, and some water vapor is imported from the southern boundary in the upper in the TianShan; (2) the water vapor passages are mainly located in the Eurasia of the mid-latitude westerlies in the north、the west and the western TianShan, but their exact locations are different. The water vapor transport mainly come from the Mediterranean and the Black Sea in the north, and their contribution ratios are 58.8%; The water vapor transport mainly come from the Southwest of the Caspian Sea in the west, and their contribution ratios are 70.8%; The water vapor transport mainly come from The Black Sea and the Southeast of the Caspian Sea in the western TianShan, total contribution ratios are 72.9%; The water vapor mainly come from the India and the Iran in the TianShan, and their contribution ratios are 64.2%. (3) The geopotential height anomaly shows “+ -”from south to north, and shows “- + -”from west to east in the different regions, but the intensity、the range and the location of the anomaly center are different, this difference leads to difference of influential regions.