【摘要】 综合考虑大气降水-植被生长-海拔相互作用等多元成因，以新疆地区2001—2019年的MODIS数据、TRMM降水数据以及该地区数字高程模型(digital elevation model, DEM)数据为遥感数据源，计算降水集中指数(precipitation concentration index, PCI)、温度植被干旱指数(temperature vegetation dryness index, TVDI)以及DEM等参数，利用主成分分析建立了改进的综合干旱监测模型。利用该模型对研究区进行时空分析，结果表明：干旱发生频率在空间上主要呈现中部高四周低的特点，研究时段内约47.7%的区域发生了干旱，其中32.3%的干旱区其干旱频率可达60%以上，主要集中于塔里木盆地以及吐鲁番盆地；研究区旱情变化趋势存在较大差异，3—9月线性回归斜率正值数值远大于负值，根据结果预测研究区2020年干旱情况主要表现为春旱和夏旱。更多还原

【Abstract】 An improved and comprehensive drought monitoring model was developed in this study. Given multi-genetic types such as the interaction of atmospheric precipitation, vegetation growth, and elevation, multiple data sources were selected for the model, including EOS-MODIS data, TRMM precipitation data, and the region SRTM-DEM(digital elevation model) data from 2001 to 2019 in Xinjiang. The parameters including precipitation concentration index(PCI), temperature and vegetation drought index(TVDI), and DEM were calculated, and the principal component analysis(PCA) method was employed to establish the model. Then, the model was used to analyze the spatio-temporal characteristics of drought in the study area. The analytical results show that the annual occurrence frequency of drought in the study area from 2001 to 2019 was high in the middle part and low in the surrounding areas. In addition, drought struck 47.7% of the study area, and the occurrence frequency of drought reached 60% in 32.3% of the drought regions. Meanwhile, drought was concentrated in the Tarim and Turpan basins. The changing trends of drought in the study area differed greatly. For the linear regression slope of drought from March and September, the absolute values of the positive slope were far greater than those of the negative slope. Based on this, it can be predicted that the drought in the study area mainly included spring and summer droughts in 2020.更多还原