Analysis of Dryness/Wetness over China Using Standardized Precipitation Evapotranspiration Index Based on Two Evapotranspiration Algorithms

On the basis of a high-resolution global meteorological dataset from Princeton University and monthly radiation data from the National Centers for Environmental Prediction (NCEP), two types of Standardized Precipitation Evapotranspiration Index (SPEI) were established using precipitation in addition to two sets of potential evapotranspiration (PET) data, in which PET was calculated by Thornthwaite and Penman-Monteith formulas, respectively. On the basis of these results, the dryness/wetness trends over China during 1949-2008 measured by two SPEIs and their differences are analyzed. In addition, the applicable regions of the two SPEIs are evaluated. It is determined that a dryness trend for the entire country was indicated by the two SPEIs. Seasonally, the most significant dryness trend occurred in spring. Geographically, drought clearly increased in northern China but decreased in southern China, with a dividing line approximately along with the Yangtze River valley. A significant dryness trend was apparent over the areas of western Inner Mongolia, North China, northeastern China, and eastern Sichuan, whereas the opposite was indicated for northern and western Xinjiang. Moreover, different grades of drought events increased across the entire country. Among them, the moderate drought event showed the most significant increasing trend, and the occurrences of moderate and extreme drought events were the most frequent after the late 1990s for the entire study period. Spatially, the regions with an obvious upward trend of drought events corresponded to the regions with a downward trend of SPEI. It is noted that the difference between the two types of SPEI was largest in winter and spring because of differences in PET. In northern China, the contribution of an aerodynamic term to PET in the Penman-Monteith formula was determined to be important in winter and spring and hence gave more reasonable results.