Detection of precipitation variability based on entropy over nearly 50 years in Xinjiang, northwestern China

Abstract

Based on precipitation data of 53 meteorological stations from 1960 to 2008, the entropy method was used to analyze spatial variability of precipitation in Xinjiang, China, over monthly, seasonal, and annual timescales. The spatial distribution of precipitation variability was significantly affected by topography and was zonal on all timescales. The nonparametric Mann-Kendall test was used to analyze changes in the distributions. A precipitation concentration index was developed to categorize the variability of annual precipitation. Summer variability contributed less to annual variability than that of other seasons. Various months contributed to annual mean variability differently across the years. Overall, the variability of precipitation was shown to increase north of Xinjiang, especially in mountainous regions, where the increase was statistically significant (P = 0.05). South of Xinjiang, the variability increased only slightly, consistent with the distribution of precipitation.