基于信息熵的新疆降水时空变异特征研究

基于信息熵理论对新疆降水序列的时空变异性进行研究。利用边际熵研究不同时间尺度降水序列的变化特征。利用分配熵和强度熵分别研究降水量和降水天数年内和年代际(10a)分配情况。利用改进的Mann-Kendall趋势检验法分析新疆降水过程不确定性变化趋势。研究表明:(1)新疆降水量与降水天数年内分配不均匀性主要表现为由南向北减小的空间分布特征;(2)新疆不同尺度的降水序列不确定性具有明显空间特征;(3)越降水越稀少的地区,降水量与降水天数变异性就越大。本研究对该区域降水时空变异研究与水资源规划具有重要意义。

Information entropy-based analysis of spatial and temporal variation in precipitation in Xinjiang

The precipitation distribution is becoming highly unstable due to climate change and intensifying human interference. Therefore, estimations of the precipitation distribution are extraordinary important for understanding the hydrological cycle and are crucial for water resource management. In this study, variation in spatial and temporal characteristics of precipitation in Xinjiang was evaluated based on information entropy theory. Variation in the precipitation sequence for various time scales was confirmed using marginal entropy. Furthermore, the distribution in precipitation amounts and the number of days with precipitation within a year and among decades were systematically analyzed using apportionment and intensity entropy measures, respectively. A modified Mann-Kendall test was applied to detect trends in precipitation uncertainty in Xinjiang. This analysis provided a few key findings: (1) The spatial variation in both the amount of precipitation and the number of days with precipitation within a year decreased from South Xinjiang to North Xinjiang, indicating that South Xinjiang has obvious uncertainty with respect to both characteristics. Moreover, the spatial and temporal distribution of precipitation amounts and the number of days with precipitation within a year were similar; the greatest variability in precipitation amount was observed during 1965-1973 and the greatest variability in the number of days with precipitation was observed in 1965, 1997, and 2007. (2) The uncertainty for various scales of the precipitation sequence in Xinjiang has obvious space structures. For example, South Xinjiang had greater uncertainty than North Xinjiang in the spring and autumn, while there was no obvious spatial distribution in the summer and winter. (3) The sparser the precipitation in a region, the greater the variation in the amount of precipitation and the number of days with precipitation. (4) Based on the apportionment disorder index for annual precipitation changes, most meteorological stations observed significant decreases, except the Ruoqiang Station, and these stations are mainly located in the southern area of South Xinjiang, the northern area of North Xinjiang, and the northern Tianshan area. This study improves our understanding of the spatial and temporal variation in precipitation and has implications for water resource management.