论文部分内容阅读
新疆洪水灾害近年来有加剧趋势,但其特征与机理尚未有深入探讨。论文利用核估计和Bootstrap方法深入分析新疆塔里木河(塔河)流域洪水发生率的非平稳性及不确定性,同时采用广义可加模型(GAMLSS)构建洪水发生频率与协变量(大气环流因子、降水和气温指标等)的关系并定量辨识主要影响因子。研究表明:1)塔河流域洪水在1960年代左右和1990年代左右两个时期高频发生,两个洪水高发期之间洪水发生次数多为2或3次,且与极端降水发生次数较为吻合;2)洪水发生率呈现显著非平稳性,从1990年左右洪水发生次数持续上升,并达到峰值,表明洪水发生频率及强度呈加剧趋势;3)冬季AMO和AO是影响新疆塔河流域洪水发生的重要因子,而冬季NAO和SOI则是影响塔河流域5个州的洪水发生次数最为显著的大气环流指标。论文研究可为新疆塔河流域洪灾预测与预警及流域洪水管理提供关键理论依据。
The flood disaster in Xinjiang has been aggravating in recent years, but its characteristics and mechanism have not yet been discussed in depth. In this paper, by using kernel estimation and Bootstrap method, the inhomogeneity and uncertainty of flood occurrence rate in Tarim River basin are analyzed in depth. At the same time, the generalized additive model (GAMLSS) is used to construct the flood frequency and covariate (atmospheric circulation factor, Precipitation and temperature indicators, etc.) and quantitative identification of the main impact factors. The results show that: 1) the floods in the Tarim River Basin occurred at high frequency in the 1960s and the 1990s, and the number of floods between the two floods was mostly 2 or 3 times, which was in good agreement with the occurrence of extreme precipitation; 2) The incidence of floods showed significant non-stationary. The frequency of occurrence and intensity of floods continued to rise and peaked from about 1990, indicating that the frequency and intensity of floods tended to aggravate. 3) Winter AMO and AO were the floods that affected the Tarim River basin in Xinjiang The winter NAO and SOI are the most important atmospheric circulation indexes that affect the flood occurrence in the five states of the Tarim River Basin. Thesis research can provide a key theoretical basis for flood forecasting and early warning and watershed flood management in the Tarim River Basin in Xinjiang.