Li Jinhang, Zhao Mengdie, Han Yuping
North China University of Water Resources and Electric Power, Zhengzhou, China.
Zhejiang University of Water Resources and Electric Power, Hangzhou, China.
PLoS One. 2025 Jul 23;20(7):e0327082. doi: 10.1371/journal.pone.0327082. eCollection 2025.
As the largest river in China, exploring the evolution characteristics of rainfall in the upper, middle, and lower reaches of the Yangtze River and analyzing the frequency of wet and dry encounters play a significant role in flood and drought prevention in the Yangtze River Basin. This study employs the Mann-Kendall test, 5-year moving curve, Pettitt test method, and Morlet wavelet analysis method to analyze and diagnose the trend, abrupt changes, and periodicity of rainfall in the Yangtze River Basin. By fitting the optimal rainfall marginal distribution and Copula joint distribution model, the probability of wet and dry period encounters and their changing patterns in the upper, middle, and lower reaches of the Yangtze River are calculated and analyzed. Based on the results of wet and dry encounters, a Bayesian network model for flood and drought management is constructed and simulated. The study shows that: (1) There is no significant change in the overall rainfall pattern in the Yangtze River Basin, with periodic changes of 14, 33, and 10 years in the upper, middle, and lower reaches respectively, and abrupt change years in 1997, 1986, and 1979. (2) Apart from the middle and lower reaches of the Yangtze River, where the number of encounters between wet and dry years reaches up to 15 times, the frequency of asynchronous and synchronous wet and dry periods is generally evenly distributed across the upper, middle, and lower reaches. (3) Based on Bayesian prior inference and simulation results, the probabilities of simultaneous floods and droughts occurring in the upper, middle, and lower reaches are 53% and 47%, respectively. When the upper, middle, and lower reaches are all in wet or dry years, the probability of floods and droughts in the Yangtze River Basin is 91%. The research results provide a theoretical basis for scientifically grasping the changes in water conditions in the Yangtze River Basin and for flood and drought prevention.
作为中国最大的河流,探究长江上、中、下游降雨的演变特征并分析干湿遭遇频率,对长江流域的防洪抗旱具有重要意义。本研究采用曼-肯德尔检验、5年滑动曲线、佩蒂特检验法和莫雷特小波分析法,对长江流域降雨的趋势、突变和周期性进行分析与诊断。通过拟合最优降雨边缘分布和Copula联合分布模型,计算并分析长江上、中、下游干湿期遭遇的概率及其变化模式。基于干湿遭遇结果,构建并模拟了用于防洪抗旱管理的贝叶斯网络模型。研究表明:(1)长江流域整体降雨格局无显著变化,上游、中游和下游分别存在14年、33年和10年的周期变化,突变年份分别为1997年、1986年和1979年。(2)除长江中下游干湿年遭遇次数达15次外,上、中、下游干湿期异步和同步出现的频率总体分布较为均匀。(3)基于贝叶斯先验推断和模拟结果,长江上游、中游和下游同时发生洪涝和干旱的概率分别为53%和47%。当上、中、下游均处于湿年或干年时,长江流域发生洪涝和干旱的概率为91%。研究结果为科学把握长江流域水情变化及防洪抗旱提供了理论依据。