Ecological flow is a key determinant of river ecosystem stability. Under climate change and increasing human activities, accurate quantification of ecological flow is increasingly important. Using hydrological observations from four stations, Tian’e, Qianjiang, Wuxuan, and Wuzhou, ecological flow was quantified for the Xijiang River mainstream. To address runoff nonstationarity driven by climate change and human activities, this study calibrated a random forest model using meteorological driving factors from the natural period and applied it to simulate quasi-natural runoff during the change period. The simulation results supported the assumption that the climate runoff relationship in the natural period was transferable for quasi-natural runoff reconstruction. This study used the Kolmogorov-Smirnov (K-S) test, Anderson-Darling (A-D) test and Chi-Square (C-S) test to evaluate five probability distributions to determine the optimal distribution of monthly ecological flow estimates. Then, the month-by-month frequency method was used to calculate the monthly ecological flow of each site at 90% guarantee rate, and the Tennant method was used to evaluate the rationality of the flow in each period of the year. The optimal fitting distribution was mainly the Generalized Extreme Value (GEV) and the P-III distribution. After runoff change points, ecological flow guarantee rates decreased during the change period relative to the natural period, with larger declines at the Tian’e and Qianjiang stations in the upper reaches of the basin and the most pronounced reductions during the flood season from July to October. These findings inform ecological flow management in the Xijiang River Basin by identifying priority reaches and sensitive months, and they provide a transferable framework for assessing ecological flows in regulated river systems to support sustainable freshwater ecosystem protection and water resources management.
Li et al. (Tue,) studied this question.