Abstract: Climate change and high-intensity human activities have altered the natural hydrological cycle of the basin, resulting in alteration in the hydrological sequence and severe impacts on the eco-hydrological system of the basin. The calculation of the ecological instream flow based on hydrological alteration conditions has become one of the hot topics in eco-hydrological research in the changing environment. The Liaohe River at the northwest of Poyang Lake in Jiangxi Province was taken as an example in this study. Based on the current situation of hydrological alterations in the changing environment, a variety of hydrological time series mutation test methods were used to detect the change-points, and the potential causes of hydrological alterations in the Liaohe River were also discussed. Then, runoff restoration was carried out by using hydrological models. On the basis of restored runoff, three kinds of goodness-of-fit methods, including Kolmogorov-Smirnov (K-S), Anderson Darling (A-D) and probability plot correlation coefficient (PPCC) methods, were used to determine the optimal probability distribution function of monthly runoff, which was used to calculate ecological instream flow under the hydrological alteration conditions. The results show that: (1) The hydrological alteration of the Wanjiabu Station in the Liaohe River occurred in 1972, and the annual runoff in the basin increased by 12% after the change-point. The increase in precipitation and the decrease in evapotranspiration were the main driving factors of the hydrological alteration. (2) The runoff was restored through the Distributed Time-Variation Gain Model (DTVGM). The simulated runoff values had a high fitting degree of fit with the measured values during the calibration and validation periods, and the correlation coefficient (r) and Nash-Sutcliffe efficiency (NSE) were larger than 0.78 during both periods, which indicates that runoff reduction based on DTVGM is feasible. (3) After the runoff restoration, the three goodness-of-fit methods were used to select the optimal distribution function from five widely-used probability distribution functions for each month, and the monthly river ecological instream flow was calculated based on the selected optimal distribution function. Compared with the Tennant method, the minimum monthly flow method and the 7Q10 method, the ecological instream flow based on the optimal distribution function was more certain and reasonable. The analysis shows that the calculation method of ecological instream flow considering hydrological alteration conditions can better reflect the influence of hydrological alteration on the runoff process, and this study provides an important basis for the adaptive management of water resources in changing environment. Under the background of hydrological alteration caused by climate change and human activities, the calculation of the ecological instream flow considering the hydrological alteration conditions can better reflect the impact of hydrological alteration on the ecological flow process. The research results can provide data support for water ecological protection and water resource management in the Liaohe River Basin, but can also provide scientific basis for water resource planning and allocation in a changing environment.