Under the background of "four water and four regulations", scientifically quantifying the future water conservation in towns or small basins in a changing environment is an important foundation for water resource management, water conservancy planning and water ecological protection. The scale transformation of water conservation is one of the difficulties in the study of ecological hydrology at present, and the research of higher fine scale is an important entry point of scale transformation. In this paper, taking Bahe River Basin at the northern foothills of Qinling Mountains as the research area, SWAT model, CA-Markov model, ArcGIS and RClimDex were used to evaluate the monthly discharge simulation capability of the NEX-GDDP-CMIP6 high-resolution dataset, and the optimal climate model was selected to simulate the future land use change. The evolution trend of water conservation in the Bahe River Basin in the next 30 years was analyzed in order to provide basic data for ecological protection, water resource management and land use protection in the Qinling Mountains. The results showed that:(1) NEX-GDDP-CMIP6 high-resolution data sets, such as ACCESS-CM2, ACCESS-ESM1-5 and CanESM5, had better simulation effect on the mean monthly flow, but worse simulation effect on extreme flow. NESM3 climate model has better simulation effect than other 8 climate models. (2) NESM3 climate model underestimated the extreme precipitation and overestimated the duration of dry days, and it had higher ability to simulate monthly scale precipitation data than annual and daily scales. (3) Kappa coefficient test showed that CA-Markov model had good applicability in land-use change in the Bahe River Basin. (4) Under the combined background of land use and climate change, water conservation in the Bahe River Basin showed a decreasing trend from 2023 to 2050. The results were of great significance for water resources management in the Bahe River Basin, water conservation and ecological protection in the Qinling Mountains.