The east Tibet and west Sichuan ecological and water conservation zone, a notable ecological vulnerable area with low anti-disturbance capability and climate sensitivity, is located in the northern part of Southwest Alpine Canyon Region of China. It is important to study the spatio-temporal distribution characteristics and influencing factors of water yield for the water resources protection, conservation, development and utilization. We simulated the water yield by using the Annual Water Yield module of the InVEST model, mainly based on the data set of surface cover products, climate, bedrock depth, soil, and topography from 2000 to 2020, and analyzed the explanatory power of the factors contributing to the spatial variability of water yield by combining with GeoDetector (GDM), and for the factors with q＞0.1 were introduced into the multiscale geographically weighted regression (MGWR) to analyze their influence on water yield in different geographic locations of the region, and the spatio-temporal variation characteristics and dominant factors of water yield were obtained by using the Theil-Sen trend analysis and the Mann-Kendall significance test, and in the meantime.Hurst index was used to predict the upward or downward trend of water yield in the future for a short period of time and the fluctuation level of water yield during the study period at different spatial locations was assessed. The results showed that: 1) the spatial distribution of precipitation and water yield presented a distribution pattern of “higher in the east and west, lower in the middle”, the greatest amount of precipitation in the Minjiang River basin, the greatest amount of evapotranspiration in the Jinsha River basin, and the amount of water yield in the Nujiang River basin was more than other three basins. 2) except for precipitation and evapotranspiration, the main factors affecting water yield were climate factors (annual mean humidity and annual mean wind speed), topographic factors (DEM), soil type, vegetation factors (NDVI and NPP) ,and social factors (LULC and HAI), especially precipitation, evapotranspiration, DEM,NDVI, NPP and surface cover type were the dominant influencing factors on water yield. 3) There were strong synergies between precipitation, DEM and HAI, and strong trade-offs between evapotranspiration, NDVI and NPP to water yield services. 4) The water yield fluctuation level was higher in the southern part of the region and in the short-term future, the water yield showed a decreasing trend in the 95.30% of the study area. 5) Increasing the artificial vegetation coverage blindly should not be the primary solution to increase water yield in the region, and the priority was to pay attention to of natural forest protection and rocky desertification prevention.