To quantitatively deconstruct the impacts of multiple factors, such as climate change, human activities, and vegetation dynamics, on the evolution of watershed water cycles, a decoupling framework for multifactor influences on the water cycle in the Wei River Basin was constructed. Partial least squares structural equation moeling (PLSSEM), Pearson correlation analysis, and trend analysis methods were used to deconstruct the direct and indirect effects of climate change, human activities, and vegetation dynamics on runoff changes in the Wei River Basin from 1982 to 2020. The results indicated that from 1982 to 2001, the combined effects of climate change and human activities on runoff in the Wei River Basin exhibited a balanced situation, with the comprehensive effect value of climate change and human activities being 0.313 and -0.315, respectively. From 2002 to 2020, the comprehensive effect value of human activities (-0.667) significantly surpassed that of climate change (0.319), becoming the dominant driving factor of runoff evolution in the basin. From 1982 to 2001, the mechanism characteristics of runoff reduction are precipitationdominated and multifactor collaborative, with the direct effect of reduced precipitation and the combined effects of temperature increase, vegetation restoration, and human activities jointly driving the process of runoff attenuation. During the period from 2002 to 2020, the mechanism characteristics of runoff increase are precipitationdominated and engineeringreduction, with the significant increase in precipitation being the dominant factor driving the rise in runoff. However, the direct effect of engineering regulation suppressed the rising trend of runoff, resulting in only a slight recovery trend in runoff.