In this study, a coupled water–energy balance equation at an arbitrary time scale was proposed as an extension of the Budyko hypothesis. The second mixed partial derivative was selected to represent the magnitude of the interaction. The extended hydrological sensitivity method was used to quantitatively evaluate the impacts of climate change, anthropogenic activities, and their interaction on dry season runoff in the Lhasa River. In addition, an ABCD model, which is a monthly hydrological model included a snowmelt module, was used to calculate the change in soil water and groundwater storage. The Mann–Kendall test, Spearman’s test, dynamic linear model (DLM), and Yamamoto’s method were used to identify trends and change points in hydro-climatic variables from 1956–2016. The results found that dry season runoff increased non-significantly over the last 61 years. Climate change, which caused an increase in dry season runoff, was the dominant factor, followed by anthropogenic activities and their interaction, which led to varying degrees of decrease. This study concluded that the methods tested here performed well in quantifying the relative impacts of climate change, anthropogenic activities, and their interaction on dry season runoff change.

Arid ecosystems have exhibited greening trends in recent decades. There is no consensus on how underlying surface changes influence streamflow across vegetation gradients. We investigated this issue for the four typical arid mountain basins using a 30-year runoff database and the Budyko framework to quantify the contributions of climate and underlying surface changes to streamflow variations during summer periods. Results showed that in the poor vegetated basins, i.e., Heizi Basin and Kuche Basin, the underlying surface change has increased summer streamflow by 14.01 and 35.67 mm, respectively; climate contributed only −7.32 and 1.86 mm to summer streamflow changes, respectively. Comparatively, in the well-vegetated basins, i.e., Huangshui Basin and Kaidu Basin, climate change dominated summer streamflow variations by increasing 21.50 and 24.65 mm, respectively; the underlying surface change only increased summer streamflow by 3.72 and 1.56 mm, respectively. Additionally, the decomposition results were extended to monthly scale (from June to September) to reveal the effects of climate and underlying surface changes on monthly streamflow. This study deepens our knowledge of runoff responses, which can provide important references to support water resources management in other regions that receive water from mountains.