Climate and land use regulate vegetation-runoff interaction pathways in the three-river source region

Alpine ecosystems are critical for water regulation but highly sensitive to climate change. In the Three-River Source Region (TRSR) of the Qinghai-Tibet Plateau, changes in temperature, precipitation, and large-scale ecological restoration have significantly altered vegetation phenology-including the start (SOS), end (EOS), and length (LOS) of the growing season, as well as vegetation growth status (GS). These shifts affect hydrological processes such as evapotranspiration, soil moisture, snowmelt, and runoff. However, the combined effects of climate and ecological changes on runoff remain poorly quantified. This study uses Partial Least Squares Structural Equation Modeling (PLS-SEM) to evaluate the relative impacts of climate, vegetation phenology, and land cover on runoff variability from 1982 to 2018. Results show: (1) Vegetation phenology exhibited notable interannual and spatial variation, with GS increasing at an average rate of (0.5 ± 0.01) × 10^-3 per year; (2) Precipitation had a strong direct effect on runoff (path coefficient = 0.82), amplified by indirect effects via GS, evapotranspiration, and soil moisture, totaling 1.28; (3) Temperature had a strong negative direct effect (-0.71), but positive indirect effects partially offset this, resulting in a smaller overall negative total effect of -0.3. (4) High-density forest exerted a positive total effect of 1.08 on runoff by offsetting its high water demand through regulating soil moisture and evapotranspiration; (5) GS was a key mediator linking climate to runoff. These findings highlight the central role of vegetation dynamics in alpine hydrology, showing that GS modulates soil moisture and evapotranspiration, thereby mediating climate impacts on runoff. Understanding these interactions is vital for adaptive environmental management to sustain water resources and ecosystem function in alpine headwaters.