Decoupling climate and vegetation impacts on hydrological processes in semi-arid regions using an improved grid-scale Budyko model

The Upper and Middle Hailar River basin (UMHRB) in China. In the context of global warming, vegetation variation and climate change (CC) have profound and significant impacts on hydrological processes, particularly evapotranspiration ( ET ), in semi-arid basins. This study employed an improved grid-scale Budyko model to propose a novel method for decoupling the independent effects of CC and vegetation variation on hydrological processes and revealed the mechanisms of their influence on evapotranspiration ratio. 1) The improved grid-scale Budyko model, integrated with remote sensing data and eddy covariance technology, significantly enhanced the accuracy of ET simulation; 2) This study first identified a nonlinear relationship between vegetation coverage ( M ) and Budyko parameter ( ε ): when ε < 1, a positive correlation indicates that vegetation enhances ET under humid conditions; conversely, when ε > 1, a negative correlation suggests that vegetation limits ET in arid regions by restricting access to deep soil water; 3) The study quantitatively separated the direct and indirect effects of climate change on ε , revealing that increased aridity and evaporative demand directly elevate ε , while vegetation degradation under dry conditions weakens its regulatory capacity, indirectly amplifying water loss; 4) By classifying the basin based on the monthly phase offset between precipitation and vegetation activity (fPAR), the study identified areas where soil moisture or precipitation dominantly regulates evapotranspiration. • An improved grid-scale Budyko model is employed to enhance ET simulation accuracy. • Remote sensing data and eddy covariance technology are integrated to improve model performance. • A nonlinear relationship between vegetation cover and the evapotranspiration ratio is identified. • The independent effects of climate change and vegetation variation on hydrology are decoupled.