Quantifying Hydrological Impacts of Ecological Restoration Projects in a Yellow River Tributary Using Remote Sensing and SWAT Modeling

Ecological and water conservation projects, such as Returning Agricultural Land to Forest (RAF) and Returning Agricultural Land to River (RAR), have significantly altered land surface conditions and hydrological processes in many river basins. However, quantifying their spatial and temporal impacts remains challenging due to the complexity of land-use/cover change (LUCC) and the need for high-resolution data. This study focuses on the Qin River Basin, a major tributary in the middle reaches of the Yellow River, where RAF and RAR projects have been extensively implemented. We integrate multi-source remote sensing data, including the China Land Cover Dataset (CLCD) and SRTM DEM, with the Soil and Water Assessment Tool (SWAT) to simulate hydrological responses from 2010 to 2018. The model performed robustly (NSE: 0.70–0.72, R²: 0.71–0.79) and revealed that RAF reduced total runoff by 3.00%, with spatially heterogeneous effects: surface runoff increased in northern subbasins, lateral flow decreased in central regions, and groundwater flow rose dramatically (2366.67%). RAR scenarios showed that converting agricultural land to water bodies enhanced runoff components, with greater efficacy on slopes <15° compared to <6°. The study demonstrates the critical role of remote sensing in capturing LUCC dynamics and highlights the importance of spatially explicit planning for sustainable water resource management in semi-humid basins under intensive human intervention. Our approach provides a scalable framework for integrating remote sensing into hydrological modeling to assess and optimize ecological restoration strategies in data-scarce or heterogeneous regions.