Precisely quantifying the independent runoff effects of different land use transitions is a critical challenge in addressing complex human-land interactions and formulating targeted watershed management strategies. This study proposes a scenario-controlled framework named DHRFLUT (Decoupling Hydrological Responses from Land Use Transitions), which simulates individual transition scenarios reflecting regional development trends by establishing a land use transition decoupling mechanism. By coupling with the SWAT model, the runoff depth response coefficient (RDRC, mm/%) was quantified at the sub-basin scale, resulting in a decoupled RDRC matrix. In a case study in the Upper and Middle Reaches of the Wei River Basin on China's Loess Plateau, urban expansion was identified as the dominant driver of runoff change, with RDRC values significantly exceeding those of vegetation transitions. Coupled topography and climate are key drivers of spatial heterogeneity in the runoff response. Specifically, precipitation gradients dominate the runoff effects of agricultural land-to-pasture and pasture-to-forest transitions, whereas geomorphic features (e.g., minimum elevation and elevation range) govern the spatial differentiation patterns of RDRC for pasture-to-agricultural land and agricultural land-to-forest transitions. Ultimately, the integrated RDRC matrix developed in this study enables the transition from qualitative understanding to quantitative management of runoff responses, thereby providing a core quantitative tool for assessing the potential hydrological impacts of land planning schemes. This research not only offers a universal methodological framework for decoupling the runoff effects of land use transitions but also provides scientific evidence for harmonizing ecological restoration with urban development and advancing sustainable watershed management.