A remote sensing-based method for river discharge estimation in data-scarce regions

River discharge is a key hydrological parameter essential for understanding the water cycle, managing water resources, and assessing the impacts of climate change. Although remote sensing techniques have advanced rapidly, estimating discharge in data-scarce regions remains challenging due to the lack of high spatiotemporal resolution data on surface-water extent and underwater topography. To address this limitation, this study proposes an approach that combines multi-source satellite observations with numerical topography simulation. Based on the morphological similarity between exposed and underwater topography, this study developed a novel approach for simulating cross-sectional topography, which was subsequently integrated with Manning formula to estimate discharge. The method was applied to four representative hydrological stations across different river systems in China. The results indicate that, at Guide Hydrological Station, the relative accuracy of underwater topography simulation reached 73.1% using 5m DEM, increasing to 76.3% with a 0.1m DEM. Across all four hydrological stations, the temporal river discharge estimations yielded an overall accuracy of approximately 80%. Sensitivity analysis further indicated that higher-resolution DEM data can substantially improve discharge estimation. The overall findings reveal that the proposed framework offers a technically feasible approach for local site-specific discharge monitoring, particularly in ungauged or inaccessible regions where traditional observations are unavailable.