Improving runoff estimation in hydrological models using remote sensing and climate data reanalysis in the Dittaino River Basin (Eastern Sicily, Italy)

Study region: The study focuses on the Dittaino River Basin, located in eastern Sicily, Italy. This Mediterranean watershed is characterized by water scarcity issues that challenge hydrological modeling and water resource management. Study focus: This study evaluates the use of ERA5-Land reanalysis precipitation data and Sentinel-2/Landsat-8 satellite-derived land cover information to improve runoff simulations using the HEC-HMS model. The methodology was tested on 14 rainfall events (2015–2018) using different input configurations: standard data (SI), reanalysis precipitation data (CR), satellite-derived land cover data (SD), and a combination of both (CR&SD). Model performance was assessed through calibration and validation against observed streamflow data. New hydrological insights for the region: Results demonstrate that ERA5-Land precipitation considerably improves runoff simulations, with a Nash-Sutcliffe Efficiency (NSE) of 0.63 and a Percent Bias (PBIAS) of −16.02 %, confirming its validity as an alternative to ground-based rainfall observations. The CR dataset exhibited a stronger influence on model performance than SD, while the combined CR&SD dataset provided the most balanced results, reinforcing the value of data fusion. The study highlights the complementary impact of precipitation and land cover datasets in runoff modeling and underscores the importance of using multi-source data for improving hydrological simulations. These findings provide practical implications for flood risk mitigation, water resource planning, and irrigation management in Mediterranean semi-arid regions.