The value of Sentinel-3 snow cover fraction data in improving hydrological simulations

Runoff and snow simulations can be enhanced using multi-spectral snow cover mapping. The main focus of this research is to evaluate the potential of a new snow cover fraction (SCF) product to improve hydrological simulations. The SCF product derived from combined Sentinel-3 SLSTR and OLCI observations based on a multispectral unmixing method at a spatial resolution of 0.00200° × 0.00200°. This study first assesses the accuracy of a new snow cover fraction (SCF) product against in-situ snow depth measurements at climate stations. It then investigates the impact of assimilating this product on runoff and snow simulation using a conceptual semi-distributed hydrological model. For this purpose, the hydrologic model is calibrated with and without snow cover product using multi-objective calibration and single-objective calibration schemes, respectively. Based on the multi-objective calibration, both runoff and snow are optimized, whereas the single-objective calibration approach focuses on runoff alone. We evaluated the proposed framework across 188 catchments in Austria. The results showed a strong agreement between SCF and snow depth measurements, with a median overall accuracy of about 95%. The analysis of the results demonstrated that the added value of incorporating snow products into model calibration is more pronounced for snow simulation than for runoff estimation. Hence, runoff and snow simulations improved in 39% and 84% of catchments during the validation period, respectively. The findings reveal that our approach enhances the model’s efficiency to more effectively capture snow cover dynamics, which supports more consistent water balance simulations and provides a stronger basis for modeling of snow-induced high-flow conditions.