A comprehensive understanding of river dynamics requires quantitative knowledge of the grain size distribution of bed sediments and its variation across different temporal and spatial scales. Several techniques are already available for grain size assessment based on field and remotely sensed data. However, the existing methods are only applicable on small spatial scales and on short time scales. Thus, the operational measurement of grain size distribution of river bed sediments at the catchment scale remains an open problem. A solution could be the use of satellite images as the main imaging platform. However, this would entail retrieving information at sub-pixel scales. In this study, we propose a new approach to retrieve sub-pixel scale grain size class information from Copernicus Sentinel-2 imagery building upon a new image-based grain size mapping procedure. Three Italian gravel-bed rivers featuring different morphologies were selected for Unmanned Aerial Vehicle (UAV) acquisitions coupled to field surveys and lab analysis meant to serve as ground truth grain size data. Grain size maps on river bars were generated in each study site by exploiting image texture measurements, upscaled and co-registered with Sentinel-2 data resolution. Relationships between the grain sizes measured and the reflectance values in Sentinel-2 imagery were analyzed by using a machine learning framework. Results show statistically significant predictive models (MAE of ±8.34 mm and R2=0.92). The trained model was applied on 300 km of the Po river in Italy and allows to detect grain size longitudinal variation and to identify the gravel-sand transition occurring along this river length.Our proposed approach based on freely available satellite data calibrated by low-cost automated drone technology can provide reasonably accurate estimates of surface grain size classes, in the range of sand to gravel, for bar sediments in medium to large river channels, over lengths of hundreds of kilometers.