 This paper presents a novel approach to generate a high-resolution surface water mask from Landsat 8 imagery and OpenStreetMap data. The authors firstly developed a method to extract water from flat areas using a canny-edge filter and Otsu Thresholding. Then, they extended their method to hilly areas by adding a supervised classification step. Finally, they compared the accuracy of their new water mask with the one obtained from OpenStreetMap and the one derived from SRTM data. Their results showed that the new water mask has a higher accuracy than the one obtained from OpenStreetMap and the one derived from SRTM data. This article was authored by Genadii Donchets, Yap Skellicans, Hassel and Semius, and others.