WATER RESERVOIRS MONITORING THROUGH GOOGLE EARTH ENGINE: APPLICATION TO SENTINEL AND LANDSAT IMAGERY

Abstract. Water reservoirs are subjected to increasing hydrological stresses, therefore continuous and accurate monitoring of these resources is essential ensure their sustainable management. This work proposes a methodology remotely monitor the surface extent water through analysis satellite multispectral Synthetic Aperture Radar (SAR) images. In particular, segmentation strategy was implemented within Google Earth Engine (GEE) distinguish bodies from surrounding land measure extension, by applying three different approaches Sentinel-1, Sentinel-2, Landsat-8 imagery. The first approach based on use Automatic Extraction Index (AWEI) self-adaptive Otsu’s thresholding method, second image conversion RGB (Red-Green-Blue) HSV (Hue, Saturation, Value) parametric threshold, third SAR imagery an empirically selected threshold. A “static” validation developed scratch standard metrics were computed evaluate accuracy approaches. average values F1 scores Sentinel equal 0.95, 0.90, 0.84 for approaches, respectively. same metric Landsat 0.95 0.93 approach. best approach, i.e. AWEI-based then applied in which effects 2022 drought particularly significant: Sawa lake (Iraq), Poyang (China), Po river (Italy). results visually highlighted good performance segmenting areas.