Intercomparison between sentinel-1, sentinel-2, and landsat-8 on reservoir water level estimation

Remote sensing technology has made it possible for the surface water level to be updated accurately and often to prepare against water scarcity and drought events under the influence of anthropogenic activities. This research provides deeper insights into the detection and estimates of the surface water level with Sentinel-1 Synthetic Aperture Radar (SAR) data, Sentinel-2/MSI, and Landsat-8/OLI data in the Shimen Reservoir, located in northern Taiwan. This research uses data from Sentinel-1A, Sentinel-2, and Landsat-8 missions to demonstrate how well different water indices, such as the Normalized Difference Water Index (NDWI), the Normalized Difference Vegetation Index (NDVI), and the Modified Normalized Difference Water Index (MNDWI), can be used to estimate Water and water surface area. Estimated water levels from the multispectral and SAR images directly compared with DEM elevation cross-section profiles. Assessment of water level and surface water area results shows that the accuracy of SAR is comparable to that of NDWI/Landsat-8. Due to land interaction, multispectral water indices’ accuracy was better in detecting the inlet branches than either SAR Sentinel-1A VH or VV polarization. The results reconfirmed that SAR sentinel-1A data can be the best alternative for monitoring water levels in the reservoir when multispectral images were not available. Sentinel-1 data could provide similar accuracy on surface water delineation like multispectral images.