Combined use of Sentinel-2 and Landsat 8 to monitor water surface area dynamics using Google Earth Engine
Xiucheng Yang, Yang Chen, Jingzhe Wang
Abstract
Detecting the water surface area and its temporal changes is essential in water resource management. Multispectral satellite data are applied extensively to monitor the surface water dynamics because of their repeated coverage. This letter monitors waterbody dynamics in a high temporal resolution from Sentinel-2 and Landsat 8 images by using the Google Earth Engine (GEE) platform. Water index, automatic threshold and postprocessing of noise elimination are used to extract the water pixels from the background. Taking a part of the middle reaches of the Yangtze River (100km×100km) as a case study, the surface water dynamics are estimated by 47 dates from February 2017 to November 2019. The popular JRC Global Surface Water database includes 8 qualified monthly maps for the study area in 2017 and 2018, while our results display 32 maps covering 18 months. The surface water dynamic maps obtain an overall accuracy of 92.5% (using random sample points) and 82.0% (using manual-drawing thematic map). Also, water occurrence is calculated based on the water dynamics to indicate the permanent, seasonal and flooding areas. The high temporal dynamics of surface water are meaningful for the analysis of ephemeral lake and inundation extent.