Continued decrease of open surface water body area in Oklahoma during 1984–2015

Changes in Water Surface Area during 1989–2017 in the Huai River Basin using Landsat Data and Google Earth Engine

“…Due to the mixed distributions of water and grasses in the wetlands, the wetland vegetation is the main factor leading to the classification errors of surface water bodies [51]. It has been suggested that combining mNDWI, the normalized water body index (NDVI) and enhanced vegetation index (EVI) can perform better and more stable than the individual index in delineating water [17,18,[29][30][31]. Accordingly, this study used the combination of three indexes to extract surface water areas, which are EVI, NDVI, and mNDWI (Equations (1)-(3)).…”

“…Satellite-based methods have advantages compared to the traditional methods in surface water mapping due to the low cost, high frequency, and repeatable observations. In recent decades, regional, continental, and global-scales surface water areas have been investigated using the advanced very high resolution radiometer (AVHRR) [5,6], the moderate-resolution imaging spectro-radiometer (MODIS) [7], Landsat [8][9][10][11][12][13][14][15][16][17][18], Sentinel satellite images and so on. Meanwhile, many satellite-based approaches have been developed to detect surface water.…”

“…Meanwhile, many satellite-based approaches have been developed to detect surface water. The surface water detection algorithms can be roughly divided into general feature classification methods and thematic water surface extraction algorithms [17,19]. General feature classification methods include spectral mixture models [20,21], maximum likelihood classification [22], artificial intelligence methods [15,23,24], etc.…”

“…The multi-band water indices are superior to the single-band because it takes advantage of different reflectivity differences of spectrum information between water and other land covers. Meanwhile, the water indices also have the advantages in accurate, easy, rapid and reproducible extraction of the surface water information to capture the dramatic intra-annual and inter-annual water variability, and were successfully used to extract surface water using remotely sensed data [17][18][19][29][30][31].The surface areas of water bodies have significant inter-annual and intra-annual variations. Therefore, there are a lot of uncertainties using a single image to map water surface areas [32][33][34].…”

“…Google Earth Engine (GEE) is a representative cloud-based platform that can process the data in an intrinsically-parallel way with high-performance and consists of a multi-petabyte remote sensed data, which is preprocessed to ready-to-use and to efficiently access [37]. GEE has been successfully used to map regional, national and global land cover in urban land [38,39], coastal tidal flats [31], mangrove forests [40], crop planting areas [30,41], forests [40] and open surface waters bodies [13,17,18,42].…”