Active‐Passive Surface Water Classification: A New Method for High‐Resolution Monitoring of Surface Water Dynamics

Abstract

This study develops a new, highly efficient method to produce accurate, high‐resolution surface water maps. The “active‐passive surface water classification” method leverages cloud‐based computing resources and machine learning techniques to merge Sentinel 1 synthetic aperture radar and Landsat observations and generate monthly 10‐m‐resolution water body maps. The skill of the active‐passive surface water classification method is demonstrated by mapping surface water change over the Awash River basin in Ethiopia during the 2015 East African regional drought and 2016 localized flood events. Errors of omission (water incorrectly classified as nonwater) and commission (nonwater incorrectly classified as water) in the case study area are 7.16% and 1.91%, respectively. The case study demonstrates the method's ability to generate accurate, high‐resolution water body maps depicting surface water dynamics in data‐sparse regions. The developed technique will facilitate better monitoring and understanding of the impact of environmental change and climate extremes on global freshwater ecosystems.

Construction of High Spatial-Temporal Water Body Dataset in China Based on Sentinel-1 Archives and GEE

“…Sentinel-1 data in GEE is only processed through primary processing such as noise removal, calibration and geocoding. As we all know, layover and shadow will bring some errors to the extraction of water based on radar images [17]. In this study, compositing ascending and descending SAR scenes have reduced this error but do not fully eliminate it.…”

“…Some studies have shown that synthetic ascending and descending SAR scenes reduce some errors caused by radar shadows or layover, but do not completely eliminate them [17]. Using optical sensors to detect surface water will also encounter the problem of terrain shadows, and there has been a lot of research performed using slope dataset to solve this problem [6,11,27,28].…”

“…Synthetic Aperture Radar (SAR) data have the advantage of being unaffected by clouds, so they can be used to regularly monitor surface water [14,15]. Yet, due to the difficulty of SAR data acquisition at large scale and the complexity of data processing, in the past, water body monitoring and mapping using SAR data has mainly focused on individual bodies of water or specific small areas [16][17][18]. With the free distribution of Sentinel-1 data, as well as the release and application of the Google Earth Engine (GEE) platform, it is possible to carry out dynamic monitoring and mapping of water bodies on a large scale.…”

ReaLSAT, a global dataset of reservoir and lake surface area variations

“…Recently, Landsat imagery has been leveraged to identify not only the location of the world's waterbodies but also their sizes and dynamics. Previous efforts to map global inland waters fall into two categories: 1) vector-based mapping, which provides static polygons of the world's lakes and lacks any temporal dynamics [5][6][7][8][9] , and 2) raster-based mapping, which divides the Earth's surface into pixels and documents the change in the presence of water over time in those pixels [10][11][12][13][14][15][16] . Both these categories of approaches provide an incomplete view of the world's lakes in that the former does not indicate the dynamic nature of surface water, and the latter tracks the presence of water at the level of pixels but fails to associate these pixels with lakes.…”

Long and short‐term assessment of surface area changes in saline and freshwater lakes via remote sensing

“…Most recently, Slinski et al . (2019) developed a new method, called as active‐passive water classification method, to map surface water dynamics and applied it to the Awash River Basin in Ethiopia that has freshwater lakes, saline water lakes and reservoirs between years 2014 and 2017. Year 2015 was a term during which the basin experienced a major East African regional drought, whereas; localized flood events were highly noticeable in 2016.…”