Progress and prospects in satellite remote sensing monitoring of terrestrial surface water

Terrestrial surface waters, which include rivers, lakes, reservoirs, marshes, and wetlands, are distributed in the terrestrial surface layer and are crucial components of the Earth's system and essential for human survival. Additionally, they serve as an indicator of changes in climate and human activities. In recent years, influenced by a combination of natural and anthropogenic factors, surface waters have undergone significant changes in their range, morphology, water volume, and physicochemical properties. Over the past century, increased demand for water resources, water pollution, and inadequate management has exacerbated the conflict between humans and water resources. Consequently, there is an urgent need for macroscopic, rapid, and effective observations of surface waters, including quantification of their volume, extraction rates and dynamic changes. Remote sensing has become an indispensable means of acquiring information on surface water resources. Compared to traditional field surveys, remote sensing offers advantages such as large-area coverage, low cost, high frequency, and rich spectral information, significantly facilitating understanding of the spatial distribution and changes of surface waters. In recent decades, with the rapid development of sensors, water indices, machine learning algorithms, and cloud computing environments, surface water research has entered an unprecedented era of big data. Driven by the development of various sensors such as optical, thermal infrared, microwave, and lidar, massive amounts of observational data with multiple spatiotemporal resolutions are being captured and applied in various fields of surface water research. However, a systematic review and discussion on the characteristics, key methods, and development trends of overall satellite remote sensing monitoring of surface water is still missing. This paper presents a comprehensive review of existing research on surface water remote sensing monitoring, covering data sources, methods, main content, and development trends, identifying the challenges and opportunities faced by surface water remote sensing research. Research findings indicate that over the past decade, surface water remote sensing monitoring has entered a golden period of development, especially attributed to the wide availability of optical data, such as from Landsat and Sentinel-2, and the release of global surface water data products, such as Global Surface Water (GSW), which have provided robust support for large-area surface water research. However, emerging big data methods such as deep learning still lag behind in surface water research. Due to the limitations of optical remote sensing monitoring methods, less research has been conducted on special types of surface waters like reservoirs and wetlands compared to large lakes. Surface water research is very active in China, the United States, Europe, and Canada, while it remains relatively scarce in economically underdeveloped and arid regions. Future research needs to focus more on the dynamic changes of surface water in special regions, improve the identification of challenging surface water bodies (such as thermokarst lakes and glacier lakes) and strengthen the application of emerging big data methods like deep learning to promote comprehensive understanding of surface water, contributing to Earth system research and sustainable development.