Surface water

Abstract

Inland water is the primary water source that supports most lives on land. Its change affects the rise and decline of lives. On the other hand, the spatiotemporal variation of water distribution not only reflects but also affects the energy balance on land. The size and distribution of inland water surfaces are the most dynamic components in the water cycling of the Earth system. Therefore, accurate information about inland water distribution plays an important role in understanding water cycling on Earth. In this article, we provide an overview of water surface mapping based on remotely sensed data in optical spectral region on board of satellite platforms. First, we introduce some basic characteristics of water, including the spectral, temporal, and topographical characteristics. Generally speaking, water has higher reflectance in the visible (VIS) spectral range than the short-wave infrared (SWIR) range. Yet the overall reflectance of water is low, compared with other land cover types like bare soil, except for shallow and/or turbidity water bodies. Water is a dynamic land cover type on Earth. In short time (days), it can change into ice/vegetation/mud and sand, and in long term, it can change into agricultural and impervious surfaces. Mostly, water is distributed on flat terrains with low slopes (< 8°). We introduce three difficulties in optical water mapping: the spectral confusion between low/high-reflectance water and nonwater types, such as the confusion between water and mountain/cloud shadow, and the spectral mixing problems. Next, algorithms designed for image enhancement and optical water mapping are introduced, including spectral indices for water, target detection methods, spectral mixing analysis techniques (consisting of endmember extraction and spectral unmixing), and some basic concepts of image classification that have been used for water extraction purposes. This has been followed by a brief review on some state-of-the-art global surface water products. Finally, we provide a perspective on optical water mapping in the future.