A review of satellite-derived terrestrial evapotranspiration: theories, methods and products

Accurately estimating terrestrial evapotranspiration (ET), the second-largest hydrologic flux in the terrestrial water cycle, is vital for understanding global water and carbon exchanges. It is difficult to measure and estimate terrestrial ET at regional and global scales. Satellites have provided us an effective tool to estimate regional and global terrestrial ET in recent decades. In this article, we provide a comprehensive review of the basic theoretical foundations, methods and products of satellite-derived terrestrial ET. The basic theoretical foundations for estimating terrestrial ET are the Monin–Obukhov similarity theory (MOST) and other budding theories (e. g. , Maximum entropy production theory, and generalized Hamilton theory). The six major categories of satellite-derived terrestrial ET methods (simple statistical methods, thermal infrared (TIR) -based surface energy balance methods, spatial contextual methods, biometeorological ET methods, integrated ET methods, and other ET methods) have made great progress in parameterization of the models, but each method has its merits and limitations. Currently, these methods provide a variety of satellite-derived terrestrial ET products (MOD16, GLASS, GLEAM, FLUXCOM, BESS, PMLV2, ETMonitor, METRIC, Hi-GLASS, SSEBop, OpenET and ECOSTRESS). However, an intercomparison of most of these products revealed that no single satellite-derived terrestrial ET product provides the most accurate terrestrial ET estimation across all plant functional types. Further advances in satellite-derived terrestrial ET will highlight the development of novel satellite-based ET theories for heterogeneous underlying surfaces, advanced ET models that integrate visible and near-infrared (VNIR), TIR and microwave satellite data, and adaptive high-accuracy satellite ET products for user requirements. This article contributes to the understanding of the ET process and satellite-derived terrestrial ET estimation, potentially aiding in the assessment of regional and global water resource variability.