The Three-Temperature Model to Estimate Evapotranspiration and its Partitioning at Multiple Scales: A Review

Abstract. Remotely estimating evapotranspiration (ET) and partitioning of its two components, soil evaporation (E) and vegetation transpiration (T), have been challenges for ET studies and their applications. In this study, we reviewed the three-temperature model (3T model) and its applications for thermal remote sensing under multi-scale conditions. There are two submodels in the 3T model: the soil evaporation submodel and the vegetation transpiration submodel. E and T can be separately estimated with these two submodels, and then ET can be obtained by putting E and T together. One of the most significant advantages of the 3T model is that only a minimum number of parameters are required. The necessary parameters for E estimation are net radiation, soil heat flux, surface temperature (drying soil and dry soil), and air temperature, while the parameters for T estimation are net radiation, surface temperature (canopy and dry canopy), and air temperature. Verifications and applications were carried out using ground-based and space-based thermal remote sensing from chamber scale to catchment scale. Verifications were carried out by comparing the estimated results with estimated ET from the Penman-Monteith method and measured ET from weighing lysimeter, Bowen ratio, eddy covariance, and water budget methods. Results showed that the "3T model + thermal remote sensing" approach is a reliable and practical method for remotely estimating ET, especially in arid and semi-arid regions. Therefore, it can be applicable for irrigation water management, water and energy budget monitoring, thermal environment monitoring and evaluation, and other applications.