Advances in a two-source energy balance model: Partitioning of evaporation and transpiration for cotton

Accurate partitioning of the evaporation (E) and transpiration (T) components of evapotranspiration (ET) in remote sensing models is important for evaluating strategies aimed at increasing crop water productivity. A two-source energy balance (TSEB) model designed for row crops solves the energy balance of the soil-plant canopy-atmosphere continuum using surface brightness temperature. By solving the energy balance of the soil and canopy separately, the TSEB model can calculate E and T, which cannot be done with single-source models. However, few studies have tested the TSEB model where E or T measurements were available, which until recently has impeded its advance. This paper reviews recent physically-based advances of the TSEB model. The advances were tested using measurements of E, T, and ET by microlysimeters, sap flow gauges, and weighing lysimeters, respectively, at Bushland, Texas, USA for irrigated cotton having a wide range of canopy cover. Root mean square error (RMSE) and mean bias error (MBE) were 0.54 mm d^-1 and - 0.19 mm d^-1, respectively, between measured and calculated E. RMSE and MBE were 0.87 mm d^-1 and 0.31 mm d^-1, respectively, between measured and calculated T. This was deemed an improvement over previous TSEB model versions, which overestimated E and underestimate T, resulting in RMSE and MBE up to 3.8 mm d^-1 and -3.5 mm d^-1, respectively. Ongoing research includes testing the TSEB model using different remote sensing platforms, from ground-based to satellite scales.