Estimation of evapotranspiration components across China based on a modified Priestley–Taylor algorithm with assimilated soil moisture data

Consisting of evaporation from wet surfaces (E) and transpiration through plants (T), evapotranspiration (ET) is an integral part of earth’s ecological and climate systems. Since the different ET components have their certain water resource function and ecological significance, accurate estimation of regional ET components is essential to better understand water cycle and surface energy budget. Incorporating soil relative humidity (SRH) into remote sensing ET algorithm, this study presented an update to the widely-used Priestley Taylor-Jet Propulsion Laboratory (PT-JPL) ET algorithm to incorporate spatially explicit monthly SRH control on soil evaporation (E_s) and canopy transpiration (T). The updated algorithm (i.e., PT-SM) was evaluated using 17 eddy covariance towers across different biomes, and 24 hydrological catchments across different climatic regions of China, respectively. The PT-SM model shows increased R² and NSE, and reduced RMSE and Bias, with the greatest improvements occurring in water-limited regions. SRH incorporation into E_s can improve ET estimates by increasing R² and NSE by 3% and 17%, respectively, and RMSE and Bias were reduced by 13% and 26%, respectively, while SRH incorporation into T would improve ET estimates by raising R² and NSE by 6% and 27%, respectively, and RMSE and Bias were reduced by 32% and 63%, respectively. We apply the algorithm to the whole China using SRH data at depths [10-cm, 20-cm, and 50-cm] and a resolution of 0.5° × 0.5° assimilated by the farmland soil moisture observation, the NASA’s Gravity Recovery And Climate Experiment (GRACE) solutions and observed precipitation. The mean annual of total estimated ET increased from the northwest to the southeast, with E_s/ET and T/ET roughly presenting opposite spatial distribution characteristics.