Comparison of three evapotranspiration models with eddy covariance measurements for a Populus euphratica Oliv. forest in an arid region of northwestern China

The accurate estimation of evapotranspiration (ET) in arid regions is important for improving the water use efficiency of vegetation. Based on successive observations from May to October of 2014, we estimated the ET of a Populus euphratica Oliv. forest during the growing season in an extremely arid region using the PM (Penman-Monteith), SW (Shuttleworth-Wallace) and SSW (an improved canopy transpiration model) models. Estimated ET values were compared with those of the eddy covariance measurements. Results indicated that the actual ET of the P. euphratica forest was always overestimated by the PM model. The accuracy of the SW model was higher than that of the PM model. However, some data were not easily obtained because of the complicated structure of the SW model. The newly proposed SSW model gave the most accurate ET values, and its accuracy was higher at hourly than at daily time scale. In conclusion, the SSW model is more suitable for sparse vegetation system at large scales in extremely arid regions.