Evapotranspiration of a Populus euphratica forest during the growing season in an extremely arid region of northwest China using the Shuttleworth-Wallace model

Thus far, measurements and estimations of actual evapotranspiration (ET) in extremely arid areas are still insufficient. Based on successive observations from June-September 2014, we simulated ET of a Populus euphratica Oliv. forest during the growing season in an extremely arid region of northwest China using the Shuttleworth-Wallace (S-W) model. Simulated ET values were compared to those of the eddy-covariance (EC) method on a 1 h interval. With a root mean square error (RMSE), relative error (RE) and mean absolute error (MAE) of 0.192, 3.100 and 0.165 mm h(-1), respectively, model performance was not satisfactory. In particular, on days with strong winds (Sep. 11-13), deviations between simulated and observed ET values increased to 0.275, 0.878 and 0.251 mm h(-1), RMSE, RE and MAE respectively. These values were significantly greater than those in other study periods and were most likely owing to sharp increases in wind speed. As a result, there were substantial advective effects, which is not consistent with the assumption of the S-W model that there are no advective effects or mesoscale circulation patterns induced by surface discontinuities.