Evaluating of eight evapotranspiration estimation methods in arid regions of Iran

Evapotranspiration (ET) plays an important role in efficient irrigation and water resources management. ET can be accurately estimated via the FAO-Penman-Monteith method (ET_FPM), which is known as the standard reference method for the ET estimation. However, this model requires more detailed meteorological data than are generally available at most weather stations. Alternative methods for predicting daily ET should be found which are simpler to work with less input data without scarifying prediction accuracy. The main purpose of this study was to determine a reliable alternative ET models that requires fewer input data compared with the ET_FPM as a reference method. Here, five radiation-based models including Makkink (ET_MAK), Priestley and Taylor (ET_PT), Abtew (ET_ABT), Jensen-Haise (ET_JH), McGuinness and Bordne (ET_MB), and three temperature-based models including Hargreaves and Samani (ET_HS), Hamon (ET_HAM), Linacre (ET_LIN) were evaluated by comparing them with PET_FPM on a daily and growing season scale, using long-term data from 13 meteorological stations in northeast Iran. The statistical analysis revealed that the ET_HS, ET_JH, ET_HAM, and ET_LIN were the best methods for predicting daily ET in 54%, 15%, 15% and 15% of areas, respectively. In general, the temperature-based methods outperformed the radiation-based-methods in the study area. The cumulative values of daily ET during the growth periods showed that the Jensen-Haise method performed the best for the warm growing season compared to other alternative methods, while the Hargreaves-Samani method was the best prediction method for the cool growing season. Nevertheless, divergence between estimations of the best alternative methods and the reference method showed that the best ET alternative methods might be unreliable in some regions, and could not be recommended for estimating crop water requirements. Accordingly, the spatiotemporal variability in predictability performance of ET models should be taken into account prior to use.