Assessing the physical and empirical reference evapotranspiration (ETo) models and time series analyses of the influencing weather variables on ETo in a semi-arid area.

Accurate estimation of irrigation requirement is necessary for conserving the quantity and quality of water resources. Generally, irrigation requirement is estimated by calculating reference evapotranspiration (ETo). In this study, radiation-based, temperature-based, and combination-based ETo models were assessed based on the monthly averaged weather data between 1987 and 2017. The combination-based Standardized ASCE Penman-Monteith (ASCE PM Std.) was selected as the benchmark model due to its global acceptance and accuracy. Results showed that the combination-based Penman models were ranked as the top models among the other ETo models. However, if some weather variables are missing, the Priestly-Taylor model followed by the Makkink and Turc models (all as radiation-based models) were the next recommended ETo models.The performance of the temperature-based models and some other radiation-based models (FAO24 Radiation and Jensen-Haise) were not satisfactory. Trend and change point detection analyses on air temperature, relative humidity, and wind speed showed that the study area is getting warmer and drier, which indicate that ETo would increase in the study area. Therefore, it is recommended to use the ETo models that consider the majority of the weather variables that influence ETo. The results of this study could serve as a reliable guide for selection of appropriate ETo models to protect water resources in arid and semi-arid areas. .