Assessment of reference evapotranspiration (ETo) methods under data scarcity scenarios (case study: Kermanshah)

Many studies have evaluated different methods to estimate evapotranspiration as one of the most importantparameters of water resources and irrigation management in recent decades. The result of these studies confirmed that the FAO Penman-Monteith (PM) is the standard method for calculating reference evapotranspiration. This method requires numerous meteorological data such as air temperature, solar radiation, air humidity and wind speed. Since in most regions of Iran, the weather station is not available or all of the meteorological data is not measured, hence investigation of methods to estimate evapotranspiration is necessary in these situations. In this study, data scarcity scenarios, (when one, two and three meteorological variables are not available: relative humidity, wind speed and sunshine hours), have been evaluated. Also, the FAO-56 manual recommended methods (PM-R), PM-CI (computation-then-interpolation), PM-IC (interpolation-then-calculation) and Hargreaves–Samani (HS) were compared using different statistical parameters. The results showed that in absence of wind speed (when one or two meteorological parameters are not available), the PM-IC method estimated more accurately (R2> 0.96), while precision of the PM-R method is higher (R2≈ 0.99) in absence of other weather parameters (sunshine hoursand air humidity). Also, the HS method, temperature-based methods (absence of three meteorological data), provided a more acceptable result (PBIAS = -3.091, ME = -0.156, MAE = 0.337) than other methods.