Comparative study of Hargreaves's and artificial neural network's methodologies in estimating reference evapotranspiration in a semiarid environment

The Penman-Monteith equation (PM) is widely recommended because of its detailed theoretical base. This method is recommended by FAO as the sole method to calculate reference evapotranspiration (ETo) and for evaluating other methods. However, the detailed climatological data required by the Penman-Monteith equation are not often available especially in developing nations. Hargreaves equation (HG) has been successfully used in some locations for estimating ETo where sufficient data were not available to use PM method. The HG equation requires only maximum and minimum air temperature data that are usually available at most weather stations worldwide. Another method used to estimate ETo is the artificial neural network (ANN). Artificial neural networks (ANNs) are effective tools to model nonlinear systems and require fewer inputs. The objective of this study was to compare HG and ANN methods for estimating ETo only on the basis of the temperature data. The 12 weather stations selected for this study are located in Khuzestan plain (southwest of Iran). The HG method mostly underestimated or overestimated ETo obtained by the PM method. The ANN method predicted ETo better than HG method at all sites.