Application of ANFIS and GP Models to Estimate Monthly Reference Crop Evapotranspiration in Northwest of Iran

Evapotranspiration is one of the important components of hydrological cycle whose accurate estimate is needed for design and management of irrigation systems، simulation of crops products، and programming water resources management. In this research، to predict monthly reference evapotranspiration، ANFIS and GP models were employed and 38 years (1973-2010) of data were collected from six synoptic weather stations located in the northwest of Iran. At first، monthly reference evapotranspiration was estimated by FAO-Penman-Montieth method for the selected stations and was considered as the output of GP and ANFIS models. Then، a regression equation between effective meteorological parameters and evapotranspiration was fitted and different input patterns for the models were determined. Relative humidity as the less effective parameter was deleted from input of the models. Also، in this study، to investigate effect of “memory” on prediction of evapotranspiration، one، two، three and four months lags were used as the input of the models. Results showed that both models estimated monthly evapotranspiration with high accuracy، but ANFIS model was better than GP model.