Estimating Daily Reference Crop Evapotranspiration Using Artificial Intelligences-Based ANFIS and ANN Techniques and Empirical Models

Evapotranspiration، as a major component of the hydrologic cycle، is important in water resources development and irrigation planning. This paper aimed at investigating the abilities of Adaptive Neuro-Fuzzy Inference System (ANFIS) to estimate daily reference evapotranspiration (ET0). The daily climatic variables such as air temperature، relative humidity، wind speed and solar radiation from two weather stations (Salvatierra and Zambrana) in Spain equipped with electronic sensors for collecting of climatic data، were used as inputs to the Neuro-Fuzzy model to estimate ET0. Comparisons were made among the estimates provided by the ANFIS، Artificial Neural Networks (ANNs) and following the empirical models: Hagreaves – Samani، Ritchie، Makkink and Turc. The comparisons revealed that the ANFIS models (with RMSE between 0. 276-0. 437 mm) could be employed successfully in modeling evapotranspiration process. The ANNs (with RMSE between 0. 298-12. 5 mm) were also found to perform better than the empirical models in this regard.