Modeling Reference Evapotranspiration Using Evolutionary Neural Networks

The ability of evolutionary neural networks (ENN) to model reference evapotranspiration (ET0) was investigated in this study. The daily climatic data, solar radiation, air temperature, relative humidity, and wind speed of three stations in central California, Windsor, Oakville, and Santa Rosa, were used as inputs to the ENN models to estimate ET0 obtained using the FAO-56 Penman-Monteith equation. In the first part of the study, a comparison was made between the estimates provided by the ENN and those of the following empirical models: the California Irrigation Management System, Penman, Hargreaves, modified Hargreaves, and Ritchie methods. Root-mean-squared error, coefficient of efficiency, and correlation coefficient statistics were used as comparing criteria for the evaluation of the models' accuracies. The ENN performed better than the empirical models. In the second part of the study, the ENN results were compared with those of the conventional artificial neural networks (ANN). The comparison results revealed that the ENN models were superior to ANN in modeling the ET0 process. DOI: 10.1061/(ASCE)IR.1943-4774.0000333. (C) 2011 American Society of Civil Engineers.