Evapotranspiration is a fundamental requirement in the planning and management of irrigation projects. Methods of predicting evapotranspiration (ET) are numerous, but the Food and Agriculture Organization (FAO) of the United Nations adopted the FAO Penman-Monteith (PM) equation, as the method which provides the most accurate results for the prediction of reference evapotranspiration (ET0) in all regions and for all weather conditions. The main identified drawback in the application of this method is the wide variety of weather parameters required for the prediction. To overcome this problem, artificial neural networks (ANNs) models have been proposed to simulate the nonlinear, dynamic ET0 processes. This paper highlights both the traditional empirical PM method, and the enhancement obtained by the utilisation of ANN techniques in predicting ET0.