Journal of Korean Society of Agricultural Engineers ISSN:1738-3692(Print) 2093-7709(Online) 한국농공학회논문집

Evapotranpiration (ET) is one of the important factor in Hydrological cycle and irrigation planning. In this study, temperature-based artificial neural network (ANN) model for daily reference crop ET estimation was developed and compared with reference crop evapotranpiration (ET₀) from FAO-56 Penman-Monteith method (FAO-56 PM) and parameter regionalized Hargreaves method. The ANN model was trained and tested for 10 weather stations (5 inland stations and 5 costal stations) and two input climate factors, maximum temperature (T_max), minimum temperature (T_min), and extraterrestrial radiation (RA) were used for training and validation of temperature-based ANN model. Monthly reference ET by the ANN model also compared with parameter regionalized Hargreaves method for ANN model applicability evaluation. The ANN model evapotranspiration demonstrated more accordance to FAO-56 PM evapotranspiration than the ET₀ from parameter regionalized Hargreaves method(R-Hargreaves). The results of this study proposed that daily reference crop ET estimated by the ANN model could be used in the condition of no sufficient climate data.