Gene Expression Programming Capability Assessment in Estimating the Potential Evapotranspiration Compared to Hargreaves-Samani Equation

Due to the water scarcity in areas with hot climates, an accurate estimation of evapotranspiration has a significant impact on water resources management and planning. In this paper, the gene expression programming (GEP) approach was used to estimate the daily evapotranspiration in Yazd (hot and dry climate) and Jiroft (warm and humid climate) synoptic stations. The results of this method along with the results of empirical relation of Hargreaves-Samani were validated with those of the empirical relation of Penman Mantis FAO-56. Statistical indices of coefficient of determination (R2), root-mean-square error (RMSE) and mean absolute error (MAE) were employed to evaluate the efficiency of the utilized methods. The results indicated that GEP model (with R2 = 0.951 and RMSE = 0.66 mm/day for Yazd and R2 = 0.958 and RMSE = 0.65 mm/day for Jiroft) had higher performance compared to Hargreaves-Samani relation. Therefore, the use of GEP as a model with high capabilities is recommended for estimation of evapotranspiration in areas with hot-dry and warm-humid climates