Modeling of daily pan evaporation of Lake Eğirdir using data-driven techniques

Gene Expression Programming (GEP) and Radial Basis Function Network (RBFN) models are developed to estimate daily pan evaporation which is an important parameter in hydrological and meteorological studies. Meteorological parameters used to estimate daily pan evaporation from Lake Eğirdir at the southwestern part of Turkey are air temperature (T_a), water temperature (T_w), solar radiation (R_C) and relative humidity (R_h). The GEP and RBFN models are developed by using different input combinations with these four parameters. The GEP model with T_a, T_w, R_C and R_h parameters has the highest R² (0.812) and the lowest MSE (0.0066 mm/day) values for testing set. The R² value of the best RBFN model is obtained as 0.794 with T_a T_w and R_h. It can be seen that the GEP model was slightly more superior to the RBFN model. It is concluded from the results that GEP and RBFN can be proposed as an alternative to daily pan evaporation measurement.