Efficiency Comparison of Fuzzy Regression Models with the Penman-Monteith Method in Estimating of Monthly Reference Evapotranspiration of Neyshabour Plain

In this study, fuzzy linear and fuzzy least-squres regression approach was employed to estimate the monthly reference evapotranspiration of Neyshabour plain. The data used, including maximum temperature (Tmax), minimum temperature (Tmin), mean temperature (Tmean), relative humidity (RH), solar radiation (Rs) and wind speed (U2), were obtained from synaptic meteorological station of Neyshabour. Three different scenarios were designed to estimate the evapotranspiration for either fuzzy linear or fuzzy least-squres regression models. Mean absolute error (MAE), root mean square error (RMSE), and the coefficient of determination (R2) were used to evaluate the performance fuzzy regression models and its comparison with FAO-56 Penman-Monteith. Results indicated that the fuzzy linear regression model in January and the fuzzy least squares regression model in October had the highest and lowest accuracy with R2 of 0.903 and 0.502, respectively. Among the new proposed models, the fuzzy linear regression under scenario FLR1 (Inputs included Tmax, Tmin, RH and U2) had the highest accuracy, however, in both regression models, despite having lower input parameters (Tmean, RH and Rs), the second scenario, was comparable with other and therefore it can be used in data deficit conditions as an optimal approach in determining ETo for irrigation planning and water resource management.