Prediction of evapotranspiration based on climatic data with Fuzzy Logic Method

In this study, Evapotranspiration (ET) which is one of the most important parameter of hydrologic cycle estimation and calculation methods are investigated. Modelling ET mathematically is hard, due to this reason Adaptive Neuro – Fuzzy Inference system (ANFIS) and Multi Linear Regression (MLR) methods are used for modelling because of these methods are in use for nonlinear modelling. Close to 75 percent of data set including 2287 daily ET, Solar Radiation (SR), Air Temperature (T), Wind Speed (U) and Relative Humidity (RH) meteorological parameters are used as training set and remaining 571 daily data as test set. Data set is gotten from De Soto County, Florida, USA station. In the first part of the study, ANFIS and MLR methods are used for the investigation of parameter effect on ET. In the second part of the study, empirical Hargreaves – Samani, Ritchie, Penman Monteith and Turc formulas are applied to the data set. For the comparison of ANFIS, MLR and empirical equations results, determination coefficient (R), Mean Absolute Error (MAE) and Mean Square Error (MSE) statistics are used. As a result it is understood that using SR, T, U, RH combination as input for ANFIS shows better performance than MLR method and empirical equations.