Estimation of groundwater level of Dörtyol region-Hatay province by fl, multiple li̇near regression and arma models

Accurate and reliable estimation of groundwater level is important for the development and management of water resources. In this study, the performance of autoregressive moving averages (ARMA) models, Multiple Linear Regression and Fuzzy Logic (BM) models in predicting groundwater level were investigated. In the first part of the study, 3 different models for ARMA models (ARMA 1-1, ARMA 3-3, ARMA 5-5) using the underground water level time series, 2 for the Multiple Linear Regression model (MLR 3-3), (MLR 2) For BM models 2 models (BM 1-1), (BM2) were applied. As a field of application, it is applied for wells of State Water Works (DSİ) 5512 of Dörtyol Region of Hatay Province. In the continuation of the study, the hydrological parameters such as Precipitation (P), average air temperature (T), relative humidity (RH), wind speed (WS) monthly average changes are taken as input and with the addition of a model in which meteorological effects are involved and the change of groundwater level (GWL) It was. For the Hatay-Dörtyol region, the 147-month-old hydrological data set between 2000 and 2015 was taken from the General Directorate of Meteorology (MGM) and the groundwater level was estimated. In the evaluation of the model and measurement results, as the statistical performance evaluation parameters, the coefficient of determination (R2), mean square error (MSE), mean absolute error (MAE) were used. As a result of this study, ARMA,MLR and FL models performed well for YASS estimation. In particular, the UN model yielded better results than the ARMA and MLR model.