Comparison of ANN, ANFIS and Regression Models to Estimate Groundwater level of Neyshaboor Aquifer

Groundwater and water resources management play a key role in sustainable water resources management in arid and semi-arid areas. Neyshaboor plain is one of the most plains of Khorasan Razavi province، which has an important role in agricultural production. Unallowable discharges of the resources is causing that water table has 74 cm drawdown. Purpose of this study is evaluated of classic models and expert systems (Artificial Neural Network ANN and Adaptive Neurao-Fuzzy Inference Systems ANFIS) in prediction of groundwater table. In this study، effected parameters in water table example (monthly precipitation and discharge) detected and raster maps was gained by geostatistical methods. Data bank was gained by Arc GIS software from raster maps to training and testing expert systems. Regression equations of effective parameters in water table were resulted from data bank. Results showed that ANFIS models often had the most accuracy to predict monthly and regression models had the lowest performance. The ANN models had suitable accuracy on summery months.