Water Table Prediction by Using Time Series Models and Adaptive Neural Fuzzy Inference System

Choubin, Bahram; Malekian, Arash; Sajedi Hosseini, Farzaneh; Rahmati, Omid

doi:10.22059/ijswr.2014.36606

Water Table Prediction by Using Time Series Models and Adaptive Neural Fuzzy Inference System

Document Type : Research Paper

Authors

¹ PhD Candidate, Sari University of Agriculture Science & Natural Resources, Sari, Iran

² Assistant Professor, Faculty of Natural Resources, University of Tehran, Karaj, Iran

³ MSc. Student, Sari University of Agriculture Science & Natural Resources, Sari, Iran

⁴ PhD Student, University of Lorestan, Lorestan, Iran

10.22059/ijswr.2014.36606

Abstract

Modeling in arid regions to better manage water resources is very important. Groundwater is an important water resource in arid regions. The purpose of this study was to assess the performance of adaptive neuro fuzzy inference (ANFIS) and time series models to predict the water table. In this study, groundwater levels of Shiraz plain for one month ahead were forecasted by using time series models and ANFIS model. In the ANFIS model has been used Gamma and M-test for determine of the optimal input combination and training and testing data length. Performance of different models was compared with the parameters of the error and Taylor diagrams. ANFIS model results showed that this model with membership function of Π-shaped has better performance than the rest of membership functions. Performance comparison of the models indicated very suitable performance of the ARIMA (2, 1, 2) model than ANFIS models with different membership functions.

Keywords