Estimation of Water Table Depth Using Wavelet-ANFIS: A Case Study

Abstract

Predicting fluctuation in groundwater level (GWL) is extremely significant in managing water resources problems and hydrological phenomenon. The objective of present study is comparison of ANFIS and Wavelet-ANFIS models to forecast GWL for altered period of prediction and finding model effectiveness based on their performances. mean absolute error (MAE), root mean square error (RMSE) and correlation coefficient (R2) statistical tools were utilized to evaluate accurateness of the aforesaid models. Results indicate that Wavelet-ANFIS model with R2 of 0.9628 performed better than ANFIS model with R2 0.9247. It can be seen that wavelet transform helps in improving the efficiency of ANFIS model to forecast GWL. It is observed that the forecast done by Wavelet-ANFIS is more precise than those by ANFIS.