Adaptive Network- based Fuzzy Inference System-Genetic Algorithm Models for Prediction Groundwater Quality Indices: a GIS-based Analysis

The prediction of groundwater quality is very important for the management of water resources and environmental activities. The present work has integrated a number of methods such as Geographic Information Systems (GIS) and Artificial Intelligence (AI) methodologies to predict the groundwater quality in Kerman plain (including HCO^3, concentrations and Electrical Conductivity (EC) of groundwater). In this research work, we investigate the abilities of the Adaptive Neuro Fuzzy Inference System (ANFIS), hybrid of ANFIS with Genetic Algorithm (GA), and Artificial Neural Network (ANN) techniques, and predict the groundwater quality. Various combinations of monthly variability, namely rainfall and groundwater levels in the wells, were used by two different neuro-fuzzy models (standard ANFIS and ANFIS-GA) and ANN. The results obtained show that the ANFIS-GA method can present a more parsimonious model with a less number of employed rules (about 300% reduction in the number of rules) compared to the ANFIS model and improves the fitness criteria and so the model efficiency at the same time (38.4% in R^2 and 44% in MAPE). This work also reveals the groundwater level fluctuations and rainfall contribution as two important factors in predicting indices of groundwater quality.