Prediction of groundwater level using artificial neural network as an alternative approach: a comparison assessment with numerical groundwater flow model

Over-exploitation of groundwater in arid and semi-arid regions requires an understanding of dynamics for effective management strategies. This study uses an artificial neural network (ANN) to predict groundwater levels using initial water level, pumping, and hydrologic and meteorological input variables. Sensitivity analysis assessed predictors' impact on accurate groundwater level prediction. The accurate prediction of the ANN model was examined through different performance evaluation criteria and the result achieved is satisfactory as it is compared to reasonable statistical indicator values. A comparison of results simulated by the ANN model and numerical groundwater flow model (MODFLOW) was carried out and it was found that the prediction of the ANN model was consistently superior to the numerical model. Therefore, data-driven models such as ANN can provide better predictions of groundwater levels that aid groundwater resource management.