Modeling saltwater upcoming with artificial neural networks

Salt-water upcoming and intrusion into freshwater groundwater systems can result in serious water quality problems, affecting potable supplies in coastal areas around the world. In this study, artificial neural networks (ANN) were developed to accurately predict highly time-variable specific conductance values in a real-world unconfined coastal aquifer. Unlike physical-based models, which require hydrologic parameter inputs, such as horizontal and vertical hydraulic conductivities and porosity, ANNs can "learn" system behavior from easily measurable variables. In this study, the ANN input predictor variables were the initial specific conductance (a measure of dissolved ions) measured at a monitor well, total precipitation, mean daily temperature, and total pumping extraction. The ANNs predicted conductance at a single monitoring well located near a high capacity municipal-supply well over time periods ranging from 30 days to several years. Model accuracy was compared against measured/interpolated values and predictions made with linear regression (I-R), and in general, achieved excellent prediction accuracy. The ANNs were also used to conduct a sensitivity analysis that quantified the importance of each of the four predictor variables on final conductance values, providing valuable insights into the dynamics of the system. The results demonstrate that the ANNs technology can serve as a powerful and accurate prediction and management tool, minimizing degradation of ground-water quality to the extent possible by identifying appropriate pumping policies under variable groundwater system and weather conditions.