Development of a novel method to monitor the temporal change in the location of the freshwater–saltwater interface and time series models for the prediction of the interface

The monitoring and prediction of temporal change in the location of the freshwater–saltwater interface including the groundwater level (GWL) is the essential information for effective management of coastal aquifer, especially for managing the saltwater intrusion. A simple and novel method for monitoring temporal change in the location of the freshwater–saltwater interface was developed and evaluated by applying to a coastal aquifer in Jeju Island, resulted in obtaining time series data of the freshwater–saltwater interface level (FSL) successfully. And applicability of artificial neural network-based time series models to prediction of the GWL and the FSL is evaluated using the observed data from the site. The prediction result shows that the input configuration including autoregressive components enhances the model performance and the recursive prediction models can be useful for the long-term prediction of the GWL and the FSL in this case study.