Water Resources Research Act Program

Details for Project ID GU_2025_KIM_1

The analysis of saltwater intrusion into a freshwater resource with deep learning tools.

Institute: Guam
Year Established: 2025 Start Date: 2026-01-01 End Date: 2026-12-31
Total Federal Funds: $23,050 Total Non-Federal Funds: Not available

Principal Investigators: Yong Sang Kim

Project Summary: Currently, the Guam Waterworks Authority (GWA) supplies 85% of the island's potable water (45 MGD) from its primary source, the Northern Guam Lens Aquifer (NGLA). To meet demand, GWA operates 100 to 120 deep vertical production wells within this aquifer. However, many of these wells have exhibited increasing chloride levels over the past several decades. WERI's chloride research, which focuses on GWA production wells, has revealed a significant upward trend in chloride concentrations in 70% of freshwater production wells. Several factors may be contributing to this increase, including: 1) higher water production rates, 2) rising sea levels since 1993 and 3) variability in precipitation rates and intensity. Deep learning (DL) is a valuable modeling tool for analyzing historical datasets and predicting contaminant trends. Environmental field monitoring data are inherently complex due to the influence of multiple factors, such as precipitation, tides, and sea-level fluctuations. However, DL applications are expected to provide more accurate insights into chloride changes in groundwater resources. In this study, we will select production wells located in the basal zone of the Finegayan and/or Yigo-Tumon basins as target sites for chloride variation projections. By utilizing historical chloride data, the DL model will be trained to analyze chloride variation trends in the freshwater lens of the aquifer. After calibrating and validating the model, future chloride trends will be projected, and the potential impacts will be assessed.