Groundwater Sustainability in the Michigan Lowlands – Understanding the Complex Interplay of Natural Brine Upwelling, Human Activity, and Climate Change

Abstract:

This research investigates a relatively unknown, basin-scale contamination hazard threatening the long-term sustainability of groundwater resources in the Lower Peninsula of Michigan (LPM). Many studies across LPM report elevated chloride concentrations ([Cl^-]>250 mg/L) in the near-surface water, including the Michigan lowlands (west LPM), the Saginaw lowlands (east LPM) - where the upwelling of deep brines was inferred as the source - and at discrete points further inland. This begs the question: do the scattered occurrences of elevated [Cl^-] share a common underlying mechanism, namely, the upwelling of brine? We addresses this question with data-driven modeling of the [Cl^-] dynamics and hydrology across multiple scales, capitalizing on two massive statewide databases with water quality information from over 10⁶ wells and static water levels (SWL) from roughly 500,000 wells. Long-term SWL were mapped at the basin-, regional-, and local-scale to identify recharge/discharge areas. Wells with elevated [Cl^-] were overlaid to each map, revealing a consistent pattern of low [Cl^-] in recharge areas and elevated [Cl^-] in discharge areas. This multi-scale relationship provides compelling evidence for upwelling of brines, a concept supported by the analysis of 450 well samples collected in the Michigan lowlands which showed [Cl^-] generally increases with depth. We assessed temporal trends by comparing field results to past records at 249 sampling locations. Most of the sites exhibit an increase of [Cl^-] in recent years, which roughly coincides with a time period when Lake Michigan-Huron levels systematically decreased. Moreover, recent increases in [Cl^-] are focused to areas where temporal analysis of SWL indicated increased well withdrawals. The results show that the contamination is caused by upwelling brines, and suggest that it is becoming more severe in recent years. Future work will integrate multi-scale, process-based groundwater modeling with further data collection/analysis to simulate the complex interplay of human activity, natural upwelling, and climate change. Insight from the modeling effort will be used to inform resource planners and policy makers of various groundwater management options available to vulnerable communities across the Lower Peninsula of Michigan in years to come.