Calibrating a model to 30years of data to capture the accumulation of chloride from winter deicers in a shallow aquifer

• Published in: Journal of the American Water Resources Association Vol. 59; no. 3; pp. 539 - 553
• Main Authors: Cullen, Cecilia, Abrams, Daniel B, Mannix, Devin H, Kelly, Walton R
• Format: Journal Article
• Language: English
• Published: Middleburg Blackwell Publishing Ltd 01-06-2023
• Subjects: Accumulation; Anthropogenic factors; Aquifers; Bedrock; Calibration; Cells; Chloride; Chlorides; Contamination; Deicers; Deicing; Dolomite; Dolostone; Environmental protection; Geology; Glacial deposits; Government agencies; Human influences; Residence time; Sediments; Stormwater; Water pollution; Water wells; Winter; United States > US; Illinois
• ISSN: 1093-474X; 1752-1688
• DOI: 10.1111/1752-1688.13105

Winter deicers are a major source of chloride contamination to shallow aquifers in northern latitudes. In the greater Chicago area of northeast Illinois, chloride has been accumulating for decades and in many places now exceeds the United States Environmental Protection Agency secondary standard of 250 mg/L. MODFLOW‐NWT and MT3D‐USGS were used to simulate the accumulation of chloride with 30 years of data in the shallow aquifer of Will County in northeast Illinois. This aquifer is composed of unconsolidated sediments, largely from glacial deposits, overlying a fractured dolomite bedrock. To calibrate to observed heads and chloride, the model needed refined geologic features, higher chloride concentrations on cells representing commercial or industrial lands, lower chloride concentrations on cells representing roads, and point source areas to speed chloride's emergence in the aquifer. These point sources are locally significant and could represent different anthropogenic or geologic features, such as municipal stormwater infrastructure. Future simulations indicate that chloride is not at steady state in the shallow aquifer and wells are at risk of exceeding the secondary standard if winter deicing applications are not reduced. It may take decades for the full impacts of reduced deicing rates to be observed in wells, owing to the long residence time of water in the aquifer. This transient model calibration was possible because of the 30‐year dataset collected by communities and government agencies.