37-10 Groundwater flow modeling of flood-prone seepage lakes in a glacial aquifer, Chequamegon-Nicolet National Forest

Session: Innovations in Research of Groundwater-Surface Water Interactions over Multiple Spatio-Temporal Scales

Presenting Author:

Authors:

Abstract:

Recent groundwater flooding in the Chequamegon-Nicolet National Forest (northern Wisconsin) resulted in extreme and variable lake-level rises across a watershed with minimal surface hydrologic connectivity. The area contains seepage lakes and disappearing streams, but no surface water inlets or outlets, highlighting the dominant role of groundwater in controlling water levels. This study investigates the geologic controls underlying these patterns by developing a numerical groundwater flow model informed by data spanning multiple spatial and temporal scales.

The model domain includes Pigeon Lake – a 1 km² seepage lake – and its surrounding 200 km² watershed, which borders the regional divide separating the Lake Superior and Mississippi River basins. The surficial aquifer consists of sandy glacial deposits varying dramatically in thickness from absent (where crystalline bedrock outcrops in the south) to over 150 m feet in the north. This landscape is a product of a highly dynamic glacial history, shaped by repeated advancement, retreat, and outwash deposition from the Last Glacial Maximum. We hypothesize that heterogeneity in sediment types resulting from this history drives spatial variability in flooding, though sparse well data limit direct verification.

To address data limitations, we integrate observations across multiple scales. At the regional scale, aerial imagery and LiDAR datasets complement well construction reports (WCRs) by identifying lakes with prolonged flooding. These data provide spatial snapshots of changing lake conditions and help characterize the degree of groundwater-lake connectivity across the watershed. At the local scale, three years of hydrologic monitoring and stable isotope mass balance analysis at Pigeon Lake provide detailed insight into groundwater exchange that supports model calibration. The modeling also recognizes differences in temporal resolution of the various datasets. WCR water levels represent long-term averages and climatic variability but may be skewed by uneven temporal coverage, while field measurements and aerial imagery reflect instantaneous and transient conditions.

Through iterative conceptual model testing, we explore how different representations of glacial heterogeneity influence flooding risk and groundwater response times. The resulting model improves understanding of groundwater-surface water interactions and highlights the importance of varied data considerations in capturing the complexity of glacial systems.

Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025

doi: 10.1130/abs/2025AM-10457

© Copyright 2025 The Geological Society of America (GSA), all rights reserved.