EVOLVING CONCEPTUAL HYDROGEOLOGIC MODELS FOR COMPLEX GLACIAL AQUIFER SYSTEMS

Despite advances in groundwater flow and contaminant transport modeling techniques over the past two decades, a need for improved hydrogeologic models of complex glacial aquifer systems persists. Failure to effectively contain and remediate groundwater pollutants in such systems is motivating the development of revised conceptual models for the three-dimensional variability of aquifer properties and the morphology of contaminant plumes. Over the next decade, advances in flow and transport modeling within glacial strata will result from improvements in conceptual models for the spatial variability of aquifer properties and the architecture of solute plumes. The tools that will generate these advances are sequence stratigraphy, geostatistics, and geographic information systems (GIS). This presentation will illustrate conventional and emerging approaches to aquifer description, plume mapping, and groundwater modeling with examples drawn from the Pall Life Sciences (formerly Gelman Sciences) groundwater contamination site, located west of Ann Arbor, Michigan, USA.

At the Pall Life Sciences site, wastewater containing 1,4-dioxane was discharged into unlined seepage lagoons and spray irrigated across a 15 acre field from 1967 to 1985. Plumes of 1,4-dioxane-contaminated groundwater have migrated several kilometers from the site in different directions through the underlying 80m of glacial drift. Because 1,4-dioxane is readily soluble in water and does not easily degrade or adsorb to soil particles, it provides a tracer-like record of solute transport. More than 120 monitoring wells and 16 extraction wells have been drilled to investigate and remediate 1,4-dioxane in the area. In spite of attempts to contain and remove the contaminant following its discovery two decades ago, remediation activities have met with limited success; and efforts to characterize and model the aquifer continue as the deepest known plume advances toward a municipal water supply well and the Huron River. Analysis of the Pall Life Sciences dataset suggests that the utilization of geostatistics within a deterministic sequence stratigraphic framework and the application of GIS tools to subsurface investigations will improve our ability to model and predict contaminant transport in complex glacial aquifer systems.