Shallow lateral DNAPL migration within slightly dipping limestone, southwestern Kentucky

Abstract

Dense non-aqueous phase liquid (DNAPL) migration from source areas has mostly been governed by gravitative movement along beds dipping ∼1° NW that show diffuse-flow, semi-confined characteristics. This shallow, phreatic migration has occurred within relatively fractured and vuggy beds of the Mississippian St. Louis Limestone at depths beneath ground of 12–15 m, over lateral distances to 230 m. Near source areas, vertical product migration is occasionally deeper than 21 m beneath ground. Migration along the top-of-bedrock (TOB—generally at depths of 1.5–4.6 m) appears to be restricted to distances of less than 46 m, and therefore is unlikely to foster significant offsite migration. However, opportunely located TOB lows can provide routes for free-product entry, into the main lateral-migration beds, that are not located beneath source areas. Preferred migration along layer diplines appears to reflect: (1) bedding–plane partings; (2) interconnected, layer-bounded joints; (3) minor solutional enhancement of the above two features, especially the first; (4) sufficiently large and interconnected pore spaces, dissolution pits, and vugs; and (5) any possible solution conduits (for which there is no direct evidence, but some local occurrences are hinted at by site hydraulic characteristics). Local deflection of migrating free product away from the bedding dipline can be expected where (1) NW-trending bedding, joint, or vug-network pathways are poorly developed or absent in proximity to such available pathways trending along other directions, or (2) clay-filled depressions in the TOB surface, deeper than about 12 m beneath ground, act as barriers to downdip migration in the uppermost bedrock. Where such critically deep TOB lows might be filled with relatively permeable sands, lateral DNAPL migration from the uppermost bedrock into the soil mantle could occur along the updip slope of the TOB low. The directions of groundwater flow and transport (to the W) and DNAPL migration (to the NW) are divergent, which probably led to a plan-view aqueous plume that, over decades, had advanced toward the NW and W while tapering toward the migration limit to the NW.