Characterization of Preferential Ground-Water Seepage From a Chlorinated Hydrocarbon-Contaminated Aquifer to West Branch Canal Creek, Aberdeen Proving Ground, Maryland, 2002-04

By Emily H. Majcher² , Daniel J. Phelan¹, Michelle M. Lorah¹, and Angela L. McGinty<sup>3

1</sup>U.S, Geological Survey, Baltimore, Maryland
²Geosyntec, formerly of the U.S. Geological Survey
³EA Engineering, Science, and Technology, Inc., formerly of the U.S. Geological Survey

Abstract

Wetlands act as natural transition zones between ground water and surface water, characterized by the complex interdependency of hydrology, chemical and physical properties, and biotic effects. Although field and laboratory demonstrations have shown efficient natural attenuation processes in the non-seep wetland areas and stream bottom sediments of West Branch Canal Creek, chlorinated volatile organic compounds are present in a freshwater tidal creek at Aberdeen Proving Ground, Maryland. Volatile organic compound concentrations in surface water indicate that in some areas of the wetland, preferential flow paths or seeps allow transport of organic compounds from the contaminated sand aquifer to the overlying surface water without undergoing natural attenuation. From 2002 through 2004, the U.S. Geological Survey, in cooperation with the Environmental Conservation and Restoration Division of the U.S. Army Garrison, Aberdeen Proving Ground, characterized preferential ground-water seepage as part of an ongoing investigation of contaminant distribution and natural attenuation processes in wetlands at this site. Seep areas were discrete and spatially consistent during thermal infrared surveys in 2002, 2003, and 2004 throughout West Branch Canal Creek wetlands. In these seep areas, temperature measurements in shallow pore water and sediment more closely resembled those in ground water than those in nearby surface water. Generally, pore water in seep areas contaminated with chlorinated volatile organic compounds had lower methane and greater volatile organic compound concentrations than pore water in non-seep wetland sediments. The volatile organic compounds detected in shallow pore water in seeps were spatially similar to the dominant volatile organic compounds in the underlying Canal Creek aquifer, with both parent and anaerobic daughter compounds detected. Seep locations characterized as focused seeps contained the highest concentrations of chlorinated parent compounds, relatively low concentrations of chlorinated daughter compounds, and insignificant concentrations of methane in shallow pore water samples. These seeps were primarily along the creek edge or formed a dendritic-like pattern between the wetland and creek channel. In contrast, seep locations characterized as diffuse seeps contained relatively high concentrations of chlorinated daughter compounds (or a mixture of daughter and parent compounds) and detectable methane concentrations in shallow pore water samples. These seeps were primarily along the wetland boundary. Qualitative thermal infrared surveys coupled with quantitative verification of temperature differences, and screening for volatile organic compound and methane concentrations proved to be effective tools in determining the overall extent of preferential seepage.

Hydrologic and physical properties of wetland sediments were characterized at two focused and one diffuse seep location. In the seeps with focused discharge, measured seepage was consistent over the tidal cycle, whereas more variability with tidal fluctuation was measured in the diffuse seep location. At all locations, areas were identified within the general seep boundaries where discharge was minimal. In all cases, the geometric mean of non-zero vertical flux measurements was greater than those previously reported in the non-seep wetland sediments using flow-net analysis. Flux was greater in the focused discharge areas than in the diffuse discharge area, and all fluxes were within the range reported in the literature for wetland discharge. Vertical hydraulic conductivity estimated from seepage flux and a mean vertical gradient at seeps with focused discharge resulted in a minimum hydraulic conductivity two orders of magnitude greater than those estimated in the non-seep sediment. In contrast, vertical conductivity estimates at a diffuse seep were similar to estimates along a nearby line of section through a non-seep area. Horizontal hydraulic conductivity appeared to be negatively correlated with increasing depth below land surface at both focused seep locations on the basis of cone penetrometer tests and particle-size analysis. When these hydrologic properties were extrapolated to focused and diffuse seep locations within West Branch Canal Creek, seep areas were estimated to account for about 1 percent of the total discharge area of the wetland. This 1 percent of wetland discharge area is estimated to contribute more than 20 percent of the total ground-water discharge to the creek, however, indicating that the seeps play an important role in the transport of ground water (and dissolved volatile organic compounds) to surface water.

Lithologic descriptions of wetland sediments in focused seep areas were consistent with descriptions elsewhere in the wetland. There was no visible evidence in the sediments of preferential pathways for ground-water discharge. Physical properties of seep sediments collected from cores showed characteristics of both organic and mineral sediments and were characteristic of organic clays with a relatively high fraction of fines. In seep sediment cores, the variation in physical properties with depth was consistent with hydrologic properties; however, the increased seepage flux in seep locations could not be explained on the basis of the physical properties analyses conducted during this study.

In the seeps with focused discharge, natural attenuation (mainly anaerobic biodegradation) is affected by the increase in vertical seepage flux and resulting decrease in residence time in the wetland sediments. This results in less reducing conditions, an accumulation of chlorinated volatile organic compounds in shallow pore water, and in some cases, a change in the microbial community. Results of surface-waterquality sampling alone were not indicative of seep location or pore water concentration at the ground-water/surface-water interface, indicating that thermal and passive pore water sampling methods such as those used in this study are more appropriate for identifying areas of preferential discharge to surface water than surface-water sampling alone.

Despite the large areal footprint of overlapping volatile organic compound plumes in the Canal Creek aquifer beneath the West Branch Canal Creek wetlands and stream bottom, natural attenuation appears to efficiently reduce concentrations in the organic-rich wetland sediments in most of the wetland area. The identification of discrete areas of preferential discharge of contaminated pore water to surface water provides an alternative approach to remediation of source zones contained within or nearby the sensitive wetland ecosystem by reducing the overall treatment area to 1 percent of the total discharge area. Targeted, passive treatment at the ground-water/surface-water interface could provide treatment immediately prior to discharge to surface water. Some limited, passive remediation in seep areas coupled with natural attenuation could preserve the wetland ecosystem and reduce the loading of volatile organic compounds to West Branch Canal Creek.