VOLATILE ORGANIC CONTAMINANT TRANSPORT AND DEGRADATION IN A WETLAND SEEP AREA: TRACER AND DIRECT INJECTION TEST RESULTS

Seep areas that have groundwater-flow rates 10 to 100 times faster than in non-seep areas have been identified at an army base in Maryland where chlorinated volatile organic contaminants (VOCs) discharge from a sand aquifer through 4-m-thick wetland sediments. With the purpose of evaluating bioremediation methods, laboratory experiments and field sampling, including a conservative tracer test and a preliminary direct injection test of donor biostimulation, have been used to characterize groundwater flow, contaminant transport, and biogeochemical processes controlling VOC degradation at one seep.

Contaminant concentrations (about 30 mg/L total VOCs) and groundwater discharge in the seep were consistent during tidal cycles and seasonally, although vertical VOC distribution across the seep area was variable. Bromide and rhodamine dye were injected 3.6 m below land surface and monitored vertically and within a 1.2-m radius. Dye was not observed at land surface, indicating the absence of high velocity preferential flow paths. Bromide dispersed upward relatively uniformly, although preferential horizontal flow to the south/southeast was detected. The seepage velocities were 0.05 to 0.08 m/day based on bromide breakthrough, consistent with 0.06 m/day calculated from hydraulic conductivity and head gradient. Using porosity of 40 to 60 percent, 68 to 100 percent of the bromide mass was accounted for within the test area. The tracer results suggest that diffuse flow is the main mechanism of groundwater movement through the seep, verifying suitability for the direct injection test. Flow-through column experiments with seep sediment showed that degradation could be enhanced by donor addition, but degradation rates were highly dependent on donor concentrations. Direct injection results confirmed the enhancement of degradation rates with donor addition. Greatest enhancement occurred in the south/southeast sites where bromide tracer showed a preferential flow, which may have more effectively distributed the donor. These sites also were the only locations that showed pre-existing methanogenic conditions and may have supported a dechlorinating microbial community that could be stimulated more rapidly than in iron-reducing locations.