Long-Tube Testing to Evaluate Settling for Enhanced Sedimentation Using Engineering Controls

摘要

Background/Objectives. Enhanced natural sedimentation (ENS) was considered as aremedial strategy for a Superfund site in the floodplain of a major river, where mercury isan ecological constituent of concern (COC). Long-tube testing was employed to estimatesettling rates for the enhanced sedimentation pilot study, in which a berm and gate structure(engineering controls) was constructed around an oxbow lake to trap flood watersoriginating from the river and containing suspended solids. The deposition of clean sedimentduring flooding, enhanced by the berm and gate structure, could then potentiallycap the sediments over time. Long-tube testing was performed on water collected fromthe river to estimate settling rates and physically and chemically characterize the sedimentprior to construction of the berm and gate structure.Approach/Activities. Approximately 250 gallons of water were collected from the riverduring a flood event for bench-scale testing. Water samples were classified prior to testingto determine that variation among buckets was within acceptable limits. The longtube test procedure was a modified form of that presented in the U.S. Army Corp of EngineersManual EM1110-2-5027 so that the tube length was representative of the waterdepth throughout most of the lake. The testing consisted of filling an 8.5-foot columnwith sample water, agitating the sample in the tube, allowing sediments to settle, and collectingwater samples from varying column heights over time. The water samplescollected were analyzed for total suspended solids (TSS). At the end of the test, waterwas decanted from the tube, and the sediments were collected and analyzed for mercury,percent moisture, density, and grain size. Each test was performed with 4 replicates (4tubes). The results of tests showed that greater than 50 percent of TSS had settled 6 hoursafter ceasing agitation, and 85 percent or more of the sediments had settled after 48hours. The sediments were non-detect (<0.101 mg/kg) for mercury, and the concentrationof sediment in the river water samples was 310 mg/L.Results/Lessons Learned. The settling times estimated from the long-tube test were usedto optimize operation of a berm and gate structure for implementation of ENS. When aflood event began and the berm was overtopped, the gate was closed to retain floodwaters.The gate remained closed for at least 48 hours, based on the long-tube test results, toallow clean sediment to deposit over the site. After 48 hours, the gate was opened incrementallyto decant the floodwater into the river with minimal disturbance to the newlydeposited sediments. Data collected to date indicate that ENS appears to have reducedmercury concentrations in surficial sediment in the vicinity of the gate.