David W. Ostendorf

Biodegradation of hydrocarbon vapors in the unsaturated zone

The time-averaged concentration of hydrocarbon and oxygen vapors were measured in the unsaturated zone above the residually contaminated capillary fringe at the U.S. Coast Guard Air Station in Traverse City, Michigan. Total hydrocarbon and oxygen vapor concentrations were observed over a 13-month period. Supplementary grain size, porosity, and moisture content data support the assumption of a uniform, homogeneous site geology which, in view of the planar hydrocarbon source term, abundant oxygen, and sparse data base, is suitable for simple analytical modeling. In the assumed absence of advection, leaching, and transience, the analysis is a straight-forward balance of gaseous diffusion and biological degradation coupled stoichiometrically in the two reacting constituents. Volatilization is shown to be a significant transport mechanism for hydrocarbons at Traverse City, and biodegradation prevents the escape of appreciable contamination to the atmosphere for most locations at this site. Little oxygen is expected to reach the water table because of the aerobic biodegradation process in the unsaturated zone.

A model for stripping multicomponent vapor from unsaturated soil with free and trapped residual nonaqueous phase liquid

We present a model for the multicomponent vapor transport due to air venting in an unsaturated zone in the presence of free and trapped phases of residual nonaqueous phase liquid (NAPL). On the microscale the soil particles are assumed to form spherical aggregates with micropores filled with immobile water, trapped phases of NAPL and air. The interaggregate space is occupied with mobile air, and a thin film of free NAPL adheres on the aggregate surface. While the free NAPL can readily be in equilibrium with macropore vapor, the mass transfer from immobile phases in aggregates is rate-limited by aqueous diffusion. This model enables us to predict the vapor concentrations of various chemical species and the free NAPL saturation over the macroscale, based on the detailed understanding of the aqueous concentrations of the species and the trapped NAPL saturation within the aggregates. The model is compared favorably with some experimental data of sparging multicomponent vapor out of an intact core taken from a contaminated site. The distinctive features of multicomponent transport, clearly exhibited by the data, are further examined in the simulations of a hypothetical case of three-aromatic vapor transport under a radial flow field. It is found that while the vapor concentration of the most volatile component drops monotonically with time, those of the less volatile may rise as their mole fractions in the NAPL increase. The vapor concentration of a heavy component may have a local maximum at the evaporation front of the free NAPL. In the case of radial flow the free NAPL has two receding evaporation fronts. Condensation of the heavy component downstream of the far front causes a temporary increase of its total concentration there. With trapped NAPL and soil aggregation the macroscale transport is retarded, and the effluent concentrations end up in noticeable tailing.

Soil gas transport above a jet fuel/solvent spill at Plattsburgh Air Force Base

We calibrate a stoichiometrically coupled soil gas diffusion model with spatially resolved observations of oxygen, carbon dioxide, total hydrocarbon, and trichloroethylene vapor concentrations in the unsaturated zone above a weathered jet fuel/solvent spill at Plattsburgh Air Force Base in upstate New York. The calibration suggests that aerobic microorganisms in the capillary fringe degrade jet fuel vapor at a steady rate of 9.5 μg hydrocarbons (m−2 s−1). The solvent does not degrade in the fringe, however, and the model and data estimate a steady evaporation rate of 1.2×10−2 μg TCE (m−2 s−1). Barometric pumping slightly alters the steady concentration profile at Plattsburgh, although the transient advective flux is the same order of magnitude as the steady diffusive flux. We derive a simple perturbation theory for the second-order transient concentration corrections and include it in the calibration. The perturbation theory is valid at Plattsburgh because the soil is uniform and permeable with a relatively deep capillary fringe.

Cation exchange in a glacial till drumlin at a road salt storage facility

We use laboratory and field data to calibrate existing geochemical and transport models of cation exchange induced by contamination of an unconfined aquifer at a road salt storage facility built upon a glacial till drumlin in eastern Massachusetts. A Gaines and Thomas selectivity coefficient K models the equilibrium sodium and divalent cation distribution in the groundwater and solid matrix, while an existing method of characteristics model describes the advective transport of total dissolved cations and sorbed sodium. Laboratory isotherms of split spoon soil samples from the drumlin calibrate K with an average value of 0.0048 (L/g)(1/2) for a measured cation exchange capacity of 0.057 meq/g dry soil. Ten years of monitoring well data document groundwater flow and the advection of conservative chloride due to outdoor storage and handling of road salt at the site. The monitoring well cation data and retarded transport model offer an independent K calibration of 0.0040 to 0.0047 (L/g)(1/2): the consistency of the field and laboratory selectivity coefficient calibrations endorse this application of the Gaines and Thomas and method of characteristics models. The advancing deicing agent plume releases divalent cations from the till into the groundwater, so that monitoring well samples do not reflect the chemical composition of the road salt. In this regard, dissolved divalent cation milliequivalent concentrations are as high as 80% of the total dissolved cationic concentrations in the salt contaminated monitoring well samples, far greater than their 2.5% level in the road salt stored at the site. Cation exchange can thus obscure attempts to hindcast stored road salt sodium water table concentration from monitoring well sample stoichiometry, or to predict sodium impacts on groundwater or receiving stream quality downgradient of the well.

Slug Tests in the Presence of Background Head Trends

We extend Bouwer and Rice (1976) slug test theory to incorporate background head trends that may be important in incompressible material of low permeability k. The extension, which features a convolution integral of the background head, is closed form for linear trends. A sensitivity study suggests that a rising background head can diminish the head changes associated with a slug-out test and underestimate k if it is ignored, as does falling background trend with a slug-in test. A falling background head can reinforce slug-in test head change and, if ignored, can overestimate k, as does a rising background head with a slug-out test. The simple extension is verified by field tests in glacial till and stratified drift deposits in eastern Massachusetts.

Aerobic Biodegradation of Petroleum-Contaminated Soil: Simulations from Soil Microcosms

The capacity of natural bacteria to aerobically degrade hydrocarbon vapors was measured and modeled to assess the potential of bioventing to reduce exhaust vapor treatment requirements at a petroleum spill site. Five sets of aerobic soil microcosms from the vadose zone of a Massachusetts Highway Department contaminated right-of-way were dosed with different initial petroleum vapor standard concentrations, then monitored by gas chromatographic analysis over a 55-day period. The five sets yielded an average maximum reaction rate of 20 μg/m3 (soil gas)-sec, which compared favorably with studies of light hydrocarbon vapor degradation in sandy soils from other sites. The calibrated rate was incorporated into a steady-state bioventing model that simulated the evaporation of 34 000 L of petroleum over a 170-year natural release period and an 8-year accelerated release period for 10-day residence time. Aerobic degradation for a 10-day residence time reduced exhaust vapor concentrations by over 100 percent for nat...

Chloride Dispersion across Silt Deposits in a Glaciated Bedrock River Valley

Soil and groundwater from the Neponset River floodplain deposit that receive high concentrations of deicing agents from nearby highways were investigated. The silty sand floodplain is separated by a silty aquitard from the underlying aquifer that serves as a public water supply. We made a transport-based assessment of the capacity of the aquitard to protect the underlying aquifer. One hundred seventeen soil samples and 469 groundwater samples collected during a period of 4 yr from boreholes and 10 wells grouped in two well clusters were analyzed for dissolved Cl concentration. The soil characterization and groundwater monitoring results agreed, showing a very slow change in subsurface Cl contamination with time. These data also calibrated a vertical one-dimensional advective-dispersive transport model across the deposits. Advective transport dominated only in the top 3.37 m of the floodplain deposit, with dispersion being the main transport mechanism below this depth. Due to the silty nature of the aquitard, dispersion rather than diffusion was the main transport mechanism into the floodplain-aquitard system. Soil and groundwater quality data confirmed a Cl concentration at the floodplain surface near the highway runoff drainage outlets of 2450 mg L. The model estimated a vertical dispersivity at the site of 8 mm and a vertical hydrodynamic dispersion coefficient of 3.71 × 10 m s. These data confirmed the aquitards capacity to contain deicing agents, protecting the underlying aquifer from contamination.

Case Study of Steady Oxygen Concentration Gradients in a Groundwater Plume from a Highway Infiltration Basin

We measure and model the steady transport of specific conductivity and dissolved oxygen through a groundwater plume from a highway infiltration basin in southeastern Massachusetts. Specific conductivity is treated as a conservative surrogate for runoff contamination, and the data calibrate a 0.27-m vertical dispersivity α of the aquifer and the bottom streamline elevation of the plume, which falls to an 8-m depth below the water table. The dissolved oxygen degrades as a first order reactant in the plume to levels below 1 mg/L, with a decay constant λ of 0.12  day−1 . The latter may be attributed in part to the historical use of an alternative de-icing agent calcium magnesium acetate on the highway, since acetate is a readily biodegradable substrate for microorganisms. The calibrated kinetics suggest that plume microbes and geochemistry degrade oxygen over two orders of magnitude faster than their ambient groundwater counterparts, which impose a linear decrease of dissolved oxygen concentration below the p...

CMA Induced Organic Enrichment and Oxygen Depletion from Highway Runoff

We measured dissolved oxygen, acetate, bicarbonate, and iron concentrations in groundwater near a state highway in southeastern Massachusetts. A closed drainage system sedimentation basin receives runoff containing calcium magnesium acetate (CMA), an alternative deicing agent. The data suggest that CMA impairs groundwater quality by increasing organics and iron in an anoxic plume downgradient of the basin. Drift plowed beyond the closed system is degraded by vadose zone aerobes in the highway shoulder upgradient of the basin. Open drainage systems with biologically active vadose zones reduce the oxygen demand before the runoff reaches groundwater.

Near Surface Soil Vapor Clusters for Monitoring Emissions of Volatile Organic Compounds from Soils

ABSTRACT The overall objective of this research was to develop and test a method of determining emission rates of volatile organic compounds (VOCs) and other gases from soil surfaces. Soil vapor clusters (SVCs) were designed as a low dead volume, robust sampling system to obtain vertically resolved profiles of soil gas contaminant concentrations in the near surface zone. The concentration profiles, when combined with a mathematical model of porous media mass transport, were used to calculate the contaminant flux from the soil surface. Initial experiments were conducted using a mesoscale soil remediation system under a range of experimental conditions. Helium was used as a tracer and trichloroethene was used as a model VOC. Flux estimations using the SVCs were within 25% of independent surface flux estimates and were comparable to measurements made using a surface isolation flux chamber (SIFC). In addition, method detection limits for the SVC were an order of magnitude lower than detection limits with the SIFC. Field trials, conducted with the SVCs at a bioventing site, indicated that the SVC method could be easily used in the field to estimate fugitive VOC emission rates. Major advantages of the SVC method were its low detection limits, lack of required auxiliary equipment, and ability to obtain realtime estimates of fugitive VOC emission rates.