A.J.A. Unger

Mechanisms Controlling Vacuum Extraction Coupled With Air Sparging for Remediation of Heterogeneous Formations Contaminated by Dense Nonaqueous Phase Liquids

The numerical model CompFlow is used to study the mechanisms controlling vacuum extraction, coupled with air sparging, as a means for remediation of heterogeneous formations contaminated with dense nonaqueous phase liquids (DNAPLs). Two dominant mechanisms are demonstrated to control this remediation technology. First, at early times, the gas phase directly contacts the DNAPL, particularly in the unsaturated zone, causing relatively rapid transfer of contaminant from the nonaqueous phase to the gas phase and subsequent removal by the vacuum extractor. Second, at later times, remediation is controlled by the transfer of contaminant from the nonaqueous phase to the aqueous phase below the water table. During this time the vacuum extractor pumps both liquid and vaporized water in the aqueous and gas phases. This causes the contaminant that is dissolved in the aqueous phase to migrate vertically upward across the permeability layers toward the vacuum extractor where it is removed. This intermediate to late time removal mechanism is shown to be controlled by contaminant dissolution, which is a slower transfer process than the direct DNAPL vaporization that occurs at early time. Our analysis indicates that as long as both air and water are actively flushed through the DNAPL zone, both early-time vaporization and intermediate- to late-time dissolution are effective mechanisms leading to the removal of the DNAPL. We show that it may be possible to design the remedial system so as to reduce its performance sensitivity to geologic heterogeneity. A lack of sensitivity of a remedial design to heterogeneity is highly desirable because a robust design implies that the degree of site characterization required for reasonable success will be less than that needed for a less robust scheme.

Variable spatial and temporal weighting schemes for use in multi-phase compositional problems

Abstract Variable spatial and temporal weighting of the advective contaminant mole fraction term is explored as a means of reducing numerical dispersion of contaminant plumes in a multi-phase compositional simulator. The spatial schemes considered are upstream, central and a non-linear flux limiter, while fully-implicit and Crank-Nicolson time weighting are examined. The performance of each weighting scheme, in terms of stability of the Newton iteration and computational cost, is assessed for simplified problems designed to be representative of various aspects of more complex subsurface remediation problems. Results indicate that for problems with an homogeneous permeability field, the non-linear flux limited along with fully-implicit weighting gives superior performance to any other combination of spatial and temporal weighting schemes. For heterogeneous permeability fields, the macrodispersion imparted by heterogeneity dominates numerical dispersion so that smearing of contaminant mole fraction fronts does not appear to be a serious problem.

Application of system dynamics for developing financially self-sustaining management policies for water and wastewater systems

Recently enacted regulations in Canada and elsewhere require water utilities to be financially self-sustaining over the long-term. This implies full cost recovery for providing water and wastewater services to users. This study proposes a new approach to help water utilities plan to meet the requirements of the new regulations. A causal loop diagram is developed for a financially self-sustaining water utility which frames water and wastewater network management as a complex system with multiple interconnections and feedback loops. The novel System Dynamics approach is used to develop a demonstration model for water and wastewater network management. This is the first known application of System Dynamics to water and wastewater network management. The network simulated is that of a typical Canadian water utility that has under invested in maintenance. Model results show that with no proactive rehabilitation strategy the utility will need to substantially increase its user fees to achieve financial sustainability. This increase is further exacerbated when price elasticity of water demand is considered. When the utility pursues proactive rehabilitation, financial sustainability is achieved with lower user fees. Having demonstrated the significance of feedback loops for financial management of water and wastewater networks, the paper makes the case for a more complete utility model that considers the complexity of the system by incorporating all feedback loops.

Coupled Vadose Zone and Atmospheric Surface-Layer Transport of Carbon Dioxide from Geologic Carbon Sequestration Sites

Geologic CO 2 sequestration is being considered as a way to offset fossil fuel–related CO 2 emissions to reduce the rate of increase of atmospheric CO 2 concentrations. The accumulation of vast quantities of injected CO 2 in geologic sequestration sites may entail health and environmental risks from potential leakage and seepage of CO 2 into the near-surface environment. We are developing and applying a coupled subsurface and atmospheric surface-layer modeling capability built within the framework of the integral finite difference reservoir simulator TOUGH2. The overall purpose of the modeling studies is to predict CO 2 concentration distributions under a variety of seepage scenarios and geologic, hydrologic, and atmospheric conditions. These concentration distributions will provide the basis for determining aboveground and near-surface instrumentation needs for CO 2 sequestration monitoring and verification, as well as for assessing health, safety, and environmental risks. A key feature of CO 2 is its large density (ρ = 1.8 kg m −3 ) relative to air (ρ = 1.2 kg m −3 ), a property that may allow small leaks to cause concentrations in air above the occupational exposure limit of 4% in low-lying and enclosed areas such as valleys and basements where dilution rates are low. The approach we take to coupled modeling involves development of T2CA, a TOUGH2 module for modeling the multicomponent transport of water, brine, CO 2 , gas tracer, and air in the subsurface. For the atmospheric surface-layer advection and dispersion, we use a logarithmic vertical velocity profile to specify constant time-averaged ambient winds, and atmospheric dispersion approaches to model mixing due to eddies and turbulence. Initial simulations with the coupled model suggest that atmospheric dispersion quickly dilutes diffuse CO 2 seepage fluxes to negligible concentrations, and that rainfall infiltration can cause CO 2 to return to the subsurface as a dissolved component in infiltrating rainwater.

A Noniterative Technique for the Direct Implementation of Well Bore Boundary Conditions in Three-Dimensional Heterogeneous Formations

A noniterative algorithm for handling prescribed well bore boundary conditions while pumping or injecting fluid in a three-dimensional heterogeneous aquifer is described. The algorithm is formulated by superimposing conductive one-dimensional line elements representing the well screen onto the three-dimensional matrix elements epresenting the aquifer. Storage in the well casing is also naturally accommodated by the superposition of the line elements. The numerical algorithm is verified by comparison with results obtained from the solution of Papadopulos and Cooper (1967). A large-scale example problem involving groundwater extraction from a partially penetrating pumping well located in a highly heterogeneous confined aquifer is presented to demonstrate the utility of the approach.

Influence of alternative dissolution models and subsurface heterogeneity on DNAPL disappearance times

Abstract The numerical model CompFlowC++ is used to examine the relative importance of different formulations of a forward dissolution rate model, within the range of parameters measured from laboratory and field studies, for predicting the mean and uncertainty in dense nonaqueous phase liquids (DNAPL) dissolution times in heterogeneous sandy aquifers. We also investigate whether uncertainty in the DNAPL distribution due to imprecise knowledge of the form of the nonaqueous phase relative permeability curves and the variance of the heterogeneous permeability field are more important factors affecting the dissolution of a DNAPL source zone than the formulation and parameter values of the forward dissolution rate. Results indicate that the mean dissolution time is particularly sensitive to the exponent associated with the nonaqueous phase saturation term, in the kinetic formulation of the forward dissolution rate expression, while the uncertainty is sensitive to the Reynolds number associated with aqueous phase flow. Uncertainty in the DNAPL distribution due to imperfect knowledge of parameters controlling multiphase flow or the geostatistical nature of the aquifer exert less influence on the dissolution behaviour of the DNAPL source zone than the formulation of, or parameters associated with, the forward dissolution rate.

Simulating the evolution of an ethanol and gasoline source zone within the capillary fringe.

Blending of ethanol into gasoline as a fuel oxygenate has created the scenario where inadvertent releases of E95 into soil previously contaminated by gasoline may remobilize these pre-existing NAPLs and lead to higher dissolved hydrocarbon (BTEX) concentrations in groundwater. We contribute to the development of a risk-based corrective action framework addressing this issue by conducting two laboratory experiments involving the release of ethanol into a gasoline source zone established in the capillary fringe. We then develop and apply the numerical model CompFlow Bio to replicate three specific experimental observations: (1) depression of the capillary fringe by the addition of the gasoline fuel mixture due to a reduction in the surface tension between the gas and liquid phases, (2) further depression of the capillary fringe by the addition of ethanol, and (3) remobilization of the gasoline fuel mixture LNAPL source zone due to the cosolvent behaviour of ethanol in the presence of an aqueous phase, as well as a reduction in the interfacial tension between the aqueous/non-aqueous phases due to ethanol. While the simulated collapse of the capillary fringe was not as extensive as that which was observed, the simulated and observed remobilized non-aqueous phase distributions were in agreement following ethanol injection. Specifically, injection of ethanol caused the non-aqueous phase to advect downwards toward the water table as the capillary fringe continued to collapse, finally collecting on top of the water table in a significantly reduced area exhibiting higher saturations than observed prior to ethanol injection. Surprisingly, the simulated ethanol and gasoline aqueous phase plumes were uniform despite the redistribution of the source zone. Dissolution of gasoline into the aqueous phase was dramatically increased due to the cosolvency effect of ethanol on the non-aqueous phase source zone. We advocate further experimental studies focusing on eliminating data gaps identified here, as well as field-scale experiments to address issues associated with ethanol-BTEX biodegradation and sorption within the development of a risk-based corrective action framework.

Simulating the fate and transport of TCE from groundwater to indoor air.

This work provides an exploratory analysis on the relative importance of various factors controlling the fate and transport of volatile organic contaminants (in this case, TCE) from a DNAPL source zone located below the water table and into the indoor air. The analysis is conducted using the multi-phase compositional model CompFlow Bio, with the base scenario problem geometry reminiscent of a field experiment conducted by Rivett [Rivett, M.O., (1995), Soil-gas signatures from volatile chlorinated solvents: Borden field experiments. Groundwater, 33(1), 84-98.] at the Borden aquifer where groundwater was observed to transport a contaminant plume a substantial distance without vertical mass transport of the contaminant across the capillary fringe and into the vadose zone. Results for the base scenario model indicate that the structure of the permeability field was largely responsible for deflecting the groundwater plume upward towards the capillary fringe, permitting aqueous phase diffusion to transport the TCE into the vadose zone. Alternative permeability realizations, generated as part of a Monte Carlo simulation process, at times deflected the groundwater plume downwards causing the extended thickness of the saturated zone to insulate the vadose zone from exposure to the TCE by upward diffusive transport. Comparison of attenuation coefficients calculated using the CompFlow Bio and Johnson and Ettinger [Johnson, P.C. and Ettinger, R.A., (1991), Heuristic model for predicting the intrusion rate of contaminant vapors into buildings. Environmental Science and Technology, 25, 1445-1452.] heuristic model exhibited fortuitous agreement for the base scenario problem geometry, with this agreement diverging for the alternative permeability realizations as well as when parameters such as the foundation slab fracture aperture, the indoor air pressure drop, the capillary fringe thickness, and the infiltration rate were varied over typical ranges.

Numerical study of the hydromechanical behavior of two rough fracture surfaces in contact

Flow in fractures is traditionally modeled by characterizing the aperture distribution with some deterministic function or set of stochastic parameters. Other models generate the aperture distribution by the closure of two stochastic surfaces. The objective of this research is to develop a model where the aperture distribution is determined during the closure of two random elastic surfaces with complete hydromechanical interaction. Because stress and strain conditions required to generate a given aperture distribution are calculated during closure, the model is used to couple the mechanical and hydraulic characteristics of the fracture. Stochastic realizations of clay fracture surfaces are generated by measuring one-dimensional profiles of a fracture surface. Next, the spectral representation of the profile is related to the fractal dimension of the fracture. Using the fractal dimension determined from one-dimensional clay profiles, an equivalent two-dimensional fractal surface is generated. Conceptually, each surface consists of linear elastic rectangular asperities resting upon a linear elastic half-space. During closure, asperities that come into contact deform and punch into the half space creating mechanical interaction between all the asperities on the grid. Once we determine the aperture distribution at an applied stress level, a hydraulic gradient is applied across the fracture and fluid flow is determined. Nodal pressures created by flow deform the aperture distribution coupling hydraulic to mechanical behavior. Stress versus relative closure results indicate that stress increases nonlinearly with relative closure. Fluid pressures in the aperture distribution exert a significant influence on the mechanical characteristics of a fracture. Fluid discharge through the fracture decreases exponentially with an increase in relative closure. Flow calculated in the rough walled aperture distribution deviates increasingly from the parallel plate model with the geometric mean aperture as the percent contact area increases. The deviation results from an increase in tortuosity and channelling of the flow field in the aperture distribution. We can use this model to develop stress and flow versus relative closure constitutive relationships for a single fracture as a function of fracture surface geometry.

Development of a system dynamics model for financially sustainable management of municipal watermain networks.

This paper develops causal loop diagrams and a system dynamics model for financially sustainable management of urban water distribution networks. The developed causal loop diagrams are a novel contribution in that it illustrates the unique characteristics and feedback loops for financially self-sustaining water distribution networks. The system dynamics model is a mathematical realization of the developed interactions among system variables over time and is comprised of three sectors namely watermains network, consumer, and finance. This is the first known development of a water distribution network system dynamics model. The watermains network sector accounts for the unique characteristics of watermain pipes such as service life, deterioration progression, pipe breaks, and water leakage. The finance sector allows for cash reserving by the utility in addition to the pay-as-you-go and borrowing strategies. The consumer sector includes controls to model water fee growth as a function of service performance and a households financial burden due to water fees. A series of policy levers are provided that allow the impact of various financing strategies to be evaluated in terms of financial sustainability and household affordability. The model also allows for examination of the impact of different management strategies on the water fee in terms of consistency and stability over time. The paper concludes with a discussion on how the developed system dynamics water model can be used by water utilities to achieve a variety of utility short and long-term objectives and to establish realistic and defensible water utility policies. It also discusses how the model can be used by regulatory bodies, government agencies, the financial industry, and researchers.