1999 Conference Proceedings
Simulating Air Entrapment, Mobilization and Oxygen Dissolution for Fluctuating Water Table Conditions
M. D. White, M. D. Williams, M. Oostrom
Pacific Northwest National Laboratory, Richland, WA
Proceedings of Nineteenth Annual American Geophysical Union Hydrology Days, Fort Collins, CO
Abstract
In-situ redox manipulation is an innovative permeable barrier technology for the remediation of redox-sensitive groundwater contaminants (e.g., chromate, chlorinated solvents, uranium, and technetium). The technology involves the reduction of naturally occurring iron in aquifer sediments through the injection of an aqueous soluble reducing reagent. Reduced iron in the aquifer then affects the removal of redox-sensitive contaminants migrating through the permeable barrier region. In addition to its capabilities for removing contaminants from the groundwater the permeable barrier reacts with other oxidizing species in the groundwater; notably dissolved oxygen. This study is concerned with determining the fate of an anoxic groundwater plume in an unconfined aquifer with a fluctuating water table. In particular, the study is directed at predicting dissolved oxygen concentrations in the groundwater near the Columbia River for the purposes of assessing the potential impact of the reducing barrier on aquatic organisms. Attenuation processes for anoxic plumes include oxygen diffusion from the vadose zone and lower aquitard, oxygenated recharge water, and air entrapment and dissolution within the fluctuating water table elevations. A model for numerically simulating the reoxygenation of an anoxic plume in an unconfined aquifer is presented. The model considers oxygen migration and dissolution for variably saturated conditions with entrapment, mobilization and dissolution of entrapped air for fluctuating water table conditions. Model simulations are compared against an intermediate-scale flow-cell experiment with a fluctuating water table as a validation exercise. This exercise was designed to test capabilities for accurately modeling air entrapment and oxygen dissolution processes. Results from the intermediate-scale experiments showed that entrapped air within the fluctuating water table elevations cause significant reoxygenation and contribute to increases in dissolved oxygen concentration deep in the flow system. Simulation results in terms of entrapped air saturations and dissolved oxygen concentrations showed reasonable agreement with the experimental observations, demonstrating the applicability of the numerical model. In a field-scale application of the numerical model, concentrations of dissolved oxygen downgradient from a permeable redox barrier were predicted using two conceptual models (i.e., with and without a fluctuating water table) for a subsurface site on the Hanford Reservation near the Columbia River. Results from these simulations were used to quantify the relative importance of the various reoxygenation mechanisms and to provide a conservative estimate of dissolved oxygen concentrations downgradient from the demonstration site.
Conditional Simulation and Upscaling of Soil Hydraulic Properties
M. L. Rockhold1, C. J. Murray2, M.J. Fayer2
1Department of Bioresource Engineering, Oregon State University, Corvallis, OR
2Pacific Northwest National Laboratory, Richland, WA
Proceedings of the International Workshop on Characterization and Measurement of the Hydraulic Properties of Unsaturated Media, Riverside, CA
Abstract
One of the difficulties associated with modeling flow and transport processes at the field scale is determining effective model parameters. Soil hydraulic properties are often characterized by measurements on a few, relatively small core samples. The size of grid blocks required for numerical modeling of field scale problems may be orders of magnitude larger than the size of these core samples due to computational constraints. In addition, the spatial variability of soil hydraulic properties is usually not characterized adequately by the few sparse measurements that are typically available. A method is presented that uses both hard and soft data to parameterize numerical flow and transport models. The method utilizes geostatistical indicator techniques for spatial interpolation of field-measured water contents and porosities, and a conditional simulation method based on similar media scaling for estimating hydraulic properties from a set of scale-mean parameters and the initial water content and porosity distributions. An upscaling algorithm is used for determining effective model parameters. The method is used to estimate parameters for three-dimensional simulations of an injection experiment conducted in southeastern Washington State. During the experiment, water and radioactive tracers were injected into heterogenous unsaturated sediments. The water plume was monitored using a neutron probe in 32 wells arranged radially around the injection well. Comparisons of the modeling results with field data were good, indicating that the conditional simulation and upscaling method provides an efficient, systematic means for estimating effective soil hydraulic properties for field-scale model applications.