USGS Science⌕ Search
Home/Archive/USGS

USGS · 70016180

A comparison of two- and three-dimensional stochastic models of regional solute movement

A.M. Shapiro·V.D. Cvetkovic
View original source ↗Export metadata ↓

Abstract

Recent models of solute movement in porous media that are based on a stochastic description of the porous medium properties have been dedicated primarily to a three-dimensional interpretation of solute movement. In many practical problems, however, it is more convenient and consistent with measuring techniques to consider flow and solute transport as an areal, two-dimensional phenomenon. The physics of solute movement, however, is dependent on the three-dimensional heterogeneity in the formation. A comparison of two- and three-dimensional stochastic interpretations of solute movement in a porous medium having a statistically isotropic hydraulic conductivity field is investigated. To provide an equitable comparison between the two- and three-dimensional analyses, the stochastic properties of the transmissivity are defined in terms of the stochastic properties of the hydraulic conductivity. The variance of the transmissivity is shown to be significantly reduced in comparison to that of the hydraulic conductivity, and the transmissivity is spatially correlated over larger distances. These factors influence the two-dimensional interpretations of solute movement by underestimating the longitudinal and transverse growth of the solute plume in comparison to its description as a three-dimensional phenomenon. Although this analysis is based on small perturbation approximations and the special case of a statistically isotropic hydraulic conductivity field, it casts doubt on the use of a stochastic interpretation of the transmissivity in describing regional scale movement. However, by assuming the transmissivity to be the vertical integration of the hydraulic conductivity field at a given position, the stochastic properties of the hydraulic conductivity can be estimated from the stochastic properties of the transmissivity and applied to obtain a more accurate interpretation of solute movement. ?? 1990 Kluwer Academic Publishers.

Explore related subjects

Transport in Porous MediaArticle

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

A.M. Shapiro, V.D. Cvetkovic. 1990. A comparison of two- and three-dimensional stochastic models of regional solute movement. https://doi.org/10.1007/bf00714626

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2013

Determination of water retention in stratified porous materials

Predicted and measured water-retention values, θ ( ψ ), were compared for repacked, stratified core samples consisting of either a sand with a stone-bearing layer or a sand with a clay loam layer in various spatial orientations. Stratified core samples were packed in submersible pressure outflow cells, then water-retention measurements were performed between matric potentials, ψ , of 0 to -100 kPa. Predictions of θ ( ψ ) were based on a simple volume-averaging model using estimates of the relative fraction and θ ( ψ ) values of each textural component within a stratified sample. In general, predicted θ ( ψ ) curves resembled measured curves well, except at higher saturations in a sample consisting of a clay loam layer over a sand layer. In this case, the model averaged the air-entry of both materials, while the air-entry of the sample was controlled by the clay loam in contact with the cell's air-pressure inlet. In situ, avenues for air-entry generally exist around clay layers, so that the model should adequately predict air-entry for stratified formations regardless of spatial orientation of fine versus coarse layers. Agreement between measured and predicted volumetric water contents, θ , was variable though encouraging, with mean differences between measured and predicted θ values in the range of 10%. Differences in θ of this magnitude are expected due to variability in pore structure between samples, and do not indicate inherent problems with the volume averaging model. This suggets that explicit modeling of stratified formations through detailed characterization of the stratigraphy has the potential of yielding accurate θ ( ψ ) values. However, hydraulic-equilibration times were distinctly different for each variation in spatial orientation of textural layering, indicating that transient behavior during drainage in stratified formations is highly sensitive to the stratigraphic sequence of textural components, as well as the volume fraction of each textural component in a formation. This indicates that prolonged residence times of water, nutrients, and pollutants are likely within finer-textured layers, when ψ conditions have resulted in drainage of underlying coarser-textured strata.

Transport in Porous Media↗
USGS2010

Applying dispersive changes to Lagrangian particles in groundwater transport models

Method-of-characteristics groundwater transport models require that changes in concentrations computed within an Eulerian framework to account for dispersion be transferred to moving particles used to simulate advective transport. A new algorithm was developed to accomplish this transfer between nodal values and advecting particles more precisely and realistically compared to currently used methods. The new method scales the changes and adjustments of particle concentrations relative to limiting bounds of concentration values determined from the population of adjacent nodal values. The method precludes unrealistic undershoot or overshoot for concentrations of individual particles. In the new method, if dispersion causes cell concentrations to decrease during a time step, those particles in the cell having the highest concentration will decrease the most, and those with the lowest concentration will decrease the least. The converse is true if dispersion is causing concentrations to increase. Furthermore, if the initial concentration on a particle is outside the range of the adjacent nodal values, it will automatically be adjusted in the direction of the acceptable range of values. The new method is inherently mass conservative.

Transport in Porous Media↗
USGS2008

Transport of viruses through saturated and unsaturated columns packed with sand

Laboratory-scale virus transport experiments were conducted in columns packed with sand under saturated and unsaturated conditions. The viruses employed were the male-specific RNA coliphage, MS2, and the Salmonella typhimurium phage, PRD1. The mathematical model developed by Sim and Chrysikopoulos (Water Resour Res 36:173–179, 2000) that accounts for processes responsible for removal of viruses during vertical transport in one-dimensional, unsaturated porous media was used to fit the data collected from the laboratory experiments. The liquid to liquid–solid and liquid to air–liquid interface mass transfer rate coefficients were shown to increase for both bacteriophage as saturation levels were reduced. The experimental results indicate that even for unfavorable attachment conditions within a sand column (e.g., phosphate-buffered saline solution; pH = 7.5; ionic strength = 2 mM), saturation levels can affect virus transport through porous media.

Transport in Porous Media↗