Modeling dispersion in three-dimensional heterogeneous fractured media at Yucca Mountain

Highly resolved numerical simulations are conducted to evaluate the longitudinal and transverse dispersivities proposed for use in the larger-scale Yucca Mountain saturated zone (SZ) site-scale model. Two different stochastic continuum models (SCM) that define the spatial variability of permeability...

Full description

Saved in:
Bibliographic Details
Published in:Journal of contaminant hydrology 2003-04, Vol.62, p.577-594
Main Authors: McKenna, Sean A, Walker, Douglas D, Arnold, Bill
Format: Article
Language:English
Subjects:
Approximation
Boolean algebra
Computer simulation
Dispersion
Dispersions
Extreme values
Forecasting
Fracture mechanics
Fractured rock
Geological Phenomena
Geology
Geostatistics
Heterogeneity
Hydrology
Mathematical analysis
Mathematical models
Models, Theoretical
Mountains
Nevada
Permeability
Probability theory
Radioactive Waste
Refuse Disposal
Repositories
Stochastic continuum model
Storage facilities
Tracking
USA, Nevada, Yucca Mt
Variability
Water Movements
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Highly resolved numerical simulations are conducted to evaluate the longitudinal and transverse dispersivities proposed for use in the larger-scale Yucca Mountain saturated zone (SZ) site-scale model. Two different stochastic continuum models (SCM) that define the spatial variability of permeability are inferred from the observed fracture characteristics and the measured permeabilities. These models are created with a combination of indicator geostatistics and boolean simulation that allow for modeling different correlation lengths and anisotropy ratios at different permeability thresholds as well as the inclusion of large, high-permeability features. Longitudinal and transverse (horizontal and vertical) dispersion through the permeability realizations is evaluated for both distributed and focused source geometries using groundwater flow and streamline particle tracking. These numerical results are compared to behavior predicted by an analytical solution and to dispersivities estimated by an expert panel. Early time transport results are significantly non-Gaussian due to the strong heterogeneity of the fractured medium. At late times, travel distances of 23 correlation lengths, the longitudinal and transverse horizontal dispersivity results are well approximated by the analytical solution and the expert elicitation estimates. The calculated transverse vertical dispersivity values are smaller than those estimated from the analytical solution. Inclusion of high-permeability features of the same size as the model domain with a distributed planar source creates extreme values of the longitudinal and transverse horizontal dispersivity.
ISSN:0169-7722
1873-6009
DOI:10.1016/S0169-7722(02)00189-4