Field investigation of the Waste Isolation Pilot Plant (WIPP) site (New Mexico) using a nonstationary stochastic model with a trending hydraulic conductivity field

Authors
Citation
K. Seong et Y. Rubin, Field investigation of the Waste Isolation Pilot Plant (WIPP) site (New Mexico) using a nonstationary stochastic model with a trending hydraulic conductivity field, WATER RES R, 35(4), 1999, pp. 1011-1018
Citations number
28
Categorie Soggetti
Environment/Ecology,"Civil Engineering
Journal title
WATER RESOURCES RESEARCH
ISSN journal
00431397 → ACNP
Volume
35
Issue
4
Year of publication
1999
Pages
1011 - 1018
Database
ISI
SICI code
0043-1397(199904)35:4<1011:FIOTWI>2.0.ZU;2-N
Abstract
Verification of a stochastic model which models the phenomena of flow and t ransport in a nonstationary conductivity field [Rubin and Seong, 1994] is a ttempted with field data from the Waste Isolation Pilot Plant (WIPP) site n ear Carlsbad, New Mexico. The nonstationarity is manifested as a spatial li near trend in the mean log conductivity field. Analysis of the log conducti vity and the hydraulic head data shows that the site corresponds to Rubin a nd Seong's alpha(2) model, where the mean flow direction and the gradient o f mean log conductivity are normal to each other. The experimental semivari ogram of the log conductivity obtained from field data strongly suggests th at it is nonstationary and can be fit with a Gaussian covariance function. The alpha(2) model is successful in predicting the experimental semivariogr am of the head obtained from the field data, especially in capturing the fi nite sill characteristics. The stationary model predicts an unbounded semiv ariogram at large distances. Solute travel times are also investigated by c omparing the travel time cumulative distribution function (CDF) resulting f rom the stationary and the nonstationary hydraulic conductivity models. The stationary model predicts a higher CDF at early times and hence overestima tes the very early travel times to the regulatory plane of compliance than the alpha(2) model.