HR: 1340h
AN: H53F-1488    [Abstracts]
TI: Measurement of Relative Permeability Using Single-Step Transient Flow Centrifuge Experiments
AU: * van den Berg, E H
EM: evandenb@utk.edu
AF: The University of Tennessee. Department of Earth and Planetary Sciences, 1412 Circle Drive, Knoxville, TN 37996-1410, United States
AU: Perfect, E
EM: eperfect@utk.edu
AF: The University of Tennessee. Department of Earth and Planetary Sciences, 1412 Circle Drive, Knoxville, TN 37996-1410, United States
AU: Mayes, M A
EM: mayesma@ornl.gov
AF: Oak Ridge National Laboratory. Environmental Sciences Division, P.O. Box 2008, MS-6038, Oak Ridge, TN 37831, United States
AB: Steady-state flow centrifuge methods have been widely used to determine the relative permeability function, k_rw, of porous media. The centripetal acceleration greatly accelerates the speed at which equilibrium conditions are found in drainage or imbibition experiments. While equilibrium conditions are required by steady- state flow centrifuge methods, transient flow centrifuge methods utilize the time variant adaptation of state variables (e.g. fluid saturation, pressure or flux) to sudden changes in boundary conditions. Transient flow centrifuge experiments were conducted using a commercial rock core ultracentrifuge to displace wetting fluids from an initially fully-saturated sample by non-wetting fluids. The cumulative production of wetting fluid leaving the sample was recorded using an automated digital camera recording system. The time series of cumulative wetting fluid production forms the input for the objective function of inverse numerical modeling efforts and for the independent estimation of k_rw using an analytical solution for the transient flow of fluids through a porous medium in a centripetal field (the Hagoort method). In this study, single-step transient outflow experiments were run on Berea sandstone cores. The angular velocity was 942 sec^-1. Independent laboratory techniques were used to measure the saturated water content, θ_s, and the intrinsic permeability, k_sat. Besides these measurements, synthetic production data were generated with forward simulations of Hydrus-1D using the standard van Genuchten parameters for 'Sand' and 'Silt' soils. The models simulated acceleration of the centrifuge rotor during the first ~95 seconds of the experiment by increasing the angular velocity stepwise in 2.5 second intervals to 733 sec^-1. The data were interpreted using a modified version of the Hagoort method in which numerical derivatives are used to compute the slope of the production curve in order to calculate relative saturation and the corresponding k_rw function. We also used Hydrus-1D to inversely estimate the parameters of the van Genuchten relative permeability model, θ_r, a, n, and l. Systematic variation of the initial parameters values used in the inverse simulations showed that some of the parameter estimates obtained with inverse simulation were non-unique. For both the synthetic simulations and the measurement data, the difference between the original k_rw curve and the k_rw curve obtained with inverse simulations was most sensitive to the initial estimate of the shape parameter, n. Estimation of initial parameter values could be augmented by the good correspondence between the k_rw curves obtained with the Hagoort method and those used and obtained in numerical (inverse) simulations. Transient flow centrifuge experiments and their interpretation are expected to have applications in predicting infiltration and remediation of environmental contamination within the vadose zone.
DE: 1865 Soils (0486)
DE: 1866 Soil moisture
DE: 1875 Vadose zone
DE: 5114 Permeability and porosity
SC: Hydrology [H]
MN: 2007 Fall Meeting