Estimation of Dispersion Coefficients in Porous Media

Publisher Website
Google Scholar
Abstract
A procedure has been developed by which dispersion coefficients may be estimated in general porous media. The formulation utilizes the Taylor hypothesis which states that the dispersion coefficient tensor equals the covariance of the pore velocity fluctuations times the Lagrangian time scale tensor. Velocity fluctuations in the porous formation are simulated using a three‐dimensional groundwater flow model with random variations in hydraulic conductivity. Simulated pore velocity fluctuations in simplified formations agree well with theory. The form of the Lagrangian time scale tensor in formations where the mean pore velocity is much greater than the velocity fluctuations has been developed. Dispersion coefficients estimated by the proposed method have been shown to agree well with experimental results in isotropic formations. Comparison of predictions with experimental results in an anisotropic formation indicate order of magnitude agreement. The proposed methodology is expected to be applicable in cases where molecular diffusion effects are negligible. The formulation is expected to be extremely useful in those cases where dispersion measurements are unavailable or where hydrodynamic dispersion is to be separated from chemical attenuation.

This publication has 12 references indexed in Scilit: