2013 Volume 122 Issue 1 Pages 170-191
It is well known that longitudinal wave velocities and attenuation coefficients in fluid-saturated porous rocks vary with frequency. Evidence comes from experimental measurements and the poroelastic theory using Biot's model. A number of experimental studies in the literature report either direct or indirect evidence of elastic wave dispersion in water-saturated porous rocks. Biot's model predicts a low-frequency sound speed, followed by an increase to a higher sound speed beyond a transition frequency. Biot's model of large-scale mechanisms explains well the dispersion observed in high-porosity rocks, but the predictions of Biot's theory for dispersion are usually lower than measured values by some orders of magnitude. Measurements show that the Biot's model alone does not adequately explain observed velocity dispersion. As a result, much research has been devoted to modifying some aspects of the theory to include additional dispersion mechanisms. Viscoelasticity due to the creation of a local fluid flow approach was devised to represent an additional dispersion mechanism to Biot's model, because Biot's theory ignores the effects of fluid distribution heterogeneity within rocks on their seismic properties. This approach focuses on losses that occur due to the local flow of pore fluid in an individual pore when it undergoes deformation caused by passing longitudinal waves. Several applications of local flow effects can be found in the literature and the squirt-flow is generally known as a dispersion mechanism resulting from pore pressure relaxation caused by viscous flow. However, it is difficult to construct a quantitative model of local flow phenomena because it has to address rock on a microscopic pore-scale level. So far, there has been no consensus on the role of a mechanism that adequately predicts the observed dispersion properties of real rocks. The main purpose of this paper is to provide further insights into the nature of squirt-flow phenomena. Sedimentary rock and crystalline rock specimens were tested using longitudinal waves for differences in velocity dispersion phenomena observed in each specimen, and we examined whether we can quantitatively explain the observed magnitude of dispersion in a variety of rocks. The measurements and theoretical argument show that the squirt-flow model can be applied to observed experimental data. Moreover, it is clear that we can estimate the permeability of rock specimens and artificial porous media using seismic wave dispersion characteristics based on the squirt-flow model.