In Fig. 2, we compare the prediction of Johnson (2001) for KU to our own for a consolidated sandstone (frame properties as determined in the Appendix with k=100 mD, c=10, ) in which phase 1 is saturated with water and phase 2 is taken to be spherical regions saturated with air. The two estimates have identical asymptotic dependence in both the limits of high and low frequencies. In the cross-over range, the physics is not precisely modeled in either approach. However, even in the cross-over range, the differences in the two models is slight.
Figure 3 gives the P-velocity and attenuation for a model in which the frame properties correspond to k=10 mD, c=15, and . Phase 2 is saturated by air and is taken to be isolated spheres of radius a=1 cm. Phase 1 is saturated with water. The volume fraction v2 occupied by these 1 cm spheres of gas is as shown in the figure. Even tiny amounts of gas saturation yields rather large amounts of attenuation and dispersion.