A numerical analysis of seismic waves for an anisotropic fault zone

In this study we examine the effects of anisotropy on the seismic wavefield in a fault zone from computation of the synthetic seismograms for a simple fault zone model and a variety of seismic wave sources. The fault zone is modeled by a homogeneous vertical layer with transverse isotropy, induced by cracks, sandwiched between isotropic half-spaces (host rocks). The symmetry axis of the transverse isotropy is horizontal and perpendicular to the fault zone strike. We calculate the synthetic seismograms for this anisotropic fault zone model using a semianalytical method, the propagator matrix method. The synthetic seismograms show a later phase arriving after the main shear-wave in the horizontal component perpendicular to the fault zone strike at most stations near the fault zone. It is the slower shear-wave (qS2) and its reverberation. The amplitude of this phase and the time delay from the main shear-wave arrival are proportional to the degree of anisotropy, which suggests that observing such phase in field measurements may imply the presence of an anisotropic fault zone. We also perform the shear-wave splitting measurements by applying the cross-correlation method to the synthetic seismograms for various sources. For a strike-slip source, the synthetic seismograms show that the wavefield is more affected by the velocity structure than by the degree of anisotropy, which makes it difficult to estimate the anisotropic (shearwave splitting) parameters. For normal and dip-slip fault sources with the strike parallel to or striking against the fault zone, the effects of anisotropy is so dominant that the anisotropic fault zone can be detected. These results suggest that the determination of the anisotropic properties in the fault zone would require an appropriate station deployment and the source type information.