Effects of anisotropy in southwest Germany on the propagation of surface waves

Abstract

This paper discusses the effects of anisotropy in southwest Germany on the propagation of surface waves. A model for the anisotropy has been compiled previously by Fuchs from the combined information of petrological and seismological data. Seismic evidence for the anisotropy has been derived from the observations of azimuthal variation of Pn velocity and from a P-velocity depth function for a refraction seismic profile. The purpose of this paper is to present the effects of this model of the seismic anisotropy in southern Germany and similar ones on the dispersion and polarization of surface waves. For quasi-Rayleigh and quasi-Love waves the phase velocity variations with azimuth amount to about 2%. The period of maximum variation is 20 s for Rayleigh waves and about 30 s for Love waves. Anisotropy causes the particle motion for both generalized surface waves to be elliptical, where the plane of the ellipse is inclined with respect to the direction of particle motion in isotropic media. For Love waves the angular variation of the polarization plane is about 6° maximum, whereas for Rayleigh waves it amounts to 16°. The sign of the azimuthal effect of dispersion is independent of the period for both wave types. This means for all periods the same directions are fast or slow. In contrast, there is a change in sign for the polarization effect, which causes the plane of polarization to be inclined in the opposite direction for long-period waves as compared to rather short periods. For rather simple models it is shown that the period of the polarization node depends linearly on the depth of the anisotropic layer and that independent of the layer thickness, the change in the polarization direction occurs where the wavelength is about five times as large as the mean depth of the anisotropic layer. Observed dispersion data for surface waves in the area of southern Germany do not yet allow us to derive a detailed model of anisotropy as it has been compiled from body-wave data. This is due to the lack in azimuthal coverage, reduced resolution for analog data, the lack in overlapping period ranges for the observations and the travel-time differences caused by heterogeneities crossed by the ray paths.