Detecting Anisotropy from Back-Azimuth Amplitude Dependence of Sp Converted Waves

Abstract

Many researchers have used S-to-P (Sp) converted waves to detect the Moho discontinuity, the lithosphere-asthenosphere boundary (LAB) and mid-lithospheric discontinuities (MLD). The anisotropy of Earth’s lithosphere is typically constrained with shear-wave birefringence.  Both theory and reflectivity computations, however, argue for a substantial influence of anisotropy on the initial amplitude of the Sp converted wave.  The effects of compressional anisotropy on initial Sp amplitudes are stronger than the effects of shear anisotropy for anisotropy with a tilted axis of symmetry, a geometry that is often neglected in birefringence interpretations.  This Sp behavior is not typically studied, but it has the potential to test the hypothesis that the seismic lithosphere-asthenosphere boundary (LAB) is caused by a transition in anisotropic layering at the base of Earth’s tectonic plates.We develop and apply multiple-taper correlation estimates for Sp receiver functions, applicable to either SV or SH incoming polarization, or for a linear combination of SV and SH. In the context of incoming SV-polarized body waves, e.g., SKS phases, algorithms from multiple-taper Ps RFs can be borrowed to apply moveout corrections before the Fourier transform to target a particular interface depth in the crust or mantle.  With synthetic seismograms, we find that SH “receiver functions” can be computed from incoming SV waves, promising a diagnostic detecting SKS birefringence and to estimate an average splitting signal from a station. The SV and SH waveforms can be “unsplit” in the frequency domain by the estimated average birefringence to reconstruct the S waves that impinge the lithosphere from the deep mantle.  We will report analyses with data from permanent stations of the Global Seismographic Network and the USGS ANSS.