Crustal Anisotropy from the Birefringence of P-to-S Converted Waves: Bias Associated with P-Wave Anisotropy

Many researchers have used the birefringence of P‑to‑S converted waves from Moho discontinuity to constrain anisotropy Earth’s crust. However, this practice ignores substantial influence that has on initial amplitude wave, which adds splitting acquired during its propagation seismometer. We find large variations in Ps estimates with back-azimuth occur theoretically presence P‑wave anisotropy, normally accompanies S‑wave anisotropy. The are largest for crustal a tilted axis symmetry, geometry is often neglected interpretations, but commonly found simulated globally-distributed P‑coda datasets 36 distinct 4‑layer models combinations elliptical shear or compressional and also incorporated higher-order anisotropic Backus parameter C. tested both horizontal symmetry-axis geometries tradeoff associated phases at top bottom thin high‑ low‑velocity basal layer. computed composite receiver functions (RFs) harmonic regression over back azimuth, using multipletaper correlation moveout corrections epicentral distances 471 events, simulate realistic data set. estimate radial transverse RFs, strategy similar previous studies. can be useful indicator bulk only under restricted circumstance, either media no if symmetry throughout. In other, more-realistic cases, inferred fast polarization estimated synthetic RFs tends drift back-azimuth, form weak penalty-function minima, return times depend thickness an layer, rather than accumulated within it.