Seismic receiver function interpretation: Ps splitting or anisotropic underplating?

Abstract

Crustal anisotropy is crucial to understanding the evolutionary history of Earth's lithosphere. Shear wave splitting of Moho P-to-S converted phases in receiver functions (RFs) have been often used to study crustal anisotropy. Harmonic variation of Moho Ps phases in delay times are used to infer splitting parameters of averaged anisotropy in the crust. However, crustal anisotropy may distribute at various levels within the crust due to complex deformational processes. Layered anisotropy requires careful investigation of the distribution of anisotropy before interpreting Moho Ps splitting. In this study, we show results from stations ARU in Russia, KIP in the Hawaiian Islands and LSA in Tibetan Plateau, where layered anisotropy is constrained well by intracrust Ps conversions at high frequencies using a harmonic-decomposition technique. Anisotropic velocity models are inferred by forward-modeling decomposed RF waveforms. We suggest that the harmonic variation of Moho Ps phases should always be investigated to check for anisotropic layering using RFs with frequency content above 1 Hz, rather than simply reporting averaged anisotropy of the whole crust.