Comments on 'Global anistropic phase velocity maps for higher mode Love and Rayleigh waves' by Visser, Trampert and Kennet

A recent paper by Visser et al. analyses large numbers of surface wave seismograms for velocity anisotropy by assuming the waves separate into two independent sets of modes with (implicitly) the Rayleigh- and Love-wave particle motion expected from propagation in isotropic substrates. Visser et al. claim they present 'global anisotropic phase velocity maps for higher mode Rayleigh waves and Love waves'. I show that surface wave modes in an azimuthally varying anisotropic Earth are generalized modes with generalized elliptical particle-motion in three-dimensions that cannot be separated into independent sets of Rayleigh- and Love-modes with sagittal and transverse-horizontal particle motion. Inversion would lead to two independent sets of anisotropic elastic constants which would be inappropriate for modelling an anisotropic Earth.