Evolution of Earth Structure and Future Directions of 3D Modeling

Abstract

It is no longer adequate to treat the Earth as a nearly spherically symmetric body with simple receiver, source and attenuation corrections tacked on. The aspherical velocity structure is now being determined by surface wave and body wave tomographic techniques and it has been found that heterogeneities are present at all levels. In the upper mantle the lateral variations in velocity are as large as the variations across the radial discontinuities. There is good correlation of velocity with surface tectonic features in the upper 250 km but the correlation rapidly dimishes below this depth. The focusing and defocusing effect of these lateral variations can cause large amplitude anomalies and these effects can be more important than attenuation. Velocity variations in the mantle can be caused by temperature, mineralogy and anisotropy, or crystal orientation. The largest variations are caused by anisotropy and relaxation phenomena such as partial melting and dislocation relaxation. There is increasing evidence for anisotropy in the upper mantle and this must be taken into account in Earth structure modeling. Both azimuthal and polarization effects are important. Layering or fabric having a scale length less than a wavelength will show the statistical properties of the small scale structure. Global maps of heterogeneity and anisotropy show that if anisotropy is ignored the data will be mapped into a false heterogeneity. Azimuthal anisotropy compounds the off-great-circle problem. The absorption band concept predicts that Q should be higher at short periods than at long periods and that there should be large lateral and radial variations in Q. The t* controversy is probably related to shifts in the absorption band. If velocity is anisotropic then Q should be as well. Evidence is starting to suggest that there is a Love wave, Rayleigh wave discrepancy in Q, suggestive of Q anisotropy.