3-D Joint Regional Inversion of Teleseismic Body Waves and Rayleigh Waves

Teleseismic body waves and surface waves have distinctly different and complementary strengths and weaknesses when attempting to resolve regional scale structures using seismic tomography. Body waves are best for resolving lateral velocity variations, but because of the near vertical incidence of teleseismic waves, they have only limited ability to resolve variations in velocity with depth. Furthermore, teleseismic body wave tomography can resolve only relative velocity variations within the model space, not absolute velocities. Surface waves are better at resolving changes in velocity with depth, but because they average over structure along horizontal paths, they have more difficulty constraining small lateral variations. Also, surface waves provide information on absolute velocities, a feature that makes them particularly valuable for helping to constrain possible interpretations of the observed velocity structure. We expand the 2-D method of West et al. (2004) to 3-D to perform a joint inversion of body and surface waves. Differential surface wave arrival times for a range of frequencies are treated as "travel times," and paths are adjusted to allow for perturbations in the wavefront caused by structural heterogeneity prior to its arrival in the model space. We apply this method to the seismic dataset collected as part of the Southern Africa Seismic Experiment (SASE). This is an ideal dataset with which to test this methodology, since both body waves (James et al., 2001; Fouch et al., 2004) and surface waves (Larson et al. 2006) have been investigated previously, and these studies indicate that the data quality is high and the azimuthal distribution is good. By jointly inverting these two datasets, we can better constrain the absolute velocity structure in the area, and gain more insight into the nature and extent of the lithospheric root beneath the Kaapvaal Craton.