Source stacking followed by cross-correlations: application to global waveform tomography using the Spectral Element Method

Accurate synthetic seismic wavefields can now be computed in 3D earth models using the spectral element method (SEM), which helps improve resolution in full waveform global tomography. However, computational costs are still a challenge. These costs can be reduced by implementing a source stacking method (Capdeville et al., 2005, GJI), in which multiple earthquake sources are simultaneously triggered in only one teleseismic SEM simulation. One drawback of this approach is the perceived loss of resolution at depth, in particular because high-amplitude fundamental mode surface waves dominate the summed waveforms, without the possibility of windowing and weighting as in conventional waveform tomography.

This can be addressed (e.g. Romanowicz et al., 2013, Fall AGU), by computing the cross-correlation wavefield between pairs of stations before each inversion iteration. While the Green's function between the two stations is not reconstructed as well as in the case of ambient noise tomography, this is not an issue, since the same processing is applied to the 3D synthetics and to the data, and the source parameters are known to a good approximation. By doing so, we can separate time windows with large energy arrivals corresponding to fundamental mode surface waves, and apply a weighting scheme to bring out the contribution of overtones and body waves.

Here we present the results of testing this approach for a synthetic 3-component waveform dataset computed for 273 globally distributed events in a toy 3D radially anisotropic upper mantle model which contains shear wave anomalies at different scales. We compare the results of inversion of 10,000 s stacked time series using source stacking, and cross-correlations with and without weighting.

In the case of real data, the issue of missing records can be minimized by applying cluster analysis to optimally grouped seismic sources into a small number of groups, therefore still reducing the number of SEM computations required by at least an order of magnitude. We show first results of application at the global scale, including first and second orbit surface waves down to 60s period, starting with a 1D model. We compare the resulting global 3D radially anisotropic shear velocity model obtained after several iterations to other existing models constructed using conventional approaches.