2D Global Rayleigh Wave Attenuation Model Using Finite Frequency Focusing and Defocusing Theory

Abstract:

We have developed an efficient technique to process and measure surface-wave amplitude and phase from a large collection of seismic waveforms. These amplitude and phase data sets are used to jointly invert for 2D phase velocity and attenuation maps. As demonstrated by Dalton and Ekstrom (2006), correcting for the effects of focusing and defocusing by elastic structure is crucial in order to obtain reliable attenuation structures.

A robust theory that can reliably predict focusing-defocusing effects and is insensitive to the details of making the phase velocity maps is preferred. Great circle ray theory can give useful predictions for the focusing-defocusing effects if careful attention is paid to how the phase velocity model is smoothed. However, the predictions of the finite frequency kernels are more robust at the low-intermediate frequency range (below 25mHz) and suggest that they are better suited as a basis for inversion.

We invert for the phase velocity, attenuation, source, and receiver terms simultaneously. Our models provide 60-70% variance reduction to the raw data though the source terms are the biggest contribution to the fit of the data. The attenuation maps show structures that correlate well with surface tectonics and the age-dependent trend of attenuation is clearly seen in the ocean basins. We have also identified problematic stations and earthquake sources as a by-product of our data selection process.

Although our approach was developed for a global study, it can be extended to regional studies. Our first regional-scale application of this approach is to the Atlantic upper mantle.