Investigation of core-mantle boundary topography using adjoint methods.

Using spectral-element and adjoint methods, we investigate body waves interacting with the Earth's most dramatic interface, the core-mantle boundary (CMB). Intermediate-to-high frequency seismograms are com puted incorporating topography models. We analyse the sensitivity of many seismic phases interacting with the interface. The study aims at showing effects of CMB structure on synthetics and highlights difficulties of imaging this region due to strong trade-off between velocity variations and topography.

There are many studies trying to understand and map velocity along this boundary and above, so-called the D^'' layer. How ever, the mantle area wherein seismic waves travel before and after interacting with the boundary makes the identification of isolated relevant seismic phases cumbersome. Moreover, observed traveltimes can be hard to interpret as a result of topography alone, due to approximate modelling and inversion of traveltime data.

Synthetic waveforms computed at dominant periods of 6-18 seconds are used in order to observe time shifts due to topography and calculate the adjoint sensitivity kernel. This is done by selecting a time window around the theoretical arrival of each phase. We focus on diffracted, core reflected and refracted P and S waves. The sensitivity kernel depicts first Fresnel zones of these phases and others that may contribute to narrow time window, although unpredicted by ray theory. We perform comparisons between time shifts due to topography models made on full-waveform synthetics to ray theoretical predictions to assess methods usually deployed for imaging CMB. This shows that for most phases ray theory performs well enough with some accuracy loss.

We propose that using relevant seismic phases simultaneously in full-waveform inversion may improve CMB topography imaging. However, it seems necessary to jointly invert for velocity variations due to D^'' layer, which is so far poorly understood and presents a challenge when it comes to identifying its effects on traveltime data. In our further research, a FWI workflow is being developed and aims at addressing these issues. It is envisioned that an improved image of the core-mantle boundary will be obtained.