On the Improvement of SKS Splitting Measurements by the Simultaneous Inversion of Multiple Waveforms (SIMW)

Seismic anisotropy in the Earth's upper mantle is mainly due to preferred orientation of anisotropic minerals such as olivine, caused by plastic deformation due to shear stress and strain in the lithosphere and asthenosphere. Large-scale azimuthal anisotropy can be observed seismologically e.g., by using core-refracted shear waves (SKS, SKKS) for splitting measurements. They provide constrains on the strength and the orientation of S-wave anisotropy in the upper mantle beneath a recording station. However, the depth resolution is generally poor, and results are often interpreted under the assumption of a single anisotropic layer. In order so resolve multiple-layer anisotropic structures, measurements from different backazimuths are necessary. For many regions on Earth, well-recorded SKS- or SKKS-phase observations are not available at least for different backazimuths, due to missing earthquake recordings at the required distance range or due to a too low signal-to-noise ratio. We propose a new method, which allows the simultaneous inversion of several SKS splitting measurements from different earthquakes of the same source region, measured at a single seismic station. The individual SKS waveforms are concatenated to form one new waveform, which is then inverted with the Silver & Chan method to determine the splitting parameters dt and phi. We apply our method to recordings at the stations of the Norwegian Seismic Array (NORSAR). Our results demonstrate that this stacking method allows us a stable determination of splitting results even for regions with low-amplitude, or noisy signals.