Vertical and Small-scale Lateral Varying Seismic Anisotropy in the Upper Mantle Underneath the Upper Rhine Graben, Central Europe

The analysis of long-term recordings (partly > 20 yr) of shear wave splitting of XKS (SKS, SKKS, PKS) phases at six (semi-) permanent broadband seismological recording stations in the Upper Rhine Graben (URG) area, Central Europe, indicates vertical and small-lateral variations in the seismic anisotropy. We measure the splitting parameters, the fast polarization direction and the delay time, using SplitLab (Wüstefeld et al. 2008) with the plugin StackSplit (Grund 2017) and calculate the splitting intensity. The analysis of SKS-SKKS pairs have no significant discrepancies what indicates no major contribution for the lowermost mantle to the observed anisotropy.

To explain the shear wave splitting measurements, we tested structural anisotropy models with one layer with a horizontal as well as a tilted symmetry axis and two layers with horizontal symmetry axes using the MATLAB Seismic Anisotropy Toolbox (Walker and Wookey 2012). Beside fitting the whole backazimuth (BAZ) range, we also limit the modeling to station-specific BAZ ranges. The consideration of differences and similarities in the splitting pattern across the recording stations allows us to propose a structural 3-D models of the anisotropy in the upper mantle underneath the URG. ... mehrDifferences are found between the east and the west sides of the URG and between the Moldanubian Zone in the south and the Saxothuringian Zone in the north. Inconsistent splitting parameters and the observation of numerous null observations, especially below the southern URG may be at least partly related to scattering of the seismic wavefield or a modification of the mantle material.

To test this anisotropy 3-D model, synthetic seismograms using AxiSEM3D (Leng et al. 2016, 2019) are generated. Besides comparing synthetic and observed seismograms, we measure the shear wave splitting of the synthetic phases. Comparing observed and modeled splitting parameters and splitting intensity will help us to update and improve the 3-D anisotropy model.