Revisiting moment tensors in Switzerland: Unraveling source characteristics in Central Alps and their foreland

Studies on moment tensors (MT) and focal mechanisms are of great importance for assessing regional and local seismotectonic processes, especially when a high-quality, dense network is in operation. However, common MT inversion methods are largely restricted to magnitudes > 3.5. In order to lower the completeness of MT catalogs, improved Green’s functions and/or hybrid inversion techniques are needed. In this study, we revisit small-to-moderate earthquakes, which occurred in Switzerland and surrounding regions by means of various MT inversion methods and assess the potential to improve completeness of MT catalogs in Central Alps region. To accomplish this, we implement state-of the art methods for MT inversion using either full waveform data or combinations of first-motion polarities with amplitudes and amplitude ratios. Methods based on full waveform inversion considered in this study are ISOLA (Sokos & Zahradnik 2013) and Grond (Heimann et al. 2018), as well as techniques based on amplitudes and/or polarities (HybridMT (Kwiatek et al. 2016), MTfit (Pugh & White 2018)), which can solve MTs for smaller magnitude earthquakes. Hence, the combination of multiple techniques allows to compute full or deviatoric MTs for a broader range of magnitudes and enrich the existing catalogs.

We first apply these methods to recent earthquake sequences occurred in the Central Alps between 2019 and 2021. During that period, several earthquake sequences, like the one associated with the 2021 M4.1 Arolla earthquake, occurred and show complexity on the waveforms, due to their shallow focal depths. In addition, several of the standard MT solutions calculated by the Swiss Seismological Service (SED) for these earthquakes indicate complex moment tensors with unusually high percentage of the CLVD component. To check whether such CLVD component is real and not an artifact caused, for instance, by unmodeled heterogeneities, we invert for full and deviatoric MTs using multiple 1D velocity models and algorithms. Additionally, we perform MT inversions for several earthquakes either within selected earthquake sequences or regional background seismicity. The resulting MT solutions are compared to existing high-quality focal mechanisms computed using first motion polarities as well as to high-precision double difference locations. Uncertainties of MT solutions are estimated using bootstrap-based methods. This work contributes towards an enriched high-quality focal mechanisms database for Switzerland, which could be used to revisit the regional to local stress field at unprecedented resolution and provides new insights into the complexities of active fault systems in the Central Alps region.

References:

Heimann, S., Isken, M., Kühnn, D., Sudhaus, H., Steinberg, A., Vasyura-Bathke, H., Daout, S., et al. (2018) Grond - A probabilistic earthquake source inversion framework., GFZ Data Services. doi:10.5880/GFZ.2.1.2018.003

Kwiatek, G., Martínez-Garzón, P. & Bohnhoff, M. (2016) HybridMT: A MATLAB/Shell Environment Package for Seismic Moment Tensor Inversion and Refinement. Seismol. Res. Lett., 87, 964–976. doi:10.1785/0220150251

Pugh, D.J. & White, R.S. (2018) MTfit: A Bayesian Approach to Seismic Moment Tensor Inversion. Seismol. Res. Lett., 89, 1507–1513. doi:10.1785/0220170273

Sokos, E.N. & Zahradnik, J. (2013) Evaluating Centroid-Moment-Tensor Uncertainty in the New Version of ISOLA Software. Seismol. Res. Lett., 84, 656–665. doi:10.1785/0220130002