Shear wave splitting and synthetic S wave tomography at Mt. Etna volcano

In volcanic systems, seismic anisotropy is a common phenomenon, typically attributed to the presence of eruptive fissure, dikes, sills and microcracks/pores, whose preferential orientation depends on the local stress field. A common tool for observing seismic anisotropy is the measurement of shear wave splitting (SWS), the splitting of shear waves into two quasi shear waves with orthogonal polarization directions and different propagation speeds when entering an anisotropic medium.  The relationship between seismic anisotropy and the density and orientation of fluid-filled cracks makes SWS an excellent tool for studying the volcanic stress field and associated volcano dynamics (Savage et al. 2010; Araragi et al. 2015; Johnson et al. 2015; Mroczek et al. 2020; Nardone et al. 2020). However, SWS observations only provide path-integrated information, so the interpretation of anisotropic features from these observations is limited. In contrast, body wave tomography studies that have the potential to give insights into the 3D distribution of anisotropy are often conducted assuming isotropy as this simplifies the seismic inversion strategy. However, P-wave (Bezada et al. 2016; VanderBeek and Faccenda 2021) and S-wave (VanderBeek et al. 2023) tomography experiments have shown that the assumption of isotropy in the presence of anisotropic structure can generate significant velocity imaging artifacts, potentially resulting in misinterpretation of true thermal and compositional heterogeneities.

Here we present a study of seismic anisotropy beneath Mt. Etna, one of the best monitored active basaltic volcanoes in the world. Our preliminary SWS measurements of local earthquakes between 2006 and 2016 (following the automated method of Hudson et al. (2023)) provide evidence for strong anisotropy at Mt. Etna. This is supported by previous SWS studies (Bianco et al. 2006; Nardone et al. 2020), as well as by P-wave anisotropic tomography (Lo Bue et al. 2024). The well-established sensitivity of S waves to fluids suggests that in volcanic environments S waves should be particularly sensitive to anisotropy due to preferentially aligned fluid-filled cracks. To quantify the potential bias in seismic imaging caused by the neglection of anisotropy, we have performed seismological synthetic experiments and compared synthetic isotropic tomography results from an isotropic and an anisotropic model (based on prior imaging of Mt. Etna by Del Piccolo et al. (in review)). Our results give new insights into the importance of incorporating seismic anisotropy in the study of the subsurface structure and dynamics of active volcanoes with S wave tomography.