Tomographic Image Interpretation and Central-Western Mediterranean-Like Upper Mantle Dynamics From Coupled Seismological and Geodynamic Modeling Approach

The Central-Western Mediterranean (CWM) is one of the most complex tectonic setting on Earth. Episodes of slab rollback, break-off and tearing, the opening of back-arc extensional basins (i.e., Liguro-Provencal, Alborean, Algerian and Tyrrhenian basins), the presence of large mountain ranges, active volcanoes and violent earthquakes have made the Mediterranean an ideal environment to study a wide range of geodynamic processes and an important target for seismological studies (e.g, seismic tomography). Here we build a geodynamic model which, although it does not reproduce its exact tectonic structure (e.g., due to the limits of the numerical method, approximations in the initial setup, etc), presents multiple and geometrically complex subduction systems analogous to those found in the CWM. The tectonic evolution of this model is estimated with petrological-thermo-mechanical 3D simulations, then, we dynamically compute the upper mantle fabrics and seismic anisotropy as a function of the strain history and local P-T conditions. After comparing the model with SKS splitting observations in order to quantify the discrepancies with the true Central-Western Mediterranean, we use the elastic tensors predicted for the modeled configuration to perform 3D P-wave anisotropic tomography by inverting synthetic P-wave delay times. Using the geodynamic model as reference, we evaluate the capabilities of a recently developed seismic tomography technique to recover the isotropic anomalies and anisotropy patterns related to a complex subduction environment in different conditions, such as poor data coverage and bad data quality. We observe that, although P-wave tomography still remains a powerful tool to investigate the upper mantle, the reliability of the retrieved structures strongly depends on data quality and data density. Furthermore, the recovered anisotropic patterns are consistent with those of the target model, but in general an underestimation of the anisotropy magnitude in the upper mantle is observed. In the light of future developments, our study suggests that by combining micro- and macro-scale geodynamic simulations and seismological modeling of seismic anisotropy it will be possible to reproduce, at least to a first order, the tectonic evolution of real study regions (e.g., the Mediterranean) thus providing fundamental constraints on the processes that have contributed in shaping their current geological scenario.