Effect of lithospheric structure and rheology on flow patterns beneath the European Alps

Lithospheric and mantle dynamics beneath the European Alps have been linked to surface deformation and uplift, but the deep structure itself remains a matter of debate. Some tomographic studies suggest a subduction polarity flip along strike, whereas others infer that there are two vertical and/or partly detached slabs. Both scenarios may affect the dynamics of the lithosphere and flow in the upper mantle in distinct ways. We set out to explore the dynamic predictions of these models so as to understand the range of likely scenarios, and also to guide future geological and geophysical studies. We perform high-resolution, regional scale models of coupled mantle-lithospheric deformation. These 3D geomechanical models are discretized with finite differences and realized with a parallel code (LaMEM). It includes an internal free surface and a detailed, isostatically balanced crustal/lithospheric model based on seismological interpretations. Quasi-instantaneous computations are performed for various geometries (a single vertical slab, a slab that is partly detached, and two slabs with opposite vergence) and we vary both, the densities and effective viscosities of the slabs. We find that the effect of geometry is much smaller than the effect of effective viscosity of the slabs. If the viscosity contrast between slab and asthenosphere is relatively small, mantle flow is predominantly vertical, whereas significant (anticlockwise) toroidal flow occurs at the eastern edge of the Alps for larger viscosity contrasts (> 500). This toroidal flow is not very sensitive to the detailed geometry of the sub-Alpine slab configuration but is rather caused by the slab roll-back underneath the northern Apennines. Partial slab breakoff, however, has a discernable effect with higher average elevations above locations where the slab detached. In some of our models, we obtain Alp-parallel flow, which roughly corresponds to inferences from recent studies on azimuthal seismic anisotropy. Our results suggest that future seismological studies should not only focus on the Alpine chain itself but also include the wider region. This study was funded by ERC Starting Grant 258830