Gravitational and teconic forces controlling post-collisional deformation and the present-day stress field of the Alps. Constraints from numerical modelling

(pdf)

We perform numerical modelling to investigate the mechanisms leading to the post-collisional tectonic evolution of Alps. We model the lithospheric deformation as a viscous thin-sheet with vertically-averaged rheology and coupled with surface mass transport. The applied kinematic boundary conditions simulate the convergence between the Adria indenter and the European foreland during the last 35 Ma. Model predictions of elevation, lithospheric structure, erosion/sedimentation pattern and vertical-axis rotation are compared with observations of the planform shape of the chain, topography, crustal thickness, distribution of rock exhumation and sediment thickness, and paleomagnetic rotations.
Thickening of the lithosphere in the Alpine region is shown to be highly sensitive to the assumed viscosity law, to the strength contrasts between the different regions and to the surface mass transport. Modelling results indicate that the large-scale deformation of the Alps during the post-collisional phase is mainly controlled by accommodation of convergence in a weak orogenic zone caught between a nearly rigid Adria plate and a strong European foreland.
Modelling of the present-day stress field shows that: 1) the present rotation of Adria is responsible for the change of extension direction from strike-perpendicular in the western Alps to strike-parallel in the east; and 2) the strike-perpendicular extension observed in the western Alps is likely due to lateral variations of gravitational potential energy. The results suggest a NNE shift of about 700 km of the Euler pole of Adria relative to Europe from its mean position during post-collisional deformation to the present-day.

a) Topography of the Alpine region and location of the modelling area (white rectangle) and of the geologic map (black rectangle); b) Geologic map of the study area showing temperature of metamorphic rock samples (black contours, in °C; every 100°C would imply exhumation between 3 and 5 km) at approximately 30 Ma. Yellow/orange colours show sediment thickness in the Molasse (Oligocene to present) and Po basins (Pliocene to present). Numbers in circles identify the following geological units: 1: Mont Blanc/Aiguilles Rouges massif; 2: Aar Massif; 3: Engadine Window; and 4: Tauern Window.

Model results at the last stage (35 Myr, corresponding to the present) for Model A (first column): temperature-dependent viscosity and constant effective strain rate of 4 10-16 s-1 thorough the entire model domain; Model B (second column): temperature- and strain rate-dependent viscosity; Model C (third column): as B but incorporating erosion and sedimentation, Kd=1000 m2/yr. Upper panels show the actual topography in the study area (color shading) and the model-predicted topography (contours). Central and lower panels show the predicted Moho depth and lithospheric mantle thickness. Lowermost-right panel shows the amount of erosion/sedimentation predicted in Model C (negative values for erosion and positive for sedimentation). Note that the three models incorporate the same boundary conditions, and therefore the same amount of mass input and total lithospheric shortening. Differences in topography and thickness distributions are due exclusively to the different rheology assumed (compare models A and B) and to the effects of surface transport (compare models B and C).

Horizontal components of the principal strain rate directions, predicted by thin-sheet Model C (black bars for compression and white arrows for extension) and measures compiled by Kastrup et al. (2004) (black arrows for compression and grey arrows for extension). Contours are Moho depth, with 2 km interval (Waldhauser et al., 2002).

Horizontal components of the principal strain rate tensor predicted for the last stage (present day) of Model C (arrows for extension and bars for compression). Insets: 1 corresponds to the boundary condition without Adria movement and applied on panel a); and 2 present rotation of Adria, the velocity boundary conditions are computed assuming an Euler pole from a geodetic study and applied on panels b) and c). The strain rates are obtained considering: a) the only driving force comes from lateral variations of GPE, no Adria convergence (inset 1); b) by applying in an homogeneous lithosphere (no lateral variations of crustal and lithospheric thickness, therefore, no lateral contrast of GPE) a velocity boundary condition computed from the present Euler pole (inset 2); and c) by applying to the last stage (including the lateral variations of GPE) the velocity boundary condition computed from the same Euler pole (inset 2). Contours correspond to the predicted elevation.

The present-day strike-perpendicular extension observed in the western Alps can be explained as driven by lateral variations of GPE in this area. Ongoing rotation of Adria relative to Europe causes the observed rotation of extension direction from perpendicular to the chain in the western Alps to parallel to it in the eastern Alps. In this way, we obtain a change in tectonic regime from normal-faulting in the southwestern Alps to thrusting in the eastern Alps. Whereas explaining this present stress regime requires a rotation pole similar to that derived from geodetic measurements (9.10°E, 45.36°N), a rotation pole further to the south (7.1°E,39.3°N) is needed to reproduce the distribution of mean post-collisional Alpine deformation. Consequently, we propose that the Euler pole of Adria (relative to Europe) has shifted NNE-ward about 700 km from the location responsible for mean post-collisional deformation to the present-day situation.