Final Report Summary - RHEOLITH (Rheology of the continental lithosphere, a geological, experimental and numerical approach)
The surface of our planet is permanently moving and the interaction between moving plates leads to deformation. Long-term deformation produces mountain belts or rift zones, short-term deformation produces earthquakes. One of the key-issues for understanding these first order processes is the mechanical behaviour of continental lithosphere and why it behaves so that deformation is localized along preferential zones. RHEOLITH was designed to approach this question through the association of three groups of specialists: field geologists who described the studied shear zones, experimentalists who worked on the mechanical behaviour of rocks at small-scale and numerical modellers who had in charge the 2D or 3D modelling of the lithosphere at different scales. In order to be efficient the project required the installation of a new radiochronology laboratory and a new press for deforming rocks at high pressure and high temperature in ISTO (Orléans) as well as a fast computer in Paris. The main results of the project are the following. Strain localization is a complex process that depends upon the scale considered. At the scale of a hand-specimen or smaller, the reduction of the grain size and the interactions of fluids with rocks, two phenomena that weaken rocks, are efficient to localize deformation. This is rendered more complex by the role played by metamorphic transformations, for instance in subduction zones where the transformation of smectites into illites induces an unstable behaviour, typical of earthquakes. At larger scale, two main mechanisms lead to strain localization. The first one is changing conditions along plate boundaries or below plates, which imposes the creation of localized shear zones. Laterally changing subduction is very efficient to initiate the formation of large faults, such as the North Anatolian Fault, in the upper plate of subduction zones, and the asthenospheric mantle flowing underneath plates is also an efficient driver of deformation. The second mechanism is the intrinsic heterogeneity of the continental crust (including presence of magmas) and probably also, but to a lesser extent, of the mantle. At all scales this heterogeneity is a first-order localizing factor for deformation. The contrasts of resistance introduced by this heterogeneity is more efficient than most other small-scale localizing factors and it governs the deformation regime of mountain belts and rifts to a large extent. We have dated shear zones and we can now say that some of the large shear zones we have studied have been active during a period that is comprised between 7 and 15 Ma depending on the context. This is an important constraint to model the behaviour of deforming zones. The time constant of deformation processes is indeed a key to understand and model the mechanical behaviour of plate boundaries and deforming zones. The role played by mantle, for instance by mantle plumes when they interact with the lithosphere, has been enlighten through RHEOLITH. The mantle can no longer be considered a passive medium that accommodates the deformation of the continental crust. Instead we have shown that it plays a first order role in the deformation of rifts using high-resolution 3D numerical models and deeply penetrating seismic profiles provided by oil companies. The relative motion between the mantle and the crust drives some major deformations at the scale of continents. The two branches of the east African Rift can be explained by interaction of a plume with an heterogeneous continental lithosphere and the behaviour of subducted pieces of lithosphere when they retreat or are torn apart has a profound influence on the deformation history of the overriding plates of subduction zones. RHEOLITH has finally shown that the best possible test of numerical models is when they are able to reproduce not only the observed geometries but also the tectonic history on the long term and the succession of tectonic events through time.