Mechanical and Microstructural Evolution of Ductile Shear Zones: Implications for the Deep Structure of Lithospheric Faults

We offer three new concepts that help place constraints on the mechanics and width of plate-boundary shear zones below the brittle-ductile transition. 1. Lithospheric shear zones operate at approximately constant stress at any given depth (temperature). This is because shear zones form by microstructural changes that cause weakening and hence strain localization. These changes occur when the ambient stress reaches the yield strength σy of intact rock. As a result, the cumulative width w of shear zones reaches a value such that they can accommodate plate motion at a flow stress equal to σy. 2. Exhuming shear zones preserve a record of the stress-temperature profile through the deforming crust. Increasing strain localization as the rocks cool, and quenching of the microstructure outside the narrowing shear zone, allow preservation of the microstructure and mineral chemistry at various stages in their evolution. If the flow stress in the shear zone at any depth is a measure of the yield strength of the surrounding rock, we can use this information to construct strength-depth profiles through the lithosphere. 3. Dislocation creep causes dynamic recrystallization and grainsize reduction. This may result in a switch to grainsize-sensitive creep, which is the main cause of weakening and strain localization in shear zones. At constant strain-rate, this results in a stress drop, which may be followed by grain growth, preventing a permanent switch in mechanism. If shear zones operate at constant stress, however, dislocation density in the deforming grains remains the same after the switch, so that dynamic recrystallization and grain-boundary migration driven by dislocation strain energy continue at the same rate as before. This inhibits grain growth driven by surface energy, so that the deformation mechanism switch is permanent. We calculate shear zone widths at depth in the lithosphere based on these concepts. We use a stress-temperature profile obtained from the Whipple Mountains, California, to constrain the strength of the ductile quartz-rich middle crust, and we estimate stress-T-width profiles through the lower crust and upper mantle using published flow laws. We conclude that the cumulative width of shear zones making up the San Andreas Transform is likely to reach about 6 km in the mid-crust, narrows to a few tens of meters in feldspar-dominated lower crust and olivine-dominated uppermost mantle, and widens to 80 km at 45 km depth. Below about 55 km, we predict that weakening mechanisms do not operate, and hence will not contribute to strain localization.