Spacing and strength of active continental strike-slip faults

Abstract

Parallel and evenly-spaced active strike-slip faults occur widely in nature across diverse tectonic settings. Despite their common existence, the fundamental question of what controls fault spacing remains unanswered. Here we present a mechanical model for the generation of parallel strike-slip faults that relates fault spacing to the following parameters: (1) brittle-crust thickness, (2) fault strength, (3) crustal strength, and (4) crustal stress state. Scaled analogue experiments using dry sand, dry crushed walnut shells, and viscous putty were employed to test the key assumptions of our quantitative model. The physical models demonstrate that fault spacing (S) is linearly proportional to brittle-layer thickness (h), both in experiments with only brittle materials and in two-layer trials involving dry sand overlying viscous putty. The S/h slope in the two-layer sand–putty experiments may be controlled by the (1) rheological/geometric properties of the viscous layer, (2) effects of distributed basal loading caused by the viscous shear of the putty layer, and/or (3) frictional interaction at the sand–putty interface (i.e., coupling between the viscous and brittle layers). We tentatively suggest that this third effect exerts the strongest control on fault spacing in the analogue experiments. By applying our quantitative model to crustal-scale strike-slip faults using fault spacing and the seismogenic-zone thickness obtained from high-resolution earthquake-location data, we estimate absolute fault friction of active strike-slip faults in Asia and along the San Andreas fault system in California. We show that the average friction coefficient of strike-slip faults in the India–Asia collisional orogen is lower than that of faults in the San Andreas fault system. Weaker faults explain why deformation penetrates >3500 km into Asia from the Himalaya and why the interior of Asia is prone to large (M>7.0) devastating earthquakes along major intra-continental strike-slip faults. Our new approach of estimating absolute fault strength may be useful in future studies of continental deformation and earthquake mechanics.