On the rotation and frictional lock-up of normal faults: Explaining the dip distribution of normal fault earthquakes and resolving the low-angle normal fault paradox

Abstract

Classical rock mechanics predicts that normal faults should form at 60–65°, rotating to 30–40° before locking, so widespread slip at around 20° is considered paradoxical. Furthermore, the dip distribution of normal fault earthquakes has a distinct, unexplained peak at ~45°. For both problems, some combination of low friction, high fluid pressures, stress rotation and efficiency of low-angle slip have been suggested but provide at best a partial solution. A simple quantitative model for normal fault rotation (iterating between slip on faults and distributed strain) predicts that faults spend more time at lower angles. Combining this result with Mohr-Coulomb analysis of the range where seismogenic normal faults should lock up predicts the dip distribution of seismogenic normal faults and matches both the range and 45° peak observed for normal fault earthquakes. As even the lowest dips of normal fault ruptures fall within the limits of fault reactivation, slip on seismogenic low-angle normal faults is not paradoxical. A new Mohr-Griffith solution to the limits of fault reactivation extends the analysis into the top few km, where low-angle slip is best constrained by field observations, and shows that low cohesion normal faults can remain active at <20° without recourse to very low friction, high fluid pressures or stress rotation. This analysis thus resolves the long-standing paradox of slip on low-angle normal faults. Where continued rotation or changes in mechanical properties cause low-angle faults to lock in the shallow subsurface, a new fault propagates upward from the point of lock-up, transferring a hanging wall slice to the footwall, to be rafted up and out as a rider block. By recording the dip/depth of lock-up, riders atop detachments constrain the mechanics of faulting and of lock-up.