Dynamic path selection along branched faults: Experiments involving sub‐Rayleigh and supershear ruptures

Carl‐Ernst Rousseau,Ares J Rosakis

doi:10.1029/2008jb006173

Abstract

Building upon previous laboratory earthquake experiments of dynamic shear rupture growth taking place along faults with simple kinks, new and complex fault geometries involving cohesively held fault branches are studied. Asymmetric impact at the specimen boundaries controls the incoming shear ruptures, which are manipulated to propagate at either sub‐Rayleigh or supershear velocities. High‐speed photography and dynamic photoelasticity are used with a model material, Homalite‐100, to monitor incoming and outgoing rupture propagation, acceleration, deceleration, or arrest at the vicinity of the branch location. Differences and similarities of rupture velocity history between cases involving faults with either simple kinks or branches, on the one hand, and sub‐Rayleigh and supershear incoming ruptures, on the other, are highlighted and explained. Results of the experiments show a clear general bias toward large branch inclination, smaller branch angles appearing to be overshadowed and suppressed by the stress field associated with the main fault. Of great interest, also, is the sustenance of rupture propagation along a branch by the Mach cone, when the initial rupture is supershear driven. Generally, higher rupture speeds favors larger arrays of branching angles to be triggered. A companion analysis by Templeton et al. (2009) featuring detailed numerical simulations of these experiments provides further insight into the observed phenomena.

Highlights

[1] Building upon previous laboratory earthquake experiments of dynamic shear rupture growth taking place along faults with simple kinks, new and complex fault geometries involving cohesively held fault branches are studied
[2] Dynamic interfacial fracture is generally identified as one of the primary causes of failure in welded, bonded, or otherwise joined structures that are subjected to impulse loading
The dynamic fracture theories, which are developed for the study of such engineering structures and are validated in the laboratory level can, in principle, be transferred to much larger scales pertaining to natural earthquake events

Summary

Introduction

[2] Dynamic interfacial fracture is generally identified as one of the primary causes of failure in welded, bonded, or otherwise joined structures that are subjected to impulse loading. [8] Motivated by such theoretical and numerical works, Rousseau and Rosakis [2003] initiated experimental studies of fault bends from which they developed models to describe the behavior of a propagating rupture as it travels along a non planar path These experiments featured incoming shear ruptures propagating at various speeds along cohesively bonded interfaces. The existence of a path of lower fracture toughness makes these specimens fracturewise inhomogeneous but does not, in any sense, affect their continuum mechanics description This is the condition that prevails across young earthquake faults, which are essentially preferential paths of lower toughness relative to the Earth’s crust that trap ruptures, compelling them to proceed only along their prescribed routes [Rosakis, 2002]. The magnitude of maximum in-plane shear stresses in the material is given by the following relation [Dally and Riley, 1991]: gauge attached to the steel buffer, upon being impacted, triggers the camera to begin recording the event

Experimental Observations

Findings