2019 AAPG Annual Convention and Exhibition:

Datapages, Inc.Print this page

Faulting Events: Moving Beyond Frictional Thinking


Many active faults in the modern world move in a spasmodic manner, via discrete events of slip (whether these are seismic or not). By inference, similar episodic events must apply to some, perhaps much, of the faulting in the geological record. Friction-based stability arguments are normally used to assess the initiation of a slip event, and concepts for rate-dependent reductions of slip-zone resistance may be important in explaining the continuation of slip motion in each event. But neither of these approaches explains why a faulting event stops, yielding a finite amount of slip. And, by themselves, neither friction-like idea explains why faulting events are repetitive, thus allowing the accumulation of slip magnitudes that are larger than the size of any conceivable single event. We examine a key concept that is adopted in the existing approaches, namely that the stress state around the fault is fixed. We argue that this cannot be true during any type of slip event, whether the slip occurs by frictional processes or via gouge-zone rate-dependent hyper-plasticity. The stress state cannot be uniform for the initiation of the event, nor can it be uniform afterwards, which also means that the stress state is not uniform for the initiation of a subsequent event. The arguments in support of these conjectures extend, from a quasi-static energy-budget analysis, to full process simulations. These methods reveal that during fault movement, the normal stress on the fault is reduced from its pre-faulting value, across some of the moving fault surface, due to relaxation of the fault-parallel component of the elastic-strain state and consequent Poisson effects on the normal component. At larger scale, a form of arching occurs, ‘locking’ the fault in places while the slip occurs on the part of the fault under the arch. Numerical simulations are able to explore the consequences of these changes in terms of the development of non-elastic strains and associated permanent alterations of the fault-rock properties. The simulations develop emergent patterns of permanent strains that closely resemble those observed in experimental studies of shear, and in fault outcrops. This process understanding leads to the conclusion that a faulting event is dependent on a prior heterogeneous accumulation of elastic strain (and thus stress change) to create a dilated region along a part of the fault into which the strain-energy of an adjacent compacted region is transferred during the slip event. Slip magnitudes are dependent on the amount of dilation that can accumulate before a slip event occurs.