DOES LONG-TERM FAULT MEMORY CONTRIBUTE TO THE SPACE-TIME VARIABILITY OF CONTINENTAL INTRAPLATE SEISMICITY?

Large earthquakes in continental interiors show complex space-time variability quite different from behavior at plate boundaries. Fault zones tend to have clusters of earthquakes separated by long quiescent periods, and seismicity seems to roam between faults. As a result, traditional concepts like recurrence intervals and seismic gaps can be of only limited use, posing major challenges for hazard assessment. Although the mechanics involved are unclear, it seems likely to involve interactions between fault systems that jointly accommodate slow tectonic loading. Here, we explore the possible contribution of another process that we call Long-Term Fault Memory (LTFM). LFTM is a modified earthquake cycle model where the probability of an earthquake increases with time until one happens, after which the probability decreases, but not necessarily to zero. In this way, the probability can remain higher than the long-term average and clusters of events can result. We have used LFTM to simulate paleoearthquake records from Pallett Creek on the San Andreas Fault and the Cascadia Subduction Zone. Here we apply LTFM to records from continental interiors, including a particularly long record from the Cadell fault in intraplate Australia. We will explore similarities and differences between the application of LTFM to plate boundaries and intraplate regions.