3D Ruptures Simulations Across Stepping Faults; Comparing the Slip Weakening and Rate-State Friction

Kayla A. Kroll, James H. Dieterich, Keith B. Richards-Dinger, & David D. Oglesby

Submitted August 15, 2018, SCEC Contribution #8737, 2018 SCEC Annual Meeting Poster #197

Large earthquakes commonly involve the rupture of two or more faults sections including cases where the ruptures jump across gaps between disconnected sections. Several large earthquakes including the 3 November 2002 Mw7.9 Denali, Alaska, the 4 April 2010 Mw7.2 El Mayor-Cucapah, Mexico, and the 14 November 2016 Mw7.8 Kaikoura, New Zealand events all ruptured numerous fault sections, including disconnected sections. Because the total rupture length is a primary determinant of earthquake magnitude, understanding how earthquake ruptures are able to jump across fault gaps is of critical importance for estimates of seismic hazard. For example, to develop a catalog of possible earthquake rupture events in the system of active faults in California, the UCERF3 forecasts employ a set of parametric rules to define the type of gaps earthquake rupture might jump. Here, we explore the effect of RSF on the rupture of parallel fault stepovers using homogeneous initial stress conditions. We compare our results to more traditional dynamic rupture models that employ SWF. Results are evaluated in terms of the slip distribution, re-nucleation location, and maximum jump distance. In this context, we also investigate the effect the normal stress change on the evolution of the state variable and effect of time-dependent nucleation on delayed rupture jumps. We perform this analysis for fault systems classified by high (S=0.49) and low (S=1.25) initial stress states. We show that results from RSQSim compare favorably with those from FaultMod. We find that the average slip is inversely related to offset in both simulations. Average source and receiver fault stress drops vary considerably with larger offsets. Receiver fault re-nucleation locations lie within lobes of increased static stress for extensional and compressional offsets. For models with low stresses, the maximum jump distance is ~0.5 km for both simulation methods. For models with high stresses, FaultMod simulations jump larger offsets (~3km) than those produced with RSQSim (~2 km).

Prepared by LLNL under Contract DE-AC52-07NA27344.

Key Words
RSQSim, stepovers, jumping ruptures, earthquake simulation, dynamic rupture

Citation
Kroll, K. A., Dieterich, J. H., Richards-Dinger, K. B., & Oglesby, D. D. (2018, 08). 3D Ruptures Simulations Across Stepping Faults; Comparing the Slip Weakening and Rate-State Friction. Poster Presentation at 2018 SCEC Annual Meeting.


Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)