Background

Rupture speed may have strong implications in seismic hazard (Das, 2007). Supershear rupture propagation (i.e. vr bigger than vs) along predominantly homogeneous faults radiate mach-waves caring potentially-destructive energy much farther than seismic waves radiated if rupture propagates with a subshear velocity (e.g. Bernard and Baumont, 2005; Dunham and Bhat, 2008). As a consequence, when vr > vs, radiated energy travels efficiently and induces important stress perturbations in the neighboring fault region and high frequency radiation into the far-field (Bizzarri and Spudich, 2008). This has been recently supported by the identification of specific aftershock patterns around supershear fault segments in real earthquakes (Bouchon and Karabulut, 2008). While those fault segments remains remarkably quiet after a mainshock, clusters of aftershocks on secondary structures are activated. The post-earthquake quiescence of supershear fault segments suggests that friction is relatively uniform there, whereas the activation of off-fault structures is explained by transient perturbations of mach-waves (Bouchon and Karabulut, 2008). These waves are thus carriers of valuable information about the stress-breakdown process and may reveal some friction parameters on specific fault regions (Cruz-Atienza and Olsen, 2009). Subshear ruptures do not shear this interesting advantage of supershear earthquakes, since high frequencies in the former case decrease too fast (Cruz-Atienza et al., 2009). 

The mechanical condition of faults is directly related to the energy partition during rupture and the way energy is radiated. Friction thus plays a major role and should differ among segments with different rupture speeds. Laboratory experiments with rocks and changes of the Gutenberg-Righter law b-value in the aftershock activity surrounding subshear and supershear faults suggest that supershear segments are characterized by low-friction levels (Amitrano and Schorlemmer, 2009). On the other hand, direct observations of supershear ruptures in real and laboratory earthquakes have been done mostly in the past few years (e.g. Archuleta, 1984; Rosakis et al., 1999; Bouchon et al., 2001; Bouchon et al., 2002; Bouchon and Vallée, 2003; Dunham and Archuleta, 2004; Ellsworth et al., 2004; Aagaard and Heaton, 2004; Xia et al., 2005). These observations demonstrate that supershear rupture transients in real earthquakes happen more often than expected by the traditional thought. For this reason, in this work we will perform an integrated analysis involving the seismic cycle (i.e. stick-slip loading), the statistics of foreshocks and aftershock, friction changes on the fault and their relationship with rupture speed, all that integrated in a 3D dynamic (i.e. spontaneous) rupture model.

Goals

1) Identify possible segmentation of seismicity patterns (e.g. b-value) along faults with heterogeneous friction properties along the seismic cycle (i.e. stick-slip and creeping behaviors spontaneously induces by tectonic loads).
2) Identify possible relationships between those friction-dependent patterns and rupture speed to explain observation in real earthquakes.

Methodology

To this purpose we will implement a 3D dynamic-rupture slip- and rate-dependent friction model subject to a constant-rate tectonic load. Such seismic-cycling model will spontaneously evolve producing micro-earthquakes (i.e. creeping-like processes), intermediate- and large-size ruptures with characteristic rupture properties, as hypocenter locations and rupture speeds. The numerical model will initially consist in a high-performance parallel finite difference SGSN code (Olsen et al, 2008; Dalguer and Day, 2007) coupled with an interface charged of reconstructing the initial condition for the next event from residual information of the last event (e.g. fault tractions). We will test both barrier and asperity models, as well as mixtures of these source models, using self-similar stochastic generators with specific correlation lengths that will mimic the fault roughness. All results will be statistically analyzed under the Gutenberg-Righter law perspective on the fault plane. We will further analyze the evolution of off-fault stresses and deformations during the seismic cycle to indentify possible triggering mechanisms associated with supershear fault segments.

Collaborators

David AMITRANO², Michel BOUCHON² and Denis LEGRAND<sup>1

1</sup>Departamento de Sismología, Instituto de Geofísica, UNAM
²Laboratoire de Géophysique Interne et Tectonophysique, UJF
 

References

 •    Amitrano, D. and D. Schorlemmer (2009). Friction of faults inferred from earthquake statistics. Submitted to Nature.

•    Bernard, P. and D. Baumont (2005). Shear Mach wave characterization for kinematic fault rupture models with constant supershear rupture velocity. Geophys. J. Int., 162, 431–447 doi: 10.1111/j.1365-246X.2005.02611.x.
•    Bizzarri, A. and P. Spudich (2008). Effects of supershear rupture speed on the high-frequency content of S waves investigated using spontaneous dynamic rupture models and isochrone theory. J. Geophy. Res., 113, B05304, doi:10.1029/2007JB005146.
•    Bouchon, M. and H. Karabulut (2008), The Aftershock Signature of Supershear Earthquakes. Science, 320, 1323–1325, DOI: 10.1126/science.1155030.
•    Cruz-Atienza V. M. and K. B. Olsen (2009). Supershear mach-waves expose fault friction. Submitted special issue on "Supershear Earthquakes", Elsevier, edit. S. Das and M. Bouchon.
•    Cruz-Atienza V. M., K. B. Olsen and L. A. Dalguer (2009). Estimation of the Breakdown Slip from Strong-Motion Seismograms: Insights from Numerical Experiments. Bull. Seismol. Soc. Am., 99, 6, 3454–3469, doi: 10.1785/0120080330.
•    Dalguer, L. and S. Day (2007). Staggered-grid split-node method for spontaneous rupture simulation. J. Geophy. Res., 112, B02302, doi:10.1029/2006JB004467.
•    Das, S. The Need to Study Speed (2007). Science, 317, 905-906.
•    Dunham, E. and H. Bhat (2008). Attenuation of radiated ground motion and stresses from three-dimensional supershear ruptures. J. Geophy. Res., 113, B08319, doi:10.1029/2007JB005182.
•    Olsen, K.B., S.M. Day, L.A. Dalguer, J. Mayhew, Y. Cui, J. Zhu, V.M. Cruz-Atienza, D. Roten, P. Maechling, T.H. Jordan, D. Okaya and A. Chourasia (2009). ShakeOut-D: Ground motion estimates using an ensemble of large earthquakes on the southern San Andreas fault with spontaneous rupture propagation, Geophysical Research Letters, 36, L04303, doi:10.1029/2008GL036832.