Theoretical studies indicate (e.g., Weertman, 1980; Ben-Zion and Andrews, 1998; Ben-Zion 2001; Ampuero and Ben-Zion 2008; Brietzke et al. 2009) that ruptures on bimaterial faults that separate different lithologies have larger slip-velocity and slip in the direction of particle motion in the compliant solid (referred to as the preferred direction). This can have significant effects on the seismic shaking hazard at major metropolitan areas (e.g., Los Angeles, San Francisco, Istanbul) near large bimaterial faults (e.g., the San Andreas and North Anatolian faults). The asymmetric along-strike rupture behavior on bimaterial faults is expected to produce enhanced dynamic triggering in the preferred propagation direction (Figure 1). This is also expected leads to strongly asymmetric damage across the fault, with more damage on the stiffer side (Ben-Zion and Shi 2005). Analyses of seismic fault zone waves and geological signals in the structures of the San Andreas and San Jacinto faults indicated damaged fault zone layers that are asymmetric as expected for bimaterial ruptures in those fault zones (Lewis et al. 2005, 2007; Dor et al. 2006a,b; Wechsler et al. 2009). Some evidence for asymmetric dynamic triggering on the Northern San Andreas fault has been documented by Rubin and Gillard (2000) and Schorlemmer and Ben-Zion (2008). Here we examined the along-strike symmetry properties of seismicity along many individual fault zones in California.

Ilya ZaliapinDepartment of Mathematics and Statistics

University of Nevada Reno

Introduction and motivation Clarifying the genuine properties of seismicity patterns remains an extremely challenging problem because of the inherent complexity of the earthquake process combined with the limited and noisy available data. Purely statistical studies tend to analyze seismicity associated with large spatial domains to increase the amount of data. This approach, however, can mix different populations of earthquakes (e.g., those on large plate-bounding faults vs. small intra-plate faults or seismicity on cold vs. hot regions). Here we attempt to increase the information content of the available data by establishing correlations between spatio-temporal patterns of seismicity and geophysical properties of the crust. In particular, we examine the relations between asymmetric dynamic triggering patterns along faults and velocity structure images.

Figure 1: Strain Model Contour plot of second invariant of strain rate model reflecting secular deformation constrained by GPS velocities and Quaternary fault slip rates (used faults shown by thin lines).

Annual MeetingSeptember 12-16, 2009

Palm Springs, CA

Figure 2: Fault zones. Indexes in squares indicate aftershock-dominated zones; they are not analyzed. After Powers (2009).

Figure 1. Particle velocities at a given time generated by a rupture pulse (small red bar in the white box on the left, shown in larger detail on the right) propagating to the right along a right-lateral strike-slip fault (thin red horizontal line) between different elastic solids. The rupture produces considerably larger seismic radiation to the right [Ben-Zion, 2001].

Figure 4: Observed values of the ratio and confidence regions according to a truncated Pareto model.

Yehuda Ben-ZionDepartment of Earth Sciences

University of Southern California

Summary We attempt to establish correlations between spatio-temporal patterns of seismicity and geophysical properties of the crust. In this study, we examine the relations between asymmetric dynamic triggering patterns along faults and velocity structure images. We use the regional southern California (CA) catalog of Lin et al. (2007) and the catalogs of Power and Jordan (2009) for various specific fault-zones in CA. The analyses are based on the earthquake clustering technique of Zaliapin et al. (2008) that employs the Baiesi-Paczuski (2004) distance between earthquakes and allows one to distinguish between the clustered and homogeneous parts of an earthquake catalog. The initial results indicate the presence of asymmetric triggering in early-time close aftershocks along the the San Andreas and San Jacinto faults and absence of asymmetry for the Garlock fault.

Methodology (1) Catalog: We analyze seismicity located along well-defines faults, which makes the interpretation of results particularly straightforward. Specifically, we work with 52 seismic zones defined by Powers (2009), and only analyze the earthquake location along respective faults (Figure 2).

(2) Clustering: Triggering analysis of seismicity is commonly complicated by the problem of associating events with possible predecessors (mainshocks or parents). We estimate seismic clusters using the method of Zaliapin et al. (2008). In this method, the distance ij between event i and a later event j is measured by the Baiesi-Paczuski metric

where tij and rij are time and spatial distance between events and mi is the magnitude of the first one. We further define the normalized distance R and time between events as

Here R measures distance in mainshock’s rupture lengths. The joint 2D distribution of the normalized times and distances reveals earthquake clusters (Figure 3). For each cluster that consists of a mainshock (parent) and n aftershocks (descendants) we define the asymmetry index

Large positive (negative) values of IA indicate that aftershocks tend to happen with positive (negative) displacement relative to the respective mainshock. In our approach triggering is only estimated within statistically significant clusters.

10 ibmij ij ijt r

(1 )10 ; 10 ; / 0.0152; 0.42i ip bm pbmij ijt r d R r p

Results

Figure 4: Asymmetry results for different fault zones (zone index is from Figure 2).

Zone 1: No asymmetry detected Zone 51: NW (positive) asymmetryZone 5: SE (negative) asymmetry

Panel a: Each point represents a single descendant event; parents are stacked at

the origin. Red points – clustered seismicity, which is used for triggering analysis.

Panel b: Each point represents a single cluster, formed by some of the red

points in panel (a)

Panel c: Time-space map of a selected cluster from panel (b) (see black arrows). Each point represents an earthquake; red dot – parent event, blue

dots – descendants used to compute the asymmetry index. The Y-axis shows the distance in km along the fault zone shown in Figure 2.

Panel a: see details in Zone 1 part

Panel b: see details in Zone 1 part

Panel a: see details in Zone 1 part

Panel b: see details in Zone 1 part

3-sigma error bars

Mean asymmetry index averaged over 496 different choices of clusters (varying R and thresholds that define red points in panels (a) above)

Empirical 10% and 90% quantiles

Zone 1: The lack of significant asymmetry is consistent with lack of pronounced velocity contrast across

the Garlock fault.

Zone 5: The “negative” asymmetry is consistent with evidence [e.g., Fuis et al., 2001] for “reversed” velocity contrast along the Mojave section of the SAF (with the NE block having faster velocities) and observed asymmetric rock damage in that section [Dor et al., 2006a,b]

Zone 11: The “negative” asymmetry along the San Jacinto fault is consistent with seismic

imaging of the velocity contrast there [Scott et al. 1994] and asymmetric rock damage based on

trapped waves [Lewis et al., 2005] and geological signals [Dor et al., 2006a; Wechsler et al. 2009 ]

Zone 51: The “positive” asymmetry is consistent with seismic imaging on the velocity contrast along the Parkfield section of the SAF [e.g.,

Thurber et al., 2006; Ben-Zion et al., 1992]. The results in zone 48 to the left are consistent with

similar findings of Rubin and Gillard [2000].

Zone 15: The “positive” asymmetry is consistent with seismic imaging on the

velocity contrast along the southern SAF [SCEC velocity model].

A

mean( )1

std( )

RI n

R

Figure 3: Three examples of asymmetry analysis: positive (zone 51) and negative (zone 5) asymmetry and no asymmetry (zone 1).

Zone 5: Negative triggering Zone 5: Negative triggering Zone 51: Positive triggering

Zone 51: Positive triggering

REFERENCES

[1] Ampuero, J.-P. and Y. Ben-Zion, 2008, Cracks, pulses and macroscopic asymmetry of dynamic rupture on a bimaterial interface with velocity-weakening friction, Geophys. J. Int., 173, 674–692, doi: 10.1111/j.1365-246X.2008.03736.x.

[2] Baiesi, M and M. Paczuski, 2004, Scale-free networks of earthquakes and aftershocks. Phys. Rev. E, 69, 066106.

[3] Ben-Zion, Y., 2001, Dynamic Rupture in Recent Models of Earthquake Faults, J. Mech. Phys. Solids, 49, 2209-2244.

[4] Ben-Zion, Y. and D. J. Andrews, 1998, Properties and Implications of Dynamic Rupture Along a Material Interface, Bull. Seism. Soc. Am., 88, 1085-1094.

[5] Ben-Zion, Y., S. Katz and P. Leary, Joint inversion of fault zone head waves and direct P arrivals for crustal structure near major faults, J. Geophys. Res, 97, 1943-1951, 1992.

[6] Brietzke, G.B., Cochard, A. and Igel, H., 2009, Importance of bimaterial interfaces for earthquake dynamics and strong ground motion, Geophys. J. Int., 178, 921-938.

[7] Dor O., T. K. Rockwell and Y. Ben-Zion, 2006a, Geologic observations of damage asymmetry in the structure of the San Jacinto, San Andreas and Punchbowl faults in southern California: A possible indicator for preferred rupture propagation direction, Pure Appl. Geophys., 163, 301-349.

[8] Dor O., Y. Ben-Zion, T. K. Rockwell and J. Brune, 2006b, Pulverized Rocks in the Mojave section of the San Andreas Fault Zone, Earth Planet. Sci. Lett., 245, 642-654.

[9] Fuis, G.S., Ryberg, T., Godfrey, N., Okaya, D.A., and Murphy, J.M., 2001, Crustal structure and tectonics from the Los Angeles basin to the Mojave Desert, southern CA. Geology, 29, 15-18.

[10] Lewis, M.A, Z. Peng, Y. Ben-Zion and F. Vernon, 2005, Shallow seismic trapping structure in the San Jacinto fault zone, Geophys. J. Int., 162, 867–881, doi:10.1111/j.1365-246X.2005.02684.x.

[11] Lewis, M.A, Y. Ben-Zion and J. McGuire, 2007, Imaging the deep structure of the San Andreas Fault south of Hollister with joint analysis of fault-zone head and direct P arrivals, Geophys. J. Int., 169, 1028–1042.

This research is supported by SCEC 2009-2010 project

“Correlation between seismic clustering properties

and regional physical conditions”

[18] Wechsler, N., T. K. Rockwell and Y. Ben-Zion, 2009, Analysis of rock damage asymmetry from geomorphic signals along the trifurcation area of the San-Jacinto Fault, Geomorphology, doi:10.1016/j.geomorph.2009.06.007.

[19] Weertman, J., 1980, Unstable slippage across a fault that separates elastic media of different elastic constants, J. Geophys. Res., 85, 1455-1461.

[20] Zaliapin, I., A. Gabrielov, H. Wong, and V. Keilis-Borok, 2008, Clustering analysis of seismicity and aftershock identification, Phys. Rev. Lett., 101.

[12] Lin, G., P. Shearer, and E. Hauksson, 2007, Applying a three-dimensional velocity model, waveform cross-correlation, and cluster analysis to locate southern California seismicity from 1981 to 2005, J. Geophys. Res. 112, B12309.

[13] Powers, P. M. and T. H. Jordan, 2009, Distribution of Seismicity Across Strike-Slip Faults in California, J. Geophys. Res, in review.

[14] Power, P. M., 2009, SEISMICITY DISTRIBUTION NEAR STRIKE-SLIP FAULTS IN CALIFORNIA, PhD thesis, University of Southern California.

[15] Rubin, A. and D. Gillard, 2000, Aftershock asymmetry/rupture directivity along central San Andreas fault microearthquakes, J. Geophys. Res., 105, 19,095-19,109.

[16] Schorlemmer, D. and Y. and Ben-Zion, 2008, Directivity effects of fault velocity contrast on triggered seismicity, Seism. Res. Lett., 79 (2), 295.

[17] Thurber, C.H., Zhang, H., Waldhauser, F., Hardebeck, J., Michael, A. & Eberhart-Phillips, D., 2006, Three-dimensional compressional wavespeed model, earthquake relocations, and focal mechanisms for the Parkfield, California, region, Bull. Seismol. Soc. Am., 96(4B), S38-S49, doi: 10.1785/0120050825.