Subsections

Nonvolcanic Tremor Evolution in the Parkfield, CA Region

Robert M. Nadeau and Aurélie Guilhem

Introduction

Nonvolcanic tremors have generally been observed in transition zones between freely slipping and locked faults, and in many locations, tremor activity is also seen to increase with detectable transient fault zone deformation (slow-slip events) (Liu and Rice, 2007). These associations suggest that tremor activity is closely related to the processes responsible for generating earthquakes. We analyzed tremor activity in the Parkfield region of California between 27 July 2001 and 21 February 2009, during a period when two strong earthquakes occurred in the region: the 6.5$M_{w}$ San Simeon earthquake in 2003 and the 6.0$M_{w}$ Parkfield earthquake in 2004 (Figure 2.24).

Figure 2.24: Study region with 1250 well-located tremors (black dots). Thirty kilometer square boxes (black) define the Monarch Peak (MP) and Cholame tremor zones. Color contours give regional shear-stress change at 20 km depth from the Parkfield earthquake (green segment) along the San Andreas Fault (SAF). The thrust type San Simeon earthquake rupture is represented by the grey rectangle and hatched line labeled SS. The currently locked Cholame segment is $\sim$63 km long (solid portion of arrow) and is believed capable of rupturing on its own in an $\sim$M7 earthquake. The grey lines within the Cholame box bound the west quadrant where quasi-periodic episodes predominate. White star indicates the epicenter and gray stars the foreshocks of the 1857 Ft. Tejon earthquake.
\begin{figure}\centering\epsfig{file=nadeau09_1_1.eps, width=8cm}\end{figure}

Detection, Size, Location

Using borehole seismic data from the High Resolution Seismic Network (HRSN) at Parkfield, CA, 2198 tremor events (12,547 minutes of activity) were detected during the period (Figure 2.25), with event activity lasting from 3 to nearly 21 minutes. Significant activity also occurred below the detection threshold. The average seismic energy released during the tremors was typically equivalent to a 0.5$M_{w}$ earthquake, and energies ranged over 1.5 $M_{w}$ units. Data from five seismographic networks (BP, BK, NC, CI, and TA) were combined to locate the tremors (Figure 2.25). Approximately 90% of the tremors occurred adjacent to the locked Cholame segment of the San Andreas fault (SAF). An additional 5-10% occurred $\sim$65 km to the northwest in the vicinity of Monarch Peak, CA. Tremors in both zones occurred at between $\sim$15 and 30 km depths.

Evolution

A sudden increase and subsequent decay of tremor activity (aftertremors) began immediately after the 2003 San Simeon and 2004 Parkfield earthquakes (Figure 2.25A). In addition, approximately 3 weeks before the Parkfield earthquake, an unusually strong episode of activity (foretremor) lasting $\sim$5 days also occurred (Nadeau and Guilhem, 2009).

The decay of tremor activity after the Parkfield earthquake continued until mid-2006 when annual rates began again to increase, reaching $\sim$181% of the pre-San Simeon level during the final year of our analysis (Figure 2.25A). Before the Parkfield earthquake, tremor activity in the Monarch Peak tremor zone (Figure 2.24) was low (Figure 2.25B). However, beginning $\sim$10 days after the Parkfield event, activity in this zone increased significantly and has continued at an elevated rate. After the Parkfield earthquake, persistent episodes of quasi-periodic tremor also emerged that are reminiscent of episodic tremor observed in some subduction zones (Figure 2.25A) (Rogers and Dragert, 2003).

The episodic component of the tremors are most persistent and regular in the western quadrant of the Cholame tremor zone that is bounded by the SAF to the northeast and by the seismic-aseismic transition into the locked Cholame segment to the southeast (Figures 2.24, 2.25C and D,). This localization of tremor behavior suggests that there are differences in the process generating tremor in different subregions of the SAF and that tremor in the Cholame zone is distributed both normal-to and along-strike of the SAF at depth.

Figure 2.25: Time histories of tremor activity. (A) Fifteen-day (grey) and 45-day (black) smoothed rate histories for all tremors in the study area. Times of the San Simeon and Parkfield earthquakes are SS and PF, respectively. Intense foretremor (Fore-Tremor) activity occurred $\sim$3 weeks before PF. Fifteen- and 45-day peak values just after PF are 31.1 and 18.9 minutes/day, respectively. White bars are interpreted times of quasi-periodic episodes. Label ? is time of a weak or missing episode. (B) History of tremor in the Monarch Peak zone (45-day smoothed). (C) History since 2005 of activity (5-day smoothed) in the western quadrant of the Cholame zone (Figure 2.24) (423 tremors; 2835 cumulative minutes of activity) showing strong episodic behavior. (D) Same as (C) for Cholame tremors outside the west quadrant (416 tremors; 2423 cumulative minutes) showing significantly less episodic behavior.
\begin{figure}\centering\epsfig{file=nadeau09_1_2.eps, width=8cm}\end{figure}

Stress Change

Static Coulomb- and shear-stress changes in the tremor zones from the 2003 San Simeon and 2004 Parkfield earthquakes on SAF aligned planes were small (6 to 14 kPa), but increases in tremor rates correspond with the increases in Coulomb- and shear- stresses. Static normal-stress changes from the earthquakes do not clearly correspond to the tremor rate changes, suggesting that normal-stress change may not play a significant role in stimulating the SAF tremor. The Coulomb stress changes that stimulated the SAF tremors are roughly an order of magnitude smaller than those typically reported for the triggering of earthquakes (King et al., 1994). This suggests that tremors are a more sensitive indicator of stress change than earthquakes.

Implications

The periodic tremor and persistent elevated activity after the Parkfield quake are not consistent with expectations of exponentially decaying post-seismic stress after an earthquake (Savage and Langbein, 2008). Because the SAF tremor changes have persisted for over 4 years since Parkfield, they could be signaling a shift in the process of deformation and stress accumulation beneath this hazardous portion of the SAF. Faulting from the great $M_{w}$ 7.8 1857 Ft. Tejon earthquake appears to have propagated from the Monarch Peak area southeastward along the SAF for $\sim$350 km, through Parkfield and the Cholame tremor zone (Toppozada et al., 2002) (Figure 2.24). The Cholame segment is now fully locked and last ruptured 152 years ago as part of the great 1857 event.

Acknowledgements

The HRSN is funded by USGS grant 07HQAG0014. Research was supported was by USGS grants 06HQGR0167 and 08HQGR0100, and by NSF grants EAR0537641 and EAR0544730.

References

King, G.C.P., R.S. Stein and J. Lin, Static Stress Changes and the Triggering of Earthquakes, Bull. Seism. Soc. Am., 84, 935-953, 1994.

Liu, Y. and J.R. Rice, Spontaneous and triggered aseismic deformation transients in a subduction fault model, J. Geophys. Res., 112, B09404, 2007.

Nadeau, R.M. and A. Guilhem, Nonvolcanic Tremor Evolution and the San Simeon and Parkfield, California, Earthquakes, Science, 325, 191-193, 2009.

Rogers, G. and H. Dragert, Episodic Tremor and Slip on the Cascadia Subduction Zone: The Chatter of Silent Slip, Science, 300, 1942-1943, 2003.

Savage, J.C. and J. Langbein, Postearthquake relaxation after the 2004 M6 Parkfield, California, earthquake and rate-and-state friction, J. Geophys. Res., 113, B10407, 2008.

Toppozada, T.R., D.M. Branum, M.S. Reichle and C.L. Hallstrom, San Andreas Fault Zone, California: M $\geq$5.5 Earthquake History, Bull. Seism. Soc. Am., 92, 2555-2601, 2002.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail: www@seismo.berkeley.edu
© 2007, The Regents of the University of California