Subsections

Locating Nonvolcanic Tremors Beneath the San Andreas Fault Using a Station-pair Double-Difference Location Method

Robert M. Nadeau, Haijiang Zhang (Massachusetts Institute of Technology), and Nafi Toksoz (Massachusetts Institute of Technology)

Introduction

It has been a challenging task to locate nonvolcanic tremors because of their lack of impulsive wave arrivals. To help overcome these difficulties, we have developed a station-pair double-difference (DD) location method to determine absolute tremor locations by directly using the station-pair travel time differences measured from cross-correlating tremor waveform envelopes. To account for velocity model inaccuracy, multiple tremors are located together to invert for station corrections. The new method is applied to tremors in the Parkfield region of central California occurring between 27 July 2001 and 21 February 2009, and the resulting locations are compared to the catalog locations of Nadeau and Guilhem (2009). The comparison shows that the new tremor locations more clearly delineate the spatial and temporal distribution of tremor activity in the area and improve our understanding of tremor origin and process.

Station-pair double-difference location method

The DD location method, developed by Waldhauser and Ellsworth (2000), has been widely used to locate earthquakes using differential arrival times at common stations from pairs of `events'. We use the same concept applied to the case where differential times on pairs of `stations' from common events are accurately calculated (using the station-pair differential travel times directly to locate the tremor events). In addition, because station corrections are included in our inversion, our approach is a `multiple-event' location method. Our method uses station-pair cross-correlation delay times measured from tremor waveform envelopes between different stations. It should also be applicable for locating low frequency earthquakes (LFEs) when similar travel time difference information is available. We applied our location method to station-pair delay time measurements from nonvolcanic tremors occurring beneath the San Andreas Fault around the Parkfield and Cholame area (originally detected and located by Nadeau and Guilhem (2009)). More detail on the station-pair DD method can be found in Zhang et al. (2010).

Results and Discussion

Figure 2.48 compares station-pair DD locations with original catalog locations (Nadeau and Guilhem, 2010). The DD located tremors are shifted northeast and deeper (Figure 2.48C and D), relative to the original catalog. The average shift is 3.4 km in X and 3.7 km in depth. The northeast shift may be due to bias introduced by our use of a 1-D velocity model in this region of strong lateral velocity contrast. The shift in depth is likely due to the station-pair DD location method's more accurate determination of tremor depths, which avoids the coupling effect between depth and origin time. The DD locations are also more clustered, and more substructure can be seen than with the catalog locations (Figure 2.48G and H). Overall, the location uncertainties of the station-pair DD tremor locations are about half those of the catalog tremors (Figure 2.48G and H; also see Auxiliary Material in (Zhang et al., 2010)).

Figure 2.48E and F show histograms of tremor depths from the catalog and station-pair DD locations. In this area, earthquakes locate in the upper $\sim$15 km of the Earth's crust, and the Moho depth is $\sim$25 km. Hence, our results suggest that tremors predominate below the seismogenic zone, in the ductile lower crust, and that a distinct gap in depth between the seismogenic and tremor zones exists.

Nadeau and Guilhem (2009) also observed periodic episodes of tremors at Cholame after the 2004 Parkfield M6 earthquake that where concentrated and more periodic in the southwestern portion of the tremor zone. Our results confirm these observations (see Figures 2c and d of Zhang et al., 2010), and they indicate that the process generating tremors in this area may be structurally controlled and vary across the SAF.

Acknowledgements

Supported through USGS grants 06HQGR0167, 07HQAG0014, 08HQGR0100, and 08HQGR0101; DOE contract DE-FC52-06NA27325; NSF grants EAR-0537641 and EAR-0544730; and Chinese government executive program SinoProbe-02.

References

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

Waldhauser, F. and W.L. Ellsworth, A double-difference earthquake location algorithm: method and application to the Northern Hayward Fault, California, Bull. Seism. Soc. Am. 90, 1353-1368, 2000.

Zhang, H., R.M. Nadeau and N. Toksoz, Locating nonvolcanic tremors beneath the San Andreas Fault using a station-pair double-difference location method, Geophys. Res. Lett., 37, L13304, doi:10.1029/2010GL043577, 2010.

Figure 2.48: Comparison of catalog and station-pair DD tremor locations. (A) Map view of 1246 tremors (red/gray dots) and 64 stations (triangles) used in this study. Tremor catalog locations are from Nadeau and Guilhem (2009). The Cartesian coordinate system used in location is shown with a tick (marked as `+') interval of 10 km on X and Y axes. Faults are shown as black lines. Parkfield and Cholame are marked as white dots. (B) Map view of new tremor locations determined by station-pair DD location method. Station corrections in seconds are also shown. (C) Catalog tremor locations (blue/dark-gray) are shown in X-depth and Y-depth sections. (D) DD tremor locations are shown in X-depth and Y-depth sections. (E) Depth distribution of catalog tremor locations. (F) Depth distribution of DD tremor locations. (G) Across-fault cross sections of catalog tremor locations and the associated 95% uncertainty bounds at Y=34 and 40 km (within 1 km of both sides of the section). (H) The same as G but for DD tremor locations. In both panels (C) and (D), background seismicity from Thurber et al. (2006) is shown as light gray dots and the inferred LFE locations of Shelly (2009) are shown as the near horizontal red/light-gray dots at $\sim$25 km depth.
\begin{figure*}\begin{center}
\epsfig{file=nadeau10_1_1.ps, width=10cm}\end{center}\end{figure*}

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