Temporal Variations in Crustal Scattering Structure near Parkfield, California, from Receiver Functions

Pascal Audet


The accurate determination of crustal velocity structure in the region surrounding a fault is an essential component in the investigation of fault processes since it yields important information on the composition and state of the crust (e.g. anisotropy, pore-fluid pressure, etc). In addition, temporal variations in crustal architecture (i.e. 4-D imaging) can provide key constraints on the dynamics of faulting. One method that provides accurate point measurements of crustal velocity structure is based on the characterization of the scattering structure beneath a recording station using teleseismic events, i.e. the so-called receiver function method. The technique is based on the deconvolution of the source time function, approximated by the P-wave train, from the S-components of motion. The resulting seismograms represent an approximation to the Earth's Green function, and are used to determine the depth and velocity of discrete crustal layers. A novel use of the receiver function method is proposed here which makes use of decade-long high-quality records from permanent broadband stations to estimate temporal variations in crustal scattering structure.

4-D imaging using receiver functions

The conventional receiver function method has been applied to data from the Parkfield broadband seismic station PKD, and results demonstrate the existence of a lower crustal, low-velocity layer with strong ($\sim$15%) anisotropy (Ozacar and Zandt, 2009). Using 12 years of data at station PKD ($\sim$1000 events, Figure 2.15), we calculate the power spectral density (PSD) for each individual, unfiltered receiver function and bin into 12 month-long, 90% overlapping segments within which we calculate the median PSD and total power by integrating the PSD. Results show a clear change in PSD and spectral power after the 2003 $M_{w}$ 6.5 San Simeon earthquake with a subsequent decrease ($\sim$2 dB) in the post-seismic background PSD level (Figure 2.16). Careful investigation of both event distribution and instrument response functions estimated from noise records reveal that the change is not due to uneven source distribution or instrumental bias. Moreover, this change is observed only for events coming from the back-azimuth range 0$^{\circ}$-200$^{\circ}$ (dominated by events originating from South-Central America), which sample the crust along strike and southwest of the San Andreas Fault near Parkfield (Figure 2.15).

Figure 2.15: Moho piercing points (grey and black dots) of receiver functions around station PKD near Parkfield, California, with respect to study region (inset). Events from eastern azimuths (0$^{\circ}$-200$^{\circ}$- black dots) sample the crustal structure along the strike of the San Andreas Fault.
\epsfig{, width=8cm}\end{center}\end{figure}

Figure 2.16: Temporal variations in receiver functions at station PKD for back-azimuths 0$^{\circ}$-200$^{\circ}$. Top panels show power-spectral density (PSD) of receiver functions binned within 90% overlapping, 12-month windows for radial (a) and transverse (b) components. Panel (c) shows the corresponding variations in total power (black line - radial; grey line - transverse). Distribution of events with respect to back-azimuth and slowness of incoming wave-fields is presented in (d,e). Vertical lines indicate times of the San Simeon (2003) and Parkfield (2004) earthquakes.
\epsfig{, width=14cm}\end{center}\end{figure*}

Variations in PSD levels do not constrain the depth of this change; however, given its azimuthal dependence, it is unlikely that it originates at shallow levels. Its manifestation at periods of 0.5-3 s where PSD is highest suggests a depth-integrated effect that may be related to crack opening, non-linear damage, or permeability-enhanced pore-fluid flow. Reduced PSD levels imply a decrease in scattering energy from structure, i.e. smaller impedance contrasts across interfaces, which is consistent with redistribution of crustal pore-fluids and the breaking of impermeable barriers due to shaking (i.e. damage). The fact that the Parkfield earthquake did not have a similarly strong effect on the PSD levels may reflect the total re-equilibration of pore-fluid pressures in the crustal column after the San Simeon earthquake.

Such questions may be addressed by complementary methods that provide depth resolution (e.g. time-domain receiver functions, ambient-noise tomography, etc.) and using different data sets. These results are consistent with time-varying crustal S-velocities from ambient noise correlation studies and the coincident modulation of tremor activity near Parkfield following the 2003 San Simeon and 2004 Parkfield earthquakes (Brenguier et al., 2008). This preliminary study indicates that the receiver function method bears the potential for investigating 4-D faulting processes, albeit only as a diagnostic tool.


This work was funded by the Miller Institute for Basic Research in Science (UC Berkeley).


Brenguier, F., M. Campillo, C. Hadziioannou, N.M. Shapiro, R.M. Nadeau, and E. Larose, Postseismic relaxation along the San Andreas fault at Parkfield from continuous seismological observations, Science, 321, 1478-1481, 2008.

Ozacar A. A., and G. Zandt, Crustal structure and seismic anisotropy near the San Andreas fault at Parkfield, California, Geophys. J. Int., in press, 2009.

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