Model Results

The model plane is vertical with a strike of 319 degrees. We discretize the plane into 300 2 km x 1 km elements and perform a least squares inversion while seeking a smooth distribution of right-lateral slip. The results are shown in Figure 13.10. The slip model contains two high-slip asperities: one just above the hypocenter (red star in Figure 13.10) and a larger asperity to the northwest. Much of the slip in our model occurs within the area outlined by aftershocks as shown in Figure 13.10. Langbein et al. (2005) note that the relocated aftershocks occur in the same clusters and streaks as the pre-earthquake background seismicity, including a streak which is visible in the aftershocks at about 5 km depth. One interpretation of microseismicity streaks is that they occur at the boundaries of creeping and locked asperities of the fault surface. Assuming that in an earthquake the strong asperity makes up its slip deficit by slipping more than the surrounding creeping sections, the microseismic streak would be expected to bound areas of high coseismic slip. The streak of aftershocks at 5 km depth occurs near the top of the northwest asperity, and could be interpreted as weakly bounding it.

Figure 13.10: Results of inversion for slip during the 2004 Parkfield earthquake and two days of postseismic slip. Red star is earthquake hypocenter, black circles are double-difference relocated aftershocks (Langbein et al., 2005), and colored triangles are surface offsets measured by creepmeters. X-axis datum is GPS station Carr Hill (CARH)
\epsfig{file=ingrid05_1_2.eps, width=8cm}\end{center}\end{figure}

Creepmeter measured surface offsets are plotted as colored triangles in Figure 13.10 for comparison. Many of the creepmeters on the main SAF trace went off scale either during or soon after the earthquake. Between Sep. 30 and Oct. 2, these creepmeters were brought back on scale and an offset was manually measured. The triangles in Figure 13.10 represent the total creep measured when the creepmeters were brought back on scale. The top row of our model matches the overall magnitude and some of the features of the creepmeter surface slip distribution. Both show more surface slip 0-10 km north of Carr Hill that to the south.

The total moment of our model is $2.56\times10^{18}$ Nm; equivalent to $M_w$6.2. Seismic modelling of the coseismic rupture has estimated a moment magnitude $M_w$6.0, which would require $1.32\times10^{18}$ Nm ($M_w$6.0) of the slip in our model to have occurred within the first two days of the postseismic period. An analysis of additional InSAR and GPS data spanning three months post-earthquake, suggests that $1.48\times10^{18}$ Nm ($M_w$6.1) of moment was subsequently released postseismically (Johanson et al., submitted). Johanson et al. (submitted) find that aseismic slip could account for as much as 70% of the total (coseismic and postseismic) moment release of the Parkfield earthquake.

Berkeley Seismological Laboratory
215 McCone Hall, UC Berkeley, Berkeley, CA 94720-4760
Questions or comments? Send e-mail:
© 2005, The Regents of the University of California