Interseismic Creep on the Concord Fault from PS-InSAR and SBAS

Ingrid Johanson, Roland Bürgmann, Alessandro Ferretti (TRE, Milan) and Fabrizio Novali (TRE, Milan)


The Concord fault (CF) is part of the San Andreas fault (SAF) system in California's San Francisco Bay Area. Its long-term slip rate of $7\pm2$ mm/yr (geodetically determined) represents about one-fifth of the total SAF system rate. The Concord fault also creeps at a rate of 2.5-3.5 mm/yr, determined from measurements at two alinement arrays (McFarland et al., 2009). The alinement arrays also showed time-variable slip, with creep events occurring every 3-5 years. Measuring creep and its variability on the CF is important, not just for understanding the fault's earthquake potential, but also because it may give us insight into how slip is transferred onto the Concord fault. The similarity in creep rate between the Northern Calaveras Fault (NCF) and CF, noted by Galehouse and Lienkamper (2003) is one line of evidence for linking the two.

We use PS-InSAR (Permanent Scatterer Interferometric Synthetic Aperture Radar) and the SBAS (Small Baseline Subset) technique to construct time series of ground motion around the CF and measure the creep rate along several profiles (Figure 2.7). The analysis is possible because of the extensive set of ERS data available through the WInSAR and GeoEarthScope archives. The PS-InSAR method identifies and integrates individual points that act like point scatterers and have stable phase measurements in a set of interferograms with a common master scene (starting date) (Ferretti et al., 2001) . The SBAS technique uses a set of interferograms with various master and slave scenes and provides time series of deformation at any consistently coherent pixel (Berardino, 2002).

Figure 2.7: Overview map showing mean velocity of each PS-InSAR and SBAS dataset (B,C,D). Dashed boxes in subfigure A are the areas included in the swath averages, which are then projected onto the solid black centerlines. White triangles are alinement arrays CSAL and CASH.
\begin{figure}\centering\epsfig{file=ijohanson10_1_1.eps, width=8cm}\end{figure}

Figure 2.8: An example of a swath average profile for descending Track 70. Grey circles are the actual PS-InSAR points within the swath. Black circles are the calculated swath averages, with one sigma error bars. The dashed black line and offset is fit to the data by minimizing the least-squares residuals.
\epsfig{file=ijohanson10_1_2.eps, width=6cm}\end{center}\end{figure}

Figure 2.9: Range change across the trace of the Concord fault through time for all three datasets for one profile. Track 70 is shown in red, track 342 in green and track 478 is shown in blue. The secular creep rate as determined from a weighted linear inversion is printed next to each series.
\epsfig{file=ijohanson10_1_3.eps, width=7cm}\end{center}\end{figure}

Figure 2.10: Calculated creep rates at each profile are shown according to their location along strike (red squares). Creep rates from alinement array measurements are shown as triangles.
\epsfig{file=ijohanson10_1_4.eps, width=7cm}\end{center}\end{figure}

Measuring creep from InSAR time series

We use 2 PS-InSAR datasets from descending tracks 70 and 342 from the European Space Agency's ERS1 & 2 satellites, using 46 and 30 acquisitions, respectively, and spanning from 1992 through 2001. The data were processed using the PS-InSAR method of Ferretti et al. (2001) to produce time series of range change (change in distance between the ground and satellite) for each point as shown in Figure 2.7B & C. We also use a set of 22 ERS scenes from ascending track 478, arranged into a time series using the SBAS technique (Figure 2.7D).

To measure creep on the Concord fault, we look at motion along several profiles crossing the Concord fault. We use swath averaging to construct a profile at each time step of the InSAR time series (e.g. Figure 2.8). Within each swath, shown as dashed lines in Figure 2.7A, points within 0.25 km bins perpendicular to the fault are averaged together and projected onto the centerline. For example, all points between 0.5 and 0.75 km west of the fault are averaged together to provide one point on the profile, and their standard deviation provides an estimate of the profile point's uncertainty.

A swath average profile is constructed for all acquisition dates in the three datasets and on each of the five profiles, for a total of 490 profiles. A linear inversion is performed on each profile to obtain the offset at the fault trace, produced by the shallowly creeping fault (dashed line in Figure 2.8). Each offset value represents the amount of creep on the Concord fault since the beginning of the time series.

Long-term creep rates

By fitting a line to the offsets, we can obtain a measurement of the secular range change rate at each profile (Figure 2.9). Using multiple viewing geometries (descending and ascending tracks) allows us to uniquely separate the range change into horizontal and vertical components. We assume horizontal motions are due to fault creep (Figure 2.10) and vertical motions represent hydrologic processes. All horizontal motion is taken to be parallel to the local strike of the Concord fault at each profile, as would be the case for pure strike-slip. The inversion of fault creep rate ($V_{SS}$) is set up as shown below.

RC=V_{SS}\left[\sin^2{\theta}\cos{\alpha} +...
RC_{T478} \\
\nonumber \\

The results for secular creep-rate ($V_{SS}$) for each profile are shown in Figure 2.10. The results are very consistent with the long-term creep rates obtained from the alinement array surveys. It would appear that creep rates increase toward the ends of the Concord fault trace; however all the creep rate measurements from InSAR profiles are within uncertainty of the alinement array derived rates (2.5-3.5 mm/yr).


Berardino, P., G. Fornaro, R. Lanari, and E. Sansosti, A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms. IEEE Trans. Geo. Rem. S., 40 2375-2383, 2002.

Ferretti, A., C. Prati, and F. Rocca, Permanent scatterers in SAR interferometry, IEEE Trans. Geosci. Remote Sens., 39, 8-20, 2001.

Galehouse, J.S. and J. J. Lienkaemper, Inferences drawn from two decades of alinement array measurements of creep on faults in the San Francisco Bay Region. Bull. Seismol. Soc. Am., 93(6), 2415-2433, 2003.

McFarland, F., J. Lienkaemper, and S. J. Caskey, Data from theodolite measurements of creep rates on San Francisco Bay region faults, California; 1979-2009, USGS Open file report 09-1119, 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