The 100-km wide San Andreas fault system in northern California is composed of three sub-parallel right-lateral faults: the San Andreas, Ma'acama, and Bartlett Springs faults. The San Andreas fault has been essentially aseismic since it last ruptured from San Juan Bautista to Cape Mendocino in the great 1906 San Francisco earthquake. No major earthquakes have occurred during the historical record on the more seismically active Ma'acama, and Bartlett Springs faults, northern extensions of the Hayward-Rodgers Creek and Calaveras-Concord-Green Valley faults in the San Francisco Bay area, but the slip deficit on the Ma'acama fault, based on a high slip rate found by previous geodetic studies, may now be large enough to generate a magnitude 7 earthquake, posing a significant seismic hazard.
In Fall 2002 we will resurvey about 40 monuments that form 4 profiles across the northern San Andreas fault system (Figure 24.1). Most of the monuments were last observed in 1993 or 1995, so the new observations will significantly improve estimates of their relative motion and models of average interseismic strain accumulation, including possible spatial variations along the fault system, and will form the basis for future observations to detect temporal variations in deformation. The monitoring network will provide roughly 10-station profiles every 50 km from Pt. Reyes to Cape Mendocino. We will reprocess the GPS observations using more modern techniques and apply recently developed modeling techniques to provide a more self-consistent description of deformation across northern California and strain accumulation on the northern San Andreas fault system.
The northern segment is the youngest portion of the San Andreas fault system, forming in the wake of the northward propagation of the Mendocino triple junction (MTJ) where the Pacific, North America, and Gorda (southern Juan de Fuca) plates meet. The Pacific plate moves northwestward about 35-40 mm/yr relative to central California across a broad 100 km zone in northern California, with the additional motion accommodated in eastern California and the Basin and Range province making up the total relative Pacific-North America, about 50 mm/yr predicted by NUVEL-1A.
Geodetic measurements, which are particularly useful for detecting deformation and strain on deep structures throughout the seismic cycle, currently provide only weak constraints on average strain accumulation for much of this region. This is particularly true north of Point Arena, where the San Andreas fault lies offshore and bends 20° clockwise as it comes onshore at Point Delgada before terminating at the Mendocino triple junction near Cape Mendocino. The Ma'acama and Bartlett Springs faults also undergo clockwise rotation to a lesser degree and appear to reactivate older subduction structures as the triple junction migrates northwestward. This complex geometry and other features, such as surface creep on the Ma'acama fault, suggest the deformation may be spatio-temporally variable, similar to deformation that has been observed, and is now being monitored by continuous, frequent survey-mode GPS, and other geodetic observations, on the southern extensions of these faults in the Bay Area that share many of the same characteristics. Understanding how this spatio-temporal variability affects strain accumulation on the faults is critical for assessment of the timing and hazards posed by future earthquakes.
In Freymueller et al. (1999) we reported on data collected in 1991-1995, primarily from the Ukiah and Willits profiles (Figure 24.1). GPS velocities from this network place some constraints on the total slip rate on the San Andreas fault system, which we estimate to be 39.6+1.5-0.6 mm/yr (68.6% upper and lower confidence intervals from a nonlinear inversion are indicated by subscripts and superscripts). Slip rates on the individual faults are determined less precisely due to the high correlations between estimated slip rates and locking depths, and between slip rates on adjacent faults. Our estimated slip rate on the San Andreas fault is lower than all geologic estimates, although the 95% confidence interval overlaps the range of geologic estimates. Our estimate of the Ma'acama fault slip rate is greater than slip rate estimates for the Hayward or Rodgers Creek faults, its continuation to the south. The Ma'acama fault most likely poses a significant seismic hazard, as it has a high slip rate and a slip deficit large enough to generate a magnitude 7 earthquake today since there have been no significant earthquakes on the fault in the historical record.
In our Freymueller et al. (1999) study, we used a single station constrained to a VLBI-derived velocity to define the velocity reference frame, and were unable to properly account for the far-field motions in terms of angular velocities. We will reanalyze the observations using a more robust, globally defined reference frame that will facilitate application of the angular velocity-fault backslip model (Murray and Segall, 2001), and provide a more self-consistent description of northern California deformation. The method can accommodate the observed creep on the faults, and we will also extend the technique to 3D models using backslip on shallow rectangular dislocations in order to investigate the effects of the fault orientations rotating more clockwise at the northernmost segment.
We will investigate the uncertainties of our strain accumulation models using the bootstrap techniques, which provide realistic uncertainties for nonlinear optimization problems. The correlations between slip-rates and locking depths are highly correlated, so that higher slip rates on the San Andreas fault, for example, can tradeoff with lower slip-rates on the Ma'acama fault with almost equal misfit. We will test methods for adding geologic and other information, using Bayesian techniques, to test whether the additional information can reduce the correlations and provide better resolution on other parameters. For example, the previously determined slip-rate on the San Andreas fault ranged from 10.5-22.6 mm/yr at 95% confidence with locking depths ranging 4.7-44.6 km. By applying constraints derived from other seismic, geodetic, and paleoseismic observations, the estimated parameters can be much better resolved (Figure 24.2).
This project is supported by the USGS, through the NEHRP External Grants Program.
Freymueller, J. T., M. H. Murray, P. Segall, and D. Castillo, Kinematics of the Pacific-North America plate boundary zone, northern California, J. Geophys. Res., 104, 7419-7442, 1999.
Murray, M. H., and P. Segall, Modeling broadscale deformation in northern California and Nevada from plate motions and elastic strain accumulation, Geophys. Res. Lett., 28, 4315-4318, 2001.