next up previous contents
Home: Berkeley Seismological Laboratory
Next: Inversions for buried slip Up: Ongoing Research - Local Previous: Parkfield Studies

Hayward Fault Studies

T. V. McEvilly, R. A. Uhrhammer, R. Burgmann, R. M. Nadeau, and R. W. Clymer; BSL, P. Hipley; Caltrans, L. Hutchings; LLNL


Introduction

The Hayward fault along the eastern side of San Francisco Bay is arguably one of the most hazardous faults in the world when one considers the probability of an earthquake together with the density and nature of development along it. The M$\sim$7 October 21, 1868 Hayward fault earthquake produced a 30-to-40-km-long surface rupture, and slip of up to 2 m may have extended at depth from Warm Springs to near Berkeley [Yu and Segall, 1996]. No other large historic earthquake has been unequivocally assigned to the Hayward fault. An 1836 earthquake, long assumed to have ruptured the northern Hayward fault, may actually have occurred on the San Andreas fault to the southwest, implying that the northernmost Hayward fault has not ruptured for at least 160 or perhaps even 220 years [Toppozada and Borchardt, 1998]. Williams [1992] documents at least six Hayward fault ruptures during the past 2100 years in a trenching study on the southern Hayward fault.

Long-term Hayward fault slip rate estimates of $\sim$10 mm/yr. suggest that more than a meter of slip potential has accumulated since the most recent events, making the Hayward fault capable of M$\geq$6.5 events in the near future [Lienkaemper et al., 1991; Savage and Lisowski, 1992]. Accordingly, the Hayward fault has been assigned the highest probability for a destructive earthquake in the Bay area in next 30 years with an estimated recurrence interval of 167$\pm $67 years [(WGCEP), 1990]. Estimated 30-year probabilities are 28$\%$ for the northern Hayward fault segment, and 23$\%$ for the southern part. Estimated cost and loss of life from such an event approach 10's of billions dollars and several thousand deaths, respectively. It is an extremely threatening fault, demanding a better understanding of its pathology. At present we know essentially nothing about its segmentation, and only that it creeps along much of its trace. Initial steps have been taken in modernizing the seismographic and geodetic instrumentation along the fault zone, and this project will use these accumulating data in new ways to better define the deformation in space and time.

Hayward Fault Seismicity

We have initiated the computationally intensive search for characteristic sequences on the Hayward fault, where seismicity resembles the Parkfield clustering, at a lower rate. The initial processing defines the 'Clustering Signature' (CS) for the seismicity on any seismogenic fault segment, and this signature reveals the pathology of that fault in terms of waveform similarity within the population of events, a unique characteristic of the fault zone that defines the degree of clustering in the earthquake process. In Figure 16.1 we illustrate CS with three examples calculated with waveforms from NCSN stations at Parkfield and at two places on the Hayward fault. Even at reduced detection completeness (M$\sim$1.3) of NCSN without the borehole network data, we see highly similar events (coherency $\ge $ 0.5) in the NCSN catalog. Seismicity on the Hayward fault resembles the 'clusters of clusters' pattern seen at Parkfield.


  
Figure: Cluster Signatures from NCSN waveform archives for earthquakes at Parkfield (PRK), the northern Hayward Fault (northern HF), and southern Hayward / Mission faults (southern HF/Mission). In this analysis, comparisons of waveform similarity between all possible pairs of earthquakes are made. The similarity of the earthquakes, as manifest in maximum waveform cross-correlations, are shown on the ordinate ($\beta $) axis and the distance between the two events is given on the abscissa. Contoured values show the relative number of event pairs with given similarity separated by a given distance. Hence spatial clusters of very similar earthquakes (including all repeating earthquakes) plot in the upper left hand quadrant, while less similar widely separated earthquake pairs plot in the lower right hand quadrant. Note that theoretically for completely uncorrelated waveforms, $\beta $ should be equivalent to the maximum correlation of white noise (ie. 0.25).
\begin{figure}
\begin{center}
\epsfig{file=bob99_2_1a.ps, width=7cm}\end{center}\end{figure}

The fraction of total seismicity made up by the repeating events on a fault segment is proportional to the slip rate in the model of Nadeau and McEvilly (1999), and the data support the model. The fractions are 48$\%$, 13$\%$ and 6$\%$, respectively, for Parkfield, southern Hayward/Mission and northern Hayward fault segments, within the NCSN catalog. If the data base from the borehole network at Parkfield (detection threshold ML 0) is used to determine the fraction, it is about 67$\%$. At Parkfield, where fault slip is around 25 mm/yr, the recurrence time for ML$\sim$0.0 is 3 to 6 months. Scaling to 10 mm/yr on the Hayward fault yields a value of 7 to 15 months repeat times. For the borehole stations, the improved signal-to-noise ratios being attained are pushing down the detection threshold to somewhere between ML 0.0 and ML -0.6 for local events. The expected rate of seismicity within a 10 km radius of Berkeley is 3 per month at ML 0.0 or 10 per month at the ML -0.6. We conclude that adequate data are expected to be available for this exercise from the Hayward fault borehole network, accessible from the optical mass-store archive at NCEDC. Our early search is identifying sequences of repeating events consistent with our expectations.

Slip Rate from Earthquake Recurrence

This method has recently been developed in our research program using data from another borehole network along an active fault zone. Studies at very high resolution of microearthquakes at Parkfield, CA since 1987 revealed a systematic organization in space and time, dominated by clustering of nearly identical, regularly occurring microearthquakes ('characteristic events') on 10-20 m wide patches within the fault zone (Nadeau and McEvilly, 1997, 1999). At Parkfield, more than half of the 5000+ events in the 1987-1998 catalog exhibit this trait. In general, recurrence intervals (0.5 to 2 yr.) scale with the magnitude of the repeating events for the magnitude range available (Mw 0.2 to 1.3). Clusters of these characteristic events occur throughout the slipping fault surface. Scalar seismic moments were estimated for 53 of the repeating sequences and combined with equivalent estimates from 8 similar but larger event sequences from the Stone Canyon section of the fault and the main Parkfield M6 sequence. These estimates show that seismic moment is being released as a function of time in a very regular manner. Measurements of the moment release rate, combined with an assumed tectonic loading rate, lead to estimates of the seismic parameters source area, slip, and recurrence interval. Such parameters exhibit a systematic dependence upon source size over a range of 1010 in seismic moment which can be described by simple scaling relationships (Nadeau and Johnson, 1998). What emerges from this analysis of moment release rates is a quantitative description of an earthquake process that is controlled by small strong asperities that occupy less than 1$\%$ of the fault area. A 26-months period of greatly increased activity during the study interval (M4.2, 4.6, 4.7 and 5.0 events and their aftershocks) accompanied changes up to 50$\%$ in previously stable recurrence intervals. Langbein et al. (1998) report on evidence in deformation measurements for a slip rate increase in 1993 at Parkfield. Nadeau and McEvilly (1999) show that it is possible under reasonable assumptions to infer the spatial distribution of variations in slip-rate on the fault surface from the changes in recurrence intervals for the characteristic event sequences. The analysis requires an assumption of constant area for the repeating sources - one easily supported by the lack of any detectable change in the waveforms (over the 100 Hz bandwidth) associated with the change in recurrence interval.

Improving Inversion for Fault Slip

We model the measured displacement field with rectangular displacement discontinuities with uniform slip in an homogeneous elastic half-space. A number of modeling and inversion tools have been developed that will enhance our ability to model these new data [e.g., Murray et al., 1993, 1996; Burgmann et al., 1994; 1997]. The addition of subsurface slip constraints to the inversion is fundamental to this proposed study. Even with the InSAR-quality range-change data available crossing the Hayward fault, determination of the vertical extent (creep patch width) of the creeping zone is nearly impossible to recover if the locked zone is deeper than about 3 km. Slip distribution at depth is elusive, and asperities are cannot be imaged even near the surface, without subsurface slip information. The next M6.5+ event on the Hayward fault will likely nucleate typically near the lower bound of the seismogenic zone, a depth of 8-10 km, where resolution of the slip distribution from surface observations is very poor. Constraints from recurrence-determined slip rates at a few depth points offers promise for substantially sharpened resolution near the all-important base of the slipping zone. We are working to improve the slip inversion data set by adding to the growing InSAR data base a combination of microearthquake-determined slip rates throughout the fault zone and enhanced GPS coverage from a tight rotating deployment of single-frequency L1 receivers in the center of our study region.

The Mw 4.1 El Cerrito Earthquake

The largest local earthquake that occurred along the northern segment of the Hayward Fault during the past year was a Mw 4.1 event which occurred at 4:16 AM local time on December 4, 1998 at a depth of 6.8 km along the Hayward Fault approximately 6.3 km northwest of the Berkeley campus. This earthquake was recorded by all the northern Hayward Fault Network stations and by the Caltrans bridge borehole sites.

Ground motions recorded at BDSN stations in the vicinity of the northern Hayward Fault are shown in Figure 16.2. The relative ground velocity traces are ordered from top to bottom by epicentral distance. Note the variation in the seismic wavefield with distance, azimuth, and side of the Hayward Fault. The crustal structure across the northern Hayward Fault zone, in the vicinity of Berkeley, is rather complex with a lateral bedrock velocity contrast of approximately 30$\%$ across the fault zone with the SW side fast and also with an attenuation contrast across the fault zone with the NE side having higher attenuation as evidenced by the variability in the observed seismic ground motions shown in Figure 16.2. For example, the observed waveforms at BKS (6.8 km away on NE side) and at BRK (5.8 km away on SW side) show considerable variation in the high-frequency spectral content, in the low-frequency coda duration, and in the S-P interval (a 60$\%$ change for a 20$\%$ increase in distance. The observation that the first motions at BKS and BRK are both dilational is an indication, given the right-lateral strike-slip source mechanism and compressional quadrant path to BKS, that the initial P-wave recorded at BKS is laterally refracted along the fault zone.


  
Figure 16.2: Ground motions at nine locations in the vicinity of the northern Hayward fault for the 12/04/98 Hayward Fault earthquake. The traces are vertical-component relative ground velocities and they are ordered top to bottom by epicentral distance. All stations are in boreholes except for SMCB which is in a 4 m posthole and BKS and BRK which are standard BDSN broadband stations in surface vaults.
\begin{figure*}
\begin{center}
\epsfig{file=bob99_2_2a.ps, width=15cm}\end{center}\end{figure*}

Bedrock ground motions at seven locations along the Bay Bridge for the Mw 4.1 earthquake are shown in Figure 16.3. Note that the character and phasing of the ground motion varies considerably from east to west (top to bottom in Figure 16.3) as the wavefield propagates across San Francisco Bay. Knowledge of the bedrock ground motions is of paramount engineering importance for seismic design of the bridge.


  
Figure 16.3: Ground motions at seven locations in bedrock along the Bay Bridge from the Mw 4.1 earthquake located on the northern Hayward Fault approximately 13 km NNE of Yerba Buena Island (the middle trace). Velocity records transverse to the bridge are shown and all traces are to the same scale.
\begin{figure*}
\begin{center}
\epsfig{file=bob99_2_3a.ps, width=15cm}\end{center}\end{figure*}

References

(WGCEP), W.G.C.E.P., Probability of large earthquakes in the San Francisco Bay Region, California, U.S. Geol. Surv. Circ., 1053, 1-51, 1990.

Burgmann, R., R. Arrowsmith, T. Dumitru, and R. McLaughlin, Rise and fall of the southern Santa Cruz Mountains, California, deduced from fission track dating, geomorphic analysis, and geodetic data, J. Geophys. Res., 99 (B10), 20181-20202, 1994.

Burgmann, R., P. Segall, M. Lisowski, and J. Svarc, Postseismic strain following the 1989 Loma Prieta earthquake from GPS and leveling measurements, J. Geophys. Res., 102, 4933- 4955, 1997.

Langbein, J., R.L. Gwyther and M.T. Gladwin, Possible increase in fault slip rate at Parkfield in 1993 as inferred from deformation measurements from 1986 to 1997, Seism. Res. L., 69, 151, 1998.

Lienkaemper, J.J., J.S. Galehouse, and R.W. Simpson, Creep response of the Hayward fault to stress changes caused by the Loma Prieta earthquake, Science, 276, 2014-2016, 1997.

Murray, M.H., J.C. Savage, M. Lisowski, and W.K. Gross, Coseismic displacements: 1992 Landers, California, earthquake, Geophys. Res. Lett., 20, 623626, 1993.

Murray, M. H., Marshall, G. A., Lisowski, M., and Stein, R. S., The 1992 M=7 Cape Mendocino, California, earthquake: Coseismic deformation at the south end of the Cascadia megathrust,

Nadeau, R. M. and L. R. Johnson, Seismological Studies at Parkfield VI: Moment Release Rates and Estimates of Source Parameters for Small Repeating Earthquakes, Bull. Seism. Soc. Am.,. 88, 790-814, 1998.

Nadeau, R.M. and T.V. McEvilly , Fault slip rates at depth from recurrence intervals of repeating microearthquakes, Science (submitted), 1999.

Savage, J.C., and M. Lisowski, Inferred depth of creep on the Hayward fault, central California, J. Geophys. Res., 98, 787-795, 1992.

Toppozada, T.R., and G. Borchardt, Re-evaluation of the 1836 "Hayward fault" and the 1838 San Andreas fault earthquakes, Bull. Seism. Soc. Am., 88, 140-159, 1998.

Williams, P.L., Geologic record of southern Hayward fault earthquakes, Div. Mines Geol. Spec. Pub., 113, 171-179, 1992.

Yu, E., and P. Segall, Slip in the 1868 Hayward earthquake from the analysis of historical triangulation data, J. Geophys. Res., 101 (B7), 16,101-16,118, 1996.


next up previous contents
Next: Inversions for buried slip Up: Ongoing Research - Local Previous: Parkfield Studies

The Berkeley Seismological Laboratory, 202 McCone Hall, UC Berkeley, Berkeley CA 94720
Questions and comments to www@seismo.berkeley.edu
Copyright 1999, The Regents of the University of California.