ML20127D838

Potential for Earthquake Rupture of Northern Calaveras Fault

ML20127D838
Person / Time
Site:	Aerotest
Issue date:	09/04/1992
From:	Lindh A, Oppenheimer D INTERIOR, DEPT. OF, GEOLOGICAL SURVEY
To:
Shared Package
ML20127D815	List: ML20127D812 ML20127D838
References
NUDOCS 9209150189
Download: ML20127D838 (13)

Text

- ,

De Potential for Earthquake Rupture of the Northern Calaveras Fauh by David II. Oppenheimer and Allan G. Lindh' ABSTRACT Since 1969 virtually no seismicity has occurred on the Calavens fault north of the Calaveras Reservoir and r.outh of Danville. By compt.rison with the seismic behavior of the Calaveras fault south of the reservoir and the fauh segment of the San Andreas fauh that ruptured during the Loma Prieta earthquake, we infer from this absence of earthquakes that the northern Calaveras fauh is locked and accumulating clastic stmin energy. He slip rate of this segment of the fauh is approximately 6 mm/yr, and there is no indication of surface cTeep. He seismic gap is approximately 40 km long and the width of the fauh inferred from the maximum depth of earthquakes ranges from 10 to 15 km. We consider the potential of an earthquake occurring on this segment over the next 30 years for two equally plausible rupture scenarios. If the entire northern Calaveras fauh ruptures at one time, the expected magnitude (M) would be approximately M 7, but there is insufficient infomtation at present to calculate reliable probabilities. Ahematively, the fault could rupture in two or more M 6 earthquakes, as suggested by the occturence of three modente earthquakes between 1858 and 1864, with a probability of one or more events of M > 6 of 0.33. Ahhough the probabilities have large uncertainties, we believe the geologic and geophysical data indicate that the northem Calaveras fauh is a significant seismic hazard for the San Francisco Bay region.

INTRODUCTION ne earth science community is not able at present to provide precise forecasts of the time, place, and inagnitude oflarge earthquakes. Yet earthquakes present considerable risk to society, and despite our limited understanding of the earthquake process, we are continually urged to make educated assessments of eanhquake hazards. Publications by the Working Group on Califomia Eanhquake Probabilities (WGCEP) (1988) reached consensus opinion on the probability of large earthquakes in Califomia through a formal process that considered the available infonnation on known, active faults. A subsequent report was released in 1990 that considered the probabilities of earthquakes in the San Francisco Bay region in light of the occurrence of the 1989 magnitude

- (M) 7.0 lema Prieta earthquake and newly published information on the Rodgers Creek fauh (WGCEP,1990). nat report considered the earthquake potential of the IIayward, Rodgers Creek, and San Francisco Peninsula segment of the San Andreas fault. We will not revisit those topics in this report; rather, we consider the potential earthquake liazard of a fauh segment not considered by the WGCEP - the Calavens fault north of the Calaveras reservoir (" northern" Calaveras) (Fig.1) We will review the geologic and geophysical data that bear on the likely size of the rupture zone, the interval of time since the last earthquake, and the slip rate of the fault.

We will then use this information to forecast the probability of an eartlquake occurring on the northem Calaveras fauh over the next 30 years.

' 4 ,Cahfomia 94025 Both at U.S. Geological Survey,345 Middlefield Road MS 977, Menlo Pi.

1 k g 9 920904 PDR

a. .

MICROSEISh0 CITY AND MAIN SHOCK RUPTURE ARFA Maps showing fauhs with late Quaternary or Holocene displacement are one of the primary tools .

in forecasting the locations oflarge earthquakes (Schwartz and Coppemnkh,1984). Dese maps, however, depict only the two-dimensional aspect of fauh stmcture; the third dimension, depth, has remained more difficult to quantify until the advent of earthquake netwtuts beginning in the late 1920's. Seveal features of the background seismicity reflect conditions in the crust that ,

control main shock mpture. The depth interval in which this seismicity occurs indicates the width .

of the brittle, or t.eismoEenic, region of the crust, and consequently the maximmn wkhh of faults that could dynamicaDy nrpture during large earthquakes. I In addition analyses of recent earthquake sequences in Califomia indicate that the location partems of background microseismicity can also delineate the lateral extent of the rupture zone of I

the expected main diock. For exunple, studies of seismicity on the Calaveras fauh south of Calaveras reservoir (' southern" Calaveras) and of the segment of the San Andreas fault that ruptured during the Loma Prieta earthquake show that the region of the fauh that slipped during the main shock was aseisade both before and after the main shock (Hartzell and Heaton,1986; Oppenheimer et al.,1990; Oppenheimer,1990). Conversely, Bakun et al. (1986) have shown s that the cumulative amount of slip produced by microseismicity is much less than the amount of slip expected from long-term slip rate estimates; consequently, they concluded that regions of a fauh zone that exhibit microseismicity slip primarily through aseismic creep processes. Aus, continuous monitoring ofseismicity over extended time periods can image seismic ' gaps" that 2 may be sites of potential main shock ocanTence. 'Ihese gaps provide an estimate of the dimensions of the mpture zone, and hence, the magnitude of future earthquakes. ,

If the above modelis appropriate, creeping sections of a fault should have microcarthquakes that locate on the plane of the fanh. Likewise, their focal mechanisms should also be consistent with the sense of slip anticipated for the main shock. Conversely, locked fauhs could exhibit background seismicity, but any microcarthquakes should primarily occur adjacent to, but not on, the fault anticipated to mpture during a main shock. Because of the prevailing state of fanh nonnal compressive stress in central Califomia, the focal mechanisms of off-fauh seismicity is frequently different from the mechanism of the anticipated main deck. *Ihus, to discriminate

" locked" fauli runes from creeping fauh zones, it is necessary to examine the locations and focal mechanisms of the backgramd seismicity.

TIIE NORTIIERN CALAVERAS SEISMIC GAP Seismicity has been recorded in the cast San Francisco Bay region by the U.S. Geological Survey (USGS) since 1969 (Fig. 2). We estimate the epicentral location accuracy to be approxhnately 1.5 km in this region due to a seismic station spacing of 715 km and the use of locally specific velocity models (F. Klein, USGS, written conumm.,1991). He locations of earthquakes recorded during the past two decades image several seismic gaps throughout the region that could potentially rupture in M > 6 earthquakes. We do not believe our undentmding of the slip rates and historical mpture for most of these gaps is sufficient to enable us to provide meaningful estimates of the likelihood of mpture for each gap. Nonetheless, the sudden onset of the 1986 Mt. Lewis (M i5.8),1980 Livennore (M 5.9), t and 1989 Loma Prieta earthquake sequences demonstrates that it is not unusual for main shocks to occur on fauh segments which 2

m_. . _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _

)

are not well imaged by background seismicity, but which are known to be active either through geologic investigadons or by the occurrence of seismicity on adjacent segments, in c_otrast to the fauh segments described above, the slip rate and historic record of main shock ocarrtence for the seismic gap on the noithem Calaveras fault is perhaps adequate to attempt a preliminary estirnate of the probability of rupture on this segment. The gap extends approximately 40 km from the Calaveras Reservoir to the vichity of DanviDe (Figs. 2 and 3); the vertical width of the seistnogenic zone appears to increase from 13 km near Calaveras Rese voir to 18 km near Danvi!)e. By comparison with the similar size of the mpture area of the Loma Prieta earthquake, the northern Calaveras seismic gap is capable of sustaining an earthquake with a magnitude as large as M 7, ne endpoints of the seismic gap also coincide with complications in the fauh geometry. The southern end of the se6 ment occurs at the intersection of the southem

- Calaveras and the Mission Tsuks; the latter is poorly expressed in the surface geology ahbough a trend of seismicity is obscaved in the vicinity of the mapped fauh (Wong and IIemphill-11aley.

1992; Andrews aat Oppenheimer,1992). The northern end of the segment coincides with the northern termination of the Calaveras fauh (Simpson et al.,1992), as manifest in the surface geology, where the displacement of the Calavens fault is probably transfened to the Franklin andbr Concord fauhs. We believe that most of the Calaveras displacement is m. ken up primarily by the Concord fault because the step-over region that connects the two fauhs is seismically active (Oppenheimer and Macgregor-Scott,1992).

Slip Rate No creep has been detected on the surface fauh trace of the northem Calaveras fauh (Prescott and Lisowski,1983; Galchouse,1991), but the slip rate recorded on short (<5 km) baselines is approximately 2.5-3.5 mm/yr (Moster,1977; Prescott et al.,1981; Burford and Sharp,1982; Prescott and Lisowski,1983). Im.aediately south of the segment, the slip rate measured acauss the Calaveras Reservoir geode"c network (7 km aperture) is approximately 6 mm/yr (Prescott et

- al.,1981). North of the gap, creep observations across the Concord fault indicate slip rates of 2.7-4.0 mm/yr (llarsh and Burford,1982; Galehouse,1991). Because of the proximity of the Hayward and Calaveras faults, it is difficah to model how stain accumulation is partitioned as a function of depth, but it is likely that these short base-line observations refled shallow slip.

llowever, strain accumulation at seismogenic depths on locked sections of the northem Calaveras fauh is not precluded by these obseivations, and the shallow slip rates of 2.5-6 mmfyr can be M red a minimum estimate of the slip rate at depth.

Another estimate of the slip rate on the northem Calavens fauh can be obtainw! indirectly by considering the regional slip budget. The 17 mmfyr slip nte across the southern Calaveras fault l - (Savage er.rl.,1979) is partitioned into 9 mmfyr across the Haywara fauk (Lienkacaper et al.,

1991), perhaps a few mmlyr on the Mt. Lewis-Greenville fault system, leaving about 6 mm/>T on the northem Calaveras fauh. If we further assume that approximately 2 unnlyr of slip occurs across the San Gregorio fauh and 19 mm/yr across the San Andreas fauh (WGCEP,1990), then the total slip rate across the central San Francisco Bay area is approximately 40 mmlyr. This total agrees well with the DeMets et al. (1990) NUVEL-1 model for the San Andreas system and is consistent with the Lisowski et al. (1991) observation of 38 mm/yr across the south Bay. The Lisowski er al. observation is a lower bound value since their tri-lateration array did not extend far enough to the east or to the west to be assured of detecting all the displacement. Moreover, Kelson et al. (1992) obtain a geologic slip rate estimate of 8i3 mm/yr at the Leyden Creek trench 3

I

4.

. . .. s ,

I .

site on the northem Calaveras fauh. Considering all the available data, we estimate tlat the slip rate for the northem Calaveras fauh is 6 2 mmlyr.

Mkroseismicity ne focal mechanisms of the background seismicity indicate that the carthquakes are occurring on a variety of fauh types and orientations (Fig. 4), consistent with the hypothesis that the northern Calaveras fault is locked. At the southem end of the segment near the Sunol Valley, three fault types predominate - reverse fauhs and vertical, right-lateral, strike-slip fauks, both of which strike parallel to the Calaveras fault, and vertical, right lateral, stdke slip faults trending nonh-south. He few reliable focal mechanisms that could be determined for the cental portion of the gap indicate that seismicity is occurring predominantly on vertical, dght. lateral, strike-slip fauhs parallel to the Calaveras fault. Ilowever, these earthquakes are distnluted within a region 5 km from the trace of the Calaveras fatdt, far wider than could be attributed to earthquake location error. This difTuse epicentml ditibution suggests that these earthquakes are occurring on small, subsidiary fauhs adjaccat to the Calaveras. At the nonhem cod of the segment near Pleasanton, reverse and strike-slip fanhing is occurring in a fashion simihr to the fauhing near Sunot, and the epicentral distributions we t >c, diffuse to be able to correlate with the focal mechanisms. In summary, the mechanismr. Indicate tLat some of the background seismicity may be occuning on the Calaveras fauh, asstadng that the fauh is vertical, but the lack of definitive epicentral trends and the variety of other mechanisms suggests to us that most of these earthquakes occur off the Calaveras fault and that the Calavens is locked.

In support of this hypothesis, the recent geologic investigations by Kelson et al. (1992) at Leyden Creek indicate that several slip episodes occurred within the past 1800 years. In general, surface mpture is not observed for earthquakes smaller than Af 6 in Califomia. Rus, the presence of a mapped fault with evidence of discrete slip events at the surface, historical seismicity associated with the faub (discussed below), the hck of observable creep, a slip rate of 6 mm/yr, and the hck of microseismicity are all consistent with the hypothesis that the northem Calaveras fault is locked and accumuhting chstic strain energy at depth that could be released during moderate-to-brge canhquakes.

RUPTURE PROBABILITY To assess the likelihood for a main shock on this segment, it is necessary to know the date of the hst main shock. Seismographs were installed in the San Francisco Bay region beginning in the 1

early 1900's, and information on earthquake locations and magnitudes before that time are based on published accounts of the severity and location of damage, Rese isoseismallocations are probably accurate to 25 km at best (Toppozada, et al.,1981), except where surface rupture is observed, and the magnitude uncertainties are about 0.5 units. In 1861 an earthquake occurred at the northern end of the northern Calavens fault (Fig.1), producing a fissure 13 km in length.

Toppozada et al. (1981),has assigned the earthquake a Af 5.6, but if the fissure represents surface rupture rather than shaking effects or post-seismic creep, the 13 km length indicates that the earthquake probably exceeded Af 6. Earthquakes also occurred in the vicinity of this gap in 1858 (Af 6.1) and 1864 (AI 5.7), but it is not known whether they ruptured the Calaveras fauh, the nearby llayward fauh, or a blind fault that is not curretly genenting microseismicity. The historical record indicates that the northem Calaveras fault has probably not mpnited in a Af 7 earthquake since 1830, and certainly not aller 1849, 4

_ _ - _ _ _ _ _ _ _ _ _ - _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ ~

• g j' e j .

l L

We have argued kbove that the seismological evidence pennits the hgerpretation that the north l

Cahvens fauh consists of one locked 40 km long segment, capable of failure in a single M 7 l earthquake. Ilowever, there is no independent Ecologic evidence for such an interpretation fauh geomorphology (Simpson et cL,1992) suggests, instead, that the noi1hern Cahver more frequent M 6+ events, two or more of which would be requh ed to fill the 40 km gap. Th rupture scenario is equally plausible in consideration of occurence of the three earthquak between 1858 and 1864. We have no basis on which to distinguish between diese two hypotheses, and so consider them both in the following discussion of mpture probabilitie Because lbe slip history of the fauh is not yet known, we must make several assumptions in order to calcuhte the probabilities. In particular, we requhe an estimate of the expected recurrence interval and the thne of hst rupture. For the first hypothesis of a single M 7 on the northern Cahveras fauh, we know only that the time oflast rupture precedes at htest 1849.

Moreover, since we do not yet have any geologic evidence to provide a recurrence interval, w ruust estimate the repeat time from the slip rate and the amount of slip expected to occur duri M 7 earthquake. Here too, we have little data on the amount of slip per event. We can assume that the rupture length, L, of 40 km would genemte slip, D, of approxi nately 1.1 m, using the relation: D = 2.8 4X 10 L of the WGCEP (1988). Attematively, we can anticipate that the slip would be nearly 2 m, consistent with estimates of the total displacement from geodetic data (Lisowski et al.,1990) in the lema Prieta carthquake. Combined with the slip rate estim 2 uun/yr, this resuhs in an expected recuTence interval of either 333 or 187 yeam, depend which slip value is assmned. 'Ihese uncestainties cater into probability cakubtions through coefficient of variation (WGCEP,1990), and we would assign a value of 0.4 to the coefficient as the WOCEP did for t!e two Hayward segments. However, without a reliable date of last rupture, the probability calcuhtions would probably not be reliable. Qualitatively, we would estim probability that a M 7 on the northem Cahvens will occur in the next 30 years to be estimates of 0.22-0.28 on the four other faults in the San Francisco Bay region (WGCEP,199 If the northern Cahvens fails in more frequent M 6 events, we can calcable the probabilities based on the Lindh (1988) statistical model that assumes that earthquake recurrence can p be described by a Gaussian probability distribution with a mean recurrence time and coeffici variation of the recuumco time. Unlike the M 7 scenario, we know when the last M 6 earthqua occurred on the northern Calaveras fauh. However, we have no direct information on recurrence, On the southern Calmras fault,t fhe M 6.2 Morgan Hill segment 2 and # 5.9 Coyote Lake segments have recurrence times of 73 and 82 years, respectively (Oppenheimer et the WGCEP (1990) estimated its slip rate to be 14.4 mmlyr. Adjusting this -80 year repeat for M 6 events for the 6 mm/yr slip rate of the northern Cahveras segment, we estimate the recurrence interval of the northern segment to be abotr 192 years. If we further assume that th entire northern Cahveras failed between 1858 and 1864, we can use 1860 as a reasonable estimate of the time of the most recent event for both sub-segments. 'Ihe resuhing estimate of time of the next event is then 2052158, assuming a coefficient of variation of 0.3 to reflect th better estimate of the time of the hst event under this hypothesis. We estimate a probability o 0.18 that a M 6 eventwill occur on each segment within the next thirty years, and a probab 0.33 that an M 6 will occur on one or more segments in that same interval of time.

A discussion of rupture likelihood, however, should be considered in the context of how failur of neighboring faults may influence the state of stress on the Calavens fauh. Since 183 5

- '~" .__-_m__ _____ _

t .. .

Af>7 earthquakes have occurred in the Bay area the 1836 and 1868 events on the northem and soutirrn liayward fauh, the 1906 San Andreas earthquake, and the 1989 Loma Prieta enthquake. De amotmt of stress change is a function of the amount and distribudon of slip postulated for these historic events and the distance from the Calaveras fauk. Consequently, the 1836 and 1868 earthquakes had a significant effect on the state of stress on the Cahveras fauh due to their proximity, but the 1906 earthquake also had a significant effect due to its greater unount of slip. He first-order effect of these earthquakes was to impose left lateral shear stirss on the northem Calaveras fanh segment, inhibiting the occurrence of earthquake activity on the fauk for perhaps as long as a centmy, depending on the slip-rate assumed for the fault (R.

Simpson, USGS, written commun.,1992). De tuodeling also indicates that the northern Calaveras fauh has probably recovered from the stress reduction of these earthquakes, consistent '

with the occunence of modcrate seismicity surrounding the northem Calaveras fauh since 1955 (rolid stan, Fig.1). Ellsworth et al. (1981) reached similar conclusions in their analysis of the seismicity following the 1906 earthquake. Recently, Oppenheimer et al. (1990) discussed the northward progression of moderate events on the Cahveras fauh since 1974 (solid star hbeled 1988 in Figure 1 is northenunost of these eveets) and speculated diat this progression might continue with a An event on the llayward fauh. It is unknown whether tids northward progression will lead to a faDure of the llayward fault or perhaps the northem Calaveras segment, but the halo of large earthquakes in the East Bay region suggests that the level of stress throughout the region may again be high enough to produce significant earthquakes on either of these fauhs.

CONCLUSIONS From our analysis and interpretation of the seismic, geodetic, and geologic data along the northc rn Calaveras fauh, we believe a seismic gap exists between the Calaveras Reservoir and the city of Danville. Several studies indicate that this section of the fauh has a slip rate of approximately 6 nun /yr. By comparison with the microseismic characteristics of the southem Calaveras fauh and the Loma Prieta segment of the San Andreas fault, we propose that the nonbern Calaveras fault is locked and accumulating elastic strain energy that could resuh in the occurrence of an earthquake as large as M 7. The limited 162-year history of earthquakes along p this seismic gap indicates an attemative rupture scenario in which this gap could ruptme in a series of smaller (Af 6+) earthquakes. Until geologic investigations reveal the rupture history of the northem Calaveras fault,it is not possible to determine reliable probabilities for the occunence of a Af 7 earthquake on this fauh segment.110 wever, we calcuhte the probability of one or more AI > 6 earthquakes occmring in the next 30 years to be 0.33. He assumptions required to calculate these probabilities resuh in considerable uncertainty, but we believe the evidence is sufficient to consider the northem Calaveras fault as a significant seismic hazard to the entire San Francisco Bay region.

ACKNOWLEDGMENTS We thank Fred Klein, Billixttis, David Schwartz, and Ivan Wong for their careful resiews of the manuscript, and Bob Simpson for providing us with the resuhs of his dislocation modeling. We also acknowledge the many people who have recorded and processed the earthquake data recorded by the Northem Califomia Seismic Network of the U.S. Geological Survey since 1969.

REFERENCES 6

I .

1

, , 3 Andrews, D.J., and D.H. Oppenbehner,1992, The Mission fauh: A magnitude 6+ thrust or resme slip earthquake can be expected, this vohune.

Bakun, W.II., G.C.P. King, and R.S. Cockerbam,1986, Seismic slip, Aseismic slip, and the mechanics of repeating earthquakes on the Calaveras Fault, Califonda, in Luthquake Source Mechanics, Geophysical Monograph 37, p.195 207, American Geophysical Union, Washington, D C.

Bolt, B.A., and RD Miller,1975, Catalogue of earthquakes in northern Califomia and adjoining areas, University of Califorma Press, Berteley.

Burford, RO., and R.V. Sharp,1982, Slip on the Ilayward and Calaveras faults determir.ed from offset powerlines, Confereace on Earthquake Hazards in the Eastem San Fnncisco Bay Area, Proceedings, rpecial Publication 62, p. 261-269, Califomia Depastment Of Conservation, Division Of Mines And Geology, San Fmacisco.

DeMets, C., RG. Gordon, D.F. Argus, and S. Stein,1990, Current plate motions, Geophysical Joumal of the Interior,101, p. 425-478.

Ellsworth, W.L., A G. Lindh, W.H. Prescott, and D.G. Herd,1981, ne 1906 San Fnncisco earthquake and the seismic cycle, in Eanbquake Prediction, an International Review, Geophysical Monograph 4, p.126-140, American Geophysical Union, Washington, D.C.

Galchouse, J.S.,101, Heodolite measurements of creep rates on San Fnncisco Bay region faults,11.S. Geological Survey, Open-File Repc'191 352, p. 375-384.

Harsh, P.W., and RO. Burford,1982, Alignment. array measurements of fault slip in the eastem San Francisco Bay area, Califomia, Conference on Eanbquake Hazards in the Eastem San Francisco Bay Area, Proceedings, Special Publication 62, p. 251-260, Califomia Department Of Conservation, Division Of Mines And Geology, San Francisco.

Hartzell, S.H., and T.H. Heaton,1986, Rupture history of the 1984 Morgan Hill, California, carthquake from the inversion of strong motion records, Bulletin of the Seismological Society of America, 76, p. 649474.

Jennings, C.W., RG. Strand, T.H. Rogers, M.C. Stinson, J.L. Bumett, J.E. Kahle, R. Streitz, and R.A. Switzer,1975, Fault Map of California, scale 1:750,000, Califomia Division of Mines and Geololgy, Califomia Geology Data Map Series, Map 1.

Kelson, K.I., W.R. Lettis, and G.D. Simpson,1992, Late Holocene paleeseismic events at ,

L Leyden Creek, northern Calaveras fault, this volume.

Klein, F.W.,1989, User's guide to HYPOINVERSE, a prognun for VAX computers to solve for earthquake locations and magnitudes, U.S. Geological Survey, Open-File Report 89-314, 44 p.

' Lienkaemper, J.J., G. Borchardt, and M. Lisowski,1991, Historic creep rate and potential for t.eismic slip along the Hayward fault, Califomia, Journal of Geophysical Research, %, p.

18,261-18,283.

Lindh, A.G.,1988, Estimates oflong-tenn probabilities for large earthquakes along selected fauh segments of the San Andreas fault system in Califomia, in S.K. Guha and A.M. Patwardhan, editors, Earthquake Prediction - Present Status, Pune. India, University of Poona, p. 189-200.

7

y. - - - _ _ _ - . - - - - - - - - - - _ _

e

. . . 5 Lisowski, M., W.H. Prescott,, J.C. Savage, and M.J. Johnston,1990, Geodetic estimate of coseismic slip during the 1989 Loma Prieta, Cahfomia, esthquake, Geophysical Research Letters,17, p.1437-1440.

Lisowski, M., J.C. Savage, and WJf. Prescott,1991, 'nie velocity field along the San Andreas fauh in central and southern California, Joumal of Geophysical Research, %, p. 8 69-8389.

Mosier, J.B.1977, Results of triangulation for earth movernent study at Califomia aqueduct fault crossing sites: Section VI, north and south sites, U.S. Department of Commerce, NOAA, National Geodetic Survey,28 p.

Oppenheimer, DJI.,1990, Aftershock slip behavior of the 1989 lema Prieta, Califomia earthquake, Geophysical Research Letters,17, p. I 199-1202.

Oppenbehner, D.H. and N. Macgregor-Scott,1992, The seismotectonics of the eastern San Francisco Bay region, this volume. -

Oppenheimer, D.H., W.H. Bakun, and A.G. Lindh,1990, Slip partitioning of the Calaveras fauh, Califomia, and prospects for future earthquakes, Joumal of Geophysical Research,95, p.

8483-8498.

Prescott, W.H., and M. Lisowski,1983, Strain accumulation along the San Andreas fault system east of San Francisco Bay, Califomia, Tectonophysics, 97, p. 41-56.

Prescott, W.H., M. Lisowski, and J.C. Savage,1981, Geodetic measurement of crustal deformation on the San Andreas, Hayward, and Calavens fauhs near San Francisco Bay, Joumal of Geophysical Research, 86, p.10853-10869 Reasenberg, P.A., and D.ll. Oppenheimer,1985, FPFIT, FPPLOT, and FPPAGE: FORTRAN computer programs for calculating and displaying carthquake fault plane solutions, U.S.

Geological Survey, Open-File Report 85-739,109 p.

Savage, J.C., WJL Prescott, M. Lisowski, and N. King,1979, Geodolite measurements of defonnation near llollister, Califomia, Joumal of Geophysical Research,84, p. 7599-7615.

[

Schwartz, D.P., and K.J. Coppersmith,1984, Fault behavior and characteristic earthquakes:

Examples from the Wasatch and San Andreas fauh zones, Joumal of Geophysical Research,89, p 5681-5698.

Shnpson, G.D., W.R. Lettis, and K.L Kelson,1992, Segu.entation model for the northem Calaveras fauh, Calaveras Reservoir to Walnut Creek, this volume.

Toppozada, T.R., C.R. Real, and D.L Parke,1981, Preparation of isoseismal maps and summaries of reported effects for pre-1900 Califomia earthquakes, Califomia Division of Mines and Geology Open File Report 81-11 SAC,182 p.

Wong, LG., and M.A. Hemphill-Haley,1992, Seismicity and fauhing near the Hayward and Mission rauhs,this vohune.

Working Group on Califomia Earthquake Probabilities (Agnew, D.C., C.R. Allen, LS. Cluff, J.H. Dieterich, W.L. Ellsworth, R.L Keeney, A.G. Lindh, S.P. Nishenko, D.P. Schwartz, K :

Sieb, W. 'Hiarcher, R.L Wesson),1988, Probabilities oflarge earthquakes occurring in Califomia on the San Andreas fauh, U.S. Geological Smvey Open-File Report 88 398,62 p.

8 i

r; -

. ' I Working Group on California Earthquake Probabilhies (J.II. Dieterich, C.R. Allen, L.S. Cluff, C.A. Comell, W.L Ellsworth, LR. Johnson, A.G. Lindh, S.P. Nishenko, C.II. Scholz, D.P.

Sciswartz, W. natcher, P.L Williams),1990, Probabilities oflarge earthquakes in the San Francisco Bay Region, California, U.S. Geological Survey Circular 1053,51 p.

FIGURE CAPTIONS Figure 1. Location of Quatemary faults in the castem San Francisco Bay region (Jennings et al.,

1975) and emthquakes (stars) above M 5.0 since 1830. Instnunentally located earthquakes are shown as solid stes. Earthquake locations and magnitudes (in parentheses) are from Topporada et al. (1981), Doh and Miller (1975), and the USGS. Rectangular box corresponds to region shown in Figure 4. CCF-Concord fault, CVF-Calavera1 fauh, GVF-Green Valley fault.

!!WF-Ilayward fauh, FNF-Franklin fauh, CR Calaverr Reservoir, DV-Danville, LV-Livennore, ML-Mount 12wis,ILPleasanton, SB-StJsun Bay, SFB San Francisco Bay.

Figure 2. Seismicity recorded by the USGS from January,1%9 through January,1992. He earthquake locations (Klein,1989) have an RMS travehime enor kss thm 0.3 s, at least 6 P-wave arrival times, and horizontal and vertical uncertainties of 2.5 and 5.0 km, respectively.

'Ibe dashed polygon from A to A' depicts the selection region for the cross section shown in Figure 3. Itachured region indicates extent of hypothesized gap. See Figure 1 for fauh names and locations of cities cited in text.

Figure 3. Cross section of seismicity within polygon of Figure 2 along the Green Valley -

Concord - Calaveras fauh systeau. 'Ihere is no vertical exaggerstion. Seismicity gap extends froru Calaverns Reservoir to approximately Danville (~40 km). Comparison with map view of seismicity in Figure 3 shows that inost of the seismicity within the gap occurs adjacent to the Calaveras fauh. In particular, the seismicity cluster near Danville occurs 5ithin a step.over region to the Concord fauh, the chister at 82 km occurs east of Calaveras fauh, and the seismicity frt>m 86-90 km ocaus in the vicinity of the Mission fauh. IIachured area indicates area of fauh crpected to rupture in a moderate-to-large cathquakes.

Figure 4. Lower hemisphere, equal area focal mechanisms calculated fran P-fat motions (Reasenberg and Oppetiheimer,1985) for earthquakes within polygon shown in Figure 2 and rectangular region shown in Figure 1. All mechanisms have at least 25 frstanotion readings.

Compressional quadrants are shaded solid. Earthquake locations are depicted by small dots.

9 L

l

.... .a..i.... ....l.........i.........i.........

b 10' \%4 .

\ \

/ -

y& / '1 -

. A 3 _

38 W 18s (s.o) -

A 1955 (5.4) isto(5.3)

% s

\ -

1990 .1)  :

50' gy nao(5.8) _

} "80 (51) igg,(s.,8gg

( sesi (5.s) g {

\

- k g LV 1980 (5.4) s

\ \ -

5 1868 (6. 2'

., \ -

~

40' l SFB Wi h 7) 4 x' -

A 18 (5.3 1 903 (5.5)

' 58 (8- ) ML 30' i A1 56(.. - _

. Y\

• 1986 (5.7)  :

\ ,

\ . .

1o ggh 1988 (5.3)  :

20'

.\

, . . . i . . . . i . . . .

i ........i

. 955 (5.5)

. . \. . . . . . . i . . . . . . . . .

20' 10' 122 50' 40' 30' 4

F73 1.

p m u.re........il m....... ....... . . ..... .

(.p( 's

• MAGS

s4 \ \ # + '+

10' i \+1+'s i 5 ig 4 -

• 4.5+

' >* 1

+

/ - MAGNITUDES

. \

p

'+ ;' ..

0.0 +

3'B' '

,+

+

j*- -

s

\

+7

+

+

+

+

1.0 +

2.0+

.- +

,p \

4 * +++

+ 3.0+

2 , g j , ,

[ + + }+ + 3 3

+  %

, f 4.0 +

+- ' 5.0 +

50' i

-+ s *

. . j, * *I '+ t

+

5 *[ A c's.. J W+ t'

• g+

eC */ \

+ 4 40' -

• 4 .

s

• 1. ,A-[#s;.*+.g.N T  :

' *+

~

!+ ,

+

- + +

30' ( #f + g + -

L

' s.-

+- 4 s .

.~.

. .;s+

+

., .i.

+

Y + . . .k . . .

l- 20' . . Y' > . A . . . - . . . . . , . . . . ' . . . .$ . . . . . r c . . , . 3 . ? . *.

. 20' 10' 122' 50' TO' 30' F7s . "

t l

l- .

~

r' .,

. '_, 1 N.

(E.

g -i.

. .g 4

q*

. L. -

P

5

,I,

- * ' k i }: .b i kAt. .k-fhj;.  :. E u

b .-

is

wn

, y E

R ++ .

g

.g- .

s.te .

i$:

+*'

% . o e

4 #+O t Y+ "

. . . . *o

- ++ h O

g;: *. . o

~ *"

++

se

? ++

i w..

.g i.. $ ,,,,,,

o 2 i i (KX) iud 30

- -. . . . , . - ~ - - ~ . - . - ~

s a_ - -.

-4 .(-

t- s .

' (.Al A O. - O. -

3 's A B l A i 1 d A A 8 L  !

[- - g* .

W f .-

. . s -

.& .) .s.'* t .ds '.

=

,.Q.T..6% .I .. . .., .

• . : ' .p "* ..

. ~X i .

a . .

N- , *g/ *;

• a 0

~

Q s. r, * * *

• o . : '. : . .Q .

.(*

a.~,. , p' i rs.: . . .' . .

. ;aw . .

(p.) . /

. - ~

i.. .

h

., i '. '

p -

,, , ,j.k). ..

..y ,, ,i,w . , .

.. 'y , , , ,

. A1,,4 , s. .' ... . * ... . *

• 7 ,,..

~

% * \ *.' .

.f . ,

e .

-e e 'e e

.g*,, , g

. ,. g ' 8 8,.

. l .

. ~.

_ (,j1 = y /<  ;,i  ;

"3

.g-

..*i k , . .j,., , .. -

- * . . . v *# e1.. ., .

• 9

?; ',

.y .. -. , ' , . -'

, q, .  ; e-

..,s

, . s . .

s'.'... f

- .r .

. . . ' +.* . s ei

, , , . . ," .. . . . .. . f' . ... , .

.  : a e '

s .

• E . , ,8% *'.' '. p ), .

. , ."7 ,' j_ 8 83 ,,

e a e i  ; e i i s g i e i i I

(L l

h'

~

.C e

r