Content uploaded by Allan Lindh
Author content
All content in this area was uploaded by Allan Lindh on Sep 01, 2015
Content may be subject to copyright.
A preview of the PDF is not available
Probabilities of Large Earthquakes in the San Francisco Bay Region, California
Abstract and Figures
In 1987 a Working Group on California Earthquake Probabilities was organized by the U.S. Geological
Survey at the recommendation of the National Earthquake Prediction Evaluation Council (NEPEC). The
membership included representatives from private industry, academia, and the U.S. Geological Survey. The
Working Group computed long-term probabilities of earthquakes along the major faults of the San Andreas
fault system on the basis of consensus interpretations of information then available. Faults considered by the
Working Group included the San Andreas fault proper, the San Jacinto and Imperial-faults of southern
California, and the Hayward fault of northern California. The Working Group issued a final report of its
findings in 1988 (Working Group, 1988) that was reviewed and endorsed by NEPEC.
As a consequence of the magnitude 7.1 Loma Prieta, California, earthquake of October 17, 1989, a
second Working Group on California Earthquake Probabilities was organized under the auspices of NEPEC.
Its charge was to review and, as necessary, revise the findings of the 1988 report on the probability of large
earthquakes in the San Francisco Bay region. In particular, the Working Group was requested to examine the
probabilities of large earthquakes in the context of new interpretations or physical changes resulting from the
Loma Prieta earthquake. In addition, it was to consider new information pertaining to the San Andreas and other faults in the region obtained subsequent to the release of the 1988 report. Insofar as modified techniques and improved data have been used in this study, the same approach might also, of course, modify the probabilities for southern California. This reevaluation has, however, been specifically limited to the San Francisco Bay region.
This report is intended to summarize the collective knowledge and judgments of a diverse group of
earthquake scientists to assist in formulation of rational earthquake policies. A considerable body of information about active faults in the San Francisco Bay region leads to the conclusion that major earthquakes are likely within the next tens of years. Several techniques can be used to compute probabilities of future earthquakes, although there are uncertainties about the validity of specific assumptions or models that must be made when applying these techniques. The body of this report describes the data and detailed assumptions that lead to specific probabilities for different fault segments. Additional data and future advances in our understnding of earthquake physics may alter the way that these probabilities are estimated. Even though this uncertainty must be acknowledged, we emphasize that the findings of this report are supported by other lines of argument and are consistent with our best understanding of the likelihood for the occurrence of earthquakes in the San Francisco Bay region.
Figures - uploaded by Allan Lindh
Author content
All figure content in this area was uploaded by Allan Lindh
Content may be subject to copyright.
... El primero fue encontrar fallas colombianas que tuvieran el mismo tipo de movimiento y la misma longitud que las fallas bien caracterizadas; esto debido a que Wells & Coppersmith (1994) señalaron que existe una relación matemática con la cual se puede obtener la magnitud de un sismo en función del tipo y longitud de la falla que lo genere, paso denominado analogía geológica. El segundo paso es la analogía sismológica, que consiste en encontrar la magnitud máxima probable de un sismo para cada una de las fallas estudiadas de conformidad con la ley de Recurrencia de Gutenberg-Richter o mediante las relaciones de magnitud-área de Hank & Bakun (2002) y Ellsworth (2003). ...
... Como se indicó anteriormente, la analogía sismológica se realiza teniendo en cuenta la magnitud máxima de un sismo para las fallas estudiadas. El cálculo de la magnitud máxima se puede hacer de dos maneras: para fallas que no se encuentran entrelazadas con otras fallas este valor se calcula por medio de la ley de recurrencia de Gutenberg-Richter, mientras que para las fallas que sí están unidas a otras fallas el cálculo de la magnitud máxima probable de un sismo se hace por medio de las relaciones de magnitud-área de Hank & Bakun (2002) y Ellsworth (2003). Según esto, sólo para cinco fallas bien documentadas fue posible el uso de la ley de recurrencia Gutenberg-Richter porque su comportamiento era independiente de las fallas vecinas; el resto de las fallas bien documentadas se analizaron por medio de las relaciones magnitud-área. ...
Los estudios de amenaza sísmica en Colombia cuentan con vacíos de información debido a dos causas principales: instrumentación sísmica insuficiente en el tiempo y en el espacio, y desconocimiento del com-portamiento de las fallas geológicas. Por esta razón, se caracterizaron ciertas fallas que afectan ciudades de Colombia por medio de una analogía geológica y sismológica, con algunas de las fallas geológicas mejor caracterizadas del mundo. Esta analogía se realizó teniendo en cuenta las relaciones empíricas de magnitud-geometría desarrolladas por Wells & Coppersmith (1994), por medio de las cuales puede esti-marse la magnitud máxima de un sismo en función de la longitud y el tipo de la falla que lo produce. Por otro lado, desde el punto de vista sismológico, la estimación de las magnitudes máximas de los sismos producidos por las fallas estudiadas se realizó por medio de la ley de recurrencia de Gutenberg-Richter y las relaciones magnitud-área de Hanks & Bakun (2008), y Ellsworth (2003). De esta manera, las fallas colombianas que se caracterizaron por medio de la analogía son Buesaco, Cimitarra, Cocora, Cucuana, Garrapatas, Irlanda, Oca, Potrerillos, Sibundoy y Vianí; y las fallas bien documentadas utilizadas en esta analogía fueron Shelter Cove,
... These intensities are derived from ShakeMap scenario calculations (WGCEP, 2003). The gray regions in Figure 2(A) represent earthquakes for which shaking intensity at the Civic Center is less than V on the Modified Mercalli Intensity (MMI) scale (Richter, 1958) and there is unlikely to be damage. Above MMI V the likely damage increases with the severity of shaking, from light (V: unstable objects displaced), to strong (VII: broken furniture and damage to masonry), to violent (IX: masonry seriously damaged or destroyed, frames displaced from foundations). ...
The geological, geomorphological and seismotectonic characteristics of Australia’s North West Shelf (NWS) and adjoining eastern Sunda and western Banda arcs were investigated to examine the nature of tectonic processes and the geological and seismological hazards that occur in this region. The NWS previously has been considered “geologically stable” and part of the Australian Stable Continental Region. However, the results of this study demonstrate that this part of the Australian continental margin is tectonically active and there is a broad range of dynamic geological hazards.
Key outcomes of this research show that active faults extend at least 2,000 km along the northwestern continental margin of Australia and these faults pose significant seismic hazards. For the first time in Western Australia there are sufficient data to model fault sources using the probabilistic seismic hazard analysis (PSHA) approach. However, there are alternative approaches to modelling seismic sources that can produce significantly different ground motion results. The PSHA results from two alternative modelling approaches were compared to assess the differences in Peak Ground Acceleration (PGA) at a range of return periods. Model A uses only areal source zones and Model B uses areal source zones and fault sources on the NWS. For Model A the peak ground acceleration (PGA) values for 1,000, 5,000, and 10,000-year return periods are approximately 0.1 to 0.14g, 0.28 to 0.35g, and 0.38 to 0.45g, respectively. Model B produced changes to the spatial pattern and amplitude of PGA within about 40 km of a fault. In the near field, Model B yields PGA values 1.2, 1.6 and 2.0 times the Model A values at the 1,000, 5,000, and 10,000-year return periods, respectively.
The PGA values determined using the PSHA approach are of sufficient amplitude to be a trigger for submarine landslides on susceptible slopes. This implies that seismically induced slope failures have the opportunity to occur at frequencies of 10-3 to 10-4 per year. The occurrence of geological hazards at these frequencies is significant in a risk-based framework.
On 24 September 2013, an Ml 3.6 earthquake struck in the Gulf of Valencia (Spain) near the Mediterranean coast of Castelló, roughly 1 week after gas injections conducted in the area to develop underground gas storage had been halted. The event, felt by the nearby population, led to a sequence build-up of felt events which reached a maximum of Ml 4.3 on 2 October.
Here, we study the role of static stress transfer as an earthquake-triggering mechanism during the main phase of the sequence, as expressed by the eight felt events. By means of the Coulomb failure function, cumulative static stress changes are quantified on fault planes derived from focal mechanism solutions (which act as both source and receiver faults) and on the previously mapped structures in the area (acting only as stress receivers in our modeling). Results suggest that static stress transfer played a destabilizing role and point towards an SE-dipping structure underlying the reservoir (or various with analogous geometry) that was most likely activated during the sequence. One of the previously mapped faults could be geometrically compatible, yet our study supports deeper sources. Based on this approach, the influence of the main events in the occurrence of future and potentially damaging earthquakes in the area would not be significant.
Northern Taiwan offers a rare opportunity to quantify the rate and style of deformation in the transition zone, from the collision in the southeast, subduction, rifting, to orogenic collapsing in the northwest. The updated GPS velocity field collected from 2002 to 2014 and leveling vertical velocity field from 2004 to 2010 were inverted by a 3-D block model in this study to infer the tectonic block rotations and fault slip rates in reference to the Eurasian plate. Our optimized modeling results show that the clockwise rotation rate of 20.8° ± 2.8°/Myr in the Hualien block is caused by the coupling of the waning arc-continent collision between the Luzon volcanic arc and the Eurasian plate, and the northeastward migration of the block toward the Ryukyu trench. In Ilan domain, the rapid clockwise motion was derived at a rate of 34.9° ± 65.9°/Myr in southern part of the plain. We proposed that the rapid clockwise rotation results from the asymmetry rifting at the southwestward extension of the Okinawa trough. An another new key finding is highlighted in the capability of maximum magnitude ∼6.2 is indicated on the Suao and the Choshui faults in the Ilan plain.
Northern Attica in Greece is characterized by a set of north dipping, subparallel normal faults.
These faults were considered to have low tectonic activity, based on historical earthquake
reports, instrumental seismicity and slip rate estimates. This study presents new data for one
of these faults, the Milesi Fault. We run GIS based geomorphological analyses on fault offset
distribution, field mapping of postglacial fault scarps and ground penetrating radar profiling to
image hangingwall deformation. The first palaeoseismological trenching in this part of Greece
allowed obtaining direct data on slip rates and palaeoearthquakes. The trenching revealed
downthrown and buried palaeosols, which were dated by radiocarbon. The results of our
investigations show that the slip rates are higher than previously thought and that at least four
palaeoearthquakes with magnitudes of aroundM6.2 occurred during the last 4000–6000 yr.We
calculate an average recurrence interval of 1000–1500 yr and a maximum throw rate of ∼0.4–
0.45 mm a−1. Based on the new geological earthquake data we developed a seismic hazard
scenario, which also incorporates geological site effects. Intensities up to IX must be expected
for Northern Attica and the southeastern part of Evia. Earthquake environmental effects like
liquefaction and mass movements are also likely to occur. This scenario is in contrast to the
official Greek seismic hazard zonation that is based on historical records and assigns different
hazard zones for municipalities that will experience the same intensity by earthquakes on the
Milesi Fault. We show that the seismic hazard is likely underestimated in our study area and
emphasize the need to incorporate geological information in such assessments.
Northern Attica in Greece is characterized by a set of north dipping, subparallel normal faults. These faults were considered to have low tectonic activity, based on historical earthquake reports, instrumental seismicity and slip rate estimates. This study presents new data for one of these faults, the Milesi Fault. We run GIS based geomorphological analyses on fault offset distribution, field mapping of postglacial fault scarps and ground penetrating radar profiling to image hangingwall deformation. The first palaeoseismological trenching in this part of Greece allowed obtaining direct data on slip rates and palaeoearthquakes. The trenching revealed downthrown and buried palaeosols, which were dated by radiocarbon. The results of our investigations show that the slip rates are higher than previously thought and that at least four palaeoearthquakes with magnitudes of aroundM6.2 occurred during the last 4000-6000 yr.We calculate an average recurrence interval of 1000-1500 yr and a maximum throw rate of ~0.4- 0.45 mm a-1. Based on the new geological earthquake data we developed a seismic hazard scenario, which also incorporates geological site effects. Intensities up to IX must be expected for Northern Attica and the southeastern part of Evia. Earthquake environmental effects like liquefaction and mass movements are also likely to occur. This scenario is in contrast to the official Greek seismic hazard zonation that is based on historical records and assigns different hazard zones for municipalities that will experience the same intensity by earthquakes on the Milesi Fault. We show that the seismic hazard is likely underestimated in our study area and emphasize the need to incorporate geological information in such assessments.
New mapping of two active transpressional fault zones in the California Continental Borderland, the Santa Cruz‐Catalina Ridge fault and the Ferrelo fault, was carried out to characterize their geometries, using over 4500 line‐km of new multibeam bathymetry data collected in 2010 combined with existing data. Faults identified from seafloor morphology were verified in the subsurface using existing seismic reflection data including single‐channel and multichannel seismic profiles compiled over the past three decades. The two fault systems are parallel and are capable of large lateral offsets and reverse slip during earthquakes. The geometry of the fault systems shows evidence of multiple segments that could experience throughgoing rupture over distances exceeding 100 km. Published earthquake hypocenters from regional seismicity studies further define the lateral and depth extent of the historic fault ruptures. Historical and recent focal mechanisms obtained from first‐motion and moment tensor studies confirm regional strain partitioning dominated by right slip on major throughgoing faults with reverse‐oblique mechanisms on adjacent structures. Transpression on west and northwest trending structures persists as far as 270 km south of the Transverse Ranges; extension persists in the southern Borderland. A logjam model describes the tectonic evolution of crustal blocks bounded by strike‐slip and reverse faults which are restrained from northwest displacement by the Transverse Ranges and the southern San Andreas fault big bend. Because of their potential for dip‐slip rupture, the faults may also be capable of generating local tsunamis that would impact Southern California coastlines, including populated regions in the Channel Islands. High‐resolution bathymetry delineates major active seafloor fault zonesTranspression along steep faults produces basin inversion and basement upliftActive offshore fault systems may sustain long complex fault rupture
WGCEP90 estimated the Hayward fault to have a high probability (0.45 in 30 yr) of producing a future M7 Bay Area earthquake. This was based on a generic recurrence time and an unverified segmentation model, because there were few direct observations for the southern fault and none for the northern Hayward fault. To better constrain recurrence and segmentation of the northern Hayward fault, we trenched in north Oakland. Unexpectedly, we observed evidence of surface rupture probably from the M7 1868 earthquake. This extends the limit of that surface rupture 13 km north of the segmentation boundary used in the WGCEP90 model and forces serious re-evaluation of the current two-segment paradigm. Although we found that major prehistoric ruptures have occurred here, we could not radiocarbon date them. However, the last major prehistoric event appears correlative with a recently recognized event 13 km to the north dated AD 1640–1776.
We adopt a lognormal distribution for earthquake interval times, and we use a locally determined rather than a generic coefficient of variation, to estimate the probability of occurrence of characteristic earthquakes. We extend previous methods in two ways. First, we account for the aseismic period since the last event (the “seismic drought”) in updating the parameter estimates. Second, in calculating the earthquake probability we allow for uncertainties in the mean recurrence time and its variance by averaging over their likelihood. Both extensions can strongly influence the calculated earthquake probabilities, especially for long droughts in regions with few documented earthquakes. As time passes, the recurrence time and variance estimates increase if no additional events occur, leading eventually to an affirmative answer to the question in the title. The earthquake risk estimate begins to drop when the drought exceeds the estimated recurrence time. For the Parkfield area of California, the probability of a magnitude 6 event in the next 5 years is about 34 per cent, much lower than previous estimates. Furthermore, the estimated 5-year probability will decrease with every uneventful year after 1988. For the Coachella Valley segment of the San Andreas Fault, the uncertainties are large, and we estimate the probability of a large event in the next 30 years to be 9 per cent, again much smaller than previous estimates. On the Mojave (Pallett Creek) segment the catalog includes 10 events, and the present drought is just approaching the recurrence interval, so the estimated risk is revised very little by our methods.
The 430-km-long segment of the San Andreas fault that ruptured in the great 1906 earthquake (M 8) has been seismically quiet at the M 6 level for the last 75 years (Ellsworth et al., 1981). The historic record of earthquakes in the 19th century for the southernmost 100 km of the rupture, extending from San Francisco to just north of San Juan Bautista, indicates that it was the site of three or four M 6 or greater earthquakes in the 110 years preceding 1906: 1800?; 1838, M 7?; 1865, M 6.5; 1890, M 5.9 (Toppozada, 1980). We believe that these earthquakes are plausibly accounted for by repeated slip on the southernmost 50 km of fault, just to the northwest of San Juan Bautista .
The cyclic nature of strain accumulation and release in great earthquakes on the San Andreas fault appears to have modulated the historic pattern of seismicity in northern coastal California. The region experienced many more strong earthquakes in the half century preceding the 1906 earthquake than in the half century following it. Fedotov and Mogi recognized a similar seismic cycle accompanying great earthquakes in the subduction zones of the Western Pacific. The cycle consists of an extended period of quiescence after a major earthquake, followed by a period of increased activity that leads to the cycle-controlling earthquake and its foreshocks and aftershocks. The timing of the next great earthquake cannot be forecast with precision at present, although it appears to be decades away. The reemergence of M≥5 earthquakes since 1955 within the latitudes of the 1906 surface rupture to the north of San Jose, after an extended period of quiescence following 1906, lead us to conclude that the region is entering the active stage of the cycle in which events as large as M 6, to perhaps M 7 can be expected.
A simple statistical model is presented for the recurrence of Characteristic Earthquakes on specific segments of the San Andreas Fault System. The model is the simplest possible, and assumes that after the occurrence of a Characteristic Event on a given segment, the expected time of the next event on that segment is a random variable with a Gaussian distribution. The mean recurrence time is just the ratio of the strain accumulation rate to the strain drop in the last event. The variation about that mean time is a function of the information available on average recurrence on the given segment, plus any variation due to irregularities in the failure process itself. The model assumes that there is a characteristic stress threshold for each segment at which failure is likely to occur (Byerlee and Brace. 1968 ), and as such is an explicit application of the elastic rebound theory of Gilbert ( 1884 ) and Reid ( 1910 ). In asserting that earthquake recurrence is, on the average, a quasi-periodic process, the model can be considered a stochastic variant of the time-predictable model of Shimazaki and Nakata ( 1980 ).
This is an account of the effects of the most costly earthquake to hit the United States since 1906, which took place in the Santa Cruz Mountain segment of the San Andreas Fault, some 25 miles long. The southwest side moved 6 feet northwestward and 4 feet upward in relation to the northeastern side. It was thus rather different from typical San Andreas displacements and was also distributed over a wide band. Most of the damage occurred in areas of unconsolidated sand and mud. Detailed information of the shock, its extent, aftershocks and surface displacements are also given, with impressive illustrations of building damage, the effects of ground liquefaction, landslides and highway deformation. Some future dangers are emphasised, especially since the lessons that could have been learned from the 1906 earthquake have been ignored. -A.Scarth
The fault has long been recognized as the source of the destructive San Francisco earthquake of 1906 and of the similarly large Fort Tejon earthquake of 1857, as well as the smaller (M=7.1), but also destructive, Loma Prieta earthquake of 1989. This volume is addressed to a varied audience, but especially to earth scientists who wish to gain a brief overview of the San Andreas fault system. The 10 chapters review geologic, geomorphic, geophysical, geodetic, and seisologic information about the San Andreas fault system. -from Author