New Southern California Earthquake Probabilities

by Kenneth Campbell, Thomas Heaton, and Ronald Eguchi

Recent Southern California earthquakes have caused experts to increase their probability estimates for a large event in that region. New methods of rapid earthquake notification in conjunction with rapid damage assessment tools are being developed to aid post-earthquake response and recovery.

Introduction

In June of 1992, the Landers and Big Bear earthquakes rocked the Southern California areas of Yucca Valley and Big Bear Lake. Although these earthquakes were felt throughout Southern California, they caused relatively little damage. Scientists believe that the significance of these earthquakes extends well beyond the damage that did occur.

As a result of the earthquakes, scientists from various California agencies and universities and the U.S. Geological Survey (USGS) have increased their probability estimates for a large (M7+) earthquake in the greater Southern California area to be as high as 12% in the next year. Prior to the Landers Earthquake, the estimated annual probability was 4%.

Fault structures that are likely candidates for a large earthquake include the Mojave shear zone northeast of Los Angeles, the San Bernardino Mountains and Coachella Valley segments of the San Andreas Fault, and the northern San Jacinto Fault.

If a large earthquake (M8+) were to occur along the southern stretch of the San Andreas Fault, the expected number of casualties and damaged buildings could be enormous. In 1980, reports by the Federal Emergency Management Agency (FEMA) estimated these totals for a late afternoon earthquake to be 14,000 dead; 55,000 seriously injured; and $17 billion in building and contents damage. Current studies are now attempting to update these estimates, based on more accurate data on building exposure and vulnerability.

In order to mitigate the damage potential, scientists at the California Institute of Technology (Cal Tech) and the USGS in Pasadena are preparing for large earthquakes. They have set up a special system that will calculate and broadcast the magnitude and location of moderate to large earthquakes in the Southern California area within minutes after the event. If incorporated into damage assessment models, these data can identify areas of severe building damage and large numbers of casualties. Having such information in the immediate hours after an earthquake would facilitate effective response and recovery efforts.

Landers and Big Bear Earthquakes

At 4:58 A.M. on June 28, 1992, the most powerful earthquake (M7.6) to occur in California in the last 40 years rumbled through Southern California. The size of this shock makes it the third-largest California earthquake in this century. The epicenter was located near the small town of Landers, approximately 110 miles east of Los Angeles; 30 miles north of Palm Springs; and 15 miles northwest of the April 23, 1992, Joshua Tree Earthquake (M6.3).

At 8:04 A.M., a second earthquake (M6.7), centered near Big Bear Lake in the San Bernardino Mountains, struck approximately 20 miles west of the initial shock. Because of the location and timing of this second earthquake, scientists have classified it as an aftershock of the Landers event.

In total, these two earthquakes were responsible for one death, over 400 injuries, and approximately $100 million in damage to property, roads, and water pipeline systems. Hardest hit were the desert communities of Landers, Yucca Valley, and Joshua Tree, as well as Big Bear in the mountains.

Ad Hoc Working Group

In response to the Landers Earthquake, an ad hoc working group was formed to evaluate how the Landers-Big Bear earthquake sequence might have affected future occurrences of large earthquakes in the Southern California area. The working group included members of the National Earthquake Prediction Evaluation Council, the California Earthquake Prediction Evaluation Council, and the Southern California Earthquake Center.

Specifically, this group was asked to assess the likelihood of a large earthquake in the next few years, and compare the resulting probabilities with earlier estimates made by members of the Working Group on California Earthquake Probabilities. In essence, there were four factors that led the ad hoc working group to increase the earlier probability estimates.

1. Since 1985, there has been a significant increase in the number of earthquakes recorded in Southern California, particularly for larger-magnitude events (M>5)(see graph on page 13). There also appears to be a shift in the location of these events—from the Imperial Valley and along the San Jacinto Fault to the San Gabriel Valley and close to the San Andreas Fault.

2. A comparison of the mean earthquake recurrence interval with the time since the last event shows that several segments of the San Andreas Fault are either "due" or "overdue" for a large event.

3. Because of the orientation of the fault rupture in the Landers-Big Bear earthquake sequence, it appears that stress has increased on portions of the San Andreas and San Jacinto faults. Calculations suggest that while stress on the Mojave segment of the San Andreas Fault may have been relieved, stress on the San Bernardino Mountains and Coachella Valley segments has probably increased, pushing them closer to failure. In addition, stress has probably increased on the San Jacinto Valley segment of the San Jacinto Fault.

4. Finally, the aftershocks that followed the Landers-Big Bear earthquakes migrated toward the San Andreas Fault. A few aftershocks were actually observed within the mapped fault zone near Yucaipa, and others were within several miles of the fault near Desert Hot Springs.

The ad hoc group's estimates suggest that the probability of a large (M7+) earthquake in Southern California is somewhere in the range of 5% to 12% in the next year, depending on the specific assumptions used in the calculations. Based on a 200-year recorded history in Southern California, the yearly probability of a large earthquake should be about 4%.

The ad hoc group further concluded that if the overall high rate of seismic activity experienced since 1985 persists, there is a 47% chance of a magnitude 7+ earthquake in Southern California within the next five years. Unfortunately, there is no real way to predict when we will return to a period of more normal (lower) seismic activity.

In light of these increased probability estimates, the ad hoc working group made two recommendations for further action.

1. The California Office of Emergency Services (OES) should intensify loss reduction and public information efforts.

2. The California OES should plan for a magnitude 6 or greater earthquake on the San Andreas Fault.

The second recommendation does not mean that such an earthquake will definitely occur in the near future, but the likelihood of such an occurrence is high enough to warrant the response plans' being in place. As part of these response plans, it is recognized that a magnitude 6 event on the San Andreas Fault could be a foreshock to a much larger earthquake. Therefore, if this event occurs, the California OES is prepared to issue an alert indicating that a larger earthquake could follow within days.

Regional Impact of a Large San Andreas Earthquake

There have been numerous studies over the past 20 years attempting to quantify the regional impact of a large San Andreas earthquake. Early studies performed for FEMA (1980) suggest that depending upon the time of day, the number of fatalities in a large (M8.3) San Andreas event could range from 3,000 to 14,000, and the number of hospitalized individuals could be as high as 55,000. These estimates are considered to be upper bounds because of the conservative assumptions used in characterizing the seismic vulnerability of Southern California buildings.

A more recent report, published by the Southern California Earthquake Preparedness Project (1987), estimates that the amount of damage to buildings in the city of Los Angeles alone could exceed $1.1 billion from a large San Andreas earthquake. This report also indicates that the number of uninhabitable living units could exceed several thousand.

Another study, conducted by the California Division of Mines and Geology (1982), estimates that a large portion of the infrastructure may be impaired in a San Andreas earthquake. Aqueducts transporting significant portions of the area's water supply cross the San Andreas Fault near Tejon Pass, and are expected to be severed and unavailable for up to six months after a large earthquake on the Mojave segment of the fault. Electric power outages immediately after the earthquake could result in a 50% loss in service. Major highway corridors, particularly through the mountains, are expected to be impacted during the initial post-earthquake period.

Improving Post-earthquake Response and Recovery

In order to enhance regional response and recovery from a large earthquake in the Southern California area, several agencies are creating post-earthquake assessment tools.

Cal Tech and the USGS have developed CUBE (Cal Tech/USGS Broadcast of Earthquakes), a joint project to systematize rapid earthquake notification. About a dozen other agencies (utilities, emergency services agencies, and railroads) are collaborating on this joint development project.

After an event, each participant receives earthquake epicentral information over a pager. A direct radio link has been established between CUBE computers and the paging system, thus ensuring reliable communications even in the event of telephone failure.

Messages are sent for all earthquakes larger than magnitude 3.5 to individuals who carry pocket pagers, and messages are sent for all located earthquakes (larger than about M1.8) to pagers connected to DOS-based PCs. The PC system captures the pager information and displays it on a map, which contains demographic and infrastructure information.

Data from 64 telemetered seismic stations are continuously analyzed to detect and locate earthquakes. A second generation of the system is currently being developed, which will use data from more than 220 stations. Fifteen strong-motion accelerometers are also monitored by the system to determine the size of larger earthquakes. The system is currently being modified so that ground motion information from these telemetered sites, such as peak acceleration and velocity, will also be transmitted on the CUBE system.

With the enhanced CUBE network in place, it will be possible to determine the shaking intensity levels at critical sites within minutes, from which it will be possible to estimate the performance of critical facilities.

Under contract with the USGS, EQE is developing an Early Post-Earthquake Damage Assessment Tool (EPEDAT) for buildings and exposed populations. This EPEDAT system will receive data from the CUBE network and will estimate the numbers of partially and totally collapsed buildings, fatalities and injuries, and displaced individuals in the affected region. EPEDAT is being designed to make these assessments within minutes after an earthquake.

Conclusions

It is inevitable that a large earthquake will occur along the San Andreas Fault sometime in the near future. The Ad Hoc Working Group on the Probabilities of Future Large Earthquakes in Southern California has estimated that the probability of a large (M7+) earthquake occurring somewhere in the greater Southern California area in the next five years could be nearly 50%. The severity of the effects of such an event will depend upon a number of factors, including its location and magnitude.

If the earthquake were to occur north of Los Angeles along the Mojave segment of the San Andreas Fault (comparable to the 1857 Fort Tejon Earthquake), damage in the Los Angeles area would be significant. If the earthquake were to occur along the southern portion of the San Andreas Fault (i.e., along the San Bernardino Mountains and/or Coachella Valley segments), Los Angeles may not be as severely affected, but overall regional damage could be even greater. Large portions of San Bernardino and Riverside counties would be severely affected if the earthquake were to occur along either of these two southern segments.

With proper emergency response planning and seismic retrofit of inadequate buildings, however, the impact of these large events can be lessened. Rapid post-earthquake assessment tools being developed will not only help to identify severely damaged areas after such an event, but will aid in providing effective response measures and minimizing recovery time.