EARTHQUAKE HAZARD IN SOUTHERN CALIFORNIA IS WIDESPREAD, ACCORDING TO MODEL THAT COMBINES MANY KINDS OF ASSESSMENT
NOTE: This release is EMBARGOED until Wednesday, April 6, 1994, when Steven Ward will present his research at a meeting of the Seismological Society of America at the Pasadena Convention Center.
PASADENA, CA--Southern Californians have long lived in fear of "The Big One." In recent years, however, scientific focus has shifted from that one huge earthquake on the San Andreas Fault, as seismologists have realized that smaller but potentially more hazardous faults crisscross virtually the entire L.A. basin. The latest example ripped Northridge in January: a magnitude 6.8 quake on an unknown fault far underground.
Now, a model of seismic hazard from Parkfield to San Diego quantifies that new reality. While the mighty San Andreas and other big faults remain players, the model suggests that 40 percent of the region's total earthquake motions could occur in widely scattered areas away from those faults--and closer to many cities.
As a result, says model creator Steven Ward of the University of California, Santa Cruz, "For most residents of southern California, the lion's share of hazard from moderate earthquake shaking over a 30-year period derives from smaller, closer, more frequent quakes of magnitude 5 to 7, rather than from large San Andreas ruptures, whatever their likelihood."
Ward, a research geophysicist at UCSC's Institute of Tectonics, will present his model on April 6 at a meeting of the Seismological Society of America in Pasadena. An article on the model will appear later this year in the Bulletin of the Seismological Society of America.
Ward's model differs from most previous attempts to define seismic hazard in a large region. Most commonly, researchers use records of past quakes and known rates of slip along major faults to estimate when future earthquakes might strike. That works well for areas where all the action happens on a few faults, but it fails for southern California's complex patchwork. For a more complete model, Ward blended two other ingredients into his stew:
Space geodesy, a relatively new and powerful way to find out how quickly strain builds up in the earth's crust. Measurements using both satellites and distant quasars reveal tiny motions in the positions of many stations on the ground relative to each other. Ward used such data compiled for the last ten years to map strain throughout the entire region.
"Synthetic" seismicity. Ward developed a computer model of the San Andreas and other faults. The model simulates thousands of years of earthquake behavior, based on the physics of how stress accumulates and releases in the earth. Its results: Quakes happen less regularly than researchers have assumed, and they often break several segments of a fault in a row.
Historical earthquake records and geological findings from the field, together with predictions from the computer model, gave Ward the constraints he needed to convert the geodetic motions into estimates of the sizes, locations, and frequencies of quakes. To assign those numbers to geographical areas, Ward used a map of 66 boxes spread across the landscape of southern California like a crazy quilt. Some of the boxes straddle sections of famous faults, but two-thirds of them cover areas with either many faults or unknown faults.
"With this approach, we can estimate the seismic hazard in each area even if we don't know exactly where the faults are," Ward says. "The space geodesy is what is really new--it supplies information where geology and seismology can't."
Ward displays the model's results as a series of maps, each of which shows the likelihood of different types of hazard in southern California. For instance, over the next 30 years virtually all of the region has more than a 50-50 chance of feeling shaking at a moderate level of 10 percent the force of gravity. Shaking twice as strong is most likely to occur at Parkfield, near San Bernardino, and northwest of Los Angeles from Ventura to San Fernandonear the site of the Northridge quake.
Another map shows that much stronger and more damaging shaking--from three-tenths to half the force of gravity--has a roughly 6 percent chance of occurring in most of the region during the next 30 years. Those odds increase in the areas listed above.
The most powerful shaking in Ward's model does occur near the San Andreas Fault and the San Jacinto Fault, which branches south off the San Andreas near San Bernardino. However, such shaking is much less likely to happen within 30 years, because large quakes on those faults are rare. Indeed, when Ward removed all quakes greater than magnitude 7 from his model, it made little difference--the hazards for moderate shaking throughout most of the region stayed largely the same.
Ward's maps show that the region's hazards are not uniform-- they vary markedly from place to place. "But most areas will not escape eventual shaking," he says. "There are faults everywhere."
Ward notes that his model does not account for different soil conditions. For instance, loose soil can amplify ground shaking, dramatically increasing the hazard. In addition, he is updating his model to include the most likely patterns of earthquake faulting in each area. A "thrust" earthquake, such as the Northridge event, can create more severe ground shaking than a "strike-slip" quake, where the ground moves from side to side.
Editor's note: You may reach Ward at (408) 459-2480 or ward@uplift.ucsc.edu. For the illustrations of seismic hazard that accompany this release, call the UCSC Public Information Office at (408) 459-2495.
Press Releases Home | Search Press Releases | Press Release Archive | Services for Journalists
Maintained by:pioweb@cats.ucsc.edu