Study Says Forecasts of Big Quakes in California Are Overstated
Dec. 08, 1992
SAN FRANCISCO (AP) _ Government forecasts of disastrous earthquakes in California seriously exaggerate the risk, scientists said Tuesday after using a computer to simulate 10,000 years of shaking on the San Andreas Fault.
The new study suggests that the risk of a magnitude-7.5 or larger quake on the southern California part of the fault is 19 percent within 30 years, said Steven Ward, a geophysicist at the University of California, Santa Cruz.
A 1988 U.S. Geological Survey forecast said the odds were at least 60 percent in 30 years. The forecast was made and endorsed by panels representing a consensus of scientists.
Ward and Saskia Goes, a Dutch graduate student who co-authored the study, said that even if the odds are lower than in the government's widely publicized forecast, Californians shouldn't be complacent.
''California is still earthquake country,'' said Goes, who presented the study during the American Geophysical Union's fall meeting. ''It's still good to be prepared.''
The study also indicated that a quake measuring about 6 on the Richter scale is about 66 percent likely within 30 years on the central San Andreas near Parkfield. In 1985, the Geological Survey made its first fairly specific prediction, saying it was 95 percent certain such a quake would happen by the end of 1992. It hasn't happened yet.
The new computer simulation didn't directly address two other government reports: a 1990 forecast that another magnitude-7 quake in the San Francisco Bay area is 67 percent likely within 30 years, and a forecast issued last week that a quake of that strength or larger may be up to 47 percent likely in Southern California within five years.
But Ward speculated that the odds for a San Francisco Bay-area quake are only somewhat overstated, while the report issued last week contains odds that may be too high.
''I hope this guy (Ward) is right, but there's no reason to believe he's more right than the other scientists are,'' said Richard Andrews, director of California's Office of Emergency Services. ''With the degree of uncertainty we have, the prudent course is to be prepared.''
Ward said the 1988 report assumed each segment of the San Andreas fault builds up stress until it snaps in a big quake, and that such quakes recur at fairly regular intervals.
His theory is that each segment transfers stress to neighboring segments in a complex way during quakes. The simulation suggested big quakes happen almost randomly, rendering the government forecasts unreliable.
''It's hard to know if his model really captures what the Earth is doing,'' said Duncan Agnew, an author of the 1988 report and a geophysicist at the University of California, San Diego.
Geological Survey seismologist Tom Heaton said he was sympathetic to Ward's arguments, but Californians still should ''assume large earthquakes will happen and could happen at any time.''
In little more than three years, three major quakes have rocked California. The magnitude-7.1 Loma Prieta quake in 1989 killed 63 people in the San Francisco Bay region. The 7.1 Cape Mendocino quake and its 6.7 and 6.6 aftershocks last April injured 356 people on the state's north coast. And on June 28, the 7.5 Landers quake and its 6.6 Big Bear aftershock killed a child and injured 402 Southern Californians.
David Jackson, a researcher at the University of California, Los Angeles, listed several reasons many scientists believe the risk has increased for a big quake on the southern San Andreas. The Landers quake added stress to fault segments nearest San Bernardino and Palm Springs, segments considered most overdue for big quakes. Also, large earthquakes tend to happen in clusters, and Southern California has had a sharp increase in quakes since 1985, he said.