Scientists Upbeat on Earthquake Predictions
Long-term earthquake prediction is closer to reality than most people think, scientists said at a symposium here last week. In more than one case, such general predictions years in advance have already taken place, they concurred.
The researchers also expressed optimism that short-term predictions--precise as to day, place and size--will eventually be commonplace.
The discussions took place on the final day of the annual meeting here of the American Geophysical Union, and, as usual in such gatherings, a few scientists had rather somber warnings for Californians.
Lynn Sykes of Columbia University, for example, said the numerous fairly strong quakes, such as the Northridge temblor, that took place near the San Gabriel and Santa Susana mountains in Los Angeles County from 1985 to 1994 might portend a very large quake in the Southland in coming years.
Sykes compared the quake clusters of magnitudes 5 and 6, on other faults than the mighty San Andreas, with a similar cluster on other faults that presaged the great San Francisco quake of 1906, which ruptured the San Andreas over a 200-mile stretch.
Sykes noted that such a cluster of quakes had also led certain scientists, including Allan Lindh of the U.S. Geological Survey, to make a successful long-term prediction in 1988 that a large quake would occur relatively soon in the Santa Cruz Mountains.
Just a year later, the Loma Prieta quake did occur in the locale named by the scientists.
Lindh attended last week’s discussions, and he said he feels there will be many research failures before short-term predictions become reliable. He guessed it might take until the year 2100 or even 2200.
Another series of warnings for California came from Caltech’s Thomas Heaton, who said present building codes may well be totally inadequate when it comes to coping with really large quakes, in the 7 and 8 magnitude range.
Heaton said he had recently reviewed plans for retrofitting San Francisco’s famed Golden Gate Bridge, and he believes the amount of displacement being allowed for the bridge foundations is five times smaller than would be required if another magnitude 8 quake were to occur in the Bay Area.
Heaton also reiterated his previously expressed reservations about seismic base isolation as a means of protecting tall buildings like Los Angeles City Hall. The base isolators also have displacement limits, and if these limits should be exceeded by prolonged heavy shaking in a huge quake, the damage to buildings, and the numbers of casualties, could be enormous, he said.
Engineers have not yet been able to explain why so many tall redwoods were snapped off two-thirds of the way to their tops in the 1906 quake. And Heaton says the same shaking forces, radiating up the high-rise buildings of today, could bend them beyond repair. The way shaking affects structures as well as trees remains incompletely understood, he said.
The Northridge quake, he added, was not a fair test of Los Angeles’ high-rise buildings, because its worst shaking occurred miles away from them, he said.
The moderator of the session, UCLA’s David Jackson, said quake scientists can take satisfaction in certain gains they have made in the last decade.
He named development of a comprehensive geodetic measuring system, with its precise measurements of ground movement, the installation of many more precise strong motion instruments, and improved understanding of how one quake can trigger another while reducing stress in other locales as among the important accomplishments.
But two scientists who urged more effort at earthquake prediction, Max Wyss of the University of Alaska at Fairbanks and Paul Silver of the Carnegie Institution, said monitoring of all kinds of possible earthquake precursors must be vastly stepped up if scientists are to achieve reliable short-term predictions.
Wyss complained that Americans have become too discouraged about progress at making predictions. He suggested that this keeps adequate government and private money from being put into research that would facilitate it.
Silver said geodetic and bore-hole measurements must be integrated and a strategy devised for long-term progress before the physics of earthquakes can be better understood. Only when this happens will prediction be consistently successful, he said.
Meanwhile, Ross Stein of the U.S. Geological Survey, who has long been involved with the earthquake prediction experiment at Parkfield, Calif., gave his colleagues one possible explanation of why scientists seemed to have failed when they predicted in 1985 that a magnitude 6 earthquake was 95% likely to occur there by 1993.
Stein said it now appears that the Coalinga earthquake of 1983 may have relieved stress around Parkfield and retarded the expected quake by several years at least.
Stein and others have developed a “stress shadow” theory in recent years, which holds that large earthquakes relieve stress in some directions while increasing it in others. For instance, they suggest, San Francisco, the peninsula and part of the East Bay may still be in a stress shadow from the 1906 earthquake, having therefore less chance of a quake in the near future.
Also, after the Landers quake of 1992, these scientists developed a stress map, which predicted correctly where most of the aftershocks from that quake would occur, and, conversely, where there would not be many aftershocks.
The discussions drew an audience of several hundred scientists and researchers, representing both government and private institutions from throughout the country.
