Ask about the likelihood of a major earthquake on the West Coast, and researchers there will often tell you when they expect it, its minimum strength, and the part of a specific fault involved. Ask a similar question of scientists in the East and they'll reply with a time and minimum strength, too. But a location? Somewhere in the eastern United States, comes the reply.
This vagueness results from gaps in knowledge that make life difficult for scientists and engineers trying to convince public officials in the East of the need to prepare for major earthquakes and to adopt building codes that lead to structures less vulnerable to quake damage.
Based on history alone, ``there is every reason to believe that there is a 95 percent certainty'' of an earthquake with a Richter magnitude of six or greater occurring in the next 20 years somewhere in the eastern US, says Robert L. Ketter, director of the National Center for Earthquake Engineering Research (NCEER) at the State University of New York at Buffalo. That would be roughly equal in magnitude to the quake that shook Whittier, Calif., last October.
But researchers say that damage from such a quake in the East could eclipse that experienced in Whittier. The underlying geology in the East allows earthquake waves to travel farther than in the West.
For a given magnitude, an Eastern quake can affect an area up to 100 times greater than a Western quake. Dr. Ketter says that some formations in the East actually amplify ground shaking. And many regions that experienced earthquakes in the past have high population densities, old housing stock, and lack seismic requirements in building codes.
The serious study of quakes east of the Rocky Mountains has come about only in the last decade, researchers say. This despite the fact that the biggest quake known to have hit the continental United States struck near New Madrid, Mo., between 1811 and 1812.
Part of the lag is due to progress in understanding Earth's crust. The 1960s saw the development of plate-tectonic theory, says Klaus Jacob, a senior research scientist at Columbia University's Lamont-Doherty Geological Observatory. The theory holds that Earth's crust is made up of vast, constantly moving plates that well up from midocean ridges. These plates grind past and ultimately slide under other plates, leading to a high rate of earthquake and volcano activity at plate boundaries.
The advent of tectonics turned those plate boundaries into Meccas for seismologists, Dr. Jacob says. For the next 10 years, scientists paid relatively little attention to what happened in the middle of a plate.
That changed when the nuclear industry needed better information about earthquake hazards near power plants, says Peter W. Basham, a geophysicist with the Geological Survey of Canada. Most of those sites are in the East.
Congress also began to realize that earthquakes were more than California's problem and earmarked money for more research in other parts of the country, the NCEER's Ketter says.
Finally, Jacob says, the Eastern engineering schools that had contributed to the West Coast's efforts to establish seismic standards for structures began to turn their attention to their own backyards.
The stresses that account for most of the Eastern quakes result from the general circulation of crustal plates surrounding the North American plate and from the pressure of new plate material pushing up and out from the mid-Atlantic ridge.
Earthquakes with a magnitude greater than six seem to be limited to reactivated faults in rift areas, such as the St. Lawrence River Valley, and the Mississippi Embayment, which lies deep under sediments beneath the southern portion of the Mississippi River, Dr. Basham says.
Hundreds of millions of years ago the crust split and spread in both these areas but stopped far short of tearing North America into new continents.
Large quakes also appear along ancient plate boundaries. One of those zones sweeps around the Atlantic and Gulf seaboards while another roughly follows the Atlantic coast between the shoreline and the deep sea bed.
Quakes with magnitudes less than six are more randomly located, Basham says, and many come as surprises because geologists had not located the faults ahead of time.
Indeed, fault-finding is one of the activities that is proving more difficult in the East than in the West.
``You can't see the faults in the East'' from the surface, says Robert Hamilton, with the US Geological Survey. ``The whole Atlantic and Gulf crust has been subsiding for a hundred million years,'' burying the faults beneath thick layers of sediment.
As a result, he says, geologists have to rely on remote sensing devices that can detect minute changes in the gravitational or magnetic fields of deeply buried rock formations.
Once a suspected fault zone has been isolated by a pattern of seismic activity, they begin the hunt for the fault itself. The fault responsible for the New Madrid quakes, invisible from the surface, was detected using these geophysical techniques, Dr. Hamilton says.
Scientists still don't understand why major quakes are not uniformly spread over the underlying formations with which they are associated, Basham says. One possibility is that, given the current orientation of the stresses on the plate, only some rift features are aligned in a direction that puts them under enough pressure to generate quakes.
Another possibility is that the major quakes occur so infrequently that records don't go back far enough to get a handle on how frequently quakes reoccur. Apparently inactive areas may just be in dormant periods.
In some areas where there have been several major quakes, the lack of geological evidence at the surface may suggest that the faults involved became active only thousands or tens of thousands of years ago, Basham continues.
To try to get a handle on repeat times in the West, geologists have dug trenches along the traces of faults to expose layers of soil and peat. By spotting layers that have been offset and by noting how large the offset is, they can estimate a quake's magnitude. Offsets at different depths indicate different quakes, and the affected soils can be dated using radiocarbon techniques to establish estimates of repeat times.
But in the East, faults leave no trace on the surface, forcing geologists to rely on evidence of soil becoming liquefied. This liquefaction is caused by severe shaking of soils in areas with fairly high water tables.
The process leaves distinct signatures that can also be uncovered through trenching. Such efforts have been used to study the effects of the New Madrid quakes, the Charleston, S.C., quake of 1886, and the Newbury, Mass., quake of 1727.
There is hope that the technique will ultimately prove as useful as trenching faults in uncovering repeat times. Stephen Obermeyer of the US Geological Survey says that in addition to the 1886 quake, a site near Charleston shows evidence of at least three quakes in the last 7,200 years strong enough to produce liquefaction.
Hamilton says scientist need to do a better job of identifying areas prone to earthquakes. ``It took 10 years to develop a good model of New Madrid'' at a cost of some $5 million to $10 million, working in a 100,000-square-kilometer area. While recognizing the high cost, he says ``there is so much to be gained that it's worth it.'' He says that the next areas he would like to see scientists focus on would be either Charleston, S.C., or the St. Lawrence River Valley.
EARTHQUAKES IN THE EAST (all had magnitudes above 4.8 on a scale from 0 to 10) 1. 1663 St. Lawrence River 2. 1727 Newbury, Mass. 3. 1732 W. Quebec/E. Ontario 4. 1755 East of Cape Anne, Mass. 5. 1811 New Madrid, Mo. 6. 1811 New Madrid, Mo. 7. 1812 New Madrid, Mo. 8. 1860 Charlevoix, Canada 9. 1869 Bay of Fundy, Canada 10. 1870 Charlevoix, Canada 11. 1884 New York City, N.Y. 12. 1886 Charleston, S.C. 13. 1895 Charleston, Mo. 14. 1897 Giles County, Va. 15. 1904 Eastport, Maine 16. 1925 Charlevoix/LaMalbaie, Quebec 17. 1929 Attica , N.Y. 18. 1929 Grand Banks, Canada 19. 1935 Temiskaming, Canada 20. 1940 Ossippee, N.H. 21. 1944 Massena, N.Y./Cornwall, Ontario 22. 1968 South Central Ill. 23. 1982 Miramichi, Canada 24. 1982 Gaza, N.H. 25. 1983 Goodnow, N.Y. 26. 1986 Painesville/Cleveland, Ohio 27. 1987 Lawrenceville, Ill.