The box at center outlines the area of the Los Angeles basin where a new, active earthquake fault has been discovered. In 1994, a moderately large temblor (marked M6.7) on an adjacent fault struck Northridge, Calif., killing 61 people and causing $35 billion in damage.

A large, active crack in the earth, capable of causing destructive earthquakes, has been discovered under Los Angeles.

The crack, or fault, does not break the Earth's surface, so it remained hidden until geologists from Harvard University and the University of California, San Diego, found it last year.

"Los Angeles is caught in a vise," says John Shaw, an assistant professor of structural and economic geology at Harvard. "It is locked between converging sections, or plates, of Earth's crust, carrying North America and part of the Pacific Ocean floor. As the plates collide, rocks beneath the city are shattered and cut by faults of many shapes and sizes."

Motion along the faults appears to be shortening or contracting the Los Angeles basin an estimated quarter inch (7-8 millimeters) per year.

"That doesn't sound like much, but it's a significantly high rate when one of the largest cities in the world is built on top of it," Shaw comments.

Stress building up along parts of the newly discovered fault, then rupturing, could cause a series of large shakers every 250 to 1,000 years, according to Shaw and fault co-discoverer Peter Shearer of the University of California, San Diego. An earthquake of comparable size hit nearby Northridge in 1994, killing 61 people and causing $35 billion in damage.

A violent breach along the entire fault at once could produce a convulsion three times the intensity of the Northridge quake. Such a cataclysm could happen every 500 to 2,000 years and result in an estimated 3,000 to 8,000 deaths and damages of about $200 billion.

The problem with trying to give a more accurate forecast is not knowing when the clock started, or when the stress along the fault was last relieved by a major temblor. "But it seems inevitable that, sooner or later, a large earthquake will occur in the Los Angeles area," Shaw says. "And it could well be in our lifetimes."

The fault finding is to be announced at a press conference on Thursday, March 4, and will be published in the March 5 issue of Science magazine.

A cross-section shows the Puente Hills fault running from Northern Orange County (right) to downtown Los Angeles (left) at depths between 2 and 9 miles. Rocks on top of the fault are moving southward relative to those below the fault, producing a shortening of the basin in which the city sits.

Finding a Fault

A number of faults cut through the Los Angeles metropolitan area, including the one responsible for the Northridge quake. However, that crack lies north of the Santa Monica Mountains, which protected the city from the worst ground shaking. Then there's the infamous San Andreas fault that runs north to San Francisco and beyond. Most of the motion between North America and the Pacific Ocean floor takes place along that system, which lies well to the east of Los Angeles. Slippage along the San Andreas caused the 1906 San Francisco quake, one of the largest known temblors in the country's history.

What sent Shaw and Shearer on their search, however, was a smaller, 1987 outburst near Whittier, only 12 miles from the center of downtown Los Angeles. "There was a lot of uncertainty about whether this quake occurred on a new fault or one we knew about," Shaw recalls.

John Shaw suspected that a 1987 earthquake, only 12 miles from downtown Los Angeles, was located on an undiscovered blind thrust fault, so called because it did not break the surface.

He and Shearer decided there was enough geological evidence to believe that an undiscovered fault lay buried under the city and the basin in which it sits. The idea got mixed reviews from more senior geologists.

What broke the case open was evidence from oil companies who had done extensive drilling and subsurface seismic exploration in the area. Few people are aware of it, but the Los Angeles basin was once one of the most prolific oil fields in the world. It is underlain by folded rocks and fractures where large pools of oil collect.

Petroleum prospectors had been interested in folds of rock beneath the Puente Hills in northern Orange Country. Shaw and Shearer were interested in the same thing: they believed the hills were created by movements of the Earth's crust along a blind, subterranean fault.

"Using oil company data, we located the source of the Whittier earthquake on what we now call the Puente Hills fault system," Shaw explains. "We got a direct image of the fault surface by a technique that is somewhat analogous to obtaining a sonogram of an unborn child." In this case, the sound waves were generated by explosives.

Subsurface maps show a huge fracture in the earth starting about 2 miles below the surface and inclined to the north at an angle of 27 degrees. The geologists traced it to a depth of 9 miles. In lateral extent, it stretches from the Puente Hills to right under downtown Los Angeles.

On top of the fault, the earth slides southward relative to the rocks below the fault. The ultimate pushing force involves enormous sections of Earth's broken crust moving toward each other.

What's Next

Shaw and Shearer are working with the Southern California Earthquake Center to make the geological data available to emergency response agencies and other local and state groups.

"The information can best be used to save lives by strengthening existing building codes and developing new codes to deal with this new threat," Shaw says.

Working with Jeroen Tromp, professor of geophysics, and other researchers at Harvard, Shaw is attempting to construct a three- dimensional computer model to map the paths and intensities of shock waves from quakes along the fault. "We want to learn where the ground will shake the most, what buildings will be affected, and how many people live in such areas."

Shearer and Shaw also plan to collaborate with earth scientists from other universities to obtain better data for predicting future quakes. "We want to drill holes in the basin and retrieve rocks and other material that can be dated," Shaw explains. "Such investigations could yield more precise information on how fast rocks move along the fault, when the last quake occurred and how big it was."

The geo-sleuths want to determine how much of California's shortening occurs on the Puente Hills fault compared with motions on other faults in the Los Angeles basin. The faster the movement along a fault, the more often it will break. At present, Shaw and Shearer suspect about one-tenth of an inch (2-3 millimeters) shortening on the fault per year. Additional field studies are expected to give a more accurate estimate of future movements.

"What has gone on in the past," Shaw says, "provides the best clues to what will happen in the future."