Sometime in the next 30 years, according to the most recent forecast from the U.S. Geological Survey, a large portion of the San Francisco Bay Area will jump more than 3 ft. in less than 30 sec., shaking the ground for perhaps 100 miles and triggering an earthquake with a magnitude of 6.7. Bridges will buckle. Apartment buildings will pancake. The dorms at the University of California, Berkeley, will roll like barrels on a wave. Water, power and transportation lines will be cut. The subway that runs under the bay could be a death trap. By the time the dust settles, more than 100,000 people will be homeless. Economic losses will total at least $35 billion. The human cost will be immeasurable.

Is there a way to prevent this catastrophe? Probably not within the next 30 years, but perhaps some day. No plans to stop a quake in Northern California are seriously being considered, and even if researchers had one they would like to try, enormous scientific and legal barriers would stand in the way. But new science, some accidental observations and a novel "microscope" to study the spawning grounds of earthquakes have for the first time made quake prevention something worth considering.

The basic science is pretty straightforward. The earth lurches from time to time because its outer shell is broken into 11 huge, solid plates floating on a layer of molten rock that has the consistency of Silly Putty. These tectonic plates are constantly jostling each other, like rafts crowded into a small pond, and it's along the boundaries where they meet that most quakes are born. The two plates that form California's infamous San Andreas Fault, the Pacific and the North American plates, are the largest on Earth. And they're moving inexorably in opposite directions.

Complicating matters is the fact that the plates don't slide past each other very smoothly. Like two giant pieces of sandpaper, they often get stuck—sometimes for hundreds of years. The pressure keeps building until something—often a juncture just a few miles below the earth's surface—snaps. Then the two plates move abruptly, sometimes great distances. A California quake in 1857 separated fences and roads by as much as 30 ft.

You might think that bolting the two plates together would fix the problem or at least buy a little time. But given the forces involved, holding the San Francisco Bay Area together for even a brief period of time would require bolts the size of the World Trade Center towers—an engineering feat that not even a modern-day Pharaoh could afford.

And where would you put them? The great San Francisco quake of 1906 cracked the earth across 350 miles, about one-third of the northern San Andreas Fault. To make things worse, California is riddled with faults that are smaller and not so well mapped as the San Andreas. For example, the 1994 Northridge quake, which registered a magnitude of 6.6 and caused $20 billion in damage, occurred on a fault no one knew was there at all.

But the biggest problem with such a plan is that even when bolted together, the plates would continue to build up stress, and that stress would have to be relieved somewhere. A much better approach would be to relieve the stress gradually—with a lot of small quakes—rather than let it accumulate. If man-made quakes could move the fault just 1.67 in. a year, the Big One on the northern San Andreas could be averted.

MORE>>

PAGE 1 | 2