"The Japanese and Taiwanese have drilled through faults," Mark Zoback, a geophysics professor at Stanford University in Palo Alto, California, said last December at a meeting of the American Geophysical Union.
But those weren't major, active faults, and the drilling wasn't conducted at the heart of their earthquake-producing zones, Zoback said.
William Ellsworth is a seismologist with the U.S. Geological Survey in Menlo Park, California. He says SAFOD's goal is to have instruments in a section of the fault that breaks.
"That's why we go into the ground, to be inside the earthquake machine where the action starts," Ellsworth said.
Earthquakes can begin hundreds of miles underground. But at Parkfield they're only 2 miles (3.2 kilometers) deep.
Getting to them is still a challenge.
"The pressure is like being under two miles of water," Ellsworth said. "And the ambient temperature is about 125°C (257°F).
"Making anything work for a long time in those conditions is a real challenge. Especially for electronics."
The final instruments will be placed in the borehole sometime in 2007. Then the SAFOD team will be able to measure small, barely detectable slippage and changes in water pressure in the fault-zone rocks.
A typical magnitude 2 quake involves only 1 inch (2.5 centimeters) of slippage along a segment of the fault about the area of a football field.
That's a very small earthquake, and the precursory signs that the scientists are looking for are likely to be very subtle. But the hope is that these data can be used to figure out what to look for to predict bigger quakes elsewhere using more conventional instruments.
"It all boils down to testing whether or not earthquakes are predictable," Zoback said.
Even if they fall short of that holy grail, the SAFOD team expects to learn much about how earthquakes occur.
That knowledge can be used to save lives and reduce damage in places like San Francisco. (See "100 Years Later, San Francisco Ripe for Another Megaquake.")
"When we simulate what is likely to happen in an earthquake, to warn people about the kind of shaking an area might experience and for disaster-response planning, there is an underlying physical model of the earthquake source," Zoback said.
"We hope to model earthquakes more precisely than before."
Already the project has yielded some interesting results.
To begin with, nobody was really sure how wide the active fault zone was: Did all of the slippage occur along a single narrow seam, or was it spread throughout a broader region?
Now the scientists know that the fault near Parkfield is actually a band of pulverized rock about an eighth of a mile (200 meters) across, with different earthquakes occurring at different places within it.
Another surprise: The fault zone is unexpectedly weak.
To understand what that means, press your hands together hard, then attempt to slide them apart sideways.
Putting a dollar bill between your palms while doing this is akin to having a strong fault that resists slippage. (Try it.) Butter between your hands is the oppositea weak fault that slips easily, like the San Andreas Fault near Parkfield.
In either case, you get movement (an earthquake). But the amount of force needed to cause it differs.
Understanding how slippage works won't prevent earthquakes, but it may help alert us to earthquake hazards at places we have not expected them.
Free Email News Updates
Sign up for our Inside National Geographic newsletter. Every two weeks we'll send you our top stories and pictures (see sample).
SOURCES AND RELATED WEB SITES