No guessing game: Texas A&M team trying to predict earthquakes

People in earthquake-prone California often talk about the “Big One,” a devastating quake that many experts say will surely strike the region sometime in the future.

A research team is now working to predict when the big one – and even little ones – might occur. Termed SAFOD (San Andreas Fault Observatory at Depth), the project involves more than 20 researchers from several major universities, labs and government agencies, including the husband-wife team of Fred and Judi Chester of Texas A&M’s College of Geosciences. SAFOD’s goal is a lofty one – to drill directly into the San Andreas Fault about two miles deep, place various types of instruments in the bored-out area, get rock samples and use the new data to extend and test models that may allow researchers to predict when the next major earthquake might hit.

It’s the first time anyone has ever drilled directly into an active fault zone where scientists think an earthquake will occur in the next couple of years. The multi-year project is part of EarthScope, a long-range program to study the tectonics of North America in partnership with the U.S. Geological Survey, and is backed with $220 million in funding from the National Science Foundation. The core samples taken will be housed and studied in the Integrated Ocean Drilling Program (IODP) core repository at Texas A&M, which has some of the world’s best expertise and resources in core handling.

The drill site chosen is near the town of Parkfield, Calif., which is between Los Angeles and San Francisco, and boasts 18 full-time residents but proudly calls itself the “earthquake capital of the world.” That claim is no boast: Parkfield has registered at least six quakes of magnitude 6 or higher since 1857, and dozens of smaller ones each year for decades because of its unique location where the Pacific and North American Tectonic Plates converge along the San Andreas Fault line.

Parkfield is located between the sites of the largest recorded earthquakes in California, the 7.9 Fort Tejon quake of 1857 near Los Angeles and the 7.8 quake that occurred in 1906 near San Francisco. “The basic problem is this: earthquakes occur at great depths, so we can’t see what’s going on when they do happen,” Fred Chester explains. “Our goal is to drill into the active earthquake zone near Parkfield, collect core samples and leave instruments down there that will collect data ’24/7’ over the next few earthquake cycles.”
Chester said earthquakes along the San Andreas occur between two and 10 miles below the Earth’s surface. The Parkfield site is ideal, he says, because of its earthquake frequency. “We know that there have been many earthquakes there in the past, and we are certain there will be many more to come in that area,” he adds. “This is the first time in the U.S. for a scientific team to drill into a place where earthquakes are generated. We want to monitor this site for 5 to 10 years and see what changes occur in stress levels, temperature and fluids before and after an earthquake.

“Previous models used to predict earthquakes have pretty much failed,” he notes. “We’re hoping this project will give us reliable information so that we can understand the physics of an earthquake at its source, and that we can develop strategies to predict the location, timing and size of earthquakes.”

Drilling into the San Andreas Fault line began this summer and uses advance directional drilling technology developed for the oil and gas industry. At most, only about 200 feet can be bored out in a day.

Drilling work will end in summer 2005 at a length of about three miles and depth of two miles, Judi Chester said. And then in 2007, more than a half-mile of core will be collected from across the fault zone. “Some of the most important work will be examining the core samples once they reach IODP at Texas A&M, and the first samples will arrive next month,” Judi Chester adds. “We’ve studied fault lines at the surface for many years, and now we can study them at the depth of the earthquake source. These core samples should provide extremely valuable information.”

The Parkfield area has averaged a magnitude 6 quake or higher about once in every 22 years, Fred Chester says, with the last one occurring just six weeks ago on Sept. 28. It’s part of a very active area: In 1983, a 6.4 quake struck Coalinga, a small town about 25 miles northwest of Parkfield. Also nearby, the San Simeon quake, magnitude 6.5, killed two people in Paso Robles on Dec. 22, 2003. “The scientific community is becoming more and more confident that earthquake prediction will someday be a reality, but we still have much to learn,” Chester adds.

“This project will be the most detailed and exhaustive ever. If nothing else, the information we’ll get about earthquakes will be of tremendous benefit not only for Californians, but for people all around the world who live in earthquake areas.”