New information about the inner workings of faults could change how experts estimate the potential for the next “big one,” according to UC Irvine and Arizona State University researchers.

A team led by Lisa Grant Ludwig, associate professor of public health at UCI, established the age of a few dry stream channels with bends created by the San Andreas fault in the Carrizo Plain, about 100 miles northwest of Los Angeles. They then determined how many earthquakes had occurred after the formation of each channel and approximately how much the fault had moved in each quake.

“The distance that a fault ‘slips,’ or moves, during a large earthquake reveals a lot about the mechanics of faults,” Ludwig says. “Getting slip measurements from past earthquakes is like getting a look under the hood of a car – it tells you a lot about the engine.”

By digging trenches across the fault, radiocarbon-dating channel sediment samples and studying historic weather data for the Carrizo Plain, the researchers got to peer into an engine of large earthquakes. And they found something unexpected: Fault slip varied from earthquake to earthquake instead of recurring in a uniform manner, as had been believed.

“The idea of slips repeating in characteristic ways is very appealing, because if you can figure that out, you’re on your way to forecasting earthquakes with some reasonable confidence,” Ludwig says. “But these results show that we don’t understand the San Andreas fault as well as we thought we did.”

The southern San Andreas is widely studied by geologists and has influenced the development of fault rupture models and estimations of earthquake probability. These concepts, which have been applied to other faults around the world, may need revision.

The most studied quake in the region is the original “big one” – the magnitude 7.8 Fort Tejon quake of 1857, which was presumed to have caused a nine- to 10-meter slip along the San Andreas fault. But Ludwig’s team discovered that it was only half as much and that slip in some prior earthquakes had been even less.

In addition, they found that none of the five large quakes in the Carrizo Plain over the past 500 years had produced slip anywhere near nine meters. The maximum was about five meters – the slip created by the Fort Tejon temblor. Since then, Ludwig says, approximately five meters of strain, or potential slip, has again built up along the San Andreas fault in the Carrizo Plain, matching that released in 1857. This suggests the potential for a similarly large earthquake in the region, she says.

“The recent disaster in Haiti is a reminder that a destructive quake can strike without warning. One thing that hasn’t changed is the importance of preparedness and earthquake-resistant infrastructure in seismically active areas around the globe,” Ludwig adds.

UCI’s Sinan Akciz and Gabriela Noriega and Arizona State’s Olaf Zielke and J. Ramon Arrowsmith also participated in the study. Results appear online Jan. 21 at Science Express.