Sitting in the bottom of a 3-foot-deep trench with a hard hat shaped like a cowboy hat, earthquake geologist Lisa Grant Ludwig scanned the bank of tan earth in front of her. Rainstorms had left layer upon layer of sand, silt and pea-sized gravel, going back hundreds of years.
Ludwig was looking for disruptions in the horizontal pattern, evidence that the San Andreas fault had moved in ground-cracking earthquakes.
There it was, a V-shaped fissure that slashed through orderly stripes of sediment. She studied its shape for several silent minutes. Ludwig pressed her fingers against the soil to test the fineness of the grain, as if the history of the San Andreas were written in Braille.
“I’ve said to people, ‘Here’s the San Andreas fault,’ ” she said. “They say, ‘Why isn’t it more obvious?’ That’s why we have to spend a lot of time out here.”
Ludwig, an associate professor at UC Irvine, has dug nearly two dozen trenches in the Carrizo Plain, a remote saltbush- and tumbleweed-studded grassland about 120 miles northwest of Los Angeles. Her particular spot, the Bidart Fan, is a flat area where streams rolling down the hills nearby spread out and dry up.
She has returned again and again to this place. Every time she comes back with more sophisticated questions, Ludwig said, the Carrizo yields answers. Recently, she and her colleagues have been looking to refine information about the magnitudes of and the time spans between quakes in the last few hundred years, and her most recent data will probably challenge conventional wisdom about the Carrizo.
“This is a scientific gold mine,” she said. “You’re going to keep working a gold mine.”
Inhabitants of the Carrizo Plain have long recognized its tendency to shake. Early Spanish explorers labeled places where they felt earthquakes with the word “temblor,” including the mountain range east of the plain. (“Carrizo” means reed grass.)
A remnant of the vast grasslands that used to cover the San Joaquin Valley, the area is one of the classic places to see disruptions caused by the San Andreas fault.
The landscape clearly shows the warping caused by two pieces of the Earth’s crust grinding past each other in a northwest-southeast direction.
Scarce rainfall has meant that few trees or other vegetation obscure the low bluffs and sinkholes alongside the fault. Visitors can easily spot stream channels that run straight down from the mountains and shift abruptly to the right when they cross the San Andreas.
Human settlement has also been generally sparse, though tribes of Native Americans, then cattle ranchers and dry farmers have inhabited the plain. As a result, the layer cake of sediments underfoot has remained well-preserved.
Since 2001, much of the plain has been protected as a national monument, managed by the federal Bureau of Land Management, the state Department of Fish and Game and the Nature Conservancy.
Scientists started putting the pieces together at the turn of the 20th century, naming the San Andreas fault and linking the crack threading through the Carrizo to the Ft. Tejon earthquake in 1857 and the San Francisco earthquake in 1906.
Previous studies of features here, including the Bidart Fan and, about three miles away, the dry stream bed known as Wallace Creek, showed hundreds of feet of motion on the fault over thousands of years, said J. Ramon Arrowsmith, a geologist at Arizona State University and one of Ludwig’s collaborators. But there was little detail on the pacing, or how they happened on a time scale of hundreds of years, so Ludwig’s group looked for a place that preserved a more richly layered package of sediments.
Wallace Creek “effectively looks like steady motion,” Arrowsmith said. “At Bidart, we can see individual snapshots.”
They want to know how the earthquakes moved the plates, he said. “Was it a bunch of little guys and a big one? Was it a bunch of medium-sized ones?”
Ludwig dug her first trench in the Bidart Fan, named after the family that owns the land, about 20 years ago.
In her first trenches, Ludwig saw traces of seven earthquakes. But she had to find chunks of charcoal to date the sediment around the ruptures and could find enough to put time brackets around only five temblors. Those quakes appeared to have taken place between about 1200 and 1857.
The work raised questions about the intervals between earthquakes on the Carrizo because the first four quakes appeared to be clustered, occurring roughly every 100 years. Then there was a gap of about 400 years before the 1857 quake.
Improvements in radiocarbon dating recently allowed Ludwig to process tiny samples she saved from her first dig, refining the dates for the five earthquakes and an average period between the quakes of about 140 years.
Ludwig thought she could find more quakes and more accurate dates if she dug new trenches and took advantage of new dating technologies.
She also thought she could figure out how big the prehistoric earthquakes were.
Many scientists had theorized that the estimated magnitude 7.9 temblor in 1857 was typical of quakes in the Carrizo and had expected this part of the fault to rupture about every 200 to 400 years. Ludwig wasn’t sure that was true.
The new trenches, which she started opening in 2005, have started to provide some answers. She has returned to the fan every summer since then with a group of colleagues who include Arrowsmith, UC Irvine postdoctoral scholar Sinan Akciz and a rotating group of students.
Although Ludwig provided just a sketch of their results because they are hoping to publish them in a scientific journal soon, she said they are confident they have accurate dates for at least six earthquakes in the last 800 years. The new trenches also show a shorter average interval between quakes, about 100 years.
By measuring the distance small streams have been offset and correlating them to ruptures they see in the trenches, the group has found that many of the quakes appear to be significantly smaller than the Ft. Tejon temblor, but as large as or larger than the magnitude 6.7 Northridge earthquake in 1994.
“When I first started working here, there were a lot of pages missing,” Ludwig said. “We’re finding those pages now.”
David Schwartz, a U.S. Geological Survey scientist who is not involved in the group’s study but has been observing it, said the results will probably lead to a “fundamental revision of how that entire part of the San Andreas is moving.”
It will likely also change the Uniform California Earthquake Rupture Forecast, which estimates probabilities for earthquakes in California, Schwartz said.
“It’s been 150 years since the most recent earthquake -- the one in 1857 -- on the Carrizo and, even though there’s a lot of uncertainty, you could view us as being much closer to reaching the time when it might slip again,” he said.
Ludwig and her collaborators believe they are on the last leg of fieldwork this summer. But one recent afternoon, a few weeks into their dig, they still had issues to work out.
Ludwig showed Arrowsmith and Akciz the trench diagram she had been making that day. With her pinkie, she pointed to several slanting fissures cutting through the stripes.
“I can’t tell where the tops are, where these earthquakes end,” Ludwig said.
They needed to figure out how close to the surface the fissures went so they could date the layer of sediment right on top of them to learn the most recent date before which each quake occurred.
She pointed to a symbol with whiskers she had drawn near the top of one of the fissures. Squirrels had burrowed into sediments softened by earthquakes and blended the stripes they needed for dating.
“It’s those critters,” she sighed. “That’s right where I’d like them not to dig.”