Sediment cores from the 1960s are shaking up recent conclusions some scientists have drawn about Cascadia earthquakes, which occur along the subduction zone that stretches off the coast from Vancouver, British Columbia, to Northern California. A new Geology paper published online on 29 July (doi:10.1130/G35902.1) says that the Cascadia earthquake hazard is more complicated than what the recent research has shown.

Specifically, the new paper “question[s] the confidence in some of the estimates of earthquake size and occurrence” along the Cascadia margin, lead author Brian Atwater, a geologist at the U.S. Geological Survey (USGS), told Eos.

Earthquakes along the Cascadia subduction zone margin have periodically rocked the coasts of Washington, Oregon, and Northern California over the past 10,000 years or longer. Studying the margin is pivotal to understanding the earthquake hazard risk, which affects everything from constructing highways and bridges in the region to setting insurance premiums.
While Atwater was involved in discussions during the recent update of the USGS’s National Seismic Hazard Maps, he unearthed dissertations from the 1960s and found something surprising in the descriptions of sediment cores made by their authors: Little evidence of the number of turbidites—sediments left over by ancient mudflows called turbidity currents—in the northern part of the Cascadia margin that were found by more recent reports.

The Geology paper specifically questions some of the conclusions made in a 2012 USGS report, Turbidite Event ­History—­Methods and Implications for Holocene Paleoseismicity of the Cascadia Subduction Zone, that detailed the Cascadia earthquake hazard on the basis of turbidites. Both the 2012 report and a Tectonics paper from 1990 (doi:10.1029/TC009i004p00569) that also used some of the 1960s era cores concluded that turbidite records along the entire Cascadia margin point to regularly recurring ­margin-wide megathrust earthquakes about every 500 years. However, because Atwater found little record of turbidites in the 1960s sediment cores—which were taken in the northern region of the fault—evidence for earthquakes that stretch the entire length of the subduction zone seems to be lacking.

The issue is not whether the Pacific Northwest is subject to big earthquakes, Atwater said. “The question is what sort of numbers you put on that hazard,” he said.

“A number of fundamental assumptions were made to underpin turbidite paleoseismology, and we’re simply calling into question those assumptions,” Bobb Carson, who was a grad student at the University of Washington when he took some of the core samples in the 1960s, told Eos. Carson is now an emeritus professor in the Earth and Environmental Sciences Department at Lehigh University in Bethlehem, Pa.

The disparity between the turbidite records found by Atwater and those found by the 2012 and 1990 reports could be attributed to the fact that the northern region of the Cascadia subduction zone lacks the large amounts of old sediment found farther south, off the coast of southern Oregon and Northern California, Carson said.

“What you would like is for turbidite deposits to correspond one-to-one with earthquakes if you are going to use them as proxy records,” Atwater said. “If sediment supply can limit production of turbidites in one place and enhance it [in] another, that becomes a complicat[ing] factor.”

Chris Goldfinger, the lead author on the 2012 report detailing the hazard of the Cascadia margin based on turbidite records, disagrees with Atwater’s findings. Goldfinger said that the descriptions of the old cores used by Atwater and his team are unreliable and that better technology has allowed scientists to see things in the cores today that researchers in the 1960s may not have.

“The very same cores that were shown to have a poor record [of turbidites] actually have a very good record” of turbidites, Goldfinger said.

However, turbidity currents can be caused by other events, including floods or slope failures, said David Piper, a research scientist with the Geological Survey of Canada who reviewed drafts of the Geology paper. In other words, according to Piper, in regions where there is not much sediment to deposit in the first place, it may not be possible to confidently attribute a turbidite to an earthquake.

Thus, the Geology paper calls into question the accuracy of earthquake records sourced from turbidites. “[There is] not in many cases a particularly good relationship between even submarine landsliding and large earthquakes, and there certainly isn’t a good systematic relationship between large earthquakes, landsliding, and turbidity current deposition,” Piper said.

Tark Hamilton, a geosciences professor at Camosun College in Victoria, British Columbia, who was not involved with the 2012 report or the Geology paper, said that because the region in question is sediment starved, the old cores cannot necessarily be used as negative evidence either. The turbidites that were found in the cores could be attributed to other factors, but “it doesn’t mean there [were not] earthquakes.”

“There is a record out there,” Hamilton said, “[but] it’s more difficult to tease out. There are good people working to bring it forward, and we will get the answer collectively.”

—JoAnna Wendel, Staff Writer

© 2014. American Geophysical Union. All rights reserved.

© 2014. American Geophysical Union. All rights reserved.