Around 1,100 years ago, seismic activity rattled what is now the Puget Sound region, where Seattle is located. The event unleashed landslides, raised cliffs, flooded forests, and formed lakes. The land bears the scars of this seismicity on multiple faults. But it hasn’t been clear whether these faults ruptured within hours, months, years, or even a century of each other. Now, an analysis of tree rings has pinned the activity on two adjacent faults to a 6-month period. Shaking could have come from two closely spaced earthquakes or a single quake on both faults that exceeded the magnitude thought possible in the Puget Sound region.
“This is a blockbuster paleoseismic result,” said Rich Briggs, a research geologist at the U.S. Geological Survey (USGS) who was not involved with the study. Typically, scientists glean clues about the timing of ancient earthquakes from radiocarbon dating of plants or charred wood. But it can be hard to determine how these recorders relate to an earthquake, he said. “Usually, we’re dealing with decades to centuries of uncertainty in the ages of fossil earthquakes.”
This team instead looked to the annual growth rings of trees that died during the Puget Sound event when they were buried beneath mud or submerged by water. The researchers rounded up Douglas fir samples previously harvested from five areas near the Seattle and Saddle Mountain Faults. Earlier work showed these faults may have been the source of shaking. At a sixth site—Price Lake in North Cascades National Park—divers used a chainsaw under nearly zero visibility to cut sections of now submerged wood that had been buried beneath mud.
The locations of the trees with respect to the faults showed that seismicity on the Seattle Fault could have killed trees at five of the sites, whereas those at Price Lake could have been affected only by the Saddle Mountain Fault.
Critically, many of the samples contained the bark of the tree. Bark, which is on the outside of the last growth ring, marks the end of a tree’s annual records. “We could tell exactly when the tree died,” said Bryan Black, a dendrochronologist at the University of Arizona and lead author of the study.
Black and his colleagues used the patterns in the trees’ rings, which vary in thickness depending on climate conditions, as a barcode of their lives. “We found that those patterns matched across our six sites,” Black said. “Those trees lived together and, importantly, died together.”
All the trees had lived through a cosmic event—possibly a solar storm—that caused a global spike in radiocarbon between 774 and 775 CE. This event allowed the team to validate their dates across all the trees. And by comparing the patterns with that of a sample from an unaffected fir in Vancouver, the team determined that the trees felled by quakes must have died after their growth season. That narrows the window of time for the seismicity to 6 months straddling the end of 923 and the beginning of 924 CE, the team reported in Science Advances. Because the trees across all six sites died at once, both the Seattle and Saddle Mountain Faults must have ruptured during that period.
Shaking Up Earthquake Plans
On the basis of the sizes of the faults, the team estimated that they produced two quakes of magnitude 7.5 and 7.3 within 6 months of one another or one 7.8 magnitude quake that spanned both faults. “That’s a very violent large quake that’s on par with what happened in San Francisco in 1906, for example,” Black said.
This research is “really important,” said Diego Melgar, a seismologist at the University of Oregon in Eugene. People have been worrying about the Seattle Fault for decades, he said. The study is “telling us that really big earthquakes have happened in the past and we have every reason to believe that they’ll happen in the future as well.” Knowing about the timing and intensity of the quake(s) could help scientists understand the future hazard.
Both scenarios the team reported are worrisome, Briggs said. A single large quake can cause very high ground motions that need to be considered when designing buildings. And a large quake on the heels of another earthquake could deal a blow to buildings and infrastructure that have already been weakened, as witnessed earlier this year in Türkiye and Syria.
“Neither of these is a better option,” Briggs said. “They’re both different options that need to be understood and modeled.” Simulations based on the physics of earthquakes could help to reveal what these faults are capable of.
The study raises many questions for future research, and it’s not yet clear what this means for building design, said Susan Chang, a geotechnical engineer at the Seattle Department of Construction and Inspections.
“If it’s a double event or a series of events as opposed to one event, that has a lot of implications for emergency management and performance of buildings,” Chang said.
New designs for tall buildings may need to factor in the possibility of a large shock or back-to-back quakes. Chang said she expects this new information will also be considered the next time that USGS scientists revise their ground motion maps, which the city bases its building codes on.
—Carolyn Wilke (@CarolynMWilke), Science Writer
This news article is included in our ENGAGE resource for educators seeking science news for their classroom lessons. Browse all ENGAGE articles, and share with your fellow educators how you integrated the article into an activity in the comments section below.