Oregon scientists found two West Coast faults have ruptured together three times in 1,500 years

Geologists at Oregon State University have found that two of the West Coast’s biggest fault systems have ruptured within minutes to hours of each other three separate times over the past 1,500 years, a pattern that undercuts the assumption that the Cascadia subduction zone and the northern San Andreas fault behave as independent hazards. The study, built from roughly 3,100 years of seafloor sediment cores, was published in the journal Geosphere and re-reported this week as new modeling attention turns to the Pacific Northwest’s earthquake risk. Chris Goldfinger, the Oregon State marine geologist who led the work, said the synchronized events describe a scenario worse than a Cascadia rupture on its own.

Three ruptures, minutes to hours apart

Goldfinger’s team identified three instances in the sediment record where the Cascadia megathrust and the northern San Andreas fault broke within minutes to hours of one another, with the most recent pairing dated to around 1700, the same year geologists have long tied to the last great Cascadia earthquake. Two faults that stretch along roughly two-thirds of the West Coast, in other words, have a documented history of triggering each other rather than rupturing in isolation.

People are used to hearing the “Big One” — a Cascadia rupture alone — described as the region’s worst-case earthquake, Goldfinger told Oregon State University’s newsroom. “It turns out it’s not the worst case scenario,” he said. A joint rupture, his team’s sediment data suggest, would be — and the historical record now shows it has happened before, not just once.

Reading the pattern in seafloor sediment

The evidence comes from marine sediment cores drilled off the Oregon and Northern California coasts, which preserve turbidites — the disturbed layers of mud and sand that slide down the seafloor when a large earthquake shakes them loose. Goldfinger’s group compared turbidite sequences from Cascadia sites with a separate record from the northern San Andreas, dated the layers with radiocarbon methods, and looked for pairs of disturbances close enough in age to indicate the two faults moved together rather than coincidentally. The peer-reviewed paper, titled “Unravelling the dance of earthquakes,” describes those matched pairs as evidence of “partial synchronization” between the two systems rather than a single shared fault plane.

Three matches emerged from that comparison, out of a record spanning roughly 3,100 years — a low enough frequency that the authors describe the linkage as occasional rather than routine, but frequent enough to register as a real pattern rather than noise.

A compressed emergency across four cities

What makes the finding consequential is less the rupture mechanics than the emergency-response math. ScienceDaily’s account of the study quotes Goldfinger describing the resource strain of a single Cascadia earthquake, then the added strain of a doubled event: an earthquake on one of the two faults alone, he said, “would draw down the resources of the whole country to respond to it.” A joint rupture would go further, putting “San Francisco, Portland, Seattle and Vancouver all in an emergency situation in a compressed timeframe,” in Goldfinger’s words — four metropolitan areas, spread across two countries, needing federal disaster resources at nearly the same hour rather than in sequence.

The Cascadia subduction zone alone runs from northern California to southern British Columbia, according to the U.S. Geological Survey, placing Portland, Seattle and Vancouver within its shaking radius even without an accompanying San Andreas rupture near San Francisco. A synchronized event, by Goldfinger’s account, would ask federal and state responders to cover all four metropolitan areas at once rather than sequentially.

Gaps the fault record still leaves open

The study does not establish that one fault mechanically triggers the other, only that the timing lines up closely enough, three times, to rule out coincidence as the sole explanation. Goldfinger’s team stops short of assigning the pairing a specific recurrence probability, and the paper calls for additional core sites to test whether the synchronization pattern holds further back than 3,100 years.

Oregon State’s Cascadia sediment work has run for decades, and the USGS’s own Cascadia subduction zone database draws on the same class of turbidite records to estimate rupture intervals for the megathrust in isolation. Folding a second fault into that hazard picture is new, and Goldfinger, whose faculty page at Oregon State’s College of Earth, Ocean, and Atmospheric Sciences lists three decades of paleoseismology work on the West Coast, said the next step is determining how often — not just whether — the two faults have moved in tandem before 1700.

For now, the 1,500-year window holds three confirmed pairings and no fourth, leaving open whether the pattern is accelerating, random, or simply too rare for a 3,100-year record to resolve with confidence.

Emergency planners in Oregon, Washington, British Columbia and Northern California have spent years building response models around a Cascadia-only scenario, stockpiling supplies and mapping evacuation routes for a single subduction-zone rupture. Goldfinger’s data do not replace those plans, but they add a second, harder question underneath them: whether the region’s mutual-aid agreements, which typically assume neighboring states can send help across a shaken area, would hold up if the northern San Andreas broke at the same time and pulled California’s own resources toward its own damage.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview