The Cascadia Subduction Zone is unusually quiet for a megathrust fault. Spanning more than 600 miles from Canada to California, the fault marks the convergence of the Juan de Fuca and North American plates. While other subduction zones produce sporadic rumblings as the plates scrape past each other, Cascadia shows very little seismic activity , fueling assumptions that the plates are locked together by friction.
The subduction zone — miles offshore and deep underwater — is difficult to observe. Most data collection is based onshore, which limits the breadth and quality of results. The lack of earthquakes further complicates efforts to understand its behavior and structure.
In a new study, the first to monitor strain offshore over an extended period of time, University of Washington researchers report that the plates may not be fully locked. Based on 13 years of ground motion data from sensors in different regions, the study shows the northern portion of the fault is locked and quiet, but the central region appears to be more active. There, researchers observed signs of a shallow, slow-motion earthquake and detected pulses of fluid flowing through subterranean channels, which may relieve pressure from the fault.
The findings, published Feb. 27 in Science Advances , may alter expectations of how this area will respond to a large earthquake. Similar features in other places have stopped a rupture that might have otherwise continued along the entire fault line.
“It’s preliminary, but we think that variable fluid pathways in Cascadia will change the behavior of large earthquakes on the fault,” said co-author Marine Denolle , a UW associate professor of Earth and space science.
The Juan de Fuca plate is advancing toward the North American plate at a rate of approximately 4 centimeters a year . But because the plates are stuck together, that motion generates pressure. Eventually, the tension building at the boundary will exceed what the plates can tolerate. When they eventually slip free, an earthquake will spread along the boundary.
Megathrust earthquakes, which occur at boundaries where one plate slides beneath another, rock the Pacific Northwest every 500 or so years. Researchers dated the last one to 1700, and estimates suggest a 10 to 15% chance that the entire fault will rupture, producing an earthquake that could exceed magnitude 9, within the next fifty years. The results from this study do not alter those odds, but the dynamics captured might influence the severity of the eventual earthquake.
A recent survey of the seafloor found that the fault can be separated into at least four geologically distinct segments. Each one may be insulated from a rupture in another region. In this study, the researchers took a closer look at two of the regions by analyzing data from three monitoring stations, one near Vancouver Island and two off the coast of Oregon.
“We wanted to understand strain changes in different regions offshore,” said lead author Maleen Kidiwela , a UW doctoral student of oceanography. “We used the seismometers to measure how the seismic velocity varies underneath each station.”
Seismic velocity is a term used to describe the rate at which ambient noise travels through a material. Because the speed of sound depends on what it is moving through, tracking seismic velocity can give researchers a window into processes occurring beneath the ocean floor.
“When you compact something, you can expect the sound waves to move through it faster,” said Kidiwela.
The steady increase in seismic velocity observed at the northern site told the researchers the rock was compacting, which supports the theory that the two plates are locked in place.
The central region displayed a different pattern. For two months in 2016, seismic velocity decreased. The researchers attribute this drop to a slow-motion earthquake on the shallow edge of the oceanic plate that relieved some of the pressure at the fault.
Other drops in seismic velocity, recorded between 2017 and 2022, were linked to fluid dynamics. Subduction squeezes liquid out of rocks and pushes it toward the surface. The study found that other faults, running perpendicular to the subduction zone, may act as pathways for letting trapped fluid out.
“During a megathrust rupture, one of the ways that an earthquake propagates is through fluid pressure. If you have a way to release these fluids, it could help improve the stability of the fault, and potentially impact how the region behaves during a large earthquake,” Kidiwela said.
Pulling data from just three sites, the researchers observed complex dynamics that may have gone overlooked. Future work will greatly expand this effort. UW researchers received $10.6 million in 2023 to build an underwater observatory in the Cascadia Subduction Zone.
“Finding this link between fluids coming to the shallow subduction zone is pretty unique, as is the evidence that the fault is not completely locked,” said co-author William Wilcock , a UW professor of oceanography and one of the scientists involved with the observatory. “It suggests that we need more instruments there, because there may be more going on than people have been able to figure out before.”
Additional co-authors include Kuan-Fu Feng from the University of Utah.
This study was funded by the Jerome M. Paros Endowed Chair in Sensor Networks at the University of Washington and the National Science Foundation.
For more information, contact Kidiwela at seismic@uw.edu .
Science Advances
Data/statistical analysis
Active protothrusts and fluid highways: Seismic noise reveals hidden subduction dynamics in Cascadia
27-Feb-2026