Thirteen Years of Offshore Data Suggest the Cascadia Subduction Zone Is Not Fully Locked

The Cascadia Subduction Zone has a reputation for being quiet - dangerously quiet. Spanning more than 600 miles from northern California to British Columbia, the fault where the Juan de Fuca plate dives beneath North America produces almost no small earthquakes, unlike most subduction zones elsewhere in the world. That silence has long been interpreted as evidence that the two plates are locked tightly together by friction, building stress that will eventually release in a megathrust earthquake. The last one struck in January 1700, generating a tsunami that reached Japan.

What happens to accumulated stress when the plates finally slip - whether a rupture propagates along the entire fault length or stops short - depends partly on the physical properties of different sections of the fault. A study published in Science Advances by University of Washington researchers uses 13 years of data from offshore sensors to show that the fault is more variable than the locked-and-quiet picture suggests.

What seismic velocity can tell you about what is happening underground

The study's approach relies on an indirect measurement technique. Seismometers can detect not just earthquakes but ambient noise - the constant background vibration of oceans, weather, and human activity - and measure how fast that noise travels through the rock. Seismic velocity changes when the stress state or fluid content of rock changes: compressed rock transmits sound faster; rock that is fractured or saturated with fluid transmits it more slowly.

Lead author Maleen Kidiwela, a UW doctoral student, and colleagues analyzed data from three monitoring stations: one near Vancouver Island in the northern Cascadia zone, and two off the coast of Oregon in the central region. They tracked how seismic velocity changed over 13 years, looking for patterns that might reflect processes occurring beneath the ocean floor.

Two regions, two different stories

At the northern Vancouver Island site, seismic velocity increased steadily over the monitoring period - a pattern consistent with ongoing rock compaction as the plates press together. This supports the standard picture of two locked plates accumulating stress. That story appears to hold in the north.

The central Oregon sites told a different story. For two months in 2016, seismic velocity dropped noticeably. The researchers attribute this to a slow-motion earthquake - a "slow slip" event - on the shallow edge of the oceanic plate that released some accumulated pressure without producing perceptible ground shaking. Slow slip events have been documented at other subduction zones and are thought to periodically relieve stress that would otherwise contribute to eventual rupture.

Additional velocity drops recorded between 2017 and 2022 were linked to a different process: fluid dynamics. When subducting rock is compressed, water is squeezed out and pushed upward. The data suggest that faults running perpendicular to the subduction zone may be acting as channels for this pressurized fluid to escape, relieving pressure in the process.

What this means for earthquake risk

The findings do not change the overall probability of a large Cascadia earthquake. Researchers estimate a 10 to 15 percent chance that the full fault ruptures in the next fifty years, potentially producing a magnitude 9 or greater earthquake. That estimate stands.

What the new data could change is the anticipated behavior of such an event. In other subduction zones, regions with active fluid systems and slow-slip behavior have interrupted ruptures that would otherwise have propagated further. "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. Whether the central Cascadia region with its fluid venting would act as a barrier to full-fault rupture is an important and open question that the current data cannot definitively answer.

The value of more instruments

Three monitoring stations across a 600-mile fault produced complex findings. The UW team received $10.6 million in 2023 to build a more comprehensive underwater observatory in the Cascadia zone. Co-author William Wilcock, a UW oceanography professor involved with the observatory, noted that the current results suggest the value of that investment: "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. It suggests that we need more instruments there, because there may be more going on than people have been able to figure out before."