Tides raised by the moon and the sun press on Earth’s crust with a force of only a few kilopascals, roughly a gentle hand pushed against a table. A study in the Journal of Geophysical Research: Solid Earth argues that this is enough to set off slow earthquakes on faults such as the ones beneath Cascadia, if the timing of the push matches the fault’s own rhythm.
Yishuo Zhou, Ankit Gupta, Hideo Aochi, Alexandre Schubnel and Harsha S. Bhat of the Laboratoire de Géologie at the École Normale Supérieure in Paris, together with Satoshi Ide of the University of Tokyo and Pierpaolo Dubernet, built a computer model of a fault and tested when a tidal nudge produces slip. Their paper, “Theoretical constraints on tidal triggering of slow earthquakes,” was accepted on September 11 and published online on September 20, according to the journal record. The abstract closes by saying observed and model-predicted tidal correlation patterns may help constrain the frictional strength of the plate interface.
A hand-press force and a swing pushed on the beat
Slow earthquakes release energy over days rather than seconds, so nobody feels them. The authors write that although the stresses from tides “are very small, typically of the order of a few kilopascals and comparable to the pressure from a gentle hand press, they have been observed to trigger slow earthquakes on some faults,” a passage quoted in ScienceAlert’s report on the study. The puzzle is why a force that small matters on a fault carrying far larger stresses.
Their answer is resonance. The model is a spring-block system with rate-and-state friction, a standard way to describe how rock surfaces stick and slip, and it shows that when the period of the tide matches a fault’s natural response timescale, slip is amplified, “much like pushing a swing at the right rhythm makes it move higher,” in the paper’s own comparison. The abstract states the result in plainer terms: even small stress perturbations can trigger periodic as well as temporally complex slip events.
The paper itself, available through its DOI record, reduces the problem to a small set of variables. Two dimensionless numbers govern when that happens, according to the preprint posted in February: a normalized tidal period between about 2 and 70, and a normalized stress amplitude of at least 0.2. Inside that window, slip phase-locks to the tidal cycle. The authors constrain tidal stress amplitudes to between 0.1 and 100 kilopascals when solid-Earth tides and ocean loading are combined, and the fault’s instantaneous frictional strength to between 0.5 and 500 kilopascals.
What Cascadia shows
The observation that the model is built to explain comes from the field. In southwest Japan and in Cascadia, tremor rates peak at 12-hour and 24-hour intervals, matching the cycle of the tides. The Pacific Northwest has long been a testing ground for that link. Heidi Houston, in a 2015 Nature Geoscience paper, compared seismic data with tidal stress calculations for 31,000 tremors generated by six large slow-slip events in Cascadia between 2007 and 2012 and found that tremor’s sensitivity to tides rises as slip accumulates during each event.
Cascadia’s slow slip, known as episodic tremor and slip, is well documented, and it runs from British Columbia to northern California. The Pacific Northwest Seismic Network says events recur about every 14 months in northern Washington and about every 22 months in central Oregon, and that an average event in northern Cascadia produces around 2 centimeters of slip on the fault at depth. Nearly every year, it adds, a ground shift equivalent to a magnitude 6 earthquake occurs below the Seattle area through these processes, in the network’s explainer on slow slip.
Forecasting claims, and their limits
The practical hope is diagnostic. Because the model ties the timing of tremor within a tidal cycle to fault properties, seismologists could work backward from detected tremor to estimate frictional strength on a subduction interface. The authors say the constraints could help assess the rupture potential and spatial extent of future megathrust earthquakes, and the PNSN notes that slow slip may transfer strain to the locked zone, “bringing it incrementally closer to failure.”
The study does not claim a forecast. It models isolated fault patches, not the large multi-segment ruptures that define a magnitude 9 event, so its findings speak to how strain is stored along Pacific Rim subduction zones more than to when the next big rupture will come.
Whether real Cascadia patches sit inside the resonance window is a measurement problem: it needs independent estimates of friction on those patches, and the paper’s model applies to a fault only when that friction and the tidal period line up. The open question for Cascadia is whether its tremor timing, matched against tides, gives a number that can be checked.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
