First Author

Even if they have not experienced them, most people are familiar with earthquakes. The violent tremors occur when pressure builds up between tectonic plates at a fault line, causing the plates to slip, releasing energy in the form of seismic waves. However, there are other types of tremor that occur deeper inside Earth — 25–40 km down — that cannot be felt at the surface. Since their discovery, the origin of these 'non-volcanic tremors' has puzzled researchers. On page 1048, Amanda Thomas and her colleagues at the University of California, Berkeley, report a connection between extremely small stresses induced by tides and non-volcanic tremors detected near the San Andreas fault in Parkfield, California. Thomas tells Nature more.

When and where were non-volcanic tremors discovered?

In 2002, in a region beneath the eastern coast of southwest Japan. In 2003, more were detected in Cascadia, beneath the coast of Canada's Vancouver Island. These are both subduction zones, where Earth's crustal plates collide, and one moves underneath the other. At these sites, tremors occur in conjunction with ultra-slow-motion slip of the plates that occurs at regular intervals and is affected by the rise and fall of ocean tides.

How did you get involved in the work?

In 2003, my co-author Robert Nadeau detected tremor-like signals originating near the San Andreas fault, which is not a subduction zone. The tremors were similar to the subduction-zone tremors but with some fundamental differences. For example, the Parkfield tremors were less dominated by periodic episodes and had smaller energies. And no one has found evidence of slow-slip events occurring there.

So what causes the tremors in Parkfield?

We calculated that shear stress produced by tidal activity could trigger them. Tides generate extremely small stresses, several orders of magnitude lower than those caused by ambient pressure. Yet, when it is riding on top of the huge background stress of the plates pushing against each other, the tidal-induced shear stress is enough to push the system over the stress threshold required for shear failure. The same mechanism may apply to subduction zones.

Why are the findings important?

These tremors may be key to understanding fundamental processes that occur at the roots of faults. Some people have suggested that non-volcanic tremors could signal accelerated slip between plates, and thus increased danger of earthquakes. But we don't have enough data to make that connection.