Earth’s volcanoes are often located in seismically excitable parts of the world. Just take the so-called Ring of Fire, which is technically a horseshoe-shaped region that traces the edges of tectonic plates around the Pacific basin. This area hosts 90 percent of the world’s recorded earthquakes and 75 percent of all active volcanoes.
In such seismic hotspots, eruptions and earthquakes are often taking place at roughly the same time—but that’s exactly what you would expect. Despite frequent breathless speculation online, you can’t automatically assume that there’s a connection between a given quake and a subsequent eruption.
“The volcano may have already been preparing to erupt, or it’s already been erupting for a long time,” says volcanologist Janine Krippner.
Still, the question of whether earthquakes can cause volcanic eruptions is a serious research topic that experts have been investigating for centuries. And multiple lines of evidence from recent studies suggest that a connection could potentially exist in certain situations. So, where do scientists currently stand on the issue? We’ve got you covered.
Joining up some dodgy dots
Atsuko Namiki, associate professor of geosciences at Hiroshima University, highlights a few geophysical studies with data that suggest a connection. A 1993 paper, for example, links a magnitude 7.3 quake in California to volcanic and geothermal rumblings immediately afterward. And a 2012 study reckons that a magnitude 8.7 earthquake in Japan in 1707 forced deeper magma up into a shallow chamber, triggering a huge blast at Mount Fuji 49 days later.
Even the ever-cautious U.S. Geological Survey says that sometimes, yes, earthquakes can trigger eruptions. The agency suggests that some historical examples imply that an earthquake’s severe ground shaking, or its ability to otherwise change the local pressure surrounding the magmatic source nearby, can trigger volcanic unrest. They cite the magnitude 7.2 earthquake on Hawaii’s Kilauea volcano on November 29, 1975, which was quickly followed by a short-lived eruption.
But there are problems. First, as the USGS stresses, the triggering mechanisms for such events are not well understood, and papers linking quakes to later eruptions can really only speculate.
Second, it’s possible that the timing in all these examples was just a coincidence. Geologists must understand the specific triggering and rule out chance before a connection can be definitively made–and Earth’s geological complexities make both extremely difficult.
Darwin’s accidental deception
Statistical analyses are attempting to tackle the chance problem head-on. A 1998 Nature paper investigated whether magnitude 8.0 or larger quakes could trigger explosive volcanism up to 500 miles away from the epicenter within five days. Using data from the 16th century to the present, its authors found that these types of eruptions happened four times more often than chance alone could explain.
Similarly, a 2009 paper used historical data to show that that magnitude 8.0 quakes in Chile are associated with significantly elevated eruption rates in certain volcanoes as far as 310 miles away. The problem is that these sorts of historical data aren’t exactly great.
“Major earthquakes and large volcanic eruptions are both relatively infrequent events, and scientists have only been reliably keeping these records for the last half century or more, depending on the region,” says Theresa Sawi, an undergraduate researcher in geophysics at the University of California, Berkeley.
Many data points in the past come from fairly ambiguous news reports and journal entries. David Pyle, a professor of volcanology at the University of Oxford, points out that one of the earliest writers to link earthquakes and eruptions was none other than Charles Darwin.
In 1840, Darwin gathered eyewitness information on some minor changes at Chilean volcanoes following the powerful quake there in 1836. It’s unclear if any eruptions took place, but “nonetheless, all of these 'events' ended up in the catalog of volcanic eruptions and now appear to offer evidence for earthquake triggering,” Pyle says.
Squeezing out toothpaste
Sawi is a coauthor on a more recent statistical analysis in the Bulletin of Volcanology that tries to circumvent this issue. This study focused only on more scientifically robust data from 1964 onward, and it looked at smaller quakes of at least a magnitude 6.0 that took place 500 miles from a volcanic eruption.
Sawi’s study found that there was a 5 to 12 percent increase in the number of explosive eruptions two months to two years following a major quake.
The team identified 30 volcanoes that may have at some point undergone a potentially triggered eruption. On a scale of days, the team found no evidence for triggering that couldn’t be explained by chance alone. That result actually goes against one of the findings of a 2006 review featuring Michael Manga, a coauthor on the new paper.
“It's nice to see researchers not being afraid to make conclusions that go against their previous work,” said Oliver Lamb, a volcanologist at the University of North Carolina at Chapel Hill. “This is how science should work, really.”
Curiously, Sawi’s study found that there was a 5 to 12 percent increase in the number of explosive eruptions two months to two years following a major quake. This jump is both surprising and interesting, according to Lamb, but it’s also pretty small.
Jackie Caplan-Auerbach, an associate professor of seismology and volcanology at Western Washington University, says that the paper “actually highlights how unlikely it is that a quake could trigger an eruption.”
How, then, could this long-term trend be explained? What may happen during those months is that ruptures caused by quakes open up new pathways for viscous magma to follow, gradually, to the surface. The shaking, over time, could also create additional bubbles in the magma, which increases its pressure–a bit like shaking up a can of soda.
Perhaps the movement of rock can squeeze magma bodies like a tube of toothpaste, Sawi says, slowly forcing the magma out via volcanic exit routes. Or the quake may stretch the rock around a volcano’s magma reservoir, which would prompt gases to bubble out of the molten rock and increase the pressure in the reservoir.
Honey, I shrunk the volcano
Caplan-Auerbach suspects that if a quake does trigger an eruption, then the volcano has got to be primed and ready to go when it strikes. But while it might seem “intuitively reasonable that large earthquakes might trigger activity at a volcano that is poised to erupt, the empirical evidence for this link is rather thin,” Pyle says.
Some scientists, like Namiki, are hoping to find such evidence. She and her colleagues design models of volcanic systems in the lab and shake them about to examine how triggering could physically take place.
In a 2016 study, his team used syrups with varying crystal numbers, bubble quantities, and so forth to simulate various magma reservoirs. They found that at the resonant frequency, the frequency at which an object can naturally vibrates, the back-and-forth sloshing of the “magma” was most prominent. Bubbles joined up, and the frothy foam atop collapsed. In a real volcano, this would allow hot gases to readily escape from the magma, increasing the reservoir’s pressure and potentially pushing the volcano to erupt.
In 2018, the team also published a study of a gel model of a volcano injected with fluids simulating different types of magma. They found that shaking the model caused the fluids to move around faster than they otherwise would. However, where the fluids went was tied to their buoyancy and storage depths. Less buoyant fluids moved sideways or downward, which in a real volcano would make an eruption less likely. But bubbly fluids at shallow depths ascended, something that could lead to an eruption.
Keeping their eyes to the ground
It’s certainly not straightforward, and Namiki notes that skepticism about eruptions triggered by earthquakes is perfectly natural. However, Eleonora Rivalta, group leader of earthquake and volcanic physics research at GFZ Potsdam, suggests that the mood may be slowly shifting toward the possibility of a connection.
“While the wider scientific community may still be a bit skeptical, many volcano geophysicists are now convinced volcanoes may indeed react to earthquakes with a variety of responses,” she says. She emphasizes, however, that the smoking gun is still missing–specifically, a clear demonstration of how exactly an eruption was triggered at a specific volcano by a specific earthquake.
There are other avenues to explore outside statistics and lab simulations. Pyle suggests that if certain volcanoes are thought to be triggered by quakes, then the volcanic debris they eject could hold clues about the state of the magma reservoir prior to their outburst. That may reveal if the quake really did make a significant difference, or it may show that they were primed to erupt anyway and the quake just accelerated the countdown.
For Sawi, the way forward is clear: “Increased monitoring of volcanoes worldwide, especially those historically under-studied volcanoes, would help provide the data needed to begin recognizing patterns and, yes, triggers that could indicate a heightened probability of eruption.”
Editor's Note: This story has been updated to reflect Oliver Lamb's most current affiliation.