by Will Chase


Can one natural disaster influence the risk of another? It's well known that natural disasters can cause others in their immediate vicinity, for instance, hurricanes are often accompanied by flooding, and earthquakes are followed by aftershocks. But what about longer distance interactions? Could one earthquake trigger another on the other side of the world? hey click me for more info!

Our story begins in California in June of 1992. The dots below represent earthquakes recorded in the ten days from June 18th to June 28th. Their size and color correspond to their magnitude: larger earthquakes are bigger and more red. There were 44 quakes during these ten days, though most were small and wouldn’t have been felt by people. But in the early hours of June 28th, that changed.

At 5:00 on the morning of June 28th, a magnitude 7.3 earthquake struck, waking much of Southern California. The epicenter was located near Landers California, in the Mojave Desert. Thanks to the remote location, the damage was minimal.

As expected, the mainshock was followed by a smattering of aftershocks which occurred in the immediate vicinity of the mainshock, but what played out over the next ten days was entirely unexpected.

Survey stations across the Western United States recorded a flurry of seismic activity in the ten days following the Landers earthquake. Seismologists recorded 192 earthquakes outside of the aftershock zone during this time—four times the amount in the ten days proceeding the Landers quake. Even more surprising was that these earthquakes occurred at great distances from the aftershock zone, with many recorded hundreds or even thousands of miles from the mainshock epicenter.

Placing each earthquake on a timeline before and after the mainshock reveals the striking pattern. There is an especially large burst of activity in the three days following the mainshock, after which things taper off. This event was one of the first reports of remotely triggered earthquakes to gain wide recognition. Most scholars were convinced that this increase in seismic activity was triggered by energy sent out from the Landers quake, but many questioned how relevant it was due to the small size of most of the triggered quakes.

Most earthquakes are small (M2.5–M4.0). These quakes are not typically felt by people, and are only recorded by sensitive seismographs. Earthquakes of magnitude 4.0–5.0 can cause shaking that might knock things off your shelf, but they’re unlikely to cause structural damage. What we’re really concerned about are quakes greater than M5.5; only one of the 192 earthquakes following the Landers mainshock met this criterion.

Over the next 15 years, there were numerous papers published on remote triggering of earthquakes. It became well accepted that large earthquakes often trigger small earthquakes at long distances. However, the question of whether source events could remotely trigger large earthquakes (M5.5+) remained contentious. Several researchers argued that the largest earthquakes (M8.0+) could remotely trigger M5.5+ quakes, but not all were convinced. Scientists squabbled over statistical details and seismic mechanisms until 2012, when we got a unique glimpse into the possibilities of remote triggering.

The spring of 2012 was an unusually quiet time for large earthquakes. In the ten days from April 1st to April 11th there were only five M5.5+ earthquakes. Note that from here on we will only consider earthquakes larger than M5.5.

On the afternoon of April 11th 2012, a massive M8.6 earthquake struck off the coast of Sumatra. The temblor triggered many aftershocks (not shown here). This quake was unique for its size and mechanism. To date, this is the largest ever recordedstrike-slip earthquake.

In the ten days following the Sumatra mainshock, 44 earthquakes M5.5+ were recorded. The statistics were clear: this nine-fold increase was highly significant, and most researchers agreed the likely explanation was remote triggering by the M8.6 Sumatra earthquake. This massive triggering of large earthquakes was unprecedented, and speculation and investigation into a possible mechanism or explanation started immediately.

When earthquakes strike, they send out seismic waves that travel through the earth’s crust. These waves often propagate out from the mainshock in a '+' pattern meeting at the opposite side of the earth (the antipode). Analysis of the Sumatra quake showed that nearly all of the triggered earthquakes fell within the seismic waves from the mainshock.

The 2012 Sumatra quake showed that it was possible for a large earthquake to remotely trigger other large earthquakes, but the question remained: how common was this? Many argued that the Sumatra quake was an outlier, and rigorous statistical analysis was needed to prove otherwise. Recently, a small group of researchers at Oregon State University carried out an analysis of 47 years of earthquake data to look for statistical evidence of remote earthquake triggering. Their method took each large mainshock (M8.0+) and then divided the globe into slices of ten degrees each emanating outward from the mainshock (the 30 degrees surrounding the mainshock was disregarded as it was already known to have elevated earthquake levels). For each slice, they compared the rate of M5.0+ earthquakes in the three days following the mainshock to a historical average to determine a relative rate of earthquakes.

The researchers found that in the three days following the largest earthquakes (M8.0+), the rate of M5.5+ earthquakes at certain locations was elevated as much as twice the average. Two main regions—around 80 degrees and in the 30 degrees surrounding the antipode—have significantly elevated rates. The area around the antipode is particularly intriguing, as that is typically where seismic waves from the mainshock converge. Although suggestive, skeptics may argue that these results merely show a correlation—without causation—between large mainshocks and elevated rates of M5.5+ quakes. However, there have been a handful of documented cases of remote triggering of large quakes with a demonstrated seismic mechanism that more convincingly show causation.

On September 11th 2008, a M6.6 earthquake struck off the coast of Halmahera, Indonesia. The quake sent a burst of seismic waves towards Japan...

Twenty-one minutes after the Papua New Guinea mainshock, a M6.9 eathquake was recorded off the Eastern coast of Hokkaido, Japan. The Hokkaido quake hit mere seconds after the seismic waves from the Papua New Guinea mainshock landed. Detailed analysis showed that the Hokkaido quake was directly triggered by the seismic waves from the Papua New Guinea quake. Though this is perhaps the most compelling example, there have been a handful of other studies demonstrating similar results of direct remote triggering via a seismic mechanism.