They're calling it a "rogue earthquake."
On April 11 of this year, one of the 10 biggest earthquakes ever recorded struck off the coast of Indonesia. It was felt from Bangladesh to Australia.
You may not have even heard of this magnitude 8.6 quake. It barely made the news in the U.S. because it did very little damage. Two people died, but there was no massive tsunami.
To understand why this quake was so big, yet not catastrophic, you need to know this: There are two kinds of major quakes in the world. One type happens when two plates of the Earth's crust slide past one another horizontally. That's what happens along California's San Andreas Fault.
The other kind of major quake occurs when two plates collide, and one slips beneath the other with a jolt.
"We expect that the largest earthquakes occur in that kind of a setting, called a megathrust," says Greg Beroza at Stanford University. In fact, he says, since seismologists started recording earthquakes about 120 years ago, every quake this huge has been a megathrust quake.
Or that was true, until last month.
The magnitude 8.6 quake off of Indonesia was actually the side-slipping kind, called a strike-slip. Fortunately, that kind of quake is much less likely to cause a tsunami because sideways motion doesn't generate big waves the way up-and-down motion does (unless it triggers an undersea landslide). That's why it caused relatively little damage.
But how could this sideways-moving quake be so huge? Beroza says the rupture was very deep — with an average depth of 30 miles. At that depth, the rock is very hot. The friction from a rupture would make it even hotter.
"When slip starts to occur, it generates heat, and that heat weakens the rock, and in turn allows more slip. It's called a thermal runaway," he says.
If that thermal runaway happened here, it could explain how this sideways-moving quake got to be so big.
Beroza and co-author Jeffrey McGuire, at the Woods Hole Oceanographic Institution, write in Science magazine about another surprising feature of this earthquake: It was mapped about 60 miles away from the fault zone where two tectonic plates meet.
"What made this earthquake so unusual is that it was not on a plate boundary — at least not on a commonly recognized plate boundary — and it was huge," Beroza says.
That makes it the biggest quake ever recorded that was not on a fault. It dwarfed the best-known example of quakes like that: the 1811 and 1812 quakes centered near New Madrid, Mo. Nobody measured that one at the time, but historical reconstructions estimate the magnitude at around 7.7.
Beroza suspects the April quake in Indonesia — along with a companion quake the same day, magnitude 8.2 — are part of a gradual process that is making new faults and redefining the boundary between plates.
"You have to start them somewhere," he says.
That interpretation is debatable. Kerry Sieh, who runs the Earth Observatory of Singapore, suspects the quake was actually on an existing fault, just not an obvious one.
Sieh was jolted to attention on April 11 because he's been expecting another quake, akin to the 2004 megathrust quake off Indonesia that generated the devastating Christmastime tsunami.
This wasn't the quake he was expecting, but Sieh says it built up even more pressure in the rocks off the coast of Indonesia. So he says it has primed the system for another dangerous tsunami in the years or decades to come.