There is no natural disaster sneakier than an earthquake. Hurricanes can be predicted and tracked weeks in advance, and even tornados, monsoons and blizzards at least have seasons. But earthquakes strike entirely without warning. Now, however, a new study suggests that we may want to brace for a surge of quakes in the year ahead, and the reason for the danger is an unlikely one: the rotation of the Earth has slowed slightly.
While accurately forecasting earthquakes is impossible, a backward look through the seismic record allows geologists to detect some distinct patterns. In the new study — which was presented at the annual meeting of the Geological Society of America, in Seattle, and published in Geophysical Research Letters — geologists Roger Bilham of the University of Colorado, Boulder, and Rebecca Bendick of the University of Montana, tracked the incidence of magnitude 7 or greater earthquakes worldwide since 1900. While in most years there is an average of just 15 such major shake-ups — already more than enough — there have been evenly spaced intervals in the past 117 years in which the annual total jumped to between 25 and 30.
A little more than a century on a planet that is more than 4 billion years old is not exactly a representative time sample, but Bilham and Bendick noticed something else about these volatile, quake-prone periods. They seem to follow periodic slowdowns in speed of the Earth’s rotation. Our solid planet is a lot less solid than it seems, and that’s true not just of its oceans and air, but of its outer core, which is about 1,200 mi. (2,200 km) thick and is composed mostly of liquid iron and nickel. That molten ooze tends to slosh about, following a pattern that oscillates more or less predictably over time, much the way — on a vastly smaller and more fleeting scale — water sloshing in a bucket will move back and forth in a repeating cycle.
Such motion deep inside the Earth slightly changes the planet’s rate of spin, adding to or subtracting from the 24-hour day by about a millisecond — a change that is regularly recorded by atomic clocks. When a slowdown occurs, the molten core continues to strain outward, obeying Newton’s fundamental law that objects in motion will try as hard as they can to remain in motion.
That outward pressure slowly propagates through the rocks and plates and faults that lie above it. Bilham and Bendick calculate that it takes five to six years for the energy sent out by the core to radiate to the upper layers of the planet where quakes occur, meaning that after the atomic clock notices a slowdown you’ve got five to six years before you’d better buckle up.
The last such time the planet slowed was in 2011, and recent events suggest a troubling pattern again playing out: the magnitude 7.1 quake that struck Mexico City on Sept. 19; the 7.3 event on the Iran-Iraq border on Nov. 12; and the 7.0 off New Caledonia on Nov. 19.
Not only does the new study suggest when there could be an uptick in quakes, it also points to where: near the equator, within a latitude of 30º north or south. It makes sense that this would be the danger zone because any given point along the equator — the planet’s widest point — rotates up to 1,000 mph (1,600 k/h) faster than a point closer to the poles, so a slowdown in the overall spin would be more powerful along that midline. The Iran-Iraq quake occurred at about 33º north latitude, exceeding that cartographic limit, but not by much.
None of this says that 2018 will definitely be a more geologically unstable year, and it certainly doesn’t pinpoint where any possible quaking will occur. It does say that the maddeningly imprecise science of earthquake prediction has at least gotten a tiny bit more precise. For disasters with such deadly stakes, even that small improvement makes a difference.