Researchers practice deploying a wave glider that will help measure seismic strain off the Mexican coast.

LIQUID ROBOTICS, A BOEING COMPANY

Underwater network hunts for mysterious slow quakes

On a dimly lit loading dock at the National Autonomous University of Mexico (UNAM) in Mexico City, geophysicist Víctor Cruz-Atienza heaves the lid off a wooden crate to reveal a device that could help solve a geophysical mystery. Peeking through the packing material is the yellow body of a wave glider, an aquatic drone that's the size and shape of a chunky surfboard with underwater wings and that can trace programmed paths across the ocean's surface. Next week, Cruz-Atienza and colleagues from Mexico and Japan will embark on a cruise to place sensors on the sea floor off the Mexican state of Guerrero. Then they'll release the wave glider to roam overhead, gathering sensor data that could indicate whether stresses that might one day produce a big earthquake are building up on the sea floor—or quietly dissipating.

The sensors are part of a $6 million network, funded jointly by Mexico and Japan, that includes nearly 70 new seismometers and GPS stations on land, as well as on the sea floor. What it reveals could affect tens of millions of people. Most of central Mexico's Pacific coast roils with earthquakes as the ocean's tectonic plates force their way underneath the continental plate of North America. But a 130-kilometer stretch of Guerrero's coast has been seismically silent for more than 100 years (see map, below).

Researchers have long feared that stress is mounting in the so-called Guerrero gap, and that the accumulated energy could one day unleash an earthquake of magnitude 8 or more. That could topple buildings as far away as Mexico City, which is still reeling from a magnitude-7.1 earthquake on 19 September that killed more than 300 people. But recently, a new hypothesis has emerged: Perhaps most of the Guerrero gap's pressure has already been relieved by puzzling "slow slip events," in which swaths of Earth's crust shift by several centimeters—not in seconds, as in ordinary earthquakes, but over weeks or months.

Slow slip events were discovered less than 2 decades ago in the Cascadia subduction zone off the coast of Washington, Oregon, and British Columbia in Canada; they have since been observed in subduction zones around the world. "They're pretty widespread and a really fundamental part of how faults work," says Laura Wallace, a geophysicist at the University of Texas in Austin and the research company GNS Science in Lower Hutt, New Zealand. But they can easily pass unnoticed.

Plugging the gap

New underwater sensors off the coast of the Mexican state of Guerrero will reveal whether slow earthquakes are relieving stress in a feared seismic gap, bypassed by major earthquakes for more than a century.

CREDITS: (MAP) J. YOU/SCIENCE; (DATA) VLADIMIR KOSTOGLODOV

"Guerrero is one of the best places to do slow earthquake science," says Yoshihiro Ito, a seismologist at Kyoto University in Japan and Cruz-Atienza's co–principal investigator. Vladimir Kostoglodov, a UNAM geophysicist, already has observed what appears to be pressure-relieving slow slip there: Landbased GPS stations near Guerrero's coast gradually shift northeast over several years, a sign that the continental plate is deforming and building up stress under the pressure of the oceanic plate. But every 3.5 to 4 years, the stations are carried back to nearly their starting points by what appears to be a slow slip event, moving an average distance of about 10 centimeters over 6 months. So much land shifts during these events that they can release as much energy as a magnitude-7.6 earthquake—but slowly and harmlessly.

Yet slow slips can be "a double-edged sword," Wallace says, relieving stress in one place only to displace it and boost quake risk elsewhere. In 2014, a slow slip event redistributed seismic stress just northwest of the Guerrero gap and triggered a magnitude-7.3 earthquake, according to a 2016 paper in Nature Geoscience. Slow slip events have also preceded major quakes in Japan and Chile.

Scientists hope the new seafloor network will give them a detailed look at whether the terrestrial slow slip events they've observed extend to the marine trench where the oceanic and continental plates meet. If not, an earthquake there remains a risk—and would likely trigger a tsunami that could devastate Acapulco and other coastal communities. "Better understanding the likelihood of tsunami-generating earthquakes on the offshore plate boundaries is critically important. And what these guys are doing [in Mexico] is really the only way to get at that," Wallace says.

The research cruise, which will run from 10 to 23 November, is timed to what should be the beginning of Guerrero's next slow slip event, 3 years and 9 months after the last one. "We're hoping that we can catch it," says Vala Hjörleifsdóttir, a seismologist at UNAM. She, Cruz-Atienza, and Ito will deploy the wave glider, which will determine the precise position of a point on the sea floor using data transmitted by sonar from the sensors plus positional information from GPS satellites.

When the researchers return next year, they will be able to see whether that point has moved, reflecting the deformation of the sea floor due to the grinding of the tectonic plates—and, if they're lucky, the reverse action of a slow quake. If slow slip has kept the point from moving northeastward, it would show that such events are relieving pressure near the trench—and Mexico can breathe a little easier.