Recent monitoring of the seafloor off the coast of New Zealand has revealed that episodes of slow rupture of Earth’s crust can occur in the same shallow portion of a fault zone where tsunami-generating earthquakes originate. The new finding deepens scientists’ suspicions that the type of slow-motion, or “silent,” earthquake that the researchers detected, known as a slow-slip event, may signal or even trigger the onset of tsunami-generating earthquakes.
“There seems to be a link potentially with where these really shallow slow slip events happen and where tsunami-generating earthquakes happen,” said Laura Wallace, a research scientist at the University of Texas at Austin. Wallace is the lead author on a paper published yesterday in Science about the new observations.
In undersea trenches in many parts of the world, one of the Earth’s tectonic plates pushes beneath another, or subducts, in an inching, lurching process that builds up stress in the subduction zone that can catastrophically release as an earthquake, which in turn can trigger a tsunami. However, in slow-slip events that are common at subduction zones, sliding of the plates at a more rapid pace than usual (although still much slower than the sudden shift of an earthquake) relieves stress in the fault zone over a period of typically days to months.
In the new research, Wallace and her colleagues studied a subduction zone called the Hikurangi margin off the east coast of New Zealand’s North Island, where the Pacific Plate slowly slides under the Australian Plate. In 1947, two earthquakes at shallow depths within this margin sent tsunamis crashing onto New Zealand’s shore, damaging buildings and roads.
Slowly Slipping Plate
Scientists have long tracked tectonic plate movements at subduction zones—for example, in Japan, Costa Rica, and the U.S. Pacific Northwest—by using land-based GPS monitors on the overlying plate that can detect motions deep below. However, gathering data on plate behavior at the relatively shallow, undersea trench that lies offshore and is where the leading edge of the overlying plate meets the subducting plate has proven to be a difficult task requiring other sorts of sensors. GPS observations typically reveal slow-slip events in subduction zones at depths 25–50 kilometers deeper than the trench, Wallace said.
Onshore near the Hikurangi margin, GPS measurements have shown that slow-slip events occur roughly every 18 months, the larger ones taking place once every 4–5 years, according to Wallace. To find out if slow slip could be observed at shallow depths and offshore near the trench, she and her team picked a yearlong window of time when they would mostly likely detect a slow-slip event. Then, in May 2014, they deployed an array of 15 ocean bottom seismometers and 24 seafloor pressure gauges on the seafloor to record possible events.
Between May 2014 and June 2015, the pressure detectors revealed vertical movements of the ocean floor by, in essence, weighing the overlying water column: Higher pressure meant that the seafloor sank and more water pressed down, whereas a lower pressure indicated a rising seafloor, which displaced water and decreased the pressure.
After analyzing the data, the researchers found a slow-slip event that lasted 2 weeks and moved the seafloor upward 1.5–5.5 centimeters—a vertical movement associated with 15–20 centimeters total slippage along the plate boundary. That shift equates to 3 to 4 years of normal plate movement, Wallace said.
The newly revealed slow-slip rupture took place in the same shallow portion of the subduction zone where the 1947 tsunami-generating earthquakes had originated. If this slip had occurred suddenly, rather than over the course of 2 weeks, it would have clocked in as a magnitude 6.8 earthquake, the researchers report.
“Our results clearly show that shallow, slow-slip event source areas are also capable of hosting seismic rupture and generating tsunamis,” said Yoshihiro Ito of Kyoto University in Japan, who coauthored the study.
Future Earthquake Monitoring
Earlier this year, another Science paper reported that slow-slip events often occurred before an earthquake of magnitude 5 or higher hit. In fact, a swarm of slow-slip events preceded the devastating 9.0 magnitude earthquake and tsunami that hit Japan in 2011.
Scientists have typically detected slow slip at subduction zones at tens of kilometers beneath the trench, where temperatures reach 350°C–450°C and pressures are high, Wallace said. They suspected that slow-slip events also occurred in shallower regions of trenches, less than 10–15 kilometers deep, where pressures and temperatures are lower and tsunami-generating earthquakes originate.
The new findings indicate that slow-slip events can, indeed, happen “over a massive range of conditions,” from warm temperatures and high pressures within the crust to shallow locations, cooler crustal temperatures, and lower pressures, Wallace continued. “This is important to know because we don’t really understand yet why these slow-slip events happen.” She added that slow-slip events might trigger earthquakes in a subduction zone by providing stress relief in one area that causes stress buildup somewhere else, leading to a sudden rupture.
“These data should aid in better understanding these somewhat enigmatic shallow tsunami earthquakes,” said Roland Burgmann, a seismologist at University of California, Berkeley, who wasn’t involved in the research.
Wallace and her colleagues plan to investigate the Hikurangi margin in the future by drilling into the seafloor to figure out what causes slow-slip events in the first place.
Having now observed slow slip close to the epicenter of the 1947 earthquake underscores the need for monitoring, Wallace said, “because there’s potential for a slow-slip event to trigger an earthquake that could generate a big tsunami.”
—JoAnna Wendel, Staff Writer
Editor’s Note, 6 May 2015: For a detailed look at research efforts off New Zealand’s coast that led to the results described above, read the Eos.org project update.
Citation: Wendel, J. (2016), Undersea data tie slow fault slip to tsunami-causing quakes, Eos, 97, doi:10.1029/2016EO051993. Published on 6 May 2016.