Fiber-optic cables laid across Swiss glaciers have successfully detected seismic signals produced by the formation of glacier crevasses, highlighting the technology's potential for monitoring icequakes. The research results were presented at the annual meeting of the Seismological Society of America. Glacier crevasses are critical to glacier stability because they can channel meltwater into the glacier bed, speeding up ice flow and causing increased melting. However, on heavily crevassed glaciers, extreme conditions make deploying traditional seismic sensors challenging.

Drone footage of crevasses on the Gorner Glacier in Switzerland. Photo credit: Tom Hudson

Tom Hudson of ETH Zurich explains that the seismic signals produced by icequakes are very different from those produced by tectonic earthquakes caused by shear forces, and from chemical or nuclear explosions involving a rapid release of energy. He said the cracks were "a source of cracks that opened purely in one direction only."

The new study, conducted by Hudson, along with Andreas Fichtner of ETH Zurich (who presented the study at the meeting) and colleagues, "is a real-life example of using fiber optics to detect this type of crack-opening rupture-type seismic activity in the subsurface," Hudson said. "That's about as close as we can get to the earthquake source. Our crack earthquake occurred within ten meters of the fiber optic cable, which is very rare."

This success suggests that fiber optic detection may be useful in monitoring similar cracks that may occur in rocks in carbon storage reservoirs or geothermal energy systems.

"Because ice is a simpler seismic medium than rock, its velocity structure is well known and we can really explore the source physics," said Hudson. "If we can do it in this simpler environment, then we hope that maybe we can start thinking about doing it in more complex environments."

Researchers deployed a dense, two-dimensional grid of fiber optics in the crevasse of Gornergletscher, Switzerland's second-largest glacier. Hudson said the weather conditions during the team's deployment were very lucky. They deployed the cable during the transition from winter to summer, so there was no snow and the researchers could avoid the danger of falling into covered crevices.

One of the major challenges in using fiber optic cables to collect seismic data is ensuring that the cables are in good contact, or "coupled," with the ground to which they are laid. "It's still high enough during the day that the fiber heats up and melts slightly into the glacier, because the fiber is black compared to the ice. Then, when the fiber melts, the temperature is low enough that it freezes in place overnight," Hudson explained.

"So we actually get the best coupling one could hope for in terms of fiber melting and freezing," he added.

The team detected and located 951 icequakes, whose seismic waveforms contained strong oscillations or wakes after the arrival of seismic surface waves. These oscillations can occur when there is water within a crevasse, and the movement of the water during an earthquake can produce a resonant signal. But analysis by Hudson and his colleagues suggests the oscillations are more likely caused by "resonances created when seismic waves bounce back and forth between multiple crevasses in a crevasse zone," Hudson said.

The researchers also compared data from the fiber optic grid to data from more traditional seismic node deployments. Fiber optic cables can deliver almost 20 times the amount of data as an array of nodes. "While there were some data processing challenges, the amount of data was much larger, which allowed us to essentially see the complete wavefield in the data itself, which is unusual," said Hudson.

He noted that another advantage of using fiber-optic cables is that they are sensitive to a wider range of signal frequencies, including low-frequency signals that last for hours or even days, allowing seismologists to measure the bending of ice over time.

Hudson hopes to use fiber optics to measure the velocity structure of the ice and develop 3D images of its subsurface, he said.

"I'd really like to quantify the extent and density of cracks and see how damaged the ice is in this area," he explained, "so we can know that icequakes are triggered by cracks. We haven't quantified the number and size of cracks yet, so that's hope for the future."

Compiled from /scitechdaily