Barry Arm Landslide
The Barry Arm landslide in Alaska’s Prince William Sound Christian E. Zimmerman / Public Domain
To live during an age of rapid climate change is to be witness to environmental collapse.
Sometimes these environmental disintegrations can be subtle, like the slow decline of an endangered keystone species or the gradual weakening of ocean currents. Other times, the collapse of the natural world sends signs that can’t be ignored—stronger blazes, deadlier storms, record-breaking heat waves, devastating droughts.
And on September 16, 2023, scientists witnessed a climate-change-related collapse unlike any other.
Researchers around the world spotted an unusual oscillating sine wave sketched across seismographs. Later described as a monotonous hum, this signal originated from somewhere in eastern Greenland. The rumble traveled unnoticed below the feet of the planet’s eight billion human inhabitants, and it reached Antarctica within an hour. The seismic activity of moderate earthquakes usually takes only minutes to dissipate, but for some weird reason, this signal continued emanating for another nine days.
“The seismic signal was something no one ever observed before,” says Kristian Svennevig, a researcher at the Geological Survey of Denmark and Greenland who studied this strange signal. “This is a new natural phenomenon thrown at us by climate change.”
Scientists didn’t initially know what created this signal. It was so strange, seismologists simply labeled it an “unidentified seismic object.” But after a year of scientific detective work, Svennevig, along with an international multidisciplinary team of scientists, pieced together the unprecedented geological story that unfolded in the remote watery inlet called the Dickson fjord.
The glacier in this particular fjord thinned so much that it could no longer support the rock above it. Known as “glacial debuttressing,” this melting process suddenly and violently unleashed 25 million cubic meters (883 million cubic feet) of rock and ice. Debris roughly the same volume as 10,000 Olympic-sized swimming pools plunged into the water below, creating a 650-foot-tall tsunami. Thankfully, the fjord’s finger-like geography largely contained the wave’s energy and formed a seiche—a standing wave that moves in the same way water sloshes back and forth in a bathtub. Undulating against the fjord’s walls every 90 seconds, the seiche essentially became a giant resonating chamber vibrating at a frequency of around 11 millihertz. At first this standing wave sloshed roughly 23 feet up the fjord’s walls, but over the course of nine days it dissipated to only an inch or so before disappearing entirely. The rhythmic undulation of this dissipating seiche—slamming massive amounts of water back and forth against the fjord’s rocky walls—is what ultimately caused the nine-day-long monotonous hum observed around the world.
“Had you suggested last summer if a wave could last for nine days in a fjord like that, no one would have believed it,” Svennevig says. “But it happened—we have to re-equilibrate our perception of what could happen because of climate change.”
While the Dickson fjord’s strange seiche signal may be a seismic outlier, collapses like these are becoming more common. In 2015, the Taan fjord landslide on the other side of the Arctic in Alaska collapsed creating a 600-foot-tall tsunami, and just two years later, the Karrat fjord landslide in west Greenland generated a 325-foot-tall tsunami that ultimately claimed the lives of four people in Nuugaatsiaq, a small fishing village. Svennevig says that anthropogenic climate change is speeding up the global retreat of glaciers worldwide and increasing the frequency and magnitude of natural hazards on these slopes.
While scientists were caught unaware by the Dickson fjord landslide and other similar collapses, a fjord in Alaska called Barry Arm features a slow-moving landslide that gives experts an unusual opportunity to study a natural disaster in the making. Although it’s difficult to pinpoint when this dangerous dance with gravity first began, a large section of the fjord has been sliding for at least a century. Geologists, seismologists and climatologists have flocked to the scene in an effort to understand this phenomena, making Barry Arm, in essence, a key Arctic laboratory.
Welcome to Barry Arm
The Barry Arm fjord is located in the Chugach National Forest southeast of Anchorage. The fjord in the northwestern corner of Prince William Sound features glassy, calm water with imposing rock walls looming above it. The dramatic valley is accented by snow-capped peaks casting their shadows over a finger-like body of water half-encased in ice. This is the retreating Barry Arm glacier, and this disaster-in-progress was first spotted in 2019 by an artist kayaking in the fjord who noticed odd fractures on a cliff looming above the water. Although local geologist Bretwood Higman first dismissed Barry Arm as a landslide at first after analyzing the artist’s photos, he later realized his mistake. “It turns out I had made a Geology 101 error. … I hadn't zoomed out enough, and I missed the big picture,” he later told NASA. This potential landslide was far bigger than anything that had ever been reported.
Scientists from universities around the United States signed an open letter calling for immediate investigation of the site, and soon a wide network of teams across local, state and federal agencies turned their gaze on the fjord to assess its instability. The results? The land was collapsing in slow-motion. The analysis estimated that the landslide had shifted at least 600 feet from 2009 to 2015, and that in the event of a total collapse, communities and natural areas near Barry Arm could be destroyed.
In a worst-case scenario, an estimated 17.5 billion cubic feet of rock and ice would plunge into the water below—some 20 times more material than what fell in the Dickson fjord. Because of Barry Arm’s geology, this landslide wouldn’t produce a seiche but would create a 650-foot-tall wave. And in its likely path of destruction lies Whittier—a quirky town where many of its 260 or so inhabitants.
“It’s such a beautiful place in northwestern Prince William Sound … then you just have this nasty-looking slope hanging out there,” says the U.S. Geological Survey’s (USGS) Dennis Staley, who’s worked on Barry Arm since 2021. “Once you know it’s there, it’s hard to not see it.”
Barry Arm Landslide Graphic
Annotations show landslide areas of the Barry Arm fjord. Gabe Wolken (June 26, 2020) / Public Domain
Since the landslide’s discovery, various state and federal organizations—including the USGS, the National Tsunami Warning Center, the Alaska Division of Homeland Security and Emergency Management, the Alaska Division of Geological and Geophysical Surveys, the Alaska Earthquake Center, and the National Oceanic and Atmospheric Administration—began collecting data on the landslide using satellite-based radar, airborne lidar, aerial photography, local seismometers and remote cameras.
Even though Barry Arm is a serious threat, Staley says it’s also an incredible scientific opportunity. Because of the slope’s slow-moving collapse, scientists can study the landslide’s historical rates of deformation and see how those changes relate to the retreat of the Barry Arm glacier. However, an ideal opportunity doesn’t mean the Barry Arm fjord is a particularly forgiving laboratory.
“There’s very few places to land a helicopter, you can’t hike up to it, there’s a tremendous amount of rockfall off the top, there are these massive tension cracks going across it,” Staley says. “[We] installed a seismometer on the landslide itself … and it didn’t even last eight months … it was taken out by an avalanche in April.”
The immediate goal of the interagency effort is to develop a way to detect and locate tsunami-generating, or tsunamigenic, landslides as they happen and warn imperiled Alaskans of any impending waves—all in less than five minutes.
That’s easier said than done. Alaska is one of the most seismically active places on the planet. Here the Pacific plate slips, or subducts, under the North American plate, and earthquakes—lots of earthquakes—are the natural result. Some 75 percent of all earthquakes in the U.S. greater than magnitude five occur in Alaska, and an earthquake occurs in the state every ten minutes on average. The state also experiences two types of landslides: those with glacial thinning like the one in Barry Arm and also shallow landslides caused largely by permafrost degradation. Understanding the dynamics at play at Barry Arm, as well as sifting through all the seismic noise to pinpoint these tsunami-generating landslides as quickly as possible, will one day save lives as more and more people move to these largely uninhabited spaces.
The dangerous conditions at Barry Arm and various other landscapes around the world have been millennia in the making. For the past 15,000 years, glaciers have moved across landmasses and formed mountain ranges, steep valleys, deep lakes and towering fjords. As the ice retreats to the poles, many rock walls become free. Some haven’t been exposed to air for thousands of years, and others up to a million years. Because of our limited perception of these landslides—confined in data to only the past two centuries—Svennevig says we need to be prepared for events previously considered unimaginable.
Research seismologist Ezgi Karasözen’s job is to detect these future landslides and protect communities vulnerable to subsequent tsunamis. She first studied earthquakes in Turkey and Iran, another seismic hotspot, and arrived at the Alaska Earthquake Center to do the same, until the discovery of Barry Arm shifted her focus.
“The Alaska Earthquake Center installed seismometers on Barry Arm to see what signals we were dealing with, and that’s when I started working on this project,” Karasözen says. “I like that I can actually do research that could help a community in real time.”
In midsummer 2021, Karasözen, along with her colleague Michael West, traveled to Barry Arm, set up cameras and took notes of what they were hearing. Large glacial groans sounding like ominous, deep-throated booms echoed through the fjord as the glaciers melted in the midsummer heat. These rumblings were joined by other seismic activity common in this part of the world, making the scientists realize that sifting through these geologic noises wasn’t going to be easy.
Because of the need to provide rapid detection and warning to surrounding communities, Karasözen and her team decided to keep the approach to detecting landslides as simple as possible. She decided to focus on long-period seismographs, which detail activity at lower frequencies, to help separate landslide events from other seismic background noise. Because this signal also looks similar across seismic stations, Karasözen uses this technique to essentially pinpoint where a landslide occurs.
Karasözen and her team began running tests across a larger area centered on Barry Arm in the summer of 2023, and they detected seven landslides in a few months. A year later, she says, the algorithms pinpointed around 30 of them. The landslides deposited anywhere from roughly 3.5 million to around 141 million cubic feet of rock and ice. None of them created a tsunami, but history suggests that that streak won’t last long.
“It only takes one event to be really, really bad,” Karasözen says. “You can’t just say, ‘Let’s just let it happen and hope for the best.’”
Although the algorithm successfully detected dozens of landslides, it also proved adept at detecting some earthquakes with long-period signals similar to landslides. Karasözen and her team are now working with their growing data set of confirmed landslide signals to refine the algorithm so it isolates landslides from the typical rumblings of Alaska’s dynamic geology.
An emerging science
Karasözen’s hope is that this continually improving algorithm will be valuable for not just the Barry Arm region but the entire state, and possibly beyond—in Greenland and Norway. In fact, Staley’s own work with the USGS has sent him to Norway, a country that’s very familiar with fjords, to share data and techniques on how to study landslides and protect an area that’s likely to see a boom in human activity as warming occurs.
As scientists develop algorithms, compare notes and take increasingly detailed data using airborne lidar and space-based satellites, Svennevig says that this relatively new field still has plenty of room for simple, old-school techniques.
He explains that there’s still one tried-and-true field method that can uncover evidence of prehistoric landslides, known as paleo-landslides, that could possibly help prepare us for what comes next. “Where I see a huge task for me as a geologist.” Svennevig says, “is just walking around and looking at rocks.”