Pular para o conteúdo
Hub Nexus

Wrangell - St Elias National Park & PreservePágina 40 de 98

Flat Creek: When a Large Section of a Glacier Collapsed in Wrangell-St. Elias

On July 15, 2015, a Bucknell University geologist flying over the eastern Wrangell Mountains noticed a messy debris deposit on the south bank of the White River, at the mouth of a small tributary valley known locally as Flat Creek. It looked large, recent and out of character with the braided glacial stream deposits that dominate the area. Photos of it revealed another landscape-altering event in Wrangell-St. Elias National Park and Preserve that had gone unnoticed for several years by park visitors and staff. Mylène Jacquemart of the University of Colorado Boulder and Michael Loso of the National Park Service describe how the debris turned out to come from the collapse of a large section of a glacier.

Piecing the story together

Anyone close enough to witness the event could easily have been killed; luckily, no one was there to see it happen. It had not gone entirely unnoticed, though. A long-time seasonal resident on the opposite side of the White River remembered hearing and feeling a large shock from the direction of Flat Creek in mid-summer 2013, and later found an impressive deposit, rich in ice and mud, burying white spruce forest almost to the river. The resident also recalled an even larger flow in 2015.

Anecdotes and a few photographs showed that some debris-flow-like event had happened at least twice on Flat Creek. A U.S. Geological Survey seismologist confirmed the timing: seismometers near Barnard Glacier, 40 miles (65 km) south-southeast, had recorded the events because they shook the ground hard. They show two main slope failures, the first on August 5, 2013 and the second almost exactly two years later, in 2015. The second was a very large detachment followed by two smaller events, all within half an hour. In 2016, while the researchers were piecing the story together, another event was seen. Two NPS rangers flying a routine hunting patrol on August 10, 2016 happened to look down just in time to see an icy debris avalanche moving down the river channel at almost 25 mph (40 km/h), and they filmed it. It was much smaller than the earlier events, which the seismic data support, and extremely wet, carrying more ice than mud.

The writer John McPhee, in a 1988 essay, described a debris flow as resembling fresh concrete, water mixed with solid material mostly larger than sand, some of it car-sized. The 2013 and 2015 events fit that fairly well from their deposits, except that, according to the resident, they also held ice. The 2016 flow, mostly ice, hardly fit at all. The authors say they might call it a slushalanche, a term with no precedent in the scientific literature, and the ice and water forced two central questions: what causes these flows, and where do the ice and water come from?

What the satellites showed

Frequent cloud makes satellite images hard to compare here, but Planet Labs' small satellites pass over very often, and their high-resolution data (pixels of 16.4 feet, 5 m) tracked the changes. Vast areas of white spruce forest were overrun. In the lower valley the 2013 flow left a narrow deposit 2.3 miles (3.7 km) long and 260-820 feet (80-250 m) wide, in places more than 65 feet (20 m) thick, and it deposited debris atop a riverside knob the authors informally named West Hill, which stands more than 300 feet (100 m) tall. The 2015 flow was even bigger: it overtopped the whole hill, ran over the 2013 deposit, destroyed more spruce forest and covered almost 3,700 acres, about 2,800 American football fields.

The most notable finding was that large parts of the glacier at the head of Flat Creek disappeared in the two events. A satellite image from August 11, 2013 shows the front third of the glacier tongue missing and large pieces of ice beside the river below. In 2015 the ice in the glacier's central trough disappeared altogether. Digital elevation models from 2012, 2014 and 2016 bracket the events, and comparing them showed that about 244-395 million cubic feet (6.9-11.2 million m3, roughly 4,000 Olympic swimming pools) washed downstream from the upper basin in 2013, and more in 2015. The losses were concentrated almost entirely in glacier-covered parts of the watershed and reached up to the highest part of the accumulation zone. Some of the volume must have been bedrock or sediment, since the flows carried so much of it, but the authors conclude that the events, including 2016, whose ice they could not trace to a source, involved the nearly instantaneous collapse of a large, full-thickness section of Flat Creek Glacier.

What the field team found

In summer 2018 the team made the first on-site investigation. Its first strong impression was of how big the events were: a runout of approximately 7 miles (11.26 km) from a glacier crest only about 4,900 feet (1,500 m) above the valley floor. Because the glacier's extent matches the outline of the failure, the authors argue this was not a glacier caught in a rock avalanche but a glacier that detached from its bed, taking rock and debris with it. They know of only three places in the world with similar catastrophes, with long, low-angle runouts after a sudden glacier collapse: Kolka Glacier in the Russian Caucasus in 2002, two glaciers in Tibet in 2016 and several glacier detachments at Iliamna, a volcano in Cook Inlet, Alaska.

Hardly anything larger than a toy truck survived in the deposit; the car-sized boulders of McPhee's description are almost missing. Part of the reason, the authors think, is weak rock in the headwall, dark grey and reddish-brown mudstone mapped as the early Permian Hasen Creek Formation. The headwall lines up with the late Quaternary Totschunda Fault, though they found no seismic trigger, and the bedrock is full of faults, igneous intrusions and possibly thermal alteration that further weaken it. The deposit holds molards, conical debris mounds that newer evidence suggests often form when boulder-sized pieces of weak bedrock cemented by ice are carried downslope, thaw and fall apart. That points to permafrost, and the field team found frozen ground in mid-July at a depth of 31.5 inches (80 cm) and a ground temperature of only 34.2 °F (1.2 °C) at 27.5 inches (70 cm). One measurement is not conclusive, but a permafrost map suggests permafrost in more than 90% of the Flat Creek basin, and studies have tied warming permafrost to a potential increase in landslide activity. Stream-cut outcrops of ice-rich debris held mostly individual clasts of massive ice, which the team interpreted as glacier ice carried down by the flows.

What remains unexplained

The part of the glacier that failed appears to be among its least steep. The loss was too sudden to be a surge, too rapid to compare with typical climate-driven melt, and involved too much low-angle ice for an ice avalanche. The exposed mountainside, presumably the failure plane, slopes about 20 degrees. A drone flown toward the release zone, because the team did not go onto the headwall for safety, showed heavily folded and fractured black mudstone with crumbling rock and ice everywhere. They can say with confidence that each event was a variable mix of glacier ice and rock that formed a wet slurry flowing at high velocity like a debris flow. They cannot yet say whether the ice collapse set off the bedrock or the other way round, or where the water came from, since Flat Creek is very small and ice avalanches rarely, if ever, produce enough water to sustain a wet flow. Preliminary climate data show that 2013 and 2015 were very warm, with above-freezing temperatures likely producing a lot of meltwater even at high elevations. At the end of the 2018 season they left time-lapse cameras in the valley, and they are analyzing weather and satellite data, rock, sediment and tree-core samples, and running models of the flow.

The precedents are sobering: the authors say over 120 people were killed and a village obliterated in the 2002 Caucasus collapse, and nine people and hundreds of herd animals in the 2016 Tibet event. Both were later called extraordinary, which partly reflects how incomplete the understanding of glacier and debris-flow dynamics is. The authors ask whether accelerating glacier mass loss worldwide comes with new and potentially more dangerous forms of retreat, a question they call urgent in a park with over 3,000 glaciers covering over 11,000 square miles (more than 29,000 km2).

Sources