As frozen debris field slides toward Dalton Highway, state makes plans to move road

FAIRBANKS — The state plans to move a 2,000-foot section of the Dalton Highway in 2017, hoping to delay a collision of geologic proportions.

A moving mass of frozen debris — about 80 feet deep, 600 feet wide and a mile long — continues to inch downhill toward the highway, both winter and summer, sweeping away everything in its path. Researchers theorize that the giant blob is sliding year-round atop a layer of water that remains a liquid despite the permafrost because it is under high pressure.

It may take five to 10 years for the "frozen debris lobe" to reach the road at its current pace of several inches per week, but a similar formation three miles away has been clocked at 125 feet a year, almost fast enough for someone to hear the rocks move.

If the debris pile near Mile 219 of the Dalton Highway accelerates to that speed for some reason, the road and the trans-Alaska pipeline could quickly be placed between a rock and a hard place.

The road realignment plan set for 2017 would be in conjunction with a much larger road reconstruction effort from Mile 209 to 235, a spokeswoman for the Department of Transportation and Public Facilities said.

Moving the road would provide a temporary solution while the more difficult challenge of responding to a force of nature draws more study.

The road is about 140 feet from the leading edge of the debris — a frozen accumulation of sand, gravel, rock, silt, old pieces of wood, ice and other materials that continues to puzzle scientists.


There may be hundreds of these formations in the Brooks Range, but those found in the Dalton Highway corridor — 43 so far — are drawing the most interest because of what they could do to the road. There are 23 frozen debris lobes within a mile of the uphill side of the highway, all potential road hazards.

If the state simply waits for the moving debris pile to reach the highway, it would require 440 loads a year with a 50-ton dump truck to keep the lanes open, according to one estimate. But that does not take into account the possibility that removing the debris could be like taking a pile of rocks out of a stream, causing it to flow faster.

Engineer and geologist Margaret Darrow, an associate professor at the University of Alaska Fairbanks, said that one way to think of the debris lobe is as a "slow-moving landslide in permafrost."

Her interest in this topic originated after initial investigations by Ronald Daanen, a permafrost researcher and hydrologist, who took a close look at the abundance of misshapen trees found on the debris piles.

"It pushes the trees down flat, it rolls up the organic mat like a carpet," Darrow said in a presentation at UAF last winter. "And when you climb up on top of the surface, you'll find that trees just are not straight. They get knocked out of vertical, they correct themselves, they get knocked again, they correct themselves."

The tree-ring record in one old tree cut as part of a drilling research effort showed that it had been knocked out of kilter and righted about every 30 to 50 years for the last 250 years.

Darrow and Daanen believe the debris lobe closing in on the highway is moving faster than it did decades ago, based in part by measurements in recent years and comparisons with aerial photos over the last half-century. Climate change may be a factor.

"When we explained this problem to DOT last year, we told them that this thing is faster than you think and it's bigger than you think," Daanen said in a phone interview Monday.

"This one is close to the road, but there are more of these features along the haul road, 23 or so, that are upslope of the highway," he said.

He said people often suggest that bombing the debris lobe would solve everything, but he said that idea has some big drawbacks — first because it would be an enormous blast and second because it would expose a big piece permafrost to additional melting and create other environmental problems.

He said that debris lobes are a geological phenomenon of long-term consequences, as mountains fall apart and the landscape changes.

"Rocks and soil pile up at the bottom of the slope and when there is too much it has to go somewhere," he said. "In this case it kind of follows a slight valley, like a streambed down a hill."

"It keeps accumulating at the top because the mountain doesn't stop eroding," he said.

Daanen and Darrow have a theory, based on drilling conducted at the site, that there is a layer of water inside the debris pile, perhaps 65 to 75 feet down at some points, that is a key factor in its movement. They suspect the water remains in a liquid state because of a combination of pressure and temperature.

"It's water that should be ice, but it isn't," he said. The permafrost in the debris lobe is about 30 degrees, two degrees warmer than the surrounding area, but still below freezing. Some of the water in the lobe doesn't turn to ice, however.

Darrow said that engineers who have discussed mitigation plans have considered a variety of challenges, starting with the water.

"With any good landslide you get rid of the water, that's the first step," she said. "Well, where do you put the drains? How effective will the drains be before they shear off because the thing is moving."


Some engineers have suggested separating the top half of the lobe from the lower half by excavating a large trench across it and draining the water.

"It would probably lead to a quagmire of thawing permafrost," she said. Plus, the surface of the lobe is so cracked and uneven that getting equipment into place would be a problem.

Others have said that thermosyphons, designed to lower the ground temperature, could be used to lower the permafrost temperature, which could be expensive and difficult. Yet another other idea is to build a rock wall as a buttress to stop the advance of the lobe.

"Will that be enough resistance to stop the lobe? We would need to model that to find out," she said.

Darrow said that moving the highway is the preferred option at the moment, mainly because not enough is known about what else would work and how much it would cost. Even the road shift is not a cure-all, however.

"This will work for now, given the movement, the rate that we have," said said. But if the lobe speeds up, it could reach the road before it is moved. "And what about the other frozen debris lobes? What if they step up movement?"

The answer remains a mystery, along with the exact mechanism driving the advance of the frozen debris lobes.

Dermot Cole

Former ADN columnist Dermot Cole is a longtime reporter, editor and author.