When the ground collapses like a soufflé: Studying the effect of thermokarst on the Arctic

December 28, 2009

By Emily Stone 

This is the thermokarst failure on a stream leading into the Toolik River on the day Breck Bowden and Michael Gooseff discovered it in 2003. (Courtesy of Michael Gooseff)

Scientists Breck Bowden and Michael Gooseff were flying in a helicopter near Toolik Field Station in 2003, scouting for good field sites for river research when they spotted something peculiar. 

Unlike the crystal clear Kuparuk River nearby, the Toolik River was a muddy brown, an unusual site in the Arctic where tundra streams don’t pick up much sediment because the ground is usually frozen. The men had the helicopter fly upstream to investigate. After 40 kilometers they saw the culprit: a small stream leading into the river had a huge, narrow crater on its shore that was dumping sediment into the river. 

“It was a severe gash on what was otherwise a nondescript hillside,” said Gooseff, an assistant professor of civil and environmental engineering at Pennsylvania State University. 

The feature is what’s known as a thermokarst failure. Thermokarst occurs when ice in the usually solid permafrost melts and the land gives way like a soufflé. When this happens on flat ground, the melted water pools into a thermokarst lake. When it happens on a slope, as was the case along the Toolik, the water rushes downhill and usually into a nearby body of water and the ground slumps after it, causing what’s called a thermokarst failure. 

The helicopter landed by the gash. Bowden, a professor of watershed science and planning at the University of Vermont who is over six feet tall, was engulfed in it when he stood at the bottom. He guesses that it had formed within a few days of their arrival. 

Standing there, he remembers looking around at all the other tiny streams that led into the region’s big rivers and thinking, “It wouldn’t take many of these on the landscape to have a fairly big impact.” 

Bowden is now leading a project that includes Gooseff and 15 other principal investigators to discover what exactly these thermokarst features are doing to the landscape and river networks, and how they form in the first place. They’ve already established that there are many more of them than there were 25 years ago, the likely result of rising temperatures in the Arctic. As more thermokarst failures develop, researchers want to know how the additional nutrients dumped into rivers will affect aquatic ecosystems, how they’ll impact the plant communities that grow back after a thermokarst landslide, and how they’ll change the amount of carbon dioxide and methane being released into the atmosphere — all crucial questions in the study of climate change. 

Breck Bowden explores the thermokarst failure on a stream leading into the Toolik River on the day he and Michael Gooseff discovered it in 2003. (Courtesy of Michael Gooseff)

Downstream of the same thermokarst feature as it looked this summer when the group of researchers began their multi-disciplinary study of the phenomenon. Photo: Emily Stone

The group, which includes another couple dozen graduate students and technicians, is in the first year of a four-year, $5-million grant from the NSF’s Arctic System Science Program. They spent their first field season at Toolik this past summer picking their research sites and setting up equipment to monitor the changes happening in and near the thermokarst failures. 

Previous research by Bowden, Gooseff and some of the other collaborators established that, at least in the area around the field station, there are many more thermokarst failures than there were in the early 1980s. The group did an aerial survey of 600 square kilometers around the station and compared their observations to an aerial survey that was done in the same area around 1980. They found 34 thermokarst, two-thirds of which were new. This data doesn’t necessarily correspond to the rest of the Arctic since different soil conditions, slope and climate affect thermokarst formation, but it does suggest that the features are growing in at least one large swath of Alaska. 

The concern is that these features can have an outsized impact on the environment. 

Bowden is interested in the nutrients that are usually held frozen in permafrost — what he describes as “brown concrete” — that are released into rivers when that permafrost melts. The addition of ammonium, nitrate and phosphate means that aquatic microbes and plants have much more to eat and can flourish in areas where their populations were previously limited by a lack of food. This can dramatically alter a river’s ecosystem. 

He has set up water monitoring stations on rivers above and below thermokarst failures to compare the sediment and nutrients in the water before and after the thermokarst soil reaches it. Earlier research showed that the Toolik River thermokarst failure delivered more sediment to the river than was dumped into the Kuparuk River over the course of 18 years from a 132-square-kilometer section of watershed. 

Other researchers in the group are looking at how plants react to thermokarst failures. “We have a suspicion that what they evolve into is a shrubby community,” Bowden said, instead of the low tundra grasses that dominate the region. 

This is important because shrubs hold on to more sunlight than grasses, which warms the soil below. This can in turn release more stored carbon out of the warming soil. Additionally, warmer soil releases more nutrients for microbes to use as fuel. Microbes then release more methane into the atmosphere, which is a powerful greenhouse gas. Scientists in the thermokarst group are measuring this CO2 and methane release. 

Other researchers are looking at remote sensing and computer modeling, as well as interviewing Native Alaskan communities nearby to learn about their memories of where thermokarst have occurred in the past. 

Gooseff is taking a step back to try to figure out what causes the thermokarst failures in the first place. He has placed water and temperature sensors at various depths in and near several thermokarst features, as well as instruments above ground that measure rain, snow, sun and wind. His post doc, Dr. Sarah Godsey, set up cameras to take pictures of the thermokarsts every hour. Their goal is to be able to correlate the weather and soil data with physical changes in the permafrost and landscape. 

Bowden notes that thermokarst are not a new occurrence. Scientists have been aware of them for years, and engineers have long studied them in the context of building roads, homes and pipelines. 

“It is a natural phenomenon, but it appears to be one that is accelerating,” he said.

Emily Stone is a freelance writer from Chicago, Illinois. She spent a week at Toolik Field Station in 2009 as an MBL journalism fellow.


Happy Times!

December 25, 2009

On Bailey! Cash Littrell, son of Levi and PFS' Karla College, gives Santa a hand. Photo: Karla College

May the peace and joy of the holiday season be with you throughout the coming year.

From all of us at Polar Field Services–two- and four-legged alike!


The (Glacial) View From Within

December 22, 2009

Before Dr. Alberto Behar of the NASA Jet Propulsion Laboratory tossed an enclosed camera into the gaping hole that bored into Jakobshavn Glacier last summer, he took a moment to listen to the roar of the river plunging into the ice.

“It sounds like a jet engine,” said Behar last week from the American Geophysical Union annual meeting in San Francisco. “If you fall in there, forget about it. You’re not going home.”

Scientists lower a probe into a moulin in the Pakisoq region in western Greenland. All photos courtesy NASA

While a human would be hard pressed to survive a “dip” in a moulin—a narrow, tubular shaft in a glacier that guides water from the surface to the glacial base—Behar and co-investigator Dr. Konrad Steffen of the University of Colorado, Boulder, are hoping to design a high-tech, video and camera-equipped probe that will. Specifically, they recently completed the fourth field season of a NASA-funded study on moulins.

Their team is developing a tool to measure the depth of moulins and, ideally, track the path of water from the surface to the ocean. The best-case scenario would be if one of the probes dropped into a moulin emerged later in the ocean and “could tell us where it had been,” said Behar in the press conference.

Behar and Steffan call their tools “expendable rovers.” The size of a pocket book, these solar-powered systems are modeled after the Antarctic Ice Borehole Probe, which studied ice streams in West Antarctica, the Amery ice shelf in East Antarctica and the Rutford ice stream in West Antarctica.

The Greenland version is modified specifically to explore the moulin environment. It consists of two high-resolution charge-coupled device cameras (a side-viewing digital camera and a downward-viewing video camera), lights, associated electronics and an inclinometer that measures the tilt of the moulin chute.

A watertight probe can withstand the immense water pressure in a moulin.

Images are sent in real time through a tether one kilometer long (about 3,300 feet) to a receiving station at the surface. The station has a video display, computer and digital tape recording devices.

Earlier this year, working in bitter cold, slushy, windy conditions (minus 10 degrees Celsius, or 14 degrees Fahrenheit), the scientists deployed the probe in two locations of a moulin. Once the probe descended to 110 meters (361 feet), it encountered horizontally flowing water and debris about one to two meters (3.3 to 6.6 feet) deep.

In this particular moulin, the water flows out in well-developed channels to the edge of the ice sheet. At the time of the experiment, the scientists measured the water flow rate of the surface melt rivers feeding the moulin at approximately 15 cubic meters a second (about 238,000 gallons a minute).

“This was very interesting and is evidence that we need an integrative plan on how to study these in a more sustainable way,” said Behar. “We need to get multi-year funding.”

Alberto Behar on the edge of a moulin in 2006.

Behar and Stefan have been developing a moulin probe since 2006, and Behar offered the following synopsis for each year of field work in the Pakisoq region.

  • 2006 The team used an ice borehole camera that shot an image about 100 meters down a Moulin. However, the camera was heavy and proved to be difficult to work with.
  • 2007 The team returned with a Sony HD video recording camera in a watertight Lexan enclosure. They sent the camera into the Moulin, but the images were hard to interpret. “It was a lot like fishing,” said Behar. “We found the crevasses are much more complex than we had thought.”
  • 2008 The team developed a simple device with a tracker GPS modem that had temperature sensors and could measure the pressure. They expected it to follow the water pathway, emerge and call home. They never heard from it again.
  • 2009 The team developed a live video feed camera system with a fiber optic cable. The camera transmitted images to special glasses (Behar calls them “Blade Runner-esque”), and the viewer could watch the camera’s progress.

“This was an exciting, important first look into a place that’s not well understood but could have an important role in understanding the dynamics of this region,” said Behar. “We’re excited by the possibilities this technology holds, not only for future studies of Earth’s icy regions, but also for future missions to explore extreme ice and liquid environments on other planets, such as the Martian polar ice caps and Jupiter’s moon Europa.”

Next year, University of Colorado scientists will use ground-penetrating radar to accurately measure the glacial ice thickness at this location. These data will help scientists better interpret their findings and plan future tests.

Scientists expect moulins to shed light on complex glacial dynamics, which are not well understood and are responding rapidly to climate change. Previous NASA measurements in the Pakisoq region using global positioning system data show the ice there moves an average of about 20 centimeters (8 inches) a day, accelerating to about 35 centimeters (14 inches) a day during the summer melt. Scientists suspect the moulins may affect—directly or indirectly—that rate of advance.

In Greenland, the surface of the ice sheet moves at varying speeds, on both seasonal and shorter-term time scales. Seasonally, glacial water penetrates to the glacier bed through significant thicknesses of cold ice. However, early in the melt season and at other times, there can be periods when water flows rapidly into glacial drainage systems, resulting in sudden new flows of water out of the glaciers. In the middle of the melt season, surface melting resumes after periods of cold weather, which can partially close sub-glacial channels.


The Shortest Day

December 21, 2009

Winter solstice in Denali National Park, Alaska. Photo courtesy Denali Education Center: http://www.denali.org/

So the shortest day came, and the year died,
And everywhere down the centuries of the snow-white world
Came people singing, dancing,
To drive the dark away.
They lighted candles in the winter trees;
They hung their homes with evergreen;
They burned beseeching fires all night long
To keep the year alive,
And when the new year’s sunshine blazed awake
They shouted, reveling.
Through all the frosty ages you can hear them
Echoing behind us – Listen!!
All the long echoes sing the same delight,
This shortest day,
As promise wakens in the sleeping land:
They carol, fest, give thanks,
And dearly love their friends,
And hope for peace.
And so do we, here, now,
This year and every year.
Welcome Yule!

Susan Cooper, The Shortest Year


What Lies Beneath

December 18, 2009

For years, scientists thought that melted water beneath Greenland’s coastal glaciers such as the Jakobshavn and Helheim lubricated the giant sheets of ice above, accelerating their plunge into the ocean and contributing to loss of sea ice. Turns out, that was an over-simplified explanation, said Ian Howat, assistant professor of earth sciences at Ohio State University.

Speaking in a press conference Wednesday at the annual meeting of the American Geophysical Union (AGU), the NASA-funded, CPS-supported scientist explained that the subsurface dynamics beneath glaciers is significantly more complex than previously thought.

“In the science community it’s been accepted that basal lubrication due to increased melting and warming is responsible for accelerating glacial advance and breaking off,” said Howat. “We’re finding out that’s not true.”

A calving glacier drops huge ice chunks into the sea. Photo: Martyn Clark, National Snow and Ice Data Center

Specifically, a complex, subglacial “plumbing” system involving the ocean, meltwater, and ice evolves, which drives the glacial calving. In fact, early evidence from Howat’s research suggests that ocean changes have a greater impact on the rate at which outlet glaciers spill into the sea than does meltwater.

Much of the melt water comes from early summer hot temperatures, which melt the glacier’s surface. The water flows through cracks in the ice to the ground surface.

Ian Howet in the field. Photo: Ohio State University

In the early summer, the sudden influx of water overwhelms the subglacial drainage system, causing the water pressure to increase and the ice to lift off its bed and flow faster—up to 100 meters per year, he said. The water passageways quickly expand, however, and reduce the water pressure so that by mid-summer the glaciers flow slowly again.

Inland, this summertime boost in speed is very noticeable, since the glaciers are moving so slowly in general. But outlet glaciers along the coast, such as the Jakobshavn, are already flowing out to sea at rates as high as 10 kilometers per year — a rate too high to be caused by the meltwater.

“So you have this inland ice moving slowly, and you have these outlet glaciers moving 100 times faster. Those outlet glaciers are feeling a small acceleration from the meltwater, but overall the contribution is negligible,” Howat said.

His team looked for correlations between times of peak meltwater in the summer and times of sudden acceleration in outlet glaciers, and found none. So if meltwater is not responsible for rapidly moving outlet glaciers, what is? Howat suspects that the ocean is the cause.

Through computer modeling, he and his colleagues have determined that friction between the glacial walls and the fjords that surround them is probably what holds outlet glaciers in place, and sudden increases in ocean water temperature cause the outlet glaciers to speed up.

However, Howat said meltwater can have a dramatic effect on ice loss along the coast. It can expand within cracks to form stress fractures, or it can bubble out from under the base of the ice sheet and stir up the warmer ocean water. Both circumstances can cause large pieces of the glacier to break off, and the subsequent turbulence stirs up the warm ocean water, and can cause more ice to melt.


Christmas North of the Arctic Circle

December 17, 2009

Photo: Karl Newyear

Up at Summit, darkness continues to descend as the winter solstice on 21 December marking the shortest day in the northern hemisphere still approaches.  For the crew of five taking care of NSF’s research station on the Greenland ice sheet, the solstice and Christmas holiday a few days later are major bright spots on the calendar.  Here, PFS station manager Karl Newyear offers a glimpse into the strange and familiar world of Summit at Christmas time.

As we enter the middle of December, Christmas preparations are underway at home (not counting the decorations that began showing up in stores before Halloween!), but how are the holidays celebrated in Greenland?  Christianity didn’t arrive on the island until around 1721 when Hans Egede from the joint kingdom of Denmark-Norway arrived and began to convert the native Inuit.  However, here more than many places on Earth the cultural history is closely tied to the annual solar cycle and of course the winter solstice falls just a few days before Christmas.  The gateway city for Summit Station, Kangerlussuaq, is just north of the Arctic Circle and the solstice is one of only a few days in which the Sun doesn’t rise above the horizon.  At Summit Station, however, we are in the middle of 74 consecutive days without direct sunlight.  Therefore, holiday festivities take on added significance to provide some variety and color to our routine.  Christmas also marks the mid-point of our stay on top of the ice cap so we’ve got a number of reasons to mark the date.

Contemporary Greenlandic Christmas traditions are primarily derived from the Scandinavian cultures, which is to say they are very similar to those of the US.  And with Summit’s current residents all hailing from the US, our preparations are familiar with a few accommodations to local conditions.  We can’t go out and cut our own Christmas tree, or even go to the local nursery to buy one.  Instead we’ve got a small artificial tree in our lounge.  We don’t have a fireplace, so we’ve hung our stockings on a world map next to the tree.  We’re still looking for the colored lights to hang outside so that Santa knows where to find us.  Summit Station is only about 1100 miles from his workshop at the North Pole, so we’ll watch for eight tiny reindeer (plus Rudolph) on their training runs.

The nearest church is several hundred miles away, and there aren’t any neighbors that we can sing carols to, so some traditions from home will have to be skipped this year.  And of course science never takes a holiday so we’ll have a few work tasks to complete before enjoying the day.

Happy Holidays to all our friends and families back home!


Rapid Coastal Erosion Correlated to Diminishing Sea Ice

December 16, 2009

Retreating sea ice leaves the Alaskan coast vulnerable to the full force of the ocean. Photo: Benjamin Jones, USGS

Rapid erosion of the northern coastline of Alaska midway between Point Barrow and Prudhoe Bay is accelerating at a steady rate of 30 to 45 feet a year, according to CPS-supported scientists presenting a study at the annual American Geophysical Union meeting this week in San Francisco. As the coast erodes, frozen blocks of silt and peat that contain 50 to 80 percent ice topple from bluffs into the Beaufort Sea during the summer.

The acceleration is caused by a combination of large waves pounding the shoreline and warm seawater melting the base of the bluffs, said CU-Boulder Associate Professor Robert Anderson, a co-author on the study. Once the blocks fall they melt within days and sweep silt material out to sea.

Anderson, along with collaborators Cameron Wobus of Stratus Consulting and Irina Overeem of CU’s Institute of Arctic and Alpine Research (INSTAAR) have studied the coastline for the past two summers with Office of Naval Research support. Equipped with two meteorology stations, a weather station, time-lapse cameras, detailed GPS and wave sensors outfitted with temperature loggers, they documented the summer ocean/shore dynamic.

Triple Whammy

Declining sea ice, warming sea water, and increased waves create a “triple whammy” that expedites erosion. For the majority of the year, the Beaufort Sea is covered with sea ice that disconnects from the coast during the summer. These ice-free summer conditions are lasting for longer periods of time, allowing warmer ocean water to lap the coast and weaken the frozen ground. And the longer that sea ice is not connected to the coastline, the further the distance grows between the ice and the shore.  This open-ocean distance between the sea ice and the shore, known as “fetch,” increases both the energy of waves crashing into the coast and the height to which warm seawater can come into contact with the frozen bluffs, said Anderson.

The shoreline bluffs are made up of contiguous, polygon-shaped blocks, primarily made of permafrost and measuring roughly 70 to 100 feet across. Ice “wedges” (created by seeping summer surface water that annually freezes and thaws) are driven deep into the cracks between individual blocks each year. The blocks closest to the sea are undermined as warm seawater melts their base, and eventually split apart from neighboring blocks and topple during stormy conditions, said Anderson.

Impacts of Erosion

As the coastline submits to the ocean, old whaling stations, military and oil related infrastructure and entire towns threaten to fall into the sea. In addition, the loss of sea ice alters ocean systems and diminishes habitat for creatures like the polar bear.

According to a 2009 CU-Boulder study, Arctic sea ice during the annual September minimum is now declining at a rate of 11.2 percent per decade. This year, only 19 percent of the ice cover was more than two years old — the least ever recorded in the satellite record and far below the 1981-2000 summer average of 48 percent.