Out of the Crevasse Field

On December 9, 2004, Robert Bindschadler, a glaciologist at NASA’s Goddard Space Flight Center, got a phone call from a colleague in New Zealand. A team of technicians was trying to complete an overland traverse across the frigid vastness of Antarctica, from McMurdo Station near the coast to the South Pole. And the team had run into some trouble.

Bindschadler, a veteran of Antarctic research, knew all about the trouble the team faced: crevasses. Antarctica’s ice moves slowly, but it always moves. As a result, it often splits apart to form crevasses—cracks in the ice that can be tens of meters deep, tens of meters across, and many kilometers long. If every one were a visible, gaping hole, crevasses might merely be annoying, but when they hide beneath the snow, these cracks in the ice become deadly dangerous.

  Page 2

(Image in title graphic courtesy Ted Scambos, National Snow and Ice Data Center)

  Snow bridge over crevasse

“Snow blows around in Antarctica a lot,” Bindschadler explains. “It builds these snow bridges, so as a crevasse opens up, the snow sort of stretches across the top of it. It can be a few meters thick, or it can be centimeters thick.” A lightweight skier might be able to cross a thick snow bridge safely by distributing his or her weight over a wide area, but heavy vehicles don’t stand a chance. “So that’s the danger: that you walk right on top of a snow bridge and you just break through and you fall.”

In their approach to the Transantarctic Mountains, the Antarctic traverse team members had run into an area filled with crevasses they hadn’t anticipated, as many as 10 in a single mile. They called George Blaisdell, operations manager for the U.S. Antarctic Program, who then called Bindschadler. Blaisdell posed a simple question: Can you help the team find a way out?


A sinister Antarctic crevasse lurks under a snow bridge. Nicknamed Mongo, this crevasse was 32 feet (9.8 meters) wide and 82 feet (25 meters) deep. Though thick enough to support a person, the snow bridge could not support a heavy vehicle. The traverse team had to fill the crevasse with snow so that heavy vehicles could cross it safely. (Photo courtesy Russ Alger, Cold Regions Research and Environmental Laboratory, National Science Foundation)


Beating a Path to the Pole


The Antarctic traverse team was on its way to South Pole Station in a proof-of-concept trek from McMurdo Station. The National Science Foundation (NSF), which operates both stations, commissioned the traverse, and Bindschadler had worked on NSF-funded projects in the past. “It costs a lot to support people at the South Pole. The primary cost is just the fuel to keep them warm,” says Bindschadler. Ships regularly supply McMurdo on Antarctica’s coast with fuel. “But the last, most difficult part is getting the fuel from McMurdo to the South Pole,” he says. Historically, South Pole Station has been supplied by costly airplane flights. According to a report by the McMurdo Area Users Committee, aircraft used to transport fuel consume a gallon for every gallon they deliver.

Added to the high fuel demand is the disadvantage that cargo transported south must fit into the airplanes’ fairly limited storage compartments. Large pieces of big telescopes, for instance, barely fit.

“The South Pole has a 9,500-foot elevation, very low water vapor, and the cleanest air in the world,” says Blaisdell. “Because it’s at a pole, everything rotates around it. All those things make it an ideal place for astronomy.” Several types of telescopes operate at the pole, and construction is underway on a neutrino telescope that will be a cubic kilometer in volume. Transporting heavy materials for construction projects like this may be much easier on the ground than in the air.

  page 1Page 3

Map of Antarctica

Reaching the South Pole from McMurdo Station, which lies along the coast of Antarctica, means negotiating challenging terrain on the Ross Ice Shelf and in the Transantarctic Mountains, which line the coast on the McMurdo side of the ice shelf. (Image adapted from CIA World Fact Book)

  Neutrino telescope parts

Another reason researchers wanted an overland—or oversnow—traverse was that airplanes in Antarctica are limited. Finding a different way to supply South Pole Station would free up flights to do other work.

Once planners had settled on the concept of an oversnow traverse, they had to pick the best route. In 1958, Sir Edmund Hillary completed an adventurous motorized trek to the South Pole by driving up the Skelton Glacier near McMurdo. While Hillary’s route had its merits, especially the proximity to McMurdo Station if anyone ran into difficulty going up the glacier, researchers eventually settled on a different route. They chose to go across the Ross Ice Shelf to the base of the Leverett Glacier, up the Leverett, and across the Antarctic Plateau.


In a mile-deep hole drilled in the ice by hot water jets, scientists have installed a piece of a new neutrino telescope at South Pole Station (left). Two tanks that are part of the neutrino telescope’s detector array sit on the ice surface (right). These and other large pieces of equipment may be more easily transported overland than by airplane in Antarctica. (Photos courtesy Berkeley Lab Research News)

  Transantarctic Mountains

One reason for this choice was vehicle performance. Reaching the Antarctic Plateau by crossing the Transantarctic Mountains at the start of the journey would force the vehicles to climb to the highest elevations carrying the most fuel. Once they are at the higher elevation, vehicles also operate less efficiently. By deferring the trip across the mountains until later in the journey, the traverse team could reduce the amount of fuel (and therefore, the weight) the vehicles had to carry up the mountains, and keep them running smoothly for a longer period.

Bindschadler was one of the researchers the Antarctic Traverse planners had consulted for the best possible route, and he liked the Leverett Glacier. A small glacier not carrying much ice, it was less likely to hold many crevasses. It also had a low, gentle grade, making it easier for the team to climb. Though disconnected from the traverse project after helping choose the route, Bindschadler stayed interested in how his recommendation would work out. By the time Blaisdell called him in December 2004, it didn’t appear to be working out as well as he hoped.


The Transantarctic Mountains roll into the distance. The Antarctic Traverse team followed a route that negotiated these rugged, high mountains later in their trip, when their vehicles were less weighed down with fuel. (Photo courtesy National Oceanic and Atmospheric Administration Photo Library)


Rules of Engagement


In December 2004, the traverse team found itself stalled near the base of the Transantarctic Mountains. Everyone had known they would encounter crevasses, but no one had anticipated the sheer number of them. Except for the route they had taken into the crevasse field, the team members couldn’t find a reliable way out.

  page 2Page 4
  Crevasse scouting and exposed crevasse

In the field, Antarctic researchers and technicians use a couple of different methods for detecting crevasses. One is to drive over the terrain in a lightweight snowmobile equipped with radar that can sense open spaces under the snow. Such a light vehicle isn’t likely to exert enough pressure to break through a snow bridge. Big, heavy vehicles like tractors use long booms on the front with ground-penetrating radar equipment mounted on them. The heavy vehicles move slowly enough to stop if the radar detects a crevasse.


A researcher dragging a sled of gear scouts for crevasses on the Ross Ice Shelf (left). An exposed crevasse gapes at the surface. In front of the hole, someone has plunged two flagged poles and an ice axe (right). (Photos courtesy Ted Scambos, National Snow and Ice Data Center)

  Traverse camp

“Likewise, at night when the team stops to camp, the ground-penetrating radar covers a 200-foot-diameter circle and surveys that entire area. The vehicle drives around in that circle, and that track can be fairly readily seen. That becomes the ‘safe area.’ So at night, if people want to ski or walk around, or just sit out and read a book, that’s safe. They don’t go outside that ring,” says Blaisdell.


Taken with a remote-controlled airplane, this picture shows the traverse team’s camp on the Antarctic Plateau, after a successful ascent of Leverett Glacier. Tracks in the snow outline the camp’s “safe” area. (Photo courtesy John Penny and The Antarctic Sun)


“What we’re doing is extremely different than what Hillary or Scott or Amundson or some modern-day adventurers have done. We’re not doing this to say we’ve been someplace and had an adventure; we’re doing this to set up a system for routine supply of critical goods to the places that need them,” he says. “This is not about being a hero or taking risks.” The trek was organized, start to finish, to minimize that kind of drama.

Although they were never in mortal danger, Blaisdell explains, the team was frustrated. At the base of the Transantarctic Mountains, the team’s forward progress ground to a halt as they encountered and avoided crevasse after crevasse. No one wanted to turn back; everyone wanted to cross the mountains. But even though they had the ability to keep themselves from tumbling into any crevasse in their path, they had no way of knowing which path would be the least trouble. What the team needed was a different perspective.

  View out of a crevasse

At McMurdo Station, researchers trained to rescue each other from crevasses. Ted Scambos took this picture while waiting to be “rescued” from the void. (Photos courtesy Ted Scambos, National Snow and Ice Data Center)


The View from Space


Bindschadler started using the perspective of remote sensing over 25 years ago. “My first season in Antarctica, I didn’t realize I also came with my preconceived notions about how one studies glaciers, based on what I had done in my post-graduate work. I knew Alaskan glaciers were big, but Antarctica was just a totally different scale, and you could not be very effective in the time you had to spend there unless you included satellite data,” he recalls.

Satellite remote sensing complements fieldwork both before and after a trip to Antarctica. “It gives you the big picture before you go. You can see what’s uniform and what’s changing and decide where you want to do fieldwork,” Bindschadler explains. “After you get back, it helps you put your measurements into a broader context. So remote sensing permeates every aspect of the planning, the conduct, and the analysis of the fieldwork.”

  page 4
  Crevasse field

Based on his decades of experience with studying frozen landscapes with satellite data, Bindschadler suspected that a remote-sensing perspective could guide the traverse team away from the most crevasse-filled areas and enhance their safety.

The snow bridges that cover crevasses often sag. The sags are so slight that, from the ground, they can be hard to detect, but Bindschadler knew from experience the right satellite might be able to find many of them from space. After taking Blaidsell’s call, Bindschadler and his associate, Patricia Vornberger, began poring over the two most promising satellite data sets: radar data from the RADARSAT Antarctic Mapping Project (RAMP) and multi-spectral data from the Advanced Spaceborne Thermal Emission and Reflection Radiometer on NASA’s Terra satellite.

In 1997, Canada’s RADARSAT-1 satellite collected images that led to the first high-resolution map of the entire continent of Antarctica. RADARSAT also offered an advantage not available from every satellite: it could see through clouds. Over the past several years, RAMP data have done a reliable job of finding crevasses in Antarctica, so it was logical for Bindschadler to turn to these data again.


A field of crevasses fades into the distance on the Ross Ice Shelf. The smooth ice on the right is frozen to the bedrock. The rough ice on the left and in the distance is part of the Bindschadler Ice Stream (formerly named Ice Stream D). The strain of movement causes crevasses to form in this ice. (Photo courtesy Ted Scambos, National Snow and Ice Data Center)

  Comparison of RAMP and ASTER images

The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor flies on NASA’s Terra satellite. The high-resolution sensor is a cooperative effort between NASA,  Japan’s Ministry of Economy, Trade and Industry, and Japan’s Earth Remote Sensing Data Analysis Center. “By pure serendipity, we had some ASTER images of the same area because we were studying the dynamics of the nearby Whillans Ice Stream,” Bindschadler says. “Since we had them in our lab, we could quickly take a look at them.” While Bindschadler searched the RAMP data, Vornberger concentrated on ASTER data. The two quickly discovered that ASTER data showed crevasses where RAMP didn’t.

A close look at the ASTER imagery shows a series of crevasses on the ice shelf. From space, each crevasse looks almost like a giant had pressed his thumbnail into the surface of the ice. RAMP imagery of the same area shows larger ice shelf features, but the crevasses aren’t apparent.

ASTER has a higher resolution than RAMP (15 meters compared to 25 meters), and this may have given it an edge in finding crevasses. Likewise, ASTER “sees” the same kind of light that the human eye sees, whereas RAMP looks at a different part of the electromagnetic spectrum. So ASTER’s images may be more intuitive to our eyes.

Armed with Bindschadler and Vornberger’s crevasse predictions, the traverse team was back on track. “We gave them the waypoints we felt very confident were safe for surface travel,” Bindschadler says. “They followed those, but they didn’t trust us completely. I don’t blame them because they had already had some unpleasant surprises.”

Blaisdell confirms that satellite imagery is no silver bullet. “It’s safe to say that if you see a crevasse feature on satellite imagery, you’ll find a crevasse. But you cannot assume that if the satellite doesn’t see a crevasse, there won’t be one. For a traverse team, satellite imagery is one of the many tools we use.” Silver bullet or not, Blaisdell estimates that the ASTER imagery saved the traverse team a tremendous amount of time. “If ASTER couldn’t have seen the crevasses, it would have taken us probably a full season to probe our way through that area using ground tools,” he says.


Although RAMP has a history of finding large crevasses, this RAMP image (top) shows no crevasses while an ASTER image of the same region shows many (bottom). (Images courtesy Robert Bindschadler, NASA GSFC)


Once out of the crevasse field, the traverse team marched steadily up the Leverett Glacier and toward the pole. By 8:30 p.m. on January 4, 2005, all the tractors in the caravan had driven up onto the polar plateau. Everyone wanted to continue all the way to South Pole Station, and they had the fuel to get there. But they didn’t have enough for the round trip to the pole and all the way back to McMurdo. The U.S. Antarctic program could have flown the fuel to South Pole Station, but bad weather put the supply flights behind schedule. The traverse team had to turn back.

Team members were disappointed at not reaching the pole, but the traverse team had nevertheless far exceeded its own expectations, thanks in a large part to ASTER data. Bindschadler explains, “This shows the value of using imagery to assess the safety of an area, especially new areas we haven’t seen before. Satellite imagery is a perfect tool that’s underutilized. We really need to appreciate that fact and use it more often.”

  • References
  • Bindschadler, R., and Vornberger, P. (2005) Guiding the South Pole Traverse with ASTER Imagery. Journal of Glaciology 51(172), in press.
  • McMurdo Area Users Committee. (2001) Science Advantages of an Oversnow Traverse to Resupply S. Pole. (PDF document) Accessed August 26, 2005.
  • Hutchison, K. (2005) Trailblazers’ Long Haul to the Plateau. The Antarctic Sun February 6, 2005.
  Antarctic traverse team

The Antarctic Traverse team members pose for a picture at the southernmost point of their trek. (Photo courtesy John Wright and The Antarctic Sun)