Some like it cold

As graduate student Kiya Riverman plants her crampons into solid ice, she inspects the surrounding walls. No cracks, so she knows she is safe here. She unhooks her carabiner from the rope used to rappel fifty feet down into the ice cave where she now stands, inside a glacier on the Norwegian archipelago of Svalbard. She pulls out a scratch pad and pen and begins sketching the shape of the tunnels before her, tunnels carved by water melting at the glacier's surface and cutting its way down, forming deep canyons whose walls later closed in above them.

Nearly 4,000 miles away, senior scientist David Pollard sits at his desk on the University Park campus, next to whimsical pictures drawn by his children, and types numbers into his computer. He hits Enter to start a simulation which, after it finishes running in several days, will yield the most accurate prediction yet of what Antarctica's ice sheets will look like in 500 years. Pollard’s model shows how the warming from increasing greenhouse gases will melt large portions of the ice sheet that now caps the pole, and how that meltwater, dispersed into the ocean, will contribute to a rise in global sea level.

The explorer and the modeler. They work in very different modes, yet Riverman and Pollard share a passion for understanding what makes glaciers and ice sheets move, shrink, and grow, and their efforts are closely linked. Both are members of a cadre of scientists known as Penn State Ice and Climate Exploration, or PSICE.

Riverman's goal in exploring ice caves is to track the path of glacier meltwater as it descends from surface to bedrock. Once it meets bedrock, this water can provide lubrication that speeds  the glacier's slide toward the sea. The data Riverman collects are an invaluable commodity for computer modelers like Pollard. His work, in turn, can determine what Riverman or another explorer will look for next.

This close collaboration between modelers, observational scientists, and others focused on the ice has allowed PSICE to make a number of significant advances in glaciology since its formation in the early 1990s.

Bridging boundaries

The PSICE group originated when Charles Hosler, then dean of Earth and Mineral Sciences, created the College's Earth System Science Center. Hosler and his successor, John Dutton, helped lay the groundwork for researchers to pursue a broad, holistic approach to understanding how Earth’s systems — air, ocean, water, ice, and more — function naturally, and how human activity might be affecting those systems. They hired climate scientist Eric Barron, now Penn State’s President, as the center's first director. One of Barron’s first hires, in turn, was a young glaciologist named Richard Alley.

"He was working to assemble a broad-based, comprehensive group of Earth system researchers, and wanted coverage of ice,” says Alley, now Evan Pugh University Professor of Geosciences. “I was hired to provide that."

Alley had cut his teeth, glaciologically speaking, on Greenland ice cores, tubes of ice the diameter of a two-liter soda bottle that totaled two miles in length and had been painstakingly drilled from the Greenland ice sheet beginning in 1989. These cores, composed of layers of snow accumulated and buried over thousands of years that gradually densified into ice, provide a unique window into the history of Earth’s climate.

As a graduate student, Alley had gone to Greenland to help learn where to drill, and soon after arriving at Penn State he was part of the team that extracted the cores. He and several collaborators then developed a more accurate method for dating the cores, vital to establishing a record of changes in snowfall, temperature and other parts of the climate over more than 100,000 years.

One of the major revelations to come out of the ice core data was that Earth’s climate can change drastically in a relative heartbeat. “We’ve seen changes from a climate like that of northern Alabama to conditions similar to southern Maine that occurred over a period of about ten years,” Alley says.

When he started in the field, atmospheric carbon dioxide levels were not considered a matter of pressing urgency. Today, as the Earth’s warming intensifies due to carbon dioxide and other greenhouse gas emissions, it’s becoming imperative for glaciologists to understand what that warming means for the future of ice sheets, glaciers, and the world’s oceans.

All in the family

In 1992, the PSICE group doubled when Sridhar Anandakrishnan joined as a research scientist working with Alley. Now a professor of geosciences, Anandakrishnan uses his background in electrical engineering to develop devices that can monitor subtle changes in the ice sheets using a combination of seismic and radar waves, GPS coordinates, and other data. His networks of devices have provided researchers a way to see through glaciers without cutting into them.

The two have continued to build the group in the years since, adding a half dozen researchers with complementary skillsets. “PSICE studies how ice interacts with the land, how ice interacts with erosion, the properties of ice as a material, and ice as a historical archive,” says Alley. “But our main focus recently has been to understand how ice is a driver of sea-level rise.”

It takes a special brand of hardiness to sign on for a three-month season in Antarctica or Greenland, to live in a tent and work on the ice, long grueling days at the mercy of sun and wind and cold.

“When the wind blows it is unrelenting; there’s nothing to stop it,” writes Don Voigt, a senior research associate and veteran of 18 seasons on the ice. “It’s a rare opportunity to have sunburn, windburn, and frostbite all at the same time.”

Not surprisingly, then, the team’s members share a special bond. Their energy and camaraderie, Riverman says, is what drew her to Penn State in the first place. “It's a dynamic group of people who are excited about figuring out how this part of the world works,” she says. “We meet every week for pizza, spit-balling ideas. It’s science at its finest. I came to one of those meetings and I knew I wanted to come study here.”

Alumni from the group, now scattered around the world, continue to collaborate with Alley, Anandakrishnan, and other PSICE researchers. Riverman, who received her Ph.D. in May 2017, plans to do the same.

"Richard and Sridhar have built a glaciology family that keeps on growing," she says.

History on ice

As Anandakrishnan sees it, a glacier is like a diary.

"It writes down in its body all the dirt and dust that fall on it. All the gases that it's 'breathing,' if you will, and all the temperatures it's experiencing," he says. "We can go back and read that diary, page by page, and we can say ‘here's when it was dirtier,’ or ‘this ice over here is radioactive, so it could have been from years right after World War II.’"

Fresh ice near the surface traps bubbles of gas, which provide even more insight into Earth's past.

"You can pull the bubbles out and put them into a gas analyzer as you would with air today, and that's a powerful tool that's unique to glaciers,” Anandakrishnan explains. “With most Earth systems, animals breathe gases, or use oxygen and interact, so you don't have a pristine view of what the air looked like. But there are no animals on ice sheets. Once snow and air and dirt are captured in ice, they're there forever."  Senior scientist Todd Sowers, another PSICE member, analyzes the gas from frozen bubbles to build long and exquisitely accurate histories of the composition of past atmospheres.

The bubbles can be used in other ways as well, as Alley realized when working with Matthew Spencer, a graduate student during the early 2000s. The density of bubbles in an ice core, for example, provides a gauge of temperature and other factors.

Alley and Spencer developed a method for gathering information from air bubbles without destroying them in the process. Other innovations the team has developed build on existing data-gathering techniques, such as GPS, radar, and satellite imagery. Recently, geosciences grad student Nicholas Holschuh worked with PSICE members to devise a better way to use radar data to map the interiors of ice sheets. “The way a radar wave refracts tells us something about the nature of the ice through which it’s traveling,” Alley says.

As new approaches help to fine-tune our understanding of how glaciers change and move over different time scales, the PSICE team is finding, to its surprise, that the path to understanding the future of Earth’s glaciers lies in combining understanding of short-term changes with the earlier work on long-term factors such as ice ages.

On the move

As snow accumulates on a glacier, the glacier is pushed by its own weight outward or downward, toward the sea. The conventional wisdom is that this movement takes place over extended time periods – “at a glacial pace," as it were. But the phrase is a misnomer.

"There are two ideas that come out of that phrase — that glaciers' movement is slow, which is not true, and steady, which is also not true," says Anandakrishnan. "In some places their movement is steady and slow, but what we've found is that as the glacier approaches the edges of the continent it sits on, it starts to flow faster."

Anandakrishnan was one of the first glaciologists to confirm that glaciers undergo change over periods of hours as well as centuries. He and his team had set up a network of GPS monitors in West Antarctica to track the movement of glaciers. West Antarctica is a rapidly changing area of the continent covered with an ice sheet so vast that, if it one day collapses into the ocean as some scientists predict, it could raise global sea level by as much as 13 feet.

He and his colleagues noticed that Bindschadler Ice Stream, a fast-flowing river of ice in the area, would move and then suddenly stop at regular intervals. As they began to pore through their data, the team noticed an odd correlation—the changes in speed were linked to the tides. Specifically, when the tide comes in it pushes on a glacier, sometimes enough to stop its movement entirely.

"It's a relatively small force,” says Anandakrishnan, “but we found that it was enough to change the speed of these glaciers quite a bit, not just at the coast but also far inland."

Not all glaciers stop dead in their tracks in response to the tides, however. Some are barely affected at all.

The difference has to do with how “slippery” glaciers are, Anandakrishnan says. The giant masses sit on a material that isn't just solid rock. As a glacier moves, it breaks off pieces of the bedrock, which gets turned into a muddy mix with help from subsurface heat and friction and meltwater from above.

"If the bedrock is very bumpy, the force is dissipated due to frictional influences," Anandakrishnan says. " It's like pushing a person in sneakers compared to someone on roller skates."

The PSICE group has also made headway in understanding the effect of meltwater lakes, pools that form when the sun melts the surface layer of ice and snow. This meltwater then drains through the ice sheet to the bed. Its path down through the ice can have varying effects on the speed and slipperiness of a glacier. 

"We came up with a theory on how surface meltwater can drill its way through a thick, cold ice sheet and then did some modeling to see what type of impact this process of lubrication could have on the long-term evolution of the Greenland Ice Sheet," says Byron Parizek, associate professor of mathematics and geosciences at Penn State DuBois and part of the PSICE team.

"We predicted relatively minor changes unless the meltwater accessed the bed in regions where the ice was frozen to the bed." Subsequent observations by other scientists have thus far confirmed that the impact of surface meltwater on ice flow speeds in Greenland is not as straightforward as the old adage "just add water."

To test the meltwater penetration-mechanism hypothesis, Sarah Das, a former PSICE grad student, traveled to Greenland to witness a drainage event. The amount of water she saw draining was staggering— "twice the volume of Niagara Falls in less than one hour," says Alley—and the team was able to confirm its hypothesis.

An uncertain future

What ice sheets will look like 10, 50, or 100 years from now—and whether they will melt into the sea as Earth continues to warm—is a question at the forefront of climate science. To try to answer it, glaciologists rely on computer models that simulate future climate—models informed by observations from the field.

For years, David Pollard had been working on a model that simulates both Antarctica's past and its future. Replicating past conditions on Earth, which scientists can confirm from ice cores and other evidence, is critical for ensuring that a model is accurate.

Pollard was struggling to get his model to conform with some aspects of the historical record  when he heard Alley describe for the first time the huge ice cliffs that would arise if the margins of the West Antarctic Ice Sheet continued to retreat deep into the interior, where the bed is far below sea level but thick ice towers far above. "This led us to add large ice cliffs that can fail structurally into our model," Pollard says.

These failure events result when the ice can't support its own weight. A giant chunk falls off into the ocean, exposing a new cliff face. When Pollard and his collaborator, Robert DeConto of the University of Massachusetts, incorporated cliff failure and hydrofracturing, the effects of water wedging open cracks the way Sarah Das’s lake did in Greenland, into their model, they were able to simulate documented levels of past sea level rise. Theirs was the first model to do so.

That same model drew a lot of attention last year when Pollard and DeConto applied it to the future, and published results of a simulation suggesting that, given a continuation of current trends, global sea level could rise by up to 50 feet by the year 2500, with ice crumbling from the larger East Antarctic ice sheet as well as West Antarctica.

Such a change, though catastrophic, would not be unprecedented. "In past ice ages, global sea level has dropped and then risen about 100 meters with the growth and then shrinkage of huge ice sheets over North America, Europe, and Asia," says Pollard. "The question is, how much will the remaining ice sheets melt and raise sea level over the next thousand years?"

It's a complex question, one whose answer will require both ingenuity and a mix of minds. Parizek says that the PSICE group's accomplishments over the years are an ongoing example of that.

"Every time someone comes back from the field with a new data set, we look at and interpret it to see if we have accounted for the physical processes in our models that likely led to the unexpected behaviors in nature," he says. An updated model, in turn, can suggest a new direction for experiments in the field.

This accumulation of small, complementary steps and varying perspectives adds up to a growing understanding of how ice sheets and glaciers change over a day, a year, and a century. And an understanding that stretches across these multiple timescales, PSICE scientists believe, will be key to predicting how much and how quickly the ice sheets will contribute to sea-level rise.

 “Right now we're trying to reduce the uncertainties by learning more about the flow of the ice in Antarctica," Alley says. "Then we can get together and make informed decisions about the best solution.”

"It's a zero-sum game between oceans and glaciers," adds Anandakrishnan. "Every time you change the size of a glacier, it will immediately affect sea level. There's nowhere else where this water can go."

This story first appeared in the Fall 2017 issue of Research/Penn State magazine.