Polar science can be challenging, but satellites give scientists access to otherwise inaccessible regions.

The smell of penguin guano is one that’s not easy to forget – slightly musky, more than a little fishy, overall grossly unpleasant. And it sticks. To your clothes, your hair, even your skin. As a penguin biologist, it’s a smell I’ve grown to love. It means I’m in my favorite place on Earth: the Antarctic. But Antarctica is changing before our eyes. And it all starts with the ice.

Ice has been in the news. The collapse of the Larsen-C ice shelf in July 2017 is a notable example. Despite its remoteness, scientists had been diligently monitoring the Larsen ice shelf, due to concerns about its instability. And the best way to monitor an ice shelf? From space.

The Larsen C ice shelf is enormous – covering 19,300 square miles (50,000 square kilometers) – and is fairly inaccessible. Thus, to monitor the state of the ice, scientists often rely upon satellite data, which can provide information on the extent and sometimes the structure of the ice. In July 2017, the satellite imagery was particularly critical – NASA’s AQUA satellite captured a massive calving of the ice shelf , creating a 2,240 square mile (5,800 square km) iceberg, and shrinking the Larsen C ice shelf by 10 percent.

When visiting the polar regions, you can’t help but be struck by the scope of ice. Glaciers tumbling down mountains and falling into the sea. This is the ice which took thousands of years to create, with snow slowly accumulating, year after year. The weight of each new layer compressing the snow beneath and squeezing all the air out until it becomes crystal-clear and ice blue. This is what ice sheets, like the now-famous Larsen Ice Shelf, are made of. But there are other forms of ice in the polar regions. Icebergs – bits of glacier which tore away and crashed into the ocean. And then there’s the sea ice.

As opposed to the other types of ice, which formed from layers of snow being compressed into glaciers and then tearing away into the ocean, sea ice is different. This is the ice which was formed from the ocean freezing. That, in itself is remarkable. The ocean has a salinity of about 35 parts per thousand. The dissolved salts in ocean water actually lowers the freezing point. Additionally, as the water freezes, salt is excluded and is expelled into the ocean or condensed into brine pockets, while the ice is left much fresher than the surrounding seawater.

Because sea ice floats, it’s able to move, sometimes out of the polar regions. According to Marika Holland, researcher at the National Center for Atmospheric Research (NCAR), the movement of sea ice has important consequences for the global freshwater budget. Holland notes that in the Arctic, especially, we’ve been seeing a net movement of sea ice (and thus freshwater) out of the Arctic basin and into the North Atlantic. This loss of freshwater from the Arctic has even larger implications – these changes in sea ice are closely linked to the projected increases in precipitation and river runoff within the region – important factors influencing both ecological and social systems.

So we know that the type of ice is important, as is its movement and timing. But how do we study something as ephemeral and widespread as sea ice?

One of the most widely used methods to track changes in sea ice involves the use of satellite imagery. In 1972, NASA began taking sea ice measurements using a passive-microwave instrument aboard the Nimbus-5 satellite. Since that time, with the exception of a two-year gap in the mid 1970s due to instrument failure, NASA and the National Snow and Ice Data Center (NSIDC) have been tracking sea ice with satellites nearly continuously. Now, the NSDIC uses data from several satellites, most of which are collected by the Defense Meteorological Satellite Program (DMSP). Data from these satellites offer scientists near-real-time estimates of sea ice extent and location.  Through these observations, we know that last summer, the Arctic sea ice extent was the eighth lowest on record and the average sea ice extent in both the Arctic and Antarctic are currently tracking well below previous years.

But what can we do with the data?

Ice grows, melts, and moves, sometimes quickly. Understanding how that happens is critical to understanding polar ecosystems.

Sea ice creates important habitat and foraging grounds for a diverse array of species. Heather Lynch is an associate professor at Stony Brook University; she studies the population dynamics of Antarctic penguins. “Sea ice is the nursery for Antarctic krill, which is the key species for all life in the Southern Ocean, including penguins,” she notes. “Without sea ice, there would be no krill, and the loss of Antarctic krill would have unconscionable implications for the Southern Ocean ecosystem.”

Lynch adds that it is not only the amount of sea ice that is important, but also the timing. Casey Youngflesh, a Ph.D. candidate working with Lynch, clarified that each spring, the loss of sea ice drives a phytoplankton bloom, which krill feed on. To understand how penguin breeding is being affected by a changing climate, Youngflesh tracks both sea ice and penguin abundance at breeding colonies from the satellite imagery. “The timing of this break up thus dictates when food is available for penguins,” Youngflesh said. “A later break up of sea ice means food isn’t readily available for penguins during the critical breeding period.”

And that’s exactly what happened last year on the Antarctic Peninsula. Ron Naveen, president and founder of Oceanites, a nonprofit studying penguin populations in the region, explained that, “if the ice isn’t there or it’s radically diminished, like it was last season in the Peninsula (by more than 60,000 square miles), that year’s class of krill will be impacted negatively.”

Youngflesh says less food equals less penguins.

And this is true on the other end of the world, as well. Ian Stirling, from the University of Alberta, has studied polar bears and seals throughout his 50-year career. But if you ask him about the most important factors affecting these species, he’s quick to cite the impact of sea ice.

“The average date of breakup (in Hudson Bay) is three weeks or more earlier than it was 35 years ago,” said Stirling. “And we have the data to show how this has affected [polar] bears on the west coast of Hudson Bay during that time: Polar bears are 30-40 kilograms of fat lighter than they were 40 years ago; reproduction is very low.”

Now in the twilight of his career, you can find Stirling aboard tourist vessels several months a year, visiting the Arctic and Antarctic. His goal is to figure out how polar bears use the ice around Svaldbard in early summer.  But in his quest to answer that question, Stirling was struck by something even more profound. “(When I began studying them), we couldn’t even access many of the bears – there was too much ice. Now, in Svaldbard, many fjords have no sea ice at all by January and February,” he said.

In the Arctic, sea ice doesn’t only affect the bears and seals. Veronica Padula is an Earthzine Writing Fellow and Ph.D. candidate at the University of Alaska studying seabird foraging ecology. She explains, “Many seabirds hunt at the ice edge – knowing where they hunt means knowing where we can find them and what they’re eating.”

But as the patterns of sea ice change, these birds are moving. One study found that Little Auks (Alle alle) in the Franz-Josef Land archipelago have completely shifted their foraging grounds from the offshore sea ice margin to inland grounds at the foot of retreating glaciers, and these changes were associated with a decrease in adult body mass.

Tracking how sea ice has changed is critical to understanding its impacts on ecosystems. But perhaps an even harder question is how it will continue to shape the poles in the future. That’s Marika Holland’s goal.

Holland,  uses high resolution satellite imagery data to model sea ice dynamics and make predictions about future sea ice. She’s trying to understand how sea ice varies over seasons, decades, and even centuries, and how those changes affect ecosystems and larger-scale climate systems of the Earth. But modelling sea ice is no easy feat. While there are physical laws which determine its heat budget and movement, understanding the finer scale dynamics is more challenging. It is the fine-scale details, like how the variability in how snow cover is distributed across the ice , which can have profound effects on sea ice models. And it is these details which they’re trying to resolve now.

Holland notes that this dependency upon multiple lines of evidence and methods of data analysis create “a very collaborative field because we’re all fitting the pieces together.” The pieces she’s focused on include improvements in computational power and the physics of fine-resolution dynamics.

She notes that “if any part of our model is wrong, the whole model will be wrong; it’s an iterative process – one we’re constantly improving.”

But recently, their concern has been raised about the satellite data used to make these models. Four satellites which the NSIDC rely on to maintain the nearly 50-year time series are aging; at eight, 11, and 14 years old, these satellites are well past their anticipated lifespan of just five years. Walt Meir, a sea ice specialist with the NSIDC recently told Nature magazine, “Every day it’s more and more risk. If one of those goes it will get to be nail-biting time, and certainly if two of them go.”

Currently, there are no clear plans for the United States to launch a new satellite carrying the microwave sensors the NSDIC currently relies on until 2022.

And polar scientists are concerned. “Changing sea ice has massive implications for not only polar ecosystems, but also the rest of the world,” says Heather Lynch. “We need these satellites to help us predict what’s coming and mitigate those impacts,”

Catherine Foley is 2017 Earthzine Writing Fellow and Ph.D. candidate in ecology and evolution at Stony Brook University where she is studying the population and spatial ecology of Antarctic penguins and seals. Follow her on Twitter @cmrfoley.