In Alaska’s tundra desert, freshwater fish survival depends on a lesser-known — and thinning — permafrost feature

On the North Slope, all eyes are on aufeis.

Fewer than 30 miles from the shores of the Arctic Ocean, where the Shaviovik River wove as blue, frozen braids across the Coastal Plain, the peculiar wanderings of one wolverine — a scavenging male nicknamed Nimbus — sparked the beginnings of a mystery in April 2022.

Five cold weeks passed that spring at Kavik River Camp, an abode built just outside the Arctic

National Wildlife Refuge and the home base for several wildlife ecologists from the University of Alaska Fairbanks. In the shadow of the Brooks Range, the team spent the season outfitting live-trapped wolverines with GPS collars as part of a study to better understand the mammals’ movement patterns.

“Anywhere that they spent more than a couple hours was of potential interest to us,” said Tom Glass, a UAF research ecologist. Almost immediately, Nimbus’s travels, which appeared as recurring lines and dots on Glass’s computer screen, caught his eye.

“I saw the data start to come in from this animal, and he was returning to this same site repeatedly over the course of days,” Glass said. “He might go out to a surrounding area and spend an hour or two away from that place, but he’d always come back.”

Nimbus journeyed with a determined fervor, trekking over and over again — no less than 17 times over a span of 50 days — to a stretch of land known as Shaviovik Spring. Though the spring itself spans more than a mile across the Shaviovik River, Nimbus always settled for hours within the same 60-yard radius.

The behavior was typical of a wolverine that had found and was gradually eating a moose, muskox or caribou carcass. But when Glass and a colleague ventured to the spring via snowmachine, expecting to find bones and last leftovers of frozen flesh, they came upon something much different: heaps of frozen, glittery, fish scale-filled poop.

“I had never seen that before,” Glass said. “I’d been doing this kind of work for three years on the North Slope, following around a lot of wolverines. I’d never seen them eating fish.”

The unexpected discovery inspired further investigation. When Glass eased his vehicle across the frozen river’s floodplain to the middle of the channel, he was taken aback when he saw — sticking a Go-Pro camera through a layer of newly formed ice — nearly 100 dead Arctic grayling frozen inside, a layer of liquid water just beneath them.

Nimbus had been digging holes in the shallow ice and eating trapped fish, the spring his own well-stocked freezer.

“It was a morbid scene, and I was shocked,” Glass said, recalling when he first saw the footage on his computer back at camp. “I ran over to my coworker’s bunkhouse and showed her the video, and she was shocked, too. It was this very surprising, eye opening moment.”

All eyes on aufeis

The die-off Glass and his colleagues discovered and later wrote about in the journal Ecology speaks to the fragility of Arctic freshwater ecosystems on Alaska’s North Slope, where temperatures rarely exceed single-digits and often linger well-below zero degrees Fahrenheit through the winter and spring.

“Studies have estimated that the aquatic habitat on the Coastal Plain of the Arctic Refuge shrinks to just five percent of what it is during the summer,” said Julian Dann, a graduate fellow with the Alaska Climate Adaptation Science Center.

The only fresh water that remains liquid throughout these seasons is found either deep in lakes or pooling as small springs that receive a consistent flow of warm groundwater. These liquid springs, few and far between across the region, often hug the edge of larger river systems. Because they are the only locations where Arctic grayling and Dolly Varden populations are able to survive the winter, researchers like Glass often dub these perennial springs “fish oases.”

These oases typically remain liquid year-round, though their shallower, downstream shores tend to gradually freeze in the frigid conditions. Thin, sheet-like stripes of ice, called aufeis, thicken as they layer against each other. Over the course of a winter, aufeis fields can grow tens of miles in length and width, and may freeze to be several meters thick.

Because of their massive size, aufeis endures on the Arctic landscape. Unlike snow, which disappears soon after the weather warms, aufeis melts gradually throughout the summer to create consistent freshwater flows. Fish rely on these passages to connect to main river stems, spawning grounds and feeding sites. In unglaciated basins, aufeis is the second-largest source of summer surface freshwater storage, a crucial resource for human communities.

Glass later wrote about his experience at Shaviovik spring in the The Revelator: “Old reports we found mentioned thousands of fish preparing to overwinter here in autumn, and we learned that it’s known to the local Iñupiat, who harvest these fish for food, as the ‘Place Where the Land Sweats.'”

Historically, Dann said, aufeis fields have accumulated more mass in the winter than they lose in the summer. But aufeis coverage in some locations is changing. Communicated in a series of Alaska Climate Adaptation Science Center funded studies published as recently as April 2025, Dann has worked to improve satellite imaging that distinguishes between aufeis and snow. Analyzing data collected between 2010 and 2021 across 83 aufeis sites in northeast Alaska, he showed that the seasonal extent of aufeis fields is shrinking.

While the exact cause is unknown, Dann has hypothesized that warmer winters means it takes longer for aufeis to freeze, resulting in longer fields with thinner ice, which melt faster and more completely in the spring.

This positive feedback loop — less aufeis forming in the winter, more melting in the summer — means less aufeis overall, which threatens water availability and the survival of fish in the summer. In the winter and spring, thinner aufeis fields and shallower bodies of water are more sensitive to dynamic temperature changes, resulting in melting and freezing that never occurred in the past.

This may explain what occurred at Shaviovik Spring, the Shaviovik river’s only known overwintering habitat for freshwater fish.

“In the springtime, when frozen rivers start thawing out, they don’t flow based on the gravel or riverbed underneath,” said Randy Brown, a fish biologist with the U.S. Fish and Wildlife Service who has been studying North Slope river systems for decades.

“The meltwater goes right over the top, and wherever there might be a fissure in that ice, it’ll run it right down to the gravel,” Brown said. “It’ll start flowing and it melts that ice on the top of the aufeis field.”

That a fish oasis had suddenly become a trap clearly shows how variable seasonal temperatures influence water availability, aquatic life and mammalian survival.

“This morbid find, equally fascinating and disturbing … connected seemingly distant corners of the ecosystem — fish, ice, wolverine, people — and reminded us that even little things, like a slow spring or a cold winter, can tip the balance,” Glass wrote.

The age of water

“Hydrologists speak of Alaska’s Coastal Plain the same way experienced pipefitters might size-up an old toilet: its plumbing is relatively straightforward.

What begins as rain in the Brooks Range gradually seeps underground and drains northward, filling lakes, rivers, springs, and forming aufeis fields as it flows towards the Beaufort Sea.

For years, it was assumed that the underground portion of this water’s journey was exceptionally long. Early UAF research, based on data collected between 1985 and 2009, indicated that it took at least 1,000 years for groundwater to move completely through the system, from mountains to coast.

But the small group of scientists focusing on aufeis in recent years have found alarming and conflicting realities. The actual number, said Josh Koch, a research hydrologist with the U.S. Geological Survey, may be closer to 20 years.

“Traditional literature talks about the old age of water coming out of underground aquifers and into these aufeis fields,” Koch said. “But using some new methods, we were able to show that, in fact, a lot of that water has only been in the ground for a short amount of time.”

Underground aquifers in other parts of Alaska are massive and extensive, Koch said, sometimes encompassing the size of entire states. Researchers knew this wasn’t the case on the North Slope, though a lack of long-term data collection meant no one was exactly sure what lay beneath their feet.

In the spring of 2021, Koch and his colleagues used chemical tracers to follow and measure the age of water as they settled in aquifers beneath Alaska’s Arctic Slope. The tracers contained a signature of the atmosphere from when the water fell as rain, and seeped underground.

“By looking at the chemistry of that water, we could sort of back-calculate when that water must have recharged into the aquifer,” Koch said.

Koch and his colleagues were surprised to learn how the aquifers beneath Arctic National Wildlife Refuge were fed by short, shallow flows. The amount of water below ground was limited, traveled short distances before returning to the surface, and completed its journey through the system in only a few decades.

“We went into this study asking: Can we quantify how old the water is, how regionally connected water may be, and how much availability there is?” Koch said. “We learned that water is young and very limited. There’s no regional deep aquifer.”

These findings, published in March 2024, show the vulnerability of one of the world’s largest wetlands, home to fish and 200 species of migratory birds, as well as Alaska Native communities who rely on healthy streams for subsistence foods and water.

The prevailing 1,000-year timeline, scientists thought, would allow time for any pollutants to filter out of the water as it traveled for centuries through cool, frozen subsoils. But smaller volumes of groundwater, which follow shorter paths underground, are more susceptible to pollution and warming — an especially important conclusion as the Arctic National Wildlife Refuge, which is experiencing atmospheric warming, is also being considered for oil and gas development.

“With the short flow paths, we know that there’s a risk of contamination, and so any contaminants that do get into the subsurface are going to show up in our surface waters quickly,” Koch said. “There could be impacts for what that means for water temperatures as well.”

Contradictingly, while the Coastal Plain’s underground aquifer is currently smaller and shallow, warmer seasons and permafrost thaw are causing it to expand. As soils thaw, there’s more volume underground to store groundwater. Eventually, this water reaches the surface and forms aufeis.

“In some places aufeis fields are decreasing in size, but in other places they’re actually increasing,” Koch said. “We’re really challenged in the ways we can learn about the subsurface. These aufeis fields are these great windows into what’s happening below ground.”