When Dr. Steven Amstrup, chief scientist for Polar Bears International, began studying polar bears in Northern Alaska in the 1980s, he seldom saw bears on land there. The sea ice just offshore was located over waters teeming with food, and bears would spend their time out on the ice dining on seals and other prey species.

“I remember my early years going to Alaska, and if I was on the North Slope, like in Prudhoe Bay or Barrow, Alaska, I could look north in the summertime and the sea ice was right there,” Amstrup says. “Now, at those same times of year, the sea ice is hundreds of miles offshore, and the remaining ice is over deep, unproductive water.” That’s unproductive, as in not teeming with food.

Nowadays, the region’s polar bears have two choices: follow the ice out to deeper, unproductive waters for the summer – an option yielding little food – or try finding scarce food on land. “The ultimate situation is the same: a longer and longer fasting period,” Amstrup says.

Climate change is affecting Alaska’s polar bears and other polar bears around the globe, and it’s affecting the bears’ populations in different ways and over different timelines. That helps explain why Amstrup and his colleagues examined 19 polar bear subpopulations to analyze how each is being affected by changing local conditions. They’ve found that unless greenhouse gas emissions are reduced, most polar bear subpopulations, but not all, could be gone by 2100.

“As the world warms and the ice responds by melting, we are going to see polar bears in different parts of their range responding at different times,” says Amstrup, whose team published its results in Nature Climate Change. Amstrup has led polar bear research projects since 1980, and his research led to the species being listed in 2008 as threatened under the Endangered Species Act.

A new approach to learning about polar bears

Researchers previously had tended to analyze polar bears as a single group, often extrapolating data from widely studied subgroups to the population as a whole. However, Amstrup’s team worked to create individualized estimates for each subpopulation to account for the variety of climates, habitats, ecosystems, and sea ice ecoregions they encounter.

Collecting data about polar bears has always been a huge challenge. Amstrup explains that one common approach to polar bear research involves flying helicopters over sea ice – or looking for Hudson Bay bears on land during the summer – and shooting immobilizing darts at bears before physically examining them. A multi-year recapture approach is commonly used to see how the bears’ body condition has changed over time, and scientists observe changes in the bears’ physical condition and health over time.

“As you can imagine, the answers from work like that come along pretty slowly,” Amstrup says. “You can only catch so many bears, or they’re budget limited. How long can I keep a helicopter with me? How long are the environmental conditions safe?” He points out the need for solid sea ice to walk on and good daylight, leaving relatively short windows of opportunity, typically in the fall and spring.

Flying aerial transects to count bears is another common research method. “[It] gives you a snapshot of how many bears might be around in any one year,” Amstrup says, noting the method is limited by expense and the small areas that can be realistically surveyed. “When you think of polar bears living over the whole Arctic and you’re just going to go out and do aerial surveys over Hudson Bay – that’s a pretty small part of it,” he says.

Advancing understanding of polar bears through ‘basic energetics’

The challenges inherent with these research methods led the researchers to try a different approach to analyze the future of polar bears, using basic energetics, fasting thresholds, and predictions for future sea ice.

“The real issue for polar bears is how many ice-free days can polar bears tolerate because polar bears can catch their food only from the surface of the sea ice,” Amstrup says. “And, as the sea ice goes, as the sea ice disappears, they have less and less time on the ice to catch their food, and they have a longer and longer time to fast – a longer time when they’re food deprived.” He describes the approach as “basic energetics,” noting “a bear that’s this fat can only last this long without food.”

Researchers set out to analyze how many ice-free days may occur in the future to determine how long the bears will have to go without food – and if they will be able to survive fasting that long. They analyzed the bears’ “fasting impact threshold,” which the Nature Climate Change paper describes as an estimate of “the threshold numbers of days that polar bears can fast before cub recruitment and/or adult survival are impacted and decline rapidly.”

Some previous studies used a single fasting threshold estimate for all polar bears – typically around 180 days. This study delved deeper, developing a “dynamic energy budget” to analyze how long bears of different sexes and life stages may be able to fast.

Not surprisingly, the team found cubs the most vulnerable (at 117 days), followed by yearlings. Adult males require a lot of food, but their leaner bodies cannot store as much energy, leaving them a projected threshold of around 200 days. Solitary adult females were found to be the most resilient with a fasting threshold around 255 days, and mother bears were found to encompass a wide range of fasting thresholds.

However, the study authors note these figures could vary by months. They also point out that poor ice conditions make it more difficult for bears to eat enough food and store an adequate amount of fat before their fast, meaning they may enter the fasting period in suboptimal condition. They also mention that future ice conditions are likely to be very different from any condition polar bears have ever encountered in the past.

The study authors report also that some subpopulations may already be precipitously close to the brink, saying, “recruitment and survival impact thresholds may already have been exceeded in some subpopulations.” However, if “business as usual” greenhouse gas emissions continue, a high number of polar bear subpopulations will be lost by 2100. That result would “jeopardize the persistence of all but a few high-Arctic subpopulations by 2100,” according to the paper. A moderate reduction in emissions would help, but the paper notes it “is unlikely to prevent some subpopulation extirpations within this century.”

Subpopulations around the globe

Around the globe, different populations of bears face their own unique challenges. In 2008, Amstrup published a paper describing four “Arctic sea ice ecoregions” where polar bears live. These four regions vary in their drift patterns and in the timing and patterns of sea ice freezing and breaking up.

The “Seasonal Ice Ecoregion” (SIE) in central and eastern Canada melts every summer, when the bears go onshore, though there isn’t much food available onshore. The “Divergent Ice Ecoregion” (DIE) stretches from coastal Alaska to Svalbard, and ocean currents carry ice offshore. In summer, there is usually a gap between the ice and shore, but that gap is much larger than it used to be, and the ice can be far from shore. The “Convergent Ice Ecoregion” (CIE) is where much of the ice from the DIE ends up. Bears here can now generally stay on ice all year. The “Archipelago Ecoregion” (AE) covers the far north Canadian Arctic ocean channels, and it is expected to be one of the last areas where polar bears can survive.

They found in 2020 five subpopulations currently have no risk of reproductive failure. But even with moderate mitigation, by 2080 only one of those subpopulations – the Queen Elizabeth Islands subpopulation in the CIE – would still have no risk of reproductive failure.

The other two CIE subpopulations which currently have no risk of reproductive failure, Northern Beaufort Sea and East Greenland, would have “likely” and “very likely” reproductive failure, respectively, without mitigation. Moderate mitigation would reduce the risk of reproductive failure by 2080 to “possible” in the Northern Beaufort Sea and “likely” in East Greenland.

The other two subpopulations currently with no risk of reproductive failure, the Barents Sea and Laptev Sea subpopulations, both in the DIE, would face “inevitable” and “likely” reproductive failure, respectively, by 2080 with business as usual, with “very likely” and “possible” reproductive failure under a moderate mitigation model.

Researchers found the SIE to be one of the places where bears are most at risk, with reproductive failure in the Southern Hudson Bay subpopulation “inevitable” by 2080 without action. “We don’t have a lot of time for the bears in the southern part of the Seasonal Ice Ecoregion,” Amstrup says. “We need to make changes very quickly… the forecast is really not good at all, but we can change it – that’s within human control.”

Scientists have found the relationship between the global temperature and sea ice to be fairly linear, meaning we are not past an unstoppable tipping point. However, Amstrup notes that society can’t stop the impact of greenhouse gas emissions on a dime: Even if emissions were stabilized immediately, it would take a few decades for the sea ice to stabilize.

“Sea ice does stabilize, but about 20 or 30 years after you stabilize greenhouse gas rise,” Amstrup says, emphasizing prompt action is crucial. He points out that if some subpopulations may face reproductive failure in a couple decades as a result of declining sea ice, “we really need to be acting now.”

Polar bears, like canaries in the coal mine, a harbinger

“We can’t wait until 2040 to decide,” Amstrup emphasizes.

While saving polar bears is on its own a key reason to take immediate action on greenhouse gas emissions, Amstrup emphasizes that the bears are just one of many reasons to take action.

“The real takeaway is that these polar bears are messengers for us about what we’re doing to the world, and if we allow polar bears and their habitat to disappear, we will be profoundly changing the global climate … in ways that we haven’t experienced as humans.”

However, Amstrup also notes there are many other reasons to reduce emissions as soon as possible. It’s not just about the polar bears. “If [we] allow the world to continue to warm it clearly doesn’t bode well for polar bears,” he cautions.

“The bigger concern is if we allow those polar bears to get to the point that we’re projecting they will reach by the middle of this century, I don’t think global managers and policy makers are going to have time to be concerned about polar bears, because human climate refugees are going to be the big issue,” Amstrup says. “The future will see far more issues than what we’ve got now, and there’s going to be very little time, I think, to be concerned about polar bears.”