Polar bears reign over the sea ice surface covering the top of our planet. They evolved from brown bears (ie, grizzlies) to take advantage of a novel hunting opportunity: capturing the seals that come to the ice surface to rest, bask in the sun and rear young. To this day, polar bears are tied to the sea ice – all spend the majority of their lives on the ice, and in some regions, they are born and die on the ice, probably never setting foot on land.
However, in much of the Arctic, ice melts in summer, restricting their opportunities to catch seals. Furthermore, climate change is raising Arctic temperatures and altering currents, causing a decline in sea ice area and thickness, most noticeably in summer.
In our recent paper in Science, my colleagues and I sought to understand polar bear physiology during this critical season.
We found that bears appear to exhibit a typical mammalian response to food limitation in summer, including a slow, moderate decline in both body temperature and activity. This contrasts with some previous research, which had suggested that bears enter a hibernation-like state in this season to help cope with limited food availability – an ability that may have offered a bit of relief from consequences of summer ice loss.
Our data suggest that it’s unlikely polar bears will be able to physiologically adapt to extended food deprivation linked to the loss of sea ice.
Helicopter darting polar bears
We sampled 30 bears in the Southern Beaufort Sea, one of 19 subpopulations that are spread across every region of the Arctic where sea ice is a dominant feature.
It is thought that roughly two-thirds of polar bears live in areas where the sea ice melts partially or completely during summer, while the remaining third are found where the ice persists. (However, a recent study concluded that if we do not reduce our rate of carbon emissions, over the next century this persistent ice will probably melt seasonally as well.)
In regions where the ice partially melts, bears must follow the retreating ice north or spend several months on shore. We studied bears in both habitats by capturing them via helicopter darting. We fitted them with collars carrying location transmitters and motion sensors, and surgically implanted tiny temperature sensors inside their abdomens.
Bears on shore and ice had similarly low activity in late summer. Bears on the sea ice may have been food-deprived, as the ice retreats to deep water thought to harbor few seals. Bears on shore are usually thought to be food-deprived as well, although in our study area, bears have learned to scavenge the remains of huge bowhead whales left after subsistence harvest by Inuit communities.
By becoming less active, bears can conserve their energy to wait out the summer. However, these savings likely pale in comparison to a hibernation-like state: the amount of each day the bears spent being active dropped from about 25% in spring to 12% to 22% in late summer, whereas bears in winter hibernation are only active for 1% to 2% of their day.
We were surprised to find that bears on the sea ice continued to exhibit high, steady movement rates even as they spent less time being active.
How could that be? Our data suggest that as bears rest on the sea ice, the ice itself continues to drift at a rate similar to the walking of a bear, making it appear as if the resting bear is still in motion.
Not hibernating, just hungry
Bears that spent their summer on the sea ice had data loggers from May to October (except for bear 20529 – more on her in a moment), and over that period, their body temperatures gradually fell by less than a degree Celsius. Many mammals exhibit a similar decline if they go without food for weeks on end.
This contrasts with the abrupt drop in temperature that occurs during winter hibernation, bringing us back to bear 20529. We could not recapture her in October, but we did the following spring. By then, she had two new little cubs, telling us that she had spent the previous winter hibernating in a maternal den, like all pregnant female polar bears. Her temperature logger had records through late January, when the memory became full. Her data illustrate the hibernation state: a steep drop from 37°Celsius to 35°C in the first week of December, followed by strict regulation at this new “setpoint.”
Critically, previous research with captive bears has shown that this new setpoint allows bears to substantially reduce their metabolic rate, and burn much less energy on a daily basis.
Something else surprised us about the data from 20529, as well as from other bears: bouts of body temperature as low as 22°C, generally lasting less than 12 hours. Comparing body temperature with bear location, activity and the ambient temperature, we realized that these bouts seemed to occur only when a polar bear was swimming. They swim a lot, in cold Arctic water (previously, we published the observation of one bear swimming continuously for nine days); but their fur loses 90% of its insulation value when wet, and their body fat is not evenly thick like the blubber of a seal or a whale. As a result, we think that when polar bears swim, they temporarily allow the outermost regions of their body core to cool off, helping to create an internal, insulating shell to protect the most vital organs.
Our data leave us with a new understanding of how polar bears function in the icescape that is their world.
Glimpsing how bears live on the summer sea ice is especially exciting – we knew almost nothing about this going into the project, because of the difficulty of sampling there – we had to use a military icebreaker to carry our crew and helicopters deep into the Arctic to reach them.
These data also tell us that polar bears in summer appear to function as typical, food-limited mammals. Our findings support previous research showing that if bears maintain their energy needs in summer, ice loss will eventually reduce their body condition, survival, and ultimately, abundance.