Sea ice loss associated with climate warming is driving increased land use by polar bears. While on land polar bears have been thought to primarily fast and rest while waiting for the return of their preferred habitat, the sea ice upon which they hunt ice-dependent seals (primarily ringed and bearded seals). Projected increases in polar bear use of and time on land have led to predictions that recruitment and adult survival will be significantly diminished across much of the polar bear’s circumpolar range without aggressive mitigation of greenhouse gas emissions. However, polar bears have also been observed feeding on terrestrial foods including berries, waterfowl and seabirds and their eggs, and caribou. Such observations have raised the question of whether terrestrial foods could compensate for reduced access to seals as polar bears are increasingly reliant on summer land use.

In the southern extent of their range, polar bears have long relied on land as the sea ice melts completely in the summer. In the past decade this has kept polar bears on land for about 130 days. Although previous studies have measured changes in body mass and composition of polar bears while summering on land, data has been lacking that simultaneously measures changes in body mass and composition with measures of behavior, activity, movement, diet, and energy expenditure. Such inclusive data are essential to evaluate how consumption of terrestrial foods and differences in behavior might impact the ability of polar bears to survive on land.

Advances in biologging technologies including remote video camera collars with tri-axial accelerometers have enabled new opportunities to continuously monitor the diet, behavior, movement and activity of individual bears. We collected GPS locations of bears every 5 minutes, tri-axial acceleration continuously at 16 Hz, and 5 second video clips every two minutes during daylight hours.

We combined these technological advances with measures of metabolism by tracking dosages of stable isotopes of hydrogen and oxygen. Together this provided a detailed ecophysiological understanding of how individual behavior and diet influence overall energy expenditure and changes in body mass and composition with implications for surviving the summer onshore period. We applied these techniques to twenty polar bears of different sex and ages on land near western Hudson Bay during a three-week period. The western Hudson Bay provided an ideal area to apply these techniques because the bears are concentrated in relatively high densities within Wapusk National Park and Environment and Climate Change Canada has a long-term research program monitoring polar bears in this area.

We found polar bears in this region exhibited diverse responses to summer land use. Approximately one-third of the bears were highly sedentary, resting as much as 98% of time, and moving as little as 15 km over three weeks. Those bears that were largely sedentary fed minimally, but they were able to minimize their energy expenditure and burned calories at rates similar to bears during hibernation.

In comparison, the remaining two-thirds of the bears were more active and moved greater distances, including one individual that moved 330 km (205 miles) over three weeks. Of these more active bears, three made long swims covering 54, 120, and 175 km (34, 75, and 109 miles) in total. As a result, active bears had significantly greater energy demands relative to less active individuals.

For example, the daily number of calories burned by adult male polar bears in our study varied by 18,000 kcal. To put this in perspective, after accounting for differences in body weight, that’s equivalent to an adult male polar bear needing to eat 45 hamburgers per day if it is active, but only 11 hamburgers per day if it is inactive. Active individuals spent more time eating primarily feeding on berries, grasses, kelp, and bird carcasses, with some individuals spending up to 40% of daylight hours feeding on berries.  Although terrestrial foods partially compensated for the higher number of calories active bears used, they did not prevent the bears from losing weight. Ultimately, all the bears, except one individual who found a marine mammal carcass on land, lost about 1 kg (2.2 lbs) per day on average, which highlights that none of these behavior strategies were beneficial for extending the period in which polar bears can survive on land.

Surprisingly, behavioral strategies (active versus inactive) did not appear to be driven by body condition, sex, or age. Instead, they appeared to be driven by individual variation. For example, the largest bear in our study, an adult male weighing 583 kg (1285 lbs) was more than 4 times more active than the other 4 adult males we monitored and swam a total of 54 km (34 miles). This bear burned 22,300 kcal/day.

In comparison the least active adult male weighed 446 kg (983 lbs) and burned only 4,300 kcal/day. Two bears found marine mammal carcasses during their swims (a beluga carcass and a seal carcass) yet they fed from these minimally which suggests that polar bears are unable to feed on large prey in open water.

The results from this study highlight the potential of remote biologging technology to provide a detailed understanding of wildlife ecophysiology and spatial ecology to better understand their ability to respond to climate driven habitat loss.  Our findings indicate that although polar bears exhibit remarkable plasticity in behavior, terrestrial food resources are insufficient to allow polar bears to adapt to longer summers on land and they remain at risk of starvation with forecasted declines in Arctic sea ice. Our next step will be to use these data to predict the effects of continued sea ice loss on the reproduction and survival of specific polar bear populations in different parts of their range.