Pilfold, N.W., A.E. Derocher, I. Stirling, and E. Richardson, 2014, Polar bear predatory behavior reveals seascape distribution of ringed seal lairs, Popul. Ecol. (56) pg. 129-138, doi: 10.1007/s10144-013-0396.
Observing animals in their natural environments is limited in part by how remote the location. For example, studies of the ringed seal (Pusa hispida) (figure 1), particularly their breeding distribution, have been limited to their near-shore habitats. A seven year study by Canadian scientists utilize aerial polar bear (Ursus maritimus) tracking and the well established predator prey relationship between polar bears and ringed seals to collect information about ringed seal pup populations in offshore sea ice environments.
Ringed seal pups are the primary prey for polar bears. It is hypothesized that the population sizes of the two species are correlated. The primary feeding season for polar bears is during the spring when ringed seals reproduce. This leads to higher pup kill rates during years of high natality. Scientists in this study used polar bear tracks from hunting ringed seal pups to infer information about the relationship between changes in ringed seal natality and ringed seal lair distribution.
Ringed seal dens for resting and nursing are beneath the snow surface and are called ‘subnivean lairs’. Lairs are found in areas where ice and snow have accumulated; it is important they are built to last through the end of breeding season. Under the snow the ringed seals are protected from predators and the elements.
Unfortunately, lairs are not completely predator proof. Polar bears find lairs with pups by smell, and then access them by breaking through the roof with digging and force. The hunting attempts on the subnivean liars make a visible impression from the air, so by following the polar bears’ hunting paths, researchers were able to observe the distribution of ringed deal pup lairs in near and offshore environments.
In the eastern Beaufort Sea, within 150 km of the Canadian coast (figure 2), aerial helicopter observations of ringed seal pup kills and attempted hunts were made between April and mid May from 2003 to 2011.
The study area had fast ice near the coast and pack ice offshore. In the snow, the attempts by polar bears to attack seals in their subnivean lairs were identified by disturbed ice and, in some cases, by blood and pieces of seal carcass (figure 3); the locations were recorded with GPS. Lairs were described by location, type of ice, and underlying bathymetry; the Information used for the characterization was collected by charts and satellites. In the cases when carcasses could be recovered, seal age was determined by teeth annuli counts (when available) and species were confirmed by DNA analysis.
Researchers investigated how habitat use during years of low natality compared with use during years of high natality, and applied resource selection functions to model how the probability of predation on ringed seal in lairs varied with habitat changes related to natality.
It was observed that when natality was low the polar bear attacks on pups were concentrated (65%) on the near-shore fast ice, at an average of 36 km from shore (figure 4). In contrast, during periods of high natality pup attacks were observed at an average of 46 km from shore and there were a similar number of attacks on fast ice and pack ice. The model predicted the location with the highest probability of a polar bear attack on a seal pup when natality is low is near shore in fast ice, which agrees with previous studies.
The researchers offer two hypotheses to explain the observations. First, it is possible that high natality forces the utilization of less ideal habitats for lairs to accommodate space with higher breeder density, and second, they speculate that females who give birth in lairs in offshore pack ice are more likely to have the pups get killed.
Perhaps a less obvious significance of this research is that it is an example of how we can access information about the ‘hard to reach’ cryptic species of the world by studying other species with which they are closely ecologically knit.