Polar bear attacks are extremely rare says new study — but the data are incomplete

I did considerable research on polar bear attacks for my thriller of a novel EATEN — which many readers are finding a welcome change from the numbers-and-statistics approach of science — and I have to say that a recently published scientific summary of this phenomenon (1880-2014) authored by biologist James Wilder and colleagues left me speechless (Wilder et al. 2017).


with permission, see EATEN cover.

By attempting to generate information that could be assessed with statistical methods, the authors ended up with data that is so skewed and incomplete that it fails to provide a plausible assessment of the risk to humans of attacks by polar bears. In my opinion, acknowledging that well-reported attacks on Europeans (or recorded by them) make up the bulk of the data used in the paper does not adequately address the weakness of the authors’ conclusions that attacks by polar bears are “extremely rare.”

The paper also focuses much attention on the potential for increases in polar bear attacks on humans due to sea ice loss (blamed on global warming) but ignores totally the increased risk stemming from the larger proportion of adult males that now exist in protected populations. Adult males frequently steal the kills of younger bears and in recovering (i.e. growing) populations, relatively more adult males potentially generate more young males that are nutritionally stressed and at risk of attacking humans (see discussion below).

Finally, no supplementary data is provided to show which records of attacks were included in the study, and no information is provided about how to access the database. How is that possible in this day and age?

Much is made in the paper of the negative effect of polar bear attacks on conservation objectives and the perceived increase in attacks associated with recent sea ice loss. These points were picked up by activist organization Polar Bears International (“Save Our Sea Ice!“) in a press release issued yesterday (11 July, pdf here). This has already generated the desired media attention (here and here, likely more to follow, like this) which is predictably focused on predictions of more polar bear attacks on humans due to global warming.

I have a feeling Inuit and other native inhabitants of the Arctic will not be impressed.

Dismissal of indigenous experience

Thousands of years of interactions between indigenous Arctic people and polar bears are dismissed in this paper with this opening statement (my bold):

“Polar bears (Ursus maritimus) have evolved to exploit the biologically productive Arctic sea ice niche by using it as a platform to prey upon marine mammals (Amstrup 2003). Before European exploration, this habitat specialization likely kept them separated from most people, and thus helped reduce human-bear conflicts. However, the extent of human-polar bear interactions began to change in the sixteenth century with the advent of widespread maritime exploration.”

How could any Arctic scholar not know that formerly-abundant Inuit inhabitants of North American and Greenland — as well as other indigenous Arctic peoples across Eurasia — traditionally hunted the same marine mammals as polar bears, using the same sea ice platforms as the bears? Many Inuit still do. Inuit in the past also lived as near to the coast as possible, in communities where marine mammal meat and fat was consumed and cached for future use — all activities potentially attractive to hungry bears (and still problematic for modern native inhabitants of the Arctic).

And how could any Arctic scholar not know that the cut-off date for the beginning of the study (1880) largely marks a time of decimation of Inuit people across the Arctic from disease and other impacts of European occupation, or that during the 20th century many traditionally nomadic Inuit were forced into permanent settlements (Schreiber 2013) which almost certainly changed their interactions with bears?

This means that except for well-reported incidents in the last few decades, virtually all attacks on the people most likely to encounter polar bears are dismissed as insignificant by the authors of this paper: they discount the almost perpetual danger from polar bear attack that Inuit people endured — and still endure in many areas — because those people existed in “relatively low numbers” and then come to the laughable conclusion that polar bear attacks are extremely rare.

Although the Arctic has been inhabited by Indigenous people in relatively low numbers for thousands of years, the first recorded polar bear attack we found dates to 1595 when 2 members of William Barent’s second expedition were reportedly killed and eaten by a polar bear in the Russian Arctic (de Veer 1876).” [pg 2, my bold]

Bottom line: The data used in the paper are seriously skewed and in my opinion this totally invalidates the authors’ conclusions that attacks by polar bear are “extremely rare.”

No access to the database

Because no access to the database is provided (either as a link or supplemental data), it is unclear if Wilder’s accounts include any of the incidents, discussed previously here, of the reports of winter polar bear attacks on the indigenous people who lived on St. Lawrence Island in the Northern Bering Sea (the “Chukchi Sea” polar bear subpopulation) in the late 1800s.

Distinguished 19th century naturalist Edward Nelson (in his recently-found 1877-1881 journals) recounts stories about polar bear attacks in winter (reprinted by the Smithsonian’s Arctic Studies Center):

“The Eskimo of Saint Lawrence Island and the American coast are well supplied with firearms which they use when bear-hunting. In winter, north of the straits, the bears often become thin and very savage from lack of food.

“A number of Eskimo on the Alaskan coast show frightful scars obtained in contests with them in winter. One man, who came on board the Corwin [Nelson’s ship], had the entire skin and flesh torn from one side of his head and face including the eye and ear, yet had escaped and recovered.

One incident was related to me which occurred near Point Hope during the winter of 1880-’81. Men went out from Point Hope during one of the long winter nights to attend to their seal nets, which were set through holes in the ice.

While at work near each other, one of the men heard a bear approaching over the frosty snow, and having no weapon but a small knife, and the bear being between him and the shore, he threw himself upon his back on the ice and waited. The bear came up and for a few moments smelled about the man from head to foot, and finally pressed his cold nose against the man’s lips and nose and sniffed several times; each time the terrified Eskimo held his breath until, as he afterwards said, his lungs nearly burst.

The bear suddenly heard the other man at work, and listening for a moment he started towards him at a gallop, while the man he left sprang to his feet and ran for his life for the village and reached it safely.

At midday, when the sun had risen a little above the horizon, a large party went out to the spot and found the bear finishing his feast upon the other hunter and soon dispatched him. Cases similar to this occur occasionally al1 along the coast where the bear is found in winter.” [my bold]

Chances are the Point Hope incident Nelson described in detail was included in the Wilder database (and this may be why the study begins in 1880) but unlikely that the devastation wrought by the bears on local hunters before that time made the cut. Restricting the incidents included to those with particular details leaves out a huge body of data.

Conservation trumps human life?

This quote from the paper [pg. 3, my bold] sums up the concern about the effects of polar bear attacks on conservation efforts:

“To date, polar bear attacks on humans have been rare. When they do occur, they evoke negative public reaction, often to the detriment of polar bear conservation. In some communities, those negative reactions can persist for decades and result in less social tolerance for polar bears and increased defense kills. Recurrent conflicts not only undermine the well-being of people and wildlife, they also negatively affect local support for conservation. Therefore, the effective management of human-bear conflict is an essential precondition for the coexistence of bears and people across the Arctic.”

While no one wants to see polar bears killed unnecessarily, it could be said that the conservation concerns in this paper are emphasized at the expense of describing the real risk of polar bear attacks on Arctic residents (due to the lack of unreported modern and historic Inuit attacks and because the authors set the threshold for inclusion in the database so high that many reported attacks could not be included).

This paper runs the risk of being interpreted by Inuit and other Arctic residents as saying they have little reason to be afraid of polar bears and if they are attacked, it’s their own fault. If perceived that way, any valid information the paper contains is likely to be dismissed out of hand by the people it was meant to inform.

Demographic change effects ignored

The authors blame declining sea ice for an apparent recent increase in young male bears in poor condition (the age and sex class responsible for the majority of attacks on humans) but make no mention of the well-known phenomenon of food stealing perpetrated by mature males against younger, smaller individuals (Stirling 1974).

Total hunting bans and restricted hunting of polar bears across the Arctic have resulting in growing populations that almost certainly contain more mature males than existed for most of the 20th century. As a consequence, more food-stressed young bears surely appear every year, which means more 1-3 year old males that might be desperate enough to attack humans.

This phenomenon is likely the basis for Ian Stirling’s comment in 1976 about the total ban on hunting in Norway (which I discussed in an essay about human/polar bear conflicts, with references):

“Dr. Stirling felt that complete cessation of hunting, such as exists in Norway, may increase bear-man conflicts. Dr. Reimers replied that the careful harvesting of polar bears was probably desirable, but the total ban now in effect was largely an emotional and political decision rather than a biological one. Last year four bears were killed in self-defense.” [1974 PBSG meeting “Norway – progress reported by [Thor] Larsen”; Anonymous 1976:11]

Blaming lack of sea ice for all food-stressed young bears is an unscientific cop-out.

Overall impression

Not only are the data reported in the Wilder et al. paper irrevocably skewed towards polar bear encounters with Europeans (or recorded by them) but in their zeal to implicate sea ice loss in recent attacks, the authors fail to even mention a biologically valid explanation for the apparent increase in hungry young male polar bears that have attacked people over the last few years. While the goal of improving the deadliness of conflicts reducing deadly attacks on people by polar bears in the Arctic is admirable and necessary, I’m afraid that in the long run, this paper may do more harm than good because so much has been ignored.


Anonymous. 1976. Polar Bears: Proceedings of the 5th meeting of the Polar Bear Specialists Group IUCN/SSC, 3-5 December, 1974, Le Manoir, St. Prex, Switzerland. Gland, Switzerland IUCN. http://pbsg.npolar.no/en/meetings/

Schreiber, D. 2013. Immobilizing polar bears/Inuit: Productivity and Interspecies wildlife management in the Canadian Arctic. Anthropologica 55:157-176. [not available online, contact me for a copy]

Stirling, I. 1974. Midsummer observations on the behavior of wild polar bears (Ursus maritimus). Canadian Journal of Zoology 52: 1191-1198. http://www.nrcresearchpress.com/doi/abs/10.1139/z74-157#.VR2zaOFmwS4


Wilder, J.M., Vongraven, D., Atwood, T., Hansen, B., Jessen, A., Kochnev, A., York, G., Vallender, R., Hedman, D. and Gibbons, M. 2017. Polar bear attacks on humans: implications of a changing climate. Wildlife Society Bulletin, in press. DOI: 10.1002/wsb.783 http://onlinelibrary.wiley.com/doi/10.1002/wsb.783/full

Comments are closed.