Dr. Ian Stirling recording observations of polar bears

Photo: © Ian Stirling

Dr. Ian Stirling observing polar bears outside on a warm day in early summer at Cape Liddon in Canada's High Arctic.

The Amazing Breeding Behavior of Polar Bears

By Dr. Ian Stirling



15 Mar 2019

With the exception of family groups, most polar bears on the sea ice are solitary, traveling extensively over home ranges that may vary in size from a few thousand to hundreds of thousands of square kilometers. This makes extended observations of the undisturbed behavior of wild polar bears difficult to obtain but extremely valuable when opportunities arise.

For several years, beginning in 1973, I led a study to observe the undisturbed behavior of wild polar bears—hunting, interacting, sleeping, and otherwise just being bears. Collectively, scientists were already accumulating vast amounts of information on polar bear movements by observing where tagged polar bears were recaptured relative to where they were first captured and, later, from the use of satellite radio collars. Similarly, we obtained a great deal of new information from body measurements and specimens collected from bears while they were briefly immobilized for tagging.

However, I felt it was still critical to understand what polar bears spent their time doing when they were not disturbed by scientists in helicopters or Inuit hunters. In other words, I wanted to just let the bears show us, at their own speed, what it meant to be a wild undisturbed bear on the sea in the Arctic.

To do this, we used telescopes from cliff-top observation cabins at Radstock Bay, Devon Island, in the Canadian High Arctic that were high enough that the bears were unaware of our presence. Because of 24-hour daylight in the late spring and early summer, we could watch the bears continuously. Often, we had to take turns between watching inside (and sometimes out in the cold) and sleeping in an unheated tent outside the observation hut. We recorded observations of individual bears non-stop, as long as they didn’t step behind distant land forms or have their visibility obscured by fog.

A cliff-side observation cabin and sleeping tent on Cape Liddon, Devon Island, in spring.

Photo: © Ian Stirling

A cliff-side observation cabin and sleeping tent on Cape Liddon, Devon Island, in spring

In the spring of 1997, we were extremely fortunate to have the unique and rare chance to witness the complete behavioral sequence of a mating pair of polar bears. Prior to what we learned from our direct observations, it was generally thought that the breeding season was restricted to a period from about early April through the middle of May. 

Superficially, one might think the breeding behavior would be quite straightforward: the bears simply have to meet up out on the sea ice at the right time, mate, and sometime later cubs are born. In reality, however, the behaviors and physiological adaptations are highly evolved and definitely not straightforward.

From what we were fortunate enough to learn in 1997, we were able to understand brief observations of bears on the sea ice in Svalbard which enabled me to confirm that breeding behavior continues from the spring through June, and possibly longer.

Finding mates

For starters, most adult females keep their cubs with them for 2 ½ years. Thus, on average, in most years, only about a third of the adult females are even available for breeding. Consequently, adult male polar bears are tasked with locating females that might be available to breed, but range at low densities over vast areas of sea ice. 

Thus, males may travel long distances in search of a mate, pausing only to carefully sniff the tracks of any bears they cross in their travels. This is because adult females appear to have glands in their paws that leave a chemical scent capable of informing an adult male that she is potentially available for breeding. When a male detects such a track, he immediately starts following the female until he catches up. 

Building trust

However, polar bears don’t just start mating soon after they first meet. Remember that individual adult bears of both sexes are widely dispersed and travel and hunt independently over vast areas on the sea ice most of the time. Thus, female polar bears don’t ovulate spontaneously as the females of many mammals, including humans, do, because, if they did and there was no suitable male close by, the egg would simply die and be wasted. Thus, the females have what is known as induced ovulation which means they will not actually release an egg to be fertilized until they have been accompanied by an adult male for long enough, and with sufficient interaction between them, to stimulate release of the egg at a time when being fertilized is highly likely. 

Even after an adult male has located a potential reproductive female, both still have to overcome their biggest behavioral barriers: fear and aggression. Under normal circumstances all adult females, and especially those accompanied by cubs and yearlings, flee from any adult male they meet because of the risk he will kill and eat the cubs. Sometimes an adult male will even kill and eat the female herself, as shown in the photo below.

Dr. Ian Stirling examining female polar bear cannibalized carcass

Photo: © Ian Stirling

Dr. Ian Stirling examines an adult female polar bear that was killed and cannibalized by an adult male.

Not surprisingly, the behavioral process necessary to allow sufficient mutual trust to develop so mating can follow is protracted. The first thing a male does after catching up with a lone female in spring is to try to herd her to a location where the chance of encountering another potential competing male is lower, such as an area of rough sea ice, a small isolated bay, the side of a hill on land, or an island. Then, he tries to keep the female in the same restricted area for about a week, while they interact with each other constantly, pausing only to sleep for 7-8 hours each day.

At first, the pair alternately run toward or away from each other, while still being careful not to become too widely separated. They do little or no feeding during this period. As the days pass, the female slowly becomes more trusting of the male and allows him to be closer to her. In return, his behavior becomes progressively less threatening until finally they start simply standing near each other, following each other back and forth, until she finally allows him to make non-threatening physical contact.

The female may reciprocate by initiating non-threatening physical contact with the male. When you remember that normally a female would not allow a male to be anywhere near her for fear of possibly being killed, the development of sufficient trust on her part to allow his intimately close presence represents an essential, but huge, physiological and behavioral adjustment, even if only for long enough to facilitate mating to occur. The male also undergoes a similarly large behavioral change that allows him to behave in a sufficiently unthreatening manner to allow mating to take place.

Mating behavior

Once sufficient trust has been established, after a week or so of interactions, the female will allow the male to mount. Mating then carries on for several periods, sometimes in excess of two hours at a time, for several days in a row. This protracted process appears to facilitate circumstances that are reliable enough physiologically to allow ovulation to occur and the egg to be fertilized.

After mating ceases, the male and female still remain together for an additional day or two, exhibiting tolerant, mellow behavior toward each other that is totally unlike their behavior at all other times of year. These behaviors may include slowly walking together in no particular direction, lying near each other in the snow, or separating, sometimes for as much as a kilometer or more, and then coming back together again, after which the male may again touch the female’s neck or chest with his nose.

Male polar bear nuzzling female

Photo: © Mick Brown

An adult female polar bear showing no fear response while being nuzzled by an adult male.

Sometimes, they will simply lie in the snow together (but not touching). Such behavior may give the impression the bears may have actually developed a longer-term attachment to each other. However, it is only a temporary behavioral and physiological adaptation, critical for reproduction to occur, but short-lived. Finally, and unceremoniously, they simply diverge and walk away in different directions.

Repeating the process

Once alone again, the male then begins to search for another female. If he finds one, the process starts all over again. Since only one complete behavioral sequence has ever been recorded in detail, we don’t know if all mating events require a similar commitment of time though it appears from partial observations that they probably do. Considering how long it may take for a male to locate a potential adult female, and then behaviorally engage through the amount of interaction time required before mating may occur, the maximum number of matings a male could undertake in a single year is probably in the range of 4-6.

Because the intensity of competition between adult males is significant and, since larger males are more likely to out compete smaller ones for mating privileges, through natural selection over the longer term, adult males have evolved to be roughly double the size of females.

Until recently, the polar bear’s breeding season was generally thought to take place between early April and mid-May.  Probably, a significant contributing factor to determining the mating period was that little hunting by Inuit, or field research by scientists, takes place between late May and the open water season in summer. However, with the advent of ecotourism in Arctic marine areas, particularly in Svalbard, ecotourism ships now regularly enter polar bear habitat from early April through the break-up of the sea ice in early summer. As a result, some of the natural history guides have reliably documented breeding behavior of polar bears through the end of June. Although it is likely that the frequency of mating behavior is significantly reduced by that time, the confirmation of breeding behavior over such a long period is an important revelation. 

Because actual mating constitutes such a small part of the total behavioral breeding sequence (≤ 2% of the total time), most casual observers seeing two bears together at distance and behaving as described above, simply did not realize they were observing some part of the sequence of normal mating behavior. Many likely assumed that two polar bears of different size walking about or interacting together were a mother and cub (or were simply confused). However, now that we understand more about the duration, variability, and the overall pattern of the behavioral sequences of “mating pairs,” it has been possible to confirm the full period over which breeding of polar bears in the wild may occur.

Female polar bear follows male after breeding

Photo: © Mick Brown

An adult female follows an adult male after breeding is completed. In the past, such pairs were sometimes mistaken for a mother with a cub.

Delayed implantation

The evolutionary adaptations of polar bears to facilitate successful reproduction are not limited simply to ensuring copulation.  After the female has completed mating, the fertilized egg undergoes a small number of divisions before entering a state of dormancy known as delayed implantation. This means that the fertilized egg develops no further but sits in the uterus in a dormant state until autumn when the female’s body “decides” in about October whether it has enough fat stored to be able to have cubs. If the answer is yes, the fertilized egg(s) will then implant and growth of the cubs begins, with birth of the tiny cubs taking place only a couple of months later in about December. If the answer is no, the fertilized egg is either shed or reabsorbed but we don’t know which at the moment.

There are likely many more almost unbelievable adaptations of polar bears to their polar environment that have yet to be understood.  Some will require largely opportunistic extended observations of undisturbed polar bears in wild situations before their significance can be appreciated. However, the more we are able to observe the behavior of undisturbed wild polar bears, the more we will come to understand their incredible adaptations to life in one of the most difficult environments in the world.

Dr. Ian Stirling is a member of the Scientific Advisory Council for Polar Bears International. He is also Research Scientist Emeritus for Environment and Climate Change Canada and Adjunct Professor of the Department of Biological Sciences, University of Alberta.


I am particularly thankful for the long-term support of my behavioral studies of wild free-ranging polar bears from: Department of Biological Sciences at the University of Alberta, Environment Canada, Natural Sciences and Engineering Research Council of Canada, and the Polar Continental Shelf Project. I am also grateful for the opportunities to observe undisturbed polar bears in the pack ice while working for ecotourism companies: Oceanwide Expeditions, One Ocean Expeditions, and Quark Expeditions. I also thank Dennis Andriashek and Cheryl Spencer for their invaluable assistance observing and analyzing the field observations on reproductive behavior.

References for those who might wish to follow up the subject in more detail:

Derocher, A., M. Anderson, Ø. Wiig and J. Aars. 2010. Sexual dimorphism and the mating ecology of polar bears (Ursus maritimus) at Svalbard. Behavioral Ecology and Sociobiology 64:939–946.

Smith, T. G., and J. Aars. 2015. Polar bears (Ursus maritimus) mating during late June on the pack ice of northern Svalbard, Norway. Polar Research 34:25786. (Available here.)  

Stirling, I., Spencer, C., Andriashek, D. 2016. Behavior and activity budgets of wild breeding polar bears (Ursus maritimus). Marine Mammal Science 32: 13-37. (Available here.)

Wiig, Ø., I. Gjertz, R. Hansson and J. Thomassen. 1992. Breeding behaviour of polar bears in Hornsund, Svalbard. Polar Record 28:157–159.