Reproduction in the female polar bear is similar to that in other ursids. They enter a prolonged estrus between March and June. In the polar basin, the peak of estrus as evidenced by turgidity of the vulva and vaginal discharge seems to be in late April and early May. Ovulation is thought to be induced by coitus (Wimsatt 1963; Ramsay and Dunbrack 1986; Derocher and Stirling 1992). Implantation is delayed until autumn, and total gestation is 195�265 days (Uspenski 1977), although during most of this time, active development of the conceptus is arrested. Young are born by early January (see below), but stay within the shelter of the den until March or early April (Amstrup and Gardner 1994). Litters of two cubs are most common over most of the polar bear range. Litters of three cubs are seen sporadically across the Arctic, and were most commonly reported in the Hudson Bay region (Stirling et al. 1977b; Ramsay and Stirling 1988; Derocher and Stirling 1992). Young bears will stay with their mothers until weaning most commonly in early spring when the cubs are 2.3 years of age. Female polar bears undergo a lactational anestrus and are available to breed again after weaning. Therefore, in most areas, the minimum successful reproductive interval for polar bears is 3 years (see below).
Newborn polar bears have hair, but are blind and weigh only 0.6kg (Blix and Lentfer 1979). The growth of cubs is very rapid, and they may weigh 10�12 kg by the time they emerge from the den in the spring. After leaving the den, the rapid growth continues, and cubs may increase their weight by an order of magnitude between den exit and their first birthday (S. C. Amstrup, unpublished data). Cubs can double their weight between their first and second birthdays. Cubs receive an especially rich milk from their mothers. The milk of polar bears typically has a higher fat content than that of other bears, and in general the milk of bears is richer in fat and protein than the milk of other carnivores (Jenness et al. 1972). Polar bear milk is more similar to that of pinnipeds than it is to milk of most terrestrial mammals (Jenness et al. 1972; Ramsay and Dunbrack 1986). Although polar bears may nurse cubs through their second birthday, some females apparently stop allowing cubs to suckle sometime after their first birthday. The contribution to growth from milk during the second year of life is much lower than during the first year (Arnould and Ramsay 1994). Arnould and Ramsay (1994) noted that fat content begins to decline fairly early in lactation, but the biggest differences are between the first and the second year of the cubs's lives. Mean fat content of milk provided to cubs of the year was 31.2 ± 1.6%, whereas the fat content of milk fed to yearlings was 18.3 ± 2.4%. The energy contribution from milk is a significant contributor to the observed rapid growth of cubs and comes at a significant cost to mother bears (Arnould and Ramsay 1994).
The exact timing of birth may vary across the range of polar bears. Harington (1968) reported births as early as 30 November with a median date of 2 December. Derocher et al. (1992), reported, based on progesterone spikes in the blood of pregnant bears and the implied date of implantation, that births of Hudson Bay bears probably occur from mid-November through mid-December. Messier et al. (1994) suggested that polar bears give birth by 15 December. In contrast, many pregnant female polar bears in the Beaufort Sea did not enter dens until late November or early December (Amstrup and Gardner 1994; S. C. Amstrup, unpublished data). Unless those bears were giving birth immediately on den entry, a later date of birth can be assumed. One captive female in Barrow, Alaska gave birth on 27 December, corroborating that assumption (Blix and Lentfer 1979). Similarly, L�n� (1972) reported that implantation of the conceptus into the uterus of the polar bear began in November, around the peak of den entry in the Beaufort Sea. The timing of implantation, and hence that of birth, is likely dependent on body condition of the female. Condition of the female, in turn, depends on a variety of environmental factors. The interaction between environmental and physiological factors that control births is clearly an area in need of further research.
Testes of male polar bears reside in the abdomen for most of the year. They descend into the scrotum in late winter, and remain there through May. Descent of the testes permits spermatogenesis, which is thought to occur from February to May (Erickson 1962; Lentfer and Miller 1969; L�n� 1970). L�n� (1970) reported that male/female pairs were observed as early as 8 March and as late as 20 June. According to histological examination of testes and ovaries, L�n� (1970) further concluded breeding could last into July. Deteriorating ice conditions preclude scientific observations in most polar bear habitats by June, so the frequency of summer breeding cannot be easily documented.
Lentfer and Miller (1969) concluded, from presence of mature spermatozoa in epididymides, that male polar bears in Alaska may be able to breed as early as 3 years of age. Presence of sperm also guaranteed reproductive capability until at least age 19 years (Lentfer and Miller 1969). A recent study in Greenland found that 2 of 7 two-year old males, 5 of 10 three-year-olds, and 4 of 9 four-year-olds had some spermatazoa in epididymides (Rosing-Asvid et al. 2002). Although spermatazoa occurred at low density in the younger bears, all bears ? 5 years old, except for one very thin individual, had produced abundant spermatozoa and appeared capable of breeding. Lentfer et al. (1980) observed males 3�11 years old in consort with estrous females, confirming at least the age of earliest breeding ability for male polar bears. It should be noted, however, that excessive hunting in Alaska just before and during the time those observations were made had all but eliminated prime males (aged >10 years) from the population (Amstrup et al. 1986). Subsequently, few male bears that young have been observed with females. Since 1980, the proportion of prime males in Alaskan waters has been high (Amstrup 1995). Presently, large males weighing 400�500+ kg are abundant in this region. Three and 4-year-old bears typically weigh 250 kg, and would not be able to compete successfully for mates with the now-abundant large males. Currently, young males must have very low reproductive output despite their apparent reproductive potential.
Productivity of polar bear populations appears to be largely dependent on numbers and productivity of ringed seals. For example, in the Beaufort Sea, ringed seal densities are lower than in some areas of the Canadian High Arctic or Hudson Bay. As a possible consequence, female polar bears in the Beaufort Sea usually do not breed for the first time until they are 5 years of age (Stirling et al. 1976; Lentfer and Hensel 1980). This means they give birth for the first time at age 6. In contrast, across many areas of Canada, females reach maturity at age 4 and produce their first young at age 5 (Stirling et al. 1977b, 1980, 1984; Ramsay and Stirling 1982, 1988; Furnell and Schweinsburg 1984).
Craighead and Mitchell (1982 :527) reported that in grizzly bears "reproductive longevity approximates physical longevity." Female polar bears, on the other hand, may show a reproductive senescence long before the end of their lives. Derocher et al. (1992) calculated an average age of first breeding in the Hudson Bay area of 4.1 years. Productivity, assessed by estimated pregnancy rates, remained high between 5 and 20 years of age and declined thereafter (Derocher et al. 1992). Unfortunately, long-term monitoring of individual polar bears is uncommon and data addressing senescence are few. One 32-year-old female in the Beaufort Sea was monitored for the last 25 years of her life and seen annually during her last 10 years. This bear was in extraordinary condition nearly every autumn. Although she was not recaptured during the autumn of her 30th year, she was observed standing next to a 400-kg female that was captured that season. The 30-year-old female appeared larger, but still did not enter a den that autumn. Despite her apparent excellent physical condition, she last produced cubs at age 22, suggesting a prolonged reproductive senescence. Some contrary evidence also is available. One 29-year-old female in the Beaufort Sea was clearly in estrus (based on turgidity of the vulva) and traveling with an adult male in the spring of 2001. Derocher et al. (1992) also indicated that some females retained reproductive competency throughout life. The reproductive longevity of brown bears and polar bears appears to be fertile ground for further research.
Derocher and Stirling (1994) noted that litter size varied with maternal age, increasing until age 14 years, after which it declined. Heavy hunting reduces numbers of prime-age and older polar bears of both sexes (Amstrup et al. 1986). If such changes occurred without density-dependent increases in reproductive performance for young animals, overharvesting could have the additional population-depressing effect of actually reducing reproduction at low population densities rather than increasing it. Polar bears in the Hudson Bay area were heavily harvested into the 1970s, but numbers there appear to have increased since then (Prevett and Kolenosky 1982; Derocher et al. 1997). Litter size, litter production rate, and other reproductive factors can be expected to change with population size relative to carrying capacity. It also changes in a response to hunting pressure and other population perturbations. Hence, comparisons among populations or within populations over time must take into account the status of the population relative to natural and anthropogenic features of the environment.
In most parts of the Arctic, female polar bears cannot complete a reproductive cycle more frequently than every 3 years. The interbirth interval is determined by the length of time cubs are attended by their mothers, which most commonly is 2.3 years (Stirling et al. 1976, 1980; Lentfer et al. 1980; Amstrup et al. 1986; Amstrup and Durner 1995) (Fig. 27.9). L�n� (1970) concluded that in the Svalbard area, most cubs were weaned by about 17 months of age. Likewise, Ramsay and Stirling (1988) reported that during the 1970s and early 1980s, a significant proportion of female polar bears in the Hudson Bay region weaned their cubs at about 1.3 years of age. After weaning her cubs in the spring of their second year (at age 1.3 years), a female bear could breed again that same spring and achieve a 2-year reproductive interval.
The historically shorter reproductive interval of polar bears living in Hudson Bay (Stirling et al. 1977b) meant that they were more prolific than most other populations of polar bears. Captures of many hundreds of female polar bears and their young in Alaska, Canada, and Svalbard have suggested geographic differences in litter size, litter production, onset of maturity, and reproductive interval. For example, mean litter sizes of cubs and yearlings in Alaska were 1.63 and 1.49, respectively (Amstrup 1995). In Svalbard, these values were 1.81 and 1.32, respectively, whereas litter sizes of polar bears in Hudson Bay during the early 1980s were 1.9 and 1.7 for cubs and yearlings, respectively (Ramsay and Stirling 1988, Derocher and Stirling 1992). Annual litter production rates as high as 0.45 litters/female have been reported for polar bears in the Hudson Bay area (Derocher and Stirling 1992). Nearly half of the females in that population were annually producing a litter of cubs at that time. By comparison, only one fourth of the female polar bears in the Beaufort Sea produce a litter of cubs each year (a litter production rate of 0.25) (Amstrup 1995). That is, in Hudson Bay, each female had a litter nearly every other year, but in the Beaufort Sea, each female produced a litter only every fourth year. Because polar bears in Hudson Bay also produced larger litter sizes, these differences in litter production rates translated into a much higher overall reproductive rate there than in the Beaufort Sea. Female polar bears in the Beaufort Sea produced only 0.40 cub/year, whereas in the Hudson Bay area they produced up to 0.90 cub/year at the time those studies were conducted (Derocher and Stirling 1992). Reproductive rates in most other areas appear to be more similar to those in the Beaufort Sea than in Hudson Bay.
FIGURE 27.9. The 3-year reproductive cycle typical of polar bears throughout most of their range. Exact timing of birth, shown here as 1 January, is not well known and may vary geographically from November to January. Within geographic regions, the timing of birth also may depend on the condition of the female on den entry. SOURCE: Modified from L�n� (1970). Click image to enlarge.
In assessing reproductive intervals, it is critical to confirm weaning, as opposed to mortality of cubs. Many polar bear cubs die in their first year of life (Amstrup and Durner 1995). Those females can breed again in the year of the loss (if it occurs early enough in the spring) or the next year. The breeding frequency, by itself, might suggest a short reproductive interval when it is actually prolonged by poor cub survival. In addition to documenting that tagged females were no longer accompanied by yearling cubs in the spring, Ramsay and Stirling (1988) also captured many weaned yearlings in the autumn of their second year (approximately 1.8 years of age), confirming that many females in the Hudson Bay region actually did have a 2-year reproductive interval.
L�n� (1970), Stirling et al. (1977b), and Ramsay and Stirling (1988) reported on populations that may have been well below carrying capacity due to unregulated hunting (Stirling et al. 1977b; Larsen 1986; Derocher and Stirling 1992, 1995a). Likewise, breeding intervals in the Hudson Bay area have increased, possibly in response to increased relative density of bears in the area (Derocher and Stirling 1992, 1995b). Annual litter production rate in the Hudson Bay region declined from 0.45 litter/female in the period from 1965�1979 to 0.35 during 1985� 1990 (Derocher and Stirling 1992). A higher proportion produced cubs every 3 years in the latter period. The inverse of the litter production rate is the interbirth interval. That increased from 2.22 years in 1965� 1979 to 2.86 years in 1985�1990. Simultaneously, cub mortality from spring to autumn was significantly higher in the latter period (Derocher and Stirling 1992). The proximate factor associated with all of these trends was the declining weight of adult females during this 25-year period (Derocher and Stirling 1992). Age of first successful reproductive effort increased, although pregnancy rates did not change noticeably. An increasing age of maturation may indicate that a population is approaching carrying capacity. Age of maturation in mammals is often associated with attainment of a threshold body mass (Sadleir 1969) which could be more difficult to attain as competition for resources increases.
A delay in reaching that threshold mass may signal densitydependent influences on the population. Such influences, however, also could result from environmental changes that reduce carrying capacity rather than from increases in polar bear numbers. The documented declines in body weights of females, declines in numbers of independent yearlings, and protracted reproductive intervals appear to be closely related to earlier deterioration of the sea-ice of Hudson Bay (Stirling et al. 1999). The sea-ice extent in the Arctic has been declining throughout the past two decades (Gloersen and Campbell 1991; Vinnikov et al. 1999). Declining Arctic sea-ice cover by itself is difficult to link with polar bear reproductive performance. The timing of melt of the sea-ice in Hudson Bay, however, is more easily connected. Polar bears there, especially pregnant females, depend heavily on the spring and early summer foraging for seals to carry them through the ice-free period (late summer to autumn). Pregnant females, unlike other polar bears in Hudson Bay, remain ashore in autumn when ice returns, and may be food deprived for up to 8 months. Those females must secure sufficient fat stores during the spring and summer to see them through that long period of food deprivation (Stirling 1977; Derocher and Stirling 1992). The mean date of sea-ice break-up in the late 1990s was >2 weeks earlier than it was in the 1970s and early 1980s (Stirling et al. 1999). Earlier break-up and the shortened foraging period accompanying it may mean a significant reduction in the fat stores female polar bears can accumulate before denning. This hypothesis is strengthened by the observation of a transient increase in condition of females coming ashore during the early 1990s when cooler than normal temperatures resulted in later break-ups (Stirling et al. 1999).
Evidence of the critical link between availability of seal prey and reproduction in polar bears is also available in more northerly parts of the range. Weights of females and their reproductive output in the Beaufort Sea decreased markedly in the mid-1970s following a decline in ringed and bearded seal populations (Stirling et al. 1976, 1977b; Kingsley 1979; DeMaster et al. 1980; Stirling et al. 1982; Amstrup et al. 1986). The strength and longevity of declines in reproductive parameters varied both geographically and temporally with the severity of ice conditions that reduced numbers and productivity of seals (Amstrup et al. 1986).
|
