Atlantic walrus (Odobenus rosmarus rosmarus) COSEWIC assessment and status report: chapter 9

Biology

General

The age of a walrus is determined from the number of growth layers in the cementum of the lower canines (Mansfield 1958; Garlich-Miller et al. 1993; Stewart et al. (ed.) 1993), which in captive walruses corresponds closely to known age (0-15 years; Fay 1982), but age validation studies have not been conducted on wild Atlantic walruses. Over 35 cemental growth layers have been counted in wild Atlantic walruses, suggesting that layering of the cementum occurs throughout the life of an animal (Mansfield 1958). Growth layers in the mandible are not reliable indicators of age in mature walruses and underestimate the ages of males that are more than 19 yrs old and females over 9 yrs, probably due to resorption and slower bone growth (Garlich-Miller et al. 1993).

Male walruses grow larger than females and geographical differences are apparent. The asymptotic standard body length of male Atlantic walruses taken from northern Foxe Basin in 1983-93 (315.2 cm ±3.8 (SE), n=103) was significantly greater (P<0.05) than that of females (276.6 cm ± 3.4, n = 90) (Garlich-Miller and Stewart 1998). The growth patterns reported for standard length were similar to those reported from northwest Greenland (Knutsen and Born 1994). Animals from both populations were significantly larger and reached full maturity older than those sampled from northern Hudson Bay in the 1950s (Mansfield 1958). The Foxe Basin and northwest Greenland walruses grew to lengths similar to that of the Pacific walrus, but did not attain the same body mass. The predictive equation relating log mass (M, kg) to log standard length (SL, cm) for the northern Foxe Basin walruses was: Log10M = - 3.74 + 2.58(Log10SL), n=25, r2=0.98 (Garlich-Miller and Stewart 1998).  The largest intact male weighed about 1100 kg and the largest female about 800 kg (Garlich-Miller 1994).


Reproduction

Walruses are gregarious and polygynous, and mature males compete intensely for females (Le Boeuf 1986; Sjare and Stirling 1996; Fay 1981). Relatively little is known about the reproductive behaviour of the Atlantic walrus, since mating occurs in the water in remote areas from February through April (Born 1990, 2003; Sjare and Stirling 1996). Detailed observations of breeding behaviour were made at a high-Arctic polynya surrounded by fast ice (Sjare and Stirling 1996), where males competed for and defended access to females for up to five days. A female was likely to mate with the male that was attending the herd when she became receptive (Sjare and Stirling 1996). Sea ice stability may be an important determinant of breeding behaviour. The mating system in fast-ice habitat differs from that of Pacific walruses breeding in pack ice, which exhibit a lekking behaviour in which several mature males sing from small defended territories very near a herd. It is not known whether the behaviour of Atlantic walruses breeding in pack ice is more like that observed in Pacific walruses,

A low reproductive rate makes the walrus vulnerable to over-hunting and to environmental perturbations. Female Atlantic walruses mature between 5 and 10 yrs of age. Of 79 females from northern Foxe Basin, all aged 7 yrs or older had ovulated, but not all became pregnant (Garlich-Miller and Stewart 1999). Age at first pregnancy ranged from 5 to 12 years. Males mature between 7 and 13 yrs (Born 2003).

Females give birth on average once every three years (Mansfield 1958; Born 1990; Garlich-Miller and Stewart 1999). Most mature females ovulate every second year (Born 1990; Garlich-Miller and Stewart 1999). However, some females live well past their reproductive prime, so not all adult females in a population are fertile (Reeves 1995). Reported pregnancy rates among fecund females are 0.29 in Foxe Basin, 0.35 in northern Hudson Bay (Mansfield 1958), and 0.37 in Greenland waters (Born 1990). Garlich-Miller and Stewart (1999) found a pregnancy rate of 0.33 and birth rate of 0.30 in northern Foxe Basin.

Implantation of the embryo is delayed, occurring in late June–early July (Born 1990; Garlich-Miller and Stewart 1999). Active gestation lasts about 11 months. In the Thule area of Greenland, young are born between early April and mid-July, but mostly in late May and early June (Born 1990). Most Atlantic walrus pregnancies involve a single foetus but twins have been reported in Pacific walruses (Fay et al. 1991). Mansfield (1973) estimated the gross annual production rate, or proportion of newborns in the population, at 11%. Recent counts at high-Arctic uglit in August, not corrected for sex and age segregation among uglit, suggest a calf production of about 10% (Stewart 2002). Generation time, calculated after Pianka (1988) (i.e. the mean of age at first reproduction and age at last reproduction), would be about (7 + ~35) / 2 i.e. 21 yrs.

Young walruses will suckle for up to 25–27 months (Fisher and Stewart 1997). Female walruses take their calves to sea when they forage for food (Kovacs and Lavigne 1992); the young can nurse in the water (Loughrey 1959; Miller and Boness 1983). Calves moult in their first summer and each summer thereafter (Mansfield 1958). The birth-rate is low, but females, and the herd as a whole, are intensely protective of the young, so calf survival is high relative to that of other pinnipeds.


Survival

Survival rates for Atlantic walruses are unknown. Hunting by humans is the greatest cause of mortality. Survival of the young is probably high, owing to the intense maternal care they receive, although they are vulnerable to trampling when a herd is stampeded. Mortality from predation is probably low, given the animal’s large size, aggressive defence and dangerous tusks. But competitive fighting during the breeding season may increase the natural mortality of males (Fay 1985). Present contaminant burdens are unlikely to affect survival rates. Mortality due to parasites, diseases, and accidents is unknown.

Trampling

Young walruses and those in poor condition are vulnerable to trampling if herds are stampeded onshore or offshore. These stampedes typically result in few deaths. However, in one incident at St. Lawrence Island in the Bering Sea, where at least 537 Pacific walruses died, trampling may have been one cause of death (Fay and Kelly 1980). Some of the animals examined there had been attacked by killer whales (Orcinus orca), which may have stampeded the large herd ashore and caused the death by trampling of many smaller or weaker individuals.

About 400 carcasses also washed ashore from various sources and about 15% of the total mortality consisted of aborted foetuses. The latter likely resulted from physical trauma but an infectious or toxic agent could not be ruled out. Mortality on such a scale has not been reported for Atlantic walruses but stampedes do cause some mortality (Loughrey 1959).

At some ‘rocky’ summer haul-out sites tusk breakage may be a problem if animals startle and stampede into the water (B. Sjare, DFO, pers. com. 2005).

Ice entrapment

Evidence for the entrapment of walruses in ice is scant. Walruses can break ice with their tusks to keep holes open and can climb out onto the ice using their tusks, and large male Pacific walruses can break through ice up to 20 cm thick by ramming it from below with their heavy, dense skulls (Bruemmer 1977). Walruses can travel over the ice for at least 6 km when they are frozen out (Calvert and Stirling 1990), typically in a straight line regardless of obstacles (Freuchen 1921).

Predation

Polar bears (Ursus maritimus) prey on Atlantic walruses; some die in the attempt (Loughrey 1959; Killian and Stirling 1978). Walruses are most vulnerable to bears when they are frozen out of their breathing holes or must rely on a very limited area of open water for breathing and haul-out, particularly where rough ice provides cover for stalking bears (Calvert and Stirling 1990). Sub-adults are more vulnerable than adults, which are aggressive and possess large tusks for defence. While predation rates are unknown, polar bears are important predators of walruses in the central Canadian high Arctic in late winter and early spring.

Killer whales also kill walruses (Lowry et al. 1987), but this does not seem to be common. Greenland Inuit report that killer whales are afraid of walruses and will imitate the sound of an enraged bull to deter the approach of these walruses if they encounter them while hunting offshore (Bruemmer 1977).

Diseases and parasites

The susceptibility of walruses to mortality from disease is not well understood. Fay (1982) reviewed information on parasites and diseases in the Pacific walrus. Serological testing of 210 walruses from sites in the eastern Canadian Arctic did not find antibodies to influenza A virus, which can cause high mortality among seals and has been found in ringed seals (Phoca hispida) and belugas (Delphinapterus leucas) (Nielsen et al. 2001b). If walruses were susceptible to this virus, the absence of seropositive animals could mean that they had not been exposed to the virus or that all of the infected animals had died. There is serological evidence for sporadic infection of walruses in the Igloolik area with a bacterium of a Brucella sp., which can cause reproductive failure in marine mammals (Nielsen et al. 1996, 2001a). Morbillivirus antibodies are common in walruses from the eastern Canadian Arctic (Nielsen et al. 2000), indicating exposure to the phocine distemper virus (PDV) or a related virus. The pathology of all these viruses in walruses is unknown.

Walruses are commonly infected by the helmith parasite Trichinella nativa Britov and Boev 1972, which causes trichinellosis (or trichinosis) in humans (Campbell 1988; Pozio et al. 1992; Serhir et al. 2001).Footnote 2 Outbreaks of trichinosis due to eating uncooked walrus meat are recurrent among Inuit in the eastern Canadian Arctic, with recent outbreaks at Salluit in 1983 (Viallet et al. 1986) and 1987 (MacLean et al. 1992), the Kivalliq Region in 1989-95 (Heinzig 1996), Inukjuak in 1997, Qikiqtarjuaq in 1999 (Serhir et al. 2001), and Repulse Bay in 2002 (Hill 2003). The pathology of this parasite in walruses is unknown.


Anatomy and physiology

Recent studies have described the cranial bones and their significance for hauling out, feeding, and accommodation of the sensory organs (Kastelein and Gerrits 1990); the muscles and their role in feeding and haul-out behaviour (Kastelein et al. 1991); the eyes (Kastelein et al. 1993) and tongue and their functions in walrus ecology (Kastelein et al. 1997); and the ears and their function in aerial and underwater hearing (Kastelein et al. 1996).

Walruses are well adapted to cold and ice. They reduce heat losses during extreme cold by constricting blood flow to their peripheral vascular system and vice versa (Ray and Fay 1968). Their thick skin (2–4 cm) and blubber (1-15 cm; Fay 1985) enable them to sleep on the ice at –31°C with a strong wind blowing (Bruemmer 1977). They huddle together and reduce their exposed skin surface by curling into a “foetal position” when it is cold.

Benrischke (2002) described the placental anatomy and histology of a Pacific walrus. Little is known of the endocrinology of the species.

Lead isotope ratios in walrus teeth are proving useful for differentiating groups of walruses and studying movements (Outridge et al. 1997; Outridge and Stewart 1999; Stern et al. 1999), but whether these differences indicate management stocks, populations, COSEWIC designatable units, or merely feeding groups, is uncertain.


Movements/dispersal

Walruses can travel long distances by swimming or by riding ice floes but their seasonal movements in the Canadian Arctic are poorly known. A walrus tagged in east Greenland was recaptured at Svalbard, a straight-line distance of about 700 km (Born and Gjertz 1993). Wintering areas have been documented within the range of each of the putative populations.

South and East Hudson Bay population

There is no evidence for a concerted movement of walruses into or out of southeastern Hudson Bay. Instead, there are local seasonal movements between the rocky sites where animals haul out during the ice-free period and their wintering areas (Freeman 1964). In both the Belcher and Sleeper archipelagos, walruses are present at the floe edge in winter and move into the islands and onshore as the pack dissipates in summer (Fleming and Newton 2003; Peter Kattuk, Mayor of Sanikiluaq and Zach Novalinga, Sanikiluaq Environmental Committee, pers. comm. 1993). The winter whereabouts of animals that summer along the Ontario coast is unknown, and the question of whether they move between this area and the Belchers is unanswered.

Lead isotope signatures in the teeth suggest that some males have moved between Foxe Basin and eastern Hudson Bay (Stewart et al. 2003), but these isotopic records spanned several years and do not necessarily indicate seasonal movements.

Northern Hudson Bay–Davis Strait population

Seasonal movements of the Northern Hudson Bay–Davis Strait population are poorly understood. Where environmental conditions permit, some animals remain year-round, apparently moving inshore and offshore in response to changes in the ice. Others appear to undertake significant seasonal migrations. Evidence for the extent of these movements is circumstantial, as it is based on local observations. Whether the wintering and migratory animals represent different genetic populations is unknown (Stewart 2002).

Walruses occupy the north side of Chesterfield Inlet in the spring, are absent near the community in summer, and are present in the Chesterfield Inlet–Roes Welcome Sound area in winter (Brice-Bennett 1976; Fleming and Newton 2003). They occur in Wager Bay when ice is minimal, and Inuit indicate that they prefer areas with strong currents. Walruses are common in the Repulse Bay area but less so when the summer ice concentration remains high. Their presence also depends on the strength of the current, which varies each summer. When the current is stronger they sometimes approach within 60 km of Repulse Bay in the fall; they are sometimes seen at the floe edge in winter.

Walruses are present year-round in northern Hudson Bay and western Hudson Strait (Orr and Rebizant 1987). Tagging studies in the mid-1950s at Bencas, Coats, and Southampton Islands, using harpoon-head tags (147 tagged, 4 recaptured), revealed only local movements (Mansfield 1958; Loughrey 1959). However, hunters report seasonal movements in response to changing ice conditions (Orr and Rebizant 1987). Walruses occur off the floe edge along the south and east coasts of Southampton Island and west and southwest coasts of Foxe Peninsula in winter, favour the floating pack ice of Evans Strait and Hudson Strait in late spring and summer, and move ashore to uglit as pack ice dissipates. In the fall they are concentrated at or near uglit on Bencas, Walrus, Coats, Mills, Nottingham, and Salisbury Islands and on western Foxe Peninsula. There is a similar shoreward movement of walruses in the Repulse Bay area in the fall (September); some winter in Roes Welcome Sound (Fleming and Newton 2003).

Inuit from Akulivik and Ivujivik have seen walruses moving northward from Hudson Bay into Hudson Strait in the fall (Figure 4; Reeves 1995; Fleming and Newton 2003). Walruses remain in the Ivujivik area year-round but are seldom seen near Akulivik in summer (Fleming and Newton 2003). Akulik and Pilik Islands, which do not appear on maps, are important sites for these animals. In the early 1990s, Ivujivik hunters would go to Akulik when they did not see walruses elsewhere in winter.

There is a general westward movement of walruses through Ungava Bay and Hudson Strait in summer to Nottingham and Salisbury Islands, with a return movement in the fall (Degerbøl and Freuchen 1935; Loughrey 1959). Currie (1963) described an influx of walruses to the southeast coast of Akpatok Island in Ungava Bay as soon as ice conditions permitted in June or early July, and their subsequent dispersal in late July or August northwest past Cape Hopes Advance into Hudson Strait, with a return migration following the same general route but further offshore the cape in September and October. Smith et al. (1979) observed a large influx of walruses, apparently from Hudson Strait, into the Hall Peninsula area in mid-September. Some walruses are present year-round near Nottingham and Salisbury Islands, where strong currents maintain polynyas through the winter (Kemp 1976; Orr and Rebizant 1987).

The presence of animals far offshore in the pack ice of Davis Strait suggests that some walruses move between southeast Baffin Island and western Greenland, perhaps in response to changing ice conditions (Vibe 1967; Born et al. 1995). The walruses present on West Greenland sea-ice in winter no longer use land haul-outs in West Greenland when the ice disappears from West Greenland waters in summer. Movement of walruses from West Greenland wintering to south-east Baffin summering areas is therefore likely. A female walrus, accompanied by a calf, tagged in West Greenland waters in spring of 2005 was observed in southeast Baffin about a month later (R. Dietz, pers. comm.).

Foxe Basin population

Most seasonal movements of walruses in Foxe Basin are apparently local in response to changing ice conditions (Mansfield 1958; Loughrey 1959). Movements have been observed between summering areas around the islands in northern Foxe Basin, particularly the Spicers, and wintering areas along the floe edge that forms along the north side of Rowley Island and extends southward, parallel to the Melville Peninsula, to about 67°30’N (Loughrey 1959; Orr et al. 1986). There is also some north-south movement by walruses in northern Foxe Basin (Anderson and Garlich-Miller 1994). Recent analyses of lead isotope signatures in the teeth of male walruses harvested by Hall Beach (Stewart et al. 2003) support assertions by Degerbøl and Freuchen (1935) that some animals from this population move to Southampton Island and by Loughrey (1959) that some go to Hudson Strait; they do not indicate that these are seasonal movements. Indeed these animals may travel further south to the Sleeper Islands where hunters from Inukjuak often kill walruses. Significant seasonal movements of animals through Fury and Hecla Strait are thought unlikely (Loughrey 1959; Mansfield 1959; Davis et al. 1980; Garlich-Miller cited in Stewart 2002).

Baffin Bay (High Arctic) population

The substantial migrations of walruses that Freuchen (1921) and Vibe (1950) described, northward in the spring along the west coast of Greenland and southward in the fall along the east coast of Baffin Island, no longer seem to occur (Born et al. 1995). However, some walruses cross from Greenland to Ellesmere Island in the spring and presumably return in fall.

Walruses move westward along Lancaster Sound into the central Canadian Arctic archipelago as the ice edge recedes in spring (Degerbøl and Freuchen 1935; Tuck 1957; Greendale and Brousseau-Greendale 1976). The main migration occurs from mid-June to mid-July, mostly along the north side of Lancaster Sound, although some animals stray deep into Pond, Milne and Admiralty Inlets (Schwartz 1982). Some enter bays and inlets along the south coast of Devon Island; others continue westward into Barrow Strait, north into Wellington Channel, or south into Prince Regent Inlet (Read and Stephansson 1976; Riewe 1976; Davis et al. 1978). They move ashore as ice dissipates. Hunters from Resolute suggest that there is a concerted and brief eastward migration out of the central Canadian high Arctic via Lancaster Sound in the fall (Stewart 2002): an animal tagged in August 1993 at Bathurst Island was killed by Inuit in early June 1994 at Milne Inlet on Baffin Island, about 750 km by sea to the east (Stewart 2002). Another animal using the haul-out in 1993 had wintered near Dundas Island (76°05’N, 94°58’W) (B. Sjare cited in Stewart 2002). There is also a westward influx of walruses from Baffin Bay into Jones Sound in early August (Davis et al. 1978). However, walruses also appear to winter every year in the Cardigan Strait–Fram Sound and Penny Strait–Queens Channel areas, at the Hell Gate and Dundas Island polynyas, and in other areas with small polynyas or rotten ice (Riewe 1976; Davis et al. 1978; Killian and Stirling 1978; Stirling et al. 1983; Sjare and Stirling 1996).

The movements of walruses from this population are being studied with satellite-located radio tags (R.E.A. Stewart, DFO, pers. comm. 2003; see also Nicklen and Lanken 2002).

Nova Scotia–Newfoundland–Gulf of St Lawrence (Maritime) population

No information has been found describing the historical patterns of movement of this population.


Nutrition and interspecific interactions

Atlantic walruses feed mostly on benthic prey at depths of 10 to 80 m (Vibe 1950; Mansfield 1958; Born et al. 2003). Some dives can last 24 minutes (Gjertz et al. 2001). They identify suitable prey using their sensitive whiskers (Loughrey 1959; Fay 1981; Kastelein and van Gaalen 1988; Kastelein et al. 1990). Disturbed bottom sediments suggest that prey are identified by rooting with the snout and then excavated using jets of water from the mouth (Oliver et al. 1983). Bivalves are sucked out from their shells; by retracting and depressing its tongue a walrus can create a vacuum of –87.9 kPa in air, or –118.8 kPa in water (Kastelein et al. 1994, 1997). The walrus has good control over its tongue muscles and over both the intensity and duration of suction.

Walruses feed preferentially on bivalve molluscs but also eat a variety of other species (Degerbøl and Freuchen 1935). The stomachs of Atlantic walruses taken by Inuit in July (n = 105) and September (n = 2) from northern Foxe Basin contained benthic invertebrates, including bivalves, gastropods, holothurians, polychaetes, brachiopods, amphipods, isopods, priapulids, and sea urchins (Fisher 1989; Fisher and Stewart 1997). Vibe (1950) and Mansfield (1958) found similar invertebrate prey in the stomachs of animals they sampled. In July, Mya truncata contributed 81.4% of the total gross energy to the diet and Hiatella arctica 7.5% (Fisher 1989; Fisher and Stewart 1997). Holothurians and the polychaete Nereis sp. contributed 3.5 and 2.8%, respectively. Walruses less than 3 years old consumed mostly milk, but also ate some benthic invertebrates. Seasonal feeding patterns are not well known. The two stomachs collected in September both contained more food than the fullest stomach from July, suggesting that walruses may feed more intensively in the fall. In September, M. truncata contributed 59.9% of the total gross energy to the diet and the bivalve Serripes groenlandicus contributed 37.9%. Males and females have similar diets but the females have a more efficient digestion (Fisher 1989; Fisher et al. 1992).

Walrus stomachs often contain only the feet or siphons of bivalves (Vibe 1950; Mansfield 1958; Fisher 1989; Welch and Martin-Bergmann 1990; Fisher and Stewart 1997), but direct observations in the wild show that they remove most of the soft parts of their bivalve prey (Vibe 1950; Born et al. 2003). This suggests that the viscera are removed and spat out or else very quickly digested.

Over a 97-hr feeding cycle the estimated daily gross energy intake by a 1200 kg male Atlantic walrus in the wild off east Greenland was 214 kJ per kg of body mass (95% CI: 153-275 kJ) (Born et al. 2003). During this period the animal dove 412 times and consumed an average of 53.2 bivalves (SE=5.2, range 34-89, n=10) per dive. This corresponds to the ingestion of 57 kg (95% CI: 41-72 kg) wet weight of bivalve biomass per day. This is higher than the daily food and energy intake reported from studies of captive walruses (Fisher et al. 1992; Kastelein et al. 2000). Feeding rates in captivity--and probably also in the wild--vary with age, sex, reproductive status, and season (Kastelein et al. 2000).

Atlantic walruses are also known to eat ringed seals, bearded seals (Erignathus barbatus), fishes and squids (Vibe 1950; Mansfield 1958). Observations on Pacific walruses suggest that most seal-eating is predation rather than scavenging (Lowry and Fay 1984). Inuit recognize the livers of seal-eating walruses by their ‘cooked’ appearance and avoid eating them (Loughrey 1959; Fleming and Newton 2003), as they contain toxic concentrations of Vitamin A (Bruemmer 1977; Reeves 1995). Atlantic walruses also prey on seabirds (Gjertz 1990; Donaldson et al. 1995) and scavenge dead whales (Degerbøl and Freuchen 1935; Mansfield 1958). Some large bulls will eat young walruses (Degerbøl and Freuchen 1935).

Bearded seals and Pacific walruses compete for clams in some areas (Lowry and Frost 1981); the same is likely true for Atlantic walruses. The presence of walruses tends to drive away ringed seals (Reeves 1995).


Behaviour/adaptability

Walruses hauled out on the land spend most of their time resting, often lying in contact with one another (Salter 1979a; Miller and Boness 1983). This inactivity enables them to maintain high, stable temperatures in their skin and appendages, which may be crucial during the moult, and possibly for the healing of wounds and the survival of young calves (Fay and Ray 1968; Ray and Fay 1968). Walruses are more active in the water; foraging trips can last 72 h between haul-outs (Born et al. 2003).

Conflicts are common at uglit, where animals must gain and defend space to avoid being crowded (Miller 1975, 1976, 1982; Salter 1979a,b; Miller, and Boness 1983), but less so in the water. Large body size and long tusks characterize dominant animals. Tusks are used in threat displays by both sexes and are important in fighting. Females with calves favour the central and seaward areas of the ugli, where the calves are better protected from polar bears (Miller 1982; Miller and Boness 1983). Males in mixed herds tend to occupy the inland locations. In the water, males tend to be separated from females with offspring, possibly owing to differences in food requirements and to time and energy budgets related to nursing (Miller and Boness 1983).

Conflicts observed on the haul-out beaches in summer and fall are minor compared with the serious battles that take place in the breeding season (Sjare and Stirling 1996; B. Sjare, DFO, pers. comm. 2005). Most of the fights in the breeding season occur in the water and go undetected (there have been few studies of breeding behaviour). The proportion of males injured in the breeding season might not be large, but those actively breeding or struggling to establish themselves sustain serious puncture wounds, gashes, loss of eyes and tusk breakage. In addition, breeding males lose a lot of weight through singing almost continuously and not feeding during February, March and April.

Walruses use a wide variety of vocalizations both in and out of the water to communicate threats, submission, and distress and to maintain contact between females and calves (Miller and Boness 1983; Miller 1985; Stirling et al. 1987; Sjare and Stirling 1996; Sjare et al. 2003; Stirling and Thomas 2003). During the breeding season, males vocalize (sing) underwater to communicate dominance and attract females. Sjare and Stirling (1996) described the breeding behaviour of Atlantic walruses at the Dundas Island polynya.

When hunted, animals of both sexes aid the injured. Herds that consist only of adults flee, but those with young will return to protect the calves and to encourage them to escape (Burns 1965 in Miller 1985). Adoption may be widespread and important to Pacific walruses (Fay 1982), but has not been studied in Atlantic walruses.

The ability of Atlantic walruses to re-colonize areas where populations have been depleted is unknown. The rarity of animals along the Atlantic coast of Canada since the maritime population was extirpated suggests that re-colonization is exceedingly slow at best.

Sensitivity to disturbance

The level of response by walruses on land to aircraft overflights depends upon the distance and altitude of approach (Salter 1979a). Walruses on Bathurst Island at Brooman Point (75°31’N, 97°24’W) raised their heads to locate the source of the disturbance when a Bell 206 helicopter was up to 8 km away, oriented toward the sea when it was within 1.3 km and sometimes escaped into the water immediately thereafter.

Fright or interest responses caused by disturbances may impact population dynamics by causing stampedes, interfering with feeding and increasing energy expenditures--particularly among calves, and by masking communications, impairing thermoregulation and increasing stress levels (Stewart et al. (ed.) 1993). Prolonged or repeated disturbances may cause walruses to abandon uglit (Salter 1979a). Their adaptability to non-threatening man-made disturbances, such as ecotourism, is unknown.

Diving behaviour

Gjertz et al. (2001) used dive recorders to study the diving behaviour of 9 male Atlantic walruses at Svalbard. On average the animals spent 56 h in the water followed by 20 h hauled out on land. They dove to a maximum depth of 67 m and stayed submerged up to 24 min. Foraging dives to a mean depth of 22.5 m lasted an average of 6 minutes. Satellite-linked data loggers indicate that walruses can dive to at least 180 m (Stewart and Fay 2001).

While slow and awkward on land, walruses are good swimmers. Their cruising speed seldom exceeds 6 or 8 km/h, but they can accelerate to about 30 km/h for a short time when chased (Bruemmer 1977).

Adaptation to captivity

Walruses have been maintained successfully in captivity for many years (Fisher et al. 1992; Kastelein et al. 1994, 2000; Kastelein 2002; McIntyre 2002). Few captive-born walruses have been raised by their mothers, which are so protective of their young that they tend to spend more time defending the calf than nursing it (Kastelein 2002). Feeding formulae are now available that enable walrus calves to be hand-raised without nutritional deficiencies. The tusks of captive animals wear down on abrasive surfaces in aquaria, and are routinely removed surgically (McIntyre 2002; R.E.A. Stewart, pers. comm. 2003).

Page details

Date modified: