A large owl, similar in size to the Great Horned Owl (Bubo virginianus). Definitive plumage largely white, more or less barred and spotted with dusky blackish-brown. Ear tufts indistinct or rudimentary. Eyes rather small (for an owl) with conspicuous golden irises. Feathers nearly conceal bill. Toes and claws partly concealed by thick feathers. Adult males generally whitest overall, sometimes lacking barring altogether. Adult female and first-year male difficult to distinguish, and there is some plumage overlap. First-year females are the most heavily barred, usually having barred upper breasts and crowns. Ageing/sexing generally not safely accomplished in the field. But in-hand examination of flight feather molt patterns can help determine age classes.

Nearly unmistakable if seen well. Snowy Owls are the only all-white or mostly white large owls with dark barring. Perhaps confusable with the pale Arctic form of Great Horned Owl, but not that species prominent ear tufts; Snowy Owl lacks ear tufts. Frequently confused with Barn Owl by beginning birders, but Barn Owls have a heart-shaped face and buffy orange upperparts.

Breeding Range

Fig. 1 . In North America in the w. Aleutians (Attu, Buldir), Hall Island (Bering Sea), and from n. Alaska and throughout the Canadian Arctic Islands south to coastal w. Alaska (Hooper Bay), n. Yukon, n. Mackenzie, s. Keewatin, ne. Manitoba (Churchill), n. Quebec and n. Labrador.

Winter Range

Fig. 1 . From breeding range in North America south to s. Canada and n. United States, sporadically as far south as central California and the Gulf states from Texas to Florida. Also recorded in Bermuda. (See Populations: population status.)

Breeding Range

From n. Greenland, n. Scandinavia, n. Russia, s. Novaya Zemlya, and n. Siberia south to the limits of tundra in Eurasia and the Commander Islands. Rarely in British Isles. (See Cramp 1985for details).

Winter Range

Irregularly from breeding range south to Iceland, British Isles, n. continental Europe, central Russia, n. China, and Sakhalin. Accidental in the Azores, Mediterranean region, Iran, nw. India and Japan.

An owl from the late Miocene of Italy, the size of the modern Snowy Owl and provisionally named Strix perpasta, may represent an extinct genus (Olson 1985); not clear if it is related to the Snowy Owl. Nyctea scandiaca remains have been recovered from prehistoric Pleistocene deposits in North America in Alaska (Cape Prince of Wales), Little Kiska Island, St. Lawrence Island, and Illinois (Brodkorb 1971); also from European prehistoric sites: England, Spain, France, Italy, Germany, Czechoslovakia, Poland, Austria, and Hungary. An extinct, somewhat stronger and more robust Snowy Owl, with 2% longer and 6% thicker long bones than the present-day owl, is known only from glacial deposits in s. France; described by C. Mourer-Chauviré as Nyctea scandiaca gallica (Voous 1989).

Paleolithic people of Europe used owls as subjects in their cave art. The outline of a pair of Snowy Owls and their chicks was etched into the rock face of a cave, Les Trois Frères, Ariège, France, at a time when the Arctic climate extended farther south than today and rendered much of France suitable as a breeding ground (Burton 1973). Evidently the Snowy Owl is the oldest bird species recognizable in prehistoric cave art (P. Stettenheim pers. comm.).

Monotypic. No designated subspecies, probably because of nomadic habits and dispersal over broad areas (see Demography and Populations: range).

Formerly in genus Nyctea, based on plumage, weak osteological (skull) characters (Ford 1967), and DNA (Sibley and Ahlquist 1990). Now, on the basis of new genetic evidence (Wink and Heidrich 1999), considered closely related to Bubo and nested in that genus (Banks et al. 2003).

Not predictable. Migratory movements relate in ways not fully understood to the abundance of prey species, thought by some researchers to vary considerably from region to region across the polar tundra.

The classic explanation of the species’ irruptions implied that the owl’s principal prey, lemmings, crashed along a broad tundra front following a four-year peak, forcing the owls to wander in search of prey. The fact that many Snowy Owls appear in s. Canada usually one winter in four suggests that these irruptions are linked to the lemming cycle, although admittedly the flights sometimes occur in two successive winters, the second involving smaller numbers (Fig. 3; Godfrey 1986).

Recent observations on wintering Snowy Owls challenge the implication of the classic hypothesis that a coordinated lemming-owl cycle is geographically synchronous continent-wide. Although these owls show unpredictable periodic irruptions in eastern and western North America, a large majority appear to be regular migrants over much of their winter range, notably in a central zone including the northern Great Plains where they are especially abundant (Figs. 1, 3; Kerlinger et al. 1985). Lacking hard evidence that lemmings fluctuate synchronously over vast areas (Maher 1970), Kerlinger and associates believed that the microtines occupied a mosaic of tundra patches varying in size. Owl dispersion even in the favored winter midlands is patchy, and the abundance of owls within patches may vary markedly between seasons. Dispersion is thus thought to be linked to local availability of their primary prey (Kerlinger and Lein 1988b).

Looking at the geographic dispersion of lemming abundance and Snowy Owl breeding during any given year yields a mosaic. Peak numbers of varying lemmings (Dicrostonyx groenlandicus) at high latitudes in North America appear to be far less than peak numbers recorded at lower latitudes where both the varying lemming and brown lemming (Lemmus trimucronatus) occur in the same areas, or in those areas where extraordinary numbers of only brown lemmings occur at times (Parmelee 1972). Also, lemming fluctuations appear to be less regular at the higher latitudes where owl clutches average smaller, judging by a long-term study on Bathurst Island at 76°N in the Canadian Arctic. Compounding the problem of synchronizing owl breeding with lemming peaks is the fact that at least a few owls may breed during periods when lemmings are only moderately abundant. Nevertheless, the crux of the phenomenon is a mobile breeding population of owls that move nomadically, breeding where and when their prey is abundant.

Not so easily explained are the large, geographically synchronous winter irruptions that are observed in eastern and western North America (Fig. 3). Breeding areas for the birds involved in these irruptions must necessarily be on the order of millions of hectares, far greater than can be attributed to lemming mosaics (Kerlinger et al. 1985). Other factors, possibly weather, i.e., the relationship between snowfall and temperature conditions on the breeding grounds, could be important. Herein lies a crucial, little known area in the species’ makeup that merits additional research.

Apparently do not wander aimlessly on the winter grounds, at least during regular migrations to the Great Plains. Many that winter in s. Alberta defend territories (Boxall and Lein 1982b), as they do in Wisconsin (Keith 1964); winter site fidelity demonstrated through banding of individuals (Oeming 1957, Follen and Leupke 1980).

Sex and age classes overlap little on winter range. On average, immature males winter farthest south, adult females farthest north, with adult males and immature females in between (Kerlinger and Lein 1986). This arrangement probably due to “social dominance;” observations suggest females dominant over males, mediated by territoriality. Primarily first-year birds (nonbreeders) occupied areas subject to irruptive movements east and west of the northern Great Plains.

No information.

Breeding Range

Open tundra from near tree line to the edge of the polar seas. Nests from near sea level to inland mountain slopes, but usually much lower than 300 m elevation as reported by Murie (1929), except in Norway where lemmings occur only on mountains at 1,000+ m (Watson 1957). In North America, the owls prefer high rolling tundra with numerous promontories which serve as perches and nest sites. In low-arctic habitats, the dominant vegetational type is often dense, hummocky dwarf shrub meadow. In high-arctic habitats, the dominant upland plants (willow, saxifrage, heather, lichens) are low, spreading, and often widely dispersed over extensive areas commonly referred to as desert tundra. Owls also use lowland salt grass meadows and poorly drained freshwater wet meadows, especially for hunting.

Spring And Fall Migration

No specific habitat described although the owls are widely believed to prefer open to wooded areas.

Winter Range

Prime winter habitat in the northern Great Plains closely matches the species’ traditional tundra habitat and prey species (Kerlinger et al. 1985). Usually perch on the ground or slight rises, sometimes on buildings, telephone poles, fence posts, or almost any prominent structure designed by humans. Frequently seen in settlements—from small northern villages to the heart of metropolitan centers in the south. Also in marshes and on dunes.

Main Foods Taken

Usually mammals, from small rodents to large hares; also birds, ranging from small songbird nestlings to medium-sized geese. Fish and other small aquatic animals less often. Probably opportunistic feeders, but likely specialize when local prey are abundant.

Microhabitat For Foraging

In arctic, areas occupied by rodents, notably lemmings. Sites are conspicuous, pockmarked by the holes of lemming burrows which in summer are excavated in turf (often beneath rocks); in winter, through layers of snow. At openings where burrows come to the surface, lemmings are especially vulnerable to the owls. In desert tundra where turf is scarce, the vegetation with its lemming burrows is concentrated at hummocky or rocky prominences. Owls using these sites in turn fertilize the surrounding nitrogen-poor soil. In time these sites become luxuriant green spots in an otherwise austere landscape (Parmelee et al. 1967)—important microhabitats not only for plants, lemmings and owls, but also for weasels, foxes, and the natives who trap them. Not known if the owls ever maintain a lemming diet during a polar winter.

On winter grounds, the owls select habitats where prey is most available (Boxall and Lein 1982a). Likewise on the breeding grounds, the owls choose areas with high lemming densities. Owl density varies not only in different valleys on Baffin Island, but also in different parts of one valley (Watson 1957), as well as other areas in the Canadian Arctic (DFP).

Food Capture And Consumption

Appear to have extraordinary vision in locating and catching prey. Equally adept at hearing, judging by birds’ ability to locate lemmings in turf and snow. These attributes need to be tested. Virtually nothing is known about their sense of smell, or other means they may have in locating prey. Commonly pursue prey through flight and capture in the air, on the ground, and on the water. Also pounce on prey from a perch, or while standing or walking. Several times Audubon (1840) observed a Snowy Owl catching fish while lying lengthwise belly down on a rock beside a water hole. Small fish were devoured near the hole, larger ones carried off.

Of 51 hunts observed in s. Alberta, “sit-and-wait” technique was witnessed in 50, hovering only once; overall success rate (capture) 43%, adult females significantly more successful than juvenile females (Boxall and Lein 1982a). Juveniles also inept at handling prey. Watson (1957) frequently observed hovering on the breeding grounds and concluded that breeding males have no difficulty catching prey when the latter are abundant.

On Baffin Island, Snowy Owls hunt in all weather and at all hours during the continuous summer light, although they are seemingly less active about noon and midnight (Watson 1957). On Greenland, they hunt mostly in early morning and late afternoon (Manniche 1910). More observations on foraging times are needed for both breeding and nonbreeding periods. Not known if these owls hunt at night, or even by moonlight, during winter darkness.

Hungry owls eat lemmings and mice head first, swallowing them at one gulp (Watson 1957). After a few have been eaten whole, others are carefully picked and eaten in small pieces, starting at the head. Commonly before eating, owls look at prey, raising the head higher and standing high on the feet before making the first bite. After feeding, both sexes sometimes wipe their bills and faces. Snow is used often in cleaning (P. Kerlinger pers. comm.).

Major Food Items

Lemmings (varying and/or brown), where and when abundant in the Arctic (Gabrielson and Lincoln 1959). At times they appear to be exploited exclusively (Sutton and Parmelee 1956, Watson 1957), or nearly so (Parmelee et al. 1967). Murie (1929), however, stated that this owl’s food varied with the character of the nesting ground, rodents being taken almost exclusively in the uplands while waterfowl were taken to some extent in the marshy areas. Seven nesting pairs observed by Dufresne (1922) suddenly switched from rodents to ptarmigan when the latter became abundant after hatch.

When lemmings are scarce or absent, a wide variety of prey is consumed. During a nine-year study in the Shetland Islands, rabbits were the preferred prey, but during rabbit lows waders and other birds formed a major part of the diet (Robinson and Becker 1986). Nonbreeding individuals may subsist in areas that lack mammals entirely. On Agattu Island in the Aleutians, pellet analysis showed that Ancient Murrelets (Synthliboramphus antiquus) composed 68% of the diet, other alcids 6.5%, and ducks 13.4% (Williams and Frank 1979).

Quantitative Analysis

On the wintering grounds, a wide variety of live prey, also the remains of dead. In sw. British Columbia, grebes and ducks were 80% of the owls’ diet with the Horned Grebe (Podiceps auritus; captured on the water) most prevalent (Campbell and MacColl 1978). Prey selected by size within a range of 400–800 g body weight. Where mammals abundant in Maine, rats and mice comprised 35%, snowshoe hares (americanus) 20%, and passerine birds only 10% of diet (87 stomach samples; Mendall 1944).

In s. Alberta, deer mice (Peromyscus maniculatus) were the most abundant species in 100 pellets collected during 1976–78 (Boxall and Lein 1982a). Meadow voles (Microtus pennsylvanicus) were only half as numerous, but the two rodents comprised about 90% and 75% of the total prey individuals. Lesser prey included ground squirrels, weasels, jackrabbits, a few passerines, and partridge. Biomass calculations, however, showed that mice and voles together formed only 62% of the calculated prey biomass in 1976–77 and only 28% in 1977–78. Richardson ground squirrels (Spermophilus richardsonii) formed 14% and 28% of calculated prey biomass in the two seasons respectively, while during one winter Gray Partridge (Perdix perdix) alone comprised 33%. Also in Alberta, Kerlinger (pers. comm.) observed that owls were abundant where partridge were abundant. Dietary differences between seasons were thought to be related to differences in weather.

Males prey almost exclusively on rodents, females on a wider range of prey, including larger species (Boxall and Lein 1982a). Diet of different age cohorts not to differ greatly.

Lemming density in the most representative type of tundra at Barrow, AK, varied from one per 2.47 ha to 42 per 2.47 ha (or 0.04 to 17.0 lemmings per ha) three years later (Thompson 1955). During a peak lemming year on Baffin Island, Watson (1957) found the proportion of lemmings eaten to be far less than might be imagined. Taking into account that peak lemming populations increase from early to late summer, the daily intake of three adults was less than 0.1% of the lemming population. Even in August, 3 adults plus 15 young still consumed less than 0.2% of the lemming population daily.

Watson (1957) estimated that adults eat 3 to 5 full-grown lemmings daily, or the equivalent of 600 to 1,600 lemmings per year. This amounts to 150–350 g per day or 55–130 kg per year. From early May to early Sep, a pair would probably eat 400 to 1,100 full-grown lemmings; including their brood of nine owlets, the total number of lemmings eaten in their territory would amount to 1,900 to 2,600. Males appear to need more food than females during nesting, presumably because of their more active role in securing food while the females incubate and brood.

Later studies (Gessaman 1972) indicate that Watson’s estimate was low with respect to the daily food requirements of a free living bird, especially during winter where ambient temperatures can drop below –40°C. Gessaman found that the daily requirements for a confined owl are 200–400 g of lemming at air temperatures between –4.5°C to –40°C. The average gross energy intake of birds caged outdoors when the average air temperature was –34°C was equivalent to 400 g of lemming. Wild owls with more activity should require more than 400 g/d.

Less is known about daily and annual intake beyond the range of lemmings where a large variety of foods may be consumed. Beamer (1937) estimated that a wild owl would eat 3,000 to 5,000 Microtus in a year, amounting to 60–150 kg.

No quantitative studies. Captive birds drink water, especially after feeding (Watson 1957). Not known if owls can substitute snow for water. Adults and young eject pellets after stretching head and neck and wide gaping of the bill. Although nest is fairly clean throughout the egg-laying period, it becomes increasingly soiled following hatching of young, not only through defecation and pellets, but also through a surplus of decaying lemming carcasses. Vegetation if present is much reduced, but the richly manured ground is rapidly recolonized, judging by layers of pellets from different nestings (Watson 1957). The pellets contribute to the bulk and composition of the turf.

Needs further study. Given the opportunity, some nesting owls switch diet from rodents to ptarmigan when the latter become available. Captive birds select mice over rats (Watson 1957). Will cache surplus lemmings at perches apart from nests (as many as 26 uneaten carcasses have been found; Parmelee 1972), but function of perch cache is not clear. Perhaps male must procure a certain number of lemmings in order to bring the female into breeding condition.

Development

No information.

Vocal Array

Whether the diurnal Snowy Owls rely less on acoustical signals than other owls is a subject not well researched. The distinction between song and call in this species also merits further study, especially with reference to hooting. Although both sexes are known to hoot, it appears to be rare in females while common in males. Hooting probably serves as song at times, especially when mutual hootings occur among territorial males that may even face one another in threat posture (Watson 1957). Sutton (1932) witnessed as many as a dozen males hooting at once. One of these birds seen close up lifted his head, swelled out his throat enormously, elevated his tail until it stuck almost straight up, and gave four long, low hoots, bowing violently each time he hooted. Booming hoots answered from the ridges far and near. Hoots described as a local hoo, hoo, usually double, but sometimes six or more, the last often the loudest, with 1- or 2-second intervals after each (Watson 1957). Variants a deep hoo, and a long drawn-out hooooo . Hoots sometimes thought to have a ventriloquial quality, so that they seemed to come from high in the air or from the ground (Sutton 1932). Also, they carried a great distance (heard up to 7 mi by Sutton). Males hooted not only from ground or perch, but also commonly during flight.

Taylor’s (1973) observations confirm those of Sutton (1932) and Watson (1957). He reported that the “territorial hooting display” or song is similar to that of the Great Horned Owl, and that hooting volume is loudest in the context of territorial defense. Taylor’s drawings of the attitudes of males during territorial hootings (Fig. 4) are similar to Sutton’s description though somewhat less exaggerated.

Other than hootings, owls give several vocalizations, usually when disturbed near their nests. Different renditions and interpretations occur in the literature, all of which warrant more thorough study. One of the loudest and most often heard by Thaxter (1875) is the so-called watchman’s rattle, likely the rick, rick, rick heard by Wheelwright (in Witherby et al. 1952), and kre, kre, kre, kre, kre which reminded Watson (1957) of a Corncrake (Crex crex). Sutton and Parmelee (1956) rendered it as a barking ha, how, quack, quock, or quawk, sometimes repeated in wild volley, usually by the male although both sexes used it. The call is given commonly during flight and also from a perch. It appears to function as a threat, often becoming louder and more frequent as a human approaches a nest.

Females frequently give a call described as a whistling or mewing call (Watson 1957), which occurs before and after feeding by the male, during distraction display, and in displacement coition. Watson heard this call from females only, and Sutton and Parmelee (1956) heard it from only one of many males they observed. A crowed ca-ca-oh is given by attacking females but not recorded for males (Sutton and Parmelee 1956). Males give a low, rapid, cackling ka-ka-ka-ka-ka-ka in displacement coition or when alighting on a favorite perch (Watson 1957). A similar though often higher pitched ke-ke-ke-ke-ke is given by females before displacement coition, at the start of pre-coition display, and when feeding young.

Both sexes hiss when threatened. Although chicks cheep soon after hatching, they do not begin hissing until about two weeks old (Watson 1957). A juvenile call seldom reported is a shrill squeal given by unfledged owlets that have left the nest and are dispersed over the tundra. For a human to hear these rending screams issuing from a multitude of hiding places, and not to know their source, is a perplexing experience. Likely this call functions in helping the male find his scattered young; before each young he drops a freshly killed lemming (Parmelee 1972). Presumably these calls persist until the parent-offspring bond breaks.

Phenology

Except for Evans (1980), who observed wintering birds within a small defended area conducive to territorial behavior, few details have been reported concerning vocalizations outside the breeding season. He stated that during daylight, owls maintain territories by vocalizations accompanying agonistic displays, chases, and even attacks. Intrusions elicit high-pitched, drawn-out screams from a resident owl (“territorial screaming display” of Evans), accompanied by an attitude similar to that described by Taylor (1973) for the territorial hooting display. At maximum display it resembles the display described for hooting males by Sutton (1932) with particular reference to the tail, which is held nearly perpendicular to the ground. Flying intruders elicit stronger responses than perched owls.

Wintering birds also utter a series of 10 to 15 soft grating sounds accompanying a territorial screaming display (Evans 1980). This may be the sound described by Sutton (1932) as “heavy teeth grinding together.” Owls also scream without an agonistic display. Perched birds may give extremely soft melodious warbling sounds (2–3 s) and a low-pitched, two-syllable grunting sound after flying from the perch (Evans 1980).

Daily Patterns

Highly vociferous at all hours during the irregular nestings, but otherwise mostly silent year-round.

Sexual Differences

In addition to differences in vocal array, males are generally much more vocal than females, at least during nesting. This difference may be related to the male’s greater share in defense of nest and territory (Watson 1957).

Places Of Vocalizing

Usually an often used prominence within the territory.

As in many owl species, Snowy Owls snap their mandibles loudly when threatened. Owlets only 8 to 10 days old engage in bill snapping when handled (Watson 1957).

Walking, Hopping, Climbing

Generally sedentary when not flying, often remaining at a commanding perch for long, uninterrupted spells. Capable of walking and hopping, particularly while quartering the ground in search of prey, also while engaged in distraction behavior such as injury feigning.

Flight

Described as “strong, steady, and direct but not rapid...The long downward stroke of the large wings is a rather deliberate, long sweep; but the upward stroke is quick and rather jerky; it often sails on horizontal wings for some distance, especially as it sweeps upward to alight on some eminence” (Bent 1938). Flight is suggestive of buteo or harrier, and the jerky, swaying movement, given by the quick upstroke and slow downstroke of the wings, is a good field character even at a great distance (Watson 1957). However, when in pursuit of flying prey, the flight is not jerky, but rather strong and swift, almost falcon-like (P. Kerlinger pers. comm.). Not known whether the flight and body feathers are better adapted to a diurnal existence than to a nocturnal one that utilizes soft, velvety feathers for silent flight.

Swimming

Do not swim as a rule, although they can swim at least short distances. Parmelee et al. (1967) observed one large but flightless young that leapt into the water of a lake, and by using both wings made it across the deep waters to the opposite shore “some 50 feet away,” where it escaped pursuit.

Maintenance Behaviors

Little reported. Watson (1957) mentioned a pair of captive birds that spent much time preening, wing stretching, and cleaning their bills and faces by wiping them. Incubating females preen and adjust their feathers, yet their plumage often becomes deeply stained in sharp contrast to the nearly immaculate plumage of males, which do not incubate. Presumably the owls bathe on the breeding grounds, but details are needed on this.

The daily body temperature of any individual varies less than 1°C, while the mean body temperature of different owls may vary over a range of 2.5°C (Irving and Krog 1954,Siegfried et al. 1975, Gessaman 1978). The average daily body temperature in January for one of Gessaman’s owls ranged from 39.5° to 40.0°C (mean was 39.7°C over 7 d), for another 39.3° to 40.5°C (40.0°C, 5 d), and for another 40.8° to 41.5°C (41.1°C, 3 d). Its large size permits this owl to carry more insulation of feathers; its thermal conductance is the second lowest recorded for an avian species and is equivalent to that of arctic foxes (Alopex lagopus; Gessaman 1972). These data indicate that the Snowy Owl regulates body temperature well in winter. Daily patterns of heart rate in captive birds in winter resemble those reported for other large birds (Gessaman 1978).

Sleeping And Roosting

Little reported. Watson (1957) noted that a male spent little time hunting (estimated at 20% in July) and most of his time sleeping or resting in the long polar daylight. Incubating females rarely leave the nest, generally only when disturbed.

Daily Time Budget

No information, but such studies highly desirable.

Physical Interactions

In the presence of humans, individuals range from docility to skittishness in nonbreeding areas, and from timidity to forceful strikes on the nesting grounds. Males are the most important defender of nest and offspring (Wiklund and Stigh 1983), but individual females will also strike blows, especially in the presence of young. Serious wounds to humans have been reported (Barth 1950, Sutton and Parmelee 1956). Most aerial or dive-bombing attacks are initiated from the rear. They are most intense near the nest or favorite perch, but can occur far from the nest, fully a kilometer in one case (Parmelee 1972).

Information is sparse on the ratio of aggressive pairs to timid ones: four of 10 pairs noted by Sutton and Parmelee (1956) were considered shy, while Watson (1957) recorded only four of 19 shy, two of them nonbreeders. When striking humans, Snowy Owls use their talons, which may be opened or closed on contact. They also use their wings to some extent and, when handled, their bills. Individuals will attack most anything they consider threatening to their eggs or young. A male dive bombed repeatedly a pair of arctic wolves (Canis lupus); one of which was so annoyed as to leap high at the plunging owl (Parmelee 1972).

Close interactions between territorial owls on the breeding grounds have not been reported often although Taylor (1973) noted them once and thought that they might occur regularly early in the year. Reported encounters between nonbreeding Snowy Owls are equally scarce, except that Evans (1980) observed not only frequent threatening behaviors but also chases and mid-air clashes with clasped talons (three cases).

Some evidence suggests close encounters with other species. Snowy Owls wintering in Alberta regularly sustain injuries, mostly from unknown collisions (Kerlinger and Lein 1988a). A bizarre encounter took place between a Snowy Owl and a Great Horned Owl near Sebeka, MN, resulting in the death of both birds whose side-by-side remains were photographed for a local newspaper (D. Johnson pers. comm.).

Communication

In threat, a Snowy Owl lowers the front of its body, stretches its head low and forward, with wings partly extended, and feathers of head, back and flanks raised (Watson 1957). Both young and captive adults exaggerate this display when cornered. If forced back, birds finally lie on their backs, striking with talons and bill. On the breeding ground it is usually the male that postures, although when hooting the wings and feathers are less elevated. Females posture to a much lesser extent both on and off the nest.

Displacement behaviors, noted by several observers, have been perhaps best described by Watson (1957). Displacement coition occurs commonly when a female leaves the nest upon being disturbed. For whatever reason, the male follows her in flight and then attempts copulation on the ground, usually only once but occasionally several times. Completed displacement coition appears to be exceptional. Female initiation of the act has been observed, but considered rare. Intensity of displacement is related to the distance from the nest, and to the time of year.

Displacement feeding occurs among both sexes on the breeding grounds, but usually by the male during spells of great tension. Vigorous tearing of the turf and shaking vigorously pieces of turf accompany the act. Males engage in displacement hunting by hovering in “typical hunting flight” when their nesting territory including perches is invaded. Whether these activities are truly distraction displays or, according to Taylor (1973), simply displays that occur regularly as part of courtship, remains unknown. Distraction displays do occur on the wintering grounds, for Evans (1980) observed an owl biting wood pieces from a pole, which he interpreted as redirected aggression.

Individual Distance

The smallest occupied areas for breeding pairs reported for North America are about 0.5 mile or slightly less than 1 km apart (Hart l880, Brandt 1943, Pitelka et al. 1955a). Complicating the density estimates are nonbreeders that may utilize the areas occupied by nesting pairs, sometimes up to five or more pairs per square mile (= about 2.6 km2; Pitelka et al. 1955a).

Territoriality

On the breeding grounds males establish territories from late Apr (depending on locality) and defend them through vocalizations (hooting) and threat postures. Not determined is whether combat ever takes place among the owls in the nesting territories as it does under certain conditions in the winter territories.

Breeding density depends on the abundance of prey. Pairs are usually widely spaced but not necessarily uniformly distributed over the tundra even under the most favorable conditions (Parmelee et al. 1967). Breeding pairs are fewer in number and more widely dispersed when their prey is only moderately abundant, and none breed during extreme lows in areas traditionally used for nesting. Under certain adverse climatic conditions, fledging production can be maintained by increasing territorial size, thereby compensating for a lower availability of rodents (Wiklund and Stigh 1986).

Whether the same individuals of one or both sexes remain faithful to the same breeding ground is a moot question. P. Kerlinger (pers. comm.) and the author suspect that the species is a nomadic breeder, as apparently is the case with some other owls that are dependent on fluctuating abundance of certain prey (see Lundberg 1979).

Important North American predators that compete with Snowy Owls for lemmings and other prey are the Rough-legged Hawk (Buteo lagopus), Golden Eagle (Aquila chrysaetos), Peregrine Falcon (Falco peregrinus), Gyrfalcon (F. rusticolus), Pomarine Jaeger (Stercorarius pomarinus), Parasitic Jaeger (S. parasiticus), Long-tailed Jaeger (S. longicaudus), Glaucous Gull (Larus hyperboreus), Short-eared Owl (Asio flammeus), Common Raven (Corvus corax), gray wolf, arctic fox, and ermine (Mustela erminea). Of the birds, any can nest fairly close to Snowy Owls; e.g., Sutton and Parmelee (1956) observed an owl’s nest on the top of a high cliff not far above a Peregrine Falcon’s nest; the two species engaged in skirmishes when disturbed. Several avian species that compete for lemmings may coexist. An area about 2.5 km near Barrow, AK accommodated one pair of nesting Snowy Owls, three pairs of nesting Short-eared Owls, 19 pairs of nesting Pomarine Jaegers, and an additional 8 to 10 pairs of nonbreeding Pomarine Jaegers (Pitelka et al. 1955a). The difference between resident and nonbreeding nomadic predators should be stressed (Watson 1957). Control of a lemming increase probably depends largely on the density of nomadic predators that assemble in an area, as the resident population by itself would not suffice (Watson 1957)—a challenging concept that merits further study.

In winter, five Snowy Owl territories ranged in size from about 0.7 –2.5 km2 at Horicon Marsh, WI (Keith 1964). A maximum density of 2.3 owls per 100 ha was found on Wolf Island, Ont. (Weir 1973). Near Duluth, MN, Evans (1980) observed as many as 6 owls at one time in an area only 40 ha in size; the site was near grain elevators with abundant rodents, hardly typical of the surrounding open country and farmland where owls were less numerous.

Immature males establish territories early in the season only to be displaced later by females (Evans 1980). By mid-Dec essentially all territories are held by females and an occasional adult male, and then maintained until spring departure. Boxall and Lein (1982b) also found that females maintained winter territories following the disappearance of males.

Mating System

Mostly monogamous, with only a few reported exceptions. One male bred, successfully fed, and defended two females; all 15 young produced by the two females fledged (Watson 1957). Matings of one male with two females also noted by Y. Hagen (inJohnsgard 1988) and Robinson and Becker (1986). Conversely, a female may mate with two males simultaneously (Mikkola 1983).

Pair Bond

Among courtship displays, conspicuous over the breeding ground is the “aerial display” described by Taylor (1973; see Fig. 5). The exaggerated wingbeat with a pronounced upswing of the wings has long been observed. Less often mentioned is the brief delay between beats that causes a “sinking” or loss of altitude that is quickly recovered with the next power stroke. The length of the resulting undulating flight depends on the distance between perches or patches of exposed ground. The display ends in a gradual climb followed by a gentle vertical descent to the ground with wings flapping or set in a “V” position. The flight is oriented toward, or away from, but past the female who is usually perched; flying females may also elicit the aerial display. The displaying male often carries a lemming in his bill, sometimes in his claws. Lemming carrying gradually diminishes as the season progresses. By July even the display is rarely observed.

Following the aerial display, the male alights on exposed turf, not infrequently at old nest sites, and then drops the lemming or other prey on the ground before performing his ground display. In the beginning he remains fairly erect, keeping the wings partly spread with wrists high (see Fig. 6). He then orients his back and sides, rather than his front, toward the female, who usually flies or hops in closer. As the display continues, the male leans farther forward with head lowered and tail partly fanned (Fig. 6). According to Taylor, the flat surfaces of the raised wings reflect light and cause a flashing effect that is visible from afar. Ground displays usually last 1 to 2 min, occasionally as many as 5 min. Rarely observed is the aerial exchange involving the passing of a lemming from male to female during flight, as noted once by Sutton and Parmelee (1956). The two birds swung gracefully upward as they met in midair, and the male passed the lemming from its claws directly to those of the female. Conceivably these displays evolved in consequence of the species’ diurnal habits, reminding one of similar ones in falcons.

A female assumes essentially the same posture before, during, and after copulation (Taylor 1973; see Fig. 7 .). While perched, she leans fairly far forward with partly raised tail, wings held loosely at sides, and body feathers generally slightly ruffled. This attitude elicits displays by males not only before but also following copulation. Evidently vocalizations may or may not be a regular part of these behaviors, a subject that merits additional study. Males mounting females have often been witnessed, but details of the copulatory act are lacking, especially on what constitutes successful sperm transfer.

Mate fidelity is poorly understood. Owing to between-season fluctuations in breeding, long-term banding operations on traditional nesting grounds are highly desirable. This species is generally thought to have a weak pair bond, in contrast to many owls (Johnsgard 1988). There is no indication that a male and a female form a pair for more than one season (Voous 1989). Nomadism almost certainly would preclude mate fidelity (P. Kerlinger pers. comm.). Exceptions may occur under unusual circumstances, as at Fetlar, Shetland Islands, where the species breeds rarely and males are disproportionately fewer than females. According to Robinson and Becker (1986), the same pair of owls (based on plumage details) nested there from 1967 to 1974; in 1973 and 1974 the male was bigamous.

Extra-Pair Copulations

No information.

Degree Of Sociality

Appears to be antisocial wherever breeding birds are concentrated. However, nonbreeding birds that occurred in local concentrations at Barrow, AK, did not react to each other aggressively and were considered not to be antisocial by Pitelka et al. (1955b). In winter the owls may be either highly territorial (Evans 1980) or at times nonterritorial despite the proximity of several individuals (P. Kerlinger pers. comm.).

Play

No information.

Interactions With Other Species

Although predators generally try to avoid one another, the relationship between such predators is mostly one of mutual kleptoparasitism in all places and all seasons of the year (Voous 1989). On the other hand, certain potential prey species, notably geese, appear to profit from the presence of nesting Snowy Owls, conceivably because of the protective influence of the owls in driving off other predators such as foxes. Greater Snow Geese (Chen caerulescens atlantica) and Lesser Snow Geese (C. c. caerulescens) have been found nesting in the immediate vicinity of nesting Snowy Owls, even though unlimited nesting areas are available to the geese and the owls usually commence nesting before their arrival (Parmelee 1972).

Snowy Owls, like some other predatory birds, evidently seldom take prey close to their nests, although Robinson and Becker (1986) saw a female owl leave her nest and kill an oystercatcher that had landed within 20 m of her. The same owl ignored passerines close by.

Kinds Of Predators

Humans are probably this owl’s principal predator. Snowy Owls have been used for food from at least the last glaciation period, judging by numerous bones found in cave deposits. The species’ rareness in northern Europe may be due to persecution by man as much as to the contraction of arctic habitats (Voous 1989). Inuits and other northern peoples still hunt the owls for food. Humans also kill them for trophies, to protect game animals (Dufresne 1922), and perhaps out of curiosity (Gross 1947). Nonhuman predators listed in the literature include foxes and jaegers. Potential predators almost certainly include wolves and dogs, and most any avian predator capable of killing the owl. In one case, an incubating female was killed by two Pomarine Jaegers (Bailey 1948).

Manner Of Predation

Humans secure adult Snowy Owls chiefly through hunting and trapping. Flightless young are gathered on the tundra. Predators other than humans prey mostly on the species’ eggs and young.

Response To Predators

In defense of nests, the male stands guard while the female incubates or broods. While the male is usually the first to confront humans and other potential predators, both sexes display agonistic responses. These are often followed by dive-bombing pursuits that may include strikes, usually by the male, sometimes by the female. Failing in this, either sex may engage in distraction displays.

Pair Formation

Little information. On the breeding ground, presumably not before late Apr to mid May (Voous 1989; Fig. 2), but pairs may actually form on the wintering grounds in the northern Great Plains. Incipient courtship activities as early as midwinter in s. Alberta (Boxall and Lein 1982c). Whether pairs actually bond there and arrive together on the breeding ground is debatable.

Selection Process

Presumably the male establishes the territory and the female initially selects the nest site, but this subject merits further study.

Microhabitat

At the onset of laying, the female visits one of several possible sites, usually on some windswept prominence, not necessarily the highest in the vicinity. In high, rolling country such hillocks are numerous and widespread. In flat, marshy areas they may be decidedly scarce, but nevertheless essential, if only amounting to a few hummocks or frost-heaved polygons. Nests on top of large boulders are exceptional in some areas (Murie 1929) but not in others (Watson 1957).

Site Characteristics

In addition to a sufficient food supply, the site must be snow-free, not subject to flooding, and command a view of its surroundings.

Construction Process

Nest constructed solely by female, so far as known. On turf or bare ground, female scratches out a scrape with her claws. Here she moves about and even twirls, until a distinct but shallow hollow results. No insulating lining of feathers or vegetation is added, though such material may accidentally fall into the depression.

Dimensions

No information.

Microclimate

Nests often placed in windswept places where snow and moisture tend not to accumulate. Cold does not appear to be a factor as long as the female covers eggs or small young. Eggs and small chicks are totally dependent on insulative protection of the female’s incubation patch, as an unlined scrape has no insulative value.

Maintenance Or Re-Use Of Nests

Scrapes may be used repeatedly over long periods (Watson 1957). At an unusual breeding ground in the Shetland Islands where owls bred for nine consecutive years, Robinson and Becker (1986) imply that a “main nest” was used repeatedly.

Nonbreeding Or Unfinished Nests

For reasons unknown, the female may work at one scrape and then abandon it for another nearby. Parmelee et al. (1967) observed a female that worked over a scrape during a three-day period before abandoning it.

Shape

Short elliptical or subelliptical (Harrison 1984).

Size

Measurements of 56 eggs show four extremes: 60.5 x 47 mm, 58 x 47.5, and 50.6 x 4l.7; mean measurements for all were 56.4 x 44.8 mm (Bent 1938). Mean measurements given by Harrison (1984) are 57.4 x 45.2 mm. No geographic differences in size across the Northern Hemisphere; 20% smaller than those of northern Eagle Owls (Bubo bubo) and 8% smaller than those of northern Great Horned Owls (Voous 1989).

Mass

Not well known (for fresh eggs), in part because incubation begins with the first egg. In a clutch of seven eggs, each weighed a mean of 58 g (Watson 1957). An eight-egg clutch had embryos that ranged in size from small to large (feathers); weights of the eggs ranged from 58.1 to 64.5 g (mean 62.1; DFP). Weights given by Y. Hagen (in Watson 1957) range from 60 to 65 g and probably approximate the weight of fresh eggs, but this subject merits further study.

Clutch mass equivalent to 25–43% of the female’s body mass—a significant physiological achievement and exceptional for large owls and large raptors in general (Voous 1989).

Eggshell Thickness

No data.

Color

White or creamy white (Bent 1938, Harrison 1984). As incubation proceeds, the whitish shells often become stained.

Surface Texture

Smooth to somewhat granular. A few “corrugated lines” are noticeable in some specimens (Bendire 1892, DFP). Some eggs lack gloss (Bent 1938) but others show a slight gloss (Harrison 1984).

Egg Laying

Interval between completion of a nest scrape and laying of first egg not reported, nor has the precise time of egg laying. Eggs apparently laid almost any time during the 24-h period of continuous daylight.

Eggs usually laid at intervals of about 2 d. Mean intervals range from 41 to 50 h (Schaanning 1907; Robinson and Becker 1986), occasionally longer (Parmelee et al. 1967). At one nest, 4 d elapsed between laying of the eighth and ninth eggs (but not between the ninth and tenth eggs), with consequent 4-d interval between hatching of the eighth and ninth chicks. Tulloch (1968) noted similar disruptions, including a 5 d interval between layings, possibly related to inclement weather. This subject merits further study.

Egg attentiveness by the female commences immediately with the laying of the first egg. Throughout the egg-laying period, the male guards the site. Replacement clutches suspected but not confirmed. Small clutches appearing late in the season suggest replacements, but late clutches also relate to bigamous matings (Watson 1957), possibly also to young birds (Robinson and Becker 1986). Intraspecific egg dumping not observed.

Onset Of Incubation

Female alone incubates, commencing with the first egg. No exception reported.

Incubation Patch

Female develops a conspicuous incubation patch—an enormous, flabby, highly vascularized featherless area of pink belly skin used in covering the eggs (Parmelee 1972).

Incubation Period

Mean period for the first egg over a 9-year period reported as 31.6 days, slightly longer (32 d) for later eggs (Robinson and Becker 1986). Watson (1957) and Taylor (1973) reported 32 to 33 d. The period for the fourth, seventh and eighth eggs at one nest, and for the ninth and tenth eggs at another, reported as 32 to 33 d (Parmelee et al. 1967). Extreme records of 27 d (Schaanning 1907) and 37 to 38 d (Sutton 1932) not confirmed.

Parental Behavior

The duration of the female’s attentive period probably varies individually. Toward the end of incubation, females become more intermittent in their incubation (Taylor 1973). On the other hand, one female noted by Parmelee et al. (1967) remained at the nest as long as eggs (or young) were present; another continued to incubate an addled egg (no embryo) as late as 1 Aug after the young had left.

Hardiness Of Eggs And Effects Of Egg Neglect

Eggs are subject to below-freezing temperatures, particularly in May. When left unattended during severe weather, they will freeze hard and crack open (Parmelee 1972). Drifting snow caused one female to abandon four eggs, whereas no desertions occurred at nearby nests that were essentially snow-free, emphasizing the importance of windswept sites (Taylor 1973). Neglect of certain eggs results in mortality, likely more through accident than design. One female, brooding chicks during inclement weather, failed to cover an egg that should have hatched on 18 July; although the chick within the shell was still alive on the 21st, it died before hatching (Parmelee et al. 1967).

Preliminary Events And Vocalizations

No information.

Shell Breaking And Emergence

Starring may occur either at the end or side of the shell. A small circle of fine cracks appears, followed by a gradually enlarging and outward lengthening of the cracks. Pipping occurs when the chick finally punches a hole through the weakened circle of cracks. Several hatchings observed during the afternoon hours, but presumably they occur at most anytime in the nearly continuous polar daylight (DFP). The period from the first cracking of the shell until the first small hole was at least a day in three cases; the period from first visible crack to the final emergence of the chick was well over a day but less than 2 d in three cases, and less than 1 d in another (Watson 1957). Periods up to 3 d have been observed (DFP), indicating that the period from starring to emergence may be variable.

Eggs generally hatch at 2 d intervals except when egg laying has been interrupted, resulting in longer periods between hatchings. Female apparently gives no assistance during hatching; no evidence that she carries away eggshells. Some shell remains are trampled into the ground within the nest scrape or close by. Captive females also eat shell remains (Watson 1957), and presumably wild birds do likewise.

Condition At Hatching

Semialtricial; newly hatched chicks leave the shell wet, blind, and seemingly helpless. Eyes are closed at hatching and start to open usually by the fifth day, occasionally later. Within a few hours the newly hatched chick is covered nearly completely with rather short, white, fluffy down that extends to the base of the bill and claws, but with a bare patch at tarsal joint. Cere pale blue, bill bluish, palest on the horny part of upper mandible, becoming darker with age; eyelids bluish, irises blue-white to pale-grayish, becoming yellower with age (Sutton 1932).

Weights (to nearest 5 g) of seven newly hatched chicks ranged from 35 to 55 g (mean 46 g; Watson 1957). Mean weight of three was 44.7 g (Parmelee et al. 1967). Linear measurements lacking for live chicks. One preserved specimen has an approximate body length of 100 mm, bill from cere 11.0 mm, middle toe claw from sheath 5.5 mm; another had a body length of 99 mm, bill 10.00 mm, claw 5.0 mm.

Growth And Development

In the most complete study to date, some chicks lose much weight the first day (as much as 45% before recovering), while others increase greatly from the start (Watson 1957). Some chicks increase their weight by 60%–70% in a day. Nearly all grow rapidly by the third day, but increase less thereafter, falling to about 20% by the eighth day, and to about 6% in the fourth week. Some 28-day-old young weigh 1,300–1,600 g.

Some 4-day-old chicks show noticeably enlarged tarsal pads, claws, and bill, which still retains an egg tooth (DFP). Chicks about 9 days old have prominent bills (16.5 mm from cere, no trace of egg tooth) and claws (14.0 mm from sheath of middle toe). First signs of incoming additional down feathers appear along all of the major feather tracts in some 4-day-old young (DFP). By as few as 8 d, or as late as 12 d, the short white down is largely replaced directly by long, fluffy, dusky colored down with grayish tips, giving the owlet a dark appearance touched here and there with frosty patches, particularly in the head and body regions. Incoming remiges are short and sheathed nearly to the tip in some chicks 10 d old. Feathers emerge from their sheaths in some chicks 11 and 12 d old (Watson 1957). Growth of the gray down and feathers varies among chicks (DFP).

Nestlings react to handling by bill snapping and hissing, and are easily immobilized (tonically immobile) by turning them over on their backs or simply picking them up (Watson 1957). In such a state the body and legs stiffen, and even the stretched wing remains motionless. The eyes in some are closed, but those of others remain open. Fairly large young handled by Parmelee (1972) were so stiff that they were easily balanced on a point of rock. The significance of this behavior is not well understood, but evidently it occurs in other owl species as well.

Threat postures are first noticeable 20 to 25 d after hatching and commonly after 28 d (Watson 1957). Little information is available on intersibling conflict, a condition less likely in this species than in other owls because the young scatter over the tundra long before fledging. Young move about actively within the nest scrape where small young often find shelter beneath their older siblings. Nestlings are capable of leaving the nest when only 14 d old, though most remain a few days longer. Whatever their age of final departure, they are quick and agile afoot at that time. Flight happens slowly, usually long after the nest has been abandoned. The owlets spring from the ground and beat their wings furiously, only to fall back many times before accomplishing even a short flight (Parmelee 1972).

Brooding

Female broods the young from hatching until they abandon the nest. The asynchronous hatching results in brooding young and incubating eggs at the same time. Neither brooding of young by the male nor brooding outside the nest by either sex has been reported. Unless the brooding female is disturbed, she remains at the nest without relief from the male except for feedings. Female alone broods. No exceptions reported.

Feeding

Newly hatched young may eat partially digested food taken from the adult (Collett 1921). For the first two days owlets are fed on lemming soft parts, including heart, liver and testes, which the female tears out and feeds in small pieces (Watson 1957). Some 5 d old young may be fed parts containing bones, and those 10 d old may be given partially dissected lemmings. Some week-old owlets already bring up long pellets of lemming fur and bones. Although young 14 d old are capable of handling their food, most feed themselves very little. The female may cast up pellets that are fed directly to the larger nestlings (Robinson and Becker 1986). Dispersed young handle undissected lemmings and other prey.

Male brings food to the incubating/brooding female, who in turn feeds the young. Young away from the nest are fed by the male, at least until the female abandons the nest. More information is needed to determine the role of the female following her activity at the nest. Female dissects prey into small portions before feeding small young. Older young are given larger portions up to the time they receive whole prey. Male feeds only whole prey to their young. More information is needed on the parent offspring break.

No data available on frequency of feeding trips, amounts of food brought per trip, and apportionment of food to young. The fact that siblings do not always gain body mass equally suggests unequal apportionments of food and possibly sibling aggression. On the other hand, the successful departure from a nest by all of its siblings indicates that uneven growth rates matter little when sufficient food is available.

Nest Sanitation

Female defecates away from the nest, which remains quite clean before the young begin to hatch. Thereafter, the nest becomes increasingly soiled through defecation (no fecal sacs) by the young, disgorged pellets, and particularly unused lemming carcasses and other prey that accumulate and spoil. Flies gather at nests that become soiled after hatching of young. Mosquitoes attack owlets, particularly about the eyes and face (Sutton and Parmelee 1956, Watson 1957). Not determined is whether such attacks are simply annoyances, or vectors of disease. A detailed study on the invertebrates at a Snowy Owl’s nest is needed.

Little information. Females and broods of bigamist matings remain apart and do not interact within the males’ territory (Watson 1957).

Departure From Nest

On Baffin Island, Parmelee and Sutton (1956) observed that owlets left the nest when 14 d old, although Watson (1957) found that the period ranged from 18 to 25 d; most young left about 25 d. On Victoria Island, Parmelee et al. (1967) found that most young left at 14 to 16 d, some remaining until 22 d of age. Having left the nest, young may (Parmelee et al. 1967, Robinson and Becker 1986) or may not (Watson 1957) return. Although some owlets leave the nest and wander when only 14 d old, some of them apparently return to the nest one or more times before abandoning it for the last time. A final departure date of 25 to 26 d seems reasonable; many days pass before even these late-departing young fledge. Under the care of both parents, the scattered young sometimes reconvene and band together (Parmelee 1972).

Fledglings make frequent short flights, especially down inclines, before they are capable of making sustained buoyant ones. Owlets are unable to fly strongly with good control until over 50 d old (Watson 1957, Parmelee et al. 1967, Robinson and Becker 1986). At one nest with 11 eggs, 54 d elapsed between the laying of the first egg and hatching of the last young; allowing another 50 d for fledging, the interval from first egg to fledging of the last young totaled 104 d (Parmelee 1972). The period would be considerably shorter for nests with small clutches.

Once the young leave the nest (at about 25 d old), they are fed entirely by the parents for at least another five weeks, when they first began to hunt for themselves (Watson 1957). They are probably partly fed for at least another week or two, as they are still poor at hunting. During the five weeks after leaving the nest, an owlet eats roughly 7 kg of lemmings. Including lemmings eaten in and out of the nest, a brood of nine owlets will have consumed as many as 1,500 lemmings by the time they reach independence.

Little information on independent, fledged young on the breeding grounds. See Migration; Populations: range.

Age At First Breeding

Not known. Inasmuch as Pitelka et al. (1955a) observed nonbreeding individuals thought to be immatures one year old, at least two, possibly more years are required for sexual development. Under certain, perhaps unusual conditions, some individuals breed annually when not dependent on lemmings. A male that was mated to two females bred for nine consecutive seasons until he disappeared (Robinson and Becker 1986). Considering the irregularity of lemming abundance, it is conceivable that some individuals breed only once in every three to five years.

Clutch

Highly variable, depending largely on the availability of food. The clutch size is generally small (3–5) when food is limited, but large (7–11) when plentiful. Clutches in excess of 11 eggs (as many as 16 reported) need confirmation. Nesting is aborted when food is extremely scarce. Normally one clutch per breeding season. Repeat nestings following early loss of eggs is suspected but not confirmed.

Annual And Lifetime Reproductive Success

Difficult to determine due to nomadic behavior of birds. Number of young produced per pair/nest/clutch may be highly variable among neighboring pairs. One nest produced 11 young from 11 eggs (at least 10 fledged) while another nearby produced only four young from 10 eggs (Parmelee et al. 1967). Under very favorable conditions, reproductive success can be extraordinary. Of 32 eggs observed by Watson (1957), 31 hatched and all 31 young fledged. This observation shows that an entire group of nesting females under surveillance can succeed. Females not known to produce more than one brood per season, the limiting factor perhaps being the extra-long breeding season.

Maximum known age in the wild is 9 yr, 5 mo (Glutz and Bauer 1980; DFP). A captive bird in Switzerland lived for at least 28 years (Schenker 1978), an age that may be attained in the wild. Inasmuch as the Snowy Owl is simultaneously nomadic, migratory and irruptive, the species may have reached a terminal point in its evolution, unless it is able to develop other means of avoiding winter, life-wasting migrations, and irruptions (Voous 1989).

Many Snowy Owls that move southward from arctic regions are mistakenly assumed to die from starvation. Although this may prove to be the case during irruptive migrations of young in western and eastern sections of North America, there is no evidence that this is so in the N. Great Plains. In Alberta, 45% of the specimens examined had moderate to heavy fat deposits, and traumatic injuries were the major cause of mortality (Kerlinger and Lein 1988a). Causes of death or injury were collisions with unknown objects (46.5%), automobiles (14.1%), utility lines (4.2%) and airplanes (1.4%); also gunshot wounds (12.7%), electrocution (5.6%), fishing tackle (1.4%). Only 14.1% was believed due to starvation. Even as far south as Kansas, a Snowy Owl fed on rodents at a lumberyard for nearly a month before being accidentally electrocuted (Parmelee 1972). Gross (1947) inferred that individuals seen far from land at sea never live to return, but this is a moot question, difficult to resolve.

Prolonged periods of inclement weather subject young to diseases. An owlet that died in the nest in the Shetland Islands was infected with pneumonia and containedStaphylococcus, and another harbored the fungal infection Aspergillosis, both conditions indicating dampness (Robinson and Becker 1986).

Natal Philopatry

Not well documented. A prefledged Snowy Owl banded by Parmelee et al. (1967) in 1960 was shot by a native hunter three years later within 70 km of its natal ground on Victoria Island in the Canadian Arctic. It was not determined whether this individual was breeding, although in this same area at least one pair raised young that season. Another prefledged owlet banded on Victoria Island in 1960 (Parmelee 1972) was recovered nine years later at Southampton Island in November far from the natal site; not determined was whether this individual had bred where recovered. Conceivably, nomadism would be selected over philopatry if the owls are truly nomadic in their breeding as appears to be the case.

Fidelity To Breeding Site And Winter Home Range

Not documented for breeding sites, but some evidence exists for winter site fidelity (Oeming 1957, Follen and Leupke 1980).

Dispersal From Breeding Site

Seven young siblings were banded by Parmelee (1972) on Victoria Island on 18 July 1960. One was recovered in e. Ontario near Clyde Forks on 19 October 1961, another near Hudson Bay at Attawapistat, Ontario on 8 May 1962, and a third, on the opposite side of the world, at Sakhalin, USSR, on 18 February 1962—all three recoveries within seven months of one another. This extraordinary dispersal of young also explains in part why the species is nomadic and monotypic.

Home Range

No information.

Numbers

Although populations have been estimated on the breeding grounds, most cover a relatively small area rather than an entire island or region. In perhaps the most meaningful census covering a broad area, Manning et al. (1956) estimated 15,000 to 20,000 owls in the Canadian Arctic on Banks Island (about 64,000 km2) during a reproductive high, but only 2,000 during a low. The number of individuals seen per hour of observing was multiplied by a figure related to the conspicuousness of the species to estimate the number per square mile. Considering the conspicuousness of the owls, aerial reconnaissance used for censusing waterfowl and other wildlife would be desirable.

Wintering owls recorded at 235 sites of Christmas Bird Counts, generally in low numbers (Root 1988). Highest abundance (0.21 individuals/party hour) recorded near Green Bay, WI, where the owls were seen in 80% of the years examined. Winter censuses are disrupted by periodic irruptions of owls, especially outside the northern Great Plains.

Seventeen irruptions between and inclusive of the years 1882 and 1883 and 1945 and 1946 took place in North America with intervals of 3 to 5 years (mean 3.9 yr). Numbers of owls were estimated in the thousands (Gross 1947). For reasons unknown, during such times the owls fly far from land over sea where they take refuge on ships. The distribution map based on Root’s (1988) analysis substantiates the hypothesis by Kerlinger, Lein, and others that the majority of Snowy Owls in North America winter regularly in the northern Great Plains.

Trends

Except for an apparent decline of populations in northern Europe (Voous 1989), information is lacking for long term shifts in the numbers or distribution of this species.

The most meaningful conservation policy in recent years has been protective measures that prohibit the shooting and trapping of owls for food, sport, or trophies. Harvesting of owls for food, feathers, claws (ornaments) by northern native peoples may impact local populations, but probably does not have the continent-wide potential for destruction as in the highly populated regions farther south. Other than shooting/trapping, the policies presently being enacted to protect large birds in general from electrocutions, airplane strikes, etc., will benefit the Snowy Owl.

Juvenal Plumage

Feathers of body and greater, median, and lesser wing coverts of very loose and downy structure especially on head and underparts. When not fully grown this plumage has general appearance of down as distal portions of rami appear as though they were separate filaments. In this stage much white nestling down still adheres to tips of rami. Upperparts, greater, lesser and median wing coverts and underparts dark mouse brown, feathers tipped and speckled grayish white, those of scapulars and mantle not of such loose structure as rest and faintly barred whitish; facial disk and chin white, feathers tipped brown; toes white; wing feathers, primary coverts and tail feathers of normal structure, white, barred brown like adult female but tips of innermost secondaries and primary coverts considerably mottled brown (Witherby et al. 1952). Flight feathers were still sheathed at the base of a 40-day-old, flightless female juvenile (1,800 g, wing chord 320 mm, tail 135 mm, culmen from cere 25 mm). Dark down predominated ventrally, but white winter feathers were starting to replace the down. Specimen shows down feathers still attached to the tips of incoming winter feathers (Parmelee et al. 1967). The juvenal body plumage, greater, lesser and median coverts are molted but often not completed before October. Remiges, rectrices and primary coverts are not molted (Witherby et al. 1952).

Basic I Plumage

First-year males with white bib of 2–8 cm present and back of head primarily white. First-year females with barring present in all areas of plumage except for white facial disk, tarsi, feet and wing linings; white bib, if present, less than 4 cm and back of head primarily barred. Females also show more rows of bars on underside (6 rows) and upperside (3 or more rows) of tail than do males (3 or less rows). Both sexes show moderate or extensive mottling on distal portions of greater and/or median secondary coverts (Josephson 1980).

Basic II Plumage

Not clearly defined. Specimens examined by Josephson (1980) showed mixed characteristics of first-year birds and adults, suggesting that second-year birds retain characteristics of the first basic plumage to a variable extent while displaying many features typical of adults. Not clearly defined. If Josephson’s (1980) suggestion proves correct concerning a basic II plumage, this molt probably takes place during the summer when birds are two years old.

Definitive Basic Plumage

Plumage of adult males entirely white or white except for narrow, sparse, pale gray, or brown barring on breast, back, wings, head and/or tail. Plumage of adult females with moderate to extensive barring present on breast, wings, head, and/or tail; little if any mottling on distal portions of greater and/or median secondary coverts; white bib of less than 8 cm present and/or back of head primarily white (Josephson 1980).

The darkest males and the palest females are virtually alike in color, but the whitest birds, sometimes practically pure white, are always males, and the most heavily barred ones are always females (Bent 1938). In addition to their heavier barring, incubating females often have a highly soiled plumage, which easily identifies a breeding individual on the nesting grounds. Adults have one complete annual molt, beginning in July and lasting into early fall (Bent 1938).

Bill And Gape

Short, strong and much compressed. Upper mandible strongly curved and projecting at tip, which is sharply pointed. Lower mandible broader and shorter with rounded tip. Blackish horn colored. Nostrils covered by bristle feathers of disk (Witherby 1952).

Iris

Brilliant golden yellow, at times almost orange (Sutton 1932).

Legs And Feet

Tarsus and toes thickly covered with feathers (with long, threadlike rami). Three toes directed forward and one backward. Claws long, much compressed, strong, curved, and tapering to sharp point. Blackish horn color similar to bill (Witherby et al. 1952).

Linear

Ranks first in over-all size among the 18 species of owls in North America (Burton 1973). First-year birds and adults of the same sex do not differ significantly in wing length or in length of central rectrices, but both are significantly larger for females than for males regardless of age (Josephson 1980). Most published accounts on measurements combine age groups (Table 1.) .

Probably the best information to date relating to fat conditions and body weights derives from Kerlinger and Lein’s (1988a) study of live or freshly killed birds rather than salvaged specimens that may have lost weight due to desiccation, injury, etc. For wintering owls in Alberta, those with no fat or light fat deposits are not distributed randomly among age-sex classes. Immature males have no fat or light fat deposits more often than expected and moderate to heavy fat deposits less often. Adult females show moderate to heavy fat deposits more often than expected, and no fat or light fat deposits less often. Deviations for adult males and immature females account for less than 15% of the chi-square values

Mean weights reported by Kerlinger and Lein (1988a) are significantly greater than those based primarily on museum specimens by several other researchers. In their sample, males (n = 23) range in weight from 1,606–2,043 g (mean 1,806 ± 30), females (n = 21) 1,838–2,951 g (2,279 ± 57). Approximate minimum weights of birds with at least moderate fat deposits are estimated about 1,575 g for males and 1,950 g for females. Adults tend to be heavier and fatter than immature birds. For practical considerations relating to survival, Kerlinger and Lein recommended that the weights of captive males should exceed 1,800 g and that of females 2,200 g before being released in the wild. These are thought to be approximate mean weights for healthy birds during winter. Needed at present are many more weights of live or freshly killed specimens taken during the breeding season. One nonbreeding adult male weighed 1,630 g on 27 Jul 1962, and one breeding adult female weighed 2,025 g on 28 Aug 1960 when collected by Parmelee et al. (1967) in the Canadian Arctic.

For invaluable assistance I thank the following individuals and organizations: Tom English (Bell Museum, University of Minnesota, Minneapolis); Mark R. Fuller (Patuxent Wildlife Research Center, Laurel, MD); J. A. Gessaman, Utah State University, Logan, UT; David Gill (Vertebrate Ethology Section, Canadian Museum of Nature, Ottawa, ON); David H. Johnson (Oregon Cooperative Wildlife Research Unit, Oregon State University, Corvallis); Paul N. Kerlinger (Cape May Bird Observatory, Cape May Point, NJ); Lloyd F. Kiff (Western Foundation of Vertebrate Zoology, Los Angeles); Jean Parmelee (Bell Museum, University of Minnesota); Susan M. Sivard (New York Historical Society, New York, NY); and Gary A. Voelker (Bell Museum, University of Minneapolis).