Demography and Populations
Welcome to the Birds of North America Online!
You are currently viewing one of the free species accounts available in our complimentary tour of BNA. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this species.
For complete access to all species accounts, a subscription is required. Subscriptions are available for as little as $5 for 30 days of complete access! If you would like to subscribe to BNA, please visit the Cornell Lab of Ornithology E-Store or call us at 877-873-2626 (M-F, 8:00-4:00 ET).
Measures of Breeding Activity
Age At First Breeding, Intervals Between Breeding
Not known for wild individuals. Probably sexually mature in the first year of life, but the ability to acquire territory and resources needed for breeding is likely dictated by age, experience, and hunting ability (DWH). Detailed mark-recapture studies are needed to confirm age of first breeding.
During the breeding season in Barrow, AK, Pitelka et al. (1955b) observed local non-breeding concentrations ranging from small groups to up to 20-30 owls in an area of 20-30 ha; the owls did not appear aggressive towards each other. Additionally, in Barrow, AK, over 22 yr DWH observed loose groups of non-breeding owls roost in separate areas from breeders. Based upon knowledge of plumage traits, these owls are believed to have been younger, in perhaps their first year of life; most appeared to be males. If males need 3-4+ yr to attain Definitive Basic plumage, then it appears they may not breed for the first time until several years old. In Barrow, AK, only males with almost pure plumage breed (n = 239 nests), implying that females mate only with males > 4 yr old.
Due to nomadic lifestyle and fluctuating food resources, individuals may or may not breed in consecutive years. Few data. Based on satellite data, one female may have bred in two consecutive years -- moving from Alaska to Russia and to Canada -- while others may have not, or have failed at breeding (Fuller et al. 2003). In Canada, confirmation of satellite-marked females (n = 7) proved breeding in consecutive years, and for one individual, 3 consecutive years (Therrien et al. 2011a). No data on consecutive breeding of males.
One clutch per breeding season. No re-nesting confirmed. The limiting factor determining one clutch / brood per season is probably the long time required to nest, and the long developmental period of young owls. Additionally, clutch size varies and depends on prey density, usually lemmings, but exceptions occur. Clutch sizes are small (3–5) when food is limited, but larger (7–11) when food is plentiful (DP).
Mean clutch from Hooper Bay, Alaska, 7.5, range 5-11 (n = 24) (see Pitelka 1955a, Table 1). Clutch size from a small sample in Barrow, AK, in 1953 was 6.7, range 4-9 (n = 7) Pitelka 1955a). Extreme clutches of 13-15 from Barrow, AK (Brower, in Bailey 1948) have been questioned, but see below. In Barrow, AK, clutches ranged from 3-10 (n = 239) over 22 yr of study (DWH).
Of 11 nests found in Frobisher, Baffin I., Canada, in 1953, complete clutch (n = 5) averaged 9, range 9-10 (Sutton and Parmelee 1956). See Parmelee et al. (1967) for average of 9.8 in 1960 on Victoria I, Canada, (in Voous 1988). In another study on Baffin I. in 1953, mean clutch size was 8 (range 7-9, n = 6; Watson 1957). Mean clutch size from Southhampton I., Canada: 8 (range 5-10, n = 11; Parker 1974). On Ellesmere I., Nunavut, mean 8.4 (range 8-9, n = 14; Hart 1880, in Watson 1957).
On Wrangel I., Russia, clutch size ranged from 3-14, with mean annual variation ranging from 5.4-9.4 (Litvin and Baranyuk 1989, in Vaughn 1992). See Watson (1957), Witherby (1952), Cramp (1985), and Holt et al. (1999) for additional European data.
Annual And Lifetime Reproductive Success
Few data, needs study. Most complete data set from Barrow, AK, 1992-2014 (DWH; Table 3). During those 23 yr, 260 nests found on a 213 km2 study area. Snowy Owls bred in 13 of those years, ranging from 4-54 nests annually. Over these 13 yr, complete reproductive information is available for 239 nests. Clutch size ranged from 3 to 10, pooled mean = 6, and pooled mode 6. Hatching success varied from 39-91%, and nestling success (i.e. those that departed their nest on foot at about 3 wk of age) also varied from 31-87%. Fledging success varied from a minimum that was visually confirmed (48%) to a maximum - those that probably fledged (65%). The maximum number is probably closer to the real value, as those that were not visually confirmed were usually a few days from fledging.
Elsewhere, at one nest of 11 eggs, 10 young fledged; at another nest of 10 eggs, 4 fledged (Parmelee et al. 1967). Of 32 eggs observed by Watson (1957), 31 hatched and 31 young fledged.
No information on lifetime success. In unusual circumstance, a pair of Snowy Owls bred on Fetlar I, UK, from 1967-1975 (Robinson and Becker 1986). This was the first breeding record for Britain. Under intensive observation and protection, the owls were monitored for 9 breeding seasons. During winter 1975, the male disappeared and breeding ceased. Based upon nest observations, researchers believed the pair was the same two individuals for the entire 9 yr. If true, this represents the only records of mate and site fidelity known for Snowy Owls. The “pair” produced 49 eggs of which 44 hatched, and 23 young fledged (2.5 per / nest annually). Of six males that survived into their first winter, none returned to Fetlar.
A few females, however, remained. In Norway, only 35 chicks fledged (were ringed) out of 70-75 eggs from 12 nests (Hagen 1960).
Life Span and Survivorship
Few data; needs study. Difficult to determine due to limited information, lack of long-term breeding studies, and few long-term wintering studies. Nonetheless, winter studies provide hints of life span and survivorship (see Fidelity to Breeding Site and Winter Home Range, below). A captive Snowy Owl in Switzerland lived for at least 28 yr (Schenker 1978), and a captive after-hatch-year owl from the Owl Rehabilitation Foundation, Vineland, Ontario lived ~ 30 yr, and produced fertile eggs until its death from West Nile Virus (K. McKeever, pers. comm). Some individuals may live 15-20 yr in the wild. A female recaptured in Alberta, Canada, in 2013 proved to be > 20 yr old (see Fidelity to Breeding Site and Winter Range).
In a breeding season study from Bylot I., Canada, Therrien et al. (2011a) estimated annual survival over 3 yr at 85 - 92%, for 12 females. Several other unpublished studies of satellite-tracked owls provide additional short-term survival information. All above studies lack long-term results, however.
Historically it was thought that most Snowy Owls migrating south from Arctic breeding grounds died from starvation on their wintering grounds. Suggesting this myth were: the ‘crashes' or population declines in small mammals (i.e. lemmings) that drove the migrations, the energetic demands of migration, and competition among the owls on their wintering grounds. It was also believed that many Snowy Owls arrived on their wintering grounds in semi-starved condition (Gross 1947). This popular thinking, buttressed by some media reports and a small body of scientific literature, has endured (see Kerlinger and Lein 1988b, Holt and Zetterberg 2008). Bent (1938), however, provided convincing evidence that this myth is false; he cited several sources to prove that many Snowy Owls live through the winter and return in following years to the same or other wintering sites.
Recent data suggest survival is the norm; there is no evidence that large numbers of owls die in the northern great plains of Canada or the U.S. (Kerkinger and Lein 1988b), or even in intermountain valleys of the northwestern U.S. (Holt and Zetterberg 2008). Some adults are known to return to the same wintering areas in ensuing years, far south of their breeding range (Oeming 1957, Follen and Luepke 1980; and M. Blom, D.L. Evans, H. Pletz, T. McDonald, N. Smith, M. Stoffel, and D. Zazelenchuk, pers. comm.).
At Logan Airport, Boston, MA, most owls appear in good condition on their wintering grounds, particularly during irruption migrations when the bulk of the wintering population is young owls in their first year of life (N. Smith, pers. comm.).
Of the Snowy Owls that do die on their winter grounds, most are killed by humans, including many of those that are apparently starving. Of 71 wintering Snowy Owls where cause of death could be determined, 61 (86%) died from trauma (shooting, collisions with autos and unknown objects, and electrocution), with only 10 (14%) from apparent starvation (Kerlinger and Lein 1988b). Based on live-trapped individuals, the authors believed that Snowy Owls regularly sustain injuries, but generally heal.
In British Columbia (Campbell and Preston 2009), 177 Snowy Owl deaths were separated into 24 categories. Of these, shooting accounted for 25% of all deaths, assumed starvation for 13%, with 12% “found dead.” Interestingly, museum and private collecting accounted for 11.7%, while other categories such as collisions, electrocutions, trapping, lead poisoning, and others were < 8%.
Of 438 band encounters reported by the USGS banding lab, 150 (34.2%) were found dead from unknown causes; 52 (11.9%) shot; 31 (7.1%) hit by motor vehicles; 24 (5.5%) found dead or injured on highways; 17 (3.9%) collisions with towers or wires; 12 (2.7%) caught in traps not intended for birds; 9 (2.1%) hit by aircraft; and 3(0.6%) entangled. The remaining 146 (33.3%) were recovered injured from miscellaneous causes. No encounters reported starvation, exhaustion, or weather, as causes of death. See also Conservation and Management for mortality at airports and human efforts to remove the species from these locales.
During the breeding season prolonged periods of cold, damp, and wet weather subject nest-bound and nest-departed chicks to several sources of mortality. For example, of two fledglings that died on Fetlar I., UK: one was diagnosed with pneumonia and contained Staphylococcus, and the second was diagnosed with the fungal infection Aspergillosis (Robinson and Becker 1986). In Barrow, AK, nest-departed young are vulnerable to starvation, leading to hypothermia and pneumonia, during prolonged (2-3 d) rain (DWH).
Few data, needs study; but see Dispersal from Breeding Site, below. In Barrow, AK, about 750 chicks were banded as of 2014, with almost none resighted or recaptured to date.
Fidelity To Breeding Site And Winter Home Range
Few data; needs study. Adult females outfitted with satellite transmitters left Barrow, spent following summer in Russia, returned and passed by Barrow en route to Canada (Fuller et al. 2003). In general, the species appears to have low fidelity to nesting sites, as befits a bird that depends on sporadic irruptions in availability of prey.
Some banded individuals known to return to wintering areas in ensuing years (Oeming 1957, Follen and Leupke 1980, M. Blom, D. Evans, H. Pletz, T. McDonald, N. Smith, M. Stoffel, and D. Zazelenchuk, pers. comm.). In Duluth, MN, from 1974-2012, D. Evans banded 419 Snowy Owls during winter in a study area (~ 65 km2) that had persistently available prey (Norway rats). During years with high numbers of rats, owl captures averaged 15.2/yr; when rats were less plentiful, owl numbers declined (9.3/yr).
Evans recaptured 43 of 419 owls previously banded at this site. These 43 owls were wing-tagged and provided positive identification. Of these, 38 returned for 1 consecutive winter, 8 for 2 consecutive winters, 4 for 3 consecutive winters, and 1 for four consecutive winters. Nine returned in non-consecutive years. Additionally, one owl caught in Feb 2011, wintered the following two winters near Edmonton, Alberta, Canada, approximately ~1700 km northwest of the Duluth / Lake Superior site. In this study, zero adult males returned at any time, while 6 (4.7%) of 129 juveniles males returned a subsequent winter; and 16 (11.3%) of 141 juvenile females returned a following winter, and 19 (17%) of 112 adult females returned in a following winter. Two additional wing-tagged females returned in subsequent winters but their wing tags could not be read and their original age unknown. Collectively, sightings of these wing-tagged owls have been recorded in Minnesota, N. Dakota, S. Dakota, Illinois, Indiana, Michigan, Wisconsin; and Alberta, Manitoba, Nunavut, Ontario, and Quebec, Canada.
In another winter study from 1981-2013, N. Smith from Massachusetts recaptured 17 of 452 owls previously banded during winter from Logan Airport, Boston. These owls returned in ensuing winters; 11 returned the following season; 3 two yr later; 1 six yr later, 1 10 yr later, and 1 16 yr later. Additionally, one satellite-tracked owl outfitted at Logan Airport, in February 2012, returned to Nunavut, west of Baffin I., and then in Nov 2012, returned to Logan Airport (N. Smith unpubl. data).
On 23 Jan 2013 near Calmar (Edmonton area), Alberta, Canada, Hardy Pletz recaptured a female owl that was banded by him on 17 Jan 1994 as AHY near Millet (Edmonton area). This owl proved to be at least 20 yr old and about 9 km from its 1994 banding site. This probably represents the oldest wild Snowy Owl known. See also Life Span and Survivorship.
Not known if site fidelity occurs on breeding territories. Satellite tracking data showed that 3 females left Barrow, AK, for Russia for a season, returned through Barrow the following season and moved on to areas around Banks and Victoria islands, Canada. One returned to Barrow the following year, as identified by the satellite transmitter harness still attached to the owl. These long distance movements and apparent breeding in widely disparate locations, complicates the question of site fidelity, as researchers really do not know where home is for many Snowy Owls.
Dispersal From Breeding Site
A Snowy Owl chick banded in 1960 was shot by a native hunter in 1963 within 70 km of its natal ground on Victoria I., Canada (Parmelee et al. 1967). Its sex and breeding status were unknown. In this same area, however, at least one pair raised young that season. Another chick banded on Victoria I. in 1960 was recovered in 1969 at Southampton I., Canada. Its sex and breeding status were not known (Parmelee 1972).
Seven nestlings were banded by Parmelee (1972) on Victoria I., Canada, on 18 July 1960. One was recovered in e. Ontario near Clyde Forks, on 19 Oct 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 Feb 1962—all three recoveries within 7 mo of one another. This extraordinary dispersal of young also explains in part why the species is considered nomadic and monotypic.
DWH (this study) and others attached the first satellite transmitters to Snowy Owls 1999 (Fuller et al. 2003). Six adult Snowy Owls (5 females: 1 male) outfitted with PTT satellite transmitters. Five of six dispersed from Barrow, AK after the breeding season. The male left the study area and his signal was lost when > 100 km south of Barrow, but still in Alaska. One female was killed in the study area. Of the four other females, all dispersed from the study area from late Aug to late Nov (Fuller et al. 2003). These data provided additional information to buttress banding recoveries, of the remarkable travel routes, distances, and wintering sites of some Snowy Owls (National Geographic 2002, Fuller et al. 2003).
For example, female 57, left her Barrow, AK, breeding area in late Aug 1999, flew coastally and crossed the Chukchi Sea to Russia. She spent 4 mo travelling mostly coastal areas. By Mar 2000, she flew overland and north to Chuan Bay on Russia's north coast, and then proceeded westward along the coast and stayed in one area from Apr to late Aug. She may have nested, but we were not able to confirm this. She was 1,928 km from her 1999 nest site in Barrow, AK. She then reversed her route somewhat in late Aug 2000, and stayed around Chuan Bay for 2 mo, then moved inland for Dec 2000 through Apr 2001, and then flew directly overland and over the Chukchi Sea back to the Barrow region by early Apr. She continued east along the coast to Canada, and shifted north to Banks I., by 20 May 2001. She possibly nested, as indicated by her transmitter locations. She was 1,548 km from her possible summer nesting in Russia in 2000. She travelled at least 3,476 km over 2 yr, and possibly nested 3 yr in a row.
Female 54 left her Barrow breeding site in late Nov 1999, and wandered in Alaska until Dec, then flew over the Chukchi Sea and south to settle in and around the Bering Sea and its islands. She remained in this ice-covered area for 2.5 mo. In Feb 2000, she moved overland and eventually westward along the Russian coast, similar to owl 57's route. Her movements indicate she probably did not nest. She then reversed some of her route past Chuan Bay, and overland south to the coastal Chukotsky Peninsula where she stayed from Dec 2000 to Mar 2001. The by 10 Apr 2001 she passed by Barrow, where she bred in 1999. She proceeded eastward (as did female 57) to Canada and then moved north to Victoria I. It does not appear she nested.
Female 81 left her Barrow breeding site in late Oct 2000, and moved south along the Alaska coast, crossed the Bering Sea and settled along the Chukotsky coast from Dec 2000 to Mar 2001. By early Apr 2001, she moved north to the Barrow region and continued east along coastal Canada, and finally moving north to Banks I. by late May, staying there until late Sep. It does not appear she nested.
Female 80 had a totally different strategy. She left her breeding site on 9 Sep and moved east to the U.S. / Canada border and then overland, south to coastal Gulf of Alaska and settled from early Nov 2000 to late Mar 2001 along the coast. She reversed her route in moving north and arrived back at the U.S. / Canada border by early Apr 2001. She stayed there until late May, and then moved north to Banks I. by early Jun, and then to Victoria I. by Aug. She probably did not nest.
Similar results reported by Therrien et al. (2011a). During a 3-yr period, most satellite marked females (n = 17) remained in the Arctic (n = 15), at > 55°N, while two moved to temperate sites. Furthermore, birds moved on average 1100 km (n = 13) in 2007/2008, and 2,900 km (n = 4) in 2008/2009, from their previous summers breeding sites.
In summary, satellite data show these owls covered about 1/3 of their Holarctic breeding range and did not breed or winter in consecutive years in the same areas, and that they may be transient or resident on their winter areas. Perhaps two nested >1000 km between nests in consecutive years, and two did not. All four owls used similar return routes, suggesting they were familiar with their original route. All wintered above 60°N, and 3 of 4 moved / migrated in an east to west direction.
These data also give some indication of how quickly the owls may evaluate an area and decide to stay or move on – likely linked to food. For example, lemming sampling at Barrow, AK for over 22 yr showed that the lemming population was low in 2001. Three of the four owls passed through Barrow in mid-Apr and continued east. Thus, it appeared to take them only a short period of time to decide if lemmings were abundant for breeding.
Few data, needs study. See Behavior: spacing.
Given the nomadic habits of Snowy Owls, population estimates from the breeding and wintering grounds are difficult to gather with any measure of precision. Thus, the data reported below should be taken with caution. Only two studies (Wrangel I., Russia, and Barrow, AK) provide long-term (> 20 yr) consistent breeding season data for the same areas. Fewer than 10 studies provide long-term winter season data for North America (See Banding Records, Mortality and Disease, Range sections, above). On Wrangel I., Russia, I. Menyushina has been studying Snowy Owl since 1986. Although the overall the status of the Snowy Owl population on Wrangel I., has not been evaluated, amplitude variation from year to year is quite variable (Menyushina 1997). The second study from Barrow, AK, derives data from 22 breeding seasons with similar amplitude variation (DWH). See Trends.
Furthermore, the estimates cited below are based upon data from published literature where most studies have been conducted for one or few season, and where chance circumstances have led to discovery of Snowy Owl breeding locations while investigating other species. Additionally, significant recording gaps in time and space, present huge problems in extrapolation over wide geographic areas. Finally, the unpredictable and non-synchronous nature of lemming and vole “cycles”, over wide geographically areas, and the nomadic nature of breeding Snowy Owls, complicate reliable population estimates.
North American Population
In North America, Millsap and Allen (2006) used data from Partners in Flight (Rich et al. 2004) to estimate Snowy Owl populations. They modified the Rich et al. (2004) population estimate by eliminating the visibility correction factor of 50% for Breeding Bird Surveys. Thus, for 2003-2004, Millsap and Allen (2006) estimated the total number of North America Snowy Owls at 72,500, with about 0.30 (n = 21,750) juveniles, as derived from species-specific population models.
On a more regional basis, on Banks I. (about 64,000 km2) in the Canadian Arctic, the population may have reached 15,000 to 20,000 in 1951 (Manning et al. 1956). Whereas Manning et al. (1956) estimated a mean population of 5,000 for 1952 and 1953 with about 2,000 in 1953. Similar population estimate from year to year reported for other Canadian islands (Also Behavior; spacing).
Aerial transects were conducted on 12 Queen Elizabeth Islands, NWT, Canada, in July 1985 and 1986 (Miller 1987): in 1985, 314 Snowy Owls were counted on 9 islands. Extrapolation from the entire survey area yielded an estimate of 932 individuals (Miller 1987). In 1986, only 3 Snowy Owls were counted but only 86% of the land mass was covered in 1986 (Miller 1987).
Based upon previous surveys, Miller (1987) extrapolated Tener's (1963) results of 13 Snowy Owls seen on aerial surveys from 5 Queen Elizabeth Islands, and concluded 166 individuals occurred on the entire area, yielding 5.5 times as many Snowy Owls for his 1985 survey, compared to the 1961 survey (Tener 1963). Similarly, Miller (1987) retro-extrapolated 20 Snowy Owls seen on 3 other Queen Elizabeth Is. from aerial surveys (Tener 1963), and concluded 383 Snowy Owls occurred in 1961 verses 7 Snowy Owls for 1986 surveys (Miller 1987). Miller (1987) concluded lemmings were the driving force behind highs and lows in the Snowy Owl populations on the Queen Elizabeth Islands.
The number of breeding pairs in Canada was estimated at 10,000-30,000 in the early 1990's (Kirk et al. 1995).
The only United States breeding population is in Alaska. There, aerial surveys vary greatly from year to year so population estimates and growth and decline rates are of limited value (see Alaska Natural Heritage Program, Snowy Owl report at; www.AKNHP.uaa.alaska.edu). In Alaska, the AKNHP lists the Snowy Owl globally as G5 (secure) while the state of Alaska Department of Fish and Game lists it as; of S3 / S4 (vulnerable / secure) and SGCN (species of greatest conservation need) (see AKNHP, and ADFG 2005, 2011).
During annual waterfowl surveys along the Alaska Arctic coastal plain, population estimates for Snowy Owl were recorded for 6 yr. Results varied greatly: 0 to 4,048 individuals were estimated. The change in magnitude varied from 268% to > 4,000%. The authors concluded the variation could not be explained by mortality, but likely reflected the owls' numerical response to changing prey densities such as lemmings (Brackney and King 1991).
In a separate population estimate, Harwood et al. (2004) conducted 5 aerial transects in sw. Alaska during an unusual concentration of summering Snowy Owls. They used the number of owls detected along transect length to derive estimates with 95% confidence intervals. Their five transect results were: 1) 1618 km, 15 owls detected, 79 ± 48 owls estimated; 2) 1618 km, 25 owls detected, 131 ± 59 owls estimated; 3) 779 km, 17 owls detected, 185 ± 86 owls estimated; 4) 779 km, 21 owls detected, 229 ± 143 owls estimated, and; 5) 1618 km, 27 owls detected, 142 ± 63 owls estimated.
Aerial transects were conducted in 1972, 1973, and 1974 on 3 Queen Elizabeth islands, NWT, Canada (Miller et al. 1975). Snowy Owl densities: 7.18 / 100 km2 in 1973, 10.28 / 100 km2 in 1974 (Miller et al. 1975). The distance between Snowy Owls on the same transect on the 3 islands was 5.5 km (range 0.8-14.4 km) in 1973; 5.2 (range 0.4-15.6 km) in 1974, and 3.9 (range 0.4-11.6 km) in 1972 (Miller et al. 1975).
Additional aerial transects on 12 Queen Elizabeth Islands were conducted in July 1985 and 1986 (Miller 1987). In 1985, Snowy Owls were counted on 9 islands, and extrapolation from the entire survey area yielded an estimated mean density of 45 owls/1000 km2 (Miller 1987). Collared lemmings appeared to be abundant in these areas. In 1986, only 3 Snowy Owls were counted but only 86% of the land mass covered in 1985 was covered in 1986 (Miller 1987). Extrapolation resulted in an estimate of 7 Snowy Owls for the area covered, or 0.4 owls/1000 km2 (Miller 1987). Zero lemmings were found despite a search for them on one island.
The above data are interesting, but one needs to be careful with extrapolations over large landmasses. If lemmings are as important to Snowy Owls as believed (DWH), and asynchronous in distribution as suggested by Miller et al. (1975), Miller (1987), Pitelka 1993, DWH), then Snowy Owls are unlikely to be evenly distributed throughout large geographic areas.
Probably not Globally Threatened (IUNC 2013), Birdlife International (2013). However, estimating global Snowy Owl populations is extremely difficult and estimates should be interpreted with caution. In 2004, the world Snowy Owl population was estimated at 290,000 individuals (Birdlife International 2004). Data from Partners In Flight (PIF) from 1998-2007, and updated in 2013, version 2.0, estimate the global / world population at 200,000, and North American population at 100,000 (PIFSC 2013).
Population estimates from Scandinavia are perhaps easier to estimate in breeding years, as so few Snowy Owls do breed: Finland = 0-100 pairs; Norway 1-20 pairs; Sweden 1-50 pairs (Koskimies 1998, in Potapov and Sale 2012). According to Birdlife International, the European part of Russia is estimated at 1,300-4,500 pairs, and in Greenland 50-1,000 pairs (in Potapov and Sale 2012).
In Russia, an aerial survey in 1994 extrapolated results to suggest 4,871 (± 1,024 SE) Snowy Owls between the Indigirka and Kolyma rivers, covering 99,280 km2, and the following year 1995, the same aerial survey reported 635 Snowy Owls (Poyarkov et al. 2000, in Potapov and Sale 2012). For further analysis of the data in Russia, see Golovatin and Paskhalniy (2005), in Potapov and Sale (2012). See also Dial et al. (2012) for microsatellite markers and usefulness for population genetics.
Few data but populations reported to be in decline in northern Europe (Voous 1988, Marthison et al. 2011). And, although not threatened worldwide, the Snowy Owl is believed to be declining in the western Palearctic, with a notable vulnerable or endangered status in the Scandinavian countries (see Marthinsen et al. 2008).
Little information for North America, but Snowy Owls are believed to have declined by 52% since the mid-1960's (see Berlanga et al. 2010). In Barrow, AK, a decline in number of nesters has occurred over the past 22 yr but the amplitude varies greatly from year to year. Causes for declines remain unknown on breeding and wintering grounds. Given the dependency of this species on lemming populations for breeding, factors affecting lemmings will have direct consequences on Snowy Owls (see Conservation and Management). Long-term breeding studies from the same locations, and long-term winter monitoring through Christmas Bird Counts, may provide useful data for overall population trends (Root 1988).
Food supply clearly regulates local populations. In Duluth, MN, from 1974-2012, D. Evans banded 419 Snowy Owls during winter in a study area (~ 65 km2) that had persistently available prey (Norway rats). During years with high numbers of rats, owl captures averaged 15.2/yr; when rats were less plentiful, owl numbers declined (9.3/yr).