Golden Eagle Aquila chrysaetos Scientific name definitions
Version: 2.0 — Published September 17, 2020
Account navigation Account navigation
Movements and Migration
Welcome to Birds of the World!
You are currently viewing one of the free accounts available in our complimentary tour of Birds of the World. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this account.
For complete access to all accounts, a subscription is required.
The Golden Eagle is highly mobile throughout all stages of life after fledging. Nearly all movement is via flight, though individuals are capable of walking surprisingly long distances. Flight relies on thermal or orographic updraft, with flapping avoided except when necessary. Telemetry studies provide most of the information on the movements of individuals, although surprisingly large numbers of birds are counted at some migration count sites. Much of the knowledge about movement behavior comes from North America and Europe (e.g., 11, 252), and comparatively less is known about movements of birds in Africa and Asia.
Dispersal and Site Fidelity
Natal Dispersal and Philopatry
Initial movements from a nesting site are strongly influenced by the migratory tendency of the population from which it originates. As such, we discuss separately migratory and non-migratory birds.
Dispersal in Migratory Populations
Post-independence movements of migrant individuals from interior Alaska include short to long-distance migrations that span a broad region of western North America (220, see Movement and Migration: Timing and Routes of Migration). A similar study in Norway of 25 nestlings from a partially migratory population recorded post-fledging movements of 10–1,500 km from the nest, with a median date of departure of 21 October (253). Direct comparison of these two studies is difficult because of the partially migratory characteristics of the Norwegian population. Many of the eagles termed “dispersers” in that study traveled south and returned to their natal territory the following spring.
Dispersal in Non-migratory Populations
Post-independence movements at temperate latitudes involve nonlinear wandering beyond the natal area that sometimes follows topography (254, 104). First-year eagles banded (ringed) in Snake River Canyon, Idaho, dispersed from natal areas in nearly all directions (254). Most individuals encountered did not move beyond the boundaries of adjacent states. In fact, 78% of band encounters were < 100 km, and only 1% of band encounters were > 1,000 km from banding locations (254).
All 66 Golden Eagles telemetered as nestlings in the Colorado Plateau and southern Rocky Mountains left their natal territory within one year of fledging (104). Most (67%) moved < 120 km from their natal nest in the first and second year of life. Another 17% moved moderate distances (120–500 km), and 13% moved long distances (> 500 km). Two eagles did not easily fit in the above classification scheme, as they regularly left and returned to their hatching site. In general, long distance travelers were from more arid areas and had lower survival rates than did eagles that moved short distances. Eagles that moved relatively short distances (< 350 km) traveled in apparently random directions. However, birds that moved longer distances tended to travel south, southeast, or east, to areas in northern Mexico or, in one case, Arkansas.
Natal philopatry has been rarely recorded. Radio-telemetry and banding data from non-migratory eagles in Scotland suggest that as eagles reach breeding age, they tend to return to natal areas (255). Eight Golden Eagles banded in the Snake River Canyon nested 1–12 territory widths from their natal territories (254; USGS, unpublished data). Five of those eagles whose identity could be confirmed bred 7 to 65 km from their natal nests (mean 34 km ± 21 SD; 254; USGS, unpublished data).
There is some evidence that male may be more likely to nest near their natal areas than are females (256). Six of seven eagles marked as nestlings in Snake River Canyon and later encountered there as breeders were identified as males (254; USGS, unpublished data). The female who returned to nest in the Snake River Canyon moved 44 km, farther than the average distance moved by males (32 km), but less than the maximum distance moved by a male (65 km). Two males banded in Utah bred 48 and 93 km from their natal nests (American Eagle Research Institute, unpublished data). A female banded in southwestern Montana first nested 61 km from its natal nest (R. Crandall, unpublished data), and three females banded in the Colorado Plateau-Southern Rocky Mountain region of the southwestern United States bred 19, 66, and 124 km from their natal nests (256). In contrast, a genetic study in Scotland suggested that long-distance dispersal was male-biased (257).
Adult Fidelity to Breeding Site and Dispersal
Adult eagles often show fidelity to breeding sites (176, TAM). Most telemetered adults repeatedly use the same nesting territory for the entire time their telemetry unit functions (TEK, CLM, TAM), but tagged birds occasionally switch territories (176; G. Hunt, personal communication). One marked individual remained on the same nesting territory for ≥ 12 years in Idaho (Snake River Canyon, USGS, unpublished data). Closely related Imperial Eagle and Spanish Eagle are similarly faithful to nesting territories, with generally low rates of territory switching (258, 259). White-tailed Eagle, which is more distantly related and has different ecology, appears to switch nesting territories more frequently (260).
There are few records of eagles moving from one nesting territory to another (i.e., breeding dispersal). This behavior may be under-appreciated and the reason for these movements are not well-understood. A radio-tracked female used nesting territories 15 km apart in consecutive years in Snake River Canyon (USGS, unpublished data). Three individual territory-holders (one female, two males) tracked by radio telemetry switched territories in California. These birds settled in different territories within 8 km of their old ones, generally 10 days to 3 months after leaving their original territories. Habitat change near the original nesting site may have precipitated one of these moves (G. Hunt, personal communication).
Fidelity to Overwintering Home Range
Both residents and migratory individuals often show fidelity to wintering areas (261). For example, 10 eagles wintering in central Idaho were recaptured within 1.6 km of sites where they were first captured 1–5 winters earlier (EHC, T. Craig, unpublished data). Similarly, two adult migrants from Montana (n = 1 male) and Alaska (n = 1 female) returned to their respective wintering sites in Idaho for ≥ 2 consecutive years (EHC, M. Fuller, T. Craig, L. Schueck, J. McKinley, USGS unpublished data). Similarly, telemetered migratory eagles in eastern North America regularly returned to wintering sites they used the prior year (TAM, TEK, unpublished data).
Younger individuals appear to show more variable fidelity to summer and winter ranges than do older birds. Pre-adult eagles typically do not show strong winter site fidelity until the third or fourth winter (TAM, CLM, unpublished data). In many cases juveniles and some second winter eagles settle farther south and possibly explore other wintering sites during their first or second northward migration. In subsequent years they tend to settle in an area explored during a prior spring migration. However, as additional telemetry data are collected, variations on this strategy are apparent (CLM). For example, preliminary results based on a sample of ~75 individuals of all age classes (HY to adult) telemetered in Alaska suggests that there is substantial individual variation in fidelity to winter ranges. In this sample, some individuals used the same winter range each year, some individuals return to the same general area each year, and others use different winter ranges in different years (Alaska Department of Fish and Game, National Park Service, and U.S. Fish and Wildlife Service, unpublished data).
Populations may be non-migratory, partially migratory, or short- medium- or long-distance migratory. Island populations (e.g., the United Kingdom) are generally non-migratory, and birds tend to stay on or near territories year-round. Birds from non-migratory mainland continental populations may make longer-distance movements from territories more frequently than do their island counterparts. Migrants may cover short distances (e.g., within Scandinavia) or extremely long distances (e.g., from northern Alaska and Canada to southern United States and Mexico), and may travel in a north-south or south-north direction in either season.
Individuals from northern breeding areas are usually migratory (32, 262), and northern territory holders tend to migrate longer distances than individuals nesting farther south (263, 264, 265). Northern eagles can migrate south > 5,000 km from breeding to wintering areas (266, 220, 267, 252). In eastern Canada, nearly all individuals are migratory (TAM, TEK). In western North America, it appears that individuals that nest south of 55°N often are not migratory, and those north of there often are migratory (the authors). In Scandinavia and Alaska, populations are partially or wholly migratory, likely in response to prey availability. For example, some eagles from Alaska may overwinter in interior and northern parts of the state when snowshoe hare (Lepus americanus) are abundant (172, CLM, EHC; T. Booms, personal communication). Individuals from south-central Alaska (T. Booms, unpublished data), western Alaska (CLM) and the Gaspé Peninsula of Quebec may stay close to northern breeding ranges some years and migrate south during other years (179, 265).
Migration patterns of eagles in arid areas of the southwestern North America appear different from those farther north. In these populations, pre-adults travel through New Mexico and western Texas at the beginning and end of winter (November and March; 268). Tracking data suggest that northward movements of these birds may be considered migratory (103, 104). Published telemetry studies suggest extensive overlap in western North America between individuals from migratory and non-migratory populations (see Distribution: Non-breeding Range). There is no information as to whether eagles from arid areas of southern Europe and the Middle East engage in similar behaviors.
Timing and Routes of Migration
Telemetry data suggest that Golden Eagle leaves northern areas of North America as early as August or as late as November. Date of departure appears to vary by age and social class. Specifically, non-territorial individuals often leave earlier than territorial individuals, failed breeders leave earlier than successful breeders, and young-of-the-year and territorial adults leave last (CLM, 265).
Individuals from Denali National Park and Preserve, Alaska, leave in September or October, soon after the onset of hibernation of ground squirrels and hoary marmot (Marmota caligata; 269, 220). Several cohorts of juveniles from this area initiated migration between mid-September and early October (mean departure date = 24 September ± 5 d [n = 43]; 269, 220). Those birds reach their southern wintering areas 28–58 days after departure (mean 44 d ± 9 d [n = 26]), arriving on winter ranges throughout western North America between 26 October and 19 November (mean 7 November ± 7 d [n = 16] 220). Three second-year eagles in this study arrived on winter ranges from 1 to 8 October in 2000 (mean 5 October ± 3.9 d) (CLM).
Eagles tracked by telemetry from summering grounds in eastern Canada departed between 11 August and 17 December (265). Mean departure date was 15 October ± 15 d (SD) (n = 26) for adults; 5 October ± 23 d (n = 18) for sub-adults; and 12 October ± 15 d (n = 2) for juveniles (TAM, 265). Duration of migration of these birds is highly variable with large overlaps among age classes (265). Some individuals of each age class wander and use short-term stopover sites for several weeks during autumn migration (TAM). These eagles reached their wintering grounds in 17–153 d (mean 53 ± 28 d; n = 44) (TAM). Adults reached the wintering grounds in 52 ± 28 d (n = 24), arriving between 23 October and 16 February (mean 9 December ± 28 d). Pre-adults reached the wintering grounds in 56 ± 30 d (n = 18), arriving between 16 September and 2 February (mean 28 November ± 32 d). In another smaller study, radio-tagged adults (n = 4) from eastern Hudson Bay took 26–40 d to reach their wintering areas in central Michigan, eastern West Virginia, northwestern Pennsylvania, and northeastern Alabama, arriving in November or December (186).
Substantial (> 1,000 individuals) numbers of migrating Golden Eagle are counted at multiple count sites in the eastern and western United States during autumn migration. Some of these count sites have been operating for many years or decades (i.e., Hawk Mountain Sanctuary, Pennsylvania; Bridger Mountain in western Montana; Mt. Lorette in southern Alberta), and others only for a few years (i.e., Big Belt Mountains in Montana). Together, counts at these sites provide insight into the magnitude and timing of Golden Eagle migration. In general, autumn flights occur from early September through December at count sites in southern Canada and the northern United States. Peak flights are in October in western North America, but in November in eastern parts of the continent.
Median dates of peak passage as recorded at migration count sites in western North America tend to be in October, with adults often migrating a few days later than pre-adults. In Glacier National Park, Montana, maximum passage rates occur in early October (270). The largest numbers of migrating Golden Eagles are detected at Mount Lorette in the Front Ranges of the Rocky Mountains in southern Alberta (243). Median passage dates for all migrating Golden Eagles detected at this site from 2012 to 2018 range from 12–15 October (P. Sherrington, unpublished data; and Eagle Watch). Eagles arrive on wintering areas in New Mexico and western Texas in October, reaching peak numbers from December to February (268).
Finally, there appears to be within-day variation in timing of fall migration. In western Montana, younger birds are seen more frequently early the morning and late in the day than are older birds (271).
Exhibits age-structured seasonal timing of migration, with adults leaving wintering areas earlier than juveniles. This is reflected in both telemetry and observational data, and in eastern and western North America.
Telemetry data suggest that adults begin to migrate from wintering areas in southwestern Idaho between about 4 March and 13 April (mean 19 March ± 12 d [n = 8]; ECH; M. Fuller, USGS, T. Craig, unpublished data). Juvenile eagles originally telemetered in the nest in Denali National Park and Preserve, Alaska, depart wintering areas between 6 April and 8 May of their first winter (mean 21 April ± 12 d [n = 12; 220]). These first-year eagles return to northern latitudes from 10 May to 13 June, up to 12 weeks after territory holders in the same region (220). Travel time between Idaho wintering areas and breeding areas in Alaska and northwestern Canada, ranges from 6 to 27 days (mean 17 d ± 7 d [n = 8]; EHC, M. Fuller, T. Craig, unpublished data). Travel time between wintering areas and summering areas in Alaska and northwestern Canada ranges from 22 to 47 days (mean 35 d ± 6 d [n = 12]; CLM).
In eastern North America, there is little temporal overlap in timing of spring migration between territorial and non-territorial individuals, and no overlap between adults and juveniles (265). Adults migrate earliest, beginning migration between 4 February and 13 April (mean 10 March ± 11 d; n = 47); subadults migrate from 14 February to 15 May (mean 29 March ± 18 d; n = 41), and juveniles on their first northbound migration migrate from 22 March to 10 May (mean 19 April ± 13 d; n = 18) (TAM). Adults arrive on their breeding grounds after 29 d ± 18 d (range 4–92 d; n = 26), between 20 February and 21 July (mean 13 April ± 21 d; n = 28). These arrival dates are likely determined in part by latitude and breeding status. Subadults arrive on their summering grounds after 32 d ± 23 d (range 7–123 d; n = 24) between 3 March and 12 July (mean 1 May ± 25 d; n = 24). Juveniles on their first northbound migration arrive on their summering grounds after 37 d ± 14 d (range 20–55 d; n = 7), between 7 May and 23 June (mean 30 May ± 14 d; n = 8). These later migrating pre-adults travel different migratory pathways than the adults (239, 265), and this can explain why age-ratio data from spring migration count sites may not be accurate indicators of demographic trends.
Observational data suggest that Golden Eagle leaves wintering areas in the southwestern United States in March (268). Median passage date at Sandia Mountains, New Mexico (1985–2008) was 20 March ± 3 d (± 95% CI; 272). Farther north in the Rocky Mountains, maximum passage rates occur during the second and third weeks of March in Glacier National Park, Montana (270) and from 8 to 27 March at Rogers Pass, Montana (273). Farther north again, at Mount Lorette in western Alberta, spring migration spans ≥ 92 days (274). The first migrants are seen in mid-February, followed by large movements of pre-adults from mid-April to May (274). Peak passage there is from 19 to 25 March (275). Finally, over two seasons of observations at Gunsight Mountain in south-central Alaska, peak passage of breeding age Golden Eagles occurred between from 22–31 March and, in one of those years, 95% of all birds were detected before 10 April (276).
Count data also indicate that adults in western North America usually migrate earlier in spring than pre-adults. Median passage date from 1992–2001 for adults at Sandia Mountains, New Mexico was 9 March ± 2 d, significantly earlier than for pre-adults, whose median passage date was 5 April ± 4 d (277). The proportion of pre-adults migrating at Glacier National Park increased with time from mid-March to mid-April (270). Adults move through northwestern Alberta earlier than pre-adults, with adults common in March and younger birds more frequently seen in April (275). Finally, observations during spring migration in 2016 and 2017 at Gunsight Mountain in south-central Alaska suggest that most birds were adults that pass by over a period of 3 to 4 weeks starting about the second week of March (278; P. Fritz, personal communication). This suggests that counts of migrants at Gunsight Mountain are composed primarily of territory-holders and these timing data may not be applicable to other age classes who migrate later, possibly across a broader area (CLM).
Territory holders and older pre-adults in Alaska arrive earlier than other individuals, with most coming to the breeding range in a narrow window from late February to early April (172, 131, CLM, Alaska Department of Fish and Game [AFG], National Park Service [NPS], USFWS, unpublished data). A second pulse of migrants, usually younger birds, return to Alaska from late March into early June.
In eastern North America, median passage dates at Tussey Mountain, Pennsylvania are 10 March (9–11 March; 95% CI) for adults and 19 March (19–21 March 95% CI) for pre-adults (N. Bolgiano, unpublished data). Golden Eagle migration through the Mackinac Straits, Michigan occurs from 7 March–17 May, with median passage of adults from 9 March–3 April (range 7 March–27 April) and median passage date of pre-adults from 17 April–9 May (range 28 March–17 May) (E. Pike, personal communication).
Major migratory routes of North American Golden Eagles occur along the Pacific Coast Ranges, the Rocky Mountains, the Appalachian Mountains, and through the Great Lakes region (252). Details about these migrations vary by season. This seasonal variation in use of routes is likely driven by corresponding variation in availability of updrafts that facilitate flight and in the motivation for migration. For example, in the Appalachians, thermals become more available as spring progresses adults are driven to return to nesting grounds sooner than are subadults (246, 265).
Migration routes across the interior and south-central regions of Alaska and the southern portion of the Yukon, Canada, tend to run east-west and northwest-southeast (247). However, starting in northern British Columbia, migration routes are predominantly north-south to northern Mexico. Autumn and spring migration corridors in the northwestern portion of North America in Alaska, Yukon and Northwest Territories generally tend to be broad (220, 252, 279, 247; NPS, USFWS, AFG, unpublished data; Figure 2). For example, the migration corridor spans about 400 miles across the southern Yukon, northern British Columbia, and northwestern Alberta (220, 279). At 55°N latitude, the migration corridor spans ~300 miles from central British Columbia to the eastern slopes of the Rocky Mountains in Alberta. The corridor becomes broader again 49°N latitude at the border of Canada and the coterminous United States.
There is also good evidence of a migration corridor along the coastline of southeastern Alaska and western British Columbia (220; NPS, USFWS, USGS, unpublished data; TEK, TAM, EHC). Numbers of eagles migrating through this area are probably far less than those migrating inland to the east of the Coast Mountains (CLM).
Data from telemetry and years of visual observations clearly show that large numbers of migrating eagles concentrate at specific sites. In Alaska and Canada, these sites include the Matanuska-Susitna Valley and Gunsight Mountain in south-central Alaska (280, 276; B. Dittrick and T. Swem, personal communication); the Mentasta Mountains in eastern interior Alaska (281); the Shakwak and Tintina Trenches, Kluane Lake and Ruby Range regions in southern Yukon, Canada (Sinclair et al. 2013); the northern Rocky Mountains including the Cassiar Mountains in northern British Columbia; and the northern Rocky Mountains and Canadian Rocky Mountains in western Alberta, including Mt. Lorrette in the southern portion of the Canadian Rocky Mountains in Alberta (Rocky Mountain Eagle Foundation, unpublished data). There are also a number of concentration areas in the conterminous United States (for details, see 282).
Telemetry studies have provided detailed information on autumn migration routes in eastern North America (186, 265, 252). Adults nesting in Quebec and Labrador typically migrate south, concentrating along the north shore of the Gulf of St. Lawrence and along the ridges of the Appalachian Mountains. Some individuals, especially younger birds, move down the Atlantic coast. Golden Eagles from western Quebec migrate through the Great Lakes region, passing by the Mackinac Straits, Michigan, through the Michigan Upper Peninsula into Wisconsin or along the north shore of Lake Erie. Individuals from Ontario and Manitoba typically migrate south to the north shore of Lake Superior and pass by Hawk Ridge, Duluth, Minnesota, but some pass through the Mackinac Straits as do birds nesting farther east. Migration through the Midwest south of the Great Lakes is typically dispersed but appears to follow forested and riparian corridors (TAM).
Spring migration corridors in the eastern United States occur primarily along the ridges of the Appalachian and Allegheny Mountains, especially early in the season. The corridor widens as spring progresses due to higher thermal activity that promotes soaring away from ridgelines (246, 265). Regularly recorded at migration count sites in Pennsylvania, western Maryland, Virginia, West Virginia, North Carolina, Tennessee, and along the shores of Lake Ontario in New York (186, 283). Spring migration also occurs in the Great Lakes region. At Whitefish Point, Michigan, increasing numbers were observed for several years after 1986 (284). The highest spring counts in eastern North America occur at the Mackinac Straits in Michigan and at Tussey Mountain in central Pennsylvania. Golden Eagles are rare at coastal migration count sites (i.e., Sandy Hook, New Jersey; Cape Henlopen, Delaware; Plum Island, Massachusetts) and at migration count sites in Massachusetts, Vermont, and Maine (283, 285).
Age-Related and Inter-Seasonal Differences in Migratory Routes
Migration pathways of younger age classes, particularly juveniles, from Alaska may be more diverse and dispersed than those of older age classes (220, Alaska Department of Fish and Game, U.S. Fish and Wildlife Service, National Park Service, unpublished data). Route fidelity is variable, with some individuals using the same route repeatedly across years and others changing routes over time (CLM; T. Booms, personal communication).
Spring migration routes of territorial individuals are often similar to routes taken during autumn migration (CLM). Spring migration routes of non-territorial individuals are often similar to autumn routes, but routes become more dispersed at the northern end of the migration corridor as eagles approach the terminus of their migration in Alaska (220; U.S. Fish and Wildlife Service, National Park Service, unpublished data). First-winter birds from Denali National Park and Preserve, Alaska tended to reverse their autumn migration route until reaching central Alberta, where they veer east of the Rocky Mountains, flying north through Alberta, across central Yukon, and to summer ranges throughout Alaska and the northern Yukon (220).
Telemetry data suggest that more experienced (≥ third-year) eagles in eastern North America follow a similar route in spring as they do in autumn, but tend not to concentrate along the southern shores of the Great Lakes. Migration becomes more dispersed in April and May when thermals are stronger, allowing birds to stray from ridgelines, and when young, inexperienced birds are migrating north. For these reasons, more individuals, especially of younger age classes, are counted at Lake Ontario migration count sites (Braddock Bay and Derby Hill, New York) during April. Most older Golden Eagles are counted in March at ridge top migration count sites in the Appalachian Mountains (265).
Age Ratios along Migration Routes
Age ratios of eagles recorded at autumn migration count sites vary geographically. Mean pre-adult:adult ratio are 0.27 in Alberta (P. Sherrington, unpublished data), 1.03 at Hawk Mountain, Pennsylvania (Hawk Mountain Sanctuary, unpublished data), 1.2 ± 0.8 at Bridger Mountain, Montana (± 95% C.I.; 286), 1.38 ± 0.37 at Wellsville Mountains, Utah (287), 2.19 ± 0.361 at Goshute Mountains, Nevada (288), and 2.3 ± 0.42 in Manzano Mountains, New Mexico (289). Most individuals observed at coastal migration count sites in the eastern United States are juveniles (290) adults and subadults tend to migrate farther inland (5).
Pre-adult:adult ratios in spring are 0.08 at Rogers Pass, Montana (273), 0.12 in Alberta (P. Sherrington, unpublished data), 0.16 at Gunsight Mountain, Alaska, (but individuals were difficult to age accurately; 276), and 1.56 ± 0.46 (95% CI) in the Sandia Mountains, New Mexico (272). Differences between pre-adult:adult ratio in fall and spring at raptor migration count sites might reflect age-specific differences in survival rates, migration routes, or detection probability, or even the difficulty of accurately aging a migrating eagle. Alternatively, these differences also may reflect differences in flight strategies of earlier migrating older birds (who use more orographic updraft to subsidize flight and thus are more visible to migration counters) versus later migrating younger birds (who migrate when thermals are more prevalent and thus may be harder for counters to see; 246, 265).
Flight Behavior During Migration
The Golden Eagle is a diurnal migrant. It does not migrate in flocks (271), but may join small “kettles” near thermals (245); at eastern migration count sites it is often observed in groups of 2–3 individuals (TAM) and in eastern Alaska up to 30 individuals have been noted using thermals during spring migration (CLM).
Up to 137 eagles/hr have been counted in northwestern Montana in autumn (270). Mean number of eagles seen/hr (± 95% C.I.) in autumn are 0.06 at Hawk Mountain, Pennsylvania; 0.25 ± 0.03 in the Manzano Mountains, New Mexico; 0.39 ± 0.03 at the Goshute Mountains, Nevada; 0.45 ± 0.12 at the Wellsville Mountains, Utah; 5.2 ± 0.56 at Bridger Mountain, Montana; and 5.4 near Mount Lorrette, Alberta (291, 292, 293, 289; Hawk Mountain, unpublished data; P. Sherrington, personal communication). Peak hourly counts of > 100 Golden Eagles were made repeatedly in early October in eastern Alaska (281). Within-season variation in hourly rates of passage is known to be related to movement of weather systems in eastern North America. Count data suggest that eagle passage rates peak one day after a cold front and subsequently decline steadily for the next three days (294).
Golden Eagle migration responds to the physiographic environment over which the birds pass. Golden Eagles use orographic and thermal updraft (262) along ridges , but also migrate over large flat or featureless terrain using thermal updraft and continuous gliding flight (270). Flight altitude and behavior of migrating eagles varies with weather and topography (5, 295). Golden Eagles fly lower over summits and cliffs and typically subsidize flight with orographic updraft. In contrast, when flying over side slopes, gentle hills, and flats, they use thermal updraft to subsidize flight and can obtain altitudes to just below the boundary layer (296, 297). Use of thermals decreases when higher wind speeds break down thermals but create orographic updraft (295). During spring and autumn migration in northern Canada, tends to avoid flying over high elevation ice fields and snow-covered mountains (247).
Congruent with their reliance on updraft, the Golden Eagle tends to avoid crossing water that generates little updraft, and large water bodies appear to divert migrating eagles. In eastern North America, Golden Eagles tracked by telemetry appear to get “trapped” on large peninsulas, including the Gaspé Peninsula and the thumb of Michigan. These birds sometimes circle the peninsula many times before retreating or making a long water crossing (TAM). When crossings do occur, they almost always occur near islands.
In eastern North America, the longest water crossing recorded by telemetry was of an adult female crossing 32 km of Lake Superior during November (TAM). Ten individuals crossed 20–26 km of open water in the Gulf of St. Lawrence. These crossings were only in spring and may have been over partially or fully-frozen waterways. Three individuals crossed two times each. Two females, a third-year and a fourth-year, appeared to die crossing 21 and 22 km of open water. Telemetry data suggest they were within 500 m and 1 km of shore when they went into the water (TAM). The cause of death is unknown but could have been caused by exhaustion or interspecific interactions. As an example, observers watched a juvenile Golden Eagle that was attacked and forced into the water by a Bald Eagle as it migrated across Braddock Bay, Lake Ontario, New York (J. Soprano, personal communication).
Migration includes periods of wandering and stopovers (243), but many individuals migrate with little or no wandering (298, 186, 220, CLM). Adults from eastern Canada migrate more directly to their wintering areas than do younger birds (265). Some eastern individuals, particularly those migrating through the Midwest to the southeastern United States, stop for long periods of time (> 3 weeks) before proceeding to their final wintering area (TAM). One adult male captured in Pennsylvania and tracked for 4 years spent the first month of each winter in southwestern Virginia and then migrated north to Pennsylvania to spend the rest of the winter.
Eagles tracked by telemetry in eastern North America traveled faster during spring than during fall (265). Over the entire fall migration, including stopover periods, adults traveled at a rate of 66 km/d ± 8 SE, subadults 63 km/d ± 7 SE, and juveniles 74 km/d ± 28 SE. During spring, adults traveled 110 km/d ± 10 SE, subadults 101 km/d ± 10 SE, and juveniles 102 km/d ± 19 SE. When actively engaged in migration, ignoring periods of stopover, during fall adults traveled 98 km/d ± 7 SE, subadults 95 km/d ± 9 SE, and juveniles 106 km/d ± 19 SE. During spring, adults actively migrated at a rate of 134 km/d ± 7 SE, subadults 130 km/d ± 9 SE, and juveniles 129 k/d ± 11 SE. During fall, adults spent 56 ± 5% of days migrating, subadults 54 ± 3%, and juveniles 57 ± 11%. During spring, adults spent 74 ± 4% of days migrating, subadults 69 ± 4%, and juveniles 66 ± 8%. The maximum distance traveled in one day was 508 km during spring and 412 km during fall (299).
Migratory connectivity geographically links individuals and populations among life cycle stages (300). When migratory connectivity is strong, individuals remain close to each other in all seasons of the year (i.e., individuals that summer in the same area also winter together). When connectivity is weak, individuals that are together in one season are not together in other seasons (i.e., individuals that summer in the same area do not winter together, or, conversely, eagles that winter in the same area do not summer together).
There have been no formal assessments of the strength of migratory connectivity of North American or Eurasian populations. However, it is possible to draw some inference about migratory connectivity from literature describing Golden Eagle migration (301, 252, 279). Twenty-eight eagles hatched and telemetered in Denali National Park and Preserve, Alaska, wintered at latitudes from central Alberta to New Mexico (220). However, of those birds tracked into the following year, all summered north of the Brooks Range, Alaska. Fifty-two Golden Eagles captured at wintering locations spanning the Appalachian Mountains from New York to Alabama summered in habitats that covered the northern two thirds of Quebec and much of Labrador (144). These trends suggest low levels of migratory connectivity of Golden Eagles within each of these regions. However, when considered at a continental scale, there appear to be moderate levels of connectivity (252). For example, Golden Eagles that summer in northeastern Canada also spend the winter with other eagles from the region in the eastern United States and do not appear to interact with birds that summer in Alaska and northwestern Canada. Furthermore, eagles from central Canada (Manitoba, Ontario) tend to winter farther west than do eagles that summer in northeastern Canada (TAM, unpublished data).
Control and Physiology of Migration
There are few data on proximate cues underpinning Golden Eagle migration. The tendency to migrate is strongest at northern latitudes (262). Departure from northern breeding areas tends to coincide with the first lasting snowfall, freeze-up, decreasing prey abundance, or the presence of north winds (186, 269). Poor winter foraging conditions due to low jackrabbit abundance in the Great Basin may stimulate migratory movements among otherwise sedentary eagles (291).