Montana Field Guide

The Northern Goshawk is a fairly large hawk with a long tail having a broad, dark sub-terminal band and three to four narrower dark bands, rounded wing tips, and a conspicuous pale eyebrow. The sexes are similar with adults having a dark crown, blue-gray back, white underparts with fine, dense gray barring and conspicuous white undertail coverts. The eyes of adults are deep ruby-red and the feet are yellow. Immature Northern Goshawks are brown above, buffy below, with dense, blurry streaking. The undertail coverts are dark-streaked and the tail has wavy dark bands bordered with white and a thin white tip. The eyes of immature Northern Goshawks are yellowish, deepening in color to red as they mature. The total length is 53 to 66 cm, with females averaging larger than males (Squires and Reynolds 1997).

For a comprehensive review of the conservation status, habitat use, and ecology of this and other Montana bird species, please see Marks et al. 2016, Birds of Montana.

The Northern Goshawk is the largest and heaviest bodied of the three North American accipiters. Goshawks are clearly larger than Sharp-shinned Hawks (Accipiter striatus). Male Northern Goshawks can be of similar size to female Cooper's Hawks (Accipiter cooperi), but Northern Goshawks have broader wings and a relatively short tail compared to Cooper's Hawks (Squires and Reynolds 1997). Their ventral surface is pale rather than rust-colored as well. They can be distinguished from both Sharp-shinned and Cooper's Hawks by their whiteish underside as well as a boldly patterned head with a strong, white superciliary line above the eye (Sibley 2000). Juvenile Northern Goshawks can be distinguished from juvenile Cooper's Hawks by their conspicuous pale superciliary line (Squires and Reynolds 1997), overall buffy wash appearance on the breast and belly (Sibley 2000), uneven tail-bands creating a zig-zag pattern when the tail is spread (Squires and Reynolds 1997) and overall paler and more patterned upperside (Sibley 2000). Northern Goshawks can be discerned from falcons by their shorter, more rounded wings, and alternating flap-and-glide flight pattern.

Earliest, peak and latest migration dates for Northern Goshawks in Montana are not well established for either spring or fall, and spring movements are less obvious than fall migration. Northern Goshawks are much less common at migration watch sites than are the other two accipiters, and their numbers vary more from year to year (Mueller et al. 1977). The number counted annually from late August through late October at the Bridger Mountains watch site ranged from two to 96 between 1991 and 2010 (mean = 32). West of the divide at the Jewel Basin watch site, 50 were counted from 31 August to 1 November 2008, 30 from 30 August to 25 October 2009, and 35 from 28 August to 23 October 2010 (Dan Casey, personal communication). The highest numbers occurred after the middle of October in all three years, and the lower counts in 2009 and 2010 may have resulted from the count periods ending sooner owing to heavy snows (Dan Casey, personal communication). The species is generally considered a year-round resident or partial migrant in Montana as Northern Goshawks have been observed in transit during every month of the year (Montana Bird Distribution Committee 2012). In the northeastern quarter of Montana it is an uncommon migrant and winter resident only. Movements are apparently dependent on prey availability and is often only to lower elevations or into more open habitat types (Squires and Reynolds 1997), which may explain the presence of Northern Goshawks wintering at locations within the state east of documented breeding locations. Migration routes throughout the west are poorly defined and are believed to occur over a broad area (Squires and Reynolds 1997). Beebe (1974) believed significant numbers from Alberta, as well as Montana breeding birds, might migrate south along the Rocky Mountain Front Range. Irruptive movements of northern birds to the south occur in approximately 10-year cycles and are apparently dependent on Snowshoe Hare and grouse population lows (Squires and Reynolds 1997).

Goshawks nest in a variety of forest types in Montana, including Douglas-fir and Western Larch west of the Continental Divide, Lodgepole Pine in Beaverhead County, and Ponderosa Pine in Powder River and Carter counties. They prefer mature and old-growth forests with a preponderance of large trees, a dense canopy, and a relatively open understory (Hayward and Escano 1989, Squires and Reynolds 1997, Clough 2000). An exception to this generality is in Beaverhead County, where nests commonly occur in Lodgepole Pine stands with an average tree diameter of only 13 cm, although the birds usually place their nests in larger trees within these stands (Kirkley 1996). The nest is a bulky platform of sticks placed near the main trunk of a large tree from 6-20 m off the ground, usually in the lower part of the canopy. Forest stands where Northern Goshawks nest in Montana tend to be mature large-tract conifer forests with a high canopy cover (69%), relatively steep slope (21%), and little to sparse undergrowth (Kirkley 1996). Hayward and Escano (1989) examined nest-site characteristics at 17 territories in western Montana and northern Idaho in 1983. The birds nested preferentially in mature and old-growth stands of conifers that had a closed canopy (75-85% canopy cover) and a large forest opening within 1 km of the nest. Nest heights ranged from 7-17 m, and most nests were placed next to the main trunk in the lower one-third of the canopy. All Northern Goshawk nest trees reported by Kirkley (1996) were either Lodgepole Pine or Douglas-fir with an average DBH (diameter at breast height) of 33.6 cm and average height of 21.9 meters. In another study conducted in Montana, Douglas-fir, Ponderosa Pine and Grand Fir were the trees selected most often for nest building (State of Idaho HCA/CS Dev. Team 1995). Nests were constructed an average 10.9 meters above the ground and were usually located near water (232 m) or a clearing (85 m) (Kirkley 1996). Range-wide nest site characteristics are similar. Almost no information is available regarding Northern Goshawk foraging strategies in Montana. It is known they hunt in closed canopy habitats as well as more open landscapes and over 50 species of identified prey indicate they are generalists in terms of prey selection. Little information exists concerning Northern Goshawk non-breeding or wintering habitat in the state. However, in the Bozeman area, birds coming into the valley are found in forested or thickety areas. Multiple observations of wintering Northern Goshawks are documented in the north-central and northeastern areas of Montana (Montana Bird Distribution Committee 2012), possibly indicating movement toward areas of higher prey availability.

Goshawks are generalists when it comes to foraging, hunting beneath the forest canopy in dense and open stands and at forest-grassland and forest-shrubland ecotones, where they take a wide variety of prey that includes forest grouse, woodpeckers, corvids, lagomorphs, and squirrels. During winter in Montana, many birds move to grasslands, shrublands, and valley-bottom riparian areas, where they hunt Gray Partridges, Ring-necked Pheasants, Sharp-tailed Grouse, and other prey that frequent open country (J. Marks, pers. obs.). Parratt (1964) observed a nest in Glacier NP in 1956 that was placed in a spruce about 18 m off the ground and contained four young that fledged on 12 August; small mammals (especially chipmunks) were the main prey items. Mammals made up the bulk of the diet of 19 pairs in the Flint Creek Mountains in 1997 and 1998 (Clough 2000), with the major prey items being Snowshoe Hares, Columbian Ground Squirrels, and Red Squirrels. Avian prey included Ruffed Grouse, Dusky Grouse, Northern Flickers, and Gray Jays. Generally, Northern Goshawks forage during short flights alternating with brief prey searches from perches. Elsewhere across their range, dominant mammalian prey includes five species of tree squirrels, four ground squirrels, and lagomorphs. Frequently killed birds include three galliformes, four corvids, six woodpeckers (Piciformes) and the American Robin (Turdus migratorius) (Squires and Reynolds 1997). During the nesting season, the diet can vary with prey availability. For example, as more fledgling passerines become available, they make up a greater portion of the diet (Linden and Wikman 1983, Reynolds and Meslow 1984). The ratio of mammalian prey to avian prey in the diet during the breeding season (in percent) in Arizona was 76:24 and 62:38 (Boal and Mannan 1994, Reynolds et al. 1994). This ratio in Nevada was 67:32 (Younk and Bechard 1994) in New York was 39:61 (Grzybowski and Eaton 1976) and Oregon was 42:59 and 45:55 (Bull and Hohmann 1994, Reynolds and Meslow 1984). Non-breeding season food habits are unknown for North American populations. In Sweden, birds dominate the diet during the nesting season (86 percent of prey), whereas in winter, Red Squirrels (Sciurus vulgaris) comprise the bulk of the diet (79 percent) (Widen 1987).

Northern Goshawks are considered to be decreasing in numbers near Fortine (Weydemeyer 1975). All other ecological information regarding this species, from studies in other areas of the species' range, indicate nesting densities of most western U.S. populations range from 6.6 to 10.7 pairs per 100 square kilometers (Squires and Reynolds 1997). The single nesting density estimate for the eastern U.S. is 1.17 pairs per 100 square kilometers (Kimmel and Yahner 1994). Home ranges during nesting vary from 95 to 3500 hectares depending on sex and habitat characteristics (Squires and Reynolds 1997). Home ranges of males are typically larger than those of females (Hargis et al. 1994, Keane and Morrison 1994, Kennedy et al. 1994). Exclusive of nesting areas, home ranges of adjacent pairs are not defended and may overlap (Squires and Reynolds 1997). The core area encompasses the nest site and constitutes 32 percent of the home range (Kennedy et al. 1994). Individuals typically enlarge or sometimes shift location of home ranges after breeding (Hargis et al. 1994, Keane and Morrison 1994). Home ranges of non-breeders are poorly known, but may be larger than those of breeders (Squires and Reynolds 1997). In North America, winter home ranges are unknown. In Sweden, winter home ranges of males and females were similar and averaged 5700 hectares (Widen 1989). In California, 76.5 percent of males and 71.4 percent of females returned to the same nesting area in subsequent years. Males were significantly more likely to return to previously inhabited territories in consecutive years than females (Detrich and Woodbridge 1994). In Arizona, 80 percent of nest areas examined in two consecutive years were re-used the second year by one or both members of the pair banded the first year (Reynolds et al. 1994). Sixty to 72 percent of adults located in consecutive years retained the mate from the previous year (Detrich and Woodbridge 1994, Reynolds et al. 1994). Dispersal of young is not well documented. Detrich and Woodbridge (1994) recaptured two adult females, banded as nestlings 5 to 7 years prior, 16 and 24 kilometers from their natal sites. Three females, banded as nestlings and recaptured as breeding adults, moved an average of 21.5 kilometers from their natal sites, and another female, captured as a breeding adult seven years after being banded as a nestling, moved 100 kilometers from its natal site (Squires and Reynolds 1997). Little is known regarding survivorship in the U.S. In Arizona, annual survivorship of male and females more than 1 year old was estimated to be 68.8 percent and 86.6 percent, respectively (Squires and Reynolds 1997). In Yukon, Canada, an observed population decline was attributed to increased mortality of eggs, nestlings, immatures and adults, as well as to dispersal following a precipitous decline in number of Snowshoe Hares (Doyle and Smith 1994). The maximum lifespan of a wild bird is 11 years (Squires and Reynolds 1997). The sex ratio is 1:1 prior to fledging and among adults (Mueller and Berger 1968, Reynolds et al. 1994). The BBS is inadequate for assessing population trends, as are counts at migration watch sites (Boal et al. 2005).

Information on Northern Goshawk reproduction in Montana is limited. Eggs are laid from early April to late May, young fledge from late June to mid-August and are independent of parental care by early September. Only one brood is raised per year. Most Northern Goshawk studies have focused on habitat and nest-site characteristics. Very few studies provide information on aspects of nesting biology or reproductive effort/success. The Montana Natural Heritage Program's Point Observation Database indicates most documented Northern Goshawk nests had 1 to 3 chicks/juveniles. This information must be used with caution because in many instances these were just observations of nests with no methodological survey techniques applied. Kirkley (1996) did have a total of 13 young fledge from five active nests and Pilliod (1994) had two nestlings but only one fledgling from a single nest. Egg dates are probably similar to those reported for Wyoming: May 10 to June 17 (Johnsgard 1986). Young have been observed in June and August (Davis 1961). Information from other portions of this species' range indicate that usually one clutch is produced per year, occurring from late April through early May (Squire and Reynolds 1997). However, some individuals may not breed during cold, wet springs (DeStefano et al. 1994).

Egg-laying may begin later at higher elevations and during cold, wet springs (Henny et al. 1985, Younk and Bechard 1994). The average clutch is typically two to four eggs and rarely one to five (Squires and Reynolds 1997). The average clutch size of 44 North American clutches is 2.7 eggs (Apfelbaum and Seelbach 1983). Eggs are laid every two to three days and incubation usually begins after the second egg is laid. Incubation, conducted principally by the female, takes 28 to 38 days and hatching is asynchronous. Principally the female performs brooding and feeding of nestlings. The male brings food to the nest. The young begin flying at 35 to 42 days and become independent at about 70 days (Boal 1994, Squires and Reynolds 1997). Northern Goshawks maintain one to eight alternate nests within a nest area (Squire and Reynolds 1997). Alternate nests range from 15 to 2066 meters apart (Reynolds and Wight 1978, Woodbridge and Detrich 1994). Nesting by subadults is more frequent in expanding populations and less frequent in stable populations (Reynolds and Wight 1978).

The Northern Goshawk is currently listed as a sensitive species in all National Forests and on all BLM lands in Montana. It is also a Forest Plan Management Indicator Species in the Beaverhead-Deerlodge National Forest. Northern Goshawks are a Species of Management Concern in U.S. Fish and Wildlife Service Region 6 (USFWS 1995). However, Maj (1996) reports Northern Goshawk populations in Region 1 are increasing or stable in most forests. The USFWS was petitioned in 1997 to list goshawks under the Endangered Species Act but determined that such listing was not warranted. An independent committee that was formed to review the status of goshawks in the western U.S. concluded that existing data were not adequate for assessing population trends, and that assessing the status of goshawks solely using distribution of late-successional forests is not appropriate (Boal et al. 2005). The latter conclusion resulted from the fact that goshawk habitat use is not restricted to mature and old-growth forest, even though goshawks nest selectively in these habitats in many areas. On balance, more information is needed on population trends and habitat relations, especially with regard to how forestry practices influence these issues.

Although shooting is not considered an important mortality factor (Squires and Reynolds 1997), the problem has not disappeared entirely given that a radio-tagged goshawk was shot in Montana as recently as Jan 2011 (J. Kirkley, personal communication). Timber harvest is the main threat to nesting birds because of their preference for large trees and forest stands with high canopy cover (Squires and Reynolds 1997), although loss of forest stands to fire and bark beetles could degrade goshawk habitat. The effects of timber harvest on large-scale population trends remain unknown (Squires and Reynolds 1997, Boal et al. 2005).