Northern Prairie Wildlife Research Center
Habitat and History. The Swainson's hawk occurs in the western half of North America where it adapted to the prairie (Palmer 1988b). This hawk was widely distributed and probably the most abundant buteo in the prairie pothole region before settlement (Houston 1978; Houston and Bechard 1983). The extensive conversion of native grasslands to cropland and the expansion of the aspen parkland have altered nesting habitats and reduced habitat of the major prey species, especially Richardson's ground squirrels (Spermophilus richardsonii; Houston 1978; Bechard 1982; Gilmer and Stewart 1984; Houston and Bechard 1984; Schmutz 1989). Nevertheless, in Alberta, Schmutz (1989:368) found that "Swainson's hawks were largely opportunistic" and "preferred cultivated areas over grassland regardless of the extent of cultivation." Todd (1947:393) reported that Swainson's hawks were "common and generally distributed" in southern Saskatchewan in 1932, but Farley (1932:25) reported that in south-central Alberta, "the Swainson's hawk is not nearly as plentiful now as it was twenty years ago." Houston and Bechard (1983) presented convincing evidence that populations declined substantially in the aspen parkland of Saskatchewan. However, Schmutz (1984) found little evidence of a decline in Alberta. Schmutz (1989) reported a 57% increase in the nesting population of Swainson's hawks from 1982 to 1987 in a study area in southeastern Alberta following an increase in abundance of Richardson's ground squirrels.
Swainson's hawks often nest in tall shrubs or low trees and are not averse to nesting near farmsteads or other areas of human activity (Schmutz et al. 1980; Gilmer and Stewart 1984; Schmutz 1984; Bechard et al. 1990). Because of these factors, their nests were particularly vulnerable to egg collectors during the early 1900's (Bechard and Houston 1984). Nests and adults were especially vulnerable to destruction (Cameron 1908; Judd 1917; Williams 1946); shooting hawks was a common practice until the 1950's (Farley 1932; Kirkpatrick and Elder 1951; Bird 1961; Salt and Salt 1976). These activities probably reduced the abundance of Swainson's hawks in the region (N. Criddle 1929).
Population Structure. Swainson's hawks are strongly territorial and space their nests widely (Dunkle 1977; Schmutz et al. 1980; Rothfels and Lein 1983). Estimated densities of nests in several areas are available. Lokemoen and Duebbert (1976) found an average density of 0.004 occupied nests/km2 in a 269-km2 area in north-central South Dakota during 1973 and 1974. Gilmer and Stewart (1984) estimated an average density of 0.007 occupied nests/km2 in a 16,519-km2 area of south-central North Dakota during 1977-79. Schmutz et al. (1980) found an average density of 0.18 nesting pairs/km2 in a 335-480-km2 grassland area in southeastern Alberta during 1975-77, and Rothfels and Lein (1983) reported a density of 0.15-0.22 nesting pairs/km2 in a 150-km2 area in southwest Alberta during 1979-80. Schmutz (1984) estimated a density of 0.05 occupied nests/km2 in the prairie region of Alberta in 1982.
Distribution and Abundance. Swainson's hawks were present each year in all study areas except in western Minnesota (Fig. 18). The percentage of study areas with common or more abundant Swainson's hawks did not differ between Canada and the United States (Appendix Table 6). In Canada, they were common in seven study areas (44%) and uncommon in eight (50%). Abundance was undetermined in one (6%) study area (Holden) where four nests were found (Appendix Table 15). In the nine study areas in the United States where nests of raptors were censused, the abundance of Swainson's hawks was common in four (44%), uncommon in two (22%), and undetected in three (33%). The species composition data in the eight Central Flyway study areas from which no data on nests were available indicated the size of the populations there were similar to the size of populations in other study areas in the prairie (Appendix Table 14).
Swainson's hawks were common or more abundant in more study areas in the prairie than in the aspen parkland (Appendix Table 7). The three study areas in which they were not detected were in aspen parkland. Nest densities in completely searched study areas (excludes 1983 data) averaged 0.03 occupied nests/km2 (SD = 0.04) in the aspen parkland, but 0.11 occupied nests/km2 (SD = 0.05) in the prairie (calculated from data in Appendix Table 15). The highest density of occupied nests was 0.24/km2 (six nests) in one study area (Ceylon) in the prairie in 1984 (Appendix Table 15).
There were notable exceptions to the pattern of highest densities of Swainson's hawks in study areas in the prairie. For example, in 1984 nest densities of 0.12/km2 in two study areas (Hay Lakes, Inchkeith) in aspen parkland were the highest densities in any study area, but no occupied nests were in one study area (Cartwright) in prairie (Appendix Table 15). The high density of nests in the first area (Hay Lakes) was unexpected because it was one of the most heavily wooded areas (Table 1) and in the center of the aspen parkland of Alberta. The second study area (Inchkeith) had little woodland and was near the interface of the prairie and aspen parkland. Although no nests were in the third study area (Cartwright; Appendix Table 15), Swainson's hawk were frequently observed there (Appendix Table 14).
Swainson's hawks were 0-74% of the sightings of large hawks in study area-years and were the most commonly observed large hawks in 11 (33%) study areas (Appendix Table 14). All but one study area in which they were the most frequently seen hawks were in the prairie. Populations in areas studied >1 year varied little between years (Appendix Table 15). Occupied nests were widely spaced in all study areas; only once were two occupied nests within 0.8 km of each other.
Fig. 18. Density of occupied Swainson's hawk nests (nests/10km2) in study areas in the prairie pothole region, and ratings of abundance of Swainson's hawks in each area. Results from each area are for >=1 year, 1983-88 (Appendix Table 15); results from areas studied in > 1 year were averaged.