USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Distribution and Abundance of Predators that Affect
Duck Production: Prairie Pothole Region

Factors of Abundance and Distribution

Inter- Intraspecific Interactions

Relations between predator species range from antagonistic to beneficial. Of particular consequence to duck production in the prairie pothole region is that gray wolves suppress the abundance of coyotes (Mech 1970; Berg and Chesness 1978; Fuller and Keith 1981; Carbyn 1982) and coyotes suppress the abundance of red foxes (N. Criddle 1929; Sargeant 1982; Voigt and Earle 1983; Sargeant et al. 1987a; Harrison et al. 1989). The primary mechanism for suppression seems to be interference competition (Voigt and Earle 1983). Hence, extirpation of the gray wolf from the region by the early 1900's was a major causative factor of subsequent changes in the distribution and abundance of other canid species in the region.

Effects of interspecific interactions among other predator species are largely undetermined but may also be considerable. For example, red foxes and presumably coyotes and hawks prey on Franklin's ground squirrels (Choromanski-Norris et al. 1989), and red foxes prey on weasels (Latham 1952). Red foxes, striped skunks, and minks use burrows dug by other species. Raccoons, which are excellent tree climbers, prey on eggs and nestlings of American crows (Chamberlain-Auger et al. 1990) and black-billed magpies (Jones 1958). Conversely, great horned owls use platforms of other raptors for nesting (punstan and Harrell 1973; McInvaille and Keith 1974; Houston 1975; Bohm 1980). The antagonism between great horned owls and American crows is well-known (Houston 1960; Craighead and Craighead 1969), and great horned owls occasionally prey on certain other owls and hawks (Errington et al. 1940; Craighead and Craighead 1969; Dunkle 1977) and on striped skunks (Seton 1953). These are complex relations about which little is known. Understanding interspecific relations may help explain changes in the composition of predator populations in the prairie pothole region and may have predictive value.


Data from our track surveys allowed us to examine relations among the abundance of several carnivore species (Table 2). Canids had the most complete occupancy of study areas of all studied species and species-groups. All study areas and probably all quarter sections of each area were occupied by either coyotes, red foxes, or both during all study area-years. Tracks of one or both species were found in 93% of all twice-searched quarter sections and in 62% of those searched once. The coverage of study areas by these species conformed to that expected from the spatial arrangement of sympatric coyotes and red foxes in North Dakota described by Sargeant et al. (1987a) and included: (1) an entire study area with tracks of one species; (2) a contiguous block of quarter sections, sometimes almost entire study area, with tracks of both species (areas of interspecific territory overlap); and (3) contiguous quarter sections with tracks of one species separated by quarter sections with tracks of both species (study area that crossed interspecific territory boundaries; Fig. 22). When the percentage of quarter sections of study areas with coyote tracks increased, the percentage with fox tracks decreased (Table 2), which supports findings of others that red foxes tend to avoid living in coyote territories (Voigt and Earle 1983; Sargeant et al. 1987a; Harrison et al. 1989).

gif -- Graph of spatial arrangements

Figure 22. Examples of spatial arrangements of coyotes and red foxes in four selected study areas in the prairie pothole region as revealed by presence or absence of tracks found in quarter sections during two systematic searches for tracks in April-June (only one search was conducted in the Sharon study area), 1983-84.

Circumstantial evidence suggests that coyotes may suppress raccoon populations in the prairie pothole region (Cowan 1973; Stelfox 1980), and Clark et al. (1989) found that coyotes occasionally kill raccoons. In our study areas, indices from tracks revealed a decreased abundance of raccoons with an increased abundance of coyotes (weak correlation; Table 2), giving some credence to the hypothesis that coyotes suppress abundance of raccoons. However, coyotes were numerous in a study area (Yorkton) with the highest raccoon index in Canada (Figs. 5 and 8). Possibly, the greater abundance of abandoned farm buildings in that study area offered denning and resting raccoons additional protection from coyotes.

Coyotes (Godin 1982) and badgers (Gilbert 1960; Sargeant et al. 1982) occasionally prey on striped skunks, and there is evidence of mutual (Lehner 1981) albeit sometimes fatal attraction (Rathbun et al.1980) between coyotes and badgers. However, we found no relation between track indices of coyotes and striped skunks, badgers and striped skunks, and coyotes and badgers Cable 2). No evidence suggested that interactions between other pairs of sympatric mammalian predator species affected their abundance. The concomitant increases of track indices of red foxes and raccoons Cable 2) were probably a reflection of the increasing abundance of these species with a decline in populations of coyotes. We found no literature to suggest red foxes and raccoons mutually benefit one another.

Table 2. Correlations between occurrence of tracks (% of searched quarter sections with tracksa) of pairs of mammalian carnivore species in study areas in the prairie pothole region, 1983-88b.

Species Pairsc r P
Coyote - Red Fox -0.86 <0.001
Coyote - Raccoon -0.64 <0.001
Coyote - Striped Skunk <0.01 0.988
Coyote - Badger 0.24 0.258
Red Fox - Raccoon 0.63 <0.001
Red Fox - Striped Skunk 0.14 0.504
Red Fox - Badger -0.11 0.619
Raccoon - Striped Skunk -0.10 0.639
Raccoon - Badger <0.01 0.992
Striped Skunk - Badger -0.38 0.060

aIncludes only study area-years where two searches for tracks were made.
bThe mink was excluded because of its low occurrence or absence in most study areas.
cNumber of study area-years equals 25 for all species.


Published evidence of interspecific segregation of nests of studied raptors was greatest by sympatric great horned owls and red-tailed hawks (McInvaille and Keith 1974; Schmutz et al. 1980), even though red-tailed hawks frequently nest near great horned owls (Orians and Kuhlman 1956; Hagar 1957; Dunstan and Harrell 1973; Houston 1975). Schmutz et al. (1980) observed aggression by Swainson's hawks toward both ferruginous hawks and red-tailed hawks. They also found that ferruginous hawks that nested within 0.3 km and Swainson's hawks that nested within 0.2 km of nests of another buteo species were less successful in fledging young than conspecifics that nested farther from nests of another buteo species. Rothfels and Lein (1983) reported interspecific territoriality among sympatric Swainson's hawks and red-tailed hawks.

Our tests of interspecific relations between all two-way combinations of the three raptor species (Swainson's hawk, red-tailed hawk, and great horned owl) revealed only the red-tailed hawk and great horned owl tended to nest near each other and no species tended to avoid nesting near another (liable 3). The observed number was three times greater than the expected number of nests of red-tailed hawk within 0.5 km of an occupied nest of a great horned owl. This tendency was not uniformly strong in all study area-years and was most pronounced (Holder) in 1983 and (Hawley, Hitterdal, Litchville) in 1988.

We do not believe interspecific attraction occurs between red-tailed hawks and great horned owls. Conversely, these species exhibit mutual antagonism; great horned owls occasionally prey on nestling red-tailed hawks (Craighead and Craighead 1969; Luttich et al. 1971; McInvaille and Keith 1974) and red-tailed hawks occasionally attack great horned owls (Smith 1970; Palmer 1988a). Red-tailed hawks clearly have strong philopatry and build sturdy nest platforms to which they often add nest material in successive years (Luttich et al. 1971). Great horned owls use nests of other species, especially of red-tailed hawks (Orians and Kuhlman 1956; Dunstan and Harrell 1973; Houston 1975; Petersen 1979). Because they seldom migrate and nest much earlier than the migratory red-tailed hawks (Bent 1961; Stewart 1975), great horned owls often usurp seasonally vacated nests of red-tailed hawks. A returning red-tailed hawk pair, finding its nest occupied, probably builds a new nest nearby. The lack of a significant tendency by great horned owls and Swainson's hawks to nest near one another (Table 3) may be because Swainson's hawks nest later and closer to the ground (Schmutz et al. 1980; Palmer 1988b), build less sturdy nests (A. B. Sargeant and M. A. Sovada, personal observation), and are more vulnerable to aggression by great horned owls than red-tailed hawks (Ounkle 1977).

Table 3. P-values and odds rations from Mantel-Haenzel tests (Fleiss 1973) of inter- and intraspecific relations of nest spacing by avian predator species in study areas in the prairie pothole region, 1983-88.a,b

Species pair or species P-value
Nc Oddsd Associatione Homogeneityf
American crow-great horned owl 17 0.41 0.193 0.950
Red-tailed hawk-great horned owl 12 3.38 0.002 <0.001
Swainson's hawk-great horned owl 18 0 0.252 >0.999
American crow-red-tailed hawk 19 0.35 0.005 0.992
Swainson's hawk-red-tailed hawk 11 0.75 0.763 0.988
American crow-swainson's hawk 20 1.35 0.260 <0.001
American crow 25 2.02 <0.001 <0.001
Swainson's hawk 19 0.46 0.299 >0.999
Red-tailed hawk 18 0.12 0.002 >0.999

aAssociation refers to an overall tendency to avoid or aggregate.
bHomogeneity indicates the consistency of the association across study area-years.
cNumber of study area-years.
dOdds < 1 indicates avoidance of nesting near each other; odds > 1 indicates aggregation of nests.
eP-value based on chi-squared distribution with 1 degree of freedom.
fP-value based on chi-squared distribution with N-1 degrees of freedom.

Our examinations of interspecific interactions revealed that across all study area-years American crows tended to avoid nesting within 0.5 km of red-tailed hawks (Table 3). Although no such relation was detected between American crows and Swainson's hawks, the test for homogeneity of the association revealed a tendency by American crows to aggregate nests around nests of Swainson's hawks in some study area-years but to avoid or have no nest site association with nests of Swainson's hawks in other study area-years (Table 3).

Our examinations of intraspecific interactions were restricted to examinations of spacing of nests by several avian predator species. Previous investigations revealed American crows may cluster or space nests (Good 1952; Bent 1964), but that the raptor species we studied, except the northern harrier, are territorial and tend to space their nests widely (Bent 1961; Craighead and Craighead 1969; Schmutz et al. 1980; Rothfels and Lein 1983; Hamerstrom et al. 1985). In our study, American crows tended to aggregate their nests, whereas red-tailed hawks avoided nesting near each other (Table 3). The tendency by American crows to aggregate their nests was not consistent across all study area-years but was particularly strong in two (Moore Park and Inchkeith) in 1984 and in four (Shamrock, Inchkeith, Earl Grey, Yorkton) in 1985. Red-tailed hawks consistently avoided nesting near each other in all study area years. Swainson's hawks did not demonstrate intraspecific relations in nest spacing (Table 3) in contrast to findings of Rothfels and Lein (1983). Data on great horned owls were too limited to permit meaningful analysis.

Previous Section -- Factors of Abundance and Distribution - Direct Human Influences
Return to Contents
Next Section -- Population Trends

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 19:48:58 EST
Sioux Falls, SD [sdww54]