Coyote movements, habitat use, and vulnerability in central Alberta

A 2-year radiotelemetry study revealed no migratory movements of coyotes (Canis latrans) between agricultural and adjacent boreal forest areas in central Alberta. Resident coyotes were predominantly adults and essentially maintained nonoverlapping home ranges averaging 12.1 km2 in size. Nonresident coyotes, predominantly subadults, frequented home ranges of resident coyotes but may have avoided contact with them. Mobility of coyotes increased after mid-late February. Dispersal occurred between late January and late March and between July and November. Coyotes preferred forested over open areas and avoided open areas close to travelled roads during daylight hours. Nonresident coyotes were located closer to agricultural carrion than resident coyotes, suggesting a greater vulnerability to hunters and to coyote control using carrion as bait. However, annual survival rates of resident and nonresident coyotes after 1 January were identical (S, = 0.38). J. WILDL. MANAGE. 49(2):307-313 Todd and Keith (1976) found a positive relationship between coyote densities in mid-late winter and availability of dead livestock on agricultural areas in central Alberta. A seasonal migration of coyotes between forested areas in summer and agricultural areas in winter was hypothesized to explain an apparent coyote population decline on agricultural areas and a concomitant increase on nearby forested areas during late winter. Mobility of coyotes is restricted by deep, soft snow in forested areas (Murie 1940, Ozoga and Harger 1966). In contrast, wind sweeps much of the snow from open fields, and stronger crusts form in open agricultural areas. Greater mobility coupled with an abundance of dead livestock enhance agricultural areas as coyote habitat in winter. This study was conducted to determine if coyotes migrate seasonally to forested areas in summer and to agricultural areas in winter in central Alberta. We also monitored daily movements and spatial relationship of coyotes in an agricultural area to assess their effect on coyote control during winter months. We thank M. G. Hornocker, A. W. Todd, and A. W. Hawley for critical advice. H. T. McVeety provided field assistance. We also thank landowners near Westlock, Alberta, for their cooperation. SPresent address: Alberta Agriculture, J. G. O'Donoghue Building, Edmonton, Alberta T6H 5T6, Canada. This content downloaded from 157.55.39.212 on Wed, 08 Jun 2016 05:14:28 UTC All use subject to http://about.jstor.org/terms 308 COYOTE MOVEMENTS * Roy and Dorrance J. Wildl. Manage. 49(2):1985 STUDY AREAS AND METHODS We radiocollared coyotes on an agricultural area in winter and on a forested area in summer from January 1977 to December 1978. The 332-km2 agricultural study area was centerd in a mixed farming community near Westlock, Alberta. The area was boreal parkland transition forest (Moss 1955) and was described by Wetmore et al. (1970) and Todd and Keith (1976). Ninety-two percent of the area was cultivated or cleared pasture. The remainder was forest dispersed along water courses and in small woodlots and was mostly grazed by cattle. The 100-km2 forested study area was located 30 km northwest of the agricultural area in the boreal-cordilleran transition forest zone (Moss 1955). This forested area was bordered to the north and west by patches of cleared land. Between the two study areas, land use changed progressively from predominantly forested to predominantly agriculture. Coyotes were captured with snares in winter and leg-hold traps in summer and were classified as pups, yearlings, and adults by incisor wear (Gier 1968). Ages estimated by incisor wear were compared with cementum layer counts of sectioned canine teeth from 75 coyotes collected by hunters and trappers on or near the agricultural study area during November-March. Radiolocations of coyotes were obtained on both study areas biweekly in 1977 and on 6 consecutive days every month in 1978 to determine if a migration occurred. Daily movements were assessed by radiolocations during January-March on the agricultural study area. Time of radiolocation was alternated as follows: 2 days from 0800-1600 hours, 2 days from 2400 to 0800 hours, and 2 days from 1600 to 2400 hours. Home-range size was delimited by the minimum convex polygon method (Mohr 1947). The effect of sallies by coyotes (Hibler 1977) was reduced by including only 95% of radiolocations closest to the range center of the coyote (Bowen 1982). We plotted the cumulative total of the area used by coyotes against successive increments of 10 radiolocations to obtain an observation-area curve. An asymptote of the observation-area curve was obtained when the area did not increase by more than 10% over 3 months. Coyotes were classified as residents if an asymptote of their observation-area curve was obtained, or if no dispersal movement occurred over 3 months, and as nonresidents if no asymptote was reached or if dispersal movements occurred. Movements greater than 5 km with no return for over 10 days were termed dispersal movements because the upper 95% confidence limit of the length and duration of 112 sallies calculated from Hibler (1977) was 5 km over 10 days. Mobility was measured as the distance between successive radiolocations a maximum of 48 hours apart. Coyote habitat was classified as forested or open and _ or >0.4 km from travelled roads. This distance was perceived as a maximum distance coyotes could be easily seen by hunters on roads. Mobility and distance to carrion were compared by analysis of variance. Other comparisons were made by nonparametric tests (Siegel 1956). Survival rates were calculated as described by Trent and Rongstad (1974).