Food habits of Mexican Spotted Owls in Arizona

Food habits of Mexican Spotted Owls in Arizona. -The Spotted Owl (Strix occidentalis) is most common in mature and old-growth coniferous forests throughout much of its range (Forsman et al. 1984, Laymon 1988, Ganey and Balda 1989a, Thomas et al. 1990). Proximate factors underlying habitat selection in Spotted Owls are understood poorly. Abundance and availability of food, however, may be a key factor influencing habitat selection in birds in general (Hilden 1965) and in owls in particular (Southern and Lowe 1968). In addition, prey abundance has been implicated as an important influence on reproduction in many owls (e.g., Rusch et al. 1972, Lundberg 1976, Wendland 1984). In many areas, only two or three species of prey comprise 70-90% of prey biomass in Spotted Owl diets (Thomas et al. 1990:205-215). Understanding the ecology of these key species of prey could contribute greatly to understanding the ecology of the Spotted Owl. Despite this possibility, little is known about the diet of the Mexican subspecies of the Spotted Owl (S. o. lucida), which inhabits the southwestern United States and Mexico (Forsman et al. 1984). Previous reports suggest that Mexican Spotted Owls prey heavily on woodrats (Neotoma spp.) and also take other small mammals, birds, reptiles, and invertebrates (Ligon 1926, Marshall 1957, Kertell 1977, Wagner et al. 1982, Ganey et al. 1988, Duncan and Sidner 1990). These reports are based on small samples from restricted geographic areas, however, and may not reflect the full range of the diet in the southwest. Information on food habits of the Mexican Spotted Owl was gathered at locations scattered throughout Arizona from 1984-1990. This is the first broad survey of food habits of Spotted Owls in the Southwest, and could aid in designing future studies on both feeding ecology of Mexican Spotted Owls and the prey populations on which they depend. Methods. -Pellets and prey remains were collected beneath Spotted Owl roost and nest sites in habitats ranging from mid-elevation (1 125 m) rocky canyons to high-elevation (2930 m) mixed-conifer forests. Remains were collected opportunistically whenever they were found. All remains collected during a visit to a site were treated as a single sample because remains of large vertebrates can appear in more than one pellet (Forsman et al. 1984). If some pellets appeared markedly older than others, however, I identified two samples as containing old and recent pellets, respectively (Forsman et al. 1984). Older pellets were dry, dusty, and partially disintegrated. All prey remains collected were dated and grouped by five geographic regions: the San Francisco Peaks, approximately 3 km north of Flagstaff; Walnut Canyon, approximately 4 km southeast of Flagstaff; Black Mesa, on the Navajo Reservation in northeastern Arizona; the White Mountains of east-central Arizona; and southeastern Arizona (including the Santa Catalina, Santa Rita, Huachuca, and Chiricahua Mountains). Pellets were collected from three to 11 pairs of owls within each geographic area. Lumping of pellets from adjacent pairs may have masked variation in dietary composition among individual pairs (Laymon 1988), but was necessary because pellets collected from areas of overlap between adjacent pairs could not always be unambiguously assigned to a particular pair. Also, sample sizes were too small to characterize the diet adequately for most individual pairs. Vertebrate prey were identified by comparison with specimens in the Northern Arizona University Museum of Vertebrates (NAUMV), using skeletal remains or feathers. The minimum number of individuals of each species was determined in each sample by counting skulls, right or left rami, or other identifiable skeletal remains (Forsman et al. 1984). Invertebrates were identified and counted using mandibles, femurs, wings, or fragments of exoskeleton. In samples containing beetle legs, I estimated the minimum number of