Publication Date


Type of Culminating Activity


Degree Title

Master of Science in Raptor Biology



Major Advisor

James R. Belthoff, Ph.D.


Relatively little is known about the distribution, abundance, and population trends of many species of owls. Frequently, studies that describe owl distributions are accomplished by modeling characteristics of the habitat such as land cover and topographic features. However, the potential to model owl distributions as a function of species interactions has been largely unexplored. I investigated how habitat characteristics as well as species interactions shape owl distributions in the Western United States. Using occurrence data collected between 2009 and 2010 in the Boise National Forest, Idaho, I developed species distribution models for Flammulated Owls (Otus flammeolus) and Northern Saw-whet Owls (Aegolius acadicus) and spatially applied these models in a Geographic Information System to delineate habitat suitability. I considered land cover and topographic variables by selecting the best representative spatial scale from 0.4-km, 1-km, or 3-km-radius plots centered on point-count locations (N = 150). Flammulated Owls occupied 27 (18 %) point-count locations and occurred in areas with a higher proportion of Douglas-fir (Pseudotsuga menziesii) at the 0.4-km scale, less diverse land cover composition at the 1-km scale, and they associated with south-facing aspects at the 3-km scale. Saw-whet Owls occupied 45 (30%) point- count locations and were associated with relatively flat landscapes at the 0.4-km scale and locations containing larger proportions of non-forested area. At the 1-km and 3-km scales, Saw-whet Owls occurred in areas with south-facing aspects and those containing a higher proportion of ponderosa pine (Pinus ponderosa), respectively. To investigate the role of species interactions in determining owl distributions, I examined patterns of cooccurrence between woodpeckers (woodpeckers excavate cavities in living and dead trees, which provide nesting sites for secondary cavity nesters such as owls) and sympatric cavity-nesting owls using two-species occupancy models. Specifically, I tested the hypothesis that cavity-nesting owl occupancy was conditional on the presence of one or more common species of woodpecker: Hairy Woodpeckers (Picoides villosus), Lewis’s Woodpeckers (Melanerpes lewis), Northern Flickers (Colaptes auratus), Pileated Woodpeckers (Dryocopus pileatus), and Red-naped Sapsuckers (Sphyrapicus nuchalis). Additionally, I examined the pattern of co-occurrence between cavity-nesting owl species to help understand the nature of their possible competitive interactions. As snags may also be an important component of cavity-nesting owl occurrence, I modeled owl occupancy as a function of snag number and density and pattern of species co-occurrence while accounting for imperfect detection (i.e., the possibility that an individual may go undetected during surveying even when present). The average number of snags per hectare (12.5 ± 0.2, N = 150) and mean diameter at breast height of snags (35.17 cm ± 0.08, N = 150) had no effect on cavity-nesting owl occupancy. There was no support for the hypothesis that cavity-nesting owl occupancy was conditional on the presence of woodpeckers. Likewise, presence of Saw-whet Owls neither excluded nor facilitated Flammulated Owl occupancy. Thus despite the possible value in understanding the occupancy of the nocturnal owl community by examining the diurnal woodpecker community, relationships between woodpeckers and either Flammulated Owls or Sawwhet Owls do not appear strong enough to warrant such an approach.