Publication Date

12-2016

Date of Final Oral Examination (Defense)

10-28-2016

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Raptor Biology

Department

Biology

Major Advisor

James R. Belthoff, Ph.D.

Advisor

Jesse R. Barber, Ph.D.

Advisor

Julie Heath, Ph.D.

Advisor

Christopher J. McClure, Ph.D.

Abstract

Barn Owls (Tyto alba) are killed by vehicle collisions in greater numbers than any other North American bird of prey. Interstate-84 (I-84) in southern Idaho, USA has among the world’s highest known rates of Barn Owl-vehicle collisions. Little is known about Barn Owl occupancy in this region, so it is unclear if owls are killed in proportion to their abundance, or if they are equally abundant in segments with lower mortality and somehow escape collisions. Furthermore, studies of Barn Owl movements and behavior are limited. I was interested in understanding (1) factors that affect Barn Owl occupancy in two seasons: early- and post-breeding, (2) Barn Owl colonization of sites from early-breeding to post-breeding season, (3) examining the relationship between model-predicted occupancy, based on the factors I assessed, and roadway mortality established from observed Barn Owl mortality locations, and (4) examining Barn Owl behavior, movements and habitat selection, particularly in relation to I-84. I conducted nighttime point counts for Barn Owls in southern Idaho during the early- and post-breeding seasons (Jan – Mar and Sept – Nov 2014, respectively). I detected Barn Owls during 52 of 666 (7.8 %) point counts and at 39 of 222 (17.6 %) locations in the early-breeding season and during 50 of 201 (24.8 %) point counts and at 31 of 67 (46.3 %) locations in the post-breeding season. During the early-breeding season, probability of Barn Owl detection was 0.32 ± 0.06 (SE). Detection increased with playback of Barn Owl calls and with increasing Julian date, percent moon illumination, and cloud cover. Barn Owl occupancy increased with increased proportion of crops and presence of trees, and it decreased as background noise level (dBA) increased. Probability of detection of Barn Owls was higher in the post-breeding season (0.45 ± 0.07). Detection increased with playback of Barn Owl vocalizations, increasing Julian date, and decreasing background noise. During the post-breeding season, Barn Owl occupancy was positively related to stream length and negatively related to proportion development and increasing distance from the Snake River. Of the potential models I assessed to describe colonization between seasons, there were two top models. The first model indicated that colonization of sites from the early- to post-breeding season declined with increasing terrain roughness while the second suggested that colonization increased with increasing cumulative stream length but decreased with distance to the Snake River and proportion development. Understanding factors influencing occupancy of Barn Owls will facilitate more effective conservation of this species in southern Idaho, especially in light of potential population declines related to roadway mortality.

Using data from standardized roadkill surveys, I also compared road mortality locations of Barn Owls to model-predicted occupancy estimates to understand how occupancy may be influencing mortality along I-84. Using the previously created occupancy models, I generated predicted occupancy at point-count locations which I then paired with the nearest interstate segment (1- and 5-km lengths) to examine the potential effects of occupancy and season on the likelihood of dead Barn Owls. The likelihood that 1-km segments near point count locations included a dead Barn Owl increased with occupancy and was greater during the early-breeding season. For 5-km segments, there was an interaction between occupancy and season, with occupancy having a greater positive effect on mortality during the early-breeding season than in the post-breeding season. However, a substantial proportion of variation in roadway mortality at both scales (96 % and 56% at 1 and 5 km scale respectively) was not explained by occupancy and season, so factors such as geometric roadway features, traffic patterns, fluctuations in small mammal abundance, and owl behavior near the interstate likely also influence mortality rates and locations.

Finally, to address questions of Barn Owl behavior and movements in relation to I-84, as well as habitat use I studied four adult male Barn Owls that were tending nests during February 2015. Two of these nests were within 3 km of the interstate, whereas the other two were more than 25 km away. I examined the efficacy of GPS data loggers for tracking Barn Owls and assessed six recapture methods to retrieve the data loggers. I obtained location data that spanned approximately two weeks of activity for each owl during the nesting season. I recaptured all instrumented males and found that manual- or laser-break triggered trap doors mounted on the nest box were most effective for recapturing Barn Owls. Within home ranges, the probability a Barn Owl used a site was higher near trees and minor roads and lower as distance from the nearest stream and the nearest major road increased. Barn Owl use of areas also increased as terrain roughness increased. Relative use of development, hay/pasture, and sage steppe land cover were less than for cultivated crops, while owls used grassland/herbaceous and wetland land cover more than cultivated crops. The two male Barn Owls that nested within 3 km of the interstate never moved closer than 1 km even though their maximum movements ranged up to 3 km. Thus, it is possible major roadways function as barriers to adult owl movements during the breeding season because they avoid roads, but they remain susceptible to road mortality in their more rare attempts to cross.

Download

Share

COinS