Date of Award

Summer 8-19-2017

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science (MS)

Department

Wildlife Ecology and Wildlife Conservation

Advisor

Erik J. Blomberg

Second Committee Member

Joseph Zydlewski

Third Committee Member

Brian J. Olsen

Additional Committee Members

Daniel G. McAuley

Abstract

Migration poses risks and energetic demands to individuals that may be greater than those experienced during non-migratory periods. Most migratory birds require stopover sites to rest and recuperate energy spent during migratory flights, and stopover locations can alleviate risks and provide supplemental energy en route to the animal’s end destination. An individual’s stopover duration is contingent first on energy acquisition that is constrained by resource availability, and secondarily on environmental conditions such as weather that may facilitate or constrain continued migration. From 2010 to 2013 I conducted a radio-telemetry study of a short-distance migrant, the American Woodcock (Scolopax minor), on the Cape May Peninsula, New Jersey, an important stopover site for American Woodcock during fall migration (October – January). My research objectives were to (Chapter 1) describe diurnal land-cover type characteristics used by American Woodcock and to evaluate second-order resource selection during the fall migration period, and (Chapter 2) to evaluate the influence of individual (age, sex, mass) and environmental (weather, moon illumination) variables on stopover departure rates and test whether resource selection affected the timing of departure. In this latter case, I specifically asked whether the decision to depart based on weather conditions changed depending on individual resource selection. I radio-marked 271 individuals and collected 1,949 locations, and used GIS and generalized linear mixed models to compare land cover types and other landscape characteristics between used and available locations. I found that American Woodcock selected portions of the Cape May Peninsula with greater proportions of deciduous wetland forest, old fields, and shrub-covered wetlands, and avoided deciduous forest, coniferous forests and sites further away from potential roosting fields. I used these results to develop a predictive model of habitat distribution at Cape May as both continuous and discrete surfaces, which predicted 17.5% of the Cape May Peninsula as potential woodcock habitat with 90% classification accuracy based on a withheld validation dataset. To evaluate stopover departure, I used Cormack-Jolly-Seber survival analysis to estimate daily woodcock detection and residency probabilities, where the probability of departure from Cape May was defined as 1 – residency. Woodcock detection probabilities varied among and within years, ranging from 0.06 (+/-0.01 SE) to 0.98 (+/- 0.004 SE). I found woodcock departed more frequently when individuals had higher average resource selection, an evening tailwind and higher temperatures. However, there was also an interaction between resource selection and wind direction; individuals with lower resource selection values did not depart as regularly with a tail wind. This suggests that individuals were better-able to take advantage of favorable weather conditions when they used higher quality habitat. The results in my thesis not only help fill a knowledge gap in the annual cycle of American woodcock, but further contribute to greater understanding of avian migratory stopover ecology.

Download

Share