Graduation Term


Document Type


Degree Name

Master of Science (MS)


School of Biological Sciences

Committee Chair

Angelo P. Capparella


During spring and fall seasons, Neotropical migratory passerines travel nocturnally across the Western Hemisphere between their wintering and breeding grounds, often encountering man-made threats. One hazard that has gained considerable publicity in recent decades is the communication tower. While there have been many tower kill studies recorded, there has been no attempt at predicting the risk of towers based on their different attributes (e.g., height, type of light, landscape placement) spatially on a regional scale. The objective of this study was to create seasonal mortality profile maps in GIS for Illinois, based on tower attributes combined with key factors such as bird movement, altitude, and weather patterns as predictors of the set of factors that best determine the avian mortality risk at different towers seasonally throughout the state. Data from previous tower kill studies, compiled into a meta-dataset, was used to determine the relative importance of the various attributes and factors. Principal Component Analyses grouped tower height and land elevation into the first tower factor, and cloud cover, cloud ceiling and percentage of days with >50% cloud cover into the first weather factor. Stepwise multiple regressions selected the first tower factor and tower light color for fall, and the first tower factor and wind speed for spring. No interactions were found between stepwise selected factors in ANCOVA analyses. Formulas produced to apply the selected factors to GIS yielded spatial indices of tower mortality risk for towers within coverage of five radars for the fall (36% of total towers) and spring (61% of total towers). This method is the first attempt at spatially evaluating avian mortality risk at communication towers for any state, and can be refined as an environmental impact assessment to improve safety policies for migratory birds in the communications industry.


Page Count