Abstract
The objective for this study was to measure the auditory and visual physiology of Golden and Bald Eagles in order to use eagle sensory capabilities to inform the design of potential deterrent stimuli that could be used to reduce eagle/turbine collisions with wind turbines. The rationale for this approach is that sensory systems of any organism will limit the capability of that organism to perceive aspects of the world around it. Moreover, species can differ dramatically in their sensory physiology so it is important to examine these characteristics in the species of concern, rather than relying on data from similar birds. Our project consisted of two main phases. The first phase was the acquisition and analysis of visual and auditory information from Golden and Bald Eagles in rehabilitation centers. This was performed in order to identify light and sound stimuli tuned to sensitive areas in the eagle’s sensory systems. The second phase of the project was to present these stimuli to both species of eagles in a behavioral experiment to identify which stimuli would be the most effective in changing the behaviors of the eagles.
Results of phase one indicated that the visual system of the Golden Eagle strongly absorbs ultraviolet light, making it unlikely the Golden Eagle (and most likely the Bald Eagle) will detect ultraviolet light signals. The Golden and Bald Eagles have differences in the sensitivities of their visual systems to light within the eye, but mathematical models indicate that both species are able to detect indigo/blue and orange/red light produced by LEDs (light emitting diodes) very well. We also found that both species of eagles have a blind spot above their head. This blind spot is particularly large in Golden Eagles due to a pronounced brow ridge above the eyes. This blind spot will result in the inability of a Golden Eagle to see something in front of it when its head is pointed down during flight – as might happen while hunting (i.e. searching the ground for prey). As such, the blind spot may increase the chance of collision with wind turbines if the eagle is actively hunting. This problem is less pronounced in Bald Eagles because their blind spot is smaller than in the Golden Eagles and their foraging strategy is different.
Results of phase one also indicated that the auditory systems of the Golden and Bald Eagles respond differently to a variety of sounds (static tones, static chords (i.e. stacked tones), and sounds with dynamic changes through amplitude modulation or frequency modulation). Both species’ auditory systems responded strongly to tones across a wide range of frequencies (0.5 – 5kHz ), however the Bald Eagles’ auditory system was much better at processing complex sounds with dynamic rapid changes in amplitude or frequency modulation than the Golden Eagle. All of these sounds were then played with two types of noise in the background (white or pink). White noise more closely resembles the sound of wind and pink noise more closely resembles wind turbines or other sources of anthropogenic noise. Most sounds were more strongly masked by pink noise than by white noise, but several sounds (especially sounds with rapid modulation changes) showed little or no masking, indicating these were good candidate signals. However, even though rapidly changing sounds are less subject to noise masking, they are also less strongly processed by the Golden Eagle auditory system. This tradeoff does not exist in Bald Eagles because individuals of this species are very good at processing rapidly changing sounds. Given that Golden Eagle populations are at greater risk than Bald Eagle populations, we suggest that the most efficient alerting sound stimuli used in deterrent systems should be complex sounds that do not change very rapidly.
We identified candidate light (indigo/blue and orange/red LED lights) and sound (sinusoidal frequency modulated sound, linear frequency sweeps, amplitude modulated sound, and a mistuned harmonic stack) stimuli that both eagle species sensory systems are highly sensitive to. Results of phase two, in which we presented these stimuli to eagles in a behavioral experiment, indicated that eagles behaviorally responded to all the stimuli presented, but at varying degrees. The Golden Eagles, especially, elicited higher rates of visual exploratory behavior with a flashing blue light stimulus and all sound stimuli. We therefore recommend the use of these stimuli in field-testing of light/sound eagle deterrent systems on wind turbines. The eagles showed lower rates of behavior over the course of an experiment, suggesting either that they habituated to our stimuli or were initially stressed by the setup of the behavioral tests. These results underscore the need to test for habituation effects. Nonetheless, habitation to the stimuli in these field tests would likely be reduced by the use of random presentations of the four sounds and if possible random presentation of the candidate lights.
Report by Purdue University.