Abstract
Collisions with wind turbines can be a problem for many species of birds. Of particular concern are collisions by eagles and other protected species. This research study used the laboratory methods of physiological optics, animal psychophysics, and retinal electrophysiology to analyze the causes of collisions and to evaluate visual deterrents based on the results of this analysis. Bird collisions with the seemingly slow-moving turbines seem paradoxical given the superb vision that most birds, especially raptors, possess. However, our optical analysis indicated that as the eye approaches the rotating blades, the retinal image of the blade (which is the information that is transmitted to the animal’s brain) increases in velocity until it is moving so fast that the retina cannot keep up with it. At this point, the retinal image becomes a transparent blur that the bird probably interprets as a safe area to fly through, with disastrous consequences. This phenomenon is called “motion smear” or “motion blur” and is well known in human visual perception.
Based on this analysis, we devised a variety of patterns intended to give the retina more time to “rest” between successive stimulations by the blades. These patterns include various staggered-stripe patterns on a three-blade array, as well as a single black blade paired with two white blades. Several of these patterns increased the visibility of the blades. Nevertheless, above a critical retinal-image velocity, even these patterns lost their visibility advantage and became blurred. Using our data, we were able to model the distances at which patterns maintain their visibility for different turbine diameters and rotation rates. Although it seems counterintuitive, the stimuli lose their visibility at greater distances from the larger-diameter, slower-rotating turbines than from the smaller, faster types. Thus, an anti-motion-smear pattern that maintained its visibility as close as 20 m for a small, fast turbine would lose visibility closer than 50 m for a large, slow turbine.
Another series of experiments involved the use of single-colored blades instead of single black blades. In addition, various chromatic and achromatic single blade types were evaluated against naturalistic backgrounds composed of color photographs of wind turbines in various types of sky and foliage configurations. Although these studies indicated that color contrast was a critical variable (i.e., that the effectiveness of a colored blade depended on the color of the background against which it was viewed), the applicability of their results to the real world of birds is limited. First, the background was always stationary, whereas in nature, the background seen by a flying bird is always moving. Second, the colors in photographs may be accurate for the human eye, but avian color vision is quite different from human color vision, and color photographs may not accurately represent the colors seen by birds and may not have appeared natural to them. Given these uncertainties, in those conditions in which the background color changes rapidly depending on the moment-to-moment view point of the bird, black would probably be the best compromise color, even though it was not as highly visible compared to colors such as blue and green against a fixed background in a laboratory simulation.
Finally, we tested a series of devices applied to the blade tips to deter collisions from a lateral approach to the blades. Attaching tips to two of the three blades clearly improved visibility against the neutral background, but less so against a naturalistic background. Although the results of the naturalistic background studies were inconclusive, they did suggest that such devices might be effective under certain circumstances.
It is important to note that these studies have only evaluated the visibility of anti-motion-smear blade patterns and not their ability to deter a flying raptor from approaching them. Deterrence is a psychological property of a pattern that can only be evaluated in an awake, behaving bird, not in an anesthetized bird, as was the case in our studies. The deterrent effect of these patterns can best be tested in a field setting by observing the behavior of the birds and by determining the before and after fatality rates at turbines that have been treated with various patterns versus untreated, fatality-matched turbines.
In our opinion, the results of the laboratory visibility simulations were sufficiently encouraging. We recommend a field test of a single-blade, solid black pattern or a single-blade, thin-stripe pattern as the next step to determine whether the patterns are effective in reducing fatalities.