Migrating wildlife in Vermont: radar and thermal infrared imaging studies
Cooperative Agreement (No. 50181-6-J069A) between the U.S. Fish and Wildlife Service and the Board of Trustees of the University of Illinois, with subcontract participation by Boston University.
Utility-scale wind turbines are about the height of a 40-story building and can cause harm to flying birds and especially bats in some areas of North America and Europe. Because high areas are prime candidates for wind power generation and little is known about the behavior of these animals encountering mountainous terrain in North America, this study seeks to provide needed data on flying wildlife over and among the mountains of Vermont, USA. A substantial additional goal is to investigate the power of modern transportable radar to distinguish bats from birds, thereby laying a foundation for meaningful monitoring of these two kinds of flying vertebrates. Both goals will inform government agencies, the wind industry, consultants, and others seeking to bring wind energy to the forefront in North America while minimizing harm to wildlife.
An earlier report (<A HREF="http://atkins.inhs.uiuc.edu/~rlarkin/VT_phase_1_report_final.pdf">Larkin and Kamen 2007</A>) listed desiderata to obtain information on migratory birds and bats in mountainous Vermont. This field study, using instrumentation radars, marine radar, and thermal imaging cameras, addresses those issues. Many recent reviews and workshops on wildlife and wind energy have concluded that: (a) low-flying animals are at risk, (b) radar is the most suitable tool for obtaining information on how birds and bats behave approaching mountainous and man-made structures, (c) combining several observational techniques is essential for advancing understanding, and (d) guessing the taxonomic identity of flying animals is unacceptable. This study is producing specific information applicable to Vermont and other mountainous areas of the eastern USA.
The study uses complementary techniques synergistically. Different techniques which, when used alone, suffer problems, make up for each others’ deficiencies when used together, particularly if one individual flying animal is observed by two or more techniques at the same time. Key points have been discussed extensively at national and regional meetings on wind power and wildlife and are further discussed in review articles and reports (see Larkin and Kunz publications).
Our study includes:
• Adequate taxonomic identification (insects, passerines,. other birds, bats, etc.)
• Local, focused field work at night in late August and early September. (Most birds and presumably bats migrate at night.) Observations during the day also may be useful.
• Observation of animals at appropriate spatial scale, large enough to compare behavior at different locations but small enough not to be obstructed by intervening topography. “Different locations” might include above a ridge vs. in a valley vs. on the ridge slope, or in a mountain pass vs. over adjacent areas approaching and beyond it.
• Measurement of winds aloft on a scale appropriate to the terrain. Detection and measurement of foehn winds, upslope winds, and orographic rotors may be needed.
• Measurement of height of animals
• Measurement of geographic position of animals accurate enough to calculate their height above the ground
• Measurement of migratory flux at the height where wind turbines would likely be placed.
• Sophisticated thermal infrared observations by personnel from the laboratory of Prof. Thomas Kunz of Boston University.
• Additional simultaneous observations using marine radar to document the usefulness of that technique against a backdrop of intensive data from instrumentation radar.
Reference:
Larkin, R.P. and Kamen, B. 2006. Vermont Bird Migration Study: Phase 1 of 2; Feasibility of using WSR-88D radar Data to Characterize Bird Migration in the Mountainous Landscape of Vermont. Final report to US Fish and Wildlife Service Region 5, Hadley, MA.
An area of northern Vermont in autumn viewed looking SSW from above migrating animals. A valley and lake on the right contrast with a mountainous area with snow. The E-W ridgeline is slightly over 2 km long. The specific location was chosen because snow cover shows the topography clearly, not necessarily as a favorable location for this work. Image courtesy of Google Earth. (This specific area is not one selected in 2008; it is shown as an example.)
Red illustrates passage rates of migrating birds and bats and their height distributions over different terrain. The expected number of birds and bats that would encounter wind turbine rotors will be given for topographic situations of concern. Because the radar records wing beats of each animal, the critical biological information is included.
A pencil-beam radar with its narrow beam at 20 degrees from the horizon. Here the operators point the beam in alternation in three directions (azimuths) to count migrating birds and bats at all heights over different terrain. As an example, a small section of the beam (bright red) counts birds and bats at left to quantify compression of birds above the ridgeline (Phase 1 Report). The count is compared with other sections of the same count, including higher (farther) sections above the ridgeline and lower (closer) sections where birds and bats may be ascending to clear the ridge. In the same time period, the count is also compared in ABCABC fashion with counts over the valley (right) and counts of animals that have not encountered the high ground (foreground). The red areas are approximately to scale, showing the ca. 2.7-km reach of this radar for migrating songbirds. The best formulas for calculating actual passage rates from such data come from the publications of V.A. Drake.
Light Blue illustrates direct three-dimensional records of the behavior that may put birds and bats at risk, or not. The three tracks show wildlife
A tracking radar with three realistic but hypothetical tracks of of wildlife encountering Vermont mountains. Wider lines indicate greater height of flight. Solid line A is a bird or bat ascending to follow a ridgetop for about 1 km; a north-south ridge would pose a hazard for such an animal depending on its height over the ridge top. Solid line C is a bird or bat climbing over a ridge and descending again on the other side; this animal's behavior would result in a compressed layer of flying animals at the ridge top. Solid line C is a bird or bat encountering the same topographic obstruction but finding a way around it without changing height; a turbine on the right slope would be very poorly located for animals behaving this way. The tracking radar can follow birds or bats at distances well beyond this illustration but, like any radar, cannot follow them behind obstructions; the dashed continuations would not be part of the data.
An advanced thermal infrared (TIR) camera, used by researchers at Boston University, mounted coaxially with the antenna of the Illinois Natural History Survey radars. The TIR camera can visualize and help identify flying animals even at night without disturbing their flight, thus informing the radar data about the taxonomic identity of the target. The two instruments are complementary, the TIR camera operating at limited range, provides near-optical resolution and sensing heat produced by flight muscles, and the radar operating both close and at great distances to count animals and measure heights as they negotiate mountainous terrain. See also video clips at http://www.bu.edu/cecb/wind/video/ .
This site in northern Vermont allowed two INHS radars to document nocturnal movements of flying wildlife in 2008. Birds and bats flew over and along parallel ridges (background), providing detailed tracks and distributions of heights above the terrain. We are analyzing the data to learn where wind turbines may be placed to minimize their impact on migrating animals. The Vermont Agency for Natural Resources and Essex Timber assisted with access to the site.