American offshore wind turbines have the gross potential to produce over 2,000 gigawatts of power—more than double the amount of energy consumed by the United States each year. Offshore wind is consistent, abundant, and reliable, making it a promising source of energy near heavily populated coastal areas. When planning offshore wind projects, collecting data on the abundance of birds and bats informs project siting. Post-construction monitoring is also important to ensure project effects are sustainable and understood.
Exactly how different bat and bird taxonomic families behave in the presence of turbines and the potential effects of offshore wind farms on bat and bird flight and site use patterns has not been fully characterized. As a result, researchers at Pacific Northwest National Laboratory (PNNL) are working to ensure that offshore wind safely coexists with wildlife. By better understanding the different species affected—where they live and where and when they fly—developers can accelerate the environmentally responsible siting and permitting of offshore wind farms.
Continuously monitoring offshore locations provides a complete picture of the animals’ behavior patterns at those sites. However, limited accessibility, harsh weather conditions, and nocturnal migrations of bats and birds make monitoring offshore locations difficult. Using thermal video, Shari Matzner found a way around these limitations. Thermal video works both day and night and in low-visibility conditions. Even better, it is a mature, cost-effective technology.
Matzner recorded animals off the Washington coast with a thermal video camera and developed a software application called ThermalTracker. The open-source software quantifies the flight tracks of birds and bats by the time of day, direction of travel, relative size of the animal, and more. This information can be used to determine which taxonomic families of these animals live offshore. ThermalTracker characterizes and quantifies these data using a standard desktop computer. ThermalTracker’s postprocessing output can be readily analyzed using common statistical software like Microsoft Excel.
After developing ThermalTracker, Matzner tested it for accuracy—and found the software detected 81% of all the birds and bats. The missed detections occurred when animals were too far from the camera or the weather conditions were too cloudy. ThermalTracker’s data processing can be performed in roughly the same amount of time as the video duration, implying real-time processing is possible. When human experts reviewed the same footage, it took, on average, five times longer to generate the same data as ThermalTracker.
Two ThermalTracker models were developed for species identification. The first combines flight track data with the time of day, allowing researchers to classify observed animal species such as gull, tern, swallow, or bat. The second ThermalTracker model identifies different North American sea bird and bat taxonomic families based on wing-beat frequencies. The two models work together to accurately identify animals.
The next stage of ThermalTracker’s development will incorporate stereo vision to estimate the location of an animal and its approximate size to improve identification. An updated algorithm detects animals as video is being recorded and only saves video when a bird or bat is detected. With less data to store, the system can be used for long-term observation and provide more complete information.
Biologists at the nonprofit Biodiversity Research Institute (BRI) are currently testing the system off the coast of Maine.
“Developing technology to detect bird and bat activity at terrestrial and offshore wind farms will promote a better understanding of the nature of wildlife risks—or lack thereof—at any type of wind farm and reduce uncertainty about the potential for unintended impacts during operation,” said BRI Deputy Director Wing Goodale. “These cameras could provide a reliable method of detecting bird and bat response to offshore wind projects, where it is not possible to conduct wildlife monitoring by traditional means.”
Matzner and her colleagues will use the BRI field biologists’ notes to refine ThermalTracker algorithms later this year.