Bioacoustics and artificial intelligence: A new chapter for northern spotted owl monitoring and research

Abstract

Northern spotted owl (Strix occidentalis caurina), a species listed under the Endangered Species Act, have been monitored for nearly three decades using callback surveys and mark-and-recapture methods. Callback surveys can result in negative effects on the owl yet are widely used on federal, state, and private forest land prior to activities that range from timber sales to restoration projects. Despite conservation efforts, spotted owl populations are declining due to habitat loss and invasive barred owls (S. varia); and as barred owl populations increase, spotted owls are becoming harder to detect with conventional methods.Damon Lesmeister, a research wildlife biologist with the USDA Forest Service, Pacific Northwest Research Station, is developing an approach to monitoring spotted owl populations using passive bioacoustic technology and machine learning (a form of artificial intelligence). The method is proving to be a robust way to detect spotted owls with less impact on their declining populations. The method applies to multiple scales—from the entire range of the spotted owl to a specific patch of forest—and holds promise for studying more vocalizing species.Lesmeister is collaborating with multiple scientists and wildlife managers to establish a science-based protocol for using bioacoustic monitoring in all forests of the Pacific Northwest. By 2023, the approach is expected to replace the conventional methods for monitoring spotted owls.

Citation

Kantor, Sylvia; Lesmeister, Damon; Davis, Ray. 2022. Bioacoustics and artificial intelligence: A new chapter for northern spotted owl monitoring and research. Science Findings 245. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 6 p.