Skip to content

Latest commit

 

History

History
3 lines (2 loc) · 1.44 KB

README.md

File metadata and controls

3 lines (2 loc) · 1.44 KB

Passive Acoustic Recording as A Potential Tool for Monitoring Individual Barred Owls

Many studies have suggested that vocal individuality of owls can be used to monitor populations using automatic approach; however, few have explicitly tested passive acoustic monitoring in this role. In this study, we assessed the possibility of identifying individual Barred Owls (Strix varia) based on their calls collected by a passive acoustic monitoring project. We set up autonomous recording units at the John Prince Research Forest (54° 27'N, 124° 10'W, 700 m a.s.l) in 2021 from Feb. to April. During this period, we collected 454 Barred Owl calls from 10 Barred Owl individuals. We measured 30 call features, 12 temporal features and 18 frequency features, from each call. Using forward stepwise discriminant function analysis, the model correctly identified 84.4% of the calls based on a 5-fold cross validation. The model achieved a Kappa statistic of 0.77, which showed substantial agreement between predicted individual versus observed individual. The most important discriminating features include total call length, interval between the 4th and the 5th syllables, interval between the 6th and 7th syllables, and the duration of the 8th syllable. Our results suggest that passive acoustic monitoring is an effective tool for monitoring individual Barred Owl and this study inform the potential of passive acoustic monitoring to be a tool for owl population censusing.