https://orcid.org/0000-0002-1563-575X
![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Passive acoustic monitoring (PAM) has become increasingly popular in monitoring biodiversity. It produces large amounts of data and can provide a foundation for understanding the long-term consequences of environmental degradation. However, extracting biological information from such extensive datasets can be challenging and requires advanced computational skills. Herein, we introduce a streamlined workflow for detecting acoustic signals of three critically endangered birds: Cherry-throated Tanager (Nemosia rourei), Alagoas Antwren (Myrmotherula snowi), and Blue-eyed Ground-dove (Columbina cyanopis). As these species are among the world’s most endangered birds, locating new populations is a top priority. We chose potential templates based on the acoustic parameters of the vocal repertoire and evaluated their performance using soundscapes with known composition (gold standard data). To evaluate the efficiency of the templates, we used precision and recall metrics and found that achieving high precision rates comes at the cost of recall rates. Although we used gold standard data to calibrate our algorithm, large-scale validations have revealed the limitations as some templates have exhibited significantly lower precision values. The use of binomial models helped reset precision values to 90%. Our workflow can process large amounts of data efficiently, helping to monitor populations of these critically endangered species, locate new populations and evaluate population dynamics.
Acknowledgements
We wish to thank all field personnel that work hard to keep the lights on. CONICET provided a post-doc fellowship to CBdA.
Disclosure statement
No potential conflict of interest was reported by the author(s).