In pinyon-juniper forests in the Southwest, hidden recorders capture the fluted whistles of a pinyon jay — part of a new effort to track ecological changes through sound.
Researchers are combining machine learning and expert review to test wide-scale acoustic monitoring in the Gila, Kaibab and Cibola National Forests. In 2022, their recording units automatically captured over 50,000 hours of audio. These devices were scattered throughout the three forests, silently listening.
After recording, the researchers ran the audio data through BirdNET, an AI-powered database that uses deep learning to recognize over 6,000 bird species. This database is increasingly used throughout the country for large-scale avian monitoring.
Once BirdNET identified the species, experts double-checked the accuracy. They confirmed that the method can currently identify at least 52 species. The pilot study identified six “priority” species, including pinyon jays and Mexican spotted owls, with high levels of confidence. The recordings even captured several lonesome Mexican gray wolf calls drifting through the night.
By better understanding the acoustics of the forest, land managers can track bird populations more efficiently — and at a fraction of the cost — of traditional surveys. It’s a reminder that listening carefully with the right tools can help us understand how a landscape is changing.
This Earth Note was written by Danika Thiele and produced by KNAU and the Sustainable Communities Program at Northern Arizona University.
