Conserving bird populations requires knowledge of bird distribution and habitat use at all stages of their life cycle, including during migration. Historically, researchers have primarily relied on in-person observations for such monitoring, but because birds are frequently detected by sound, advances in automated digital audio recorders (ARUs) offer opportunities to expand monitoring techniques. ARUs can be deployed in the field for long periods of time to efficiently increase both spatial coverage and total amount of monitoring effort, especially in places and times where access is difficult and costly for human observers. ARUs have been evaluated for comparability to human observers during the breeding season in many habitats. However, birds behave differently during migration than during the breeding season, and monitoring birds during migratory periods is a current challenge in ornithology. Our study compares in-person point count observations to extended duration ARU observations during spring migration on the southern shore of Lake Superior, to assess and refine the applications of ARUs in migratory stopover habitat. We analyzed ARU recordings by conducting desk-based "listening point counts,” which can be completed by any observer with the skills to identify species aurally.
Results/Conclusions
When comparing identical 10-minute count periods, ARUs detected significantly fewer species than in-person point counts. However, when ARUs randomly sampled 1-minute intervals from throughout a longer 5-hour period, rather than sampling 10 consecutive minutes, ARUs detected multiple species that were never detected during in-person point counts and detected the arrival of some migrant species up to 2 days earlier in the season than did in-person point counts. Our results indicate that, even without complex automated species detection methods, skilled point count technicians can use ARUs to more completely sample some bird communities for which bird presence and/or detectability changes throughout the day. Listening to randomly-selected 1-minute intervals closed the gap in species richness estimates between ARUs and in-person point counts. Our results indicate that ARUs may provide a better representation of the bird community using a particular location during migration and at other times when birds may display less short-term site fidelity than during the breeding season. We demonstrate a relatively easy, immediate application for ARUs during migration, offering increased access to locations which may otherwise be inaccessible during the migration period due to weather and road conditions.