Comparison of acoustic recorders and field observers for monitoring tundra bird communities

Abstract

Acoustic recorders can be useful for studying bird populations but their efficiency and accuracy should be assessed in pertinent ecological settings before use. We investigated the utility of an acoustic recorder for monitoring abundance of tundra-breeding birds relative to point-count surveys in northwestern Alaska, USA, during 2014. Our objectives were to 1) compare numbers of birds and species detected by a field observer with those detected simultaneously by an acoustic recorder; 2) evaluate how detection probabilities for the observer and acoustic recorder varied with distance of birds from the survey point; and 3) evaluate whether avian guild-specific detection rates differed between field observers and acoustic recorders relative to habitat. Compared with the observer, the acoustic recorder detected fewer species (βMethod = −0.39 ± 0.07) and fewer individuals (βMethod = −0.56 ± 0.05) in total and for 6 avian guilds. Discrepancies were attributed primarily to differences in effective area surveyed (91% missed by device were >100 m), but also to nonvocal birds being missed by the recorder (55% missed <100 m were silent). The observer missed a few individuals and one species detected by the device. Models indicated that relative abundance of various avian guilds was associated primarily with maximum shrub height and less so with shrub cover and visual obstruction. The absence of a significant interaction between survey method (observer vs. acoustic recorder) and any habitat characteristic suggests that traditional point counts and acoustic recorders would yield similar inferences about ecological relationships in tundra ecosystems. Pairing of the 2 methods could increase survey efficiency and allow for validation and archival of survey results. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.