Monitoring Grassland Bird Communities with Acoustic Indices

Several researchers have tried to find relationships between acoustic indices and vocal animal communities to use acoustic indices as a passive monitoring method, as human-derived surveys are expensive, time-consuming, and suffer from observer bias. However, supplanting manual surveys with acoustic indices is a daunting task, considering effective indices for biological monitoring need to differentiate biologically relevant sounds from the broader soundscape, including anthropophony and geophony. The objective of our study was to test how well acoustic indices can be applied to avian community monitoring within a temperate grassland ecosystem in North America. We collected avian community data and calculated six commonly used acoustic indices from recordings in an intact lowland tallgrass prairie in the Central Platte River Valley of Nebraska throughout the avian breeding seasons of 2019–2021. Singular acoustic indices had only weak correlations with all community metrics. However, multivariate models including multiple acoustic indices showed potential for monitoring grassland bird abundance when anthropophony was considered. Fragmented grassland remnants likely experience significant anthropophony that needs to be accounted for when monitoring avian populations. Additionally, multivariate models incorporating several indices may provide a more accurate prediction of avian biophony than individual acoustic indices.