Citation
Abstract
Acoustic monitoring has the potential to be a very effective method of surveying birds, particularly in remote areas where traditional survey coverage is limited by availability of appropriately skilled surveyors. In order to assess its suitability — particularly for augmenting long-term abundance monitoring — this study compares data collected by acoustic recorders and by surveyors on existing sites (1-km squares) of the UK Breeding Bird Survey (BBS). Twenty-eight volunteers in Scotland deployed an acoustic recorder on their BBS survey square during spring 2023. An automated classifier (BirdNET Analyzer) was run on all recordings to detect birds and identify them to species. A random subset of recordings was checked manually in order to evaluate classifier performance. We discuss the performance of BirdNET in the context of its suitability for acoustic monitoring in Scotland and compare BirdNET outputs with recent BBS data collected from the same sites. We also consider ongoing and potential future work in this area.