Acoustic indices predict recovery of tropical bird communities for taxonomic and functional composition

Abstract

Quantifying the success of biodiversity restoration is a major challenge in the UN Decade on Ecosystem Restoration. We evaluated the potential of acoustic indices to predict the recovery success of bird communities within abandoned agricultural areas. Using audio recordings from a lowland tropical forest region, we identified 334 bird species and calculated established acoustic indices. Community composition was analyzed using Hill numbers, accounting for incomplete sampling. Acoustic indices effectively predicted independent species data (R2 = 0.59-0.76), capturing not only taxonomic but also functional and phylogenetic composition. Taxonomic composition was best predicted for common and dominant species, while functional and phylogenetic compositions were more accurately predicted for rare and common species. In addition, community composition was strongly influenced by the surrounding habitat. Our findings demonstrate that a small set of acoustic indices, once validated by stratified ground truth data, provides a powerful tool for assessing restoration success over large tropical areas, even for functional composition of rare tropical birds.