Acoustic monitoring of anurans and birds in tropical biomes

Abstract

Passive acoustic monitoring (PAM) is increasingly popular in ecological research, butrecording and analyzing large amounts of data is still a critical bottleneck for the longterm monitoring of multiple species. We evaluated how temporal and spatial samplingeffort affects species diversity estimates using a set of 14,045 1-min recordings fromvarious neotropical birds and anuran communities. Our goals were to evaluate (i) thedaily vocal activity cycle of birds and anurans, (ii) the effect of temporal structure (e.g.,number of minutes listened each hour; continuous versus intermittent recordings) ondetermining the species composition, and (iii) the species–area relationship, and howthe number of recorders affects species richness estimates. Based on sampling coverage and completeness, we (iv) evaluate manual inspection schedules for birds andanurans across four biomes of Brazil. We found marked diel variation in vocal activity between taxonomic groups, indicating that birds and anurans are more efficientlydetected during early periods of the day and night, respectively. For proper diversityestimates, biomes with higher biodiversity required longer inspecting periods and alarger number of replicates, irrespective of taxa. Although fewer recordings per hourare less informative than full-hour sampling, species diversity is better estimatedwhen inspected minutes are interspersed over longer periods than inspecting minutes recorded over shorter timespans. Based on our findings, we recommend how to setPAM programs over highly diverse ecosystems.