Listening in the dark: acoustics indices reveal bat species diversity in a tropical savannah

Abstract

ABSTRACT Surveying biodiversity using bioacoustics has become increasingly common worldwide, although it is mostly concentrated in temperate regions. The variety of automatic recorders, the development of free analytical tools, and several acoustic indices have increased the number of studies worldwide. The bioacoustic approach is essential for application in poorly surveyed regions with the pressure of human activities, such as the Brazilian cerrado. We tested the association of four bat diversity metrics (taxonomic, phylogenetic, and two functional diversity metrics, being one based on morphological and the other on acoustical traits, with five commonly used acoustic indices. We used a dataset of 608.4 h obtained from 30 sampling points in three protected areas in Central Brazil. Using Flexible Discriminant Analysis, we identified 21 bat species used in our subsequent analysis. The Entropy index was the best predictor of taxonomic and phylogenetic diversity, whereas the Acoustic Complexity Index was the best predictor of functional morphological diversity. We concluded that acoustic indices are suitable for estimating the diversity of insectivorous bats in the cerrado. However, we registered only part of the bat community, and bats can vary seasonally masking the real diversity of the study area; thus, this method should be used parsimoniously.