Acoustic beta indices exhibited a weak correlation with both vocal species composition dissimilarity and environmental factors dissimilarity in a subtropical forest

Abstract

Monitoring biodiversity through analyzing acoustic signals of ecosystems, as well as their relationship with environmental factors has garnered significant attention in fields like landscape ecology and biodiversity conservation. However, validation studies centered around only one of the acoustic diversities, the alpha index, while largely left out the beta one. Therefore, we evaluated the efficacy of acoustic beta indices as proxies of dissimilarity in vocal bird composition, and investigated the relative importance of vegetation characteristics dissimilarity, topographic dissimilarity, and geographical distance on acoustic beta diversity. To get acoustic beta indices of birds, we deployed 16 autonomous acoustic recorders across a biodiversity monitoring platform within a subtropical evergreen broad-leaved forest during bird breeding season. We performed bird species aural identification and investigated the relationship between vocal bird composition dissimilarity and eight acoustic beta indices with Mantel test. In addition, we obtained environmental factors included vegetation characteristics and topography variables through unmanned aerial vehicle light detection and ranging (UAV LiDAR) and from census data of the forest dynamics plots. The relative importance of vegetation characteristics, topography, and geographical distance on acoustic beta diversity was investigated with multiple regressions on distance matrices (MRM) and generalized dissimilarity modelling (GDM). Our results showed that the eight acoustic beta indices were significantly positively related to each other. There was weak correlation between acoustic dissimilarity and vocal bird composition dissimilarity, indicating that acoustic diversity is an independent facet of biodiversity. This study provided the first evidence that the combination of vegetation characteristics dissimilarity, topographic dissimilarity, and geographic distance account for few variation in the acoustic beta indices within the subtropical forest. Notably, vegetation characteristics dissimilarity emerged as relatively more important to the acoustic dissimilarity compared to topographic dissimilarity. Further research examining the relationship between specific acoustic spectral characteristics and potential influencing factors is highly encouraged.