Exploring spatio-temporal variation in soundscape saturation of an African tropical forest landscape

Monitoring biodiversity calls for efficient monitoring techniques. Soundscape analysis is a landscape-level approach to acoustic monitoring whereby acoustic indices are calculated as proxies for biodiversity based on all sounds occurring in a landscape, i.e. a soundscape. However, it is not fully understood what the influence is of environmental, anthropogenic and temporal heterogeneity on soundscapes and what soundscapes mean in terms of species diversity. Here, we use soundscape saturation, a proxy for acoustic diversity which calculates the degree of use of acoustic niches. We study how soundscape saturation responds to forest management and seasonality in the tropical forests of Gabon, while taking into account distance to rivers, human accessibility, signs of hunting, elevation and slope. To study the contribution of medium to large mammals to soundscape saturation, we relate soundscape saturation to the abundance of several animal guilds as estimated with camera trap data. Soundscape saturation was higher in a multi-use forest vs. logging concessions, in the rainy reason vs. the dry season, and in proximity of rivers and human access points. Elevation, slope, and hunting were less important covariates. We did not find strong evidence for a relationship between soundscape saturation and the abundance of several mammal guilds derived from camera traps. The lack of this relationship highlights the opportunity for a joint approach of eco-acoustics and camera trapping due to complementary species coverage. Our study provides the first soundscape saturation study in African tropical forests and the first to combine soundscape saturation with camera trap data. Soundscape saturation is considered a useful additional index for quantifying biodiversity across a range of land-use types. Future studies may aim to improve species inference and the site-specific interpretation of saturation levels by investigating species-specific contributions to, and spatiotemporal variation in, acoustic activity