As technological advances continue to pave the way for near real-time monitoring of complex dynamic systems, the questions we can ask about the influence of global change on ecological communities continue to develop. Using a combination of acoustic recording, machine learning and remote sensing, we can gain insight into community composition and dynamics. We use ecoacoustic data collected as part of the OKEON (Okinawa Environmental Observation Network) project, a scientific monitoring network consisting of malaise traps, camera traps, bioacoustic recorders and weather stations across 24 sites in Okinawa, Japan, to explore how urbanisation and future climate warming may affect Okinawa’s avian communities and sensory environment. We analyse the temporal dynamics of various bird species’ vocal activity and calculate acoustic diversity across sites using the Shannon entropy of multiple frequency bands, and the variability of this diversity metric over time as a metric of stability. This allows us to examine the diversity and stability of ecoacoustic patterns along a strong urban-rural gradient and explore how they are influenced by surrounding land-use and environmental conditions.
Results/Conclusions
Our results reveal clear patterns of biogenic vs anthropogenic noise across Okinawa. We found correlations between forest land cover and detection rates of two culturally important species in the island soundscape – the endemic Okinawa Rail and the Ruddy Kingfisher. Our soundscape indices followed both diurnal and seasonal cycles, but with clear differences among sites. Whilst urban sites were overall more acoustically diverse, the biotic component of the soundscape was markedly less diverse. Preliminary analyses of soundscape stability also reveal differences in the temporal stability of ecological communities along a land-use intensification gradient, with urban land-cover increasing the variability of bioacoustic signalling and ecoacoustic diversity. Yet these results were scale-dependent, with surrounding land-use correlating more strongly with acoustic diversity at larger scales surrounding each site. Overall, our results highlight the utility of remote acoustic monitoring practices that, when combined with other real-time monitoring techniques as we have done, can provide a holistic picture of the resilience of ecological communities to global change, their dynamics and their stability.