@article{oai:oist.repo.nii.ac.jp:00000313,
 author = {Ross, Samuel R. P. -J. and Friedman, Nicholas R. and Dudley, Kenneth L. and Yoshimura, Masashi and Yoshida, Takuma and Economo, Evan P.},
 issue = {1},
 journal = {Ecological Research},
 month = {Nov},
 note = {Ecologists have many ways to measure and monitor ecosystems, each of which can reveal details about the processes unfolding therein. Acoustic recording combined with machine learning methods for species detection can provide remote, automated monitoring of species richness and relative abundance. Such recordings also open a window into how species behave and compete for niche space in the sensory environment. These opportunities are associated with new challenges: the volume and velocity of such data require new approaches to species identification and visualization. Here we introduce a newly-initiated acoustic monitoring network across the subtropical island of Okinawa, Japan, as part of the broader OKEON (Okinawa Environmental Observation Network) project. Our aim is to monitor the acoustic environment of Okinawa’s ecosystems and use these space–time data to better understand ecosystem dynamics. We present a pilot study based on recordings from five field sites conducted over a one-month period in the summer. Our results provide a proof of concept for automated species identification on Okinawa, and reveal patterns of biogenic vs. anthropogenic noise across the landscape. In particular, we found correlations between forest land cover and detection rates of two culturally important species in the island soundscape: the Okinawa Rail and Ruddy Kingfisher. Among the soundscape indices we examined, NDSI, Acoustic Diversity and the Bioacoustic Index showed both diurnal patterns and differences among sites. Our results highlight the potential utility of remote acoustic monitoring practices that, in combination with other methods can provide a holistic picture of biodiversity. We intend this project as an open resource, and wish to extend an invitation to researchers interested in scientific collaboration.},
 pages = {135--147},
 title = {Listening to ecosystems: data-rich acoustic monitoring through landscape-scale sensor networks},
 volume = {33},
 year = {2017}
}