Acoustically assessing apes: chimpanzee conservation with passive acoustic monitoring

With one million species at risk of extinction, there is an urgent need to regularly monitor threatened wildlife. In practice this is challenging, especially with wide-ranging, elusive and cryptic species or those that occur at low densities, such as chimpanzees (Pan troglodytes). Monitoring is needed to establish conservation actions, but also to assess population trends. For conservationists, key questions concern species distributions and densities. These data must be regularly updated, allowing conservation planners to effectively execute and assess conservation efforts. Conservationists benefit from methods that are time- and cost-efficient and simultaneously provide accurate and precise data. I evaluate passive acoustic monitoring (PAM) as a tool for detecting, localising, and estimating densities of chimpanzees. I compare with results from camera traps (CT), a more common method. I deployed two arrays of acoustic sensors in Issa Valley, Tanzania: one comprising twelve non-GPS-synchronised acoustic sensors across the whole study area for nine months to estimate chimpanzee presence/absence and density. I simultaneously deployed 53 CT for methodological comparison and used spatial capture-recapture (SCR) and distance sampling (DS) methods. Another acoustic array comprised four custom-built GPS synchronised acoustic sensors, deployed for a 3-month period around a single valley (~2km2), to localise chimpanzees. I found chimpanzee detectability varied over seasons. It is five times faster than an equivalent method using CT. Furthermore, I found that the estimated density of calling chimpanzees with acoustic SCR was lower than density derived from SCR with CT data, but within the 95% CI obtained with DS and CT data. Playback sounds were localisable with 27 ± 21.8m error and chimpanzee calls were localisable within 52m from the location of a researcher following the calling individuals. These results suggest PAM is a promising non-invasive method for chimpanzee monitoring. Despite the current challenges to automate data analysis, improvements of automatic call detection are promising. I anticipate that PAM will become more common in the conservationist’s toolbox for loud calling terrestrial species monitoring such as chimpanzees, gibbons, orangutans, wolves or elephants