Passive Acoustic Cue Counting Surveys for Vocal Primates: A Field Test on the Critically Endangered Nosy Be Sportive Lemur (Lepilemur tymerlachsoni)

Abstract

Passive acoustic monitoring, which uses autonomous audio recorders in the field, offers a powerful, efficient and cost-effective survey method that samples continuously and non-invasively. But despite its potential, its application remains atypical in primate research. We conducted a field test of passive acoustic density estimation in a near range-wide survey of the Critically Endangered Nosy Be sportive lemur (Lepilemur tymerlachsoni). We used the sound level of target calls detected in audio recordings as a proxy for radial distance as part of a conventional cue-counting distance sampling framework. While the passive acoustic survey recorded a large number of calls (9,560 detected calls from 250 hr of audio recordings) and the sound-level-based distance prediction model performed well, the resulting density estimates using single calls as cues were implausibly high and had low precision, which we attribute to the inherent challenges of estimating the cue rate. Results were improved, however, when using call bouts, rather than single calls, as the unit of analysis (i.e., as “cues”). Concurrent line transect surveys provided a more plausible density estimate of 303.7 individuals/km2 (95% confidence interval = 189.6–486.3, coefficient of variation = 0.18), corresponding to a global abundance estimate of c. 11,300 individuals. Our results highlight the potential and limitations of passive acoustic density estimation and provide critical information for the conservation of one of the world’s most endangered primates. (Figure presented.)