Distance Sampling
07 Dec 2015

While visiting University of Sussex recently, I was talking with conservation biologist Mika Peck http://www.sussex.ac.uk/profiles/76093. They are interested in the use of acoustic recordings for species richness surveys which obtain information about presence/absence but not about species abundance. I made the comment that acoustic methods cannot easily deal with abundance because it is not easy to determine if a single individual is calling multiple times or many individuals are calling once only. Mika referred me to a protocol called distance sampling http://distancesampling.org/whatisds.html.

Quoting from the website, distance sampling is a widely used methodology for estimating animal density or abundance. Its name derives from the fact that the information used for inference are the recorded distances to objects of interest (usually animals) obtained by survey lines or points. In the case of point sampling (for example the often used five-minute point-count protocol for birds), the radial distances from the point to detected animals are recorded. A key underlying concept is that the probability of detecting an animal decreases as its distance from the observer increases. Much of distance sampling methodology is concentrated on detection functions, which model the probability of detecting an animal, given its distance from the transect.

The relevance of this to acoustics is that the signal-to-noise ratio (SNR) of a recorded bird call can be taken as a proxy for distance. One could then use distance sampling techniques to estimate abundance from SNR information. Also of significance is that distance sampling software is available in R and are open source (available here).