Estimating tapir densities using camera traps

Abstract

Population size is a key concept in evolution, ecology, and conservation. However, counting moving animals in the forest is not an easy task, especially if we are talking about tapirs, which are very secretive animals with crepuscular or nocturnal activity patterns. Characteristics that make the counting even more difficult. To solve this difficulty, statisticians and biologists have developed special techniques known as distance sampling, which involve walking long linear transects. On each transect, the perpendicular distance from the observer to the observed animal is recorded. This data is then used to calculate a detectability function,  and then the density is calculated by applying sophisticated mathematical techniques in a widely used software called Distance. 

Recently, TSG member Mathias Tobler and collaborators [1] have used a novel way to solve many of the issues involved in counting lowland tapirs using camera traps. They placed the camera traps on a regular grid using the estimate of home range size, data that we now have for all tapir species in different environments, to set the minimum distance between cameras. By assuming that tapirs have activity centers related to the home range, and by applying very sophisticated statistical techniques (Bayesian statistics), they were able to estimate the density of lowland tapirs, addressing the problem of low detectability in a very elegant way.