Density estimation of sound-producing terrestrial animals using single automatic acoustic recorders and distance sampling

Abstract

Obtaining accurate information on the distribution, density, and abundance of animals is an important first step toward their conservation. Methodological approaches using automatic acoustic recorders for species that communicate acoustically are gaining increased interest because of their advantages over traditional sampling methods. In this study, we created and evaluated a protocol to estimate population density, which can be used to compute abundance of terrestrial sound-producing animals from single automatic acoustic recorders and using an automatic detection algorithm. The protocol uses cue rates from the target species, environmental conditions, and an estimate of the distance of the individual to the recorder based on the power of the received sound. We applied our protocol to estimate the density of a Hawaiian forest bird species (Hawaiˊi ˊAmakihi [ Chlorodrepanis virens ]) on the island of Hawaiˊi, USA. We validated our approach by comparing our density estimates with those calculated at the same stations using a traditional point-transect distance sampling method based on human observations. Overall density estimates based on recorded signals were lower than those based on human observations, but 95% confidence intervals of the two density estimates overlapped. This study presents a relatively simple but effective protocol for estimating animal density using single automatic acoustic recorders. Our protocol may easily be adapted to other sound-emitting terrestrial animals.