Do trails affect relative abundance estimates of rainforest frogs and lizards?

Abstract

AbstractThe selection of a sampling protocol is critical to study amphibian and reptile communities and in many instances researchers have combined the use of visual encounter surveys (VES) conducted on trails and off trails. The effect of human‐made trails on relative abundance estimates of amphibians and reptiles has been assessed in a few temperate locations, but data are lacking for tropical sites. Our study was designed to address this issue by comparing abundance estimates of frogs and lizards on and off trails in a lowland rainforest in south‐eastern Perú. We used nocturnal VES to sample frogs and lizards along transects established on trails and off trails in two different forest types. We found that the observed relative abundance estimates of frogs and lizards were affected by the location of transects (on trail vs. off trail) and the type of forest (floodplain forest vs. terra firme forest). We also found an interaction between the two main effects, indicating that the effect of transect location with respect to trails varies as a function of habitat. Observed frog abundances were higher on trails than off trails, indicating that studies that include VES on trails will bias relative abundance estimates in contrast to studies that include only VES off trails. We suggest that transects should be established only off trails, especially for monitoring studies because trail use by humans can have a strong influence on observed animal abundance.