Considering sampling approaches when determining carnivore diets: the importance of where, how, and when scats are collected

Abstract

Understanding the diets of carnivores is often required to inform their management, to conserve their prey, or to minimize depredation of domestic animals. Scat analysis is one of the oldest and most commonly used methods for determining carnivore diets and many accuracy issues associated with scat analysis have been addressed in other studies. Little attention, however, has been given to the questions of where, how, and when scats are sampled in the field and how these factors can affect conclusions about diet. Based on a review of 64 articles, the two most common sites for collecting scats from grey wolves (Canis lupus) are from wolf homesites (i.e., dens and rendezvous sites) and from along roads. Collections from such areas are often combined and interpreted as depictions of a seasonal diet, with no acknowledgment of the inherent problems associated with sampling at different sites. Rather than combining these two samples of scat, however, we were interested in comparing them. In our field study, we supported our hypothesis that the frequency of prey types differs significantly between wolf scats collected along roads and scats collected at wolf homesites and, therefore, would tell different stories about summer wolf diet if scats had only been collected from one of these sites. This difference is likely due to a combination of three interactions of wolf ecology with our sampling design: local prey availability, timing of scat deposition, and movement behavior. This difference is especially of interest when considering three findings from our review: descriptions of sampling design are often omitted, scats collected from different sites are often combined into a single estimate of diet, and scats are often collected opportunistically. Given the results from our field sampling, these three common practices may significantly affect how carnivore diet data should be interpreted. Furthermore, using our own idiosyncratic study as an example, we discuss other common assumptions in the scat diet literature, including assuming that diet is constant across packs and years. To increase the reliability of diet estimates from scats, researchers should clearly articulate where, how, and when they collected scat samples and should discuss the assumptions of the chosen sampling design so that potential biases and inaccuracies can be assessed.