Optimization of sampling effort in carnivore surveys based on signs: A regional-scale study in a Mediterranean area

Abstract

Understanding species distribution is central to ecology, evolution and wildlife conservation. However, detecting species presence in the field may be challenging, especially for scarce, elusive and wide-ranging animals such as mammalian carnivores. Here we undertook a large-scale methodological study to assess scat sampling effort required to detect five mesocarnivore species in a Spanish Mediterranean area. Our goals were to: (1) evaluate the optimal effort (i.e. transect length) needed to detect scats of each species; and (2) examine if this sampling effort depends on environmental factors. First, we constructed “curves of first detection” for each carnivore species and evaluated the most cost-effective sampling effort. Second, we used Generalized Linear Mixed Models (GLMMs) to relate the transect length to first detection of each species in each positive transect to a comprehensive set of environmental variables. Our results suggested that sampling effort in this study, namely the transect length needed to detect a given species, can be substantially shortened for two species (red fox and stone marten) with a relatively low reduction in the information on their presence. The most efficient scenario corresponded to a transect length of 1.5km, where red foxes and stone martens would be detected in 72.95 and 65.43% of the positive transects, respectively. However, different sampling approaches would be advisable for the remaining species. We also found that the sampling effort was weakly affected by environmental constraints. Our study may contribute to the design of more efficient scat-based carnivore sampling methodologies elsewhere. In particular, we propose that our results can be successfully extrapolated to other areas of the Iberian Peninsula with close abundances of the species considered and similar environmental characteristics.