A test of motion‐sensitive cameras to index ungulate densities: group size matters

The use of species detection rates gathered from motion‐sensitive cameras as relative abundance indices (RAIs) could be a cost‐effective tool to monitor wildlife populations; however, validations based on comparisons with reference methods are necessary. We considered 3 ungulates, wild boar (Sus scrofa), roe deer (Capreolus capreolus), and fallow deer (Dama dama), and compared 2 different RAIs with independent indices of density obtained through feces counts across 3 summers (2019–2021) in a protected area of central Italy. We estimated the number of detections per day (RAIevents), and the number of individuals per day (RAIindividuals) from remote camera videos. Both indices were correlated with density estimates, yet only RAIindividuals correctly ranked interspecific densities. Values of RAIevents for the most abundant and gregarious ungulate (i.e., wild boar) were biased low and were lower than those of fallow deer. The uncertainty of RAIs was acceptable for the 2 most abundant study species (CVs ≤ 25%) but was greater for roe deer. At the intra‐specific level, density estimates and RAIs showed comparable but slight inter‐annual variation. Our results support the use of RAIs derived from motion‐sensitive cameras as a promising and cost‐effective tool to monitor ungulate populations, and researchers should incorporate group size into monitoring. We advocate the necessity of field tests based on comparison with locally reliable reference methods to validate the use of motion‐sensitive cameras. Cameras could be used to estimate relative abundance indices (RAIs) as cost‐effective tools to monitor wildlife populations, but validation tests based on comparisons with reference methods are necessary. For 3 ungulate species, we found that RAIs based on species' group size correlated with independent estimates better than traditional RAIs based on the raw count of species' detections. For gregarious species, RAIs based on camera detections should consider group size.