Study design and body mass influence RAIs from camera trap studies: evidence from the Felidae

Camera trapping is a widespread tool used to study mammals and to derive abundance and density estimates of wild populations. For the majority of Felidae species, individuals are identifiable from camera trap images because of their unique coat patterns; therefore, camera trapping is used extensively for felids to build individual capture histories. For those species not individually identifiable, one can be tempted to use trap success rate as a relative abundance index (RAI) to inform conservation and management decisions, although robust scientific approach can be used for this scenario. Indeed, recent studies have shown that RAIs can be biased by species-specific behaviour, camera trap deployment and by the body size of the species. Here, we investigate how camera trap rates (henceforth RAIs) vary between species within the Felidae and whether these RAIs are related to body mass or camera-trapping approach (whether the species is targeted or not) accounting for likely confounding factors such as trapping effort, inter-trap distance and number of deployed cameras. We explored all the literature available from primary scientific databases dealing with camera trap studies in Felidae. The literature search resulted in 179 studies from which 513 records for 30 felid species from 53 countries were obtained. Across species, the number of detections (accounting for trapping effort) was positively related to species log-body mass and to the targeting of species in the survey. Hence, comparing RAIs across species of different body size and within species with different camera-trapping approaches might be misleading. Accounting for variability in species body mass is of primary importance when designing multi-species camera trap survey, and in the subsequent analysis of data, in addition to conducting trials to test for species-specific camera-trapping responses.