Environmental field conditions and sampling effort affect the molecular identification success of livestock predators

For centuries, wolf depredation on livestock has triggered human–wildlife conflicts throughout Europe. Free-ranging dogs, however, are increasingly abundant and may also act as predators of livestock herds. This calls for combined efforts aimed at identifying the causes of depredation events and optimizing the procedures towards predators’ identification. Here, we analysed 56 livestock depredation events in central Portugal, an area where conflict between wolves and human populations takes place. We estimated the mean minimum sampling effort (number of swabs) required to detect at least one potential predator and examined how extrinsic factors (i.e. time, meteorological conditions and vegetation cover) drive sample degradation and predator identification success. Free-ranging dogs were the only putative predator detected in most attacks (66%). Results indicate that a minimum of three swabs are needed to detect at least one predator, but using at least four would substantially increase the detection rate. We found that the longer the interval between an attack and sample collection and the higher the local humidity, the lower is the probability of identification success. On the other hand, higher temperatures and precipitation levels seem to be associated with a higher probability of success. The unexpected positive effect of precipitation may be linked to specific environmental contexts (i.e. higher precipitation levels in colder weather may still favour sample conservation). As identification success depends on time and weather conditions, the time span between a depredation event and sample collection should be reduced whenever possible, and sufficient samples should be collected to ensure an adequate detection success.