Data from: Wildlife whodunnit: forensic identification of predators to inform wildlife management and conservation, collected in northern, Washington, USA from 2017-2021

• Main Authors: Taylor Ganz, Melia DeVivo, Ellen Reese, Laura Prugh
• Format: Dataset
• Language: unknown
• Published: Knowledge Network for Biocomplexity 2022
• Subjects: carnivore; DNA; forensic analysis; genetic; mortality; predation; predator identification; saliva; ungulate; Cervus canadensis; Odocoileus hemionus; Odocoileus virginianus; Canis lupus; Canis latrans; Puma concolor; Lynx rufux; Ursus americanus; Lynx
• Online Access: https://search.dataone.org/view/urn:uuid:cddc83b0-04a5-4b4a-b072-5438d4de3ce1

Genetic evidence at predation mortality sites is rapidly improving predator-prey interaction studies and can provide information beyond field-based investigations alone. However, factors contributing to the retention of genetic evidence have received limited investigation in a field setting, and researchers have yet to leverage genetic evidence to improve traditional field investigations. Using data from 61 mortality investigations of mule deer (Odocoileus hemionus), white-tailed deer (O. virginianus), and elk (Cervus canadensis), we evaluated factors influencing predator DNA amplification success and misidentification of predators in field investigations. We found that predator DNA was detected more for prey with higher body mass (18.5% increase per standard deviation [23.1 kg] in carcass body mass above the mean [32.8 kg]). Predator DNA was also 27.0% more likely to amplify when collected from kill sites that had not undergone a freeze-thaw cycle between the mortality and the investigation. The delay between the kill and the investigation, the swabbing surface, and the amount of precipitation did not influence amplification of predator DNA. Misidentifications of the predator based on the field ID were not influenced by the investigation delay or investigator confidence level, suggesting that investigators should collect genetic evidence even when they feel certain about the predator. Errors in identifying the predator during the field investigation increased for prey with smaller body mass, and the predator was actually more likely to be misidentified than correctly identified for fawns and calves <~21 kg. Black bears (Ursus americanus), bobcats (Lynx rufus), cougars (Puma concolor), coyotes (Canis latrans), and wolves (Canis lupus) were equally likely to be missed in a field investigation, but bobcats tended to be falsely assigned more than expected and cougars were falsely assigned as the predator less than expected. Using genetic evidence as validation, we showed how patterns of predation and the field signs left by predators differed for some species depending on the size of the prey. Our findings should help researchers and managers to optimize their use of genetics to enhance field investigations.