Predator species identification from saliva at kill sites with limited remains

Thesis (Master's)--University of Washington, 2016-06Kill site investigations can provide valuable data about elusive predator species and predator-prey interactions for wildlife scientists and managers alike. However, the predator species must be definitively identified as the one responsible for the predation event. The traditional method of visually analyzing prey remains to identify predator species is prone to observer bias and may be particularly challenging in systems with congeneric predators. Often, so few remains are left behind that visual analysis is rendered impossible. Other potential evidence left by predators, such as scat or tracks, may be difficult to reliably link to the responsible predator or require adequate substrate conditions. Continuing advancements in molecular techniques provide a more objective option that is also feasible on limited remains: swabbing for predator DNA from saliva. This underutilized method has primarily been applied to larger carcasses that provide ample surface area to collect epithelial saliva cells. This study demonstrates the usefulness of a saliva-swabbing method for smaller prey with minimal surface area, including kill sites with almost no prey remains or even just a radio-collar left behind. This is also the first study to compare saliva-swabbing success by sample type (carcass remains versus the radio-collar). From 2010-2013, I employed forensic techniques to increase certainty about predator species identification at snowshoe hare (Lepus americanus) kill sites for the ultimate purpose of elucidating predator-prey interactions. Predator saliva at kill sites was sampled with foam buccal swabs, stored in lysis buffer for shelf-stable preservation, and tested for predator mitochondrial DNA using the polymerase chain reaction (PCR) and species-specific primers. This saliva-swabbing method yielded definitive and objective predator species identification for the majority (58.5%) of sampled kill sites (N = 31/53). Not only were small amounts of remains (carcass and/or fur) often able to yield DNA from predator saliva, but saliva swabs from radio-collars proved to be significantly more successful for predator species identification (4.04 odds ratio). Saliva-swabbing also provided identifications at 65.5% of kill sites lacking unambiguous predator sign and contributed significantly more predator identifications throughout the year when compared to tracking methods. The expansion of saliva-swabbing methods to smaller prey, radio-collars, and limited remains allows for more definitive predator identifications at kill sites than have been possible with commonly employed methods, thus augmenting the potential to understand and manage for predators and pre