Assessing the potential impacts of a changing climate on the distribution of a rabies virus vector

Common vampire bats (Desmodus rotundus) occur throughout much of South America to northern México. Vampire bats have not been documented in recent history in the United States, but have been documented within about 50 km of the U.S. state of Texas. Vampire bats feed regularly on the blood of mammals and can transmit rabies virus to native species and livestock, causing impacts on the health of prey. Thus cattle producers, wildlife management agencies, and other stakeholders have expressed concerns about whether vampire bats might spread into the southern United States. On the other hand, concerns about vampire-borne rabies can also result in wanton destruction at bat roosts in areas occupied by vampire bats, but also in areas not known to be occupied by this species. This can in turn negatively affect some bat roosts, populations, and species that are of conservation concern, including vampire bats. To better understand the current and possible future distribution of vampire bats in North America and help mitigate future cattle management problems, we used 7,094 vampire bat occurrence records from North America and species distribution modeling (SDM) to map the potential distribution of vampire bats in North America under current and future climate change scenarios. We analysed and mapped the potential distribution of this species using 5 approaches to species distribution modeling: logistic regression, multivariate adaptive regression splines, boosted regression trees, random forest, and maximum entropy. We then projected these models into 17 "worst-case" future climate scenarios for year 2070 to generate hypotheses about how the vampire bat distribution in North America might change in the future. Of the variables used in this analysis, minimum temperature of the coldest month had the highest variable importance using all 5 SDM approaches. These results suggest two potential near-future routes of vampire bat dispersal into the U.S., one via southern Texas, and a second into southern Florida. Some of our SDM models support the hypothesis that suitable habitat for vampire bats may currently exist in parts of the México-U.S. borderlands, including extreme southern portions of Texas, as well as in southern Florida. However, this analysis also suggests that extensive expansion into the south-eastern and south-western U.S. over the coming ~60 years appears unlikely.