Funding by USGS and Pennsylvania Game Commission.
Chronic wasting disease (CWD) is a transmissible spongiform encephalopathy that was first detected in 1967 in a captive research facility in Colorado. In the northeastern United States, CWD was first confirmed in 2005 in New York and West Virginia, and has also been found in Maryland, Virginia, and Pennsylvania. We examined demographic and environmental factors in the central Appalachian region to assess the spatial distribution of CWD in white-tailed deer (Odocoileus virginianus). The objectives of our study were to (1) apply Bayesian hierarchical modeling to harvest location data of white-tailed deer tested for CWD in the region since 2005, (2) identify model(s) that best described the spatial distribution of CWD, and (3) map probability of CWD infection.
Demographic covariates included age and sex, and environmental covariates included elevation, slope, riparian corridor, percent clay in the soil, and percent of three habitat types (developed, forested, open). For each deer, environmental covariates were extracted within 6 km2 grid cells as this size reflected our estimate of the 99% size of home range for white-tailed deer in the region using Brownian Bridge Movement Models. The model with the most support contained random spatial effects and percent habitat and accounted for 94.4% of the overall weight for the candidate set of models. Percent forest cover appeared to have the strongest correlation with the distribution of CWD in the region, with increased risk of CWD occurring in areas that had lower amounts of forest cover. Our results will assist resource managers in understanding the spatial distribution of CWD not only within the study area, but also in surrounding areas where CWD has yet to be found. Efficiency of disease surveillance and containment efforts can be improved by allocating resources used for surveillance into areas that are at a greater risk for infection.
A historical over-view of CWD in the northeast is in press in the Journal of Fish and Wildlife Management and the primary modeling manuscript will be submitted in late-summer 2014.
