Landscape Ecology of Chronic Wasting Disease in Virginia, USA

Abstract

Wildlife diseases often occur under quantifiable and consistent patterns, which can be understood to statistically predict their occurrence and spread across landscapes. Chronic wasting disease (CWD) is a neurodegenerative disease in the deer family Cervidae caused by a prion, a pathogenic and misfolded variant of a naturally occurring protein. Managing and controlling CWD is imperative for conservation of ecologically and economically important cervid species, but unclear transmission mechanisms within landscapes complicate evidence-based management. Gaps of information in the landscape ecology for CWD are particularly pronounced for areas with recent disease emergence and spread, such as within the CWD cluster in the Mid-Atlantic United States. Thus, I identified current gaps in information and sought to fill neglected areas of research, specifically focusing on landscape determinants for CWD occurrence and spread in the state of Virginia. In chapter 2, I conducted a scoping study that collected and synthesized decades of CWD research and identified trends with respect to statistical and mathematical modeling methods used, connectivity within the CWD research community, and the geographic areas from which studies were performed. In chapter 3, I investigated landscape determinants for CWD in Virginia using remote sensing landscape data and an epidemiological dataset from Virginia Department of Wildlife Resources (DWR) using diverse algorithms and model evaluation techniques. Finally, in chapter 4, I modeled landscape connectivity between confirmed CWD cases to examine potential paths and barriers to CWD spread across landscapes. My results indicate that landscape ecology was rarely incorporated throughout CWD's 50+ year history. I provide evidence that remotely-sensed landscape conditions can be used to predict the likelihood of CWD occurrence and connectivity in Virginia landscapes, suggesting plausible CWD spread. I suggest areas of future work by explicitly identifying gaps in CWD research and diagnostic methods from which models are based, and encourage further consideration of host's ecology in modeling. By integrating remotely-sensed data into my modeling framework, the workflow should be easily adaptable to new study areas or other wildlife diseases.