Improved predictions and forecasts of chronic wasting disease occurrence using multiple mechanism dynamic occupancy modeling
2022
Cook, Jonathan D. | Williams, David M. | Porter, William F. | Christensen, Sonja A.
Diseases spread and persist according to complex interactions among hosts, pathogens, and their shared environments. In actively managed wildlife diseases, unanticipated disease spread and persistence can challenge mitigation efforts, necessitating modeling approaches that elucidate and integrate multiple mechanisms. Dynamic occupancy models (DOMs) can predict and forecast species occurrence patterns; however, they have not been extended to explain the complex spread and persistence patterns of chronic wasting disease (CWD). We used a DOM to predict and forecast the persistence, and localized and long‐distance spread of CWD in white‐tailed deer (Odocoileus virginianus) according to distinct ecological relationships that we hypothesized were important. We estimated large differences between the probabilities of disease persistence and localized and long‐distance CWD spread in Wisconsin, USA. Chronic wasting disease was more likely to persist in areas with low clay content in soils, to spread locally as the amount of localized disease increased, and to occur in areas far removed from prior positives in places with a higher proportion of agriculture. Across all landscape compositions, the probability of occurrence decayed by distance from prior positives in a manner consistent with transient movement distances of white‐tailed deer. Finally, a multi‐mechanism DOM was able to produce spatially accurate predictions and forecasts of disease occurrence using only data collected during the first 2 days of the firearm harvest season. Thus, our multi‐mechanism DOM offers advantages over simple linear models that can improve our understanding of the complex ecology driving CWD dynamics and lead to better surveillance and management strategies moving forward.
显示更多 [+] 显示较少 [-]