95th ESA Annual Meeting (August 1 -- 6, 2010)

OOS 43-7 - Population forecasting: Assimilating models and data to understand dynamics of brucellosis in the Yellowstone bison population

Thursday, August 5, 2010: 10:10 AM
401-402, David L Lawrence Convention Center
N. Thompson Hobbs1, Chris Geremia2, P.J. White2, John Treanor2, Rick Wallen2 and Jennifer A. Hoeting3, (1)Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, CO, (2)Yellowstone Center for Resources, National Park Service, Yellowstone National Park, WY, (3)Department of Statistics, Colorado State University, Fort Collins, CO

Conserving the Yellowstone bison represents one of the greatest success stories in the history of wildlife management in North America.  However, success has not come without challenges. The most significant problem confronting those who manage the population today is conflict created by the risk of spread of brucellosis from bison to cattle.  Modern, adaptive management aimed at minimizing this risk requires a model capable of forecasting the population’s dynamics and its disease status. We used a discrete time, stage structured model assimilated with time series of monitoring data and results of detailed process studies to forecast changes in abundance and diseases status of the Yellowstone bison population.  A Bayesian, hierarchical approach allowed use of data from multiple sources, supported estimation of process variance and observation error, and provided true forecasts, that is, predictions with confidence envelopes.


Model forecasts suggested changes in policy could enhance conservation of bison while reducing risks of transmission of brucellosis from bison to livestock.  Episodic, large removals to regulate the size of the population could be replaced by culling a much smaller number of animals annually, thereby mimicking chronic, natural mortality.   Targeting sero-negative animals for culling rather than sero-positive ones could reduce the probability of disease transmission.  A forecasting model can enhance management and policy by allowing evaluation of alternative actions.  These evaluations are made honest by comprehensive assessments of uncertainty.