Estimating wildlife disease dynamics in complex systems using an Approximate Bayesian Computation framework.

Area of Science:

• Wildlife disease ecology
• Conservation biology
• Statistical modeling

Background:

• Emerging infectious wildlife diseases pose significant challenges for conservation and management.
• Predicting disease dynamics is difficult due to complex host-pathogen interactions and limited empirical data.

Purpose of the Study:

• To demonstrate the utility of Approximate Bayesian Computation (ABC) for reconstructing disease dynamics in wildlife populations with sparse data.
• To model the impact of bovine tuberculosis (bTB) on the lion population in Kruger National Park.

Main Methods:

• Utilized an Approximate Bayesian Computation (ABC) statistical framework.
• Reconstructed disease dynamics of bovine tuberculosis in lions using limited empirical data.
• Incorporated host demographics, behavior, and disease transmission into the model.

Main Results:

• Bovine tuberculosis (bTB) is likely to infect a substantial portion of the lion population.
• Lions act as spillover hosts for bTB, with long disease latency periods being common.
• Projected lion population decline of approximately 3% over 50 years, stabilizing at a new equilibrium in the absence of aggravating factors.

Conclusions:

• The Approximate Bayesian Computation (ABC) framework offers a novel tool for wildlife managers to model complex disease systems.
• ABC facilitates the inference of unknown parameters in wildlife disease ecology, even with limited empirical data.
• The study provides crucial insights into the long-term effects of bovine tuberculosis on lion populations, informing conservation strategies.