Ecology & Environmental Sciences PhD
Decision support systems and natural resources management schemes that are based on underlying ecological processes have the potential to be more sustainable, reduce reliance on off-system inputs, and avoid undesirable outcomes due to biological feedbacks. Our goal is to develop a formal approach to agroecology based multi-pest management. In this context, our work focuses on modeling biological causal networks where decision are made under the presence of multiple confounded variables. The pest complex [cheatgrass (Bromus tectorum), wheat stem sawfly (Cephus cinctus), and Fusarium crown rot] was chosen as a case study to evaluate a Bayesian decision theoretic model that combines information on multi-trophic agricultural pests into a single crop management recommendation. Our model can be viewed as a variant of path analysis in that it accounts for each variables direct and indirect effects.