Managers are faced with considerable uncertainty about the systems they manage; yet, they are still expected to optimize weed management. A wide variety of mathematical models can be used to improve the efficacy and implementation of biocontrol for invasive plants. In this article, applied insights that have been generated by mathematical models of biological control systems are described. First, the ways in which models can be used to help choose appropriate agents for biocontrol are outlined. In particular, a novel example showing how demographic and dispersal models can be linked to address the life-history stage to target to reduce either local abundance or spatial spread of weeds is presented. Second, the use of stochastic optimization models to improve biological control release and redistribution strategies is reviewed. The discussion outlines the way in which learning about uncertainties can be explicitly included in the management process. These two topics demonstrate the use of ecological models for improving the targeting and implementation of biological control of weeds.
Additional index words: Optimization, release strategies, spatial spread, uncertainty.
Abbreviation: AAM, active adaptive management.