Robust surveillance and control of invasive species using a scenario optimization approach
Uncertainty about future outcomes of invasions is a major hurdle in the planning of invasive species management programs. We present a scenario optimization model that incorporates uncertainty about the spread of an invasive species and allocates survey and eradication measures to minimize the number of infested or potentially infested host plants on the landscape. We demonstrate the approach by allocating surveys outside the quarantine area established following the discovery of the Asian longhorned beetle (ALB) in the Greater Toronto Area (GTA), Ontario, Canada. We use historical data on ALB spread to generate a set of invasion scenarios that characterizes the uncertainty of the pest's extent in the GTA. We then use these scenarios to find allocations of surveys and tree removals aimed at managing the spread of the pest in the GTA. It is optimal to spend approximately one fifth of the budget on surveys and the rest on tree removal. Optimal solutions do not always select sites with the greatest propagule pressure, but in some cases focus on sites with moderate likelihoods of ALB arrival and low host densities. Our approach is generalizable and helps support decisions regarding control of invasive species when knowledge about a species' spread is uncertain.