Prediction and error in multi-stage models for spread of aquatic non-indigenous species

Invasions of ecosystems by non-indigenous species (NIS) are occurring at increasing rates globally (Gollasch, 2006; Ricciardi, 2007; Hulme, 2009). Proactive efforts to reduce invasions are the most cost-effective management option (e.g. Leung et al., 2002; Finnoff et al., 2007), although managers may be unwilling to ‘risk’ a preventative approach because of the high uncertainty inherent in preventative practices relative to postestablishment control (Simberloff, 2003; Finnoff et al., 2007). Thus, a key challenge for invasion biologists exists with respect to forecasting dispersal and establishment of NIS to inform the most appropriate management decision (see Lodge et al., 2006). Department of Biological Sciences, University of Alberta, Edmonton, AB, T6G 2E9, Canada, Centre for Mathematical Biology, University of Alberta, Edmonton, AB, T6G 2G1, Canada, Department of Mathematical and Statistical Sciences, University of Alberta, Edmonton, AB, T6G 2G1, Canada, Great Lakes Institute for Environmental Research, University of Windsor, 401 Sunset Avenue, Windsor, ON, N9B 3P4, Canada