Modeling dynamics of native and invasive species to guide prioritization of management actions

Action to achieve biodiversity conservation is usually expensive, and resources are limited relative to conservation goals. Prioritizing management investment therefore is essential if important goals are to be achieved. New software, the "IslandsDSS," has been developed to prioritize the mix of management actions that will optimally mitigate biodiversity loss. Here, we present novel temporally dynamic models of species population growth, interaction, and management efficacy that have been incorporated into the software. We have analyzed the sensitivity of these models to uncertainty in four parameters: maximum rate of population growth (r(max)), coefficient of species interaction (alpha(ij)), quantity of food resources required to maintain species equilibrium (J(i)), and the coefficient of management efficacy (theta(i)). We focused on the projected abundance of species by simulating interactions among one to four species, both invasive and native, on a hypothetical arid-tropical island that is 1000 ha in size and consists of five evenly distributed habitat types. Sensitivity analysis revealed significant variation in species abundance due to uncertainty in rmax (coefficient = 51.34; P < 0.001) and alpha(ij) (N-i = -16.48; P = 0.43; N-j = -2.33(2); P = 2.00(-16)), a significant but potentially stabilizing effect of modeling multiple species simultaneously (coefficient = -65.80; P = 2.00(-16)), and mirroring by species response trajectories of threat mitigation trajectories. There are several benefits of using temporally dynamic models of species responses to threat mitigation in systematic conservation planning including increased accuracy in estimates of the cost of management; locally relevant understanding of lag-times between threat establishment and unacceptable impacts on valued species; understanding of threat abundance and required intensity of control for biodiversity features to persist; site-and species-specific understanding of time to eradication and threat recovery when management is interrupted; and an improved understanding of the opportunity cost, in terms of threat levels and responses of native species, for islands not selected for management. Our models and associated software are based on decades of ecological research, potentially useful in a wide range of situations, including islands, the mainland, and marine regions, and we suggest that they provide managers with novel and powerful tools to efficiently prioritize conservation actions via the new systematic conservation planning software, " Islands DSS."