Margaret Eppstein: Predicting Naturalization versus Invasion using Stochastic CA Models

Natural Sciences lecture series

Margaret Eppstein of the University of Vermont discusses invasion vs. naturalization in plant models

Margaret talked about what makes plants invasive. Several ideas already exist to explain some of the phenomena: the enemy release hypothesis, evolution of increased competitive ability, the biotic resistance hypothesis and propagule pressure. Models include both positive feedback effects, like pollinators and the presence of symbionts, and negative feedback effects like predators and pathogens. A characteristic like soil chemistry can be both a positive and negative feedback effect.

However, the standard Lotka-Volterra models typically used fail to account for frequency-dependent feedback effects, as well as typically making the assumption of a mean field, meaning that the use of space is ignored.

Margaret has developed a new model that incorporates propagule pressure, and both frequency dependent and independent effects, as well as spatial scale. Her results show that it's important to consider space when modeling whether an exotic species can establish when invading a new environment, as well as both frequency-dependent and independent interactions. This is then modeled as a stochastic CA with a configurable cell neighborhood.

This new model may be a practical way to assess the invasive potential of new species, or to estimate range limits of invasive species.