Detecting regime shifts in artificial ecosystems

Abstract

Ecosystems are subjected to a range of perturbations that have the potential to induce relatively sharp transitions in states. These can be referred to as regime shifts or critical transitions. They may be driven by perturbations that vary over a wide range of spatial and temporal scales, from responses to deforestation within a small field to responses to the gradual increase of carbon dioxide in the Earth s atmosphere. Here we investigate potential early warning signals that may presage regime shifts in model ecosystems. We hypothesise and model a relationship between biodiversity and community structure that influences ecosystem structure. We argue that Artificial Life methodologies have potential to make substantial contributions to efforts searching to predict large changes in ecosystems and other elements in the Earth system, as there is a recognised limitation in empirical data and ability to conduct experiments in the real-world. Consequently simulation and exploration of the low-level mechanisms that give rise to regime shifts in artificial in-silico ecosystems represents a useful line of enquiry.