Image

A fundamental goal of community ecology is to understand and predict patterns of biodiversity within and across ecological communities.

One approach is to construct detailed individual-based species community models in which the many underlying processes - e.g. birth, growth, competition, death, migration - are represented explicitly. However, the inherent complexity of such models makes them difficult to parameterise and limits their usefulness.

In this talk I'll discuss a novel (and simpler) statistical approach inspired by how complex systems are modelled in physics, based on the principle of maximum entropy. In this approach many of the underlying mechanistic details are essentially treated as random noise (“chance”) and only a few key dynamical constraints - e.g. community resource balance, available space - are modelled explicitly (“necessity”).

I'll discuss how this approach has led to new insights into some long-standing questions concerning the relationships between diversity, productivity and stability in plant communities.

Prof. Roddy Dewar’s postgraduate and postdoctoral research was in theoretical physics (Edinburgh and Oxford). He then morphed into a plant and ecosystem modeller, holding positions at the Institute of Terrestrial Ecology (Edinburgh), UNSW (Sydney) and the Institut National de la Recherche Agronomique (Bordeaux) before joining ANU in 2008. His current research lies at the interface between physics and biology, and aims to uncover and exploit common organisational principles governing the emergent behaviour of complex systems such as plants, ecosystems, turbulent fluids and planetary climates.

This seminar is part of the Plant Biology Seminar Series.