The linear food chain of high school biology textbooks has been replaced by the food web, the increasingly complex network of trophic interactions in an ecosystem. This complexity, however, masks a number of robust statistical properties. Yet much research to date has concentrated upon predicting and testing the existence of such patterns rather than on the factors giving rise to them. In order to study such a complex problem, I incorporate concepts and methods from fields not traditionally aligned with ecology---such as chemical engineering and statistical physics---to develop an integrated approach.

By following this approach, I have demonstrated that there are key universal features common to all ecosystems, independent of variables such as the population and type of species present, assembly history, or particular environment. These universal features include a number of important descriptors of the structure of food webs, as well as the empirically observed mechanism of prey selection. My research provides insight into, for example, how best to develop dynamic ecosystem models to probe the effect of exotic species.