Using replicated empirical networks to understand drivers of ecosystem structure and stability

University Of California-Santa Barbara, Santa Barbara CA

Investigators

Abstract

Science still does not understand the basic drivers of ecosystem structure (e.g., complexity) or how an ecosystem's structure affects its ability to resist disturbance. Understanding these relationships is necessary for predicting the response of ecosystems to many kinds of disruption (e.g. species extinction, invasion, habitat loss). This conceptual shortcoming thus prevents the efficient use of scarce conservation resources by limiting our ability to identify and conserve systems most sensitive to human disturbance. By comparing ecosystem structure across a series of islands that vary in ecosystem size and productivity - this project will examine the extent to which these basic properties drive ecosystem structure and how that structure determines ecosystem stability. This work should lead to broad management insight on the importance of environmental characteristics in determining the impact of species removals and invasions. This project will train multiple graduate students, undergraduates, and a postdoctoral researcher, with a focus on underrepresented groups. Furthermore, through partnerships with the university Kids in Nature program, the researchers will also engage grade school children, and train elementary school teachers in ecology and invasive species. Ecological complexity makes it difficult to identify general patterns in nature, such as community stability. To understand what drives stability, one could measure changes in systems over time across environmental gradients. Another approach has been to consider how system structure affects community stability. However, it is unlikely that structure and environmental gradients are independent, because environmental factors might alter system structure as well as their stability. To understand how ecosystem size and productivity influences ecosystem stability, 23 high-resolution food-webs from a series of Pacific islets that vary independently in size and productivity will be assembled and compared. A suite of mathematical modeling approaches will then create predictions for how variation in food-web structure affects food-web stability. These model-generated predictions of stability will then be tested against observed changes in food-web structure before and after the removal of a common omnivore, the rat (Rattus rattus). These results can then be used to evaluate the relative importance of ecosystem size, productivity, and food-web structure, in predicting system stability. Cumulatively, these efforts will help identify not only how environmental characteristics structure communities but also the extent to which they drive system-level responses to perturbation.