Margaret Palmer, University of Maryland
As human impact on natural systems has increased, there has been a parallel increase in the need and interest in restoring ecosystems. Ecosystems are dynamic, living systems whose attributes depend on complex interactions between physical and biological processes, biota, and the context within which they are embedded. Yet, most restoration has focused on restoring structure – single species like salmon or stable river channels whose appearance mimics past images. I argue that the pre-occupation with restoration of structure, including certain combinations of species or levels of diversity, has hindered progress in improving the success of restoration efforts and in advancing theory to support restoration efforts. The focus should be on restoration of ecosystem processes, particularly their resilience where the latter is defined as the magnitude of disturbance that a system can absorb without a loss of desired ecosystem services or states. Note, this definition assumes there is not a single “right” state; instead, humans act as selective agents by imposing their desires on the restoration enterprise. Once these desires are identified, even if they are structural, the focus should be on restoring biophysical processes such as nutrient transformation, flow regimes, or propagule dispersal that support the desired endpoints. Unfortunately, our understanding of the link between process and structure is rudimentary for most ecosystems. I provide examples of these points using river ecosystems and data on the response of structural and functional attributes to environmental variability and propose hypotheses that I believe we should be spending our time testing.