Restoration through eradication? Removal of an invasive bioengineer restores some habitat function for a native predator.

Invasive aquatic macrophytes increase structural complexity in recipient systems and alter trophic and physical resources; thus, eradication programs that remove plant structure have potential to restore some impaired ecological functions. In this study we evaluate how an invasive ecosystem engineer, Atlantic smooth cordgrass (Spartina alterniflora), interferes with the movement and foraging activity of a mobile predator, Dungeness crab (Cancer magister), and whether removal of aboveground cordgrass structure rapidly reestablishes access to foraging habitats. By 2004, smooth cordgrass had invaded >25% of crab foraging habitat in Willapa Bay, Washington (USA), and transformed it into a highly structured landscape. However, by 2007 successful eradication efforts had eliminated most meadows of the cordgrass. In order to investigate the effect of smooth cordgrass on the habitat function of littoral areas for foraging crabs, we integrated field, laboratory, and statistical modeling approaches. We conducted trapping surveys at multiple sites and used a hierarchical model framework to examine patterns in catches prior to and following cordgrass removal (i.e., before-after control-impact design, BACI). Prior to eradication, catches of Dungeness crabs in unstructured habitats were 4-19 times higher than catches in adjacent patches of live cordgrass. In contrast, the results of post-eradication trapping in 2007 indicated similar catch rates of crabs in unstructured habitats and areas formerly invaded by the cordgrass. Subsequent laboratory experiments and video observations demonstrated that the rigid physical structure of smooth cordgrass shoots reduces the ability of Dungeness crabs to access prey resources and increases the risk of stranding. Taken together, these findings suggest that eliminating the structural complexity of invasive macrophytes may rapidly restore some ecological function (i.e., foraging area) for migratory predators like Dungeness crab. However, restoration of affected areas to a preinvasion state will also depend on long-term patterns of succession in invaded areas and the degree of persistence of physical changes that continue to alter biotic characteristics of the habitat. Our work highlights: (1) the efficacy of employing multiple methods of inquiry to evaluate causal relationships through mechanisms of interaction, and (2) the importance of targeting particular ecological functions when identifying both short- and long-term goals of restoration efforts.