Abstract
Streams are being subjected to physical, chemical, and biological stresses stemming from both natural and anthropogenic changes to the planet. In the face of limited time and resources, scientists, resource managers, and policy makers need ways to rank stressors and their impacts so that we can prioritize them from the most to least important (i.e., perform 'ecological triage'). We report results from an experiment in which we established a periphyton community from the Huron River (Michigan, USA) in 84 experimental 'flumes' (stream mesocosms). We then dosed the flumes with gradients of six common stressors (increased temperature, taxa extinctions, sedimentation, nitrogen, phosphorus, and road salt) and monitored periphyton structure and function. A set of a priori deterministic functions were fit to each stressor-endpoint response and model averaging based on AICc weights was used to develop concentration-response best-fit predictions. Model predictions from different stressors were then compared to forecasts of future environmental change to rank stressors according to the potential magnitude of impacts. All of the stressors studied altered at least one characteristic of the periphyton; however, the extent (i.e., structural and functional changes) and magnitude of effects expected under future forecasts differed significantly among stressors. Elevated nitrogen concentrations are projected to have the greatest combined effect on stream periphyton structure and function. Extinction, sediment, and phosphorus all had similar but less substantial impact on the periphyton (e.g., affected only structure not function, smaller magnitude change). Elevated temperature and salt both had measurable effects on periphyton, but their overall impacts were much lower than any of the other stressors. For periphyton in the Huron River, our results suggest that, among the stressors examined, increased N pollution may have the greatest potential to alter the structure and function of the periphyton community, and managers should prioritize reducing anthropogenic sources of nitrogen. Our study demonstrates an experimental approach to ecological triage that can be used as an additional line of evidence to prioritize management decisions for specific ecosystems in the face of ecological change.