Abstract
Determining the causes of biological impairment in urban stream settings presents unique challenges because there are many potential stressors associated with human development. A rigorous, scientifically based process is more likely to identify influential stressors that can be reduced to improve stream condition. We used the U.S. Environmental Protection Agency's (U.S. EPA) CADDIS (Causal Analysis/Decision Information System) stressor identification process to assess eight candidate causes in the urban Soos Creek Basin in Washington State. The eight candidate causes capable of negatively affecting the abundance and diversity of benthic macroinvertebrates are: flow alteration, increased fine sediments, reduced habitat complexity, elevated water temperature, low dissolved oxygen, elevated nutrients, increased ionic concentration, and toxic pollutants. We assembled multiple lines of evidence, as well as the consistency of that evidence and agreement with other assessments. We evaluated the influence of natural and cumulative anthropogenic stressors on macroinvertebrate communities by comparing various chemical, physical, and biological measures at sites in the Soos Creek Basin with regional reference sites. Of the stressors evaluated, flow alteration, increased fine sediments, and loss of habitat complexity were the most probable causes of biological impairment, with multiple biological metrics responding predictably across levels of impairment. Key findings from this study include: the use of specific community alterations as evidence in causal assessment, demonstration of links in a complete causal pathway, and the use of multiple models to show which pathway is likely stronger. In addition to the value to the specific case, the analyses increased our understanding of the responses of stream invertebrate communities in urban environments. Ultimately, demonstrating the utility of causal assessment in a practical situation provides greater confidence that mitigation efforts aimed at improving biological health of urban stream communities will have detectable desired effects while also providing a baseline from which the effectiveness of management practices can be evaluated.