Detecting the response of fish to habitat alterations in freshwater ecosystems

Rushed efforts to repair freshwater ecosystems are obstructing the scientific assessment of successes and failures. Existing ecosystem science is ignored as ecologists plan restoration projects. The problem of detecting fish responses to habitat alterations is reviewed via linked issues: expectations, measures, variability, and scale. Habitat alterations directed at single species populations usually affect other ecosystem components. In the fish community, no single population is the reference point for detecting responses. A set of 15 alternate responses highlights the need for an array of ecological indicators if alternatives are to be distinguished. Potential measures are surveyed. Patterns and sources of spatial and temporal variability are identified. Integrative indicators such as production and community indices may be more suitable for detecting changes as they exhibit less variability. The spatial and temporal scales of ecosystem responses are usually greater than the scales of human interventions and assessment. Guidelines for designing assessments are recommended. Management actions are experiments and must be assessed as such. Learning requires systematic observation with replicates and controls. Restoration plans must focus more on ecosystems and less on preferred species. Given the scales of experimentation required, larger opportunities for useful science must be created by wider coordination, consultation, and cooperation.