748 North American Journal of Fisheries Management 23:748–759, 2003  Copyright by the American Fisheries Society 2003 Responses in Fish Community Structure to Restoration of Two Indiana Streams ASHLEY H. MOERKE* AND GARY A. LAMBERTI Department of Biological Sciences, University of Notre Dame, Notre Dame, Indiana 46556-0369, USA Abstract.—Stream restoration has accelerated in the Midwestern United States during the past decade, but the effects of restoration on stream biota are rarely evaluated. From 1997 to 2000, we studied the responses in fish communities to the attempted restoration of two channelized streams (Juday Creek and Potato Creek) in northwestern Indiana, each of which received two new meanders to a 1-km reach of stream length. The restored meanders of Juday Creek also received major im- provement to instream habitat, bank stabilization, and silt control. In contrast, Potato Creek received only reconnection of the stream to historical meanders. Fish were monitored for 3 years after re- construction by use of electroshocking and salmonid redd surveys. In Juday Creek, trout size-class distribution broadened and redd construction increased in the restored reaches. However, most fish metrics for reconstructed reaches did not surpass the levels in the channelized reaches after 3 years. Continued sedimentation from upstream sources, which reduced habitat quality, likely counteracted the positive effects of the restoration. In contrast, unanticipated geomorphic changes in Potato Creek led to decreased current velocity and highly altered fish community structure. The American brook lamprey Lampetra appendix, a sensitive species, was not collected after restoration, and the fish community changed from rheophilic species to highly tolerant, slow-water species. Overall, changes in fish community structure revealed strengths and weaknesses in contemporary stream restoration approaches, findings that will aid future restoration efforts. At a global scale, aquatic and riparian ecosys- tems are being impacted by human activity at a greater rate than at any other time in history (NRC 1992). The Midwestern United States exemplifies this trend. Since the 1870s, Illinois and Indiana have lost more than 85% of their wetland area, with land use now dominated by agriculture (Dahl 1990; USDA 1992). Rapid agricultural and urban development in the Midwest has contributed to surface water pollution, draining of wetlands, and stream channelization. All of these activities can adversely affect habitat diversity and biological communities in streams and reduce water quality for human uses. Fish communities have often been used to detect stream impairment because they are sensitive to a range of biological, physical, and chemical dis- turbances (Karr 1981). The structure of a fish com- munity is influenced on a local scale by water depth, current velocity, size of substrate particles, cover, and temperature (Rabeni and Jacobson 1993), which also may alter biological interac- tions. Anthropogenic activities such as channeli- zation alter many of these physical features and eliminate the natural pool-riffle sequence of a stream, which may affect the abundance and dis- * Corresponding author: moerke.1@nd.edu Received January 23, 2002; accepted October 24, 2002 tribution of juvenile and adult fishes (Bayless and Smith 1964; Jones 1975). Stream restoration often attempts to reverse anthropogenic degradation by increasing habitat diversity (Gore et al. 1995) and thus favors different organisms, including fishes of various sizes and species. Despite the need to restore degraded habitat and improve water quality of Midwestern streams (NRC 1992), little infor- mation exists on the effects of restoration practic- es. Many studies have demonstrated the impacts of habitat degradation on fishes in coolwater streams (e.g., Elser 1968; Tarplee et al. 1971; Chapman and Knudsen 1980) and conversely the effects of small-scale habitat improvement (e.g., Saunders and Smith 1962; Hunt 1976). However, quantitative assessment of ecological responses to larger scale stream restorations are rare (NRC 1992; Kondolf 1998). As a result, many well- intentioned projects have either failed or caused further damage to the ecosystem (Iversen et al. 1993; Kondolf 1998). Thus, improvements to con- temporary restoration approaches require basic re- search to document ecological responses. Stream restoration can be viewed as large-scale experi- ments that may provide opportunities to assess our current understanding of stream ecosystem struc- ture and function and to test our abilities to suc- cessfully repair degraded ecosystems. Our objective was to determine the effects of