Long-Term Effect of Instream Habitat-Improvement Structures on Channel Morphology Along the Blackledge and Salmon Rivers, Connecticut, USA DOUGLAS M. THOMPSON Department of Physics, Astronomy and Geophysics Connecticut College New London, Connecticut 06320, USA ABSTRACT / Habitat-improvement structures on the Black- ledge and Salmon rivers date back to the 1930s and 1950s. Forty of these structures were investigated to determine their long-term impact on channel morphology. These structures include designs that continue to be used in modern restora- tion efforts. During the intervening period since these struc- tures were introduced, several major floods have affected the two channels. The floods include three flows in excess of the 50-year event, including the flood of record, which has an es- timated recurrence interval of almost 300 years. Despite the extreme flooding, many structures were discovered in varying conditions of operation. Grade-control structures and low-flow deflectors generally create some low-flow habitat (P = 0.815) but do not produce the depth of water predicted by design manuals (P  0.0001). Unintended erosion has developed in response to many of the channel modifications especially along the outside of meanders. In addition, the mode of failure of grade-control structures has created localized channel wid- ening with associated bank erosion. Meanwhile, cover struc- tures have produced a 30% reduction in streamside vegeta- tion with over 75% less overhead cover than unaltered reaches. Based on these results, it is important for prospec- tive designers to carefully consider the long-term impacts of instream structures when developing future channel-restora- tion projects. Despite the increased demand for successful chan- nel-restoration techniques, few studies have docu- mented the long-term response of rivers to the em- placement of habitat-improvement structures. The Blackledge and Salmon rivers in Connecticut, USA, both underwent periods of restoration activity dating back to the 1930s. The Blackledge River also underwent a second major phase of channel relocation work in the 1950s. Fortunately, these rivers still contain good exam- ples of historic restoration techniques including many varieties of habitat-improvement structures. Many of the basic designs for individual structures are still widely used in modern restoration efforts. Because many of the structures were emplaced more than 40 – 60 years ago, they have endured an extended pe- riod of normal flow and several extreme floods. Clearly, the Blackledge River and Salmon River sites offer a unique opportunity to evaluate the longevity and effec- tiveness of different types of habitat structures in gravel- and cobble-bedded channels. Therefore, a study was undertaken to determine the current state of habitat- improvement structures and their geomorphic impact on the channel. Based on these results, even where structures remain in these types of forested rivers, tra- ditional restoration techniques do not replicate natural channel processes and most currently fail to create the improvements envisioned by restoration designers in the 1930s and today. Historic Channel Restoration with Instream Structures Channel-restoration projects can take many forms but often focus on the enhancement of aquatic habitat represented by the pool and riffle sequence (Morris 1995, Barinaga 1996). Modern approaches suggest that creation of pool and riffle habitat is one of the most important criteria for evaluating the success of channel- restoration projects (Kondolf and Micheli 1995). Riffles tend to support benthic invertebrate populations as well as young and small fish, while pools can support larger fish and provide critical habitat during low flow (Gordon and others 1992). Studies recommend mini- mum water depths of 20 – 60 cm to ensure adequate low-flow habitat for fish (Arthur 1936, White and Bry- nildson 1967). Although restoration techniques often attempt to replicate natural features, many researchers have noted that the restoration of geomorphic form does not necessarily result in the restoration of the appropriate geomorphic process (Sear 1994, Morris 1995, Barinaga 1996, Goodwin and others 1997, Poff and others 1997). Because most restoration techniques KEY WORDS: Pool-riffle morphology; Stream restoration; Grade-con- trol structure; Deflector; Cover structure DOI: 10.1007/s00267-001-0069-0 Environmental Management Vol. 29, No. 1, pp. 250 –265 © 2002 Springer-Verlag New York Inc.