Taxonomic and ecological tradeoffs associated with small dam removals

Dams and impoundments are widely recognized as having dramatic, negative impacts on freshwater ecosystems and are frequently implicated in the plight of endangered riverine biota (Dudgeon, 2000; Lydeard et al., 2004; Strayer and Dudgeon, 2010; Burkhead, 2012; Haag and Williams, 2013). As such, their removal is a desirable and necessary component of stream restoration projects. Despite this general sentiment, there is a practical and financial need to prioritize among dams for removal and assess both the benefits and costs to stream ecosystems. Palmer et al. (2005) and others have called for standards to evaluate stream restoration success and specified five key traits of successful restoration projects. The first criterion they identified was that restoration must be guided by an image of a more dynamic, healthy river. Second, restoration should result in measurably improved stream condition and a more resilient self-sustaining system. Nearly all dam removal projects meet these criteria both inherently and quantitatively. In addition, Palmer et al. (2005) suggested that stream restoration should not cause irreparable harm to the ecosystem and that preand post-restoration monitoring data be made available to the public. While these may seem like eminently attainable and transparent goals, the reality is that financial restrictions limit the temporal scale of preand post-removal monitoring and restrict the ability to describe recovery intervals accurately, which may in turn impede an accurate assessment of the long-term effects of dam removals. Popular concepts of dam impacts are often focused, quite naturally, on their upstream effects on stream habitats. Dams transform free-flowing reaches to lentic habitats, restrict downstream sediment movement, and dramatically alter temperature, nutrient and productivity dynamics. Dams also alter downstream habitats and much work has documented physicochemical and biotic changes in downstream reaches. Much of what is known about dam impacts is derived from studies of high hydroelectric dams on large (i.e. >6th order) streams (Baxter, 1977; Graf, 2006). However, low-head dams (i.e. those <7.5m height) greatly exceed hydroelectric dams in number and thus affect a much broader range of stream sizes and ecosystem types (Graf, 1999; Csiki and Rhodes, 2010). Removing small dams has become a major part of stream and faunal restoration projects in North America. However, because so little is known about the effects of smaller dams on stream biota and ecosystems, obvious solutions may turn out to be problematic for other species. For example, because rivers across the globe have