Movements of Nonnative Brook Trout in Relation to Stream Channel Slope

—We provide new insights on the ability of naturalized brook trout Salvelinus fontinalis to ascend steep, headwater streams in the western USA. We tested hypotheses that upstream movements by brook trout are limited or absent in reaches of steep streams and are more prevalent and longer in gradually sloping streams. We compared brook trout movements in headwater streams in Idaho at sites with varied channel slopes (averages of ,1–12%). After eradicating fish from 200-m stream sections, we assessed immigration of marked fish into these sections. Contrary to our hypothesis, upstream movements were more prevalent than downstream movements during the summer, even in steep streams. Marked brook trout ascended stream channels with slopes of 13% that extended for more than 67 m and 22% for more than 14 m; they also ascended a 1.2m-high falls. Nearly vertical falls, rather than steep slopes per se, apparently inhibited upstream movements. Our hypothesis that upstream movements would decrease with increasing channel slope was partially supported; fish did not move as far upstream in steep as in gradual sites, and upstream movements through steep channels were dominated by larger fish (.135 mm total length). Immigration by marked fish smaller than 95 mm was uncommon in all sites. Slopes up to 13% do not ensure against upstream dispersal, although other mechanisms may inhibit brook trout invasion in steep channels. In very steep channels, fewer dispersing fish and slower upstream movement rates may increase the time required for successful invasion and reduce its likelihood of occurrence. Brook trout Salvelinus fontinalis, a char native to eastern North America, have been introduced to cold water streams and lakes throughout western North America (MacCrimmon and Campbell 1969; Meehan and Bjornn 1991) and have successfully invaded many waters beyond where they were intentionally stocked. They are presently the second most widely distributed salmonid species (native or introduced) in the interior Columbia River basin, surpassed only by introduced rainbow trout Oncorhynchus mykiss (Thurow et al. 1997). Brook trout have been implicated in reducing populations of some native salmonids (Fausch 1989; papers in Howell and Buchanan 1992; Leary et al. 1993; Dunham et al. 1999), as well as other ver* Corresponding author: [email protected] 1 Present address: U.S. Forest Service, Southern Research Station, 1000 Front Street, Oxford, Mississippi, 38655-4915, USA. Received October 5, 1998; accepted October 13, 1999 tebrate and invertebrate fauna (Dawidowicz and Gliwicz 1983; Bradford 1989; Bechara and Moreau 1992; Bradford et al. 1993). Use of barriers to prevent brook trout invasion or reinvasion of streams is increasingly prevalent, but not, perhaps, always justified. Recently, concerns over declines in and local extirpations of bull trout S. confluentus and cutthroat trout O. clarki have led to expensive attempts to eradicate brook trout from streams and lead to the construction of barriers to prevent their reentry (Dambacher et al. 1992; Thompson and Rahel 1998). Artificial barriers have also been considered as a means to prevent brook trout from invading places they have not previously occupied (Kershner 1995; Thompson and Rahel 1998), even though the ability to predict future invasions is limited. Barriers may hinder movements of native fishes, however, which could disrupt adaptive migration and dispersal patterns (Gowan et al. 1994) and exacerbate declines in native fishes by increasing population fragmen-