Genetic Diversity in Steelhead before and after Conservation Hatchery Operation in a Coastal, Boreal River

—Allelic variation at 10 DNA microsatellite loci was assayed in scale samples of steelhead Oncorhynchus mykiss (n 5 333) collected between 1976 and 2003 from the lower main stem of the Kitimat River, British Columbia. Our objectives were to (1) investigate the genetic diversity of wild steelhead populations in the river before hatchery stocking began in 1984 (baseline samples: 1976–1977, 1983–1984; n 5 145) and (2) assess the potential genetic impacts of interbreeding of returning hatchery adult fish with wild spawners over almost 20 years of large-scale hatchery operation (1987–2003; n 5 188). The annual target number of wild broodstock adults used for hatchery production was 40 (20 of each sex) but varied from 9 to 39 among years. The level of population subdivision (u) among Kitimat River samples was low (0.004) and not significantly different from 0. Tests of u between prehatchery and posthatchery operation indicated no significant changes. Similar results were obtained using other measures of genetic differentiation (principal components analysis of microsatellite allele frequencies and Cavalli-Sforza genetic distance). Our data, however, did indicate a slight but significant reduction in allelic richness after hatchery stocking. Pairwise tests for genetic differentiation among samples from different yearclasses were nonsignificant. We conclude that for the current management regime there is little apparent impact of hatchery practices on either the genetic structure or variation within the lower main-stem Kitimat River steelhead, but there may be a reduction in rare alleles. The practice of using substantial numbers of wild fish and multiple year-classes in the hatchery may have minimized genetic changes via genetic drift. Releases of cultured fish into wild populations and their subsequent interbreeding may have genetic effects on natural fish populations, but the effects can be complex and unpredictable (e.g., Hindar et al. 1991; Utter 1998; Reisenbichler 2004). The ecological basis for the wide range of outcomes—from no detectable effect to complete displacement (see review by Hindar et al. 1991)— is not always clear. Introduced nonnative fish may reproduce less successfully, presumably because they are not well adapted to the new, local environments (e.g., Chilcote 2003; Kostow et al. 2003; McLean et al. 2003). In the instances where ge* Corresponding author: [email protected] Received March 15, 2005; accepted June 27, 2005 Published online January 16, 2006 netic effects on performance traits have been documented, they mostly appear to be negative and tied to the genetic introgression of nonnative fish, hatchery-reared fish, or both with the wild populations (e.g., Chilcote et al. 1986; Reisenbichler and Rubin 1999; Kostow et al. 2003). Therefore, the precautionary principle (e.g., Heywood 1995) implies care with respect to the management strategy of stocking fish into wild populations. Conservation-based fish hatchery programs attempt to ensure that the broodstock comes from the local (wild) populations and are collected annually in certain numbers to guard against genetic drift (e.g., Adkison 1995; Caughley and Gunn 1996). A major concern is to conserve the population structure and diversity of wild fish populations, but there is also a demand for hatchery production with its potential for altered directional