Residualization of Hatchery Steelhead: A Meta-Analysis of Hatchery Practices

Fisheries Centre (FC), University of British Columbia - Vancouver, Vancouver, British Columbia, Canada
North American Journal of Fisheries Management (Impact Factor: 0.95). 10/2012; 32(5):905-921. DOI: 10.1080/02755947.2012.711269


Freshwater residualization, whereby anadromous juvenile salmonids fail to emigrate seawards within the primary migration period, causes considerable economic and ecological management concern. Previous studies have attempted to identify possible factors contributing to residualization, including both fish-related and release methodology-related attributes, in order to develop measures to reduce it. Here, we synthesize 48 previous estimates of the residualization rates of hatchery-reared steelhead Oncorhynchus mykiss from 16 studies and evaluate the cross-study effects of several factors that can be controlled by hatchery managers. The proportion of fish in hatchery release groups that residualized ranged from 0% to 17% (average, 5.6%). Characteristics of the release process were dominant in affecting residualization rates, while characteristics of individual steelhead primarily determined which, but not how many, individuals residualized. Releases of fewer fish and those located closer to the ocean or to a confluence with a major river produced fewer residuals than larger releases located further upstream. Acclimation ponds also appeared to reduce residualization, but there was no evidence of a release date effect across locations and years. Within a release year, individuals from endemic broodstock had higher residualization rates than those from hatchery-propagated broodstock while smaller individuals and larger males were more likely to residualize than individuals of intermediate size (similar to 213 mm fork length). To meet management objectives of reducing steelhead residualization, we recommend releases closer to an ocean or large river, particularly for releases of relatively few fish, in conjunction with the use of acclimation ponds. Management effort should focus on selective harvesting of hatchery residuals, a process which may be supported by rearing and release strategies. These objectives may trade off with conservation objectives; straying risk and genetic effects should especially be taken into account.

Download full-text


Available from: Michael Melnychuk, Oct 02, 2015
1 Follower
30 Reads
    • "Melnychuk, unpublished data), and by tailoring release times to avoid seasonal periods of predator aggregations (Mace 1983; Wood 1987). Other release strategies can be used to help reduce the frequency of residualization (Hausch and Melnychuk 2012) and to reduce potential impacts on wild populations (Naish et al. 2007). "
    [Show abstract] [Hide abstract]
    ABSTRACT: We observed large survival differences between wild and hatchery-reared steelhead trout (Oncorhynchus mykiss) during the juvenile downstream migration immediately after release, which persisted through adult life. Following a railway spill of sodium hydroxide into the Cheakamus River, British Columbia, a short-term conservation hatchery rearing program was implemented for steelhead. We used acoustic telemetry and mark-recapture models to estimate survival of wild and (or) hatchery-reared steelhead during 4 years of the smolt migration, with both groups released in 2008. After adjusting for estimated freshwater residualization, 7%-13% of wild smolts and 30%-40% of hatchery smolts died in the first 3 km of the migration. Estimated survival from release to ocean entry was 71%-84% for wild fish and 26%-40% for hatchery fish and to exit from the Strait of Georgia system was 22%-33% for wild fish and 3.5%-6.7% for hatchery fish. A calculated 2.3-fold survival difference established during the downstream migration was similar to that after the return of adult spawners, as return rates were 8.0% for wild fish and 4.1% for hatchery fish. Contrary to current understanding, a large proportion of salmon mortality in the smolt-to-adult period, commonly termed "marine mortality", may actually occur prior to ocean entry.
    Canadian Journal of Fisheries and Aquatic Sciences 06/2014; 71(6):831-846. DOI:10.1139/cjfas-2013-0165 · 2.29 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: When rainbow trout (Salmo gairdneri) fry (underyearlings) were introduced into laboratory channels at three different prey levels and permitted to emigrate voluntarily, their density remained highest at the highest prey level. The distribution of fry was positively associated with a gradient in prey abundance. Both territory size and frequency of aggressive encounter varied inversely with prey level; the higher the prey level, the smaller the territory and the lower the frequency of aggressive encounter. Emigration from the channels was neither as rapid nor as marked when prey level was reduced, compared to when fry were initially introduced to different prey levels. However, frequency of aggressive encounter significantly increased when the prey level was decreased and significantly decreased when the prey was increased.
    04/2011; 31(7):1201-1209. DOI:10.1139/f74-143
  • [Show abstract] [Hide abstract]
    ABSTRACT: Drawing on acoustic telemetry this study identifies and describes local and regional scale survival and movement patterns of Central California Coast steelhead (Oncorhynchus mykiss), including their potential utilization of newly restored tidal marsh habitats in the Napa River system. Between April 8th and May 5th of 2010, 20 steelhead smolts ranging in fork length from 164 to 305 mm were collected, tagged with acoustic transmitters, and released in the upper Napa River (above tidal influence). We found no effect of release date (P < 0.001) or size (P < 0.005) on survival estimates based on model likelihoods. Cumulative survival from smolt release location to the Golden Gate Bridge over approximately 77 river kilometers (RKM) was 0.60 (SE = 0.16). Reach-specific survival was lowest in the initial 30 km reach (0.70 SE = 0.1). Survival was higher in San Pablo Bay (0.89 SE = 0.1) and San Francisco Bay (0.96 SE = 0.2). Sixty percent of the fish that entered the ocean were detected on a line of acoustic monitors at Point Reyes approximately 60 km north of the Golden Gate. Average movement rates of smolts were highest in San Pablo Bay (36.6 km∙d-1 SE = 3.3) and San Francisco Bay (28.9 km∙d-1 SE = 6.6). Smolts migrated more slowly in the river (9.0 km∙d-1 SE = 0.9) and ocean (4.1 km∙d-1 SE = 1.2). However, smolt movement rates in the river were dependent on location. Average movement rates of smolts were greatest shortly before their exit from the Napa River (83 km∙d-1 SE = 13.2). Fish were not detected within the recently reconnected former salt production ponds (North, Central, and South units) adjacent to the Napa River. Based on the detection patterns of fish throughout the study area, it appears that most fish were moving at relatively high rates and were not exploring off-channel habitat.
    Environmental Biology of Fishes 02/2012; 96(2-3). DOI:10.1007/s10641-012-0092-4 · 1.57 Impact Factor
Show more