First page Back Continue Last page Overview Graphics
Re-visiting Araki et al. 2007: The Hood River Study
Notes:
For those who are familiar with the Araki et al. 2007 paper in Science, arguably the most significant finding of that study was that Hood River steelhead that were spawned and reared in the hatchery that had two hatchery parents had lifetime relative fitness that was only 60% that of hatchery reared fish that had one wild parent and one hatchery parent. Because all fish were reared in a common hatchery environment, the difference in relative fitness presumably has genetic basis.
This is figure 2b from Araki et al. 2007. I want to walk you through the studies included in this graph and describe them and add some additional data points that also meet the criteria for inclusion in this figure. Reg Reisenbichler had taken this approach in a paper a few years back. I like the approach. Even though there are many variables that differ among the various studies, it still may be a good way to get a sense of whether the weight of evidence suggests increasing fitness loss with the duration of a hatchery program.
Relative fitness is on the Y axis and generations that the hatchery population was in captivity is shown on the x-axis.
The first two data points are from the Hood River, and show hatchery fish compared to wild fish. So, as represented on this graph they indicate the combine effects of hatchery rearing and possible genetic effects. The fitness loss attributed to genetic effects in the comparison where all fish were reared in the hatchery was about 40%.
Fleming et al. (2000) compared a farmed atlantic salmon stock comprised of several different populations to a different wild population. The hatchery fish were reared to adulthood in net pens. Relative adult-to-parr fitness was 19% and relative lifetime fitness was 16% thus nearly all of the reduced fitness occurred during breeding and early life history. This is consistent with others studies finding poor breeding success in salmon that are reared in captivity throughout their life cycle, and leaves the possibility of a strong environmental component to the fitness reduction.
Kalama R. The next two data points are from the Kalama River studies (Leider et al. 1990 on summer-runs and yet unpublished work by Hulett et al for winter run steelhead).
The winter run stock was a mix of Elochoman, Chambers Creek, Cowlitz stock propagated at Beaver Creek on the Elochoman River and
The summer run stock was the Skamania stock from the Washougal River. Both of these studies show substantially reduced lifetime fitness, and are confounded by this problem of stock origin and the possibility that a lack of local adaptation may have contributed to their poor performance.
Hansen et al. (2002) studied two brown trout populations that received more than 10 years of intensive stocking of brown trout from the same non-local, domesticated hatchery population. They conducted admixture analyses and the results indicated an estimated 6% contribution from the hatchery population in the Karup River, suggesting poor reproductive success and that’s the data point reflected in this graph. The other population, the Skjern River, had a much higher level of introgression (about 64%), and the authors note that some wild fish were brought into the hatchery for broodstock, so the introgression may have occurred there. In that case the high level of introgression reflects high reproductive success of H x W hybrids.
Ford et al. (2006) also studies lifetime RS of coho salmon in Minter Creek, and found no significant differences in relative fitness. However, in this study the naturally spawning population was composed of a large percentage of hatchery origin spawners, so there was presumably high levels of gene flow from the hatchery population into the natural population.
The final two points that could be added are from Araki et al. 2006, which was also conducted in the Hood River. The low point represents lifetime fitness of winter steelhead from the Big Creek population, and the higher point represents lifetime fitness from the Skamania summer-run steelhead. These two This situation has a lot in common with the Kalama River studies in that both included non-local hatchery broodstocks that had been stocked into these river systems for many years prior to the initiation of these studies.