Carcassman, I understand why you are making this statement about steelhead losing productivity very rapidly based on results primarily from the Hood River studies by Araki and Blouin, and Cooper. Based on their life history, I suspect that they might be more susceptible to hatchery domestication in comparison to other species. I would however suggest taking another very close those papers, before saying that there is overwhelming evidence of loss of fitness in a single generation. The two papers that I am suggesting that you take a closer look at would be the 2007 Science article, “Genetic Effects of Captive Breeding Cause a Rapid Cumulative Fitness Decline in the Wild” (along with the supporting on-line material for that article) and the 10 June 2009, Biology Letters in the Royal Society Journal, “Carry-over effect of captive breeding reduces reproductive fitness of wild-born descendants in the wild.” In particular you should look at the statistical tests and the results of those statistics which suggest a different conclusion to me. Both of these are cited regularly to come to the same conclusion you put forward in your post. However, their results (as opposed to the conclusion reached particularly in their summaries) indicate something different to me.

Specifically in the 2007 Science article, if you look at Table 2B on page 102 or the journal, you will see that the point estimate of relative reproductive success (RRS) for 1st generation captive reared fish for all study years combined (1995 – 2000) is 0.85 (relative to wild fish). This is the result cited by some to show 1st generation loss, however, if you look at the confidence intervals around that point estimate you will see that they included 1.0, the value for wild fish. Based on what I had to learn to even get out of undergraduate quantitative sciences, the conclusion in this case should be that the reproductive success of 1st generation hatchery fish was indistinguishable from that of wild fish, the same conclusion that they reached in their first two papers published in Conservation Biology in 2007, albeit with fewer return years. The fact that the report does not make that clear, even though (as stated in their supporting on-line material) the authors chose the test “with the maximal power to detect lower fitness of captive-bred fish,” is a bit of a concern for me. Let me put it simply: The authors chose a test most likely to identify a statistical difference. They found no statistical difference in the first generation, but went on in the paper, and particularly the summary (abstract) as if they had. In my experience, I would certainly call this at least bad writing. If the cynical me came out a bit, I might even call it bad science. That is too bad, because there is at least one very clear conclusion from this paper. That is that there is a very significant loss of fitness with the 2nd generation of captive rearing. In this case, the point estimate of RRS was 0.31 (P<0.001), significantly different than wild fish. This result alone should be a clear warning as to how to select broodstock when developing integrated hatchery programs for steelhead.

In the second study (the 2009 Biology letters), the authors looked at the RRS of naturally spawning fish that had one captive-bred parent and two captive-bred parents. The point estimates for RRS were 87.1% (1 captive-bred parent) and 37% (2 captive-bred parents). In this case the result for one captive-bred parent was non-significant (P = .414), but the results for two captive-bred parents was significant (P = .006). Using these results the authors, in their abstract, concluded a loss of fitness for both groups and “a significant carry-over effect of captive breeding which has negative influence on the size of the wild population in the generation after supplementation” even though the authors themselves note on page 2 of their paper that the difference for fish with one captive-bred parent was not significant. Call me crazy if you will, but with these results, I would have concluded that the RRS of fish with one captive-bred parent was indistinguishable from fish with both wild parents, but that there was a clear reduction of RRS when fish with two hatchery-bred parents spawned naturally. That risk would need to be looked at more carefully when contemplating a hatchery program.

The results that I have summarized here are taken straight from the papers of these authors, and I have told you what I have concluded. Because of this, I would not be willing to state conclusively what you stated above. What I will say, is that based on what I see of the results is that in these studies there was a clear loss of fitness when naturally spawning fish had been in the hatchery for two generations and also that the RRS of offspring of fish that had two parents go through a hatchery was significantly less than wild fish. I can’t come to the other conclusions of the authors, and therefore, cannot conclude the dire consequences suggested by them.

I’ll leave it up to you as far as what you want to conclude, but again, I suggest looking at all the papers by these authors a bit more carefully. Don’t get me wrong. The results from these studies have to be considered and need to be understood better. People contemplating starting integrated hatchery programs, particularly for steelhead, absolutely need to be aware of the significant outcomes of these studies. However, I think it is also really important to see researchers attempting to tease out the factors that could be causing these results rather than making large scale conclusions based on them. I am pleased to see this happening.