Not my intent to overstate anything about this study. If I have done that, I apologize. Maybe you are reading more into my post(s) than what's really there.

Many of the so-called "segregated" winter steelhead programs out there are using early-timed Chambers stock as their primary strategy to keep hatchery fish from co-mingling with later timed wild fish. While that strategy seems completely logical, it has gone largely untested in real-life. That's what the investigators set out to study.... and this is what they found. Run-timing alone is not sufficient to prevent significant co-mingling of hatchery and wild populations. The study concludes nothing more and nothing less.

In practical terms, what this shows us is that a number these programs are probably falling well short of the the HSRG benchmark of < 5% hatchery gene flow back to the wild for a properly-run segregated program.

Sg, it's a little too late for this approach. There is no longer a steelhead program at Chambers Creek. The last releases from there were around 1996 or 1997. The program basically had no returns to the Chambers Creek rack since several years earlier (1991 or so) and had to be supported by transferring eggs back from some the its derivative populations. So even in the hatchery environment, this population wasn't productive enough to sustain itself especially after the down-turn in survival in the late 80's - early 90's.

Confirmation bias is all too common in science, which is where people accept information that supports their beliefs, and reject information that conflicts with their beliefs. Those who are affiliated with the decisons about fish management or fish consumption are particularly vulnerable.

This publication provides some unique information that might help people figure out how to recover steelhead. Anyone who knows Tom Quinn knows he is all about producing good science and using it to improve fish management. His students are held to that same standard. I have worked on this study, indirectly. The information contained in the publication should be considered in the broader body of science, and not dismissed merely because it conflicts with beliefs or agendas. Anyone who outright rejects the potential for hatchery management to be detrimental to wild populations should seriously consider whether they have a truly objective view on this issue. Please read the paper before rejecting the findings.

OncyT,

Oops, either didn't know or didn't remember that. Thanks. That's some food for thought right there however.

Sg

Seems like the same scenario is playing out in other facilities where the fabled Chambers stock is propagated.

The programs simply fail to produce enough returning adults to reliably replace the broodstock needed for the next spawn.

Sad....

Doc -
I'm not questioning the Forks Creek data or the potential use of it however the title of this thread -

"Hatchery vs wild ... segregation does NOT work" seems like a pretty strong statement and seems leaves little room to look at individual caes.

It is something like say since Puget Sound steelhead numbers continue to be depressed inspite of wide spread application of Wild Steelhead Release regulations WSR does not work. Clearly this issue are much more complex than thant we need to look them with clear eyes with the best science in mind.

Tight lines
Curt

OncyT-

A big issue at the Chambers Creek hatchery was the very high temperatures that were used historically to produce 1 year smolts was also causing high mortalies in the developing eggs (at times more than 50%). With improving fish culture methods and foods it became possible to rear those hatchery fry to acceptable smolt size in a single growing season at most WDFW hatcheries. It only made sense to de-centralize the culture of those steelhead (and of course you point of the terrible survival of south sound steelhead had to be a factor as well).

As an aside the Chambers Creek hatchery fish did not originate from Chambers Creek. They were in fact a composite stock of brood fish collected from 7 or 8 river basins. It was common to take advantage the warm water temperatures (mid-50s) found at Chambers Creek to jump start the developement of the eggs and accerlate the growth of the fry to reach acceptable smolt size. It was typical the as the fry grow to a decent size to transfer them from Chambers Creek to other rear facilities. Since essentially all the hatchery steelhead went through the Chambers Creek hatchery it was natural to refer to them as Chambers Creek steelhead.

By the way NMFS paper on the historic Puget Sound steelhead populations does not list Chamber Creek as a population.

Tight lines
Curt

Another apology...

Limited space on PP for a title to fit without getting truncated. As regards the thread title... wanted to get the gist of the study in there and still make enough of an impact to make folks look.

Obviously it worked.

I agree Curt. The failure to look at individual cases really is the problem here. The department made some assumptions about segregation of these hatchery populations. Then they went further and made the assumption that those assumptions applied in basically every stream in the western WA and planted them with Chambers Creek stock. The more we are looking at the real condition of those populations, we are finding many cases where the assumptions were wrong. I won't take the time to total up all the examples of introgression that folks said couldn't happen, but I'm beginning to think that once we look around enough, we will find more wrong cases rather than right.

Curt, where does this viewpoint stem from? Simply being an early returning fish? Because that is the the only criteria used, there was never any selection process in place for age / size classes.

Bob -
I'm well aware that the Snider Program was to enhance the early returning portion of the wild population. Given the fishing pressure on the early part of the run and the information known at the time 25 years ago that approach to enhancing the run seemed like a reasonable way to go.


However in the last decade or so folks have come to the realization that if one is concern about the wild steelhead resource that only hatchery appoaches that make sense are wither a well segregated program or a well interated one.

As the Forks study points out a successful segregated program is not always easily obtained. The best would be one that limits the interaction between the hatchery and wild fish to acceptable low levels through some combination of temporal and spacial separation of the two population. In effect being able to treat the hatchery and wild populations as two discrete populations with limited interactions.

On the other hand a well interated program requires that the hatchery and wild populations are a single population. A major underlying assumption is that in order to achieve that the brood stock used as in the hatchery program is representative of the wild population. The means avoid selection of genetically controls traits for the hatchery population that would different from the spawning wild population as a whole. While the brood stock Snider Creek program may not have been selective for size, age, etc it clearly was selective for both an early run timing as well as early spawn timing. In short the objectives of the program was not adapted to reflect the recent developments in the understanding of the wild fish needs.

Programs like one on Snider Creek further compound the problem by increasing any adeverse impacts by amplifying the contribution of the hatchery fish to the wild population and in doing so incease any non-representive selection in the brood stock. My understanding is that the Snider Creek brood fish typically represented appproximately 1% of the potential wild spawners in the Sol Duc. Further my back of the envelope calculation would seem to show that 1% of the population that ended up in the brood program would produce 1/3 to 1/2 of the smolts being produced in the Sol Duc. That makes it a near certainity that the selection the fry from the brood fish experienced in their time in the hatchery (which is much different than the selection they would have experienced in the wild) would be infused into the wild population.

I think one thing that is extremely important for folks to realize is the responsive fish management will always be an iterative process. That is today's management will be build on yesterday's and as the knowledge base increases changes will be needed. Of course there is always a need to make immediate decisions with the less than perfect knowldege available at the time (it should understood that with the passage of time those decisions will be found to be flawed. Successful management is a classic example of "adaptive management" as management changes with increasing the knowledge and understanding of the biological systems. The unwillingness to adapt to the changing knowledge is the one "mistake" that should not be forgive.

Tight lines
Curt

The results of the Skagit and Forks Creek studies do make sense to me. Cold blooded animals have temperature specific enzymes that control/assist metabolism. The enzymes are fairly temperature specific and are inherited. That is you will find populations of the same species (take cutthroat, for example) that flourish in alpine cold waters and desert hot waters. Just not the same fish.

The Skagit fish were a product of the time when hatchery steelhead were incubated and reared in warm water; they were selected for the genes that functioned in warm water. it is no wonder that they did poorly in the wild, where they incubated and reared in cold water.

The Forks Creek, and many of the currebnt programs, are incubated and reared on surface waters. the selection is for a fish whose enzymes work in that temperature. Consequently, it should not be surprising that more of them are successful in the wild.

If one wishes to segregate a hatchery stock it is not only necessary to have temporal and spatial separation but also environmental. The less well a fish fits into an stream, the less likely it will be to successfully spawn. All you want them to do is come back and bit a hook.

On a broad-brush view, hatchery salmon are generally more successful in the wild than hatchery steelhead. Hatchery salmon, by and large, are incubated and reared on surface waters and are, therefore, selecting for a temperature regime at least close to the wild. If we switch steelhead programs to surface waters we will find hatchery fish being more successful at spawning in the wild.

CM -
I'm always surprised that folks have not made a bigger case for increased wild fish escapements on those rivers with hatchery programs. For a segregated hatchery program one of the surest ways to increase the natural selection pressures operating against those fish is by having lots of wild fish competing with them.

If one wants hatchery fish in a system to support fishing one of the "cost" should be managing for increased escapement objectives - one closer to carrying capacity rather than MSY.

In regards to the Forks Creek and Skagit steelhead I still believe that the differences in spawning time of the respective wild populations and the very different hydrographs in the two basins are significant factors.

A huge difference between hatchery salmon and steelhead is the more extended and modified rearing changes required to produce a steelhead smolt over a salmon smolt. That by its very nature subjects steelhead to more hatchery selective pressures rending them less fit to survive in the very different wild environment.

Tight lines
Curt

What Smalma said: "In regards to the Forks Creek and Skagit steelhead I still believe that the differences in spawning time of the respective wild populations and the very different hydrographs in the two basins are significant factors."

And I still need to finish reading the paper to see if wild steelhead were sampled in other tributaries and the mainstem. It seems like the sample design would display the worst possible case of introgression. I'm not saying that the hatchery fish don't have negative affects because I've written before that they do. But the context of sampling and reporting matters a lot when it comes to interpreting the results. I think the reporting of conclusions in the absence of sufficient context is making this genetics issue more confusing to the layman than necessary. It's bad enough that many of the professional fishheads have joined one side of the argument or the other, not realizing they need to better understand the issue within a contextual framework that causes it to make sense.

Sg

As always the answer to a questions depends on the context, how, and reasons for the question.

Research should not play to the layman's mentality.

We basically have several case studies here that all can be interpreted to support a particular view. Nothing wrong with case studies, but you are all struggling with the inference space a case study provides, which is none! A case study is one data point, interesting and relevant to that case, but not necessarily applicable to anywhere else.

I agree, milt...those that support your personal opinion are often given the weight of significant studies, and those that don't, of course, are "politically driven by _______ agenda!" (fill in the blank with anything you don't like).

Fish on...

Todd

Name 1

None of the ones I've been fishing for about the past 30 years are as good as they were even ten years ago...but 20 years ago was pretty much lights out compared to both before and after that, during my career at least.

Fish on...

Todd

Money post of the entire thread.

That's the true take home lesson to be drawn from the paper!