Confessions of a Hatchery Heretic

A Dead Fish is a Dead Fish, Right?
October 3, 2018
SSJ Editor and Publisher Pat Hoglund Inducted into Oregon Coastal Conservation Association Hall of Fame
December 11, 2018
A Dead Fish is a Dead Fish, Right?
October 3, 2018
SSJ Editor and Publisher Pat Hoglund Inducted into Oregon Coastal Conservation Association Hall of Fame
December 11, 2018
Show all
Confessions of a Hatchery Heretic

Once upon a time hatcheries weren’t really run for the benefit of fish. That’s changing at light speed.

Confessions of a Hatchery Heretic
By Patrick McGann, SSJ Editor at Large

HATCHERIES: 3rd of a 4-Part Series

What if hatcheries could produce fish that would improve the survivability of wild runs if they did happen to reproduce? With return rates similar to wild fish? Wouldn’t you want to crank up hatchery production to what it was 30, 40, 50 years ago? Is that even possible? Yes, it is. But maybe you would be too terrified. Getting burned at the stake is not much fun.

Creativity is how all problems are solved. If creativity isn’t required, it really isn’t a problem. That’s why looking back at our wake is not how we’re going to return salmon and steelhead runs to abundance. That’s why being afraid to look the truth right in the eye isn’t either. We have to — forgive me — think outside the box.

And to justify that rusty ol’ cliché, let’s make sure we know what “the box” is. It is the “Four H” concept itself, which is a kind of brutally transparent way of framing, or corralling, an issue, no? That is, to make sure only those four things are considered the big problems, and that none of them can be seen as a solution.

There are so many huge problems beyond the Four H’s: special interests pulling management in their favor, partisan politics holding fish hostage on issues that have nothing to do with salmon and steelhead, allocation itself is a problem easily on par with any of the H’s, and of course there is Alaska.

The main purpose of hatcheries is to support fisheries, but what if people who are against hatcheries, don’t see fisheries as necessary?

Are hatcheries a problem? There is evidence they have been, or at least that hatchery fish always seem to be somewhere nearby when there is a problem, the old correlation/causation thing. There is also evidence they aren’t a problem, even when the hatcheries are not being run very well. Of course, you can’t prove a negative, but there is a dog hair thicket of studies out there. Pick your result.
Longtime and legendary Oregon guide and salmon and steelhead advocate Jack Smith laughed and told me, “On sheer bulk there are more studies showing no harm than harm.” But we both marveled at what a hulking bulk it is too.

We are told: Hatchery fish are too weak to survive but strong enough to outcompete on the spawning grounds. They pick up their bad habits from hatchery pond conditions and those bad habits aren’t erased in their offspring in wild conditions. We can select for undesirable traits but not for desirable ones. We should select for weakness rather than fitness to prevent wild/hatchery interaction in order to prevent weakness. Domestication is an either/or and can occur in one or two generations but it won’t disappear in one or two generations providing domestication didn’t doom the poor smolt in the first place, which it will certainly do, except when it returns triumphantly to spawn. Etc. Etc. Etc. There is very little adding up in all of this.

All traits and behaviors of hatchery fish are the product of selection, intentional or not, and selection has been employed since the first hatcheries that have been in operation in the Northwest more than 100 years ago. If hatchery fish are unfit, it is because that’s what somebody wanted.


We can boil down the fourth H problem simply. If you think hatcheries aren’t working, then you probably have a different notion of ‘working’ than the people who control hatchery policy.

Hatcheries exist, and always have existed, to support fisheries. Tricky word, fisheries. If you read 100 newspaper, magazine and scholarly articles on anadromous fish hatcheries, like I have in the last few months, you can see very clearly that it is difficult for a lot of reporters to discern between a fishery and a fish run. One involves fish and humans. The other involves only fish.

So any discussion of hatcheries must first deal with the question: Are fisheries necessary?

Technically, they are not. There doesn’t need to be tribal harvest. There doesn’t need to be even tribal ceremonial harvest. The world wouldn’t end if there were no more non-tribal nets in southeast Alaska or trollers working the waters off Oregon, Washington, northern California, B.C. and Alaska. The Earth would continue to spin if we told the people of the Northwest, or all of America, really, that they could no longer go out and catch their own salmon or steelhead, make their own communion with a wildness that defines the Pacific Northwest. (Did I just write that? Heretic!)

‘Technically,’ that is. The reality is, for the vast majority of people in the Northwest, whether they fish or not, agitprop of the wild fish advocates notwithstanding, that prospect is unthinkable. Economically, politically and culturally, letting these fisheries die is impossible. The bellowing from Tom Douglas alone would be intolerable. The people will not stand for it. For that reason, hatcheries are inevitable, but that’s not why they are vital.

“I’d prefer all wild runs were strong enough to support fisheries, as they used to be, but they can’t anymore,” says Smith, who is president of the Coastal Conservation Association in Oregon.

And they never will be strong enough. I think that’s a universal belief among the salmon and steelhead harvest community. The massive loss of habitat, harm caused by climate change, our inability to remove obstacles to passage, the sheer pressure exerted by extremely profitable commercial harvesters — WILD CAUGHT! — all join to prevent a reliance solely on natural propagation. We cup our hands to our mouths and urge the world to stop making so much money for a minute and let the wild salmon and steelhead thrive but the economic engine drowns us out. We will keep trying, but we shouldn’t be idiots, either. What is will be may not necessarily be what we wish.

If we did away with hatcheries, as wild-fish lawsuits have accomplished too many times, those wild runs the hatcheries were removed to protect would eventually disappear. Just look, we’ve been at this for decades now. The removal of hatcheries and hatchery fish does not restore wild runs, not to abundance, not beyond the remnant status that allowed a lawyer to get her ruling, not in any way. At some point now, the outcomes of the wild fish court victories must be accounted for. It is failure, starkly so, and almost everywhere. The elimination of hatchery fish doesn’t stem the loss of wild fish. It just doesn’t.

On the contrary, if you eliminate hatcheries, you’re eliminating wild fish. That’s the reality on the ground and it counts a gajillion for every wild-fish-only fever dream.


The mission statements of Oregon, Washington, Alaska, Idaho and British Columbia hatchery programs all say basically the same thing, that they exist to support sustainable fisheries consistent with restoration of natural fish runs and protection of listed fish.

What’s that mean? Hell if I know. But in practice, it means they can do what they’ve always done and defend the results in court. They will have to anyway.
Hatcheries, like all bureaucracies, are run primarily by pain avoidance. The sources of pain are harvesters, ideological government cost cutters and wild fish advocates,

Take chum.

Have you noticed that, in the north at least, there sure are a lot of chum swimming around these days? They’re everywhere. You can’t get away from them. Do you know why that is? Their habitat has gotten much better? Nope. Their forage supply has gotten much better? Nope. Passage has improved in their spawning streams? Nope. They’re adapting better to warmer water temperatures? Nope. They have suddenly found a way to more successfully spawn naturally? Nope. Harvest has declined on them? Nope, just the opposite. The reason we are awash in chum salmon is because you cannot target them efficiently with troll gear, or a downrigger. It is virtually impossible to concentrate on them in the open ocean. They are relatively safe from the drift nets, reef nets, set nets and purse seines of southeast Alaska and BC.

The way you catch chum salmon is with gill nets (and seines) right off the mouths of their spawning streams.


So what’s going on is that, while we are decreasing hatchery production of other species, we are growing a lot more chum salmon in Washington state hatcheries than we used to. A LOT more. You already know why. Boldt. — Ack! Firecracker in church! — And it’s dangerously complicated. Lots of log rolling going on there. Chum hatcheries and hatchery chum are a big bargaining chip in co-management in Puget Sound and Hood Canal, just like springers, sockeye, steelhead, fall Chinook and coho hatcheries are big bargaining chips on the Columbia.

Selection begins at the earliest stages of life. Everything from water temperature to tank/pond structure to where fry find their food is important.

I get that. But yeah, chum. And the hatcheries that produce them are flourishing, multiplying even. Right there is a perfect example of the modern mission of anadromous fish hatcheries: allocation. No, it doesn’t appear in the careful concise working of the mission statements, but it is there.

There are abundant chum in Puget Sound because an 800-pound gorilla wants them there. There is no such gorilla in Oregon. In fact, Oregon’s gorilla doesn’t want chum hatcheries, or any hatcheries for that matter, so the only chum runs Oregon has are damsels in distress in dire need of protection. This is the desired outcome. Desired? Yes. Someone benefits by saving damsels, or more accurately, trying to save damsels that never get saved. One year an accident. Thirty years, a policy.

And it has always been so.

Hatcheries create and maintain passion for salmon and steelhead. Without hatcheries that constituency dies and with it, the front line in the war for restoration.

The advent of the coded wire tag, invented in 1963 allowed unprecedented precision in fish tracking. Through ocean sampling, biologists could now get a fairly accurate picture of what happens after fish have out-migrated from stream to ocean. Shortly after that came the discovery of a major difference between Toutle River coho and Cowlitz River coho. The latter popped out into the ocean and turned right and the former turned left.

So, if you are releasing coho smolts into Puget Sound, Hood Canal and northern Washington coast streams, which way do you want them to turn? Left, of course. South. That way Washington trollers have a better shot at them and there’s going to be less interception from BC and southeast Alaska. If you’re releasing coho into southern Washington coastal streams and the Columbia tributaries, which way do you want them to turn. Right, north. That way they won’t be subjected to Oregon trollers and coastal sport anglers. So that’s how Cowlitz and Toutle coho genes got jumbled just about everywhere in Washington state. Was that beneficial to those runs? I don’t think that question was ever asked.


For a fisheries manager whose objective was to produce the biggest, strongest coho ever, better able to survive and thrive in the wild than even native spawners, and one that would not only not theoretically harm wild spawning fish, but would help strengthen their specific run, would you really be too concerned about which way the fish were going to turn when they reached the ocean? No.
You’d be focusing first not just on genetic diversity but on genetic suitability and then on the hormone thyroxine and the fastest most efficient way to determine when it was at its peak levels in coho fry/smolts because the higher the thyroxine level when that fish leaves the hatchery (or the stream) the better that fish is going to survive the ocean and come back and spawn almost at comparable rates to natural spawners.

In the wild, there is diversity which is a strength but also a weakness. The fittest survive. The less fit don’t. In hatchery production, we call that waste.
If domestic turkey breeders were interested in a bird that could compete in turkey fights with wild turkeys, badasses they are, you wouldn’t have a bird so obese of breast that it couldn’t stand up let alone mix it up with some gnarly tom. Instead we have Butterballs. The fish equivalent would be Atlantic salmon bred for net pens. Could we breed lean and mean turkeys that could hold their own with wild birds? Yes, of course.

Fish breeding is different from turkey breeding only in that it is faster and therefore easier to achieve the selective qualities you’re after. So, if you are after traits that are harmful, that’s what you’re going to get. And consider the fact that literally billions are being spent on anadromous fish genetics, adaptations, and breeding. Our capabilities are improving fast.

Right now, wild propagation is producing smaller and smaller fish. There are lots of reasons and lots of theories. But it’s happening and it has nothing to do with hatcheries. It is natural selection in so far as gillnets, set nets and drift nets are natural. What is Oregon, California, British Columbia and Alaska doing about smaller and smaller wild salmon and steelhead? Nothing. What can they do? If you don’t intervene in the selection process, you’re going to get what you get. The only state hatchery system that is selecting for size is Washington (Chinook) and that effort is new and tentative. The hatcheries will use any female but select sperm from larger males.

Right now, hatcheries are selecting for traits and behaviors based on cost efficiency and reduction of perceived harm to wild fish. We know those priorities are producing a less fit fish. So, we are selecting also for traits and behaviors that segregate those less fit hatchery fish from wild fish and minimize the chance of interbreeding. That in turn further degrades fitness. And that in turn produces traits and behaviors that are useful only in a courtroom where somebody is trying to eliminate hatcheries for the greater glory of their wild fish religion. This spiral is not going to end well for anyone.

If hatcheries want to produce wild-like traits, they could. The capabilities for genetic selection are so far ahead of where they were just ten years ago, it’s crazy that we are selecting weak traits and then huddling in fear that the fish we’re producing might spawn. It’s not so crazy to select for strong traits and get some sleep at night about who’s spawning with whom.

Is that horrifying? (Heretics find that particular question quite funny.)

We can make a better fish, maybe a better fish than most fish in the wild. Booga booga! The work of Satan!

It’s true that laboratories supporting fish farms, particularly in Europe, are doing some pretty scary things genetically, like the so-called “super fish” that introduces Chinook DNA into an Atlantic salmon to make it grow faster. It’s scary because . . . why? Because they are selecting for traits that we don’t want in wild fish or for that matter in the wild, that’s why. And the prospect of interbreeding could very well introduce traits that are harmful to wild fish and we don’t want that. And wild fish are what we want because they’re . . . what? Wild? What’s wild got to do with it?

Wrong question. The right question is what is it that we want in a fish? Is the question itself heresy?

We prefer wild fish because we believe they’re cheaper, bigger, stronger, better fighters, more challenging to catch and more wholesome to eat. We believe that. Like Crusaders we believe that. Like snake handlers we believe that. Like explosive-vest martyrs we believe that.

The main advantage of hatchery fish and sport harvest is fin marking. Anglers can release wild fish with some, but a minimum of harm. Most commercial net harvest can’t do that.

But hatchery fish are anything the hatchery programs and their funders want them to be. We know how to minimize maladaptation in pre-release smolts. We know how to time releases to maximize fitness based on hormone levels in the smoltification process. We know how to select for size and how to minimize inbreeding effects (which can be just as much a threat in a remnant wild run as in an unreformed hatchery). We know where to release smolts to modify the nature of their return (faster, slower and how high in a river system).

Remember the Yechout video back in 2000 about the state of Oregon clubbing hatchery coho and spilling their eggs at the Fall Creek Hatchery to protect listed coastal coho? The property rights people went whackadoodle over it and sued trying to prove that by killing those hatchery coho the state was killing listed fish. And they succeeded only in getting a ruling saying that wild fish and hatchery fish are not the same thing. And in Y2K that was true. It is true today. It will be false tomorrow.

A hatchery fish has genetic differences from a wild fish and is therefore not a wild fish the courts have said and is therefore not protected under the Endangered Species Act. They have not ruled on the genetic difference between wild fish.

Or the benefits we assign to diversity, mostly on faith.

Genetic diversity is a matter of degree. It is not an absolute. It is also a dynamic state. In merely a decade a wandering wild fish’s genes could be randomized to invisibility in a particular run. We think we know that the utter lack of diversity of the product of an old school factory hatchery makes the fish vulnerable to disease, predators, deafness and what not, but when the diversity is increased by degrees by improved hatchery practices, we know much less about it, especially when that DNA is scrambled throughout a population in not many years. So we believe. But we don’t know. How many angels can fit on the head of a pin?

On the other hand, we do know that if we release fish too soon or hold fish too long and keep too many of them in a concrete swimming pool and feed them floating fish food like goldfish in a bowl, that’s going to have a dramatic and negative impact on the return rates of those fish. But that’s not genetics. It’s a big deal, cost-wise but there’s little harm to wild fish because of that. That’s where hatchery fish get their wrap, from maladaptation. But if we improve acclimation practices, at a cost, we can dramatically improve the efficiency. But even if such a maladapted fish does spawn successfully, those negative traits will be swept away by the rearing conditions of that fish’s offspring.

So, if we can produce a fish in a hatchery that is as much or almost as much of what we want in a fish as a naturally spawned fish, who would find that pee-puddling terrible?

Religion is faith and belief, based not so much on evidence, but on dogma. (Everybody has a religion of some kind, even atheists.) In the presence of current advances in fish genetics and hatchery practices, the evidence for the superiority of naturally spawned salmon and steelhead is shrinking. But the faith persists. And questioning wild fish religion, both in and out of government, is received just as happily as Pelagius’ questioning of the need for the Church in achieving grace or Luther’s insistence that bribery isn’t very holy.


The better hatchery fish is not theoretical. It is not experimental. It is in virtual production. Not in state or federal hatcheries, where hatchery policy makers crouch in terror of wild fish lawsuits and budget committees, but in tribal hatcheries on the Columbia tributaries, especially the Clearwater and Snake. Tribal hatcheries are relatively immune from the pressures that paralyze state and federal hatcheries. And they’re not waiting around.

The Nez Perce on the Snake and Clearwater rivers have been exceptionally aggressive in hatchery reform. And why wouldn’t they be? Their culture is salmon-based and their fish have been drying up. They can hear the yacking about the dams and spill, the water temperature issues, harvest, predation . . . they’ve heard it for decades now. And they sit around and wait for somebody, anybody to do something? Or do they take matters into their own hands? One guess.

Same thing with the Warm Springs tribes and the Yakama. They have their own hatcheries. They have tribal biologists. They have brought in all kinds of experts and commissioned studies on top of studies, adding to the hulking bulk of data noise. They know how to build a better hatchery fish, so what’s stopping them?

The Nez Perce multi-year study of culturing, acclimation and release of Chinook smolts to match wild conditions, optimized through better data collection and management technologies, has been talked about for almost a decade. It’s like old hat, like, “Oh, yeah, I’ve heard about that.” But it’s been showing blistering results in return rates, not quite as good as wild fish, but close.

The point is, they’re getting ready now to move those techniques out of the surrogate test group into full production. And when that happens, their hatchery return rates are going to explode. And when that happens and a great many of those first-generation hatchery fish spawn, you’re going to see the lie put to the impact of a lack of genetic diversity vs. the impact of poor acclimation practices.

The Nez Perce, Yakama and Warm Springs tribes are doing this kind of hatchery reform across the board, working closely with WDFW, ODFW and IDFG and the feds, who are watching over the tribal shoulders. It’s the most encouraging thing I’ve seen in salmon and steelhead restoration . . . ever.

Meanwhile, the state and federal production hatcheries are bending themselves into pretzels to select weak traits to prevent hatchery/wild interbreeding and make a show to the legislators of saving money at any cost, what you’re seeing in the tribal propagation efforts is the opposite. They are decidedly refusing to segregate hatchery fish from natural spawning and they’re pulling the trigger on techniques that are more costly upfront but less so on rate of return. The Nez Perce measure their hatchery performance by natural spawning, meaning, their hatchery fish are being designed to interbreed with wild fish. Nobody is really parading that fact around, but that’s the case just the same.

Their wild runs are increasing — and from the brink, too — and hatcheries are why. In two generations, you wouldn’t be able to win a court case anywhere on the functional difference between the interbred fish and wild fish. There is no other conclusion at which any reasonable person could arrive. This is beyond merely proven. It is self evident. And it is going to help everyone.

Elsewhere, where hatchery fish are designed for weaker traits, hatchery production is down, natural production is down, NOAA Fisheries is on a hair trigger to list, hatchery fish zoom through the rivers and provide little opportunity for fisheries, Sisyphean efforts to restore habitat send economic tsunamis through the economy, and everyone is miserable. Of course it’s not working. It’s not working for anyone but the hatcheries, the bean counters and anti-hatchery people. It’s not designed to work.

My problem with the Four H framing is the assumption that hatcheries are a problem rather than a solution. That orientation is disingenuous. It focuses on poor hatchery practices that are rapidly changing and a razzle dazzle and hocus pocus of faulty cause and correlation. It isn’t science. It’s religion.

And to that religion I am proudly a heretic. The problem isn’t that there is too much hatchery production. The problem is that there isn’t enough.


Leave a Reply