Nature’s loss

The Valdez Fisheries Development Association’s pink salmon hatchery operation/VFDA photo

The massive, free-range, salmon-farming operation run by commercial fishermen in Alaska’s Prince William Sound appears to be undermining the natural spawning success of pink salmon there, according to a peer-reviewed study published in the journal “Evolutionary Applications” earlier this month.

The study reported finding “some of the smallest relative reproductive-success values ever observed in Pacific salmon” in streams filled with a mix of wild fish and hatchery strays.

Other than illustrating another sign of humankind’s massive footprint on the planet, however, the significance of this change driven by the multitude of pinks straying from the region’s industrial-size hatcheries is unclear.

Wild pink salmon stocks in the Sound remain in excellent shape with the Alaska Department of Fish and Game this year reporting the wild “pink salmon run was above average in 2021, encouraging given that wild fish were from the parent year in
2019 when spawners returned to dewatered streams amid a record-setting drought.”

The Evolutionary Applications study of hatchery fish led by state fisheries biologist Kyle Shedd with help from colleagues at Fish and Game and the Prince William Sound Science Center might also have shed some light on how those returns came back strong despite the drought faced by their parents in the spawning year of 2019.

The freshwater in the Sound’s short, steep salmon stream – or at least that in the two streams that have been the subject of intensive genetics studies – does not appear to be as productive as the brackish water where the streams meet the ever ebbing and flooding tides of the sea.

“Parents sampled upstream had 59 percent as many offspring on average as pink salmon sampled in the intertidal” area, the study reported. If those results were to hold for all streams in the Sound, low water pushing more pinks to spawn intertidally might actually increase rather than reduce production although there is clearly a limit as to how much of a drought the fish can survive.

Saltwater will kill eggs in the gravel, so they need some flow of freshwater even if it is the  little flowing beneath apparently dewatered streams. Old studies in Southeast Alaska, meanwhile, showed an interesting push-pull between environmental conditions for intertidal spawning pinks.

Warmer ocean waters, especially during cold winters, improved the survival of eggs and alevins in the gravel of intertidal areas, but eggs that didn’t get enough water to reduce the salinity of the ocean died. If the parent fish spawned too far outside the zone of brackish water, their young were doomed.

The survival problem scientists documented in the Sound by genetically fingerprinting adult fish and then looking for their fingerprints in the genes of returning young are not, however, due to environmental problems related to habit, but to human-driven problems, albeit accidental, related to salmon wandering into Sound streams to spawn instead of returning to the region’s hatcheries.

Farming fish

Alaska has a massive fish-farming program, but it does not confine the fish to pens as in Norway, Chile, Scotland and elsewhere.

Instead, the fish are born in hatcheries, raised there briefly, and then released to free-range the ocean. Net-pen farming was banned in Alaska three decades ago, and the state prefers to call what it does “ranching” rather than farming.

The returning fish are sold as “wild-caught,” which subtly distinguishes them from wild fish without overtly alerting consumers to their origin. Alaska, for a time, led international efforts to market the idea that farmed salmon were inherently unhealthier than those caught in the ocean but has now backed away from that idea.

The Alaska Seafood Marketing Institute no longer attacks farmed salmon but instead pushes Alaska fish for its:

  • “Superior flavor and texture…prized around the world”
  • Sustainability
  • Made in America origin
  • And the chance for consumers to help support “generations of fishing families and communities”

Much of the shift in strategy appears tied to the fact that the major fish processors that help fund the Institute are these days deeply involved in sales of net-pen farmed salmon, which dominate the global market.

Approximately 75 percent of the salmon eaten around the world today is farmed (not counting U.S., Japanese and Russian hatchery returns) and the percentage is growing. The net-pen farmers have enjoyed record profits as sales of fresh salmon have skyrocketed during the pandemic.

Alaska, unfortunately, has limited access to fresh-salmon markets because of the seasonability of its salmon returns.  As a result, many of the hatchery-returning fish still go into cans for which increasing market demand is projected, but primarily in lower-value markets.

“Owing to rise in consumers’ purchasing power, (the canned-salmon) market is likely to increase significantly in developing countries. Furthermore, the vast customer base in developing nations is likely to have a significant impact on segment growth,” according to Allied Market Research.

And where there are higher-valued markets for canned salmon – such as those for boneless, skinless filets – the farmers are increasingly getting into the market. Still, Alaska canned salmon, now sometimes also packed in pouches, overwhelms the list of the “The 8 Best Canned Salmon in 2022” posted by The Spruce Eats, a website for foodies. 

Unfortunately, the website’s suggestions on the best-canned salmon also come with the proviso “if fresh salmon isn’t available….”

Competition from the net-pen farmers for fresh fish makes the fishing business in Alaska a battle for companies trying to take advantage of the tens of millions of pink salmon that now return each year thanks to the annual ocean stocking of more than 700 million pink salmon.

The 49th state is a world leader in ocean ranching with a total release of “approximately 1.8 billion juvenile salmon annually,” according to the study. This is largely thanks to a government-funded hatchery program begun by the Fish and Game in the 1970s.

The hatchery salmon that aren’t pinks are almost all higher-value chum salmon once commonly referred to as “dog salmon” in the 49th state because of the hooked snouts and obvious, canine-like teeth that develop on males during spawning, but now most often marketed as “keta salmon.” 

Humpies gone wild

Humpies – as Alaskans usually call pink salmon, the males of which develop huge humps on their backs as they begin the reproductive stage of life – are by far the most abundant salmon in the state.

And though they might all look alike in Sound streams, according to the new study, they don’t all perform alike.

“Reproductive success, measured as sampled adult offspring that returned to their natal stream, was significantly lower for hatchery- versus natural-origin parents,” the study found, with female humpies performing far worse than males.

Hatchery females that decided to spawn in the wild were less than a half as productive as their wild sisters with the range of success falling from 0.47 all the way down to 0.03. Some males did much better, with the range running from 0.86 down to 0.05.

Overall, researchers concluded, the “results strongly suggest that hatchery-origin strays have lower fitness in the wild,” adding that it is still uncertain, however, whether reduced productivity is environmentally driven, and thus subject to quick changes as environmental conditions shift, “or genetically driven, and likely persistent across generations.”

The study looked only at humpies spawning in Hogan Bay Creek and Stockdale Creek from 2013 to 2016. Both are short streams. Hogan is only about a third of a mile long, and most spawning gravel is in the intertidal reach. Stockdale is slightly less than a mile long with more freshwater spawning habitat.

One of the notable differences found between hatchery fish gone feral and wild spawners was the former’s preference for freshwater spawning. That might have been part of the reason for the much lower spawning success of hatchery strays.

“It is unclear why hatchery-origin fish traveled farther upstream where reproductive success was lower,” the study’s authors wrote. “They may have experienced lower reproductive success because they were strays and were not locally adapted to the spawning habitat. Alternatively, they may have traveled further upstream to less suitable spawning habitat and avoided the intertidal zone because many of the hatchery brood sources came from upstream, freshwater sites and hatchery-origin fish imprint on freshwater sources as embryos and fry in the hatcheries.”

Study lead author Shedd said in an email exchange that there are still a lot of unknowns here, but he added that it is doubtful any loss in wild-salmon productivity can be overcome simply by increasing the number of spawners returning to hatchery-affected streams.

The way fisheries are managed in the Sound today, he said, the spawning grounds are maintained at carrying capacity. A variety of previous studies have found that over-crowding spawning areas decrease spawning success as pinks spawn atop the beds of other pinks and knock eggs out of the gavel.

“…Egg loss during spawning increases as the density of female spawners increases,” Robert McNeil reported in a peer-reviewed study in the Journal of the Fisheries Research Board of Canada in 2011.

Still, even if spawning success declines in the Sound due to straying hatchery fish, fishery managers should be able to maintain the same-size returns to wild streams by simply reducing harvests in wild-fish harvest areas given that pinks usually return in numbers significantly greater than are needed for spawning.

Salmon fishery management is entirely built around the idea that salmon returns usually exceed spawning needs and thus produce a “harvestable surplus” of fish in “sustainable fisheries.”

A perfect world

In the ideal scenario, both humans and salmon benefit from sustainable management in that by reducing the so-called “over-escapement” of salmon, the fish surviving to spawn maintain a high rate of productivity.

Hatcheries were designed to boost productivity even further by providing the fish a more stable environment early in their life cycle, and the hatcheries have been very successful at doing so in the Sound.

A peer-reviewed study published in the Canadian Journal of Fisheries and Aquatic Sciences in 2016 reported that since the North Pacific warmed in the 1980s the pink salmon harvest in the Sound “increased nearly tenfold between the low-productivity, prehatchery period and the high-productivity, full hatchery production period.

“Meanwhile, in all other management areas (in Alaska) catch increased approximately threefold. The disproportionate increase in catch observed in PWS compared with the other areas suggests a considerable contribution by the PWS hatchery program, even after accounting for potentially increased wild stock productivity following the shift to improved environmental conditions. ”

The unseen cost of this now appears to have been a loss in some wild watershed productivity due to genetic or epigenetic alterations arising from the domestication of pinks in hatcheries, but it is hard as yet to quantify the size of that loss.

Shedd said the productivity of wild-hatchery hybrid pinks is not yet fully defined, and there is a bigger question revolving around the hybridization that has now been underway for decades. Hybrids are not as successful on the spawning grounds as wild fish, the study indicated, but they are more successful than hatchery fish.

If what was found at Hogan Bay and Stockdale creeks holds true for all streams in the Sound, hybridization would decrease natural production overall, but that would be more than offset by the hatchery production which is orders of magnitude above the historic, natural production of the Sound.

How much the natural production might fall overall will “largely hinge on whether the mechanisms driving reduced reproductive success of hatchery fish are primarily due to genetic mechanisms or to non-genetic mechanisms,” Shedd said.

“For example, if hatchery fish are genetically poorly adapted to successfully reproducing in the wild, then they may pass these poorly adapted traits on to their offspring and it may take generations for (natural) selection to work to increase the reproductive success.

“On the other hand, if the reason why hatchery fish have lower reproductive success is because they are unfamiliar with the stream and don’t know where to spawn (homing wild fish may key in and spawn in the stream section where they emerged), then those effects could be erased in the next generation (poorly selected spawning areas will not produce many fish).

“These are just two examples of these mechanisms – there are lots more postulated in the paper. We are hoping that additional years of data will provide clues regarding the mechanisms and address this important question.”

And the answer to the biggest question is tied up in learning more about how the long-term spawning success of the hybrids.

“We…know that some of the hatchery strays are successfully contributing to the next generation,” Shedd said. “(But) we don’t yet know how their offspring are contributing, so that makes calculating precise proportions of hybrids difficult.”

Also unknown is how long it will take before the  Sound is fully hybridized. There remain large variations across the region as to the numbers of hatchery straying salmon, wild salmon and hybrids in streams.

“Both of our study streams located in Southwestern Prince William Sound (PWS) had high proportions of hatchery-origin spawners,” Shedd said. The research was specifically focused there for this reason. Access to a lot of hatchery strays made for better data.

“…Most of the rest of PWS has much lower average proportions of hatchery-origin spawners,  so the levels of hatchery introgression are likely much lower. I’d guess that most streams in southwest PWS likely have some degree of hatchery introgression, whereas the streams in eastern PWS, likely have very little hatchery introgression based on the low proportion of hatchery-origin” salmon observed in earlier studies, Shedd said.

Genetic studies are ongoing although some salmon scientists are now wondering if they don’t overlook a far bigger question involving hatcheries – that being the possible effect on Chinook, sockeye and coho salmon when huge numbers of young hatchery pinks and chums are dumped into the ocean every year.

Some scientists have suggested never-before-seen numbers of humpies in the North Pacific may be depressing the number of those big Chinook, the fish Alaskans call “king,” sockeyes; and cohos all along the North American West Coast by grazing ocean pastures down to the terrestrial equivalent of bare ground.

Think of the European rabbits introduced to Australia in the 1800s only to reach a population estimated at 10 billion by 1920, leading to pasture degradation that cost Australian farmers hundreds of millions of dollars in lost production. 

Scientists are only beginning to thoroughly examine the abundance of pinks in relation to the declines of other species of salmon to see if they can pin down a cause-and-effect relationship. So far, there are only correlations.

A group of scientists led by Canadian Brendan Connors in 2020  reported finding that “from 2005 to 2015, the approximately 82 million adult pink salmon produced annually from hatcheries were estimated to have reduced the productivity of southern sockeye salmon by 15 percent on average.”

They blamed food competition, but other scientists have argued the real culprit could be ocean-warming due to a changing climate creating a friendlier environment for pinks. They say more evidence is needed to link the declines in sockeye in Canada and the Pacific Northwest to the Alaska pink salmon boom.









11 replies »


    An unusually cool ‘blue blob’ in the North Atlantic is slowing glacier loss in Iceland

    A patch of unusually cold water in the North Atlantic Ocean, known as the “blue blob,” could forestall some of Iceland’s glacier melt in the next three decades, a new study says.

    The cold water, south of Iceland and Greenland, is creating more snowfall over Iceland, replenishing the island’s glaciers as they melt and run off.

    The cold blob is expected to linger until the mid-2050s, granting a temporary reprieve against glacial melt, according to the study published in Geophysical Research Letters.

  2. Through the use of coded wire tags, significant straying by hatchery bred pinks returning to PWS became evident during oil spill related studies in the early 90s. When faced with extrapolated tagging data that indicated up to 70% of some stream’s pink salmon escapement were hatchery bred fish, the State geneticist at the time preferred to pursue the theory that the tags themselves may have caused the inability of tagged fish to home properly. 70% straying created an issue with the State’s genetics policy. It took another ten years and thermal marking all hatchery fish to begin to confirm the magnitude of the straying in PWS

    Wild inter-tidally bred pink salmon were also shown to stray to some extent. However, the volume of hatchery strays inundated many small streams along their migrations through PWS. Yes, stream “escapement” numbers may be achieving seasonal goals. But wild stocks may not.

    To those who say a “meh, pink is a pink”: back in the 60’s some longer streams in PWS were shown to have distinct upstream and intertidal spawning populations, some with bimodal run timings unique to the drainage. Those are genetic traits, honed by nature over eons were tied to that stock’s success. Can they be successfully replaced by ocean ranched, hatchery bred pinks? We’ll have to check the policy on that one.

    • Your last question frames the fundamental issue there. The latest genetic works makes it clear that though all pinks look alike on the outside, they are decidedly different on the inside. Hatchery strays do not reproduce very well, although they can and do breed and produce young.

      If I were a seiner in the Sound, this would have me pondering how much longer the wild-fish targeted fisheries last. ADF&G could well find itself in a position where it needs to restrict those fisheries more and more to maintain returns to streams with declining productivity due to most spawners being strays or wild-stray hybrids.

      It’d have me looking for a way to make sure I was tied to some sort of PWSAC or VFDA “cost-recovery” contract to harvest hatchery fish.

      • If you read the statues and regulations, you would see that the Alaska hatchery system is designed to minimize strays, not guarantee that there are no strays. Even wild pink salmon cannot be guaranteed to stray. Honestly, the Alaska system has been successful at minimizing strays.

        Secondly, if hatchery fish are so darn injurious to wild fish, then why has the PWS wild pink salmon stocks not crashed yet? This hatchery program has been in plac no for close to 50 years, and the wild pink salmon are doing quite well.

        Predicting a catastrophe, and changing policies based upon a speculated catastrophe is not science. Well, maybe political science.

    • After the 1964 earthquake which raised most of the land from PWS to Kodiak and on out to Unimak Pass. This uplifting caused the formation often impassable waterfalls at the mouth of many salmon spawning creeks.

      As a consequence, pink salmon particularly straed simply to survive. This instinctual reaction probably saved the species. Over time the waterfalls were eroded and the pink salmon “strayed” back, repopulating rivers hardest hit by the earthquake.

      If I remember correctly from the archeological diggings at Chernabura Island, these massive earthquakes happen about every 600 years, and the pink salmon always recovered.

      In PWS it took until 1979, for the pink salmon to fully recover, a 15 year period.

      Straying is not only natural, but is nature’s way to prevent extinction.

      This straying is bad stigma, scientifically ignores selective facts like these.

      Some folks have an anti-hatchery bias they want to promote by flaunting their advanced degrees, rather than producing good unbiased science.

      The salmon hatchery system in Alaska are unlike any others in the world. The Alaska system was designed specifically to eliminate inherent flaws in hatchery systems in Japan, Russia, and the lower 48. These century old legacy hatchery systems are vastly different than the 40 year old Alaska system. This was not an accident. The details matter.

      So how has the Alaska hatchery system performed?

      Alaska’s salmon hatchery system is the only system in the world that brought their wild native spawning populations back from the brink of extinction, to a robust spawning population today.

      Not a single fisheries scientist can make this claim, except Alaska’s fisheries scientists, many of which were private contractors to organizations of commercial fishermen.

      The scientists who attack this system, should in reality be studying instead. It is the only system that has worked in human history.

      I know this story because I lived it, and watched it blossom.

  3. This makes me wonder if pinks in increasing numbers have a deleterious effect on the spawning beds/success of kings or reds. I would normally assume that the species spawn in differing stream regimes. Or is there any commonality of overlap?

    • Chris: Studies have found some overlaps of spawning habitat in some wateres, but I’ve never seen anything in the literature indicating anyone finding it enough of an issue to even suggest it might affect reproductive success anywhere.

      • Many years ago (35?), so many sockeyes came back to a stream that F&G was giving out netted fish. A friend of mine whose family of 4 ate salmon almost daily drove down the Seward Highway somewhere (Seward or Kenai area) with his long bed Isuzu pickup and son. They had 3 coolers. After a discussion with the F&G personnel as to the extent of their love of salmon, they were told they could have as many as they wanted. He lined the bed of his pickup with a tarp, and they filled it with salmon. He and his family were ecstatic and spoke of the incident for years. He has passed away and his family relocated to Vancouver Island so I have no way to get the particulars. My recollection is that there were so many sockeyes returning that they were digging up the beds of previously laid eggs.

    • Chris-
      I was fishing/hiking along a remote stream down on the Peninsula and saw Kings, Reds, Humpies and Chums in different stages of spawning. The Silvers were just arriving. It was interesting to see that they were all in slightly different habitats with varying depth, current and gravel.

  4. This is an agenda driven article, masquerading as science. The prose bounces around from subject to subject.

    Let’s just take a couple of facts this “scientific” piece fails to mention.

    1) Pink salmon have always naturally strayed, even thousands of years ago. Natural occurring events like landslides have killed all the salmon in an individual river. Nature’s replant the dead rivers with stray salmon from other rivers.

    2) Pink salmon hatcheries coexist alongside native streams and the native streams have robust runs of natural salmon. Where else in the whole world do you see this?

    I will await your examples of where major hatcheries coexist right alongside strong native salmon.

    • Douglas: This is one of those comments that just makes me want to ask, ‘What’s your point, dumbass?” But I always restrain myself because such behavior is counter-productive.

      What I would suggest is that you go back and read the story again, and take note of the second and third paragraphs:

      “The significance of this change driven by the multitude of pinks straying from the region’s industrial-size hatcheries is unclear.

      “Wild pink salmon stocks in the Sound remain in excellent shape….”

      Pink salmon hatcheries do exist along native streams full of salmon, and there is at this point no signs of this depressing overall salmon production. But if you read the rest of the story, you’ll find out you’re fooling yourself if you believe those streams near hatcheries contain “robust runs of natural salmon.”

      All pinks might look alike from the outside, but their genes say a bunch of the fish making for “robust runs” in streams near hatcheries are not wild fish, but hatchery fish. This situation can be replicated anywhere a hatchery is built next in the vicinity of a major pink salmon stream.

      Where else in the “whole world” can this situation be found now? I’d suggested Aniva Bay in the Sakhalin Islands of Russia and the Quinsam River in British Columbia.

      Canadian production of hatchery humpies, the last time I looked, was on the order of 10 million or 15 million per year total, which is chicken feed. But if you want to go looking for good numbers of pinks in natural waters near a hatchery producing millions of humpy fry, the Quinsam Hatchery on the Quinsam River, a tributary to the Campbell River, on Vancouver Island in Canada, would be a good bet.

      The study doesn’t suggest hatchery humpies lead to the depopulation of nearby humpy streams. What it suggests is that hatcheries lead to the displacement of wild fish by hatchery fish or to fishery harvests that remove both hatchery and wild fish, sometimes leaving other hatchery fish to make up a large part of the “wild” escapement.

      Natural straying wasn’t mentioned because it’s a given. All salmon stray to greater or lesser degrees. Hatchery straying is mentioned because it has been suggested by some in the past,, though why I don’t know, that hatchery location can prevent this, thus preventing the mixing of wild and domesticated fish.

      That is clearly not the case, and the story probably should have pointed out that when Alaskans were told hatchery locations would prevent straying they were being sold a bill of goods. Sound hatchery fish have shown up all over hell and gone, including in a lot of streams at the south end of Cook Inlet.

      Is this a good thing or a bad thing? Hell if I know.

      As you have observed, if some of thoe streams were blocked by landslides and then reopened, the straying fish could help repopulate the stream. This study suggests that would take a lot longer for hatchery fish to do than for wild fish to do, given the lower reproductive success of the former, but that doesn’t mean they couldn’t get the job done.

Leave a Reply