A group of international fisheries scientists is now pointing out the simple, ecological reality of mathematics long ignored by Alaska salmon managers: addition matters as well as subtraction.
With the human-manipulated ecosystem of the North Pacific Ocean once again oscillating wildly, they are warning that fishery management isn’t just about the removals of salmon via human harvests; it is also about the additions of billions of the little fish now dumped into the ocean each year by industrial-scale salmon hatcheries in Alaska, Japan and Russia.
“The intentional release can have wider-ranging consequences than previously thought, as the impacts can propagate through a diversity of ecological interactions,” they warn in a peer-reviewed paper published by Proceedings of the National Academy of Sciences of the United States of America (PNAS) last month.
“This recognition has sparked a discussion of how the massive introduction of native species alters short-term ecosystem dynamics. Yet, current debates overlook the fact that we have rarely assessed the community-wide impact in the long term.”
Evidence to support the theory that dumping billions of hatchery salmon into the Pacific can play havoc with wild ecosystems has been hotly debated for years now, but the researchers reporting in PNAS say they’ve been able to gather the evidence to document long-term, community-wide consequences by studying masu salmon in Japan.
Masu, or cherry salmon as they are sometimes called, are the least common of the six species of Pacific salmon. Their range is limited to northern Japan and the adjacent coast of Asia as far north as Russia’s Kamchatka Peninsula.
Their economic importance is unclear, according to the Wild Salmon Center, because masu harvests are lumped together with pink salmon harvests in Asia.
Economics has long played a role in the global willingness to overlook the consequences of environmental tampering in the form of the billions of young salmon that spill out of hatcheries every year, the authors of the PNAS paper note.
As the scientists from the Salmon and Freshwater Fisheries Research Institute in Hokkaido, Hokkaido University, Japan’s National Institute of Polar Research and the University of North Carolina put it, hatcheries have become “pervasive in natural resource management owing to the significant economic benefit.”
Alaska is the textbook example of economic benefit. The state got into the hatchery business in a big way in the 1970s to prop up the state’s commercial fishing industry as the annual territorial harvests with a long-term average near 100 million fell to 49 million salmon per year in the new state, largely due to cold waters in the Gulf of Alaska.
The goal then was to try to restore harvests to something above 100 million per year, given that Alaska territorial peak harvests had been near that number in the 1940s. The original goal was set at an arbitrary average of 143 million per year with about 36 percent of the catch to be hatchery fish.
The goal was long ago passed.
The state’s five-year, annual average harvest from 2014 through 2018 was over 200 million, and the annual average for the decade of the 2020s was 181 million per year.
Some of the increase came from better management of wild stocks, but much of it has been driven by hatchery production, primarily of pink salmon, the smallest, shortest-lived and least valuable of the species.
When the state witnessed a record harvest of 272 million salmon in 2013, pinks accounted for more than 80 percent of the catch. That year’s take of 219 million pinks more than doubled the annual, average all-species catch of Alaska salmon from the 1940s when all-species harvests were at their previous peak.
Approximately 92 million, or more than 42 percent of the 2013 pinks, were hatchery fish, according to the annual Salmon Fisheries Enhancement Program report from the Alaska Department of Fish and Game.
The report bragged that Prince William Sound, the big bay at the northeast corner of the Pacific just south of Anchorage, had become an industrial-scale salmon factory thanks to hatcheries.
“An estimated 75 million salmon returned from hatchery releases, accounting for an estimated 79 percent of the total number of salmon in the commercial common property harvest (CPH); 88 percent of the chum, 80 percent of the pink, 45 percent of the sockeye, and 39 percent of the coho salmon in the commercial CPH were hatchery-produced fish,” the report said. “In addition, hatchery-produced salmon contributed an estimated $113 million, or 68 percent, of the ex-vessel value of salmon in that” common property harvest.
“Common property harvest” is the state description for hatchery salmon intercepted by commercial fishermen before they get to the hatcheries. There are also “cost recovery” fisheries at the hatcheries that have turned those facilities, though technically “non-profit” businesses, into operations that provide good-paying jobs in some parts of the state.
The average hatchery employee now earns about twice as much per year as the average commercial fisherman, according to an economic report prepared for the Alaska Seafood Marketing Institute, and the reported $239 million per year in annual hatchery spending now tops the less than $200 million dollars per year (in today’s dollars) the Sound’s commercial fisheries were producing prior to 1981.
Hatcheries were just beginning to kick in that year when Fish and Game reported “the total return of pink salmon was the largest run in recorded history in Prince William Sound and surpassed the previous record set only two years ago by over four million fish.”
The ’81 vcatch was valued at $58 million, just shy of $200 million when corrected for inflation to 2023 dollars and proved more than four times the size of the 4.5 million catch five years prior, which had been a record for the 1970s.
Not-so-good old days
Before the state hatchery program began to gear up in the mid-1970s, Sound harvests for averaged 3 million salmon of all species per year from 1951 to 1979, according to a history compiled by fisheries biologists.
A key part of the harvest during those years was high-value and now world-famous, wild Copper River sockeye salmon of which the annual average catch averaged more than 550,000, or about a fifth of the total harvest, according to state Fish and Game data.
In even number years, when Alaska runs of pink salmon are historically weak, the catches of high-value sockeye sometimes exceeded the catches of low-value pinks in the Sound. The sockeye harvest of 1972 was 17 times larger than the pink catch. Two years later, it was still 61 percent bigger.
But as returns of hatchery pinks increased, the sockeye harvests became a smaller and smaller part of the overall catch.
Wild Copper River sockeye are now a tiny percentage of the overall Sound harvest although their numbers increased in the 1980s and 1990s as the waters of the Gulf of Alaska warmed. They peaked in the 2010s at about 2 million sockeye per year, according to Fish and Game data.
The 10-year average for the period 2012 to 2021 is now down to 1.09 million of the fish, and scientists who in the mid-2010s went looking for long-term damage from the Exxon Valdez oil spill that smeared the Sound with crude oil in 1989 found an interesting ecological correlation between wild sockeye and hatchery pinks.
“All sockeye salmon stocks examined exhibited a downward trend in productivity with increasing Prince William Sound hatchery pink salmon returns,” they reported in a peer-reviewed study publised by the Public Library of Science (PLOS) in 2017. “While there was considerable variation in sockeye salmon productivity across the low- and mid-range of hatchery returns (0–30 million), productivity was particularly impacted at higher levels of hatchery returns.
“We do not know if possible deleterious interactions between hatchery pink salmon and wild sockeye salmon in this study are from predation or competition, or whether they occur in nearshore or offshore areas,” the study said. “Pink salmon feeding may cause a general depletion of prey availability that could impact sockeye salmon without tight spatial overlap of these two species. In this regard, the apparent impact to sockeye productivity may reflect a general increase in pink salmon abundance across the northeast Pacific rather than increased abundance of hatchery pink salmon to PWS (Prince William Sound) in particular.”
The 2012 to 2021 average return of hatchery pink salmon topped 38 million per year, according to state data, although it has fallen to near 28.6 million per year over the last five years despite the hatcheries releasing nearly the same numbers of young fish.
Although hatchery operators try to maximize the chances their fish will survive once they go to sea, ocean survival has generally been declining. The hatchery return of 34 million salmon in 2020 was only 65 percent of the forecast return of 52 million, according to the state reports; a 2019 return of 43 million was 54 percent of the forecast of 79 million.
How these numbers affect wild fish numbers is an unknown, but hatchery fish maximized for a better chance of ocean survival at sea are fish that by definition are supposed to have a competitive advantage over wild fish.
Meanwhile, the interactions at sea between the large numbers of small, short-lived pinks and bigger, longer-lived wild sockeye, coho and Chinook salmon are still not well understood.
Messing with nature
There are correlations between hatchery increases in pinks and declines in these wild fish, but as state fisheries research director Bill Templin pointed out to the Board of Fisheries in 2018, when the board was considering a rollback of hatchery releases, “correlation is not causation.”
That doesn’t make the correlations any less interesting or worthy of serious study.
Under state management since the 1970s, the annual commercial harvest of Chinook – Alaska’s fabled “king salmon,” the largest, longest-lived and most valuable of the species – is now less than half of what it was in the 1970s despite hatchery boosting.
Once famous for producing kings up to near 100 pounds in size, the Kenai River south of the state’s largest city is now closed to fishing for the species and fishing for kings throughout Cook Inlet has been restricted to protect kings because the number of spawners has fallen so low.
Hatchery releases have not been directly tied the decline in the big kings, but Canadian and lower 48 scientists contend that they can document Gulf of Alaska sockeye declines tied to the boom in hatchery pinks.
Three years ago, Brendan Connors of the Insitute of Ocean Sciences with Fisheries and Oceans Canada along with colleagues including Michael Malick from the U.S. National Marine Fisheries Service’s Northwest Fisheries Science Center in Seattle and others revealed they had found “evidence” 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 approximately 15 percent, on average.”
Their study concluding that overabundant hatchery pink salmon are reducing other salmon species via competition for food at sea was published in May 2020 in the peer-reviewed Canadian Journal of Fisheries and Aquatic Sciences.
Among the other scientists involved in the study were Seattle’s Greg Ruggerone and James Irvine from British Columbia who are credited with writing the definitive paper on a North Pacific salmon abundance in 2018.
That study pegged salmon populations as the highest in recorded history but found the population dominated by pinks.
Since then, other research has raised questions as to whether hatchery production of chum salmon in Southeast Alaska might be reducing the survival of wild British Columbia chums and possibly replacing wild chums in the Alaska Panhandle.
The state’s private, officially nonprofit hatcheries primarily run by state-sanctioned associations controlled by commercial fishermen have done a phenomenal job of producing hatchery fish, but the vast majority are pinks and chums or what were once called “dog salmon” in the north.
The dog salmon are now largely marketed as “keta salmon.”
Of the 12.8 million keta caught in Alaska in 2012, Fish and Game’ Salmon Fisheries Annual Enhancement Report claims 9.4 million – or more than 73 percent were hatchery chums,
Commercially caught Yukon River chums once supported what little of an economy there was in the economically deprived Interior of Alaska, but hatcheries along the Alaska coast largely gutted that fishery before it went into a still unexplained decline which might, in part, be due to competition for food with the proliferation of the salmon, some of them hatchery fish, in the Bering Sea and western North Pacific.
Russia now releases about 1.2 billion young salmon, primarily pinks and chums, into the northeastern Pacific each year, somewhat less than the approximately 1.7 billion released by Alaska hatcheries, according to data compiled by the North Pacific Anadromous Fish Commission (NPAFC), a treat organization.
There has been considerable debate within the scientific community about the ecological consequences of these releases with a variety of scientists arguing hatchery production is detrimental to wild fish and others going even further to argue hatcheries have triggered “trophic cascades” that have at times devastated populations of some seabirds and spread chaos throughout the marine ecosystem.
Though most environment-affecting activities undertaken in the U.S. now require the completion of an environmental impact statement (EIS) assessing the impacts of anything dumped into the nation’s waters, those National Environmental Protect Act standards have never been applied to the consequences of dumping young salmon in the ocean.
When the U.S. Forest Service conducted an environmental assessment, the baby sister of an EIS, at the Hidden Falls hatchery in Southeast Alaska in 2015, the report outlined how the hatchery’s production of chum salmon had “increased significantly” since the facility was built in 1978-79, but made no attempt to assess how the approximately 45 million young salmon the hatchery was dumping into nearby waters every year affected the nearshore or greater marine environments.
Forest Service reviewers were more interested in the potential problems of increasing numbers of visitors to “a large-scale hatchery that operates in an industrial manner” and stated their preference that going forward the site be managed “as a public education opportunity to tell the story of Forest Service, Alaska Department of Fish and Game, and Nothern Southeast Regional Aquaculture Association cooperative planning for the enhancement of fisheries resources through fish hatchery and aquaculture facilities and activities in the Tongass National Forest.”
The observations are a reflection of how subtle public perceptions affect public policy. There is “good” environmental tampering and “bad” environmental tampering.
Plowing under tens of thousands of acres of wild land in order to transform that land into fields of wheat is, for instance, considered a positive improvement on unproductive nature in the eyes of most. Digging up hundreds of acres to mine copper or other minerals, on the other hand, is largely considered a violent violation of the beauty of nature.
Salmon hatcheries have always fallen into the category of good tampering even as some of those in Alaska have begun to look more and more like the net-pen salmon farms the state banned in 1990.
Alaska hatchery fish are now regularly held in net pens just like the farmed salmon in Norway, Chile and elsewhere. And in those net pens they are fed and excrete waste just like the farmed salmon in Norway, Chile and elsewhere.
“Several factors determine how long the fish are fed in the net pens before they are released,” according to the Prince William Sound Aquaculture Association. “To increase the survival rate, a key target size and release time is selected. But often, critical factors, such as the occurrence of ‘zooplankton blooms’ in the receiving waters, which provide natural feed for the fry upon their release, also influence the decision as to when to release the fry.”
The key target size and release times linked to zooplankton blooms are intended to insure the hatchery fish are bigger than wild fish at release and grow faster than wild fish after release. Both of these factors increase the odds the hatchery fish will survive at sea.
Better odds for the hatchery fish could also translate into worse odds for wild fish, which is what the researchers working with masu samlonsay they found. Or, as they put it in the study, Japanese hatcheries caused “interspecific competition (that) reduced the unenhanced species.”
And in the case of masu, this happened in a situation wherein the hatchery fish were relatively few in number.
“It is noteworthy that the scale of the (masu) release program at the protected watersheds (annual average maximum equals 240,000 fish) is comparable or even smaller when compared with those for other fishery resources,” they wrote. “For example, 649.1 million fish are released annually to supplement the pink salmon stock at the Prince William Sound in Alaska, where wild populations seem to be severely compromised due to competition with hatchery fish.”
Alaska fishery managers and the politically appointed members of the Alaska Board of Fisheries, which is assigned the task of setting state fishery regulations, have, however, refuted that conclusion.
The ocean ecosystem, Templin told the Board, is so complicated it is impossible to document a cause and effect relationship between Alaska’s massive hatchery operations any declines in numbers of wild Alaska salmon, and the Board agreed.
The thoughtful comments are appreciated. I have wondered why the focus of production has almost exclusively been on salmon and why more effort has not been placed on how we might improve or restore habitat and reproduction of key prey species on the west coast.
If up to one percent of salmon spawn in different rivers and hatcheries than their origin, after over 100 years of of hatchery produced fish spawning with wild fish, it’s a struggle to embrace the idea of superiority of one over another. Non-hatchery fish were used for broodstock during the same time period. What identified differences exist between hatchery and non-hatchery?
California, Oregon and Washington have been running their programs, constrained by the believers in wild superiority, for decades. On the chance hatchery fish competition has negative impact on non-hatchery, we remove returning hatchery fish without allowing them to continue upstream, spawn, and die in the rivers.
We deprive the rivers and non-hatchery fish of millions of pounds of rotting flesh which would otherwise replenish nutrients in river systems, We guarantee non-hatchery fish are denied benefit from massive returning salmon runs. The following year, salmon have far less to eat when they hatch. We trade potential impacts for a guaranteed impacts. Who determined the guaranteed impacts are a lower risk?
The results of creating guaranteed impacts to offset potential impacts down south are amazing. California and Oregon seasons are closed this year.
The elephant in the room is the huge number of hatchery fish mixed the much lessor number of Wild fish. ( mixed stock fishery). The harvest scoops up large sets of this mixture masking the loss of wild fish. Consistent otolith monitoring in these mixtures is rare. Stray hatchery fish bloat escapement goals giving an illusion of healthy returns of wild fish…but oops wrong fish…fish with lower resilience, less eggs incrementally replace the wild savvy.
Wild chums in SEAK and BC show this loss in the 90% gone. The state crows low value quantity over quality. It is time the state gets over its denial and admit we are unwittingly liquidating our wild populations benefitting few
Lots of good comments here about a topic that’s been on my mind for some time, as we frantically try to zero in on the factors behind our statewide decline of king salmon. I don’t doubt that climate change has quite a bit to do with the fluctuations we’re seeing in numbers and size of all the salmon species, but we need to focus on the things that we can control right now, starting with the most probable culprits. The release of billions of chum and pink salmon smolts annually at terminal sites all along the North Pacific by Russia, US and Japan to augment commercial fishing shows reckless disregard for the carrying capacity of the oceans and the health of the wild salmon runs. Its not rocket science. These fish will compete directly or indirectly with the other salmon on the high seas. We need to rein in these hatcheries now, both here in Alaska and on the other side of the Pacific. And the trawlers are another probable major factor, at least for the runs in SW Alaska. It will be interesting, and very telling, to see in the next few years whether our management institutions will take the necessary steps to limit these giant industries to save our wild runs here in Alaska.
As documented in this and other articles you have written, there are scientists working on the thought that our large hatchery releases of salmon may be displacing wild salmon stocks. If one even considers basic science taught in our elementary schools — that there is such a thing as carrying capacity — this is logical. When one considers the observations of lower sockeye salmon production from Copper River / Prince William Sound stocks following years with 30 million plus pink salmon returns it only becomes more plausible.
Considering wild king salmon that have declined throughout the entire State of Alaska — if even a portion of the declines in king salmon numbers and fish size have been caused by excessive competition with hatchery salmon — what are the chances wild king salmon production will improve as long as the same numbers of hatchery salmon continue to be released. Now that even young hatchery pink salmon are raised to a competitive-advantage size compared to wild pink salmon — the chances of wild king salmon production improving may further decline.
In talking with a senior Alaska Department of Fish and Game staff member this past week, the statement was made that Prince William Sound hatchery pink salmon release have not increased since the 2010s. Point taken — however that is after saltwater king salmon production — so if the wild king salmon decline was even partially caused by competition with hatchery salmon — are we not still exacerbating the issue by continuing to release the same excessive numbers of hatchery salmon?
Are we, as a State, continuing a king salmon crisis on an annual basis through our own hatchery operations? Are large hatchery productions in Prince William Sound, and other Alaska locations, causing serious declines in wild salmon production in the Sound and other parts of the state? If the answers to these questions even need to be debated — our hatchery production may simply be too large.
Hatchery releases have indeed been stable since the 2010s, but pen-rearing salmon before release to get them to that “optimum size” has increased. The whole idea there is to make increase their odds of survival for hatchery fish. Given that at carrying capacity, survival is a zero-sum game, some other fish are going to have to die to make room for these “optimum size fish.”
The key question then becomes simple; Which fish are dying?
We don’t have a lot of data on this subject. It hasn’t really been studied much. But there are some pretty damn strong indications hatchery fish could be displacing wild fish, and not just hatchery pinks displacing wild pinks. Hatchery pinks might be displacing Chinook, sockeye and coho as well.
You might want to take a look at average annual sockeye returns to Cook Inlet in the decade before the hatcheries reached full production in the 2010s, and the average annual sockeye returns to Cook Inlet in the years since then. Correlation is not causation, but the correlation sure ought to stir a whole lot more effort at investigation.
let me know if you think I’ve got the right concept:
Even though the number of hatchery fish released has not increased sine the 2010s — the raising to optimum size before release– may have as much or more impact than if additional fish had been released.
Do you know when pink salmon production started getting ramped up through Cook Inlet Aquacluture in Lower Cook Inlet? Do you know what level that production is now? and if it is still in the process of getting ramped up to some specific level of production?
Andy: You have the concept right. Anything that is done to increase survival is effectively the same as adding more fish.
CIAA production has been all over the place over time with a fair bit of experimentation as to size. CIAA released 8.2 million, 0.76 gram pinks at Tutka in 2012, and it was up to 55 million last year, but the fish were about half the size, 0.35 grams. They released 86 million at 0.70 grams in 2019, clearly trying to take advantage of the odd-year strenth of pink runs.
The 2015 release was of 11.2M, 1.07 gram fed pinks and 1M, 0.28 gram unfed pinks, which gives you some idea of the sort of manipulation going on here. You go through all CIAA’s numbers here: https://www.ciaanet.org/reports/releases/
There are tradeoffs between releasing fed and unfed fry, the big one being that it costs money to feed the fish. But CIAA has been pushing for more feeding opportunities. I’m guessing your familiar with the net-pen operations the organization now runs: https://craigmedred.news/2018/05/11/runaway-net-pens/
I personally find it more than a little ironic that the state of Alaska banned net-pen “farming” of salmon and now we’re heaving into growing salmon in net pens to further the “ranching” business.
When is a farm not a farm?
How many fry have not been in ocean due to absence of returns in BC and WN? We chose not to release pinks in SE and chose to do chum,coho and kings.The chums have been straying and producing new runs while evolving into WILD fish
How do the experts propose to get Russia to moderate releases?
I do not think enough research has been done to come to any firm conclusions on this topic but it certainly could indicate a need to moderate some of our releases while others may increase carrying capacity regionally.
Does anyone really think that wild King salmon across all of the northern Pacific Ocean watersheds have collectively forgotten how to eat?
Kenai kings used to eat enough food in the ocean in the 1980s to stay out in the ocean many for five years and return typically more than 50 pounds, now few if any return at that size and age.
To listen to top fish researchers explain – Food chains are so complex in the ocean… It is so tough to tease apart fish growth patterns in the ocean…
It is just a black box and we will never be able to understand what happens in the ocean environment… Likely it’s just climate change causing the collapse of wild Kings, Chums, Sockeyes… yeah, no, even though there are more salmon than ever recorded in history swimming in the northern Pacific Ocean.
Fascinating is the lack of interest in really examining the largest and longest existing biological sampling program in history – fish scales of all species of salmon are collected every year in every ADFG managed salmon fishery, extending back prior to statehood.
Is it really too hard to do basic fishery research to see how growth patterns for salmon, recorded in the fish scales, have changed through time since the introduction of hatchery fish?
Salmon fish scales are akin to tree rings and document ocean growth patterns – per individual, per species, per watershed, through time. Too bad we live in a time when there are those thingy thinky computer mabobs…
One challenge in using salmon scales to study relationships between growth and survival is that very often scales are only sampled from fish that have survived to return to a fishery/hatchery/stream. So we don’t have as many scale samples from fish that did not grow enough to avoid predation and/or starvation. Assessing growth trends only from individuals who were able to successfully eat, grow, and survive may not get at the hatchery competition question. This is part of the reason why recent high seas sampling and surveys are so important.
Are not all salmon escapement goals in essence based on the total run return, of fish that both return for escapement and are harvested?
If useful biological and management information, numbers upon which supposed sustainable modern salmon management is based upon, come from returning / surviving fish, then the argument that somehow the scale information is flawed, maybe fatally, seems politely, like hogwash.
Let’s see the complete salmon scale data set in a relational database, by year, by species, by watersheds, to see if useful information can be gleamed through time.
Do large returns of pink salmon, such as the last decade of odd returns of pinks in the gulf of Alaska, impact the growth of other salmon species – are their alternating cycles of growth for kings and reds that spend a minimum of three years in the ocean?
If there is no food competition, then it will show up in the fish scales; if there is, it will show up in the fish scales – of course with the caveat that impact was felt by the fish that lived successfully to the river of origin.
But at least it is worthwhile to solve that part of the salmon ecology equation.
This paper provides some interesting context but differs substantially from the pink and chum salmon releases into the North Pacific. Masu salmon rear for at least a year in the release streams. It isn’t hard to imagine some intra- and interspecific competition given the relatively small amount of habitat they would share for the rearing period of their lifecycle. This paper doesn’t provide strong evidence to your arguement.
It doesn’t, Ed, if you assume the ocean is one giant habitat. It’s not. Young salmon appear to be largely contained by currents, ocean rivers so to speak, which puts them in a similar situation as those masu salmon.
It’s a mistake the believe ocean habitat is all the same because the surface of water all looks the same. The habitats beneath the surface mimic those of land – some good, some bad, some in between.