Site icon Conservation news

Spamming streams with hatchery salmon can disrupt ecosystems, study finds

A coho salmon spawning.

A coho salmon spawning in the Salmon River. Image by Bureau of Land Management via Flickr (CC BY 2.0).

  • In a new study, researchers found that releasing hatchery-bred native masu salmon into freshwater streams in Hokkaido, Japan, destabilized the local ecosystems.
  • Overall, the study found that the total number of fish, and number of different species, both declined in the long term due to greater competition for resources like food and preferred feeding spots.
  • Masu salmon populations also did not increase in the long term, the research found.
  • With hatchery releases increasing in many areas — and for many species — the findings add to the ongoing debate over their wider effects on wild fish populations.

Every year, 22 million sockeye salmon begin life some 420 kilometers, or about 260 miles, inland from the Alaskan coast, in plastic bins. They’re at the Gulkana hatchery, the largest sockeye salmon hatchery in the world — but just one of countless hatcheries around the globe that release native fish into rivers, lakes and oceans to augment wild stocks.

Fish hatcheries, like the species they breed, come in many shapes and sizes. Some, like the Gulkana facility in the U.S., pump out millions of fish each year to support commercial fisheries. Others, like the Nechako White Sturgeon Conservation Centre hatchery in Canada, focus on conservation by rebuilding and maintaining genetic diversity in threatened populations. Some hatcheries use eggs and milt (sperm) from wild fish; others use hatchery fish as parents. But no matter what the species of fish or the purpose of the hatchery program, the released fish enter the ecosystem en masse and interact with an existing community.

Pacific salmon are economically and culturally important species for nations across the northern Pacific Ocean, and hatchery programs positioned to boost commercial fisheries have increased dramatically over the last 150 years. Since the early 1990s, approximately 5 billion hatchery salmon have been released every year, primarily by the United States, Japan and Russia. Chum (Oncorhynchus keta) and pink (Oncorhynchus gorbuscha) salmon are by far the most common hatchery salmon species, but others are also bred and released from hatcheries. For example, in Japan, masu salmon (Oncorhynchus masou masou) are a highly sought-after fish, and hatchery managers release approximately 10 million hatchery-bred fry — a juvenile life stage of salmon — into freshwater each year. Typically, the released masu fry stay in streams for at least a year before the majority migrate to sea (as smolts), returning a year later to freshwater streams to spawn (as adults).

Masu salmon in Hokkaido, Japan.
Masu salmon in Hokkaido, Japan. Image by Akira Terui.

But hatchery programs may have unintended consequences, according to a recent study in the Proceedings of the National Academy of Sciences. On Hokkaido Island in northern Japan, researchers found that the release of hatchery masu salmon destabilized stream communities, lowered the number of other fish species in streams, and didn’t lead to a long-term increase of masu salmon.

Study lead author Akira Terui, a freshwater ecologist at the University of North Carolina at Greensboro in the U.S., told Mongabay the findings show that releasing fish into natural environments, even when using native species, sometimes can have “harmful or unexpected consequences to the entire ecosystem in the long term.”

Release of hatchery salmon on Hokkaido has been ongoing since the 1950s to support commercial fisheries. The streams are protected — no angling or habitat alteration is allowed — and the Hokkaido Research Institute monitors all stream fish populations.

To understand how the periodic influx of hatchery fish might impact the whole steam community, the researchers first created a mathematical model that included parameters like how many hatchery fish were released, how fast species grew, and how many fish the ecosystem could theoretically support, a concept known as the carrying capacity. Then they compared the model’s prediction with 21 years of long-term data from 97 streams on Hokkaido where hatchery fish had been released.

Chum eggs collected from spawning salmon
Chum eggs collected from spawning salmon at the Chehalis River Hatchery in British Columbia, Canada. Image by Mirko Diaz.
Rows of incubation trays where eggs are reared and hatched
Rows of incubation trays where eggs are reared and hatched at Big Qualicum River Hatchery in British Columbia, Canada. Image by Sam James.

The findings showed that in the long run, streams that had a greater number of released hatchery fish had fewer fish overall, and less species variation (a measure called species richness). That’s because adding more fish than the ecosystem could handle interfered with the way species coexisted and shared resources such as food or preferred feeding spots. There was more competition between different species that had a similar diet or other needs — for example, between the masu salmon and rainbow trout (Oncorhynchus mykiss) for drift insects. Competition for limited resources also intensified between fish of the same species. Generally, hatchery masu are more aggressive than their wild counterparts, the study noted, which may have given them an advantage over wild masu salmon.

Overall, Terui said the model and data both showed that “the competition effect overwhelms the positive aspects of the fish release,” particularly when carrying capacity of the ecosystem was low. In other words, he said, “nothing good happened.”

Study co-author Hirokazu Urabe, a researcher at the Salmon and Freshwater Fisheries Research Institute in Hokkaido, told Mongabay by email that he was very surprised by the results, and that the study has received a lot of attention in Japan. He said the findings show that encouraging natural spawning by restoring habitat should be a management priority. Hatchery releases should be used with caution, and only after careful consideration of the carrying capacity of the freshwater environment, he added.

A stream in Hokkaido, Japan.
All streams in the study area in Hokkaido, Japan were strictly protected. Image by Bong Grit via Flickr (CC BY-NC-ND 2.0).

Jason Hwang, vice president of the Pacific Salmon Foundation, a Canadian environmental organization that preserves and restores Pacific salmon (Oncorhynchus spp.), who was not involved in the research, told Mongabay the study highlights that salmon hatcheries can have extremely complex ecosystem-wide impacts. However, he said different factors will determine ecosystem impacts, so it isn’t possible to make generalizations about those impacts. For instance, he said, differences in salmon species, hatchery release management strategies, and even the natural environment, will affect the way an ecosystem reacts to hatchery releases.

Hwang said that in British Columbia, salmon hatcheries use various management strategies to avoid excessive pressure on freshwater ecosystems. Right now, wild salmon populations in the Canadian province are generally low, he said, so researchers use historical data to estimate how many fish the freshwater habitat could sustain. Then they look at how many wild salmon are returning to spawn, and plan releases so that fry numbers in the freshwater ecosystem are in line with historic levels. Or, in some places, hatchery salmon are released in freshwater when they are more mature (as smolts), so that they’re nearly ready to migrate to sea, he said. That minimizes the amount of time hatchery salmon share freshwater resources with wild salmon or other fish.

But research suggests that hatchery-released Pacific salmon and wild fish also compete for resources in the ocean. A 2018 study in Marine and Coastal Fisheries found that for the three most abundant Pacific salmon species — chum, pink and sockeye (Oncorhynchus nerka) — hatchery salmon make up 40% of the 5 million metric ton total salmon biomass in the Pacific Ocean. The authors noted that while interactions between wild and hatchery salmon are still unclear, we can already see the effect that abundance of Pacific salmon is having on various species in the marine ecosystem, and that “there are indications the carrying capacity of the ocean may have been reached.”

Hwang said it’s the very large hatcheries releasing millions of salmon — and not the small-scale hatcheries — that contribute to that high salmon abundance, which might be increasing competition and pushing the carrying capacity in the ocean. Still, he noted, “all hatchery production needs to be aware that the ocean does not have infinite productivity.”

A chum salmon
A chum salmon in Allison Springs, Washington. The three most abundant Pacific salmon species — chum, pink and sockeye — hatchery salmon make up 40% of the 5 million metric ton total salmon biomass in the Pacific Ocean. Image by Roger Tabor/USFWS via Flickr (CC BY-NC 2.0).
A harbor seal skinning a chum salmon.
A harbor seal skinning a chum salmon in Whatcom Creek Estuary, Washington. The study highlights that salmon hatcheries can have extremely complex ecosystem-wide impacts. Image by Andrew Reding via Flickr (CC BY-NC-ND 2.0).

Michael Price, scientific director of Skeena Wild, a conservation trust in British Columbia and a postdoctoral fellow at Simon Fraser University, who wasn’t involved in the study, pointed out another way hatchery programs can impact wild populations: fishing pressure. Large salmon hatcheries increase stocks to the level where the number of salmon in the Pacific supports large commercial fisheries. That means more fishing fleets in the ocean. The problem is, hatchery and wild salmon are out in the ocean together — and they both ended up getting caught, Price told Mongabay. That puts pressure on wild populations.

Price said one way to alleviate that pressure would be to mark all hatchery fish by clipping the small adipose fin on their backs. Fishers could then retain hatchery fish only and release wild salmon, in what’s called mark-selective fishery. That’s already being done in some places, like Washington state in the U.S., where virtually all coho (Oncorhynchus kisutch) and Chinook salmon (Oncorhynchus tshawytscha) are marked,  Some, though, have raised concerns about mark-selective fishery’s efficacy as a management tool, due to the unintended mortality of released fish and other factors.

While carefully managed hatcheries can be a tool for rebuilding populations, Price said that until we can stop commercial fisheries from intercepting vulnerable wild populations, hatcheries probably end up doing more harm than good. “I think we need to really start scaling back on hatcheries and focusing just on the natural world and rebuilding populations in another way,” he said.

Terui said we need to pay more attention to how all species, including those that aren’t economically important, interact, and prioritize the health of the whole ecosystem.

“Environmental restoration or preservation is something to come first,” he said. “Afterwards, some of the [release] programs might work, depending on the species attributes.”

Banner image: A coho salmon spawning in the Salmon River. Image by Bureau of Land Management via Flickr (CC BY 2.0).

Read more: As sea lice feast away on dwindling salmon, First Nations decide the fate of salmon farms


Terui, A., Urabe, H., Senzaki, M., & Nishizawa, B. (2023). Intentional release of native species undermines ecological stability. Proceedings of the National Academy of Sciences, 120(7), e2218044120. doi:10.1073/pnas.2218044120

Ruggerone, G. T., & Nielsen, J. L. (2004). Evidence for competitive dominance of pink salmon (Oncorhynchus gorbuscha) over other salmonids in the North Pacific Ocean. Reviews in Fish Biology and Fisheries, 14(3), 371-390. doi:10.1007/s11160-004-6927-0

Ruggerone, G. T., & Irvine, J. R. (2018). Numbers and biomass of natural‐ and hatchery‐origin pink salmon, chum salmon, and sockeye salmon in the North Pacific Ocean, 1925-2015. Marine and Coastal Fisheries, 10(2), 152-168. doi:10.1002/mcf2.10023

In Chile’s Patagonia, another salmon plant angers water defenders

Exit mobile version