top of page
  • FBU

Net loss: the high price of salmon farming

By: Mark Kurlansky September 15, 2020 The Guardian The salmon has always been a barometer for the health of the planet. Now industrial-scale farming is bringing pollution, plagues of sea lice and threatening the future of wild salmon. When a fish is in crisis, the public wants to blame the fishermen. It is preferable to blaming ourselves. But a fish whose only problem was overfishing, a fish stock that could be saved simply by a ban on all commercial fishing, would be very rare. It would be an enviably easy problem to fix.

The salmon is as magnificent an animal as anything on the Serengeti – beautiful in its many phases; thrilling in its athleticism; moving in its strength, determination and courage – and it would be a tragedy if it were to disappear. All that is true, but a more important point is that if the salmon does not survive, there is little hope for the survival of the planet.


The salmon, though it belongs only to the northern hemisphere, has always been a kind of barometer for the planet’s health. That is because anadromous fish – fish that live part of their life in freshwater lakes and rivers and part of it in the sea – offer a clear connection between marine and terrestrial ecology. Most of what we do on land ends up impacting the ocean, but with salmon we are able to see that connection more clearly.

Our greatest assaults on the environment are visible in salmon. Complex as the problem of survival is for most fish, few species are faced with as many difficulties as salmon. This is partly because it is central to the “food web” (now that we understand the importance of biodiversity and the interdependence of species, this term has replaced the more familiar “food chain”) and partly because of a complicated life cycle that depends on both marine and inland habitat. In 2005, a group of scientists studying the survival prognosis for Pacific salmon concluded that 23% of all salmon stocks in the world were at moderate or high risk of complete extinction. For Atlantic salmon, the situation is even more desperate.

There is a growing realisation, greatly promoted by the United Nations Food and Agriculture Organization (FAO), that we have to start producing much more protein to sustain a growing world population. The FAO believes that we cannot afford to plough up more land for agriculture and we need to derive more protein from the sea. This is clearly not going to be accomplished with wild fish, already struggling under the effects of climate change. One way to make the sea productive is fish farming. But fish farming currently creates as many problems as it solves.

A tourist viewing the dramatic sheer mountain fjords of Norway and the island-strewn mouths of the rivers or passing the lochs of Scotland might easily never notice the presence of fish farms. They might see an unobtrusive area with metal poles sticking about a metre out of the water and perhaps some netting over the top of floating wooden walkways; it doesn’t look like much. Invisible from across the loch are a million salmon below. A visitor looking closely might see a few salmon leaping, because salmon always leap.

But passersby see very little because most of the farm – other than a few tubes to a feeder, the top rim of a fish pen that goes down some 50 metres (164ft), and usually a barge – is below the surface of the water. The pen needs to be placed in deep water because it cannot rest on the bottom and should have more than 30 metres of swiftly moving ocean current rushing under it. The pen is large enough for about 200,000 fish – as some opponents point out, packed in tightly. But fish generally don’t mind a crowd, believing there is safety in numbers. Even in these packed pens, there is open water as well as crowded areas.

In Norway, where fish farming is big business, a farm might have eight or 10 pens, which would hold more fish than the wild Atlantic salmon population of the entire world. Originally fish farms in Norway were not this large. Salmon farming started in Norway in the 70s as an extension of normal agriculture, and a farmer would get a licence for one pen that would be placed in the sea as close as possible to the farm. But farmers discovered that fish farming required a great deal of capital and most gave up. At the same time, they found that if you had the capital and could farm on a large scale, it was enormously profitable. So large companies started buying up licences from failing farms, meaning it was soon a large-scale industry. A few one-licence farms still exist, but most salmon farming operations are owned by large multinational companies.

The first complaint against salmon farming is that it is polluting. A pen with 200,000 fish produces an enormous amount of waste. In nature, animal waste is not harmful; in fact it is often beneficial. But large concentrations of it can be destructive. The waste of wild fish swimming around is not harmful, but the waste of hundreds of thousands staying in the same spot is. What was needed, the big companies that took over came to understand, was a spot in the sea where strong currents would carry and disperse the waste. A spot such as the Bay of Fundy, between Nova Scotia and New Brunswick in Canada and Maine in the US, and has the strongest tides in the world, was ideal. But a few shallow- or slow-water farms still operate. Chile is often singled out by environmental groups such as Greenpeace for lax regulations and poorly located farms.

The vexing problem that fish farming companies have yet to solve is escaping fish. If they are in the Atlantic, the escaped farmed fish mix with wild. The males usually die off because they are not strong enough to compete with wild males in the tough and often violent process that is salmon spawning. But the females lay eggs and wild males fertilise them. These escapes are frequent enough that wherever there are Atlantic salmon being farmed near wild Atlantic salmon, they are mixing. Farmed genomes, often Norwegian, are showing up in the wild salmon of Norwegian, Scottish, English and Irish rivers. They turn up in the genetics of wild fish in Nova Scotia, New Brunswick and Maine. More than a third of the wild catch from the Faroes are actually escaped farmed fish. This is another reason why some so-called “wild” fish from honest merchants turn out to be farmed when DNA testing is done on the commercial catch.

Farmed salmon are not greatly different from one pen to another. They have been selected for fast growth, and growing fast seems to be their major skill. They do not have all the special survival skills of the wild stock. Although fast-growing, they only grow for a short time and never achieve the size of the more slow-growing wild salmon. This is one of the reasons that they do not reproduce at the same rate as the larger wild fish. A salmon living in the wild that has a farmed parent or even grandparent is much less likely to survive at sea, and, in fact, sea survival has declined in places with farming.

Even fish farmers agree that this is a cause for concern. In Norway, at any given time, there are about 400m farmed fish and only 500,000 wild fish. Tor Eirik Homme, director of feed and nutrition for a large Norwegian farming company, Grieg Seafood, says, “I understand their fear”.

The risks of escaped farmed Atlantic salmon in the Pacific, where there are far more wild salmon, are not entirely understood. Being from different genera, an Atlantic salmon will not mate with a Pacific salmon. But even farmed Atlantics spawning with each other does not guarantee that the offspring will be successful; after all, a major argument against farming in the Atlantic is that farmed fish lack survival skills.

In the summer of 2017, an accident destroyed one of eight Atlantic salmon farms in Washington state. A pen owned by a Canadian company came apart and released 250,000 farmed Atlantic salmon into the wild. They ran to Puget Sound and up into rivers. It is not certain what damage these alien fish could do in the wild. In the Atlantic, such an accident would be devastating, but most of the escaped salmon disappeared within months and are thought to be dead. The accident caused enough fear, however, that salmon farming was banned in the state. All salmon farming is to be phased out in Washington state by 2025.

Fish escape from pens simply through accidents, usually human error, though occasionally a seal might tear a net. In the Washington state case, inadequate maintenance allowing a buildup of mussels and other sea life made pens susceptible to collapse. Sometimes equipment can tear holes in nets. Fish have also escaped because of damage caused by storms. This may become an even greater problem because, due to climate change, we are living in an age with an increasing number of violent storms. The pens are being made from stronger Kevlar netting. “We are improving,” says Homme. “In the early days sometimes some smolts [a salmon’s intermediary stage] would be smaller than the mesh size,” he continues, “one fish escaped is far too many.”

The fish farming industry, so far, has been mainly looking for sticking plasters for this problem. An improvement would be to give up pens and use enclosed units, or perhaps even give up working at sea. Farmers have a number of reasons why they are reluctant to turn to such alternatives, most notably the added expense, but they may be forced to act because of a much more difficult problem: sea lice.

Sea lice are saltwater crustaceans, smaller than a fingernail but still clearly visible. There are 37 genera of sea lice, all parasites, of which two target salmon. Before there were fish farms they did not pose a significant problem. They roamed the ocean looking for salmon, which make up a tiny minority of the fish population. One or two might attach themselves to a salmon, and the fish would live with the parasites until it returned to the river. Sea lice cannot live long in fresh water, so they fall off and die in the river. In fact, anglers are pleased when they catch a fish with one or two lice still on it because it means the fish only recently entered the river and are still in their most robust state.

Until farming, sea lice survived but never found huge schools of salmon on which to feed. Now they find salmon farms with hundreds of thousands of salmon trapped in one spot. As Homme says, “Now it’s party time”. The lice eat the salmon’s skin. It is difficult to penetrate the scales so they attack the head and neck. They will completely skin a fish’s head and then it will die of exposure. Farmers find the dead fish with raw skinned heads at the bottom of the pens. It is not unusual to lose a quarter of a pen. The lice also attack smolts. While it takes more than 50 lice to kill an adult, it only takes about 10 to kill a smolt. Sea lice are a huge financial loss for fish farmers.

From their base in the fish farms, sea lice can also easily spread out and attack nearby wild salmon. Wherever there are fish farms near wild runs, the wild runs are also plagued by sea lice. The problem is less intense in the Pacific where the wild population still greatly outnumbers the farmed fish. Nevertheless, wild salmon fisheries on the US west coast that are near salmon farms also report their fish being nearly overwhelmed by sea lice. In the Atlantic, biologists believe that some salmon are taking additional quick trips into the rivers to lose the lice. But this is a dangerous strategy. A week away from the sea means a week missing out on the high-nutrition ocean diet needed so that they will be large and strong for their spawning journey.

River managers, biologists, conservationists and environmentalists have been demanding that fish farms be moved away from wild runs, sealed off, or brought out of the sea on to land – many of the same answers being proposed to deal with escapes.

There are several inland salmon farms but this has its own environmental cost. Is it a benefit to the environment to stop using the natural force of the ocean to circulate water and instead burn energy pumping water? This also becomes a more costly way to produce, and farmed salmon would no longer be a low-cost product.


Open net pens at sea are very cheap to build and operate. Ben Hadfield, the managing director of Marine Harvest Scotland [now Mowi], asks: “Do you want to take a low carbon-efficient protein and move it on land and increase its carbon footprint? Meat [farming] is extremely inefficient and I don’t want to do that to fish farming.”

It has been suggested that a closed containment fish farm could produce its own energy from the fish waste it extracted. This technology is available and might make closed farming more appealing. However, Homme at Grieg Seafood says: “Cost for land-based farming is another level and also it requires a lot of land. We need to produce food in the sea because that is the most space available.”

Because of the high cost and questionable environmental impact of these potential alternatives, fish farmers want to try almost everything else first. They are trying to breed lice-resistant fish. Short of that, they wash the fish, brush them, warm the water for 20 seconds. But these measures will also kill a few fish. Farmers have tried to chemically kill off the lice with what is essentially an anti-crustacean poison. It will kill sea lice but can also kill shrimp or lobster. At the beginning of this century there was a huge outcry among lobstermen in the Bay of Fundy who claimed these chemicals, used by fish farmers, were also killing their lobsters. If these chemicals spread to krill, a tiny crustacean that is critical to the diet of salmon and many other fish in the ocean, it would be devastating. Questions have also been raised about eating salmon treated with these chemicals. In any event, farmers have largely stopped doing this because the lice, over time, develop immunity to the chemicals.

Fish farmers then turned to an older idea that was used against insects before pesticides were invented. Known as biological control, the idea is to find animals that will destroy the unwanted pest. Fish farmers found two Norwegian natives: wrasse and lumpsuckers.

Lumpsuckers are tiny globe-like fish that buzz around a pen beating their fins so furiously that they resemble an insect. They attach to surfaces with a sucking mechanism on their bellies. Wrasse, a colourful fish the size of a small grouper, is the more efficient lice killer, but the waters of northern Norway are too cold for them and so lumpsuckers are needed for northern farms. Neither lumpsuckers nor wrasse can live off sea lice, but if they are fed and well taken care of, they will spend their time eating the lice off salmon as an added treat.

According to Brit Hjeltnes, the deputy director of fish health at the Norwegian Veterinary Institute, placing these fish in farms is a very dangerous practice because introducing wild animals to farm animals always carries the risk of the wild bringing new diseases.

But a different problem arrived first. In capturing enough lice-fighters to protect their salmon, the farmers were overfishing wrasse and lumpsucker stocks. Farming in closed containers instead of open pens would create some problems, but it could also eliminate the issues of escapes and sea lice and the spread of diseases. The spread of diseases has many observers troubled.

It is not surprising that large numbers of animals kept closely together are a kind of Petri dish for growing dangerous pathogens. The farmers seek to protect their fish from these pathogens the same way humans gird themselves against communicable disease: the fish are vaccinated. While smolts, the fish are anaesthetised in a bath and shot in the belly. The basic vaccine had seven components, but as new pathogens are found, additional vaccines are developed.


Farmers have been fairly successful against bacteria, but more and more viruses are being discovered. These diseases, if not controlled in the pen, might spread to the wild population.

The use of a variety of antimicrobials – medication to fight bacteria, viruses and fungi – is worrisome to consumers. This same issue has come up with beef, milk and other farm products. Dangerous fungi, bacteria and viruses with resistance or even immunity to all known medicines have been appearing and rapidly spreading. One dangerous fungus, Candida auris, was found in Japan and within 10 years has spread to 30 countries including the US.

These so-called “superbugs” are becoming a major health concern, and scientists believe they are caused by humans overusing medications. Some believe that medicating farm animals, salmon included, can overexpose humans through food and contribute to this problem.

It is also reasonable to suspect that wild fish are dying from fish-farm diseases, but this is extremely difficult to prove because the bodies of most wild fish that die are consumed by predators and never found. When the escaped salmon in Puget Sound in Washington state were examined, numerous viruses were found but none that were deemed a threat to local wild fish. Even without proof, the spread of disease by fish farms is a volatile issue.

In British Columbian fish farms, Norwegian eggs were thoroughly screened and tested for diseases. The business grew fast. In 1984 there were 10 farms and by 1986 there were 40. The industry continued to grow, and by 2003 there were 121 farms in British Columbia. But the chosen sites for farming are very close to British Columbia’s most important wild salmon runs. That and the number of farms have raised great public concern.

One of the original complaints against fish farming was that it is unsustainable. The industry likes to claim that it has developed a sustainable way of providing affordable fish. Salmon farming companies such as Marine Harvest are enormously profitable. Just their Scotland operation can see £58m in a good quarter.


Salmon are carnivores and eat other fish. In the early days of fish farming, fish was ground into meal and pressed into pellets that looked like dog food. It would take 10lb of wild fish (dry weight) to make a 10lb farmed fish (wet weight). So where was the gain? The fish for fishmeal tended to be scooped up by huge factory trawlers, one of the most wasteful types of fisheries.

The accusation of being unsustainable may have hit home for companies such as Marine Harvest with the rising cost of fishmeal. Forty per cent of the cost of raising a farmed fish is the feed. Farming companies would like to reduce that by turning their salmon into vegetarians, but this is not easy because salmon have short intestines designed for digesting meat but not well-suited for plants. The old, entirely fish-based feed would be prohibitively expensive today. Soy is a large component in modern fishmeal. Originally raw soy was used, but now a protein concentrate has replaced it. Protein is what salmon need and no matter what is done to soy, it does not have the 70% protein of fishmeal. Cargill, the American multinational food giant, makes different feeds at different prices. The higher the fish component, the more expensive the feed; but high-priced food might be cheaper in the long run, because the salmon grow to market size more quickly.

The goal is to get more protein with less fishmeal. The search is on for high-protein plants, but they are rare and often expensive. Marine Harvest, which produces 23% of the world’s farmed salmon, uses a feed of 14% fishmeal, 8% fish oil, 20% rapeseed oil, and more than half soy, corn and wheat gluten. But it is recognised that farming soy is not a sustainable way to produce fish either There is research into producing protein from algae grown by fermenting plants. There is also research on the ability of some insects to convert carbohydrates into protein. The black soldier fly larvae are already being used as protein in some animal feed.

Farmed fish are less likely to spread poisons when they are not fed fishmeal. Wild fish pick up contaminants from the ocean, many of which are concentrated in the cold northern waters where the small fishmeal species live. But it is really a trade-off. If farmed salmon eat more plants, they have less risk from heavy metals and PCBs, but more risk from pesticides.

Salmon farming is likely to survive – enough people want the jobs, the profits, or the fish – but it has many battles ahead. On 19 November 2015, the US Food and Drug Administration, to the horror of some and excitement of others, ruled that genetically engineered salmon was fit for human consumption. Further, the FDA ruled that these salmon did not have to be labelled as genetically engineered, which is the highly controversial US policy for genetically engineered crops. As such, salmon is leading the way to a scary new world. It is the first genetically engineered animal to be allowed in stores. AquaBounty Technologies, so far a small company in Massachusetts, spent almost 20 years fighting for this approval. If this Frankenfish, as it is sometimes called, were to mix with a wild population, it might have disastrous repercussions, perhaps even worse than other farmed salmon. Legal challenges began immediately following FDA approval. The genetically engineered salmon is a farmed salmon that has been manipulated to grow to market size in about half the time that it takes a normal farmed salmon. Is it the farmed salmon of the future? The plan is for it to be raised inland in contained tanks to minimise exposure to wild animals. If these genetically modified animals succeed, it may be difficult for other farmed salmon to compete. Or the old type of farmed salmon may gain a special cachet. But if genetically modified fish are produced on a large scale inland with a low carbon footprint, and making the claim that unlike the sea-farmed fish they do not harm wild fish, this might move the entire salmon farming industry away from the shore. We are learning in the increasingly challenging struggles to save the environment that simply banning things yields very little. We have to find viable solutions. There are good reasons for fish farming and real dangers to it as well. So the question is how to make it work. And it is a dilemma that we desperately need to solve. This is an edited extract from Salmon: A Fish, the Earth and the History of a Common Fate, published by Oneworld on 1 October. To order a copy, go to the Guardian Bookshop. Follow the Long Read on Twitter at @gdnlongread, and sign up to the long read weekly email here. This article was amended on 22 September 2020 to correct the spelling of Ben Hatfield to Hadfield, and to include the new name for Marine Harvest Scotland.





5 views0 comments
bottom of page