The way they live, the food they eat, and the effect on us

A typical open-net salmon farm in Bergen, Norway with around 800,000 fish crowded into twelve cages, leaving the fish vulnerable to parasites and pathogens. Waste—excess feed, chemical residue, and fecal matter—is flushed out of the cages by the ocean but settles on the seabed below and forms a layer of slime that smothers marine life.


Excerpted from Salmon Wars: The Dark Underbelly of Our Favorite Fish by Douglas Frantz and Catherine Collins. Copyright © 2022 by Douglas Frantz and Catherine Collins. Published by Henry Holt and Company, an imprint of Macmillan Publishing Group, LLC. All rights reserved. Reprinted with permission.

Consumers have been told that salmon is a healthy and environmentally friendly food. Doctors recommend eating salmon for protein, nutrients, and heart-healthy omega-3 fatty acids. The U.S. Department of Agriculture suggests two servings of fish a week. The salmon’s packaging may show a fish leaping upstream in pristine rivers and boast that it is a certified-natural product: “organic,” “sustainable,” “naturally raised.” There are many salmon choices and little information to help consumers choose from among them, but shouldn’t any fillet you buy be just fine? What’s the problem?

To start with, in almost every case, the salmon in front of you spent its life in a cage and the marketer’s claims are false. “Organic?” There is no USDA-approved definition of organic salmon, so that term is misleading at best. “Sustainable?” When farmed, salmon are carnivores raised on a diet heavy in small wild fish ground into meal and oil, which makes salmon inherently unsustainable. “Naturally raised?” Nothing is natural about feed laced with chemicals and antibiotics or fish swimming in crowded, parasite-plagued cages for two years or longer. Outside the alternative reality of marketing, the slabs of reddish flesh at the seafood counter have nothing to do with pristine waters or muscular salmon navigating upstream and everything to do with the industrialization of food in today’s world.

Now, few Atlantic salmon remain in the wild anywhere. Rivers that once saw tens of thousands of returning fish currently count returns in double or single digits, even though such groups as the Atlantic Salmon Federation and the North Atlantic Salmon Conservation Organization are working against heavy odds to restore salmon and their rivers (see “Releasing Rivers” by Clinton B. Townsend, Natural History, April 2014). Look at the fine print on the label of the Atlantic salmon in your local market and you may see that it is farmed—if you are able to find a label that identifies the origin of the fish. In fact, 90 percent of the salmon consumed by North Americans is farmed Atlantic salmon, raised in feedlots floating on the ocean and flown in from Canada, Scotland, Norway, and Chile; the remaining 10 percent is mostly wild-caught Pacific salmon from Alaska, one of the few places where wild salmon are still fished commercially.

We didn’t always eat farmed salmon; we used to have a choice. There was a time when wild Atlantic salmon was known as the “king of fish.” Ice Age humans painted images of salmon on cave walls in Dordogne, France. Caesar’s legions brought the taste for salmon back to Roman markets. In North America, salmon was a principal food and cultural icon for indigenous people, and it sustained early settlers from Europe. For millennia, millions of Atlantic salmon migrated three thousand miles from the freshwater rivers of what is now the northeastern United States and eastern Canada to the western coast of Greenland, where they fed and matured for years before following Earth’s magnetic fields, their own genetic coding, and a strong sense of smell to return to their precise river of origin to spawn and create new generations. Variations on this journey were repeated across Europe, Scandinavia, and Russia, creating a sustainable food source and an enduring wonder of nature.

So, why do we find ourselves eating unsustainably farmed salmon?

There are many culprits in the king’s demise. Beginning during the Industrial Revolution in the late 1700s, waste was dumped directly into rivers and streams, and the seemingly inexhaustible stocks of salmon began to decline across Europe. By the mid-1800s, numbers were reduced further by commercial fishing and the construction of dams and mills that destroyed habitats and blocked salmon rivers. Within another century, and because of these continued activities, salmon that once numbered in the millions were nearly extinct in Europe and parts of Scandinavia—foreshadowing the disappearance that has left the rivers and streams of New England and Atlantic Canada nearly empty of salmon today. In recent decades, the climate crisis has warmed the oceans and rivers, industrial and municipal pollution has poisoned waterways, deforestation and chemicals have spoiled habitats, and intensive overfishing has decimated wild populations. And in the past forty years, a new threat emerged in the form of industrial-scale salmon farms in fragile coastal regions along salmon migration routes. The primary means of farming salmon is in large cages suspended in the ocean, known as open-net farms. Once seen as a means of taking pressure off overfished wild salmon, these farms turned out to pose a new, man-made danger.

An underwater view of the crowded conditions of open-net cages at a salmon farm on the west coast of Ireland


As wild Atlantic salmon disappear, these floating feed-lots have made salmon one of the world’s most popular and inexpensive fish and have created a $20 billion global industry. In Asia, North America, the United Kingdom, and Europe, what was once a luxury in restaurants or reserved for special occasions at home is eaten at millions of meals a day; a decade ago, salmon replaced tuna as the most popular fish in the American diet, second only to shrimp in seafood consumption.

But availability and cheapness come at great cost. In the industrial-scale farms in coves and bays off the coasts of Norway, Scotland, Chile, and Canada, the only barrier between the cages that harbor millions of salmon and the environment is a net that allows the ocean to flush the pens. Excess feed, chemical residue, and fecal matter form a layer of slime on the seabed below the farms, smothering marine life and plants. Parasites and pathogens proliferate in the crowded cages and spread disease to wild fish. Hundreds of thousands of farmed fish escape each year, competing with wild salmon for habitat and food and interbreeding to produce hybrid fish too weak to survive.

According to a study in 2003 by a coalition of advocacy groups known as the Pure Salmon Campaign, the average salmon farm contains around 800,000 fish in twelve cages and produces fecal matter roughly equivalent to a city of 65,000 people. A 1989 study of Norwegian salmon farms estimated that the organic waste in the fjords that is generated by the production of 150,000 tons of fish equaled 60 percent of the waste generated by Norway’s total population of 4.7 million people; today, Norway produces more than six times as much farmed salmon, and a lot more waste. Sewage and other waste cause far-reaching dam-age to the environment, contaminating the seabed and nearby marine life. A city must treat its sewage, but the farms dump the excrement and excess feed on the seabed. Waste beneath farms turns the ocean floor toxic, consuming oxygen needed by marine life and dispersing contaminants through the water. A 2014 study in Scotland found a reduction in biodiversity up to two hundred yards away from salmon cages; other studies described wider impacts on marine life and wild salmon.

Just as agribusiness turned to hyperintensive farming of cattle, chickens, and pigs on land, salmon farming exploded from small operations to industrial-scale feedlots on water. This unchecked expansion occurred because regulation has remained weak.

Humans aren’t the only ones being harmed by the status quo. Farmed salmon face their own staggering health risks from disease, parasites, and predators. In Norway, the government reported that 52 million fish died before harvest in 2020; the previous year, the figure was 53 million. In Scotland, the mortality rate for farmed salmon quadrupled between 2002 and 2019, according to government figures. In 2019 in Newfoundland, more salmon died in cages than were harvested. Norway and Scotland are the only salmon-farming countries that release mortality statistics, but estimates are that between 15 and 20 percent of all farmed salmon worldwide die each year before they can be harvested. By comparison, the U.S. National Chicken Council reports that its average mortality rate is 5 percent, and cattle feedlots average 3.3 percent.

Getting farmed salmon to the world’s tables does not have to be this way. There are innovators raising Atlantic salmon in land-based, closed-containment facilities where chemicals and antibiotics are unnecessary and where there is no threat to wild salmon. Studies show that consumers are willing to pay a premium for products raised or manufactured without damaging the environment or endangering their health. Land-based salmon farmers are trying to leverage that sentiment into a market for a more environmentally friendly and healthier product. Challenges remain for this disruptive new technology, but it offers hope for the future of the industry.


Aquaculture farmer holding pellets of fish feed made from wild-caught fish and other ingredients. Approximately four to five tons of whole fish are required to produce each ton of fish meal.  


The aquaculture industry feeds billions of farmed fish every day, fueling the global demand for small fish, such as anchovies, sardinella, and herring, to provide fish meal and oil. Nutritionists who design feed for fish have long relied on fish meal and fish oil because they contain an almost perfect balance of the forty or so essential nutrients that salmon and other fish need to be healthy and grow. In processing plants, forage fish are cooked, pressed, dried, and ground to make fish meal; oil is removed in the pressing stage, and excess water is discharged. The meal and oil are then mixed with other ingredients to form dry pellets for salmon feed. Studies show that approximately four to five tons of whole fish are required to produce each ton of fish meal. Based on that formula, the six million tons of fish meal produced annually for aquaculture requires 24–30 million tons of fish, taking protein from people in lower-wealth countries to feed people in wealthier countries and disrupting the marine food chain by depriving larger fish, seabirds, and marine mammals of staples in their diet.

Some types of farmed fish, such as tilapia and grass carp, are herbivores and rely on 100 percent vegetarian feed from crops and other agricultural by-products. However, Atlantic salmon are carnivores and require animal protein. As demand for farmed salmon has grown, the price of wild-caught fish meal and oil has more than doubled in the past twenty years. Economics, rather than environmental concerns, is now pushing feed companies to explore alternative protein sources. Major feed companies have steadily reduced the percentage of wild-caught fish in feed in recent years, partly by expanding the use of plant-based ingredients. Two decades ago, as much as 90 percent of salmon feed comprised marine ingredients; today the percentage hovers between 25 and 30 percent, depending on the manufacturer, and efforts are underway to develop feed that is free of marine ingredients.

Yet, even as feed companies try to reduce the amount of wild-caught ingredients in feed, the rising demand for salmon continues the pressure on forage fisheries. The feed industry is aware of the criticism of its use of wild caught marine ingredients, but its representatives see demand increasing in response to the need to feed the world’s growing population. The Food and Agriculture Organization of the United Nations predicts a need for 20 million tons of additional seafood before 2030. “This means that there’s a need for an additional 25–30 million tons of additional feed ingredients in this decade,” according to Petter Martin Johannessen, director general of the London-based International Fishmeal and Fish Oil Organisation, a private group that represents the marine feed industry. Johannessen and other industry participants in a 2021 webinar acknowledged the increasing public concern over wild-caught ingredients and advocated developing more sustainable marine ingredients.

Aquaculture’s early efforts to develop alternatives to marine ingredients in feed focused on soybeans, corn, and other grains, but problems emerged in the health of salmon that could translate into problems for the people who eat them. Farmed salmon get their omega-3 fatty acids from oily forage fish; the more oily, small fish in the diet, the higher the level of omega-3, a major health advantage for salmon. Oils from soy and similar plants do not provide omega-3 fatty acids, and levels of omega-3 in farmed salmon have declined as the percentage of plant protein and oils in feed has increased in the past decade, reducing the health benefits of salmon for humans.

Research on alternatives for sources of protein and omega-3 is being pursued from university labs and start-upcompanies to a repurposed sugar-processing plant in England and rice paddies in California. Much of the work has focused on insect larvae, oil from algae, and by-products from land animals, such as chicken meal, feather meal, poultry oil, and blood meal. Trimmings and bones from fish processing are also ground into meal and substituted for wild-caught fish.

Cargill, the Wayzata, MN-based agribusiness giant, is a major supplier of feed to salmon farmers. The company has been experimenting with feed formulas that use soy and insect larvae as substitutes for wild-caught fish. Early results reduced the wild-caught content in salmon feed to 27.6 percent in 2018 compared with 31.7 percent in 2017. Cargill expanded its use of fish trimmings and animal by-products, which would otherwise be wasted. But the company said “poor consumer perception” limited the use of by-products. The shift to both plant-based proteins and animal by-products may lead to another problem. Some researchers have found evidence that linking seafood production to terrestrial agriculture poses health risks to consumers through indirect exposure to air, water, and soil contaminated by chemicals used in industrial farming.

At Norway-based Skretting, another large aquaculture feed producer, researchers developed salmon feed that contains no fish meal or fish oil. The feed product, named “Infinity,” relies on algae oil to provide the omega-3 fatty acids that make salmon healthy for consumers. Skretting researchers focused on a suite of novel ingredients—algae oil, insects, yeast, and food processing by-products—to lower the environmental and social footprint compared to using wild fish to create feed. “If we are to increase food production sustainably and ensure that the rising world population has access to essential food, it is important that we seek to reduce the amount of human resources that are used in fish and shrimp diets,” said Sophie Noonan, Skretting’s global communications manager.

Fish nutritionist Frederick “Rick” Barrows spent fourteen years with the U.S. Department of Agriculture, experi-menting with replacements for wild-caught fish in salmon and trout feed. After retiring in 2016, he founded Aquatic Feed Technologies in Bozeman, Montana to continue his research on such protein substitutes as larvae, algae, and bacteria. His science-based reasoning is straightforward: “Fish do not require fish meal. They require the nutrients that fish meal happens to contain. That is why fish meal has been used so much in aquaculture. If you take the fish meal out, you must supplement with other ingredients to get the necessary nutrients, so you need other protein sources.”

Manufacturing feed that uses alternative protein sources is more demanding, requiring more ingredients and pre-cise formulas. But the result can be a sustainable feed that reduces the demand for forage fish at the same time that it grows healthy salmon. Fish nutritionists are testing dried black soldier fly larvae as an alternative to fish meal. The people working with the soldier fly larvae point out that salmon in the wild eat insects. Single-cell proteins, such as mold and yeast, have also been tried, with mixed results. Algae, a diverse group of aquatic organisms at the bottom of the marine food chain, have shown promise as an alter-native source of the omega-3 fatty acids found in fish oil. A study in the journal Aquaculture found that replacing some fish meal and fish oil with algae meal boosted weight gain and maintained fatty acid levels in striped bass, a carnivorous fish like salmon.

Since 2016, Barrows has also been chief science officer at F3 Future of Fish Feed, a collaboration between NGOs, academic researchers, governments, and the private sector. The goal is to encourage feed companies to develop replacements for wild-caught fish globally. Central to the organization’s effort is the F3 Challenge, a series of three contests to produce alternatives that the industry will buy and use. The first contest challenged feed companies to develop and sell seafood-free feed for freshwater fish that are not carnivorous, such as tilapia and carp, using innovative proteins. The winner of the $200,000 prize was China-based Guangdong Evergreen Feed Industry Company, which sold 84,000 metric tons of a new feed that used only soy bean meal, peanut meal, and rapeseed meal. F3 estimated that Guangdong’s innovative feed saved 350 million forage fish.

The second contest required developing fish oil replacements that mimicked the fatty acid profile of forage fish. This $200,000 prize went to a Netherlands-based joint venture, Veramaris, which produced an algae-rich oil at pilot plants in Slovakia and Nebraska. The company said its algal oil contained twice as much omega-3 as fish oil and could help reverse the decline of those essential fatty acids in farmed salmon, a decrease directly linked to diets more reliant on plants. F3 estimated that the 850 metric tons of algal oil used spared the equivalent of 2 billion forage fish. Veramaris worked closely with Norwegian salmon-farming companies; the largest, Mowi, said it would test the algal oil in its feed. Anette Zimowski at the Norwegian Seafood Council said companies are looking for alternatives, such as algal oil and insects, but she said vegetable ingredients, such as soy meal, currently make up 60 to 70 percent of aquaculture feed.
The third contest posed the greatest difficulty—to develop substitute feeds for aquaculture’s greatest consumers of forage fish: salmon, shrimp, and other carnivorous species. The goal was to accelerate the development and adoption of alternatives to fish meal and fish oil, with a $100,000 prize for the winner in each of the three categories of carnivorous fish. While the rules did not specify ingredients, the feeds were expected to diverge from the heavy reliance on marine ingredients in favor of the novel proteins being developed. Contest results are expected in the fall of 2022.

The F3 Challenge was designed to stimulate and reward innovations that could lead to large-scale reductions in the use of forage fish. But even prizes worth hundreds of thousands of dollars are a drop in the ocean for the $20 billion salmon-farming industry. Economics will drive the reduction of fish meal and oil in aquaculture feed. If the price of fish meal continues to go up as fisheries continue to be depleted, and the price of alternative proteins goes down, the feed profile will change, and farmed fish might ultimately have a sustainable food that would allow forage fisheries and the people who depend on them to recover. Skretting’s Noonan said novel ingredients offer a path toward sustainability, but they must be scalable and cost-effective, the default position for an industry focused on the bottom line.   --DF, CC

view counter
view counter

Recent Stories

The way they live, the food they eat, and the effect on us

A true but unlikely tale

Story and Photographs by William Rowan

Increasing day length on the early Earth boosted oxygen released by photosynthetic cyanobacteria.

Genomic evidence shows that Denisovans and modern humans may have overlapped in Wallacea.