About 10 years ago my friend Charles Johnson, the Vermont state naturalist, called me with a question about salmon. I was surprised and flattered to have Charles ask me about anything natural, for Charles holds an encyclopedic knowledge about the natural world and a deep but nonpedantic environmental ethic. Usually I call him. When I was writing a book about New England’s forests, I called Charles, and he made bog ecology, which is as complicated as calculus, seem as plain as pancakes. And when a woman I’d fallen in love with told me she had always wanted to see moose, I called Charles and asked him where to find moose. “Victory Bog,” he said. At Victory Bog my love and I found moose, and three years later we married.
Now Charles was calling me. “Got a question,” he said. “We’ve got some people over for dinner”—from the background came a rowdy banter—“and we were having this discussion. We’re wondering is it OK to eat salmon?”
This explained the call. As Charles knew, I am an avid salmon angler, and I had just written a book called The Great Gulf, about decimated ocean fisheries, and several articles about salmon. This made me a sort of salmon-expert-for-the-day. Charles, meanwhile, was as confused as most eco-conscious people are about the shifting fates and statuses of the world’s saltwater fish.
“How do you mean ‘OK’?” I asked.
“Well, you know—ecologically,” said Charles. “Are salmon fisheries sustainable?”
“Depends,” I said. “What kind of salmon?”
“Atlantic.”
“I was afraid of that,” I said. “Atlantics are lovely fish. But any Atlantic salmon you buy in a store came from a farm.”
“Oh,” said Charles. “Is farmed salmon not so good?”
“Did you already eat this fish?” I asked.
He laughed. “That bad, huh?”
A little while later, when I had explained it all, he said, “I’m embarrassed that I didn’t know this.”
I assured him—this was 10 years ago—that few people did. “Probably even most state naturalists don’t know this.”
“Maybe,” he said. “I better go. Dessert’s ready. And I gotta hide a salmon carcass.
“But do me a favor, will you?” he asked. “Don’t tell anybody about this.”
The “Is It Ok?” Algorithm
Now, 10 years later, many people, both foodies and greenies, still struggle to answer Charles Johnson’s question: Is it OK to eat farmed salmon? If the answer seems elusive, it’s partly because things change—ecosystems change, the fishing and aquaculture industries change (or not), and we learn more about how those industries affect fish populations, ecology and economies. But the answer is slippery mainly because it depends so heavily on what the meaning of the word “OK” is. OK is personal. The most satisfying answer comes from what you might call your own Is-That-Food-OK Algorithm—a weighting of variables that will be as simple or complex as the criteria you bring to it.
For some, taste trumps all. Others give weight to price or health or local economic or environmental impact. There are plenty of good reasons to eat salmon: It tastes good. It’s easy, fast and aesthetically pleasing to prepare. It’s incredibly healthy; no common fish delivers more of the omega-3 fatty acids that help keep arteries clear and hearts strong. The past decade has shown that these fatty acids may also strengthen the immune system and eyesight, and even improve mental health. These pluses have helped inspire Americans to more than triple their consumption of fresh and frozen salmon in the last 15 years, from 50,000 metric tons in 1990 to 180,000 in 2004.# The only fish we eat more of are shrimp and canned tuna.
But the main reason we’re eating more salmon is because a burgeoning worldwide salmon-farming industry has almost quadrupled the supply of salmon in the last two decades, making farmed salmon obtainable almost anywhere, anytime, for under $8 a pound. These farmed salmon make up about 80 percent of the huge increase in U.S. consumption since the late ’80s. The “Is Farmed Salmon OK?” question therefore relies at least partly on the impact of that expanding salmon aquaculture industry.
How does one sort this out? We’re talking about food here, and values, so I won’t presume to tell you how to weigh things. But I do know that the variables in the salmon equation have changed since Charles Johnson phoned me a decade ago. In particular, we know a lot more now than we did then about salmon’s health benefits—and a lot more about how salmon farms affect the environment and wild fisheries. Time to recalibrate the algorithm.
Life Stories
The algorithm starts with the fish’s rhythm—the arc of its life. Imagine before you a salmon steak freshly plated: orange, warm, aromatic. You ordered wild tonight, so your fish is one of the five species of Pacific salmons, genus Oncorhynchus; it must be Pacific because the fishery for wild Atlantics (Salmo salar, a different genus altogether) is long kaput. (More on that shortly.)
The Pacifics most commonly caught—chum, coho and sockeye, chinooks and pinks (the pinks mostly for canning)—occupy huge ranges. Chum and sockeye run up rivers and off coasts from California to Kyoto. Cohos run thick from British Columbia up through Alaska. And the magnificent chinook, three feet long and sandbag heavy—also known as king, tyee, blackmouth—swim sparse in San Francisco, profusely in the Columbia River and Puget Sound, and in Alaska run abundant, climbing as far as 2,000 miles up the Yukon to spawn. All Pacific salmon are anadramous, spending their youngest days in rivers and their youths and adult lives at sea. When it comes time to spawn, most Pacific Northwest salmon will do so only in their natal streams, making each watershed’s population distinct and vulnerable to extinction if a dam or other insult wipes them out. In the Pacific Northwest and western Canada, where dams have knocked down some populations, the fisheries are faring poorly. In Alaska, the wild-salmon fishery is one of the world’s best-managed and populations are robust.
Let’s say you ordered an Alaskan king. Imagine, on an autumn day a few years go, this fish’s mother, three feet long, greenish with a red snout and almost brutally muscular, surging up the Yukon River. She swims for 60 days up the Yukon, then finds a gravelly spot, swishes out with her tail a shallow depression (called a redd), and with a splash and a thrash lays as many as ten thousand orange-red roe. Presently an attending male, perhaps several attending males, thrash and splash too, releasing skeins of milt over the roe.
The eggs hatch three to five months later. The fry, louse-sized at first, stick close to the bottom of the river. Their mottled stripes make them all but invisible from above, so that you can look right at them (as I have many times) and not see them till you’ve spooked them. Leisurely, the fry gobble larvae and small crustacea on the bottom; sometimes they flash to the surface, cobra-quick, to grab a bug or an emerging nymph. Most are themselves gobbled.
Yours is one of the few—1 in 500, 1 in 1,000—that makes it. He lives a year or two, grows large enough to move downstream. After a couple weeks he reaches a brackish estuary where he remains for a few days acclimating to saltwater. Then he enters the North Pacific. He eats with an open mind—other fish, mollusks, and lots and lots of krill and other planktonic crustacea that have feasted on red algae. This diet turns his flesh pink and rich in omega-3 fatty acids. When he is six or seven years old—as young as two in some species—he will need this fat to fuel his swim upriver to spawn. Once he enters the river he will not eat, and he may have to swim for days. Some fish swim only a dozen or a few score miles to spawn; others will swim hundreds. Alaska’s Copper River salmon, an exceptionally fatty and tasty fish, swims only 300 miles, but in doing so climbs 4,000 vertical feet. Your fish might swim 1,200 miles before it finds its spawning site, wiggles over its redd, and then dies. Or, it would have had it not been caught off the coast a few days ago, iced, boxed and shipped to the restaurant where it now greets you.
But perhaps you ordered a farmed salmon—that is to say, an Atlantic species. Virtually all Atlantic salmon sold in North America come from farms. They could hardly do otherwise, for overfishing and the damming and pollution of rivers in Europe and North America have destroyed the once-great wild fishery for Salmo salar, native to the North Atlantic and all its shores. Wild Atlantic salmon are now on the Endangered Species List. Because they grow faster than most of the Pacific species and better tolerate the crowding of the net pens, farmed Atlantics account for about 85 percent of all farmed salmon worldwide. (Some farms raise Pacific chinooks because they sell for upward of $20 a pound.)
The fish before you, then, represents the union of egg and sperm surgically extracted, quite possibly at different facilities, and shipped separately to its birthplace, which is likely a hatchery and fish farm in British Columbia or Washington, or possibly Scotland or Japan, or perhaps Chile or New Brunswick or New Zealand.
No matter where it’s hatched and raised, this salmon is still an Atlantic salmon. Once fertilized, the egg is incubated and hatched. The fry spends a year or so in freshwater tanks gradually made salty. Then your fish, some 8 to 10 inches long, is vaccinated against an array of infectious diseases, which are a constant threat in the crowded net pen into which the fish is now placed. The pen—a bowl-shaped net, essentially, with its rim at sea level—is about 30 to 100 meters across. It holds tens of thousands of fish. Instead of roaming the seas and swimming hundreds of miles to spawn, these fish never travel more than a few yards. Here in its pen your salmon fattens up on fish pellets. The pellets contain fish meal, fish oil, perhaps grains, and invariably an additive to pinken the fish’s flesh—usually astaxanthin, a carotenoid derived from commercially grown red yeast or algae. Given bountiful food and little work, your fish grows apace. When it’s
a couple of years old and a couple of feet long and weighs 8 or 10 pounds it is netted, killed and packaged, iced and shipped.
Weighing the Many Variables
From these life histories devolve all variables relevant to the “OK?” question. How much does the fish cost? How easily can I get it? How do farmed and wild fish differ in taste, chemical load and nutrition value? What is the fish’s relation to local, regional and global environments and economies?
Over the last 20 years these questions have been asked in many different forums—papers in scholarly journals like Aquaculture and Fisheries Research; government reports; articles in Audubon, The Economist and EatingWell; at fish markets and dinner parties; in puzzled phone calls. Some of the answers change little. For instance, wild Atlantic salmon populations, despite millions spent on dam removal and habitat improvement, remain deeply in trouble. Every year only a few hundred return to the New England rivers that once had millions, and in Canada thousands swim where tens of millions once teemed. Pacific species are doing better, and in fact are increasing in some areas. They’re doing splendidly in Alaska, with record catches and population estimates.
As for taste: flavor varies more in wild fish than in farmed, for wild Pacific fish differ more in their life histories and physiology and fat content than do Atlantics, which are genetically more alike and which all eat similar diets heavy on fish oil. Young Pacific chum, for instance, generally aren’t as fatty and tasty as Atlantics, while sockeye and chinooks with long or arduous spawning runs offer extraordinary flavor.
The markets, meanwhile, remain steady. For nearly a decade, wild Pacific salmon, for instance, have cost about two to three times the $6 or $7 a pound that Atlantic costs. You can pay well north of $20 for wild Pacific, too, if you want your salmon overnighted from Alaska.
Availability is similarly plain. While you can get Atlantics anywhere year-round, wild fish availability is more exacting. For starters, wild salmon are seasonal, with the huge majority of fresh fish available between May and October, when fishermen can capitalize on the predictable concentration of fish as they move toward the coast to spawn. A few online sites offer it later in the season, and many offer frozen throughout the winter months. But in general, eating a fresh wild salmon means eating it in summer or fall.
In most big stores you can confirm the fish is wild—and where it’s from— from the country-of-origin labels (or COOL) now required for all fish. The label, either on the package or on hand for whole fish, will tell you its origin and whether the fish is farmed or wild.
Small fish and specialty shops, however, aren’t required to carry COOL labels, so in those cases it’s caveat emptor. A depressing New York Times story in 2005 found that for six of the eight fish and specialty shops visited, salmon sold as wild were actually farmed, according to a lab that the Times paid to analyze the fish for tell-tale levels of the commercially grown carotenoids added to pinken the meat. Such substitution seems most likely during the off-season, when “wild” salmon, though rare on fishing boats, seems uncannily abundant in stores. Then, in particular, insist on a label or ask your local shop specifically where the “wild” fish came from and phone the supposed source to confirm.
Though I’m chemophobic, I’m ready to declare the health concerns about farmed salmon too insignificant to fret about, or at least too close to fight about. Farmed salmon hold about 2 to 10 times the levels of PCBs, DDT, dioxins, pesticides, mercury and other suspected carcinogens that most wild salmon do, apparently because the rich meal they eat contains bits of oily fish in which these contaminants tend to concentrate. Yet the levels of those chemicals are still so low, as Harvard Medical School nutrition specialist George Blackburn, M.D., Ph.D., noted in a recent article in the Journal of the American Medical Association, “that it’s not going to cause harm.” Meanwhile, numerous studies have shown that omega-3 fatty acids reduce heart attack risk; in one study, consuming 700 mg of omega-3 fatty acids a day (the equivalent of eating 10 ounces of salmon once a week) cut mortality in Italians with coronary artery disease by 20 percent.
Keep in mind, the generally higher levels of mercury often found in farmed salmon should swing the scale toward wild fish if you’re pregnant, nursing, young or particularly brain-protective. That aside, however, the tremendous omega-3 benefits—reduced heart attack risk, better immunological, neurological and even psychological health—easily outweigh the tenuous risks.
The Bigger Picture
Farmed salmon’s environmental effects are less flattering. I wish it were otherwise. I love the idea of sustainable farm-raised fish. That’s why I cook a lot of catfish—it’s splendid just oiled and floured, better yet grilled with some Southwestern or Asian seasoning, and my pleasure is enhanced because I know it is raised in sustainable closed systems that emit little pollution.
The news on salmon farms, however, gets worse by the year. The industry’s expansion since 1980 has put hundreds of salmon farms on cold-water coasts in Europe, Asia and North and South America, raising hundreds of millions of fish. About 85 percent of these are Atlantic salmon. A decade ago, we knew that Atlantics sometimes escaped from pens, but it was not clear how many escaped or whether they messed with wild salmon or wild-salmon habitat. Now we know that millions escape, that they disrupt feeding and spawning behavior of wild salmon in both the Atlantic and Pacific oceans, that they enter and even colonize streams, and that they directly compete with both native Atlantic and Pacific salmon. Though they don’t crossbreed with Pacifics, they can disrupt their spawning. Confirmed too are suspicions that farms spread disease and parasites to wild salmon stocks. All this adds to the tremendous pressure already on wild salmon stocks from habitat loss and overfishing.
This, then is the part of the formula that has changed for the worse: the environmental variables associated with farmed salmon look far less attractive than they did a decade ago, even as the industry has almost tripled in size.
An emphatic illustration of this emerged as I was writing this article. This one involved sea lice. Sea lice routinely parasitize wild adult salmon without killing them. But because the lice can’t live in fresh water, they die as the wild fish move upstream to spawn, leaving juvenile salmon unmolested during their early growth periods in the stream. Even after they enter the sea, young salmon usually run into sea lice only occasionally, at exposures they can handle.
But sea lice infest salmon farms heavily, thriving in the crowded conditions. We knew this 20 years ago, but were told they wouldn’t spread to wild populations. By 10 years ago research had proven that the lice did spread to wild populations, and by five years ago we knew that high concentrations of lice from fish farms were killing some wild fish.
But did they kill enough salmon to harm whole populations? The industry said “No.” But a paper in the journal Science published last December demonstrated convincingly that sea lice from salmon farms in British Columbia’s Broughton archipelago, a collection of islands and channels between North Vancouver Island and the mainland, are exterminating entire native runs of pink salmon.
The salmon industry at Broughton mirrors the industry’s worldwide development. Broughton got its first farms in the late 1980s, and by 2005 it had over 20 farms holding millions of Atlantics. These farms were releasing tens of millions of sea lice. Over the past five years these sea lice, floating in great density in the channels in which the salmon farms float, have forced virtually every pink salmon leaving the Broughton archipelago to swim through an infested gauntlet to try to reach the ocean. The wild pink fingerlings, their skin still unscaled, pick up two, three, a half-dozen lice apiece—tolerable to adults, perhaps, but not to these finger-sized smolts. “They are not equipped to survive this,” says Alexandra Morton, director of the Salmon Coast Field Station and a co-author of the paper, “and they don’t.” This infestation is wiping out the wild pink salmon runs whose rivers flow from the mainland into the Broughton archipelago. Analysis of returns and comparisons to nearby rivers without farms show the lice are killing 80 percent of the pinks that run the sea-lice gauntlet. At that rate, and with the pinks’ two-year-long life cycle, the several river populations of pink salmon that must run the Broughton gauntlet could be 99 percent extinct by 2011.
Maybe this is not so bad. This is only one river basin, and the salmon are pinks, which, being low-fat and lacking that salmony taste, mostly get smoked, salted or canned. And the pink-salmon populations in other rivers will be unaffected, at least as long as no one builds salmon farms in their river mouths. Yet the either-or nature of this disaster deeply disturbs me. It could hardly be more clear: we must either close those farms or kiss those runs of wild salmon goodbye.
I’m increasingly convinced that the larger issue of farmed versus wild salmon poses a similar choice. The withering array of injuries that salmon farms inflict on wild salmon forces a sort of long-range consumer decision. This is not like deciding whether you want free-range versus conventional chicken for tonight’s dinner; that’s a decision with limited echo. To decide that you may as well eat farmed Atlantic tonight, however, is to decide, in a very real sense, that you may as well eat farmed salmon, and farmed salmon only, forever. You may feel differently. But that just doesn’t sit well with me. For now, anyway, I’ve eaten my last farmed salmon.
The promising part—there is hope here—is that forsaking farmed salmon for wild might actually press the salmon aquaculture industry to change. The technology exists now to raise farmed salmon in floating, solid-sided tanks—a container more floating aquarium than net—that avoid almost all the drawbacks of net pens. The fish can’t escape; pests can’t enter or escape; water can be filtered and cleaned of waste.
This low-impact solution would realize aquaculture’s true potential, just as good catfish operations do. Companies in British Columbia, Iceland and Norway are testing prototypes. The catch is that even if they work, they’ll be costly, producing fish pricier than other farmed fish (though possibly cheaper than wild), and it’s not clear whether consumers will pay the premium. Yet investors might find the necessary courage if enough consumers stop buying conventionally farmed salmon.
The consumers, of course: that would be you and me. Holding out for lower-impact farmed salmon—and eating only wild salmon in the meantime—is a sacrifice of sorts. It means eating fresh salmon seasonally and paying more for it when you do. It means eating fish that, because they come from various places and live varying lives, will be less consistent in flavor than farmed salmon are; sometimes they’ll taste better, sometimes not. But in the realm of things, that set of trade-offs sounds pretty good to me. I’d even say it sounds OK.
David Dobbs writes on science, culture and the environment for The New York Times Magazine, Audubon and other publications.
READER'S COMMENT: