Something's Fishy

Americans adore fish hatcheries. At their home waters, anglers greet stocking trucks the way kids greet ice cream trucks. At the hatchery itself there's the zoo experience, but beyond that there's the warm feeling of encountering all that rippling energy soon to recharge fresh and salt water, the perception of humankind making nature right again with artificial milk and honey. The Bonneville Hatchery, part of the Bonneville Dam complex in Cascade Locks, Oregon, is even on the National Register of Historic Places. Its 58 concrete raceways (consisting of descending rectangular pools) annually produce about 10 million fall chinook salmon, 400,000 coho salmon, and 250,000 steelhead trout.

On the damp, cold morning of October 30, 2004, the facility is crowded with people as well as fish. Mothers tow snack-munching kids; lovers stroll hand in hand; foreign students photograph one another in front of display tanks. Everywhere there's a low murmur of appreciation. There's a line at the pellet dispenser. Insert a quarter and you get a handful of processed fish chow to throw at adult rainbow trout, kept here for display. They surge upward, making the raceway boil, and when their dark backs cleave the surface, I can see their deformed fins—pectorals and caudals, scarred and rounded by cement, each dorsal nipped to a fleshy stump by the ravenous horde. When I walk along the raceways, the fish follow me like city pigeons. The salmon and steelhead are smaller because they'll be dumped into the Columbia River system as smolts—the juvenile stage that acclimates to salt water, sweeping tail first to the sea. When they go through generating turbines like the ones at Bonneville Dam, they get disoriented and become easy targets for predator fish. In the impoundments there is little current to guide survivors downstream, so they mill around, feeding more predator fish. Trucking and barging them around dams has proven a dismal failure.

Despite its enormous popularity, there's a problem with the Bonneville Hatchery: It doesn't work. Hatcheries, like drugs, can be healing, or hurtful and addictive. They can and do serve as genetic reservoirs, saving species from extinction until habitat is repaired, and they can and do provide fishing in waters where there was never a chance of natural reproduction. But when their mission is to replace habitat degraded by industrial and municipal pollution; siltation from watershed disturbances; loss of streamside shade; and, especially, the construction of dams, they fail.

The Bonneville Hatchery and hundreds of others built to “mitigate” the loss of salmon and steelhead caused by habitat destruction are of the latter sort. Despite their prodigious output—and, to some extent, because of it (although dams are the major culprit)—runs have been declining for more than a century. In the late 1800s Washington State hatcheries produced roughly 4.5 million juvenile chinooks a year; by 1950 they were producing 30 million; by 1968, 93 million. By the 1980s and early 1990s the Columbia alone was being doused with 100 million to 120 million hatchery smolts a year. At that point, wild and hatchery runs in the system had dwindled from perhaps as many as 16 million returning adults to about 2 million. Throughout the Northwest, 159 wild stocks (races adapted to specific rivers) faced extinction. Despite fluctuations caused by varying ocean conditions, runs continue to decline. Even today some hatcheries still measure success not by numbers of adult fish that come back from the sea but by numbers of smolts they stock. As Yogi Berra might say, the bottomless pit just keeps getting deeper.

“A colossal failure of adaptive management” is how Kurt Beardslee, director of the native-fish advocacy group Washington Trout, describes the saturation bombing of Northwest rivers with hatchery fish. “The [state, federal, and tribal] hatchery bureaucracy continues to make the same mistakes over and over and over again, each time expecting a different result,” he says. “That's one definition of insanity.”

Trout and salmon are not turnips, and if they are to survive in the wild, they cannot be produced like turnips. They are sentient beings with learned survival skills and physical adaptations molded by their environment. For example, to migrate from their natal streams to the Gulf of Alaska or, on the Atlantic side, to Greenland, salmon depend on the sharp, full fins they lose in hatcheries. Wild juvenile salmonids are largely subsurface feeders. Reared on pellets, they become surface feeders, vulnerable to predatory fish and birds. When salmonids detect a moving shadow, they need to flee, not rush forward in hope of getting fed.

In order to fill space and maximize growth, traditional hatcheries take eggs and milt from the first fish that show up in the river, thereby selecting for early returning fish, so that runs naturally staggered throughout most of the year are compressed into weeks. When eggs and milt are taken from hatchery brood stock, progeny can be warped by inbreeding. Wild fish compete for mates and spawning sites, so the fittest are favored. Traditional hatcheries mix eggs and milt randomly. Wild fish have evolved to fit conditions in their rivers of origin; if, for example, a river is fast and steep, the native stock may be lean, with noticeably large fins. But traditional hatcheries play musical chairs with stocks, taking them from one watershed and dumping them in another.

Salmonids reared in traditional hatcheries are selected for domesticity. They adapt to crowded conditions because there is no alternative; they learn not to seek cover because there isn't any. In short, they are molded to be everything wild fish are not.

In a 1970 experiment Montana fish managers stopped stocking a section of the Madison River with hatchery trout. Four years later large trout (three years and older) were up 942 percent. How could this be? After all, the more turnip seeds you stick in the ground, the more turnips you harvest. But again, salmonids aren't turnips. Those produced by hatcheries are relatively short-lived, but because they are reared in conditions that make them fight viciously for food, they survive long enough to eat wild juveniles, outcompete wild adults, and often spread diseases in the process. As a result of the Madison River experiment, Montana no longer stocks its rivers.

All this is not exactly breaking news. John Cobb of the old U.S. Bureau of Fisheries complained about “an almost idolatrous faith in the efficacy of artificial culture of fish for replenishing the ravages of man and animals,” and noted that “nothing has done more harm than the prevalence of such an idea.” That was in 1917. Have we learned anything since? Well, some of us have, and some of us haven't.