EVERY year, fisheries biologists in Washington, Oregon and California watch over several hundred million Pacific salmon eggs, rearing the hatchlings in concrete ponds until they are old enough to begin their arduous journey to the sea. This huge effort 鈥� costing tens of millions of dollars a year 鈥� began nearly 60 years ago as a way to restore commercial salmon production lost when the government dammed rivers and diverted water for irrigation projects. Today, well over half the salmon caught in western rivers began life in a hatchery, not a natural stream.
But the hatcheries鈥� success has a dark side. Many biologists fear that through competition and genetic change, hatcheries may deliver the coup de grace to many wild-spawning salmon populations. 鈥淭o me, that鈥檚 a prescription for disaster,鈥� says Robin Waples, who heads the Pacific salmon conservation programme for the US National Marine Fisheries Service (NMFS) in Seattle. 鈥淭hose natural populations really are the foundation of the whole system, and if we don鈥檛 conserve them, we鈥檙e going to be in trouble in our children鈥檚 generation or our grandchildren鈥檚.鈥�
In response to this threat, fisheries biologists have begun to argue that hatcheries should serve not as permanent factories churning out replacements for wild fish, but as short-term bridges to supplement native populations until they can be restored to health. Only a few such 鈥渟upplementation hatcheries鈥� exist so far, and no one knows if they will damage the genetic make-up of wild salmon populations.
Advertisement
But in a seminal move, the NMFS strongly endorsed the concept of supplementation hatcheries last month in a plan to help restore threatened populations of salmon in the Snake River, a tributary of the Columbia River (see Diagram). The report is now open for public comment, but if adopted it will be the first recovery plan for endangered salmon that commits officials to agree a policy for hatcheries within a strict timetable.
鈥淲e have seen a dramatic shift in our objectives,鈥� says Richard Carmichael, a fisheries biologist with the Oregon Department of Fish and Wildlife. 鈥淢aintaining genetic diversity has become a much more important objective in fisheries management. Supplementation hatcheries play a role in achieving that objective. Conventional hatcheries don鈥檛.鈥� The biologists argue that even if rivers can be restored so that they can support large numbers of fish, without native genes salmon would lack the genetic adaptations they need to thrive.
Salmon display a complex homing behaviour, determined partly by genes, that takes them back to their natal streams to spawn. As a result, salmon populations show much tighter genetic tailoring to their local environments than other endangered species such as wolves. Indeed, almost every attempt to transplant wild salmon from one river to another has failed.
So, each river鈥檚 wild salmon represent a unique evolutionary jewel, irreplaceable on any timescale relevant to humans. According to a study carried out in 1991, at least 106 major salmon populations in Washington, Oregon and California have gone extinct since European settlement, and another 214 are now at risk. 鈥淎t this point we can鈥檛 afford to write off any [more] native stocks,鈥� says Willa Nehlsen, a biologist with the Pacific Rivers Council in Portland, who headed the 1991 study.
Most of the 214 at-risk populations are in trouble because logging and other human activities have destroyed spawning and juvenile rearing habitats. Fishing, too, takes a heavy toll. In 1991, for example, the industry inadvertently caught between 60 and 65 per cent of all adults of the threatened chinook salmon that entered the Snake River that autumn. The river鈥檚 fish were vulnerable because they mingled with healthier populations and hatchery fish in commercial fishing areas, says Waples.
But in about half Nehlsen鈥檚 214 cases, hatchery fish contribute more directly to their wild relatives鈥� decline. Young fish dine well at hatcheries, and often live in water that is warmer than that in natural streams. As a result, they tend to be larger than wild fry when released into rivers and outcompete the smaller fish for food and territory, says Reginald Reisenbichler, a population biologist with the National Biological Service in Seattle.
Genetic violence
And the problems do not end there. Hatchery fish do genetic violence to wild populations as well, says Reisenbichler. When most hatchery brood stocks originated several decades ago few biologists bothered to select their fish from local populations. After all, the salmon would be returning to spawn with human assistance in concrete tanks, so what did local adaptations matter?
But every year, large numbers of these hatchery salmon 鈥� by now changed by generations of natural selection for hatchery conditions 鈥� stray further upstream to interbreed with wild salmon. And experiments show that these hybrids are less fit than wild fish.
Fishery managers hope that supplementation hatcheries will eliminate many of these problems. Carmichael manages one of these new-style hatcheries on the Imnaha River, a tributary to the Snake in northeast Oregon. When the hatchery was founded in 1982, biologists took their brood stock from the Imnaha to avoid introducing foreign genes. And by carefully controlling water temperature and feed, Carmichael has learnt to produce hatchery fish that match their wild cousins in size.
Even as his hatchery bolsters the natural population鈥檚 numbers, Carmichael oversees an intricate balancing act to maintain the genetic integrity of the wild fish. Instead of releasing the young fish directly from the hatchery to swim down river each year, he transfers them (with fins clipped for easy recognition later) to holding ponds in the river for a month, giving them 鈥渁 taste of the wild鈥� so that when they return as adults they will home to this spawning ground instead of the hatchery.
When they do return, Carmichael and his colleagues capture each fish at a weir and either keep it for the hatchery or release it upriver to spawn. Every year, at least 30 per cent of the hatchery brood stock comes from adults who began life in the wild, thus knitting wild and hatchery stocks together into a single genetic unit.
To further guard against genetic corruption from the hatchery, the researchers make sure that no more than half the fish they release upriver to spawn are hatchery-reared. Carmichael鈥檚 balancing act has been successful enough that today he can detect no behavioural or genetic differences between his hatchery salmon and the wild population.
鈥淲e know that because of the hatchery programme, we have more fish in the Imnaha basin,鈥� says Carmichael. 鈥淏ut that doesn鈥檛 mean it鈥檚 more successful because success means not just getting more fish back to the basin, but producing more natural fish.鈥�
Creeping change
But the longer such programmes operate, the more likely it is that some genetic change will creep into the wild population, says Waples. Whether such change will be trivial or critical remains to be seen. 鈥淚t will take us a number more years to get the kind of data we need to answer that question,鈥� says Carmichael. The five-year life cycle of chinook salmon means that he has seen only one or two complete generations, and other supplementation hatcheries 鈥� a few dozen at most 鈥� are still in their infancy. No one has yet run a supplementation hatchery long enough to try the ultimate test 鈥� weaning a native population off its dependence on the hatchery to become a healthy, self sustaining salmon population.
Because of the uncertainties, most experts advocate a go-slow approach. 鈥淚f conservation is our concern,鈥� says Reisenbichler, 鈥渦ntil we get answers to these questions, we should recognise a risk [from supplementation] and protect a substantial proportion of our populations from that risk. If we want the benefits of supplementation 鈥� which is usually done so people can [catch] more fish 鈥� let鈥檚 protect 20 or 30 per cent of our wild populations from that risk. Indeed, the NMFS鈥檚 policy so far has been to treat supplementation hatcheries as experimental.
The new programmes are also proving expensive. 鈥淚t鈥檚 very manpower-intensive to maintain these supplementation hatcheries, because we deal with fish on an individual basis,鈥� says Carmichael. 鈥淚 think that鈥檚 going to be a limiting factor.鈥� The ponds for holding juvenile fish before release and weirs to capture the adults can cost several hundred thousand dollars apiece. This may be cheap for an experimental programme, but the cost would be horrendous if applied to every troubled fish population along the coast.
An even bigger cost, however, arises from the need to restrict the size of the hatchery to keep it in balance with the wild-spawning population. Carmichael鈥檚 hatchery, for example, set an original goal of releasing 490 000 young chinook salmon a year to compensate for losses from dam construction on the Snake River. But in practice, the hatchery has rarely released more than half that number because the wild population has been too small to provide enough hatchery brood stock.
Shortfalls like this have convinced many fisheries managers, Carmichael included, that the new hatchery concept will never replace conventional hatcheries completely. People who fish for fun or money expect there to be plenty of fish 鈥� even if they are genetically foreign. Hatcheries on the Columbia River alone contribute tens of millions of dollars a year to fishing industry earnings.
Paradoxically, however, the fishing industry is one of the strongest advocates of supplementation hatcheries. 鈥淚n the short run we may take a loss, but in the long run it鈥檚 a gain because you don鈥檛 crash the system by overrunning the wild stock with poorly adapted hatchery stock,鈥� says Glen Spain of the Pacific Coast Federation of Fishermen鈥檚 Associations, the largest organisation of commercial fishermen on the West Coast. The federation has even helped to develop a supplementation hatchery in California鈥檚 Trinity River.
But changes in hatcheries alone cannot save wild salmon populations. The Snake River recovery plan suggests a long list of measures to improve conditions for salmon, from changing the way dams are operated to altered logging practices. Experts agree that wild fish populations will never return to full health until the US deals with these issues. And that will prove a much harder and more costly effort than any changes in hatchery regimes.