Scientists Say Real Salmon Recovery Depends on Improving Freshwater Habitat

Scientists say the federal government's newly announced draft "Hatchery Listing Policy" opens the door to political manipulation of decisions about listing and delisting of salmon under the Endangered Species Act. The policy requires the inclusion of hatchery fish in status reviews and listing determinations, even while scientific evidence mounts that hatcheries are ineffective at restoring wild salmon and in fact can be a significant detriment to recovery.

"The Endangered Species Act requires that listing of species be determined by the "best available scientific information," said Chris Frissell, salmon research ecologist and Senior Staff Scientist with the Pacific Rivers Council. "This new policy ignores a large body of science and a century of management experience affirming that wild salmon recovery will depend on how we manage habitat and fishing, not hatchery fish."

"The science tells us clearly that it's habitat restoration and the health of wild salmon bred in streams and rivers that will determine salmon recovery," said Frissell. "Counting fish that originated from hatcheries is an ecological delusion - it has always masked what is really happening to wild salmon populations and the rivers they depend on to survive."

Lawyers working with the Pacific Rivers Council on the hatchery policy issue believe the federal government's interpretation of recent federal court orders about hatchery fish in listing decisions is incorrect, and is a thinly veiled attempt to politicize the species listing decisions under the Endangered Species Act. "We now know that any apparent gains provided by hatcheries come at the expense of wild fish, and do nothing to improve the survival of salmon in rivers, coastal, and ocean waters," said Frissell, PRC's Senior Scientist. "The overwhelming weight of scientific evidence is that hatchery fish contribute to the decline of wild stocks."

Recent research led by independent scientists Gretchen Oosterhout and Chuck Huntington indicates that Oregon coastal coho salmon populations are unlikely ever to be restored by fish hatcheries, even if they were "reformed" as state-of-the- art "conservation" hatcheries. The new study underscores that recovery of these dwindling salmon runs depends on the protection and restoration of freshwater habitat.  The research highlights that even though survival of hatchery fish in the natural environment may be lower than that of wild salmon, those hatchery fish that manage to survive can displace wild fish from the ecosystem. Even under the best of circumstances, supplementing natural salmon populations with hatchery fish is unlikely to restore wild salmon.

"Improvements" in hatchery management intended to increase survival of hatchery-bred fish only intensify the basic problem," said Frissell, a member of the Oosterhout study's scientific advisory panel, because all fish are competing for the small fraction of freshwater and coastal habitat where survival is good. "The irony is that any increased survival of hatchery fish will likely push more wild fish out of the limited habitat available."

"Our conclusion is that even the best hatcheries we can design pose a clear risk of harm, but promise little or no benefit to populations of salmon breeding in the wild," said Dr. Oosterhout. "This finding is justified based on a comprehensive literature review and bolstered by computer modeling using extensive field data for coho populations, but the same conclusion is likely valid for salmon conservation everywhere."

As lead petitioner in the coast-wide coho petition initiated over ten years ago, PRC supports continued federal protection on the basis that coastal coho populations are not sufficiently recovered, and long- term trends in habitat deterioration have not reversed. While it supports continued state leadership in recovery efforts, PRC finds that de-listing discussions are premature - in part because Oregon has inadequately funded high-priority restoration activities, shifted resources to hatchery projects of dubious merit, and made little or no progress to reduce logging-related harm on state and private lands.

Scientists working on recovery admonish against declaring coho salmon recovered based on the recent salmon boom. As Huntington, a consulting biologist, explains: "Even our most recent, relatively high coho numbers along the Oregon Coast remain at a small fraction of historic abundance.  Furthermore, it simply isn't appropriate to measure the status of a population based only on production during good ocean years like we've had very recently. The persistence of individual populations of Oregon Coast coho will depend on how well the fish can perform in the freshwater habitat available to them during the next extended period of poor ocean conditions. The ability of the coho populations along the Oregon coast to survive cycles of such conditions is the true test, and will depend on the presence of high-quality freshwater habitat."

Available data indicate that the productivity of currently suitable freshwater habitat may be near its limit. "Within areas that have been monitored, smolt production of Oregon Coast coho has increased in response to the recent increase in adult escapement, but appears now to be nearing the capacity of existing freshwater habitat," said Huntington.

"The problem is that hatchery fish are essentially a political commodity," said Frissell. "If governments believe they can buy their way out of real salmon recovery and avoid the hard job of habitat protection with more or "better" hatchery fish, they will do that. At its core, the new hatchery policy embraces the same scientifically bankrupt management delusions that led to the decimation of salmon runs on both coasts of North America."

QUESTIONS AND ANSWERS ABOUT HATCHERIES AND SALMON RECOVERY

• What was the Oosterhout and Huntington hatchery-wild study about?
  
  

  The study's purpose was to answer the question, Under what circumstances could hatchery fish stocking contribute to the recovery or viability of natural populations of Oregon Coast Coho salmon? Independent researchers adapted an existing life-cycle model for coho salmon to enable computer simulation of interactions between wild and hatchery fish in a river basin. The new model examined the potential effects of proposed state-of-the-art "conservation" hatchery designs intended to reduce negative effects on wild fish while boosting failing returns of salmon. Nicknamed "CoRAS," the model is believed to be the first to simulate the population dynamics between Oregon coastal hatchery and wild coho salmon.

• What did the study find?
   

  The study found that hatchery fish pose a risk of harm but little or no wild population benefit.

  Simulated supplementation of a Coho population on the Oregon Coast -- even by an ideally-conceived-and-operated conservation hatchery -- rarely showed a net benefit to salmon abundance, and in those few instances that it did, benefits were marginal and short-lived (i.e., a few years). The results suggest that freshwater and near-shore ocean conditions cap the survival of both wild and hatchery-released salmon. Because wild fish tend to do well in freshwater habitat wherever it is of high enough quality, adding hatchery fish to the freshwater system can push out some wild fish, replacing wild fish with hatchery-origin fish that may pose a risk of compromising the genetic integrity of the wild population. Neither wild nor hatchery fish survive and reproduce well in poor-quality habitat.

  The study was focused on Oregon Coast Coho salmon as an example. The extent to which the results directly translate to other salmon species or regions remains to be investigated. However, the basic tenets and conclusions of the study are clear and unambiguous, and they likely are broadly applicable to salmon conservation everywhere.

• What does this study mean for public policy?
  
  

  One clear implication of the study is that improvements in hatchery fish survival, much sought-after by management agencies in recent years, come at a cost. The more hatchery fish that survive after release to reproduce in the natural environment, the more they may act to displace wild fish, so that any short-term benefit the "improved" hatchery fish may confer to the total population may be offset by their displacement of wild individuals. The result is that some naturally-produced wild fish are traded away for costly hatchery fish.

  Oosterhout and Huntington found that of the various management interventions simulated in the model runs, only freshwater habitat restoration offered significant and permanent gains in salmon abundance. In contrast to hatchery programs, habitat restoration poses no genetic or ecological risk to salmon populations. The results consistently suggest that wild populations respond rapidly to improvements of their habitat without the addition of hatchery salmon.

  The Pacific Rivers Council finds there are at least three important public policy implications:

  1. Status determinations, recovery plans, and recovery goals for salmon should not assume that hatchery-origin salmon are an integral component of the wild population.
  2. New, often very expensive "conservation hatchery" or "supplementation" programs, at best, likely will be only marginally more effective than past practices in providing any benefit to natural fish populations.
  3. Protection and restoration of freshwater habitat are the critical steps necessary for the recovery of natural salmon populations.

  In sum, wild production and habitat restoration determine recovery.

• Is there any way that hatcheries could be changed to provide more of a benefit to a Coho recovery program on the Oregon Coast or elsewhere for other at-risk populations?
  
  

  The study was based on an extensive review of the peer-reviewed and gray literature, and included extensive peer review of the simulation model and the analyses conducted. To avoid any hint of bias, the authors tried to err on the side of making the idealized conservation hatchery successful. Therefore, it appears unlikely that any foreseeable change in hatchery management could change the conclusions of this study.

  For example, the authors assumed that:

  *Hatchery managers in the simulations had perfect information about both wild and hatchery fish abundance and distribution, and perfect control that real managers could never attain.

  * Hatchery managers had the knowledge and ability to release fish from a hatchery at some idealized "appropriate time," whatever that might be, which would presumably be at a point that would maximize fitness of the hatchery fish and minimize adverse effects on wild fish.

  * Hatchery managers knew exactly how to sample the wild population, in time and space, for the fraction that were taken as broodfish, and were able to carry out that sampling without errors.

  * The river basin simulated in the analysis had habitats of good, fair, and poor quality, so some reaches were often unoccupied, and thus could potentially benefit from supplementation by hatchery fish as recolonizers.

  * The river basin simulated was not so degraded that neither the wild population nor the wild + supplemented population was so small as to be unviable.

  * Reproductive performance of hatchery fish was affected primarily in the first generation, with genetic effects on their offspring substantially diminished, and essentially gone after the second generation; this is a conservative assumption based on what is known about possible genetic effects of hatchery domestication.

• Why did the authors conclude that habitat restoration is a more effective means of salmon recovery than hatcheries?
   

  Hatchery fish are a temporary subsidy, and any effect they have on the population recovery stops as soon as the last hatchery fish (or their offspring) return. After that, the population merely equilibrates to the population abundance that the habitat conditions can support-the same abundance that un-stocked, wild populations achieve. By contrast, the effects of habitat restoration improve survival rates for all fish over the long term, permanently raising the cap on potential survival and abundance.

QUESTIONS AND ANSWERS ABOUT OREGON'S COASTAL COHO

Status of Oregon's Coastal Coho Salmon

• There are so many more salmon spawning this year, doesn't that mean the fish have recovered?
   

  No, it is not appropriate to measure the status of a population based on the outcome of good ocean years. That is like testing a soldier's bravery in battle by how well he or she gets along at a cocktail party. Rather, the persistence of individual populations of Oregon Coast coho will depend on how well the fish can perform in the freshwater habitat available to them during the next extended period of poor ocean conditions, not on their current abundance. The ability of a population to survive the most adverse conditions Mother Nature can dole out is the true test.

• But aren't coho numbers up significantly?
  
  

  Absolutely, particularly for adult spawners. The abundance of adult Oregon Coast coho has rebounded dramatically in the last few years from all-time lows, largely due to substantial increases in marine survival rates associated with a shift in ocean conditions. It appears that favorable natural variation in freshwater environmental conditions may also have contributed to the magnitude of the improvement.

   In addition to higher numbers of spawning adults, juvenile survival rates also are up, but not as much. Juvenile survival is an important measure of the extent to which habitat conditions are limiting productivity.

• Don't these improvements demonstrate that stream habitat is recovering?
   

  Recent large-scale increases in the abundance of Oregon Coast coho are consistent with what might be expected with improved ocean conditions, low harvest rates and decreased competition from hatchery fish on spawning grounds. They are likely to have had little to do with recent habitat restoration efforts, although such efforts may be affecting increases in coho abundance within localized areas.

  Contact PRC for more information about coho populations.