The conveners heartily thank all participants for the spirit of cooperation that pervaded the meeting. Discussion focused on the commonality of purpose in fostering abundant salmon populations, with as much quality habitat and genetic fitness and diversity left in place as possible to help them to face an uncertain future. General agreement on this and other important fundamental principles emerged. Vigorous debates on difficult strategic issues also highlighted important information gaps. The meeting closed with a plan to create a working committee to press forward in building a consensus-based hatchery reform programme.

Underlying all the discussion was a sense of apprehension over the uncertain future for wild salmon populations, and a recognition that societal values and expectations have shifted in the closing years of the last century. Optimistic predictions of major increases in salmon "production" have been replaced by anxiety over possible extinctions of naturally spawning salmon in large geographic areas. Hatcheries that were built to provide enhanced fishing opportunities are now being called upon to save important salmon populations from extinction. There was general agreement that hatchery reform was needed in light of these changing expectations and needs.

In addition, a consensus emerged:

• That hatcheries and other forms of artificial enhancement cannot readily replace damaged or lost freshwater habitat,
• That we humans can never fully understand enough of the complexities of natural ecological and genetic systems to be able to maintain them artificially.
• That there can be no substitute for diligent maintenance of

• high-quality natural habitat,
• healthy freshwater and marine ecosystems, and
• abundant, naturally reproducing salmon populations with their genetic fitness and diversity intact.

Several participants also stressed that restoration of damaged habitat, though valuable if properly done, was also far more expensive and less reliable than avoidance of the damage.

There was also general agreement that (i) hatcheries need to be viewed as components of complex ecological and genetic systems, (ii) hatchery evaluations need to focus onimpacts on these systems. To this end, the more traditional estimates of juvenile and adult production and survival rates need to be augmented. Enhanced returns of hatchery fish might, e.g., come at a cost of reduced returns of other populations or reduced genetic fitness or diversity in wild fish populations.

Carrying capacity was also a recurring theme. There was general recognition that to overload a small stream with coho fry from a hatchery may do more harm than good. Some participants also raised concerns of similar problems in overloading marine habitat with hatchery output. . In addition, several participants stressed the extreme variability and uncertainty surrounding carrying capacity and associated survival rates, and emphasized that this uncertainty must be fully accounted for in hatchery reforms.

This was the subject of lively debate. In the discussion, "strategic intervention" was interpreted as a limited use of artificial enhancement, typically including some sort of intervention in the reproductive cycle, to attempt to preserve a fish population from extinction. As currently envisaged by management agencies represented at the workshop, strategic intervention would take place when important populations were identified as facing a serious risk of extirpation. This intervention would be part of a recovery plan. This plan would delineate both an ongoing assessment strategy and specific rules for termination.Specifically, the effects would be carefully monitored with mandatory re-evaluations of the programme after each salmon life cycle, and the intervention would cease as soon as rebuilding targets were reached. Removals from the natural population would be aimed at minimizing the risk of the loss of important genetic information in that population, and young salmon would be returned to the wild as soon as feasible.

All participants recognized the lack of clear theoretical guidance and policy for when to intervene. In the conveners’ opinion, there are enough uncertainties surrounding the effectiveness of intervention strategies to question their value as conservation tools. Research on the ultimate value of such efforts is urgently needed in light of escalating numbers of declarations of threatened or endangered salmon populations. Those who supported strategic intervention listed the following as important criteria for deciding when to intervene: absolute population size, geographic extent of the population(s) at risk, rate and amount of the decline, and genetic structure and uniqueness. They also stressed that it was important not to wait until genetic losses had occurred. Participants also stated that special social and cultural values associated with individual populations might trigger earlier intervention.

Others cited evidence of potential harm from well-intended strategic interventions, and argued against their deployment. Some also pointed to the potential value of other, potentially less invasive conservation tools such as cryopreservation. The disagreement over strategic intervention highlighted key information gaps as outlined below.

Major information gaps came to light repeatedly throughout the meeting. The following were identified as critical for making more informed decisions in hatchery reform.

• Goals: There is an overriding need for a comprehensive reassessment of the role of hatcheries in fish management. This would ideally lead to a hatcheries component in an overarching management framework. The comprehensive reassessment should eventually lead to a careful re-examination of the role of individual hatcheries, the setting of measurable goals for each hatchery, and the creation of a monitoring and evaluation scheme for assessing success in achieving these goals.
• Strategic Intervention: We need to reduce the uncertainty over whether a proposed intervention will do more good than harm. To this end, we need both

• Ecological Impacts: Ecological interactions in general were identified as high-priority items for further study. Of these, the following were highlighted:
• Mixed-stock fishery impacts.
• Carrying-capacity limits. Two key aspects were identified.
• The value of a better understanding of the causes of recent, large fluctuations in marine survival.
• The potential value of experimental manipulations in hatchery output to probe their ecosystem impacts. In this regard, participants noted the need for a thorough review of existing evidence, the need for an advance assessment of feasibility, including the probability of detecting a substantial carrying-capacity effect if one were present (i.e. an evaluation of statistical power), and a pilot study, possibly in a freshwater environment. Participants stressed the need for broad consensus and cooperation in an experiment that involved the manipulation of hatchery output across a broad geographic scale. For example, any experiment designed to probe hatchery impacts on the Strait of Georgia would require cooperation amongst all agencies involved in managing fisheries and hatcheries in the combined Strait of Georgia and Puget Sound regions. Furthermore, such an experiment would impact aboriginal, commercial and recreational fishing interests, hatchery workers, and others. It would require a major effort in public relations to bring these groups onside.
• Genetic Impacts: Of the many unanswered questions, the following emerged as the most important.
• What is the comparative, lifetime success of hatchery vs. wild fish in lifetime reproductive success? Some experiments are being initiated by the U. S. National Marine Fisheries Service, but more are needed in a variety of geographic regions and on different species.
• How soon do important genetic differences appear between hatchery and wild fish?
• If these genetic differences infiltrate the wild gene pool, how long do they persist after a hatchery operation has been terminated?

• Zonation Issues: A map of enhancement-free salmon populations would help to assess present status. In addition, the issue was raised of creating special zones within which no salmon culture would be undertaken. The map would be a useful tool in helping to identify candidate areas for such a designation.

The group recognized the special difficulties in implementing hatchery reforms —even beyond those associated with other aspects of fish management. Both the Puget Sound and Coastal Washington Hatchery Reform initiative in Washington State and the Selective Fishing Strategy in British Columbia were praised as models to emulate. ] Key features that were highlighted were:

• Active participation at all stages of development and implementation from
• government agencies,
• hatchery operators,
• First Nations,
• local communities,
• commercial and recreational fishing interests,
• nongovernmental conservation organizations,
• independent scientists, and
• facilitators and communicators.

It is critical to bring all interested parties into the process to avoid alienation and polarization. In addition, the role of independent scientists in providing complementary expertise and perspective was stressed, as was the importance of involving people with the ability to facilitate consensus building and to communicate to the public the sometimes subtle, complex problems that need to be addressed. It is also important to recognize that both the hatchery reform initiative in Washington State and the Selective Fishing Strategy in British Columbia required a major commitment of resources. A similar process for hatchery reform in British Columbia would cover a much larger geographic area than the one in Washington State, and would involve more interest groups than the Selective Fishing Strategy. It would require even more resources. Participants stressed repeatedly the value, not only of creating a structure that could foster such a cooperative effort, but also of creating a network for continuing the discussions begun at this workshop. In particular, British Columbia participants stressed the value of learning from the experience in Washington State. The Think Tank session ended with a commitment to create a small working committee to pursue this vitally important objective.