I would like to extend the discussion about straying in the context
of ongoing activities between Indian tribal fishery agencies and
the State. Both groups have been working on jointly developing
a policy on wild salmonids that addresses straying, gene flow,
and other genetic issues. The issue of salmon straying has important
genetic implications for the conservation of wild stocks, harvest
management, and hatchery production. Hatchery production, usually
using non-local stocks, has been used to meet harvest management
objectives (harvest augmentation). In contrast, the chief goal
of hatchery supplementation, like that used for Snake River stocks,
is to restore or increase productivity of wild stocks. Increasingly,
the occurrence of declines in wild stock abundance has prompted
consideration of alternative rebuilding strategies, with much
focus on strategies involving supplementation. This move to emphasize
natural productivity and to maintain or increase harvestable surpluses
reflects a transition from the context that most fishery managers
have been working within for many decades. It is important to
keep in mind that salmon management, with all of its inherent
complexity, is now undergoing a major transition.
The key factor when considering the genetic effects of straying
is maladaptive gene flow, not just the physical presence of non-local
or hatchery-origin fish in natural populations. However, direct
measures of gene flow are at best difficult to obtain, and the
number of strays into a spawning population may be all that we
have as surrogate estimates of gene flow. Given that we need
better ways of estimating the actual numbers of non-local fish
spawning in the wild, we must focus on reproductive overlap between
hatchery fish spawning in the wild and wild fish. Estimates of
this overlap will provide one of the best surrogate measures of
actual gene flow in the absence of direct measures, and thus overlap
is a key part of State and tribal discussions on wild salmonids.
Several assumptions about life-history patterns and gene flow
have been made during our wild-salmonid policy discussions. One
is that some level of straying and gene flow occurs naturally
between wild populations, but just how much and under what circumstances
is uncertain. Another assumption is that human-produced elevated
rates of natural straying and unintended straying from non-local
hatchery sources are undesirable because of the potential loss
of genetic variability within and among populations, and because
outbreeding may decrease fitness and productivity. More information
on the effects of outbreeding depression, as Gary Graves pointed
out, is urgently needed to assist in the development of rational
management policies. Another important assumption in the evolving
policy is that various management strategies can be used to increase
homing or to decrease rates of straying. Several state monitoring
and evaluation programs are under way to understand better the
causes and effects of straying.
With respect to general guidelines in formulating Washington's
wild salmon policy, the intent of the State and tribal fishery
agencies is to manage gene flow to maintain genetic diversity
and to conserve local adaptations, productivity, and evolutionary
potential. More information is needed on the means to achieve
these rather abstract goals. The underpinnings of our wild salmonid
policy are clearly set out in legislation passed by the Washington
State Legislature in 1993, which directs the State to ensure that
department actions and programs are consistent with the goals
of rebuilding wild stock populations to levels that permit commercial
and recreational fishing opportunities (Washington State Legislature
1993, p. 73). This general directive and the Department's response
to developing a wild salmonid policy under the State Environmental
Policy Act (public review and comment) and in collaboration with
tribal managers is intended to allow flexibility in responding
to the specific requirements of different watersheds, regions,
and species. A fundamental conceptual difficulty has been to
decide whether a fitness-based approach or a diversity-based approach
is more appropriate; to date we have leaned toward the latter.
Issues related to fitness would be indirectly addressed.
In general, our diversity-based approach requires first a description
of genetic diversity within and between populations of each species.
Part of this need was met in the wild salmon and steelhead stock
inventory (WDF 1993). In addition, we are attempting to describe
the hierarchical components of genetic, life-history, and ecological
diversity within each species. Our approach and methods are similar
to those used by NMFS to make determinations regarding Evolutionary
Significant Units (ESUs). However, as a practical, conservative
approach, we will identify smaller units to manage for genetics,
population status, and maintenance. Secondly, we expect to manage
for specific levels of gene flow (e.g., reproductive overlap)
between hierarchical population units; that is, we intend to manage
unnatural gene flow between population units. This includes,
for example, gene flow resulting from stock transfers and other
practices, and from overlap resulting from hatchery fish that
escape harvest and spawn with wild fish. Surrogate measures of
gene flow will be required.
We are aware that reproductive isolation between subpopulations
can occur in different ways. Some groups of fish may spawn at
the same time, but in different places. Others may spawn at different
times, but in the same area. Again, accommodating these subtleties
of life-history patterns and migratory and spawning behavior is
important. Although simply measuring the number or percentage
of hatchery fish straying into a particular stream may be feasible,
it will lead to imprecise estimates of gene flow. A better approach
would be to estimate overlap, and to target the amount of allowable
overlap between population units to be scaled to the relative
genetic similarity between hatchery and wild fish in a particular
watershed. We are, however, still uncertain as to what levels
of gene flow or surrogate measures (e.g., overlap) should be allowed
between hatchery and wild fish under various circumstances and
are hoping the panel will help clarify the options and their corresponding
risks. For our policy approach to be implementable and therefore
successful, we need to be able to measure and to monitor suitable
parameters; we are aware that various theoretical predictions
are possible, but we need to have practical tools so we can monitor
progress toward meeting policy objectives. Once the policy has
been completed, implementation plans will consider a broad range
of strategies. Monitoring plans will include a systematic review
of hatchery practices to identify and activate improvements needed
to achieve compliance with wild salmonid policy goals.
Washington State Legislature, House Committee on Fisheries and
Wildlife. 1993. Second Engrossed Substitute House Bill 1309,
p. 69-74. 1993 final legislative report: Fifty-third Washington
State Legislature, 1993 regular session, first special session,
Washington State Legislature, Olympia, WA, 340 p. (Available
from House Office of Program Research and Senate Research Center,
Olympia, WA 98501.)
Washington Department of Fisheries (WDF), Washington Department
of Wildlife (WDW), and Western Washington Treaty Indian Tribes
(WWTIT). 1993. 1992 Washington State salmon and steelhead stock
inventory (SASSI). Wash. Dep. Fish Wildl., Olympia, 212 p. and
5 regional volumes. (Available from Washington Department of Fish
and Wildlife, 600 Capitol Way N., Olympia, WA 98501-1091.)
Comment: Marta Nammack: I noticed that both Gary Graves and
Steve Leider say that the burden of proof should be on the scientists
to show that theoretical concepts are important. I think the
burden of proof should be shifted to the proponents of hatchery
programs to show that their activities do not influence wild populations.
Steve Leider: Point well taken.