Abstract
Here we review the science of hatchery reform, describing the benefits of hatcheries, the risks of hatcheries, and the major operational options available to hatchery mangers for balancing the benefit-risk trade-off. Our review is a component of a larger evaluation of the Washington Department of Fish and Wildlife’s (WDFW) hatchery and fishery reform policy, adopted in 2009, that embraced the recommendations of the Hatchery Scientific Review Group (HSRG). Thus, we explicitly aim to synthesize new and emerging science available in the last ten years, towards the goal of helping inform whether a revision to the policy is warranted. We focus on Pacific salmon and steelhead hatcheries, with strong emphasis upon examples from Washington State to illustrate key concepts.
Hatchery benefits have received much less research attention than hatchery risks. The goals of hatcheries are typically classified as providing harvest opportunities or improving the conservation status of natural populations. Harvest opportunities, which are heavily subsidized by hatcheries in Washington State, provide economic benefits, sociocultural benefits, and satisfy legal obligations, including fulfilling the treaty rights of Native American Indian tribes. We provide a brief overview of these benefits, but they are not the focus of this paper. Regarding conservation, hatcheries have proven successful at preserving unique genetic lineages. Hatchery-origin fish also commonly spawn naturally in rivers throughout Washington State, both intentionally and unintentionally. Depending on the circumstances, this can have short-term conservation value to population demographics, but must be carefully weighed against long-term genetic risks. However, the evidence that hatcheries have increased the abundance of natural-origin adult salmon, a fundamental metric for evaluating population status, is generally sparse, likely because key assumptions regarding habitat carrying capacity and hatchery operational objectives (e.g., high proportion natural-origin broodstock) are rarely met.
Hatchery risks include fishery risks, ecological risks and genetic risks. Fisheries targeting abundant hatchery runs can unintentionally increase mortality of co-mingled natural populations. We describe three factors that contribute to fisheries risks: constraints on implementing mark-selective fisheries that remove only hatchery-origin fish; an asymmetry between lost harvest opportunity and the conservation gain of curtailing fisheries; and uncertainty in the harvest benchmarks due to the frequency of naturally spawning hatchery-origin fish.
The ecological risks of hatcheries, especially competition for limited foraging resources and increasing predation on natural populations, have been recognized for some time. Retrospective analysis of population monitoring data highlight the potential for hatcheries to reduce the productivity and abundance of natural populations via ecological interactions. However, at the population or watershed scale, ecological impacts of hatcheries are difficult to separate from other factors affecting productivity and abundance, including the genetic risks of hatcheries.
Several hatchery reform measures are intended to minimize ecological interactions. These include releasing smolting rather than pre-smolt fish, volitional rather than forced releases, the use of acclimation ponds rather than direct releases, delaying releases until after the migration of threatened species that hatchery-reared fish might consume as prey, and concentrating hatchery releases geographically distant from natural populations. Isolated case studies have demonstrated the effectiveness of some of these approaches, but research on ecological interactions lags far behind the attention devoted to genetic risks of hatcheries. Importantly, research suggests the potential for ecological interactions in marine environments shared between multiple hatchery and natural populations, yet very little is known about the likelihood or magnitude of population-scale ecological impacts of hatcheries.
The major genetic risks of hatcheries include the loss of genetic diversity within and among populations, and the loss of fitness due to selection for traits favorable in the hatchery but deleterious in the wild. Despite risks of inbreeding and loss of alleles in small populations, there are several examples in which conservation hatcheries have maintained or even increased within population genetic diversity through thoughtful management. However, at larger regional scales, some hatchery practices, including a legacy of intentional transfers of eggs and broodstock between watersheds, have contributed to genetic homogenization and reduced genetic diversity among populations. Studies comparing the number of offspring produced by hatchery-origin fish and natural-origin fish when both groups spawn in the wild (relative reproductive success, RRS) have demonstrated a general pattern of lower reproductive success of hatchery-origin fish. Researchers have highlighted domestication selection as a plausible cause for the lower RRS of hatchery-origin fish, although unequivocal, empirical, population-scale evidence for a genetic basis to fitness loss remains rare.
Minimizing fitness loss by managing gene flow between a hatchery population and its companion natural population has been a fundamental focus of hatchery reform. Through a combination of regional conservation prioritization, assessing the current status of the natural population, and specifying whether or not the hatchery intentionally exchanges genetic material with the natural population, the HSRG advanced a method for setting fitness loss risk thresholds. These thresholds track the proportion of hatchery-origin fish spawning naturally (pHOS) and the proportion of natural-origin fish used in hatchery broodstock (pNOB). Importantly, these metrics are surrogates for true, realized fitness risks. However, they have strong conceptual support from genetic models, including novel exploration of existing models presented in this paper. Furthermore, recent empirical RRS research has highlighted that hatchery programs employing 100% natural origin broodstock can provide demographic conservation benefits while maintaining genetic diversity with minimal or no apparent genetic fitness loss.
While recognizing the social, political, economic and legal considerations that factor into decisions regarding hatchery policy, we reached the following conclusions regarding hatchery reform in Washington State.
Overarching themes
- Hatchery reform is but one of several factors requiring careful planning and aggressive implementation needed to achieve meaningful recovery of salmon populations
- Hatchery reform is largely aimed at reducing risk in a relative but not absolute sense
- In WDFW’s hatchery system, a focus on efficiency and maximizing abundance prevents widespread implementation of risk reduction measures
HSRG Recommendations
- The principles of reducing pHOS and increasing pNOB to achieve fitness gains in wild populations are well-founded, and should be fundamental goals in any hatchery reform management action
- Program size requires more careful scrutiny and scientific justification because it affects virtually every aspect of hatchery risks
- The HSRG’s phased approach to recovery has strong conceptual merit, but its implementation has resulted in an absence of stricter, conservation oriented PNI goals for many populations
- We recommend crafting a stand-alone monitoring and adaptive management plan for each hatchery program that quantifies both benefits and risks, and explicitly links hatchery performance metrics to potential operational changes
Knowledge gaps and major assumption of current hatchery management
- The absence of a landscape-level, replicated experiment prevents empirical assessment of hatchery reform effectiveness
- Hatcheries have potential for large magnitude ecological impacts on natural populations that are not well understood, not typically evaluated and not measured
- Understanding the role of life history diversity on hatchery-wild ecological interactions and ecosystem stability is a significant research need
- We recommend a more rigorous, consistent and intentional evaluation of cumulative hatchery effects across multiple hatchery programs operating within a geographic region
Overall, our review supports the fundamental concepts and approach of the HSRG, but also identifies knowledge gaps and challenges to coordinated, robust implementation of scientific principles at a statewide scale. In order to advance hatchery reform as a comprehensive program for developing scientifically defensible hatchery programs, these issues warrant dedicated, programmatic initiatives.