In-season management policies for Kuskokwim River Chinook salmon

Objective 1. Reconstruct historical in-season run and fishery dynamics of Kuskokwim River Chinook, chum, and sockeye salmon stocks to inform the closed-loop simulation models.
Objective 2. Evaluate trade-offs associated with choosing among a candidate set of in-season harvest control rules for the mixed-species salmon fishery of the Kuskokwim River.
Objective 3. Elicit agency and stakeholder input on objectives and options, and facilitate technology transfer of in-season modeling tools.
Objective 4. Develop a probabilistic Bayesian in-season run forecasting tool to assist with management decision-making

We met these objectives through the development of two manuscripts, one of which has been
published in the Canadian Journal of Fisheries and Aquatic Sciences (CJFAS), and one that is in
preparation. In Appendix A we present a manuscript that describes and management strategy
evaluation (MSE) of in-season harvest management for the Kuskokwim River Chinook salmon
subsistence fishery in western Alaska. The MSE tested four primary management strategies that ranged
in their complexity and information needs. Our findings showed that all assessed strategies can perform
well, but that the more complex strategies tended to perform better when the incoming run was small.
Additionally, the optimal settings (i.e., aggressive or conservative with respect to fishing opportunity) of
each strategy depended on run size, with conservative settings favored in smaller runs. This analysis
necessitated the construction of an operating model that attempted to realistically depict an annual
run and fishery for Chinook and Chum/Sockeye Salmon in the Kuskokwim River. The model was
parameterized based on the best available scientific information as well as stakeholder input (Appendix
B), and was tested via simulation to assess the degree to which it was able to recreate historical
patterns in subsistence harvest (Appendix C). In Appendix D we present a manuscript that will be
published in CJFAS in 2019 that assesses the performance of two Bayesian information-updating
procedures to predict the run size during the season. The Bayesian updating approach we developed
provided a probabilistic expression of run size beliefs, which was used as the basis for the development
of an online run prediction and risk assessment tool that was disseminated to Kuskokwim salmon
managers prior to the 2018 salmon run. The tool was used to assist with the structuring of harvest
management deliberations during the 2018 run.

The report took longer to write than anticipated due to time demands on the Principle Investigators.

Management of salmon fisheries to simultaneously achieve both conservation (e.g., ensuring adequate escapement) and fishery performance (e.g., meeting subsistence harvest needs) objectives is challenging because of uncertainty and annual variability in run size and timing. Managers faced with balancing these objectives in western Alaska Chinook salmon fisheries are particularly challenged because of historically small runs in recent years. As Chinook stocks have declined on the Kuskokwim River, the need to improve the efficacy, transparency, and predictability of in-season management techniques has increased. We propose to address these needs by 1) articulating and evaluating in-season harvest control rules that seek to meet subsistence salmon harvest needs from the suite of available species while limiting exploitation of Chinook stocks during weak run years, 2) holding a series of technical workshops with agency staff and key stakeholder representatives to elicit feedback on potential management control rules and objectives, and 3) developing in-season run assessment tools that utilize preseason forecasts coupled with in-season assessments informed by the Bethel Test Fishery. The technical workshops will also provide an opportunity for information transfer among workshop participants and allow further refinement of models and precise consideration of stakeholder objectives used to evaluate the performance of alternative management policies.

In addition to model development and analysis, we participated in four NFWF-funded Capacity
Building Workshops that were arranged by AYKSSI and facilitated by the Quantitative Fisheries Center at
Michigan State University. We attended four of these stakeholder meetings at which we gave
presentations on in-season management and other related topics. The technical workshops also
provided an opportunity for information transfer among workshop participants and allowed further
refinement of models and precise consideration of stakeholder objectives used to evaluate the
performance of alternative management policies.