Council debates salmon risk-management tools: risk tables, forecast metrics, and p‑star buffers

The council considered two different approaches to handling uncertainty in preseason salmon abundance forecasts: (1) salmon risk tables — structured rubrics that assemble ecosystem indicators and assessment performance to guide precautionary responses; and (2) a statistical approach proposing two standardized forecast‑performance metrics (c for bias and sigma for precision) that could support bias correction and p‑star buffering of forecasts before management calculations.

Dr. Will Satterthwaite presented the statistical approach, describing c as "the median ratio between the post season estimate of abundance compared to the pre season forecast of abundance" and sigma as the log‑scale standard deviation around that ratio. He explained that applying a bias correction or a p‑star (precautionary multiplier) could reduce the frequency of under‑escapement and overfished status, but would also reduce mean harvest in some scenarios. "There does seem to be an upper limit to how accurate we can really expect forecasts to be," he told the council, arguing for transparent reporting and a repeatable method for translating concern about forecast reliability into management advice.

Advisory bodies differed in emphasis: the SSC urged standardized reporting of c and sigma in preseason reports; the SAS urged investment in improving forecasts rather than adopting buffering schemes that could materially reduce harvest; the EWG and EAS supported further work on risk tables but warned that implementation would require staffing and close coordination with co‑managers. Council members agreed the discussion was useful and should inform workload planning ahead of November peer-review and harvest‑policy work.