Advisory Panel seeks four more sablefish model configurations

The North Pacific Fishery Management Council's Advisory Panel voted unanimously Thursday to recommend that sablefish stock assessment scientists test four more model configurations before the council sets 2027 and 2028 catch limits in December. The assessment authors' two new models would put the Alaska-wide acceptable biological catch (ABC, the scientific ceiling on harvest) 13% or 32% below the 2026 level of about 47,000 metric tons, though their slides say 2027 figures will likely be higher with population growth and new recruitment data. Fleet members say they are seeing more sablefish than in decades.

In 2025 the federal longline survey shifted from covering the Gulf of Alaska every year to alternating between the Gulf and the Bering Sea and Aleutian Islands. AP member Nels Evans said that cut the share of the survey footprint sampled each year from 75% to 50%. The 2026 survey found the Aleutian index up 26% and the Bering Sea index down 14%. Asked how that yields a 32% decline, council staffer Sara Cleaver said the model changes produced "a decline that is not necessarily indicative of a change kind of in the population."

The authors, and in September the groundfish plan teams, called Model 26.2, which fits each region's survey separately rather than carrying old values forward, the "best available science"; an independent review concurred. The Scientific and Statistical Committee suggested another configuration, and plan team coordinator Diana Stram said as many as four will be reviewed in November, with no final pick made yet. A phase-in of an overall ABC happens only when it rises, Stram said: "We only do it on the way up and not on the way down."

The AP asked for Model 26.1 with carryover apportionment; a three- and five-year phase-in of Model 26.2; Model 26.2 with dome-shaped Bering Sea selectivity; and a 26.2 sensitivity run using 26.1's natural mortality value. The motion also backs restoring comprehensive annual surveys and using commercial size and catch-rate data.

Linda Behnken, executive director of the Alaska Longline Fishermen's Association, said the uncertainty "comes from a lack of data, not having an annual survey anymore, and because the stock assessment authors are not getting commercial catch data." She put the possible loss at "the $50 million that could cost the industry if that's what happens here." AP co-vice chair Chelsae Radell, speaking for shoreside processors, said that had a 30% cut in the Gulf applied in 2026, processors would have to decide which plants stay open.

"Due to the unprecedented cuts that are proposed through this— through these new models to the 2027 ABC and no biological concern for the stock, I believe that this is warranted," Evans said. Panel chair Shannon Carroll said supporting it was not about second-guessing the scientists but about what is lost when surveys and data fall short.

The full council takes up groundfish harvest specifications Saturday. The plan team reviews the models in November; the council decides in December.