The NOAA IR serves as an archival repository of NOAA-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by NOAA or funded partners.
As a repository, the NOAA IR retains documents in their original published format to ensure public access to scientific information.
i
Performance gains from incorporating dynamic numbers by length-within-age in fishery assessment models
-
2024
-
-
Source: Fisheries Research 276 (2024) 107039
[PDF-3.78 MB]
Details:
-
Journal Title:Fisheries Research
-
Personal Author:
-
NOAA Program & Office:
-
Description:Simulation testing is fundamental for evaluating which features to include in a next-generation stock assessment model. One feature, contemplated but rarely adopted, is to extend the standard age-only population array to estimate population numbers dynamically by both age and length-within-age. This allows the model to capture changes over time in the shape of the length-at-age distribution when mortality is asymmetric with respect to fish body length. Here we use simulated data from an individual-based fishery model to evaluate three assessment models. Two are versions of Stock Synthesis (SS3), one employing a standard formulation and a second using ‘platoons’ to account for dynamic numbers by length-within-age. The third model, using a ‘slice partition’ to model numbers by length-within-age, was designed for assessing fisheries with a knife-edge truncation of the catch at a regulated legal minimum length. Platoons produced substantially better estimates of mean length-at-age, selectivity, and biomass than SS3-age for the simulated fishery scenarios with a higher level of individual growth variation. SS3 with platoons also produced better estimates than the slice model for these high growth variation scenarios, primarily when selectivity was estimated and was not knife-edged. The three assessment models performed well, and similarly, for lower individual growth variability, except that the slice model performed modestly better under knife-edge selectivity for which it was designed. Incorporating dynamic population numbers by length-within-age can be ranked as a medium priority for a next generation assessment model. Incorporating platoons in SS3 showed a clear performance gain, with moderate improvement evident, when there is high individual variation in growth. If length-within-age is incorporated, platoons are preferable to slices for implementation flexibility: by specifying the use of only one platoon, the model reduces to a standard age-based form, leaving all other model features unchanged.
-
Keywords:
-
Source:Fisheries Research 276 (2024) 107039
-
DOI:
-
Format:
-
Document Type:
-
Funding:
-
Rights Information:Accepted Manuscript
-
Compliance:Submitted
-
Main Document Checksum:
-
Download URL:
-
File Type:
