Stock assessment scientists responsible for summer flounder should investigate how differences among model results arise and whether such differences indicate changes needed in the models or assumptions used. Participants in the Stock Assessment Workshop should try to test the effects of shrinkage and should investigate the conflicting trends in the youngest year classes. As stated in the 1998 NRC report Improving Fish Stock Assessments4:
Because there are often problems with the data used in assessments, a variety of different assessment models should be applied to the same data. . . . The different views provided by different models should improve the quality of assessment results. (p. 113)
This advice is borne out by the current committee’s reassessment of summer flounder data.
In the summer flounder case, the NMFS assessment could be improved by analyzing the same data using different models. The differences obtained should help analysts learn about problems in the data, problems in using the ADAPT model with these data, or problems with the assumptions used in the NMFS ADAPT model (e.g., related to shrinkage and changes in selectivity over time).
The committee also investigated concerns about the summer flounder stock assessment by analyzing data using the VPA method a number of times and changing single assumptions for different runs. The most significant finding for the upcoming SAW to consider relates to the number of age classes used to tune the assessment model. Survey data on age classes 5 and older typically are excluded from the analyses for all but the NEFSC winter survey (available only since 1992). This is due in part to the difficulty some assessment procedures have in dealing with sparse data, observed zero values, or missing data, which are more common for older age classes. Committee analyses determined that the assessments can be sensitive to the number of age classes included in assessments. Therefore, future NMFS assessment activities for summer flounder should include more year classes (up to age 7) to tune the model and decrease the possibility of missing changes in fishing mortality on older ages. This will be particularly important as the summer flounder population recovers and includes a greater proportion of older fish.
Much more detailed analyses, and additional findings and recommendations, will be presented in the full report, Improving the Collection, Management, and Use of Marine Fisheries Data.
National Research Council. 1998. Improving Fish Stock Assessments. National Academy Press, Washington, D.C.