Further examinations of the SBT operating model to explore new tagging model and grid specifications

Abstract

Stock assessments and constant catch projections were conducted using new Operating Models (OMs; sbtmod21 and sbtmod22, which have different tagging models) developed by the CCSBT ESC. The current analysis showed that: 1. a new candidate for the tagging model (incorporated in sbtmod22) led to higher M0 (natural mortality at age 0), lower M10 (natural mortality at age 10) and lower omega (non-linearity of the CPUE-abundance relationship) estimates than the previous tagging model which is used in sbtmod21, and estimated lower current stock abundance relative to the virgin unfished biomass, 2. a high S (S=0.5; S is the proportion of longline overcatch attributed to the reported effort) led to a lower M10, but the overall results were scarcely different from those for the base assumption (S=0.25), and 3. a slight change of assumptions regarding the Indonesian fishing selectivity impacted on M estimates substantially (leading to low M0 and high M10), which indicates poor ability to explain the Indonesian catch-at-age data when using a low M10 as pointed out during the 2008 SAG meeting.