I made five sensitive runs of CASAL models for stock status assessments of Dissostichus eleginoides in research block C in Division 58.4.4 b on the assumption of constant year class strength and no IUU fishing occurrence in the block. The five models are as follows: 1) catch at length model as before (Len model), 2) age structured model using single age-length key (ALK) (AgeSing model), 3) age and sex structured model using single ALK (AgeSexSing model), 4) age structured model using annual ALKs (AgeAnn model), and 5) age and sex structured model using annual ALKs (AgeSexAnn model). Von Bertalanffy parameters were not estimated in these models according to the recommendations in the last WG-FSA meeting. In addition, proportion mature for ages was calculated on the assumption of logistic curve fitting for the proportion of maturity index with stage 2+ and incorporated in the models.
The median MPD estimates of the initial B0 and B2013 were highest (539 and 487 tonnes) in AgeSexSing model and lowest (488 and 436 tonnes) in Len model. The median MCMC estimates of the initial B0 and B2013 were highest (840 and 790 tonnes) in AgeSexSing model and lowest (740 and 680 tonnes) in Len model. Thus, in each of MPD and MCMC calculation, stock size did not differ largely among five models. However, MCMC estimates were significantly larger than MPD ones for each model.
There was still some evidence of lack of convergence in MCMC posterior trace and in the median MCMC jump statistics for initialization B0 in each model. Hence, the MCMC results are not yet robust, which may be caused by factors associated with growth parameters and IUU fishing.
This report does not yet address the investigation of estimation of IUU fishing and sensitivity of scenarios to alternative selectivity by the IUU fleet (para. 6.93 (iii) and (iv) of SC-CAMLR XXXII, ANNEX 6, 2013). I need advice about the methods for these tasks during the WG-SAM meeting so that I can revise report for the following WG-FSA meeting.