CCAMLR

The exploratory fishery for Dissostichus spp. has now been operating for fifteen years in Subarea 88.1 and for ten years in Subarea 88.2. This report summarises the timing, depth, and location of fishing together with the catch of Dissostichus spp. and bycatch species by year for the period 1997–98 to 2010–11. During the 2010–11 fishing year most of the catch in Subarea 88.1 came from the slope SSRUs 88.1H and 88.1I, and SSRU 88.1K. About 63% of the catch limit in the north was taken from 88.1C with the rest taken from SSRU 88.1B. The catch limit from the shelf was taken mainly from 881J. Most catch from Subarea 88.2 was taken from SSRU 88.2E but over 100 t was taken from each of SSRU 88.2D and 88.2F, and SSRU 88.2C was fished for the first time.

Unstandardised raw catch and effort data examined during this study suggest that the median toothfish catch per set has generally increased over the course of the fishery whilst the median number of hooks per set has remained relatively constant. Other indicators of CPUE including unstandardised catch per set and catch per hook have varied considerably over time in each of the fisheries showing no trend. There is no evidence of any truncation of the overall length frequency distribution, but some evidence for a reduction in median fish length in SSRUs 88.1H and 88.1I over the last 1–2 years as a mode of smaller (c. 100 cm) fish is selected by teh fishery, most likely  related to fishing depth. We conclude that from data examined from the fishery to date there is no evidence for substantial changes in population structure or abundance at the regional (Subarea) or local (SSRU) level.
 

We provide an update of the Bayesian sex and age structured population stock assessment model for Antarctic toothfish (Dissostichus mawsoni) in the SSRU 88.2E, using revised catch, catch-at-age and tag-recapture data from selected trips and all vessel trips. The 2011 reference case model resulted in a 30% higher estimate of initial biomass than for the 2009 base case. The main reason for this increase was likely to be because of the increase in the proportion of selected trips in the 2011 assessment.

Model fits to the data were adequate, with the tag-release and recapture data providing the most information on stock size, although there were conflicts between information in the recapture data from different years. Monte-Carlo Markov Chain (MCMC) diagnostics suggested no evidence of non-convergence in any of the models. The selected trips model run using the revised maturity ogive MCMC estimated initial (equilibrium) spawning stock abundance B0 d at 10 100 t (95% credible intervals 8 720–11 770 t), and current (B2011) biomass estimated at 82.7% B0 (95% credible intervals 80.0–85.1%). Estimated yields from this model were 484 t.

SSRU 88.2E has historically been assessed and managed as a separate area because it is geographically separated from the Ross Sea fishery and little fishing had been carried out in the other parts of Subarea 88.2. Furthermore, it is generally considered more conservative to manage small fisheries separately to avoid potential problems with localised stock depletion. However, over the past 2–3 years there has been an increase in fishing in SSRUs 88.2CDFG and it may now be more appropriate to carry out a combined stock assessment based on SSRUs 88.2C–G. This combined stock assessment is developed in a separate paper.
 

This paper describes the first Bayesian sex and age structured population stock assessment model of Antarctic toothfish (Dissostichus mawsoni) in Subarea 88.2 covering SSRUs 88.2C, D, E, F, and G (SSRUs 88.2C–G). For modelling purpose, we split the region into two fisheries: south of 70.2° S where smaller fish are typically caught and effort has been light; and north of 70.2° S where larger fish are caught and most of the effort has been concentrated. For our reference case we used a selected trips data set and fitted to tag and fishery data from the northern fishery only.
Model fits to the data were adequate, with the tag-release and recapture data providing the most information on stock size. Estimated initial equilibrium mid-season spawning stock biomass (B0) for model R1 (2011 reference case) ranged from 9820 to 13 350 t, with the initial biomass estimated to be 11 420 t. The yield was estimated at 544 t, slightly lower than the combined current catch limits for the region of 575 t.
We examined several sensitivities including (i) unaccounted mortality due to lost gear, (ii) catch and tag data from the southern fishery, and (iii) tag data from all trips. The inclusion of the unaccounted mortality from lost gear had little impact on estimates of biomass. The inclusion of tag data from the south fishery led to a 10% increase of biomass, due to the low numbers of tags there, whilst including tag data from all vessels led to a 20% increase in biomass.
We believe that the SSRU 88.2C–G model provides a more realistic representation of the observations and catch data for the area than the SSRU 88.2E only model. The SSRU 88.2C–G area included tags that may move between SSRU 88.2E and adjacent SSRUs, and it better accounts for the age frequency structure of the north and south areas.
 

We provide an update of the Bayesian sex and age structured population stock assessment model for Antarctic toothfish (Dissostichus mawsoni) in the Ross Sea region (Subareas 88.1 and SSRUs 88.2A–B), using revised catch, catch-at-age, and tag-recapture data for the 2010–2011 seasons. The 2011 reference model using the selected trips tag data gave a similar, but slightly higher estimate of initial biomass than the 2009 base case. Retrospective analysis suggests that this is partly as a result of the increased number of vessels in selected data set and partly as a result of the 2010 and 2011 observations. Two sensitivity models are presented; the first considers the effect of including possible unaccounted mortality from lost gear, and the second uses tag release and recapture data from all vessel trips.

Overall, model fits to the data were adequate, and, as in previous assessments, the tag-release and recapture data provided the most information on stock size. Monte-Carlo Markov Chain (MCMC) diagnostics suggested little evidence of non-convergence in the key biomass parameters, although there was some evidence of non-convergence in the annual shift parameters for the shelf fishery. MCMC estimates of initial (equilibrium) spawning stock abundance (B0) for the 2011 reference model were estimated as 73 870 t (95% credible intervals 69 070–78 880), and current (B2011) biomass was estimated as 80.0% B0 (95% credible intervals 78.6–81.3).The estimated yield, using the CCAMLR decision rules, was 3282 t
 

359 Antarctic krill catches were analysed for by-catch, it took 48.5% of observed catches and 31.3% of total catches. Fish by-catches were usual in the sector 48.1 (89.1% of cases); the most of by-caught fishes belonged to families Nototheniidae (NOX) (66.3%). Fish by-catches in the sector 48.2 were recorded in 50% examined catches; the most frequent were fishes of families Myctophidae (LXX).
 

This paper outlines a proposal to develop a field guide for Antarctic fishes which  would use photographs of live fish wherever possible.

Difficulties in distinguishing between certain gonad stages is of common concern for observers onboard vessels fishing for Dissostichus eleginoides. To minimise errors in interpretation of maturity scales of D. eleginoides a comprehensive identification guide has been prepared. It is proposed that this guide forms part of the CCAMLR observer manual.  
 

In this paper we update the assessment of D. eleginoides in the northern half of Subarea 48.4 presented in 2010. The CASAL model is improved through the incorporation catch at age data, as an alternative to the catch at length data that were the only such data available in 2010. The catch at age model generates a yield of 52 tonnes based on CCAMLR decision rules.
In 2006 the northern subdivision of 48.4 was set north of Saunders Island after consideration of oceanographic and bathymetric data, and for the purpose of the experimental work leading to a stock assessment. Data collected in the second part of the experiment (2009 – 2011) show that although CPUE of D. eleginoides is very low there, Saunders Island is an area of overlap between D. eleginoides and D. mawsoni, and is the likely effective southern limit of the D. eleginoides population. It should probably be incorporated into the assessment and management of D. eleginoides in the northern half of 48.4, but to date insufficient information has been acquired, including from tagging, to establish the likely rates of exchange between Saunders and the other northern areas in 48.4.
Given current knowledge and uncertainties on both Dissostichus spp stock status, we suggest that further data on the D. eleginoides population at Saunders island should be acquired, with fishing controlled by maintenance of current north-south area boundaries, application of a D. eleginoides 52 tonne catch limit to both areas with a sub-division of 5 tonnes allocated specifically for D. eleginoides in 48.4 South. We also propose that directed fishing on D. mawsoni should not be permitted in 2011/12, but that a move-on rule to limit bycatch of D. mawsoni should apply. Some options for a move-on rule are discussed.