CCAMLR

In 2011/2012, a French vessel conducted its first exploratory fishing campaign in division 58.4.3a. In order to achieve a robust stock assessment that would provide advice on catch limits according to CCAMLR decision rules, it is suggested to continue the research program for the next three years. This paper is a revised proposal of WG-SAM-12/14 encouraged by 2012 SAM working group for a resubmission. The present research plan proposes to continue the mark-recapture experiment and to collect detailed catch, effort and biological data with a nearly same survey design as in the 2012. It will include a total of 28 research hauls. The objective of the experiment is to better understand the distribution of the metapopulation of toothfish centered on the Kerguelen-Heard-Crozet sector.

In 2011, AAD, the Fisheries Research and Development Corporation and Australian fishing industry conducted a joint project to study the reproductive dynamics of Patagonian toothfish (Dissostichus eleginoides) around Heard Island and the McDonald Islands (HIMI, Division 58.5.2). Gonadosomatic data was available for 12 563 individuals. The histology of male and female gonads was also analysed from specimens collected in 2011. This data confirms that toothfish spawn between July and August across a large latitudinal range along the deep slope of the Kerguelen Plateau. A comparison of macroscopic staging of gonad development with histological analysis showed macroscopic characteristics often failed to distinguish spent or resting mature animals (as confirmed by the presence of post-ovulatory follicles) from first maturing animals, and that even large mature fish may not spawn every year. As a result, the conventional method of estimating size at maturity, with stage 3 or above (according to the Kock and Kellerman (1991) scale) being assessed as mature, did not provide realistic estimates. A modified method, defining stage 2 or above as  ‘mature’, offset by two years to account for the duration of the maturation process, provided more realistic estimates. The revised ogives were shown to have a significant impact on the estimates of stock status relative to those used previously that were derived from studies of other populations of toothfish.

Between 2010 and 2012, AAD, Australian Fisheries Management Authority and Australian fishing industry provided funds to conduct high throughput ageing of toothfish otoliths. 5780 otoliths, collected during the 2000-2011 fishing seasons in Division 58.5.2 and outside the CCAMLR area at Macquarie Island were processed. The inclusion of these datasets have improved the estimates of key parameters population dynamics of toothfish including growth rates, mean recruitment levels and interannual variability, as well as allowing more precise estimates of the selectivity of the different fishing gears used in these fisheries. The inclusion of catch at age data is shown to substantially improve the precision of the stock assessment in Division 58.5.2 compared to a scenario using only catch at length data. The ongoing collection and processing of otoliths to allow the inclusion of population specific estimates of catch at age data remains a high priority for toothfish fisheries. A costing breakdown for otolith processing at the AAD indicates that initial set up of laboratory equipment and salaries are the main expense in developing an ageing program.

France presented at the CCAMLR XXVIIth meeting, an action plan aiming to reduce incidental seabird mortality by half in the French EEZ included in the statistic division and subarea 58.5.1 and 58.6, for the period 2008/2009 to 2011/2012.

France demonstrated at the CCAMLR XXXth meeting, to have successfully fulfilled its commitments and to have obtained encouraging results concerning the global diminution of seabird incidental mortality.

In addition to the sharp decrease in incidental catch of seabirds in the French EEZ (see statistical report), this document presents new proposals that France would like to implement in order to pursue its efforts to minimize incidental captures.

The Commission aims to protect vulnerable marine ecosystems (VMEs) and has established several Conservation Measures that limit bottom fishing and require data collection by fishing vessels.  Members have made substantive and important scientific contributions to support the Commission’s objective of protecting VMEs, but work comparing the effects of different gear types has been relatively limited.  We used data collected by the longline fishery in Subarea 88.1 to compare the impacts of Autolines and Spanish longlines on VME taxa.  We found that the probabilities of observing positive bycatches and the amounts of positive bycatch decreased with depth, but we also found that the rate at which positive bycatches decreased with depth was greater for Autolines than for Spanish longlines.  Since we only analyzed data from areas where there was substantial spatial overlap of sets made with each gear type, we attributed the difference in rates at which positive bycatches decreased with depth to a difference in the rate at which VME taxa drop off Autolines relative to Spanish longlines while the gear is being hauled.  We adjusted for this differential drop off to infer that Autolines have about 7 to 20 times greater impact on VME taxa per hook (or about 3.5 to 10 times greater impact per set).  We believe that biased reporting (e.g., under-reporting by vessels using Spanish longlines) would change our specific findings but not our conclusion that Autolines have a substantially greater impact on VME taxa per unit fishing effort.  Limiting the use of Autolines, which are in more intimate contact with the sea floor than Spanish longlines, may provide a precautionary approach to mitigating impacts on VMEs.

The annual Random Stratified Trawl Survey was undertaken in Division 58.5.2 in the vicinity of Heard Island during April-May 2011. This paper provides a preliminary assessment of mackerel icefish population structure, abundance and yield for the area of Division 58.5.2 to the west of 79^o 20’ E using standard CCAMLR methods. A pattern of 5 year classes being present simultaneously in the population, which was first detected in 2011, has persisted, however a large cohort of fish, estimated to be 2+, dominate the population in both abundance and biomass. A new length weight model is estimated which indicates fish in the 2012 survey were heavier on average than predicted by the previous length weight model. Sensitivity testing indicates that the approach of using the lower 95th percentile of the biomass estimate effectively accounts for uncertainty in growth and mortality, relative to scenarios using the mean biomass estimate. A catch limit of 679 t is recommended for the 2012/13 season.

The annual random stratified trawl survey at HIMI was completed in March to May of 2012. A codend liner, usually in place for the entire survey, was used only in the icefish strata, Gunnari Ridge, Plateau Southeast and Plateau West, and in a repeat of the hauls in Ground B. Catches in this RSTS were generally high in comparison with previous years, with the total fish catch being twice as high as the long term average (2006-2011). The catches of Dissostichus eleginoides for 2012 were twice those of 2011 and about one and a half times higher than the long-term average. Champsocephalus gunnari catches were twice the long term average. Catches of all the major fish species were higher in 2012 than the long-term average, except for those of Channichthys rhinoceratus which was 10% lower. The catch of Lepidonotothen squamifrons was four times higher than the average. A slightly higher than average weight of skates was also taken in 2012.

The catch of invertebrates in the 2012 survey, 4 t, was about half that of the long term average. The most abundant taxa in the catches were poriferan sponges, medusa (jellyfish), anemones and corals. Catches of individual taxa were generally much lower than in previous years, except for coral, with a reported catch of 0.6 t, being significantly greater than the long term average (50 kg). A possible explanation for the generally lower catches is that the smaller mesh codend liner was not in place for most of the survey.

Lengths and biological measurements were taken for many thousands of fish, 1095 pairs of otoliths collected and 524 toothfish were tagged.

The unicorn icefish, Channichthys rhinoceratus and the grey rock cod, Lepidonotothen squamifrons are among the most common species bycaught in the toothfish and mackerel icefish trawl fisheries at Heard Island and McDonald Islands (HIMI) (Division 58.5.2). These species are rarely taken in the longline fishery, but form the most abundant bycatch of the trawl fisheries after skates (Bathyraja spp.). In the toothfish fishery both species are taken, whereas C. rhinoceratus is much more common in the bycatch of the mackerel icefish fishery than L. squamifrons. Over the 15 years of the fisheries, approximately 20 t of C. rhinoceratus and 9 t of L. squamifrons were caught annually. The commercial and research data analysed in this study shows that both species are widespread over the plateau in waters of <1000 m. In most areas they occur at densities <100 kg.km^-2, but each species has separate areas of aggregation where densities exceed 1 t.km^-2. The annual take of these species is well below the precautionary bycatch limits set by CCAMLR, move-on rules apply, and a substantial part of their distribution occurs within the HIMI Marine Reserve, and therefore current bycatch levels are likely to be low risk. However, we intend to conduct a study of the population dynamics of these species to further evaluate the performance of current management measures.

Ocean circulation has been identified as a major process controlling the distribution of biological material in marine systems.  Large-scale transport by the Antarctic Circumpolar Current (ACC), the Ross and Weddell Gyres, and the Antarctic Coastal Current can promote spatially complex population structure in the Southern Ocean through advection.  Antarctic silverfish (Pleuragramma antarcticum), a pelagic, neutrally buoyant notothenioid fish species, are distributed around the shelf systems of Antarctica and are considered an important species rivaling krill as prey for many birds, seals, whales, and other fish. We asked whether silverfish are distributed in independent, discrete populations along the shelf systems of the Southern Ocean or whether the large-scale circulation has led to connectivity among populations.  Hypotheses were tested by measuring the chemistry, trace elements and stable isotopes, in silverfish otoliths, and comparing the chemistry with simulated particle transport using a high resolution circulation model.  The results showed strong heterogeneity indicating four separate populations: i) in the Ross Sea, ii) on the southern Antarctic Peninsula in Marguerite Bay and off Charcot Island, iii) off Joinville Island, and iv) around the South Orkney Islands. This suggested that silverfish are not transported by the ACC, shelf processes on the West Antarctic Peninsula (WAP), or along the Weddell Front.  Using the circulation model, we built spatially explicit predictions of advective supply to areas along the WAP, and examined how interactions between silverfish life history and the Antarctic Coastal Current may structure assemblages over the continental shelf.