CCAMLR

The ASPM assessment of the Prince Edward Islands toothfish resource by Brandão and Butterworth (2005) that permitted annual fluctuations about a deterministic stock-recruitment relationship is updated to take account of further catch, GLM standardised CPUE and catch-at-length information that has become available for the years 2005 and 2006. The assessment allows for a second fleet to accommodate data from a pot fishery that operated in 2004 and 2005. Updated biological parameter values for Subarea 48.3 are incorporated and lead to less optimistic results. The resource is estimated to be at about 40% of its average pre-exploitation level in terms of spawning biomass. It is suggested that it would be prudent to restrict annual legal catches to 500 tonnes or less, unless a large proportion of the catch is to be taken by pots (which avoid the cetacean predation associated with longlining). Specific issues raised at WG-FSA-SAM 2006 about this assessment are addressed.

In this paper we analyse data on the distribution of catches, the size of krill and bycatch of larval fish taken by the Saga Sea fishing with continuous trawls in 2006, and the Atlantic Navigator which fished in 2005 using both continuous trawling and conventional trawling. The sizes of krill caught by both conventional and continuous trawling techniques appear to be very similar. Although there are few data for comparison, the bycatches of larval/juvenile fish in the continuous trawl also appear to be the same as those from conventional trawls.

Two vessels, the Argos Helena (UK) and the San Aspiring (New Zealand), fished around the South Sandwich Islands (Subarea 48.4) in 2006 following one vessel, the Argos Helena, fishing in 2005. During this time animals were tagged in order to continue a mark-recapture programme started in 2005 to assess the toothfish population. We present an analysis of the catch and tagging data and a proposal for continuing research in this area in 2007.

This paper presents new results and estimation bias analyses for a Bayesian mark-recapture model, applied to the South Georgia toothfish tagging data. The updated model incorporates the data from all the analysed cohorts in the estimation routine, and we examine the potential estimation bias, using simulated data. The results confirm earlier findings, with lower estimates of natural and fishing mortality, for the given age range, than are assumed and predicted by the current assessment, respectively. Also, we find some apparent estimation bias, which is similar, but lower than is seen in similar mark-recapture models of this type. Given the increasing number of releases and recaptures in all the current tagging programs, we suggest that this type of modelling approach can serve as both a useful tool for estimating key parameters like natural mortality, as well as a method for comparing parameter estimates of the exploitation rate predicted by the full assessment models, currently being implemented.

1. As requested by SAM-2006, we provide a reference assessment for toothfish in Subarea 48.3. We provide sensitivity runs that include revised maturity and catch weighted proportion at length data. We additionally provide a sensitivity run which examines uncertainty in IUU catches from 1995, as requested by the Joint Assessment Group.
2. Also as requested by SAM-2006, we provide a review of CPUE trends and construct a model that uses age-based observations, specifically estimates of catches at age derived from otolith examination of a random sample of the catch.
3. All the assessments suggest that the SSB is now at or close to 50% of its unexploited level. Simple MPD projections suggest that catches in the region 3300 – 3600 t would be consistent with CCAMLR decision rules.

Minimizing the time gear is available to seabirds by increasing the sink rate of longlines is one method to effectively reduce seabird catch rates. We fished normal unweighted longlines (UW) and integrated weight longlines (IW-50 g/m) with and without paired streamer lines (PS) in seabird mitigation trials targeting Pacific cod in the Bering Sea, Alaska, USA. Our objective was to evaluate IW as a practical seabird mitigation device using multiple criteria: catch rates of seabirds and fish, seabird behaviour, sink rate of the gear and breaking strength.
All mitigation technologies dramatically decreased seabird bycatch rates while having little to no effect on fish catch rates. Sink rate measurements demonstrated that IW longlines reduce risk to seabirds by minimizing the distance astern that birds have access to sinking baits (~50% reduction). This study and our data from 2000 and 2002, also clearly show that the sink rate varies dramatically by vessel.
Through independent tests, we determined that IW breaking strength is 5% less than UW for new gear and 13% less for gear fished for five months; however, this difference did not manifest itself in the context of commercial fishing as increased line separations.
We conclude that IW longlines deployed with paired streamer lines is the best mitigation practice available for autoline longline systems and make recommendations for modifications to CM 24-02 and CM 25-02.

The biennial South Georgia Groundfish Survey was undertaken in January 2006 on board the FPRV Dorada. Sixty-three random trawls were undertaken, with 12.3 tonnes of mackerel icesfish caught, which produced biomass estimates of between 91000 and 117000 tonnes. The catches were dominated by fish of 20-25 cm (putative 2+), but unusually large numbers of fish of ~ 50 cm were also caught to the N and NE of South Georgia. Fish in spawning condition (Stage IV) were found at the eastern end of the island and at Shag Rocks. Spent fish were found throughout the survey area, suggesting that spawning may occur from December. The diet of icefish was dominated by Antarctic krill (82 % by weight), with Antarctomysis sp. (A. ohlini and A. maxima) and Themisto gaudichaudii the other main prey species Toothfish catches (548 kgs) were dominated by the same single cohort that was first detected as putative 1+ fish in 2003. The toothfish diet was dominated by Patagonotothen guntheri. Acoustic observations were made in association with both demersal and pelagic data and will help discriminate icefish from other scatterers.

We satellite tracked four adult southern giant petrels (Macronectes giganteus), four adult northern giant petrels (Macronectes halli), and two fledgling chicks of each species during the summer of 2005-06. Adults of both species spent time in areas under CCAMLR jurisdiction. Adult southern giant petrels spent 37% of their overall time at sea in statistical sub-area 58.4.1, and 14% in sub-area 88.1, extensively utilising these regions during long foraging trips to the ice-edge zone during incubation. Northern giant petrels were tracked during chick rearing. While 3 of four tracked birds did not enter CCAMLR waters, one bird undertook a longer trip into division 58.4.1, where it spent 6% of its overall time at sea. We consider it possible that this trip may have been indicative of trips undertaken earlier in the season (prior to hatching). Both southern and northern giant petrel fledglings crossed the Pacific Ocean, travelling east towards the South American Continental Shelf. Southern giant petrel chicks took a more southerly route, traversing statistical subareas 88.1 and 88.2 along this course, while the more northerly route taken by northern giant petrel chicks did not take them into CCAMLR waters.

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, CPUE, catch-at-age and tag-recapture data from New Zealand and all vessels. The updated reference case resulted in a slightly higher estimate of initial biomass than in 2005, though this was mostly due to a different assumption of the level of natural mortality (M=0.13 y-1), and the revised length-weight and growth relationships. There was little difference between model runs using the New Zealand tag-release and recapture data and the all vessel tag-release and recapture data (B0= 10 300 t and 9 530 t respectively).
Model fits to the data were adequate, with the tag-release and recapture data providing the most information on stock size, but none of the data sets had much information about the maximum size of the stock. Monte-Carlo Markov Chain (MCMC) diagnostics suggested no evidence of non-convergence. Reference case MCMC estimates of initial (equilibrium) spawning stock abundance (B0) were very uncertain, with the median estimated as 10 300 t (95% credible intervals 5 340–25 210 t), and current (B2006) biomass estimated as 91.4% B0 (95% C.I.s 83.4–96.5%).
Estimated yields, using the CCAMLR decision rules, were estimated to be 352 t for the reference case and 366 t for the all vessels case, assuming a future fishing selectivity equal to the maturity ogive.