CCAMLR

Through generalised linear models, the gonadosomatic index of Antarctic toothfish in the Ross Sea region was shown to vary with latitude, length and month. Limitations of the scientific observers’ staging data were highlighted and GSI is recommended as a better indicator of maturity status in Antarctic toothfish. Reports of histological analyses of a small number of Antarctic toothfish showed vitellogenic fish with low GSI values, in some cases below 1%. In most studies there was little if any difference in the GSI values of fish at different maturity stages. However these studies were carried out on a limited number of fish, most of which were from the southern area, therefore not on obviously spawning fish. Histological analysis using a hindcasting assessment of 683 samples collected in December to February showed most fish on the shelf had not spawned that year, most fish in the north had spawned and the shelf contained a mixture of fish that had spawned or not, with length at 50% maturity of about 137cm. The equivalent GSI at 50% maturity was in the range of 1.1 to 1.4%. These results were used to estimate GSI thresholds for fish that had spawned in the previous season; which were set at 1% and 1.5% GSI, but are limited to only female fish in the sampled areas for December through February. Based on both histology and GSI data, most fish found in the northern areas, about a third of those found in the slope areas and very few of those found in the southern areas had spawned. As the GSI values of fish caught in the northern area were never very low, it is expected that all the fish in the northern areas spawn every year when in that area. If they were resting in the north it is expected residual GSI would be lower. Conversely, as the GSI values of fish caught in the south were very low, fish caught there are not expected to spawn in the current year, nor are they expected to have migrated back from the north; or their residual GSI would be higher, in the order of 1% or more. Therefore any movement would have to be between the north and the slope areas, with only a proportion of fish coming back to the slope since only a third there are mature. Any other movements would have to be outside of the fishing season, for example a yearly northern migration during spawning season only. Lengths at 50% maturity of Antarctic toothfish were calculated for each SSRU or area in the Ross Sea region; they varied from about from 89 to 150cm for females and 36 to 184 cm for males, from north to south respectively. Length at 50% maturity was also calculated for female fish from the slope, which was similar to the value calculated from GSI of 137cm. Uncertainty in the oocyte development cycle may create biases in different histological assessment methods which may influence estimates of length at maturity or GSI thresholds. Length distributions are also known to be spatially heterogeneous. A population-wide length at 50% maturity can therefore not be determined without the help of a spatially-explicit population model and any length at maturity value should be treated with caution. Gonadosomatic index has shown promise as a potential index of Antarctic toothfish maturity. However, further work is recommended in order to improve the current knowledge of toothfish maturity...continued

This paper documents recent and ongoing developments in New Zealand’s Exclusive Economic Zone that are relevant to the work of WG-IMAF. These developments include refinement of the requirement of fishing practices to reduce seabird bycatch by surface longline vessels and the introduction of regulations to reduce seabird bycatch by bottom longline vessels. The bycatch of seabirds by fisheries within New Zealand’s EEZ in recent years that either breed or forage within the CCAMLR convention area (Convention seabirds) is detailed. This paper includes a tabulation of observed captures by species and bycatch estimations by fishery for all seabirds. This paper also contains description of the recent and ongoing seabird mitigation trials that are underway in New Zealand. These include trawl offal management techniques, such as batching and mincing, and the blue dyed bait trial for surface longline fishing. We also discuss a new version of the Productivity-Susceptibility Assessment technique for risk assessment and its potential for use in the CCAMLR context.

1. So far a total of 929 D. eleginoides have been tagged and released at Subarea 48.4 and 25 tagged fish have been recaptured, including 23 during the latest season. 2. A preliminary assessment of D. eleginoides stock size in the Northern Area of 48.4 estimates a vulnerable biomass of between 1,000 to 2,000 tonnes. 3. The UK proposes to continue the mark-recapture experiment in Subarea 48.4 in the 2008/09 fishing season so as to allow for a full assessment of D. eleginoides stock size in the Northern in 2009. 4. Additionally, the UK proposes to commence a mark-recapture experiment in the southern area of Subarea 48.4. This has the aims of providing the data required for assessments of the population structure, size, movement and growth of both Dissostichus eleginoides and Dissostichus mawsoni in the southern area of 48.4. 5. A catch limit of 75 tonnes is proposed for D. eleginoides the northern area, where D. mawsoni will remain a bycatch species. A combined catch limit of 75 tonnes is proposed for D. eleginoides and D. mawsoni in the southern area. 6. It is proposed that catch limits are introduced for bycatch species in the Northern Area of 48.4, set at 16% Macrourus spp. and 5% Rajid spp of the catch limit for Dissostichus species. In the Southern Area, catches taken at these proportions would trigger a move-on rule. 7. The UK has submitted a complementary proposal to IMAF to amend CM 24-02, to bring the mitigation requirements for 48.4 into line with the IMAF risk assessment, such that daytime setting would be permitted if bottle tests are undertaken, and the fishing season is extended to run from 1 December to 30 November.

The provisions for fishing season and mitigation measures in Conservation Measure 24-02 (2005) regulating fishing in Subarea 48.4 do not currently conform to the IMAF Risk Assessment advice given in CCAMLR-XXIV/BG/26. The UK proposes text that should be added to CM 24-02 to bring the measure into line with the Risk Assessment, which would allow fishing outside season April – September if it is conducted in accordance with CM 24-02. A small change, to recognise 48.4, is required to the first paragraph of CM 24-02.

1. Paired trials were undertaken to compare toothfish catch rates and by-catch composition on Spanish system longlines with those on ‘trotlines’ fitted with cetacean exclusion devices (cachalotera nets). 2. CPUE (kg/1000 hooks and kg/set) was greater for trotlines than Spanish system lines when cetaceans were present. This difference increased with increasing abundance of cetaceans highlighting the potential of this gear to reduce cetacean depredation. 3. By-catch to catch ratio was greater by number for Spanish system lines when cetaceans were present during hauling. 4. A large percentage of skates caught on trotlines fitted with cachalotera nets were dead or suffered life-threatening injuries. Toothfish were also in much poorer condition to the extent that they were unsuitable for tagging and releasing. 5. Consequences of a move to trotlines would therefore lead to: i. Some positive benefits by reducing cetacean depredation. ii. Significant problems in understanding the meaning of CPUE for trotlines, and consequently in understanding CPUE trends unless there is an extended period of overlap between Spanish system, autoline and trotline use. iii. Potential increases in skate and toothfish injury, leading to problems with skate discard survivorship and achievement of tagging goals.

Age and growth of spiny icefish, Chaenodraco wilsoni, were investigated using counts of annual growth increments from sagittal otoliths. Samples were collected during research surveys by benthic trawl carried out off Joinville-D’Urville Islands (Antarctic Peninsula) in February-March 2006 and January 2007. A total of 218 specimens were selected for the study, consisting of 120 females and 98 males. The age of fish was estimated by counting annuli on transverse sections obtained by grinding and polishing whole otoliths embedded in epoxy resin in moulds. The precision of age estimates between readings was tested applying both the average percent error (APE) and the coefficient of variation (CV). The estimated age range was 1-5 for males and 1-4 for females of C. wilsoni. Applying the von Bertalanffy growth function to the age-length data, a growth curve was obtained for each sex. The estimated values of VB growth parameters L∞ and k were respectively 32.7 cm and 0.81 for females and 32.7 cm and 0.68 for males. Age at sexual maturity was estimated to be about 2 years for females and 2.5 years for males. Compared to other channichthyids, the spiny icefish exhibited a high growth rate until they reached sexual maturity and a considerable short life span. The fish population of C. wilsoni caught in the studied area consisted mainly of adult specimens between 1 and 3 years of age, with very few older fish.

Catch limits currently in place for macrourids in Subareas 88.1 and 88.2 are defined as being equal to 16% of the catch limit of Dissostichusspp. in these subareas. The 16% was based on the ratio of the by-catch limit for macrourids to the catch limit for Dissostichus spp. in Division 58.5.2 in 2002/03 (CCAMLR-XXI, para 11.53). The bycatch limit for macrourids in Division 58.5.2 had in turn been based on the extrapolation of catch rates from a trawl survey of M. carinatus on Banzare Bank. Two bottom trawl surveys (IPY_CAML and BioRoss) have recently been carried out by New Zealand on parts of the continental slope of the Ross Sea in SSRU 88.1H. Whitson’s grenadier (M. whitsoni) was the most abundant species caught during both surveys, with mean catch rates in some strata being considerably higher than those recorded for M. carinatus on Banzare Bank. Because the strata covered by the New Zealand surveys represented only 25% of the continental slope of the Ross Sea, the data could not be used to calculate a biomass estimate for the entire slope directly. Instead, indicative biomass estimates for the rest of the area were made by extrapolating the more recent IPY-CAML survey catch rates across the entire slope. The extrapolations were carried out in two ways: (i) assuming M. whitsonidensities were constant across the entire slope and (ii) assuming M. whitsonidensities were proportional to the commercial macrourid CPUE from the fishery. The uncertainty of the extrapolated biomass was estimated through a bootstrap procedure. Indicative estimates of M. whitsoni biomass for the Ross Sea slope ranged from 26 892 t (cv = 29%) to 41 823 t (cv = 28%) depending on assumptions. Applying estimates of gamma (γ) =0.01439 and 0.01814 gave indicative estimates of yield in range 386–602 t and 487–759 t respectively. These indicative yield estimates provide tentative support for the bycatch limit of 374 t currently in place for M. whitsoni on the Ross Sea slope (SSRUs 88.1H–L and 88.2A–B).

during February and March 2008 as part of the International Polar Year. The main aim of the survey was to carry out a Census of Antarctic Marine Life in this regionfocussing on sampling the pelagic and benthic habitats on the shelf, slope, abyss, and seamounts in the Ross Sea region (CCAMLR Subarea 88.1). The shelf and slope were stratified by depth and at least three random trawls completed in each stratum. The trawl survey of the shelf area focused mainly on icefishes, notothens, Antarctic silverfish, and glacial squid whilst the survey of the slope area was designed to target the macrourid M. whitsoni as it is the main speciestaken as bycatch in the toothfish fishery, and the most important prey item in the toothfish diet. Biomass estimates and catch rates by station are presented for the eight most abundant teleost species along with scaled length frequencies of those species.

An ongoing mark and recapture experiment for skates was commenced by New Zealand fishing vessels in the Ross Sea (CCAMLR Subareas 88.1 and 88.2) in 1999/2000. Based on preliminary results from this programme showing an unquantifiable degree of the survivorship of returned skates, the CCAMLR Scientific Committee gave approval in 2004 for licensed vessels to cut live skates from the line (while in the water) as an alternative to either retaining all aboard or discarding dead skates, as a skate mortality mitigation measure. Initial opinion when the skate tag and recapture programme was first implemented was that some form of in-water release would be the ideal to improve survival chances. Subsequently a number of problems have become evident with this method. The Working Group on Fish Stock Assessment (WG-FSA) noted in 2004, when recommending skate release as an option to the Scientific Committee, that it might be difficult to detect tagged rays skates if they are cut off at the sea surface rather than being brought on board. The WG-FSA recommended that should the tag identification rate be low, a relaxation of the requirement to cut all rajids from the line on specified vessels and/or for specified time periods be adopted. Consistent with this recommendation, commencing in the 2006/07 season, New Zealand fishing vessels were given permission by the New Zealand government as part of the national research plan, to trial an alternative technique. The new method (for simplicity in this paper called Method 2), required the crew to bring the skate aboard carefully, remove the hook and snood, take and record biological (and potentially other meristic data) from a sub-sample when possible, and release live skates in a timely fashion in a manner most likely to ensure survival. The advantages of this method have been the more effective scanning of the captured skate for existing tags, greater accuracy in the assessment of ideal candidates for live release, the capability to collect supporting meristic and biological data for release candidates, correct species and sex identification, and the ability to more accurately place tags on the skate body – improving the quality of skate tagging. Following favourable results aboard by New Zealand vessels using Method 2 during the 2006/07 season and following a discussion and recommendation from WG-FSA 2007 an amendment to CM 33-03 (2007) was made stating that: ‘Unless otherwise requested by scientific observers, vessels, where possible, should release skates and rays alive from the line by cutting snoods, and when practical, removing the hooks’. This paper documents the evolution of skate release methods both for tagging and for live release for New Zealand autoline vessels. It describes the current Method 2 systems in use for two New Zealand vessels. A brief analysis based on data collected from the skates treated in this manner during the 2007/08 fishery in the Ross Sea is included in the document to highlight the additional advantages in data collection possible using Method 2. This paper is intended to inform WG-FSA with additional information collected over the last two fishing seasons in preparation for the ‘Year of the Skate’ prior to a final decision on skate release protocols. A proposal is made for a comparative experiment using two tag types to inform a final decision on a standardised CCAMLR tag type for Rajid mark and recapture programmes carried out within CCAMLR Dissostichus spp. exploratory fisheries.

Ontogenetic, inter-annual and regional variations in diet were investigated in mackerel icefish, Champsocephalus gunnari, in three successive summer seasons around South Georgia. Stomachs from 2239 icefish (130-560 mm total length) were examined. A bootstrapping technique was used to calculate confidence intervals for an index of relative importance (% IRIDC) of prey categories. Diet varied significantly between years and age classes but there was little regional difference in diet. In general diet was dominated by krill, Euphausia superba and by the amphipod Themisto gaudichaudii. Smaller (younger) fish tended to prey on a higher proportion of T. gaudichaudii and small euphausids such as Thysanoessa sp. and took smaller quantities of E. superba. In a season of poor krill availability (summer of 2003-2004) the proportion of krill in the diet, stomach fullness and fish condition (indicated by length-weight relationships) were significantly lower than in the other summer seasons. The poor krill season was followed by a large reduction (> 80%) in the estimated annual biomass of C. gunnari the following year (2005). This may have been a result of mortality of age 2+ and 3+ fish, which were more krill dependent than 1+ fish. Younger fish appear to have survived, leading to an increase in the estimated population biomass in 2006.