CCAMLR

During a longline survey on BANZARE Bank (Division 58.4.3b) in May 2008, an in-line video camera recorded the interactions between the main line and the benthos during setting and hauling. Footage revealed that after setting the line moved very little, despite cross currents. During hauling, the section of line visible to the camera swept sideways, interacting with sessile benthic organisms before leaving the seafloor. Based on an estimate of the rate of sideways movement, the total time sideways movement occurred before the line lifted off the seafloor, the total length of the line and the total time to retrieve the line, we estimate that up to 0.122 km2 of seafloor was swept by the line while it was being retrieved. Consequently the area of seafloor affected by demersal longlines is shown to be comparable with that of demersal trawls. We also provide evidence that benthic fauna that are vulnerable to longline gear may not be retained at the surface, but can fall off the longline hooks before reaching the surface. Hence surface observations of some benthic bycatch are likely to underestimate of the actual level of interaction.

In May 2008, a randomised longline survey, consisting of 15 standardised sets over 2 strata covering areas of commercial fishing activity was conducted from aboard the Australian flagged longliner FV Janas BANZARE Bank within CCAMLR Statistical Division 58.4.3b. Catch rates were very low, ranging between 0 and 225 kg.1000 hooks-1, consistent with toothfish being depleted to low densities across the majority of the surveyed area. Catches of toothfish consisted of both species of Dissostichus, with D. mawsoni found across the entire survey area while D. eleginoides was mainly found on the shallow areas on the western part of the Bank. Data on size distribution and size at maturity indicate that the D. mawsoni population is almost entirely large mature fish with a bias towards males, which are shown to mature at a smaller size than females. As the population does not contain juvenile lifestages, we hypothesise that the fish encountered originate from nearby populations in east Antarctica where juveniles of both sexes are caught. The D. eleginoides population consisted of approximately equal proportion of males and females, and few mature individuals were detected. Contrasts in the population characteristics of the two species of Dissostichus indicate that species specific management should be considered for this area. Data on major bycatch species Macrourus carinatus and Raja taaf, including size distribution, sex ratio and size at maturity, are also presented.

A survey of mackerel icefish, Champsocephalus gunnari, was undertaken in Division 58.5.2 in the vicinity of Heard Island in July 2008 to provide the information for an assessment of short-term annual yield in the 2008/2009 CCAMLR season. This paper provides a preliminary assessment of yield for the area of Division 58.5.2 to the west of 79o 20’ E using standard CCAMLR methods. The strong year class detected in last year’s survey is now fully recruited as the 2+ cohort, and dominates the population.

Tagging of the three species of skates taken as by-catch in the Heard Island and McDonald Islands toothfish fishery, Bathyraja eatonii, B. irrasa and B. murrayi has taken place over the last eight years with nearly 6,000 skates tagged. The recapture rate has been low, less than one percent. Movement after tagging shows that these fish are recaptured only short distances from where they were released. The mean distance between release and recapture was 4.8 nautical miles, even though the majority had been at liberty for several years. In order to enhance the ease of identification of skates and to ensure good quality of the data, one page identification sheets were developed for each species which will be used by observers on board fishing vessels.

A non-hierarchical seabird identification key has been produced to improve the identification of seabirds that come into contact with or are seen in the vicinity of fishing operations in the Heard and McDonald Islands region. This paper provides the background and reasoning behind the development of this key.

This paper presents a framework to assess and quantify the likely cumulative impact on potential VMEs from bottom fishing activity. The approach has been designed to facilitate standardized application across gear types and areas to allow comparisons between fisheries employing different bottom impacting fishing methods. Details of the New Zealand preliminary assessment using this framework are available in document CCAMLR XXVII-26. This paper illustrates the utility of the standardized approach and provides a methodological template for possible wider adoption within CCAMLR or elsewhere. Specific examples from the New Zealand assessment are provided for illustrative purposes.

The New Zealand fishing company Sanford Limited has initiated a research project to assess the effectiveness of artificial bait as a means of reducing the incidental catch of Macrouridae (rattails) and other bycatch species in the toothfish autoline longline fishery. A company vessel carried out some initial experimental work in the western Ross Sea (CCAMLR Statistical Subarea 88.1) in 2007 to assess basic requirements such as ease of use, bait longevity on the hook, and catchability (defined here as a measure of a baited hook to attract and catch a given species of fish) of target and incidental species. This is a preliminary report based on subsequent trials carried out aboard the vessel San Aspiring operating in the waters of South Georgia and the South Sandwich Islands (CCAMLR Statistical Subareas 48.3 and 48.4) during early 2008. During the period 12 April and 14 May 2008 seventeen trial lines were set in both Subareas; a total of 137,000 experimental hooks. The evaluation was based on alternating sections (magazines or mags - there are approximately 1024 hooks per magazine on San Aspiring) of line with control mags baited with squid, the vessel’s preferred bait for toothfish and alternating mags using an artificial or reconstituted bait (Norbait 800C™). Results from Subarea 48.4 indicated that catches of both the target Patagonian toothfish and bycatch of Macrourus whitsoni were reduced using Norbait when compared to the conventionally used squid bait. The reduction in rattail catches however was proportionally much greater than that of the toothfish and points to a potentially useful means of limiting Macrourid bycatch. This is similar to results obtained by another Sanford vessel San Aotea II in the Ross Sea in early 2007. This was not the case in Subarea 48.3, the other area where trials were carried out. In this area although Norbait reduced the incidental bycatch of Macrourus, an overall virtually identical reduction in the target catch of Patagonian toothfish negated any benefit. These reductions effectively meant an increase in fishing effort and time to achieve the same target catch result. Other than the obvious geographical differences between the two trial Subareas there were differences in both the Macrourus species caught and the size of toothfish caught in each. These factors may individually or in combination provide some explanation for the observed dissimilarities. There are a large numbers of variables involved in such comparative trials making definitive analysis difficult and precluding clearly defensible results. In such cases the collection of large amounts of base data in all circumstances and environments is the only way of reducing uncertainty and understanding the variability. For this reason it is necessary to stress that these results are preliminary and further work is indicated. Although increased and continuing data collection is an ideal, it must be highlighted that there are both clear and concealed costs involved in undertaking such trials. Obvious expenses are incurred in the purchase and transport of the trial bait. There are however hidden costs incurred in the additional time on the grounds and additional gear deployed to catch the same amount of fish when the bait used is less effective than the current standard.

We examined a methodology for assessing potential risk of interactions between fisheries and species of special interest (seabirds and marine mammals) by applying a Productivity-Susceptibility Analysis to a data set of species distribution, biological information and fishing effort. This type of Level Two Ecological Risk Assessment (ERA) has been used across fishery management regimes to identify areas or species requiring additional management or monitoring. Our study indicated that the risk of species interactions is clustered with greatest likelihood in a few species, where there is particularly strong overlap between fishing effort and species ranges. We tested the sensitivity of the analysis to changes in weighting of distributional density, fishing data type (area or point data), and to adding a factor of population. We found the outcomes of the PSA analyses were robust to these effects. However, adult survival rates did influence the rankings, and were identified as a key parameter requiring careful estimation. Relative risk rankings within the longline, trawl, troll and set net fisheries examined indicated that Procellaria petrels, the coastal Hector’s dolphins (Cephalorhynchus spp.), giant petrels (Macronectes spp.), Pterodroma petrels, and some albatrosses carried the highest relative risk. For setnet fishing, dolphins and shags were also ranked relatively highly. We examined the small statistical areas where most risk across all species applied cumulatively. We identified areas where there is greatest potential for non-target take to be occurring and which are therefore candidate areas for intensified observer monitoring and mitigation of risks.

The aim of this paper is to review the recent management of the Ross Sea toothfish fishery (including the 3-year experiment), to identify key operational and research objectives for the fishery over the next 5–7 years in relation to Article II of the Convention, and to develop an operational framework to achieve those objectives. The paper focuses primarily on Antarctic toothfish, as catches of Patagonian toothfish are negligible, and covers Subareas 88.1 and 88.2. We begin by summarising the operational management and conduct of the fishery up to the 2004–05 fishing year (prior to the start of the 3-year experiment). This includes the reasons why the 3-year experiment was initiated and the key objectives of the experiment. We then go on to summarise the operational changes which formed the framework of the 3-year experiment, and to review the success and/or any problems associated with each of those changes. Next we identify key operational and research objectives for the fishery over the next 5–7 years in relation to Article II of the Convention. As part of this process we identify uncertainties in our current knowledge which need to be addressed to fulfil the requirements of Article II. These include, for example, uncertainty in the biological parameters and stock assessment of Antarctic toothfish, uncertainty in its ecological relationships with predators and prey, and uncertainty over other ecosystem effects of fishing. Finally, we provide recommendations on the development of an operational framework for the fishery.

Skates are an important bycatch of the toothfish fishery in the CCAMLR area and have been identified as priority taxa for which assessments of status are required (e.g., SC-CCAMLR XXIII 2004, paragraphs 4.172, 4.177 and 4.199). While Dunn et al. (2007) and Agnew et al. (2007) have developed preliminary assessment models for skates, they also highlighted that further information was required before a full assessment can be carried out. In 2007, WG-SAM recommended (CCAMLR XXVI WG-SAM paragraph 8.10) a review of data requirements and a “Year of the Skate” for 2008–09 whereby data collection effort on bycatch will be concentrated on skate species in that year in order to inform a full skate assessment. This paper discusses improvements to the fishery derived data that may be required to better inform an assessment of Ross Sea skates. We propose options for the appropriate collection of such data from the fishery and a revised skate tagging protocol. These changes were piloted in the 2007–08 season by a subset of vessels fishing in the Ross Sea. The results from the pilot study are useful to inform modifications to data collection systems that are required in 2008–09, for the “Year of the Skate”. Note that we do not consider other information requirements such as determining biological parameters.