CCAMLR

The Southern Ocean is a globally important marine region, providing a range of ecosystem services which support human life, health and well-being, including the provision of marine living resources, and the regulation of global climate and sea level. Assessing ecological processes in terms of the services they provide translates the complexity of the environment into functions which can be more readily understood, for example by policy-makers and non-scientists. Ecosystem-based management requires the consideration of a wide range of objectives, and the language and concepts of ecosystem services may help to provide a common currency for balancing these objectives. However, the importance of the Southern Ocean ecosystem is generally under-represented in assessments of ecosystem services at the global scale, reflecting the spatial separation of Southern Ocean ecosystem services and their beneficiaries. Equally the concept of ecosystem services is not generally used within the Antarctic Treaty System (ATS), creating a potential disconnect between global and regional policy. Nevertheless, decision making processes within the ATS are in many ways pre-adapted to deliver evidence-based policy which takes the current and future value of multiple ecosystem services into account. Also, much of the evidence gathering work which has been conducted to support decision making in the Antarctic context could relatively easily be adapted to fit an ecosystem services evaluation framework.
Here we provide a brief review of Southern Ocean ecosystem services, and outline preliminary work towards an assessment of their distribution, status and value. The benefits of assessing Southern Ocean ecosystem services in this way include (i) emphasising their global importance; (ii) facilitating comparisons of individual services across the ATS; (iii) facilitating consideration of the full suite of ecosystem services under the ATS; (iv) allowing comparability with global governance frameworks. This has particular relevance to the work of CCAMLR, given its responsibility for the maintenance and sustainable provision of living resource services from the Southern Ocean ecosystem, and the increasing need to communicate its role in ecosystem based management to a global audience of stakeholders.

In June 2010 the CCAMLR ASAM working group examined the acoustic methodology applied to the CCAMLR 2000 synoptic survey data to generate a quality checked krill biomass for area 48 following the acoustic protocol identified in ASAM 2009. The total biomass of krill in the Scotia Sea was estimated from acoustic and net data collected during the international multi-ship krill biomass in the Scotia Sea in 2000 to be 60.3 million metric tonnes. This report aims to document the parameters utilised during the assessment and include additional information of krill density by strata and transect published for the original methodology, but omitted from the ASAM2010 report (SC-CAMLR-XXIX, Annex 5).

 This paper provides a brief status report on the ongoing analysis of KRILLBASE.

One of the main issues in the management of the krill fishery is finding a spatial subdivision of catches that allows CCAMLR to achieve its objectives for both the fishery and the ecosystem. This requires a framework of spatial areas over which catches can be subdivided. WG-EMM has devised an initial framework of small scale management units (SSMUs) in subareas 48.1 to 48.4 based on the spatial structure of the ecosystem. This framework is hierarchical. It recognises a distinction between coastal (shelf and shelf-break) areas and oceanic areas. WG-EMM devised thirteen coastal SSMUs with an average area of 34,690 km2, separating discrete concentrations of land-based predator colonies. WG-EMM also devised just four SSMUs, with an average area of 758,809 km2, to cover the remaining 87% of the four subareas. The oceanic SSMUs are the location of 72% of the estimated krill biomass, 61% of the estimated consumption of krill by predators and 10% of the cumulative total fishery catch in the three subareas (48.1 to 48.3) where the fishery operates.
There is ample evidence of structure in oceanic areas from both ecological and oceanographic studies, and from the concentration of krill catches in limited parts of the oceanic SSMUs. A more detailed application of knowledge about ecological structure is likely to allow more, finer-scale SSMUs to be devised for oceanic areas. This would: (i) allow a greater range of options for the subdivision of catches; (ii) afford oceanic predators a greater level of protection from localised fishery impacts; and (iii) allow more realistic evaluation of management strategies in terms of consequences for both the fishery and the ecosystem.
Krill distribution during the CCAMLR 2000 synoptic survey is an indicator of ecosystem structure. Relationships between krill distribution and physical environmental characteristics could be used to identify units of coherent structure in oceanic areas. British Antarctic Survey scientists are currently conducting this analysis. Because the synoptic survey only provides a single snapshot of structure it will be necessary to further investigate the consistency in this structure over intra- and inter- annual timescales by (a) testing for such consistency in longer time series of data on physical characteristics which have spatial relationships with the distribution of krill and (b) directed survey or fishing effort to test hypotheses about where concentrations of krill are likely to occur in oceanic areas. The fact that the physical influences on this structure are often dynamic (e.g. frontal systems), presents an additional challenge. Nonetheless, it is important to address the disparity between the scales of coastal and oceanic SSMUs to provide CCAMLR with the tools necessary to manage these habitats in a consistent way.

The Generalised Yield Model was used to estimate fishing mortality and spawning stock biomass reference points for the krill fishery in CCAMLR Area 48 consistent with the catch trigger level of 620,000 tonnes. Projections were run with various increased levels of recruitment variability to analyse the sensitivity of the estimates of the reference points to recruitment variability.
The estimates of F and SSB reference points for the krill stock in Area 48 consistent with the catch trigger level are 0.0159 (95 % CIs: 0.00750 – 0.0357) and 97.7 % SSB0 (80 % CIs: 71.6 – 135 %) respectively. In comparison, the F and SSB reference points for the precautionary catch limit of 5.61 million tonnes are 0.186 (95 % CIs: 0.0762 – 0.630) and 75.0 % (47.9 – 113 %) respectively.
The probability of stock depletion increases substantially with increased recruitment variability, though in absolute terms remains negligible. The uncertainty surrounding median estimates of F and SSB reference points for the catch trigger level increases with increased variability in recruitment, though the median estimates are unaffected.

The model terminated prematurely with a 40% increase in recruitment proportion SD. It is likely that this is due to a design feature of the Generalised Yield Model in place to prevent projections from running with potentially significant bias in projected recruitment resulting from poor parameterisation of the beta function.

During April 2011, a multi-national group of scientists with expertise on Antarctic krill Euphausia superba and environmental sciences attended a workshop aiming to evaluate new knowledge on the impact of climate change and increasing fisheries on Antarctic krill and Antarctic ecosystems, and possible repercussions for resource management. The workshop was organised by the Institute of Marine Resources and Ecosystem Studies (IMARES) in the Netherlands, and funded by the European Commission and the Dutch government. The scientific evaluation focused on major agents of climate change, such as ocean warming, sea ice loss, and ocean acidification. It was concluded that the cumulative impact of climate change on krill is probably negative. To be able to account for climate change-induced ramifications on Antarctic krill and ecosystems, the adaptive capacity of the fisheries management of CCAMLR must be enhanced. To achieve this, critical knowledge gaps in the biology and ecology of Antarctic krill need to be closed. Research needs to be intensified on recruitment processes in Antarctic krill, under-ice and benthic habitat use, their capacity to adapt to environmental change, their ecosystem function, as well as the energy demand and food consumption of krill-dependent predators. With respect to CCAMLR’s ecosystem-based management approach, several recommendations were agreed on during the workshop. In particular, it was concluded that current precautionary management measures need to be maintained, until sufficient knowledge exists about the population levels of sustainability. It was further agreed that increasing the efficiency of CEMP is fundamental for a solid science-based management of the fishery.

Two successful experiments on definition of Antarctic krill mortality have been carried out aboard Polish vessel Dalmor II according to the pattern submitted in paper SC-CAMLR-XXVIII/BG/10. Approximately 230 to 130 kg of krill are percolated and punched through per an hour of trawling if the average catch per an hour of trawling is about 8 tons and towing speed is 2.6-2.7 knots. The key parameters of trawling which influence the value of the Antarctic krill mortality rate are: the trawl configuration and the mesh size of its segments; the speed of the vessel and the towing duration.

Combination of conventional and continuous techniques during krill fishery at the Russian commercial vessel "Maxim Starostin" allowed to compare the size structure of the Antarctic krill (Euphausia superba) caught by these two techniques. Results of comparison are discussed in relation to the selectivity of fishing gears and the space-time variability of krill. Differences between size composition of krill caught by conventional and continuous techniques of fishery which could be connect with the trawls selectivity weren’t found. At the same time results of the analysis didn’t indicate that there were no differences between selectivity of conventional and continuous trawls. We assume that the possible influence of differences between selectivity of fishing gears were exceeded by significant space-time variability of Antarctic krill.

A krill net sampling survey was carried out west of the Antarctic Peninsula in January 2011 to collect data on krill distribution, abundance, demography, spawning and recruitment success. The survey was a joint German and US effort. While the US AMLR Survey section covered the area between Elephant Island and the western entrance of Bransfield Strait, the “Polarstern” survey grid followed back to back and extended southwest beyond Adelaide Island/Marguerite Bay. 177 quantitative net samples were taken and analyzed for post-larval and larval krill, Euphausia superba, as well as for salps, Salpa thompsoni.

The results of this survey represent the most complete picture of the spatial distribution of krill abundance, demography, and production on the western side of the Antarctic Peninsula conducted since the late 1980s.

Automated camera systems deployed at Adélie penguin breeding colonies provide daily measurements that allow high resolution temporal availability functions to be estimated, which in turn can be used to correct population estimates for availability bias. However, such frequent data are time consuming and expensive to process, and it is of interest to determine if such data could be subsampled with significant loss of information content. To this end, a simulation study was undertaken to examine how the frequency of sampling attendance at Adélie penguin breeding colonies would affect models of attendance for correcting population counts for potential attendance bias. Generalised additive models of simulated time-series were shown to adequately recover known structure for sampling periodicities up to five days. Most precise estimates of attendance ratios to correct non-optimal population counts for availability bias are obtained from higher frequency sampling, with a trade-off observed between sampling frequency and precision. Subsampling at periodicities of six days or greater did not adequately recover known simulated model structure and cannot be recommended.