FULL ASSESSMENT REPORT Olympic Seafood Antarctic Krill Fishery

Public Certification Report

Olympic Seafood AS

Authors: Geir Hønneland, Andrew I.L. Payne, Lucia Revenga, Sigrun Bekkevold Report No.: 2015-03, Rev. 4 Certificate No.: F-DNV-184887 Date: 2015-08-20

Project name: Full assessment report DNV GL - Business Assurance Report title: Olympic Seafood Antarctic Krill Fishery Det Norske Veritas Certification Public Certification Report AS Customer: Olympic Seafood AS, PO Box 234 Veritasveien 1 6099 Fosnavåg 1322 HØVIK, Norway Contact person: Bjørnar Kleiven Tel: +47 67 57 99 00 Date of issue: 2015-08-20 http://www.dnvgl.com Project No.: PRJC-504869-2014-MSC-NOR Organisation unit: ZNONO418 Report No.: 2015-03, Rev. 4 Certificate No.: F-DNV-184887

Task and objective: The ojective of this report is the initial assessment of Olympic Seafood Antarctic krill fishery against the Marine Stewardship Council Fishery standard: Principles and Criteria for Sustainable Fishing.

Prepared by: Verified by:

Andrew I.L. Payne Anna Kiseleva Independent P1 expert DNV GL GSR

Lucia Revenga Independent P2 expert

Geir Hønneland Independent P3 expert and team leader Sigrun Bekkevold Team member

☒ Unrestricted distribution (internal and external) Keywords: ☐ Unrestricted distribution within DNV GL MSC Fisheries, initial assessment, Olympic Seafood Antarctic krill ☐ Limited distribution within DNV GL after 3 years ☐ No distribution (confidential) ☐ Secret

Reference to part of this report which may lead to misinterpretation is not permissible.

Rev. No. Date Reason for Issue Prepared by Verified by Approved by

0 2015-02-18 Preliminary Draft Report for Client Andrew I.L. Payne, review Lucia Revenga, Geir Hønneland, Sigrun Bekkevold

1 2015-03-18 Peer Review Draft Report Same as above

2 2015-05-12 Public Comment Draft Report Same as above

3 2015-06-30 Final report Same as above

4 2015-08-20 Public Certification Report Same as above

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page i

Table of contents

ABBREVIATIONS ...... 1

LIST OF SYMBOLS AND REFERENCE POINTS ...... 2

LIST OF FISH SPECIES ...... 2

1 EXECUTIVE SUMMARY ...... 4 1.1 Assessment team 4 1.2 Assessment timeline 4 1.3 Scores for separate Principles 4 1.4 Main strengths and weaknesses of the client’s operation 4 1.5 Determination with supporting rationale 5 1.6 Conditions 5

2 AUTHORSHIP AND PEER REVIEWERS ...... 6 2.1 Assessment team 6 2.2 Peer reviewers 8

3 DESCRIPTION OF THE FISHERY ...... 9 3.1 Unit(s) of Certification and scope of certification sought 9 3.2 Overview of the fishery 10 3.3 Principle One: Target Species Background 11 3.4 Principle Two: Ecosystem Background 23 3.5 Principle Three: Management System Background 50

4 EVALUATION PROCEDURE ...... 53 4.1 Harmonised Fishery Assessment 53 4.2 Previous assessments 53 4.3 Assessment Methodologies 53 4.4 Evaluation Processes and Techniques 56

5 TRACEABILITY ...... 63 5.1 Eligibility Date 63 5.2 Traceability within the Fishery 64 5.3 Eligibility to Enter Further Chains of Custody 66 5.4 Eligibility of Inseparable or Practically Inseparable (IPI) stock(s) to Enter Further Chains of Custody 66

6 EVALUATION RESULTS ...... 67 6.1 Principle Level Scores 67 6.2 Summary of Scores 67 6.3 Summary of Conditions 68 6.4 Determination, Formal Conclusion and Agreement 68

REFERENCES ...... 69

APPENDIX 1 SCORING AND RATIONALES ...... 77 Appendix 1.a - MSC Principles and Criteria 77 Appendix 1.1 Performance Indicator Scores and Rationale 78 Appendix 1.2 Conditions 145

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page ii

APPENDIX 2. PEER REVIEW REPORTS ...... 146 Peer Reviewer 1 146 Peer Reviewer 2 158

APPENDIX 3. STAKEHOLDER SUBMISSIONS ...... 185

APPENDIX 4. SURVEILLANCE FREQUENCY ...... 189

APPENDIX 5. CLIENT AGREEMENT ...... 190

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page iii

ABBREVIATIONS

ACC Antarctic Circumpolar Current ASOC Antarctic and Southern Ocean Coalition APA Antarctic Protected Area ARK Association of Responsible Krill harvesting companies ASMA Antarctic Specially Managed Area ASPA Antarctic Specially Protected Area ATS Antarctic Treaty Secretariat CCAMLR Commission for the Conservation of Antarctic Marine Living Resources CEMP CCAMLR Ecosystem Monitoring Programme CITES Convention on International Trade in Endangered Species CPUE Catch per Unit Effort ERS Electronic Reporting System ETP Endangered, Threatened and Protected Species F Fishing Mortality FAO Food and Agriculture Organisation of the United Nations FV Fishing Vessel GSGSSI Government of South Georgia and South Sandwich Islands GYM Generalised Yield Model HACCP Hazard Analysis and Critical Control Point IUCN International Union for the Conservation of Nature IUU Illegal, Unregulated and Unreported Fishing IWC International Whaling Commission LTL Low Trophic Level M Natural Mortality MPA Marine Protected Area MRAG Marine Resource Assessment Group (London) MSC Marine Stewardship Council OCCAM Ocean Circulation and Climate Advanced Modelling project PCL Precautionary Catch Limit PUCL Precautionary Upper Catch Level RBF Risk Based Framework SGSSI South Georgia and South Sandwich Islands SLED Sea Lion excluder device SMOM Spatial Multispecies Operating Model SSMU Small-Scale Management Unit TAC Total Allowable Catch TED Target Eligibility Date UoC Unit of Certification VME Vulnerable Marine Ecosystem VMS Vessel Monitoring System WG-EMM Working Group on Ecosystem Monitoring and Management WG-FSA Working Group on Fish Stock Assessment WOCE World Ocean Circulation Experiment WSSW Weddell Sea Surface Water WWF World Wildlife Fund

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 1

LIST OF SYMBOLS AND REFERENCE POINTS

B0 Pre-exploitation estimate of the stock biomass Blim Minimum biomass below which recruitment is expected to be impaired or the stock dynamics are unknown BMSY Biomass corresponding to the maximum sustainable yield (biological reference point); the peak value on a domed yield-per-recruit curve Bpa Precautionary biomass below which SSB should not be allowed to fall to safeguard it against falling to Blim Btrigger Value of spawning stock biomass (SSB) that triggers a specific management action F Instantaneous rate of fishing mortality Flim Fishing mortality rate that is expected to be associated with stock ‘collapse’ if maintained over a longer time (precautionary reference point) Fmp Management plan target fishing mortality FMSY F giving maximum sustainable yield (biological reference point) Fpa Precautionary buffer to avoid true fishing mortality being at Flim when the perceived fishing mortality is at Fpa M Natural Mortality MSY Maximum Sustainable Yield SSB Spawning Stock Biomass

LIST OF FISH SPECIES

Species English name Bathydraco marri Deepwater Dragonfish Bathydraconidae Chaenodraco wilsoni Spiny Icefish Chaenocephalus aceratus Blackfin Icefish Champsocephalus gunnari Mackerel Icefish Channichthys rhinoceratus Unicorn Icefish Chionodraco myersi Myers Icefish Chionodraco rastrospinosus Ocellated Icefish Chionodraco sp. Icefish nei Cryodraco antarcticus Long-fingered Icefish Cryodraco spp. Electrona antarctica Antarctic lanternfish Euphausia superba Antarctic Krill Eusirus perdentatus Natantian decapods nei Gempylidae Snake Mackerel Gymnodraco acuticeps Antarctic Dragonfish Gymnoscopelus bolini Lanternfish Gymnoscopelus nicholsi Nichol’s lanternfish Gymnoscopelus braueri Lanternfish Icichthys australis Southern driftfish Lampanyctus achirus Lanternfish Medusae Jellyfish Myctophidae Lanternfish Muraenolepsis microps Smalleye moray cod Neopagetopisis ionah Crocodile Icefish Notolepsis coatsi Antarctic Jonasfish Notothenia gibberifrons Humped Rockcod

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 2

Notothenia kempi Striped-eyed Rockcod Notothenia neglecta Yellowbelly Rockcod Notothenia rossi Marbled Rockcod Notothenia squamifrons Grey Rockcod Nototheniidae Rockcods Nototheniops larseni Painted Rockod Pagetopsis macropterus Icefish Parachaenichthys georgianus Paradiplospinus gracilis Slender escolar Paranotothenia magellanica Magellanic rockcod Patagonotothen brevicauda Patagonian Rockcod Pleuragramma antarcticum Antarctic Silverfish Protomyctophum tensioni Tenison’s lanternfish Protomyctophum choriodon Gaptooth lanternfish Pseudochaenichthys georgianus South Georgia/ Blackfin Icefish Ragovitzia glacialis Teuthoidea Squid spp. Themisto gaudichaudii Amphipod Trematomus bernacchii Emerald Rockcod Trematomus lepidorhinus Slender Scalyhead

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 3

1 EXECUTIVE SUMMARY This report provides details of the MSC assessment process for the Olympic Antarctic Krill Fishery for the client Olympic Seafood against the Marine Stewardship Council’s Principles and Criteria for Sustainable Fishing v1.3. The assessment team used the default assessment tree as defined in the MSC Certification Requirements v1.3. A rigorous assessment of the wide-ranging MSC Principles and Criteria was undertaken by the assessment team and a detailed and fully referenced scoring rationale is provided in the assessment tree provided in Appendix 1.1 of this report. The assessment process began in September 2014. The report is prepared by DNV GL.

1.1 Assessment team The assessment team for this fishery assessment was Dr Geir Hønneland who acted as team leader and primary Principle 3 specialist, Dr Andrew I.L. (Andy) Payne who was primarily responsible for evaluation of Principle 1, Ms Lucia Revenga Giertych who was primarily responsible for evaluation of Principle 2 and Ms Sigrun Bekkevold who was DNV GL manager and chain of custody responsible.

1.2 Assessment timeline

Announcement of Initial Assessment: 2 September 2014

Site Visit and Stakeholder Consultation: 8–9 December 2014

Expected Date of Certification: August 2015

Actual Eligibility date: 14 May 2014

1.3 Scores for separate Principles

Table 1. Final Principle Scores

MSC Principle Score

Principle 1: Sustainability of Exploited Stock 90

Principle 2: Maintenance of Ecosystem 96.7

Principle 3: Effective Management System 93.1

1.4 Main strengths and weaknesses of the client’s operation 1.4.1 Strengths

 The krill fishery is operating at catch levels well below what would generally be regarded as a precautionary upper level relative to the best estimates available of stock size.  Bycatch is negligible, and there is virtually no interaction with species other than the target krill or very small quantities of retained species. Direct effects or interactions between the fishery and ETP species are non-existent. The gear can only impact the habitat in the case of gear loss, which has never happened.  The client's current and future vessels use and will use a trawl with a fine mesh panel across the entrance that prevents bigger fish from entering the main body of the net.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 4

 There is a well-established, well-functioning management regime and enforcement system for the fishery, including requirements of 100% observer coverage and catch reports after each haul.  The client works actively with, and provides financial support to, scientific research, contributing to knowledge production beyond that provided by CCAMLR and participating states.  The fishery is managed within a precautionary and ecosystem approach.

1.4.2 Weaknesses

No particular weaknesses are identified for the client. The main challenge for management of the fishery is that no synoptic survey of Antarctic krill has been conducted since 2000.

1.5 Determination with supporting rationale

The Olympic Seafood Antarctic krill fishery achieved a score of 80 or more for each of the three MSC Principles, and did not score under 60 for any of the set MSC Criteria. The assessment team therefore recommended the certification of the Olympic Seafood Antarctic krill fishery for the client Olympic Seafood. There were no conditions as non of the performance indicators was given a score lower than 80.

Following this decision by the assessment team, and review by peer-reviewers and stakeholders, the determination was presented to DNV GL Business Assurance decision making entity that the fishery has passed its assessment and should be certified. The Technical reviewer at DNV adheres to the recommendation of the assessment team and approves the certification of the Olympic Seafood Antarctic krill fishery for the client Olympic Seafood without any conditions.

1.6 Conditions

No conditions have been set for the fishery.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 5

2 AUTHORSHIP AND PEER REVIEWERS

2.1 Assessment team

Name Role Qualifications Andrew I. L. Expert for Dr Andrew I. L. Payne is an honours graduate of the University of Payne Principle 1 London who completed post-graduate degrees at the Universities of Stellenbosch and Port Elizabeth in South Africa. He worked in Namibia for five years, South Africa for 25 years (eventually leaving in 2000 as Director of the Sea Fisheries Research Institute), and retired in 2013 from the Centre for Environment, Fisheries and Aquaculture Science (Cefas), UK, where he was first Science Area Head for Fisheries and then "roving" international fisheries consultant in which role he inter alia managed a large commercial contract evaluating sites for future nuclear power stations to be built in the UK, and the Fisheries Science Partnership, an initiative bringing scientists and fishers together in a common aim to produce information of use to those charged with managing Europe's fish stocks. Most of his research work was conducted in South Africa, and he has published widely in the scientific literature, mainly about fisheries management and demersal fish in particular. He was an active player in the Benguela Ecology Programme, was involved in drafting South Africa's first democratic fisheries policy (which later became enshrined as the Marine Living Resources Act), and was a leading player in the establishment of the Benguela Current Large Marine Ecosystem project and the BENguela Environment, Fisheries, Interaction, and Training (BENEFIT) project, the latter two concentrating on three countries, Angola, Namibia and South Africa. From 2003 to 2011, he was Editor-in-Chief (and from 2000 to 2003 editor) of the ICES Journal of Marine Science, was the founding editor/editor-in-chief (and now international panel member) of the (South) African Journal of Marine Science, and is Series editor of the Springer book series Humanity and the Seas. He has also conducted peer expert review of fisheries in Argentina, South Africa and the USA, and was involved in the EU's TACIS project on Sustainable Management of Caspian Fisheries, among other EU projects. He has conducted several accreditation reviews for the Marine Stewardship Council, the full ones being for the Antarctic krill continuous pumping fishery (AkerBiomarine; twice, the second being a recertification assessment) and for Russian pollock, has acted as expert reviewer of the reports on US Limited Entry Groundfish Trawl fishery and SA hake trawl recertifications, and has twice acted as condition-meeting evaluator for the SA hake trawl fishery. Recently too, he was part of a three-man international team that formally evaluated the ICCAT bluefin tuna research programme. Finally, he has personally written/edited one book − "Oceans of Life off Southern Africa", and lead-edited and contributed to two more − "Management of Shared Fish Stocks", and "Advances in Fisheries Science; 50 years on from Beverton and Holt", the latter two both for Cefas.

Lucia Expert for Lucia Revenga is a marine scientist, specializing in Fisheries Biology, Revenga Principle 2 who holds degrees in Marine Science and in Environmental Science. Giertych For five years she worked with TRAGSA for the Spanish General Marine Secretariat, doing research on the biology and stock status of species such as bluefin tuna, skipjack tuna, albacore, mackerels, sardine, eels, prawns, Norway lobster and halibut. She has also taken part in oceanographic surveys focusing on investigating vulnerable marine ecosystems. Since 2011 she works for IFAPA (Institute for Research and Training in Fisheries) as a Fisheries Biology teacher for fishermen. She also conducts research on local

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 6

fishing activities with the aim of increasing community awareness of the conservation of coastal ecosystems and encouraging sustainable fishing practices. She has been involved as a P2 expert on the Danish Swedish Nephrops assessment and on the Aker BioMarine Krill Fishery reassessment. Earlier employment involved investigation of ISO-14000 and ISO-9000 norms. Geir Team leader Dr Geir Hønneland is Research Director at the Fridtjof Nansen Hønneland and expert Institute in Oslo, Norway, and adjunct professor at the University of for Principle Tromsø, Norway. He holds a PhD in political science from the 3 University of Oslo and has studied international fisheries management (with main emphasis on compliance issues), international environmental politics and international Arctic politics. Among his recent books are Arctic Politics, the Law of the Sea and Russian Identity (Palgrave, 2014), Making Fishery Agreements Work (Edward Elgar, 2012), International Environmental Agreements (Routledge, 2011), Arctic Politics and International Cooperation (Routledge, 2007) and Law and Politics in Ocean Governance: the UN Fish Stocks Agreement and Regional Fisheries Management Regimes (Martinus Nijhoff, 2006). He worked in the Norwegian Coast Guard from 1988 to 1994, where he was certified as a fisheries inspector. Geir also has a wide range of evaluation and consultancy experience, e.g. for the FAO and OECD, relating to responsible fisheries management. He has been involved in MSC assessments since 2009 (covering cod, haddock and herring fisheries in the Northeast Atlantic and krill in the Southern Ocean) and was certified as MSC Team Leader in 2014.

Sigrun Responsible Sigrun Bekkevold is a principal consultant at DNV GL Business Bekkevold for Chain of Assurance and holds a Master of Science in industrial chemistry and Custody and biochemistry from the Norwegian University of Science and DNV GL Technology in Trondheim. She has 25 years’ experience leading project projects for sustainable development of the marine sector. Her main manager focus has been on research, innovation and business development within the total utilization of fish. This has included compiling strategies, action plans, feasibility analyses and market analyses, organizing project teams, performing mass-flow analyses, networking with industry, research and authorities, evaluating regulatory issues and the communication of results. She has held the position of general manager in the RUBIN Foundation, aiming for value adding and better utilization of fish byproducts. RUBIN has been owned by the seafood industry in Norway and supported by the Ministry of Fishery and Coastal Affairs and the Norwegian Seafood Research Fund. Its work includes the whole value chain, from fishing vessel to the market. Quality and traceability is important in the work. In DNV GL, she has been project manager and chain of custody responsible officer for the initial assessments of the Sweden and Norway Skagerrak and Kattegat sprat fishery and for Sweden Skagerrak and the Norwegian Deep Cold Water prawn fishery. She is also project manager for surveillance assessments for the North East Arctic cold water prawn fisheries.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 7

2.2 Peer reviewers

Name Role Qualifications

Graham M. Peer Dr Graham M. Pilling is currently a senior fisheries scientist at the Pilling Reviewer 1 Secretariat of the Pacific Community (New Caledonia). With >20 years’ experience analysing, assessing and providing scientific advice on tropical, temperate and polar marine and freshwater ecosystems and fisheries, he fulfills the role pf underpinning the decisions of regional and subregional tuna management bodies in the Western and Central Pacific Ocean. Until 2010 he was the Head of the Seas and Oceans Group at the Centre for Environment, Fisheries and Aquaculture Science (Cefas) in Lowestoft UK, an organization that provides stock assessment and ecosystem advice to the UK Government, ICES and the EU. He has chaired EU STECF (SGMED) and FAO (GFCM and data-poor fisheries) meetings developing stock assessments for Mediterranean fish stocks, developed fisheries management plans taking ecosystem and ETP species issues into account, and designed and developed models to simulate the long-term impacts of uncertainty in stock biology and assessments on fisheries management, and methods to assess and manage data-poor fisheries, including non-target stocks. In addition, he has acted as a member of the MSC Assessment teams for a many fisheries around the world, covering both Principles 1 and 2. These have included Antarctic krill and the New Zealand hake, ling and blue whiting fisheries (the latter as both P1 and/or P2 expert). He has also peer-reviewed a number of MSC Assessment reports.

Denzil G. M. Peer Dr Denzil G. M. Miller has a PhD in marine biology from the Miller Reviewer 2 University of Cape Town and is a Professorial Fellow at the University of Wollongong. He was a senior specialist scientist in the South African Fisheries Department from 1979 to 2002. He received the South African Antarctic Medal in 1995 and the prestigious Duke of Edinburgh Conservation Medal in 2007 for his contribution to Antarctic conservation and management. He has published >65 peer-reviewed journal papers, book and book chapters on krill biology, as well as on marine resource biology, policy, management and conservation issues. He served as Chair of the CCAMLR (Commission for the Conservation of Antarctic Marine Living Resources) Scientific Committee (SC-CAMLR) from 1996 to 2000 and was CCAMLR Executive Secretary from 2002 to 2010. He was Convener of the CCAMLR Krill Working Group from 1987 to 1994, and has led 15 research cruises to the Subantarctic and Antarctic. He chaired the FAO’s Regional Fisheries Bodies Secretariat Network from 2003 to 2009 and has been an expert advisor to the FAO on various marine issues. He took a lead role in the setting up of three Regional Fisheries Management Organizations (RFMOs), and served on the NAFO and WPCFC performance review panels, as well as a peer reviewer for the 2014 Aker Biomarine Krill Fishery MSC certification. He is currently Director: Antarctic Tasmania and Science Research Development and became an Honorary Member of the Order of Australia (AM) in 2011.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 8

3 DESCRIPTION OF THE FISHERY

3.1 Unit(s) of Certification and scope of certification sought 3.1.1 Scope of Assessment in Relation to Enhanced Fisheries The assessment team confirms that the fishery under assessment meets the scope requirements defined in MSC Certification Requirements Version 1.3, 14 January 2013 (CR 27.4).

Principle 3, Criterion A1: The fishery is not conducted under a controversial unilateral exemption to an international agreement.

Principle 3, Criterion B14: The fishery does not use destructive fishing practices such as poisons or dynamite. 3.1.2 Scope of Assessment in Relation to Enhanced Fisheries

This is not an enhanced fishery. 3.1.3 Scope of Assessment in Relation to Introduced Species Based Fisheries (ISBF)

Antarctic krill is not an introduced species.

3.1.4 Scope of Assessment in Relation to Risk Based Framework (RBF)

The RBF has not been used for this fishery.

3.1.5 Unit of certification

The unit of certification is defined as:

Species: Antarctic krill (Euphausia superba) Geographical range of Southern Ocean, CCAMLR area 48.1, 48.2, 48.3, 48.4 fishing operations: Method of capture: Pelagic trawl Stock: Antarctic krill (Euphausia superba) Commission for the Conservation of Antarctic Marine Living Management: Resources, CCAMLR (www.ccamlr.org) Authorizing member: Norway Client group includes Olympic Holding AS and affiliated Client group: companies and vessels

3.1.6 Rationale for Units of Certification According to the MSC Certification Requirements v1.2, the proposed Unit of Certification shall include the target stock(s), the fishing method or gear and the practice (including vessels) pursuing that stock. The MSC Certification Requirements Guidance V1.1 specifies that the Unit of Certification is “The fishery or fish stock (= biologically distinct unit) combined with the fishing method/gear and practice (= vessel(s) pursuing that stock”.

This Unit of Certification was used because it is compliant with client wishes for assessment coverage and in full conformity with MSC criteria for setting the Unit of Certification.

3.1.7 Other Eligible fishers There are no other Eligible Fishers and the Certificate Sharing Mechanism is not applicable.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 9

3.2 Overview of the fishery 3.2.1 Client name and contact information

Olympic Seafood AS Contact person: PO Box 234 Bjørnar Kleiven 6099 FOSNAVÅG Phone: +47 700 81 950 Norway E-mail: [email protected]

3.2.2 Client group

Olympic Holding AS currently includes:  Olympic Seafood (Management company)  Emerald Fisheries AS (Fishing company) owning the vessel Juvel. Reg. M-361-HØ. Crewing: 50 Total employees: 100

The Olympic Holding AS company Olympic Group has a second krill license. A new vessel will be built and commence krill harvesting in 2017/2018 (expected). The same fishing and production system will be installed on the new vessel.

3.2.3 Species and fishing practices

See Section 3.3.3.

3.2.4 History and fishing areas

See Section 3.3.3.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 10

3.3 Principle One: Target Species Background

Principle 1 of the Marine Stewardship Council standard states that:

A fishery must be conducted in a manner that does not lead to overfishing or depletion of the exploited populations and, for those populations that are depleted, the fishery must be conducted in a manner that demonstrably leads to their recovery.

Principle 1 covers all fishing activity on the entire target species stock, not just the fishery undergoing certification. However, the fishery under certification would be expected to meet all management requirements, such as providing appropriate data and complying with controls, therefore demonstrably not adding to problems even if the problems will not cause the certification to fail.

In this section of the report, a referenced summary is provided covering: a. an outline of the fishery resource, including life history information; b. an outline of the status of the stock as indicated by stock assessment, including a description of the assessment methods, standards, and stock indicators, biological limits, etc. c. a history of the fishery for the stock and its management.

Antarctic krill, the single target species, is a key lower trophic level (LTL) species, so management has to take account of this and the P1 text covers the subject in detail.

3.3.1 Background information

3.3.1.1 Introduction The Antarctic krill is a key member of the Southern Ocean ecosystem and provides a crucial energy link between primary production and higher predators such as baleen whales, seals, fish, birds and cephalopods. Few organisms in the region do not depend on or are not impacted by krill, a stock believed to have the largest biomass of any species on Earth. As much as 152–313 million tonnes of krill are thought to be consumed by its various predators each year (Miller and Hampton 1989). Krill are circumpolar and in the austral summer bounded on the north by the Antarctic polar front and by pack ice in the south, although no-one really knows how far under the pack ice krill distribution extends, rendering survey estimates of biomass tenuous.

Being primarily herbivorous, especially in the warmer periods between austral spring and autumn when extensive phytoplankton blooms are found in the Southern Ocean, krill feed mainly on phytoplankton, but do take advantage of other food resources to survive when such food is scarce, e.g. in winter when they have been recorded foraging on ice-algae. Krill actively search for food aggregations and feed rapidly to exploit locally good food concentrations.

There have been many studies on krill made to date, and they cover all aspects from spatial distribution through life history, biology and ecology, but this section of the report synthesizes current knowledge as a backdrop to discussing stock status and fisheries. Key information on krill is contained in the Discovery Reports of the 1930s, various sightings and commercial activities (generally on other stocks and species) throughout the 20th century, FIBEX acoustic study reports and the results of other surveys, most of them other than the Discovery cruises having been relatively recent. Summary literature consulted to compile this summary includes the works of McWhinnie et al. (1976), Miller and Hampton (1989), Everson (2000), Miller (2003) and Nicol (2006), although specific aspects relating to certain life history characteristics are outlined in a wealth of other literature, some referenced as appropriate in the text below. 3.3.1.2 Distribution The circumpolar spatial distribution of krill is underpinned by the main currents of the southern oceans, of which the Antarctic Circumpolar Current (Miller and Hampton 1989) is probably the most influential. Notwithstanding, concentrations seem to be densest in Convention Area 48, and it is there where most of the fisheries for krill have focused their activities; only small catches have been taken from other areas. The results of various surveys have shown that krill density in the western Atlantic sector of the Southern Ocean, which apparently contains about half of the whole Southern Ocean krill stocks and as a consequence of the geography there extends

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 11

substantially farther north than elsewhere (Atkinson et al. 2004), is more than twice as large as in the Indian Ocean sector, and there is also some evidence that krill abundance is even lower in the Pacific Ocean sector than in the Indian Ocean sector.

Bottom topography (notably the shallow continental shelves and the extensive areas around Southern Ocean islands) and concomitant hydrographic features probably govern the spatial distribution of krill, ensuring its enhanced biomass around the Antarctic Peninsula in Area 48. Current-induced circulations, especially those associated with the Antarctic Circumpolar Current, underpin the circumpolar distribution of krill, notably from the known spawning and nursery areas around the Peninsula, and variations in the strength of those water flows contribute to the variability in species recruitment and biomass on a seasonal and annual basis (Sushin and Shulgovsky 1999, Meredith et al. 2008). Meredith and his co-workers also show how krill distribution and biomass fluctuations influence the distribution and sizes of most of the other species and stocks that depend heavily on the species. On a smaller scale too, the distributions of commercial krill aggregations in Area 48 are influenced strongly by the variations in these water movements, including counter- or compensatory flows that carry krill to areas not so influenced by the known flow patterns. Hotspots of krill abundance do exist, however, and Atkinson et al. (2008) state that retention of the stock in habitats that are sufficiently productive to support such a large, though fluctuating, biomass is key to the success of the species in the whole area.

Krill are generally considered to be relatively passive components of ocean systems, but there is good evidence that adults are efficient swimmers that can swim against currents for long periods. Therefore, overall krill distribution is likely influenced by adults being able to remain in favourable (i.e. productive) habitats for long periods (Miller and Hampton 1989). The scientific literature and evidence based on commercial-size aggregations is not yet definitive in terms of whether krill follow repetitive diurnal migrations; some stocks seem to do so, whereas others may not. What is not questioned, however, is that krill can exist in layers and patches ranging from a few square metres across, through shoals, schools, swarms and up to superswarms covering >100 km2, and it is this aspect of their behaviour that probably renders them appropriate for commercial harvesting. Solitary or dispersed groups of krill are often found during research surveys, generally in a seasonally repetitive pattern, but such animals are not sought commercially. The fishery-targeted krill are generally those in aggregations dense enough to sustain the type of fishery, usually around islands, over shallow shelves or associated with areas where water masses mix, and their swarming and schooling behaviour has been explained scientifically as an adaptive strategy to avoid selective predators such as fish and seabirds, as well as to increase their own efficiency in finding food. Aggregations of krill may last a matter of hours or even longer, perhaps days or weeks, and their components vary considerably in shape, size and state of sexual maturity.

For any stock with a distribution as extensive as that of krill, it must be possible that there are several stocks, but to date for krill, formal stock separation by any means, including genetics (Siegel 2000), has not been proven. 3.3.1.3 Krill growth Several techniques have been used to estimate krill age and longevity, but research is ongoing and the outputs still require verification (reliable age determination of most crustaceans is still scientific work in progress). Krill apparently attain a maximum length of >60 mm and an age of >5 years. Modal progression in length-frequency data (de la Mare 1994a, 1994b) and laboratory experimentation have been used to determine the growth rate and show that growth is highly seasonal and almost certainly depends on food availability. For instance, around the Antarctic Peninsula, the peak season for krill growth is the austral summer (January/February), and it is then that somatic fat reserves are accumulated and the fishery targets the animals for the valuable Omega 3 somatic oil. The fat content of krill, and hence the oil yield, drops again as the ice advances north as autumn advances into winter, and it is then very low from May/June on.

During winter, food availability in the form of plankton is much lower, so many have assumed that little feeding by krill takes place and that body size might then shrink, although a reduced metabolic rate then would be likely too. Laboratory experimentation has revealed krill surviving for 200 days or more without food but with some body shrinkage. However, as mentioned above, krill in the wild in winter have been observed feeding on ice-algae and other zooplankton components of the water column.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 12

3.3.1.4 Reproductive biology and life history Gravid females usually concentrate over the continental slope, less developed ones over the inner shelf or closer to the ice-edge, and spawning females mainly in open water. Spawning apparently lasts from late November to late March, but the onset of krill spawning varies spatially and between years. The extent of winter sea-ice and the duration of ice cover together seem to influence the onset of each spawning season, with spawning success tending to be lower after mild winters.

Scientific study has shown that spawning biomass is a reasonable indicator of egg production, which itself is a factor governing regional recruitment. Egg production is a product of female abundance, the percentage of females reproducing in a season, the number of spawning episodes (krill are batch-spawners) and the number of eggs released per spawning episode, with each female producing 6000–10 000 eggs per episode. It is not known why, but not all females reproduce every year, the proportion of subadult and adult females reproducing within a season apparently varying from <20% to virtually 100%.

The number of batches of eggs released per season depends on the interbrood period, which itself varies with location and year from a minimum of 6 d to a maximum of 50 d in midsummer, almost certainly depending on food availability and environmental conditions. The number of spawning episodes per season ranges from 3 to 9 and the number of eggs released per episode is likely related to the mean body size of krill in the spawning aggregation, although that relationship is still tenuous.

Female Antarctic krill generally spawn from an age of about 2 years near the sea surface and the eggs sink to hatch in deeper water, thought to be between 800 and 2500 m deep. The larvae then develop through various stages as they ascend in the water column. Feeding commences after 21–30 d at depths of 30–100 m at the fourth development stage, but the total development time from egg to final larva takes about 130 d. Male krill seem to mature at an age of about 3 years. Although krill can live for several years, the biomass of krill aged 5+ years is relatively small, <1% of the total stock surveyed, meaning that older krill do not contribute substantially to overall population structure. 3.3.1.5 Population dynamics Realistic estimates of the rate of natural mortality (M) for krill lie in the range 0.66–1.35 year–1. Based on the method of Hoenig (1984), longevity of between 7 and 5 years would be consistent with natural mortality rates of between 0.62 and 0.87 year–1, respectively.

Estimates of recruitment or an index of it are derived from survey data, and although no stock– recruitment relationship has been demonstrated for krill around the Antarctic Peninsula to date, year classes have been shown to be stronger in some years than others. One likely factor explaining the variability in krill recruitment is the extent of sea-ice, Loeb et al. (1997) showing for instance that salps compete with krill for the phytoplankton food when sea-ice extent is low. In any event, krill recruitment varies massively by region, indicating that separate geographic estimation of recruitment levels would be revealing. In some years, for instance, favourable sea- ice extent and high local levels of productivity have stimulated excellent recruitment in the dominant Atlantic Ocean sector, whereas recruitment in the Indian Ocean sector has been poor.

3.3.2 State of the Krill Stock

To be certified by the MSC, a stock needs to be subject to rigorous numerical stock assessment (based on catch rates and/or survey data) and fishing and biomass levels against reference points have to be considered in order to evaluate whether the fishery is operating sustainably. Crucial under this part of the evaluation is that emphasis is to be on total stock rather than the part of the stock in the UoC fishery area alone, but for Antarctic krill with its massive estimated total biomass and circumpolar distribution centred on Area 48, the evaluations and assessments have tended over the years to concentrate on that area. The Olympic Seafood fishery for krill operates (with a different technology) in the same area as another Norwegian fishery that is now in its second period of certification. Both certification and recertification reports for that fishery went into great detail on stock assessment, reference points and fishing levels against total stock size, and indeed little changed in the five years between evaluations. A valuable means of developing a time-series of stock data, as mentioned above, is to conduct regular surveys. Unfortunately, therefore, the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 13

lack of a fully synoptic survey of krill since 2000, even just in Area 48, is disappointing; such effort would, of course, be hugely expensive and require the participation and funding of many countries. What has happened, though, is that some individual CCAMLR member countries have conducted annual or occasional surveys, often using a fixed grid of positions, of their own fishing area of a smaller geographic extent, using mainly commercial vessels. These surveys are a valuable contribution to knowledge of stock size.

The stock assessments of krill reviewed annually at CCAMLR are therefore always new assessments of old data, supplemented by whatever information has been collected inter- sessionally, sometimes with fresh assumptions or different interpretations of parameters, supported by research findings that appear in the scientific literature. The fishery is, however, operating at catch levels well below what would generally be regarded as a precautionary upper catch level relative to the best estimates available of stock size (i.e. a precautionary catch limit or PCL), although those estimates themselves vary extensively depending on the assumptions being made and the values for key parameters used. Of note here is the evaluation by Kinzey et al. (2013) that shows that because of the uncertainty inherent in an assessment based on supplying the Generalized Yield Model (GYM) with a value of M of 0.8 and recruitment variability generated using a Beta distribution for proportional recruitment of krill, caution will be crucial if management allows the trigger level to increase towards the PCL. For the purposes of the evaluation in this report, however, it is deemed sufficient to present an updated summary of standard stock evaluation outputs, many based on knowledge tabled in the past (for instance when this fishery was first certified). The key issue here is considered to be the likely effect of fishing levels on stock size(s) over time, and, as is shown in Section 3.3.2.5 below, the management rigour and controls applying to the stock are such that scientific confidence is good that current catch levels will not affect the total krill biomass adversely even if extraneous ecosystem and oceanographic/climate come into play. 3.3.2.1 Introduction Area 48 is here considered to be a single management unit, although recruitment is shared with other areas (Pacific, Area 88; Indian, Area 58). Given the limited and certainly inconsistent rates of exploitation of krill elsewhere than Area 48, managing Area 48 as a single separate stock is deemed to be appropriate. Making this assumption does not, of course, preclude the future possibility of managing some small areas as self-recruiting single stocks if it is found that seasonal immigration from other areas is not essential in maintaining those small stocks. Miller (2003) and others identified a few possible candidates for “separate self-sustaining small krill stock” status, generally based on aggregations identified from surveys and in catch data, but to date no substantive proof of the existence of such stocks has been forthcoming. The management areas currently in place (Subareas 48.1–48.6; Fig. 1) are based on CCAMLR’s own precautionary management regime for krill, given that is unlikely that krill, or krill products, move extensively between smaller areas and following understanding of the oceanography of the Convention Area.

Figure 1: The location of subareas in Area 48 (South Atlantic) and (inset) the whole CAMLR Convention Area

Source WG-EMM-13/37 Rev. 1.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 14

Scientific and management concern has been raised about the possibility of there being localized depletion, and it is deemed essential to ensure that commercial catches leave adequate biomass to support local predator populations. This has led to an overall TAC, and trigger levels of catch (see below) for Area 48 divided among smaller-scale subareas to ensure control over depletion in each subarea. CCAMLR does provide reasonably accurate catch data since 1980 on defined small- scale management units (SSMUs; see the extensive Table 3 of WG-EMM-14/58, but note that that document is still in draft stage and not formally released) within the focal fishing areas of 48.1– 48.3, but TAC and trigger level management of such SSMUs is still not carried out. 3.3.2.2 Stock assessment There is no formal (annual or less regular) assessment of Antarctic krill. Catch reporting to the management authority is good, but comprehensive age data and regular, verifiable survey data are lacking, as stated above. Reliable age data for invertebrates generally are scarce (although length data for the krill fishery are available and have been used – see below). The CCAMLR scientific community therefore uses a different process to determine stock health. A precautionary catch limit (PCL) is based on the potential yield of the stock as determined from the synoptic survey carried out in 2000, and it then requires that strong and effective monitoring systems be in place. CCAMLR set the initial limit to yield on the basis of estimated potential yield, and the software used to estimate that value, a Generalized Yield Model (see Constable and de la Mare, 1996), uses standard population model equations, but allows the population dynamics to represent particular stocks by setting appropriate parameter values for growth, natural mortality and recruitment.

Miller (2003) provides the life history information and parameterization used in the GYM for krill, with growth and mortality estimates drawn from length frequency data (de la Mare 1994a; 1994b). The GYM accounts for expected population production and random effects around it through variable recruitment, and the projection is used to calculate the appropriate TAC as a proportion of unexploited biomass (B0) based on the harvest control rule. Unexploited biomass is estimated from surveys.

Integrated assessments, lessening the reliance on expensive synoptic surveys, and management strategy evaluation to test quota allocations have both been suggested as a means of improving the scientific evaluation supporting management, but remain a future hope rather than something that can be used immediately to improve stock assessment. 3.3.2.3 Biomass and abundance CCAMLR’s multi-ship acoustic survey of krill in Area 48 in 2000, for which the rationale and protocols are described in CCAMLR Science, Vol. 8 of 2001, remains the only fully synoptic survey available for Antarctic krill biomass in Area 48 and provides the initial biomass estimate for the harvest control rule. Lacking a swimbladder, krill biomass is difficult to estimate acoustically (but see below), but novel methodology for the time was used to provide an initial estimate of krill standing stock (B0) of 44.29 million tonnes. That value was used as a proxy for krill pre- exploitation biomass in the GYM, estimating krill sustainable yield based on the decision rule.

Subsequently, different models and target strengths have been applied to the same survey data, and a wide range of estimates of B0 has been produced, from as little as 37.29 million tonnes (Demer et al. 2007) to as high as 207.98 million tonnes (Heywood et al. 2006). Clearly, the level of uncertainty in these estimates is great, but for the potential yield parameter used in management, a low estimate is still used.

Biomass varies seasonally and annually, probably explaining the year-on-year variation in the estimates of biomass determined from surveys undertaken at different times of the year. To date, however, commercial catch per unit effort data, annual or seasonal, have not provided a good index of abundance for comparative purposes given that fishing vessels tend to seek and target aggregations of a size depending on their own company targets for product. 3.3.2.4 Monitoring Total catch of krill by live weight is generally estimated from processed product, which is weighed. Inconsistencies in the conversion factors being applied between and even within nations (for example, see Table 5 of WG-EMM-13/37 Rev. 1) and a lack of information on discard practices across the whole fishing fleet has raised some concern that the quantity of krill being removed from the system is not documented sufficiently accurately. This concern gives rise to regular queries, for example, about catch estimates and ecosystem effects of the krill fishery. Observer

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 15

coverage in the krill fishery is relatively good (though not in all cases nationally independent), however, and in the UoC fishery, it is 100% (and carried out by non-Norwegians), so this evaluation team is confident that krill fishing is being well monitored generally. Although Leape et al. (2009) have formally questioned the validity of the conversion factors being applied in some krill fisheries, catches, even if they were to be subject to raising by as much as four times, are still very low relative to the precautionary catch (trigger) level set.

Other concerns raised about the effect of krill fishing have highlighted the selectivity of the different gears being deployed by different fleets, the bycatch of fish and fish larvae, seabird warp strikes and incidental seal mortality as potential areas of concern. Also, the incidental mortality of krill from trawling (krill that pass through the net but may be damaged and die) has not been estimated.

Conservation measures for data reporting in the krill fishery do not require collection of biological information (sex and size information), but the UoC fishery does allow such data to be collected and analysed, and actively supports such research effort. The Scientific Committee and WG-EMM have over the years commented upon the implications of new technologies in the krill fishery, in particular expressing concern that the various fishing technologies being employed on krill may be capturing different components of the krill population, with some methods having a greater ecosystem impact than others. Ongoing monitoring, preferably independently of fishing nation, and verification are therefore crucial to effective management of the resource.

The CCAMLR database now holds scientific observer data from a large number of trips/deployments for each fishing season. Bycatch of any organisms in the whole krill fishery generally is low, and virtually nil in the UoC fishery, and that fishery anyway has in place effective large animal bycatch mitigation measures. For instance, the net carries a sea lion excluder device (SLED) of 200-mm-square nylon mesh spliced onto the floor, roof and sides of the front meshes of the net being deployed. Interactions of the gear with seabirds and penguins is recorded as negligible too. 3.3.2.5 Harvest strategy Article II of the CAMLR Convention lays the foundation for an "ecosystem approach" to management, requiring that harvested populations be monitored and assessed, that significant ecological interactions between harvested and other species be defined and quantified, and that levels of depletion be estimated in order to monitor their recovery, if necessary. Moreover, management specifically has to take account of the low and pivotal position of krill, as a LTL species, in the Antarctic trophic structure, so sustainable exploitation needs to take account of its interactions with other species (Miller 2003).

WG-EMM's terms of reference have been prioritized and include requirements to provide scientific advice that helps to account for major uncertainties. Emphasis is given to ecosystem assessment and status, but although CCAMLR‘s role has focused on conservation, the Commission is also tasked with allowing the development of fisheries within its Convention area. Therefore, although adequate protection needs to be afforded to krill-dependent predators at critical times and in specific areas, it is deemed that such protection should not exert unnecessary, or unreasonable, restrictions on the fishery (SC-CCAMLR 1993).

Given the dynamics of the stock and its concentrations, krill stock boundaries are not delineated other than into Subareas (as stated above), but CCAMLR strategy is to manage fishery expansion in accord with its own management objectives (WG-EMM-08/46). Its primary controls remain catch limits and it does not set MSY as a target, because sustainable harvesting levels would almost certainly be well below a single-species krill MSY. The precautionary limits are therefore set commensurate with information available on stock dynamics.

In setting a krill PCL, CCAMLR uses a potential yield estimate. Although that estimate does not take account of the potential effects of harvesting krill on its dependent predators, the decision rule does. CCAMLR does monitor the populations of krill predators, making catch reports and supporting periodic, fisheries-independent surveys of local biomass. Monitoring includes a long time-series of krill density and recruitment indices, although whether these will be sufficient to detect the impacts of fishing in a timely manner has not been tested.

The krill fishery overlaps with areas where foraging, land-based predators, particularly penguins, capture krill to feed their young during t h e rearing phase. Using a modelling approach and

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 16

robust assumptions, Mangel and Switzer (1998) suggested that the required level of krill forage by penguin offspring, and adult foraging behaviour and relative local reproductive success could be inversely correlated with the fraction of the total krill biomass caught by the fishery. Notwithstanding, every effort is being made to correlate data on potential krill fishery/predator (not just penguins) overlap, with a view to identifying areas and times of likely most significant common use, and ultimately to developing spatial catch quotas for krill.

Conservation measure CM 51-01 (2010) requires that although the total combined catch of krill In Subareas 48.1–48.4 (little fishing is undertaken currently in 48.5 or 48.6) be limited to 5.61 million tonnes in any fishing season, ‘until the Commission has defined an allocation of this total catch limit between smaller management units, …, the total combined catch in [the same areas] shall be further limited to 620 000 tonnes in any fishing season.’ That value is the so-called trigger level applied currently. Predator stocks are further protected from the fishery by Conservation measure CM 51-07 (2011) which requires that the same krill catch be distributed in such a way as not to be inadvertently or disproportionately affected by [krill] fishing activity. The measure goes on to recognize that large catches up to the trigger level from areas smaller than the subareas (i.e. SSMUs; see Table 3 of WG-EMM-14/58) be avoided, and that distribution of that trigger level needs to provide for flexibility in the location of fishing in order to (i) allow for interannual variation in the distribution of krill aggregations, and (ii) alleviate the potential for adverse impacts of the fishery in coastal areas on land-based predators. Thence, no more than 25% of the trigger level catch (i.e. 155 000 t) can be taken from Subarea 48.1 annually, 45% (i.e. 279 000 t) from each of 48.2 and 48.3, and 15% (i.e. 93 000 t) from 48.4. Those percentages clearly add up to >100%, but it would be unlikely for more than one to be reached at a time. The aim of the measures (and strategy) is clear, however, to protect local availability of food to predators.

Catches have remained well below the overall trigger level, but individual subarea triggers have been reached occasionally recently (one was reached in 2014) and fishing in some areas stopped, so the strategy does seem to be effective.

Additional provisions within the various measures in place for the krill fishery deal with (monthly) data-reporting procedures, including ongoing requests to provide haul-by-haul data from the fishery and to carry scientific observers on krill vessels. 3.3.2.6 Harvest control rule The harvest control rule for krill in Area 48 is subsumed within the precautionary catch limit that has the objective of constraining exploitation to a safe level. The limit is based on krill potential yield estimated using the GYM, so is a proportion of pre-exploitation biomass (B0). Estimates of krill recruitment variability, growth and natural mortality are used in a stochastic simulation to determine and review the likely effects of various levels of harvesting on the population. The GYM estimates the probability density of possible krill population sizes at various levels of fishing mortality (F), including in the absence of fishing, by simulating many times with different random input values for recruitment. Outputs from the simulations are used in the decision rule.

The two rules used in the approach are the recruitment criterion, i.e. that the probability of the spawning biomass falling below 20% of median pre-exploitation spawning biomass after 20 years should not exceed 10%, and the predator criterion, i.e. that the median spawning biomass should not fall below 75% of pre-exploitation spawning biomass after 20 years. The recruitment criterion therefore effectively includes a limit reference point of a general recruitment-overfishing threshold, and the predator criterion a target reference point for spawning biomass; together they address Convention Article II objectives 3a and 3b. The 20-year period specified in the approach relates to Convention Article 3c and allows integration of expanding exploitation levels with sustainability and precaution, so is good management practice; it certainly allows for several generations of the stock. Taking these two criteria into consideration in the model, the lowest value is accepted as the catch limit, and that value is reviewed annually and, of course, is well above the trigger level currently set for krill catches.

Having a precautionary catch trigger level (overall and for subareas) in place prevents the decision rule having any practical effect on management and catches until SSMUs are accepted formally and applied. The overall PUCL value of 620 000 t is almost 100 000 t above the historical maximum annual catch in Area 48 (528 201 t in 1981/82; WG-EMM-14/58), but is nevertheless considered by WC-EMM to be an arbitrary choice, less than that of alternative proposals made when it was originally established. The structuring of SSMU quotas is still deemed by the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 17

Commission to be necessary to render the management procedure fully consistent with ecosystem requirements (Article II of the Convention), but although in 2000 the Commission estimated development would take 5–10 years, practical issues of reporting, observer coverage, etc, still need to be dealt with before such a system becomes feasible. Ongoing modelling, including that related to the ecosystem, is and has been used to examine possible principles for subdividing the PUCL among SSMUs (i.e. smaller than the subareas), e.g. based on spatial distribution of historical catches, predator demand, krill biomass, krill biomass minus predator demand, krill availability indices that can be updated on a regular basis, and fishing strategies that rotate catches within and between the smaller areas.

Ultimately, various scenarios will likely be applied to the ecosystem models to evaluate their performance in allocating quota among SSMUs, and performance measures are being derived to cover the status of krill, predator populations and the fishery over relevant time- scales. This effort is deemed crucial to improving the management system covering the krill fishery as a whole, but with current total catches so low relative to potential yield, there is still some resistance to implementing further controls on the fishery at this stage. 3.3.2.7 Uncertainty and the precautionary approach The current assessment process is inclusive of parameter (fishery and ecosystem) uncertainty, as well as structural (model) uncertainty to the extent that different models have been developed and evaluated, but the robustness of the decision rule to broader uncertainties has not been explored robustly. For example, the extent to which long-term changes to parameters, particularly those caused by changes in krill/predator distribution and particularly climate and the environment, would be incorporated into the harvest strategy is not clear. Another uncertainty not currently incorporated in the assessment and decision rules is implementation uncertainty, caused, for example, by inaccurate reporting of catches. Of course, such uncertainty can be minimized by implementing appropriate management measures or explicitly representing the uncertainty in the models.

There are naturally uncertainties associated with the fishery (see Leape et al. 2009, Johnston et al. 2009), so the question does arise whether the catch limit is sufficiently precautionary to preclude there being a great risk of damage to the ecosystem. The PUCL, which is anyway well above the historical maximum catch, is justified as precautionary primarily on the basis of the potential yield estimate.

The current average annual catch in Area 48, which is now rising slowly as new vessels begin to operate in the fishery, is still generally only about 30% of the PUCL, and that itself is just 11% of the potential yield (catch limit) estimated currently. Clearly, though, whether or not the trigger level is precautionary in terms of stock dynamics depends on the yield estimated from the GYM, which has its own uncertainties (Kinzey et al. 2013).

The harvest control rule used to calculate the potential yield is probabilistic, but it does not take account of all possible uncertainties. Recruitment variability is covered through a Beta distribution for proportional recruitment of krill, but seasonal changes in stock size, natural fluctuations in abundance and long-term changes in productivity (for instance those caused by climate change) are not accounted for. Reference points and risk parameters are generic despite the considerable research activity undertaken on the species. For instance, the predator criterion refers to the simulation median after 20 years and is set half-way between SSB50%, which is a BMSY proxy, and unexploited stock size; it is conservative relative to a single-species criterion and does reserve biomass for predators. The recruitment criterion refers to 10% of simulations after 20 years falling below 20% of the spawning stock, but again, there is no firm justification for 10%, 20% or the 20 years used in the rule. The last value is, though, some three times the longevity and eight times the generation time of krill. The decision rule will decrease the exploitation rate the longer is this interval, so considering natural mortality of the species, 20 years would seem to be suitably precautionary. Also, if such a TAC were to be applied, the criteria in place mean that the reference points will likely be approached over that 20-year period, which should allow plenty of time to monitor the performance of the rule.

Current levels of exploitation are precautionary, but the range of values calculated and documented in the literature for B0 do raise concern. All are based on the level of unexploited biomass calculated from the single synoptic survey carried out in 2000; although there is apparently no likelihood of this survey being conducted again soon, some national efforts are being made to resurvey certain sectors regularly in a rigorous manner, including the area covered

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 18

by the UoC fishery. Differences in the various estimates generally exceed sampling standard errors and confidence ranges, however, suggesting that overall uncertainty is likely being underestimated, and the most-recent CCAMLR-approved estimate of unexploited biomass is taken from near the lowest of the range of values. There is confidence is in the management control system, though, if it is considered that the overall and individual subarea trigger levels are still well below the PUCL calculated on the basis of the GYM.

Elsewhere in this report, mention is made of uncertainties and concerns related to the use of varying conversion factors for krill, from processed to live weight, and to the potential for discards not to be taken into account. Such issues are almost certainly not relevant in terms of the UoC fishery, and despite total annual catches rising slowly, it would seem highly unlikely, given current controls and conservation measures, that catches will exceed the trigger levels in the short term. In the longer term, however, if catches continue to rise and further catching capacity is introduced, such issues are going to have to be addressed.

The monitoring system instituted by CCAMLR is probably adequate to detect any problems that may arise in the fishery overall, and in the UoC fishery, the observer, reporting and data-collection systems in place are as good as or better than those in place for many certified fisheries around the world. Hence, it is unlikely that any of the uncertainties within the total krill fishery are so serious as to mitigate against certification of the UoC fishery.

3.3.3 The Fishery and its Management

3.3.3.1 History Miller (2003) outlines the history of the development of the krill fishery. It started on a small scale in the 1960s, but it was not until the early 1970s that catches started to rise. Most commercial catches are made in the top 200 m of the water column, generally where the water temperature gradient is steep, and most at about 50 m deep (Naganobu et al. 2008). According to the latter author too, commercial operations in Area 48, the main commercial area, have tended to focus on oceanic fronts.

From the early 1970s, total catches of krill (not just in Area 48, though that area dominated the catches) rose steadily, from 19 785 t in 1973/74 to 528 201 t in 1981/82, with by far the most taken by the former Soviet Union, but they then quickly collapsed, likely because of technical problems associated with the discovery of high levels of fluoride in the exoskeleton of krill. When these technical problems were resolved with the introduction of new methods of processing and especially peeling, catches rose again, to attain a regular annual yield throughout the late 1980s and early 1990s of 300 000–400 000 t. However, with the break-up of the Soviet Union then, catches declined to annual values of <100 000 t, and it is only recently that they have started to rise again. Over the years, participating nations other than Russia and the Ukraine (both parts of the former Soviet Union), which dominated the developing krill fishery, have included Korea (initially in Area 58, but more recently focused in Area 48), Japan Poland and Chile, but other than Poland and Japan, they have not all maintained a consistent presence. Also, Norway in particular, the US and China, plus a few other countries that make small but inconsistent catches (according to the CCAMLR database, a total of 20 countries) have shown or are showing interest in the fishery. Norway in particular has developed its fishery and technology and has identified good markets for the products, so Norwegian catches now constitute the main commercial take of krill.

Miller (2003) considers the potential for krill harvesting outside the CCAMLR Convention Area to be small; few catches have been made or recorded at all to the north. CCAMLR Subareas 48.1, 48.2 and 48.3 dominate the catch history, especially recently, and catches in other areas (e.g. Areas 58 and 88) have historically been small. Catches tend to be concentrated on shelf-breaks, for instance north and west of the South Shetland Islands, in a broader area north of the South Orkneys, and around South Georgia. Fishing patterns are linked to the distribution of sea-ice, operations generally commencing as the ice edge retreats south in spring and diminishing as the ice edge spreads north in winter. South Georgian waters (Subarea 48.3), however, remain virtually ice-free in winter, so krill fishing there is possible virtually year-round, but that area is subject to its own rigorous management regime on top of the CCAMLR system. For more than a decade (since 2001), the formal fishing (reporting) season of CCAMLR has been 1 December of one year to 30 November of the next (Conservation Measure [CM] 32-01), but historically, krill

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 19

fishing in 48.3 tends to start later in the calendar year, as 48.1 and 48.2 are covered by the expanding ice field.

Market limitations seem to be the major constraint on the fishery developing further (Nicol and Foster 2003, Nicol et al. 2011; and see below), but there is still interest in the fishery by many CCAMLR member countries, with ongoing and sometimes collaborative international research and surveys being undertaken. Different fishing techniques are currently practiced by the fleet, but all involve some form of midwater trawling (using pelagic trawls or trawls with a beam) on layers or aggregations of krill. 3.3.3.2 Notifications and catch reporting To improve management and control, and for any country to be able to participate in the krill fishery, CCAMLR requires annual (in advance of each season) notifications of vessels and their potential catches, gear and fishing area (Conservation measure CM 21-03). However, to date and notwithstanding CCAMLR’s attempts to bring the notification system to an acceptable and realistic standard, it has not been operating as effectively as hoped for. Annual notifications of potential fishing activity have tended to exceed not only the realized catches (with Area 48 dominating notifications), but also the so-called trigger levels at which subareas or even whole areas would have to be closed to protect the stocks (see above). CCAMLR rules on notifications, and perhaps a greater cost to prospective notifiers, are expected to solve the current rather loose system and support better management rigour in the fishery overall. However, the situation for 2014/15, the season during which this report is being produced, was that six countries (Chile, China, Korea, Norway, Poland and Ukraine) had still notified intention to fish for krill in Area 48 (48.1–48.4, although Korea notified no proposed fishing in 48.4), amounting to a possible maximum catch of 611 000 t, very close to the PUCL. Clearly, the system is not as effective as it should in terms of alerting management to likely annual catches.

Figure 2: Total annual catches of krill 1972/73–2012/13 in Area 48 (note that “1973” refers to the CCAMLR 1972/73 fishing season). The annual catch exceeded 200 000 t from 1978/79 to 1982/83, from 1984/85 to 1991/92, in 2009/10 and since 2012/13 (it exceeded 200 000 t in 2013/14 too, but those catches have yet to be validated and included in the report). Nearly six million tonnes have been taken since the start of the fishery, more than half by the former USSR. The total catch of Antarctic krill peaked in 1981/82 at 528 201 t.

Source WG-EMM-14/58 (seen in draft)

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 20

Total annual catches of krill from the Convention Area are now on a slow upward trajectory (Fig. 2), but although in all areas catches are well below peak historical values, there is still a need for participating member countries to become more realistic and perhaps less frivolous in their advance notifications of annual fishing activities. Virtually all catch reporting is gratifyingly swift and accurate, so there is confidence that management is maintaining control of catching activities, especially in Area 48, but a realistic notification system would render a good management system better. So far, an individual subarea was closed in terms of its own trigger level having been reached only in 2009/10, 2012/13, and 2013/14. Such an ability to close a subarea quickly is reflective of good reporting and management control, lending confidence to a belief that the system of managing krill catches is working effectively. Note, however, that two of the seasons when subarea trigger levels were closed were the most recent two completed seasons, so it is hardly surprising that there has been recent dialogue at CCAMLR about the relative levels of subarea trigger levels and even the overall PUCL. For now, though, the situation remains the same as it has been for several years.

3.3.3.3 Gear and product Krill fishing tends to be most successful near known geographic features or in waters associated with specific hydrographic properties (e.g. fronts), and vessels use echo-sounders to locate aggregations of appropriate size for their own operation, including to determine shape and orientation; searching speed is generally ~10 knots. They then set their gear to haul through the aggregation (at ~2 knots), but over time take cognizance of krill quality (oil content, etc) to maximize product value. In the midwater trawl system of the UoC fishery, care is also taken when deciding on aggregation size and fishing depth not to block the net by making catches that are too large for swift processing when brought aboard the vessel.

Krill aggregations tend to be smaller and catch rates lower during the earlier part of the fishing season, but both increase with day length during summer. As the season progresses into autumn and day length again shortens, sea-ice cover spreads north, the southern fishing grounds (e.g. Subarea 48.1) become less accessible, and total catches drop. However, variations between years in sea-ice cover render the pattern inconsistent between years. Fishing is around the clock if weather conditions, krill availability and ice cover allow, but later in each fishing season, night- time fishing tends to stop as the krill aggregations migrate diurnally and disperse.

Miller (2003) notes that most (>95%) krill swarms are <1 t, making fishing them subeconomic for most commercial operations, but there seem nevertheless to be sufficient aggregations of appropriate size in the focus Subareas 48.1, 48.2 and 48.3 to support the current level of fishing activity there. Fine-mesh pelagic trawls are deployed by the fleet and are brought aboard the vessels after each tow, brought to the side of the vessel and then pumped out, or fished continuously for up to four weeks with the krill being pumped continuously from the codend (another Norwegian fishery). Various net configurations are used by the fleets operating in the CCAMLR Convention Area currently, likely having some influence on gear selectivity (Krag et al. 2013), but such information is generally recorded by observers and submitted to the Commission; independent observer coverage throughout the fleet is not yet complete, but it is 100% in the UoC fishery. The latter fishery uses an effective sea lion excluder device (SLED) to block the entrance of the net to sea mammals and aquatic seabirds, and no such predator has been entrapped in the net during the UoC’s period of operating in the fishery (one Weddell seal was entangled with the gear on retrieval, but it freed itself and returned to the water).

The current main limitations on catching are likely processing capacity and attendant quality issues and of course market availability. The operator wants the best possible (fresh) product, to maximize product production, so tow-length, which tends often to be as short as 30 min, is decided by aggregation size and density. Product quality is always better if the raw material is processed soon after being caught and brought aboard. The continuous pumping system of the other Norwegian operator in the area has slightly different characteristics in terms of preferred size and density of krill aggregations.

Several krill products are produced from the catch in the UoC fishery, but the most lucrative is the small production of Omega 3 oil and other organic (mainly lipid) products. Other products from the fleet are fresh, frozen and minced krill, plus krill meal, and a small portion of the product is for human consumption, but most is for use in aquaculture feed. Antarctic krill actively feeding on phytoplankton tend to be avoided by the fleet if much of the final product is for human consumption because not only is such product considered less desirable by end-consumers, but

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 21

processing time has to be shorter to preclude quality concerns. As processing technologies have improved, there has been increased demand for peeled krill for meal production and for whole animals for bait or biochemical processing (Everson 2000). Meal production is also increasing year on year and the aquaculture market for whole krill is expanding. In fisheries targeting krill for human consumption and bait, some bycatch tends to be discarded, but it is always retained and used in the meal fishery; there is no discarding in the UoC fishery 3.3.3.4 The Unit of Certification (UoC) fishery Olympic Seafood AS, based at Fosnavåg, is a Norwegian seafood and speciality ingredients company, owned by Olympic Group of Norway. The holding company owns and operates a specialized krill vessel that works only in the Antarctic, near where it is based the whole year round even if not fishing, the FV Juvel (loa 99.5 m, 5500 grt) which was launched in 2009. Most of the krill harvested is used to produce ingredients for human consumption, i.e. krill meal and oil. Concentrate, which is a byproduct of the production process on-board, is currently mostly used for aquafeed production.

The Juvel’s operations are shown graphically in Figure 3. It operates a midwater trawl (either an Omega 350-m long or an Omega 300-m long Egersund krill trawl; code TM), with the larger net of horizontal opening 10–25 m (operating usually at the top end of this range), vertical opening 23 m and a 16 m codend constructed of 100 mm knotless netting fitted with a 11 mm knotless liner (the smaller net is similar but at the front end slightly differently configured). The main net mesh size ranges from 400 mm to 144 mm in the various panels, but the net holds an inner liner of 13 sections constructed of knotless 166 mm netting. Generally, Poly-ice doors of 3200 kg are deployed (distance between doors when operating, 40 m), but a spare set of Egersund pelagic doors (3600 kg) is carried. The net is operated well clear of the seabed, resulting in no interactions with the seafloor; it is also thoroughly cleaned before deployment to deter interactions with seabirds. Over the past three completed fishing seasons, the amount of krill biomass harvested (declared) was 11 278 t (2012), 18 918 t (2013) and 15 039 t (2014). Its formally notified (to CCAMLR) catch target for the 2014/2015 season (which includes interest in all of subareas 48.1– 48.4) is 35 000 t, about double its catch over the previous three seasons and therefore well above its historical annual take.

Figure 3. Juvel’s krill net and stern view

Source: RIMFROST Group

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 22

3.4 Principle Two: Ecosystem Background Principle 2 of the Marine Stewardship Council standard states that:

Fishing operations should allow for the maintenance of the structure, productivity, function and diversity of the ecosystem (including habitat and associated dependent ecologically related species) on which the fishery depends.

The following section of the report highlights some of the key characteristics of the fishery under assessment with regard to its wider impact on the ecosystem. 3.4.1 Ecosystem characteristics and fishery interactions

The fishery takes place in the Scotia Sea (Fig. 4), which is the area of water between Tierra del Fuego, the South Georgia, South Sandwich and South Orkney Islands, and the Antarctic Peninsula, and bordered on the west by the Drake Passage. It is located in both the Southern Ocean and the Atlantic Ocean.

Figure 4: Scotia Sea location

Source: "Scotia-sea" by Gi - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons - http://commons.wikimedia.org/wiki/File:Scotia-sea.png#mediaviewer/File:Scotia-sea.png

Normally stormy and cold, these island groups all sit on top of the Scotia Ridge, which frames the Scotia Sea on the north, east and south. The Scotia Sea covers an area of some 900 000 km2.

Figure 5 below represents the profile through the Drake Passage between South America and the Antarctic Peninsula. Morphological main features in the deep sea are shown, such as the mid- ocean ridge (West Scotia Ridge) separating the Antarctic plate from the South American plate. The South Shetland Islands belong to an island belt that resulted from the subduction of the Pacific plate below the Antarctic plate. The upper graph shows salinity (parts per thousand) and temperature in surface water measured along a cruise track. The Polar Front is an oceanographic feature characterized by an abrupt change in temperature of about one degree.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 23

Figure 5: Salinity, temperature and depth in the Drake Passage.

Source: Hannes Grobe, Alfred Wegener Institute for Polar and Marine Research. Creative Commons CC-BY-SA-2.5.

The major current in the Scotia Sea is the Antarctic Circumpolar Current (ACC; Fig. 6). It comes from the west (the Drake Passage) and is immediately deflected north by the Scotia Ridge, eventually turning back east, going around South Georgia and continuing east through the Atlantic portion of the Southern Sea. The Weddell Sea Surface Water (WSSW) is effectively constricted by the southern arm of the Scotia Sea, but is able to come through the Weddell–Scotia Confluence east of the South Orkney Islands.

Figure 6: Major currents in the Scotia Sea: Red is Weddell Sea Surface Water (WSSW) and Blue is the Antarctic Circumpolar Current (ACC).

Source: http://theseasproject.weebly.com/scotia-sea.html

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 24

Figure 7: Diagrammatic representation of a basic food web within the Antarctic ecosystem.

Source: http://extrememarine.org.uk/extremeantarctica/2014/11/16/a-keystone-species/

Historically, the Scotia Sea has been used for hunting whales and seals. The extensive hunting of whales in this area (and the entire Southern Ocean) has caused a change in the food web (see Fig. 7). In the 1900s, before the days of extensive hunting of whales, baleen whales were the main consumer of Antarctic krill. However, after the decline of their populations (see Fig. 8) some of the seal populations expanded and seals became the main consumer of krill and whales became the least influential consumer of krill.

Figure 8: Population data for whales and seals in the Southern Ocean.

Source: http://theseasproject.weebly.com/scotia-sea.html . Adapted from 'Biology of the Southern Ocean'

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 25

The Scotia Sea Islands are described as the Scotia Sea Islands tundra ecoregion. They are rocky and partly covered in ice and snow year-round, but support tundra vegetation despite their harsh weather conditions, consisting of mosses, lichen and algae, and flying seabirds, penguins and seals feed in the surrounding waters.

Flying seabirds include four species of albatross: black-browed, grey-headed, light-mantled and wandering. There are only five species of bird that remain on land on the islands, and they include two endemic species: the yellow-billed pintail duck and the South Georgia pipit. Other flying birds include the Southern giant petrel, with sizeable colonies on Bird Island.

Penguin species found in the region include large numbers of king penguins, especially on South Georgia, chinstrap, macaroni, gentoo, Adélie and rockhopper.

The seals present in the area include Antarctic and Subantarctic fur seals, leopard seals, Weddell seals, Southern elephant seals and crabeater seals.

In order to protect these species and the unique ecosystem there are several protected areas designated in the Scotia Sea:  CCAMLR Marine Protected Area (South of South Orkney Islands)  South Georgia and South Sandwich Islands Marine Protected Area  Benthic closed areas  Antarctic Specially Protected Areas (ASPAs)  Antarctic Specially Managed Areas (ASMAs)

Most of the Juvel’s fishing activity takes place in the Bransfield Strait (see Fig. 9), between the South Shetland Islands and the Antarctic Peninsula. The basin known as the Bransfield Trough is about 400 km long and 2 km deep, and hosts a chain of submerged seamounts of volcanic origin.

Figure 9: Map of the Antarctic Peninsula and the Bransfield Strait.

Source: http://www.eoearth.org/

The South Shetlands Islands comprises a group of >20 islands (11 major islands and several minor ones). Some are volcanically active, and >80% of the land area is permanently glaciated. Nonetheless, the islands host several research stations from different countries.

The sea around the islands is closed by ice from early April to early December and the monthly average temperature is <0°C for eight months of the year (April–November). The warmest months are January and February, and the coldest month is July. Mean summer and winter temperatures are only about 1.5°C and about –5°C, respectively. The effect of the cold ocean

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 26

tends to keep summer temperatures low and winter temperatures from falling as low as they do inland to the south.

The South Shetland Islands have experienced measurable glacier retreat in recent years, but despite this they remain >80% snow- and ice-covered throughout the summer.

The Antarctic Peninsula, which lies at the other side of the Strait, has the mildest climate on the Antarctic continent. It is there that Antarctica’s only two vascular plants grow, and it is there too that wildlife such as seals, penguins and flying seabirds are most plentiful. The west coast of the Antarctic Peninsula has a maritime Antarctic climate, similar to that of the South Sandwich, South Orkney and South Shetland offshore islands.

3.4.2 Retained species

According to MSC CR v1.3, retained species are those species that are retained by the fishery (usually because they are commercially valuable or because they are required to be retained by management rules). However, CCAMLR scientific observer reports from the Olympic fishery, which has always produced 100% observer coverage, specify that the Juvel is only dedicated to the catch of krill and the processing of krill products, and that all species recorded in the catch composition apart from krill are discarded. Therefore, there are no retained species to consider in this section. 3.4.3 Bycatch (discarded species)

Since its first trip, the Juvel has had voluntary 100% observer coverage, which at present is mandatory at 50%. CCAMLR Conservation Measure 51-06 (2012) regarding the General measure for scientific observation in fisheries for Euphausia superba, reaffirms the need for adequate monitoring and management of the krill fishery and recommends observer coverage and the use of the Scientific Observers Manual, according to the CCAMLR Scheme of International Scientific Observation.

The Scientific Observer Manual lists in Annex I the observer’s tasks, which include, among others: - sampling of catches to determine biological characteristics, - recording biological data by species caught, - recording bycatches, their quantity and other biological data, - recording entanglement and incidental mortality of birds and mammals, - recording the procedure by which declared catch weight is measured.

The manual defines the sampling methodology in Part II, section 11. The observer is instructed to select a haul, and to ensure that all large fish are removed from the conveyor during that haul and retained for subsequent weighing and identification. At the same time, the observer needs to take a 25 kg sample, remove all fish and record the total mass of each fish species. Then he/she needs to take a 10 kg subsample from the remaining krill sample and to sort carefully through it, removing and recording the total mass of each fish species. Following that, it is necessary to take two 1 kg subsamples from the remaining krill sample and to sort through each of these, again removing and recording the total mass of any remaining fish species (paying particular attention to larval fish). When taxonomic identification at sea is not possible, samples have be photographed and retained for further study. Part II, section 12, describes the observation methodology of interactions of seabirds and marine mammals with fishing operations, detailing periods and duration of these observations.

CCAMLR Scientific observer cruise reports submitted officially record these catch details for all species and a summary of the biological data collected. The comprehensive information on length, weight, sex or maturity of the sampled individuals is recorded in the observer’s electronic logbook. As for interactions with seabirds and marine mammals, these reports describe interactions with the gear or with the vessel, and the consequences of these interactions, along with sightings of these species from the deck.

During the most recent completed season (2013/14), 12 vessels from five CCAMLR Members took part in the Antarctic krill fishery. The total catch was 205 853 t, 74% of which was taken from Subarea 48.1. That subarea reached 98% (152 402 t) of its interim catch limit of 155 000 t on 17

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 27

May 2014, so it was closed for the rest of the season, an earlier closure than the previous two closures in the same subarea (June 2013 and October 2010).

As the Juvel does not retain on board any species apart from krill, all the species present in the catch (apart from the target species) are considered as bycatch. The factory is fitted with a mincing machine, which grinds the bycatch into a fine paste before it is discharged together with the stick water from the krill. As mentioned above, observers sample the catch composition in different hauls, identifying and quantifying all bycatch species present in the catch.

Table 2 lists the bycatch species noted as present in the Juvel’s observer reports from the years 2010 – 2014, most of which, but not all, are larvae. Observer reports from the years 2010 and 2011 list the number of individuals kept as samples whereas the reports from 2012, 2013 and 2014 refer to the number of individuals sampled during bycatch sampling of the different trawls.

According to observer reports, bycatch proportions from the Juvel are <0.5%, which means that all species are minor by-catch.

Table 2: Bycatch species present in the Juvel catch (number of individuals) FAO code Species English name 2014 2013 2012 2011 2010

ANI Champsocephalus gunnari Mackerel Icefish 3 2332 3 719 6

ANS Pleuragramma Antarctic antarcticum Silverfish 6 639 2549 94 4

AQM Themisto gaudichaudii Amphipod 7 0 0 0 0

BDJ Deepwater Bathydraco marri Dragonfish 0 1 0 0 0 BTI Bathydraconidae 0 8 0 0 1

DCP Natantian Eusirus perdentatus decapods nei 0 1 0 0 0 Antarctic ELT Electrona antarctica lanternfish 0 5 0 503 0

ERN Trematomus Emerald bernacchii Rockcod 0 1 0 0 0

FIC Cryodraco Long-fingered antarcticus Icefish 1 489 149 16 1 GEP Gempylidae Snake Mackerel 0 0 1 0 0

GYA Gymnodraco Antarctic acuticeps Dragonfish 0 54 0 0 0

GYB Gymnoscopelus bolini Lanternfish 0 0 0 5 0

GYN Gymnoscopelus Nichol's nicholsi lanternfish 0 0 0 18 0

GYR Gymnoscopelus braueri Lanternfish 0 0 0 2 0

ICA Southern Icichthys australis driftfish 0 0 0 1 0 ICX Chionodraco sp. Icefish nei 0 2 0 4 138 JEL Medusae Jellyfish 0 46 0 0 0 Neopagetopisis JIC ionah Crocodile Icefish 2 1 1 100 0

KIF Chionodraco rastrospinosus Ocellated Icefish 4 3079 0 0 0 KRA Myctophidae Lanternfish 0 20 35 304 0 LAC Lampanyctus Lanternfish 0 1 0 0 0

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 28

achirus

LIC Channichthys rhinoceratus Unicorn Icefish 0 0 2 447 0 MIC Chionodraco myersi Myers Icefish 0 30 1286 0 0 Muraenolepis Smalleye moray MOY microps cod 0 0 0 56 1

NOG Notothenia Humped gibberifrons Rockcod 0 206 8 24 0

NOK Stripe-eyed Notothenia kempi Rockcod 0 67 0 0 0 Nototheniops NOL larseni Painted Rockcod 0 1 2 609 4

NOM Paranotothenia Magellanic magellanica Rockcod 0 1 0 0 0

NON Yellowbelly Notothenia neglecta Rockcod 0 16 0 0 0 NOR Notothenia rossi Marbled Rockcod 0 39 0 0 0

NOS Notothenia squamifrons Grey Rockcod 0 23 0 3 0

NOT Patagonotothen Patagonian brevicauda Rockcod 0 3 3 0 0 NOX Nototheniidae Rockcods 0 2 0 29 20 Antarctic NTO Notolepsis coatsi Jonasfish 0 96 16 1912 2

PDG Paradiplospinus gracilis Slender escolar 0 0 0 9 0 Parachaenichthys PGE georgianus 0 0 1 2 0

PMA Pagetopsis macropterus Icefish Spp 0 37 0 0 0

PRE Protomyctophum Tenison's tenisoni lanternfish 0 0 0 38 0

PRY Protomyctophum Gaptooth choriodon lanternfish 0 0 0 8 0 RGG Ragovitzia glacialis 0 5 0 0 0

SGI Pseudochaenichthys South Georgia/ georgianus Blackfin Icefish 0 6 10 23 0 SQQ Teuthoidea Squid Spp. 0 10 12 0 0

SSI Chaenoocephalus aceratus Blackfin Icefish 7 21 33 129 0

TRD Trematomus Slender lepidorhinus Scalyhead 0 2 0 0 0 Chaenodraco WIC wilsoni Spiny Icefish 3 973 4 7 0

YDB Cryodraco spp. 0 0 40 0 0

Source: CCAMLR Scientific Observer reports for the years 2010–2014.

The WG-EMM report from August 2014 summarizes the frequency of occurrence, proportion by mass and length-frequency distribution of the fish taxa recorded in fish bycatch samples as part of the CCAMLR Scheme of International Scientific Observation from 9 303 hauls collected during 60 cruises involving 18 different vessels over the period 2010–2014. The frequency of occurrence of fish ranged from 10 to 98% between vessels and there were 14 taxa for which the frequency of occurrence was >1% in any subarea (of which seven were channichthyids) and the modal size of fish was between 5 and 10 cm. The Working Group agreed that, although the provision of data on fish bycatch is improving, there is still uncertainty regarding the frequency of fish bycatch in the krill fishery and therefore that it is not possible to provide a definitive view on whether the krill

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 29

fishery could have a role in the recovery of previously overexploited stocks and on the potential interactions with currently fished stocks (e.g. icefish).

Of all species listed by observers, those in the most parlous situation in terms of stock health are apparently Nototenia rossi and Champsocephalus gunnari. Notothenia rossii, the marbled rockcod, is a commercially important species native to the Southern Ocean where it can be found at depths of 5–350 m. It can reach a length of 92 cm although most are only around 50 cm long. It was heavily fished by the Soviet Union in the 1960s and 1970s, with catches exceeding 100 000 t in some seasons, almost disappearing from around South Georgia and by 1980 depleted throughout the Southern Ocean. The mackerel icefish, Champsocephalus gunnari, is a fish found only in the Southern Ocean, near Heard and McDonald Islands, South Georgia and Kerguelen. Because of overfishing, the FAO considers the species to be ‘depleted’.

In 2012 the Marine Assessment Resources Group (MRAG) carried out an analysis of the larval fish bycatch of the Aker Biomarine vessel Saga Sea, which targets krill in the same area. Observer reports from the Juvel and the Saga Sea reveal a similar catch composition, because both vessels fish the same stock in the same area, deploying a fine mesh net to prevent the entanglement of bigger organisms. For this reason, the bycatch rates are considered comparable between the two fisheries.

For the MRAG analysis, information was taken from the Saga Sea observer reports compiled for the period 2007–2011. The results showed that myctophid (lanternfish) and channichthyid (icefish) species dominated the bycatch, with lesser levels of notothenid species present. A standardized count of all icefish, lanternfish and notothenids present per tonne sampled would sum to 1 000 individuals together. As a precautionary proxy one could consider that each larva weighs 2 g, which would give a final weight of 2 kg of fish larvae per tonne of krill taken. In other words, 0.2% of the catch composition could be considered as fish larvae. The report on bycatch assumed that the bycatches of the channichthyids and nototheniids were exclusively Champsocephalus gunnari and Notothenia rossii, the species of most concern in the analysis. It is also noteworthy that very few of the total icefish species larvae recorded during those years in subareas 48.1 and 48.2 were actually C. gunnari, the species of main concern. The report concluded that it was highly unlikely that larval fish bycatch rates of the Saga Sea posed any threat to these groundfish stocks in Area 48. Observer reports from the Juvel show a similar catch composition with bycatch proportions <0.5%.

Most of the species listed by observers accompanying the Juvel were also covered in the MRAG report on larval finfish bycatch on the Saga Sea. The species listed below represent those not considered in the MRAG report nor evaluated by CITES or IUCN, although information on their stock status can be found on the CAMLR website (their catch rates for the years 2012, 2013 and 2014 are listed in Table 2 and are considered to be anecdotal).  Antarctic dragonfish (GYA)  Cryodraco spp. (YDB)  Grey rockcod (NOS)  Gaptooth lanternfish (PRY)  Gymnoscopelus bolini (GYB)  Gymnoscopelus braueri (GYR)  Nichol’s lanternfish (GYN)  Ocellated icefish (KIF)  Pagetopsis macropterus (Icefish) (PMA)  Slender escolar (PDG)  Southern driftfish (ICA)  Yellowbelly rockcod (NON)

In terms of bird interactions, CCAMLR Conservation Measure 25-03 covers the issue of minimizing the incidental mortality of seabirds and marine mammals in the course of trawling in the Convention Area and requires the fisheries to develop gear configurations that reduce the chance of birds or marine mammals encountering the net, such as Marine Mammal Exclusion Devices. The observation methodology for interactions of seabirds and marine mammals with fishing operations is established in Part II, section 12 of the CCAMLR Scientific Observer Manual, where periods and duration of these observations are detailed.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 30

The fishing strategy, with a slow towing speed (<2 knots), quick sinking of the net on deployment, and the layout of the trawl warps, which enter the water very close to the stern of the vessel and reduce the potential for birds to strike them during fishing operations, all contribute to the sparseness of interactions recorded by observers during fishing operations.

Tables 3 and 4 list bird interactions with the vessel and the gear as recorded by observers (Table 3) during the past two years, and their IUCN status (Table 4). All species are considered of least concern by IUCN, apart from one gentoo penguin entangled with the net but released alive.

Table 3: Number of bird interactions with the Juvel and its gear, 2010–2014. (D: Dead, A: Alive) 2014 2013 2012 2011 2010 Interactions with Interactions with Interactions with Interactions with Interactions with Total gear vessel gear vessel gear vessel gear vessel gear vessel

Species D A D A D A D A D A D A D A D A D A D A D A Total

Wilson’s storm 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 2 2 petrel

Cape 0 1 0 0 4 0 0 18 0 0 2 14 0 0 0 0 0 0 1 0 7 33 40 petrel Southern 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 2 fulmar

0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 1 2 Blue petrel

Kelp gull 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1

Snow 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 petrel Chinstrap 0 0 0 2 0 0 0 0 0 4 0 0 0 0 0 0 0 3 0 0 0 9 9 penguin

Gentoo 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 penguin Total per year 0 1 1 4 4 2 0 18 0 4 4 17 0 0 0 0 0 3 1 0 10 49 59

Source: CCAMLR Scientific observer reports

Table 4: Seabirds interacting with the Juvel’s fishing gear and their CITES and IUCN status.

FAO Common name Species CITES IUCN status code CCO Wilson’s storm petrel Oceanites oceanicus No Least Concern

DAC Cape petrel Daption capense No Least Concern

FUG Southern fulmar Fulmarus glacialoides No Least Concern

HAC Blue petrel Halobaena caerulea No Least Concern

LDO Kelp gull Larus dominicanus No Least Concern

PWP Snow petrel Pagodroma nivea No Least Concern

PYN Chinstrap penguin Pygoscelis antarctica No Least Concern

PYP Gentoo penguin Pygoscelis papua No Near Threatened

Source: DNV GL

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 31

The indirect effects of the fishery on gentoo penguins were studied by Nicoll and Douglass (2012), by mapping summer foraging ranges and overlaying it on known fishing activity areas of Aker Biomarine’s Saga Sea for the period 2007–2011 (Fig. 10). The study considered that the average maximum distance travelled by penguins from their breeding colony was some 20 km and that gentoo penguins prefer to forage on the shelf where the water depth is <200 m (Kokubun et al. 2010). It concluded that there is a low overall level of overlap between foraging distribution and fishing activities of the Aker Biomarine krill fishery around the coasts of the Antarctic Peninsula and the South Orkneys.

Figure 10: Gentoo penguins foraging areas distribution.

Source: Nicoll and Douglass (2012) Table 5 lists all seabird species seen by observers during all trips of the vessel, as recorded in observer reports.

Table 5: Bird species seen by observers on the Juvel between 2010 and 2014.

Species Common name IUCN Status Catharacta maccormicki Antarctic skua Least Concern Chionis albus Snowy sheathbill Least Concern Daption capense Cape petrel Least Concern Diomedea chrysostoma Grey headed albatross Endangered Diomedea exulans Wandering albatross Vulnerable Diomedea melanophris Southern black browed albatross Near Threatened Freggeta tropica Black bellied storm petrel Least Concern Fulmarus glacialoides Southern fulmars Least Concern Halobaena caerulea Blue Petrel Least Concern Laridae Kelp gulls Least Concern

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 32

Larus dominicanus Southern black-backed gull Least Concern Macronectes giganteus Southern giant petrel Least Concern Macronectes halli Northern giant petrel Least Concern Oceanites oceanicus Wilson storm petrel Least Concern Pachyptila desolata Antarctic prion Least Concern Pagodroma nivea Snow petrel Least Concern Pelecanoides urinatrix Common diving petrel Least Concern Phalacrocorax atriceps Imperial shag Least Concern Phoebretia palpebrata Light-mantled sooty albatross Near Threatened Procellaria cinerea Grey petrel Near Threatened Puffinus grisius Sooty shearwater Near Threatened Pygoscelis adeliae Adelie penguin Least Concern Pygoscelis papua Gentoo penguin Near Threatened Pygoselis antarctica Chinstrap penguin Least Concern Thalassoica antarctica Antarctic petrel Least Concern

Source: CCAMLR Scientific observer reports.

In order to minimize these interactions, the Juvel has established a protocol that includes the following steps:

 Deckhands to go on hourly rounds to check for live birds on deck, especially during windy and snowy days.  Always inform the observer about birds found.  Return every live and unharmed bird to the air/sea.  Lights to be kept at a minimum during nights without compromising safety.

As mentioned above, CCAMLR Conservation Measure 25-03 also covers the subject of minimizing the incidental mortality of marine mammals, and requires the fisheries to develop gear configurations such as a Sea Lion Excluder Device (SLED).

As stated earlier, the Juvel is equipped with a 200 mm square nylon mesh panel that forms the SLED, spliced onto the floor, roof and sides of the front meshes, effectively closing off the net entrance (Fig. 11). This device is made from Dyneema®, which is a light and strong polyethylene based fibre. The meshes are 2 mm in diameter, which allows the net opening to widen by approximately 18 m without floats and 25 m with floats. The net is shown in an earlier section, as Figure 3.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 33

Figure 11: Juvel’s SLED (white portion of net)

Source: RIMFROST Group In order to prevent interactions with seabirds and marine mammals, offal is not discarded during shooting or hauling; the offal is retained and processed. ’Stick water’ from the factory is occasionally discharged, however, but no fresh, minced or partly ground krill is discarded. All discarded product is ground up to a paste.

According to CCAMLR Scientific Observer Manual, observers are required to record interactions and sightings of marine mammals. The following section on ETP species covers the records of marine mammals which are subject to the CITES Agreement. In this section we will cover interactions with marine mammals which are not subject to this agreement.

Observer reports for the years 2010–2014 show abundant sightings of Weddell seals (Leptonychotes wedellii) in the Gerlache Strait and also in the proximity on Greenwich Island. The Weddell seal is a relatively large and abundant true seal with a circumpolar distribution surrounding Antarctica. Because of its abundance, relative accessibility, and ease of approach by humans, it is the best-studied of the Antarctic seals. An estimated 800 000 individuals remain today. The only direct interaction with this species was on April 2013, when a Weddell seal swam onto the trawl net during hauling, but slipped off on the trawl ramp. The seal sustained no injuries and swam off apparently unharmed by the experience.

Observers have also recorded sporadic sightings of minke whales (Balaenoptera acutorostrata), currently the most abundant rorqual in the world, with a population status considered stable throughout almost all of its range (especially relative to other species of large whale). Minke whales have been recorded swimming close to the Juvel and over the net, but no entanglements have arisen. Observer reports from the years 2010 and 2011 also reveal sporadic sightings of beaked whales (Mesoplodon spp.) and Leopard seals (Hydrurga leptonyx).

Minke whales and Weddell and leopard seals are considered as Least Concern by IUCN while beaked whales are considered to be data deficient.

The Juvel has a protocol of mitigation measures that serve to prevent and reduce damage to marine mammals. These measures include:  Make sure the SLED is properly maintained.  Check all parts of the fishing gear regularly (at least once a day) for damage.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 34

 Tie up all loose ends and mend all holes.  Haul and shoot the net as quickly as possible.  Avoid as far as possible shooting and hauling the net in areas where there are large numbers of seals and penguins.  Make sure the net and the codend is cleaned before shooting.  Discharge of ‘stick water’ to be minimized during hauling and shooting.

There are also measures to follow in case a seal is on deck:  Crew to avoid blocking the seal's escape route back to the ocean.  Every care to be taken to reduce stress to and ensuring no injury to the animal when it is being returned to the sea alive.  Do not antagonize the animal.  Use netting and/or sheets of plywood as moving barriers to herd the animal back to sea.  High pressure seawater hoses can also be used.  The observer to be informed.

Indirect effects of the fishery on fish, seabirds and marine mammals would be those caused by competition for the krill resource. Murphy et al. (2007) studied seabird and seal predation in the Scotia Sea, and their dependence on krill abundance (Fig. 12). They concluded that species also look for alternative breeding options in years when krill is scarce (Fig. 13).

Figure 12: Proportional consumption of different groups of prey by the major predators in the Scotia Sea.

Source: Murphy et al. 2007.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 35

Figure 13 : Schematic illustration of alternative pathways in part of the Scotia Sea foodweb, showing shifts between (A) yYears when krill are abundant across the Scotia Sea, and (B) years when krill are scarce. Major pathways are shown as black arrows.

Source: Murphy et al. 2007.

However, removal by the fishery has been estimated to be orders of magnitude below the demand from predators and the biomass available to both predators and the fishery. Hewitt et al. (2004) estimated the annual consumption of krill in Area 48 for different predators, showing a total demand of 15 223 thousand tonnes for all of them. Murphy et al. (2007) also estimated the annual consumption of krill (106 tonnes per year) by the main krill predators in the Scotia Sea foodweb (Fig. 14). However, those estimates are based mainly on summer studies that are likely to overestimate the importance of krill in the diet.

Figure 14: Estimates of annual consumption of krill (106 tonnes per year) by the main krill predators in the Scotia Sea foodweb.

Source: Murphy et al. 2007.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 36

3.4.4 Endangered, Threatened or Protected Species (ETP)

According to MSC v1.3, ETP species are those listed in CITES Appendices. This is why most marine mammals and birds recorded by observers have already been considered as bycatch species.

The list of Endangered, threatened or protected species anticipated by CITES in Area 48 are:  Arctocephalus gazelle (Antarctic fur seal)  Australophocoena dioptrica (Spectacled porpoise)  Balaenoptera bonaerensis (Antarctic minke whale)  Balaenoptera borealis (Sei whale)  Balaenoptera musculus (Blue whale)  Balaenoptera physalis (Fin whale)  Cephalorhynchus commersonii (Piebald dolphin)  Cygnus melancoryphus (Black-necked swan)  Eubalaena australis (Southern right whale)  Hyperodon planifrons (Southern bottlenose whale)  Lammna nasus (Porbeagle shark)  Megaptera novaeangliae (Humpback whale)  Mirounga leonine (Elephant seal)  Orcinus Orca (Killer whale)  Phocoena dioptrica (Spectacled porpoise)  Physeter macrocephalus (Sperm whale)  Spheniscus humbodti (Humboldt penguin)

Following CCAMLR Observer Manual guidelines, interactions and sightings of any marine mammals and birds are recorded by the observer. Reports from the years 2010–2014 do not record any direct interaction or entanglement with ETP species, and lists the sighting of the following ETP species:

 Arctocephalus gazelle (SEA) (Antarctic fur seal) Abundant sightings of Antarctic fur seals in 2010, 2011, 2012 (near Gerlache Strait), 2013 (next to the Antarctic Peninsula) and 2014 (in the proximity of Greenwich Island). In 2013 one observer reports that 15 Antarctic fur seals were seen to play or interact with the net during hauling by swimming over and under the net’s forward and middle sections.  Balaenoptera physalis (FIN) (Fin whale) Sporadic sightings of small groups of fin whales (2-4 animals) in 2011, 2012, 2013 and 2014. They all avoided contact with the vessel.  Eubalaena australis (EUA) (Southern right whale) Reported in 2011 in area 48.3. Whales could be seen on clear days distinguished by the callosities on their heads appearing white due to the large colonies of cyamids (whale lice) and the v-shaped blowhole.  Megaptera novaeangliae (HUW) (Humpback whale) Sightings in 2010, 2012, 2013 and 2014, usually solitary or in small groups. Reports from 2012 record that on a few occasions humpback whales approached within a few metres of the vessel in the Gerlache Strait.  Orcinus orca (KIW) (Killer whale) Observer reports from 2013 say that two pods of killer whales were observed in two different occasions. They did not approach the vessel.  Eight unidentified whales (i.e. distant blow)

For the past three years, the Juvel has only operated in Subareas 48.1 and 48.2. Moreover, VMS tracks for January–May 2013 show that the vessel constrained its activity to the coasts of the Antarctic Peninsula and the South Shetlands (Fig. 15).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 37

Figure 15: Juvel’s VMS track records between 15 January 2013 and 1 May 2013.

Source: RIMFROST Group.

Figure 16: Juvel’s VMS track records between 15 January 2013 and 1 May 2013. Detail map.

Source: RIMFROST Group Indirect effects of the fishery on Antarctic fur seals and Aker Biomarine’s krill fishery were studied by Nicoll and Douglass (2012), revealing a high degree of overlap of year-round fishing operations and the summer foraging ranges of this species (Fig. 17). This extensive overlap would also be expected in Olympic’s krill fishery using the Juvel.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 38

Figure 17: Saga Sea summer and non-summer fishing effort 2007–2011 and summer Antarctic fur seal foraging ranges.

Source: Mapping selected krill predator summer foraging ranges with fishing activity of Aker BioMarine’s Saga Sea 2007-2011.

As mentioned above, CCAMLR Conservation Measure 25-03 is intended to minimize the incidental mortality of seabirds and marine mammals in the course of trawling in the Convention Area. Vessels are required to develop gear configurations to reduce the chance of interactions. The Juvel uses the SLED described above and has also established a protocol in order to avoid interactions with these species.

The International Whaling Commission (IWC) actively analyses information for the area and in 2010 held a workshop with CCAMLR to review the input of data for Antarctic marine ecosystem models.

Interactions with cnidarians or hydrozoans on the seabed, or with the seabed itself, are not expected owing to the pelagic nature of the gear.

3.4.5 Habitats

The UoC krill fishery operates with pelagic gear which should prevent any damage to the seafloor, because such gear rarely if ever touches the seafloor.

CCAMLR Observer reports record the maximum, minimum and average depth of fishing activity, along with bottom depth. Fishing average depth for the years 2010–2014 ranges from 5 to 400 m, with bottom depths ranging from 76 to 4658 m. Interactions with the seafloor could therefore only be expected in case of gear loss, which has never happened since the Juvel commenced operations. Habitat and benthic communities are not expected to be altered by the krill fishery.

Nonetheless, CCAMLR has established several measures in order to protect the habitat and benthic communities. The first Marine Protected Area, which is located in subarea 48.2, was established in 2009 through Conservation Measure 91-03, and is bounded by a line starting at 61°30'S 41°W, thence due west to 44°W longitude, thence due south to 62°S, thence due west to 46°W, thence due north to 61°30'S, thence due west to 48°W longitude, thence due south to 64°S latitude, thence due east to 41ºW longitude, thence due north to the starting point (Fig. 18). This Marine Protected Area is intended for the protection of the South Orkney Islands southern shelf. The Conservation Measure prohibits all types of fishing activities within the defined area, with the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 39

exception of scientific fishing research activities agreed by the Commission for monitoring or other purposes. Depth contours in Fig. 18 are at intervals of 1000 m.

Figure 18: CCAMLR Marine Protected Area for the protection of the South Orkney Islands.

Source: CCAMLR Conservation Measure 91-03

According to Toropova et al. (2010), Marine Protected Areas are recognized as one of the most effective means for achieving ecosystem-level conservation, so protecting marine biodiversity and mitigating key threats and pressures on marine environments and resources. They are able to achieve both conservation and fisheries management objectives, as well as providing a foundation for ecosystem-based management.

CCAMLR Conservation Measure 91-04 on the General Framework for the Establishment of CCAMLR Marine Protected Areas states that the Commission will, on the basis of the advice of the Scientific Committee, adopt a research and monitoring plan for an MPA. Every five years, Members conducting activities according to or related to the research and monitoring plan will compile a report on those activities, including any preliminary results for review by the Scientific Committee.

At the most recent CCAMLR meeting in Hobart in October 2014, participants commented strongly on the need to establish new MPAs in the region. Specifically, several countries asked for the establishment of new MPAs in the East Antarctic and Ross Sea regions, but any decision was postponed due to the lack of unanimous consensus.

The Antarctic Treaty also defines areas of special scientific or biological significance, which are designated according to CCCAMLR Conservation measure 91-02 (2012), on the Protection of the values of Antarctic Specially Managed and Protected Areas (Antarctic Specially Protected Areas (ASPAs) and Antarctic Specially Managed Areas (ASMAs)). Management plans for these areas can be found on the Antarctic Protected Areas (APA) database on the Antarctic Treaty Secretariat (ATS) website, http://www.ats.aq/devPH/apa/ep_protected.aspx?lang=e, with information on /about their management plans and the aims of their designation.

Area 48 contains the following Specially Managed and Protected Areas:  ASPA 144, Chile Bay, Greenwich Island, South Shetland Islands (Subarea 48.1)  ASPA 145, Port Foster, Deception Island, South Shetland Islands (Subarea 48.1)  ASPA 146, South Bay, Doumer Island, Palmer Archipelago (Subarea 48.1)  ASPA 152, Western Bransfield Strait, South Shetland Islands (Subarea 48.1)  ASPA 153, Eastern Dallmann Bay, Palmer Archipelago (Subarea 48.1)  ASPA 149, Cape Shirreff, South Shetland Islands (Subarea 48.1)  ASPA 151, Lions Rump, South Shetland Islands (Subarea 48.1)  ASMA 1, Admiralty Bay, South Shetland Islands (Subarea 48.1)  ASMA 3, Deception Island, South Shetland Islands (Subarea 48.1)  ASMA 7, Southwest Anvers Island, Palmer Archipelago (Subarea 48.1).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 40

In 2012 the South Georgia and South Sandwich Islands Government declared an MPA with spatial and temporal limitations in order to protect the high diversity of birds, marine dependent predators, benthic communities and marine flora and fauna which are abundant within its territory (Fig. 19). These limitations include: - seasonal closure (November 1st until March 31st) of the Antarctic krill fishery, - ban on bottom fishing deeper than 2,250 m - a 12 nautical mile No-take Zone around the island of South Georgia and around Clerke Rocks, Shag Rocks and Black Rock, - A 3 nm No-take Zone around each of the South Sandwich Islands and a 12 nm area around each of the islands that is closed to pelagic fishing. - Bottom trawling is prohibited in the Marine Protected Area and bottom fishing with other gears is only allowed between the depths of 700 and 2,250 m.

Figure 19: Chart illustrating the location of the South Georgia and South Sandwich Islands MPA Area and the additional No-take Zones. The section of the Maritime Zone south of 60ºS is not part of this MPA, but no fishing is licensed there.

Source: SGSSI Marine Protected Area Management Plan

The South Georgia and South Sandwich Islands Government has also declared several Benthic Closed Areas in order to protect its benthic biodiversity. These areas are: - West Shag - West Gully - The Northern benthic closed area - The Eastern benthic closed area - Southern Seamounts - North Georgia Rise - North East Georgia Rise - Protector Shoals - Kemp Seamount and Calderas

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 41

We must here stress that the Juvel has not fished near the South Georgia and South Sandwich Islands for the past three years, because it has focused its fishing activity on subareas 48.1 and 48.2.

As regards bathymetrical information on the Southern Ocean, the Australian Antarctic Data Centre has created some maps which can be downloaded at the Antarctic Treaty webpage (Fig. 20).

Figure 20: Bathymetry of the Southwest Atlantic.

Source: http://www.ats.aq/e/info.htm

For information about data sources used in this map refer to: http://aadc-maps.aad.gov.au/aadc/mapcat/display_map.cfm?map_id=13438

3.4.6 Ecosystems

CCAMLR’s Ecosystem Monitoring Programme (CEMP) was established in 1989 to monitor the effects of fishing on both harvested species (target species) and dependent species (predators), in order to assist CCAMLR with its task of regulating the commercial harvesting of Antarctic marine living resources in accordance with the ecosystem approach embodied in Article II (www.ccamlr.org).

The two aims of CEMP are: 1. To detect and record significant changes in critical components of the marine ecosystem within the Convention Area, to serve as a basis for the conservation of Antarctic marine living resources. 2. To distinguish between changes due to harvesting of commercial species and changes due to environmental variability, both physical and biological.

CEMP's major function is to monitor the key life-history parameters of selected dependent species to detect changes in the abundance of harvested species. ‘Dependent species’ are those marine predators for which species targeted by commercial fisheries are a major component of their diet. In the case of ‘krill-dependent species’ used in CEMP they include land-based species such as seals, penguins, petrels and albatross. This decision is consistent with the existing overlap between krill

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 42

fishing areas and the foraging ranges of these predators. Interactions with land-based predators were studied by Nicoll and Douglass (2012), but information on the potential impact of fishing on pelagic predators such as whales is still scarce.

The feedback approach to krill fisheries management adopted by CCAMLR is supposed to enable it to adjust the management measures in response to relevant new information on the interaction between krill fishery and krill predators as it becomes available. In this way, CCAMLR ensures continuous adjustment of the measures in accordance with the information available. However, at present there is no direct link between this monitoring programme and specific management objectives, and the CEMP assessment of the impacts of krill fishing on dependent species still remains to be integrated into long-term management procedures.

CCAMLR CEMP Scientific Committee developed a set of Standard Methods for monitoring predator parameters that include details of how data should be collected, formats for submission of the data to the CCAMLR Secretariat and procedures for data analysis, in order to facilitate data analysis and comparison between predator monitoring studies in the context of the different CEMP sites. Monitoring predator and prey distributions serves to identify new areas for CEMP monitoring.

Atkinson et al. (2012) described different methods for sampling krill, such as:  with nets (for historical time-series, demographic information and live krill);  acoustics (distribution, time-series, biomass and swarm-scale information);  the fishery (sustained sampling in one place and wide area and time coverage);  via predators (long time-series, demographic indices).

As CCAMLR members take part in CEMP on a voluntary basis, contributions to data gathering depend on national research programmes and priorities. In terms of environmental protection of the CEMP sites, there is no direct mechanism to protect these sites, but 7 of the 13 currently active CEMP monitoring sites south of 60oS are within ASPA or ASMA protection boundaries.

The CCAMLR Working Group on Ecosystem Monitoring and Management (WG-EMM) first met in 1995 after the amalgamation of the WG on krill (WG Krill) and the WG on CCAMLR Ecosystem Monitoring Programme (WG-CEMP), and it is responsible for the design and coordination of the monitoring programme and the analysis and interpretation of the data arising from it. According to CCAMLR, the WG-EMM shall:  Assess the status of krill.  Assess the status and the trends of dependent and related populations including identification of information required to evaluate predator/prey/fisheries interactions and their relationship to environmental features.  Assess environmental features and trends that may influence abundance and distribution of harvested, dependent, related and/or depleted populations.  Identify, recommend and coordinate the research necessary to obtain information on predator/ prey/fisheries interactions, particularly those involving harvested, dependent, related and/or depleted populations.  Liaise with WG-FSA on stock-assessment-related matters.  Develop further, coordinate the implementation of, and ensure continuity in CEMP.  Taking into account assessments and research carried out, develop management advice on the status of Antarctic marine ecosystems and the management of krill fisheries in full accordance with CCAMLR Convention Article II.  In addition, provide advice on aspects of spatial protection, including MPAs and vulnerable marine ecosystems.

The programme’s largest component is the monitoring of dependent species (predators), but it also monitors harvested species, harvesting strategies and environmental parameters, and analyses these data annually, trying to distinguish between changes attributable to harvesting and those attributable to environmental variability.

The WG-EMM has already acknowledged difficulties in differentiating the effects of fishing from those of climate change, and has reported that:  at current harvesting levels it was unlikely that the existing design of CEMP, with the data available to it, would be sufficient to distinguish between ecosystem changes attributable to harvesting of commercial species and changes attributable to environmental variability, whether physical or biological

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 43

 with the existing design of CEMP, it may never be possible to distinguish between these different and potentially confounding causal factors and that the Scientific Committee should seek advice from the Commission on the extent to which further work should be directed towards this topic  without a real ability to separate the confounding effects of harvesting and environmental variation and in the context of uncertainty, the Scientific Committee should seek advice from the Commission about the policy of how management should proceed when a significant change was detected, but no single causal factor could be attributed  one possible method that may assist in the separation of confounding effects of harvesting and environmental variation would be the establishment of an experimental fishing regime whereby fishing would be concentrated in local areas in conjunction with an appropriate predator monitoring programme.

At the most recent WG-EMM meeting held in Chile in July 2014, it was noted that the development of a Feedback Management Strategy in Subarea 48.2 is critically dependent upon the initiation of new research and monitoring, especially from fishing vessels, which will serve to support adjustment of catch limits and the spatial distribution of catches therein. CCAMLR members are encouraged to use the recently created pro forma for submission of Stage 2 Feedback Management ideas. Other priorities highlighted in this meeting were the need for further development of multispecies models to support its work in developing feedback management strategies for krill. The Working Group also noted recent progress in the development of spatially explicit and multispecies population models for toothfish and toothfish prey fitted to fisheries data, including multispecies interactions and harvest by fisheries, and that similar approaches may be useful for top predators and for krill.

The WG-EMM meeting of 2009 pointed out that climate change may induce rapid changes within the ecosystem, impacting the way indices generated by CEMP are being used to detect fisheries impacts, as the life history and demography of Antarctic krill are intimately tied to seasonal sea ice conditions, climate, and the physical forcing of ocean currents. Key spawning, recruitment and nursery areas of krill are located in the Western Antarctic Peninsula (Constable et al. 2003). The climate there is warming rapidly, and as a result, the extent and duration of winter sea ice is reducing (Parkinson 2012). Constable et al. (2003) also show that diminished sea-ice cover over the past 20 years might result in greater recruitment variability and lower overall abundance of krill in the Southwest Atlantic, whereas previously, recruitment may have been more stable and less variable. Changes in krill abundance will surely have an impact on krill-dependent predators.

The Southern Ocean is a major component within the global ocean and climate system and potentially the location where the most rapid climate change is most likely, particularly in high latitude polar regions. There, even small temperature changes can potentially lead to major environmental perturbations, and a failure of Antarctic krill recruitment will inevitably foreshadow recruitment failures in a range of higher trophic-level marine predators (Trathan et al. 2007).

Recent rapid climate change is now well documented in the Antarctic, particularly around the Antarctic Peninsula. One of the most evident signs of climate change has been ice-shelf collapse; overall, 87% of the Peninsula’s glaciers have retreated in recent decades. Further ice-shelf collapse will lead to the loss of existing marine habitats and to the creation of new habitats, with consequent changes in both ecological processes and in community structure, with changes from a unique ice-shelf-covered ecosystem to a typical Antarctic shelf ecosystem, with high primary production during a short summer. This process is likely to be among the largest ecosystem changes on the planet. The habitats previously covered under ice shelves present outstanding opportunities to undertake science related to habitat colonization. Studying these habitats when they become available would provide valuable scientific insights into how communities develop over timescales ranging from years to decades. Habitats under ice shelves have been closed to both terrestrial and pelagic community interactions over recent geological time-scales. If exposed, they would offer a range of opportunistic study sites, often with contrasting ecological scenarios. Long-term reference areas would facilitate scientific study of the effects of such changes, primarily in the absence of any effects caused by other human activities (Trathan and Grant 2013). Reid and Croxall (2001) also agree that there is need for long-term data series on krill predators, in order to understand and interpret natural ecosystem variability and how large-scale physical processes influence small-scale ecology in the Antarctic. Hewitt and Low (2000) highlighted that an extensive and well-designed monitoring programme, including both fishing and non-fishing

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 44

areas, would be key to the timely detection of local or regional adverse effects on krill or krill predators from a long-term krill decline that may be magnified by the krill fishery.

There is general agreement between scientists that there is a need for more protected areas in the region: - in areas with high species biodiversity, particularly for those predators that feed on krill, in order to improve the knowledge of how the ecosystem operates in the absence of fishing. It could be that the combination of the bathymetry, oceanography and the movements of krill could explain the biodiversity in the area. - as reference areas (with no fishing), in order to evaluate the impacts of climate change without the impact of fishing. Changes in the physical properties of the marine system are especially important for CCAMLR and include, inter alia, changes in ocean acidification (Bednarek et al. 2012), and in ocean temperature (Gille 2002) and reductions in the extent and timing of seasonal sea-ice (Stammerjohn et al. 2008) and the retreat and collapse of ice shelves, glaciers and ice tongues (Cook and Vaughan 2010; Cook et al. 2005; Gutt et al. 2010, 2013; Rignot et al. 2013).

The WG-EMM already noted in its most recent meeting that changes in pCO2 are already manifest in the Southern Ocean and that the physiological cost to krill will rise, so krill vulnerability to stress will increase.

Risk maps for Antarctic krill under projected Southern Ocean acidification show that the embryonic development of Antarctic krill in elevated seawater CO2 levels and successful hatching is impaired at CO2 levels >1000 μatm. Exposure to elevated CO2 during the first three days of embryonic development significantly retards subsequent development, even if the embryos are transferred to seawater with current ambient levels of CO2. Krill embryos appear more vulnerable to ocean acidification than other pelagic crustaceans such as copepods. pCO2 in the Southern Ocean is predicted to rise to above 1500 μatm in some parts of krill’s depth range by the year 2100 unless emissions are mitigated. Risk maps, combining modelled hatch rates and the three-dimensional circumpolar projection of future pCO2, predict that by 2100 the Weddell Sea and the waters to the east are the highest-risk areas for krill embryos. The entire Southern Ocean south of the Polar Front is predicted to be unsuitable for hatching by the year 2300, which would lead to the collapse of the krill population. Such changes highlight the need to think about future decision rules used for the management of the fishery. For instance, the present decision rules are based upon a pre- exploitation estimate of biomass (B0), but under conditions of environmental change, this may not be realistic and so alternative reference points may be required.

There is a global need to develop strategic frameworks for assessing uncertainty in ecosystem dynamics models. Such models have already been used within CCAMLR to evaluate options for managing the Antarctic krill fishery in the Scotia Sea and southern Drake Passage (Hill and Mathews 2013). However, the use of models to evaluate catch allocation options illustrates the tension between the ideal of well-constrained models and the reality of ecosystem-based management problems in which data is sparse, structure is complex and uncertainty is rife (Hill et al. 2007; Plagányi 2007).

There is tension between the parameter stability benefits of well-constrained models and the use of model conditioning to identify plausible alternative hypotheses in data-poor situations (Hill and Mathews 2013).

The krill resource has already been subject to many studies and models that try to provide a greater understanding of its role in the ecosystem, but these models may be difficult to apply to real life. Models that incorporate interspecific interaction typically have more parameters than single-species models. However, increasing complexity also leads to accumulation of uncertainties and increased difficulty in interpreting results (Fulton et al. 2003; Raick et al. 2006; Plagányi 2007; Hill et al. 2007).

Atkinson et al. (2012) reviewed different broad categories of model representing Antarctic krill, their data sources and limitations. The categories include:  models exploring specific aspects of krill biology such as life cycle, energetics or behaviour (Hofmann and Hùsrevõglu 2003; Murphy et al. 2004);

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 45

 multispecies population models, simulating either historical changes in the abundance of krill and its predators or the effects of harvesting on interacting species (May et al. 1979; Murphy 1995);  single-species population projection models, for instance to quantify regional catch limits (Constable et al. 2000);  spatial single-species models, such as that of Marin and Delgado (2001), which showed that some 80% of the krill catch was taken from within penguin foraging areas near the Antarctic Peninsula, suggesting that fisheries are in direct spatial competition with predators (Hewitt et al. 2002, 2004);  mass-balance regional foodweb models incorporating krill, such as the preliminary Ecopath with Ecosim (EwE) model of the Antarctic Peninsula ecosystem, Subarea 48.1 (Cornejo- Donoso and Antezana 2008); the model shows that phytoplankton, zooplankton and krill account for most of the mass flow, and describes the foodweb as dominated by the phytoplankton-krill-top predators chain, complemented with alternative food pathways (e.g. through Electrona antarctica);  a spatial multispecies operating model (SMOM) of krill–predator fishery dynamics, which has been used to evaluate proposed management measures for the krill fishery in the Scotia and Bellingshausen Seas (Plagányi and Butterworth 2012); the model describes the underlying population dynamics, and is used in simulations to compare different management options for adjusting fishing activities (e.g. different spatial distribution of catches), and allows the discrimination of the ecosystem impacts of different spatial fishing allocations;  models of krill transport at the maximum advection rate indicated by the Ocean Circulation and Climate Advanced Modelling Project (OCCAM), with the aim of evaluating the large- scale ocean circulation and interpreting data coming from the World Ocean Circulation Experiment (WOCE; Rintoul et al. 2001).

One of the major nursery grounds for Antarctic krill lies to the north of the Antarctic Peninsula (Siegel 1988; Brinton 1991; Spiridonov 1995; Siegel et al. 2002), close to the area of recent rapid regional warming (King 1994). Ocean currents are thought to carry krill from this area to other areas of the Southwest Atlantic (Hofmann et al. 1998; Murphy et al. 1998; Thorpe et al. 2004). Consequently, changes in the environment close to the nursery grounds have the potential to have far-reaching impacts on both local and more distant marine communities (Trathan et al. 2007). Ecosystem studies in the Southwest Atlantic have pointed to strong relationships between temperature and the abundance of Antarctic krill (Trathan et al. 2003), so climate warming needs to be taken into account in CCAMLR fisheries management strategy, as suggested by CCAMLR Scientific Committee (SC-CAMLR-XXXII, pp. 63–65).

Temporal patterns in krill recruitment suggest that there is a direct causal relationship between variability in sea-ice cover, krill recruitment, prey availability and predator foraging ecology, and that large-scale forcing associated with climate variability may be governing ecological interactions between ice, krill and their predators in the western Antarctic Peninsula and Scotia Sea regions (Fraser and Hofmann 2003).

Murphy et al. (2007) studied the main physical and biological processes important in determining the dynamics of the Scotia Sea ecosystem (Figs 21 and 22). These factors also affect krill recruitment and dispersal across the region, generating observed correlations of changes in krill density and biomass and higher trophic level predator foraging and breeding performance with sea ice and larger indices of oceanic and climatic variation.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 46

Figure 21: Schematic of the main physical processes generating variation in the Scotia Sea ecosystem. ENSO, El Niño Southern Oscillation; ACC, Antarctic Circumpolar Current.

Source: Murphy et al. (2007)

Figure 22: Schematic of the temporal and spatial scales of the main physical and biological processes important in determining the dynamics of the Scotia Sea ecosystem. The 1:1 relationship is based on the scale of physical mixing in the oceans. Note that the physical and biological processes are illustrated offset above and below this line, respectively for clarity. The shaded grey block illustrates the natural spatial and temporal scale of Scotia Sea processes. Acronyms used include PD, Population Dynamics, SST, Sea Surface Temperature and ENSO, El Niño Southern Oscillation.

Source: Murphy et al. (2007)

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 47

As reported by Smith et al. (2011), fishing low trophic level (LTL) species, even at conventional levels associated with maximum sustainable yield (MSY), can have a great impact on other parts of the ecosystem, particularly when they constitute a high proportion of the biomass in the ecosystem or are highly connected in the foodweb.

The foodweb of the Scotia Sea is highly heterogeneous, widely distributed and dynamically connected through ocean circulation. The ecosystem is dominated by the flows of the major current systems (the Antarctic Circumpolar Current and the Warm Swallow Current) and by its seasonality, manifested by the advance of sea ice across the region during winter. This unique environment is high in both nutrients and chlorophyll-a. The role of krill in the ecosystem is crucial, because the resource provides the major link between LTL production and consumption by higher trophic level predators across the Scotia Sea (Murphy et al. 2007). Different ecosystem models show that changes in primary production and detritus are responsible for most of the declines within the model, implying this is a bottom-up ecosystem (Hoover et al. 2012).

Krill attract large quantities of top predators (Howard et al. 2004) and are considered to be a keystone species (Moline et al. 2000), linking most pathways in the food chain from primary producers to top predators. In addition, krill around the Antarctic Peninsula are believed to be the main source of krill populations around the Scotia Sea (Atkinson et al. 2004), suggesting that krill are important not only in the inmediate area where the population is deemed to be large, but also to surrounding areas, and to predators there (Hofmann et al. 1998; Brierley et al. 1999; Atkinson et al. 2004). Hence, changes to the krill population at the Antarctic Peninsula will affect predators locally and across the Scotia Sea (Hoover et al. 2012).

Watters et al. (2013) used a spatially resolved, stochastic predator–prey fishery model to simulate ecosystem dynamics in the Scotia Sea and to evaluate krill management measures that were based on an allowable catch of krill and a Catch Allocation Option. The main uncertainties identified were krill natural mortality, krill movement between areas and the response of krill predators to variations in prey availability. The aim of the simulation was to evaluate potential ecosystem responses to a range of management measures, and specifically to compare the Catch Allocation Options identified. These simulations nominally represented a 20-year period of fishing beginning in 2007, followed by 20 years without fishing. The model consisted of a set of four scenarios that bracketed key uncertainties. Results were presented as estimates of the risk associated with CCAMLR failing to meet representative management objectives if it implemented a given management measure. The different Catch Allocation Options were based on: 1. the spatial distribution of historical catches during the 2002/03–2006/07 fishing seasons (hereinafter referred to as Catch); 2. the simulated spatial distribution of predator demand for krill at the start of 2007 (Demand); 3. the simulated spatial distribution of krill standing stock biomass at the start of 2007 (Stock). 4. the Catch Allocation Option (Current) based on current management (CM 51-07 in CCAMLR 2011) which limited the spatial distribution of krill catches and capped the maximum catch at the interim catch limit.

The predicted risks that krill fishing might negatively impact ecosystem productivity were generally low; the Demand option was the least risky (within the range of allowable catches considered, all predator subpopulations had a <50% chance of being depleted under the Demand option). The Catch option was the most risky, with several predator subpopulations having a >50% chance of depletion as catches increased between the interim and regional catch limits. The levels of risk under the Stock option were between those of the Catch and Demand options, and six predator subpopulations had a >50% chance of depletion at catches less than or close to the regional catch limit. The option based on Current management had low risks of negatively impacting ecosystem health.

To summarize, the simulations suggested that, in the case of the Scotia Sea krill fishery, there may be appreciable increases in risk, particularly to ecosystem health and resilience, as catches increase from the interim to the regional catch limit. However, the interim limit apparently caps these risks at roughly half the corresponding risks at the regional limit. The levels of risk at comparable catches were sensitive to the Catch Allocation Option, as was the relationship between risk and catch. The Catch option posed the greatest risks to each objective, the Stock option posed intermediate risks, and the Demand option was simultaneously “best” for both the fishery and the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 48

ecosystem. CCAMLR Conservation Measure 26-01 on General Environmental Protection during fishing regulates the disposal of plastic packaging bands, food waste and sewage, the incineration output, and prohibits the dumping or discharging of garbage and oil or fuel products or oily residues into the sea. Compliance with this measure is reported by scientific observers, and it is an important part of the Friends of the Sea Certification process.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 49

3.5 Principle Three: Management System Background Principle 3 of the Marine Stewardship Council standard states that:

The fishery is subject to an effective management system that respects local, national and international laws and standards and incorporates institutional and operational frameworks that require use of the resource to be responsible and sustainable.

In the following section of the report a brief description is made of the key characteristics of the management system in place to ensure the sustainable exploitation of the fishery under assessment.

3.5.1 Jurisdiction and management system

The fishery is managed mainly by CCAMLR, in interaction with the Norwegian Ministry/Directorate of Fisheries and the Government of South Georgia and the South Sandwich Islands (GSGSSI). CCAMLR coordinates scientific research and observer programmes, establishes TAC and distributes quotas between subareas. The Norwegian Ministry/Directorate of Fisheries issues fishery permits and performs quota control of the client vessels. GSGSSI issues permits for the vessels in the SGSSI Maritime Zone.

CCAMLR was established in 1982 with the objective inter alia of conserving Antarctic marine life. Based on the best available scientific information, the Commission agrees a set of conservation measures that determine the use of marine living resources in the Antarctic. The key institutional components of CCMLAR are the CAMLR Convention (with supportive regulations), the decision- making Commission, the Scientific Committee and the Secretariat located in Hobart, Tasmania. The Commission determines the regulatory framework applied to the management of each fishery in the Convention Area, including catch limits and seasonal or area closures and measures aimed at minimizing potential impacts of fishing activities on non-target species and the ecosystem. The Standing Committee on Implementation and Compliance, subordinate to the Commission, provides it with information, advice, recommendations on fishery monitoring and compliance. The Scientific Committee provides the Commission with the best available scientific information on harvesting levels and other management issues. In turn, the Commission is obliged by the Convention to take full account of the recommendations and advice of the Scientific Committee in making its decisions. The Scientific Committee takes into account the outcomes of research from national programmes of CCAMLR members. In addition CCAMLR has established a number of programmes to collect the data required for the effective management of the Southern Ocean, including fisheries monitoring, scientific observers on fishing vessels and ecosystem monitoring.

Norway has a well-established system for fisheries management, which has evolved over more than a century and is now codified in its 2008 Marine Resources Act. The Act provides for a formal system of cooperation between regulatory bodies of governance, such as the Ministry of Fisheries and Coastal Affairs, the Directorate of Fisheries and the Coast Guard, and further for cooperation between management authorities and scientific research institutes, primarily the Institute of Marine Research. The 2008 Integrated Management Plan for the Norwegian Sea provides for cooperation between different sector authorities, such as the Ministry of Fisheries and Coastal Affairs and the Ministry of Environment.

GSGSSI is based in Government House in the Falkland Islands and has a Director of Fisheries among its staff. It is mainly involved in the licensing of vessels that fish in the South Georgia Maritime Zone, catch monitoring at King Edward Point in South Georgia and at-sea surveillance in the Maritime Zone. A Marine Protected Area was introduced in 2012 and reinforced in May 2013, including a no-fish zone within 12 nautical miles of the coast.

3.5.2 Decision making and consultation processes

The Antarctic and Southern Ocean Coalition (ASOC) had been actively involved in marine management in the Antarctic since the establishment of CCAMLR and was given observer status in 1991. ASOC is also a key partner to the Antarctic Krill Conservation Project, which is an

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 50

international effort managed by the Pew Foundation to secure from CCAMLR an ecosystem-based fisheries management programme for krill that is highly precautionary, scientifically based and protects the unique environment of the southern polar region. Olympic is part of the Association of Responsible Krill Harvesting Companies (ARK), which has, among other things, organized meetings between industry, science, NGOs and management authorities prior to sessions in CCAMLR.

At national level in Norway, WWF is actively consulted on krill issues by Norwegian fisheries management authorities. For instance, the Ministry of Foreign Affairs invites stakeholders, including WWF, to meetings before CCAMLR meetings in order to discuss relevant issues, including for the management of krill. WWF has been invited to become part of the Norwegian delegation to CCAMLR, but has chosen to remain an independent actor.

3.5.3 Objectives and regulatory measures

All CCAMLR fisheries are managed within a precautionary and ecosystem approach, as defined by the FAO in its Code of Conduct for Responsible Fisheries, and are consistent with MSC Principles and Criteria. At national level in Norway, the 2008 Marine Resources Act, which covers all living marine resources, requires that Norwegian fisheries management be guided by the precautionary approach and by an ecosystem approach that takes into account habitats and biodiversity.

The three main objectives of the CCAMLR management system are (Article 2 of the Convention): a) prevention of decrease in the size of any harvested population to levels below those which ensure its stable recruitment; for this purpose its size should not be allowed to fall below a level close to that which ensures the greatest net annual increment; b) maintenance of the ecological relationships between harvested, dependent and related populations of Antarctic marine living resources and the restoration of depleted populations to the levels defined in sub-paragraph a) above; and c) prevention of changes or minimization of the risk of changes in the marine ecosystem which are not potentially reversible over two or three decades, taking into account the state of available knowledge of the direct and indirect impact of harvesting, the effect of the introduction of alien species, the effects of associated activities on the marine ecosystem and of the effects of environmental changes, with the aim of making possible the sustained conservation of Antarctic marine living resources.

CCAMLR‘s more specific, short- and long-term strategy for achieving these objectives is reflected in Conservation Measure 51-01 (2010). A precautionary krill catch limit of 5.61 million tonnes is set for Area 48, based on the potential yield estimate. This is well above the current catch and will allow for expansion. However, a ‘catch trigger’ (620 000 t) is set with the objective not to be exceeded until a procedure for division of the overall catch limit into smaller scale management units has been established, based on advice from the Scientific Committee. The objective of such division is to avoid possible unacceptable concentration of catch within the foraging areas of vulnerable predators. Although the trigger level is close to the highest global annual catch to date, it is significantly more than the largest annual catch to date in Area 48.

In general, CCAMLR has well-established decision-making processes. They allow for stakeholder input and clear scientific analysis of the data available within the Working Groups and Scientific Committee, and they result in conservation measures and fisheries strategies designed to achieve their short- and long-term fishery-specific objectives.

3.5.4 Monitoring, Control and Surveillance (MCS)

CCAMLR provides a clear and comprehensive monitoring system and control framework for Antarctic fisheries. Surveillance of CCAMLR fisheries is undertaken by Member States and incorporates the CCAMLR observer scheme. For the client fishery, enforcement is mainly taken care of by the Norwegian Directorate of Fisheries, which has demonstrated a consistent ability to enforce relevant regulations. Vessels are licensed on an annual basis and report catches from each haul through their electronic logbooks. In order to receive a license for the Antarctic krill fishery, Norwegian vessels are obliged to have an observer on board at all times. When entering the South Georgia Maritime Zone, vessels need to apply for a licence and pay a fee. There, all vessels are inspected by the South Georgia administration at King Edward Point before they are allowed to

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 51

start fishing. They have to report catches on a daily basis and are also inspected by a patrol vessel during fishing operation.

3.5.5 Research plan and management evaluation

A comprehensive research plan by CCAMLR exists for krill fisheries, focusing on the monitoring of krill catches, scientific observation and environmental monitoring. The CCAMLR Ecosystem Monitoring Programme (CEMP) provides cross-cutting data on the environment and predator abundance to link into fisheries data, and it targets research at an ecosystem approach to management of the krill fishery. CCAMLR conducts ongoing internal reviews of its processes and the performance of its Member States to meet the fishery-specific management requirements outlined. These requirements are reviewed annually (to fit in with the annual fisheries cycle) by the appropriate CCAMLR Working Groups (e.g. seabird mortality is analysed by the Working Group on Incidental Mortality of Associated Fauna).

CCAMLR was subject to a comprehensive external performance review during 2008, but such external review is not regular. The review was carried out by a panel appointed by the Commission composed of nine persons (see http://www.ccamlr.org/pu/E/revpanrep.htm). The purpose of the performance review was to evaluate the Commission‘s performance against comprehensive criteria and specifically against the objectives and principles set out in Article II of the Convention. The review confirms that the stock status and trends are broadly consistent with Article II of the Convention and international best practice. With particular reference to krill fisheries, the review identified the need for ongoing research into predator–prey linkages in ecosystem modelling and adequate monitoring and management within krill fisheries.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 52

4 EVALUATION PROCEDURE

4.1 Harmonised Fishery Assessment The fishery has been harmonized with the Aker Biomarine Antarctic Krill Fishery, which was recertified in January 2015. There are no differences in scoring between the two fisheries for Principle 1 or Principle 3, but there are slight differences between the two fisheries in terms of Principle 2 as a result of the differences in the gear used by the two client groups, a continuous pumping system for the Aker Biomarine fishery and a midwater trawl for the Olympic fishery. This difference implies that larval fish species present in the catch are considered as retained species in the Aker Biomarine fishery but as bycatch in the Olympic krill fishery. The bycatch Performance Indicator also takes into account interactions with other species such as seabirds or marine mammals that are not considered as ETP species, and which have been reported to be minimal in observer reports. The Olympic fishery has established a working procedure that includes two crew members who check the catch and remove anything that is not krill. This procedure contributes to the high scoring of the bycatch Performance Indicator. As for the results of the ETP Performance Indicator, records on ETP interactions have been reviewed in order to identify and quantify these interactions. The lack of direct interactions with these species contributes to the high scoring obtained for P2.

Performance Indicators Aker Biomarine Olympic Retained species 2.1.1 Outcome 80 100 2.1.2 Management 80 100 2.1.3 Information 90 100 Bycatch species 2.2.1 Outcome 100 100 2.2.2 Management 100 100 2.2.3 Information 100 100 ETP species 2.3.1 Outcome 95 100 2.3.2 Management 100 100 2.3.3 Information 95 95 Habitats 2.4.1 Outcome 100 100 2.4.2 Management 100 100 2.4.3 Information 85 85 Ecosystem 2.5.1 Outcome 100 100 2.5.2 Management 80 80 2.5.3 Information 90 90

There are no other overlapping fisheries in the area covered by the UoC.

4.2 Previous assessments There have been no previous assessments of this client.

4.3 Assessment Methodologies The basis for the MSC-certification is the standard denoted as the MSC Fishery Standard - Principles and Criteria for Sustainable Fishing, v 1.1, organized in three main principles.

 Principle 1 concentrates on the need to maintain the target stock at a sustainable level;  Principle 2 draws attention to maintaining the ecosystem in which the target stock exists;  Principle 3 addresses the requirement for an effective fishery management system in order to fulfil Principles 1 and 2. In addition, Principle 3 takes into account national and international regulations. The Principles 1–3, with pertaining criteria, are presented below. The assessment was carried out using MSC Certification Requirements v1.3. The assessment team used the default assessment tree without adjustments as defined in the MSC Certification

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 53

Requirements v1.3. The MSC Full Assessment Reporting Template V1.3 is used as basis for this report.

PRINCIPLE 1 A fishery must be conducted in a manner that does not lead to overfishing or depletion of the exploited populations and, for those populations that are depleted, the fishery must be conducted in a manner that demonstrably leads to their recovery1: Intent: The intent of this principle is to ensure that the productive capacities of resources are maintained at high levels and are not sacrificed in favour of short-term interests. Thus, exploited populations would be maintained at high levels of abundance designed to retain their productivity, provide margins of safety for error and uncertainty, and restore and retain their capacities for yields over the long term. Criteria:

 The fishery shall be conducted at catch levels that continually maintain the high productivity of the target population(s) and associated ecological community relative to its potential productivity.

 Where the exploited populations are depleted, the fishery will be executed such that recovery and rebuilding is allowed to occur to a specified level consistent with the precautionary approach and the ability of the populations to produce long-term potential yields within a specified time-frame.

 Fishing is conducted in a manner that does not alter the age or genetic structure or sex composition to a degree that impairs reproductive capacity.

PRINCIPLE 2 Fishing operations should allow for the maintenance of the structure, productivity, function and diversity of the ecosystem (including habitat and associated dependent and ecologically related species) on which the fishery depends. Intent: The intent of this principle is to encourage the management of fisheries from an ecosystem perspective under a system designed to assess and restrain the impacts of the fishery on the ecosystem. Criteria:

 The fishery is conducted in a way that maintains natural functional relationships among species and should not lead to trophic cascades or ecosystem state changes.

 The fishery is conducted in a manner that does not threaten biological diversity at the genetic, species or population levels and avoids or minimizes mortality of, or injuries to endangered, threatened or protected species.

 Where exploited populations are depleted, the fishery will be executed such that recovery and rebuilding is allowed to occur to a specified level within specified time-frames, consistent with the precautionary approach and considering the ability of the population to produce long-term potential yields.

PRINCIPLE 3 The fishery is subject to an effective management system that respects local, national and international laws and standards and incorporates institutional and operational frameworks that require use of the resource to be responsible and sustainable. Intent:

1 The sequence in which the Principles and Criteria appear does not represent a ranking of their significance, but is rather intended to provide a logical guide to certifiers when assessing a fishery. The criteria by which the MSC Principles will be implemented will be reviewed and revised as appropriate in light of relevant new information, technologies and additional consultations.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 54

The intent of this principle is to ensure that there is an institutional and operational framework for implementing Principles 1 and 2, appropriate to the size and scale of the fishery. Part A: Management System Criteria The fishery shall not be conducted under a controversial unilateral exemption to an international agreement. The management system shall:

 Demonstrate clear long-term objectives consistent with MSC Principles and Criteria and contain a consultative process that is transparent and involves all interested and affected parties so as to consider all relevant information, including local knowledge. The impact of fishery management decisions on all those who depend on the fishery for their livelihoods, including, but not confined to subsistence, artisanal, and fishing-dependent communities shall be addressed as part of this process.

 Be appropriate to the cultural context, scale and intensity of the fishery, reflecting specific objectives, incorporating operational criteria, containing procedures for implementation and a process for monitoring and evaluating performance and acting on findings.

 Observe the legal and customary rights and long term interests of people dependent on fishing for food and livelihood, in a manner consistent with ecological sustainability.

 Incorporate an appropriate mechanism for the resolution of disputes arising within the system2.

 Provide economic and social incentives that contribute to sustainable fishing and shall not operate with subsidies that contribute to unsustainable fishing.

 Act in a timely and adaptive fashion on the basis of the best available information using a precautionary approach particularly when dealing with scientific uncertainty.

 Incorporate a research plan (appropriate to the scale and intensity of the fishery) that addresses the information needs of management and provides for the dissemination of research results to all interested parties in a timely fashion.

 Require that assessments of the biological status of the resource and impacts of the fishery have been and are periodically conducted.

 Specify measures and strategies that demonstrably control the degree of exploitation of the resource, including, but not limited to:

- Setting catch levels that will maintain the target population and ecological community’s high productivity relative to its potential productivity, and account for the non-target species (or size, age, sex) captured and landed in association with, or as a consequence of, fishing for target species.

- Identifying appropriate fishing methods that minimise adverse impacts on habitat, especially in critical or sensitive zones such as spawning and nursery areas.

- Providing for the recovery and rebuilding of depleted fish populations to specified levels within specified time-frames. - Mechanisms set to limit or close fisheries when designated catch limits are reached. - Establishing no-take zones where appropriate.

 Contain appropriate procedures for effective compliance, monitoring, control, surveillance and enforcement that ensure that established limits to exploitation are not exceeded and specify corrective actions to be taken in the event that they are. Part B: Operational Criteria The fishing operation shall:

2 Outstanding disputes of substantial magnitude involving a significant number of interests will normally disqualify a fishery from certification.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 55

 Make use of fishing gear and practices designed to avoid the capture of non-target species (and non-target size, age, and/or sex of the target species); minimize mortality of this catch where it cannot be avoided, and reduce discards of what cannot be released alive.

 Implement appropriate fishing methods designed to minimize adverse impacts on habitat, especially in critical or sensitive zones such as spawning and nursery areas.

 Not use destructive fishing practices such as fishing with poisons or explosives.  Minimize operational waste such as lost fishing gear, oil spills, on-board spoilage of catch, etc.

 Be conducted in compliance with the fishery management system and all legal and administrative requirements.

 Assist and cooperate with management authorities in the collection of catch, discards and other information of importance to effective management of the resources and the fishery. The MSC Principles and Criteria presented above set the requirements for the fishery that undergoes certification. MSC’s certification methodology is based on a structured hierarchy of Subcriteria and Performance indicators. The overall performance is decided on the basis of the scoring criteria that the fishery gets during assessment. These subcriteria and performance indicators have been developed by the MSC in the form of a default assessment tree. When a fishery is evaluated the performance indicators (normally specific statements or questions) are checked out, and each performance indicator has three different ‘scoring guideposts’ that can be defined. MSC characterizes these scoring points as follows:

 Perfect practice, representing the level of performance that would be expected in a theoretically ‘perfect’ fishery (100 points).

 Exemplary or best practice (80 points).  Minimum sustainable practice (60 points).

4.4 Evaluation Processes and Techniques Site visits to the fishery were performed by the certification body (here DNV GL) and the assessment team and consultations were had with interested stakeholders. The performance indicators and the pertinent scoring systems were evaluated, and it was judged whether the fishery meets the requirements for MSC certification.

Default performance indicators and the scorings allocated in the evaluation are presented in Chapter 6.2.

4.4.1 Site Visits Relevant stakeholders were identified and stakeholder meetings scheduled and carried out in Bergen and Oslo on 8 and 9 December 2014. Persons consulted and key issues discussed during these site visits are outlined in table 6.

Table 6. Site visits conducted and key issues discussed.  Name  Affiliation  Date  Key issues Johanna Client 08.12.2014 Info about client and the fishery Tanhuanpaa, representatives  Ownership Jacob Remøy, Olympic Seafood  History of the fishery Arnt Olav  Organizational structure Aarseth, Fishing operations: Kristin Iversen  Fishing season  Fishing areas  UoC Fleet  Fishing practices: o Gears used o Fishing depth o Historical fishing levels (quotas and

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 56

landings) o Composition of catch o Information on discarding o Sampling and weighing on board o Closed areas o Loss of fishing gear Impact of fishing activity on ecosystem:  List of all bycaught fish species (species and quantities)  Any species apart from krill that is retained on board as part of the catch (species and quantities)  Bycatch of ETP species  Mitigation measures to avoid impacts on ETP species and birds.  List of commercial/non-commercial species which are usually discarded (quantities/if known)  Protected or sensitive habitats within geographical range of target stock  Loss of fishing gear, and recovery  Effect of gear used on the habitat  Reporting & registration of bycatch/discards  Sorting/separation of bycatch  Sampling Management, compliance with rules and regulations  Fishery management plans  Disputes with national/international authorities for the past 5 years.  Records of any sanctions and penalties known in 2013, 2014 (if any).  Control and surveillance: o VMS system o Landing control o Quota control o Inspections on board o Observer reports Chain of Custody start:  Review of traceability system on board and at landing  Labelling of products  First point of landing  First point of sale  Main products  Main markets

Norwegian Authorities and 08.12.2014 Management Fisheries research:  Fisheries Management & Regulations Directorate Norwegian  Consultation and decision-making process Modulf Overvik Fisheries  Mechanisms for resolution of legal disputes Directorate  Review of regulations for krill in CCAMLR area 48.1, 48.2, 48.3, 48.4

Institute for  Harvest strategy for krill Marine Institute for  Long-term objectives for the Antarctic Research, Marine Research fisheries Tor Knutsen,  Strategy for minimizing or eliminating ETP Georg Skaret bycatch  Strategy and plans for protecting sensitive habitats  Strategies in terms of scientific research,

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 57

Norwegian and international  Research programmes (Norwegian and international) for the fishery under assessment and other fisheries seeking the same stock  Control, surveillance and monitoring routines/ regulations applied to Norwegian krill fishery in in CCAMLR areas 48.1, 48.2, 48.3, 48.4 (and elsewhere?)  Logbooks: recording of landings and discards (of non-commercial species)  Significant discrepancies, if any, found at landing control for krill fisheries in the years 2012-2014.  Observed fishing patterns (gear used, fishing area, number of boats, fishing seasons)  Level of slippage/discarding in the UoC krill fishery, and, if known, in other fisheries on the same stock  Compliance with laws and regulations  Evaluation of management system

Research  Stock status, stock structure and recruitment  Review of Limit and Target reference points established for the stock  Approach to stock assessments  Research programmes for the krill fishery under assessment and krill in general in the target area  Sampling programmes (Norwegian and international) and the level of sampling  Level of discarding (composition of species, quantities)  Level of bycatch (composition of species, quantities)  Monitoring programmes for ETP species. Can the extent of interactions with ETP species be quantified?  Impact of fisheries on ecosystem  Impact of fisheries on marine habitats  Research planning Mette Authorities: 09.12.2014 Management/politics Fisheries Management & Regulations Strengehagen, Norwegian  Consultation and decision-making Odd Gunnar Ministry of  processes Skagestad Foreign Affairs  Mechanisms for resolution of legal disputes

 Review of regulations for krill in CCAMLR areas 48.1, 48.2, 48.3, 48.4  Harvesting strategy for krill  Long-term objectives for the Norwegian krill fisheries  Strategies for minimizing or eliminating ETP bycatch  Strategies regarding scientific research programmes, in Norway and internationally.  Research programmes for the krill fishery under assessment and for krill in general in

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 58

the target area (incl. other fisheries)  Strategy and plans for protecting sensitive habitats. Karoline Andaur, NGOs: 09.12.2014  Krill stock status Fredrik Myhre WWF  Fishery impact on the ecosystem o Impact on associated fish stocks o Interaction with ETP species o Impact of fishery on ETP species and habitats o Impact of fishery on ecosystem o Impact of fishery on marine habitats  Management system; institutional and operational framework for implementing Principles 1 and 2, appropriate to the size and scale of the fishery o Engagement of stakeholders o Management system  Surveying of the krill stock generally

In addition, skype meetings were held with:

 Cefas (Centre for Environment, Fisheries and Aquaculture Science, UK) o Chris Darby: discussing stock assessment and related management issues, and an overview of review processes within CCAMLR

 MRAG (Marine Resources Assessment Group, London) o James Clark: discussing observer sampling methodology as regards catch and species composition.

A main concern of scientists, NGOs and managers alike was that no fully synoptic survey has been conducted since 2000.

4.4.2 Consultations Information on the assessment process was made publicly available through www.msc.org at given stages of the assessment as outlined in Table 7. In addition, all relevant stakeholders identified at the start of the assessment (26 stakeholders) were contacted through direct e-mails and provided the possibility of monitoring the assessment process and providing feedback to the assessment team.

No stakeholder comments were submitted during the stakeholder consultancy period prior to the site visit. Information gathered during the site visits and through contact with the stakeholders after the site visit formed the main basis of the stakeholder consultancy for this assessment (see. section 4.4.1 above).

Table 7. Consultations at different stages of the assessment Date Information Media

2 September 2014 Notification of Full assessment Notification on MSC website Direct E-mail to listed stakeholders

2 September 2014 Notification of Assessment Team Notification on MSC website Direct E-mail to listed stakeholders

16 September 2014 Confirmation of Assessment Team Notification on MSC website Direct E-mail to listed stakeholders

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 59

25 September 2014 Notification of revised Assessment Notification on MSC website Direct Team E-mail to listed stakeholders

25 September 2014 Announcement of use of Default Notification on MSC website Direct Assessment Tree E-mail to listed stakeholders

Week 39 2014 Advertisement of certification + Advertisement on Invitation to contribute to www.intrafish.com assessment process Week 39 2014 Advertisement of certification + Advertisement on Invitation to contribute to www.olympicseafood.com assessment process 30 September 2014 Notification site visit Notification on MSC website Direct E-mail to listed stakeholders

14 October 2014 Notification of site visit revised Notification on MSC website Direct E-mail to listed stakeholders

30 October 2014 Confirmation of use of Default Notification on MSC website Direct Assessment Tree E-mail to listed stakeholders

30 October 2014 Confirmation of revised Assessment Notification on MSC website Direct team E-mail to listed stakeholders

16 December 2014 Notification peer reviewers proposed Notification on MSC website Direct E-mail to listed stakeholders

13 January 2015 Notification reviewers confirmed Notification on MSC website Direct E-mail

12 May 2015 Variation request: Target Eligibility Notification on MSC website Direct Date extension E-mail to listed stakeholders

12 May 2015 Variation response: Target Eligibility Notification on MSC website Direct Date extension E-mail to listed stakeholders

14 May 2015 Notification of Public Comment Draft Notification on MSC website Direct Report E-mail to listed stakeholders

14 May 2015 Revised timeline Notification on MSC website Direct E-mail to listed stakeholders

2 July 2015 Notification of Final report Notification on MSC website Direct E-mail to listed stakeholders

4.4.3 Evaluation Techniques The full assessment was publicly announced on 2 September 2014 through www.msc.org and supplemented with advertisements on www.intrafish.com and www.olympicseafood.com week 39, 2014. The assessment team produced the assessment in English to secure worldwide coverage of potential stakeholders.

At the start of the assessment, the assessment team compiled a stakeholder list based on guidance from the client and team member experience with other assessments of like fisheries. The list covers at least 26 stakeholders and has been used at every stage of the consultation process.

The site visit took place on 8 and 9 December 2014 in Bergen and Oslo. All members of the assessment team specified in section 2.1 were present at all meetings held. Stakeholder

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 60

consultations were performed in the form of direct meetings. Information on meeting participants and issues discussed is provided above, in Table 6. Some additional information was received from the stakeholders during skype meetings during and after the site visit.

The performance indicators and the pertaining scoring systems were evaluated jointly by the assessment team and all scoring was based on unanimous conclusions by the entire team during the scoring meetings which took place in Oslo on 10 and 11 December 2014.

In order to fulfil the requirements for certification the following minimum scores are required:

 The fishery must obtain an average score of 80 or more for each of the three MSC Principles, based on the weighted aggregate scores for all Performance Indicators under each Criterion in each Principle.  The fishery must obtain a score of 60 or more for each Performance Indicator under each Criterion in each Principle.

The Olympic Seafood Antarctic krill fishery achieved a score of 80 or more for each of the three MSC Principles, and did not score less than 80 for any of the set MSC Criteria. The assessment team therefore recommends certification of this fishery for the client Olympic Seafood.

As no indicators scored less than 80 there are no conditions.

Default performance indicators and the scores allocated in the evaluation are enclosed in section 6.2.

The set of scoring elements that have been considered in each outcome PI in Principle 2 are included in Table 8.

Table 8. Scoring elements

PI Scoring elements Main/not Data-deficient or component main not

2.2 Balaenoptera acutorostrata Not main Non data-deficient 2.2 Bathydraco marri Not main Non data-deficient 2.2 Bathydraconidae Not main Non data-deficient 2.2 Chaenodraco wilsoni Not main Non data-deficient 2.2 Chaenoocephalus aceratus Not main Non data-deficient 2.2 Champsocephalus gunnari Not main Non data-deficient 2.2 Channichthys rhinoceratus Not main Non data-deficient 2.2 Chionodraco myersi Not main Non data-deficient 2.2 Chionodraco rastrospinosus Not main Non data-deficient 2.2 Chionodraco spp. Not main Non data-deficient 2.2 Cryodraco antarcticus Not main Non data-deficient 2.2 Cryodraco spp. Not main Non data-deficient 2.2 Daption capense Not main Non data-deficient 2.2 Electrona antarctica Not main Non data-deficient 2.2 Eusirus perdentatus Not main Non data-deficient 2.2 Fulmarus glacialoides Not main Non data-deficient 2.2 Gempylidae Not main Non data-deficient 2.2 Gymnodraco acuticeps Not main Non data-deficient 2.2 Gymnoscopelus bolini Not main Non data-deficient 2.2 Gymnoscopelus nicholsi Not main Non data-deficient 2.2 Gymnoscopelus braueri Not main Non data-deficient 2.2 Icichthys australis Not main Non data-deficient 2.2 Halobaena caerulea Not main Non data-deficient 2.2 Lampanyctus achirus Not main Non data-deficient 2.2 Larus dominicanus Not main Non data-deficient 2.2 Leptonychotes wedellii Not main Non data-deficient 2.2 Medusae Not main Non data-deficient

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 61

2.2 Myctophidae Not main Non data-deficient 2.2 Muraenolepsis microps Not main Non data-deficient 2.2 Neopagetopisis ionah Not main Non data-deficient 2.2 Notolepsis coatsi Not main Non data-deficient 2.2 Notothenia gibberifrons Not main Non data-deficient 2.2 Notothenia kempi Not main Non data-deficient 2.2 Notothenia neglecta Not main Non data-deficient 2.2 Notothenia rossi Not main Non data-deficient 2.2 Notothenia squamifrons Not main Non data-deficient 2.2 Nototheniidae Not main Non data-deficient 2.2 Nototheniops larseni Not main Non data-deficient 2.2 Oceanites oceanicus Not main Non data-deficient 2.2 Pagetopsis macropterus Not main Non data-deficient 2.2 Pagodroma nivea Not main Non data-deficient 2.2 Parachaenichthys georgianus Not main Non data-deficient 2.2 Paradiplospinus gracilis Not main Non data-deficient 2.2 Paranotothenia magellanica. Not main Non data-deficient 2.2 Patagonotothen brevicauda Not main Non data-deficient 2.2 Pleuragramma antarcticum Not main Non data-deficient 2.2 Protomyctophum tenisoni Not main Non data-deficient 2.2 Protomyctophum choriodon Not main Non data-deficient 2.2 Pseudochaenichthys georgianus Not main Non data-deficient 2.2 Pygoscelis Antarctica Not main Non data-deficient 2.2 Pygoscelis papua Not main Non data-deficient 2.2 Ragovitzia glacialis Not main Non data-deficient 2.2 Teuthoidea Not main Non data-deficient 2.2 Themisto gaudichaudii Not main Non data-deficient 2.2 Trematomus bernacchii Not main Non data-deficient 2.2 Trematomus lepidorhinus Not main Non data-deficient

4.4.4 Risk Based Framework The Risk Based Framework (RBF) is designed for use with the default assessment tree specifically with Principle 1 and was adopted by the MSC to enable scoring of fisheries in data-deficient situations.

There are sufficient data available to estimate stock status for Antarctic krill and the impact of the fishery on ecosystem components (retained species, discarded species and habitats). Therefore, the Antarctic krill fishery under assessment is not considered a data-deficient fishery and the use of the Risk Based Framework is not invoked in this assessment.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 62

5 TRACEABILITY

5.1 Eligibility Date

The Actual Eligibility Date (AED) is the same as the latest Target Eligibility Date (TED) – 14 May 2014. The rationale for setting the AED is the same as for TED, and described below.

The Target Eligibility Date for this fishery was set for 14 May 2014.

The target eligibility date (TED) is the expected date of eligibility for products from the fishery entering the Chain of Custody and thus being permitted to bear the MSC eco label. In order to allow the client to take advantage of the opportunity to set the TED up to a maximum 6 months prior to the publication of the Public Comment Draft Report, the TED was originally set at 1 February 2015. This coincides with the start of the fishing season for krill.

However, what has become evident now is that Olympic Seafood is producing end-products for the market based on krill already fished in the 2014 fishing season, i.e. during the period February- August 2014. This applies to the krill meal produced on board, which was landed by the fishing vessel in Montevideo during the three fishing trips made during those months and stored there until they could be shipped to the plants in the USA and New Zealand for processing in 2015.

Based on a variation granted by MSC, the TED has been changed from a maximum of 6 months to a maximum of 12 months prior to publication of the PCDR. Accordingly the catch from the third fishing trip in 2014, or about 1/3 pf the krill caught in 2014, will be eligible to be labelled as MSC- certified from the certification date.

Moving the TED 6 months back implies no consequences for traceability within the fishery. The krill product landed by the client in 2014 has been in the client´s custody all the way from landing up to production of the end-product in 2015. The shipment from Montevideo occurs in sealed and labelled containers, and are not opened before processing. The krill product has been packed in bags after processing and labelled on board, which by unloading on shore are transferred to spacekraft boxes and then containers for storing in separate freezing chambers at the custom warehouse. The boxes and containers are labelled and not mixed with other products.

Olympic has full traceability of the krill products throughout the entire value chain, which applies to 2014 as well as 2015.

There is reason to believe that the fishery in 2014 would have met all requirements for being certified according to the MSC standard in the same way as the fishery at the original target eligibility date of 1 February 2015. The team has carefully studied all the on-board independent observer reports back to 2010 (originally only reports back to 2012 were perused). Their conclusion is that there is nothing in these reports or any other evidence seen that would now lead to change in the scoring of P2 indicators. Chapter 3.4 (Ecosystem background) and subsequent scoring tables in this report includes evaluations of bycatch, ETP species and habitat based on the extra observer reports from 2010 and 2011 as well as the more-recent observer reports previously perused by the team.

Further, in terms of the indicators for P1 and P3, nothing new has surfaced now (in the opinion of the expert team) to suggest that any rescoring might be necessary. Indeed, and in terms of management (P3) and assessment of the stock (P1), the team notes that the sister krill fishery of Aker Biomarine, Norway, was deemed sustainable from well before the new target eligibility date for the Olympic Seafoods fishery of about 20 May 2014. The P1 and P3 sections and scorings of the two sister fisheries are indeed very close.

More details about rationale and justification is described fully in the Variation Request and the Variation Response relative to this fishery published on the MSC website.

UMAF (Under MSC Assessment Fishery)-products stored by the fishery was not sold as MSC certified products into the supply chain before the fishery got its certification.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 63

5.2 Traceability within the Fishery Traceability up to the point of first landing has been scrutinized as part of this assessment and the positive results reflect that the systems in place are deemed adequate to ensure fish are caught in a legal manner and are accurately recorded. The report and assessment trees describe these systems in more detail, but the management system supporting traceability has the following attributes:

 Electronic logbooks reporting to enforcement bodies (The Norwegian Directorate of Fisheries and CCAMLR) every day.  A VMS (Vessel Monitoring System) in place, ensuring that vessels are operating inside the UoC.  100% coverage by independent observers on board.  No possibilities for mixture of certified with non-certified catch; no transhipment, the vessel is only fishing krill, and only in the area of the Unit of Certification.  Labelling of catch with an identification key that is traceable all the way to the end user.

5.2.1 Description of the tracking, tracing and segregation systems within the fishery The Juvel operates only on krill in the Antarctic krill fishery in CCAMLR Area 48, including South Georgia. All catches are reported continuously during the fishing operations to the Norwegian authorities and CCAMLR. Norwegian-licensed vessels are obliged to report catches from each haul through their Electronic Reporting System (ERS). For client vessels this implies reporting at a 24-h interval.

The electronic positioning system (VMS) ensures that that the vessel is fishing only within the UoC areas. Vessel movements are under surveillance of the Norwegian authorities and CCAMLR.

Olympic Seafood also adheres fully to the principle of allowing non-national observers on its vessel at all times, which is required by Norwegian law for all Norwegian-licensed vessels. Therefore, there is 100% coverage by observers on board.

Only krill products are produced on board the vessel. The plant on board is specifically designed to produce krill product, and there is no possibility for mixing products.

Olympic is the one of only two krill oil producers and marketers that operates their own vessel in the Antarctic. The Olympic also control the next steps onshore: the extraction facilities in New Zealand and the USA (joint venture with Avoca, Inc.).

5.2.2 Risk of the possibility of vessels fishing outside the Unit of Certification There is no risk of vessels fishing for krill outside the UoC. The vessel is equipped with VMS, which produces information about the position that is reviewed by the Norwegian authorities and CCMLR.

5.2.3 Risk of substitution of certified fish with non-certified fish prior to and at the point of landing There are no possibilities for mixture of certified with non-certified catch. There is no transhipment, the vessel is only fishing krill, and the vessel operates only in the area of the UoC.

5.2.4 At-sea processing of catch

Immediately after each haul the catch is processed on board by enzymatic hydrolysis and separation steps. After hydrolysis the material passes to a decanter for separation of the shell fraction, pasteurization and inactivation of enzymes, mechanical separation of excess water with soluble proteins and finally drying of the wet krill sediment containing lipids, peptides and antioxidants to obtain the Antarctic krill meal. The excess water is evaporated to a concentrate containing about 50–60% dry matter.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 64

The process facility and its on-board quality and risk management system is registered and audited by the Norwegian Food Safety Authority. The system applies Hazard Analysis and Critical Control Point (HACCP) principles.

The products produced on-board the Juvel are:  Krill meal (krill phospholipids and peptides)  Protein concentrate from krill The concentrate is stored in 1000-litre containers and the Antarctic krill meal is packed in 25 kg aluminium foil bags. The containers and bags are labelled with vessel, catch/production date, composition of product, batch number, product identification, serial number, etc. Examples of labels are shown below:

Figure 23: Krill meal Label

Figure 24: Krill concentrate label

The extraction of the oil from the Antarctic krill meal is done on shore by RIMFROST USA in North Carolina, USA, and by Olympic Biotec Ltd in New Zealand. The solvent used is pure ethanol. The products are used as food directly and as ingredients in nutraceuticals, functional foods, pet food, aquafeed and other specialized formulations.

5.2.5 Transhipment

There is no transhipment of catch in this fishery.

5.2.6 Number and/or location of points of landing

The vessel lands the products mainly in Montevideo in Uruguay. However, sometimes the vessel goes to an other port located in Las Palmas de Gran Canaria in Spain to unload the products. The reason for doing this is occasionally need for combination of unloading with repair and/or docking at the shipyard, which is difficult in Montevideo.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 65

5.2.7 Robustness of the management systems related to traceability

The system of tracking and tracing in the Olympic Seafood Antarctic krill fishery is sufficient to ensure that all the krill originating from the certified fishery, and sold as certified, could be identified prior or at the point of landing.

The team concludes that the systems in place are comprehensive and mandatory, and that enforcement provides a robust system that supports full traceability for the landed product.

Catch reports to the Norwegian enforcement bodies after each haul, VMS and 100% observer coverage ensure a high level of robustness of the management system relating to traceability; see Section 5.2.1 above.

5.3 Eligibility to Enter Further Chains of Custody Olympic are starting up a process of obtaining CoC for Olympic Seafood AS that buys the products at landing for direct sale and shipment to RIMFROST NZ or RIMFROST USA, or for temporarily storage in locked customs warehouse in Montevideo without further treatment. Also RIMFROST NZ or RIMFROST USA are in the process for obtaining CoC certification. All the goods are in the custody of Olympic Seafood AS after landing before further sale to RIMFROST NZ or RIMFROST USA. Everything is packed and clearly labelled and there is no risk for mixture with other products. Only krill caught in the manner defined in the Unit of Certification (Section 3.1) under restrictions detailed throughout the body of the final Public Certification Report for this fishery shall be eligible to enter the Chain of Custody. Chain of Custody should commence following the first point of landing, at which point the product shall be eligible to carry the MSC logo (under restrictions imposed by the MSC Chain of Custody standard). There are no restrictions on the fully certified product entering further chains of custody.

5.4 Eligibility of Inseparable or Practically Inseparable (IPI) stock(s) to Enter Further Chains of Custody

There are no IPI stocks in this fishery.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 66

6 EVALUATION RESULTS

6.1 Principle Level Scores Table 9. Final Principle Scores

MSC Principle Score

Principle 1: Sustainability of Exploited Stock 90.0

Principle 2: Maintenance of Ecosystem 96.7

Principle 3: Effective Management System 93.1

6.2 Summary of Scores Table 10. Summary of scores Principle Wt Component Wt PI Performance Indicator (PI) (L1) (L2) No. Score One 1 Outcome 0.5 1.1.1 Stock status 100 Reference points 1.1.2 90 1.1.3 Stock rebuilding n/a Management 0.5 1.2.1 Harvest strategy 95 1.2.2 Harvest control rules & tools 80 1.2.3 Information & monitoring 80 1.2.4 Assessment of stock status 85 Two 1 Retained 0.2 2.1.1 Outcome 100 species 2.1.2 Management 100 2.1.3 Information 100 Bycatch 0.2 2.2.1 Outcome 100 species 2.2.2 Management 100 2.2.3 Information 100 ETP species 0.2 2.3.1 Outcome 100

2.3.2 Management 100 2.3.3 Information 95

Habitats 0.2 2.4.1 Outcome 100

2.4.2 Management 100 2.4.3 Information 85 Ecosystem 0.2 2.5.1 Outcome 100 2.5.2 Management 80 2.5.3 Information 90 Three 1 Governance 3.1.1 Legal & customary framework 90 and policy 0.5 Consultation, roles & responsibilities 3.1.2 95 Long term objectives 3.1.3 100 3.1.4 Incentives for sustainable fishing 80 Fishery specific 0.5 3.2.1 Fishery specific objectives 90 management 3.2.2 Decision making processes system 95 3.2.3 Compliance & enforcement 100 3.2.4 Research plan 100 3.2.5 Management performance evaluation 90

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 67

6.3 Summary of Conditions There are no conditions in this fishery.

6.3.1 Recommendations There are no recommendations in this fishery.

6.4 Determination, Formal Conclusion and Agreement The Olympic Seafood Antarctic krill fishery achieved a score of 80 or more for each of the three MSC Principles, and did not score under 60 for any of the set MSC Criteria. The assessment team therefore recommended the certification of the Olympic Seafood Antarctic krill fishery for the client Olympic Seafood. There were no conditions as non of the performance indicators was given a score lower than 80.

Following this decision by the assessment team, and review by peer-reviewers and stakeholders, the determination was presented to DNV GL Business Assurance decision making entity that the fishery has passed its assessment and should be certified. The Technical reviewer at DNV adheres to the recommendation of the assessment team and approves the certification of the Olympic Seafood Antarctic krill fishery for the client Olympic Seafood without any conditions.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 68

REFERENCES » Ainley, D. G., Ballard, G., and Weller, J. 2010. CCAMLR WG-EMM-10/11. Ross Sea biodiversity. 1. Validation of the 2007 CCAMLR Bio-regionalization Workshop results towards including the Ross Sea in a representative network of marine protected areas in the Southern Ocean. http://www.ccamlr.org/en/wg-emm-10/11 » AKER BioMarine notification to CCAMLR to enter the fishery » Annex V to the Protocol on Environmental Protection to the Antarctic Treaty Area protection and management. http://www.ats.aq/documents/recatt/att004_e.pdf

» Antarctic Treaty Secretariat website. http://www.ats.aq/e/info.htm

» Arndt, J. E., Schenke, H. W., Jakobsson, M., Nitsche, F., Buys, G., Goleby, B., Rebesco, M., Bohoyo, F., Hong, J. K., Black, J., Greku, R., Udintsev, G., Barrios, F., Reynoso-Peralta, W., Morishita, T., and Wigley, R. 2013. The International Bathymetric Chart of the Southern Ocean (IBCSO) Version 1.0. A new bathymetric compilation covering circum-Antarctic waters Geophysical Research Letters, 40: 3111–3117. doi: 10.1002/grl.50413 http://hs.pangaea.de/Maps/bathy/IBCSO_v1/IBCSO_v1_digital_chart_pdfA.pdf » ASOC (The Antarctic and Southern Coalition). The need for a strategic plan for the management of the Antarctic krill fishery. Paper prepared for the 14th Meeting of CCAMLR, October/November 2007. » Atkinson, A., Nicol, S., Kawaguchi, S., Pakhomov, E., Quetin, L., Ross, R., Hill, S., Reiss, C., and Siegel, V. 2012. Fitting Euphausia superba into Southern Ocean food-web models: a review of data sources and their limitations. CCAMLR Science 19: 219–245. » Atkinson, A., Siegel, V., Pakhomov, E. A., and Rothery, P. 2004. Long-term decline in krill stock and increase in salps within the Southern Ocean. Nature 432: 100–103. » Atkinson, A., Siegel, V., Pakhomov, E. A., Rothery, P., Loeb, V., Ross, R. M., Quetin, L. B., Schmidt, K., Fretwell, P., Murphy, E. J., Tarling, G. A., and Fleming, A. H. 2008. Oceanic circumpolar habitats of Antarctic krill. Marine Ecology Progress Series 362: 1–23. » Bednarsek, N., Tarling, G. A., Bakker, D. C. E., Fielding, S., Jones, E. M., Venables, H. J., Ward, P., Kuzirian, A., Lézé, B., Feely, R. A. and Murphy, E. J. 2012. Extensive dissolution of live pteropods in the Southern Ocean. Nature Geoscience 5: 881–885. doi:10.1038/ngeo1635. » Brierley, A. S., Demer, D. A., Watkins, J. L., and Hewitt, R. P. 1999. Concordance of interannual fluctuations in acoustically estimated densities of Antarctic krill around South Georgia and Elephant Island: biological evidence of same-year teleconnections across the Scotia Sea. Marine Biology 134: 675–681. » Brierley, A. S., and Reid, K. 1999. Kingdom of the krill: the boom and bust cycle of a tiny crustacean holds the key to the health of the Southern Ocean. New Scientist 162: 36-41. » Brinton, E. 1991 Distribution and population structures of inmature and adult Euphausia superba in the western Bransfield Strait region during the 1986-87 summer. Deep-Sea Research Part A Oceanographic Research Papers 38: 1169–1193. (doi:10.1016/0198- 0149(91)90101-K) » www.ccamlr.org » CCAMLR Conservation Measure 21-03 (2013). Notifications of intent to participate in a fishery for Euphausia superba. http://www.ccamlr.org/en/measure-21-03-2013 » CCAMLR Conservation Measure 25-03 (2011) on the minimization of incidental mortalities of seabirds and marine mammals in the course of trawl fishing in the Convention Area. http://www.ccamlr.org/en/measure-25-03-2011 » CCAMLR Conservation Measure 26-01 (2009) on general environmental protection during fishing. http://www.ccamlr.org/en/measure-26-01-2009 » CCAMLR Conservation Measure 51-01 (2010) regarding Precautionary catch limitations on Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51-01.pdf

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 69

» CCAMLR Conservation Measure 51-06 (2012) regarding the General measure for scientific observation in fisheries for Euphausia superba http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51-06_3.pdf » CCAMLR Conservation Measure 51-07 (2011). Interim distribution of the trigger level in the fishery for Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51-07.pdf » CCAMLR Conservation Measure 91-03 (2009) on the Protection of the South Orkney Islands southern shelf. http://www.ccamlr.org/en/measure-91-03-2009 » CCAMLR Conservation Measure 91-04 (2011) on the General framework for the establishment of CCAMLR Marine Protected Areas. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//91-04.pdf » CCAMLR Annual Fishery Reports » CCAMLR Annual Observer reports for years 2010–2014. » http://www.ccamlr.org/en/fisheries/fishery-monitoring » http://www.ccamlr.org/en/publications/science_journal/ccamlr-science-volume-20 » CCAMLR Scientific Observer Manual. http://www.ccamlr.org/en/system/files/obsman.pdf » CCAMLR Scientific Observer Cruise Reports for 2012, 2013 and 2014. » CCAMLR-IWC. 2010. Joint Workshop to Review Input Data for Antarctic Marine Ecosystem Models report. Journal of Cetacean Research and Management 11 (Suppl. 2). http://iwc.int/cache/downloads/i5flpo5e6coog0c04g40scg0/SC-61-Rep2-JCRM11(2).pdf » CCAMLR WG-EMM 2003 report CEMP Review (pp. 143–148). http://www.ccamlr.org/es/system/files/s-sc-xxii-a4.pdf » CCAMLR WG-EMM 2007 Report. http://www.ccamlr.org/es/system/files/s-sc-xxvi-a4.pdf » CCAMLR WG-EMM 2014 Report. https://www.ccamlr.org/en/wg-emm-14

» CCAMLR XXXIII Meeting Commission Report (2014). https://www.ccamlr.org/en/ccamlr- xxxiii

» CCAMLR Secretariat. Krill fishery report: 2013 update » CCAMLR Secretariat. Krill fishery report: 2013 update. WG-EMM-13/37 Rev. 1 http://www.ccamlr.org/en/wg-emm-13/37-rev-1 » SC-CCAMLR-XXXII, climate change section (pp. 63–64). http://www.ccamlr.org/en/system/files/e-sc-32-prelim-dec.pdf » http://www.ccamlr.org/en/science/ccamlr-ecosystem-monitoring-program-cemp » http://www.ccamlr.org/en/science/working-group-ecosystem-monitoring-and- management-wg-emm » Christensen, L. B. 2006. Marine mammal populations; reconstructing historical abundances at the global scale. Fisheries Centre Research Report 14:1–161. » Convention on International Trade in Endangered Species of Wild Fauna and Flora list. http://www.cites.org/eng/disc/species.php » Constable, A. J., and de la Mare, W. K. 1996. A generalised model for evaluating yield and the long term status of fish stocks under conditions of uncertainty. CCAMLR Science 3: 31– 54. » Constable A. J., de la Mare, W. K., Agnew, D. J., Everson, I., and Miller, D. 2000. Managing fisheries to conserve the Antarctic marine ecosystem: practical implementation of the Convention on the Conservation of Antarctic Marine Living Resources. ICES Journal of Marine Science 57: 778–791. (doi:10.1006/jmsc.2000.0725) » Constable, A. J., Nicol, S., and Strutton, P. G. 2003. Southern Ocean productivity in relation to spatial and temporal variation in the physical environment, Journal of

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 70

Geophysical Research 108: 8079, doi:10.1029/2001JC001270, C4. http://onlinelibrary.wiley.com/doi/10.1029/2001JC001270/abstract » Cook, A. J., Fox, A. J., Vaughan, D. G., and Ferrigno, J. G. 2005. Retreating glacier fronts on the Antarctic Peninsula over the past half-century. Science 308: 541–544. » Cook, A. J., and Vaughan, D. G. 2010. Overview of areal changes of the ice shelves on the Antarctic Peninsula over the past 50 years. Cryosphere 4: 77–98. » Cornejo-Donoso, J., and Antezana, T. 2008. Preliminary trophic model of the Antarctic Peninsula Ecosystem (Sub-area CCAMLR 48.1) (EwE model) Ecological Modelling 218: 1– 17. http://www.ecopath.org/node/195 » Demer, D. A., Cossio, A. M., and Reiss, C. S. 2007. CCAMLR 2000 revisited. WG-EMM-07- 30. CCAMLR, Hobart. » de la Mare, W. K. 1994a. Estimating krill recruitment and its variability. CCAMLR Science 1: 55–69. » de la Mare, W. K. 1994b. Modelling krill recruitment. CCAMLR Science 1: 49–54. » Douglass, L., Beaver, D., Turner, J., and Nicoll, R. 2011. An identification of areas within the high seas of the Southern Ocean that would contribute to a representative system of marine protected areas. http://conservationgeography.org/sites/default/files/Douglass%20et%20al_CCAMLR_WS_ MPA_2011_withMaps.pdf » http://www.eoearth.org/ » http://www.eoearth.org/view/article/156142 » http://extrememarine.org.uk/extremeantarctica/2014/11/16/a-keystone-species/ » Everson, I. (ed.) 2000. Krill: Biology, Ecology and Fisheries. Blackwell Scientific, Oxford, Fish and Aquatic Resources Series 6. x + 372 pp. » Fraser, W. R., and Hofmann, E. E. 2003. A predator’s perspective on causal links between climate change, physical forcing and ecosystem response. Marine Ecology Progress Series 265: 1–15. » Friend of the Sea Certification Report of Juvel’s activity. » Fulton, E. A., Smith, A. D. M., and Johnson C. R. 2003. Effect of complexity on marine ecosystem models. Marine Ecology Progress Series 253: 1–16. » GEBCO General Bathymetric Chart of the Oceans. http://www.gebco.net/about_us/news_and_events/arctic_antarctic_mapping_meeting.htm l » Government of South Georgia and the South Sandwich Islands. 2013. The South Georgia and South Sandwich Islands Marine Protected Area management plan. http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0.pdf » Grobe, H. Alfred Wegener Institute for Polar and Marine Research. Creative Commons CC- BY-SA-2.5. » Guille, S. T. 2002. Warming of the Southern Ocean since the 1950s. Science 295: 1275– 1277. » Gutt, J., Hosie, G., and Stoddart, M. 2010. Marine life in the Antarctic. In McIntyre, A. D. (ed.). Life in the World’s Oceans: Diversity, Distribution, and Abundance. Wiley-Blackwell, Oxford, UK, doi: 10.1002/9781444325508.ch11. » Gutt, J. et al. 2013. "Shifts in Antarctic megabenthic structure after ice-shelf disintegration in the Larsen area east of the Antarctic Peninsula." Polar Biology 36(6): 895-906. » Hewitt, R. P. and Low, E. H. L. 2000. The fishery of Antarctic krill: defining an ecosystem approach to management. Reviews in Fisheries Science 8: 235–298. » Hewitt, R. P., Watkins, J. L., Naganobu, M., Tshernyshkov, P., Brierley, A. S., Demer, D. A., Kasatkina, S., Takao, Y., Goss, C., Malyshko, A., Brandon, M. A., Kawaguchi, S., Siegel, V.,

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 71

Trathan, P. N., Emery, J. H., Everson, I., and Miller, D. G. M. 2002. Setting a precautionary limit for Antarctic krill. Oceanography 15: 26-33. » Hewitt, R. P., Watters, G., Trathan, P. N., Croxall, J. P., Goebel, M. E., Ramm, D., Reid, K., Trivelpiece, W. Z., and Watkins, J. L. 2004. Options for allocating the precautionary catch limit of krill among small scale management units in the Scotia Sea. CCAMLR Science 11: 81–97. http://www.ccamlr.org/en/system/files/science_journal_papers/05hewitt-etal.pdf » Heywood, B. G., Brierley, A. S., and Gull, S. F. 2006. A quantified Bayesian maximum entropy estimate of Antarctic krill abundance across the Scotia Sea and in small-scale management units from the CCAMLR-2000 survey. CCAMLR Science 13: 97–116. » Hill, S. L., and Mathews, J. 2013.The sensitivity of multiple output statistics to input parameters in a krill-predator-fishery ecosystem dynamics model. CCAMLR Science 20: 97–118. » Hill, S. L., Watters, G. M., Punt, A. E., McAllister, M. K., LeQuere, C., and Turner, J. 2007. Model uncertainty in the ecosystem approach to fisheries. Fish and Fisheries 8:315–333. » Hoenig, J. 1984. Empirical use of longevity data to estimate mortality rates. Fishery Bulletin US 81: 898–903. doi:10.1016/j.dsr2.2003.07.012 » Hofmann, E. E and Hùsrevõglu, Y. S. 2003. A circumpolar modeling study of habitat control of Antarctic krill (Euphausia superba) reproductive success. Deep Sea Research Part II Topical Studies in Oceanography 50: 3121–3142. » Hofmann, E. E., Klinck, J. M., Locarnini, R. A., Fach, B., and Murphy, E. J. 1998. Krill transport in the Scotia Sea and environs. Antarctic Science 10: 406–415. » Hønneland, G. 2012. Making Fishery Agreements Work, Edward Elgar, Cheltenham. » Hoover, C., Pitcher, T., and Pakhomov, E. 2012. The Antarctic Peninsula Marine Ecosystem model and simulations: 1978 – Present. Fisheries Centre Research Reports 20. » Howard, S. L., Hyatt, J., and Padman, L. 2004. Mixing in the pycnocline over the western Antarctic Peninsula shelf during Southern Ocean GLOBEC. Deep Sea Research II 51: 1965–1979. » IBCSO International Bathymetric Chart of the Southern Ocean. http://www.ibcso.org/ » http://www.ibcso.org/documents/IBCSO_AGU2012_ePoster.pdf » IUCN. www.iucn.org » IUCN. 2012. Antarctic Treaty Consultative Meeting XXXV. Hobart, 2012. (IP34) Using ASMAs and ASPAs when necessary to complement CCAMLR MPAs http://www.ats.aq/index_e.htm » Johnston, P., Santillo, D., Page, R., and Dorey, C. 2009. Gambling with krill fisheries in the Antarctic: large uncertainties equate with high risks. Greenpeace Research Laboratories Technical Note 01/2009. 12 pp. » Jolly, G. M., and Hampton, I. 1990. A stratified random transect design for acoustic surveys of fish stocks. Canadian Journal of Fisheries and Aquatic Sciences 47: 1282–1291. » Juvel’s protocol on interactions with birds and marine mammals. » Juvel’s 2013 Vessel Monitoring System track records. » Kawaguchi, S., Nicol, S., Taki, K. and Naganobu, M. 2006. Fishing ground selection in the Antarctic krill fishery: trends in patterns across years, season and nations. CCAMLR Science 13: 117–141. » King, J. C. 1994. Recent climate variability in the vicinity of the Antarctic Peninsula. International Journal of Climatology 14: 357–369. (doi:10.1002/joc.3370140402) » Kinzey, D., Watters, G., and Reiss, C. S. 2013. Effects of recruitment variability and natural mortality on generalised yield model predictions and the CCAMLR decision rules for Antarctic krill. CCAMLR Science 20: 81–96.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 72

» Kokubun, N., A. Takahashi, Y. Mori, S. Watanabe, and H.-C. Shin. 2010. Comparison of diving behaviour and foraging habitat use between Chinstrap and Gentoo Penguins breeding in the South Shetland Islands, Antarctica. Marine Biology 157:811–825. » Krafft, B. A., Skaret, G., Krag, L. A., Trathan, P., and Ying, Y. 2013. Studies of Antarctic krill, krill predators and trawl gear at South Orkney Islands, 2013. Institute of Marine Research Report 8. http://www.imr.no/filarkiv/2013/04/hi_imr-report_no_8- 2013_antarctic_krill.pdf/en » Krag, L. A., Herrmann, B., Iversen, S., Engås, A., Nordrum, S., and Krafft, B. A. 2013. A method to evaluate selection of Antarctic krill in towed fishing gears. WG-EMM-13/34. CCAMLR, Hobart. » Leape, G., Gascon, V., Werner, R., Pearl, A., and Fischer, M. 2009. Analysis on the eligibility of Aker bioMarine’s krill fishery operations for MSC certification rationale in regards to a number of the performance indicators that are not achieving the minimum SG 60 necessary for MSC certification. Antarctic Krill Conservation Project (AKCP), The Pew Environmental Group, Washington, DC. Unpublished document. » Loeb, V., Siegel, V., Holm-Hansen, O., Hewitt, R., Fraser, W., Trivelpiece, W., and Trivelpiece, S. 1997. Effects of sea-ice extent and krill or salp dominance on the Antarctic food web. Nature 387: 897–900. » Lynch, H. J., Naveen, R., Trathan, P., and Fagan, W. F. 2012. Spatially integrated assessment reveals widespread changes in penguin populations on the Antarctic Peninsula. Ecology 936: 1367–1377 http://www.esajournals.org/doi/pdf/10.1890/11-1588.1 » Mangel, M., and Switzer, P. V. 1998. A model at the level of the foraging trop for the indirect effects of krill (Euphausia superba) fisheries on krill predators. Ecological Modelling 105: 235–256. » McWhinnie, M. A., Denys, C., and Schenborn, D. 1976. Biology of krill (Euphausia superba) and other Antarctic invertebrates. US Antarctic Journal 112: 55–58. » MRAG. 2012. Analysis of larval bycatch on the Saga Sea during continuous trawling for krill in CCAMLR Areas 48 between December 2007 and September 2011. MRAG, London. 36 pp. » Marín, V. H., and Delgado, L.E. 2001. A spatially explicit model of the Antarctic krill fishery off the south Shetland Islands. Ecological Applications 11: 1235–1248. » May, R. M. 1979. Ecological interactions in the Southern Ocean. Nature 277: 86–89. (doi:10.1038/277086a0) » Meredith, M. P., Murphy, E. J., Hawker, E. J., King, J. C., and Wallace, M. I. 2008. On the interannual variability of ocean temperatures around South Georgia, Southern Ocean: forcing by El Niño/Southern Oscillation and the Southern Annular Mode. Deep-Sea Research II 55: 2007– 2022. » Miller, D. 2003. Krill Species Profile. Unpublished CCAMLR document, May 2003. CCAMLR, Hobart. » Miller, D. G. M., and Hampton, I. 1989. Biology and ecology of the Antarctic krill. A review. Biomass 9. SCAR and SCOR, Scott Polar Research Institute, Cambridge, UK. 166 pp. » Moline, M. A., Claustre, H., Frazer, T. K., Grzymski, J., Schofield, O., and Vernet, M. 2000. Changes in phytoplankton assemblages along the Antarctic Peninsula and potential food web implications for the Antarctic food web, pp. 263–271. In Davidson, W., Howard- Williams, C., and Broady, P. (eds). Antarctic Ecosystems: Models for Wider Ecological Understanding. Christchurch, New Zealand, Caxton Press. » Murphy, E. J. 1995. Spatial structure of the Southern Ocean ecosystem predator–prey linkages in the Southern Ocean food-webs. Journal of Animal Ecology 64: 333–347 (doi:10.2307/5895). » Murphy, E. J., Thorpe, S. E. Watkins, J. L., and Hewitt, R. 2004. Modeling the krill transport pathways in the Scotia Sea: spatial and environmental connections generating the seasonal distribution of krill. Deep Sea Research Part II. Topical Studies in Oceanography 51: 1435–1456 (doi:10.1016/j.dsr2.2004.06.019).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 73

» Murphy, E. J., Watkins, J. L., Reid, K., Trathan, P. N., Iverson, I., Croxall, J. P., Priddle, J., Brandon, M. A., Brierley, A. S., and Hofmann, E. 1998. Interannual variability of the South Georgia marine ecosystem: biological and physical sources of variation in the abundance of krill. Fisheries Oceanography 7: 381–390 (doi:10.1046/j.1365-2419.1998.00081.x). » Murphy, E. J., Watkins, J. L., Trathan, P. N., Reid, K.,, Meredith, M. P., Thorpe, S. E., Johnston, N. M., Clarke, A., Tarling, G. A., Collins, M. A., Forcada, J., Shreeve, R. S., Atkinson, A., Korb, R., Whitehouse, M. J., Ward, P., Rodhouse, P. G., Enderlein, P., Hirst, A. G., Martin, A. R., Hill, S. L., Staniland, I. J., Ponmd, D. W., Briggs, D. R., Cunningham, N. J., and Fleming, A. H. 2007. Spatial and temporal operation of the Scotia Sea ecosystem: a review of large-scale links in a krill centred food web. Philosophical Transactions of the Royal Society of London B. Biological Sciences 362. Doi: 10.1098/rstb.2006.1957. » Naganobu, M., Kitamara, T., and Hasunuma, K. 2008. Relationship between distribution of Antarctic krill (Euphausia superba) and environmental index MTEM-200 in the Antarctic Ocean throughout the year. WG-EMM-08/32. CCAMLR, Hobart. » Nicol, S. 2006. Krill, currents, and sea ice: Euphausia superba and its changing environment. BioScience 56: 111–120. » Nicol, S., and Foster, J. 2003. Recent trends in the fishery for Antarctic krill. Aquatic Living Resources 16: 42–45. » Nicol, S., Foster, J., and Kawaguchi, S. 2011. The fishery for Antarctic krill — recent developments. Fish and Fisheries, 13: 30–40. » Nicol, S., Worby, A., and Leaper, R. 2008. Changes in the Antarctic sea ice ecosystem: potential effects on krill and baleen whales. Marine and Freshwater Research 59: 361–382. » Nicoll, R., and Douglass, L. 2012. Project report: Mapping krill trawling and predator distribution. » http://www.nmfs.noaa.gov/pr/species/mammals/cetaceans/minkewhale.htm » Norwegian Directorate of Fisheries. 2013. Melding fra fiskeridirektøren J-236-2013 (Fisheries Directorate Regulation J-236-2013) » Norwegian Ministry of Fisheries. 2008. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act). » Norwegian Ministry of Fisheries. 2012. Fiskeriavtalane Noreg har inngått med andre land for 2013 og fisket etter avtalane i 2011 og 2012 (The Fishery Agreements Norway had Concluded with Other Countries for 2013 and Fishery according to the Agreements in 2011 and 2012), Meld. St. 40 (2012‐ 2013) (White Paper No. 40 2012‐2013). » Palomares, M. L. D., Pruvost, P., Pitcher T. J., and Pauly, D. 2005. Modeling Antarctic Marine Ecosystems. Fisheries Centre Research Reports 13 (7). http://www.seaaroundus.org/researcher/dpauly/PDF/2005/Books&Chapters/ModelingAntar cticMarineEcosystems.pdf » Parkinson, C. L., and. Cavalieri, D. J. 2012. Antarctic sea ice variability and trends, 1979– 2010. Cryosphere 6: 871–880. http://www.the-cryosphere.net/6/871/2012/tc-6-871- 2012.pdf » Peatman, T., Clarke, J. M., and Agnew, D. J. 2011. Estimation of management reference points consistent with the catch trigger level for the Antarctic krill fishery in Area 48. WG- EMM-11-17. CCAMLR, Hobart. » Plagányi, E. E. 2007. Models for an ecosystem approach to fisheries. FAO Fisheries Technical Paper 477. 108 pp. » Plagányi, E .E., and. Butterworth, D. S. 2012. The Scotia Sea krill fishery and its possible impacts on dependent predators: modeling localized depletion of prey. Ecological Applications 22: 748–761. http://www.esajournals.org/doi/abs/10.1890/11-0441.1 » Raick, C., Soetaert, K., and Grégoire, M. 2006. Model complexity and performance: How far can we simplify? Progress in Oceangraphy 70: 27–57.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 74

» Reid, K., and Croxall, J. C. 2001. Environmental responses of upper trophic-level predators reveals a system change in an Antarctic marine ecosystem. Proceedings of the Royal Society of London 268: 377–384. » Rignot, E., and Jacobs, S. S. 2002. Rapid bottom melting widespread near Antarctic ice sheet grounding lines. Science 296: 2020–2023. » Rintoul, S. R., Hughes, C., and Olbers, D. 2001. Ocean circulation and climate. Chapter 4.6 in The Antarctic Circumpolar Current System. Academic Press, London. ISBN 0-12- 641351-7 » Rintoul, S. R., Hughes, C., and Olbers, D. 2001. Ocean circulation and climate. Chapter 4.6 in The Antarctic Circumpolar Current System. Academic Press, London. ISBN 0-12- 641351-7 » SC-CAMLR. 1993. Report of the Twelfth Meeting of the Scientific Committee (SC-CAMLR- XII). CCAMLR, Hobart. 431 pp. » "Scotia-sea" by Gi - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons http://commons.wikimedia.org/wiki/File:Scotia-sea.png#mediaviewer/File:Scotia-sea.png » Siegel, V. 2000. Krill (Euphausiacea) demography and variability in abundance and distribution. Canadian Journal of Fisheries and Aquatic Sciences 57: 151–167. » Siegel, V., Bergstrom, B., Muhlenhardt-Siegel, U., and Thomasson, M. 2002. Demography of krill in the Elephant Island during summer 2001 and its significance for stock recruitment. Antarctic Science 14: 162–170. (doi:10.1017/S095410200200072X) » Smith, A. M. D., Brown, C. J., Bulman, C. M., Fulton, E. A., Jonhson, P., Kaplan, I. C., Lozano-Montes, H., Mackinson, S., Marzloff, M., Shannon, L. J., Shin, Y. J., and Tam, J. 2011. Impacts of fishing low–trophic level species on marine ecosystems. Science 33: 1147–1150. http://www.sciencemag.org/content/333/6046/1147 » Spiridonov, V. A. 1995. Spatial and temporal variability in reproductive timing of Antarctic krill (Euphausia superba Dana). Polar Biology 15: 161–174. (doi:10.1007/BF00239056) » Stammerjohn, S. E., Martinson, D. G., Smith, R. C., Yuan, X., and Rind, D. 2008. Trends in Antarctic annual sea ice retreat and advance and their relation to ENSO and Southern Annular Mode Variability. Journal of Geophysical Research 113 (C3): C03S90. » Sushin, V. A., and Shulgovsky, K. E. 1999. Krill distribution in the western Atlantic sector of the Southern Ocean during 1983/84 and 1987/88 based on the results of Soviet mesoscale surveys conducted using an Isaacs–Kidd midwater trawl. CCAMLR Science 6: 59–70. » http://theseasproject.weebly.com/scotia-sea.html » Thorpe, S. E., Heywood, K., Brandon, M. A., and Stevens, D. P. 2002. Variability of the southern Antarctic Circumpolar Current front north of South Georgia. Journal of Marine Systems 37: 87–105 https://ueaeprints.uea.ac.uk/15924/1/DS_30.pdf » Thorpe, S. E., Heywood, K. J., Stevens, D. P., and Brandon, M. A. 2004. Tracking passive drifters in a high resolution ocean model: implications for interannual variability of larval krill transport to South Georgia. Deep-Sea Research Part I Oceanographic Research Papers 51: 909–920. (doi:10.1016C/j.dsr.2004.02.008) » Toropova, C., Meliane, I., Laffoley, D., Matthews, E., and Spalding, M. 2010. Global Ocean Protection Present Status and Future Possibilities http://data.iucn.org/dbtw- wpd/edocs/2010-053.pdf » Trathan, P.N., Brierley, A.S., Brandon, M.A., Bone, D.G., Goss, C., Grant, S.A., Murphy, E.J., Watkins, J.L. (2003) Oceanographic variability and changes in Antarctic krill (Euphausia superba) abundance at South Georgia. Fish. Oceanog.,12: 569-583. » Trathan, P. N., Forcada, F., and Murphy, E. J. 2007. Environmental forcing and Southern Ocean marine predator populations: effects of climate change and variability. Philosphical Transactions of the Royal Society of London, B. Biological Sciences 362: 2351–2365. http://rstb.royalsocietypublishing.org/content/362/1488/2351.full.pdf+html?sid=a0b97f88 -bd01-4184-b2cb-afd674863298

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 75

» Trathan P. N., Fretwell P. T., and Stonehouse, B. 2011. First recorded loss of an emperor penguin colony in the recent period of Antarctic regional warming: implications for other colonies. PLoS ONE 6: e14738. » Trathan, P. N., and Grant, S. M. 2013. Precautionary spatial protection to facilitate the scientific study of habitats and communities under ice shelves in the context of recent, rapid, regional climate change. CCAMLR Science: 20: 139–151. http://www.ccamlr.org/en/system/files/science_journal_papers/Trathan%20et%20al_0.pd f » Trathan, P. N., Ratcliff, N., and Masden, E. A. 2012. Ecological drivers of change at South Georgia: the krill surplus, or climate variability. Ecography 35: 983–993. » Trathan, P. N., and Reid, K. 2009. Exploitation of the marine ecosystem in the Sub- Antarctic: historical impact and current consequences. Papers and Proceedings of the Royal Society of Tasmania 143: 9–14. » Watters, G. M., Hill, S., Hinke, J. T., and Trathan, P. 2009. The risks of not deciding to allocate the precautionary krill catch limit among SSMUs and allowing uncontrolled expansion of the krill fishery up to the trigger level. WG-EMM-09/12. CCAMLR, Hobart. » Watters, G. M., Hill, S., Hinke, J. T., Matthews J. and Reid, K. 2013. Decision making for ecosystem based management: evaluating options for a krill fishery with an ecosystem dynamics model. Ecological Applications 23: 710–725. http://www.ncbi.nlm.nih.gov/pubmed/23865224 » WG-EMM-08/46. 2008. Catch uncertainty in krill fisheries. CCAMLR Secretariat, Hobart. » WG-EMM-13/37 Rev. 1. 2013. Krill fishery report: 2013 update. CCAMLR Secretariat, Hobart. 9 pp. » WG-EMM-14/58 2014. Draft krill fishery report. CCAMLR Secretariat, Hobart. 54 pp. » http://www.worldwildlife.org/ecoregions/an1101

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 76

APPENDIX 1 SCORING AND RATIONALES

Appendix 1.a - MSC Principles and Criteria

Figure A1: MSC Principles and Criteria

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 77

Appendix 1.1 Performance Indicator Scores and Rationale Evaluation Table for PI 1.1.1

The stock is at a level which maintains high productivity and has a low PI 1.1.1 probability of recruitment overfishing Scoring Issue SG 60 SG 80 SG 100 A It is likely that the stock It is highly likely that the There is a high degree of certainty is above the point where stock is above the point that the stock is above the point recruitment would be where recruitment would where recruitment would be impaired. be impaired. impaired. Guidepost Guidepost Met? Yes Yes Yes Annual krill catches are increasing slowly year on year as more vessels start to operate within the fishery and the markets expand, but they are still well below the trigger levels set by management (overall and in most years too by subarea). That overall trigger level itself is set at just 11% of the estimated catch limit, and the catch limit itself is estimated using a precautionary recruitment criterion that renders a high degree of certainty that

Justification Justification recruitment of krill is not being impaired. B The stock is at or There is a high degree of certainty fluctuating around its that the stock has been fluctuating target reference point. around its target reference point, or has been above its target

Guidepost Guidepost reference point, over recent years. Met? Yes Yes

Annual catches were much higher 20–30 years ago, but at present are well below levels that might be thought to impact target and limit reference points, although the form of the assessment is such that these points are not currently evaluated directly. It is conceded that there are many uncertainties for management to consider and that there is no annual assessment of the stock. However, rules of management and controls on the catch are tight, and the reporting systems in place are sufficiently robust to detect any possible

Justification Justification concerns long before they might become serious.

References WG‐EMM 2011, 2012, 2013, 2014, various reports, Constable and de la Mare (1996), WG‐ EMM‐13/37 Rev. 1 (2013), WG‐EMM‐14/58 (2014; seen only in draft)

Stock Status relative to Reference Points

Type of reference Value of reference Current stock status relative point point to reference point

Target Percentage of B0 after 75% Biomass is calculated by various reference 20 years methods to be in the range 37– point 208 million tonnes, well above 75%B0 Limit Probability of biomass 20% The evaluated range of biomass is reference being less than a set stated above, and there is only an point percentage of B0 after extremely low probability of the 20 years stock being at that percentage of B0 OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 78

Evaluation Table for PI 1.1.2

PI 1.1.2 Limit and target reference points are appropriate for the stock

Scoring Issue SG 60 SG 80 SG 100 A Generic limit and target Reference points are reference points are appropriate for the stock based on justifiable and and can be estimated. reasonable practice appropriate for the

Guidepost Guidepost species category. Met? Yes Yes Target and limit reference points are defined as part of the harvest control rule, not determined directly. The fishery is managed by CCAMLR through setting a catch trigger level that is at present estimated to be just 11% of the precautionary catch limit (PCL) determined from the GYM, therefore acting as an effective risk‐limit reference point. F and B values consistent with the trigger level in the fishery are well below (F) and above (B) the

Justification Justification target and limit reference points. B The limit reference point The limit reference point is set is set above the level at above the level at which there is which there is an an appreciable risk of impairing appreciable risk of reproductive capacity following impairing reproductive consideration of precautionary

Guidepost Guidepost capacity. issues. Met? Yes Yes

The determination of reference points relative to the 620 000 t trigger level (Peatman et al. 2011; F = 0.0159, 95% CIs 0.00750–0.0357); current SSB = 97.7% SSB0; 80% CIs 71.6–135%), a level much higher than recent catches, shows consideration of precautionary issues, despite concerned comments by Kinzey et al. (2013) on the uncertainty associated with the basic parameters applied for M and recruitment variability in the GYM. The limit reference point defined indirectly as part of the harvest control rule, which allows a much larger catch than currently taken and would be far higher than the trigger level, is calculated to be at a level that would mean an extremely low risk of reproductive capacity

Justification Justification of the stock being impaired in any way. C The target reference The target reference point is such point is such that the that the stock is maintained at a stock is maintained at a level consistent with BMSY or some level consistent with BMSY measure or surrogate with similar or some measure or intent or outcome, or a higher surrogate with similar level, and takes into account

Guidepost Guidepost intent or outcome. relevant precautionary issues such as the ecological role of the stock with a high degree of certainty. Met? Yes No

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 79

PI 1.1.2 Limit and target reference points are appropriate for the stock

The target reference point is set at such a precautionary level that the stock is being maintained at a level consistent with a proxy BMSY. However, krill constitute a massive forage resource and consume a huge biomass of plankton themselves so are a crucial part of the ecosystem. Therefore, an overarching concern for management is to support the ecological role of krill, substantial biomass being maintained specifically for predation by top predators. Subdivision of the stock and quotas into SSMUs to reflect this crucial ecological role of krill, however, has not been made despite an intent to do so, and

Justification Justification management is still based on subarea maximum catches only. Hence, SG100 is not yet met. d For key low trophic level stocks, the target reference point takes into account the ecological role of the stock. Guidepost Guidepost Met? Yes

Antarctic krill is a low trophic level species, and the target reference point takes into account its ecological role through the predator criterion used in management. The reference point protecting spawning biomass is required to maintain it at a level of at least 75% of B0, although that percentage is not based currently on rigorous krill biological parameters and will be subject to updating as more information becomes available

Justification Justification through research and ecosystem analyses.

References Miller (2003), Peatman et al. (2011), Kinzey et al. (2013), WG‐EMM 2011, 2012, 2013, 2014, various reports OVERALL PERFORMANCE INDICATOR SCORE: 90

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 80

Evaluation Table for PI 1.1.3

Where the stock is depleted, there is evidence of stock rebuilding within a PI 1.1.3 specified timeframe Scoring Issue SG 60 SG 80 SG 100 a Where stocks are Where stocks are depleted, depleted rebuilding strategies are demonstrated to be strategies, which have a rebuilding stocks continuously and reasonable expectation there is strong evidence that of success, are in place. rebuilding will be complete within

Guidepost Guidepost the specified timeframe. Met? N/A N/A The Antarctic krill stock is not depleted, so no scoring is applicable. Justification Justification b A rebuilding timeframe A rebuilding timeframe is The shortest practicable is specified for the specified for the depleted rebuilding timeframe is specified depleted stock that is stock that is the shorter which does not exceed one the shorter of 30 years of 20 years or 2 times its generation time for the depleted or 3 times its generation generation time. For stock. time. For cases where 3 cases where 2 generations is less than 5 generations is less than 5 Guidepost Guidepost years, the rebuilding years, the rebuilding timeframe is up to 5 timeframe is up to 5 years. years.

Met? N/A N/A N/A

With a non‐depleted stock, there is no requirement for management to plan for rebuilding, but CCAMLR has predefined rules for stock rebuilding should such a situation eventuate. Hence, should depletion be proven in future, a rigorous rebuilding plan would be implemented immediately. Justification Justification c Monitoring is in place to There is evidence that determine whether the they are rebuilding rebuilding strategies are stocks, or it is highly likely effective in rebuilding based on simulation the stock within a modelling or previous specified timeframe. performance that they will be able to rebuild the stock within a specified

Guidepost Guidepost timeframe. Met? N/A N/A

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 81

Where the stock is depleted, there is evidence of stock rebuilding within a PI 1.1.3 specified timeframe Any rebuilding plan, if required, would stipulate the time‐frame in terms of the CCAMLR rules. Justification Justification References WG‐EMM 2011, 2012, 2013, 2014, various reports

OVERALL PERFORMANCE INDICATOR SCORE: n/a

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 82

Evaluation Table for PI 1.2.1

PI 1.2.1 There is a robust and precautionary harvest strategy in place

Scoring Issue SG 60 SG 80 SG 100 a The harvest strategy is The harvest strategy is The harvest strategy is responsive expected to achieve responsive to the state of to the state of the stock and is stock management the stock and the designed to achieve stock objectives reflected in elements of the harvest management objectives reflected the target and limit strategy work together in the target and limit reference reference points. towards achieving points. management objectives Guidepost Guidepost reflected in the target and limit reference points. Met? Yes Yes Yes Antarctic krill is exploited relatively lightly, and even when catches were higher 30 years ago, they were still very small relative to current estimated stock size. If all currently advance‐nominated catches were to be taken (they currently amount to just more than the total trigger level, but catch performance since advance notifications became mandatory dictates that they will not be taken at this point in time), they would remain small. The harvesting strategy is not robustly tested against any notable escalation in the exploitation rate, but it is unlikely that the stock would be placed at risk even if it rises sharply, and is maintained high, within the period of certification of this fishery. The target reference point of the harvest strategy aims to ensure that sufficient stock is preserved for top predators, preserving the role of krill in the ecosystem. There is a small overlap between the distribution of the krill fishery and the foraging range of predators deduced from other research, but the UoC fishery and indeed the whole fleet operating on krill do try to minimize this overlap, which tends to be seasonally variable. Quality research into this issue is continuing. The current trigger level of 620 000 t has been in place for several years, and is just 11% of the precautionary catch limit based on the GYM, which itself is based on recruitment and predator criteria within a rigorous harvest control rule. The trigger level is, however, arbitrary, although it is the lowest of several proposals made by scientists originally, and still well above the historical peak annual catches of the fishery (made during the 1980s). The precautionary catch limit is reviewed annually by CCAMLR and set at the lowest of Justification Justification recruitment‐ and predator‐targeted outcomes of the model, with the predator level yielding the lowest value currently. However, no formal catch limit is expected to be set, even if potential catches rise drastically, until formal smaller management areas (SSMUs) have been established. Research into how such SSMUs may be constructed is compre‐ hensive, but no attempt has been made recently to institute them (current suggestions are listed in Table 3 of WG‐EMM‐14/58), likely because catches are generally low and because other management issues are currently determined to be more important. However, the precautionary trigger level catch is subdivided into percentage (and tonnage) trigger levels for subareas of Area 48, so management of krill exploitation is moving towards smaller area management by default. The trigger levels for smaller subareas have been reached three times in recent years and the fishery in those areas closed for the rest of the season (notably, two of those closures were in the two most recently completed seasons). The closures of subareas generally arose because the annual geographic distribution of krill fishing operations and obviously of the available stock varies by season, depending on sea‐ ice condition and climate conditions. The fleet response to the formal CCAMLR closure was immediate and positive on each occasion.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 83

PI 1.2.1 There is a robust and precautionary harvest strategy in place b The harvest strategy is The harvest strategy may The performance of the harvest likely to work based on not have been fully strategy has been fully evaluated prior experience or tested but evidence and evidence exists to show that it plausible argument. exists that it is achieving is achieving its objectives including its objectives. being clearly able to maintain

Guidepost Guidepost stocks at target levels. Met? Yes Yes No

The harvest strategy has not been tested rigorously, but evidence is that its objectives are being achieved. The observer/monitoring programme put in place by CCAMLR for all their fisheries in the Southern Ocean (independent observer monitoring on board krill vessels plus the established CEMP monitoring of predator populations), provides no evidence, factual or anecdotal, that either krill biomass or predator populations are being impacted negatively at the current exploitation level. Market demand for krill product is also stable according to representatives of both Norwegian fisheries for krill (which dominate the total catch record for the stock), this UoC one and that already certified, and there is no unaccounted for fishing activity. It is therefore unlikely that the current strategy will fail to achieve its objectives. Catch rates are not used to model fishery performance, but several of the nations participating in the krill fishery have planned or are conducting annual or bi‐annual fishery‐

Justification Justification independent surveys of parts of the main fishing area in Area 48. Such effort is crucial given that the only fully synoptic survey of the stock took place in 2000 and there is no sign of that survey being repeated again, cost mitigating against it. The current harvesting strategy does use the findings from the 2000 survey as the basis for management advice and catch limits, but its performance is reviewed annually. The lack of factual indications that either the krill stock or the predator populations dependent upon it are currently being impacted negatively at the low level of catches beneath the conservative overall trigger level are deemed to be good evidence that the current strategy is achieving its objectives. c Monitoring is in place that is expected to determine whether the harvest strategy is working. Guidepost Guidepost Met? Yes

Monitoring of the krill and predator stocks is excellent by international standards. Independent observer coverage in the krill fishery is not 100% (although it is in the UoC fishery and the other – certified – Norwegian one) and it is disappointing that some countries do not support the independence of observers. CEMP monitoring is impressive, and regular annual effort is deployed in all of the target fishing areas to show that commercial catches of krill are not negatively impacting the predator populations, i.e. that any variations in predator populations cannot be attributed solely to the effect of the Justification Justification commercial fishery for krill. Hence, it is clear that the monitoring currently in place is sufficient to show that the harvest strategy is working effectively. d The harvest strategy is periodically reviewed and improved as necessary. Guidepost Guidepost

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 84

PI 1.2.1 There is a robust and precautionary harvest strategy in place

Met? Yes

The harvest strategy is reviewed annually, and it is also subject to slight adjustment as monitoring and research information becomes available. The written records of various meetings of WGEMM and CCAMLR prove that potentially important issues are always discussed in establishing whether the harvest strategy is working effectively or what might need to be changed. Catch limits and trigger levels for the subareas within Area 48 are reviewed annually. Currently, for instance, a suggestion has been made by one parti‐ cipating nation that the trigger levels be adjusted to reflect more appropriately the annual catch distributions. That suggestion was refuted at the most recent meeting. Catches do continue to rise slowly year on year, and one must assume too that they will continue to rise in future, but they are still well below the annual trigger level overall and only three subarea (trigger level) closures have resulted so far, all well into the season.

Justification Justification According to Watters et al. (2009), the overall management objectives may not be achieved if annual catches are made at the trigger level before smaller‐area quotas are put in place, but the establishment of the subarea trigger levels (and their implementation three times in closing subareas to harvesting) has ensured that such a situation is unlikely to arise. Management in support of the harvest strategy in this case must therefore be working, likely precluding the warning of Watters et al. (2009) that management objectives might be compromised. e It is likely that shark It is highly likely that There is a high degree of certainty finning is not taking shark finning is not taking that shark finning is not taking place. place. place. Guidepost Guidepost Met? Not relevant Not relevant Not relevant

Not a relevant issue in this fishery.

Justification Justification

References Watters et al. (2009), WG‐EMM‐14/58 (2014; seen only in draft), WG‐EMM 2011, 2012, 2013, 2014, various reports OVERALL PERFORMANCE INDICATOR SCORE: 95

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 85

Evaluation Table for PI 1.2.2

PI 1.2.2 There are well defined and effective harvest control rules in place

Scoring Issue SG 60 SG 80 SG 100 a Generally understood Well defined harvest harvest rules are in place control rules are in place that are consistent with that are consistent with the harvest strategy and the harvest strategy and which act to reduce the ensure that the exploitation rate as limit exploitation rate is

Guidepost Guidepost reference points are reduced as limit approached. reference points are approached. Met? Yes Yes The overall current level of exploitation in the Antarctic krill fishery is managed to be highly precautionary. Catches are kept within an overall trigger level that is only 11% of the catch limit, meaning that the exploitation rate does not approach either target or limit reference point, at least until quotas for SSMUs are decided and implemented. The harvest control rule is well defined and fully consistent with the declared harvesting strategy for the stock. The CCAMLR objective is to allow development of the fishery for krill up to that allowed by the harvest control rule, but the trigger level currently in place constrains catches so that

Justification Justification no reference point is approached. b The selection of the The design of the harvest control harvest control rules rules takes into account a wide takes into account the range of uncertainties. main uncertainties. Guidepost Guidepost Met? Yes No

The main uncertainties in system understanding are taken into consideration in the harvest control rule, proven by the relatively low level of catch allowed against the trigger level relative to the catch limit based on the model. Nevertheless, uncertainties associated with, for example, variations attributable to climate change and the associated oceanography are not covered explicitly, mitigating against scoring this issue above 80. However, uncertainty in recruitment and biomass levels is catered for because the harvest control rule is probabilistic and the lowest estimate of B0 is always taken, generating confidence

Justification Justification that the precautionary level of total catch allowed is appropriate. c There is some evidence Available evidence Evidence clearly shows that the that tools used to indicates that the tools in tools in use are effective in implement harvest use are appropriate and achieving the exploitation levels control rules are effective in achieving the required under the harvest control appropriate and exploitation levels rules. effective in controlling required under the

Guidepost Guidepost exploitation. harvest control rules. Met? Yes Yes No

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 86

PI 1.2.2 There are well defined and effective harvest control rules in place

The evidence here is based on effective and seemingly accurate independent observer coverage, 100% in much of the fishery. The model underlying management is based on life history parameters and the survey result, but it remains disappointing that that the only synoptic survey of the stock was 15 years ago and mainly for cost reasons is unlikely in the near future to be repeated. Biomass model input is B0, although fishing did take place subsequently to the time when B0 was established, but all evidence is that catches have been very low relative to total stock biomass. The decision rule uses a proxy for biomass at MSY, with an arbitrary adjustment to allow for the biomass necessary to support predator populations. Reported catches are the main method of controlling the catch against Justification Justification targeted maximum levels of exploitation, and they are generally well monitored and managed, although some of the conversion factors from product to green weight in some fleets vary. Overall, however, given the currently relatively low levels of catch, the tools in use are deemed to be adequate in keeping exploitation levels under control.

References WG‐EMM 2011, 2012, 2013, 2014, various reports, Demer et al. (2007), Heywood et al. 2006) OVERALL PERFORMANCE INDICATOR SCORE: 80

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 87

Evaluation Table for PI 1.2.3

PI 1.2.3 Relevant information is collected to support the harvest strategy

Scoring Issue SG 60 SG 80 SG 100 a Some relevant Sufficient relevant A comprehensive range of information related to information related to information (on stock structure, stock structure, stock stock structure, stock stock productivity, fleet productivity and fleet productivity, fleet composition, stock abundance, composition is available composition and other fishery removals and other to support the harvest data is available to information such as strategy. support the harvest environmental information), strategy. including some that may not be directly related to the current

Guidepost Guidepost harvest strategy, is available. Met? Yes Yes No Krill fleet activity information is excellent, including the (admittedly overoptimistic) notification system plus impressive use of VMS, observers, logsheets and (potentially) licensing information from the GSGSSI; the reporting system is swiftly responsive, providing accurate and immediate data on fleet distribution. The synoptic survey of 2000 and the various other surveys conducted before and after provide most of the baseline data on stock structure and abundance. However, more information on the spatial distribution of biomass is crucial if SSMU management is to be instituted as planned in future with any confidence, and that varies seasonally and annually depending upon environmental conditions. Productivity is assessed through the results of annual reviews of krill stock health, harvest strategy and catch limits are reviewed, and the catch Justification Justification composition (of krill and the minute quantities of other species taken) is determined from the observer and monitoring processes. Fishery‐independent surveys and data collections yield most of the environmental data available, but the quality and quantity of such data is always compromised by the inaccessibility of the area, so are not that comprehensive by international standards. b Stock abundance and Stock abundance and All information required by the fishery removals are fishery removals are harvest control rule is monitored monitored and at least regularly monitored at a with high frequency and a high one indicator is available level of accuracy and degree of certainty, and there is a and monitored with coverage consistent with good understanding of inherent sufficient frequency to the harvest control rule, uncertainties in the information support the harvest and one or more [data] and the robustness of control rule. indicators are available assessment and management to Guidepost Guidepost and monitored with this uncertainty. sufficient frequency to support the harvest control rule. Met? Yes Yes No

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 88

PI 1.2.3 Relevant information is collected to support the harvest strategy

Stock abundance is monitored through regular review of all catch and effort information and any newly available scientific evidence provided by the scientific community. Catch monitoring in the fishery is excellent, reports being submitted at worst monthly, and given current market stability in terms of product take‐up, there is no incentive to misreport. There is no discarding in the UoC fishery, but a small amount may take place in other fisheries for krill, particularly where the size of the animals (mainly for the purposes of human consumption) is important. Conversion factors in such fisheries are not consistent, but with the current low level of exploitation, that situation is unlikely to cause a

Justification Justification management control issue. The harvest control rule for krill fishing in place depends on reasonable knowledge of krill life history, and although there has been improvement in that knowledge base over time, some elements, such as age and early life history, remain insufficiently proven, hence precluding this item being scored above SG80. c There is good information on all other fishery removals from the stock. Guidepost Guidepost Met? Yes

Virtually the whole catch is taken from the target area specified for the UoC fishery, with very little or no krill annually being caught in adjacent areas that might hold a small part of the same stock, and none from outside the CCAMLR area, so all extractions are well documented. Fishing for krill is also virtually clean, by the UoC and other fisheries operating with similar gear, few other fish species being taken (see P2 scoring beneath). It

Justification Justification is notable too that, in this fishery, incentives to misreport do not exist. [List any references here] WG‐EMM 2011, 2012, 2013, 2014, various reports, Demer et al. References (2007), Heywood et al. 2006), WG‐EMM‐13/37 Rev. 1 (2013), WG‐EMM‐14/58 (2014; seen only in draft) OVERALL PERFORMANCE INDICATOR SCORE: 80

CONDITION NUMBER (if):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 89

Evaluation Table for PI 1.2.4

PI 1.2.4 There is an adequate assessment of the stock status

Scoring Issue SG 60 SG 80 SG 100 a The assessment is The assessment is appropriate for appropriate for the stock the stock and for the harvest and for the harvest control rule and takes into control rule. account the major features relevant to the biology of the species and the nature of the

Guidepost Guidepost fishery. Met? Yes No Antarctic krill are not subject to annual stock assessment of the nature generally applied in fisheries internationally; catches are very small relative to survey‐determined stock size, and there is a general lack of the appropriate up‐to‐date data to support such an assessment. However, the approach of using a GYM and predator and recruitment criteria is appropriate for the stock in terms of determining a precautionary upper catch limit. Note too that the trigger level set for the fishery (i.e. the maximum catch that may be taken annually) is just 11% of that upper catch limit, so is highly precautionary. It is only Justification Justification the lack of baseline up‐to‐date information on stock biology and distribution that mitigates against scoring more than 80 for this item. b The assessment estimates stock status relative to reference points. Guidepost Guidepost Met? Yes

Target and limit reference points are taken into consideration in the GYM approach and harvest control rule, but catch maxima set are well below the suggested levels allowed by the inclusion in management of a trigger level of catch. Stock status is reviewed regularly through this approach. Justification Justification c The assessment The assessment takes The assessment takes into account identifies major sources uncertainty into account. uncertainty and is evaluating stock of uncertainty. status relative to reference points in a probabilistic way. Guidepost Guidepost Met? Yes Yes No

The GYM is stochastic and uses a probabilistic decision rule, so taking uncertainty into account. Although some structural uncertainties are not considered, with the GYM based mainly on dated survey data, the precautionary nature of management, instituted through the lowest of several candidate values for B0 being used to determine the catch limit and an even more restrictive trigger level, do support the belief that uncertainty is effectively

Justification Justification covered by the system now in place. d The assessment has been tested and shown to be robust. Alternative hypotheses and assessment approaches have been

Guidepost Guidepost rigorously explored.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 90

PI 1.2.4 There is an adequate assessment of the stock status

Met? Yes

The assessment is reviewed annually and any new data are employed to ensure that it is as robust as it can possibly be. Alternative hypotheses for stock status have been explored through various different models and parameter choices (e.g. for target strength in the acoustic assessment), but at present, the most conservative estimate of B0 is still used in priming the harvest rule and ensuring sustainability. Trigger level catches (for subareas and

Justification Justification overall) remain well below assessment‐derived catch limits. e The assessment of stock The assessment has been status is subject to peer internally and externally peer review. reviewed. Guidepost Guidepost Met? Yes No

A limited amount of external peer review of the assessment does take place occasionally, but most of the (thorough) annual review is through the CCAMLR WG system. That system utilizes the numerical and analytical skills of scientists in a great many countries (>20), many well versed in stock assessments internationally, and many from countries not actually operating in the krill fishery currently.

Justification Justification

References WG‐EMM‐2011, 2012, 2013, 2014, various reports, de la Mare (1994a, 1994b), Jolly and Hampton (1990), Constable and de la Mare (1996) OVERALL PERFORMANCE INDICATOR SCORE: 85

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 91

Evaluation Table for PI 2.1.1

The fishery does not pose a risk of serious or irreversible harm to the PI 2.1.1 retained species and does not hinder recovery of depleted retained species Scoring Issue SG 60 SG 80 SG 100 a Main retained species Main retained species are There is a high degree of certainty are likely to be within highly likely to be within that retained species are within biologically based limits biologically based limits biologically based limits and (if not, go to scoring (if not, go to scoring issue fluctuating around their target

Guidepost Guidepost issue c below). c below). reference points. Met? Yes Yes Yes There are no retained species other than krill, the target species. Two crew members remove anything that is not krill from the conveyor belt into a bucket which is then sampled by the observer. There is 100% international independent observer coverage on‐ board sample catch composition according to the CCAMLR Scientific Observer Manual, identifying and quantifying both krill and any other species in the catch that might be discarded and considered as bycatch under PI2.2.1. However, the bycatch proportion is considered by the observers to represent <0.5% of the total catch. Justification Justification The team considers that there are no retained species in the fishery, so, according to MSC CR v1.3 CB 3.2.1, scores this guidepost at 100. b Target reference points are defined for retained species. Guidepost Guidepost Met? n/a

As mentioned above, the team considers that there are no retained species in the catch and according to MSC CR v1.3 CB 3.2.1 scores this guidepost at 100. Justification Justification c If main retained species If main retained species are outside the limits are outside the limits there are measures in there is a partial strategy place that are expected of demonstrably effective to ensure that the management measures fishery does not hinder in place such that the recovery and rebuilding fishery does not hinder

Guidepost Guidepost of the depleted species. recovery and rebuilding. Met? n/a n/a

This guidepost is not relevant as there are no retained species to consider. Justification Justification

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 92

The fishery does not pose a risk of serious or irreversible harm to the PI 2.1.1 retained species and does not hinder recovery of depleted retained species d If the status is poorly known there are measures or practices in place that are expected to result in the fishery not causing the retained species to be outside biologically based limits

Guidepost Guidepost or hindering recovery. Met? n/a

This guidepost is not relevant as there are no retained species to consider. Justification Justification  CCAMLR Scientific Observer Manual. References  CCAMLR Scientific Observer Reports for the years 2010–2014. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 93

Evaluation Table for PI 2.1.2

There is a strategy in place for managing retained species that is designed to PI 2.1.2 ensure the fishery does not pose a risk of serious or irreversible harm to retained species Scoring Issue SG 60 SG 80 SG 100 a There are measures in There is a partial strategy There is a strategy in place for place, if necessary, that in place, if necessary, that managing retained species. are expected to maintain is expected to maintain the main retained the main retained species species at levels which at levels which are highly are highly likely to be likely to be within within biologically based biologically based limits, limits, or to ensure the or to ensure the fishery fishery does not hinder does not hinder their their recovery and recovery and rebuilding.

Guidepost Guidepost rebuilding. Met? Yes Yes Yes There are no main retained species to consider under this PI, because the Juvel only retains krill on board. Catch composition is homogeneous, because krill tend to be in layers and patches ranging from a few square metres across, through shoals, schools, swarms and up to superswarms covering more than 100 km2. Two crew members work alongside the conveyor belt removing anything that is not krill and putting it into a different bucket for the observer to examine subsequently. The company has stated its intention to continue to retain only krill, and the Juvel has

Justification Justification always allowed 100% observer coverage, greater than the 50% CCAMLR requirement, which serves to monitor the fishery on a continuous basis and to detect any impact on retained species if any. The team considers that these measures are sufficient to meet SG100, according to MSC CRG v1.3 GCB 3.3.2. b The measures are There is some objective Testing supports high confidence considered likely to basis for confidence that that the strategy will work, based work, based on plausible the partial strategy will on information directly about the argument (e.g., general work, based on some fishery and/or species involved. experience, theory or information directly comparison with similar about the fishery and/or

Guidepost Guidepost fisheries/species). species involved. Met? Yes Yes Yes

The facts that the Juvel only retains krill, that there are crew members who check the catch on the conveyor belt, that there is 100% observer coverage to monitor catch composition, and that there is a very low bycatch percentage (<0.5%) is sufficient for the team to state confidently that the strategy will work, as it is already doing. Justification Justification c There is some evidence There is clear evidence that the that the partial strategy is strategy is being implemented being implemented successfully. successfully. Guidepost Guidepost Met? Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 94

There is a strategy in place for managing retained species that is designed to PI 2.1.2 ensure the fishery does not pose a risk of serious or irreversible harm to retained species As mentioned above, this strategy is already successfully implemented. Evidence for this can be found in CCAMLR observer reports. Justification Justification d There is some evidence that the strategy is achieving its overall objective. Guidepost Guidepost Met? Yes

The low bycatch proportion along with the fact that the Juvel only works with krill and krill products are the evidence that prove that the strategy its achieving its aim. Justification Justification e It is likely that shark It is highly likely that There is a high degree of certainty finning is not taking shark finning is not taking that shark finning is not taking place. place. place. Guidepost Guidepost Met? n/a n/a n/a

There is no shark‐finning Justification Justification References  CCAMLR Scientific observer reports. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 95

Evaluation Table for PI 2.1.3

Information on the nature and extent of retained species is adequate to PI 2.1.3 determine the risk posed by the fishery and the effectiveness of the strategy to manage retained species Scoring Issue SG 60 SG 80 SG 100 a Qualitative information Qualitative information Accurate and verifiable is available on the and some quantitative information is available on the amount of main retained information are available catch of all retained species and species taken by the on the amount of main the consequences for the status of fishery. retained species taken by affected populations.

Guidepost Guidepost the fishery. Met? Yes Yes Yes The UoC falls completely under the CCAMLR international observer programme and has always had 100% observer coverage. The well‐constructed CCAMLR scientific observer manual provides guidelines for scientific observers in terms of sampling processes, data collection and the observations to be made on board. Both the manual and the reports are available on the CCAMLR website, and the manual accompanies all observers. The reports provide accurate and verifiable information on the catch of all species, and specify that

Justification Justification there are no retained species apart from krill. b Information is adequate Information is sufficient Information is sufficient to to qualitatively assess to estimate outcome quantitatively estimate outcome outcome status with status with respect to status with a high degree of respect to biologically biologically based limits. certainty.

Guidepost Guidepost based limits. Met? Yes Yes Yes

As mentioned above, observer reports give detailed information on catch composition, which is sampled according to guidelines in the CCAMLR observer manual. This sampling provides quantitative information on the catch composition of all species and should be sufficient to estimate the outcome status of retained species if there were any. However, as the Juvel only works with krill products, no other species apart from krill is retained.

Justification Justification Bycatch has been calculated to be negligible, at <0.5% of the total catch. c Information is adequate Information is adequate Information is adequate to to support measures to to support a partial support a strategy to manage manage main retained strategy to manage main retained species, and evaluate species. retained species. with a high degree of certainty whether the strategy is achieving

Guidepost Guidepost its objective. Met? Yes Yes Yes

Information gathered by observers is considered adequate to support a strategy to manage retained species if there were any, and to evaluate its effectiveness. As all krill vessels in the Southern Ocean must comply with a minimum 50% observer coverage (though the Juvel covers 100%%), CCAMLR has access to information on catch composition of other vessels in the area which are not part of this UoC. Justification Justification

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 96

Information on the nature and extent of retained species is adequate to PI 2.1.3 determine the risk posed by the fishery and the effectiveness of the strategy to manage retained species d Sufficient data continue Monitoring of retained species is to be collected to detect conducted in sufficient detail to any increase in risk level assess ongoing mortalities to all (e.g. due to changes in retained species. the outcome indicator score or the operation of

Guidepost Guidepost the fishery or the effectiveness of the strategy) Met? Yes Yes

100% international independent observer coverage on board the Juvel ensures detailed monitoring of all species in the catch, yielding accurate figures on bycatch mortality. As only krill is retained, there are no other retained species to consider. Nevertheless, accurate information is given on catch composition and bycatch species. Justification Justification  CCAMLR scientific observer manual References  CCAMLR observer reports for the years 2010–2014. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 97

Evaluation Table for PI 2.2.1

The fishery does not pose a risk of serious or irreversible harm to the bycatch PI 2.2.1 species or species groups and does not hinder recovery of depleted bycatch species or species groups Scoring Issue SG 60 SG 80 SG 100 a Main bycatch species are Main bycatch species are There is a high degree of certainty likely to be within highly likely to be within that bycatch species are within biologically based limits biologically based limits biologically based limits. (if not, go to scoring (if not, go to scoring issue

Guidepost Guidepost issue b below). b below). Met? Yes Yes Yes According to CCAMLR Scientific Observer reports, the Juvel’s larval fish bycatch proportion makes up <0.5% of the total catch. This is why the team considers that there are no species to consider as main bycatch species. As for minor bycatch species, these have been listed in the bycatch species section (3.4.3 Bycatch), Tables 2 (fish) and 3 (seabirds). As seen in observer reports, finfish species present in the catch of both the Juvel (present UoC) and Aker Biomarine’s similarly operating Saga Sea are similar. This is because they both fish the same stock and area, using a fine mesh to prevent the entanglement of bigger organisms. The status of the majority of species listed in Table 2 in subareas 48.1, 48.2 and 48.3 has already being studied in the MRAG 2012 Analysis of larval bycatch on the Saga Sea, which identifies myctophids (lanternfish, LAN) and channichthyids (icefish, ICE) as dominating the retained species in the catch, with lower levels of nototheniid (NOT) species, which suffered from overexploitation in the 1970s. Of these, species of major concern were Champsocephalus gunnari and Notothenia rossi, which may be outside biologically based limits. The work carried out by MRAG looked at the biomass estimates for these species in the different subareas and concluded that there is no significant risk to any of their populations within Area 48. Species present in the catch that were not considered in the MRAG report are also listed in Table 2, and have not being evaluated by

Justification Justification CITES or IUCN. However, due to the small bycatch proportion, the team considers that bycatch is rare and negligible, and according to MSC CR CB3.8.3 concludes that SG100 is met for fish bycatch. In terms of the bycatch of seabirds and marine mammals, any interactions with such species are recorded in the comprehensive scientific observer reports. Table 3 lists the fatalities of seabirds after entanglement with the nets, and reveals a maximum of 4 burds killed per year. All were species of least concern according to the IUCN red list and are considered to be within biologically based limits. Interactions with marine mammals are also recorded in observer reports. In the past years only one direct interaction has been recorded, in April 2013, when a Weddell seal was entangled with the gear and released unharmed. Weddell seals are considered to be within biologically based limits. The team considers that SG100 is met owing to the low fish bycatch proportions and the good stock status of interacting seabirds and mammals. b If main bycatch species If main bycatch species are outside biologically are outside biologically based limits there are based limits there is a mitigation measures in partial strategy of place that are expected demonstrably effective to ensure that the mitigation measures in

Guidepost Guidepost fishery does not hinder place such that the recovery and rebuilding. fishery does not hinder recovery and rebuilding. Met? Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 98

The fishery does not pose a risk of serious or irreversible harm to the bycatch PI 2.2.1 species or species groups and does not hinder recovery of depleted bycatch species or species groups The team considers that there are no main bycatch species to consider under this scoring guidepost. Nonetheless, it recognizes that there are effective mitigation measures in place ensuring that the fishery is not unduly influencing the status or hindering the recovery of bycatch species. These measures are:  the precautionary trigger level for the krill fishery (which is currently just 11% of the estimated catch limit overall),  the partition of the TAC among smaller subareas to ensure control over depletion in those areas,

Justification Justification  the exclusion net of fine mesh,  the design of the net,  the marine mammal exclusion device (SLED)  the operational strategy, with quick sinking of the net  monitoring of the catch composition through 100% observer coverage. c If the status is poorly known there are measures or practices in place that are expected to result in the fishery not causing the bycatch

Guidepost Guidepost species to be outside biologically based limits or hindering recovery. Met? Yes

The status of most of the species that are caught is well known. Some have not been classified yet by CITES nor IUCN, but there is information about their population on the CCAMLR website and their catch rates have proven to be anecdotal (see Table2). There is also a partial strategy in place that would avoid any hindrance to the recovery of species such as Nototheniops rosii or Champsocephalus gunnari. Justification Justification  CCAMLR Scientific Observer Cruise Reports for the years 2010–2014. References  MRAG, 2012. Analysis of larval bycatch on the Saga Sea during continuous trawling for krill in CCAMLR Areas 48 between December 2007 and September 2011. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 99

Evaluation Table for PI 2.2.2

There is a strategy in place for managing bycatch that is designed to ensure PI 2.2.2 the fishery does not pose a risk of serious or irreversible harm to bycatch populations Scoring Issue SG 60 SG 80 SG 100 a There are measures in There is a partial strategy There is a strategy in place for place, if necessary, that in place, if necessary, that managing and minimizing bycatch. are expected to maintain is expected to maintain the main bycatch species the main bycatch species at levels which are highly at levels which are highly likely to be within likely to be within biologically based limits, biologically based limits, Guidepost Guidepost or to ensure the fishery or to ensure the fishery does not hinder their does not hinder their recovery and rebuilding. recovery and rebuilding. Met? Yes Yes Yes The team considers that there is a strategy in place for managing and minimizing bycatch which consists of the following measures:  CCAMLR Conservation Measure 25‐03, which is designed to minimize the number of incidental mortalities of seabirds and marine mammals in the course of trawl fishing in the Antarctic Convention Area.  Conservation Measures 26‐01 and 51‐01 which also contribute to prevent possible impacts of the fishery on other species apart from krill.  Use of a Sea Lion Exclusion Device (SLED).  A fine‐mesh exclusion net at the codend.  Daily check of the fishing gear, and repair if necessary.  Quick sinking of the net when shooting and hauling.  Careful cleaning of the net and codend before every shooting.  Discharge of stick water to be at a minimum during shooting and hauling.  A slow towing speed (<2 knots) that allows animals to avoid the net.  Avoidance of shooting and hauling in areas with large numbers of seals and

Justification Justification penguins. Spatial limitations around South Georgia and South Sandwich Islands (where the Juvel has not fished in the past 2 years).  Juvel’s own protocol on interactions with birds and marine mammals.  Hourly rounds to check for birds on deck, and a protocol to be followed by crew members if there are any.  Retention on board of all material captured.  Crew members at the conveyor belt removing bycatch species from the catch, which are kept in a bucket and handed to the observer for identification.  100% observer coverage which samples catch composition and monitors interactions with birds and marine mammals. Krill’s tendency to form homogeneous shoals, schools, swarms and superswarms also contributes to a clean catch with small bycatch proportions.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 100

There is a strategy in place for managing bycatch that is designed to ensure PI 2.2.2 the fishery does not pose a risk of serious or irreversible harm to bycatch populations b The measures are There is some objective Testing supports high confidence considered likely to basis for confidence that that the strategy will work, based work, based on plausible the partial strategy will on information directly about the argument (e.g. general work, based on some fishery and/or species involved. experience, theory or information directly comparison with similar about the fishery and/or

Guidepost Guidepost fisheries/species). species involved. Met? Yes Yes Yes

The 100% CCAMLR observer coverage on board the Juvel provides direct information about the fishery and the species involved. These reports include information on the target species but also identify and quantify interactions with bycatch species, such as fish, birds and marine mammals. Observer reports from the years 2010–2014 reveal a very small proportion of fish bycatch (<0.5%, which is in accordance with results from the risk assessment carried out in 2012 by MRAG using information of Aker Biomarine’s vessels targeting krill in Area 48). Interactions and sightings of seabirds and marine mammals are also recorded in observer reports and Justification Justification have proven to be minimal in the past five years. The Friends of the Sea Certification Report also mentions this small interaction.

The above reasons support a high confidence that this strategy will work. c There is some evidence There is clear evidence that the that the partial strategy is strategy is being implemented being implemented successfully. successfully. Guidepost Guidepost Met? Yes Yes

CCAMLR Conservation Measures 25‐03, 26‐01 and 51‐01, Juvel’s 100% observer coverage since its first trip in 2009, crew members separating catch from bycatch, the vessel’s protocol on interactions with seabirds and marine mammals and results from the exhaustive observer records are considered as evidence that this strategy is successfully implemented. Justification Justification d There is some evidence that the strategy is achieving its overall objective. Guidepost Guidepost Met? Yes

The small proportion of fish bycatch, along with the few interactions with seabirds (1 snow petrel dead in 2014, 4 Cape petrels dead in 2013, 2 Cape petrels, one southern fulmar and one blue petrel dead in 2012 and another Cape petrel dead in 2010) and marine mammals (one Weddell seal entangled in the gear in 2013, but released alive) can be considered as evidence that the strategy is achieving its overall objective. Justification Justification  CCAMLR Scientific Observer manual.

References  CCAMLR observer reports for the years 2010–2014‐ .  Juvel’s protocol on interactions with seabirds and marine mammals.  MRAG, 2012. Analysis of larval bycatch on the Saga Sea during continuous

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 101

There is a strategy in place for managing bycatch that is designed to ensure PI 2.2.2 the fishery does not pose a risk of serious or irreversible harm to bycatch populations trawling for krill in CCAMLR Areas 48 between December 2007 and September 2011.  Olympic krill Friends of the Sea certification report. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 102

Evaluation Table for PI 2.2.3

Information on the nature and the amount of bycatch is adequate to PI 2.2.3 determine the risk posed by the fishery and the effectiveness of the strategy to manage bycatch Scoring Issue SG 60 SG 80 SG 100 a Qualitative information Qualitative information Accurate and verifiable is available on the and some quantitative information is available on the amount of main bycatch information are available catch of all bycatch species and species taken by the on the amount of main the consequences for the status of fishery. bycatch species taken by affected populations.

Guidepost Guidepost the fishery. Met? Yes Yes Yes The fishery falls completely under the CCAMLR international observer programme with 100% observer coverage. The well‐constructed CCAMLR scientific observer manual provides guidelines for scientific observers in terms of sampling processes, data collection and the observations to be made on board. Accurate and verifiable information on all bycatch species is available through the manifold observer reports, which show small fish bycatch proportion in the catch composition (<0.5%), abundant sightings of seabirds and marine mammals and extremely sporadic interactions between these species and the fishing gear during shooting and hauling (10 seabird casualties recorded in the past five years, and one Weddell seal entangled which was released alive). All these species are

Justification Justification categorized as species of least concern according to the IUCN red list, implying little consequence for the affected populations. Due to the small bycatch proportion and the few interactions of the gear with seabirds and marine mammals, consequences for the status of affected populations are considered to be negligible. b Information is adequate Information is sufficient Information is sufficient to to broadly understand to estimate outcome quantitatively estimate outcome outcome status with status with respect to status with respect to biologically respect to biologically biologically based limits. based limits with a high degree of

Guidepost Guidepost based limits certainty. Met? Yes Yes Yes

Data on catch composition and seabirds and marine mammals sightings and interactions are collected on an ongoing basis through CCAMLR scientific observer reports and are considered to be sufficient to quantitatively estimate outcome status with respect to biologically based limits. The fish larval bycatch work, carried out by MRAG in 2012 on Aker Biomarine’s krill vessels, looked at biomass estimates for fish larval species in the different subareas with respect to biologically based limits. As the majority of the species in the catch composition for both fisheries are the same, the results of MRAG’s work are applicable to this UoC fishery.

Justification Justification Moreover, as the bycatch is negligible, it is expected that biologically based limits for such species will not be affected by the fishery. In terms of the 10 seabirds and the Weddell seal which interacted with the vessel in the past five years, they are considered to be of Least Concern by IUCN, which gives confidence that impacts on these species are negligible. c Information is adequate Information is adequate Information is adequate to to support measures to to support a partial support a strategy to manage manage bycatch. strategy to manage main bycatch species, and evaluate with bycatch species. a high degree of certainty whether the strategy is achieving its

Guidepost Guidepost objective.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 103

Information on the nature and the amount of bycatch is adequate to PI 2.2.3 determine the risk posed by the fishery and the effectiveness of the strategy to manage bycatch Met? Yes Yes Yes

100% international observer coverage on board the Juvel ensures exhaustive information on catch composition and seabirds and marine mammals interactions and abundance, which is reported to CCAMLR on a continuous basis. The CCAMLR Observer programme ensures a minimum of 50% observer coverage on all vessels, which gives CCAMLR information on the catch composition of other vessels not part of the UoC fishery fishing in the same area. This comprehensive information base is adequate to support a strategy to manage bycatch

Justification Justification species and to evaluate its effectiveness. d Sufficient data continue Monitoring of bycatch data is to be collected to detect conducted in sufficient detail to any increase in risk to assess ongoing mortalities to all main bycatch species bycatch species. (e.g., due to changes in the outcome indicator scores or the operation Guidepost Guidepost of the fishery or the effectively of the strategy). Met? Yes Yes

100% observer coverage in the UoC ensures an exhaustive monitoring of catch composition and bycatch species, which serves to assess ongoing mortality to all bycatch species. Observers not only record catch composition and fish larvae identification, but also sightings, entanglements and fatalities of seabirds and marine mammals. Whenever the observer has difficulties in taxonomically identifying some individuals, this is solved by photographing and conserving them for confirmatory identification by scientists on South Georgia or in the UK. The CCAMLR observer programme started in 2006 and will continue in future, so it should be sufficient to assure future monitoring of bycatch species

Justification Justification mortality.  CCAMLR Scientific observer manual  CCAMLR observer reports for the years 2010–2014. References  MRAG, 2012. Analysis of larval bycatch on the Saga Sea during continuous trawling for krill in CCAMLR Areas 48 between December 2007 and September 2011. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 104

Evaluation Table for PI 2.3.1

The fishery meets national and international requirements for the protection of ETP species PI 2.3.1 The fishery does not pose a risk of serious or irreversible harm to ETP species and does not hinder recovery of ETP species Scoring Issue SG 60 SG 80 SG 100 a Known effects of the The effects of the fishery There is a high degree of certainty fishery are likely to be are known and are highly that the effects of the fishery are within limits of national likely to be within limits within limits of national and and international of national and international requirements for requirements for international protection of ETP species. protection of ETP requirements for

Guidepost Guidepost species. protection of ETP species. Met? Yes Yes Yes The Juvel is equipped with a Sea Lion Excluder Device (SLED) designed for effectively closing off the net entrance in order to prevent entanglements of marine mammals. Also, all vessels operating in the Antarctic krill fishery are required to have a minimum of 50% observer coverage in order to record fishing vessel activities. The CCAMLR Scientific Observer Manual also requires the observer to record any interactions with ETP species along with ETP sightings from the vessel. There are identification guides available for the observer on the bridge of the Juvel. Since its first trip in 2009, the Juvel has always carried an independent scientific observer on board during its fishing trips. Reports from the past Justification Justification five years (since the Juvel’s first trip) show no interactions at all. The presence of an independent observer on board at all times and the lack of interactions in all years provide robust proof that the effects of the fishery are within international requirements for protection of ETP species. b Known direct effects are Direct effects are highly There is a high degree of unlikely to create unlikely to create confidence that there are no unacceptable impacts to unacceptable impacts to significant detrimental direct ETP species. ETP species. effects of the fishery on ETP

Guidepost Guidepost species. Met? Yes Yes Yes

The permanent observer coverage, along with the lack of interactions recorded in the past three years provides a high degree of confidence in that there are no significant detrimental direct effects of the fishery on ETP species. Justification Justification c Indirect effects have There is a high degree of been considered and are confidence that there are no thought to be unlikely to significant detrimental indirect create unacceptable effects of the fishery on ETP impacts. species. Guidepost Guidepost Met? Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 105

The fishery meets national and international requirements for the protection of ETP species PI 2.3.1 The fishery does not pose a risk of serious or irreversible harm to ETP species and does not hinder recovery of ETP species Indirect effects of the fishery on ETP species could be those caused by the extraction of prey in the trophic chain or by the possible overlap of fishing areas with foraging areas. Krill removals are regulated by CCAMLR Conservation Measure 51‐07 in a highly precautionary manner, which allocates trigger levels in Subareas 48.1, 48.2, 48.3 and 48.4 in order to avoid local depletion of the krill population. In terms of possible interactions with ETP foraging areas, these were studied by Nicoll and Douglass (2012) and showed that the krill fishery takes place, among other sites, in the foraging areas of Antarctic fur seals around the coast of the Antarctic Peninsula. Notwithstanding this, it is also considered that at the current harvesting rate of krill there would be no significant detrimental indirect effects on this species. However, if the catch rate were to increase, species that are constrained in their foraging areas could be affected, so this could only be allowed with adequate management provisions which would need to include a high level of precautionary protection. Justification Justification Hewitt et al. (2004) estimated the annual consumption of krill by different species in Area 48, and showed that fur seals would eat 706.7 thousand tonnes per year, whales 2 360 thousand tonnes, fish 2 963.9 thousand tonnes and penguins up to 9 192.1 thousand tonnes. These estimates add up to 15 223 thousand tonnes of krill potentially consumed annually by the different predators. As removals by the fishery have been estimated to be several orders of magnitude less than both the demand from predators and the biomass available for both predators and the fishery, it is considered that there are no significant detrimental indirect effects of the fishery on ETP species.  CCAMLR Scientific observer reports for the years 2010–2014.  CCAMLR Scientific Observer Manual http://www.ccamlr.org/en/system/files/obsman.pdf  .  Convention on International Trade in Endangered Species of Wild Fauna and Flora References list. http://www.cites.org/eng/disc/species.php  Hewitt et al. 2004. Options for allocating the precautionary catch limit of krill among small scale management units in the Scotia Sea. CCAMLR Science, Vol. 11 (2004): 81–97. http://www.ccamlr.org/en/system/files/science_journal_papers/05hewitt‐ etal.pdf  Nicoll, R., and Douglass, L. 2012. Project report: Mapping krill trawling and predator distribution. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 106

Evaluation Table for PI 2.3.2

The fishery has in place precautionary management strategies designed to:  Meet national and international requirements; PI 2.3.2  Ensure the fishery does not pose a risk of serious harm to ETP species;  Ensure the fishery does not hinder recovery of ETP species; and  Minimise mortality of ETP species. Scoring Issue SG 60 SG 80 SG 100 a There are measures in There is a strategy in There is a comprehensive strategy place that minimize place for managing the in place for managing the fishery’s mortality of ETP species, fishery’s impact on ETP impact on ETP species, including and are expected to be species, including measures to minimize mortality, highly likely to achieve measures to minimize which is designed to achieve national and mortality, which is above national and international international designed to be highly requirements for the protection of

Guidepost Guidepost requirements for the likely to achieve national ETP species. protection of ETP and international species. requirements for the protection of ETP species. Met? Yes Yes Yes CCAMLR Conservation Measure 25‐03 is designed to minimize the incidental mortality of seabirds and marine mammals in the course of trawling in the Antarctic Convention Area. Conservation Measures 26‐01 and 51‐01 also contribute to prevent possible impacts of the fishery on other species apart from krill. These measures are binding for all vessels operating in Area 48. The Juvel has also established its own rigorous protocol in order to avoid or prevent such impacts, which consists of the following measures:  Use of a Sea Lion Exclusion Device which shall be properly maintained.  Fine‐mesh exclusion net at the codend  Daily check of the fishing gear, and repair if necessary.  Quick shooting and hauling of the net.  Careful cleaning of the net and codend before every shooting.  A slow towing speed (<2 knots) that allows animals to avoid the net  Avoidance of shooting and hauling in areas with large number of seals and penguins

Justification Justification  Discharge of stick water to be minimized during shooting and hauling.  Steps to follow by crew members in case a seal is on deck (which happened once on the 12th April 2013, when a Weddell seal swam onto a trawl net during hauling, but slipped off on the trawl ramp. The seal sustained no injuries and was released alive. However, Weddell seals are not considered ETP species, but the same protocol should be used in case of entanglements with ETP species.  Hourly rounds to check for birds on deck, and the protocol to be followed by crew members in case there are any. There are also spatial and seasonal limitations around South Georgia and the South Sandwich Islands in order to protect ETP species, but the Juvel has not fished around those islands for the past three years. Both CCAMLR Conservation Measures and the Juvel’s own protocol serve as a comprehensive strategy for managing the fishery’s impact on ETP species.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 107

The fishery has in place precautionary management strategies designed to:  Meet national and international requirements; PI 2.3.2  Ensure the fishery does not pose a risk of serious harm to ETP species;  Ensure the fishery does not hinder recovery of ETP species; and  Minimise mortality of ETP species. b The measures are There is an objective The strategy is mainly based on considered likely to basis for confidence that information directly about the work, based on plausible the strategy will work, fishery and/or species involved, argument (e.g., general based on information and a quantitative analysis experience, theory or directly about the fishery supports high confidence that the comparison with similar and/or the species strategy will work.

Guidepost Guidepost fisheries/species). involved. Met? Yes Yes Yes

Scientific observers on board provide accurate information on any interactions, and are required to record the sightings of marine mammals and seabirds in the vicinity of the vessel, and to count them if possible. Observer reports are kept by CCAMLR and provide direct information about the fishery. Observer reports from the years 2010–2014 (since the Juvel’s first trip) show no direct interactions between the vessel and ETP species. The absence of interactions with ETP species give confidence that the strategy will work well in

Justification Justification future too. c There is evidence that There is clear evidence that the the strategy is being strategy is being implemented implemented successfully. successfully. Guidepost Guidepost Met? Yes Yes

CCAMLR Annual Fishery Reports, along with CCAMLR Annual Observer Reports and CCAMLR Scientific Observer Reports provide clear evidence that the strategy is being implemented successfully. Justification Justification d There is evidence that the strategy is achieving its objective. Guidepost Guidepost Met? Yes

The fact that the Juvel has never had any interaction with ETP species can be considered as evidence that the strategy is achieving its objective. Justification Justification  CCAMLR Annual Fishery Reports  CCAMLR Annual Observer reports  CCAMLR Conservation Measure 25‐03 (2011) on the minimization of incidental References mortalities of seabirds and marine mammals in the course of trawl fishing in the Convention Area. http://www.ccamlr.org/en/measure‐25‐03‐2011  CCAMLR Conservation Measure 26‐01 (2009) on general environmental protection during fishing. http://www.ccamlr.org/en/measure‐26‐01‐2009

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 108

The fishery has in place precautionary management strategies designed to:  Meet national and international requirements; PI 2.3.2  Ensure the fishery does not pose a risk of serious harm to ETP species;  Ensure the fishery does not hinder recovery of ETP species; and  Minimise mortality of ETP species.  CCAMLR Conservation Measure 51‐01 (2010) regarding Precautionary catch limitations on Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51‐01.pdf  CCAMLR Scientific Observer Manual. http://www.ccamlr.org/en/system/files/obsman.pdf  CCAMLR Scientific Observer Cruise Reports for the years 2010–2014. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 109

Evaluation Table for PI 2.3.3

Relevant information is collected to support the management of fishery impacts on ETP species, including: PI 2.3.3  Information for the development of the management strategy;  Information to assess the effectiveness of the management strategy; and  Information to determine the outcome status of ETP species. Scoring Issue SG 60 SG 80 SG 100 a Information is sufficient Sufficient information is Information is sufficient to to qualitatively estimate available to allow fishery quantitatively estimate outcome the fishery related related mortality and the status of ETP species with a high mortality of ETP species. impact of fishing to be degree of certainty. quantitatively estimated

Guidepost Guidepost for ETP species. Met? Yes Yes No According to CCAMLR’s Observer Manual it is the duty of observers to record any incident related to all ETP species present in the area, whether it is an interaction, an impact, an injury or a mortality of marine mammals and birds. The information thus obtained, together with that collected by the CEMP and research institutes (such as the British Antarctic Survey, the Norwegian Institute of Marine Research, and the South Georgia and the South Sandwich Islands Environmental Agency) are enough to make a quantitative estimate of the fishery impact on ETP species. However, it is difficult to make a reliable quantitative estimate of the outcome status of ETP species due to the fact that, although there is documentation on the changes in the relationship Justification Justification between predators and krill in terms of, for instance, penguin densities, species composition and diet changes in certain areas (Trathan et al. 2011, 2012), the effects on the recovery of whales and fur seal populations in the area is still poorly described (Christensen 2006, Nicol et al. 2008, IWC 2010), which makes it difficult to quantitatively estimate outcome status of ETP species with a high degree of certainty. b Information is adequate Information is sufficient Accurate and verifiable to broadly understand to determine whether information is available on the the impact of the fishery the fishery may be a magnitude of all impacts, on ETP species. threat to protection and mortalities and injuries and the recovery of the ETP consequences for the status of

Guidepost Guidepost species. ETP species. Met? Yes Yes Yes

As mentioned above, CCAMLR Scientific Observer Reports record the presence, abundance, interactions with and injuries to ETP species such as marine mammals and seabirds, which provide accurate and verifiable information on direct impacts of the UoC on these species. Indirect impacts have also being studied, through the Nicoll and Douglass (2012) report on the overlap of the krill fishery and ETP species foraging areas and the estimation of the annual consumption of krill by different species in Area 48 by Hewitt et al. (2004). Justification Justification These records and studies are considered to provide accurate information on the magnitude of all impacts and the consequences for the status of ETP species in the region. c Information is adequate Information is sufficient Information is adequate to to support measures to to measure trends and support a comprehensive strategy manage the impacts on support a full strategy to to manage impacts, minimize ETP species. manage impacts on ETP mortality and injury of ETP species. species, and evaluate with a high degree of certainty whether a

Guidepost Guidepost strategy is achieving its objectives.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 110

Relevant information is collected to support the management of fishery impacts on ETP species, including: PI 2.3.3  Information for the development of the management strategy;  Information to assess the effectiveness of the management strategy; and  Information to determine the outcome status of ETP species. Met? Yes Yes Yes

Information gathered through the CCAMLR observer programme is adequate to support this form of strategy and to evaluate its effectiveness. The fact that injuries to or mortalities of ETP species is nil show that the strategy is working. The International Whaling Commission also collects information on ETP species in the area, which is occasionally shared with CCAMLR. This also contributes to the achievement of

Justification Justification SG100.  CCAMLR‐IWC Joint Workshop to Review Input Data for Antarctic Marine Ecosystem Models report. J. Cetacean Res. Manage. 11 (suppl.2), 2010. http://iwc.int/cache/downloads/i5flpo5e6coog0c04g40scg0/SC‐61‐Rep2‐ JCRM11(2).pdf  Christensen L.B. (2006) Marine mammal populations; reconstructing historical abundances at the global scale. Fish Cent Res Rep 14:1‐161.  Government of South Georgia and the South Sandwich Islands. 2013. The South Georgia and South Sandwich Islands Marine Protected Area management plan. http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0. pdf  Hewitt et al. 2004. Options for allocating the precautionary catch limit of krill among small scale management units in the Scotia Sea. CCAMLR Science, Vol. 11 (2004): 81–97. http://www.ccamlr.org/en/system/files/science_journal_papers/05hewitt‐ References etal.pdf  Krafft, B.A., Skaret, G., Krag, L.A., Trathan, P.,Ying, Y. Studies of Antarctic krill, krill predators and trawl gear at South Orkney Islands, 2013. Institute of Marine Research Report, Nr 8‐ 2013.  Lynch, H.J., Naveen, R., Trathan, P., Fagan, W.F. Spatially integrated assessment reveals widespread changes in penguin populations on the Antarctic Peninsula. Ecology, 93(6), 2012, pp. 1367–1377.  Nicoll, R., Douglass, L. 2012. Project report: Mapping krill trawling and predator distribution  Nicol S, Worby A, Leaper R (2008) Changes in the Antarctic sea ice ecosystem: potential effects on krill and baleen whales. Mar Freshwat Res 59:361‐382.  Trathan P.N., Fretwell P.T., Stonehouse B. 2011 First recorded loss of an emperor penguin colony in the recent period of Antarctic regional warming: implications for other colonies. PLos ONE 6:e14738.  Trathan P.N., Ratcliff N., Masden E.A. 2012 Ecological drivers of change at South Georgia: the krill surplus, or climate variability. Ecography 35:983‐993. OVERALL PERFORMANCE INDICATOR SCORE: 95

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 111

Evaluation Table for PI 2.4.1

The fishery does not cause serious or irreversible harm to habitat structure, PI 2.4.1 considered on a regional or bioregional basis, and function Scoring Issue SG 60 SG 80 SG 100 a The fishery is unlikely to The fishery is highly There is evidence that the fishery reduce habitat structure unlikely to reduce habitat is highly unlikely to reduce habitat and function to a point structure and function to structure and function to a point where there would be a point where there where there would be serious or serious or irreversible would be serious or irreversible harm.

Guidepost Guidepost harm. irreversible harm. Met? Yes Yes Yes The krill fishery operates with pelagic gear that does not touch the seafloor, so there is little potential for damaging the benthic ecosystem. The Juvel works with an Egersund krill trawl 350 Omega form of gear. Its characteristics are detailed in observer reports. Observers also record the average fishing depth and also the maximum and minimum fishing depth for each trip. Bottom depths are also recorded. For example, observer records from the first trip in 2013 show a fishing average depth of 37 m with bottom depths ranging from 80 to 917 m. No gear has ever been reported as lost by the Juvel since its first operation in 2009. This Justification Justification has been checked and confirmed in observer reports and also by the client. The fact that the gear does not interact with the seafloor serve as evidence that the fishery is highly unlikely to reduce habitat structure and function. References  CCAMLR Scientific Observer Reports for the years 2010–2014. OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 112

Evaluation Table for PI 2.4.2

There is a strategy in place that is designed to ensure the fishery does not PI 2.4.2 pose a risk of serious or irreversible harm to habitat types Scoring Issue SG 60 SG 80 SG 100 a There are measures in There is a partial strategy There is a strategy in place for place, if necessary, that in place, if necessary, that managing the impact of the are expected to achieve is expected to achieve fishery on habitat types. the Habitat Outcome 80 the Habitat Outcome 80 level of performance. level of performance or

Guidepost Guidepost above. Met? Yes Yes Yes CCAMLR Conservation Measures 21‐03 and 51‐01 restrict the type of fishing gear to be used to pelagic gear only. Depth and fishing depth are recorded by 100% observer coverage of the fishery, and these reports show that the Juvel’s gear has never had any interaction with the seafloor. In accordance with the Antarctic Treaty System, there are different means of spatially managing and protecting the marine environment. Annex V of the Protocol on Environ‐ mental Protection recognises the outstanding values in the Southern Ocean, and offers tools for its spatial management, such as Antarctic Specially Protected Areas (ASPAs) and Antarctic Specially Managed Area (ASMAs). At present the implementation of marine spatial protection and management measures through the Antarctic Treaty Consultative Meeting (ACTM) is coast‐based and small‐scale, but these measures are expected to contribute towards effective, representative and coherent spatial protection of marine biodiversity within the Antarctic Treaty Area. In 2009 CCAMLR designated the South Orkney Islands southern shelf as its first Marine Protected Area, and there have been recent discussions at the 33rd Commission meeting on the establishment of new marine protected areas. Fishing is also restricted around the CCAMLR Ecosystem Monitoring Programme (CEMP) management sites. In order to avoid overlap between the krill fishery and krill predators (especially penguins Justification Justification and fur seals) during their breeding seasons, the South Georgia and South Sandwich Islands Government established a Marine Protected Area in 2012, with a no‐take zone around the islands, a seasonal closure of the fishery for Antarctic krill from 1 November to 31 March, a minimum 700 m depth for trawling, a ban on all bottom fishing deeper than 2250 m, and additional closed areas to protect sensitive benthic fauna and provide refugia for Pata‐ gonian toothfish. Additionally, there are nine benthic closed areas within their territorial waters. According to the Juvel’s 2013 Vessel Monitoring System tracks, the vessel constrains its fishing activity along the Antarctic Peninsula, so there is no interference with any Marine Protected Areas in Area 48. To prevent infractions, vessels position can be tracked by the VMS, and at‐sea inspections are carried out under the auspices of CCAMLR and by South Georgia Fisheries Patrol Vessels. The recent creation of these protected areas is considered to be a reaffirming strategy for managing the impacts of the fishery on habitat types. b The measures are There is some objective Testing supports high confidence considered likely to basis for confidence that that the strategy will work, based work, based on plausible the partial strategy will on information directly about the argument (e.g. general work, based on fishery and/or habitats involved. experience, theory or information directly comparison with similar about the fishery and/or

Guidepost Guidepost fisheries/habitats). habitats involved. Met? Yes Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 113

There is a strategy in place that is designed to ensure the fishery does not PI 2.4.2 pose a risk of serious or irreversible harm to habitat types Scientific opinion has been taken into account on the crucial areas associated with breeding seabird colonies when establishing and locating marine protected areas. In terms of confidence in the strategy, this is supported by the CCAMLR Ecosystem Monitoring Programme (CEMP), the international scientific observer coverage and the rigorous enforcement in the area by patrol vessels. We consider the strategy to be effective, because according to direct information from the fishery there are no interactions with any

Justification Justification habitat type. c There is some evidence There is clear evidence that the that the partial strategy is strategy is being implemented being implemented successfully. successfully. Guidepost Guidepost Met? Yes Yes

Both CCAMLR and the South Georgia and South Sandwich Islands have been involved in the protection of habitats and ecosystems, as the recent establishment of marine protected areas show. Patrol inspections, observer coverage, VMS tracks, along with regulations covering these areas, are evidence of the successful enforcement of this strategy. Justification Justification d There is some evidence that the strategy is achieving its objective. Guidepost Guidepost Met? Yes

VMS tracks and scientific observer reports show just how localized the fishery is when seeking large aggregations of krill. Interactions with the seafloor are avoided by the use of pelagic gear, and exhaustive sampling of the catch by 100% observer coverage is carried out in accordance with CCAMLR observer protocols. This sampling would show up benthic organisms in the catch if any. These factors are considered as evidence that this strategy is achieving its overall

Justification Justification objective.  CCAMLR Conservation Measure 21‐03 (2013). Notifications of intent to participate in a fishery for Euphausia superba. http://www.ccamlr.org/en/measure‐21‐03‐2013  CCAMLR Conservation Measure 51‐01 (2010) regarding Precautionary catch limitations on Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51‐01.pdf  CCAMLR Scientific Observer Cruise Reports for the years 2010–2014.  Government of South Georgia and the South Sandwich Islands. 2013. The South References Georgia and South Sandwich Islands Marine Protected Area management plan. http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0. pdf  IUCN. Antarctic Treaty Consultative Meeting XXXV. Hobart 2012. (IP34) Using ASMAs and ASPAs when necessary to complement CCAMLR MPAs http://www.ats.aq/index_e.htm  Juvel’s 2013 VMS track records.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 114

There is a strategy in place that is designed to ensure the fishery does not PI 2.4.2 pose a risk of serious or irreversible harm to habitat types

OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 115

Evaluation Table for PI 2.4.3

Information is adequate to determine the risk posed to habitat types by the PI 2.4.3 fishery and the effectiveness of the strategy to manage impacts on habitat types Scoring Issue SG 60 SG 80 SG 100 a There is basic The nature, distribution The distribution of habitat types is understanding of the and vulnerability of all known over their range, with types and distribution of main habitat types in the particular attention to the main habitats in the area fishery are known at a occurrence of vulnerable habitat of the fishery. level of detail relevant to types. the scale and intensity of

Guidepost Guidepost the fishery. Met? Yes Yes No There is adequate information to determine the risk posed to habitat types by the krill fishery, because the pelagic gear prevents any impact with the seafloor. Besides, in the past few years, the Juvel has constrained its activity to the coast of the Antarctic Peninsula, which reduces its potential for impacting wider areas. As for terrestrial vulnerable habitat types in the area, their distribution is very well known. For the marine pelagic ecosystem, several oceanographic models have been developed, including the Ocean Circulation and Climate Advanced Modelling project (OCCAM) (Rintoul et al. 2001). Development work in this respect is still being undertaken, with recent discussions at CCAMLR 33rd meeting of the Commission about the establishment of new Justification Justification marine protected areas. Information on the distribution of vulnerable benthic marine ecosystems is still scarce and there is need of increased seafloor mapping, which would improve the knowledge of the habitats where the krill fishery takes place. For this reason SG100 cannot be met. b Information is adequate Sufficient data are The physical impacts of the gear to broadly understand available to allow the on the habitat types have been the nature of the main nature of the impacts of quantified fully. impacts of gear use on the fishery on habitat the main habitats, types to be identified and including spatial overlap there is reliable of habitat with fishing information on the

Guidepost Guidepost gear. spatial extent of interaction, and the timing and location of use of the fishing gear. Met? Yes Yes Yes

As the fishery operates with pelagic gear, no interaction is expected with the seafloor. Such rare impacts on the seafloor could only be considered in case of gear loss, and would be highly localized. International 100% observer coverage serves to fully quantify possible impacts of the gear on habitat types, and observer reports show that the Juvel has never suffered such an event. Justification Justification

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 116

Information is adequate to determine the risk posed to habitat types by the PI 2.4.3 fishery and the effectiveness of the strategy to manage impacts on habitat types c Sufficient data continue Changes in habitat distributions to be collected to detect over time are measured. any increase in risk to habitat (e.g. due to changes in the outcome indicator scores or the operation of the fishery or the effectiveness of

Guidepost Guidepost the measures). Met? Yes No

Terrestrial habitats changes are well studied through CEMP. As for marine habitats, the pelagic ecosystem is monitored by CCAMLR through research surveys using remote sensing, localized sensor arrays, Argo floats and other sensors. The findings are incorporated within ecosystem and ocean current models (such as OCCAM). As regards changes in the benthic habitats, these are not monitored (regardless of the minimal impact that the krill fishery may have on them), so SG100 cannot be met. Justification Justification  GEBCO General Bathymetric Chart of the Oceans. http://www.gebco.net/about_us/news_and_events/arctic_antarctic_mapping_m eeting.html  IBCSO International Bathymetric Chart of the Southern Ocean. http://www.ibcso.org/  http://www.ibcso.org/documents/IBCSO_AGU2012_ePoster.pdf  Arndt, J.E., H. W. Schenke, M. Jakobsson, F. Nitsche, G. Buys, B. Goleby, M. Rebesco, F. Bohoyo, J.K. Hong, J. Black, R. Greku, G. Udintsev, F. Barrios, W. Reynoso‐Peralta, T. Morishita, R. Wigley, "The International Bathymetric Chart of the Southern Ocean (IBCSO) Version 1.0 ‐ A new bathymetric compilation covering circum‐Antarctic waters", Geophysical Research Letters, doi: 10.1002/grl.50413 http://hs.pangaea.de/Maps/bathy/IBCSO_v1/IBCSO_v1_digital_chart_pdfA.pdf  Government of South Georgia and the South Sandwich Islands. 2013. The South Georgia and South Sandwich Islands Marine Protected Area management plan. References http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0. pdf  Rintoul, S.R., Hughes, C. and Olbers, D. 2001 Ocean circulation and climate. Chapter 4.6 The Antarctic Circumpolar Current System. Academic Press ISBN 0‐12‐ 641351‐7  Thorpea,S.E., Heywooda, K., Brandonb, M.A., Stevens, D.P. Journal of Marine Systems 37 (2002) 87 – 105. Variability of the southern Antarctic Circumpolar Current front north of South Georgia. https://ueaeprints.uea.ac.uk/15924/1/DS_30.pdf  Trathan, P.N., Grant, S.M., Siegel, V., Kock K‐H. CCAMLR Science, Vol. 20 (2013): pp 139–151. Precautionary spatial protection to facilitate the scientific study of habitats and communities under ice shelves in the context of recent, rapid, regional climate change. http://www.ccamlr.org/en/system/files/science_journal_papers/Trathan%20et% 20al_0.pdf OVERALL PERFORMANCE INDICATOR SCORE: 85

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 117

Evaluation Table for PI 2.5.1

The fishery does not cause serious or irreversible harm to the key elements PI 2.5.1 of ecosystem structure and function Scoring Issue SG 60 SG 80 SG 100 a The fishery is unlikely to The fishery is highly There is evidence that the fishery disrupt the key elements unlikely to disrupt the is highly unlikely to disrupt the key underlying ecosystem key elements underlying elements underlying ecosystem structure and function to ecosystem structure and structure and function to a point a point where there function to a point where where there would be a serious or would be a serious or there would be a serious irreversible harm.

Guidepost Guidepost irreversible harm. or irreversible harm. Met? Yes Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 118

The fishery does not cause serious or irreversible harm to the key elements PI 2.5.1 of ecosystem structure and function Hewitt et al. (2004) estimated the annual predator demand for krill in Area 48 as orders of magnitude higher than the current fishery take. Nicoll and Douglass (2012) also agree that removals by the krill fishery have been estimated to be orders of magnitude below the demand from predators and the biomass available to both predators and the fishery. Moreover, according to Murphy et al. (2007), some species also seek alternative breeding options in years when krill are relatively scarce. According to Peatman et al. (2011), the probability of stock depletion increases substantially with increased recruitment variability, although in absolute terms it remains negligible. Kinzey et al. (2013) note that the proportional recruitment option in the General Yield Model (GYM) does not appear to be able to model recruitment consistently, and that the precautionary catch limit meeting of CCAMLR criteria relies on the maintenance of a natural mortality of no more than 0.8. The current CCAMLR Euphausia superba TAC for Area 48 is 5.61 million tonnes, with an established precautionary trigger level of 620 000 t. The lower (trigger) limit is intended, inter alia, to ensure that in terms of CCAMLR primary objectives, the impact of the fishery on krill‐dependent predators is minimized. Most of the krill catch in Area 48 is made in Subareas 48.1, 48.2 and 48.3. For the past three years, the Juvel has not fished in subarea 48.3. Trigger level limits for each of those subareas are calculated using the General Yield Model, and were first introduced by CCAMLR in 2009. The fishery must stop or move to a position where a previous notification of fishing activity has been given once the trigger level is reached in any of the subareas or in the total area. The trigger level in subarea 48.1 has been reached in the past (October 2010, June 2013, May 2014), so in those three seasons, subarea 48.1 was closed until the end of the fishing season. Krill tend to be in layers and patches ranging from a few square metres across, through shoals, schools, swarms and up to superswarms covering more than 100 km2, which makes the catch homogeneous. The Juvel’s catch is subject to thorough sampling by the observer

Justification Justification that shows that larvae in the catch are relatively few (<0.5%), and that all are identified and recorded. International observers also record any sighting of ETP species as well as interactions with seabirds and marine mammals. Records show little impacts on these species, with a maximum of 4 Cape petrel fatalities in 2013. There are often numerous birds (varying from 20 to 200 animals) accompanying the vessel, the most common being Cape petrels and snow petrels. As regards marine mammals, interactions and sightings are also recorded in observer reports. Over the past five years, one Weddell seal was entangled with the gear, as it swam onto the trawl net during hauling, but it slipped off on trawl ramp. The seal sustained no injuries and returned to the sea. The Juvel has established a protocol to serve as guidance in order to minimize impacts on these species. It is considered that at the current harvesting rate it is highly unlikely that the fishery would cause serious or irreversible harm to the key elements of ecosystem structure and function. According to the predator–prey fishery model used by Watters et al (2013) to evaluate krill management measures in the Scotia Sea, the risks that krill fishing might negatively impact ecosystem productivity are generally low. In future, however, before increasing krill harvests in Area 48 beyond the trigger level, it will be necessary to verify that parameters such as krill recruitment variability, natural mortality or others as specified by Kinzey et al. (2013) do adequately represent the range of plausible values encompassing krill population biology in the scenarios used for testing management criteria. The tight regulations established by CCAMLR, along with the precautionary catch limits, the estimated needs of predators and the 100% observer coverage serve as evidence that the fishery is highly unlikely to disrupt key elements underlying ecosystem structure and function.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 119

The fishery does not cause serious or irreversible harm to the key elements PI 2.5.1 of ecosystem structure and function  CCAMLR Conservation Measure 51‐01 (2010). Precautionary catch limitations on Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51‐01.pdf  CCAMLR Conservation Measure 51‐07 (2011). Interim distribution of the trigger level in the fishery for Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51‐07.pdf  CCAMLR Scientific Observers Manual (2011). http://www.ccamlr.org/en/system/files/obsman.pdf  Hewitt et al. 2004. CCAMLR Science, Vol. 11 (2004): 81–97. Options for allocating the precautionary catch limit of krill among small scale management units in the Scotia Sea (p. 88, Table 2). http://www.ccamlr.org/en/system/files/science_journal_papers/05hewitt‐ etal.pdf  Kinzey, D., Watters, G. and Reiss, C.S. (2013). Effects of recruitment variability and natural mortality on generalised yield model projections and the CCAMLR decision rules for Antarctic Krill. http://www.ccamlr.org/en/system/files/science_journal_papers/Kinzey%20et%20 al.pdf References  Murphy, E. J., Watkins, J. L., Trathan, P. N., Reid, K.,, Meredith, M. P., Thorpe, S. E., Johnston, N. M., Clarke, A., Tarling, G. A., Collins, M. A., Forcada, J., Shreeve, R. S., Atkinson, A., Korb, R., Whitehouse, M. J., Ward, P., Rodhouse, P. G., Enderlein, P., Hirst, A. G., Martin, A. R., Hill, S. L., Staniland, I. J., Ponmd, D. W., Briggs, D. R., Cunningham, N. J., and Fleming, A. H. 2007. Spatial and temporal operation of the Scotia Sea ecosystem: a review of large‐scale links in a krill centred food web. Philosophical Transactions of the Royal Society of London B. Biological Sciences 362. Doi: 10.1098/rstb.2006.1957.  Nicoll, R., Douglass, L. 2012. Project report: Mapping krill trawling and predator distribution.  http://www.ccamlr.org/en/publications/science_journal/ccamlr‐science‐volume‐ 20  Peatman, T., Clarke, J. M., and Agnew, D. J. 2011. Estimation of management reference points consistent with the catch trigger level for the Antarctic krill fishery in Area 48. WG‐EMM‐11‐17. CCAMLR, Hobart. https://www.ccamlr.org/es/node/65473  Watters, G. M., Hill, S., Hinke, J. T., Matthews J. and Reid, K. 2013. Decision making for ecosystem based management: evaluating options for a krill fishery with an ecosystem dynamics model. Ecological Applications 23: 710–725. http://www.ncbi.nlm.nih.gov/pubmed/23865224

OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 120

Evaluation Table for PI 2.5.2

There are measures in place to ensure the fishery does not pose a risk of PI 2.5.2 serious or irreversible harm to ecosystem structure and function Scoring Issue SG 60 SG 80 SG 100 a There are measures in There is a partial strategy There is a strategy that consists of place, if necessary. in place, if necessary. a plan, in place. Guidepost Guidepost Met? Yes Yes No CCAMLR Conservation Measure 51‐07 establishes precautionary trigger levels associated with the Area 48 subareas, in order to preserve sufficient krill for predators within the ecosystem, and to underpin any recovery from depressed levels. When these levels of catch are reached, the affected subarea is closed until the end of the fishing season, and the fishery is obliged to move to another area. This level of precautionary catch has already been reached in the past (October 2010, June 2013 and May 2014) in subarea 48.1, resulting in that subarea’s immediate closure to fishing operations. This Conservation Justification Justification Measure is considered a partial strategy to prevent the fishery causing serious harm to the ecosystem, but it can neither be considered as a plan nor a full strategy. b The measures take into The partial strategy takes The strategy, which consists of a account potential into account available plan, contains measures to impacts of the fishery on information and is address all main impacts of the key elements of the expected to restrain fishery on the ecosystem, and at ecosystem. impacts of the fishery on least some of these measures are the ecosystem so as to in place. The plan and measures achieve the Ecosystem are based on well‐understood Outcome 80 level of functional relationships between performance. the fishery and the Components and elements of the ecosystem.

Guidepost Guidepost This plan provides for development of a full strategy that restrains impacts on the ecosystem to ensure the fishery does not cause serious or irreversible harm. Met? Yes Yes No

The implementation of Conservation Measure 51‐07 takes into account available information on krill stock biomass estimations, and is expected to restrain impacts of the fishery on the ecosystem, because it sets precautionary catch levels that are expected to prevent local depletion and to preserve sufficient krill resource for predators. CCAMLR keeps an accurate catch data record since 1980 which serves to measure trends on the krill catch. However, it is considered that there should be Small Scale Management Units (SSMUs) in place with defined TACs and trigger levels, which are still not in place. Also, Justification Justification there is no management feedback policy to regulate the ecosystem impacts of fishing activities, because the data collected by the CCAMLR Ecosystem Monitoring Programme are not yet being used to develop Conservation Measures.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 121

There are measures in place to ensure the fishery does not pose a risk of PI 2.5.2 serious or irreversible harm to ecosystem structure and function c The measures are The partial strategy is The measures are considered considered likely to considered likely to work, likely to work based on prior work, based on plausible based on plausible experience, plausible argument or argument (e.g., general argument (e.g., general information directly from the experience, theory or experience, theory or fishery/ecosystems involved. comparison with similar comparison with similar

Guidepost Guidepost fisheries/ecosystems). fisheries/ecosystems). Met? Yes Yes No

CCAMLR Conservation Measure 51‐07 (2011) ensures that the fishery does not harm krill stock status, by establishing precautionary trigger levels in subareas 48.1, 48.2, 48, 3 and 48.4. This measure recognizes that catches up to the trigger levels in areas smaller than the whole of Area 48 should, at the current state of knowledge, not be exceeded. This partial strategy is considered likely to work, because there have already been closures in area 48.1. Monitoring of cumulative catch in each subarea is achieved by the reporting of catch and effort data on a haul‐by‐haul basis, as required for all CCAMLR fisheries. The data thus collected are used to monitor fishery activity, and to quantify catches of target and bycatch species, including incidental catches and any removals of Vulnerable Marine Ecosystems Justification Justification (VME) species. These data are also used to estimate fishery and biological parameters, and to contribute to the assessment of fish stocks. However, there is still need to include in krill management a sufficiently adaptive feedback tool that is flexible enough to respond in a timely manner to changes in the ecosystem in order to consider all these measures as a plan or a strategy. d There is some evidence There is evidence that the that the measures measures are being implemented comprising the partial successfully. strategy are being implemented

Guidepost Guidepost successfully. Met? Yes No

There is evidence that CCAMLR CM 51‐07 is successfully implemented, as area 48.1 has been closed to the fishery up to three times (October 2010, June 2013 and May 2014), when the precautionary trigger level was reached. As mentioned before, this Conservation Measure is considered as a partial strategy, so SG100 cannot be reached yet. Justification Justification  CCAMLR Conservation Measure 51‐07 (2011). Interim distribution of the trigger level in the fishery for Euphausia superba in Statistical Subareas 48.1, 48.2, 48.3 and 48.4. http://www.ccamlr.org/sites/drupal.ccamlr.org/files//51‐07.pdf References  http://www.ccamlr.org/en/wg‐emm‐13/37‐rev‐1  https://www.ccamlr.org/en/wg‐emm‐11/5  http://www.ccamlr.org/en/fisheries/fishery‐monitoring

OVERALL PERFORMANCE INDICATOR SCORE: 80

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 122

Evaluation Table for PI 2.5.3

PI 2.5.3 There is adequate knowledge of the impacts of the fishery on the ecosystem

Scoring Issue SG 60 SG 80 SG 100 a Information is adequate Information is adequate to identify the key to broadly understand elements of the the key elements of the ecosystem (e.g., trophic ecosystem. structure and function, community composition, productivity pattern and

Guidepost Guidepost biodiversity). Met? Yes Yes The Southern Ocean has been studied over many years through different research programmes and there are appropriate developed models which serve to understand its ecosystem. Ecosystem information is collected on a continuous basis through the CEMP areas and also through scientific observer reports. The CCAMLR Working Group on Ecosystem Monitoring and Management (WG‐EMM) collects that information and monitors changes in the ecosystem as well as the effect the fishery may be having on the ecosystem. There are also other institutions such as the International Whaling Commission, the British Antarctic Survey, the Norwegian Institute of Marine Research, the US Antarctic Marine Living Resources Program, the South Georgia and South Sandwich Islands Government and some

Justification Justification other institutions and NGOs which provide even more knowledge of the region’s ecosystem. Besides, several models have been developed which cover krill, the Southern Ocean or smaller area foodweb, the Antarctic ecosystem or the Antarctic physical characteristics. For all the above reasons we consider that information is adequate to broadly understand the key elements of the ecosystem. b Main impacts of the Main impacts of the Main interactions between the fishery on these key fishery on these key fishery and these ecosystem ecosystem elements can ecosystem elements can elements can be inferred from be inferred from existing be inferred from existing existing information, and have information, and have information and some been investigated. not been investigated in have been investigated in

Guidepost Guidepost detail. detail. Met? Yes Yes No

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 123

PI 2.5.3 There is adequate knowledge of the impacts of the fishery on the ecosystem

CEMP data analysed by WG‐EMM serve to identify trends in the monitored parameters by species and site. There is also information on the fish larvae present in krill swarms caught by the fishery, as well as information on the overlap of the fishery with the marine foraging areas of land‐based predators. Unfortunately, information on the predatory requirements of pelagic predators such as baleen whales is not that comprehensive. Also, for some authors functional relationships between the fishery and elements of the ecosystem are not yet sufficiently well understood (e.g. Kawaguchi et al. 2006). According to CCAMLR WG‐EMM (2007), over the next two or three decades, and due to a combination of historical exploitation and the effects of climate change, significant and rapid changes could be expected in the Scotia Sea ecosystem. Consequences on the krill population due to changes in the Southern Ocean’s parameters such as temperature, acidity or sea‐ice coverage are not yet fully understood. As noted in the WG‐EMM 2003 report, at current harvesting levels, the existing designated CEMPs and the data available

Justification Justification to them are not sufficient to distinguish whether the changes in the ecosystem are attributable to the harvesting of commercial species or to environmental variability, whether it be physical or biological. The team considers that main impacts of the fishery on key ecosystem elements can be inferred from existing information and that some of them have been investigated in detail, thanks to data collected in logbooks, VMS track records, observer reports and the CEMP. However, as not all of these ecosystem elements have been investigated (for instance, pelagic predator needs, the impact of climate change on krill recruitment and survival), SG100 is not met. c The main functions of the The impacts of the fishery on Components (i.e., target, target, Bycatch, Retained and ETP Bycatch, Retained and species are identified and the ETP species and Habitats) main functions of these in the ecosystem are Components in the ecosystem are

Guidepost Guidepost known. understood. Met? Yes Yes

Impacts of the fishery on target, bycatch, retained and ETP species are not only identified but also quantified through observer reports. The main functions of these components in the ecosystem are understood and have been studied through a wide range of models, including:  models exploring specific aspects of krill biology (Hofmann and Hùsrevõglu 2003; Murphy et al. 2004),  multispecies population models (May et al. 1979; Murphy, 1995),  single‐species population projection models to quantify regional catch limits (Constable et al. 2000),  spatial single‐species models (e.g. Marin and Delgado 2001), Justification Justification  mass‐balance regional foodweb models such as EwE (Cornejo‐Donoso and Antezana 2008),  spatial multispecies operating model (SMOM) of krill–predator fishery dynamics (Plagányi and Butterworth 2012),  models of krill transport at the maximum advection rate indicated by the Ocean Circulation and Climate Advanced Modelling Project, OCCAM (Rintoul et al. 2001).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 124

PI 2.5.3 There is adequate knowledge of the impacts of the fishery on the ecosystem d Sufficient information is Sufficient information is available available on the impacts on the impacts of the fishery on of the fishery on these the Components and elements to Components to allow allow the main consequences for some of the main the ecosystem to be inferred. consequences for the

Guidepost Guidepost ecosystem to be inferred. Met? Yes Yes

CCAMLR website (http://www.ccamlr.org/en/) offers information derived from observer reports, CEMP research output, WG‐EMM reports and ecosystem studies. Other institutions such as the International Whaling Commission, British Antarctic Survey, Norwegian Institute of Marine Research, US Antarctic Marine Living Resources Program, South Georgia and South Sandwich Islands Government and other institutions and NGOs also provide information on Antarctic research projects. All this information is sufficient to parameterize the ecosystem models described above, and allow for the inferring of the

Justification Justification main consequences of the fishery for the ecosystem. e Sufficient data continue Information is sufficient to to be collected to detect support the development of any increase in risk level strategies to manage ecosystem (e.g., due to changes in impacts. the outcome indicator scores or the operation

Guidepost Guidepost of the fishery or the effectiveness of the measures). Met? Yes No

Sufficient data continue to be collected to detect any increase in risk levels to both target stock and the ecosystem in which is found. These data are collected by different groups and institutions, and can be obtained inter alia from logbooks, VMS track records, observer reports and the CEMP. However, there is need to continue with this collection, as there are still uncertainties associated with the Southern Ocean, such as krill stock levels (because the last fully synoptic survey was carried out in 2000), the pelagic predator demand for krill, the possible impacts of climate change and ocean acidification, how ecosystems would operate in the absence of fishing, or how oceanographic dynamics, bathymetry and Justification Justification krill movements combine to explain the biodiversity of these areas. The clarification of these uncertainties would help to support the development of strategies to manage ecosystem impacts better.  www.ccamlr.org  CCAMLR Scientific Observer Cruise Reports for the years 2010–2014.  CCAMLR WG‐EMM 2003 report CEMP Review (pages 143 to 148) (Appendix D, paragraph 87)  http://www.ccamlr.org/es/system/files/s‐sc‐xxii‐a4.pdf  CCAMLR WG‐EMM 2007 Report. http://www.ccamlr.org/es/system/files/s‐sc‐ References xxvi‐a4.pdf  http://www.ccamlr.org/en/science/ccamlr‐ecosystem‐monitoring‐program‐cemp  http://www.ccamlr.org/en/science/working‐group‐ecosystem‐monitoring‐and‐ management‐wg‐emm  Constable A.J., de la Mare, W.K., Agnew, D.J., Everson, I. and Miller, D. 2000 Managing fisheries to conserve the Antarctic marine ecosystem: practical implementation of the Convention on the Conservation of Antarctic Marine Living

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 125

PI 2.5.3 There is adequate knowledge of the impacts of the fishery on the ecosystem

Resources. Ices J. Mar. Sci. 57, 778‐791. (doi:10.1006/jmsc.2000.0725)  Cornejo‐Donoso, J., Antezana, T., Preliminary trophic model of the Antarctic Peninsula Ecosystem (Sub‐area CCAMLR 48.1) (EwE model) Ecological Modelling 218 (2008) 1–17. http://www.ecopath.org/node/195  Hofmann, E.E and Hùsrevõglu, Y. S. 2003. A circumpolar modeling study of habitat control of Antarctic krill (Euphausia superba) reproductive success. Deep Sea Res. Part II Topical Stud. Oceanogr. 50, 3121‐3142. doi:10.1016/j.dsr2.2003.07.012  Marín, V. H. and Delgado, L.E. 2001 A spatially explicit model of the Antarctic krill fishery off the South Shetland Islands. Ecological Applications 11(4): 1235‐1248.  May, R.M. 1979. Ecological interactions in the Southern Ocean. Nature 277, 86‐89. (doi:10.1038/277086a0)  Murphy, E.J. 1995 Spatial structure of the Southern Ocean ecosystem predator‐ prey linkages in the Southern Ocean food‐webs. J. Anim. Ecol. 64, 333‐347. (doi:10.2307/5895)  Murphy, E. J., Thorpe, S.E. Watkins, J.L., and Hewitt, R. 2004. Modelling the krill transport pathways in the Scotia Sea: Spatial and environmental connections generating the seasonal distribution of krill. Deep Sea Res. Part II Topical Stud. Oceanogr. 51, 1435‐1456. (doi:10.1016/j.dsr2.2004.06.019)  Plagányi, E.E. and. Butterworth, D.S. 2012. The Scotia Sea krill fishery and its possible impacts on dependent predators: modelling localized depletion of prey. Ecological Applications 22:748–761. http://www.esajournals.org/doi/abs/10.1890/11‐0441.1  Rintoul, S.R., Hughes, C. and Olbers, D. 2001 Ocean circulation and climate. Chapter 4.6 The Antarctic Circumpolar Current System. Academic Press ISBN 0‐12‐ 641351‐7 OVERALL PERFORMANCE INDICATOR SCORE: 90

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 126

Evaluation Table for PI 3.1.1

The management system exists within an appropriate legal and/or customary framework which ensures that it:  Is capable of delivering sustainable fisheries in accordance with MSC PI 3.1.1 Principles 1 and 2; and  Observes the legal rights created explicitly or established by custom of people dependent on fishing for food or livelihood; and  Incorporates an appropriate dispute resolution framework. Scoring Issue SG 60 SG 80 SG 100 a There is an effective There is an effective There is an effective national legal national legal system national legal system and system and binding procedures and a framework for organised and effective governing cooperation with other cooperation with other cooperation with other parties which delivers parties, where parties, where necessary, management outcomes consistent necessary, to deliver to deliver management with MSC Principles 1 and 2.

Guidepost Guidepost management outcomes outcomes consistent with consistent with MSC MSC Principles 1 and 2. Principles 1 and 2 Met? Yes Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 127

The management system exists within an appropriate legal and/or customary framework which ensures that it:  Is capable of delivering sustainable fisheries in accordance with MSC PI 3.1.1 Principles 1 and 2; and  Observes the legal rights created explicitly or established by custom of people dependent on fishing for food or livelihood; and  Incorporates an appropriate dispute resolution framework. The fishery is managed mainly by CCAMLR, in interaction with the Norwegian Ministry/‐ Directorate of Fisheries and the Government of South Georgia and the South Sandwich Islands (GSGSSI). CCAMLR coordinates scientific research and observer programmes, establishes TAC and distributes quotas between subareas. The Norwegian Ministry/‐ Directorate of Fisheries issues fishery permits and performs quota control of the client vessels. GSGSSI issues permits for the vessels in the SGSSI Maritime Zone. CCAMLR determines the regulatory framework applied to the management of each fishery in the Convention Area, including catch limits and seasonal or area closures and measures aimed at minimizing potential impacts of fishing activities on non‐target species and the ecosystem. The Standing Committee on Implementation and Compliance, subordinate to the Commission, provides it with information, advice, recommendations on fishery monitoring and compliance. The Scientific Committee provides the Commission with the best available scientific information on harvesting levels and other management issues. In turn, the Commission is obliged by the Convention to take full account of the recommendations and advice of the Scientific Committee in making its decisions. The Scientific Committee takes into account the outcomes of research from national programmes of CCAMLR members. In addition CCAMLR has established a number of programmes to collect the data required for the effective management of the Southern Ocean, including fisheries monitoring, scientific observers on fishing vessels and ecosystem monitoring. Norway has a well‐established system for fisheries management, which has evolved over

Justification Justification more than a century and is now codified in the 2008 Marine Resources Act. The Act provides for a formal system of cooperation between regulatory bodies of governance, such as the Ministry of Fisheries and Coastal Affairs, the Directorate of Fisheries and the Coast Guard, and further for cooperation between management authorities and scientific research institutes, primarily the Institute of Marine Research. The 2008 Integrated Management Plan for the Norwegian Sea provides for cooperation between different sector authorities, such as the Ministry of Fisheries and Coastal Affairs and the Ministry of Environment. GSGSSI is involved in the licensing of vessels that fish in the South Georgia Maritime Zone, catch monitoring at King Edward Point in South Georgia and at‐sea surveillance in the Maritime Zone. A Marine Protected Area was introduced in 2012 and reinforced in May 2013, including a no‐fish zone within 12 nautical miles of the coast. The national and international legal documents refer to and are in compliance with relevant international agreements, such as the 1982 Law of the Sea Convention and the 1995 Fish Stocks Agreement. Norwegian and South Georgia fishery authorities liaise closely with CCAMLR. The system is considered to be effective insofar as it constitutes a coherent set of rule‐making practices at national and international level.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 128

The management system exists within an appropriate legal and/or customary framework which ensures that it:  Is capable of delivering sustainable fisheries in accordance with MSC PI 3.1.1 Principles 1 and 2; and  Observes the legal rights created explicitly or established by custom of people dependent on fishing for food or livelihood; and  Incorporates an appropriate dispute resolution framework. b The management system The management system The management system incorporates or is incorporates or is subject incorporates or subject by law to a subject by law to a by law to a transparent transparent mechanism for the mechanism for the mechanism for the resolution of legal disputes that is resolution of legal resolution of legal appropriate to the context of the disputes arising within disputes which is fishery and has been tested and the system. considered to be proven to be effective.

Guidepost Guidepost effective in dealing with most issues and that is appropriate to the context of the fishery. Met? Yes Yes No

Any issues of contention among the CCAMLR member states can be raised and discussed at the meetings of the Commission and subordinate bodies. These processes are transparent and subject to scrutiny by all member states. The system is considered to be effective insofar as no major disputes have emerged, although the assessment team has not been provided with evidence that this has been tested and proven to be the case. At a national level in Norway, there is an effective, transparent dispute‐resolution system in place, with fishermen able to take their case to court if they do not accept the rationale

Justification Justification behind an infringement accusation by enforcement authorities, or the fees levied against them. Verdicts at the lower court levels can be appealed to higher levels. There are instances from recent years that management authorities have lost cases against fishermen and accepted the verdict, which is a clear demonstration that the system works. d The management system The management system The management system has a has a mechanism to has a mechanism to mechanism to formally commit to generally respect the observe the legal rights the legal rights created explicitly legal rights created created explicitly or or established by custom of explicitly or established established by custom of people dependent on fishing for by custom of people people dependent on food and livelihood in a manner dependent on fishing for fishing for food or consistent with the objectives of

Guidepost Guidepost food or livelihood in a livelihood in a manner MSC Principles 1 and 2. manner consistent with consistent with the the objectives of MSC objectives of MSC Principles 1 and 2. Principles 1 and 2. Met? Yes Yes Yes

Within the fishery there are no indigenous people dependent upon fishing in waters managed by CCAMLR, and no indigenous inhabitants of SGSSI. According to MSC Review and Report on Compliance with the Scheme Requirements, this scoring issue still needs to be scored and might meet the SG100 level. The team interprets this to the effect that a SG100 score can be achieved if the fishery has no negative impact on people dependent on

Justification Justification fishing for food or livelihood. This is the case in the present fishery.

» CCAMLR Annual Fishery Reports References » CCAMLR website » Convention on the Conservation of Antarctic Marine Living Resources, 1982

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 129

The management system exists within an appropriate legal and/or customary framework which ensures that it:  Is capable of delivering sustainable fisheries in accordance with MSC PI 3.1.1 Principles 1 and 2; and  Observes the legal rights created explicitly or established by custom of people dependent on fishing for food or livelihood; and  Incorporates an appropriate dispute resolution framework. » Government of South Georgia and the South Sandwich Islands. 2013. The South Georgia and South Sandwich Islands Marine Protected Area management plan. http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0.pd f » Interviews with representatives of GSGSSI, the Norwegian Ministry of Foreign Affairs and the Norwegian Directorate of Fisheries » Norwegian Ministry of Fisheries. 2008. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act). » Norwegian Ministry of Fisheries. 2012. Fiskeriavtalane Noreg har inngått med andre land for 2013 og fisket etter avtalane i 2011 og 2012 (The Fishery Agreements Norway had Concluded with Other Countries for 2013 and Fishery according to the Agreements in 2011 and 2012), Meld. St. 40 (2012‐2013) (White Paper No. 40 2012‐ 2013). OVERALL PERFORMANCE INDICATOR SCORE: 90

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 130

Evaluation Table for PI 3.1.2

The management system has effective consultation processes that are open to interested and affected parties. PI 3.1.2 The roles and responsibilities of organisations and individuals who are involved in the management process are clear and understood by all relevant parties Scoring Issue SG 60 SG 80 SG 100 a Organisations and Organisations and Organisations and individuals individuals involved in individuals involved in involved in the management the management the management process process have been identified. process have been have been identified. Functions, roles and identified. Functions, Functions, roles and responsibilities are explicitly roles and responsibilities responsibilities are defined and well understood for are generally explicitly defined and all areas of responsibility and Guidepost Guidepost understood. well understood for key interaction. areas of responsibility and interaction. Met? Yes Yes Yes Overall management lines and the responsibilities of different management bodies are clear. The main responsibility for developing and promulgating the management plan for the fishery within Area 48 lies with CCAMLR, within the framework of the Antarctic Treaty. Article XVII of the Convention details the role of the Executive Secretary of CCAMLR and any other staff that they may need to appoint. Scientists appointed by CCAMLR members meet annually in Working Groups to undertake stock assessments and prepare scientific advice for the Commission. This scientific advice is reviewed annually by the CCAMLR Scientific Committee, which provides management advice to the Commission. Management policies and procedures are implemented through Conservation Measures and Resolutions. The CAMLR Convention sets out the terms under which observers can attend and participate in its statutory meetings. Within the CCAMLR Secretariat, the roles for the management of the different aspects of the fishery (compliance, data, observers,

Justification Justification etc.) are well defined and operate in a clear and efficient manner. At a national level in Norway, the roles, functions and responsibilities of the various actors, including management authorities, the Institute of Marine Research, NGOs and fisher‐ men’s organizations are clearly defined in long‐standing practice and are now codified in the Marine Resources Act. Within the SGSSI Maritime Zone, the only relevant actor is GSGSSI, which is responsible for licensing and enforcement. Their role is clearly defined and well understood by participants in the fishery. b The management system The management system The management system includes includes consultation includes consultation consultation processes that processes that obtain processes that regularly regularly seek and accept relevant relevant information seek and accept relevant information, including local from the main affected information, including knowledge. The management parties, including local local knowledge. The system demonstrates knowledge, to inform management system consideration of the information Guidepost Guidepost the management demonstrates and explains how it is used or not system. consideration of the used. information obtained. Met? Yes Yes No

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 131

The management system has effective consultation processes that are open to interested and affected parties. PI 3.1.2 The roles and responsibilities of organisations and individuals who are involved in the management process are clear and understood by all relevant parties As follows from 3.1.1 above, overall management responsibility for Antarctic krill rests with CCAMLR. Member States must comply with all regulations and requirements set (as Conservation Measures) and subsequently license their own flagged vessels. All decisions on Conservation Measures and other resolutions are made by consensus. The NGO Antarctic and Southern Ocean Coalition (ASOC) has been actively involved in marine management in the Antarctic since the establishment of CCAMLR and was given observer status in 1991. ASOC is also a key partner to the Antarctic Krill Conservation Project, which is an international effort managed by the Pew Foundation, to secure from CCAMLR an ecosystem‐based fisheries management programme for krill which is highly precautionary, scientifically based and protects the unique environment of the southern polar region.

Justification Justification Olympic is part of the Association of Responsible Krill Harvesting Companies (ARK), which has, among other thing, organized meetings between industry, science, NGOs and management authorities prior to sessions in CCAMLR. The management system demonstrates consideration of the information, but the assessment team has not been provided with documentation that consistently explains how the information is or is not used. c The consultation process The consultation process provides provides opportunity for opportunity and encouragement all interested and for all interested and affected affected parties to be parties to be involved, and involved. facilitates their effective

Guidepost Guidepost engagement. Met? Yes Yes

The process followed includes an open forum for dialogue, and encourages transparency wherever possible. As follows from 3.1.2 a) above, ASOC has observer status in CCAMLR and is also a key partner to the Antarctic Krill Conservation Project. ARK has become a platform for discussion among industry, science, NGOs and management authorities involved in the management of krill. At a national level in Norway, WWF is actively consulted on krill issues by Norwegian fisheries management authorities. For instance, the Ministry of Foreign Affairs invites

Justification Justification stakeholders, including WWF, to meetings before CCAMLR meetings in order to discuss relevant issues, including for the management of krill. WWF has been invited to become part of the Norwegian delegation to CCAMLR, but has chosen to remain an independent actor. » ARK website » ASOC website » CCAMLR website » Convention on the Conservation of Antarctic Marine Living Resources, 1982 » Interviews with representatives of the Institute of Marine Research, the Norwegian References Ministry of Foreign Affairs, the Norwegian Directorate of Fisheries and WWF » Norwegian Ministry of Fisheries. 2008. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act) » Norwegian Ministry of Fisheries. 2012. Fiskeriavtalane Noreg har inngått med andre land for 2013 og fisket etter avtalane i 2011 og 2012 (The Fishery Agreements Norway had Concluded with Other Countries for 2013 and Fishery according to the Agreements in 2011 and 2012), Meld. St. 40 (2012‐2013) (White Paper No. 40 2012‐

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 132

The management system has effective consultation processes that are open to interested and affected parties. PI 3.1.2 The roles and responsibilities of organisations and individuals who are involved in the management process are clear and understood by all relevant parties 2013)

OVERALL PERFORMANCE INDICATOR SCORE: 95

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 133

Evaluation Table for PI 3.1.3

The management policy has clear long-term objectives to guide decision- PI 3.1.3 making that are consistent with MSC Principles and Criteria, and incorporates the precautionary approach Scoring Issue SG 60 SG 80 SG 100 a Long‐term objectives to Clear long‐term Clear long‐term objectives that guide decision‐making, objectives that guide guide decision‐making, consistent

consistent with the MSC decision‐making, with MSC Principles and Criteria Principles and Criteria consistent with MSC and the precautionary approach, and the precautionary Principles and Criteria are explicit within and required by approach, are implicit and the precautionary management policy. Guidepost within management approach are explicit policy within management policy. Met? Yes Yes Yes All CCAMLR fisheries are managed within a precautionary and ecosystem approach, as defined by the FAO in its Code of Conduct for Responsible Fisheries, and are consistent with MSC Principles and Criteria.

At a national level in Norway, the 2008 Marine Resources Act, which covers all living marine resources, requires that Norwegian fisheries management be guided by the precautionary approach and by an ecosystem approach that takes into account habitats and biodiversity. Justification The SGSSI Marine Protected Area management plan is designed to ensure the protection and conservation of the region’s marine life, whilst allowing sustainable and carefully regulated fisheries, which is considered to be consistent with MSC Principles and Criteria.

» Convention on the Conservation of Antarctic Marine Living Resources, 1982 » FAO Code of Conduct for Responsible Fisheries, 1995 » Government of South Georgia and the South Sandwich Islands. 2013. The South References Georgia and South Sandwich Islands Marine Protected Area management plan. http://www.sgisland.gs/download/MPA/MPA%20Management%20Plan%20v2.0.pd f » Norwegian Ministry of Fisheries. 2012. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act)

OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 134

Evaluation Table for PI 3.1.4

The management system provides economic and social incentives for sustainable fishing PI 3.1.4 and does not operate with subsidies that contribute to unsustainable fishing Scoring Issue SG 60 SG 80 SG 100 a The management system The management system The management system provides provides for incentives provides for incentives for incentives that are consistent

that are consistent with that are consistent with with achieving the outcomes achieving the outcomes achieving the outcomes expressed by MSC Principles 1 and expressed by MSC expressed by MSC 2, and explicitly considers Principles 1 and 2. Principles 1 and 2, and incentives in a regular review of Guidepost seeks to ensure that management policy or procedures perverse incentives do to ensure they do not contribute not arise. to unsustainable fishing practices. Met? Yes Yes No There is clear evidence that the management system in place is seeking to ensure that negative incentives do not arise. The client receives no national subsidy, and the CCAMLR system has no capacity to provide subsidies of any sort. Costs related to the operation and management of the Commission are fully recovered from member states through their membership fees. Nor does the GSGSSI provide subsidies of any sort, recovering their fisheries management operational costs from licensed vessels through the charging of license fees. There is an implicit incentive to meet national (Norway), CCAMLR and GSGSSI requirements in terms of licensing, observer requirements and data reporting, to ensure Justification that Norwegian fishing operations are well placed to ensure future licensing in the fishery. Although the management system seeks to ensure that perverse incentives do not arise, the assessment team has not been provided with documentation proving that incentives are explicitly considered in a regular review of management policy.

» Interviews with representatives of GSGSSI, the Institute of Marine Research and the References Norwegian Directorate of Fisheries and WWF

OVERALL PERFORMANCE INDICATOR SCORE: 80

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 135

Evaluation Table for PI 3.2.1

The fishery has clear, specific objectives designed to achieve the outcomes PI 3.2.1 expressed by MSC’s Principles 1 and 2 Scoring Issue SG 60 SG 80 SG 100 a Objectives, which are Short and long‐term Well defined and measurable broadly consistent with objectives, which are short and long‐term objectives, achieving the outcomes consistent with achieving which are demonstrably expressed by MSC’s the outcomes expressed consistent with achieving the Principles 1 and 2, are by MSC’s Principles 1 and outcomes expressed by MSC’s implicit within the 2, are explicit within the Principles 1 and 2, are explicit Guidepost Guidepost fishery’s management fishery’s management within the fishery’s management system system. system. Met? Yes Yes Partly The three main objectives of the CCAMLR management system are (Article 2 of the Convention): a) prevention of decrease in the size of any harvested population to levels below those which ensure its stable recruitment; for this purpose its size should not be allowed to fall below a level close to that which ensures the greatest net annual increment; b) maintenance of the ecological relationships between harvested, dependent and related populations of Antarctic marine living resources and the restoration of depleted populations to the levels defined in sub‐paragraph a) above; and c) prevention of changes or minimization of the risk of changes in the marine ecosystem which are not potentially reversible over two or three decades, taking into account the state of available knowledge of the direct and indirect impact of harvesting, the effect of the introduction of alien species, the effects of associated activities on the marine ecosystem and of the effects of environmental changes, with the aim of making possible the sustained conservation of Antarctic marine living resources. The aims of these three objectives mirror and preceded the establishment of the aims of MSC Principles 1 and 2. CCAMLR‘s more specific, short‐ and long‐term strategy for achieving these objectives is reflected in Conservation Measure 51‐01 (2010). A precautionary krill catch limit of 5.61 million tonnes is set for Area 48, based on the Justification Justification potential yield estimate. This is well above the current catch and will allow for expansion. However, a "catch trigger" (620 000 t) is set not to be exceeded until a procedure for division of the overall catch limit into smaller management units has been established, based on advice from the Scientific Committee. The objective of this division is to avoid possible unacceptable concentration of catch within the foraging areas of vulnerable predators. Although the trigger level is close to the highest global annual catch to date, it is significantly more than the largest annual catch to date in Area 48. Hence, well defined and measurable long‐ and short‐term objectives are in place for the fishery, demonstrably consistent with achieving the outcomes of P1. These are explicit insofar as they are well‐established, understood and applied by users within the fishery and considered durable and unambiguous (cf. CR CB4.1.3). However, they are less well defined and measurable for P2.

» CCAMLR website » Convention on the Conservation of Antarctic Marine Living Resources, 1982 » Interviews with representatives of the Norwegian Ministry of Foreign Affairs and References the Norwegian Directorate of Fisheries » Norwegian Ministry of Fisheries. 2008. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act) » Norwegian Ministry of Fisheries. 2012. Fiskeriavtalane Noreg har inngått med andre land for 2013 og fisket etter avtalane i 2011 og 2012 (The Fishery Agreements

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 136

The fishery has clear, specific objectives designed to achieve the outcomes PI 3.2.1 expressed by MSC’s Principles 1 and 2 Norway had Concluded with Other Countries for 2013 and Fishery according to the Agreements in 2011 and 2012), Meld. St. 40 (2012‐2013) (White Paper No. 40 2012‐ 2013)

OVERALL PERFORMANCE INDICATOR SCORE: 90

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 137

Evaluation Table for PI 3.2.2

The fishery-specific management system includes effective decision-making processes that result in measures and strategies to achieve the objectives, PI 3.2.2 and has an appropriate approach to actual disputes in the fishery under assessment. Scoring Issue SG 60 SG 80 SG 100 a There are some decision‐ There are established making processes in decision‐making place that result in processes that result in measures and strategies measures and strategies to achieve the fishery‐ to achieve the fishery‐

Guidepost Guidepost specific objectives. specific objectives. Met? Yes Yes CCAMLR has well established decision‐making processes. They allow for stakeholder input and clear scientific analysis of the data available within the Working Groups and Scientific Committee, and they result in conservation measures and fisheries strategies designed to achieve their short‐ and long‐term fishery‐specific objectives. Justification Justification b Decision‐making Decision‐making Decision‐making processes processes respond to processes respond to respond to all issues identified in serious issues identified serious and other relevant research, monitoring, in relevant research, important issues evaluation and consultation, in a monitoring, evaluation identified in relevant transparent, timely and adaptive and consultation, in a research, monitoring, manner and take account of the transparent, timely and evaluation and wider implications of decisions. adaptive manner and consultation, in a Guidepost Guidepost take some account of transparent, timely and the wider implications of adaptive manner and decisions. take account of the wider implications of decisions. Met? Yes Yes No

Generally, fisheries‐specific issues identified in relevant research are included in transparent decision‐making processes within the Working Groups and the Scientific Committee, as appropriate. Where and when necessary, modifications are made by these and by the Norwegian Directorate of Fisheries and GSGSSI to the monitoring and evaluation of the fisheries (through modifications to the complex data‐recording systems and observer logbooks). A clear example of the well‐functioning responsiveness of the management system is its ability to halt the fishery within a subarea once the subarea’s

Justification Justification proportion of the “trigger level” has been caught. There is no clear evidence, however, that all issues identified in relevant research, monitoring, evaluation and consultation are responded to in the appropriate way. c Decision‐making processes use the precautionary approach and are based on best available information. Guidepost Guidepost Met? Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 138

The fishery-specific management system includes effective decision-making processes that result in measures and strategies to achieve the objectives, PI 3.2.2 and has an appropriate approach to actual disputes in the fishery under assessment. CCAMLR decision‐making is based on the precautionary approach (see 3.1.3) and the best available information by national experts working closely together in Working Groups, the Scientific Committee and the Commission. Justification Justification d Some information on Information on fishery Formal reporting to all interested fishery performance and performance and stakeholders provides management action is management action is comprehensive information on generally available on available on request, and fishery performance and request to stakeholders. explanations are management actions and provided for any actions describes how the management or lack of action system responded to findings and associated with findings relevant recommendations

Guidepost Guidepost and relevant emerging from research, recommendations monitoring, evaluation and review emerging from research, activity. monitoring, evaluation and review activity. Met? Yes Yes Yes

Formal reporting to all interested stakeholders, primarily through CCAMLR reports and website, describes how the management system responded to findings and relevant recommendations emerging from research, monitoring, evaluation and review activity. The whole CCAMLR process is based on dialogue, stakeholder involvement and formal reporting. The same is true for the Norwegian fisheries management system, which in addition to written reports and websites extensively uses informal as well as formal meetings open to all interested stakeholders to describe how available information is

Justification Justification responded to. The role of GSGSSI is the management of the fisheries in question is more limited, but they also publish information about licensing and enforcement matters on their website and in written reports. e Although the The management system The management system or management authority or fishery is attempting fishery acts proactively to avoid or fishery may be subject to comply in a timely legal disputes or rapidly to continuing court fashion with judicial implements judicial decisions challenges, it is not decisions arising from any arising from legal challenges. indicating a disrespect or legal challenges. defiance of the law by repeatedly violating the Guidepost Guidepost same law or regulation necessary for the sustainability for the fishery. Met? Yes Yes Yes

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 139

The fishery-specific management system includes effective decision-making processes that result in measures and strategies to achieve the objectives, PI 3.2.2 and has an appropriate approach to actual disputes in the fishery under assessment. The management system works proactively – within CCAMLR and its Working Groups – to settle any disagreement outside the legal system. There are no signs that the management system does not rapidly implement judicial decisions arising from legal challenges, neither from Norway nor SGSSI. As mentioned in 3.1.1 b), there are instances in Norway that management authorities have lost cases against fishermen and accepted the verdict, which is a clear demonstration that the system works. However, there are no instances of court

Justification Justification cases in the fishery‐specific management system. » ASOC website » CCAMLR Annual Fishery Reports » CCAMLR website References » Convention on the Conservation of Antarctic Marine Living Resources, 1982 » Interviews with representatives of GSGSSI, the Institute of Marine Research, the Norwegian Directorate of Fisheries and WWF

OVERALL PERFORMANCE INDICATOR SCORE: 95

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 140

Evaluation Table for PI 3.2.3

Monitoring, control and surveillance mechanisms ensure the fishery’s PI 3.2.3 management measures are enforced and complied with Scoring Issue SG 60 SG 80 SG 100 a Monitoring, control and A monitoring, control and A comprehensive monitoring, surveillance mechanisms surveillance system has control and surveillance system exist, are implemented been implemented in the has been implemented in the in the fishery under fishery under assessment fishery under assessment and has assessment and there is and has demonstrated an demonstrated a consistent ability a reasonable expectation ability to enforce relevant to enforce relevant management that they are effective. management measures, measures, strategies and/or rules.

Guidepost Guidepost strategies and/or rules. Met? Yes Yes Yes CCAMLR provides a clear and comprehensive monitoring system and control framework for Antarctic fisheries. Surveillance of CCAMLR fisheries is undertaken by Member States and incorporates the CCAMLR observer scheme. For the client fishery, enforcement is mainly taken care of by the Norwegian Directorate of Fisheries, which has demonstrated a consistent ability to enforce relevant regulations. Vessels are licensed on an annual basis and report catches from each haul through their electronic logbooks. In order to receive a license for the Antarctic krill fishery, Norwegian vessels are obliged to have an observer on board at all times. When entering the South Georgia Maritime Zone, vessels need to apply Justification Justification for a licence and pay a fee. All vessels are inspected by the South Georgia administration at King Edward Point before they are allowed to start fishing. They have to report catches on a daily basis and are also inspected by a patrol vessel during fishing operations. b Sanctions to deal with Sanctions to deal with Sanctions to deal with non‐ non‐compliance exist non‐compliance exist, are compliance exist, are consistently and there is some consistently applied and applied and demonstrably provide evidence that they are thought to provide effective deterrence. applied. effective deterrence. Guidepost Guidepost Met? Yes Yes Yes

The implementation of sanctions to deal with non‐compliance is an issue for Member States, either through flag state control (here: Norway), or, in the case of South Georgia through GSGSSI, coastal state jurisdiction over the Maritime Zone. The Norwegian enforcement agencies use a graded sanctioning system, with sanctions ranging from oral warnings, written warnings and administrative fines to formal prosecution. If the fishers do not accept the fines issued by the enforcement or prosecution authority, the case goes to court. The system has demonstrated a consistent ability over time to provide effective deterrence. There are no instances of infringements by the client vessels.

Justification Justification Sanctions within the South Georgia Maritime Zone are applied at a level appropriate for deterring IUU fishing. No instances of non‐compliance by the client vessels exist there either. Moreover, Norwegian interest in the future of the Antarctic krill fishery ensures that, as for all other Norwegian fisheries interests, all regulations on the fishery now, and almost certainly all that may be applied in future, are closely monitored and adhered to.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 141

Monitoring, control and surveillance mechanisms ensure the fishery’s PI 3.2.3 management measures are enforced and complied with c Fishers are generally Some evidence exists to There is a high degree of thought to comply with demonstrate fishers confidence that fishers comply the management system comply with the with the management system for the fishery under management system under assessment, including, assessment, including, under assessment, providing information of when required, including, when required, importance to the effective providing information of providing information of management of the fishery. Guidepost Guidepost importance to the importance to the effective management of effective management of the fishery. the fishery. Met? Yes Yes Yes

Fishers seeking certification comply with the management system, providing information on the fishery prior to (Notification of Intent) and during fishing (C1, observer data catch reports), at the levels defined by CCAMLR, the Norwegian Directorate of Fisheries and GSGSSI to provide effective management of the fishery. Justification Justification d There is no evidence of systematic non‐ compliance. Guidepost Guidepost Met? Yes

There is no evidence of systematic, indeed any, non‐compliance. The Norwegian enforcement system is generally considered to work effectively, which strengthens the argument that the lack of systematic non‐compliance is merely the result of fault in detection.

Justification Justification

» Convention on the Conservation of Antarctic Marine Living Resources, 1982 » Hønneland, G. 2012. Making Fishery Agreements Work, Edward Elgar, Cheltenham » Interviews with representatives of GSGSSI and the Norwegian Directorate of Fisheries » Norwegian Directorate of Fisheries. 2013. Melding fra fiskeridirektøren J‐236‐2013 (Fisheries Directorate Regulation J‐236‐2013) References » Norwegian Ministry of Fisheries. 2008. Act of 6 June 2008 no. 37 relating to the Management of Wild Living Marine Resources (the Marine Resources Act) » Norwegian Ministry of Fisheries. 2012. Fiskeriavtalane Noreg har inngått med andre land for 2013 og fisket etter avtalane i 2011 og 2012 (The Fishery Agreements Norway had Concluded with Other Countries for 2013 and Fishery according to the Agreements in 2011 and 2012), Meld. St. 40 (2012‐2013) (White Paper No. 40 2012‐ 2013)

OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 142

Evaluation Table for PI 3.2.4

The fishery has a research plan that addresses the information needs of PI 3.2.4 management Scoring Issue SG 60 SG 80 SG 100 a Research is undertaken, A research plan provides A comprehensive research plan as required, to achieve the management system provides the management system the objectives consistent with a strategic approach with a coherent and strategic with MSC’s Principles 1 to research and reliable approach to research across P1, and 2. and timely information P2 and P3, and reliable and timely sufficient to achieve the information sufficient to achieve

Guidepost Guidepost objectives consistent the objectives consistent with with MSC’s Principles 1 MSC’s Principles 1 and 2. and 2. Met? Yes Yes Yes A comprehensive research plan by CCAMLR exists for krill fisheries, focusing on the monitoring of krill catches, scientific observation and environment monitoring. The CCAMLR Ecosystem Monitoring Programme (CEMP) provides cross‐cutting data on environment and predator abundance to link into fisheries data and targets research at an ecosystem approach to management of the krill fishery. Justification Justification b Research results are Research results are Research plan and results are available to interested disseminated to all disseminated to all interested parties. interested parties in a parties in a timely fashion and are timely fashion. widely and publicly available. Guidepost Guidepost Met? Yes Yes Yes

The research plan and its results are disseminated to all interested parties in a timely fashion and are widely and publicly available on CCAMLR’s website. Justification Justification

» CCAMLR Annual Fishery Reports » CCAMLR website References » Convention on the Conservation of Antarctic Marine Living Resources, 1982 » Interviews with representatives of the British Antarctic Survey, the Institute of Marine Research, MRAG and the Norwegian Directorate of Fisheries

OVERALL PERFORMANCE INDICATOR SCORE: 100

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 143

Evaluation Table for PI 3.2.5

There is a system of monitoring and evaluating the performance of the fishery-specific management system against its objectives PI 3.2.5 There is effective and timely review of the fishery-specific management system Scoring Issue SG 60 SG 80 SG 100 a The fishery has in place The fishery has in place The fishery has in place mechanisms to evaluate mechanisms to evaluate mechanisms to evaluate all parts some parts of the key parts of the of the management system. management system. management system Guidepost Guidepost Met? Yes Yes Yes CCAMLR conducts ongoing internal reviews of its processes and the performance of its Member States to meet the fishery‐specific management requirements outlined. These requirements are reviewed annually (to fit in with the annual fisheries cycle) by the appropriate CCAMLR Working Groups (e.g. seabird mortality will be analysed by the Working Group on Incidental Mortality of Associated Fauna). Justification Justification b The fishery‐specific The fishery‐specific The fishery‐specific management management system is management system is system is subject to regular subject to occasional subject to regular internal and external review. internal review. internal and occasional

Guidepost Guidepost external review. Met? Yes Yes No

CCAMLR was subject to a comprehensive external performance review during 2008, but such external review is not regular. The review was carried out by a panel appointed by the Commission composed of nine persons (see http://www.ccamlr.org/pu/E/revpanrep.htm ). The purpose of the performance review was to evaluate the Commission‘s performance against comprehensive criteria and specifically against the objectives and principles set out in Article II of the Convention. The review states that the stock status and trends are broadly consistent with Article II of the Convention and international best practice. With

Justification Justification particular reference to krill fisheries, it identified the need for ongoing research into predator–prey linkages in ecosystem modelling and adequate monitoring and management within krill fisheries.

» CCAMLR website References » Interviews with representatives of the Norwegian Ministry of Foreign Affairs, the Institute of Marine Research, the Norwegian Directorate of Fisheries and WWF

OVERALL PERFORMANCE INDICATOR SCORE: 90

CONDITION NUMBER (if relevant):

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 144

Appendix 1.2 Conditions No conditions have been set for the fishery.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 145

APPENDIX 2. PEER REVIEW REPORTS

Peer Reviewer 1 Overall Opinion Has the assessment team arrived at an Yes/No Certification Body Response appropriate conclusion based on the Partiall evidence presented in the assessment y report? Justification: Enhancement/clarification has been The assessment team has reached an appropriate carried out in accordance with most conclusion. There are a few areas within the of the reviewer’s suggestions. As assessment report where the information presented he says, however, the conclusion could be enhanced or clarified. Comments related to does not change, i.e. certification these are raised in both the general comments section of the fishery is appropriate and against specific PIs below.

Do you think the condition(s) raised Yes/No Certification Body Response are appropriately written to achieve Yes the SG80 outcome within the specified timeframe? Justification: No conditions are raised and this appears No comment or response is appropriate based upon the information provided. necessary

If included: Do you think the client action plan is Yes/No Certification Body Response sufficient to close the conditions N/A raised? Justification: No conditions are raised for this fishery. No comment or response is necessary

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 146

General Comments on the Assessment Report (optional) This is an assessment of a fishery closely related to a re-certified fishery (Aker Biomarine Antarctic krill fishery). Scoring for the Olympic seafood fishery has been appropriately harmonized with that fishery. Indeed, the current report has learnt from that ‘mature’ Aker fishery certification, and as a result my comments are relatively limited.

The team has adequately highlighted the two key areas of potential concern, and assessed the implication of these for Certification, and in my opinion, reached an appropriate conclusion on the fishery. Those two key areas are:  the lack of a recent survey of krill biomass. Current management is still being based on the results of the last survey performed in 2000. Given that krill are thought to live for 5- 7 years, a number of generations have passed since that survey, and combined with the comments made within the report on climate change and sea-ice influences on recruitment, this would raise some concern. However, as the team note, the very low level of krill catch taken by the fishery, well below the trigger level, along with available scientific analysis (e.g. F and SSB levels for the krill stock in Area 48 consistent with the catch trigger level (higher than recent catch levels) are F=0.0159 (95 % CIs: 0.00750 – 0.0357) and 97.7 % SSB0 (80 % CIs: 71.6 – 135 %) respectively) reduces concern over the historical nature of the survey. The impact of uncertainties in krill biology/population dynamics on the potential consequences of the trigger catch level noted by Kinzey et al., 2013 are appropriately acknowledged.  that although the development of Small Scale Management Units for krill has been discussed for a number of years, there appears from the report to be little action in this management direction. However, the use of 'smaller' scale units (sub-units of Area 48) and the low level of total combined catch which has not reached the trigger level that might drive management action on SSMUs, and the fact that closures of the fishery have occurred where sub-regions have reached their ‘sub-unit’ limit, is acknowledged. Ongoing monitoring during annual surveillance audits will continue to update on this issue. Certification Body response: The assessment team and the peer reviewer have reached the same conclusion regarding these two major shortcomings within overall evaluation and management of the fishery, so no other comment is here necessary.

Within the report, there are a few areas that would benefit from some additions or clarifications for the reader:  the ecosystem analysis of krill fishery impacts, both at the regional level and local level. Currently the wide-range of ecosystem studies that have been undertaken are summarized, but the primary findings are mainly limited to comparing the analysis of the removal of krill to that required by predators from studies in 2004 and 2007. Scoring under e.g. PI2.5, for example, would benefit from a summary of those studies that have examined the potential impacts of trigger level krill catches on the ecosystem as a whole, and local level catches on the terrestrial predators, including work cited in the original Aker certification report (e.g. Watters et al., 2013; Watters et al., 2008; Hill et al., 2008) and any work subsequently undertaken. Currently the potential for indirect effects of the fishery on ETP (e.g. p 38) are noted, but there is little conclusion on the potential level of the effect due to catches from the UoC, and a fuller presentation of ecosystem modelling results would further support the scores given. Certification Body Response: Following the peer reviewer’s advice the team has added information under section 3.4.5 covering the results of the Watters et al. (2013) evaluation of krill management measures based on a predator–prey fishery model simulating the ecosystem dynamics of the Scotia Sea. The results of this model are also commented upon under PI 2.5.1.

 Section 3.4.3 combines information from the fishery as a whole with information from the UoC (the focus of P2 issues), which is a bit confusing. I would also suggest clarifying that the Table 2 bycatch levels represent fish larvae (this is implicit but not explicitly stated until later in the section) and summarise the larval catch levels as a % of the total catch (presented elsewhere but given the fluctuation in annual catch weights within the table, it’s not clear whether this represents variation in bycatch proportions or fluctuations in the UoC krill catch).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 147

Certification Body Response: Words have been added to the heading of Table 2, specifying that most species (but not all) are larvae (bigger individuals can be identified given the number and the weight of the catch). The Juvel’s bycatch rate is also specified here, considered as <0.5% (see the substantive text).

 Can the use of the MRAG report on fish larvae impacts be better justified, given that there appears to be differences in both the area fished and the fishing method used within the Aker fishery and the UoC? This may be as simple as clarification that the species bycatch rates are comparable between fisheries. Certification Body Response: The similarity between the MRAG analysis and the data in this report from observer reports covering fish larval bycatch composition can be identified by comparing observer reports from both vessels. This is because the two vessels fish the same stock in the same area, using a fine mesh to prevent the entanglement of bigger organisms. That is why bycatch rates are considered comparable between fisheries in the context of this report. Clarification has been added to the appropriate section of the substantive text.

 The 6 fish species listed as not being covered by the MRAG report are not discussed further. Is this because they are considered within biological limits, or because catches are considered by the team to be below levels that would cause a biological issue? Please clarify and describe the basis for the lack of concern. Certification Body Response: In terms of the status of the six species, the team notes that they have not been evaluated as threatened by CITES or IUCN. Notwithstanding, information about all species is available on the internet and there are specific studies where information is provided in the literature. Moreover, their catches are anecdotal and are presented on Table 2. The team has amended the wording in Section 3.4.3 and the scoring of PI 2.2.1.c in order to clarify this situation, and no change in scoring is deemed as appropriate.

 I suggest including Figure 11 in section 3.3.3.4 to clarify the fishing approach of the UoC. In turn, a catch history of the UoC within the fishery (preferably broken down by sub-area of area 48) would be useful to allow the reader to judge the proportion of the total krill catch taken by the UoC. Certification Body Response: Good idea of the reviewer here. The certification team have followed the advice and adjusted the figure numbering throughout.

Two minor editorial comments:  In section 3.3.2.7, please clarify the comment at the bottom of page 18 ‘…despite total annual catches not rising slowly’; my understanding from Figure 2 is that in recent years total annual catches have been rising slowly? Certification Body Response: Perceptive comment of the reviewer here. The word ”not” is redundant and has been removed from the main report.

 Double check the spelling of Nicoll and Douglass (2012) – in one place, at least, the second ‘s’ of Douglass is missing? Certification Body Response: The correct spelling is “Douglass” and this has been made consistent throughout the text

Finally, I note that while the vessel is stated not to have fished within region 48.3 (South Georgia), that region is part of the unit of certification and the report includes some of the data collection and management operations from this region to justify scores. While there are some areas where the evidence of similarity in results between the regions needs to be justified (e.g. the fish larval bycatch levels mentioned above), the described ongoing data collection and monitoring systems within the fishery in all regions of Area 48 reassure me that issues arising from the vessel moving to a new area would be identified and acted upon.

Certification Body Response: The vessel has not fished commercially within the South Georgia region, but the data collected there by that management authority reassure the certification team that the whole of Subarea 48.3 is being subject to adequate and appropriate monitoring. The reviewer is right to be reassured, as is the certification team, that the level of ongoing monitoring

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 148

throughout Area 48 is such that any issues or concerns that arise will be immediately acted upon by the management authority.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 149

Performance Indicator Review .

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.1.1 Yes Yes NA

1.1.2 Yes Yes NA I agree with the scores given. Fair comment, so no other response is necessary Under c), I suspect that the precautionary level of the TRP (75% B0) is such that the stock is being maintained above a stock size level consistent with a Bmsy proxy of 50%B0?

1.1.3 NA NA NA As noted, the stock is not No response or comment is necessary considered to be depleted so this PI is not scored.

1.2.1 Yes Yes NA

1.2.2 Yes Yes NA

1.2.3 Yes Yes NA Under a) I would suggest also The text of the scoring of 1.2.3(a) has been mentioning the other fleet activity suitably enhanced as the reviewer suggests monitoring systems (VMS, observers, logsheets, potentially

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 150

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

licensing information from GSGSSI) as calling the overoptimistic notification system on its own an ‘excellent’ basis for information on krill fleet activity seems a bit incongruous.

1.2.4 Yes Yes NA

2.1.1 Yes Yes NA

2.1.2 Yes Yes NA

2.1.3 Yes Yes NA

2.2.1 Yes Yes NA Related to comments on the main The similarity in fish larval bycatch composition text above, justification of the can be shown by comparing observer reports similarity between the Juvel and from the two vessels. The similarity is because Saga Sea operations from which they fish the same stock in the same area, using to draw comparisons of fish larval a fine mesh to prevent the entanglement of bycatch is needed; I note that bigger organisms. more justification is provided here The proportion of fish in krill schools is expected than in the main text, based on to be extremely low and as rare as in the actual the similarity of species, but a catch, so cryptic mortality would also be similar or lower quantity/rate of expected to be low. In terms of cryptic mortality

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 151

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

bycatch would be better of krill, concerns have been raised by some justification for the lack of stock authors on the CCAMLR website impact. That said, given the level (http://www.ccamlr.org/en/search/site/krill%25 of interactions, I agree that fish 20cryptic%2520mortality?advanced-search- larval bycatch may be considered field=all); however, there is currently believed to ‘rare and negligible’ be no justification to evaluate it specifically within the context of this report (there are also The question of cryptic mortality no data available). (mortality after passing through the net) of both target (krill) and In terms of the status of the other species, the other species has not been team notes that they have not been evaluated discussed – has it been as threatened by either CITES or IUCN. investigated for the fishing Notwithstanding, information is available on the system? internet and there are some data published. The team has amended the wording of Section For c), although I do not think it 2.2.1.c to clarify the situation, but does not will affect the score, the status of believe that any change in scoring is called 6 species of fish (Antarctic forneeded. Table 2 lists the low catch rates of dragonfish (GYA); Cryodraco spp thesespecies, showing that most of the catches (YDB); Grey rockcod (NOS); were made in 2013. The operational measures of Ocellated icefish (KIF); Pagetopsis the vessel and the low proportions of these macropterus (Icefish) (PMA) and species are considered sufficient to ensure that Yellowbelly rockcod (NON)) that the fishery is not hindering their recovery (if are not included within the MRAG such was warranted), so the scoring is report is not discussed in the main maintained. text. Is the status of ALL species therefore known? If not, are the Clarification has been added to the justification operational measures of the vessel for this PI. considered sufficient to ensure the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 152

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

species remain within biologically- based limits?

2.2.2 Yes Yes NA

2.2.3 Yes Yes NA Within d) Please confirm that Most observers within the MRAG contract are although the vessel has not British, but there are also some southern operated within the SGSSI region hemisphere nationals (from South Africa and (48.3) according to the report, Australasia). samples from the vessel are In terms of identification of bycatch species, this analysed in South Georgia or in is generally carried out on board, to the lowest the UK? Are observers from the possible taxon, but occasionally photographs are UK, for example, or are CCAMLR taken of “difficult” taxa and sometimes too (but samples sent to these locations rarely) samples are preserved for later regularly? confirmatory identification. Scientists on South Georgia or in the UK are consulted for such identification.

2.3.1 Yes Yes NA A large part of the text in a) PI 2.3.1 focuses on the effects that the fishery discusses mitigation approaches may be having on ETP species and its and observer guides, which are compliance with national and international not directly relevant to knowing standards for their protection. As the reviewer the effects of the fishery with a says, the team has considered the lack of high degree of certainty (although interactions recorded by the 100% observer the 100% observer coverage of coverage as sufficient to conclude that there is a the UoC certainly is) and that high degree of certainty that the effects of the

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 153

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

those effects are within fishery are within the limits of national and national/international limits international requirements for ETP species. (although the limits are not No changes in the wording are considered stated, the lack of observed necessary because the wording provided interactions means the fishery explains the lack of recorded interactions and likely meets them!). provides robustness to the records.

The information collected by the CCAMLR For c), although comparison of Ecosystem Monitoring Programme (CEMP sites) annual comsumption of krill by and by research institutes (such as British predators and fishery removals Antarctic Survey, the Norwegian Institute of provides some evidence, is there Marine Research, and the South Georgia and ecosystem work performed at the South Sandwich Islands Environmental Agency) regional scale (overall limits), and serve to identify and quantify the detrimental examinations of the efficacy of effects that the fishery may be having on ETP local catch limits on ETP species species. Such information is revised anually by that could be used to further WG-EMM and should serve to detect any support the score given? I note increase in land-based ETP species feeding examples are also given under requirements. 2.3.3a No changes in the wording are required because this topic is covered by PI2.3.3 rather than 2.3.1.

2.3.2 Yes Yes NA

2.3.3 Yes Yes NA

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 154

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.4.1 Yes Yes NA

2.4.2 Yes Yes NA

2.4.3 Yes Yes NA

2.5.1 Yes Yes NA Further support for the score could Information has been added referring to the be supplied based on the findings predator–prey fishery model used by Watters et of ecosystem modelling of the al (2013) to evaluate krill management potential effects of regional/local measures in the Scotia Sea. catch levels on ecosystems.

2.5.2 Yes Yes NA

2.5.3 Yes Yes NA

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 155

Performance Has all the Does the Will the Justification Certification Body Response Indicator relevant information condition(s) Please support your answers by referring to specific scoring issues and any relevant information and/or rationale raised improve documentation where possible. Please attach available been used to score this the fishery’s additional pages if necessary. used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.1.1 Yes Yes NA Note typo of CSGSSI in the text of Corrected – thank you for pointing this out. a).

3.1.2 Yes Yes NA

3.1.3 Yes Yes NA

3.1.4 Yes Yes NA

3.2.1 Yes Yes NA

3.2.2 Yes Yes NA

3.2.3 Yes Yes NA

3.2.4 Yes Yes NA

3.2.5 Yes Yes NA

Any Other Comments

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 156

Comments Certification Body Response

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 157

Peer Reviewer 2 Overall Opinion

Has the assessment team arrived at an Yes/No Conformity Assessment Body appropriate conclusion based on the Yes Response evidence presented in the assessment report? Justification: No comment or response is necessary The assessment team has thoroughly reviewed a substantive quantity of relevant information to underpin its assessment of the Olympic Seafood Antarctic Krill Fishery. This assessment has been carried out in conformity with MSC Principles and Criteria for Sustainable Fisheries. The assessment’s resultant conclusions are thoroughly presented, well-articulated and logical. They categorically support certification of the fishery in question.

Do you think the condition(s) raised are Yes/No Conformity Assessment Body appropriately written to achieve the SG80 N/A Response outcome within the specified timeframe? Justification: No comment or response is necessary While the assessment team did not identify any conditions for the fishery, they highlighted the long-standing (15-year) lack of a fishery-independent synoptic survey to assess the krill target stock(s) concerned. This was viewed by the team as posing a serious challenge for the fishery’s ongoing management; a point I agree with if such management is to remain robust. However, this is not a weakness that can be laid at the client’s door.

The report alludes to a lack of small-scale catch apportionment, and setting of fine-scale precautionary catch limits, to mitigate catch over-concentration. However, the applicant’s commitment to ecosystem-safe fishing, observer coverage and supporting of relevant research all go beyond the CCAMLR norm. This suggests that the client is aware of the dangers of over-concentrating fishing in particular areas where the principles of balancing ‘rational use’ and ‘conservation’, in CAMLR Convention Article II parlance, might be compromised.

If included: Do you think the client action plan is Yes/No Conformity Assessment Body sufficient to close the conditions raised? N/A Response Justification: In the opinion of the certification team, As noted, the client expends considerable effort (i.e. the client’s commitment to both scientific observer deployment, financial support of relevant research and the ongoing support of research, use of fine mesh net panels etc.) on improving dedicated monitoring surveys is well relevant knowledge and fishery practices associated with the described in the text and certainly in fishery. Most notably the client has developed protocols much of the associated literature aimed at mitigating, preventing and reducing impacts on supplied (some referenced). To adjust marine animals encountered during trawling. wording to the report now would, the The extent of the client's commitment to conducting annual, team feels, have the wrong effect of or bi-annual, surveys in the absence of fishing on the Area highlighting perceived failings of the 48 krill stock(s) it targets, is not clear. The assessment team client, when there are none. The text is views such surveys as crucial to ameliorating uncertainty therefore left as is. surrounding stock status, particularly in light of the continued lack of large-scale synoptic survey in the area concerned (Assessment Report Pl 1.2.1 on p. 83).

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com

General Comments on the Assessment Report (optional)

I would like to express my appreciation to the assessment team for a well written, thorough and easy- to-understand report. The scores, conclusions and assessments provided are convincingly-justified and fully meet MSC requirements.

Certification Body response: Such comments are much appreciated by the team and reflect, it is felt, positively too on the client’s own commitment to sustainability of their fishing operation

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com

Performance Indicator Review

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.1.1 Yes Yes N/A The stock is highly productive with a low No comment or response is probability of recruitment overfishing. necessary Under CCAMLR decision rules, the allocated assessment scores are acceptable on a stock status basis. While annual stock assessments are absent (see synoptic survey absence comments), the CCAMLR management system's precautionary elements are sufficiently robust to respond timeously to likely future impacts on fishery target/limit reference points. Assigned performance indicator score of 100 justified.

1.1.2 Yes Yes N/A Catch trigger level is well below F, and No comment or response is above the B0 target/limit reference points. necessary because the Reference points are sufficiently reviewer’s comments are precautionary to allow larger catches supportive of the scoring without risking stock's future reproductive capacity. Target reference points also account for krill's key ecological role, but CCAMLR 0.75B0 escapement percentage requires further research and analysis. Assigned performance indicator scoreof 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 160

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.1.3 Yes Yes N/A The stock is not depleted and no stock No comment or response is rebuilding is necessary. However, should necessary rebuilding be required in the future then it follows that CCAMLR rules will be applied. No assigned performance indicator scores are necessary.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 161

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.2.1 Yes Yes N/A Ongoing non-assignment of SSMUs Everything the reviewer precludes accounting for predator-fishery says here is valid – overlaps directly. This, and other, succinctly, he says (as does attendant issues, may not be critical at the team in the report) that current stage of fishery development, if any catch increase may catches remain low compared to overall raise concern about stock yield and if pro-forma percentage sustainability at current subareal apportionment of trigger catch levels of knowledge. levels continues. However, any catch Therefore, the client is increase, particularly of large spatially- encouraged to join ongoing localised catches, may alter picture. CCAMLR and individual Nonetheless, current management stratgy country efforts in collecting appears to meet its objectives. More and having analysed the information is required to examine data that will allow better functional relationships between fishery and management at smaller- krill predators (including predator-fishery scale management units to overlap(s)) at various scales, as well as in be applied. Naturally, the the context of formally using CEMP-derived team endorse this view, but data to improve discrimination between no change in wording of the annual predator variability and fishing report or the scoring is effects on the Area 48 krill stock. CCAMLR necessitated. The subject should be encouraged to fully assess, and will, however, have to be decide on, the efficacy of subareal trigger evaluated and considered at levels in ameliorating the need for smaller ongoing annual than subareal catch quotas. Equally, the surveillances should the client is encouraged to share the decision- fishery be certified. making criteria (e.g. catch rate or actual catches) that it applies when, or if, it

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 162

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

Yes Yes N/A decides to limit/cease fishing when fishery- predator overlaps occur. The client is also encouraged to become more deeply involved in annual, or bi-annual, fishery- independent surveys of Area 48 krill fishing grounds. Evidence for the effectiveness of current management strategie appears justifiable. Assigned performance indicator of 95 score justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 163

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.2.2 Yes Yes N/A Current management of the Area 48 krill The reviewer supports the fishery is highly precautionary. Harvest scoring and the justification controls are well-defined and consistent and text associated with the with the declared harvest strategy. Fishery assessment team’s evalua- development is balanced with the harvest tion and conclusion. There- control rules, while trigger levels are fore, no further comment or sympathetic to the objective that no response or enhancement of reference point is approached. A lack of full the text is necessary understanding of the system and a probablistically-determined harvest control regime, based on the lowest estimated B0, generate confidence in assigned precautionary catch levels for the time being. However, the situation may change if catches rapidly approach, or exceed, trigger levels in the forseeable future (see Nicol et al. 2011). Most notably, uncertainties associated with environmental variations mitigate against scoring harvest control rules above SG 80. Assigned performance indicator score of 80 justified.

1.2.3 Yes Yes N/A Assigned performance indicator score No comment or response is of 80 justified. necessary

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 164

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

1.2.4 Yes Yes N/A The GYM approach, with its attached No enhancement of the text predator and recruitment criteria, is or scoring is suggested by adequate for determining an upper the reviewer. He merely precautionary catch limit. A general, and summarises the conclusions ongoing, lack of fishery-independent survey of the assessment team assesment of stock biomass, along with other information, mitigates againts a score greater than SG 80. Being stochastic, the GYM applies probalistic decision rules to account for uncertainty In this regard. Current target and limit reference points are expressly accounted for in the GYM approach and harvest control rule regime. Catch maxima are set well below suggested levels attached to managing the triggers. Precautionary determination of the catch limit, and even more precautionary trigger level, further suggest that prevailing management uncertainties are addressed by CCAMLR’s approach which depends on annually reviewing availiable assessments. Limited external peer review of assessments takes place. However, internal CCAMLR scientific expertise is commensurate with world standards. Assigned performance indicator score of 85 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 165

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.1.1 Yes Yes N/A Species other than target species (krill) are No comment or response is not retained by this fishery. Assigned necessary performance indicator score of 100 justified.

2.1.2 Yes Yes N/A Only krill, as the target species, is retained No comment or response is aboard. With designated crew to check necessary catches, 100% observer coverage to monitor catch composition, there is low probability (<0.5%) of non-target catch retention failing. Supporting evidence for this conclusion is provided in CCAMLR observer reports. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 166

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.1.3 Yes Yes N/A UoC fully complies with CCAMLR observer As recognised by the scheme requirements and implements reviewer, observer reports 100% observer coverage. Observer reports provide accurate provide detailed information on catch information on catch composition sampling in accordance with composition that serves to the CCAMLR observer manual. As noted in support the high scoring of 2.1.2 above, no species other than krill are this performance indicator. retained and bycatch levels (<0.5% of the

total catch) are minimal. Accurate observer information on catch composition and bycatch species is provide, so ensuring that bycatch mortality figures are accurate With only krill being retained, this endorses accuracy of the catch composition and bycatch information collected. Observer- derived information is fully accessible to CCAMLR. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 167

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.2.1 Yes Yes N/A While also covered in later discussion Accurate observer reports (Section 3.4.3), the SG100 standard is met show little interaction with contingent on the low fish bycatch bycatch species, the status proportion taken by the UoC's Juvel. The of most of which appear to stock status of seabirds and mammals be good. Mitigation interacting with the fishery is good. measures in place Furthermore, mitigation measures appear contribute to this minimal sufficiently extensive to ensure that the interaction. fishery does not unduly affect the recovery of bycatch, or interacting species. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 168

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.2.2 Yes Yes N/A Strategies in place for managing and No enhancement of the text minimizing bycatch are extensive and UoC or scoring is suggested by has been recognised by the Friends of the the reviewer. He merely Sea for its efforts. The naturally summarises the conclusions homogeneous nature of krill's distribution of the assessment team. substantially contributes to a the ,cleanliness, of the catch and low bycatch. Equally, 100% CCAMLR observer coverage provides essential information about the fishery and species involved. The attendant strategies and consequently provided information provide evidence that the Juvel is performaing as should be expected under this performance indicator as a whole. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 169

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.2.3 Yes Yes N/A Based on small bycatch proportions and few The reviewer supports the fishing gear interactions, the consequences scoring and the justification of the fishery affecting bycatch and and text associated with the interacting seabird/mammal populations is assessment team’s evalua- considered neglible. The comprehensive tion and conclusion. information base (see 2.2.2 above) Therefore, no further adequately supports strategies to manage comment or response or bycatch species, as well as evaluate their enhancement of the text is efficacy. Ongoing implementation of the necessary. CCAMLR observer program should be sufficient for the future monitoring of by catch in the fishery. Assigned performance indicator score of 100 justified.

2.3.1 Yes Yes N/A Comprehensive and persistent observer As stressed by the reviewer, coverage, as well as absence of fishery the absence of interactions interactions over the past three years, in the past few years and robustly indicate a lack of fishery impact on the different studies ETP species. Furthermore, fishery removals estimating that krill of krill have been estimated to be removals by the fishery are significantly less than krill predator demand significatively less than krill and well within the required krill predator demand serve to escapement levels (75%) attached to support the current scoring. CCAMLR's precautionary management approach. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 170

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.3.2 Yes Yes N/A Both CCAMLR Conservation Measures (CMs) No further comments. and Juvel's own strategies provide a comprehensive approach to managing the fishery'' impact on ETP species. The presence of, and information from, scientific observers indicate effective implementation of these strategies. No interactions by the fishery with ETP species show that the CMs and strategies are working. Assigned performance indicator score of 100 justified.

2.3.3 Yes Yes N/A Available information suggests that The reviewer agrees with quantitative estimates of fishery impact on the team and highlights the ETP species are possible. However, lack of a high degree of documented changes of functional certainity on the realationships between predators and krill quantitative estimation of make it difficult to estimate ETP species the outcome status of ETP outcome status with a high degree of species that prevents this PI certainity. Both null observer reports of from obtaining a higher fishery-induced ETP mortality/injury and score. IWC information attest to achievement of SG100. Nonetheless, the attendant uncertainties highlighted attest to a lesser guidepost score. Assigned performance indicator score of 95 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 171

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.4.1 Yes Yes N/A The krill fishery uses pelagic gear meaning No comment or response is that there is little probability of damage to necessary the seabed. The Juvel has never lost fishing gear. Assigned performance indicator score of 100 justified.

2.4.2 Yes Yes N/A An absence of interactions between Juvel's No comment or response is fishing gear and the seabed, along with necessary measures (including by CCAMLR), to afford marine protected area (MPA) status to particular areas clearly affirm existence of a stragey to manage fishery impacts on habitat types. Compliance enforcement is addressed through inspections (particularly in CCAMLR Subarea 48.3 - South Georgia region) and implementation of VMS-derived positional monitoring. CEMP monitoring and other scientific research in the region support these strategies. Based on fishery reports and on observations of no habitat interactions, the strategies appear to be effective. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 172

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.4.3 Yes Yes N/A Adequate information is available to Although CCAMLR does determine the fishery impact risk(s) for attempt to collate habitat types as the pelagic gear deployed information on this subject, does not impact the seabed. However, work there is still a long way to is still underway to expand the MPA go before all vulnerable network in Area 48, while there is an marine habitat types in the absence of VME information (including Southern Ocean are seafloor mapping) to improve current identified and monitored on knowledge of the habitats where krill fishing a continuous basis. takes place. Consequently, and for these reasons, SG100 cannot be met. Despite no gear loss being attributable to Juvel, along with available studies of terrestrial habitats by CEMP, a lack of benthic habitat change monitoring also means that SG100 cannot be met. Assigned performance indicator score of 85 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 173

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.5.1 Yes Yes N/A The Juvel implements protocols aimed at The team shares the minimizing impacts on certain seal and reviewer’s concern about seabird species. The review team's overall any catch increase that may assessment of this PI appears justified in so negatively impact far as the current krill harvesting level is sustainability in the future. highly unlikely to impact key ecosystem As mentioned already, the structures and functions. CCAMLR client is encouraged to join precautionary catch subarea limits are also ongoing CCAMLR and tightly regulated and have resulted in individual country efforts in occasional subareal fishery closures. It is collecting and having crucial that any decisions to move beyond analysed the data that will the interim trigger catch levels in Area 48 allow better management in (CCAMLR CM 51-07 [2014]) are preceded future. However, at current by verification of key parameters such as harvesting rates, both the krill recruitment variability, natural team and the reviewer mortality and others specified by Kinzey et agree that the fishery is al. (2013) to ensure that testing of highly unlikely to disrupt the management criteria is realistic. Assigned key elements underlying performance indicator score of 100 ecosystem structure and justified. function to a point where there would be serious or irreversible harm.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 174

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.5.2 Yes Yes N/A CCAMLR CM 51-07 is considered, in part, to The reviewer agrees with prevent the fishery seriously damaging the the team that CCAMLR ecosystem. However, it falls short of being current management a robust management action or strategy strategy is insufficient to per se. The absence of SSMU-assigned support timely responses to catch trigger levels and of a full feedback ecosystem change. No mangement policy also means that CEMP changes in the scoring are data are not being used by CCAMLR in CM needed at this point. development. The latter intimates that management of the fishery is unikely to be sufficiently flexible to support timely responses to ecosystem changes considered sufficient, and necessary, for current management efforts to serves as a robust strategic, or plannning, framework. Despite fishery closures having been implemented, CM 51-07 provisions are not considered sufficient to meet SG100 conditions for this PI. Assigned performance indicator score of 80 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 175

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

2.5.3 Yes Yes N/A Available information appears sufficient to No comment or response is provide some understanding of key necessary because the ecosystem elements. Further, major fishery reviewer’s comments are impacts on key ecosystem elements may be supportive of the scoring inferred from existing information and have been examined using logbook data, VMS track records, observer reports and CEMP information. However, limited knowledge of some important ecological processes (e.g. pelagic predator needs, climate impacts on krill recuritment and survival etc.) means that SG100 is not met. Nonetheless available information is sufficient to parameterize the ecosystem models identified by the assessment team. These models provide a basis for inferring potential consequences of fishery-induced impact(s) on the ecosystem. Clarification of the considerable list of uncertainties identified will undoubtedly facilitate further development of strategies to manage the fishery's potential ecosystem impacts. Assigned performance indicator score of 90 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 176

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.1.1 Yes Yes N/A The management system is considerd No comment or response is effective, with attached rule and decision- necessary making practices being nationally, and internationally, coherent. The system's legal and custormary frameworks are working, although within-CCAMLR evidence for a lack of major disputes is not verifiable at this time. Howeve, it may be inferred that protracted disagreements, such as those characterizing the CCAMLR MPA debate, may be symptomatic of a 'chronic' lack of consensus subliminal to the conditions for invoking a formal dispute outlined in CAMLR Convention Article XXV. This would mean that SG100 cannot be fully met. It is unlikely that the fishery would have a negative impact on people dependent on fishing for their food or livelihoods. Assigned performance indicator score of 90 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 177

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.1.2 Yes Yes N/A Overall management responsibilities and No comment or response is lines are clear with respect to CCAMLR, the necessary Juvel's flag State and on-the-water. The attached roles are clearly defined and understood by the fishery's particpants. The assessment team was not provided with documentation able to consistently explain how information considered by the management system is used. Such lack of clarity, which probably reflects the fact that CCAMLR's official record is drafted in summary form only, appears sufficient for this guidepost to score less than SG100. Assigned performance indicator score of 95 justified.

3.1.3 Yes Yes N/A CCAMLR's precautionary and ecosystem No comment or response is management approach, and the South necessary Georgia/South Sandwich Islands MPA Management Plan are consistent with MSC Principles and Criteria. The 2008 Norwegian Marine Resources Act is also contingent on a precautionary and ecosystem approach. Assigned performance indicator score of 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 178

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.1.4 Yes Yes N/A The management is largely directed at No comment or response is ensuring that negative incentives do not required arise. There are no direct cleint subsidies, either nationally or from CCAMLR. CCAMLR costs are recovered from member States, while observer costs are the client's responsibility, as are fishing license fees. CCAMLR krill fishery notification fees are currently being reviewed. Nonetheless, there are implicit incentives in the client's flag State's (Norway) case, insofar that licensing and reporting requirements are so placed to ensure future fishery participation. A lack of essential documentation means that this guidepost scores less than SG 100. Assigned performance indicator score of 80 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 179

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.2.1 Yes Yes N/A Well-defined, measurable objectives are No comment or response is available for the fishery in the long- and required short term. They are demonstrably consistent with achieving P1 outcomes and explicit insofar as being well-established, understood and applied by fishery participants. MSC Principles 1 and 2 outcome-fulfillment is only partially demonstrated. Assigned performance indicator score of 90 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 180

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.2.2 Yes Yes N/A CCAMLR's decision-making process, No comment or response is stakeholder involvement and scientifically- required sanctioned analyses result in CMs and fisheries strategies aimed at achieving short and long-term fishery objectives (SG80 met). However, no clear evidence is available to indicate that all potential issues are being responded to appropriately, in terms of relevant research, monitoring, evaluation and consultation (SG80 met). According to the assessment team, guide post 3.2.2.(c) only meets SG80. However, it appears that SG100 would actually be met since CCAMLR CMs are based on a precautionary approach, the best scientific evidence available and close insitutional cooperation. Essential information reporting and dissemination are good. The management system is both proactive and responsive. It is able to address disagreements outside the full legal conditions attached to the Convention's provisions. Assigned performance indicator score of 95 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 181

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.2.3 Yes Yes N/A Assessment shows that CCAMLR provides a No comment or response is clear, compelling, consistent and necessary comprehensive MCS regime for the krill fisheries it manages. This includes haul-by- haul and electronic catch reporting being mandate by client's the flag State, as well as deployment of CCAMLR observers. These requirements are required to meet Norwegian fishing license conditions and underpin effective deterrence. The client's vessels are inspected by South Georgia authorities before fishing is allowed, while catches are reported daily. At-sea and in- port inspections are undertaken by the authorities. The infringement sanctions regime is clear and rigorously applied by the flag State, while the client vessel has recorded no infringements. Regulations are expected to be applied in the future and to be closely monitored/complied with. The client complies with the managemen system, and provides all necessary informationprior to entering the fishery (as per CCAMLR CM-21-03 [2014]). No evidence for regulatory non-compliance reflects the effectiv-

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 182

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.2.3/Cont. Yes Yes N/A ness of the flag State's compliance enforcement regime. Assigned performance indicator score 100 justified.

3.2.4 Yes Yes N/A A comprehensive CCAMLR krill fishery No comment or response is research plan exists. This focuses on necessary monitoring catches, scientific observation and environmental monitoring. Cross- cutting CEMP data, although not widely or directly used for management purposes, provides an important management component, and context, for determining ecosystem 'health', as well as to link environment and predator abundance data with fisheries data. Research plans/results are widely and timeously disseminated. Assigned performance indicator of score 100 justified.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 183

Performance Has all the Does the Will the Justification Conformity Assessment Indicator relevant information condition(s) Please support your answers by referring to specific Body Response scoring issues and any relevant documentation where information and/or rationale raised improve possible. Please attach additional pages if necessary. available been used to score this the fishery’s used to score Indicator support performance to this Indicator? the given score? the SG80 level? (Yes/No) (Yes/No) (Yes/No/NA)

3.2.5 Yes Yes N/A CCAMLR performance evaluations are The assessment team internally robust, especially in terms of scored 90 for this annual reviews targeting both fisheries and Performance Indicator. The related data. Only one comprehensive reviewer’s statement that a external review of the institution's score of 100 is justified performance has been undertaken. While must be an error because, this 2008 review indicated the krill stock from his words, he status and trends are consistent with obviously agrees with the CAMLR Convention Article II principles and team’s conclusions. international best practice, it identified the need to ensure taht ongoing research is directed at krill-predator ecosystem linkages, as well as at further modelling of the ecosystem. The latter highlights an ongoing need for adequate monitoring of the krill fishery as well as its precautionary management. Assigned performance indicator score of 100 justified.

Any Other Comments

Comments Conformity Assessment Body Response

No other comments and the reviewer is comfortable with the results provided by the assessment team.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 184

APPENDIX 3. STAKEHOLDER SUBMISSIONS The Public Comment Draft Report was published 14 May 2015. Comments to the PCDR was received from MSC Technical Oversight/Dan Hoggarth, Fisheries Oversight Director.

MSC comments are included in full below, followed by the explicit responses of the team.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 185

Comments from MSC

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 186

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 187

Comments from MSC

SubID PageReference OversightDescription Pi CABComment

According to these comments the wording in ch. 5.2.6 is The PCDR does not sufficiently detail the number and/ changed to: «The vessel lands the products mainly in or location of points of landing. Montevideo in Uruguay. However, sometimes the vessel goes

to an other port located in Las Palmas de Gran Canaria in Spain The response "However, sometimes [the vessel] does 15294 66 n/a to unload the products. The reason for doing this is go to other ports, e.g. Las Palmas de Gran Canaria in occasionally need for combination of unloading with repair Spain, to unload the products", does not adequately and/or docking at the shipyard, which is difficult in detail how many ports are used, even occasionally, Montevideo.” and why ports other than Montevideo would be used. Olympic confirms that they are starting up a process of The report does not mention that Rimfrost (USA) and obtaining CoC for Olympic Seafood AS, that buys the products Olympic Biotec (NZ) are in the process of obtaining at landing for direct sale and shipment to RIMFROST NZ or CoC in order to use the ecolabel on krill products. RIMFROST USA, or temporarily stored in locked customs Furthermore, these plants' CoC would also need to warehouse in Montevideo without further treatment. Also cover the movement of product after landing (i.e. RIMFROST NZ or RIMFROST USA are in the process for Montevideo, Las Palmas, etc), for example, if obtaining CoC certification. 15300 67 n/a distribution or storage subcontractors are used. All the goods are in the custody of Olympic Seafood AS after landing before further sale to RIMFROST NZ or RIMFROST USA. Everything is packed and clearly labelled and there is no risk for mixture with other products.

The report will be amended accordingly.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 188

APPENDIX 4. SURVEILLANCE FREQUENCY

The surveillance level is determined based on Table C3 and C4 in the CR requirements v1.3. According to MSC Certification Requirements Version 1.3, the overall surveillance score for this fishery qualifies for the Reduced surveillance level with on-site surveillance audits in year 2 and 4. Table 2 Fishery surveillance level (Ref. CR v1.2 table C4) Year after certification or recertification Surveillance score (from CR table C3; Surveillance level Year 1 Year 2 Year 3 Year 4 Table 12) On-site On-site On-site On-site 2 or more Normal surveillance surveillance surveillance surveillance surveillance audit audit audit audit Off-site On-site Off-site Option surveillance surveillance surveillance On-site 1 surveillance Remote audit audit audit 1 audit & surveillance On-site Off-site On-site Option recertification site surveillance surveillance surveillance 2 visit audit audit audit On-site On-site surveillance Review of new Review of new 0 Reduced surveillance surveillance information information audit & audit recertification site visit

Table 12 Determination of surveillance score (Ref. CR v1.3 table C3) Surveillance Surveillance Criteria Alternatives Rationale score Score Yes 0 Default assessment tree Default assessement tree 0 used? No 2 used Zero conditions 0 Number of open Between 1-5 1 0 No conditions conditions conditions More than 5 2 ≥85 0 Principle Level Scores 0 All principles scored > 85 <85 2 Conditions on outcome Yes 2 No conditions on outcome 0 PIs? No 0 PIs Total score 0

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 189

APPENDIX 5. CLIENT AGREEMENT

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 190

About DNV GL Driven by our purpose of safeguarding life, property and the environment, DNV GL enables organizations to advance the safety and sustainability of their business. We provide classification and technical assurance along with software and independent expert advisory services to the maritime, oil and gas, and energy industries. We also provide certification services to customers across a wide range of industries. Operating in more than 100 countries, our 16,000 professionals are dedicated to helping our customers make the world safer, smarter and greener.

DNV GL – Report No. 2015-03, Rev. 4 – www.dnvgl.com Page 191