CCAMLR’s approach to managing Antarctic marine living resources^[248]

D.G.M. Miller, E.N. Sabourenkov and D.C. Ramm
CCAMLR Secretariat
P.O. Box 213, North Hobart, Hobart, Tasmania, 7000, Australia
<[email protected]>

1. Introduction

The "ecosystem and precautionary" principles of the 1980 Convention on the Conservation of Antarctic Marine Living Resources’ (CAMLR Convention)(CCAMLR 2002a) are encapsulated in its Article II (Box 1). These principles are often cited as the Convention’s most innovative feature compared to other fisheries and management instruments adopted both before, and after, the Convention’s entry into force in 1982 (Butterworth 1986, Constable et al. 2000, Molenaar 2001).

The Commission (CCAMLR) established under Article VII of the Convention has a number of important functions. The most notable of these are outlined in paragraphs 1 and 2 of Article IX and focus on the formulation, adoption and revision of Conservation Measures^[249] based on the best scientific evidence available [Article VII, paragraph 1.(f)]. Such measures include, inter alia, setting of catch limits, effort controls, closed areas and seasons, etc.

At its first two meeting in 1982 and 1983, CCAMLR was preoccupied with procedural issues and with setting up terms of reference for its advisory Scientific Committee (the functions of which are set in Convention Article XV) and other subsidiary bodies. At that time, there was a growing and urgent sense of concern over the status of fish stocks around South Georgia and the Kerguelen Islands (Kock 1992). With CCAMLR’s "newness" there was also insufficient data with which to make scientific assessments of stocks in either area.

Consequently, the first two conservation measures adopted in 1984 established mesh size regulations for pelagic and bottom trawl fisheries around South Georgia while setting up a 12 nautical mile closed area offshore (CCAMLR 1984). In keeping with the Chairman’s Statement (CCAMLR 2002a) setting conditions for the application of jurisdiction in waters under sovereign control within the Convention Area^[250], CCAMLR accepted that the fish stocks within the French Exclusive Economic Zone (EEZ) around Kerguelen Island were subject to national regulatory measures.

Additional concern surrounded the impending likelihood that a potentially large fishery would develop for one of the Southern Ocean’s keystone species - Antarctic krill (Euphausia superba) in the foreseeable future (Miller 1991, Agnew and Nicol 1996). Given the historic precedents associated with exploitation of a number of top predators (both whales and seals) in the region, there was also a growing fear that exploitation of krill, an important food item, could compromise predators’ statuses as well as their recovery from past unsustainable levels of exploitation. Despite the relative success of the international Biological Investigations of Marine Antarctic Systems and Socks (BIOMASS) Programme, particularly the First International BIOMASS Experiment (FIBEX), knowledge of krill potential yield and sustainability continued to be severely limited (Fogg 1994). These concerns resulted in the unique formulation of paragraph 3 of Article II and the Convention being termed the "krill ecosystem convention" (Mitchell and Sandbrook 1980, Edwards and Heap 1981).

Following its rather modest start, CCAMLR soon launched a comprehensive programme of scientific work. This included the evolving development of procedures to analyse historical fishery data as well as the collection of data necessary for monitoring fisheries, stock assessment, and improving the understanding of fishery stock dynamics, biology and productivity. A most important initiative set up the CCAMLR Ecosystem Monitoring Programme (CEMP) [Paragraph 7.2 in SC-CAMLR (1985)]:

"To detect and record significant changes in critical components of the ecosystem to serve as a basis for the conservation of Antarctic marine living resources".

This objective was qualified as follows:

"The monitoring system should be designed to distinguish between changes due to harvesting of commercial species and changes due to environmental variability, both physical and biological."

Institutionally, CCAMLR recognized that adequate enforcement posed a serious challenge to ensuring effective implementation of its Conservation Measures envisioned under Article IX. The problem is compounded by the Convention Area’s size, remoteness and relative ease of access, particularly its high-seas areas (Molenaar 2001). An ambitious work programme to develop a CCAMLR compliance regime culminated in the setting up of the CCAMLR System of Inspection and Scheme of International Scientific Observation (see Section 5.2.3)(Rayfuse 2000) to improve at-sea, as well as in port, monitoring and enforcement.

In subsequent years, CCAMLR developed a comprehensive suite of Conservation Measures based on both traditional fishery management approaches as well as some to address precautionary and ecosystem concerns (e.g. CCAMLR 2002b). Currently, CCAMLR’s Conservation Measures comprise an integrated set of individual measures dealing with all aspects of modern fisheries management. The Measures are kept under constant review to evaluate performance and, if necessary, provide for revision or the adoption of new measures. This approach has resulted in CCAMLR being viewed as a pioneer in developing novel and innovative measures to address both precautionary and ecosystem principles (Everson 2002). It has also served to improve management of some important CCAMLR stocks on a global basis; an approach which required implementing both compliance and trade-related measures (Sabourenkov and Miller 2004).

With growing recognition of its achievements, CCAMLR is viewed as a useful model for fisheries management in other areas, particularly on the global high seas. Through applying its precautionary and ecosystem approach, CCAMLR has addressed the concerns that:

• All data that can be collected are. The extent and effect of uncertainties and gaps in such data are taken into account before management decisions are made. This process minimizes the risk of long-term adverse effects rather than delaying decisions until all necessary data are available.

• Account is taken of all potential relationships between harvested species, including those related thereto, and physical components (such as currents, sea temperature) constituting the Antarctic marine ecosystem.

In this paper, we document some of CCAMLR’s experiences in developing and implementing a feedback (i.e. modifiable) system of contemporary fisheries conservation and management measures for the resources for which it is responsible. The first part of the paper details three examples of CCAMLR’s efforts to implement some of Article II’s provisions. These examples have been chosen to highlight the development and evolution of attached measures dealing with: (a) uncertainty attached to new and exploratory fisheries (Section 2), (b) reduction and elimination (i.e. minimization) of seabird bycatch in longline fisheries in the Convention Area, including mitigation measures (Section 3), and (c) trade related measures to combat unregulated fishing both in the Convention Area from a global perspective (Section 4). The second part of the paper focuses on assessing CCAMLR’s achievements. It also identifies some potential threats to its future effectiveness and suggests possible solutions to counteract them.

2. Regulation of new and exploratory fisheries

2.1 Introduction

In an ideal world, all the information required for sustainable and scientifically-defensible exploitation of new fishery stocks should be in place before commercial fishing is permitted (Butterworth 1986, 1999). Comparisons can then be made of stock status before and after exploitation begins, with management action being adjusted accordingly to ensure maintenance of some desirable, or pre-ordained stock level. Reality may differ and new fisheries are often exploited - even overexploited - well before the necessary management information is available or even collected (Butterworth 1999). The precautionary approach envisioned under CCAMLR Article II attempts to balance these two somewhat contradictory elements in a way that strives to minimize risk of irreversible changes in target, or dependent, stocks (Table 1).

Table 1
Modifications and, or additions to CCAMLR Conservation Measure 112/XV providing general rules for new/exploratory toothfish (Dissostichus spp.) fisheries

2.2 History of CCAMLR’s approach to new and exploratory fisheries

Following advice from its Working Group for the Development of Approaches to Conservation of Antarctic Marine Living Resources (WG-DAC) in 1989 [Annex E in CCAMLR (1989)] and from the Scientific Committee a year later [Paragraph 9.2 SC-CAMLR (1990)], CCAMLR set up a process to address the management of new^[251] fisheries in the Convention Area [Paragraph 9.3 in CCAMLR (1990)]. It was recognized that the development of such fisheries "should be directly linked with the process of elaborating scientific advice and management" measures to ensure that fisheries development does not outpace CCAMLR’s ability to meet the objectives of Article II [Paragraph 9.3 in CCAMLR (1990)].

CCAMLR accepted the Scientific Committee’s advice (Paragraph 289 in Annex 5 of SC-CAMLR (1990)] that certain information is vital for assessing the potential yield of a new fishery and that such information should be considered before a new fishery develops further [Paragraph 9.4 in CCAMLR (1990)]. Finally, it was emphasized that development of measures directed at informing the Commission of intentions to conduct any fishery in the Convention Area was crucial to ensuring effective implementation of CAMLR Articles II and IX [Paragraph 9.5 in CCAMLR (1990)].

The following year, CCAMLR adopted Conservation Measure 31/X [Paragraph 10.3 in CCAMLR (1991)] setting out provisions requiring notification of any new fishery and attached conditions for implementation. This particular Measure has remained unchanged since its inception, now standing as Conservation Measure 21-01 (CCAMLR 2002b).

Through promulgation of Measure 31/X, CCAMLR expressly recognized that fisheries should be managed from the time they commence. Prenotification of new fisheries is thus an essential component in preparing for their management, particularly when fishing is targeting species, and, or, a fishing ground that has not previously been fished. Similar considerations apply if there is an intention for the fishery to use a new fishing technique. Thus the objective of Measure 31/X is to collect information on target, as well as dependent, species and to limit catch, effort or both.

In 1993, CCAMLR developed Conservation Measure 65/XII [now Measure 21-02 - CCAMLR (2002b)] to deal with fisheries that were no longer new but for which critical management information on topics such as those outlined in Box 2 remained lacking [Paragraph 8.39 in CCAMLR (1993)]. Following a year of fishing as a new fishery, the fishery becomes an "exploratory" fishery^[252]. Conservation Measure 21-02 allows for application of both CCAMLR’s precautionary approach and the collection of data necessary to move towards full assessment of the fishery and stock(s) concerned, while attempting to reduce the potential for "irreversible change(s)". An attached, and essential element defines a Data Collection Plan as well as the need to produce Research and Fishery Operational Plans (CCAMLR 2002b). The Scientific Committee annually reviews these plans and provides advice on prosecution of the attached fishery.

2.3 History of CCAMLR’s approach to new and exploratory fisheries

CCAMLR’s initial experience with new and exploratory fisheries dealt with a new fishery for crabs (Paralomis spinossisima and P. formosa) around South Georgia (Sub-area 48.3), notified by the United States in 1990. The approach adopted was consistent with that outlined in Conservation Measure 31/X [Paragraphs 6.7 to 6.12 in CCAMLR (1991)] and aimed at setting conservative catch limits while requiring full reporting of information on the fishery’s prosecution to the Scientific Committee. The initial Conservation Measure adopted in 1992 (Measure 60/XI) dealt with these aspects. In the ensuing years, additional refinements provided more detail on data reporting requirements culminating in a final version of the Measure (52-01) for the 2002/03 season (CCAMLR 2002b).

Following a CCAMLR-sponsored Workshop on the Longterm Management of the Antarctic Crab Fishery in early 1993, the Commission [Paragraph 4.25 in CCAMLR (1993)] accepted the Scientific Committee’s advice that an experimentally-based approach should be applied to this fishery [Paragraph 4.14 in SC-CAMLR (1993)]. This approach was geared towards answering specific questions concerning the population dynamics of Paralomis stocks in Sub-area 48.3 in general, and of P. spinosissima in particular. It initially comprised three experimental phases to be conducted over two fishing seasons. These consisted of a survey of crab distribution, a series of depletion experiments and redirecting fishing effort into areas depleted during the second phase [Paragraph 4.14 in SC-CAMLR (1993)]. The latter requirement was subsequently modified in later revisions of the Measure (Measure 75/XII) and culminated in the provisions of Paragraph 5 of Measure 52-02 (CCAMLR 2002b), which stipulate that: "vessels completing the experimental harvest regime are not required to conduct experimental fishing in future seasons".

With the emergence of interest in new and exploratory fisheries for toothfish (Dissostichus spp.) in various parts of the Convention Area, the latter part of the 1990s and recent years, have been characterized by a large number of notifications for such fisheries (Agnew 2000, Sabourenkov and Miller 2004). In addition to regulatory measures setting catch limits and other restrictions (e.g. fishing season and effort limitations) on an areal basis, CCAMLR developed a single general measure (Conservation Measure 112/XV) [Paragraph 8.33 CCAMLR (1996a)] to outline requirements for new Toothfish fisheries in the 1996/97 season. This Measure contained a number of important elements that included, inter alia, procedures to spread fishing effort, both temporally and spatially, defining data reporting requirements, bycatch limits for other fish species, mandatory deployment of CCAMLR International Scientific Observers, and by implication, longline fisheries were linked to a need for seabird incidental mortality mitigation measures (see Section 3).

Measure 112/XV went through a series of significant modifications in the ensuing years. review procedure before CCAMLR authorises fishing. All other fisheries are seen either as "closed" (i.e. not specifically open to fishing as new or exploratory fisheries, or as closed by specific CCAMLR Conservation Measures) or are classed as no longer being exploratory (i.e. since essential management and catch/effort information have become available). The most notable have been mentioned in the previous subsection and are presented in Table 1. The introduction of small-scale research units in 2000 required those prosecuting a Toothfish exploratory fishery to collect data in a more scientifically rigorous way and to spread catch and effort over a number of fine-scale rectangles [Figure 1 on page 57 of CCAMLR (2002b)]. Building on similar principles to those underpinning the exploratory crab fishery outlined above, and using experience gained from an exploratory trawl fishery in Division 58.4.3 [Conservation Measure 144/XVI - CCAMLR [1997a)], the small-scale research unit approach aims at improving data on Dissostichus spp. distribution and abundance from areas where information is limited or absent. Put simply, the approach strives to maximize the data collection potential of fishing vessels while ensuring that unacceptable damage is not inflicted on stocks for which essential management data are missing.

Following the introduction of its new and exploratory fisheries measures, CCAMLR was faced with the need to better define when fisheries are no longer exploratory (i.e. when data for management purposes become sufficient to allow a regulated fishery)[Paragraph in 5.28 SC-CAMLR (1998)] as well as the principles to be applied when fishing is resumed on opening a fishery which was previously defined as "closed" or "lapsed" [Paragraph 8.35 in CCAMLR (1996a)]. The Commission’s subsequent debate focused on the need to review the interrelationship of fisheries development stages, including for new and exploratory fisheries, to ensure that there is a coherent progression from the exploitation of an unexploited resource, through various fishery phases, to a fully- commercial fishery [Paragraph 104 in CCAMLR (1997b)]. After several iterative steps, the first major breakthrough in this process introduced a Unified Regulatory Framework (URF) for CCAMLR fisheries [Section 7 in SC-CAMLR (2000)].

This URF had three key objectives [Paragraph 10.3 in CCAMLR (2000a)] whose objectives are:

• provide clear guidance on the data and information requirements from all fisheries in the Convention Area to support development of management advice by the Scientific Committee in accordance with the precautionary and the ecosystem approaches to fisheries management

• support design of control mechanisms that will enable the collection of data and information for scientific analysis, and aim to ensure that fisheries in the Convention Area do not expand faster than the acquisition of information necessary for the development of management advice and

• streamline annual review and assessment of fisheries by the Scientific Committee and its working groups, in the face of a mounting workload created by the increasing number of fisheries in the Convention Area.

The Framework also fell within the existing regulatory requirements of the relevant Conservation Measures (Measures 31/X and 65/XII), most notably by including prior notification and establishment of Research and Fishery Operational and Data Collection Plans. This approach was then extended to all fisheries, not just those classified as "new" or "exploratory", and no longer relied on defining the stage(s) of fishery development.

The generalized URF requires preparation of a new reference document to be maintained by the CCAMLR Secretariat for each fishery in the Convention Area. Known as the Fishery Plan, this provides a comprehensive summary of information on each fishery, including a list of all attached regulatory requirements. An outline of the Plan’s envisaged functions is provided in Figure 1. It also provides a summary of fishing activity and a summary list of data lodged in the CCAMLR database for the most recent fishing season. Consolidating all this information in one place facilitates the ability of the Scientific Committee and its working groups to plan future work based on data submitted from the fishery and, or, any notifications received [Paragraph 10.5 in CCAMLR (2000a)].

The Commission agreed that in order to provide comprehensive coverage of all CCAMLR fisheries under the URF, a Fishery Plan should be maintained for all fisheries currently active, or which have been active, in the Convention Area. This would simplify the structure of two fishery types to those with, and those without, fishery plans [Paragraph 7.9 in SC-CAMLR (2000)]. In respect of the former, the regulatory and scientific requirements would be specified in the Plan. For the latter, the Commission would be required to establish entry-level conditions; an issue that has already been addressed in the context of new and exploratory fisheries.

Fishery Plans enable the Scientific Committee to develop advice on whether a new assessment of a particular fishery is required and, or, possible. It also allows the Commission to formulate Conservation Measures based on all appropriate information from the fishery. Fishery Plans have been developed for the krill fishery in Area 48 [Appendix D in Annex. 4 of SC-CAMLR (2001)] and the Icefish (Champsocephalus gunnari) fishery in Sub-area 48.3 [Appendix E in Annex. 5 of SC-CAMLR (2001)]. The Commission has subsequently agreed [Paragraph 10.2 in CCAMLR (2001)] that the next step should be to prepare Fishery Plans for other fisheries in the Convention Area. The fisheries given highest priority are those for D. eleginoides in Sub-area 48.3 and Division 58.5.2, Dissostichus spp. in Sub-area 88.1 and C. gunnari in Division 58.5.2.

One important objective of the URF is to streamline the Scientific Committee’s annual review of fisheries. In this regard, the Working Group on Fish Stock Assessment, (WG-FSA) developed a summary table for new and exploratory fisheries notifications [Table 19 in Annex 5 of SC-CAMLR (2001)], which incorporates recent information from all fisheries in the Convention Area. The table includes information on most recently reported catches, notifications of intentions to take part in the fishery and advice about the currency of the most recent assessment for each fishery. Consequently, notification becomes an essential URF component. Further, the WG-FSA fishery summary is a useful complement to the Fishery Plans and its development continues to be a key element in providing guidance to the Working Group itself, as well as to the Scientific Committee, on priorities for future assessment work.

2.4 Evaluation of CCAMLR’s approach to new and exploratory fisheries

A precautionary and positive adjunct to the assessment of new and exploratory longline fisheries has been the Scientific Committee’s efforts to objectively assess the risks attached to the prosecution of such fisheries in certain areas and, or, times in relation to potential incidental seabird mortality [e.g. Paragraph 4.67 in SC-CAMLR (1997)]. This is discussed in Section 3.3.

Sabourenkov and Miller (2004) have noted that "despite the large number of new and exploratory fisheries authorized by CCAMLR, only a relatively small number have been prosecuted" (Figure 2). In this context, the Scientific Committee has agreed that a ‘prospecting default arrangement’ should be put in place in the absence of fishing on, or a formal assessment of, such fisheries [Paragraph 7.6 in SC-CAMLR (2001)]. The management advice provided then has to be qualified as being "multi-year in the absence of surveys or fishery-based research information". For previously notified fisheries, for which notifications continue to be received, the absence of new information means that no new assessments are possible. The Commission has agreed that until new information is received, no further advice on notified, but un-prosecuted, fisheries should be developed [Paragraph 10.3 in CCAMLR (2001)]. Hence, the "prospecting default arrangement" remains the current and only advice. It should be reiterated that not only does non-prosecution of notified fisheries commit administrative resources to the processing of notifications, it serves to erode application of precaution due to growing uncertainty associated with a lack of information on the intended fishery and, or, stock(s) concerned.

3. Minimization of seabird bycatch in longline fisheries

3.1 Introduction

The 1999 FAO International Plan of Action for Reduction of Incidental Catch of Seabirds in Longline Fisheries (IPOA-Seabirds) (FAO 1999) clearly highlights concerns associated with the number of seabirds being incidentally caught and killed by commercial longline fisheries worldwide. The salient aspects of such concerns are: (a) fears over possible negative consequences of mortality levels on threatened seabird stocks, and (b) the possible impact on fishing productivity and profitability. Internationally, governments, non-governmental organizations and commercial fisher associations have all begun petitioning measures to reduce the incidental take of seabirds in such fisheries.

For CCAMLR, seabird bycatch associated with longlining has two important implications. First, many species breeding in the Convention Area, most notably Albatrosses and Petrels, have been affected by longline fisheries during winter months to the north of the Area [Paragraphs 6.7 and 7.3 in SC-CAMLR (1989) and SC-CAMLR (1990) respectively]. Second, the emergence of longline fisheries in the Area, often close to seabird breeding sites, has provided additional impetus to CCAMLR’s efforts to address the problem.

In 1989, CCAMLR became the first international organization to institute seabird incidental mortality mitigation measures. Most of these measures, and CCAMLR’s experiences therewith, provided a background to, or were incorporated into, the IPOA-Seabirds.

3.2 Rationale for CCAMLR seabird incidental mortality mitigation measures

The assessment and avoidance of incidental mortality of Antarctic marine living resources due to fishing has long been an important issue for CCAMLR in the context of the potential "direct impacts" provisions of Article II.(3).(c) (Table 1). As long ago as 1984 the Commission requested Members^[253] to document and report the number, species and, where appropriate, the age, size, sex and reproductive status of any birds or marine mammals taken incidentally during fishing operations [Paragraph 21 in CCAMLR (1984)].

In 1989, demersal longlining was introduced to the Convention Area for the first time. The fishery targeted D. eleginoides (Patagonian toothfish) around South Georgia in the South Atlantic (CCAMLR Statistical Area 48.3). CCAMLR noted with concern that experience elsewhere indicated considerable levels of risk attached to this type of fishery given its attendant potential to cause substantial seabird mortality during fishing operations; a factor compounded by the relative proximity of the fishery to land-based, seabird breeding sites on the Island [Paragraph 24 in CCAMLR (1989)].

Initial and conservative CCAMLR estimates put the number of albatrosses (most Sub-Antarctic species) being killed annually at 44 000 in tuna longline (i.e. pelagic) fisheries alone outside the Convention Area. This estimate was sufficiently high to substantiate claims that observed and serious declines in Albatross populations within the Convention Area could be attributed to this type of fishing activity [Paragraphs 6.7 and 7.3 in SC-CAMLR (1989) and SC-CAMLR (1990) respectively].

CCAMLR went on to note that Australia and Japan had experienced some success in reducing seabird bycatch in tuna longline fisheries. This success was largely attributed to deployment of streamer lines, or ‘tori [bird]’ poles, to deter birds from taking baited hooks close to the surface, particularly when lines were set during daylight. The streamer lines were situated to trail in the water aft of the setting vessel and directly over the fishing line being set. CCAMLR noted that the attendant seabird catch was significantly reduced when tori poles were deployed. In addition, there appeared to be an added benefit in that bait loss was reduced during line setting [Paragraph 7.5 in SC-CAMLR (1990)].

It was therefore agreed that all CCAMLR-sanctioned longline fisheries should be regulated to minimize incidental seabird mortality [Paragraph 5.3 in CCAMLR (1989)]. The initial step was taken in 1989 when CCAMLR adopted Resolution 5/VIII ("Protection of Seabirds from Incidental Mortality Arising From Longline Fisheries"). This Resolution urged all CCAMLR Contracting Parties to investigate and, as soon as possible, introduce measures to minimize incidental seabird mortality associated with longlining in the Convention Area.

The following year, the first CCAMLR Conservation Measure was adopted to prevent, or minimize, seabird incidental mortality associated with longline fisheries in the Convention Area. This Measure (Conservation Measure 29/X), in a substantially revised form, continues to be applied (i.e. Conservation Measure 25-02 (2002)(CCAMLR 2002b). The Measure’s provisions are outlined in Box 3.

3.3 Development of CCAMLR seabird incidental mortality mitigation measures

Since 1991, CCAMLR has continued to develop and, as necessary, revise its seabird bycatch mitigating measures. These measures have been augmented by establishing closed seasons to prohibit fishing within traditional foraging areas at times when the birds are most at risk (e.g. during land-based breeding)[see CCAMLR Conservation Measure 41-02 - CCAMLR (2002b)]. Other regulations prohibit setting longlines during daylight (Conservation Measure 25-02). A complimentary measure (Conservation Measure 25-01) was adopted in 1993 to regulate the use and disposal in the Convention Area of plastic packaging bands used to secure bait boxes in an effort to avoid entanglement by birds and marine mammals in such bands (CCAMLR 2002b).

The key elements of Conservation Measure 25-02 set out requirements for:

• a line weighting regime

• mandatory use of streamer lines

• streamer line specifications

• mandatory night setting of lines

• minimal ship lighting

• rapid sinking of bait

• mandatory use of thawed bait only

• regulation of dumping of trash and offal and

• procedures for handling seabirds caught on longlines.

Both the line weighting regime and streamer line specification possess a number of unique components. These include weight size and spacing on the fishing line, the position and height of the streamer line above the water, a minimum streamer line length and stipulation of streamer line spacing.

Implementation of provisions as complex as those in Conservation Measure 25-02 was never going to be easy. Evaluating the Measure’s performance has also proved difficult. In particular, traditional methods to boost compliance (i.e. imposition of licensing requirements, application of special permit conditions, in-port and at-sea inspection of vessels) have not been sufficient to fully evaluate the performance of individual vessels under the Measure. Nevertheless, port inspections of vessels provide some indication that vessels are able to implement Measure 25-02. For example, a number of vessels have been shown to meet the design specifications ensuring that offal is only dumped over the vessel side opposite to that where longlines are set and hauled [e.g. Paragraph 7.53 in Annex 5 of SC-CAMLR (2000)]. Similarly, most vessels carry streamer lines on board and their designs usually meet stipulated specifications. However, meeting all these conditions in port does not necessarily mean that they will be effectively applied during fishing.

At-sea inspections are similarly limited since they provide only an instantaneous picture of compliance at the time of inspection (Kock 2001). This is an obvious shortcoming when fishing voyages may last two months. An additional limitation is that at-sea inspections are usually carried out when a vessel is not fishing. Therefore, it is difficult to envisage how they will be able to evaluate the efficacy of the seabird mitigations measures being applied or whether in fact bird bycatch is minimized.

Such shortcomings motivated CCAMLR to ask dedicated scientific observers aboard longline vessels to collect the information necessary to evaluate the application of Measure 25-02. On introduction of the CCAMLR Scheme of International Scientific Observation in the 1992/93 season (Section 5.2.3), it became possible to gather such information during "normal" fishing operations. The observation of incidental marine mammal and seabird mortality associated with fishing has thus become a priority task under the Scheme (CCAMLR 2002a). It entails scientific observers collecting data on vessel operations, particularly the setting of fishing lines, deployment of streamer lines, dumping of offal, etc. The deployment of international scientific observers under the CCAMLR Scheme is now mandatory for all longline as well as trawl vessels engaged in Dissostichus and C. gunnari fisheries in the Convention Area. National observers operate in most of the maritime zones under coastal state jurisdiction within the Area.

In 1992 CCAMLR established an Ad Hoc Working Group on Incidental Mortality Arising from Longline Fishing (WG-IMALF) to deal with incidental mortality of animals during fishing operations on a more formal basis. In 2001, this Group’s name was changed to the Ad Hoc Working Group on Incidental Mortality Associated with Fishing (WG-IMAF). The group’s terms of reference include reviewing data collected by scientific observers on seabird bycatch as well as implementation and monitoring of the performance of CCAMLR seabird incidental mortality mitigation measures.

Since 1992 CCAMLR has undertaken annual assessments of seabird-related measures and of the potential impact of longline fisheries in the Convention Area on seabird populations. Results from these assessments have been used to review and amend Measure 25-02 and its predecessors (Table 2). Further CCAMLR has used scientific observer data on seabird bycatch for compliance-related purposes. In 2002 CCAMLR established a special Joint Assessment Group (JAG) with membership that comprised both enforcement (from the CCAMLR Standing Committee on Implementation and Compliance) and scientific specialists (Scientific Committee) [Paragraph 8.12 in CCAMLR (2002c)]. One of JAG’s major tasks is to develop methods to evaluate the compliance of individual vessels with complex Conservation Measures such as 25-02.

Illustrated in Table 2 is Measure 25-02, which has been revised a number of times since its inception in 1991. The first revision in 1995 elaborated on the general provision for baited hooks "to sink as soon as possible" after entering the water during line setting. Weight specifications were developed for the "Spanish" longline system, which stipulated that weights should be a minimum of 6 kg and be spaced on the line at intervals of no more than 20 m [Paragraph 3.49 in SC-CAMLR (1995) and Conservation Measure 29/XIV in CCAMLR (1995a)]. The modification was based on results from limited experiments on, and modelling of, line-sinking rates. Subsequent analyses indicated low levels of compliance with this particular provision [Paragraph 7.58 in Annex 5 of SC-CCAMLR (2000)]. A key practical consideration was the time taken to place multiple weights at short intervals on the line. In addition, 20 m spacings do not adequately allow for undulations in bottom topography, and there is an increased tendency for lines to become tangled during both setting and hauling. Such limitations necessitate slower setting speeds and mother lines need to be heavier [Paragraph 7.143 in Annex. 5 of SC-CAMLR 2000). Consequently, the weight requirements and interval settings were modified to 8.5 kg and no more that 40 m intervals in 2000 (CCAMLR 2000b). The levels of compliance immediately increased [Paragraphs 7.77-7.80 in Annex. 5 of SC-CAMLR (2001)].

TABLE 2
Development of CCAMLR Conservation Measures (CM) to mitigate incidental seabird catch during longline fishing in the Convention Area

The requirement to set lines only at night has been reiterated and clarified since CCAMLR first initiated it seabird bycatch mitigation measures. This is a relatively simple provision aimed at avoiding the setting of lines at times when the bait is clearly discernible (i.e. during daylight) or when many species, particularly White-Chinned Petrels, are active (i.e. during twilight). Initially, uncertainty over the exact "local" time associated with ambient light levels at nautical twilight resulted in some non-compliance. The problem was resolved by incorporating a cross-reference to nautical almanac tables in Conservation Measure 25-02 to define local twilight times as a function of latitude and time of the year (Box 3). This arose out of earlier, and unsuccessful, efforts to provide simple tables in the Measure itself stipulating twilight times throughout the Convention Area by selected degrees of latitude [Conservation Measure 29/XV in CCAMLR (1996b)].

Conservation Measure 25-02’s line weighting and night fishing requirements were further refined in 2002. As a result, Conservation Measure 24-02 set out experimental protocols to evaluate line sink rates as a function of line weighting [Conservation Measure 24-02 in CCAMLR (2002b)]. Fulfilling these requirements sets the precondition for granting exemption from night setting in specific areas (Sub-areas 48.6 [south of 60^oS], 88.1 and 88.2, and Division 58.4.2) under paragraph 3 of Measure 25-02. Scientific observers are responsible for implementing, and reporting on, such evaluations.

The provisions addressing offal dumping were modified a number of times (Table 2). Initially (1991), dumping of offal was prohibited during fishing. This requirement was later modified to ensure that dumping, if unavoidable, was done on the opposite side of the vessel to where lines were being set (CCAMLR 1994) and to ensure that dumping was confined to periods of line hauling only (CCAMLR 1997a). In 2000 (CCAMLR 2000b), CCAMLR finally decided that vessels unable to process offal on board, or discharge it as required, should not be authorized to fish in the Convention Area.

The requirement to use streamer lines during longline deployment was introduced in 1992 after Conservation Measure 29/X (a predecessor to Measure 25-02) had been in force for a year. This requirement has remained essentially unchanged since that time with some allowance for more flexibility in streamer line design in 1995 (CCAMLR 1995a).

As already noted, CCAMLR seabird bycatch measures have been augmented by defining periods of the year (i.e. seasons) during which longlining is permitted. In 1997, WG-IMALF carried out a comprehensive analysis of relationships between time of year and the attached risks for enhanced Albatross and Petrel mortality resulting from longline fishing in the Convention Area. Results indicated that moving the opening of the Toothfish longline fishing season from 1 March to 1 May lead to substantial benefits (particularly until such time that all vessels comply with night-time setting and streamer line requirements)[Paragraph 4.61 in SC-CAMLR (1997)]. CCAMLR agreed to delay the commencement of longline fishing until 1 April in 1998 with a compromise date of 15 April being subsequently agreed to for the following season. Since 1999 the stipulated opening date has been 1 May for most longline fisheries in the Atlantic and Indian Ocean with the season closing on 31 August (Kock 2001).

Longline fishing seasons in the Pacific Ocean Sector (CCAMLR Statistical Area 88) (e.g. high-latitude fisheries in Ross Sea) have been defined taking into account seabird distribution and periods of darkness available in which to set gear. For example, the fishing season in the Ross Sea south of 65ºS was moved from 15 to 1 December for the 1999/2000 season (CCAMLR 1999a). Daylight setting has been made subject to the results of line-weighting trials to demonstrate that vessels are able to comply with a line sink rate of >0.3 m/sec (see discussion above in respect to Conservation Measure 24-02). Any vessel catching a total of three or more birds during a single fishing season is required to revert immediately to night setting in accordance with paragraph 3 of Measure 25-02 (CCAMLR 2002b) As noted, the carrying of onboard international scientific observers remains mandatory for all vessels.

3.4 Evaluation of CCAMLR seabird incidental mortality mitigation measures

Over the past five years the seabird bycatch and bycatch rate in regulated fisheries in the Convention Area has been significantly reduced. This is attributed to a combination of improved compliance with seabird bycatch measures and by delaying commencement of fishing until the end of the breeding season of most Albatross and Petrel species (Kock 2001). By 2001, regulated longline fisheries in the Convention Area exhibited negligible levels and rates of seabird bycatch in Sub-area 48.3, low levels in the South African EEZ in Sub-areas 58.6 and 58.7 and no incidental mortality in Sub-areas 88.1 and 88.2 for four successive years. In 2002, the Scientific Committee noted that, based on reported data, levels of seabird bycatch in the Convention Area had been the lowest ever recorded [Paragraph 5.3 in SC-CAMLR (2002)] (Figure 3).

In addition to implementing measures to minimize seabird bycatch in regulated fisheries, CCAMLR also considers WG-IMAF’s advice on seabird bycatch associated with proposed new and exploratory fisheries. Each year, WG-IMAF reviews proposals from such fisheries and, taking into account the potential risk of seabird bycatch in each area concerned, recommends an appropriate application of mitigation measures. A particular consideration specifically addresses fishing season restrictions and, or, night setting requirements.

CCAMLR, in keeping with its growing concern (Sabourenkov and Miller 2004), recently endorsed the Scientific Committee’s and WG-IMAF’s view that IUU fishing in the Convention Area, combined with seabird bycatch in fisheries adjacent to the Convention Area, constitute the main threats to many seabird populations in the Southern Ocean [e.g. Paragraph 6.8 in CCAMLR (2002c)]. Estimates of potential bycatch levels associated with IUU fishing in each of the past seven years are presented in Figure 3. CCAMLR concluded that such mortality levels remain unsustainable for populations of albatrosses, giant petrels and white-chinned petrels breeding in the Convention Area with many of these species declining at rates such that extinction is possible [Paragraph 6.98 in Annex 5 SC-CAMLR (2002)]. CCAMLR remains gravely concerned with the present situation - a compelling incentive for stricter measures to combat IUU fishing (Kock 2001).

3.5 Future development of CCAMLR seabird incidental mortality mitigation measures

CCAMLR Conservation Measure 25-02 comprises a few essential elements. The most pervasive strive to reduce the probability that foraging seabirds encounter bait on the surface during the setting of fishing lines. However, no single mitigating measure is likely to eliminate all seabird mortality during longline operations. An important factor remains the setting of bait that avoids visual detection. Underwater setting of lines offers one way to achieve this. The bait would be unlikely to be detected and seized, even by birds with highly developed visual acuity. Other topics for investigation include internal weighting of fishing lines to promote rapid sinking and deployment of "stealth" (dyed) bait to reduce detection.

In 2000, the Scientific Committee advised CCAMLR that once full compliance with seabird bycatch measure is attained, together with negligible levels of seabird bycatch, relaxation of the extent of closed season provisions could be introduced in a stepwise fashion, provided that any consequences were carefully monitored and reported [Paragraph 4.42 in SC-CAMLR (2000)]. CCAMLR would need to ensure that fishers comply fully with the other seabird bycatch measures, such as offal discharge, streamer line deployment and night setting. Refining and enhancing compliance with the line-weighting regime for the Spanish longline system remains a priority [Paragraph 6.11.(i) of CCAMLR (2002c)]. The development of a line-weighting regime for autoline systems also requires additional encouragement [Paragraph 6.16.(iii) in CCAMLR (2000c)].

The search for new and effective measures to enhance avoidance of seabird bycatch continues (IFF 2002). Underwater setting and hauling of lines, more effective line weighting and deployment of "stealth" bait are avenues for future research and development. In particular, the need for vessel design modification has prompted CCAMLR to draw the attention to a requirement that designs of new or replacement vessels should take account of the following to ensure, or facilitate, reduced levels of incidental seabird mortality during longline fishing [From Paragraph 6.84 in Annex 5 of SC-CAMLR (2000)]:

• deploying longlines from an optimal position relative to the location and rotation direction of the vessel’s propeller

• equipping vessels with fish meal plants to process fish offal on board

• incorporating special attachment points for streamer lines into the design of vessels

• adopting a through-the hull line setting and hauling capability in vessel design and

• locating deck lights to minimize illumination astern as well as to meet safety requirements.

CCAMLR continues to review its seabird bycatch measures on an annual basis and these measures constitute a significant component of the IPOA-Seabirds (FAO 1999).