Finalisation of Harvest Strategies for AFMA's Small Fisheries

Finalisation of Harvest Strategies for AFMA’s Small Fisheries Natalie A. Dowling David C. Smith Anthony D.M. Smith

December 2007

Australian Fisheries Management Authority Project number 2007/834

Finalisation of Harvest Strategies for AFMA’s Small Fisheries

Natalie A. Dowling David C. Smith Anthony D.M. Smith

Finalisation of Harvest Strategies for AFMA’s Small Fisheries Dowling, N. A., Smith, D. C. and Smith, A.D.M. Published by CSIRO Marine and Atmospheric Research © Australian Fisheries Management Authority and CSIRO Marine and Atmospheric Research (2007) This work is copyright. Except as permitted under the Copyright Act 1968 (Cth), no part of this publication may be reproduced by any process, electronic or otherwise, without the specific written permission of the copyright owners. Neither may information be stored electronically in any form whatsoever without such permission. Preferred way to cite this publication: Dowling, N.A., Smith, D.C. and Smith, A.D.M. (2007). Finalisation of Harvest Strategies for AFMA’s Small Fisheries. Final report for Project 2007/834 to the Australian Fisheries Management Authority, Canberra.

Executive Summary In December 2005 the Australian Government Minister for Fisheries, Forestry and Conservation issued a Ministerial Direction to the Australian Fisheries Management Authority (AFMA) under Section 91 of the Fisheries Administration Act 1991 (FA Act). The Ministerial Direction included a requirement for the development of a harvest strategy policy for relevant Commonwealth fisheries and the implementation of harvest strategies consistent with that policy in all Commonwealth fisheries by 1 January 2007 (subsequently amended to 1 January 2008). This project was developed to assist with the successful finalisation of separate harvest strategies for six minor Commonwealth fisheries. The project continues from the FRDC project “Development of Harvest Strategies for AFMA’s ‘Small’ Fisheries" and was developed in conjunction with senior AFMA staff as a continuation of the momentum gained during the previous project. Harvest strategies were developed for the following Commonwealth fisheries: •

Bass Strait Central Zone Scallop Fishery

•

Western Deepwater Trawl and North-West Slope Trawl Fisheries

•

Arrow Squid Fishery

•

Coral Sea Fishery

•

Australian Skipjack Fishery

These fisheries are generally data-poor, have low GVP, and/or are still in an exploratory phase. Given their lack of available information, quantitative stock assessments can not be undertaken for these fisheries. The following overall design principles were therefore applied when developing harvest strategies for these small fisheries. The harvest strategies must be: •

Pragmatic (given the economic and data limitations)

•

Unambiguous

•

Cost effective

•

Transparent

•

Easy to understand for all stakeholders

•

Precautionary

•

Consistent with the intent of the Policy (in the absence of direct or proxy estimates of biomass)

•

Be adaptive - note that decision rules can be changed as more information becomes available

•

Where appropriate does not constrain industry development

•

Incorporate existing management arrangements where appropriate

•

Have a formal mechanism to review the success or otherwise of the harvest strategy.

This report details the general approach taken for developing harvest strategies for these small fisheries, and contains stand-alone sections describing the finalised harvest strategies for each fishery1. The report also includes a section providing some general guidance for undertaking the subsequent review process of the harvest strategies. Harvest strategies for each fishery are briefly summarised below: Western Deepwater Trawl and North West Slope Trawl Fisheries o The Western Trawl Harvest Strategy covers both the Western Deepwater Trawl and North West Slope Trawl Fisheries off the coast of Western Australia. These fisheries have been combined into one Harvest Strategy for efficiency as they share similar stakeholders and fishing industry members. It is appropriate to use a number of methods in the HS to control expansion of the fishery and be precautionary in the absence of much biological or economic data. o Data to estimate either biomass, MEY or MSY are not available for any key commercial species in the strategy. In the absence of this information, the strategy assumes that the current very low fishing levels are sustainable and focuses on detecting impacts of increases or changes in effort. o There is concern regarding the level of latent effort in the fisheries as they are indirectly managed by economics currently. Economic factors such as fuel costs have prevented full-scale fishing to date and will likely continue to do so for the foreseeable future. There is potential that if a species of value is found, or if other factors change (e.g. the Northern Prawn Fishery buyout), effort could increase rapidly. Nonetheless, the harvest strategy aims to ensure that any sudden expansion will not lead to overfishing. o Reference points for key commercial and high-risk species for each fishery have been developed and applied separately to functional management zones. In addition, areas for permanent spatial closures have been identified. o The data collection protocol and decision rule triggers are based on the management framework that was developed for the Great Australian Bight Trawl. The WDWTF and NWSTF have similar characteristics to this fishery in that they are multiple species trawl fisheries operating over large areas. Bass Strait Central Zone Scallop Fishery o The harvest strategy takes into account that the scallop fishery does not conform in a straightforward way to the biological assumptions underlying the classical approach to reference points. It has naturally sporadic and fluctuating abundance and 1

It should be noted that in some cases, there are minor details that still require resolution. These were unable to be resolved during MAC/RAG meetings due to the time limitations and the extent of discussion necessary to obtain stakeholder and MAC/RAG approval of more fundamental aspects of the harvest strategies. AFMA is aware of these details and they are highlighted where they occur in the report. 5

intermittent recruitment, such that the concept of a static B0 is not appropriate. The scallops aggregate in sub-populations (scallop beds) which come and go and have historically resulted in a boom-and-bust fishery, so maintaining the fishery at a fixed BTARG level is inconsistent with the nature of the species. In addition, the ability of fishers to effectively fish out a scallop bed once opened to fishing implies that the overall stock level can rapidly decrease from BTARG to BLIM.. o The fishery was closed by Ministerial Direction for at least three years (2006-2008). Surveys will need to provide evidence that stocks are capable of sustaining fishing in order to reopen the fishery. Surveys will also be used in ongoing assessment of whether to fish and to what levels once the fishery is open. o The fishery requires spatial management due to its patchy distribution and irregular, episodic recruitment. However, detailed spatial management has a high information requirement. o The strategy advocates closed area spatial management of commercial scallop stocks, where the majority of the fishery is closed to fishing and only discrete beds are open to harvesting. This contrasts with the current strategy where all areas are open but selected beds are closed. Arrow Squid Fishery o The proposed harvest strategy is designed to have minimal impact and costs if the fishery remains at current catch and effort levels, but trigger reference points invoke decision rules if the fishery expands (e.g. as a result of market changes). o The harvest strategy will also enable the fishery to exploit and capitalize on a “boom” season – i.e. a season with high squid availability, where high take within that season does not have an adverse effect on sustainability. o This strategy is based on a system of real-time within-season management. Generally the assessment approach is one of undertaking spatial and non-spatial depletion analyses, with a view to determining either season length or total catch or both for the season. o Reference points were defined based on recent catch history, with values well below historical high catch levels. These serve as checks allowing controlled expansion, where progressively more information and analysis is undertaken as fishing levels increase. More robust reference points can be developed in light of improved understanding of the dynamics. o Reference points distinguish between the different fishing methods in the SSJ, SET and GABT. Reference points are generally in line with the apportionment policy for Arrow Squid.

6

Coral Sea Fishery - Aquarium o This is a developmental fishery with temporally variable species composition and tight input controls with no qualitative or quantitative estimates of target reference points in terms of maximum economic yield. o The strategy has assessed the fishery as being low risk and uses existing management arrangements with additional reference points for further analysis of data before significant increase in effort occurs. As there are over 500 species caught in the fishery, the strategy uses key functional groups rather than individual species. Coral Sea Fishery – Sea Cucumber o The strategy has a total combined TAC, TACs on the 5 main species, and separate reference points for each of the remaining species, with the aim of mitigating against the potential for overfishing or depletion through this combination of control rules. The TACs are set conservatively acknowledging the current overfishing and the susceptibility of sea cucumber species to localised depletion. Move-on provisions augment the TACs in ensuring low risk of localised depletion. o The fishery is equivalent to a “Tier 5” fishery – that is, one with minimal information. The existing management arrangements were modified to be more precautionary. Spatial closures and move-on provisions are used due to the relatively sessile nature of the species. Coral Sea Fishery – Lobster and Trochus o The small size of the sub-fishery, its low GVP and the apparent effectiveness of existing management arrangements negate the need for a more complicated harvest strategy. The fishery is effectively a “Tier 5”. o The existing management arrangements were modified to be more precautionary and decision rules were assigned to trigger catch levels. Spatial closures and move-on provisions were important given the relatively sessile nature of the species and, in the case of trochus, highly localised recruitment. Coral Sea Fishery – Line, Trawl and Trap o The agreed framework is a general approach where reference points are set to detect changes in species composition, spatial fishing patterns, declines in overall CPUE and overarching values for total catch. Separate triggers for vulnerable species are included. This strategy is advocated as being a general framework that should identify changes in the fishery without having to nominate key species. This is appropriate due to the developmental nature of the fishery. o The strategy states that as the fishery develops, a clearer identification of key species and their sustainability in given areas should be obtained. When this occurs, the harvest strategy is likely to incorporate species-specific reference points and decision rules. Thus

7

the harvest strategy framework can be considered evolutionary in the face of the current developmental state of the fishery. Skipjack Fishery o The proposed harvest strategy outlines a monitoring approach that underpins a series of skipjack catch triggers invoking decision rules. These triggers pertain to various levels of total skipjack catch, and represent a staged approach whereby decision rules assign progressively higher analysis requirements as higher trigger values are reached. o This strategy aims to enable the fishery to expand in a controlled manner by minimizing limitations placed on fishing activities at lower trigger levels while seeking an improved understanding of local stock status in order to better inform appropriate values for the higher trigger levels. In the overall discussion section, the process of harvest strategy development is evaluated and general principles for the pragmatic development of harvest strategies for small, datapoor, low-value fisheries are considered. These include i) the development of sets of precautionary triggers for key species, with progressively higher data and analysis requirements at higher trigger levels, ii) the use of spatial management, either as the main aspect of the harvest strategy, or an augmentation alongside other measures, and iii) identifying data gathering protocols and subsequent simple analyses that enable the fishery to move to a higher level of information and hence a more robust assessment. The importance of an iterative approach to developing harvest strategies involving stakeholders and managers is emphasized.

8

Table of Contents Executive Summary

4

Introduction

14

General aspects

16

Methods

19

Western Deepwater Trawl Fishery; North-West Slope Trawl Fishery

25

Overview of the fisheries

25

Western Deepwater Trawl Fishery

25

North West Slope Trawl Fishery

25

Review of the current management of the fishery

26

Western Deepwater Trawl Fishery

26

North West Slope Trawl Fishery

26

Proxies against the Harvest Strategy Policy Reference Points

26

General description of the harvest strategy

27

Fishery Issues; justification for approach

27

Data collection

28

Harvest Strategy Overview

29

Decision Rules (see subsequent section for annotated version

31

with additional explanation and rationale) Management controls for ERA high-risk species

31

Management controls for the suite of commercial species

32

Consistency with Harvest Strategy Policy

33

Process for review

34

Appendix - Annotated description of Triggers and Decision Rules

35

and additional Harvest Strategy details Bass Strait Central Zone Scallop Fishery (BSCZSF)

45

Overview of the fishery

45

Review of current management of the fishery

45


46


47


47


48


49

9

with additional explanation and rationale) Consistency with Harvest Strategy Policy

51

Process for review

51


53

Worked example of calculating BLIM proxy

59

Survey design

61

Background notes from initial harvest strategy discussions

62

Arrow Squid Fishery (Nototodarus goulidi)

65


65


65


66


66


66


67


68


70

Process for review

71


73

and additional Harvest Strategy details Coral Sea Fishery Hand Collection Sub-fishery: Aquarium

80


80


80


81


82


82


82


83


84

Process for review

84


86

and additional Harvest Strategy details

10

Supplementary analysis

88

Coral Sea Fishery Hand Collection Sub-fishery: Sea Cucumber

90


90

Review of current management of the fishery

90


91


92


92


92


93


93

Process for review

93


95

and additional Harvest Strategy details Coral Sea Fishery Hand Collection Sub-fishery: Lobser and Trochus

97


97


97


98


98


98


99


99


99

Process for review

100

Annotated description of Triggers and Decision Rules

101

and additional Harvest Strategy details Coral Sea Fishery Line, Trawl and Trap Sub-fisheries Overview of the fishery

103 103

i) Line sub-fishery (auto-longline, demersal longline, other line)

103

ii) Otter trawl sub-fishery

104

iii) Finfish trap sub-fishery

104


11

104

i) Line sub-fishery (auto-longline, demersal longline, other line)

104

ii) Otter trawl sub-fishery

105

iii) Finfish trap sub-fishery

105


106


106


106


107


108


112

Process for review

113


114

and additional Harvest Strategy details Worked examples

120

Background to HS development: October 2006 stakeholder

124

meeting outcome Skipjack Fishery (Katsuwonus pelamis)

127


127


128


129


129


129


130


131


133

Process for review

133


135

and additional Harvest Strategy details Background to HS development: December 2006 stakeholder

140

meeting outcomes General points regarding the review process

144

Discussion

146

12

Acknowledgements

149

References

150

List of acronyms

153

13

Introduction In December 2005 the Australian Government Minister for Fisheries, Forestry and Conservation issued a Ministerial Direction to the Australian Fisheries Management Authority (AFMA) under Section 91 of the Fisheries Administration Act 1991 (FA Act). The Ministerial Direction included a requirement for the development of a harvest strategy policy for relevant Commonwealth fisheries and the implementation of harvest strategies consistent with that policy in all Commonwealth fisheries by 1 January 2007. On 11 October 2006 the Minister for Fisheries, Forestry and Conservation advised AFMA of an extension to the previous 1 January 2007 deadline for full implementation of the Commonwealth Fisheries Harvest Strategy Policy (HSP). The Minister advised that full implementation of harvest strategies consistent with the government’s revised HSP would now occur from 1 January 2008. In making this decision, the Minister recognised the practical difficulties of developing and implementing harvest strategies under the original 1 January 2007 timeframe, particularly given the final HSP was not yet completed by the original date. AFMA convened a workshop in late January 2006 for RAG and MAC chairs to consider the ‘Securing our Fishing Future’ Ministerial announcement, particularly the implementation of harvest strategies. An outcome of this workshop was the suggestion of an ‘over-arching’ group to provide advice on implementation. AFMA recognized that the smaller Commonwealth fisheries may not have the capacity to develop harvest strategies in the time available and would need assistance to do so. These harvest strategies needed to meet both ecological and economic objectives. A harvest strategy framework (HSF) had been developed and successfully implemented in the Southern and Eastern Scalefish and Shark Fishery (SESSF) during 2005 (Smith and Smith 2005). The development of the HSF in the SESSF was driven by a small core team that provided a combination of technical expertise on harvest strategies and fishery management experience. The basic framework was developed quite quickly, presented at various stages to management group meetings and the AFMA Board, and then implemented by the stock assessment team and the Resource Assessment Groups (RAGs). The effectiveness of this approach resulted in the formation of a similar core team to help formulate harvest strategies for small fisheries. The emphasis is on an iterative approach via direct engagement with stakeholder groups at management meetings. Consequently, AFMA funded a project ‘Development of Guidelines for the Commonwealth Harvest Strategy Policy and Development of Harvest Strategies for AFMA’s ‘Small’ Fisheries’. The Objectives of the project were to: 1. Develop and write guidelines for the Commonwealth Harvest Strategy Policy 2. Develop and document harvest strategies for AFMA’s small fisheries The guidelines were completed as a separate report. Dowling and Smith (2007) presented harvest strategies for the “small fisheries” primarily in the form of initial harvest strategy frameworks (i.e. progress reached approximately stage i) in Figure 2 below). While clear frameworks were developed, it was clearly acknowledged that various details and/or absolute values for reference points were not refined and/or finalized

14

in many instances, so that for each fishery a range of outstanding issues was yet to be addressed at the time of that project’s conclusion. The main priority for this (second stage) project was the finalisation of the harvest strategies, addressing the outstanding issues and ensuring that the harvest strategies were adequately detailed and defined for the January 1, 2008 implementation date. The project objectives were also to provide guidance to AFMA’s fishery managers, MACs and RAGs on how to operationalise the harvest strategies for these fishery-specific harvest strategies, and to assist with planning for longer-term management reviews and design of protocols for collection of data to inform and improve the harvest strategies. A harvest strategy is a plan that sets out the management actions necessary to achieve defined resource objectives in a given fishery. A harvest strategy should specify: •

a process for monitoring and conducting assessments of the biological and economic conditions of the fishery; and

•

rules that control the intensity of fishing activity according to the biological and economic conditions of the fishery (as defined by the assessment). These rules are referred to as control rules (sometimes also known as harvest control rules or decision rules).

However, these small fisheries are often data-poor, have low Gross Value of Production (GVP), and/or are still in an exploratory phase. Some do not have Resource Assessment Groups (RAGs), and their assessments (if any) are more closely aligned with Tier 4 (or lower) of the SESSF harvest strategy framework2 (Smith and Smith 2005), which is based on trends in catch rates only. These factors were considered explicitly in developing harvest strategies for these small fisheries. The Harvest Strategy Policy is underpinned by a target and a limit reference point, equating to BMEY (or a proxy, equivalent to 1.2BMSY)), and 0.2B0, respectively (Australian Government 2007a). However, estimates of the status of stocks relative to these biomass-based reference points are simply not available for the small Commonwealth fisheries. Further, due to their low GVP, it is unlikely that reliable biomass estimates on which to base control rules will become available in the foreseeable future. Thus the challenge has been reconciling the Harvest Strategy Policy reference points against the reality of the available data/assessments (both current and future) and the low GVP for these small fisheries. Clearly there is a trade-off between setting conservative triggers/reference points in the face of uncertainty, against the cost associated with gaining more information that is likely to allow a higher exploitation level to be set. However, the low GVP for these fisheries limits the extent to which data are able to be obtained and/or 2

The tiered approach developed for the SESSF is designed to provide for an increased level of precaution in association with increasing levels of uncertainty about stock status, such that the level of risk is approximately constant across the tiers. In this approach, each species is assigned to one of 4 Tier levels depending on the amount and type of information available to assess stock status, where Tier 1 represents the highest quality of information available (e.g. a robust stock assessment). Consistent with the above, target exploitation rates will decrease as Tier levels increase. Typically, Tier 3 fisheries obtain estimates of fishing mortality from catch curves derived from catch-atage/catch-at-length data, while Tier 4 fisheries calculate recommended biological catches from temporal trends in catch-per-unit-effort. 15

more robust assessments undertaken. Therefore precautionary triggers should be set until further monitoring and analyses have been completed. For small fisheries still in a developmental/expansion phase, with a very low risk that overfishing is occurring, harvest strategy reference points should be set in such a way to allow for continued controlled development. However, checks should be in place to detect and respond to significant changes, such as fishery expansion, contraction, changes in species composition and/or spatial shifts. In this report, harvest strategies are developed for the following Commonwealth fisheries: • • • • •

Bass Strait Central Zone Scallop Fishery Western Deepwater Trawl and North West Slope Trawl Fisheries Arrow Squid Fishery Coral Sea Fishery (with separate harvest strategies for the i) Aquarium, ii) Sea Cucumber, iii) Lobster and Trochus and iv) Line, Trawl and Trap Sub-Fisheries) Australian Skipjack Fishery

This report builds on the frameworks developed in the first project (i.e. since May 2007). Within each harvest strategy for each small fishery, it fleshes out the framework by providing more detail, attempting to resolve outstanding issues, assigning absolute values against trigger/reference points, and more explicitly articulating the details and justification of the decision rules. Basic desktop analyses were undertaken to resolve some of the above and the results are given where applicable. For each fishery, the justification for, and, where relevant, history of the development of the approach are clearly provided, noting that, in some cases, the harvest strategy approach was significantly revised in subsequent stakeholder meetings from that developed in the first project. The consistency of the approach with the intent of HS policy is directly considered and the proxies for the HSP reference points are explicitly defined. Finally, initial consideration is given to priorities for the review process for each of the harvest strategies. General aspects Small fisheries are typified as being data poor, with only basic or no assessments undertaken, and have low GVPs. They may be in a development/expansion phase or opportunistic given sporadic stock availability. There may be only a few active operators in the fishery, although there may also be a substantial amount of latent effort. A substantial challenge of the current and previous projects was reconciling the Harvest Strategy Policy reference points and risk criteria against the reality of the available data/assessments (both current and future). The Harvest Strategy Policy is geared around the assumption of available estimates of biomass, but, as noted above, these are not available for the small Commonwealth fisheries. Further, due to their low GVP, it is unlikely that reliable biomass estimates will become available in the foreseeable future. Additionally, for highly variable fisheries such as the Bass Strait scallop fishery, it is difficult to reconcile their population dynamics with the intent of the Harvest Strategy Policy, particularly that the frequency of breaching the limit reference point should not exceed one year in ten. As such, direct proxies for the HSP target and limit reference points are, at least in the interim, substituted with indirect triggers or reference points. To the greatest extent possible, these

16

are consistent with the intent of the Policy reference points while acknowledging the lack of knowledge of the stock status, or variable dynamics with those implied by the definitions of the Policy reference points. When developing harvest strategies for these fisheries, the approach was one of scientific good sense and pragmatism, incorporating existing management arrangements where appropriate, rather than relying on traditional and/or sophisticated fishery assessment techniques. In some cases, it was possible to adapt existing approaches (e.g components of the Great Australian Bight Trawl HS for the SESSF were adapted in the context of the Western Deepwater Trawl and North West Slope Trawl Fisheries and the Coral Sea Line, Trawl and Trap sectors). In some instances, there was a precedent for a given approach or analysis technique that had successfully been applied in similar fisheries nationally or worldwide (e.g. depletion analyses to manage squid fisheries; spatial management for scallop fisheries). Otherwise, the general process was to set precautionary triggers based on catch history with decision rules that aimed to improve the knowledge of the fishery and hence provide a basis to develop improved harvest strategies based on more sophisticated assessments in the future. In this context, the project team developed pragmatic harvest strategy frameworks that are designed to be precautionary and consistent with the intent of the Policy. As part of this process, various suites of approaches have emerged, and these could ultimately form the basis of a set of broad guidelines for harvest strategy development for small fisheries in general. These are discussed in more detail in the discussion section of this report. For all the fisheries, the implications of their low Tier level were emphasised to all stakeholders. Clearly there is a trade-off between setting conservative triggers/reference points in the face of uncertainty (and hence greater risk), against the cost associated with gaining more information that would allow a higher exploitation level (given the improved understanding of stock status and hence relatively less risk) to be set. Given that all of the fisheries could be considered as being at a level of information equivalent to Tier 4-5 (with 5 being minimal information), the harvest strategy approaches are forced to be precautionary by making use of trigger checks and balances (as opposed to direct proxies for HSP reference points). This involves a stepwise process of triggers invoking increasingly higher cost monitoring (over and above that of any pre-existing monitoring protocols) associated with progressively more sophisticated analyses (with the intention that, in reaching the trigger, the GVP may be somewhat higher so that the additional monitoring and analyses will be affordable by the fishery). The improved knowledge gained from the additional monitoring and analyses should ultimately allow the fishery to move to a higher Tier level. An additional challenge was that of educating and engaging the various stakeholder groups, RAGs and MACs. Even for developed and data-rich fisheries, the development of a harvest strategy is not necessarily easy for fishers and other stakeholders to understand. However, in the case of small fisheries, this process was almost entirely foreign to stakeholders, and an especially difficult one given the nature of the fisheries and their lack of familiarity with such fundamental concepts as formal stock assessments. This education of stakeholders was a critical pre-requisite to obtaining their insight and knowledge about practical and operational aspects of the fishery and how the harvest strategies might be made to work. In short, the

ultimate success of the harvest strategy is reliant on a participatory and iterative approach.

17

The indirect imposition of a harvest strategy without proper explanation and engagement would have met with strong resistance from stakeholder groups who were, for the large part, naturally sceptical and uneasy about the concept of a harvest strategy.

18

Methods A harvest strategy is a framework for setting catch and effort limits to achieve management objectives. Key components of a harvest strategy include • • •

a monitoring strategy an assessment process a decision process

The focus is on the management of key commercial species. Central to the Ministerial Direction, harvest strategies should ensure that • •

overfishing ceases currently overfished stocks are rebuilt

Harvest strategies for AFMA’s small fisheries were further developed and finalised by the project team, based on the harvest strategy frameworks developed in stage 1, in an ongoing conjunction with stakeholder groups. This continued to be an iterative process involving discussions between the project team, AFMA and stakeholders. In the first project, a review of the current management levers was undertaken for each fishery, based on the management plan or arrangements for that fishery, the status reports from the Bureau of Rural Sciences (BRS) (McLoughlin 2006), the ERA reports (Smith et al, 2007; Furlani et al., 2006a-2006e), AFMA stock assessment reports, and input from AFMA managers and industry representatives. Following initial scoping discussions with AFMA managers, options for harvest strategies were discussed and developed further at stakeholder group meetings. These revised harvest strategies were summarized in writing and returned to stakeholders for further discussion and refinement. The process is illustrated in Figure 1. The current project flagged the outstanding issues for each harvest strategy and revisited the harvest strategies with a view to resolving these, using the momentum gathered during the previous project. Stakeholder consultation was sought at management group meetings (RAGs, MACs and stakeholder group meetings), while in the interim intensive discussions were held with key AFMA managers, MAC chairs and fishery scientists. The main stakeholder/RAG/MAC engagement undertaken during Stage 2 is listed below (note that this excludes the additional extensive formal and informal discussions with AFMA managers) 3: •

Bass Strait (Central Zone) Scallop Fishery: o ScallopMAC meeting October 3, 2007, Melbourne o ~3 phone conferences with selected MAC members o Presentation to AFMA Board, October 30, 2007, Hobart

3

Project P.I. Natalie Dowling was also invited to give presentations outlining the process of harvest strategy development for small fisheries at a meeting of the Small Pelagic Fishery and at the annual Northern Australian Fisheries Management (NAFM) meeting, and is assisting with the development of a harvest strategy for the NAFM red snapper fishery

19

•

Southern Squid Jig Fishery: o SquidRAG meeting April 2, 2007 o SquidRAG meeting October 10, 2007, Melbourne

•

Coral Sea Fishery: o Stakeholder meeting March 7, 2007, Canberra – line, trawl, trap sectors; lobster and trochus, beche-de-mer sub-sectors o Stakeholder meeting August 2007, Brisbane – line, trawl, trap sectors o 2 Stakeholder phone conferences – Aquarium sub-sector

•

Skipjack Tuna Fishery: o phone conference with Ryan Murphy and Trysh Stone (AFMA) October 2007 o stakeholder meeting November 28, 2007, Canberra

•

Western Deepwater Trawl Fishery/North-West Slope Trawl Fishery: o WestMAC meeting August, 2007, Fremantle o WestMAC meeting November 22, 2007, Fremantle

In attending the RAG/MAC/stakeholder meetings, the key approach of the project team was to •

Not assume prior knowledge of harvest strategies

•

provide advice without being prescriptive or pre-emptive

•

seek maximum feedback and input from all stakeholders during harvest strategy development

As discussed above, developing harvest strategies for small fisheries presents unique challenges. The following overall design features were applied when developing harvest strategies for small fisheries. Harvest strategies must be: •

Pragmatic (given the economic and data limitations)

•

Unambiguous

•

Cost effective

•

Transparent

•

Easy to understand for all stakeholders

•

Precautionary

•

Consistent with the intent of the Policy (in the absence of direct or proxy estimates of biomass)

•

Be adaptive - note that decision rules can be changed as more information becomes available

•

Where appropriate does not constrain industry development 20

•

Incorporate existing management arrangements where appropriate

•

Have a formal mechanism to review the success or otherwise of the harvest strategy.

The technical process for the development of harvest strategies is illustrated in Figure 2. The Harvest Strategy Policy provides the formal rationale, direction and specifications for harvest strategy development. First existing management arrangements for each fishery were summarised and evaluated against the intent of the Policy. The species to which the harvest strategy should apply was determined, consistent with the Policy’s requirement that a harvest strategy must be applied to key commercial species. Available data, analyses and assessments were considered with a view to establishing Tier levels (if possible) and determining the most pragmatic approach. In some cases, additional analyses were completed to better inform the harvest strategy. From this, target and limit reference points were developed that were intended to be consistent with the intent of the Policy. For small fisheries with very little information, reference points were often simply “best guess” proxies, with little knowledge of their relative magnitude with respect to the biomass thresholds to which they were intended to correspond. However, under the data collection protocols and review process developed as part of the harvest strategy, the intention is to ultimately replace these low Tier proxies with more informed (higher Tier) proxies. Once reference points are established, the strategy can be operationalised, with Tier levels being allocated across species within the fishery as appropriate. After the development of initial options for harvest control rules (decision rules), agreement was sought on the general harvest strategy framework. Following discussion and agreement with stakeholders a full harvest strategy, including monitoring, assessment and harvest control rules, was developed around the framework. The harvest strategy must then be implemented via the Resource Advisory Groups (RAGs) and the Management Advisory Committees (MACs) (Smith et al 1999, Smith et al 2001), or stakeholder groups in the absence of a RAG or MAC. Post-implementation, a qualitative evaluation should initially be made as part of the review process for the strategy. However, the robustness and performance of the harvest strategy may also be more formally evaluated using management strategy evaluation.

21

Project team or RAG meet with AFMA managers scope project, expectations, priorities etc.

RAG and/or project team develop initial options for HS

Fishery workshops - AFMA/project team plus key stakeholders (e.g. RAGs)

RAG/project team - draft harvest strategy

Present to MACs/AFMA

Revise as appropriate

Present to AFMA Board

Figure 1 Flowchart indicating the administrative/engagement process for developing a harvest strategy

22

a. Policy setting

b. Management controls

c. Species to be considered

d. Data available

e. Analysis and assessment (existing)

f. Develop limits and targets (consistent with Policy)

g. Operationalise; allocated Tiers if appropriate based on e. & f. above

h. Develop initial options for harvest control rule (HCR)

i. Develop full harvest strategy including monitoring, assessment and HCR

j. Qualitative evaluation

j. Full Management Strategy Evaluation (MSE)

k. Implementation (via RAGs, MACs)

Figure 2 Flowchart indicating the technical process for developing a harvest strategy

23

The format of the report is a series of ‘chapters’ for each fishery or sub-fishery that contain the following: •


•


•


•

General description of the harvest strategy o Fishery Issues; justification for approach o Harvest Strategy Overview o List of Decision Rules

•

Consistency with Harvest Strategy Policy

•

Process for review

The “General description of the harvest strategy” section is intended to provide a succinct summary of the main points of the strategy and a concise description of triggers and decision rules. This would be the first point of reference for managers and stakeholders wishing to consult the document. Each ‘chapter’ also has a section called ‘Annotated description of Triggers and Decision Rules (providing extra explanation and rationale), and additional Harvest Strategy details’. This is effectively an appendix that repeats but expands on the general description. The material is supplemented with more detailed explanations to further clarify and justify the approach taken. In addition, for some fisheries this is expanded to provide extra history and background to the harvest strategy development. Typically this describes previously considered harvest strategy options and provides the rationale for their subsequent rejection.

24

Western Deepwater Trawl Fishery North West Slope Trawl Fishery Overview of the fisheries Western Deepwater Trawl Fishery The Western Deepwater Trawl Fishery (WDWTF) began in 1987 as an extension of the North West Slope Trawl Fishery, as operators extended their exploratory fishing for scampi and deepwater prawns. Following poor crustacean catches, the fishery evolved into a finfish trawl fishery of considerable species diversity (>50 species) (Moore et al. 2007a). It is best classified as a byproduct/mixed fish species fishery due to the wide range of species taken in low volumes. In recent years, bugs have been targeted and now form the majority of the catch. According to logbooks, between a third and a half of the total catch is discarded. Of the discards, about a quarter is unidentified. The fishery is characterized by two distinct areas: north and south of 27°S (Moore et al. 2007a). The fishery is predominantly an off-season diversification for Northern Prawn Fishery (NPF vessels); and as such effort is limited. There is a distinct increase in effort from June-August corresponding to seasonal closures in the NPF. In 2005-06, there was minimal effort in the WDWTF due to economic reasons (Whitelaw pers. comm.). An exploratory cruise in 2006 caught only 150kg during a 2 month survey. The GVP peaked at $2.5 million in 2002/03 but has since declined dramatically: in 2004/05 the GVP was $829000 and in 2005/06 the GVP was only $3000 (Moore et al. 2007a). The status of the WDWTF is classified as uncertain (McLoughlin 2006). The assumption is that current effort levels are sustainable However, many of the demersal species are thought to be long-lived, slow-growing, late to mature and vulnerable because of spawning aggregations, including oreo species and eteline snappers (Furlani et al., 2006b). North West Slope Trawl Fishery The North West Slope Trawl Fishery (NWSTF) was initially a deepwater prawn fishery, but prawns are now taken as a byproduct when targeting scampi. Commercial interest began following confirmation of promising scampi and deepwater prawn stocks by research cruises conducted in 1978, 1982 and 1984, and by an independent industry survey in 1983. The total catch in 2006 was 27t, of which mixed scampi comprised 71% and Australian scampi comprised 23%. The 2005/06 GVP was $679 000, down from $1.7 million in 2004/05; the average GVP between 1997/98 and 2001/02 was close to $1.1 million. The change in catch composition was attributed to seasonal variation in abundance, variation in number of vessels actively fishing between years, market demands, and a decrease in abundance as exploitation has reduced standing stocks (McLoughlin 2006). Between a third and a half of the total catch is discarded. This is generally not identified to species level but may identified using informal groupings (such as “mixed fish”). The status of the fishery is uncertain. There have been declines in the relative abundance of scampi (McLoughlin 2006). Scampi appear to have a low carrying capacity and low resilience to exploitation. Given that deepwater prawns tend to aggregate on specific grounds, they are susceptible to localised depletion (Furlani et al., 2006a).

25

Review of the current management of the fishery Western Deepwater Trawl Fishery Such a diverse range of vessels have operated in the fishery since its inception that it is impossible to characterize vessels, trawl types or fishing methods. There is no formal Management Plan; the fishery is informally managed via limited entry (11 permits of 5 years' duration subject to conditions, although only 7 vessels have operated in recent years). There are no other input controls and no output controls. Management indirectly relies on external regulating factors including the part-time participation and spatial refuge of stocks yet to be efficiently targeted. Main information sources for the fishery have been the AFMA WDWTF logbooks and some State logbooks. A stock assessment has been undertaken only for ruby snapper (Hunter et al. 2002). There is limited information regarding the biological dynamics of the WDWTF resources. Further research is needed to determine the stock status of target species. A strategic data plan to identify gaps in knowledge, improve logbook data collection and gathering basic biological information on target species is also a priority. However, the low GVP of the fishery limits the extent of research that can be undertaken. For both the Western Deepwater and NW Slope Trawl Fisheries, there is an ongoing observer program budgeted for when fishing occurs. This will collect biological data as well as catch and effort, but no funds have been allocated to process or analyse the data. North West Slope Trawl Fishery There is no formal Management Plan; the fishery is informally managed via permit conditions. Permit holders generally access the fishery on a part time or opportunistic basis as an adjunct to other Commonwealth fisheries (3-4 vessels fishing as at 2006). Fishing effort in the NWSTF is generally minimal during April, September & October while vessels operate in the NPF. The fishery is managed via limited entry (7 permits of 5 years' duration) and controls on codend mesh size (maximum mesh size of 50mm to discourage targeting of demersal finfish). There are no other input controls (including no restrictions on headrope length) and no output controls (i.e. no TACs), and no spatial or regional management, or temporal closures. Permits enable only 1 vessel to operate under each at any one time but are fully transferable between vessels. There are 3 MPAs within the area of operation of the fishery, in which commercial fishing is prohibited. Stock assessments were undertaken for the scampi fishery in 1992, 1998, 2000 and 2004 (Lynch and Garvey 2005), but adjustments need to be made for changes in fishing power if estimates of stock status are to be robust. Shot-by-shot catch and effort logbooks were introduced at the beginning of the fishery. There are some limited data from infrequent fishery-independent surveys.

Proxies against the Harvest Strategy Policy Reference Points Clearly it is not possible to set meaningful triggers for each species captured by the fisheries, particularly for Western Deepwater Trawl where catch is opportunistic and species composition is highly variable.. 26

However, by identifying a suite of key species based on industry consultation and historical catch patterns, it is argued that the entire fishery will be represented and hence managed. Additionally, the inclusion of separate triggers for highly vulnerable species, and the setting of permanent spatial closures, renders the strategy is defensibly precautionary. Applying the triggers independently to separate functional management regions acknowledges the distinct sub-regions within the extensive areas over which the fishery is conducted. Given that the fisheries are largely opportunistic, and that species composition has been temporally variable (WDWTF), or has shown a shift in target species (NWSTF), there is as yet no qualitative or quantitative estimates of maximum economic yield. For these low GVP fisheries there is a need to strike a balance between allowing for industry to capitalize on sporadic economic opportunities while still managing the fishery in a precautionary and proactive manner, consistent with the intent of the Harvest Strategy Policy. The inclusion of three levels of values for each trigger facilitates the expansion of the fishery by assigning progressively higher data and analysis requirements with higher trigger values. As such, the risk associated with further expansion is minimized. The two levels of trigger aim to do this by setting the lower triggers level at a value that will detect early changes and result in low-cost analysis to identify the reasons behind these without immediately placing limitations on the fishery. The third trigger level acts a limit reference point in the absence of further information. Should the fishery wish to further expand, it will need to invest in more detailed/robust assessments that will provide stronger justification for continued expansion and upward revision of the trigger point.

General description of the harvest strategy Fishery Issues; justification for approach There is concern regarding the level of latent effort in the fisheries (Whitelaw pers. comm.). Economics (fuel costs) has prevented full-scale fishing to date and will likely continue to do so. There is potential that if a species of value is found, or if other factors change (e.g. the NPF buyout), effort could rapidly increase. Nonetheless, the Harvest Strategy must ensure that any sudden expansion will not lead to overfishing, and it must be low-cost to implement. Fishers have good knowledge of the area and of species availability: The lack of effort is not due to lack of exploration but rather due to opportunistic activity occurring as an adjunct to other fishing interests. It is a major management challenge to develop strategies that provide safety measures in face of resource uncertainty and changing fishery dynamics. These are developing fisheries and harvest strategies should not prohibit controlled expansion and business development, but must ensure that such expansion will not lead to overfishing. Harvest strategies must also acknowledge: • • • •

That the fisheries are conducted over extensive and remote areas The limited number of licences and fewer operational fishers That the fisheries have very low GVPs with little or no research funding. That the fisheries are multispecies, with no consistent target species

27

• • • •

That fishing is opportunistic and thus it is difficult to identify key species about which a harvest strategy can be developed. The lack of data – there is little information for a quantitative analyses There are high levels of discarding, especially in the Western Deepwater Trawl Fishery The environmental impact of trawling on benthic habitats and the potential for this to limit the fishery due to real or perceived threats.

The harvest strategy is focused on a suite of key commercial species and any high-risk species indicated from the Ecological Risk Assessment (ERA) (noting that at the time of this report, the ERA was yet to progress beyond Stage 1, but that dusky shark and dogfish are the main high-risk species) (Furlani et al., 2006a and b, Smith et al., 2006). It was assumed that the subset of species chosen would indirectly control the level of fishing pressure on low value byproduct and bycatch species. Regular reviews of the catch composition from the fisheries should be undertaken to verify this assumption. WestMAC reviewed the species caught in the fishery and derived the following list of key commercial species and the depth range in which they are usually caught. WDWTF: Species Deepwater flathead Boarfish Tang snapper Ruby snapper

Depth 200-250m 200-250m 200-300m 200-300 m

Caught in State? Y Y Y

Flame snapper

200-300 m

Y

Bugs Gemfish Mirror dory Scampi Deepwater prawn Orange roughy

200-300 m 300+ m 300+ m 300-500 m 300-500 m 500-800 m

Smooth dory Alfonsino Bar rock cod Squid NWSTF: Species Scampi Deepwater prawns Squid

500-800 m 500-800 m ?

Notes

North of Geraldton (Shark Bay; South Passage; 26°S) North of Geraldton (Shark Bay; South Passage; 26°S) Found on seamounts

Found on seamounts; off Cape Leeuwin (SW corner) Found on seamounts Found on seamounts Y

300-500 m 300-500 m

Data collection It is important that detailed catch and effort logbook data are collected on a shot-by-shot basis by all operators. AFMA has recently completed an extensive summary of the available logbook data (Moore et al., 2007a and 2007b), which has informed the harvest strategy. It

28

provides standard information on season, position and depth of catches and the species composition of retained and discarded catch. Baseline biological data (e.g. otoliths, length, sex sampling) should be collected on key species as an a high-priority, ongoing requirement, as soon as is practicable. A boat-level sampling protocol is required, to ensure the practical implementation of the data collection program. For example, it makes little sense to sample an atypical single, poor-quality specimen that is collected on what is known to the periphery of its known range or current aggregation location. Similarly, there are logistical limitations to sampling when large catches are being hauled, and the protocol should specify upper limits on the maximum number of samples to be obtained per haul for each species. Note that, for the NWSTF, some logbooks already contain gross size information, which may be useful when designing the sampling protocol for the NWSTF. The data collection protocol should sit alongside the harvest strategy, to facilitate the collection of this biological information. At this point in the fishery, there may be no need to analyse this data, so the cost to the industry should be minimal. However, by collecting this information from the outset and simply archiving it, a time series of critical information on the population biology of key commercial species can be established. These samples and data can be analysed at a later date, should pre-agreed catch/effort triggers be reached (see below). This approach to data collection allows relevant information to be collected cost-effectively without pre-supposing which species may need to be assessed. However, should funding be made available, initial analyses of biological data collected for the identified key commercial species should be undertaken as a priority, so that more robust trigger points may be set. Harvest Strategy Overview The proposed harvest strategy framework is one of a range of triggers for key commercial and high-risk species for each fishery, applied separately to functional management zones. In addition, areas for permanent spatial closures would be identified. This data collection protocol and decision rule triggers was based on the management framework that has been developed for the Great Australian Bight Trawl Fishery (Moore and Knuckey 2007; Australian Government 2007b). The Western Deepwater and NorthWest Slope Trawl Fisheries have similar characteristics to this fishery in that they are multiple species trawl fisheries operating over large areas. The framework is outlined as follows: 1) Define functional management zones. •

Due to the large spatial extent of the fishery, it is likely to encompass a wide range of marine ecosystems.

•

Divide fishery into a number of management zones.

•

Zones do not necessarily have to coincide with stock boundaries, but will assist with targeting management arrangements in the different regions.

•

Western Deepwater Trawl Fishery:

29

o Two latitudinal zones were proposed: north and south of 30°S as suggested by differences in species composition of catch, with triggers and decision rules applied independently. The same trigger values and decision rules are to apply independently within each zone. •

North West Slope Trawl Fishery: o Catch species and magnitude suggest three longitudinal delineations: west of 117°E, 117°E -121°E, and 121°E -125°E. Trigger values for each region to be scaled according to relative historical catch by zone (Garvey, pers. comm.).

2) Identify permanent spatial closures that encompass significant areas of each of the regions highlighted above (which may or may not be in addition to those determined for high-risk species noted below). •

Spatial closures should be implemented that protect habitats identified in the ERA as high risk.

•

This issue will be revisited with stakeholders over the next 12 months for further discussion once habitat mapping has been undertaken and information synthesised. Representative habitats and/or community types should be identified. Specifically, the outcomes of the bioregional marine planning process and the MPA process will be informative. Permanent spatial closures are a priority but the areas need to be identified in an informed and coordinated manner.

•

In the interim, (as at December 2007) expert advice has been sought from industry who have provided coordinates for clearly delineated recommended closure areas.

•

A general description of the proposed WDTDW interim trawl closures are as follows: i) a northern closure area off Carnavon ii) a mid-region closure area off Geraldton iii) a southern closure area off Bunbury in colder waters

3)

Define management controls for ERA high-risk species •

Across both the WDWTF and the NWSTF, the ERA has identified 11 by-product species of dogfish, sharks and chimaeras as being high risk.

•

These is little available information for the remaining 12 ERA high-risk species, which are comprised of teleosts and invertebrates.

•

3 of these remaining 12 species (tang snapper, gemfish and mirror dory), are included in the key species list against which catch triggers are set in 4) below.

•

It is noted that current catches for these 12 species are extremely low across both fisheries

30

4)

Define management controls for the suite of key commercial species •

For each species, three levels of zone-specific catch triggers are to apply. In the absence of any other information and assessments, the value for each trigger is broadly based on (unless otherwise specified) •

Level 1: half the historical high annual catch

•

Level 2: the historical high annual catch, taken in each of two consecutive years.

•

Level 3: double the historical high annual catch

•

For the WDWTF, the same trigger values are to apply in each zone. The values for the NWSTF species triggers will be revised in a zone-specific context, to be scaled in proportion with the relative catch from each of the 3 proposed management zones.

•

The life history and population dynamics of the species were considered in consultations with stakeholders when developing values for the trigger points such that where it was considered that the historical high catch was extremely conservative given the life history and/or population dynamics, a higher level value was set that was still precautionary without being unnecessarily invoked

•

Proposed trigger values for each species are in the table below

Decision Rules (see ‘Annotated version with additional explanation and rationale’ at the end of this fishery section) Management controls for ERA high-risk species Mitigation measures should be formulated for each individual species. The two species for which the most information exists are endeavour dog shark and school shark. Proposed decision rules associated with triggers for dogfish (including Deania, Squalus and Centrophorus) and chimaera species (including Hydrolagus) and Squatina tergocellatai (angel shark – a bottom dweller not known to be migratory) (i.e. species with less propensity to migrate, or species where little knowledge exists regarding movements, but treated in the interim as relatively site attached) (Stevens and Daley, pers. comm.): i)

ii)

Fishery-wide 50 animal move-on provision (i.e. once trigger level exceeded, no one fishes there) (as opposed to catch trip limits which have a high associated potential for unwanted discarding), and spatial (depth) closures (due to relatively low potential to migrate)

Proposed decision rules for school shark and dusky shark (these are more migratory species)

31

i)

Fishery-wide 50 animal move-on provisions (Code of Practice) (i.e. once trigger level exceeded, no one fishes there) (idea is that move-on provisions will mitigate against triggers being reached and thereby minimise unwanted discarding), and trigger values for total catch (as opposed to trip limits per boat): a. intermediate: look at spatial catch pattern and implement temporary seasonal closures, b. limit: no catch once this trigger has been reached The aim is to implement depth closures in appropriate zones as determined by habitat (depends on data and species; won’t work for highly migratory species)

ii)

iii)

NB a move-on provision in itself is not an adequate management measure (difficult to enforce) but it is a sensible measure to augment other rules and help optimise their effectiveness Proposed decision rules for high-risk teleosts and invertebrates Two levels of triggers: i)

Intermediate trigger of 2t for each species •

This value is considered to be precautionary.

•

If reached, investigate spatial distribution of the catch to attempt to determine why the trigger has been reached o If the catch is spatially or temporally aggregated, impose a spatial and/or seasonal closure o If the trigger has been reached because a market has opened up for that species, add the species to the list of “key commercial species” and establish a suite of revised triggers. o If the catch is spatially and/or temporally patchy or random, consult with experts and if there are no concerns, report as such. o Reconsider the limit trigger level value in light of the outcomes above.

ii)

Limit trigger of 4t for each species •

No targeted fishing permitted on that species once this trigger level has been reached.

Management controls for the suite of key commercial species The following decision rules apply to each trigger level: •

Level 1: o Undertake a detailed examination/exploratory analysis of logbook data, with particular focus on the species for which the trigger has been reached

32

o To the extent possible, standardise the annual CPUE for that species to obtain an abundance index. o Consult experts with a view to refining the limit reference point (i.e. obtain a more informed estimate than that of double the historical high catch) •

Level 2: o Analyse the collected biological data o If possible, undertake a basic stock assessment, with a view to establishing stock status and revising the limit reference point (Level 3) in light of this improved information.

•

Level 3: o The limit reference point, above which targeted fishing for the species must cease o No further overall increase in effort pending expert consultation and stock assessment (if outcomes from stock assessment undertaken at Level 2 are pending or uninformative)

Consistency with Harvest Strategy Policy Given the large size of the resource, the range of habitats and gears, and the localised nature of fishing, and the possible propensity for vulnerable species to be taken largely from localised areas, a spatial approach, similar to that applied to the Great Australian Bight trawl fishery (Moore and Knuckey 2007; Australian Government 2007b), combined with a suite of staged triggers for key and highly vulnerable species is considered to be the most appropriate harvest strategy framework for these fisheries. The main issue for these fisheries is mitigating against overfishing due to an uncontrolled activation of latent effort. The suite of staged triggers should achieve this while providing for the fishery to continue at the same low catch levels without imposing any new requirements on operators. However, as higher level triggers are reached the data and analysis requirements become progressively more detailed, so that ultimately some species may move to management under higher Tier levels (e.g. assessments based on size/age data). As stated earlier, it is not possible to set meaningful triggers for each species captured by the fisheries, particularly for Western Deepwater Trawl where catch is opportunistic and species composition is highly variable. Zone-specific triggers for key species acknowledge the localized distribution of many of the species and should detect any increases in exploitation. The life history and population dynamics of the species were considered in consultations with stakeholders when developing values for the trigger points such that where it was considered that the historical high catch was extremely conservative given the life history and/or population dynamics, a higher level value was set that was regarded as precautionary. The inclusion of permanent spatial closures provides an additional precautionary measure against over-exploitation and habitat degradation, while allowing for the sustainable development and exploitation of the resources in open areas in an economically efficient manner.

33

Process for review The review process should ideally include some form of data collection and initial analyses of key species in order for more appropriate/robust trigger points to be set. The list of key species should be subject to periodic review to ensure they are appropriate against current catch patterns. Spatial and seasonal catch distribution should be periodically reviewed to ensure that the functional management zones are appropriate. No seasonal zones are assigned as yet due to the difficulty of obtaining seasonal patterns given the variable and patchy catch data. Permanent spatial closures will have to be revisited with stakeholders over the next 12 months for further discussion once habitat mapping has been undertaken and information synthesized, and the bioregional marine planning and MPA processes are completed. These areas need to be devised in an informed and coordinated manner. Management options for the highly vulnerable shark, dogfish and chimaera species will need to be refined and regularly reviewed as taxonomic issues are resolved. The triggers values for the NWSTF key commercial species will have to be finalized to be zone-specific triggers that reflect the relative catch from each area. The nature of stock assessments that may be undertaken in response to level 2/level 3 trigger levels being reached should be explicitly defined. Algorithms for CPUE standardization should be developed and tested.

34

Western Deepwater Trawl Fishery and North West Slope Fishery. Appendix - Annotated description of Triggers and Decision Rules (providing extra explanation and rationale), and additional Harvest Strategy details 1) Define functional management zones. •

Due to the large spatial extent of the fishery, it is likely to encompass a wide range of marine ecosystems.

•

Divide fishery into a number of management zones.

•

Zones do not necessarily have to coincide with stock boundaries, but will assist with targeting management arrangements in the different regions.

•

Industry can provide a breakdown of spatial catch patterns by species, which will help with zone delineations. Possible categorizations include: •

Latitude categories: at least one N-S delineation of the fishery at 26°30’S as suggested by differences in species composition of catch (corresponds with an existing State fishery boundary)

•

Depth categories: many species and ecosystems have set depth distributions; useful way of categorising the different areas of the fishery. Suggested depth stratifications include:500m

•

Fishery may also like to consider seasonal management arrangements (eg. monsoon season in the north).

Progress as at August 2007: Both fisheries: •

No depth zones were assigned at this time, given that each of the key species is caught within a limited depth range in any case.

•

No seasonal zones were assigned due to the difficulty of obtaining seasonal patterns given the variable and patchy catch data.

Western Deepwater Trawl Fishery: •

Two latitudinal zones were proposed: north and south of 30°S as suggested by differences in species composition of catch, with triggers and decision rules applied independently. The same trigger values and decision rules are to apply within each zone.

North West Slope Trawl Fishery: •

Catch species and magnitude suggest three longitudinal delineations: west of 117°E, 117°E -121°E, and 121°E -125°E. The same trigger values and decision rules are to apply independently within each zone.

35

2) Identify permanent spatial closures that encompass significant areas of each of the regions highlighted above (which may or may not be in addition to those determined for high-risk species noted below). •

There are growing concerns about the impacts of demersal trawling on structured benthic habitats and increasing evidence that recovery from these impacts is slow particularly for vulnerable habitats (eg. Kaiser et al., 2006; Løkkeborg, 2005; Watling and Norse, 1998). Due to this, there is a need to ensure that some proportion of vulnerable habitats is protected via permanent closures. Although it is unlikely that scientific information will be obtained in the near future (as at 2007) to prove the benefits to the fishery and/or stocks (if in fact there are any), these closures in effect act as a safety net for the fishery from both real and perceived threats. There are growing concerns about the impacts of demersal trawling on structured benthic habitats and increasing evidence that recovery from these impacts is slow for particularly vulnerable habitats.

•

Stakeholders endorse the benefits of a pro-active approach to define closures that address these concerns at an early stage in the development of the fishery.

•

Via such an approach, it is hoped that trawling may be conducted under management arrangements that enable the sustainable development and exploitation of the resources in an economically efficient manner.

Progress as at December 2007: Spatial closures should be implemented that protect habitats identified in the ERA as high risk. It is noted that habitat planning is occurring as part of the bioregional marine planning process, reports from which are imminent. These can be used to help map high risk habitats. Sponge habitats in particular need to be identified. Alan Williams (CSIRO) has also undertaken mapping of the area and should be consulted. An informed evaluation is required whereby all available information is synthesised and considered in the context of the fisheries. As such, this issue will be revisited with stakeholders over the next 12 months for further discussion once mapping has been undertaken and information synthesised. Representative habitats and/or community types should be identified. Specifically, the outcomes of the bioregional marine planning process and the MPA process will be informative. Permanent spatial closures are a priority but the areas need to be devised in an informed and coordinated manner. In the interim, (as at December 2007) expert advice has been sought from industry who have provided coordinates for clearly delineated recommended closure areas. A general description of the proposed WDTDW interim trawl closures are as follows: i) a northern closure area off Carnarvon ii) a mid-region closure area off Geraldton iii) a southern closure area off Bunbury in colder waters

36

It is emphasized that in the context of the harvest strategy (as opposed to the MPA or bioregional planning process), spatial closures represent a proactive approach to responsible fisheries management, that enables the sustainable development and exploitation of the resources in an economically efficient manner. It is not the main management driver and it is included with industry endorsement in good faith that the definition of closed areas will be undertaken as an iterative approach between experts and stakeholders. 3) Define management controls for ERA high-risk species •

High-risk species have the potential to limit the fishery. For example, the life-history attributes of species such as dogfish render it vulnerable to over-exploitation. This has the potential to limit the entire fishery due to over-riding conservation objectives.

•

For species of highest concern in terms of risk based on the ERA, e.g. dusky shark; upper- and mid-slope dogfish (Furlani et al., 2006a, 2006b), could implement: • Strict catch triggers (with no take being one extreme) OR •

Permanent spatial closures (determined by mapping Commonwealth and State catch distributions and/or preferable habitats)

Progress as at December 2007: Across both the WDWTF and the NWSTF, the ERA has identified 11 by-product species of dogfish, chimaeras and sharks as being high risk. Mitigation measures should be formulated for each individual species. The two shark species for which the most information exists are endeavour dog shark and school shark. These is little available information 12 ERA high-risk species, which are comprised of teleosts and invertebrates. Note that three of these remaining 12 species, namely tang snapper, gemfish and mirror dory, are included in the key species list against which catch triggers are set in 4) below. It is noted that catches for these 12 species are extremely low across both fisheries. Proposed decision rules associated with triggers for dogfish (including Deania, Squalus and Centrophorus) and chimaera species (including Hydrolagus) and Squatina tergocellatai (angel shark – a bottom dweller not known to be migratory) (i.e. species with less propensity to migrate, or species where little knowledge exists regarding movements, but treated in the interim as relatively site attached) (Stevens and Daley, pers. comm.): iii)

iv)

Fishery-wide 50 animal move-on provision (i.e. once trigger level exceeded, no one fishes there) (as opposed to catch trip limits which have a high associated potential for unwanted discarding), AND spatial (depth) closures (due to relatively low potential to migrate)

Proposed decision rules for school shark and dusky shark (these are more migratory species)

37

iv)

v)

vi)

Fishery-wide 50 animal move-on provisions (Code of Practice) (i.e. once trigger level exceeded, no one fishes there) (idea is that move-on provisions will mitigate against triggers being reached and thereby minimise unwanted discarding), AND trigger values for total catch (as opposed to trip limits per boat): a. intermediate: look at spatial catch pattern and implement temporary seasonal closures, b. limit: no catch once this trigger has been reached The aim is to implement depth closures in appropriate zones as determined by habitat (depends on data and species; won’t work for highly migratory species)

NB a move-on provision in itself is not an adequate management measure (difficult to enforce) but it is a sensible measure to augment other rules and help optimise their effectiveness In developing the management options for the high-risk shark, chimaera and dogfish species, shark experts including John Stevens and Ross Daley (CSIRO), Rory McCauley (W.A. Department of Fisheries) and Terry Walker (MAFRI) were consulted to help inform the process. Some key points raised were: •

There is almost no data on the distribution of squalids (McCauley)

•

Species misidentification is a critical issue in these fisheries. Moreover, the taxonomy of various groups is currently under review (Daley and Stevens). For example, there are two recently described endemics in the area. Centrophorus westraliensis and Squalus crassispinus that are almost certainly being confused with other species. These endemics are of even higher conservation concern than the more wide-ranging species (Stevens)

•

There is very little data on movement. To design effective closed areas, sex-specific catch data is a minimum requirement, as many chondrichthyans segregate by sex (Daley).

•

There is currently no evidence for seasonal movement of dogfish, but this does not imply seasonal movements do not occur (Stevens)

•

Concern regarding the difficulty of enforcing move-on provisions (Daley and Stevens) [Note, however, that the same can be said about trip limits given the opportunity for discarding exists]

•

Depth closures are likely to be effective for dogfishes and chimaera but would impact catch of commercial species on the slope. The listed set of species extends from the inner shelf to the mid-slope. Ideally the species list should be refined based on specimens collected to narrow the closure zone (Daley)

•

Closed areas in the Great Australian Bight for dogfish are approximately 60nm in size (Stevens)

38

•

Dogfish tend to aggregate so that incidental catches from a single trawl may be large (Stevens)

Proposed decision rules for high-risk teleosts and invertebrates Two levels of triggers: iii)

Intermediate trigger of 2t for each species •

This value is considered precautionary without being non-sensible

•

If reached, investigate spatial distribution of the catch to attempt to determine why the trigger has been reached o If the catch is spatially or temporally aggregated, impose a spatial and/or seasonal closure o If the trigger has been reached because a market has opened up for that species, add the species to the list of “key commercial species” and establish a suite of revised triggers. o If the catch is spatially and/or temporally patchy or random, consult with experts and if there are no concerns, report as such. o Reconsider the limit trigger level value in light of the outcomes above.

iv)

Limit trigger of 4t for each species •

No targeted fishing permitted on that species once this trigger level has been reached.

The following table identifies the high-risk species for each fishery from the ERA residual risk analyses: Fishery

Taxonomic group

Scientific name

Common name

WDWTF

Shark

Deania quadrispinosa

Platypus shark (a deep-water dogfish)

WDWTF

Shark

Carcharhinus obscurus

Dusky shark

WDWTF

Shark

Deania calcea

Brier shark (a deep-water dogfish)

WDWTF

Shark

Hydrolagus lemures

Bight ghost shark (a chimaera)

WDWTF

Shark

Galeorhinus galeus

School shark

WDWTF

Shark

Squatina tergocellata

Ornate angel shark

WDWTF

Shark

Squalus mitsukurii

Green-eyed dogfish

39

shark,

Tope

WDWTF

Shark

Squalus megalops

Piked dogfish

WDWTF

Shark

Centrophorus moluccensis

Endeavour dogfish

WDWTF

Chimaera

Chimaera sp. C [in Last Longspine chimaera & Stevens, 1994]

WDWTF

Chimaera

Chimaera sp. E [in Last Whitefin chimaera & Stevens, 1994]

WDWTF

Teleost

Dannevigia tusca

Australian Tusk

WDWTF

Teleost

Nelusetta ayraudi

Chinaman Leatherjacket

WDWTF

Teleost

Rexea solandri

Gemfish

WDWTF

Teleost

Nemadactylus macropterus

Jackass Morwong

WDWTF

Teleost

Zenopsis nebulosus

Mirror Dory

WDWTF

Teleost

Paristiopterus gallipavo

Yellow-spotted boarfish

WDWTF

Teleost

Pentaceros decacanthus

Big-spined boarfish

WDWTF

Teleost

Dentex tumifrons

Yellowback bream

WDWTF

Teleost

Lipocheilus carnolabrum

Tang snapper

WDWTF

Teleost

Plagiogeneion macrolepis

Bigscale rubyfish

WDWTF

Invertebrate

Hypthalassia acerba

Champagne crab

NSWTF

Invertebrate

Aristaeopsis edwardsiana

Scarlet prawn

4) Define management controls for the suite of key commercial species •

System of catch/effort triggers for key commercial species.

•

Because of the multi-species nature of the fishery, effort triggers may be more costeffective due to reduced administrative arrangements.

•

Triggers should be zone-specific, and possibly season-specific.

•

Existing catch and effort data should be summarized so that levels for triggers may be determined in an absolute sense.

•

Level 1. If an effort trigger is reached in a given area and a given season, this should initiate data analyses using the collected fishery data.

40

•

these analyses must be low-cost, but could include examining catch and effort temporal trends, size-frequency trends, catch composition analysis and/or simple CPUE standardizations.

•

Level 2. Potentially a range of effort triggers, each initiating an increasingly rigorous (and as such, more costly) analysis and assessment. However, at this point a single set of intermediate triggers is probably more appropriate.

•

Level 3. Total effort trigger and/or total catch triggers (species-specific?) above which fishing ceases.

Progress as at August 2007: For both fisheries, it was decided that individual annual catch triggers for the key species were more appropriate than effort triggers. Although the latter is more cost-effective, it is less direct and does not acknowledge the variability within the fishery. In the absence of detailed spatial catch information, for the WDWTF the same trigger values are to be applied to each of the latitudinal regions defined in 1) above. However, the values for the NWSTF species will be revised in a zone-specific context, to be scaled in proportion with the relative catch from each of the 3 proposed management zones (see below for preliminary calculations; data from Garvey, pers. comm.). Irrespective of the level 1 trigger being reached, comprehensive summaries of the logbook data for each fishery, as per those prepared by AFMA in 2007 (Moore et al. 2007), should be undertaken every two years. For each species, three levels of catch triggers are to apply. In the absence of any other information and assessments, the value for each trigger is broadly based on (unless otherwise specified) •

Level 1: half the historical high annual catch

•

Level 2: the historical high annual catch, taken in each of two consecutive years.

•

Level 3: double the historical high annual catch

Historical high catches are used in the absence of any other information. Given the recent low effort in the fisheries, the multi-species nature and temporally dynamic targeting behaviour, and the low GVP and developing state of the fisheries, the use of average catches as the basis for trigger levels was inappropriate. The life history and population dynamics of the species were considered in consultations with stakeholders when developing values for the trigger points such that where it was considered that the historical high catch was extremely conservative given the life history and/or population dynamics, a higher level value was set that was still precautionary without being unnecessarily invoked. For similar reasons, level 2 catch levels are to be achieved in each of two (or otherwise specified) consecutive years for the trigger to be reached. It was noted that orange roughy are conservation dependent, and as such the SESSF and DEW should be consulted regarding the proposed trigger levels for this species.

41

The following decision rules apply to each trigger level: •

Level 1: o Undertake a detailed examination/exploratory analysis of logbook data, with particular focus on the species for which the trigger has been reached o To the extent possible, standardise the annual CPUE for that species to obtain an abundance index. o Consult experts with a view to refining the limit reference point (i.e. obtain a more informed estimate than that of double the historical high catch)

•

Level 2: o Analyse the collected biological data o If possible, undertake a basic stock assessment, with a view to establishing stock status and revising the limit reference point (Level 3) in light of this improved information. o Possible basic stock assessments could include yield-per-recruit analyses and/or biomass-dynamic models (as per Hunter et al. (2002) for ruby snapper). The latter is used to assess the state of the stock and seek to determine the maximum sustainable yield (MSY), while the former gives direction about how to adjust harvesting size and rate so that the maximum yield from a given weight of recruits can be achieved. o Biomass-dynamic models require standardised catch rates as relative indices of abundance. Yield-per-recruit analyses require information on size-specific fishing mortality, as well as growth information, and estimates of natural mortality.

•

Level 3: o the limit reference point, above which targeted fishing for the species must cease o No further overall increase in effort pending expert consultation and stock assessment (if outcomes from stock assessment undertaken at Level 2 are pending or uninformative)

42

The following catch values for each trigger level are proposed for each key species. WDWTF: Species

Historical Level 1 high catch Trigger (kg) (year) value (t) (zonespecific)

Deepwater flathead

39975 (2002)

200

Boarfish

3668 (2001)

500 1000 250 in each of 3 in each of 3 consecutive consecutive years years

Tang snapper

14957 (2001) 53404 (2000) 4540 (2004)

5

15

20

25

100

200

50

100

200

200

400

10

20 if ERA ERA high-risk species high risk; 100 Must include discard - size issues. Needs more information if not

Ruby snapper Flame snapper

Bugs Gemfish

Mirror dory

198705 (total 100 crustaceans) (2002) 8733 5 (2002)

3640 (2002)

10

Level 2 Trigger value (t) (zonespecific) To be taken in each of 2 (or specified no of) consecutive years 500

25

Level 3 Industry comments Trigger value (t) (zonespecific)

1000

50

43

The species also intersects with the GAB fishery which takes more - WDWTF targets this species but can't access its preferred depth, i.e. low susceptibility Can achieve 10t or 20t in a single shot, but there is no market for this species (is this standpoint biologically defensible?) ERA high-risk species; hence lower Level 3 trigger Also targeted by WA state line fishery Pelagic, extremely fast swimming, hard to target - but valuable, try to catch it. Low susceptibilty. Have been historically targeted, but very little is known about these species

on state of maturity - discarding because too small to sell, and perception that taking juveniles. But if fish is actually mature at this small size, may as well take them and retain them ERA high-risk species Same depth structure as boarfish; marketing problems domestic market only, otherwise apparently healthy fishery. As this species is captured in conjunction with boarfish, the trigger catch levels must be consistent with those for boarfish. Levels have been set scaled to those for boarfish, on the basis that catch of this

species equates to ~ 5% of the catch composition when catching boarfish If the species is removed from the ERA high risk list, the trigger at each level is reached only if the specified catch is achieved in each of 3 consecutive years.

Scampi

198705 (total crustaceans) (2002) Deepwater 198705 (total prawn crustaceans) (2002) Orange roughy 154179 (2001) Smooth oreo 3624 dory (2001) Alfonsino 4342 (2001)

100

200

400

100

200

400

75

150

300

2

4

8

20

Bar rock cod

9727 (2001)

10

30t in each 3 150t targeted consecutive years catch OR 30 t for 5 or more years over previous 10 years Susceptible to trawling; also 20 50

105936 (2001)

50

100

200

17924 (2002)

9

18

36

NWSTF: Species Scampi Deepwater prawns

intersects with WA state fishery John Garvy - obtain catch by zone to inform Historic high catch obtained by subtracting prawn catch from total

Catch data by zone (west of 117°E, 117°E -121°E, and 121°E -125°E) for the NW Slope Trawl fishery from 2000-2007 (Garvey, pers. comm.) gave the following relative catches by zone. Scampi (combined species):

Deepwater prawns (combined species)

west of 117°E: 16.3%

west of 117°E: 58.4%

117°E -121°E: 66.3%

117°E -121°E: 39.2%

121°E -125°E: 17.3%

121°E -125°E: 2.4%

These values should be considered in the next round of stakeholder consultation with a view to developing zone-specific triggers that reflect the relative catch from each area.

44

Bass Strait Central Zone Scallop Fishery (BSCZSF) Overview of the fishery Scallop fishing in Bass Strait commenced in the early 1970s. Fishing is managed under three jurisdictions: AFMA manages the Central Zone fishery, while Victoria and Tasmania manage zones generally out to 20 nm off their respective coastlines. The main target species is the southern scallop, Pecten fumatus. SFRs are also issued for doughboy scallops, but these are generally not fished. Scallops may be taken by dredging or trawling (trawling is an historical concession). The CZSF is overfished and is currently inactive and closed to commercial fishing under the 2005 Ministerial Direction. Scallops are patchily distributed in beds ~3-4km long and ~12km wide. Once opened, a bed tends to be fished to economic depletion. Between 1990 and 2003, known scallop beds have been closed several times to allow stocks to rebuild. Abundance is currently (as at 2005 – no surveys have taken place subsequently) unknown, with only one known aggregation (east of Flinders Island) (McLoughlin 2006). Bass Strait has a history of boom and bust: the discovery of beds has been followed by an influx of effort that fished the beds to depletion (McLoughlin 2006). There are currently no known beds in the western zone of the CZSF, and recruitment has always been episodic in Bass Strait (successful recruitment events are strong but occasional). The highest GVP in the period from 1997 to 2005 was in 1997/98 at $7 million. Subsequently the GVP declined drastically, to $191 000 in 2005/06 (Moore and Patterson 2007). Access to the fishery is by limited entry, with entry limited to the number of permits held on 1 March, 2001. From January 1, 2005 there were 103 Statutory Fishing Right (SFR) holders, with each holder allocated one boat SFR and 3500 quota SFRs for commercial scallops and 3500 quota SFRs for doughboy scallops for every permit held on the snapshot date. Boat SFRs ceased to exist on February 1, 2007. Prior to the 2007 buy back, there were 89 SFR holders and 152 concessions pre buy back.

Review of current management of the fishery The fishery is managed under a Management Plan. The current Central Zone (CZ) management regime is such that most of the area is open, with specific beds closed. The boundaries of these closed areas are flexible every year. Pre-season surveys must be undertaken to determine size distributions and areas of beds. At least one bed in the east and one in the west must remain closed. In the absence of fishery-independent stock assessments, the Management Plan stated that AFMA would determine annual TACs of 1000 tonnes (maximum) for commercial scallops and 100 tonnes (maximum) for doughboy scallops. AFMA was able to further increase the TAC if subsequent assessments of beds were favourable.

45

A minimum size for commercial scallops is set at 90mm (≈ 3 years old), allowing for two years of spawning prior to harvesting. There is a “discard rate” of 20%, meaning that the undersize scallop discard rate should not exceed 20%. If greater than 20% of the catch (as determined at the start of season from surveys) is less than 90mm, then the area will be closed to fishing to protect small-size scallops to allow them to spawn and grow to contribute to the future yield of the bed. The default fishing season is 1 May – 20 December but there is the option to vary these dates. For a bed to be opened, surveys must meet the 90mm minimum size and less than (or equal to) 20% “discarding” rules in at least 1 fishable area (bed). Otherwise fishing is allowed across the majority of the fishery where there are no closures in place. The fishery is currently (as at 2007) information-poor. Logbook data exists, but does not extend back to the early years of the fishery (i.e. mid 1980s). There have been a number of recruitment and abundance surveys, but apart from surveys to review area closures, there has been little Bass Strait scallop research since the fishery was closed in 2005. The management arrangements as at 2007 are effectively a harvest strategy. However, these are not considered consistent with the intent of the Policy, given the following: • whole fishery open with one bed closed. There is a risk that this may protect less than the limit reference point of B20. • need to demonstrate a more defensible overall level of protection in terms of the limit reference point • The harvest strategy should apply in all circumstances and stock conditions, e.g. different responses for monitoring etc. depending on stock status. It must be emphasized that this fishery does not conform well to the biological assumptions underlying the Policy reference points. It has naturally sporadic and fluctuating availability and intermittent recruitment, such that the concept of B0 has limited relevance to the fishery. The scallops aggregate in sub-populations (scallop beds) which come and go and have historically resulted in a boom-and-bust fishery, so that the notion of maintaining the fishery at a nominated BTARG level is inconsistent with the nature of the species. Moreover, the ability of fishers to effectively clear out a scallop bed once opened to fishing implies that the overall stock level can rapidly decrease from BTARG to BLIM. Even defining a BLIM is difficult as stocks have recovered from collapse from what must be assumed to be very low levels. Combined with the intermittent nature of this fishery, this strongly suggests that the Bass St Central Zone scallop fishery may not be relied upon as a continuous source of product, and hence income.

Proxies against the Harvest Strategy Policy Reference Points Under the proposed spatial management system, there is no absolute or direct target reference point proxy. This is because the concept of B0 (unfished biomass) has little relevance in this fishery, due to the patchy distribution of the stock, and the fact that recruitment events are episodic and irregular.

46

That the percentage of viable areas remaining closed to fishing must be greater than or equal to 40% at all times, was intended to be indirectly consistent with the BMSY ≈ 0.4B0 target reference point of the late-2006 version of the draft Policy. In terms of the target reference point optimizing economic yield, as per BMEY, this would perhaps indirectly equate to an adequate number of areas being opened to enable continuous harvesting by all SFR holders during the fishing season. A proxy BLIM is defined for this fishery, as a combination of spatial and biomass criteria, whereby the fishery is or remains closed if both of these are not fulfilled. This is discussed in detail below. The spatial criteria are that there must be > 1 “viable” area in terms of scallop density, size and discard rate (area = unit sized blocks of approximately 5 nm x 5 nm); the percentage of viable areas remaining closed to fishing must be ≥ 40% at all times, and areas will be opened on a rotational or staged basis (if multiple viable areas are available.) Biomass criteria are that the surveyed biomass must be ≥ BLIM set at 500t based on scaling of historical high catches and levels of biomass from which the stock has been known to recover. If the number of viable areas available = 2 , the smaller of the two areas should generally be opened, and at the closed area must have a biomass estimate greater than 500t.

General description of the harvest strategy Fishery Issues; justification for approach •

The fishery has shown a declining “boom and bust” cycle. There was only one commercially viable area known in 2007 (east of Flinders Island).

•

The fishery was closed by Ministerial Direction for at least three years (2006-2008). Surveys will need to provide evidence that stocks are capable of sustaining fishing in order to reopen the fishery.

•

The fishery is overcapitalised as at 2007.

•

The fishery requires spatial management (due to its patchy distribution and irregular, episodic recruitment). Opening the entire fishery, with no accompanying regulations (e.g. season length, size limits), or with a large area open and small areas closed, has in the past been both biologically and economically unsustainable (Moore and Patterson 2007). However, detailed spatial management has a high information requirement.

•

The benefits of closed area spatial management strategies for scallops have been detailed in the literature and include increased protection from fishing and consequent increased abundance, mean age and size, and enhanced local reproductive potential and hence improved probability of larval export to surrounding areas (Ward et al. 2001; Gell and Roberts 2003; Halpern 2003; Beukers-Stewart et al. 2005). Closed area spatial management of commercial scallop stocks, where the majority of the fishery is closed to fishing and only discrete beds are open to harvesting is considered to optimise the potential for continuity and sustainability as compared to broader spatial scales of management, such as the existing Commonwealth strategy (Haddon et al. 2006).

47

•

The concept of B0 (unfished biomass) has little relevance in this fishery (due to the patchy distribution of the stock, and the fact that recruitment events are episodic and irregular), but there are useful proxies. For example, it is possible, although theoretically unsound, to obtain estimates of B0, Blim and Btarg based on historical high catch rates.

•

TACs should not be the only management control applied to the fishery because of uncertainties associated with survey-based estimates of exploitable biomass (e.g. patchy distribution of scallops and variations in catchability).

•

Input controls are still necessary. A minimum legal size allows for two major spawning events and improves yield via a lower number of meats per kilogram. Seasonal closures protect beds during peak settlement periods and optimise scallop condition (the seasonal closure coincides with a loss of condition over summer). Limited entry pertains to fleet capacity and economic efficiency.

•

Ultimately the preference would be to manage all scallop fisheries in Bass Strait as a single management entity. However until agreement can be reached between the Commonwealth, Tasmania and Victoria as to if and when such a transition could occur, the BSCZSC must be managed in isolation.

•

This is an intermittent fishery that should not be relied on as continuous source of income/product. It may not be possible to open the fishery every year. The fishery has never been implemented sustainably, so expectations based on the past are unrealistic. The fishery must move away from the philosophy that fishing should occur if a viable area is present. This has proven to be economically and ecologically unsustainable and is not consistent with the intent of the Harvest Strategy Policy.

Harvest Strategy Overview The harvest strategy to be adopted for the BSCZSF includes a detailed spatial management system, which will take into account the following points: o Biomass estimates will be based on surveys of areas known to contain historically fished beds. This will establish areas of high density (formerly known as “beds”) within the fishery. Each area should have the potential to be commercially fished and as such should be of a sensible minimum size. These will be divided into unit-sized spatial management areas 5 nm x 5 nm. o The fishery will remain closed unless survey results indicate that the specific criteria (see below) for opening one or more areas to fishing can be met. Staged harvesting will occur whereby unit areas are opened sequentially. Additionally, ongoing monitoring will occur on all of these known areas of high density to determine their status each season. Note that by limiting the management approach to these beds/areas of high density, the area required to be monitored is reduced to a more practical subset of Bass Strait. o All other regions of the fishery are to be open to “scientific survey” (exploratory fishing) via permits and a research allowance process. Note that this is not to be confused with the formal survey process to monitor the status of the known beds (as mentioned in the 48

previous point), which forms an integral part of the harvest strategy. Rather, this is a provision to allow some ongoing fishery access and gives some flexibility for exploring. If a bed is discovered in the course of these surveys, it may then be included in/added to the pool of defined areas of high density under which the above spatial rotational system of management applies. [This gives an incentive to monitor other areas, as opposed to the current strategy where discovery of a new bed results in its closure]. o As an “insurance policy”, a percentage of each open area may be kept closed to maintain local spawner biomass (given that the nature of the stock-recruitment dynamics is largely unknown). o The condition of scallops will also be taken into account in the decision as to whether to open a bed. o The above regime will apply independently to the eastern and western regions of the fishery. This mitigates against opening the fishery when only one viable bed is present in each of the eastern and western regions. This also provides an incentive to undertake opportunistic surveys, particularly in the west where knowledge is currently (as at 2007) poor. o The Harvest Strategy can be updated if/when the quality of available information improves. Decision Rules (see ‘Annotated version with additional explanation and rationale’ at the end of this fishery section) 1. Total allowable catch: The TAC set each year will apply to the fishery as a whole (as opposed to area-specific TACs) and will equate to the proportion of the estimated biomass that corresponds to that occurring in the areas to be opened to fishing (see item 6. below). The TAC will be adjustable due both to the variable nature of the fishery and the uncertainty associated with biomass estimation. A default TAC of 100t will apply to doughboy scallops. 2. Fishing season: The default fishing season will be 1 June to 31 January each year. However, the fishing season will be subject to review each year based on the survey results and biomass estimations. It may not be possible to open the season each year and the season times may vary. Scallop condition will also be taken into account when determining the season and the opening may be delayed to allow scallops to reach a desired condition prior to harvesting. 3. Area closures: All areas of the fishery will remain closed to fishing unless survey results indicate that criteria for opening the fishery (Decision Rule 6 below) can be met. Rotational or staged harvesting will apply.

49

4. Size limit: A shell length of 90mm is the minimum size allowed to be harvested. However, if it can be demonstrated that an area has scallops that have completed 2 major spawning events (this being the rationale for the size limit), harvesting at sizes less than 90mm may be permitted. In such cases the following criteria must be fulfilled: i) the area has been monitored for a duration of not less than 3 years, such that an indication of age is obtained whereby it is known that two spawning events have occurred, and ii) if i) has been met, information on growth rates obtained over the 3 years of monitoring should be taken into account in considering the risk tradeoff between taking scallops at a size less than 90mm versus their potential to grow to a more profitable size. 5. Discard rate: The maximum discard rate for any area to be opened to fishing is 20%. However, as noted in Decision Rule 4, if an area contains scallops smaller than the minimum size that are known to have completed two major spawning events, the 20% discard rule may be amended. 6. Criteria required for the fishery to be reopened, and below which the fishery will be closed: BLim proxy will consist of both spatial and biomass criteria that must be met before any area within the BSCZSF will be reopened to fishing. Spatial criteria: • • •

Must be > 1 “viable” area in terms of scallop density, size and discard rate (area = unit sized blocks of approximately 5 nm x 5 nm) % of viable areas remaining closed to fishing ≥ 40% at all times. Area will be opened on a rotational or staged basis (if multiple viable areas are available.)

Biomass criteria: • •

At all times, the closed areas must contain a total biomass not less than 500t, as estimated by surveys. If the viable areas available = 2, in general the smaller of the two areas should be opened, but consideration should be given to accessibility and to the life history stage of the scallops in each of the areas. At least one of the areas must have a biomass estimate greater than 500t.

50

Consistency with Harvest Strategy Policy Given o the incompatibility of B0 based reference points with this fishery given its patchy distribution and sporadic recruitment of the stock o the uncertainty associated with survey biomass estimates o the need for spatial management o the low GVP in the fishery the proposed harvest strategy provides a detailed yet practical spatial management system with inbuilt monitoring requirements (in terms of surveys) that give the best opportunity for optimising both stock sustainability and economic yield. It aligns the Commonwealth approach with the Tasmanian harvest strategy, an approach that is anticipated will maximize the chance for fishery success, and provides for greater consistency between jurisdictions. The harvest strategy provides clear rules in terms of both spatial and biomass-based limit criteria for opening areas to fishing, which can be applied irrespective of whether the fishery has been closed or open. While there is no absolute target reference point, this is not sensible in a highly variable and spatially segregated fishery. Moreover, where the dynamics are such that, it is preferable to completely “fish out” areas in succession, as this appears to provide more desirable habitat for subsequent spatfalls than partially fished areas (Haddon et al. 2006), as well as avoiding wastage of scallops due to damage from cross-sectioning an entire bed. As such, maximum economic yield is intended to be achieved by controlling the number of areas that are open to fishing within a season, and setting a TAC that approximates the available biomass in the open area(s). Additional control rules pertaining to season, size limits and discard rates aim to provide maximum opportunity for successful spawning and recruitment events.

Process for review A qualitative evaluation of this harvest strategy should consider the following: o Outcome of initial survey- is there now more than one “viable” scallop bed in Bass Strait? o Effectiveness of size of unit areas – is 5nmx5nm practical, in terms of opening these areas sequentially? o The biomass-based BLIM criterion – how does this play out in practice, when confronted with survey estimates of available biomass? Is the approach yielding values consistent with what survey outcomes suggest regarding the status of the stock? o Given the cost of ongoing surveys to assess the condition of known scallop areas, and the size of Bass Strait, will the proposed spatial management approach be economically feasible? As at October 2007, industry members of the MAC voted unanimously to oppose the proposed reconnaissance survey, as they believed it was not an economically viable exercise, even if it yielded results that could lead to the fishery being re-opened.

51

A more quantitative evaluation of the performance of the harvest strategy against the stock status and the Policy reference points cannot occur until i) a survey has been undertaken and ii) results from a survey lead to the fishery being re-opened. At such a time, the extent of evaluation able to be undertaken will depend on the GVP of the fishery and hence the available funds. Meanwhile, the status of the state-based fisheries should be monitored, as should the progress of the Tasmanian Aquaculture and Fisheries Institute Project seeking to resolve the issue of stock structure. If there is strong evidence of a single stock in Bass Strait, the harvest strategy should be reviewed with a view to adopting a unified strategy across the three management jurisdictions.

52

Bass Strait Central Zone Scallop Fishery Appendix o Annotated Decision Rules (providing extra explanation and rationale) o Worked example for calculating BLIM proxy o Survey design o Background notes for initial harvest strategy discussions Annotated Decision Rules (providing extra explanation and rationale) 1. Total allowable catch: The TAC set each year will apply to the fishery as a whole (as opposed to area-specific TACs) and will equate to the proportion of the estimated biomass that corresponds to that occurring in the areas to be opened to fishing (see item 6. below). The TAC will be adjustable due both to the variable nature of the fishery and the uncertainty associated with biomass estimation. A default TAC of 100t will apply to doughboy scallops. Rationale: given the boom-and-bust nature of the fishery, it makes more sense to have an adjustable TAC as opposed to a fixed value. Given the spatial and biomass criteria under which viable areas may be opened, and the uncertainty associated with biomass estimates, the TAC in this context is more an internal allocation method, than it is a biological control. It addresses economic objectives by being set at approximately the biomass that is available to be taken. Having an adjustable TAC accounts for the uncertainty in the estimation of biomass within a viable area. The TAC applied should match the biomass in the open beds (i.e. equate to the available biomass), as it is thought best that fishers successively and completely clean out open beds, as opposed to partially fishing a scallop bed, as this allows subsequent recruitment to occur (Haddon et al. 2006), and avoids wastage of scallops due to damage from cross-sectioning an entire bed,. For the same reason, a TAC reduction should equate to reducing the number of beds/areas that are open, or not opening any new beds once the open ones have become exhausted. TACs should not be the only management control, as exploitable biomass is never well estimated due to the patchy distribution of scallops and the variable catchability when surveying. As such, a TAC that supposedly equates to the available biomass is not robust as a sole management control. Other than the default TAC, it is assumed that doughboy scallops will be vicariously managed under commercial scallop measures. Moreover, doughboy scallops fall under DEH requirements and as such are not required to be included in the harvest strategy. 2. Fishing season: The default fishing season will be 1 June to 31 December each year. However, the fishing season will be subject to review each year based on the survey results and biomass estimations. It may not be possible to open 53

the season each year and the season times may vary. Scallop condition will also be taken into account when determining the season and the opening may be delayed to allow scallops to reach a desired condition prior to harvesting. Rationale: Despite the other criteria for opening a viable area, a fishing season is still relevant in that it allows spat to settle and scallops to reach desired condition. However, it should be noted that its effectiveness is dependent on the quality of voluntary surveys to assess condition. The key issue is to enforce a minimum 4 month closure to enable spawning and spat settlement. Scientific evidence is that settlement occurs between October and April, and as such the closure must embrace this period (and note also that as at 2007 there is little knowledge of the nature of the relationship between spawning and spat settlement, although there is some evidence of a stock-recruitment relationship (Semmens, pers. comm.)). While acknowledging this, the previous default season (1 May – 20 December) has been amended to enable optimised marketing opportunities by encompassing the Christmas period. Criteria for alteration of the default season are yet to be determined but are most likely to be based on survey indices of condition. 3. Area closures: All areas of the fishery will remain closed to fishing unless survey results indicate that criteria for opening the fishery (Decision Rule 6 below) can be met. Rotational or staged harvesting will apply. 4. Size limit: A shell length of 90mm is the minimum size allowed to be harvested. However, if it can be demonstrated that an area has scallops that have completed 2 major spawnings (this being the rationale for the size limit), harvesting at sizes less than 90mm may be permitted. In such cases the following criteria must be fulfilled: iii) the area has been monitored for a duration of not less than 3 years, such that an indication of age is obtained whereby it is known that two spawning events have occurred, AND iv) if i) has been met, information on growth rates obtained over the 3 years of monitoring should be taken into account in considering the risk trade-off between taking scallops at a size less than 90mm versus their potential to grow to a more profitable size. The over-rides on the 90mm size acknowledge the observed phenomenon of stunted growth in deep water areas of high density (Semmons, pers. comm.). 5. Discard rate: The maximum discard rate for any area to be opened to fishing is 20%. However, as noted in Decision Rule 4, if an area contains scallops smaller than the minimum size that are known to have completed two major spawnings, the 20% discard rule may be amended. This rule allows for the protection of very small scallops (i.e. not those close to the size limit). Note also that this rule that relates to the decision to open the fishery. Survey

54

outcomes determine whether or not it is fulfilled. As such, this rule may lead to areas being/remaining closed. However, scallops less than minimum size are of little commercial value. Once the fishery is operating, this rule becomes contentious. It is not possible to sort/discriminate sizes once fishing is occurring. If the other criteria for opening the fishery are met, presumably the risk of exceeding the discard rate is irrespectively minimised. There is a key compliance issue whereby the need for “zero tolerance” should be revisited in the context of the harvest strategy implementation and in implementation of AFMA’s Bycatch and Discard Workplan. 6. Criteria required for the fishery to be reopened, and below which the fishery will be closed: BLIM proxy will consist of both spatial and biomass criteria that must be met before any area within the BSCZSF will be reopened to fishing. Both spatial and biomass criteria are used because spatial has the benefit of managing this spatially disaggregated fishery at the level of historically fished areas. When more than one area is viable, it is easy to demonstrate some level of stock protection by keeping some of these areas closed. However, the notion of protecting a set number of areas, or keeping a proportion of areas open, also implies that there will always be some stock available to be fished. This is addressed via the inclusion of the biomass criteria. Moreover, the use of both spatial and biomass criteria in combination is more defensibly consistent with the intent of the Policy. The Policy states that there must be a minimum level of stock protection, defined in terms of a limit reference point (BLIM). This effectively requires the definition of a suitable proxy for the “necessary biomass” to ensure sufficient recruitment (acknowledging our poor understanding of the stock-recruitment relationship) in this fishery. Spatial criteria: • • •

Must be > 1 “viable” area in terms of scallop density, size and discard rate (area = unit sized blocks of approximately 5 nm x 5 nm) % of viable areas remaining closed to fishing ≥ 40% at all times. Area will be opened on a rotational or staged basis (if multiple viable areas are available.)

“Viable” area definition: “Viable” in this context equates to exceeding the size limit and discard rates described in 4. and 5*. Areas are defined as unit sized blocks of approximately 5 nm x 5 nm. The unit size facilitates ease of opening areas in a staged manner within a season: a new area may be opened as catch rates decline on currently open areas. Note that it is preferable to completely “fish out” areas in succession than to simultaneously open multiple areas, even when these areas encompass a single scallop “bed” (Haddon et al. 2006). Small unit areas circumvent the need to subdivide a larger area into “strips” of “paddocks” that are then sequentially fished. 55

In super-imposing 5 nm x 5 nm areas over larger viable areas, invariably some peripheral squares will contain scallops only within a small fraction of their area. However, the biomass criteria outlined below will augment the spatial criteria and thus determine whether the peripheral area may be fished. When one area come close to fulfilling the size limit and discard rate criteria by which it may be termed “viable”, but does not absolutely satisfy these, if it can be justifiably shown that this area has been surveyed for at least 3 years, and that it has the potential to become “viable” within the next 12 months, then it may be included as a candidate area for potential exploitation. Note, however, that this area may NOT be opened to fishing in that season. Rather, its inclusion in the “pool” of “viable” areas may permit the opening of a second viable area, via the fulfilment of the criterion that there must be at least one viable area in order for any area to be fished. If this marginal area does not absolutely fulfil the criteria for it to be considered “viable” within the next 12 months, then this area must be removed from the “pool” of “viable” areas. [Note: In an earlier phase of harvest strategy development, the MAC considered defining areas as spatial blocks enclosing historically fished beds. Area boundaries were to have been refined by a mapping exercise undertaken to produce a density map of the historically actively managed areas of the fishery. These areas would have been delineated via economic boundaries – i.e. it would not be economically viable to fish below certain densities. Historical information may have been subsequently augmented by survey information. The notion was that a “viable” area should be a viable recruitment entity with respect to the density and maturity of the scallops, and its size and location. However, in a subsequent MAC meeting, unit sized areas were considered preferable in that they provide a common currency for spatial management, and by setting these at a reasonably small size, they could readily facilitate staged harvesting in that the subdivision of larger areas would be circumvented.] Note that the above-defined spatial blocks were felt to be preferable as spatial units than scallop beds, due to the difficulty associated with the definition of the latter. Note also that there may be a paucity of information of historically fished beds for the western sector of the fishery. At any one time, the areas being via monitoring surveys and rotational harvesting may or may not be commercially viable. However, the non-commercially viable areas must be considered to have the potential to become commercially viable. Moreover, within any one season, it is not adequate to protect only the non-commercially viable areas. *Information on scallop density might also be considered in classifying an area as “viable”. If biomass is greater than the required Blim threshold (defined below), but densities are low, then it is questionable as to whether the area should be considered viable from both a fishery (CPUE) and biological (recruitment) standpoint. Density may also be considered when determining which viable beds are to remain closed, given that closing low-density beds is not in the best interest of future recruitment. In both contexts, the trade-off between recruitment success and the minimum density above which fishing is economically viable 56

must be considered. The opening of any bed must simultaneously be commercially viable (self-regulating in that fishing will cease within a non-viable area) without compromising the sustainable recruitment of the stock (i.e. must maintain a viable recruitment entity with respect to the density and maturity of scallops, and the size and location of viable areas). Resolving this issue is difficult given the lack of understanding of recruitment dynamics as at 2007. Biomass criteria: • •

At all times, the closed areas must contain a total biomass not less than 500t, as estimated by surveys. If the viable areas available = 2 , in general the smaller of the two areas should be opened, but consideration should be given to accessibility and to the life history stage of the scallops in each of the areas. At least one of the areas must have a biomass estimate greater than 500.

While it is more precautionary to open the smaller of the two areas if these are the only two “viable”, in practice this is impractical if the larger area is more accessible than the other, and it is not sensible if the smaller area contains younger scallops while the larger contains scallops experiencing a high level of natural mortality due to their age. It is again reiterated that the concept of B0 has little meaning for this fishery, but a proxy must be provided given the lack of plasticity of the Policy with respect to the nature of this fishery. Under the Policy, there must be a minimum level of stock protection, defined in terms of a limit reference point. This effectively requires the definition of a suitable proxy for the “necessary biomass” to ensure sufficient future recruitment (acknowledging our poor understanding of the stock-recruitment relationship). (Note also that in this context, the nature of the relationship of BSCZ scallop abundance with that of Victoria and Tasmania needs to be more explicitly considered). Historical high catch, obtained by taking an average of “high catch” years, may be scaled and to give a B0 estimate on which the limit reference point may be based. The simplest option is to assume the catch equates to total biomass, but catch should probably be up-scaled, given that it is unlikely the catch removed all biomass. Such a BLIM is certainly not defensible on its own, but it sits as one of a suite of decision rules comprising the harvest strategy, and it provides a range of values within which negotiation can occur and that may be reconciled against survey outcomes. Clearly, the range of choices for both the number of years considered to be “high catch”, and for the catchability factor by which the average catch is upscaled (implying that biomass proxies can encompass both target and limit reference points, depending on the assumed catchability), can result in a range of B0 estimates. Given this uncertainty, together with the poor quality of reporting and the fact that catch has been largely market-driven, the choice of BLIM should be refined using economic criteria. Given that the Policy advocates economic viability, it goes without saying that there should be a reasonable assumption of economic viability before the fishery is opened.

57

An upper and lower value for a proxy BLIM may be obtained by scaling historical high catches. (For years 1993-2004, this equates to 6683 t and 529 t respectively, depending on the choice of years used and the scaling factor – see worked example below). In practice, a range of values for BLIM may be calculated where the emphasis is not on using high catches but whether earlier or later catch data is used. For example, separate values for BLIM could be based on catches averaged i) ii)

iii) iv)

over the last few years of the available time series: as B0 is a meaningless concept for this fishery, calculating BLIM in the currency of recent years may be more appropriate. over the initial few years of the most recent 12 year time series: this may be considered more defensible from a theoretical/purist viewpoint when assuming this to be a B0 proxy, while still pertaining to a relatively recent regime all available years: placing equal weighting on each year and thus obtaining an overall average across a highly variable fishery over the very early years of the fishery (e.g. 1980’s): from a theoretical/purist viewpoint, this would be the closest available proxy to B0. However, in a practical sense, it is likely to result in values of such high magnitude that the current population may never exceed.

The resulting range of values provides a basis for determining a minimum biomass that must be protected. The lowest value obtained from the worked examples below was 529t, when the average catch across all years was taken. A calculation based on the average catch from 2002-2004 (i.e. “recent” years only) and assuming 100% catchability gave a value of 356t. On the basis that these values are the result of a transparent, albeit shaky (given the incompatibility of scallop population dynamics with the Policy reference points, which assume a stock in equilibrium), process by which to determine a BLIM proxy, and that scallop stocks have demonstrably recovered from depletion to levels lower than these values, the minimum level of stock protection that is required is set at 500t. This biomass criterion provides an absolute threshold of stock protection and thus augments the spatial criteria in determining whether or not the fishery may operate. It should be noted that if it is not economically viable to fish given the current state of the stock, there is no reason to open the fishery now or in the near future. (In saying this, there is a need to beware that the fishery may be economically viable below BLIM [if only for a short time]).

58

Worked example of calculating BLIM proxy It is reiterated that these are not robust estimates, due to both the issues with confounding economic factors influencing catch, and the difficulties of upscaling due to the poor estimation of catchability (between 20% and 100%, implying that biomass estimates from survey catches can embrace both target and limit reference points). Some examples of possible proxies for BLIM under different assumptions and scaling factors are provided below, ranging from 530t-6700t. Values included those derived from relaxing the criteria for “high catch” years in deriving the estimate of B0 to include most or all catch years, to directly take into account the inherent year to year variability in the fishery. The following table shows a range of BLIM proxy values, obtained by scaling average catch over a different range of years and using different values for catchability (i.e. the assumed proportion of “B0” taken by the catch. year

catch (kg)

assumed proportion of Bo 0.4 0.6 0.9 scaled values giving Bo estimates average across 1994-1997 33411250 16705625 11137083 7424722 1993-1997,2002 25230343 12615172 8410114 5606743 all years 13215718 6607859 4405239 2936826 0.2

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

average average average

2128000 8063000 7711000 5642000 5313000 818000 0 5001 14060 1419412 431719 172532

average across 1994-1997 1993-1997,2002 all years

6682 5046 2643

Justification of choice of biomass component of BLIM proxy Clearly there is much uncertainty associated with deriving a BLIM proxy for this fishery bsed on the scaling of historical high catches. At the March 2007 HS meeting in Melbourne, several other options for the biomass proxy were suggested. These are listed below, together with their rationale for rejection:

Minimum number of beds to be protected Rationale for rejection: Considering a BLIM in terms of a number of beds has the problem (noted above) of implying that there is always some stock available to be fished. It is desirable to move away from this philosophy, which has proven to be ineffective in the past management of the fishery. Also, spatial criteria alone are

59

6682250 5046069 2643144

Blim proxy assuming Blim=0.2Bo (tonnes) 3341 2227 1485 1336 2523 1682 1121 1009 1322 881 587 529

1994-1997 6682250 1993-1997,2002 5046069 all years 2643144

i)

1

difficult to defend against the intent of the Policy, particularly when there is only one identified bed.

ii) Minimum proportion of the total bed area to be protected (takes into account that beds may be of different sizes and as such should not be treated as equal units) Rationale for rejection: As for i) above

iii) Proportion of the total estimated tonnage able to be taken across all viable beds to be protected Rationale for rejection: This requires a comprehensive survey to obtain the total estimated tonnage in each year of fishing, which is unlikely to be economically viable or logistically practical. This approach also has issues with being defensible against the Harvest Strategy Policy in that it implies that a percentage of available biomass may always be taken.

iv) Investigate values associated with stable and increasing recruitment in other scallop fisheries (e.g. Georges Bank) Rationale for rejection: It was agreed that the unique local and sporadic dynamics of scallop fisheries means that values associated with stable and increasing recruitment in other fisheries are likely to not be directly applicable to the Bass Strait Central Zone Scallop Fishery. Indeed, even within the fishery, biomass values associated with stable or increasing recruitment are likely to be variable over time. Bottom Line •

Under the Policy, we are obliged to provide a defensible BLIM proxy.

•

We are limited by the available data

•

Continuous comprehensive surveys are impractical

•

A BLIM based solely on spatial units (e.g. number of beds) is undesirable in that it supports the philosophy that there is always stock available to be fished, and it is difficult to defend against the intent of the Policy.

•

Thus the current notion is to define a BLIM both in terms of spatial and biomass criteria, where the latter is based on scaling of historical high catches to obtain an approximation of unfished biomass.

Note also that the agreed biomass criterion (that all times, the closed areas must contain a total biomass not less than 500t, as estimated by surveys) replaces an earlier biomass criterion which stated that “Total surveyed biomass must be ≥ BLIM approximating 20% of B0 based on scaling of historical high catches”. This is actually inconsistent with the intent of the Policy, as it implies that fishing may occur when total survey biomass is at the limit reference point – which would immediately and undesirably drive the stock below BLIM. Additionally, the earlier criterion specified that a range of BLIM values would be presented 60

based on various scalings of historical high/average catches, and that this would be refined on economic grounds, by considering the minimal acceptable harvest when BLIM is equated to catch and divided among licences. This was replaced with the absolute value of 500t, which was based on the lower values obtained from the catch-scaling exercise, because a range of values against a BLIM is somewhat meaningless. The main issue is defining a minimum level of stock that should be protected, giving considering to the levels of biomass from which the fishery has historically recovered].

Survey design As at 2007, the fishery is assumed to be below BLIM. Prior to the opening of the fishery, a comprehensive survey is required to obtain an understanding of the current stock status and thus inform the harvest strategy. Due to resource limitation, it is proposed that this survey be undertaken as a two-stage process: 1) a broad-scale “sweeping” survey of areas known to contain historically fished beds, in order to determine general scallop availability and location, followed by 2) a more focused/directed survey to estimate scallop biomass in each location where scallops were found to be present. Doing this as a single survey would result in days being lost when a “bed” is discovered, and hence the risk that the entire fishery may not be covered in the allotted time. 1. Broad-scale “sweeping” survey to ascertain what is where: ideally with coverage of entire fishery. Overriding caveat: no survey implies zero biomass assumption implies zero fishing a. Approach industry re: willingness GIVEN a transparent survey design (i.e. give industry opportunity to show desire for best practice) b. If industry reluctant/not willing to undertake survey tows in western Bass St., then assume zero biomass from the west (and this area remains closed). This will imply more conservative overall performance indicators. Note also that it may be logistically sensible to manage the east and west of Bass St. as two separate zones [but NOT in an environmental/stock structure sense], given the historical paucity of scallops in western Bass St., together with its general inaccessibility. Based on the results of the sweeping survey: -

if ≤1 scallop bed is found, the fishery remains closed, to be revisited later with another sweeping survey

-

Otherwise:

2. More focused/directed and “opportunistic” surveys allowing estimation of biomass - Based on results of broad-scale/sweeping survey, MAC to dictate areas to be more intensively surveyed in order to obtain biomass estimates - Opportunistic: as vessels pass through an area, undertake x tows (certain areas would lend themselves to opportunistic surveying as they lie on “trade routes”) - OR direct vessels to fish in designated areas for y days, with a reimbursement of a set amount of research quota

61

- having industry undertake these surveys sets a good precedent – the Tasmanian system works due to industry faith and responsibility. Must include Victorian vessels in these surveys - must be pragmatic and transparent in terms of risk, AND must build in processes whereby there is reasonable chance of allowing industry to capitalise on transient opportunities (particularly given the natural variability inherent in scallop populations) in order to break even. Need to clearly articulate ongoing survey protocols, and protocols for opening for exploratory fishing under scientific permit. For example, major (sweeping) surveys might be required every 3-5 years (should be explicitly stated), noting that the fishery is closed unless fulfilling the criteria for opening, as determined by voluntary surveys of potentially viable areas. The sweeping surveys should not be too onerous. Exploratory fishing under scientific permit should be on the proviso that no catch is permitted to be kept. A closed fishery should provide an incentive for fishers to inform management when viable areas are discovered.

Background notes from initial harvest strategy discussions: broad options initially considered for spatial management The following outlines the broad options for spatial management that were initially formulated and presented for discussion at a stakeholder/MAC meeting in November 2006. The acknowledged requirement for spatial management (Ward et al. 2001; Gell and Roberts 2003; Halpern 2003; Beukers-Stewart et al. 2005), together with the success of the Tasmanian management system (Haddon et al. 2006), implied that the focus for harvest strategy development was always based around a spatial management regime. The existing Commonwealth spatial management system was effectively a harvest strategy, albeit with some inconsistencies against the intent of the Policy. Moreover, the existing Commonwealth spatial management system was considered to have been unsuccessful to date, and, as the fishery was currently closed, this provided an opportunity to consider alternative spatial management options, as outlined below. However, many of the existing rules relating to season, size limit and discard rate were able to be incorporated in the harvest strategy. Note that at the November 2006 stakeholder meeting, options 1a) and 1b) below were rejected outright. Option 2a) was rejected as it provided no incentive to explore and report new beds. Option 2b) would need to be robust to variability in recruitment, or able to accept this [noting that high spatfalls and abundance in known beds have fortuitously augmented the success of the Tasmanian system to date]. At this meeting, the preferred option of a closed fishery with viable areas sequentially opened to fishing, and described in detail in the “Harvest Strategy Overview” section, was formulated. 1a. Close a large area, e.g. 40% (suggested as an indirect proxy target reference point given the BMSY ≈ 0.4B0 target reference point of the late-2006 version of the draft Policy) of the historically fished area. Management could then be i. Rotational without information, i.e. fixed/pre-specified areas 62

ii. Fixed (not rotational) Advantages: • low cost of little required information, preservation of spawning biomass (spatial proxy), low management costs, including cost-effective compliance, preservation of juveniles & habitats, protects economic efficiency Disadvantages: • may not reflect standing stock (i.e. closure may need to be larger than 40% of fished area), no voluntary information, lack of updated knowledge, lack of opportunistic surveys, not adaptive if things change, if beds are very scarce, rotational harvesting may not be helpful. 1b. Close a large area – open areas rotationally based on information Advantages: • allows for learning by being adaptive, better marrying of fishery to resource, potentially increased yield, greater security of resource, more vested interest in selfregulation, incentive for information flow Disadvantages: • costs of information 2. Adaptive spatial management: The following have the common factor of focusing on the closure/opening of specific areas at the size of the scallop bed. a) Fishery open with certain beds closed (e.g. ~80-90% of total area open, as per the current Commonwealth practice) [Note that the number of beds closed should ensure an adequate level of stock protection – e.g. 40% of beds as an indirect proxy for the target reference point given the BMSY ≈ 0.4B0 target reference point of the late-2006 version of the draft Policy]. Advantages: • Relatively inexpensive due to large area of Central Zone • Would not need to change current arrangements • Cheaper to search for single bed and close it, than to search out a number of beds and choose which ones to open. Disadvantages: • Has led to boom & bust • No real incentive for fishers to survey Other Issues:

63

• •

Closed beds would have to equate to adequate stock protection to meet intent of Policy How often are there sufficient numbers of beds to enable some to be opened and some closed?

b) Fishery closed (e.g. 80-90% closed?) with certain beds open (as per the current Tasmanian management system) This is a system of informed spatial management where voluntary pre-season surveys are undertaken to determine which beds are able to be opened to fishing. Beds are then opened on a rotational basis. Advantages: • Incentives for learning • Currently works well in the Tasmanian fishery • Beds protected until ready for harvesting Disadvantages: • Relatively expensive if stock status is poor (if not the expense is negated by quota), due to the ongoing survey requirement associated with informed rotational spatial management. • Unclear if this approach is appropriate for Bass Strait, given its large size (due to the high associated survey costs) and whether multiple beds occur (it is impossible to implement rotational harvesting if the fishery is reduced to a single bed, unless the bed is sub-divided). Other Issues: • This represents a marked shift in the management approach, but given that the fishery is currently closed, this may facilitate an easier transition. • Research quota will need to be allocated so that fishers can undertake exploratory surveys. • Is Bass Strait boom and bust by nature? It has possibly exhibited this pattern only because it has been so heavily fished that a single year class only is available in each fishing event, resulting in boom and bust behaviour. • If beds are very scarce, rotational harvesting may not be helpful. Adaptive spatial management at the scale of the scallop bed (2) is preferable to closures of, or rotations about large areas (1), as it is based on more information (i.e. actual beds), than that associated with closing a large proportion of the fished area.

64

Arrow Squid Fishery (Nototodarus gouldi) Overview of the fishery . The Southern Squid Jig Fishery (SSJF) targets, and almost exclusively catches, arrow squid, Nototodarus gouldi. The fishery uses automatic jigging machines targeting 50-100m depth contours (jigs operate to a maximum depth of 120m). Little non-target catch occurs (Furlani et al., 2006c); bycatch species include barracouta, dusky shark, blue shark, shortfinned mako/blue pointer and garfish. Considerable quantities of squid are also taken by trawling in the Southern and Eastern Scalefish and Shark Fishery (SESSF) (McLoughlin 2006). The SSJF is considered to be relatively lightly fished, although limited information is available pertaining to resource size (McLoughlin 2006). While the stock status is uncertain, a 2006 workshop convened by AFMA concluded that the impact of fishing was likely to be low. This is due to i) the wide distribution of squid relative to the area where fishing activity currently occurs; ii) the low level of catch in the mid 2000s, relative to historical high levels; iii) the variable temporal CPUE pattern; iv) the squid’s high fecundity and short life cycle; and v) preliminary estimates of biomass from ecosystem suggest levels orders of magnitude greater than the current catch. Jigging occurs mostly out of two ports: Portland and Queenscliff in Victoria. Other ports, including Hobart (Tasmania) and Lakes Entrance (Victoria), are also fished sporadically based on squid availability. Given that the squid habitat is wide relative to the fishery (generally ranging across south eastern Australia), in some years localised fishing does not locate the squid aggregations.

Review of the current management of the fishery Statutory Fishing Rights (SFRs) have been issued under the Southern Squid Jig Fishery Management Plan 2005. Once nominated to a boat, these SFRs authorise the holder to use a certain number of squid jigging machines during the year. The number of machines is determined by a Total Allowable Effort (TAE) limit, set annually. The number of jigging machines determines the rate and quantity of squid that may be caught. Moon phase and weather conditions also help to regulate effort. In January 2005, prior to the introduction of SFRs into the SSJF, there were 80 Commonwealth Southern Squid Jig Fishery Fishing Permits granted, these permits were replaced by 8000 SFRs in 2006. As at October 2007, there were 6400 SSJF SFRs: operators typically use 7-8 standard jigging machines per boat which requires (in 2007) the nomination of 70-80 SFRs per boat. Operators typically nominate SFRs to their boats in lots of 100 SFRs. Only 14 fishing permits were fished during 2004 (equivalent to 1400 SFRs or 140 standard jigging machines). The maximum number of active vessels over the last 10 years was 42 in 1996. There is no Total Allowable Catch (TAC) or catch quota for the SSJF. Currently there is insufficient scientific information available to set biomass-based reference points for squid,. There is a 4,000 t catch trigger, which equates to half the highest historic annual catch by foreign squid fishing vessels off southern Australia. An annual combined catch level trigger of 6,000 tonnes is currently in place for squid taken in the SSJF and the Southern and Eastern Scalefish and Shark Fishery (SESSF). This trigger

65

provides for 4,000 tonnes from the SSJF and a combined allowance of 2,000 tonnes for the Great Australian Bight Trawl (GABT) and South East Trawl (SET) sectors of the SESSF. Under the current SSJF Management Plan, advice will be provided by a Southern Squid Jig Fishery Resource Advisory Group (SquidRAG) on an appropriate management response, should any of the trigger catch levels be reached. The current (as at 2007) combined catch is considerably less than historical high levels. Catch from jigging peaked at 1971 t in 1996-97, and was 1668 t in 2004-05. Squid catch from the SESSF trawl sectors, which can comprise more than half the total catch, peaked at 893t in 2002-03 and was 583 t in 2004-05. Catch rates are variable with no clear trends, and, based on available data, there is no evidence of within-season declines in catch-per-unit-effort (CPUE) (McLoughlin 2006). Logbooks were introduced in 1986 to collect catch and effort information. Due to low effort levels, logbook data verification has not been considered necessary. There has been no routine recording of stock indicators (e.g. reproductive size or stage) from landed squid. A catch-disposal system was introduced in 2004 to gather accurate data on squid landings for use in a possible future TAC quota allocation. This data also provides the means of validating logbook catch data. In addition, there have been extensive biological studies on age and growth, genetics, reproduction life history and distribution. Kate Stark’s (pers. comm.) recent research with the University of Tasmania is of particular relevance.

Proxies against the Harvest Strategy Policy Reference Points In the absence of biomass estimates from survey or stock assessment, in place of target and limit reference points, suites of precautionary intermediate and limit catch and effort triggers were defined based on recent catch history, with values well below historical high catch levels. These serve as checks against controlled expansion, whereby the limit trigger may not be revised higher without investing in a higher Tier level assessment, the results of which provide defensible justification for doing so. The intermediate trigger levels are not associated with “hard” decision rules to limit the fishery, but rather invoke data monitoring and/or analyses in order to better inform the fishery and potentially develop more robust triggers in light of improved understanding of its dynamics. To mitigate against over-exploitation during periods of low availability, when self-regulation is not evident, there is an additional limit trigger based on effort and catch-per-unit-effort (CPUE).

General description of the harvest strategy Fishery Issues; justification for approach It is generally agreed that current catch levels of arrow squid in Commonwealth fisheries, as at 2007, are conservative: a 2006 AFMA workshop agreed that exploitation rates have been extremely low. However, given the highly variable nature of squid populations and hence their availability (both in terms of abundance and location), managers have stated that a greater concern for a squid fishery is determining when availability of squid populations is low and avoiding overfishing at these times. The proposed harvest strategy is designed to have minimal impact and costs if the fishery remains at its status quo (as at 2007), but to reach triggers invoking decision rules if the 66

fishery escalates (e.g. as a result of market changes), or to detect possible overfishing when squid populations are low. The harvest strategy will also enable the fishery to exploit and capitalize on a “boom” season – i.e. a season with high squid availability, where high take does not have an adverse effect on sustainability. Stock structure has yet to be formally resolved, and the fishery is managed as a single stock. Given the current catches and the patchy distribution of both the squid and fishing activity, spatial closures were not considered at this stage of harvest strategy development. Depletion analyses have been commonly used to undertake real-time stock assessments for squid fisheries worldwide (see for example Barton 2002 and Basson et al. 1996). Harvest Strategy Overview A system of real-time within-season management is proposed. Suites of intermediate and limit catch and effort triggers were defined based on recent catch history, with values well below historical high catch levels. These serve as checks against controlled expansion, whereby the limit trigger may not be revised higher without investing in a higher Tier level assessment, the results of which provide defensible justification for doing so. The intermediate trigger levels are not associated with “hard” decision rules to limit the fishery, but rather invoke data monitoring and/or analyses in order to better inform the fishery and potentially develop more robust triggers in light of improved understanding of its dynamics. Triggers for i) jig catch (intermediate and limit triggers), ii) jig effort (intermediate trigger), iii) combined jig and trawl catch (intermediate and limit triggers) and iv) combined Commonwealth trawl catch (limit trigger) are proposed, as follows: 1. Southern Squid Jig Fishery o Catch (2 trigger points): i) ii)

3000t intermediate trigger (this is a level that could be reached, for example during a boom) 5000t limit trigger

o Effort: 30 standard vessel intermediate trigger (where a “standard vessel” equates to a vessel carrying SFRs equivalent to 10 standard squid jigging machines, noting that the average has generally been 7-8 jigging machines). 2. Combined Commonwealth Trawl sector fisheries (note this includes much more than the GAB and SET, even though at present (2007) the majority is caught only by those 2 sectors) o 2000t catch limit trigger 3. Combined jig and trawl triggers o Catch triggers i)

4000t combined intermediate trigger

67

ii)

6000t combined limit trigger

Limit triggers may be overridden to enable industry to take advantage of “boom” seasons, during which the stock is highly unlikely to be adversely affected by the fleet fishing at full capacity. A “boom” is defined by the following “exceptional circumstance” criteria: •

Within one lunar month o The average CPUE of the entire jig fishery increases by twofold or greater;

(average CPUE to be calculated as month-specific, based on average CPUE for that month since January 1996 [the time at which a reliable catch and effort database was established])

N.B. as a general rule, average jig CPUE is 200kg/hour (this figure could potentially be used as a proxy in the absence of information – e.g. if a baseline CPUE is unavailable)

o and catch has been documented as occurring in the middle of the day and irrespective of moon phase at night To avoid over-exploitation during periods of low availability, the following criteria form an additional trigger: •

Effort is very high, defined as in excess of 45 boats (noting that the 30 boat trigger would have been reached in the interim, but that this decision rule is focused on a different objective and a more immediate response), but average CPUE per trip is low (