Final Stock Assessment Report to PIRSA

Southern Zone Rock (Jasus edwardsii) 2004/05

A. Linnane, R. McGarvey, J. Feenstra and T.M. Ward

May 2006

SARDI Research Report Series No. RD04/0164-3

This report is the fourth version of a “living” document that is updated annually as part of SARDI Aquatic Sciences’ ongoing fishery assessment program for South Australia’s Southern Zone Rock Lobster Fishery. The report provides a synopsis of information available for the fishery and assesses the current status of the resource. The report also identifies both current and future research needs for the fishery.

1 Title: Southern Zone Rock Lobster (Jasus edwardsii) Fishery 2004/05 Sub-title: Final Stock Assessment Report to PIRSA Fisheries

South Australian Research and Development Institute SARDI Aquatic Sciences 2 Hamra Avenue West Beach SA 5024

Telephone: (08) 8207 5400 Facsimile: (08) 8207 5406 http://www.sardi.sa.gov.au

The authors warrant that they have taken all reasonable care in producing this report. This report has been through SARDI Aquatic Sciences internal review process, and was formally approved for release by the Chief Scientist. Although all reasonable efforts have been made to ensure quality, SARDI Aquatic Sciences does not warrant that the information in this report is free from errors or omissions. SARDI Aquatic Sciences does not accept any liability for the contents of this report or for any consequences arising from its use or any reliance placed upon it.

© 2006 SARDI Aquatic Sciences This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the authors.

Printed in Adelaide, May 2006

SARDI Aquatic Sciences Publication No. RD04/0164-3 SARDI Research Report Series No. 135

Authors: A. Linnane, R. McGarvey, J. Feenstra and T.M. Ward Reviewers: Dr. John Carragher, Dr. Simon Bryars & Mr. Sean Sloan Approved by: Dr. Anthony Fowler

Signed: Date: 31st May, 2006 Distribution: PIRSA Fisheries, South Australian Southern Zone Rock Lobster Fishery Management Committee, SARDI Aquatic Sciences Library Circulation: Public Domain

2

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... 5

EXECUTIVE SUMMARY ...... 6

1 GENERAL INTRODUCTION...... 8 1.1 Overview ...... 8 1.2 Description of the Fishery...... 9 1.2.1 Location and Size ...... 9 1.2.2 Environmental Characteristics...... 9 1.2.3 Commercial Fishery ...... 12 1.2.4 Recreational Fishery...... 12 1.2.5 Illegal Catch...... 13 1.3 Management of the Fishery ...... 13 1.3.1 Management Milestones...... 14 1.3.2 Current Management Arrangements...... 15 1.3.3 Management Objectives and Strategies...... 17 1.3.4 Performance Indicators and Reference Points ...... 17 1.3.5 Management action on reaching a reference point outside the historical range...... 19 1.4 Biology of Southern Rock Lobster...... 20 1.4.1 Taxonomy and Distribution...... 20 1.4.2 Stock Structure ...... 20 1.4.3 Life History ...... 21 1.4.4 Growth and Size at Maturity...... 22 1.4.5 Movement...... 23 1.5 Stock Assessment...... 24 1.6 Current Research and Monitoring Programs...... 25 1.6.1 Catch and Effort Research Logbook...... 25 1.6.2 Pot Sampling ...... 26 1.6.3 Puerulus Monitoring Program ...... 27 1.6.4 Octopus Predation Project ...... 28 1.6.5 By-catch Monitoring Program...... 29

2 FISHERY STATISTICS ...... 31 2.1 Introduction ...... 31 2.2 Catch, Effort and CPUE ...... 31 2.2.1 Inter-annual Patterns...... 31 2.2.2 Within-season Patterns ...... 34 2.2.3 Patterns across MFAs ...... 40 2.2.4 Patterns across Depths ...... 43 2.3 Mean Weights ...... 50 2.3.1 Inter-annual Pattern ...... 50 2.3.2 Within-season Patterns ...... 50 2.3.3 Patterns across MFA’s...... 51 2.4 Length Frequency...... 53 2.5 Pre-Recruit Index...... 58 2.5.1 Inter-annual Patterns...... 58 2.5.2 Within-season Patterns ...... 59 2.5.3 Patterns across MFAs ...... 60 2.6 Spawning ...... 61 2.6.1 Inter-annual Patterns...... 61

3 2.6.2 Within-season Patterns ...... 62 2.6.3 Patterns across MFAs ...... 62 2.7 Lobster Mortalities...... 64 2.7.1 Inter-annual Patterns...... 64 2.7.2 Within season Patterns...... 65 2.8 Octopus Catch Rates ...... 66 2.8.1 Inter-annual Patterns...... 66 2.8.2 Within season Trends ...... 66 2.9 Changes in Patterns...... 67 2.9.1 Season Length...... 67 2.10 Distribution of Effort ...... 68 2.11 High Grading...... 70 2.11.1 Inter-annual and within Season Trends...... 70 2.12 Settlement Index...... 70 2.13 May Fishing Trial...... 72 2.13.1 CPUE ...... 72 2.13.2 Mean Weight...... 72 2.13.3 Lobster Mortality ...... 72 2.13.4 Discussion...... 75

3 THE qR MODEL...... 76 3.1 Introduction ...... 76 3.2 Methods...... 77 3.3 Results ...... 80 3.4 Discussion...... 87

4 PERFORMANCE INDICATORS...... 89 4.1 Catch Rate...... 89 4.2 Mean Weight...... 89 4.3 Abundance of Pre-recruits...... 89 4.4 Exploitation Rate...... 90 4.5 Egg Production ...... 90

5 GENERAL DISCUSSION ...... 92 5.1 Information Available for the Fishery...... 92 5.2 Current Status of Southern Zone Rock Lobster Fishery...... 92 5.3 Research in Response to DEH Recommendations...... 94 5.4 Future Research Priorities...... 96

6 BIBLIOGRAPHY...... 98

7 APPENDIX...... 103

4

ACKNOWLEDGEMENTS

Research presented in this report was commissioned by PIRSA Fisheries using funds obtained from licence fees paid by participants in the Southern Zone Rock Lobster Fishery. SARDI Aquatic Sciences provided substantial in-kind support for the project. The report builds on previous research by Dr. Tim Ward, Mr Jim Prescott, and Dr. Rob Lewis. We thank Mr Peter Hawthorne, Mr Alan Jones, Mr Matthew Hoare and Ms Kylie Howard for collecting and collating the data. The report was formally reviewed by Dr. John Carragher, Dr. Simon Bryars (SARDI Aquatic Sciences) and Mr. Sean Sloan (PIRSA Fisheries) and approved for release by Dr. Anthony Fowler (SARDI Aquatic Sciences).

5 EXECUTIVE SUMMARY

1. This report is the fourth version of a “living” document that is updated annually as part of SARDI Aquatic Sciences’ ongoing fishery assessment program for the Southern Zone Rock Lobster Fishery (SZRLF). The report provides a synopsis of the information available to the fishery, assesses the current status of the resource and identifies both current and future research needs.

2. In 2004/05, a total of 1,051,520 potlifts was required to catch the 1,900 tonne Total Allowable Commercial Catch (TACC). This reflected an increase of 0.8% from 2003/04 but a 36% decrease in effort from 1993 (1,641,876 potlifts) when the TACC was introduced (at 1,720 tonnes). In 2004/05, >85% of the catch was taken in depths of <60 m.

3. In 2004/05, commercial licence holders took an average of 94 days to catch the TACC of 1,900 tonnes, compared to 95 days in 2003/04. This was 34% less than the average number of days (143) required by licence holders to take the 1,720 tonne TACC in 1993.

4. The annual catch per unit effort (CPUE) in 2004/05 was 1.81 kg/potlift, which is 59% above the upper limit for the reference period identified in the Management Plan (1.14 kg/pot lift). This is likely to be an underestimate of true catch rate due to increased levels of highgrading in the fishery in recent seasons. During 2004/05, CPUE was highest in depths >90 m, reaching 3.5 kg/potlift in December.

5. The mean weight of lobsters in 2004/05 was 846 g, which is 0.83% above the upper limit for the reference period identified in the Management Plan (839 g).

6. The pre-recruit index for 2004/05 (calculated for November to March inclusive) was 1.31 undersize/potlift, which is inside the range for the reference period identified in the Management Plan (1.20-1.53 undersize/potlift).

7. Outputs from the qR model suggest that recruitment levels have been above the average level for the fishery in each of the last six seasons.

8. Outputs from the qR model suggest that the biomass of lobsters in the SZRLF has been increasing since 1996. In 2004/05, it was 6,530 tonnes. This represents an increase of 64% from 1993 when the biomass was estimated at 3,986 tonnes.

6 9. Outputs from the qR model suggest that the level of egg production in the SZRLF in 2004/05 was 1,500 billion eggs, which is 47% above the upper limit (1,019 billion in 1992) for the reference period identified in the Management Plan.

10. Outputs from the qR model suggest that the exploitation rate for 2004/05 was 0.28 which is 24% below the lower limit (0.37 in 1992) for the reference period identified in the Management Plan.

11. Future predictions of biomass from the qR model suggest that the biomass will increase in 2006/07 for both quotas examined (1770 and 1900 t) in response to the high puerulus settlement in 2002/03. Biomass is predicted to decrease in 2007 due to the low settlement in 2003/04, but will increase for the next two seasons due to high puerulus counts in 2004/05 and 2005/06. However, future predictions of biomass should be cautiously considered given the strong reliance fishery dependent data.

12. Despite optimistic outputs at a zonal level, this report identifies a number of localised downward trends based on fishery dependent data. Inshore catch rates in MFAs 56 and 58 have decreased notably over the last two seasons in both the 0- 30 m and 31-60 m depth ranges. In addition, the catch rate of both spawning females and undersized lobsters in both these MFAs have decreased. While current estimates of catch rate and pre-recruit index for the zone are within the range of the performance indicators in the Management Plan, these results indicate that close monitoring of these indices on a finer spatial scale may be required if localised reductions in lobster abundance is to be avoided. Overall, as >85% of the catch in the SZRLF is taken from depths of <60 m, these results highlight the need for fishery independent data that are currently lacking from both fishery statistics and stock assessment model outputs.

7 1 GENERAL INTRODUCTION

1.1 Overview

This report is the fourth version of a “living” document that is updated annually as part of SARDI Aquatic Sciences’ ongoing fishery assessment program for the Southern Zone Rock Lobster Fishery (SZRLF). It updates the 2003/04 stock assessment by Linnane et al. (2005c) with data from the 2004/05 fishing season.

The aims of the report are to provide a comprehensive synopsis of information available for the SZRLF and to assess the current status of the resource.

The report is divided into seven sections.

The first section is the General Introduction that: (i) outlines the aims and structure of the report; (ii) describes the environmental characteristics and history of the SZRLF; (iii) outlines the management arrangements for the fishery and identifies the current biological performance indicators and reference points; (iv) provides a synopsis of biological and ecological knowledge of the southern rock lobster, Jasus edwardsii; and (v) summarises previous assessments of the SZRLF.

Section two provides a synopsis of the fishery statistics for the SZRLF for the fishing seasons between 1970/71 and 2004/05. This section examines inter-annual, within- season and spatial patterns in catch, effort and catch-per-unit-effort (CPUE) in the Marine Fishing Areas (MFAs) that comprise the SZRLF. It also compares inter- annual variations in the settlement rates of puerulus with pre-recruit indices lagged by four years. This section also analyses catch rates of octopus and dead lobsters within the fishery.

The third section presents estimates of fisheries indicators obtained from the qR model (McGarvey et al. 1997; McGarvey and Matthews 2001).

The fourth section uses information provided in sections two and three to assess the status of the fishery against the biological performance indicators and reference points defined in the SZRLF Management Plan (Zacharin 1997).

Section five is the General Discussion. It synthesises the information presented, assesses the status of the fishery and the level of uncertainty in the assessment.

8 The sixth section is the bibliography, which provides a list of research papers and reports that are directly relevant to research and management of the SZRLF and/or which are cited in this report. Section seven is the Appendix.

1.2 Description of the Fishery

1.2.1 Location and Size

The Southern Zone Rock Lobster Fishery (SZRLF) includes all South Australian waters between the mouth of the Murray River and the Victorian border and covers an area of 22,000 km2 (Figure 1-1). It is divided into seven Marine Fishing Areas (MFAs), but the majority of fishing occurs in four MFAs (51, 55, 56 and 58).

Figure 1-1 Marine Fishing Areas in the Southern and Northern Zones of the South Australian Rock Lobster Fishery.

1.2.2 Environmental Characteristics

Geology and Oceanography

The sea-floor in the Southern Zone consists mainly of reefs made of bryozoan or aeolianite limestone. The limestone matrix has eroded to form ledges, crevices,

9 undercuts and holes which provide ideal habitat for lobsters. These reefs are almost continuous separated by small stretches of sand substrate (Lewis 1981).

The salinity and temperature of the surface water over the continental shelf in the southern zone cycles seasonally, with minimum salinity and maximum temperature (35.2 ppt, 18ºC) during summer (Figure 1-2) and maximum salinity and minimum temperature (35.6 ppt, 14ºC) during winter (Lewis 1981).

The water over the continental shelf is vertically well mixed during winter. However, during summer the predominant south-easterly winds result in an upwelling of nutrient-rich, cold water (11-12ºC) which intrudes onto the continental shelf (Schahinger 1987). This results in an increase in productivity of phytoplankton (Figure 1-3) which ultimately contributes to the high densities of southern rock lobster in the SZRLF (Rochford 1977; Lewis 1981).

10 Figure 1-2 Sea-surface temperatures over the continental shelf of South Australia during late summer/early autumn, 1995. In the south-east an upwelling can be seen where cooler water (dark blue) has moved onto the inner continental shelf.

Figure 1-3 Ocean colour imagery from the SeaWiFS sensor for March 2004 showing derived daily average Chlorophyll-a in mg/ m3. Areas of red are indicate production of phytoplankton. Note the enhancement of phytoplankton production off Robe in the Otway Basin and along the SE coast (green arrow) caused by the Bonney upwelling (McClatchie and Ward 2005).

11 1.2.3 Commercial Fishery

The southern rock lobster, Jasus edwardsii, has been fished in South Australian waters since the 1890’s, but the commercial fishery did not develop until the late 1940s and early 1950’s when overseas markets for frozen tails were first established (Copes 1978; Lewis 1981). There has been a gradual change to live export since then with over 90% of the current commercial catch exported live, mainly to China.

The fishery is primarily a day fishery with lobster pots set overnight and hauled at first light. The pots are steel-framed and covered with wire mesh that incorporates a moulded plastic neck (Figure 1-4). The catch is initially stored live in holding wells on boats and then transferred to live holding tanks at the numerous processing factories.

Figure 1-4 The most commonly used pot in the SZRLF.

1.2.4 Recreational Fishery

There is an important recreational fishery for lobsters in the area of the SZRLF. Recreational fishers are allowed to use drop-nets or pots or to dive for lobsters during the same season as commercial fishers. All recreational lobster pots must be registered.

Recreational potters, drop netters (with registered pots) and divers were estimated to have harvested 118 tonnes of rock lobsters, across South Australia, during the 2001 fishing season (Venema et al. 2003). This equates to 4.7% (by weight) of the

12 combined catch of commercial and recreational fishers in South Australia. This is an underestimate of the total recreational catch of rock lobsters in South Australia as it does not include the harvest of drop/hoop netters without registered pots, fishers using other gear types or the catches of charter boats.

A new survey of recreational fishers was undertaken during the 2004/05 season (Currie et al. 2006). Based on data from registered pot fishers only, the estimated State recreational catch in the 2004/05 season was 83.17 tonnes of which 74.62 tonnes came from the SZRLF and 8.56 tonnes from the NZRLF. The number of recreational pot registrations in the South Australian Rock Lobster Fishery for 2004/05 was 5,656. The number of individual pots in use was 9,827. Future estimation of the total catch of rock lobsters by recreational fishers would be enhanced by the establishment of a comprehensive database of all recreational fishers that take rock lobsters using all methods (Venema et al. 2003; Currie et al. 2006).

1.2.5 Illegal Catch

Some illegal has, and is, undoubtedly undertaken in the SZRLF. However, as in most fisheries, the size of the illegal catch has not been quantified The implementation of systems for monitoring the Total Allowable Commercial Catch (TACC) combined with the prior reporting system has reduced opportunities for the disposal of illegal catches. It is considered unlikely that illegal fishing is currently a significant source of fishing mortality.

1.3 Management of the Fishery

The commercial SZRLF is a limited entry fishery with a total of 181 licences in the 2004/05 fishing season. The majority of boats from Port MacDonnell and Robe (Figure 1-1).

The broad statutory framework for ecologically sustainable management of this resource is provided by the Act 1982. General regulations that govern the SZRLF are described in the Fisheries (General) Regulations 2000 and the specific regulations are established in the Scheme of Management (Rock Lobster Fisheries) Regulations 1991.

13 The policy, objectives and strategies to be employed for the sustainable management of the SZRLF are described in the Management Plan for the South Australian Southern Zone Rock Lobster Fishery (Zacharin 1997).

Recreational fishers are governed under the Fisheries (General) Regulations 2000.

1.3.1 Management Milestones

Management arrangements have evolved since the inception of the fishery with the commercial fishery last reviewed in 1997. The major management milestones are shown in Table 1-1.

Table 1-1 Major management milestones for the South Australian Southern Zone Rock Lobster Fishery (Zacharin 1997).

Date Management milestone

1958 Closed season for females from 1 June-31 October and for males from 1 to 31 October 1967 Pot and boat limit introduced, no new boats to operate in the then “South- Eastern Zone” 1968 Limited entry declared, compulsory commercial catch log 1978 June, July, October closed 1980 Winter closure declared. Season from 1 October to 30 April. 1984 15% pot reduction 1987 Buyback of 40 licences (2455 pots) 1993 April closed; TACC implemented for 1993/94 season at 1720 t

2001/02 TACC increased by 50 t to 1770 t

2003/04 TACC increased by 130 t to 1900; May opened on trial basis

Development and implementation of the quota management system

Seven management options were considered for the fishery (Anon. 1995). These were: (i) individual transferable quotas, (ii) total allowable catch, (iii) reduction in the numbers of pots, (iv) gear restrictions, (v) time and area closures, (vi) changes to the legal minimum size, and (vii) a buyback scheme (Zacharin 1997). From these options, the Minister for Primary Industries introduced a competitive TACC of 1,650 tonnes for the 1993 season.

14 With the TACC set for the fishery, the most difficult task of implementing the quota management system was the development of a fair and equitable method of allocating the TACC amongst fishers. The introduction of the quota management system was a controversial and complex issue for both the state government and the Fishery Management Committee to resolve. Individual transferable quotas were introduced at the beginning of the 1993 fishing season. An outline of the evolution of current TACC allocations is provided in Zacharin (1997).

1.3.2 Current Management Arrangements

Details of the management arrangements for 2004/05 are provided in Table 1-2. The commercial fishery is currently managed by a combination of input and output controls. The season extends from October 1st to April 30th of the following year. However, in 2003/04 the month of May was also opened to fishing on a trial basis only. May was also opened on a trial basis during the 2004/05 season. There is a minimum legal size of 98.5 mm carapace length, prohibition on the taking of berried females, and several sanctuaries within which lobster fishing is prohibited. The dimensions of lobster pots, including mesh and escape gap size, are also regulated. Fishers may use only 80 pots at any one time to take lobster.

The TACC is set each year and is divided evenly between licence holders as individual transferable quotas (ITQ’s). The daily catch of individual boats is monitored via catch and disposal records. The quota in 2004/05 was 1900 tonnes.

15 Table 1-2 Management arrangements for the South Australian Southern Zone Rock Lobster Fishery in 2004/05.

Control Details

Licences 181 Season October 1st to May 31st (May opened on trial basis only) Minimum legal length (both sexes) 98.5 mm CL Egg bearing females No retention Dead lobsters Landed whole with tail split and dyed Holding of live lobsters Holding on vessel, in corfs, or on land TACC 1900 tonnes Total pots 11,923 Pot lifts per day No restriction Minimum pot allocation per licence 40 Maximum pot allocation per licence 100 Maximum pots to be fished per 80 licence Pot specifications Maximum diameter 1 m; maximum height 1 m, maximum weight 40 kg; single top entrance; mesh size 50 mm diameter or escape gaps 55x150 mm Maximum vessel length No Limit

Maximum main engine BHP No Limit

Catch and effort data Daily logbook submitted monthly

Catch and disposal records Daily CDR records

16 1.3.3 Management Objectives and Strategies

Fishery management objectives and strategies are outlined in the Management Plan for the SZRLF (Zacharin 1997). The biological and environmental objectives and strategies are particularly relevant to this present report and are described below in Table 1-3.

Table 1-3 Biological and environmental objectives of the Management Plan for the South Australian Southern Zone Rock lobster fishery (Zacharin 1997).

Objective Strategy Biological Maintain lobster population at a sustainable • adopt a precautionary approach level across the fishery • set a TACC each year

Harvest rock lobster at a size likely to provide • restrict licence no’s to ≤ 185 for adequate levels of recruitment • control recreational catch • set Legal Minimum Length Environmental Minimise the environmental impact of rock • promote environmentally sensitive lobster fishing fishing practices • promote actions that reduce fishery impacts Minimise potential conflict with other users of • identify the potential for conflict with marine resources other marine resource users • determine strategies to reduce these conflicts

1.3.4 Performance Indicators and Reference Points

Information in this section of the report is taken from the Management Plan for the SZRLF (Zacharin 1997) and has been updated by Mr Sean Sloan (PIRSA Fisheries).

Reference points are agreed quantitative measures, used to assess performance of the fishery, based on clearly defined management objectives.

Reference points begin as conceptual criteria that capture in broad terms the management objectives of the fishery. To implement fishery management it must be possible to convert the conceptual reference point into a technical reference point, which can be calculated or quantified on the basis of biological or economic characteristics of the fishery (Caddy and Mahon 1995).

17 Performance indicators

Considering the stated biological objective for the fishery (Table 1-3), several performance indicators are used to assess the stock status of the SZRLF (Table 1-4). In addition, biomass, total catch and total pot lifts are also used to assess the performance of the fishery.

Table 1-4 Main performance indicators for the South Australian Southern Zone Rock Lobster Fishery.

Performance Indicator Relates to Exploitation rate level of available lobsters taken by the fishery Catch rates directly relative to current stock abundance Egg production reflects reproductive capacity of the fishery Pre-recruit abundance provides forecasting tool on future stock abundance Mean size changes in stock structure

Reference Points

It is a key goal of the Management Plan to maintain the performance indicators within the range defined in the reference points. The historical data from the seasons 1992 through 1996 have been used to define the range of the performance indicators (Table 1-5). These data are available from commercial catch returns, catch sampling programs and a stock assessment model for the fishery, for the reference seasons 1992 through 1996.

18 Table 1-5 Biological reference points for the South Australian Southern Zone Rock Lobster Fishery.

Reference season

Reference Point 1992 1993 1994 1995 1996 Range

Exploitation rate+ 0.37 0.41 0.40 0.42 0.44 0.37-0.44 Egg production (109)++ 1 019 1 018 969 926 895 895-1019 (13%) (12%) (12%) (11%) (11%) Pre-recruit abundance+++ 1.47 1.32 1.53 1.44 1.20 1.20-1.53 Catch rates (kg.potlift-1) 0.9961 1.0146 1.1383 1.0568 0.9351 0.9351-1.1383 Mean size (kg)++++ 0.7943 0.8392 0.8316 0.8282 0.8388 0.7943-0.8392

+The exploitation rate is the proportion of the population harvested annually, determined from the dynamic qR method using annual catches by weight and number. ++Total egg production (including only legal size females) has been derived from the qR stock assessment model (McGarvey et al 1997). Percent of virgin in brackets. +++The pre-recruit index is undersize catch per unit of effort (CPUE) reported in commercial catch data summed over the months of November to March (inclusive). ++++Mean size of rock lobster landed across the fishery.

1.3.5 Management action on reaching a reference point outside the historical range.

When one or more of the reference points described above are reached or exceeded, the management committee will undertake the following actions:

1. Notify the Minister for Primary Industries, Natural Resources and Regional Development and participants in the fishery as appropriate, 2. Undertake an examination of the causes and implications of ‘triggering’ a reference point, 3. Consult with the SZRLF industry and PIRSA on the need for alternative management strategies or actions, which may include: a. changes to the TACC in subsequent years or, b. changes to the minimum size limit, or c. changes to the fishing season. 4. Provide a report to the Minister and industry, within three months of the initial notification, on the outcomes of a review of the effect of triggering a performance indicator.

19 1.4 Biology of Southern Rock Lobster

1.4.1 Taxonomy and Distribution

Southern rock lobster, Jasus edwardsii (Hutton 1875) (Figure 1-5), are distributed around southern mainland Australia, Tasmania and New Zealand (Smith et al. 1980; Booth et al. 1990). In Australia, the most northerly distribution is Geraldton in Western Australia and Coffs harbour in northern New South Wales, however the bulk of the population can be found in South Australia, Victoria, and Tasmania where they occur in depths from 1 to 200 m (Brown and Phillips 1994).

Figure 1-5 Southern rock lobster, Jasus edwardsii, in reef habitat.

1.4.2 Stock Structure

Very little evidence has been found to discriminate between southern rock lobster populations. Few genetic or morphological differences that may indicate sub- structuring have been found in the Jasus edwardsii population from southern mainland Australia, Tasmania and New Zealand (Smith et al 1980; Booth et al 1990; Brasher et al. 1992). Similarly, mitochondrial DNA analysis has failed to detect any sub-division of the population on a smaller scale and it is likely that there is some exchange of genetic material from lobsters from south-eastern Australia to New Zealand (Ovenden et al. 1992). The long larval phase and widespread occurrence of larvae across the central and south Tasman Sea, in conjunction with known current flows, point to the likely transport of phyllosoma from south-eastern Australia to New Zealand providing genetic mixing between the two populations (Booth et al 1990).

20 The above notwithstanding, it is often useful to define spatially discrete fish stocks for management purposes i.e. Northern and Southern Zones of the Southern Rock lobster fishery in South Australia. In New Zealand, clustering techniques have been used to partition rock lobster statistical areas into groups based on some characteristic of the fishery i.e. trends in catch rates, size frequency distributions and size of maturity (Bentley and Starr 2001). This is used to provide aggregations of statistical areas, that to some degree, reflect fish stocks for stock assessment purposes.

1.4.3 Life History

Southern rock lobster mate from April to July. Fertilisation is external, with the male depositing a spermatophore on the female’s sternal plates (MacDiarmid 1988). The eggs are extruded shortly afterwards and are brooded over the winter for about 3-4 months (MacDiarmid 1989).

The larvae hatch in early spring, pass through a brief (10-14 days) nauplius phase into a planktonic, leaf-like phase called phyllosoma. Phyllosoma have been found down to depths of 60 m and tens to hundreds of kilometres offshore from the New Zealand coast (Booth et al. 1991; Booth and Stewart 1992; Booth 1994; Booth et al. 1999; Booth et al. 2002). They develop through a series of 11 stages over 12-23 months before metamorphosing into the puerulus (settlement) stage near the continental shelf break (Booth et al 1991; Booth and Stewart 1992; Bruce et al. 1999). The puerulus actively swims inshore to settle onto reef habitat in depths from 50 m to the intertidal zone (Booth et al 1991).

Geographic variation in larval production may be marked. In New Zealand, it has been suggested that this may be due to variations in: (i) size at first maturity, (ii) breeding female abundance and/or (iii) egg production per recruit (Booth and Stewart 1992). Additionally, phyllosoma are thought to drift passively which, coupled with the long offshore larval period, means that oceanographic conditions, particularly currents and eddies, may play an important part in their dispersal (Booth and Stewart 1992).

Geographic patterns in the abundance of phyllosoma may also be consistent with those in puerulus settlement (Booth and Stewart 1992; Booth 1994). Correlations between levels of settlement and juvenile abundance have been found at two sites in

21 New Zealand (Breen and Booth 1989; Booth and Stewart 1993). In South Australia, it has been suggested that the strength of westerly winds, during late winter and early spring, may play a role in the inter-annual variation in recruitment to the SZRLF (McGarvey and Matthews 2001). In this study, both winds and recruitment were shown to exhibit a 10-12 year periodicity, with significant correlations between recruitment and westerly winds lagged by 5-7 years.

Figure 1-6 Phyllosoma collected in plankton tow from south coast of Kangaroo Island in February 2005.

1.4.4 Growth and Size at Maturity

Lobsters grow through a cycle of moulting and thus increase their size incrementally (Musgrove 2000). Male and female moult cycles are out of phase by 6 months, with males undergoing moulting between October and November, and females during April to June (MacDiarmid 1989).

McGarvey et al., (1999a) demonstrated that there was substantial variation in growth rates between locations in South Australia. Growth rates also varied throughout the life of individuals with the mean annual growth for lobsters at 100 mm carapace length (CL) ranging from 7-20 and 5-15 mm.yr-1 for males and females respectively. Growth rates tended to increase along the South Australian coast from south-east to north-west and were highest in areas of low lobster density and high water

22 temperature. Growth rates also appeared to be related to depth of habitat and declined at the rate of 1 mm (CL)/yr-1 for each 20 m increase in depth (McGarvey et al 1999a).

The size at which 50% of f emales are sexually mature appears spatially variable, ranging between approximately 90 mm and 115 mm (CL) (Prescott et al. 1996).

1.4.5 Movement

In South Australia, movement patterns of the southern rock lobster Jasus edwardsii were determined from 14,280 tag-recapture events from across the State between 1993 and 2003 (Linnane et al. 2005a). In total, 68% of lobsters were recaptured within 1 km of their release site and 85% within 5 km (Figure 1-7). The proportion of lobsters moving >1 km in Marine Fishing Areas (MFAs) ranged from 13 to 51%. Movement rates were noticeably high in the south-east and at Gleesons Landing lobster sanctuary off the Yorke Peninsula (refer to Figure 1-1) but patterns of movement differed spatially. In the south-east, lobsters moved distances of <20 km from inshore waters to nearby offshore reefs whereas off the Yorke Peninsula individuals moved distances >100 km from within the sanctuary to sites located on the north-western coast of Kangaroo Island and the southern end of Eyre Peninsula.

These results support findings from an earlier tag–recapture study where most recaptured lobsters had moved short distances with only a small proportion having moved distances greater than a few kilometres, up to 28 km (Lewis 1981). Similarly, larger movements were generally in an offshore direction and were from the Kingston-Cape Jaffa region which is adjacent to the Coorong. High site fidelity was also found in tagging studies conducted in Tasmania with more than 90% of tagged lobsters moving less than 5 km (Gardner et al. 2003). However, consistently greater movement did occur from areas to the north of Tasmania. All the above studies indicated that immature lobsters moved greater distances than mature individuals.

23 80

70

60 N=14,280 50

40

% Proportion 30

20

10

0 0-1 1-5 5-20 20-50 > 50 Distance (km)

Figure 1-7 Percentage of lobsters that moved within a range of distance categories based on distance from initial tagging to final recapture locations across all marine fishing areas in South Australia (from Linnane et al 2005a).

1.5 Stock Assessment

The first stock assessment for the SZRLF was conducted by Copes (1978) who plotted a yield curve of catch (tonnes) against effort (pot lifts) and applied the simplest version of the Schaefer Model to estimate yield. The results suggested that the stable catch-effort relationship for the fishery was about 1,600 t from 2,000,000 pot lifts. It is notable that Copes suggested:

“the effective intensity of effort attributable to a pot lift has changed in recent years. With greater skill and experience fishermen, with new electronic equipment, the positioning of pots has become more effective over the years, so that a pot lift has gradually become a larger unit of effort” (Copes 1978, p 41).

Lewis (1981) superimposed additional data on the yield curves generated by Copes (1978) and suggested that the potential yield from the fishery was best described by

24 curves indicating a yield of between 1,600 and 1,800 tonnes. Lewis (1981) also noted that the yield curves indicated that an effort reduction of approximately 900,000 pot lifts would result in the same total catch. At the time the report was written (1981), 1,730,000 pot lifts resulted in a total catch of 1,700 tonnes. In 2001, 910,000 pot lifts resulted in the same total catch as that attained in 1981.

Since the mid-1990s, the qR model (McGarvey et al 1997; McGarvey and Matthews 2001) has provided the basis for reporting against the performance indicators for the fishery (exploitation rate and egg production). Like most stock assessment models, the qR model has undergone a process of continuous refinement (McGarvey and Matthews 2001). Outputs of the latest version of the qR model are presented in section three of this present report.

1.6 Current Research and Monitoring Programs

SARDI Aquatic Sciences is contracted by PIRSA Fisheries Policy Group to: (i) administer a daily logbook program, (ii) collate catch and effort information, (iii) conduct pot-sampling, bycatch, puerulus and fishery independent monitoring programs and (iii) produce annual stock assessment and status reports that assesses the status of the SZRLF against the performance indicators defined in the Management Plan.

1.6.1 Catch and Effort Research Logbook

Licence holders complete a compulsory daily logbook which has been amended to accommodate changes in the fishery. During 1998, the logbook was modified to include specific details about King crab (Pseudocarcinus gigas) fishing depth when it was found that on some fishing trips, fishers split their gear between lobster and crab fishing. In the 2000 fishing season, the logbook was amended and the recording of undersize, spawning and dead lobster, along with numbers of octopus became voluntary. Logbook returns are submitted monthly and are entered into the South Australian Rock Lobster (SARL) database.

25 Details currently recorded in the daily logbook include:

1. the MFA within which the fishing took place, 2. depth in which the pots were set, 3. number of pots set, 4. weight of retained legal-sized lobsters - reported at the end of each trip or as a daily estimated weight, 5. landed number of legal-sized lobsters, 6. number of undersized lobsters caught, 7. number of dead lobsters caught, 8. number of spawning lobsters caught, 9. weight of octopus caught, 10. number of octopus caught, 11. number of giant crab pots, 12. depth of giant crab pots, 13. landed weight of giant crabs, 14. landed number of giant crabs.

Validation of catch and effort logbook data in the SZRLF can be achieved by comparing them with the catch and disposal records (CDRs) used in the quota management system. Processor records are not used for validation as lobsters may be transported to processors outside of the zone in which the lobsters were landed.

1.6.2 Pot Sampling

Since 1991, commercial fishers and researchers have collaborated in an at-sea pot- sampling program. During the life of this program there were various levels of participation, and changes to the sampling regime. The program started with commercial fishers sampling from several (usually 3) pots each day, for the duration of the fishing season. During the 1995 season, sampling was reduced to one week each month over the period of the third quarter of the moon. During the following season, sampling was done as part of an FRDC project that aimed to determine the optimal sampling strategy required to produce quantifiable and minimum variances in the mean lengths and catch rates (McGarvey et al. 1999b; McGarvey and Pennington 2001). This study demonstrated that the optimal design should incorporate a high percentage of boats, with sampling done on as many days as possible from a small fraction of the pots from each boat. During the 1997 and 1998 seasons, fishers were encouraged to follow this sampling strategy. They were supported by research staff who went to sea on commercial vessels to encourage more fishers to participate in the program and to demonstrate the methods to new participants.

26 Participation in the program is neither random nor systematic and participation in the pot sampling program varied among areas and tended to taper off as the season progressed (Prescott et al. 1999). In addition, overall participation in the program has decreased over the last number of seasons (Figure 1-8) although this increase slightly to over 30 % in the 2004/05 season. Low participation in the programme may bias catch rates and length frequencies. One solution to this problem would be to ensure that all fishers provide detailed information from a small number of pots (3-5) on every trip.

50

40

30

20

% Licence holders % Licence 10

0 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 Season

Figure 1-8 Percentage of licence holders participating in the SZRLF catch sampling program over the last 6 seasons.

1.6.3 Puerulus Monitoring Program

Larval recruitment processes may be related to changes in breeding stock abundance and seasonal, annual and geographic variation in recruitment to the fishery (Booth et al. 2002). As a result, knowledge of these processes may ultimately improve the usefulness of fishery assessment models.

The monthly occurrence of puerulus settlement in crevice collectors has been studied in the SZLRF at 5 main sites since 1990 (Linnane et al. 2005c). These sites are located at Blackfellows Caves, Livingston Beach, Beachport, Cape Jaffa and Kingston with the collectors set in groups of 10 or 12. The annual Puerulus Settlement Index (PSI) is calculated as the mean monthly settlement on these

27 collectors. This index is then related to annual pre-recruit index (PRI) and model estimated recruitment lagged by three and four years respectively.

1.6.4 Octopus Predation Project

Mortality of lobsters due to predation in pots, especially by Maori octopus (Octopus maorum) is a significant problem in the South Australian Rock Lobster Fishery (SARLF), but has generally been considered to be unavoidable, resulting in minimal effort being expended in determining the scale of the problem or investigating a solution. In 1998, a project entitled “Development and assessment of methods to reduce predation of ‘pot caught’ southern rock lobster (Jasus edwardsii) by Maori octopus (Octopus maorum)” was initiated. This project was funded by the Fisheries Research and Development Corporation (Project 1998/150). This project was initiated to quantify levels of octopus predation and investigate methods for reducing rates of lobster mortality in pots, the findings of which are published in Brock and Ward (2004) and Brock et al. (2006 a and b). Some of the main outcomes from the study are as follows:

• Since 1983, between 38,000 and 119,000 octopuses per annum have been taken in SARLF traps • Over the period 1998-2003, approximately 240,000 lobsters per annum were killed in traps, representing ~4% of the total catch. • Field studies show that over 98% of within-trap lobster mortality is attributable to octopus predation. Lobster mortality rates are positively correlated with the catch rates of octopus and lobster. • Aquarium studies showed that octopuses were primarily attracted to traps by the presence of bait as opposed to lobsters and that octopus entry into traps was ‘fortuitous’ and mediated by speculative exploration. • The presence of escape gaps did not significantly affect the predation rates of lobsters above the minimum legal size, but significantly reduced the retention and subsequent mortality of under-sized lobsters. • The presence of an escape gap did not affect legal sized lobster catch rates. • A two-chambered lobster trap was developed that in aquarium and field trials significantly reduced octopus predation on trap-caught spiny lobster by 45- 48% but which also lead to reduced catch rates of legal sized lobster. • Lobster mortality rates were positively correlated with soak-times in the Southern Zone fishery and lobster size. Minimizing soak-times is one method currently available for reducing lobster mortality rates.

28 1.6.5 By-catch Monitoring Program

A report detailing the species composition and spatio-temporal trends in by-catch from the South Australian commercial rock lobster fishery was finalised in 2004 (Brock et al. 2004). The report identifies the main by-catch species within the fishery and estimates catch rates of by-catch as determined during the 2001/02 and 2002/03 fishing seasons. It also compares the effectiveness of logbook and observer sampling strategies and comments on the appropriateness of each for application within the South Australian rock lobster fishery.

In addition to the study by Brock et al (2004), ongoing monitoring of by-catch from the SZRLF is undertaken annually by SARDI scientists during routine onboard catch sampling (Figure 1-9). The results indicate that over the last three seasons, by-catch has been dominated by crustaceans (mainly velvet and hermit crabs) and temperate reef finfish namely leatherjacket (dominated by the horseshoe leatherjacket Meuschenia hippocrepis; Figure 1-10) and wrasse species (dominated by the blue throat wrasse Notolabrus tetricus; Figure 1-11). The remainder of by-catch was composed of slimy cod and other species. A risk assessment of by-catch species associated with the SZRLF is planned as part of the new Management Plan for the fishery.

100%

90%

80%

70% Others 60% Slimy Cod sp. 50% Wrasse sp. Crustacean sp. 40% Leatherjacket sp. 30%

Percentage of total By-catch 20%

10%

0% 2002 2003 2004 Season

Figure 1-9 Species composition of by-catch from the SZRLF from 2002 to 2004 as determined from routine onboard catch sampling.

29

Figure 1-10 Horseshoe leatherjacket (Meuschenia hippocrepis).

Figure 1-11 Blue throat wrasse (Notolabrus tetricus).

30 2 FISHERY STATISTICS

2.1 Introduction

This section of the report summarises and analyses fishery statistics for the SZRLF for the period between 1st January 1970 and 31sh May 2005. For ease of reference, figures and text in this section refer to the start of season year e.g. 2004 refers to the 2004/05 fishing season. Estimates presented in this section are calculated from daily data and differ slightly from estimates based on season totals that are presented in other sections of this report. Daily data are used to describe the inter-annual and within-season patterns in catch (kg), effort (potlifts), catch-per-unit-effort CPUE (kg/potlift), mean weight (kg/lobster), undersized lobsters and spawning lobsters in the main MFAs of the SZRLF and in four depth classes (0-30, 31-60, 61-90 and >90 metres). Data obtained from the commercial pot sampling program provide the length frequency distributions of lobsters sub-divided by MFA and season. Estimates of inter-annual variations in settlement of puerulus are compared with pre-recruit indices and model estimated recruitment lagged by three and four years respectively.

2.2 Catch, Effort and CPUE

2.2.1 Inter-annual Patterns

Catch

Fishing patterns between 1970 and 1983 were highly variable and some discrepancies exist between the published catches for this period and those extracted from the SARL database. Estimates of absolute catch should thus also be viewed with some caution (Figure 2-1). The highest published catch during this period was in 1971 when approximately 2,000 tonnes were landed. The lowest published catch during this period was 1,250 tonnes in 1976.

Between 1984 and 1990 catches remained steady at around 1,500 tonnes and then rose to 1,940 tonnes in 1991 before declining again to 1,670 tonnes in 1993. In 1993, a TACC of 1,720 tonnes was introduced, but only 1,668 tonnes was harvested. From 1993 to 1997, the only year in which the entire TACC was taken was 1994. The TACC was taken from 1998 through to 2002 with a quota increase of 50 tonnes to

31 1,770 tonnes implemented in 2001. In 2003, the TACC was again increased to 1,900 tonnes. In 2004, the total reported commercial catch was 1,897 tonnes.

Effort

Estimates of effort between 1970 and 1983 should be viewed with caution (Figure 2-1). A peak in effort (2.3 million pot lifts) was recorded at the end of this period in 1983. The lowest level of effort in this period was in 1974, when 1.3 million pot lifts were recorded.

Over the next decade, effort declined steadily from 2.3 million pot lifts in 1983 to 1.5 million in 1994. Effort then rose again to 1.7 million pot lifts in 1997, before falling rapidly to the lowest recorded level of 854,000 pot lifts during the 2002 season. In the 2003, a total of 1,042,233 potlifts were required to catch the 1,900 tonne TACC. This was an increase in effort of 18% from 2002 when the TACC was 1,770 tonnes. In 2004, a total of 1,051,520 potlifts were required to catch the TACC (retained at 1,900 tonnes), an increase of 0.8% from 2003.

CPUE

CPUE during the 1970s was between 0.70 and 0.90 kg/pot lift (Figure 2-2). In 1980, the CPUE reached a pre-quota peak of 1.06 kg/pot lift and then declined to 0.77 kg/pot lift in 1983. CPUE remained steady at around 0.75 kg/pot lift from 1983-1988 before rising to 0.99 kg/pot lift in 1992 (the year prior to introduction of the TACC).

The TACC was introduced in 1993. In 1994, the CPUE had risen to 1.12 kg/pot lift but declined to 0.93 kg/pot lift in 1996. Since then, it has increased substantially reaching 2.1 kg/pot lift in 2002. In 2004, the annual CPUE was 1.81 kg/pot lift, the third highest in the history of the fishery. It should be noted that this estimation does not take into account lobsters that were high graded during the season i.e. returned to the water due to damage or having a low size related market value.

32 2500 Catch (publ.) 2500 Catch Quota Introduced Effort 2000 2000 Effort (000's pot lifts pot (000's Effort

1500 1500

1000 1000 Catch (tonnes) Catch )

500 500

0 0 70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 Season

Figure 2-1 Inter-annual trends in catch and effort in the South Australian SZRLF for seasons between 1970 and 2004.

2.2 2.0 1.81 kg/potlift 1.8 1.6 Quota introduced 1.4 1.2 1.0 0.8

CPUE (kg/potlift) 0.6 0.4 0.2 0.0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season Figure 2-2 Inter-annual trends in catch-per-unit-effort (CPUE) in the South Australian SZRLF for seasons between 1970 and 2004.

33 2.2.2 Within-season Patterns

Catch

Fishing in 1970 through 1979 was conducted all year round although the majority of the catch was taken in the summer months (Figure 2-3). A seven-month fishing season from October to April was implemented in 1980. From 1980 to 1996, the highest catches were almost always taken in the fist six months of the season before dropping off in April (Figure 2-4 to 2.5). Since 1999, the fishery has recorded high monthly catches for the period from October to January, with catches declining sharply after January as the bulk of the TACC had generally been taken by this time.

In 2003, a trial extension of the season until May 31st was approved with approximately 30 tonnes landed during that month (Figure 2-6). May was also opened in 2004 during which 16.6 tonnes were landed (refer to section 2.13). Catches in April of both 2003 and 2004 were higher than those of 2002 (where May was closed) reflecting a greater level of flexibility within the fishery in 2003 and 2004 to spread catch over time.

Effort

Between 1980 and 1993, fishing effort was consistently high from October to March before dropping sharply during April (except in 1992; Figures 2-4 to 2-5). This probably reflects a seasonal decline in catch rate and the fact that there are relatively more fishable days during summer than at other times of the year. This trend continued until after the introduction of quota in 1993 until 1996. Since 1998, effort began to decline rapidly after January, which was much earlier than in previous seasons, reflecting that the majority of the TACC had been taken by this time. In 2004, within season trends in effort, generally reflected trends in catch (Figure 2-6).

34 1970 Catch 1975 400 400 Effort 300 300 200 200

100

100 ('000s pl) Effort Catch (tonnes) Catch

0 0 No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep t Nov Dec Jan Feb M ar Apr M ay Jun Jul Aug Sep

1971 1976 400 400 300 300

200 200

100 100 Effort ('000s pl) Catch (tonnes)

0 0 No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep Nov Dec Jan Feb M ar Apr M ay Jun Jul Aug Sep

1972 1977 400 400 300 300

200 200

100 100 Effort ('000s pl) Catch (tonnes) Catch

0 0 No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep

1973 1978 400 400 300 300

200 200

100 100 Catch (tonnes) Catch ('000spl) Effort

0 0 No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep

1974 1979 400 400 300 300

200 200

100 100 Catch (tonnes) ('000s pl) Effort

0 0 No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep No v Dec Jan Feb M ar A p r M ay Jun Jul A ug Sep

Figure 2-3 Within-season trends in catch and effort in the SZRLF for the fishing seasons between 1970 and 1979.

35 400 1980 Catch 1986 400 Effort 300 300 200 200

100

100 Effort ('000s pl) Catch (tonnes) Catch

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb M ar Apr

1981 1987 400 400 300 300 200 200

100 100 Effort ('000s pl) Catch (tonnes)

0 0 Oct No v Dec Jan Feb M ar A p r Oct Nov Dec Jan Feb M ar Apr

1982 1988 400 400 300 300 200 200

100

100 Effort ('000s pl) Catch (tonnes) Catch

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb M ar Apr

1983 1989 400 400 300 300 200 200

100

100 Effort ('000s pl) Catch (tonnes) Catch

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb Mar Apr

400 1985 1990 400 300 300

200 200

100

100 Effort ('000s pl) Catch (tonnes) Catch

0 0 Oct No v Dec Jan Feb M ar A p r Oct Nov Dec Jan Feb Mar Apr Figure 2-4 Within-season trends in catch and effort in the SZRLF for the fishing seasons between 1980 and 1990.

36 Catch 1992 1997 400 400 Ef f ort 300 300 200 200

100 100 Effort ('000s pl) ('000s Effort Catch (tonnes) Catch

0 0 Oct No v Dec Jan Feb M ar Oct Nov Dec Jan Feb Mar Apr

1993 1998 400 400 300 300 200 200

100

100 pl) ('000s Effort Catch (tonnes) Catch

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb Mar Apr

1994 1999 400 400 300 300 200 200

100

100 Effort ('000s pl) Catch (tonnes) Catch

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb Mar Apr

1995 2000 400 400

300 300

200 200

100 100 Catch (tonnes) Catch Effort ('000s pl)

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb Mar Apr

1996 2001 400 400 300 300

200 200

100 100 Catch (tonnes) Catch Effort ('000s pl)

0 0 Oct Nov Dec Jan Feb Mar Apr Oct Nov Dec Jan Feb M ar Apr Figure 2-5 Within-season trends in catch and effort in the SZRLF for the fishing seasons between 1992 and 2001.

37 2002 450 400

400 CATCH 350 potlifts) (*1000's Effort 350 EFFORT 300 300 250 250 200 200

Catch (t) Catch 150 150 100 100 50 50 0 0 Oct Nov Dec Jan Feb Mar Apr Month

2003

450 400

400 350 potlifts) (*1000's Effort CATCH 350 300 300 EFFORT 250 250 200 200

Catch (t) Catch 150 150 100 100 50 50 0 0 Oct Nov Dec Jan Feb Mar Apr May Month

2004

450 400 400 350 CATCH Effortpotlifts) (1000's 350 EFFORT 300 300 250 250 200 200

Catch (t) Catch 150 150 100 100 50 50 0 0 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-6 Comparison of within-season trends in catch and effort in the SZRLF for the 2002, 2003 and 2004 fishing seasons.

38 CPUE

The within-season trend in CPUE was similar during the 1970’s, 1980’s and 1990’s (Figure 2-7). There was a distinct seasonal pattern of high CPUE during the summer months and lower CPUE at the beginning and end of the fishing seasons. During 2004, CPUE was highest in January (2.07 kg/potlift) as in previous seasons (Figure 2-8) and thereafter decreased to 1.54 kg/potlift in May.

1.5

1970 - 80 1981 - 90 1.0 1991 - 00 CPUE (kg/pot lift) CPUE (kg/pot

0.5

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Month

Figure 2-7 Within-season trends in CPUE (mean ± SE) in the SZRLF for the 1970s, 80s, 90s.

2.5

2.0 2001 2002 2003 2004

CPUE (kg/pot lift) 1.5

1.0 Oct Nov Dec Jan Feb Mar Apr May Month Figure 2-8 Within season trends in CPUE (mean ± SE) in the SZRLF from 2001 to 2004.

39 2.2.3 Patterns across MFAs

Catch

While fishing is typically undertaken in seven MFAs in the SZRLF, the majority of the catch is taken in MFAs 51, 55, 56 and 58 (Figure 2-9, Figure 2-10, refer to Figure 1-1). In the 2004 season, 98.5% of the total catch came from these four MFAs. Prior to 1983, catches were similar between MFAs 55 and 56 but since then the highest catches have been consistently recorded in MFA 55. In 2004, catches were 689, 615 and 520 tonnes in MFAs 55, 56 and 58, respectively. The catch in MFA 51 has declined from around 200-300 tonnes during the 1970’s to less than 100 tonnes over the last 5 years. Data in section 2.3.3 indicates that lobsters harvested from MFA 51 are generally larger in size and thus have low market value given the preference for smaller individuals. In 2004, just 44 tonnes were harvested in MFA 51.

Effort

The majority of fishing effort is expended in MFAs 51, 55, 56 and 58 (Figure 2-10). The greatest relative change in effort has been in MFA 51, where effort has decreased from ~300,000 pot lifts in 1972 to ~18,000 pot lifts in 2004. In MFA 55, effort increased from the 1970s to a peak of ~905 000 pot lifts in 1983 and has decreased since then to ~326,000 pot lifts in 2004. Similarly, there has been a gradual decline in effort since the early 1980’s in MFAs 56 and 58. In 2004, effort increased in MFA 51 and MFA 58 by 11 and 7%, respectively, from 2003 figures. Effort decreased in MFA 55 by 6% and remained constant in MFA 56 from 2003 figures.

CPUE

Trends in CPUE are generally similar for the main MFAs of the SZRLF (Figure 2-11). Prior to 1998, the CPUE ranged from 0.63 – 1.28 kg/pot lift. Generally, CPUE has tended to increase since 1970, with the highest CPUEs occurring in MFA 51 and the lowest in MFA 58.

During the early 1990’s, CPUE in the main MFAs rose slowly until 1998, after which CPUE increased rapidly. The CPUE in 2002 was the highest recorded for each MFA, reaching levels of 2.0, 2.4, 2.5 and 1.6 kg/pot lift in MFAs 51, 55, 56 and 58, respectively. Prior to the introduction of the TACC, the highest CPUE values

40 recorded for these MFAs were 1.3, 1.2, 1.0 and 0.9 kg/pot lift respectively. In 2004, CPUE increased in MFA 51 (from 1.9 to 2.4 kg/potlift) and MFA 55 (2.0 to 2.1 kg/potlift) and decreased in MFA 56 (1.98 to 1.91 kg/potlift) compared to 2003 estimates. CPUE remained at ~1.41 kg/potlift in MFA 58.

2%

28%

MFA 51 37% MFA 55 MFA 56 MFA 58

33%

Figure 2-9 Proportion of the catch taken from each of the major MFAs in the SZRLF in 2004.

41 for thefishing seasons between 1970 and 2004. Figure 2-11 Inter-annual trends in CPUE (±SE CPUE (kg/pot lift) fishing seasons between 1970 and 2004. Figure 2-10 Inter-annual trends inthemainMFAsofSZRLFfor incatchandeffort Catch (tonnes) 200 400 600 800 200 400 600 800 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0 0

1970 MFA 51 MFA MFA 55 1970 55 MFA 1972 1972 1974 1974 1976 1978 Yr vs MFA Eff 51 Ca MFA51 1976 1980 1978 1982 1980 1984 MFA 51 1982 1986 1984 1988 1986 1990 1992 1988 1994 1990 1996 1992 1998 1994 2000 1996 2002 1998 2004 0 200 400 600 800 1000 2000 0 200 400 600 800 1000 2002 Season 200 400 600 800

2004 200 400 600 800 of the mean) ofthe main MFAs ofthe SZRLF 0 0

1970 1972 MFA 56 MFA 1970 58 MFA 1974 1972 1976 1974 1978 1976 1980 1978 1982 1980 1984 MFA 58 MFA 56 1986 1982 1988 1984 1990 1986 1992 1988 1994 1990 1996 1992 1998 2000 1994 2002 1996 2004 0 200 400 600 800 1000 1998 0 200 400 600 800 1000 2000

2002 42 2004 lifts) pot (x1000 Effort 2.2.4 Patterns across Depths

Catch by depth

During the 1970’s the majority of the catch was taken equally from the depth ranges of 0-30 m and 31-60 m (Figure 2-12) with ~20% of the catch coming from depths >60 m. Over the next 2 decades, the proportion of the catch taken from depths >30 m increased. This was presumably influenced by the gradual adoption of modern technology within the fishing fleet that allowed deeper water to be fished more easily. Over the last three fishing seasons however, >85% of the catch has been landed from depths <60 m. In 2004, <2% of the catch was taken in depths >90 m.

In MFA 51, the majority of the catch has been taken from the 31-60 m depth range over the last three seasons (Figure 2-13). This trend continued in 2004, with 77% of the catch coming from this depth range. Similarly, over 60% of the catch in MFA’s 55 and 56 over the last three seasons has come from the 31-60 m depth range. In 2004, ~14% of the catch was taken in depths >60 m in MFA 55 while 8% of the catch was taken in depths >60 m in MFA 56.

In MFA 58, >85% of the catch has come from depths <60 m over the last three seasons. In addition, the proportion of the catch taken from the shallower depth range of 0-30 m is higher in this MFA compared to others in the zone. Over the last four seasons >45% of the catch has come from 0-30 m (Figure 2-13) which is a substantial increase compared to previous years. In the 2004 season, 14% of the catch was taken from depths > 60 m in MFA 58.

Annual CPUE by MFA and depth.

In MFA 51, annual CPUE generally increased from 1997 peaking at 1.82 and 2.12 kg/potlift in the 0-30 and 31-60m depth ranges in 2002 (Figure 2-14). CPUE decreased in 2003 in both the 0-30m and 31-60m depth ranges to 1.68 and 1.86 kg/potlift respectively before increasing to 2.09 (0-30m) and 2.52 kg/potlift (31-60m) in 2004. Similar trends in CPUE were observed over the same time period in MFA 55 with CPUE estimated at 1.95 and 2.05 kg/potlift in the 0-30 and 31-60m depth ranges respectively in 2004 (Figure 2-15).

43 In MFA 56, CPUE also increased from 1997 peaking at over 2 kg/potlift in all depth ranges in 2002 (Figure 2-16). However, since 2002, CPUE has declined in all depth ranges in MFA 56 particularly in 0-30 and 31-60m where it decreased to 1.74 and 1.93 kg/potlift respectively in 2004. Similar trends of decreasing CPUE in shallow depth ranges have been observed over the last two seasons in MFA 58 (Figure 2-17). In particular, the estimate of 1.26 kg/potlift in the 0-30m depth range in MFA 58 is the lowest on record since 1996.

Given the low percentage of overall catch taken from depths >60 m (Figure 2-12) data used to calculate CPUE in depths ranging from 60-90m are limited and should be treated with caution. Overall trends show a general increase in CPUE from 1997 to 2002 across all major MFAs followed by a decrease in MFAs 55 and 56, with a marginal increase in MFA 58 (Figure 2-14 to Figure 2-17).

Within Season CPUE by depth

Seasonal CPUE patterns with depth show that after the 1980’s, CPUE increased with depth, and that the pattern of high CPUE in summer is consistent across all depth ranges (Figure 2-18). Since the 1980’s, the highest CPUEs in each month consistently occurred in fishing depths > 90 m. Prior to 1980, CPUE from the depth range 61 – 90m was higher than for depths over 90 m. The increase in CPUE at depths > 90 m since the 1980s was most likely related to improvements in fishing efficiencies mediated by improvements in technology and boat design.

In 2004, CPUE in all depth ranges was generally highest in December/January (Figure 2-19). Overall, CPUE was highest in the 61-90 m and > 90 m ranges, but as in the previous seasons of 2002 and 2003, less than 20% of the overall catch was taken in these depth ranges (Figure 2-12).

44 100% 90% 80% 70% 60% Depth range (m) 50% 0 - 30 m 40% % Catch 31 - 60 m 30% 61 - 90 m 20% > 90 m 10% 0% 1970-80 1981-90 1991-00 2001 2002 2003 2004 Season

Figure 2-12 Percentage of catch taken from four depth ranges in the SZRLF during the 1970s, 1980s, 1990s and 2001, 2002, 2003, 2004 fishing seasons.

0 - 30 m 31 - 60 m 61 - 90 m MFA 51 MFA 56 > 90 m 100%

80%

60%

40%

20%

0%

MFA 58 MFA 55

100%

80%

60%

40% % Catch from each depth range each depth % Catch from 20%

0% 1970-80 1981-90 1991-00 2001 2002 2003 2004 1970-80 1981-90 1991-00 2001 2002 2003 2004

Figure 2-13 Percentage of catch taken from four depth ranges in the four main MFAs of the SZRLF during the 1970s, 1980s, 1990s and 2001, 2002, 2003, 2004 fishing seasons.

45

MFA 51 3.5 0-30 m 3 31-60 m 61-90 m 2.5

2

1.5

CPUE (kg/potlift) 1

0.5

0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-14 CPUE in various depth ranges from 1970 to 2004 in MFA 51.

MFA 55 3 0-30 m 2.5 31-60 m 2 61-90 m

1.5

1 CPUE (kg/potlift) CPUE 0.5

0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-15 CPUE in various depth ranges from 1970 to 2004 in MFA 55.

46 MFA 56 3.5

3 0-30 m 2.5 31-60 m 61-90 m 2 1.5 1 CPUE (kg/potlift) 0.5 0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-16 CPUE in various depth ranges from 1970 to 2004 in MFA 56.

MFA 58

2.5 0-30 m 2 31-60 m 61-90 m 1.5

1 CPUE (kg/potlift) CPUE 0.5

0 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-17 CPUE in various depth ranges from 1970 to 2004 in MFA 58.

47

3.0 1970 - 80 1991 - 00 2.5 2.0 1.5 1.0 0.5 0.0

3.0 1981 - 90 2001 2.5

CPUE (kg per potlift) per (kg CPUE 2.0 0 to 30 m 1.5 31 to 60 m 1.0 61 to 90 m 0.5 > 90 m 0.0 Jul Oct Apr Jun Jan Jul Feb Mar Nov Aug Sep Dec May Oct Apr Jun Jan Mar Feb Nov Dec Aug Sep May Figure 2-18 Mean CPUE (± SE of mean) in four depth ranges in the SZRLF during the 1970s, 1980s, 1990s and the 2001 fishing season.

48 3.0 2002

2.5 0 to 30 m 2.0 31 to 60 m 61 to 90 m 1.5 > 90 m

1.0 CPUE (kg per potlift)

0.5

0.0 Apr Oct Jan Mar Feb Nov Dec May

5.0 2003 4.5

4.0

3.5

3.0

2.5 0-30 m 2.0 31-60 m 1.5 61-90 m CPUE (kg potlift) per > 90 m 1.0

0.5

0.0 Apr Oct Jan Mar Feb Nov Dec May 4.5

4.0 2004

3.5

3.0 0-30 m 2.5 31-60 m 2.0 61-90 m > 90 m 1.5

CPUE (kg per potlift) 1.0

0.5

0.0 Apr Oct Jan Mar Feb Nov Dec May Month

Figure 2-19 Mean CPUE (± SE of mean) in four depth ranges in the SZRLF for the 2002, 2003 and 2004 fishing seasons.

49 2.3 Mean Weights

2.3.1 Inter-annual Pattern

The annual mean weight of lobsters remained relatively unchanged through the 1970’s and early 1980’s with the average lobster weighing around 0.87 kg (Figure 2-20). From 1982, the mean weight of lobsters decreased until 1991 when it reached 0.79 kg before rising to about 0.84 kg in the mid 1990’s. Mean lobster weight reached an all time low of 0.76 kg in 1999 and then increased over the next 4 seasons to reach 0.85 kg in the 2003 fishing season. In 2004, it decreased marginally to 0.84 kg. In general, the pattern of rise and fall in mean size reflects long-term patterns of recruitment, with low mean weights resulting from influxes of small lobsters into the fishable biomass and high mean weights resulting from several consecutive years of low recruitment. However, highgrading (the selection of smaller sized individuals due to higher unit value) is now a significant feature under the current quota system in the SZRLF. The result is that fishing behaviour undoubtedly affects current annual estimates of lobster size in the SZRLF. The practice highlights the need for fishery independent data in order to get a robust estimate of this statistic.

2.3.2 Within-season Patterns

Since the 1970’s there has been a consistent trend of increasing lobster mean weight as the fishing season progresses (Figure 2-21). On average, the smallest lobsters are caught in October/November and the largest later in the fishing season. The seasonal pattern in lobster mean weight was almost identical during the 1970’s and 1980’s. While the trend towards larger lobsters was the same in the 1990’s, the lobster sizes were smaller overall. Over the last four seasons, the mean weight has been lowest in November before increasing as the season progressed. In 2004, it was lowest in November at 0.78 kg and highest in May at 1.02 kg (Figure 2-22).

50 2.3.3 Patterns across MFA’s

Lobster mean weight decreases with increasing latitude from the mouth of the Murray River (MFA 51) to the Victoria/South Australia border (MFA 58) (Figure 2-23). Up to 1998, the inter-annual trends in mean weight of lobsters since 1970 have been variable in MFA 51 and 55, but have gradually declined in MFAs 56 and 58. From 1998 to 2002 mean weight generally increased across all four MFAs before decreasing over the next one/two seasons (except MFA 58). In 2004, mean weight ranged from 1.13 kg in MFA 51 to 0.71 kg in MFA 58.

1.00

0.95

0.90

0.85 0.84 kg 0.80 Mean WtMean (kg/lobster)

0.75

0.70 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 2-20 Inter-annual trends in the mean weight of lobsters in the SZRLF for the fishing seasons between 1970 and 2004.

51 1.00

0.95

0.90

0.85

0.80 1970 - 80 1981 - 90

Monthly mean weight (kg) mean weight Monthly 0.75 1991 - 00

0.70 Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Figure 2-21 Within-season trends in the mean weight (± SE of mean) of lobsters in the SZRLF for the fishing seasons between 1970 to 2000.

1.10 2001 2002 1.05 2003 2004 1.00

0.95

0.90

0.85

Monthly mean weight (kg)Monthly 0.80

0.75

0.70 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-22 Within-season trends in the mean weight (± SE of mean) of lobsters in the SZRLF for the fishing seasons 2001 – 2004.

52 1.4 MFA 51 MFA 56 1.3

1.2

1.1

1.0

0.9

0.8

0.7

0.6 1.4

1.3 MFA 55 MFA 58 1.2 Mean Weight (kg/lobster)

1.1

1.0

0.9

0.8

0.7

0.6 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-23 Inter-annual trends in the mean weights (± SE) of lobster for the main MFAs of the SZRLF for the fishing seasons between 1970 and 2004.

2.4 Length Frequency

Since 1991, when the voluntary catch sampling program began, between 5,000 and 30,000 lobsters have been measured each season. Male lobsters, which grow faster and reach larger sizes than females, range between 70 and 200 mm CL length (Figure 2-24), whereas few females are longer than 150 mm CL (Figure 2-25).

The proportion of male lobsters greater than the Minimum Legal Size (MLS) of 98.5 mm CL has increased in recent seasons (Figure 2-26 and Figure 2-27). In the 1998 season, 66% of measured lobsters were >98.5 mm CL whereas in 2004 the figure was 81%. For females, the proportion of lobsters >98.5 mm CL increased from 56% in 1998 to 70% in 2004 (Figure 2-27).

53 4 1991 (n = 2 473) 1996 (n = 8 370)

3

2

1

0

4 1992 (n = 13 376) 1997 (n = 6 935) 3

2

1

0 4 1993 (n = 10 933) 1998 (n = 6 331)

3

2

1 Frequency (%)

0

4 1994 (n = 4 031) 1999 (n =10 566) 3

2

1

0

4 1995 (n = 3 749) 2000 (n = 9 667)

3

2

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace length (mm)

Figure 2-24 Length frequency distributions of male lobsters in the SZRLF for the fishing seasons between 1991 and 2000.

54 5 1991 (n = 2 735) 1996 (n = 10 755) 4

3

2

1

0

5 1992 (n = 13 999) 1997 (n = 8 311) 4

3

2

1

0 5 1993 (n = 12 179) 1998 (n = 6 985) 4

3

2 Frequency (%) Frequency 1

0 5 1994 (n = 4 624) 1999 (n =12 066) 4

3

2

1

0

5 1995 (n = 4 667) 2000 (n = 11 715) 4

3

2

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace length (mm) Figure 2-25 Length frequency distributions of female lobsters in the SZRLF for the fishing seasons between 1991 and 2000.

55 2001 and 2002 seasons. Figure 2-26 Length frequency dist Frequency (%) 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 08 010101010101010101010200 190 180 170 160 150 140 130 120 110 100 90 80 70 2002 (n = 6555) 2002 (n 2002 (n = 5944) 2002 (n 2001 (n = 10 876) 2001 (n 2001 (n = 9437) 2001 (n Carapace length (mm) Carapace length ributions ofmale and female lobsters inthe SZRLF for the Females Males (98.5mm CL) Size Legal Minimum 56 5

4 2003 Males (N = 9,752) MLS 98.5 mm CL

3

2 Frequency (%) Frequency

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace Length (mm)

5

4 MLS 98.5 mm CL 2004 Males (N = 14,508)

3

2 Frequency (%) Frequency

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace Length (mm)

5 MLS 98.5 mm CL

4 2003 Females (N = 11,028)

3

2 Frequency (%) Frequency

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace Length (mm)

5 MLS 98.5 mm CL

4 2004 Females (N = 16,767)

3

2 Frequency (%) Frequency

1

0 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Carapace Length (mm) Figure 2-27 Length frequency distributions of male and female lobsters in the SZRLF for the 2003 and 2004 seasons.

57 2.5 Pre-Recruit Index

2.5.1 Inter-annual Patterns

Data required to calculate a pre-recruit index (PRI - mean number of undersize rock lobster per pot lift) have been recorded since 1983, but estimates prior to 1987 should be treated with caution due to incomplete participation in the system. During the early to mid 1990s the PRI as calculated from logbook data varied from 1.77 in 1991 to 1.20 in 1996 (Figure 2-28). From 1998 to 2002, the PRI remained close to, or above, the highest previous record (1991), peaking at 2.21 undersized lobsters/pot lift in 1999. In 2004, the PRI decreased to 1.31 undersized lobsters/potlift but remained within the reference range as identified in the Management Plan (1.2–1.52 undersized lobsters/potlift).

Escape gaps are not currently mandatory in the SZRLF but some fishers have voluntarily incorporated them into their pots thus affecting estimates of PRI calculated from logbook data. PRI as calculated from voluntary catch sampling data (Figure 2-29) indicates that PRI increased from 1997 to 2001 before decreasing over the next two seasons. PRI increased in 2004 to 1.04 undersized individuals/potlift.

2.5

2.0

1.5 . undersize/potlift)

1.0

0.5

0.0 Pre-recruit index (No 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 2-28 Inter-annual trends in pre-recruit index (± SE of mean) in the SZRLF for the fishing seasons between 1983 and 2004 as calculated from commercial logbook data.

58 1.4

1.3

1.2

1.1 . undersized/potlift) 1.0

0.9

0.8

0.7

Pre-recruit index (No 0.6 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Season Figure 2-29 Inter-annual trends in pre-recruit index in the SZRLF for the fishing seasons between 1994 and 2004 as calculated from voluntary catch sampling data.

2.5.2 Within-season Patterns

Within fishing seasons, the PRI is consistently highest in the first four months of the season before generally decreasing from January onwards (Figure 2-30). In 2004, the PRI ranged between 0.85 and 1.52 undersized/potlift from October to March. Thereafter, it declined to 0.45 undersized/potlift in May, which is consistent with previous annual trends.

59 2.5

2.0 1983-89 1990-00 2001 1.5 2002 2003 2004 1.0

0.5

0.0

Pre-recruit index (no. undersize per potlift) Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-30 Within-season trends in the pre-recruit index (± SE of mean) in the SZRLF for the fishing seasons between 1983 and 2004.

2.5.3 Patterns across MFAs

The PRI increases with latitude between the Coorong (MFA 51) and the Victoria/South Australia border (MFA 58) (Figure 2-31). Inter-annual trends demonstrate that there are higher numbers of undersize lobsters caught in MFAs 56 and 58 compared to MFAs 51 and 55. The PRI for MFAs 51 and 55 varied from 0.1 – 0.8 undersize/pot lift between the 1998 and 2004 seasons compared to MFAs 56 and 58, which had PRI’s between 1.5 – 3.9 undersize/pot lift over the same time period. The PRI has remained relatively unchanged over the last 10 years in MFAs 51 and 55. In MFA 56, PRI reached a peak of 2.82 undersized/potlift in 1999 before decreasing to 1.57 undersized/potlift in 2003. In 2004, PRI increased marginally in MFA 56 to 1.61 undersized/potlift. In MFA 58, PRI reached a peak of 3.93 undersized/potlift in 2002 before decreasing over the next two seasons to 3.03 undersized/potlift in 2004.

60 4 MFA 51 MFA 56

3

2

1

0

4 MFA 55 MFA 58

3

2

Pre-recruit index (No. undersize per potlift) per undersize (No. index Pre-recruit 1

0 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season

Figure 2-31 Inter-annual trends in mean pre-recruit index (± SE) in the main MFAs of the SZRLF for the fishing seasons between 1983 and 2004.

2.6 Spawning lobsters

Data on lobsters in spawning condition have been recorded since 1983. As with the PRI estimates, those prior to 1987 should be treated with caution due to incomplete participation in the reporting program.

2.6.1 Inter-annual Patterns

The number of spawning lobsters per pot lift varied from 0.08 in 1991 to 0.52 in 1999 (Figure 2-32). Since 1991, the number of spawning lobsters/pot lift increased steadily, with the highest catch rates of spawning lobsters occurring between 1998 and 2003. In 2004, the CPUE of spawning lobsters dropped to 0.30 spawners/potlift the lowest on record since 1993.

61 2.6.2 Within-season Patterns

There is a strong seasonal pattern in the number of spawning lobsters/pot lift (Figure 2-33). Hatching commences in early spring and is completed by December/January. The almost complete absence of spawning lobsters in the commercial catch after December supports this fact. In 2004, within season catch rates of spawning lobsters were comparable to previous seasonal trends.

2.6.3 Patterns across MFAs

As with the PRI, the catch rate of spawning lobsters increases along the coast from the Coorong (MFA 51) to the Victoria/South Australia border (MFA 58) (Figure 2-34). Inter-annual trends demonstrate that there are significantly higher numbers of spawning lobsters caught in MFA 56 and 58 compared to MFAs 51 and 55. The catch rate of spawning lobsters in MFAs 51 and 55 has varied from 0.07 – 0.32 spawning lobsters/pot lift between the seasons 1998 – 2003 compared to MFAs 56 and 58 which had rates between 0.45 and 0.76 spawners/pot lift over the same time period. Generally, the catch rate has increased over time in all MFAs but the rate of increase has been greater in MFAs 56 and 58. In 2004, the catch rate of spawning lobsters decreased in all four MFAs. The most notable decreases were in MFAs 56 and 58 where catch rate dropped to 0.35 and 0.40 spawners/potlift respectively. These figures are the lowest on record since 1993 for MFAs 56 and 58.

62 0.6

0.5

0.4

0.3

0.2 No. spawners/potlift No. 0.1

0.0 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 2-32 Inter-annual trends in the number of spawning lobsters (± SE of mean) in the SZRLF for the fishing seasons between 1983 and 2004.

1.6

1.4

1.2 1983-1989 1.0 1991-00 2001 0.8 2002 2003 0.6 2004 0.4

No. spawners per potlift per spawners No. 0.2

0.0 Oct Nov Dec Jan Feb Mar Apr Month

Figure 2-33 Within-season trends in the number of spawning lobsters (± SE of mean) in the SZRLF for the fishing seasons between 1983 and 2004.

63 0.8 MFA 51 MFA 56 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

0.8 MFA 55 MFA 58 0.7

No. Spawners per potlift No. Spawners 0.6 0.5 0.4 0.3 0.2 0.1 0.0 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

Season Figure 2-34 Inter-annual trends in the number of spawning lobsters/pot lift (± SE of mean) for the main MFAs in the SZRLF for the fishing seasons between 1983 and 2004.

2.7 Lobster Mortalities

2.7.1 Inter-annual Patterns

There has been a gradual increase in the number of dead lobsters/potlift since 1996, although the overall numbers are relatively low ( Figure 2-35). Figures ranged from 0.14 to 0.22 dead lobsters/potlift from 1996 to 2000 before decreasing to 0.20 dead lobsters/potlift in 2001. Since then catch rates have increased to 0.27 dead lobsters/potlift in 2004.

64 0.30 0.28 0.26 0.24 0.22 0.20 0.18

Dead lobsters/potlift 0.16 0.14 0.12 1996 1997 1998 1999 2000 2001 2002 2003 2004 Season

Figure 2-35 Inter-annual trends in CPUE of dead lobsters in the SZRLF from 1996 to 2004.

2.7.2 Within season Patterns

Over the last four seasons, lobster mortality was generally highest at the start of each season in October/November/December and decreased as the season progressed ( Figure 2-36). In 2004, lobster mortality peaked at 0.34 dead lobsters/potlift in November before declining to a season low of 0.07 dead lobsters/potlift in May.

0.40 2001 2002 0.35 2003 2004 0.30

0.25

0.20

0.15 Dead lobsters/potlift Dead 0.10

0.05 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-36 Within season trends in lobster mortality from 2001 to 2004 in the SZRLF.

65 2.8 Octopus Catch Rates

2.8.1 Inter-annual Patterns

Annual catch rates of octopus in the SZRLF have been variable over the last nine seasons ( Figure 2-37). From 1996 to 2004, CPUE of octopus has ranged from 0.025 in 2002 to 0.049 octopus/potlift in 2000. In 2004, CPUE of octopus was 0.033 octopus/potlift.

0.055

0.050

0.045

0.040

0.035

Octopus/potlift 0.030

0.025

0.020 1996 1997 1998 1999 2000 2001 2002 2003 2004 Season

Figure 2-37 Inter-annual trends in catch rates of octopus in the SZRLF from 1996 to 2004.

2.8.2 Within season Trends

Over the last four seasons, catch rates of octopus were generally highest at the start of each season in November/December and decreased as the season progressed. In 2004, octopus catch rates peaked at 0.043 octopus/potlift in November before declining to a season low of 0.005 octopus/potlift in May which is consistent with previous seasonal trends.

66 0.07

2001 0.06 2002 2003 0.05 2004

0.04

0.03 Octopus/potlift 0.02

0.01

0.00 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-38 Within season trends in octopus catch rates in the SZRLF from 2001 to 2004.

2.9 Changes in Fishing Patterns

2.9.1 Season Length

During the 1970s, individual licence-holders spent, on average, 100 days fishing each season. From 1983, the average number of days fished each season increased to a peak of 176 days fished/licence in 1991 (out of a total number of 210 potential fishing days) (Figure 2-39). In 1993, the first year of the TACC, the number of days fished/licence was 143. This rose to 153 days in 1997, but decreased over the next 5 seasons to an all-time low of 80 days in 2002. In 2003, the average number of days fished/licence increased by 15 to 95 days which also coincided with an increase in TACC. In 2004, the average number of days fished was 94.

67 200 176 days 180 153 days 160

140

120 94 days 100

fished per licence 80 Mean numberMean of days 60

40 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 2-39 Inter-annual trends in the average number of days fished per licence in the SZRLF between the 1994 and 2004 fishing seasons.

2.10 Distribution of Effort

There has been a significant decline in fishing effort over the last few years in the SZRLF. The decline in effort may manifest itself in a general decrease in the number of pot lifts across all areas and/or a contraction in the total area fished. It has been suggested by some fishers that the latter is occurring, i.e. fishers, confident of catching their ITQ, are fishing closer to home ports and thus are not utilising previously fished areas.

Without spatially explicit data on the location of fishing effort it is difficult to determine how the distribution of effort is changing. The only data available are obtained from the pot sampling program, however, the commitment to this program has been variable over time. In 1994, for example, 2,554 lobsters were measured while in 2001, the figure was 10,000 lobsters. Consequently, it is difficult to accurately infer general trends in the spatial distribution of fishing effort.

In terms of depth fished, the main trend is a reduction in effort across all depth ranges, with a slight indication that proportionally less effort may have been expended in depths greater than 60 m in recent years (Figure 2-40). In 2004, the majority of effort was expended in depths of less than 60m as is consistent with trends in recent seasons.

68 0 - 30 m 1000 MFA 55 31 - 60 m 900 61 - 90 m 800 > 90 m 700 600 500 400 300 200 100 0 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04

0 - 30 m 1000 31 - 60 m MFA 56 ) 900 61 - 90 m 800 > 90 m otlifts

p 700 600 500 1000s ( 400 300

Effort 200 100 0 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04

1000 0 - 30 m MFA 58 900 31 - 60 m 800 61 - 90 m 700 600 > 90 m 500 400 300 200 100 0 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 Season

Figure 2-40 Total effort (pot lifts) in 4 depth ranges for MFA’s 55, 56 and 58 in the SZRLF for the 1983 – 2004 fishing seasons.

69 2.11 High Grading

2.11.1 Inter-annual and within Season Trends

Based on voluntary catch returns, there has been a general increase in the level of highgrading (the selection of smaller sized or non-damaged individuals due to higher unit value) within the SZRLF over the last three seasons (Figure 2-41). In 2004, a minimum of 116.11 tonnes was highgraded compared to 161.22 tonnes in 2003. Within season trends indicate that the level of highgrading is lowest at the start of the season in October-December and increases as the season progresses. In 2004, highgrading was lowest in November at 0.13 kg/potlift and highest in May at 0.57 kg/potlift.

0.8 ) 0.7 2001 (47.37 t) 2002 (115.78 t) 0.6 2003 (161.22 t) 0.5 2004 (116.11 t) 0.4 0.3 0.2 0.1 CPUE Highgrade (kg/potlift 0 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-41 Interannual and within season trends in the levels of highgrading within the SZRLF from 2001 to 2004.

2.12 Settlement Index

The five puerulus collectors in the SZRLF are located at Blackfellows Caves, Livingstone’s Beach, Beachport, Cape Jaffa and Kingston. The annual Puerulus Settlement Index (PSI) for the SZRLF is calculated from the mean monthly settlement recorded on these collectors (Figure 2-42).

70 The PSI generally rose from 1991 to 1995 but declined over the next two seasons. It peaked again in 1998 before declining to the lowest level on record in 2001. In 2002, the PSI increased to 2.4 puerulus/collector, before decreasing to 0.78 puerulus/collector in 2003. Currently, the PSI for 2005 is 3.26 puerulus/collector, the highest on record since sampling began. Data sampling required to estimate this index was not complete at time of publication and this estimate may therefore change. Final PSI estimates for 2005 will be presented in the 2005/06 Status Report for the fishery.

Lagged PSI plotted against seasonal pre-recruit index (PRI) and estimates of recruitment from the qR model are provided in section 3.3 of this report.

4.5

4

3.5

3

2.5

2

1.5

1

0.5 Settlement Index (Puerulus/collector) Index Settlement

0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Season

Figure 2-42 Puerulus Settlement Index (PSI) (mean ± SE) in the SZRLF from 1991 to 2005 (note: data for 2005 are incomplete).

71 2.13 May Fishing Trial

In an effort to give SZRLF licence holders greater flexibility on when allocated quota can be caught, PIRSA Fisheries conducted a May Fishing Trial during the 2004 season. The 2004 trial adds to data obtained in 2003 when the May fishing trial was introduced. The following catch statistics were estimated from data derived from May fishers only in order to examine any possible biological impacts of the trial.

2.13.1 CPUE

Based on May fishers only, CPUE in 2004 was lowest in October at 1.63 kg/potlift and highest in January at 2.18 kg/potlift. The CPUE in May was 1.86 kg/potlift, the third highest for the season. (Figure 2-43). In 2004, CPUE was highest in deeper waters reaching 4.37 kg/potlift in depths >90 m in January. CPUE ranged between 0.99 and 2.18 kg/potlift in the lower depth ranges (Figure 2-44) of <60 m. The observation that CPUE was highest in deeper depths is confirmed by catch/depth profiles (Figure 2-45). From October to February over 85% of the catch on a monthly basis was taken in depths of <60 m. During March and April the proportion of catch taken from depths of >60 m was 37% and 32%, respectively. In May, 54% of the catch was taken in depths >60 m.

2.13.2 Mean Weight

Based on May fishers only, mean lobster weight in 2004 generally increased as the season progressed (Figure 2-46). Mean lobster weight in 2004 was lowest in November at 0.75 kg and highest in May at 1.03 kg.

2.13.3 Lobster Mortality

In 2004, the CPUE of both dead lobsters and octopus generally decreased as the season progressed ( Figure 2-36 and Figure 2-38). CPUE of dead lobsters was highest in November at 0.34 dead lobsters/potlift and lowest in May at 0.07 dead lobsters/potlift. Similarly, catch rates of octopus in 2004, peaked at 0.043 octopus/potlift in November before declining to a season low of 0.005 octopus/potlift in May.

72 2.5

2

1.5

1 CPUE (kg/potlift) CPUE 0.5

0 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-43 Within season trends in CPUE in the SZRLF in 2004 based on data from May fishers only.

5 0-30 m 4.5 31-60 m 4 61-90 m 3.5 >90 m 3 2.5 2 1.5 CPUE (kg/potlift) CPUE 1 0.5 0 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-44 Within season trends in CPUE by depth in the SZRLF in 2004 based on data from May fishers only.

73

100% 90% 80% 70% >90 m 60% 61-90 m 50% 31-60 m

% Catch 40% 0-30 m 30% 20% 10% 0% Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-45 Within season trends in the percentage of catch taken from various depth zones in the SZRLF in 2004 based on data from May fishers only.

1.2

1

0.8

0.6

0.4 Mean Weight (kg) Weight Mean 0.2

0 Oct Nov Dec Jan Feb Mar Apr May Month

Figure 2-46 Within season trends in mean weight in the SZRLF for 2004 based on data from May fishers only.

74 2.13.4 Discussion

Initial results from the May Fishing Trial indicate that the biological impact of May fishing in 2004 were minimal and track closely to estimates observed in 2003. Within season trends in catch rates of both dead and undersize lobsters were lowest in May, as were catches of octopus. In addition, no spawning females were caught during May, which is consistent with the seasonal trends in the reproductive biology of the species. The final May Fishing Trial is scheduled for the 2005/06 season. This additional data will help to consolidate findings presented in this report.

75 3 THE QR MODEL

3.1 Introduction

The qR model (McGarvey et al 1997; McGarvey and Matthews 2001) has been used to generate estimates of performance indicators (including exploitation rate and egg production) for the SZRLF. Outputs from the qR model have been presented in stock assessment reports for the SZRLF since 1997 (Prescott et al. 1997a; Prescott et al. 1998; Prescott and Xiao 2001).

A recent review of the stock assessment research conducted by SARDI Aquatic Sciences (Breen and McKoy 2002) concluded that the qR model is an appropriate tool for assessing exploitation rate and recruitment. The model has been refined over time, most notably during the peer review process for publication of McGarvey and Matthews (2001). Hence, outputs from the current version of the model differ from those presented in previous stock assessment reports (e.g. Prescott et al 1997a; Prescott and Xiao 2001) and in the Management Plan (Zacharin 1997). The three major changes from previous versions are: (i) the replacement of the least squares method by normal likelihoods for the fits to catches in both number and weight; (ii) the adoption of a Baranov rather than a simple bi-linear Schaefer catch relationship; and (iii) the inclusion of a puerulus-based forecasting method which is used to generate predictions of future biomass based on different assumed quota levels.

This section of the report has three objectives: (i) to use the 2004 version of the qR model to generate annual estimates of biomass, egg production, % virgin egg production and exploitation rate for the SZRLF using data for the period up to the end of the 2004 fishing season; (ii) to compare estimates of recruitment obtained using the qR model with an independent measure of pre-recruit abundance; and (iii) to predict future estimates of biomass over a five-year period under a range of alternate quota levels (1770, 1900, 2000 and 2100 t).

76 3.2 Methods

General qR Model

A detailed description of the qR model is provided in McGarvey and Matthews (2001). In summary, the qR model fits to the catch in weight (Cw, in kg) and numbers (Cn, in numbers of lobsters landed). Effort (E) is taken from logbook data and a Baranov survival model using a Schaefer catch relationship (Cn=qEN) is assumed. The model likelihood is written as a modified normal and fitted numerically. Recruitment in each year is estimated as a free parameter.

Other stock assessment models (delay-difference and biomass dynamic) that fit to catch and effort data use only catch in weight (Cw), and rely on CwPUE as a measure of relative fishable biomass. The qR model adds catches in numbers to the fitted data set. Catch-in-weight divided by catch-in-number gives the mean weight of an average landed lobster, and thus, the addition of the catch-in-number time series gives information about yearly mean size in the legal catch, otherwise available only from length-frequency data. Because catches in weight and number constitute a 100% sample, the quality of information obtained about changes in mean size from catch data is far more precise than that obtained from length frequencies, which typically constitute a 0.1% to 1.0% sample fraction. Thus, the qR model uses CwPUE as a measure of change in abundance and mean weight as a measure of change in size structure.

The qR model is age-based. This permits the estimation of yearly egg production, again as total absolute number of eggs hatched annually. However, only legal-size lobsters are estimated in the qR population description. To improve the quality of yearly egg production estimates in the Southern Zone, where some females spawn prior to reaching legal size, the qR estimates of yearly egg production were modified to include one year class of undersize lobsters. This was done by (i) using the same ogives versus age for fecundity and maturity used in previous qR model time series estimates of SZRL egg production; (ii) taking the yearly estimated recruitment number from the standard qR fit; and (iii) assuming the same constant natural mortality, M, over that year. The numbers of pre-recruits of age one year younger

NySubLeg (1- ) than those recruited NyR () in each yearly cohort, was calculated as

77 NySubLeg (1)()exp-= NyMR ×[]. For the last year of this time series, namely the most recent fishing season, where sub-legal numbers NySubLeg() last would need to be inferred from the recruits NyRlast(+ 1) to the year still to come, sub-legal numbers were set equal to those from the year prior NyNySubLeg( last )=- SubLeg ( last 1) .

The pre-recruit index described in section 2.5 of this report provides a direct measure of yearly recruitment that is independent of qR-inferred estimates. It therefore provides a rough check on the accuracy of the qR model recruitment outputs. The yearly pre-recruit index used in this section of the report is based on average undersize CPUE from the months of November to March in each fishing season.

Future Predictions

The qR model was used to estimate biomass levels over the next 5 years in the SZRLF under 4 alternate quota levels (1770, 1900, 2000 and 2100 t). When forecasting recruitment, 1000 sample time series of recruit numbers were generated for each of the next 5 years (2005-2009) based on a lognormal distribution with mean and standard error taken from the number of puerulus settled per collector in each of the last five settlement years. These forecasted recruitments were then taken as inputs (along with the estimated constant catchability) to simulate future biomass forecasts. This approach uses measurements of the number of lobster puerulus that settled onto collectors in the SZRLF, with the mean number of puerulus settled per collector from of each settlement year used as the index of future recruitment.

The puerulus season in the SZRLF runs over the 12 months from May through April of each year. To produce recruitment forecasts for the last year of the time series (2009), we require a puerulus measure from the current puerulus year (May 2005- April 2006) not yet completed. Using the puerulus counts from the first nine months as data, we separably and linearly extrapolated to estimate the settlements for the last three months, thus correcting bias that might result from omitting the months of February through April which have generally lower average settlement rates.

Analysis of previous growth estimates in the SZRLF (McGarvey et al. 1999a) suggests a 4-year time lag for lobsters to grow from settled puerulus to the last moult bringing them above legal size. This 4-year lag was confirmed this year by the close

78 yearly correlation of all three independent recruitment indices, namely from the qR model, the pre-recruit index, and from puerulus per collector four years earlier. To scale the relative measure of mean number of puerulus settled per collector up to a measure of recruitment (four years later), the qR-estimates of recruitment for the 10 overlapping years when measured puerulus would reach legal size, were used. These are the settlement years 1991-2000 (i.e., the qR-recruitment years 1995-2004).

Recruitment Forecasting Algorithm

Define a re-scaling coefficient, CP->R by

RyPR=⋅CPuerulus−> y−4

i.e. multiply the mean observed number of puerulus per collector in year y-4

( Puerulusy−4 ) by the scaling coefficient (CP->R) to give recruitment in year y (Ry ). The scaling coefficient was obtained by taking the means of both puerulus and qR- estimated R’s over the three overlapping years, i.e.

CPR−> == mean(){ Ry , y 1995 to 2003} /mean( { Puerulusy , y = 1991 to 1999})

Then for the 5 future years (y = 2005 to 2009), the assumed mean level of forecasted recruitment was given by RyPR=⋅CPuerulus−> y−4 . The yearly standard deviation for puerulus-forecasted Ry which determines the level of yearly simulated recruitment variation (in each of the 1000 monte carlo runs), was given as CP−>R times the observed standard error in the estimate of puerulus per collector for that corresponding settlement year (4 years prior).

79 Then, given a distinct puerulus-forecasted mean and standard deviation of Southern Zone recruitment for each of the 5 years to come, the forecasted recruit number for each year and monte carlo run was obtained by sampling from a lognormal distribution. The coefficient of variation (CV) is obtained in the usual way as standard deviation divided by mean in each year. For each of the 5 future years, the lognormal distribution describing the range of forecasted values of recruitment is defined by two parameters, μ and σ, which are derived from the mean and CV of recruitment using

1 σ 2 the formulae σ = ln⎡ CV2 +1⎤2 and μ = ln(mean)- . Then choosing {}⎣⎦ 2 standardised normal variates, zy, one for each year, using built-in Excel routines, the sampled recruitment in each forecast year was given by Ryy= exp(μσ+⋅z ) . This lognormal sampling procedure was repeated 1000 times using an Excel VBA macro to generate 1000 forecasted future 5-year recruitment time series.

3.3 Results

Performance Indicators from the qR model

Estimates of catch in numbers and weight from the qR model fit closely with measurements of Cn and Cw obtained from the SZRLF (Figure 3-1, Figure 3-2).

Outputs from the model indicate that the biomass in the SZRLF has been increasing since 1996, peaking at 6,856 tonnes in 2002 (Figure 3-3). In 2004, the biomass was estimated at 6,530 tonnes. Similarly, total egg production in the SZRLF has been increasing over the same period. In 2004, it was 1,500 billion eggs, which equates to 16.0% of virgin egg production (Figure 3-4, Figure 3-5).

The exploitation rate has been declining in the SZRLF since 1997 (Figure 3-6) and reached an all-time low of 0.25 in 2002. In 2004, the exploitation rate was estimated at 0.28. As the exploitation rate quantifies the proportion of the fishable biomass harvested each season, the overall trend in declining SZRLF exploitation rate reflects the fact that a stable catch is being taken from an increasing fishable biomass.

80 Comparison of estimates of recruitment from the qR model with puerulus settlement and pre-recruit indices.

The recruitment estimates from the qR model suggest that recruitment levels in the seasons 1998 through 2002 were high, averaging over 3 million recruits per annum (Figure 3-7). In 2004, recruitment was estimated at 2.8 million. The temporal trends in recruitment that are predicted by the qR model fit closely to estimates of pre- recruitment index (from commercial logbook data; R2 = 0.90) and puerulus settlement index lagged by 4 years (.R2 = 0.57).

81 3000 Reference period Fishery data 2500 qR model

2000

1500

1000

Catch numbers (*1000) 500

0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 3-1 Fit of the qR model to catch in numbers for the SZRLF, based on annual catch totals from the fishery and estimates provided by 2004 version of the qR model.

2500 Reference period Fishery data 2000 qR model

1500

1000

Catch weight (tonnes) Catch weight 500

0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season

Figure 3-2 Fit of the qR model to catch by weight for the SZRLF, based on annual catch totals from the fishery and estimates provided by 2004 version of the qR model.

82 8000 Reference period 7000

6000 6,530 t

5000

4000

3000

Biomass (tonnes) 2000

1000

0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season Figure 3-3 Estimates of biomass provided by the 2004 qR model.

1800 Reference period 1600

1400 1,500 1200

1000

800

600

400

Egg production (billion eggs) (billion production Egg 200

0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season Figure 3-4 Estimates of egg production by the 2004 qR model.

83 0.20 Reference period

0.15 0.16

0.10

0.05 % of virgin egg production % of egg virgin

0.00 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season Figure 3-5 Estimates of percent of virgin egg production by the 2004 qR model

0.5 Reference period

0.4

0.3 0.28

0.2 Exploitation rate (U) rate Exploitation 0.1

0.0 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Season Figure 3-6 Estimates of exploitation rates obtained from the 2004 qR model

84 qR recruitment 2.5 Pre-recruit index (logbook) 4.5 Puerulus settlement index (4 year lag) 4.0 2.0

3.5 1.5

3.0

1.0 2.5 qR RecruitmentqR (millions) 0.5 PRI (number undersized/pot lift)(number undersized/pot PRI PSI (puerulus/collector) 2.0

0.0 1.5 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Season

Figure 3-7 Estimates of annual recruitment obtained from the qR model, pre-recruit index (PRI) as undersize numbers per pot lift (Nov-Mar) obtained from logbook data and puerulus settlement index (PSI) lagged by 4 years.

85 Forecasts of biomass

The mean expected start-of-year biomass for the next four years, based on the mean number of puerulus settled per collector, is presented in Figure 3-8. Forecasts suggest that the biomass will increase in 2006 for both quotas examined (1770, and 1900 t) in response to the high puerulus settlement in 2002. Biomass is predicted to decrease in 2007 under both quota levels, due to the low settlement in 2003, but will increase for the following two seasons due to high puerulus counts in 2004 and 2005. The notable increase in biomass in 2009 reflects one of the highest puerulus counts on record in the SZRLF in 2005.

11000 13000 10500 12000 10000 11000 9500 10000 9000 9000 8500 8000 8000 7000 7500 6000 7000 5000 2004 2005 2006 2007 2008 2009 6500 6000

Biomass (tonnes) 5500

5000 qR-Biomass 4500 1770 tonne quota 4000 1900 " " 3500 3000 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 Season

Figure 3-8 Estimates of biomass for the SZRLF and generated forecasts at different quota levels as provided by the puerulus method for the qR model.

86 3.4 Discussion

Details of qR model development, and simulation testing of its performance has been evaluated in three peer-reviewed papers (McGarvey et al. 1997; McGarvey and Matthews 2001; McGarvey et al., 2005). The fits of the qR-model predictions to yearly catch totals in weight and in number of lobsters landed are close to the data- reported values. The likelihood measure of closeness of fit of the model to these two catch data time series is given by the estimated parameter, σ0 , (McGarvey and Matthews 2001), which quantifies the mean standard deviation of yearly residuals as a coefficient of variation, that is, the standard deviation of the sum of squared differences of model and data as a percentage of the mean data levels for catch in weight and number. The estimated value of σ0 was 7%, indicating an agreement of model and data of 7%.

Because of close agreement of model and data, most of the uncertainty in the qR model estimates lies in the assumed values of input parameters, notably (1) natural mortality, (2) mean weights-at-age, and (3) CPUE as a measure of biomass. Steady- state analysis by McGarvey et al. (1997) showed that catch under-reporting has essentially no effect on the qR estimates of exploitation rate, while all yearly values of biomass and recruitment are reduced by exactly the percentage of under-reporting. Similarly, McGarvey and Matthews (2001) and McGarvey et al. (2005) both showed that (1) the qR model estimates are quite insensitive to errors in the assumed natural mortality rate, but that these estimates (2) were, like any size-based assessment, generally sensitive to the assumed growth inputs of weight-at-age. (3) The impact of differing levels of rising effective effort, and thus of the principal assumed cause of deviation in trends of CPUE and stock biomass was tested in the Northern Zone fishery where rising effective effort is presumed to be significant (Ward et al., 2002). In the Southern Zone, where fishing practices, and the widespead occurrence of fishable habitat in the coastal zone, have not much altered in recent years, the impact of rising effective effort is not considered to be large.

Outputs from the qR model predict that the biomass, egg production and % virgin egg production during the 2004 season in the SZRLF remained high. These results support observations about the status of the fishery made in the second section of this report

87 on the basis of CPUE. There appears to be a strong correlation between estimates of recruitment from the qR model with both the independent pre-recruit index (R2 = 0.90) and lagged puerulus settlement index (R2 = 0.57) which provides evidence that the qR model is a useful stock assessment tool for the SZRLF.

Forecasts suggest that the biomass will increase in 2006 for both quotas examined (1770 and 1900 t) in response to the high puerulus settlement in 2002. Biomass is predicted to decrease in 2007 due to the low settlement in 2003, but will increase for the next two seasons due to high puerulus counts in 2004 and 2005. The large predicted increase in biomass in 2009 reflects one of the highest puerulus settlement counts on record in 2005.

88 4 PERFORMANCE INDICATORS

Current biological performance indicators for the SZRLF are catch rate, mean weight, pre-recruit abundance, exploitation rate and egg production. Upper and lower limits for catch rate, mean weight and pre-recruit abundance are identified in the Management Plan and are the highest and lowest values to occur in the reference seasons 1992 through 1996 (Zacharin 1997) (Table 1-5). For those performance indicators that are estimated by the qR model, i.e. exploitation rate and egg production, the upper and lower estimates derived for the reference years are from the most recent version of the qR model.

4.1 Catch Rate

The catch rate (CPUE) for the reference years ranged between 0.935 kg/pot lift in 1996 and 1.138 kg/pot lift in 1994 (Table 4-1). CPUE for the 2004 season was 1.81 kg/pot lift, which is 59% above the upper reference limit and one of the highest CPUEs in the history of the fishery. This is the sixth season in succession in which the CPUE has been higher than the upper reference limit. As indicated in the first section of this report, the rise in CPUE reflects the increase in lobster biomass resulting from the control of catch levels associated with introduction and enforcement of the TACC and high levels of recruitment in recent years.

4.2 Mean Weight

The mean weight of lobsters for the reference years ranged between 0.794 kg in 1992 and 0.839 kg in 1993 (Table 4-1). The mean weight of lobsters for the 2004 season (calculated from season totals rather than daily catch data as it was in section two) was 0.846 kg, which is 0.83% higher than the upper reference limit. As mentioned in section 2.3.1, fishing behaviour (particularily highgrading) undoubtedly affects current annual estimates of lobster size in the zone.

4.3 Abundance of Pre-recruits

The index of pre-recruit abundance for the reference years ranged between 1.20 undersize/pot lift in 1996 to 1.53 undersize/pot lift in 1994 (Table 4-1). The pre- recruit index for the 2004 season, calculated for months of November to March (inclusive), was 1.31 undersize/pot lift, which is inside the reference range. The

89 independent estimates of recruitment provided by the qR model broadly confirms trends in pre-recruit estimates.

4.4 Exploitation Rate

Reference points for the exploitation rate for the reference years generated by the qR model ranged from 0.37 in 1992 to 0.44 in 1996 (Table 4-1). The exploitation rate for 2004 was 0.28, which is 24% below the lower reference limit. The 2004 exploitation rate estimate remains one of the lowest figures in the history of the fishery and indicates that the current level of the TACC is achieving the goal of rebuilding the biomass.

4.5 Egg Production

Reference points for egg production derived from the qR model range from 895 billion eggs in 1996 to 1,019 billion eggs in 1992 (Table 4-1). Egg production in the 2004 season was 1,500 billion eggs, which is 47% above the upper reference limit and one of the highest in the history of the fishery. The high level of egg production in the 2004 season as estimated by the qR model reflects the continuing increase in abundance of adult lobsters in the SZRLF.

90

Biological Reference Points

2004/05 Lower Limit Upper Limit

Exploitation rate 0.28 0.37 0.44

Egg production (billions) 1,500 895 1,019

Pre-recruit abundance 1.31 1.20 1.53 (no. undersized/potlift)

Catch rate (kg/pot lift) 1.81 0.94 1.14

Mean Weight (kg) 0.85 0.79 0.84

Table 4-1 Estimates of biological performance indicators for the SZRLF in 2004/05 in relation to upper and lower limit ranges.

91 5 GENERAL DISCUSSION

5.1 Information Available for the Fishery

Stock assessment of the SZRLF is aided by documentation on the history of the management of the fishery in the Management Plan and both recent stock assessment and status reports (Prescott et al. 1997; Zacharin 1997; Linnane et al. 2005b and c). The management plan also describes the management arrangements in place at the time of this assessment and the biological reference points used for assessing the fishery. Comprehensive catch and effort data have been collected since 1970. Data collected since 1983, however, provide more reliable information on effort. Fishery stock assessments are also aided by puerulus settlement data and stock assessment model outputs. Voluntary catch sampling data have been collected since 1991 and provide critical information on length frequency, pre-recruit indices and reproductive condition of females. Data from 1994 onwards are more robust due to low levels of participation in the early years of the program.

Assessment of the SZRLF currently depends mainly on commercial catch and effort data. Future stock assessment would greatly benefit from the collection of additional fishery-independent information. A fishery-independent monitoring survey (FIMS) is currently being developed in the SZRLF. Outputs from this survey will be presented in future stock assessments for the fishery and will help to reduce the level of uncertainty in the assessment.

5.2 Current Status of Southern Zone Rock Lobster Fishery

The fishery-dependent data provided in this report clearly suggest that the status of the resource upon which the SZRLF is based has improved substantially over the last decade and consequently that the biomass of rock lobster in the Southern Zone has increased, reflecting the success of the long-term rebuilding program. This conclusion is consistent with that reached in the most recent reports (Ward et al., 2005; Linnane et al., 2005b and c) and is based on numerous lines of evidence. For example, catch rates have increased rapidly since 1996 and have been above the upper reference point stated in the Management Plan since 1999. The catch rate for 2004 (1.81 kg/pot lift) is one of the highest in the recent history of the fishery (i.e. since 1970). The pre-recruit

92 index for 2004 (1.31 undersize/pot lift) is also inside the range for the reference period identified in the Management Plan.

Similarly, the performance indicators derived from the qR model suggest that the SZRLF is currently in a strong position. Recruitment levels, as estimated by the qR model, have been above the historic average in each of the previous seven seasons. In addition, the 2004 level of biomass is one of the highest in the history of the fishery (i.e. since 1970) and egg production is currently 47% above the upper limits for the reference period identified in the Management Plan. The exploitation rate for the 2004 season (0.28) is 24% below the lower limit for the reference period identified in the Management Plan. The close yearly correlation of all three independent recruitment indices, namely from the qR model, the pre-recruit index (PRI), and lagged puerulus settlement index (PSI) is encouraging and suggests that the PSI may be a suitable indicator for predicting future recruitment within the zone.

Overall, fishery data and outputs from the qR model indicate that the biomass rebuilding strategy for the SZRLF has succeeded.

Despite optimistic outputs at a zonal level, the observed downtrend trend in a number of localised fishery dependent statistics should be highlighted. Currently, >85% of the catch in the SZRLF is taken from depths of <60 m. Data presented in this report suggests that continued exploitation of inshore stocks at this level may not be sustainable. For example, catch rates in MFAs 56 and 58 have decreased notably over the last two seasons in both the 0-30 m and 31-60 m depth ranges. In addition, the catch rate of both spawning females and undersized lobsters in both these MFAs have also decreased. For example, the pre-recruit index in MFA 56 has been decreasing since 1999 and the 2004 estimate is the lowest on record since 1996. While current estimates of catch rate and pre-recruit index for the entire zone are within the range of the performance indicators in the Management Plan, these results indicate that close monitoring of these indices on a finer spatial scale may be required if localised reductions in lobster abundance is to be avoided. Overall, these results highlight the need for fishery independent data that are currently lacking from both fishery statistics and stock assessment model outputs.

93 5.3 Research in Response to DEH Recommendations

Both current and future research needs in the SZRLF have recently been refocused by the South Australian Rock Lobster Research Committee to ensure the recommendations outlined in the assessment of the fishery, by the Department of Environment and Heritage (DEH), (Anon, 2003) are addressed appropriately. The DEH report outlines 13 recommendations to the fishery that relate to both management arrangements and environmentally sustainable fishing practices. A number of these recommendations are currently being addressed through either ongoing research or through proposed research projects. A full list of the DEH recommendations are provided in an appendix to this assessment report. Recommendation 4 requests:

PIRSA to continue to improve assessment of all components of non-commercial catch in the fishery to be factored into the annual stock assessment process and management of the fishery. This will include further periodic surveys or other data collection and analysis measures to enhance the assessments of recreational and indigenous catch in the fishery. Details of management arrangements associated with in the SZRLF are provided in Section 1.2.4 of this report. Periodic recreational catch and effort surveys are undertaken (e.g. Venema et al. 2003), the most recent of which was conducted during the 2004/05 fishing season (Currie et al. 2006). Outcomes from this survey are presented in section 1.2.4 of this report.

Recommendation 5 requests:

PIRSA within 18 months, to review the monitoring requirements for both zones of the fishery, including options for independent monitoring appropriate to the scale of fishing and status of stocks in the main fishing areas, to identify monitoring measures necessary to confirm the status of stocks and support stock recovery strategies. In order to overcome the inherent limitations of the fishery dependent catch and effort logbook program, a Fishery Independent Monitoring Survey (FIMS) was developed for trial in the southern zone rock lobster fishery (SZRLF) for the 2005/06 season. Sampling is being undertaken at the beginning (September), mid season (January) and end (May) of this fishing season along five predetermined transects across a range of depth profiles. Data will be used as input for fishery independent models with outputs used in the determination of a fishery independent estimate of lobster abundance.

94 Initially, the FIMS will be conducted in the SZRLF only. However, once the sampling protocol and data analyses procedures have been developed and refined, it is proposed that they will be applied to the NZRLF.

Recommendation 7 requests:

Performance measures and targets for the main byproduct species to be included in the revised management plans for both zones, and the catches of the main byproduct species should be reviewed as part of the annual stock assessment process A report detailing the species composition and spatio-temporal trends in by-catch from the South Australian commercial rock lobster fishery as estimated using two monitoring options was finalised in 2004 (Brock et al., 2004). The report identifies the main by-catch species within the fishery and estimates catch rates of by-catch as determined during the 2001/02 and 2002/03 fishing seasons. It also compares the effectiveness of logbook and observer sampling strategies and comments on the appropriateness of each for application within the South Australian rock lobster fishery. A workshop on the outcomes of the report is proposed for 2006 as part of the revision of the management plans for the fishery. The species composition of by-catch from the SZRLF is also monitored annually through the onboard observer programme. Results are presented in a previous section of this report.

Recommendations 10, 11 and 12 request:

PIRSA within 18 months to introduce mandatory structured reporting of all interactions between the rock lobster fishery and endangered, threatened or protected species PIRSA and industry to continue to monitor the extent of interactions between rock lobster fishery and fur seals and sea lions, and develop appropriate mitigation measures including establishment within 2 years of preliminary trigger and reference points, to minimise these interactions PIRSA within 12 months to conduct a qualitative risk assessment of the interactions between the rock lobster fishery and protected species off SA and use the outcomes of this assessment to implement further protected species mitigation measures as required In response to these recommendations, a project titled “Interactions of the South Australian fishery for southern rock lobster (Jasus edwardsii) with pinnipeds” has been submitted to the Fisheries Research and Development Corporation for funding consideration. The main objectives of the project are as follows:

95 1) To measure the interaction of the South Australian rock lobster fishery with pinnipeds 2) To assess the risks to pinniped populations arising from their interactions with the South Australian rock lobster fishery 3) To develop and assess methods for mitigating the interaction of pinnipeds with lobster pots. 4) To determine the importance of rock lobster in the diets of Australian sealions. Initial funding has been provided by FRDC for a desktop review study. This report is due for completion in 2006.

5.4 Future Research Priorities

A number of additional strategic research priorities facing the fishery have been identified through the Fishery Research Sub-Committee. The relationship between rock lobster recruitment characteristics and oceanographic conditions is currently listed as a high research priority for the zone. A project titled “Relationships between sea surface temperature, ocean colour and recruitment to fisheries in South Australia” was submitted to FRDC in 2005 for funding consideration. The objectives of the project are:

1) To calculate quantitative indices from remote sensing of sea surface temperatures (SST) and ocean colour from 20-year and 16 year datasets respectively. 2) To compile the model-based and measured recruitment indices for S.A. fisheries including rock lobster. 3) To statistically compare the time series of remote sensing indices to the time series of recruitment indices for each species with the goal of determining whether a relationship exists. Rock lobster population spatial dynamics has also been identified as having a medium-high priority for the zone. In response to this, a project titled “Improving spatial management of southern rock lobster fisheries to improve yield, value and sustainability” has been submitted to FRDC for funding consideration. The objectives of the project are:

1) To enable assessment reporting of trends in biomass and egg production by depth.

96 2) To evaluate separate deep-water quota for increase in yield and egg production. 3) To evaluate regional size limits to increase in yield and egg production. 4) To conduct field experiments and sampling to provide additional data required for alternative harvest strategy evaluation (fisher catch sampling, translocation release survival, release movement, translocation growth transition, effects of translocation on maturity and egg production parameters, density dependent growth). 5) To conduct field experiments on translocation to provide additional data required for economic evaluation (change in colour, tail width, condition, and ability to survive transport). 6) To evaluate translocation options that increase yield and egg production. 7) To evaluate and compare spatial management options by economic analysis. 8) To determine the extent of ecological community change in deep water reef habitats in response to increased harvest rates of lobsters. 9) To develop functional management and monitoring recommendations to apply outcomes.

Finally, the inherent problems associated with the use of fishery dependent data for estimating lobster abundance are widely acknowledged. Changes in fishing patterns mean that commercial catch rates are no longer a true reflection of lobster biomass. In South Australia, this is highlighted under the quota management system where highgrading (the selection of smaller or non-damaged individuals due to higher unit value) is a feature of the fishery. As a result, a Fishery Independent Monitoring Survey (FIMS) has been developed for trial in the SZRLF for the 2005/06 season. Sampling is being undertaken throughout the season along five predetermined transects that cover a range of depth profiles. Data will be used as input for fishery independent models with outputs used in the determination of a fishery independent estimate of lobster abundance.

The strategic research plan for the SZRLF will be reviewed as part of the process in developing a new Management Plan for the fishery.

97 6 BIBLIOGRAPHY

Anon. (1995) A review of the management of the South Australian southern zone rock lobster fishery. South Australian Research and Development Institute, Adelaide.

Anon. (2003). Assessment of the ecological sustainability of management arrangements for the South Australian Rock Lobster Fishery. Department of the Environment and Heritage publication, Canberra, 1-34.

Bentley, N. and P. J. Starr (2001) An Examination of Stock Definitions for the New Zealand Rock Lobster Fishery 2001/48 Ministry of Fisheries, Wellington, 1- 22.

Booth, J. D. (1994). Jasus edwardsii larval recruitment off the east coast of New Zealand. Crustaceana 66: 295-317.

Booth, J. D., A. D. Carruthers, C. D. Bolt and R. A. Stewart (1991). Measuring the depth of settlement in the red rock lobster, Jasus edwardsii. New Zealand Journal of Marine and Freshwater Research 25: 123-132.

Booth, J. D., J. S. Forman and D. R. Stotter (2002). Settlement indices for 2000 for the red rock lobster, Jasus edwardsii 2002/12 National Institute of Water Research and Atmosphere, Wellington, 1-34.

Booth, J. D., J. S. Forman, D. R. Stotter, E. Bradford, J. Renwick and S. M. Chiswell (1999). Recruitment of the red rock lobster with management implications 99/10 NIWA, Wellington, 1-103.

Booth, J. D. and R. A. Stewart (1992). Distribution of phyllosoma larvae of the red rock lobster Jasus edwardsii off the east coast of New Zealand in relation to the ocdeanography. Australian Society for Fish Biology workshop on larval biology, Australian Government Publishing Service.

Booth, J. D. and R. A. Stewart (1993). Puerulus settlement in the red rock lobster, Jasus edwardsii. 93/5 MFA, Wellington,

Booth, J. D., R. J. Street and P. J. Smith (1990). Systematic status of the rock lobsters Jasus edwardsii from New Zealand and J. novehollandae from Australia. New Zealand Journal of Marine and Freshwater Research 24: 239-249.

Brasher, D. J., J. R. Ovenden and R. W. G. White (1992). Mitochondrial DNA variation and phylogenetic relationships of Jasus spp. (Decapoda: Palinuridae). Journal of Zoology 227: 1-16.

Breen, P. A. and J. D. Booth (1989). Puerulus and juvenile abundance in the rock lobster Jasus edwardsii at Stewart Island, New Zealand. New Zealand Journal of Marine and Freshwater Research 23: 519-523.

98 Breen, P. A. and J. L. McKoy (2002) Review of current and past stock assessments for the South Australian Northern Zone Rock Lobster: Report by NIWA to the NZRL FMC Fishery NIWA, Wellington,

Brock, D. J. and T. M. Ward (2004). Maori octopus (Octopus maorum) bycatch and southern rock lobster (Jasus edwardsii) mortality in the South Australian rock lobster fishery. Fishery Bulletin 102, 430-440.

Brock, D. J., Saunders, T. M., Ward, T. M. and A. J. Linnane (2006a). Effectiveness of a two-chambered trap in reducing within-trap predation by octopus on southern spiny rock lobster. Fisheries Research 77,.348-355. Brock, D. J., Saunders, T. M., Ward, T. M. and A. J. Linnane (2006b). A two- chambered trap reduces within-trap predation by octopus on rock lobsters in aquarium trials. Fisheries Research in press. Brown, R. S. and B. F. Phillips (1994). The Current Status of Australia's Rock Lobster Fisheries. Spiny Lobster Management. B. F. Phillips, J. S. Cobb and J. Kittaka. Melbourne, Blackwell Scientific Publications Ltd.: 33-63.

Bruce, B., R. Bradford, D. Griffin, C. Gardner and J. Young (1999). A synthesis of existing data on larval rock lobster distribution in southern Australia. Final report to the Fisheries Research and Development Corporation 96/107 FRDC, Canberra, 1-57.

Caddy, J. and R. Mahon (1995). Reference points for fisheries management FAO Fisheries Technical Paper 347, 1-83.

Copes, P. (1978). Resource management for the Rock Lobster Fisheries of South Australia: A report commissioned by the Steering Committee for the Review of Fisheries of the South Australian Government.

Currie, D.R., Sorokin S.J. and Ward T.M. (2006). Survey of Recreational Rock Lobster Fishing in South Australia during 2004/05. Report to PIRSA Fisheries. SARDI Aquatic Sciences Publication No. RD04/0228-2.

Gardner, C.; Frusher, S. D.; Buxton, C.; Haddon, M. 2003: Movements of southern rock lobster, Jasus edwardsii, in Tasmania, Australia. Bulletin of Marine Science 73: 653-671.

Henry, G. W. and J. M. Lyle (2003). The National Recreational and Indigenous Fishing Survey 99/158 NSW Fisheries, Cronulla, NSW, 200.

Kanciruk, P. (1980). Ecology of juvenile and adult Palinuridae. The Biology and Management of Lobsters. J. S. Cobb and B. F. Phillips. New York, Academic Press. 2: 59-96.

Kermack, W.O., McKendrick, A.G., 1932. Contributions to the mathematical theory of epidemics. III. Further studies of the problem of endemicity. Proc. R. Soc., Series A 138, 94-122.

99

Lewis, R. K. (1981). Southern Rock Lobster Jasus novaehollandae: Zone N Review South Australian Department of Fisheries.

Linnane, A, T. M. Ward, R. McGarvey and J. Feenstra (2004). Southern Zone Rock Lobster (Jasus edwardsii) Fishery Status Report 2003/04. Status Report to PIRSA Fisheries. SARDI Aquatic Sciences Publication No. RD04/0164.

Linnane, A., W. F. Dimmlich, and T. M. Ward (2005a). Movement patterns of the southern rock lobster, Jasus edwardsii, off South Australia. New Zealand Journal of Marine and Freshwater Research, 39: 335-346.

Linnane, A, T. M. Ward, R. McGarvey and J. Feenstra (2005b). Southern Zone Rock Lobster (Jasus edwardsii) Fishery Status Report 2004/05. Status Report to PIRSA Fisheries. SARDI Aquatic Sciences Publication No. RD04/0164-2. SARDI Report Series No. 108.

Linnane, A, T. M. Ward, R. McGarvey, Y. Xiao and J. Feenstra (2005c). Southern Zone Rock Lobster (Jasus edwardsii) Fishery 2003/04. Final Stock Assessment Report to PIRSA Fisheries. SARDI Aquatic Sciences Publication No. RD03/0153-02.

Lotka, A.J. (1922). The stability of the normal age distribution. Proceedings of the National Academy of Science. 8, 339-345.

MacDiarmid, A. B. (1988). Experimental confirmation of external fertilisation in the southern temperate rock lobster Jasus edwardsii (Hutton) (Decapoda: Palinuridae). Journal of Experimental Marine Biology and Ecology 120(3): 277-285.

MacDiarmid, A. B. (1989). Moulting and reproduction of the spiny lobster Jasus edwardsii (Decapoda:Palinurudae) in northern New Zealand. Marine Biology 103: 303-310.

McClatchie, S and T. M. Ward in press. Water mass and alongshore variation in upwelling intensity in the eastern Great Australian Bight. Journal of Geophysical Research.

McGarvey, R., G. J. Ferguson and J. H. Prescott (1999a). Spatial variation in mean growth rates of rock lobster, Jasus edwardsii, in South Australian waters. Marine and Freshwater Research 50: 333-342.

McGarvey, R., M. Pennington, J. Matthews, D. Fournier, J. Feenstra, M. Lorkin and G. Ferguson (1999b). Survey sampling design and length-frequency data analysis for ongoing monitoring and model parameter evaluation in the South Australian rock lobster fishery: Final report to Fisheries Research and Development Corporation 95/138 South Australian Research and Development Institute, Adelaide, 1-119.

100 McGarvey, R. and J. M. Matthews (2001). Incorporating numbers harvested in dynamic estimation of yearly recruitment: onshore wind in interannual variation of South Australian rock lobster (Jasus edwardsii). Journal of the International Council for the Exploration of the Sea 58(5): 1092-1099.

McGarvey, R. and M. Pennington (2001). Designing and evaluating length-frequency surveys for trap fisheries with application to the southern rock lobster. Canadian Journal of Fisheries and Aquatic Sciences 58(2): 254-261.

McGarvey, R., J. M. Matthews and J. H. Prescott (1997). Estimating lobster recruitment and exploitation rate from landings by weight and numbers, and age-specific weights. Marine and Freshwater Research 48: 1001-1008.

McGarvey, R., A.E. Punt and J.M. Matthews (2005). Assessing the information content of catch-in-numbers: a simulation comparison of stock assessment methods based on catch and effort totals. pp. 635-653, In: G.H. Kruse, V.F. Gallucci, D.E. Hay, R.I. Perry, R.M. Peterman, T.C. Shirley, P.D. Spencer, B. Wilson, and D. Woodby [eds.], Fisheries Assessment and Management in Data-Limited Situations. Alaska Sea Grant College Program, University of Alaska, Fairbanks.

McKendrick, A.G. (1926). Applications of mathematics to medical problems. Proceedings of the Edinburgh Mathematical Society. 44, 98-130.

Musgrove, R. J. B. (2000). Moult staging in the southern rock lobster Jasus edwardsii. Journal of Crustacean Biology 20(1): 44-53.

Ovenden, J. R., D. J. Brasher and R. W. G. White (1992). Mitochondrial DNA analyses of red rock lobster, Jasus edwardsii, supports an apparent absence of population subdivision throughout Australasia. Marine Biology. 112(2): 319- 326.

Prescott, J., R. McGarvey, G. Ferguson and M. Lorkin (1996). Population dynamics of the southern rock lobster in South Australian waters. Final report to the Fisheries Research and Development Corporation 93/086 and 93/087, 1-64.

Prescott, J., G. Ferguson, D. Maynard, S. Slegers, M. Lorkin and R. McGarvey (1997a). South Australian southern and northern zone rock lobster. South Australian Fisheries Assessment Series 97/1 South Australian Research and Development Institute, Adelaide, 1-58.

Prescott, J., R. McGarvey, G. Ferguson and M. Lorkin (1997b) Population Dynamics of the Southern Rock Lobster in South Australian Waters 93/386 and 93/087 South Australian Research and Development Institute, Adelaide, 1-65.

Prescott, J., R. McGarvey, A. Jones, A. Peso, G. Ferguson, D. Casement, Y. Xiao and P. McShane (1998) Southern Zone Rock Lobster 97/14 South Australian Research and Development Institute, Adelaide, 1-22.

101 Prescott, J., R. McGarvey, Y. Xiao and D. Casement (1999). Fisheries Assessment Report to PIRSA for the Northern Zone and Southern Zone Rock Lobster Fishery Management Committees 99/04 South Australian Research and Development Institute, Adelaide, 1-28.

Prescott, J. and Y. Xiao (2001) Rock Lobster 2001/04, Report to PIRSA Fisheries 1- 68.

Rochford, D. J. (1977) A review of a possible upwelling situation off Port MacDonnell S.A. 81 CSIRO Aust. Div. Fish. Oceanography.

Schahinger, R. B. (1987). Structure of coastal upwelling events observed off the south-east coast of South Australia during February 1983 - April 1984. Australian Journal of Marine and Freshwater Research 38: 439-459.

Smith, P. J., J. L. McKoy and P. J. Machin (1980). Genetic variation in the rock lobsters Jasus edwardsii and Jasus novaehollandiae. New Zealand Journal of Marine and Freshwater Research 14: 55-63.

Spiegelhalter, D.J., Thomas, A., and Best, N.G., (2000). WinBUGS Version 1.3 User Manual. Medical Research Council Biostatistics Unit, Institute of Public Health, Cambridge, UK.

Venema, S., V. Boxall and T. M. Ward (2003). Survey of Recreational Rock Lobster Fishing in South Australia during 2001/02 South Australian Research and Development Institute, Adelaide, 1-42.

Ward, T. M., R. McGarvey and D. Brock (2002) Southern Zone Rock Lobster (Jasus edwardsii) Fishery 2002/03a South Australian Research and Development Institute, Adelaide, 1-70.

Ward, T M. R. McGarvey, Y. Xiao, G. Ferguson and A. Linnane (2004). Southern Zone Rock Lobster (Jasus edwardsii) Fishery 2002/03. Final Stock Assessment Report to PIRSA Fisheries. SARDI Aquatic Sciences Publication No. RD03/0153.

Zacharin, W., Ed. (1997). Management Plan for the South Australian Southern Zone Rock Lobster Fishery. Primary Industries and Resources South Australia,1-29.

102 7 APPENDIX

The following is a list of recommendations by the Department of Environment and Heritage (DEH) aimed at strengthening the effectiveness of the management arrangements for the SARLF and containing the environmental risks in the medium to long term (Anon. 2003).

Recommendation 1: PIRSA to inform the Department of the Environment and Heritage of any significant changes to the management regime of the South Australian Rock Lobster Fishery.

Recommendation 2: The current review of SA's Fisheries Act 1982 should provide for the inclusion of general community members on the two fisheries management committees. Greater efforts should also be made to increase conservation and general community involvement in stock assessments and research priority setting processes.

Recommendation 3: PIRSA to pursue complementary management arrangements with other Australian jurisdictions responsible for managing southern rock lobster fisheries to ensure that all removals and other relevant impacts on the stock are properly accounted for in stock assessments.

Recommendation 4: PIRSA to continue to improve assessment of all components of non commercial catch in the fishery to be factored into the annual stock assessment process and management of the fishery. This will include further periodic surveys or other data collection and analysis measures to enhance the assessments of recreational and indigenous catch in the fishery. .

Recommendation 5: PIRSA, within 18 months, to review the monitoring requirements for both zones of the fishery, including options for independent monitoring appropriate to the scale of fishing and status of stocks in the main fishing areas, to identify monitoring measures necessary to confirm the status of stocks and support stock recovery strategies. PIRSA to progressively implement priority actions identified in the review.

Recommendation 6: PIRSA and the SA industry to work with their Victorian counterparts to investigate and adopt appropriate measures to address quota avoidance, misreporting of catches and other illegal activities in waters near the SA-Victoria border. These measures should be built into SA's compliance strategies.

Recommendation 7: Performance measures and targets for the main byproduct species to be included in the revised management plans for both zones, and the catches of the main byproduct species should be reviewed as part of the annual stock assessment process.

Recommendation 8: PIRSA to develop within 18 months a conservative harvest strategy for the Northern Zone fishery, including a TAC to commence on 1 November 2003, that includes recovery targets and reference points, and monitoring arrangements, representative of the scale of fishing in the Zone, and stock recovery timeframes.

Recommendation 9: Priority should be given to early implementation of escape gaps in the Northern Zone, and should be mandatory in both zones by October 2004. Decisions on the dimensions of escape gaps in both zones to be based on the requirement to minimise fishery impacts on all bycatch species.

103 Recommendation 10: PIRSA within 18 months to introduce mandatory structured reporting of all interactions between the rock lobster fishery and endangered, threatened or protected species.

Recommendation 11: PIRSA and industry to continue to monitor the extent of interactions between rock lobster fishery and fur seals and sea lions, and develop appropriate mitigation measures, including establishment within 2 years of preliminary trigger and reference points, to minimise these interactions.

Recommendation 12: PIRSA within 12 months to conduct a qualitative risk assessment of the interactions between the rock lobster fishery and protected species off SA and use the outcomes of this assessment to implement further protected species mitigation measures as required.

Recommendation 13: PIRSA to develop measures to assess ecosystem impacts of the fishery. Consideration should be given to the appropriateness of reference areas that would allow comparison between fished and unfished areas.

104