Potential for Parasite Interactions Between Wild and Farmed Kingfish, Discrimination of Farmed and Wild Fish and Assessment of Migratory Behaviour

Potential for parasite interactions between wild and farmed kingfish, discrimination of farmed and wild fish and assessment of migratory behaviour

K.S. Hutson, I.D. Whittington, B.M. Gillanders, J.E. Rowntree, I. Ernst, C.B. Chambers and C. Johnston

FRDC FINAL REPORT Project No. 2003/220

Potential for parasite interactions between wild and farmed kingfish, discrimination of farmed and wild fish and assessment of migratory behaviour

K.S. Hutson, I.D. Whittington, B.M. Gillanders, J.E. Rowntree, I. Ernst, C.B. Chambers and C. Johnston

FRDC FINAL REPORT Project No. 2003/220

ISBN 978-0-86396-900-3

Cover images

Top left: An ectoparasite, Zeuxapta seriolae (Platyhelminthes: Monogenea), on the gills of Seriola lalandi (Photo: C.B. Chambers)

Bottom left: Brad Smith (Research Assistant) releasing a tagged S. lalandi near Port Augusta, Spencer Gulf, South Australia (Photo: J. Laidlaw)

Right: Kate Hutson (PhD student) sampling a farmed S. lalandi in Arno Bay, Spencer Gulf, South Australia (Photo: B.P. Smith)

This publication may be cited as:

K.S. Hutson, I.D. Whittington, B.M Gillanders, J.E. Rowntree, I. Ernst, C.B. Chambers and C. Johnston. 2008. Potential for parasite interactions between wild and farmed kingfish, discrimination of farmed and wild fish and assessment of migratory behaviour. Final Report to FRDC (Project No. 2003/220). The University of Adelaide, Adelaide. 83 pp.

The University of Adelaide

Marine Parasitology Laboratory School of Earth and Environmental Sciences DX 650 418 Darling building The University of Adelaide Adelaide 5005 South Australia

Telephone: (08) 8303 5282 Facsimile: (08) 8303 4364 http://www.adelaide.edu.au

Disclaimer: The authors warrant that they have taken all reasonable care in producing this report. Although reasonable efforts have been made to ensure quality, The University of Adelaide does not warrant that the information in this report is free from errors or omissions. The University of Adelaide 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.

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without written permission from the first author.

Authors: K.S. Hutson, I.D. Whittington, B.M Gillanders, J.E. Rowntree, I. Ernst, C.B. Chambers and C. Johnston Reviewer: M. Deveney Distribution: FRDC, FRAB representatives, Libraries and Scientific contributors Circulation: Public Domain

DEDICATION

To Reggie Godfrey

Reginald Thomas Godfrey, who died suddenly in October 2006, was regarded in the fishing community as an authority on yellowtail kingfish. He was frequently in demand for his fishing expertise. A true-blue Aussie bloke, Reggie followed in the footsteps of his father, fishing in upper Spencer Gulf between the 1950‟s to mid 80‟s. Over the past ten years he assisted the collection of kingfish broodstock for aquaculture. More recently, Reggie assisted with the objectives of this report, enabling collection of parasite specimens and otoliths from wild kingfish and helping to facilitate the tag and release programme. We will never forget the mateship and knowledge he shared.

TABLE OF CONTENTS

NON TECHNICAL SUMMARY ______11 ACKNOWLEDGMENTS ______14 BACKGROUND ______15 NEED ______18 AIMS AND OBJECTIVES ______20 OBJECTIVE 1: POTENTIAL FOR PARASITE INTERACTIONS BETWEEN WILD AND FARMED KINGFISH ______21 METHODS ______21 1.1. Wild and farmed fish collection ______21 1.2. Parasite collection ______22 1.3. Risk Assessment ______24 1.3.1. Likelihood ______24 1.3.2. Consequence ______28 1.3.3. Controlled risk ______28 RESULTS/DISUSSION ______30 1.4. Parasite species identified ______30 1.5. Risk Assessment ______30 1.5.1. Copepods______39 1.5.2. Monogeneans ______41 1.5.3. Acanthocephalans ______42 1.5.4. Cestodes ______43 1.5.5. Myxozoans______44 1.5.6. Nematodes ______45 1.5.7. Trematodes ______46 OBJECTIVE 2: DISCRIMINATION OF FARMED AND WILD FISH ______48 METHODS ______48 2.1. Natural elemental signatures ______48 2.2. Rare earth elements ______48 2.3. Chemical dye of hatchery derived kingfish ______49 2.3.1. Fish maintenance and chemical treatments ______49 2.3.2 Otolith preparation and analysis ______50 2.3.3. Statistical analysis ______51 2.4. Mark success ______52 OBJECTIVE 3: ASSESSMENT OF MIGRATORY BEHAVIOUR OF WILD KINGFISH ______56 METHODS ______56 3.1. Conventional tagging programmes and participation ______56 3.2. Fish capture ______56 RESULTS/DISCUSSION ______57 3.3. Movements ______59 3.4. Research development and interest in the community ______62

FURTHER DEVELOPMENT ______64 1.1. Improved prevention and management of parasites ______65 1.2. New zone investigation ______66 1.2.1. Sources of infection ______66 1.2.3. Wild fish migration ______67 1.2.4. Stock structure ______68 CONCLUSIONS ______69 REFERENCES ______71 APPENDIX 1: INTELLECTUAL PROPERTY ______78 APPENDIX 2: STAFF ______79 APPENDIX 3: PUBLICATIONS ARISING FROM THIS REPORT ______80 APPENDIX 4: PRESENTATIONS ARISING FROM THIS REPORT ______82 APPENDIX 5: MEDIA ARISING FROM THIS REPORT ______83

LIST OF FIGURES

Figure 1. Sample locations for wild and farmed kingfish and wild Samson fish in southern Australia...... 22 Figure 2. Tag and release localities and movement of Seriola spp...... 57

LIST OF TABLES

Table 1. Risk estimation matrix (based on AQIS 1999)...... 26 Table 2. Likelihood and consequence definitions (based on Fletcher et al., 2004)...... 26 Table 3. Metazoan parasites of wild and farmed Seriola spp. in Australia listed in alphabetical order...... 31 Table 4. Consequence of parasite establishment and proliferation (Parasite species listed in alphabetical order)...... 34 Table 5. Parasite risk analysis ...... 36 Table 6. Concentrations of chemical dyes for 6 and 24 h durations ...... 49 Table 7. Rank sum scores determined by Mann-Whitney U tests ...... 53 Table 8. Seriola lalandi >1000 mm TL tagged at Port Augusta during 2005 and 2006 ... 59

10 2003/220 Potential for parasite interactions between wild and farmed kingfish, discrimination of farmed and wild fish and assessment of migratory behaviour

PRINCIPAL INVESTIGATOR: Colin Johnston ADDRESS: Primary Industry and Resources South Australia, Aquaculture, GPO Box 1625, Adelaide, South Australia, 5001, Australia PRESENT ADDRESS: Biosecurity New Zealand, Ministry of Agriculture and Forestry, PO Box 2526, Wellington, New Zealand

OBJECTIVES: 1. Potential for parasite interactions between wild and farmed kingfish 2. Assessment of migratory behaviour of wild kingfish 3. Discrimination of farmed and wild kingfish

NON TECHNICAL SUMMARY

OUTCOMES ACHIEVED

This project greatly contributed to an increased understanding of parasite fauna of yellowtail kingfish in Australian waters. This baseline information is critical to understanding potential parasite interactions between wild and farmed kingfish. Our risk assessment enables consideration of parasite species of potential threat to the emerging industry. We found batch marking hatchery fish with fluorescent dye may be the most practical and inexpensive method to discriminate wild and farmed fish. We also provide the first data on wild kingfish migrations in Spencer Gulf, demonstrating that there may be heightened interactions between wild and farmed fish in Fitzgerald Bay in summer. This project provided training to several researchers, including one PhD student in the area of aquatic animal health and one Honours student in the area of otolith chemistry. We indicate appropriate methods to enable better management practices in the kingfish industry which will help improve the viability of kingfish aquaculture in Australia in the future.

Australian yellowtail kingfish (Seriola lalandi) aquaculture has potential to develop into a large, internationally recognised and sustainable industry. Research and development is required to support rapid growth of this emerging industry whilst ensuring efficient production and while minimizing environmental impacts. Negative interactions between wild and farmed fish may harm finfish aquaculture through increased disease management costs and poor public perception of the industry and although it remains a controversial topic, evidence is emerging in the literature that poorly managed aquaculture can negatively impact the health of wild fish stocks.

This study performs a qualitative risk assessment for 54 metazoan parasite species we found infecting wild Seriola spp. in southern Australia. We estimate risks to local sea- cage aquaculture of kingfish. Risk was estimated by considering the likelihood and consequence of parasite establishment and proliferation in kingfish sea-cage farming. Benedenia seriolae and Zeuxapta seriolae (Monogenea) were considered extremely likely to establish and proliferate. B. seriolae is currently recorded as the highest potential negative consequence for cost-effective sea-cage farming of kingfish in Australia, should the industry expand elsewhere. However, B. seriolae infections can be managed by bathing fish in either hydrogen peroxide or fresh water. Paradeontacylix spp. (Digenea) are recorded as a moderate risk for sea cage aquaculture. Absence of potential mitigation methods and parasite management for Paradeontacylix spp. (Digenea), and estimated low risk myxozoan species (Kudoa sp. and Unicapsula seriolae) suggests that these species may present the highest negative consequences for kingfish aquaculture in Australia.

Naturally occurring elemental signatures (or composition) or chemical marking of fish ear stones (otoliths) may enable aquaculture and wild-caught yellowtail kingfish to be distinguished from one another, so that in the event of aquaculture escapes, escaped fish could be identified. We analysed elemental signatures (magnesium, manganese, strontium and barium) of otoliths from wild caught and farmed fish via Laser Ablation- Inductively Coupled Plasma-Mass Spectrometry (LA-ICP-MS). A significant difference was detected between elemental signatures of farmed and wild-caught fish, but this difference was largely attributable to wild fish otoliths from Port Augusta being different from the other groupings. When comparisons between farmed and wild fish were made for each of inner and outer Spencer Gulf, aquaculture fish could be distinguished from wild fish and classified correctly with a high degree of accuracy (82-100%).

Marking fish is a reliable and cost effective method for distinguishing between marked hatchery fish and unmarked wild populations. We used fluorescent dyes (alizarin complexone, alizarin red S and oxytetracycline) to mark hatchery reared fish. Alizarin complexone and alizarin red S had the greatest mark success with 100% and 86-100% of fish receiving a good or very good mark, respectively.

To gain an indication of the migratory path of wild kingfish and Samson fish (S. hippos) and potential interaction with farmed kingfish, wild Seriola spp. were tagged between December 2004 and November 2006 in Spencer Gulf and offshore from the west coast of Eyre Peninsula, South Australia. Two hundred and forty-eight kingfish and 73 Samson fish were tagged with 28 kingfish and 2 Samson fish returns. The maximum distance moved by kingfish was 130 km and the maximum time at liberty was 442 days. Samson fish were at liberty for a maximum of 378 days and were recaptured at the original capture site. Recapture results indicate that large kingfish remain in northern Spencer Gulf for up to five months, migrate south in summer past kingfish sea-cage farms in Fitzgerald Bay and return in late Autumn. Northern Spencer Gulf may be important for aggregations of large, reproductively mature kingfish. A remarkable outcome of this research was the residual impact that the tagging programme had on sustainable fishing practices in South Australia.

This project provided specialised training in introductory and advanced parasitology, risk assessment and otolith chemistry. It had extension and application in the kingfish aquaculture industry and the recreational fishing community.

KEYWORDS: Seriola, aquaculture, parasites, otolith chemistry, conventional tagging programme.

ACKNOWLEDGMENTS

This research project was funded by Primary Industries and Resources, South Australia and the Fisheries Research and Development Corporation grant no. 2003/220 under the Innovative Solutions for Aquaculture initiative. We would like to thank the kingfish aquaculture industry for their collaboration and support, in particular South Australian Aquaculture Management (SAAM), Southern Star Aquaculture, Spencer Gulf Aquaculture, the Stehr Group and Navajo Aquaculture. We thank the Innes Brothers - Bill, Chris, Ian and Ron, and their sons Michael and Cleve – who assisted with sampling yellowtail kingfish from NSW and Scott Gray who coordinated specimen collection in Victoria. Greg Kent and Reggie Godfrey made a significant contribution to wild fish collections in South Australia. We thank all recreational fishers who contributed wild fish toward this project. Dr Robert Adlard, Prof Ian Beveridge, Dr Thomas Cribb, Dr Craig Hayward, Dr Francisco Montero, Dr Matthew Nolan, Dr Sylvie Pichelin, Dr Pat Pilitt, David Schmarr and Shookefeh Shamsi provided parasitological expertise. We are indebted to Dr Danny Tang (The Univeristy of Western Australia) and Prof Geoff Boxshall and Dr Rodney Bray (The Natural History Museum, London), who assisted with identification of crustacean and trematode parasites. Thanks to Hans Schoppe who assisted with histology. We thank SARDI Aquatic Sciences, especially Tony Fowler, Jane Ham and Paul Jennings, for obtaining some farmed and wild-caught samples for otolith work. Melita de Vries, Travis Elsdon, Andrew Munro and John Tsiros assisted with otolith analyses. We thank Marty Deveney for his unfaltering assistance with this research and for providing constructive comment to this document. Part of this research was made possible through an Australian Postgraduate Award scholarship and an ARC/NHMRC Network for Parasitology Travel Research Grant awarded to Kate Hutson.

BACKGROUND

Yellowtail kingfish (Seriola lalandi) aquaculture in South Australia has potential to expand greatly. Several factors emphasise this species as an excellent prospect for sea- cage aquaculture, including favourable biology, availability of established and emerging markets and excellent farming sites in Spencer Gulf. However, among the myriad of challenges that face this industry are two major risks that threaten growth, efficiency and, therefore, sustainability: 1) potentially devastating effects of disease, which must be managed in an efficient, sustainable and responsible manner and 2) public resistance to industry expansion due to perceived negative impacts on the environment, including escaped fish.

The potential for negative interactions between wild and cultured fish stocks has resulted in fierce public resistance to aquaculture expansion in the northern hemisphere. There are two principal concerns: 1) parasites from sea-caged fish may increase levels of parasitemia in wild fish; 2) the interbreeding of escaped and wild fish stocks can alter the genetic structure of wild fish populations (e.g. Bjørn et al. 2001; Jonsson 1997; McVicar 1997). Both factors have been identified as possible reasons for declines in wild fish populations and are used by lobbyists as grounds to limit industry expansion. However declines in wild fish populations can be due to other factors (e.g. overfishing, habitat destruction and climate change) and fish farming may be blamed unjustly (e.g. Noakes et al. 2000). For both principal concerns, research is required to help defend industry from false accusations of environmental harm and to monitor responsibly and prevent potentially negative environmental impacts.

Interactions between wild and farmed fish also concern farmers because wild fish may introduce disease to cultured fish, making effective disease management difficult. Despite a perception that wild fish are “pristine” (free from pathogens), they actually harbour a wide variety of disease causing organisms naturally. Wild fish are the most likely source of pathogens that may infect sea-caged fish and cause harmful epizootics (McVicaro 1997; Bouloux et al. 1998). Seriola spp. are susceptible to a variety of viral, bacterial and

parasitic diseases. In Japan, these pathogens and parasites are responsible for serious production inefficiencies. Monogenean parasites are among the most serious problems for the yellowtail industry (S. dumerili, S. lalandi and S. quinqueradiata) and may increase production costs by well over 20% (Ernst et al. 2002). In Australia, monogeneans have become a serious risk to sustainable development of the emerging kingfish (S. lalandi) industry in Spencer Gulf. Without efficient, effective and environmentally aware parasite management strategies, competitive and sustainable production will be impossible.

The ability to identify fish successfully is an important tool for fisheries management because it provides information that can improve stock identification, enhance stocking programs, estimate population sizes and determine fish growth. Among identification techniques, the use of calcified structures of fish, such as scales, spines, and ear stones (otoliths), have been beneficial in ways where traditional techniques, such conventional tags, have not. Coded wire tags involve the handling of individual fish, are labour intensive, cannot be applied on larval and juvenile fish and have variable retention times.

As otoliths grow, they incorporate chemicals and elements from their environment, which are absorbed across the gills or intestines into the fish‟s blood stream (Campana 1999). Naturally occurring elemental signatures in otoliths are dependent on finding differences between groups of fish. This technique has been used successfully to identify groups of fish in several stock discrimination studies (e.g. Begg et al. 1998) and also for determining the origins and movements of fish (Gillanders 2002). A critical assumption made in these studies is that, not only do otoliths reflect the environment of a fish, but that geographically distinct stocks also have distinctly different otolith compositions (Begg et al. 1998).

Distinctive marks can be imprinted onto the otolith by the immersion, injection or feeding of fish with chemical marking agents such as chemical elements (i.e. strontium) (e.g. Milton and Chenery 2001) or fluorescent dyes (Hernaman et al. 2000; Thomas et al. 1995). Marks deliberately incorporated into otoliths using trace elements or fluorescent dyes can be long lasting (Brothers 1990), large numbers of fish can be marked at once including eggs and larvae, and therefore provide a cost effective method that requires minimal handling (Beltran et al. 1995). In addition, administering marks does not affect

life history characteristics such as growth, reproduction, and predation (Bashey 2004; Leips et al. 2001).

The immersion of fish in fluorescent dyes has been highly successful, allowing large numbers of fish to be marked simultaneously with minimal mortalities. The absorption of fluorescent chemicals into the otolith can be seen as a visible increment when viewed under light of the appropriate wavelength. Several chemical marks can be made on the otolith to distinguish between batches of fish (Thomas et al. 1995). Consequently, marking fish is a reliable and cost effective method for distinguishing between hatchery fish and wild populations (Beltran et al. 1995; van der Walt and Faragher 2003). It remains uncertain, however, which fluorescent dyes are most successful for marking otoliths, as there has been mixed success among fish species and between chemical dyes in controlled experiments.

An ideal marking method should use a readily available compound to create an unambiguous mark. This compound should be affordable and applied in as minimal time as possible (Thomas et al. 1995). The most common fluorescent dyes used for marking fish otoliths are oxytetracycline (OTC), calcein, alizarin complexone (AC) and alizarin red S (ARS), all of which have unique advantages in their use in marking fish otoliths. OTC is a commonly used marker for calcified structures (including otoliths) in fish (Beltran et al. 1995; Francis et al. 1992; Secor et al. 1991). OTC is an antimicrobial medicine which binds to calcified tissues as they grow, producing a fluorescent increment when viewed under ultraviolet light. However, OTC is photosensitive, and the quality of marks has been reported to be reduced when exposed to light (Hernaman et al. 2000). Similarly, calcein has been used successfully to mark fish otoliths, however, large mark variability has been associated with autofluorescence of the otolith structure in calcein marked otoliths (Tsukamoto 1988). Successful marking of fish otoliths has also occurred using AC and ARS (Lagardere et al. 2000; van der Walt and Faragher 2003). Both AC and ARS fluoresce scarlet pink with minimal autofluorescence, allowing better detection based on a more distinctive response. Few studies have considered the use of ARS (Beckman and Schulz 1996; Lagardere et al. 2000), however, success of this chemical dye may be beneficial as it is inexpensive compared with the other fluorescent compounds.

NEED

Sea-caged fish may become infected with a variety of parasites and pathogens, and these organisms pose a significant threat to aquaculture production. Many parasites and pathogens are known from Seriola spp. including viruses, bacteria, protozoa, helminths and crustaceans (Noga 2000; Ogawa and Yokoyama 1998; Roberts 2001). However hatchery reared fish are normally free from disease and kingfish newly stocked into sea- cages acquire infections from populations of wild fish. Cultured kingfish may develop higher parasite burdens than those present in wild fish populations because conditions in sea cage aquaculture enhance parasite transmission. In most cases, parasites are host specific and can only infect a single or several closely related fish species. Monogenean parasites provide an example of host specificity: Zeuxapta seriolae and Benedenia seriolae are known to infect only Seriola spp. Parasite host-specificity allows us to predict which species of naturally occurring fish may act as vectors or reservoirs of infection for cultured fish. In the case of kingfish, the majority of parasites and pathogens experienced in aquaculture are therefore likely to originate from natural populations of the same or similar species.

Two monogenean parasites, B. seriolae and Z. seriolae, are serious health concerns for cultivation of kingfish in Spencer Gulf. However, experience in Japan indicates that many other parasites may infect kingfish and threaten production efficiency (Egusa 1983; Ogawa and Yokoyama 1998). For reasons discussed above, these potential threats are likely to originate from endemic parasites and pathogens in natural populations of wild kingfish. However, wild kingfish in Spencer Gulf have not been studied to determine their endemic parasite fauna. This information is essential to determine: 1) which potentially damaging parasites and pathogens occur locally and 2) which potentially damaging parasites may be absent. Knowledge of which high-risk parasites are present in Spencer Gulf is required to ensure adequate monitoring programmes and contingency plans are developed for the aquaculture industry. Information as to which high-risk parasites are locally absent is also required to ensure that both farmed and wild fish are protected from new introductions.

In southern Australia, there has been increased public concern over escaped farmed fish entering the wild and the effects these fish may have on wild fish populations. There have been many claims recently that large numbers of yellowtail kingfish have escaped from aquaculture ventures and may interfere with natural populations in one of four ways: 1) hatchery reared yellowtail kingfish may compete with natural populations for resources such as food and space; 2) yellowtail kingfish are carnivorous and have the potential to prey upon other marine organisms including commercially and recreationally significant species (smaller fish, prawns, cuttlefish and crabs); 3) pathogens may be transmitted to wild fish with potentially devastating effects on their populations; and 4) increased numbers of yellowtail kingfish may encourage above normal numbers of larger predators, such as sharks, into an area. A reliable, affordable and food safe method of marking large numbers of fish is therefore required to distinguish hatchery-reared fish from wild populations for yellowtail kingfish in order to increase our knowledge regarding the contributions of hatchery-reared fish into wild populations.

In Australia, yellowtail kingfish fingerlings are grown from fertilised eggs in land-based hatcheries where, through standard biosecurity practices, fish are isolated from infection by metazoan parasites. When fingerlings are moved into sea-cages, wild marine fish species may act as an initial source of parasites for farmed fish. The natural occurrence of wild kingfish and Samson fish (S. hippos) near locations where kingfish are farmed in South Australia provides an opportunity for transfer of parasites from wild to farmed populations. Despite intensive cooperative tagging programme for kingfish on the east coast of Australia (Gillanders et al. 2001) and for Samson fish on the west coast of Australia (Rowland et al. 2006), there are limited data available concerning the movements of wild Seriola spp. in South Australian waters. Understanding migrations of wild Seriola spp. is critical to understand potential interactions between wild and farmed fish in Spencer Gulf.

AIMS AND OBJECTIVES

Objective 1. Potential for parasite interactions between wild and farmed kingfish We document the parasite species of wild and farmed Seriola species in South Australian waters and assess the risk to sustainability posed by documented parasites. We use risk analyses to identify: 1) the most likely parasite species to establish and proliferate in South Australian kingfish sea-cage aquaculture; 2) parasite species with potentially negative consequences for kingfish aquaculture and; 3) parasite species which may be difficult to manage, i.e. parasite species of immediate priority for research into potential management strategies.

Objective 2. Discrimination of farmed and wild kingfish To date, two methods for distinguishing between wild and hatchery reared kingfish have been used. Gillanders and Joyce (2005) used natural elemental signatures in the otolith to distinguish between hatchery reared and wild populations of kingfish, and Fowler et al. (2003) used morphometric measurements to distinguish between reared and wild kingfish otoliths. The work carried out by Gillanders and Joyce (2005) constituted part of this research project. We also attempted to use rare earth elements, but with limited success. Here, we evaluate the most appropriate and cost effective fluorescent dye to mark kingfish otoliths.

Objective 3. Assessment of migratory behaviour of wild kingfish We conducted a small-scale tagging programme in Spencer Gulf and offshore from Eyre Peninsula in South Australia to investigate the timing and nature of Seriola migrations. We compile tag and preliminary recapture data and determine movements for Seriola spp. in South Australian waters. This research provides new information on the nature of Seriola migrations and will be a useful foundation for future assessments of the recreational fishery and interactions of wild Seriola spp. with the kingfish aquaculture industry.

OBJECTIVE 1: POTENTIAL FOR PARASITE INTERACTIONS BETWEEN WILD AND FARMED KINGFISH

An abridged version of objective 1 has been published in Aquaculture:

Hutson, K.S., Ernst, I. and Whittington, I.D. (2007). Risk assessment for parasites of Seriola lalandi (Carangidae) in South Australian sea cage aquaculture. Aquaculture 271, 85-99.

METHODS

1.1. Wild and farmed fish collection

Commercial fishers caught 25 S. lalandi at Sir John Young Banks off Greenwell Point, NSW between June and July 2003 (Figure 1). Fish ranged from 760 to 950 mm fork length (FL). Recreational fishers caught 25 S. lalandi off Killarney, Victoria in January 2004 and January 2005 (Figure 1). Fish ranged from 440 to 790 mm FL. Wild S. lalandi (n = 62) were caught by line in Spencer Gulf and offshore at Greenly Island and Kangaroo Island, South Australia between 2003 and 2005 and wild S. hippos (n = 6) were caught at Greenly Island and Rocky Island, South Australia between 2004 and 2005 (Figure 1). Wild fish ranged in size from 320 to 1410 mm fork length (FL). Farmed S. lalandi (n = 58) were captured by line from inside sea-cages in Spencer Gulf, South Australia at Fitzgerald Bay near Whyalla (n = 14), Arno Bay (n = 26), and Boston Bay near Port Lincoln (n = 18) between 2003 and 2005 (Figure 1). Farmed fish ranged in size from 282 to 652 mm FL. We attempted

Figure 1. Sample locations for wild and farmed kingfish and wild Samson fish in southern Australia.

1.2. Parasite collection

Live fish were bathed individually in 10-20 L of seawater containing 5 mg/L praziquantel for 10 min to dislodge all gill monogeneans (Mansell et al. 2005). Fish were then bathed in freshwater for approximately 10 min to remove all Benedenia seriolae (see Chambers and Ernst 2005). Fish were euthanased in this treatment with a lethal dose of Aqui-S (> 200 mg/L). If fish specimens were not obtained alive, they were bathed in freshwater after the gills had been removed and fixed in 10% formalin. All parasites were collected from the bath water from both treatments by filtration through a 75 µm sieve and fixed in 10% formalin (Chambers and Ernst 2005).

The exterior of the fish, including inside the mouth and buccal folds and in the fin sulcus, was examined for ectoparasites. Parasites were preserved in 10% formalin. The gills

were removed by dissection and examined for ectoparasites and gross pathology. The heart was removed, opened and examined for parasites under a dissecting microscope. It was then flushed with saline water solution and the settled contents were examined under a dissecting microscope. Alternatively, the heart was fixed in 10% formalin and examined later. The brain cavity and body cavity were exposed and samples of brain, muscle, spleen, liver, kidney, gall and gonad were removed and fixed in 10% formalin. Tissues were embedded in paraffin wax, sectioned at 5 µm and stained with Mayer‟s haematoxylin and eosin for routine light microscope examination.

The viscera, swim bladder and muscle were examined for gross pathology as an indication of the presence of parasites. The digestive tract was removed and the stomach, caeca and intestine were opened separately and shaken vigorously in physiological saline. The settled contents were sorted under a dissecting microscope and parasites were aspirated with a pipette. Fresh squashes were made of brain, muscle and bile and examined using a compound microscope. Bile was not examined in NSW. If parasites were detected, they were fixed in 10% formalin. Sub-samples of myxozoans from parasitised bile were frozen in the field, then thawed and measured in the laboratory. Measurements were made using a computerised digitising system similar to that described by Roff and Hopcroft (1986).

Nematodes were preserved in 70% ethanol. Fixed nematodes were cleared and mounted in lactophenol and examined using a compound microscope. Trematodes and cestodes were killed in near boiling saline solution, then fixed in 10% formalin. Fixed parasites were placed in distilled water before being stained in Mayer‟s haematoxylin overnight, then destained in 1% HCl in 70% ethanol. The trematodes and cestodes were dehydrated in a graded ethanol series before being cleared in cedar wood oil and mounted on a slide in Canada balsam. Trematodes and cestodes were examined using a compound microscope. Cestode blastocysts containing larval cestodes were mounted alive on a slide. The tentacles were encouraged to evert by applying coverslip pressure and the worms were then preserved in 70% ethanol or 10% formalin.

Representatives of parasite species are lodged in the South Australian Museum (SAMA), North Terrace, Adelaide, South Australia 5000, Australia. We used FreeCalc (Ausvet Animal Health Services, 2002) to estimate the likelihood of not detecting a parasite species in our sample of wild and farmed fish. Assumptions were: 1) if a fish was infected

the parasite would have been detected, and 2) that fish were sampled from a large population size (n = 999, 999).

1.3. Risk Assessment

We devised a qualitative five-factor assessment to estimate the likelihood of parasite establishment in farmed fish by combining current knowledge on: 1) the potential exposure of farmed hosts to parasites on wild hosts and 2) the biological pathway necessary for parasite species to infect the farmed fish species. We also developed a semi- quantitative framework to assess the potential negative consequence of establishment and proliferation of parasite species. This risk assessment estimated two parameters, likelihood and consequence, to be considered independently. The parameters were then combined to produce an estimated total level of risk associated for each parasite species. Consequence was also reviewed in light of potential mitigation measures and is herein referred to as „controlled risk‟. Risk was estimated for each parasite species identified from this survey and from previously published records in the scientific literature.

1.3.1. Likelihood The likelihood of parasite establishment and proliferation in kingfish sea-cage farming was estimated. This included: 1) an estimate of exposure of farmed fish to wild infected kingfish species considering information currently available on host species range and parasite species geographic distribution and 2) information on the biological pathways necessary for the parasite species to infect the farmed fish species. Parasite host- specificity allows us to predict which species of naturally occurring fish may act as vectors or infection reservoirs for farmed fish. For the purposes of this study, we examined wild kingfish species for metazoan parasites. It is important to keep in mind that parasites of farmed fish which exhibit direct life-cycles are likely to originate from natural populations on the same host species or genera, while those with complex life-cycles will involve intermediate host species.

Five likelihood criteria were used to estimate farmed fish exposure to the parasite species and five likelihood criteria were used to estimate the likelihood of parasite establishment through the necessary biological pathway. Using a risk estimation based on the Australian

Quarantine and Inspection Service (AQIS 1999) (Table 1), these two factors were combined to estimate the likelihood of parasite establishment and proliferation in kingfish farming in South Australia. Likelihood definitions ranged from negligible to extreme (based on Fletcher et al. 2004) (Table 2).

Estimate of exposure of farmed fish to wild parasitised Seriola species To estimate exposure we considered the likelihood of a parasite occurring in areas of kingfish farming. Wild kingfish and Samson fish are migratory species capable of moving long distances. Studies on the movement of kingfish using conventional tags show that some fish from the east coast of Australia migrate between Australia and New Zealand (Gillanders et al. 2001). It is not known, however, whether kingfish migrate from waters of Victoria, New South Wales (NSW), Queensland or Western Australia into South Australian waters. Kingfish tagged in NSW have been recaptured in Victoria and Western Australia, suggesting long migrations do occur (Woodrick, K., NSW Department of Primary Industries; unpublished data). For the purposes of this risk assessment, parasite species that have been documented only from the east coast of Australia were considered to present a negligible risk of exposure, while those only from the southern coast (i.e. Victorian coast) were considered to present a low risk of exposure to farmed kingfish.

Samson fish are also a highly mobile fish; tagging data show that fish from Perth are capable of moving up to 1100 km along the south coast of Australia (Rowland, A., Murdoch University, unpublished data). Samson fish are not known to enter Spencer Gulf where kingfish are farmed, but are captured near the entrance to Spencer Gulf. For the purpose of this risk assessment, parasite species of Samson fish were also considered to pose a low risk of exposure to farmed kingfish.

Table 1. Risk estimation matrix (based on AQIS 1999). Likelihood of parasite establishment / proliferation (2) Likelihood of exposure (1) Negligible Low Moderate High Extreme Extreme Negligible Low Moderate High Extreme High Negligible Low Moderate High High Moderate Negligible Negligible Low Moderate High Low Negligible Negligible Negligible Low Moderate Negligible Negligible Negligible Negligible Negligible Low

Table 2. Likelihood and consequence definitions (based on Fletcher et al., 2004). Likelihood level Descriptor Extreme Already occurs with high certainty High It is expected to occur

26 Moderate May occur

Low Uncommon Negligible Never heard of, but not impossible Consequence and risk level Extreme Catastrophic consequences to the entire industry, total mortality or eradication of fish is considered, trade implications at the national level High Establishment and proliferation of parasites could have serious biological consequences, prolonged high mortality rates, enterprise survival is questioned, significant economic concern to the industry Moderate Substantial seasonal morbidity and mortality rates with significant cost to the farmer to warrant intermittent concern by the industry Low Establishment and proliferation of parasite species is manageable with low economic significance to the industry Negligible Establishment and proliferation of parasite species could have no significant consequences, with low or no measurable economic effect at the enterprise level

In South Australia, kingfish are believed to spawn at the top of Spencer Gulf near Port Augusta, where they are frequently captured (Fowler et al. 2003). Given the long- range movements of large, wild kingfish (see Gillanders et al. 2001), it is likely that kingfish migrate up Spencer Gulf and pass S. lalandi farms in order to spawn at the top of the gulf. For the purposes of this risk assessment, parasites found on wild kingfish in South Australian waters outside Spencer Gulf were considered to present a moderate risk of exposure to farmed kingfish. A high risk was assigned to parasites found to be infecting kingfish in Spencer Gulf, while an extreme risk was assigned to parasites found on farmed kingfish in Spencer Gulf.

Biological pathway necessary for parasite species to infect farmed fish species We estimated the likelihood of parasite transfer from wild fish to farmed fish considering the biological pathway or route of infection. Parasites known only to infect Samson fish or other fish species and not kingfish may be host-specific (i.e. may only infect one host species). These parasites may present minimal, if any, risk to kingfish and were considered to pose a negligible risk of establishment. Parasites with complex, indirect life-cycles that require two or more specific host species for development may be limited in their ability to establish and proliferate in farmed fish because of restrictions to interactions between required host species, e.g. these parasite species may require an infected intermediate host to be consumed by the definitive host. In a sea-cage, there are limited opportunities for farmed fish to consume infected wild species unless smaller, parasitised fish or crustaceans move into the sea-cage through the netting. Parasites with two or more host species in the life-cycle, that require the definitive host to consume an infected intermediate host, were considered to pose a low risk of establishment and proliferation.

Parasite species that require only two host species to complete their life-cycle are more likely to be present in farmed fish, particularly when intermediate host species are in close proximity to sea-cages. The likelihood of infection would also increase for parasites with a direct infective stage to locate the definitive host (i.e. sanguinicolids and presumably myxozoans). Parasites requiring a two host life-cycle with direct infection of the definitive host were considered to pose a moderate risk for farmed fish. Parasites with direct life-cycles are usually found in sea-cage aquaculture as they only require a single host species and may be able to reproduce rapidly. These parasites

were considered a high risk of establishment. Parasites that have previously established on sea-caged kingfish in South Australia were considered to pose an extreme risk of establishment and proliferation.

1.3.2. Consequence Information was gathered from the scientific literature on the parasite (species or genus, if available) to indicate any potential negative consequence of establishment and proliferation in regard to sustainable aquaculture in Australia. Information was sought that directly related to host pathology, previous parasite records in Seriola spp. aquaculture, potential impact on marketability and potential impact on consumer health. Using this information, we estimated the consequence of parasite establishment and proliferation as adapted from risk criteria by Fletcher et al. (2004) (Table 2).

The consequence of potential parasite establishment and proliferation was reviewed with regard to four risk criteria including: 1) potential to cause mass mortality, 2) parasite pathology, 3) potential impact on marketability and 4) potential impact on consumer health. The first criterion was met if the parasite had the potential to cause mass mortality as demonstrated by previous experiences in Seriola aquaculture. The second criterion included parasites that have been associated with host pathology or morbidity in aquaculture. The third criterion grouped parasites that could potentially have a negative impact on consumer acceptance. The fourth criterion was met if the parasite species is known to have a negative impact on consumer health. Parasites were scored for each of these four criteria. Parasites that met all four criteria were assigned an extreme consequence, parasites that met three criteria were assigned a high consequence, parasites that met two criteria were assigned a moderate consequence, parasites that met one criterion were assigned a low consequence and parasites that meet no criteria were assigned a negligible consequence.

1.3.3. Controlled risk Consequence was then reviewed considering current management procedures available that could potentially mitigate parasite infestations in the event of an outbreak. For parasite species where management procedures are viable, we considered a lowered consequence level. Multi-factor management procedures reduced the controlled risk

level further. Parasite species for which there are no current mitigation options present the same controlled risk level as estimated previously in the consequence table.

RESULTS/DISUSSION

1.4. Parasite species identified

The parasites detected on wild and farmed kingfish and wild Samson fish during this survey and from previous published records in Australia are shown in Table 3. The microhabitat of the parasites is also indicated. Where a parasite species was found in all three regions of the digestive tract (i.e. the stomach, caeca and intestine), their microhabitat is noted as „digestive tract‟. Some parasites could not be identified to species, a result of a combination of factors including limited number of parasite specimens, potentially undescribed parasite species and inability to identify larval parasite life stages definitively. Empty cysts, suspected of being cestode blastocysts, were observed in the viscera of farmed kingfish from Fitzgerald Bay and Botany Bay and in the viscera of wild Samson fish from Greenly Island. Blastocysts of similar colour and size containing a larval cestode, Callitetrarhynchus gracilis, occurred in the viscera of wild kingfish.

Samples of wild (n = 62) and farmed (n = 58) kingfish gave close to 95% confidence of detecting parasite species at 10% prevalence. The sample of wild Samson fish (n = 6) provided a 95% chance of detecting a parasite prevalence of 40%.

1.5. Risk Assessment

Parasites were ranked as posing a negligible to extreme likelihood of establishment and proliferation from wild Seriola spp. to farmed kingfish in South Australia (Table 5). The monogeneans B. seriolae and Z. seriolae, which have previously proven problematic in S. lalandi aquaculture in Spencer Gulf, presented an extreme likelihood

Table 3. Metazoan parasites of wild and farmed Seriola spp. in Australia listed in alphabetical order.

Phylum Taxon Microhabitat Host Locality Wild Farmed Reference/Accession no Acanthocephala Australorhynchus tetramorphacanthus Intestine lalandi EC, SA Yes No Lebedev, 1967 Rhadinorhynchus spp. Intestine lalandi EC, Vic Yes No AHC29141-29142 & 34177 Cestoda Callitetrarhynchus gracilis Body cavity lalandi SG Yes Cyst: W, PL AHC29179 Nybelinia thyrsites Intestine lalandi EC Yes No AHC29134 Tetraphyllidea Type 1 Stomach lalandi EC, SG Yes No AHC29135 Type 4 Digestive tract lalandi EC, Vic, SG Yes PL AHC29136-29140 Copepoda Caligus amblygenitalis Cavities* lalandi EC Yes No C6311 C. epidemicus Body surface* lalandi EC, SG Yes No C6313

31 C. lalandei Body surface* lalandi EC, Vic Yes No C6228 & C6229

Body surface hippos SA Yes No C6314 C. spinosus Gill arch lalandi EC, Vic Yes No C6230 & C6231 Caligus sp. 1 Not determined lalandi EC, Vic, SA, SG Yes W, AB, PL C6232 hippos SA Yes NA C6315 Caligus sp. 2 Gill arches hippos SA Yes NA C6316 Dissonus hoi Nasal cavity lalandi EC, SG Yes W C6317 Nasal cavity hippos WC Yes No Tang and Kalman, 2005 Lepeophtheirus sp. Not determined hippos SA Yes No C6318 Lernanthropus paenulatus Gills lalandi EC, Vic, SG Yes No C6239 & C6309 Gills hippos SA Yes No C6319 Parapetalus spinosus Gills hippos SA, WC Yes NA C6320 Parabrachiella seriolae Buccal folds lalandi EC, Vic, SG Yes No C6321 Fin sulcus hippos SA Yes NA C6322 Parabrachiella sp. Gills lalandi EC, SG Yes No C6238 Peniculus sp. Body surface lalandi EC, Vic Yes No C6233 & C6234 Naricolax chrysophryenus Nasal cavity* lalandi EC, SG Yes AB, PL C6240, C6248-C6298

Monogenea Benedenia seriolae Skin lalandi EC, Vic, SG Yes W, AB, PL AHC29102-29103 Skin hippos SA Yes NA AHC29189 Zeuxapta seriolae Gills lalandi EC, Vic, SG Yes W, AB, PL Rohde, 1981 Gills hippos EC Yes NA Rohde, 1978 Paramicrocotyloides reticularis Gills lalandi EC Yes No AHC29105 & 29106 Myxozoa Ceratomyxa seriolae Gall-bladder lalandi Vic, SG Yes W, AB, PL AHC34173 C. buri Gall-bladder lalandi Vic, SG Yes W, AB, PL AHC34174 Kudoa sp. Muscle lalandi EC Yes No Rohde, 1976 Unicapsula seriolae Muscle lalandi EC Yes No Lester, 1982 Nematoda (larva) Anisakis sp. Stomach, caeca lalandi SG Yes No AHC34189 Stomach hippos SA Yes NA AHC34262 Contracaecum sp. Caeca lalandi EC Yes No AHC34184 Stomach hippos SA Yes NA AHC34263 32 Hysterothylacium sp. Stomach, caeca lalandi EC, SG Yes No AHC34179-34183 Stomach hippos SA Yes NA AHC34265 Pseudoterranova sp. Intestine lalandi EC Yes No AHC34178 Rhabdochona sp. Stomach lalandi Vic Yes No AHC34191 (adult) Hysterothylacium sp. Stomach lalandi SG Yes No AHC34266 Trematoda Acanthocolpidae Stephanostomum petimba Digestive tract lalandi EC, SC, SG Yes No AHC29109 Tormopsolus orientalis Stomach, intestine lalandi EC, Vic, SG Yes PL AHC29110-29113 T. attenuatus Caeca hippos SA Yes NA AHC29145 Bucephalidae Bucephalus gorgon Digestive tract lalandi EC, SC, SG Yes W AHC AHC29114 & 29116 Caeca hippos SA Yes NA AHC29149 Rhipidocotyle longicirrus Stomach lalandi EC, SG Yes PL AHC29118 & 29119 Telorhynchus sp. Digestive tract lalandi Vic Yes No AHC29121 & 29122 Didymozoidae Undetermined species Viscera lalandi EC, Vic Yes No AHC29123 & AHC34174 Hemiuridae Dinurus longisinus Stomach lalandi EC Yes No Bray et al., 1993a Ectenurus trachuri Stomach lalandi EC Yes No Bray et al., 1993a

Erilepturus hamati Stomach hippos SA Yes No AHC29152 Elytrophallus sp. Stomach lalandi EC, SC, SG Yes No AHC29126 Stomach hippos SA Yes NA AHC29154 Elytrophalloides oatesi Stomach lalandi SC Yes No AHC29125 E. humerus Stomach lalandi SG Yes AB AHC29155 Hirudinella sp. Stomach lalandi SC Yes No AHC34176 Lecithocladium sp. Stomach lalandi EC Yes No AHC29127 Lecithaster stellatus Stomach lalandi EC Yes No Bray et al., 1993b Parahemiurus merus Stomach lalandi SC, SG Yes PL AHC29128 Plerurus digitatus Stomach lalandi EC Yes No AHC29129 Lecithasteridae Aponurus laguncula Stomach lalandi SC Yes No AHC29130 & 29131 Lepocreadidae Opechona kahawai Stomach lalandi SC Yes No AHC29132 & 29133 Sanguinicolidae Paradeontacylix godfreyi Heart lalandi SG, Vic Yes No AHC28904–28908 P. sanguinicoloides Heart lalandi EC Yes No AHC28909 Heart hippos SA Yes No AHC28910 Paradeontacylix sp. Heart lalandi Vic Yes No AHC28911 33 Heart hippos SA Yes No AHC28912

Parasites identified in the present study and documented previously from wild and farmed S. lalandi and wild S. hippos from the east coast of Australia (EC), Victoria (Vic), west coast of Australia (WC), South Australian waters other than Spencer Gulf (SA) and Spencer Gulf, South Australia (SG) are included. Where a parasite species has been detected on farmed S. lalandi in Fitzgerald Bay, Whyalla (W), Arno Bay (AB) or Boston Bay, Port Lincoln (PL) is indicated. Accession numbers are given for parasites in the Australian Helminth Collection (AHC) and Marine Invertebrate Collection (C) at the South Australian Museum (SAMA). *Denotes microhabitats indicated from previous studies that were undetermined in this survey; NA not applicable. Green highlight indicates specimens that have been submitted to the museum that await accession numbers.

Table 4. Consequence of parasite establishment and proliferation (Parasite species listed in alphabetical order).

Parasite taxa Potential mass mortality Pathology Marketability Consumer health Consequence Acanthocephala Australorhynchus tetramorphacanthus - - - - Negligible Rhadinorhynchus spp. - - - - Negligible Cestoda Callitetrarhynchus gracilis - - -* - Negligible Nybelinia thyrsites - unknown - - Negligible Tetraphyllideans Type 1 - unknown - - Negligible Type 4 - unknown - - Negligible Copepoda Caligus amblygenitalis - unknown - - Negligible C. epidemicus - X - - Low C. lalandei - - - - Negligible

34 C. spinosus - X - - Low Caligus sp. 1 - unknown - - Negligible Caligus sp. 2 - unknown - - Negligible Dissonus hoi - unknown - - Negligible Lepeophtheirus sp. - X - - Low Lernanthropus paenulatus - X - - Low Parapetalus spinosus - unknown - - Negligible Peniculus sp. - unknown - - Negligible Parabrachiella seriolae - unknown - - Negligible Parabrachiella sp. - unknown - - Negligible Naricolax sp. - unknown - - Negligible Monogenea Benedenia seriolae X X X - High Paramicrocotyloides reticularis - X* - - Low Zeuxapta seriolae X X - - Moderate Myxozoa Ceratomyxa seriolae - unknown - - Negligible

C. buri - unknown - - Negligible Kudoa sp. - - X - Low Unicapsula seriolae - - X - Low Nematoda Anisakis sp. (larvae) - X - -* Low Contracaecum sp. (larvae) - X - -* Low Hysterothylacium sp. (larvae & adult) - X - -* Low Pseudoterranova sp. (larvae) - X - -* Low Rhabdochona sp. (adult) - X - - Low Trematoda Bucephalus gorgon - unknown - - Negligible Dinurus longisinus - unknown - - Negligible Didymozoid - unknown - - Negligible Ectenurus trachuri - unknown - - Negligible Elytrophalloides oatesi - unknown - - Negligible

35 E. humerus - unknown - - Negligible

Elytrophallus sp. - unknown - - Negligible Erilepturus hamati - unknown - - Negligible Lecithocladium sp. - X - - Low Lecithaster stellatus - unknown - - Negligible Opechona kahawai - unknown - - Negligible Paradeontacylix godfreyi X X - - Moderate P. sanguinicoloides X X - - Moderate Paradeontacylix sp. X X - - Moderate Parahemiurus merus - unknown - - Negligible Rhipidocotyle longicirrus - unknown - - Negligible Stephanostomum petimba - X - - Low Telorhynchus sp. - unknown - - Negligible Tormopsolus attenuatus - unknown - - Negligible T. orientalis - unknown - - Negligible Parasites are scored for four criteria, (denoted with an X) including: 1) previous mass mortality in Seriola aquaculture, 2) potential parasite pathology, 3) potential negative impact on consumer acceptance and 4) potential negative impact on consumer health. *See Discussion for comment.

Table 5. Parasite risk analysis Likelihood of parasite establishment and proliferation including: 1) estimate of exposure of farmed fish to parasitised Seriola species and 2) biological pathway necessary for parasite species to infect the farmed fish species. Consequence of parasite establishment and proliferation, potential mitigation procedures and controlled risk shown. Parasite species shown in alphabetical order.

Parasite taxa 1) Exposure 2) Pathway Likelihood Consequence Ability to mitigate Controlled risk Acanthocephala Australorhynchus tetramorphacanthus Low Low Negligible Negligible Yes Negligible Rhadinorhynchus sp. 1 Negligible Low Negligible Negligible Yes Negligible Rhadinorhynchus sp. 2 Low Low Negligible Negligible Yes Negligible Cestoda Callitetrarhynchus gracilis Extreme Low Low Low Yes Negligible Nybelinia thyrsites Negligible Low Negligible Negligible Yes Negligible

36 Tetraphyllideans Type 1 High Low Low Negligible Yes Negligible

Type 4 Extreme Low Low Negligible Yes Negligible Copepoda Caligus amblygenitalis Negligible High Negligible Negligible Yes Negligible C. epidemicus High High High Low Yes Negligible C. lalandei Low High Low Negligible Yes Negligible C. spinosus Low High Low Low Yes Negligible Caligus sp. 1 Extreme High High Negligible Yes Negligible Caligus sp. 2 Low Negligible Negligible Negligible Yes Negligible Dissonus hoi Extreme High High Negligible Yes Negligible Lepeophtheirus sp. Low Negligible Negligible Low Yes Negligible Lernanthropus paenulatus High High High Low Yes Negligible Parapetalus spinosus Low Negligible Negligible Negligible Yes Negligible Peniculus sp. Low High Low Negligible Yes Negligible Parabrachiella seriolae High High High Negligible Yes Negligible Parabrachiella sp. High High High Negligible Yes Negligible Naricolax sp. High High High Negligible Yes Negligible

Monogenea Benedenia seriolae Extreme Extreme Extreme High Yes Low Paramicrocotyloides reticularis Negligible High Negligible Low Yes Negligible Zeuxapta seriolae Extreme Extreme Extreme Moderate Yes Low Myxozoa Ceratomyxa seriolae Extreme Moderate Moderate Negligible No Negligible C. buri Extreme Moderate Moderate Negligible No Negligible Kudoa sp. Negligible* Moderate Negligible* Low No Low Unicapsula seriolae Negligible* Moderate Negligible* Low No Low Nematoda Anisakis sp. High Low Low Low Yes Negligible Contracaecum sp. Low Low Negligible Low Yes Negligible Hysterothylacium sp. (larvae and adult) High Low Low Low Yes Negligible Pseudoterranova sp. Negligible Low Negligible Low Yes Negligible Rhabdochona sp. Low Low Negligible Low Yes Negligible

Trematoda Bucephalus gorgon Extreme Low Low Negligible Yes Negligible Dinurus longisinus Negligible Low Negligible Negligible Yes Negligible Didymozoid Low Low Negligible Negligible Yes Negligible Ectenurus trachuri Negligible Low Negligible Negligible Yes Negligible Ellytrophalloides humerus Extreme Low Low Negligible Yes Negligible E. oatesi Low Low Negligible Negligible Yes Negligible Ellytrophallus sp. High Low Low Negligible Yes Negligible Erilepturus hamati Low Low Negligible Negligible Yes Negligible Lecithocladium sp. Negligible Low Negligible Negligible Yes Negligible Lecithaster stellatus Negligible Low Negligible Negligible Yes Negligible Opechona kahawai Low Low Negligible Negligible Yes Negligible Paradeontacylix godfreyi High Moderate Moderate Moderate No Moderate P. sanguinicoloides Low Moderate Negligible Moderate No Moderate Paradeontacylix sp. Low Moderate Negligible Moderate No Moderate Parahemiurus merus Extreme Low Low Negligible Yes Negligible 38 Rhipidocotyle longicirrus High Low Low Negligible Yes Negligible Stephanostomum petimba High Low Low Low Yes Negligible Telorhynchus sp. Low Low Negligible Negligible Yes Negligible Tormopsolus attenuatus Low Negligible Negligible Negligible Yes Negligible T. orientalis High Low Low Negligible Yes Negligible

*See Discussion for comment.

of establishment and proliferation. Six of the copepod species presented a high likelihood of establishment and proliferation. The consequence of parasite establishment and proliferation ranged from negligible to high (Table 4). Consequence was high for B. seriolae and moderate for Z. seriolae and three Paradeontacylix spp. Total parasite risk ranged from negligible to high (Table 6). In the mitigated assessment, we estimated that Kudoa sp., Unicapsula seriolae and Paradeontacylix spp., for which there are no current mitigation procedures, pose the greatest risk to S. lalandi aquaculture in Australia (Table 5). The effectiveness of in-feed chemical treatments (e.g. praziquantel and fenbendazole) have not been quantified for these parasite species.

Each parasite group is discussed separately below, beginning with parasite groups that have direct life-cycles. We identify routes of transfer that were considered when assessing the likelihood of parasite establishment and justify the consequence category assigned to each parasite group or species (Table 4). We also discuss current and potential parasite management practices in kingfish aquaculture, which we used to estimate controlled risks (Table 5). When considering this risk assessment, it is important to note that the likelihood of parasite establishment is dependent upon current information available on parasite presence and distribution. Despite a thorough parasite sampling technique, it is possible that some parasite species may not have been detected, especially if they occur seasonally or exist in low (<10%) prevalence in wild or farmed kingfish populations.

1.5.1. Copepods Caligid copepods have direct life-cycles consisting of free-living, free-swimming and attached parasitic stages. Severe ectoparasitic copepod infestations in aquaculture have been associated with mortalities through host osmoregulatory failure, anaemia, ulcerations or through facilitating secondary infections (Finstad et al. 2000). Caligus spinosus, which we detected on wild and farmed kingfish (Table 3), has been associated with gill disease in farmed Japanese S. quinqueradiata, where serious infestations result in anaemia (Egusa 1983). Affected fish may also become emaciated due to appetite depression, rub against the sea-cage and develop ulcerations around the mouth (Egusa 1983; Ho et al. 2001). Caligus epidemicus,

which was detected on wild kingfish in Spencer Gulf (Table 3), is known to parasitise a number of wild and farmed marine fish species in Australia and Asia (Ho et al. 2004; Johnson et al. 2004). Ho et al. (2004) suggest this species presents a threat to aquaculture because of its low host-specificity. Considering the known pathology of C. spinosus and C. epidemicus in aquaculture, we considered that they have a low consequence for kingfish aquaculture in Australia (Table 4).

Caligus lalandei, which is well known from farmed Seriola species in Japan (Ho et al. 2001) and New Zealand (Diggles and Hutson 2005), was not recovered from farmed kingfish, despite being found on wild kingfish in Victoria and on the east coast of Australia (Table 3). Interestingly, C. lalandei has not been associated with disease in aquaculture, although Ho et al. (2001) suggest it may cause a serious problem in the event of an outbreak because of its large size. However, considering that there is no known pathology associated with C. lalandei infections, we estimated that they have a negligible consequence for kingfish aquaculture in Australia (Table 4).

We detected Lepeoptheirus sp. on Samson fish and Lernanthropus paenulatus on wild kingfish and Samson fish, although these species are not known to be pathogenic in Seriola spp. aquaculture. However, lacerated tissue, erosion, desquamation and necrosis of secondary gill lamellae have been noticed near the site of attachment of L. kroyeri to sea bass, Dicentrarchus labrax, farmed in sea-cages in Greece (Manera and Dezfuli 2003). Loss of D. labrax condition was associated with L. kroyeri infection. Similarly, Lepeoptheirus salmonis has been associated with salmon mortalities throughout the northern hemisphere (Costello 1993) and management of that parasite contributes significantly to the cost of rearing salmon where it is found. Considering the pathology of species belonging to these two genera in aquaculture elsewhere, we considered that they have a low consequence for kingfish aquaculture (Table 5). However, it is important to note that Lepeoptheirus sp. is only known from Samson fish (Table 3), resulting in a negligible likelihood of this species establishing in kingfish farms (Table 5).

Crustaceans can be treated with hydrogen peroxide by bath treatment. In Japan, eradication of C. spinosus has been achieved by immersing S. quinqueradiata in

seawater containing Trichlorfon (Fujita et al. 1968). In the northern hemisphere, knowledge of parasite biology and epidemiology has assisted in the effective management of parasites on farms and has helped to minimise parasite numbers. For example, research of the population dynamics of salmon lice from a number of selected sites has demonstrated the value of fallowing and the effectiveness of single year-class stocking, as opposed to multiyear-class stocking of farm sites (Bron et al. 1993).

Crustacean species could be managed in the event of an outbreak using hydrogen peroxide bathing. Four copepod parasite species were found to present a low consequence for kingfish farming (Table 4). However, these species can be potentially managed and were estimated to pose a negligible consequence for kingfish farming in Australia in the controlled risk assessment (Table 5).

1.5.2. Monogeneans The capsalid Benedenia seriolae and the heteraxinid Zeuxapta seriolae are common pathogens of farmed Seriola spp. and have been associated with considerable production losses in Seriola aquaculture in Japan (Ogawa 1996), Australia (Ernst et al. 2002; Whittington et al. 2001) and New Zealand (Diggles and Hutson 2005) due to increased FCR, decreased growth performance and mortality. These species occur on wild and farmed kingfish and wild Samson fish in Australia (Table 3). We did not detect the microcotylid Paramicrocotyloides reticularis on farmed or wild S. lalandi in South Australia, but it has previously been documented from wild S. lalandi on the east coast of Australia and New Zealand (Diggles and Hutson 2005; Rohde 1978) (Table 3).

Hydrogen peroxide has been used effectively in the treatment of monogeneans (Rach et al. 2000) and is the South Australian kingfish industry‟s current treatment of choice for control of Z. seriolae and B. seriolae (see Mansell et al. 2005). Cycles of reinfection can also be prevented if treatments are coordinated strategically to break the parasite‟s life-cycle (Ernst et al. 2005). However, treating fish for monogenean infections by this method is labour intensive and costly. If left untreated, high numbers of B. seriolae on the body surface may render fish unappealing to consumers (Table 4), but this can be overcome by removing the parasites before sale. Therefore,

the controlled risk of B. seriolae was lowered by two consequence factors (i.e. from high to low). Z. seriolae was also estimated to pose a low consequence for kingfish farming in Australia in the controlled risk assessment (Table 5).

Paramicrocotyloides reticularis is currently not present in kingfish aquaculture in South Australia. Little is known about the biology of P. reticularis, but it likely exhibits similar biology to Z. seriolae (i.e. infects gills and feeds on blood). We considered that P. reticularis might exhibit high fecundity, given the biology of related organisms and based on the number of eggs observed in the uterus. For the purposes of this risk assessment, we propose that P. reticularis may have a similar impact on host health as Z. seriolae and be amenable to control via the treatments mentioned above (Table 5). We estimated that Paramicrocotyloides reticularis poses a negligible risk of establishment and proliferation to kingfish farmed in South Australia (Table 5) because it was not detected on wild fish in the same region (Table 3). P. reticularis would present a higher risk if the industry were to develop kingfish farms on the east coast of Australia where the parasite does occur in wild fish.

1.5.3. Acanthocephalans We found two Rhadinorhynchus spp. infecting kingfish in Victoria and NSW (Table 4). Rhadinorhynchus spp. were not embedded in the tissues, which is consistent with Costa et al. (2004) who observed R. pristis free in the intestine of Scomber japonicus. Australorhynchus tetramorphacanthus has been previously recorded from the Great Australian Bight and Tasman Sea (Lebedev 1967). Although we are unaware of any current management practice for controlling acanthocephalan infection in fish, cotton- top tamarins (Saguinus oedipus) have been treated successfully for acanthocephalans with oral albendazole (Weber and Junge 2000). It is not clear whether this would be successful for fish species or if this could be undertaken in fish used in food production.

There are few reports of acanthocephalans in farmed finfish. It is unlikely that acanthocephalans could establish and proliferate in South Australian kingfish aquaculture because of limited interaction with infected intermediate hosts. It also appears as though acanthocephalan species previously detected in wild kingfish in Australian waters may not negatively impact upon host well being, given that they are

not known to embed in the tissue. Albendazole may be a potential chemotherapeutic treatment for these parasites, but it is not currently registered or issued a permit for use in Australia. Nevertheless, establishment and proliferation of these parasites in aquaculture is unlikely as they are transferred when infected intermediate hosts are eaten by the definitive host. Therefore, we estimated the consequence of these two acanthocephalan species to be negligible for kingfish sea-cage aquaculture (Table 5).

1.5.4. Cestodes Cestodes transfer to piscivorous fish when infected intermediate hosts are eaten. We detected immature larval Callitetrarhynchus gracilis encysted in the body cavity and viscera of wild kingfish. Cysts, suspected to be cestode blastocysts, were observed in the viscera of one farmed kingfish from Whyalla and Port Lincoln and in wild Samson fish, but did not contain any cestode larva. It is not clear whether Callitetrarhynchus blastocysts are associated with a pathological host response. Adjei et al. (1986) found blastocysts containing C. gracilis in lizard fish (Saurida tumbil and S. undosquamis) adjacent to the ventral aorta and in the body cavity, but did not observe any associated necrotic tissue. We found no literature concerning potential pathology of larval tetraphyllideans and the trypanorhynch Nybelinia thyrsites documented from the digestive tract of kingfish (Table 3).

In Japan, farmed Seriola quinqueradiata fed parasitised raw fish became infected with a larval cestode, C. nipponica, which altered the appearance and reduced the marketability of the flesh (Ogawa 1996). However, when raw fish was replaced with frozen food, the parasite disappeared from farm sites. Currently, all kingfish farms in South Australia use extruded feed, a practice that contributes to the negligible likelihood of establishment of C. gracilis as estimated in the risk assessment (Table 5). Neverthless, farmed fish can be infected by cestodes, given that we found cestode cysts (presumably C. gracilis blastocysts) in the viscera.

Since we did not find any blastocysts containing C. gracilis in the flesh, this parasite is unlikely to have any impact on the marketability of kingfish (Table 4). The risk of parasite establishment and proliferation can be minimised by maintaining an extruded pellet diet. This would also reduce the potential of infection of larval tetraphyllideans

and the trypanorhynch Nybelinia thyrsites. Consequently, we estimated that cestode species discussed here present a negligible consequence for kingfish farming in Australia (Table 4).

1.5.5. Myxozoans Myxozoa are now recognised as cnidarians and most are believed to have a two host life-cycle involving fish and invertebrates (Moran et al. 1999). We recovered Ceratomyxa seriolae and C. buri from the gall-bladder of wild and farmed kingfish (Table 3). These species and Myxobolus spirosulcatus (see Maeno et al. 1995) are also known from the gall-bladder of farmed S. quinqueradiata in Japan (Yokoyama and Fukuda 2001), but there have been no apparent pathological changes or mortality associated with infection. It is believed that myxozoan parasites in the gall-bladder may cause discolouration of the liver, by blocking the normal flow of bile from the bile ducts to the gall-bladder (Egusa 1983). However, it has also been suggested that liver discolouration is related to the quality of vegetable protein used in extruded feed (Sheppard 2004).

Kudoa sp. and Unicapsula seriolae have been detected previously in the flesh of wild kingfish on the east coast of Australia (Lester 1982; Rohde 1976) (Table 3). In Japan, farmed S. quinqueradiata are infected with similar myxosporean parasites, Kudoa pericardialis and K. amamiensis (see Egusa 1983; Moran et al. 1999). Although these infections are apparently not associated with mortality (Egusa 1983) they can have detrimental effects on product quality and consumer acceptance. Infections of myxosporeans in the flesh have been associated with large, unsightly cysts or regions of lysis within the musculature (e.g. K. amamiensis in S. quinqueradiata, see Moran et al., 1999) and/or accelerated muscle degeneration and post-mortem myoliquefaction (e.g. U. seriolae in kingfish, see Lester, 1982).

Moran et al. (1999) discussed potential strategies for controlling diseases induced by myxosporeans. There are currently no chemotherapeutic treatments available, while avoiding host exposure in sea-cages appears difficult to impossible. They suggest frequent net changing may reduce accumulation of potential intermediate hosts and therefore reduce the risk of exposure to infective stages of the parasite. However, we

are unaware of any documented evidence that indicates that this can effectively manage or control myxosporean infections.

Given current information available on the distribution of Kudoa sp. and U. seriolae in Australia, these species present a negligible likelihood of establishment and proliferation in kingfish aquaculture in South Australia (Table 5). However, it is important to note that we are aware of anecdotal reports of myoliquefaction in the flesh of wild kingfish in Spencer Gulf, South Australia. Farmed Thunnus maccoyii in Spencer Gulf (Deveney et al. 2005) and wild Samson fish in Western Australia (Andrew Rowland, pers. comm.) are also known to experience myxosporean infections in flesh. Considering the potential reduction in market value and negative consumer acceptance for infected fish, we found that these species present a low consequence for kingfish aquaculture (Table 5). Clearly, Kudoa sp. and U. seriolae are very important parasite species to consider if the industry were to develop on the northern east coast of Australia where myxosporean infection is common in wild kingfish.

1.5.6. Nematodes We noted evidence of granuloma formation associated with larval nematodes in the viscera of wild kingfish in Victoria. Nematode migration and encapsulation within body tissues and visceral organs often cause the development of lesions (Dezfuli et al. 2000). Larval nematodes present a negligible likelihood of establishment and proliferation in kingfish sea-cage farming because of the current farming practice of using an extruded pellet diet (Table 5). Given that these parasites were detected in the viscera and not in the flesh, the risk of human consumption is substantially reduced. Consequently, we did not consider that these parasites present a negative consequence for consumer health or marketability (Table 4). Nevertheless, nematodes still pose a low consequence for kingfish aquaculture because of the potential for harm to the host (Table 4). Although there are no registered anthelminthics in Australia that can be used to treat nematodes in fish destined for human consumption, nematode parasites could be managed by maintaining an extruded pellet diet. Consequently, we estimated that the nematode species documented here present a negligible consequence for kingfish farming in Australia in the controlled risk assessment (Table 5).

1.5.7. Trematodes We found Parahemiurus merus, Rhipidocotyle longicirrus and Tormopsolus orientalis in farmed kingfish in Boston Bay, Port Lincoln, which may be a result of fish being fed a raw/frozen pilchard diet at the time of sampling (Table 3). At Fitzgerald Bay, Whyalla, where kingfish were fed an extruded pellet diet exclusively, specimens of Bucephalus gorgon were detected (Table 3). This indicates that farmed fish may consume some wild, infected, intermediate hosts. Trematode parasites can be easily managed in farms by feeding fish with an uninfected diet. Some farmed fish may still become infected by trematodes by feeding opportunistically on infected wild species moving through the netting, but it is unlikely that these parasite species will be able to proliferate in the farmed population. Given the lack of information available in the literature concerning the relative pathogenicity of bucephalids and hemiurids, which likely stems from the absence of apparent pathology, we could not estimate the consequence of these families for kingfish sea-cage farming (Table 4).

Sanguinicolids have been problematic in aquaculture because their intermediate mollusc or annelid host may inhabit areas close to farmed fish, such as on cage structures or sediment, and infection of the definitive host by emerging cercariae is direct. The likelihood of sanguinicolid establishment and proliferation was estimated to be moderate for Paradeontacylix godfreyi and negligible for Paradeontacylix sp. and P. sanguinicoloides (Table 5). Although the latter two species are only known from Samson fish in South Australian waters, both are known to infect kingfish elsewhere in Australia (Hutson and Whittington 2006). The likelihood of these species establishing in kingfish aquaculture should not be underestimated, as they infect at low prevalence and intensities in wild hosts and the extent of their current range is unknown.

Sanguinicolids in Paradeontacylix have been associated with mass mortalities of farmed amberjack, S. dumerili, in the Spanish Mediterranean (Crespo et al. 1992) and in Japan (Ogawa and Fukudome 1994). They are also of concern to kingfish farming in New Zealand where Paradeontacylix-like blood flukes have been detected in histological sections of the heart, brain and internal organs and have been associated with low-level mortalities (Diggles and Hutson 2005). We did not detect sanguinicolids in farmed kingfish in South Australia (Table 3).

Control of blood fluke infections may only be achieved in semi-open aquaculture systems by separating intermediate and definitive hosts, as elimination of susceptible intermediate hosts in open water is impractical and cost-prohibitive (Bullard and Overstreet 2002). We are aware that some farms in Japan use orally delivered praziquantel for treatment of Seriola spp. infected with blood fluke, however, np published data exists on the efficacy of this treatment. Identifying the intermediate host or hosts for Paradeontacylix spp. would help to determine suitable sea-cage sites for S. lalandi away from potential infection sources as the industry expands. However, the intermediate host(s) is currently unknown. Consequently, we estimated that Paradeontacylix spp. present a moderate consequence for kingfish farming in Australia in the controlled risk assessment (Table 5).

OBJECTIVE 2: DISCRIMINATION OF FARMED AND WILD FISH

METHODS

2.1. Natural elemental signatures

Gillanders and Joyce (2005) used natural elemental signatures in the otolith to distinguish between hatchery reared and wild populations of kingfish. This work constituted part of this research project.

2.2. Rare earth elements

Two experiments were conducted to determine whether otoliths of hatchery fingerlings could be marked with trace elements. The first involved exposing fish to three concentration of lanthanum (administered as lanthanum chloride) for three different time periods while the second involved exposing fish to various combinations of lanthanum, cerium and caesium for a single time period in order to batch mark otoliths. Uptake of lanthanum, cerium and caesium was high enough to enable juvenile hatchery-reared kingfish to be distinguished from farmed and wild caught kingfish via artificial elemental signatures in otoliths. However, uptake of lanthanum, cerium and caesium did not appear to increase with immersion period or concentration contrary to previous studies. This may be due to analytical technique, size of otolith or precipitation of elements. The specific details of these research results can be obtained in: Joyce, T.C. (2003) Distinguishing aquaculture and wild yellowtail kingfish via natural and aritifical elemental signatures in otolith. Honours thesis, School of Earth and Environmental Sciences, The University of Adelaide, South Australia Australia.

2.3. Chemical dye of hatchery derived kingfish

2.3.1. Fish maintenance and chemical treatments Juvenile kingfish (~ 80mm TL) were obtained from Spencer Gulf Aquaculture hatchery and stored in a 2000L tank. Fish were held for 12 days before being exposed to experimental conditions. All seawater used during experimental and rearing stages was collected twice weekly from the South Australian Aquatic Sciences Centre, South Australian Research and Development Institute (SARDI), West Beach, Adelaide.

Experiments were performed in a temperature-controlled room maintained at 23ºC with a photoperiod of 12 h of light and 12 h of dark. Thirty-eight 40 L tanks were bleached, acid washed and rinsed several times to minimise contamination: they were then filled to 40 L with seawater (salinity 32 ppm) and aerated. Three concentrations (high, medium and low) of each of the chemical dyes were added to individual tanks (Table 6). Kingfish were immersed in three concentrations of each chemical dye: Alizarin Complexone (AC), Alizarin Red S (ARS) and OCT for 6 and 24 h durations. Two replicates of each compound, concentration and time combination were tested (Table 6). In addition, two replicate tanks were left untreated and used as procedural controls. All treatments were randomly assigned throughout the controlled temperature room.

Table 6. Concentrations of chemical dyes for 6 and 24 h durations

Chemical Dye Concentration AC 10 ppm AC 25 ppm AC 40 ppm ARS 40 ppm ARS 100 ppm ARS 250 ppm OTC 40 ppm OTC 250 ppm OTC 100 ppm Control No treatment

(AC = Alizarin Complexone, ARS = Alizarin Red S, OTC = Oxytetracycline)

Prior to exposing fish to treatments, fish guts were purged by withholding food for 48 h to avoid fouling of water during the treatment. Total length and wet weight of 20 randomly selected fish were taken to represent the experimental fish. Ten fish were added to each of the 40 L treatment tanks. At the conclusion of chemical exposures, fish were transferred to individual holding tanks. As kingfish otoliths are slow growing, fish were reared for a further 25 days in seawater (32 ppm) to allow post- mark growth so that any chemical mark could be distinguished from autofluorescence on the otolith edge. Half water changes were conducted every second day. Fish were fed twice daily with SkrettingTM 2 mm dry pellet. On completion of the rearing period, fish were killed and frozen.

2.3.2 Otolith preparation and analysis Prior to fish dissection, the total length of fish was measured and random selections of 20 fish were weighed to represent the mean weight of experimental fish. Sagittal otoliths were dissected and removed. Otoliths were cleaned of adhering tissue in deionised water and air dried in microcentrifuge tubes in a laminar flow cabinet to reduce air-borne contamination. A single otolith from each fish was set in epoxy resin (Struers epofix). Transverse sections of approximately 200 m thick were taken through the primordium (central section) of each otolith, using a low speed diamond saw. Sections were polished using 9 m and 3 m lapping film. Polished sections were fixed to microscope slides using crystal bond (thermoplastic glue) and placed into clean sealable bags to await analyses. Otoliths were examined for a chemical mark using a Leica DMLB light microscope (40  magnification) equipped with a mercury lamp (ebq 100) producing ultraviolet light. Filter sets were attached to the microscope to detect the presence of a chemical mark on the otolith. The three filters used were: an excitation filter – Band Pass 355-425nm; dichromatic mirror - 455nm; suppression filter – Long Pass 470 nm which is used primarily for the detection of OTC, a Band Pass 546/12 nm; 565n m, and a Band Pass P 600/40 nm both used for detecting similar wavelengths to AC and ARS.

2.3.3. Statistical analysis One-way ANOVAs were performed on the standard length of fish in each treatment compared with the standard length of fish in control treatments to determine whether growth of fish was affected by treatments. Both percentage marked (%M) and mark quality was assessed. Percentage mark was calculated by the number of individuals found with a mark divided by the total number of fish in each treatment  100. Mark quality was determined by classifying the chemical mark on each otolith into one of four grades Largadere et al. (2000); absent (0), uncertain (1), good (2) or very good (3). A mark was classified as „absent‟ if there was no visible signs of a chemical mark, as an „uncertain mark‟ if the mark was barely visible, thin or obscured by edge effects and autofluorescence, as a „good mark‟ if the chemical complex was highly visible but did not lay down a perfect incremental ring, and as a „very good mark‟ if the mark was imprinted as a full increment around the otolith with no obstructions. From these, three independent sets of readings were made (by Justin E. Rowntree, Melita de Vries and Dr Andrew Munro, The University of Adelaide) and the grades were then averaged, and the average value for each otolith was used in statistical analysis. Mann-Whitney U tests were used to detect significant differences in mark quality among treatment groups.

RESULTS/DISCUSSION

2.4. Mark success

The differences between initial weight and length and final weight and length was highly significant (wt: F = 184.81, P = 0.000, df = 59) (length: F = 298.21, P = 0.000 df =59) indicating that fish grew under laboratory conditions in all treatments. There was no influence of treatments on the growth of fish for any treatment concentration or immersion duration. The post treatment survival for each of the three chemical complexes (not including fish that died due to equipment failure) was AC 98.8%, ARS 92.8%, and OTC 96.2%.

AC had the greatest mark success with 100% of fish receiving a good or very good mark in all treatments, marking durations (six and 24 h) and concentrations (10, 25, and 40 ppm). ARS had the second best mark success with middle to high concentrations (100 and 250 ppm, respectively) yielding 86-100% mark success. There was only a minor difference in mark success between marking durations (six and 24 h) in both medium (100 ppm) and high (250 ppm) concentration treatments of ARS. Mark success in low concentration treatments of AR (40ppm) was average to low with 29-55% of fish successfully marked. Oxytetracycline marks were the most variable and least successful mark of all chemical complexes. No mark was seen on fish immersed in low concentrations of oxytetracycline (40 ppm) for both six and 24 h durations. Medium (250 ppm) and high (500 ppm) concentrations of oxytetracycline had a low to moderate mark success (0-43%) at both six and 24 h immersions.

The top three rank sum scores determined by the Mann-Whitney U tests from all treatments (n=24) were AC treatments, AC25 (24 h immersion) (RS = 2792), both AC10 (24 h) and AC40 (6 h immersion) (RS = 2711) and AC25 (6 h) (RS = 2387). Each of the remaining AC treatments scored rank sum scores in the top seven of all treatments (Table 7). Alizarin Red S 250 scored highly in two treatments ARS250, 24h (RS = 2306) and ARS250, 6h (RS = 2220), and were the only other treatments to compare in mark quality with AC treatments. Oxytetracycline had the lowest of all rank sum scores with only one treatment (OTC500,24h) obtaining a rank score in the

top nine from all 24 treatments (Table 7). Each treatment that scored a rank score in the top eight of all treatments had only good and very good mark quality on the otolith. Treatments that received rank sum scores beyond the top eight treatments had variable chemical mark on otoliths ranging from good to absent (Table 7).

Table 7. Rank sum scores determined by Mann-Whitney U tests

No. of fish Compound Concentration Time Rank Sum Rank score % M survived AC 25 24 14 2792.000 1 100 AC 10 24 14 2711.000 2 100 AC 40 6 14 2711.000 2 100 AC 25 6 14 2387.000 3 100 ARS 250 24 13 2306.000 4 100 AC 10 6 14 2225.000 5 100 ARS 250 6 14 2220.500 6 93 AC 40 24 12 1803.000 7 100 OTC 500 24 12 1789.500 8 75 ARS 100 24 14 1730.000 9 86 ARS 100 6 14 1730.000 9 100 OTC 250 24 12 1042.500 10 42 ARS 40 24 11 998.000 11 55 ARS 40 6 14 965.000 12 29 OTC 500 6 14 965.000 12 29 OTC 40 24 14 623.000 13 0 Control Control Control 14 623.000 13 0 OTC 40 6 12 534.000 14 0 OTC 250 6 10 530.500 15 14

A high score represents otoliths with the greatest mark quality. %M indicates the number of fish that were successfully marked in each treatment. Kruskal-Wallis Test Statistic =168.995; Probability is 0 assuming Chi-square distribution with 18 df.

The three chemical dyes compared in this study, alizarin complexone (AC), alizarin red S (ARS) and oxytetracycline (OTC), each produced marks on otoliths of kingfish with varying success. Alizarin complexone produced a high mark quality for each concentration (10, 25, 40 ppm) and immersion time (6 and 24 h). In addition, marking success and post treatment survival were both 100%, for all AC treatments. These results are consistent with the findings of Lagardere et al. (e.g. 2000) who reported 100% marking success for large juvenile turbot for immersions ranging from 25-200 mg/L for 6-24 h with no effect on growth. Despite the success of AC, its use may have limited practicality given its expense ($2,700 AUS for 100 g). Chemicals

used for marking fish may also need to be registered with AP and Food Standards ANZ which may require further testing to ensure chemicals are not retained in fish flesh being used for sale.

Although fewer studies have considered the use of the less expensive ARS ($246.50 AUS for 100g) for marking fish otoliths, ARS immersion appears to be a promising technique for marking the otoliths of small fish (Beckman and Schulz 1996). The present study found that high concentration of ARS (250 mg/L) produced a mark equal in quality to AC treatments for both six and 24 h immersions with a 100% mark success and post treatment survival for 24 h immersions and 93% for six h treatments. Lagardere et al. (2000) also found that high concentrations of ARS (200-400 mg/L) produced good quality marks in the otoliths of juvenile turbot with no mortality or cessation of growth. In addition, they found that lower concentrations of ARS (e.g. 100 mg/L) did not produce a clear mark in their otoliths. Similarly, we found that low concentrations of ARS (40 mg/L) produced a marking success rate of only 29% for both six and 24 h immersions. Contrary to Lagandere et al.‟s finding, in our study, ARS concentrations of 100 mg/L were successful at producing marks on most individuals (86-100% mark success), but 24 h immersions suffered high post treatment mortality (35%). The cause of the mortalities could not be established.

Oxytetracycline was the least successful chemical dye for marking otoliths of kingfish. Marking success using OTC ranged from 0-42%. Although high concentrations (500mg/L) and 24 h immersion times of OTC were found to produce good and very good marks on the otolith, no marks were apparent within the same treatments. Brothers (1990) found that the marking of fish otoliths using OTC was not effective for all marine species and suggested that using low concentrations of OTC to mark fish might not be effective, as it binds with calcium and magnesium in seawater before being incorporated when fish are immersed. Hernaman et al. (2000) suggested that the failure to identify clear marks using OTC may arise due to one of three reasons: 1) the chemical was not incorporated into the otolith; 2) the chemical was incorporated into the otolith, but it was not possible to distinguish between treatment marks and edge autofluorescence or 3) the chemical was incorporated into the otolith but the mark was lost over time.

To date, two other methods for distinguishing between wild and hatchery reared yellowtail kingfish using natural tags have been practiced. Gillanders and Joyce (2005) used natural elemental signatures in the otolith to distinguish between hatchery reared and wild populations of yellowtail kingfish. Although significant differences between hatchery-reared fish and wild fish signatures were established, the differences depend on heterogeneity of water chemistry, salinity and temperature, which may not vary among close populations (Elsdon and Gillanders 2003). Fowler et al. (2003) used morphometric measurements to distinguish between reared and wild yellowtail kingfish otoliths; however determining such differences requires a great degree of time and expertise. In the present study, we have established a method to mark kingfish otoliths using chemical dyes. This method is cost effective, requires minimal handling, has little impact on fish growth and mortality and creates an unambiguous mark in the fish otolith. Before this method could be used at the hatchery further research would be required. These methods require fish to be immersed in solution. Although fish may be marked in baths, they still require some degree of handling, which may be stressful to fish. Further methods which rely on a natural tag would therefore be beneficial.

OBJECTIVE 3: ASSESSMENT OF MIGRATORY BEHAVIOUR OF WILD KINGFISH

An abridged version of Objective 3 has been published in Transactions of the Royal Society of South Australia.

Hutson, K.S., Smith, B.P., Godfrey, R.T., Whittington, I.D., Chambers, C.B., Ernst, I. and Gillanders, B.M. (2007). A tagging study on yellowtail kingfish (Seriola lalandi) and Samson fish (S. hippos) in South Australian waters. Transactions of the Royal Society of South Australia 131, 128-134.

METHODS

3.1. Conventional tagging programmes and participation

A tagging programme for Seriola spp. in South Australia was conducted in conjunction with the South Australian branch of the Australian National Sportfishing Association‟s tagging programme, called „Saftag‟. Tagging data were provided to Saftag for inclusion in their locally administered database. We sought additional tag records from NSW Fisheries because game fishing clubs in South Australia commonly use NSW Fisheries tags. All existing tag and recapture records for Seriola spp. from Saftag and NSW Fisheries databases for South Australian waters were consolidated.

3.2. Fish capture

Fish were captured by line at Arno Bay and Port Augusta in Spencer Gulf (Figure 2). Four charter operators fishing offshore from Port Lincoln participated in the tagging programme and captured, tagged and released fish at Yatala Reef, Greenly Island and Rocky Island in 2005 (Figure 2). In addition, small schools of kingfish were netted in northern Spencer Gulf in October 2005 and 2006 (under Primary Industries and Resources South Australia exemption no. 9901854). When a fish was landed, a nylon

headed, single-barbed dart tag was inserted into the muscle adjacent to the dorsal fin at a 45º angle, so that the barb on the tag would lock into the pterygiophores. Total Length (TL) was measured to the nearest mm, where possible.

Figure 2. Tag and release localities and movement of Seriola spp.

(Yatala Reef, west of Streaky Bay not shown).

RESULTS/DISCUSSION

A total of 248 kingfish and 73 Samson fish specimens were tagged in this study (Table 8). Seriola spp. were tagged at 15 South Australian locations (Figure 2). Eleven percent of recaptured fish showed movements of more than 5 km (Figure 2).

Table 8. Total tag recordings for Seriola lalandi and S. hippos captured in South Australia (compiled from all available data, February 1991 to December 31, 2006).

Species S. lalandi S. hippos

Total no. tagged 335 88

Mean total length in mm (range) 502 (320-1480) 1100 (910-1430)

Total no. recaptured (line & net) 27 (25 & 2) 2 (2 & 0)

Min and max days at liberty 0 – 442 302 – 378

Number released after recapture 7 1

Percent recapture (line & net) 8.0% (7.5% & 0.6%) 2.3 % (2.3% & 0%)

* Includes fish recovered by line only to date (Dec 2006) to indicate recreational fishing pressure

A total of 335 S. lalandi and 88 S. hippos specimens has been tagged between February 1991 and December 31, 2006, in South Australia (Table 8) at 15 locations (Figure 2). Of this total, we tagged and released 248 S. lalandi and 73 S. hippos specimens for this study with the assistance of charter operators and recreational fishers. Before this tag programme, only 57 S. lalandi and 1 S. hippos had ever been tagged and released in South Australian waters since the first record in 1991. In the current study, 87 large S. lalandi (>1000 mm TL) were tagged near Port Augusta in northern Spencer Gulf (Table 9). Approximately 8% of S. lalandi we tagged were recaptured across South Australia after being at liberty for between 0 to 442 days (Table 8 and 9).

Table 8. Seriola lalandi >1000 mm TL tagged at Port Augusta during 2005 and 2006

Year 2005 2006 Total

Total no. tagged 39 48 87

Mean total length in mm (range) 1212 1215 1180

(1110–1480) (1100–1250) (1110–1480)

Total no. recaptured (line and net) 3 (1 and 2) 6 (6 and 0) 9 (7 and 2)

Min and max days at liberty 8 – 442 46 – 98 8 – 442

Number released after recapture 3 2 5

Percent recapture (line and net) 10.3%

(8.0% and 2.3%)

* Includes fish recovered by line only to indicate recreational fishing pressure

3.3. Movements

Four recaptured fish showed movements >5 km (Figure 2). Two small S. lalandi (400-500 mm TL) tagged at Arno Bay (33º55′21″S, 136º36′14″E) were recaptured approximately 100 km southeast after 39 days at Wardang Island (34º27′35″S, 137º23′15″E) and approximately 130 km northeast after 49 days at Port Broughton (33º21′35″S, 137º33′21″E) (Figure 1). One small S. lalandi (550 mm TL) tagged at Whyalla in 2002 was recaptured 51 days later, 40 km north, towards Port Augusta (Figure 2). A larger individual (1110 mm TL) tagged at Port Augusta (32º 42′04″S, 137º46′17″E) moved approximately 50 km south to Point Lowly, near Fitzgerald Bay (32º 24′14″S, 137º19′16″E) and was at liberty for 46 days. Two S. lalandi (950 and 870 mm TL) tagged at Rocky Island were recaptured 0 and 18 days later, respectively, at Rocky Island. Twenty-one S. lalandi tagged at Port Augusta were recaptured at Port Augusta, including four individuals (>1000 mm TL) tagged in May 2005 or May 2006 and recaptured in the same area eight, 88, 98 and 149 days after release. A further four large fish were recaptured at Port Augusta following 326, 328, 383 and

442 days at liberty. The only two recaptures of S. hippos (910 and 1070 mm TL) tagged at Rocky Island were made at the same location 302 and 378 days later.

We observed S. lalandi schools during October 2005 and October 2006 while netting in shallow water (1 to 4 m depth) in northern Spencer Gulf. Large S. lalandi (>1000 mm) were observed alone, in pairs and in small to large schools containing from 10 to >200 individuals. Schools of large fish contained individuals of various sizes (1110 to 1480 mm TL). One fish originally captured on rod and reel (1220 mm TL) was recaptured 149 days later (1290 mm TL) in the net with another 19 individuals.

Two Samson fish were recaptured at their original capture location, around one year later (302 and 378 days). In a similar conventional tagging programme of Samson fish in Western Australia, approximately 8,850 fish have been tagged off Rottnest Island where they form spawning aggregations. Recapture data indicate these fish migrate east along the south coast of Western Australia in autumn as far as the south coast of Kangaroo Island, South Australia and return to Rottnest Island in summer to spawn (Rowland et al. 2006). It is possible that Samson fish in South Australia form part of this migratory stock. Determining whether South Australian Seriola spp. make long- range movements will require increased tagging effort.

Our understanding of the movements of large kingfish in northern Spencer Gulf is highly speculative. This study has provided the first firm data on Seriola movements in this region. We found that large kingfish probably remained in, or returned to, Port Augusta for up to five months, as shown by four individuals (>1000 mm TL) tagged in May (2005 and 2006) and recaptured in the same area eight, 88, 98 and 149 days after release. Kingfish may leave the area and return to Port Augusta seasonally, as suggested by four fish that were recaptured in the same area after 326, 328, 383 and 442 days at liberty. The absence of recaptures and landings between November 2005 and April 2006 implies that large fish may leave the area during summer, returning in late autumn/early winter. Indeed, one large kingfish captured at Port Augusta in October 2006 and recaptured near Fitzgerald Bay, Whyalla in December 2006, may indicate a seasonal southerly migration, but too little data were recovered to be certain.

Port Augusta attracts aggregations of large, mature kingfish. This area comprises a sheltered, shallow region with mangrove habitat. McGlennon (1997) suggested that some kingfish spawning may occur in Port Augusta and that spawning was imminent for some fish caught in a fishing competition in August 1996. In marked contrast, it is speculated that kingfish in NSW are pelagic spawners, moving offshore to spawn (Smith et al. 1991). Knowledge about whether kingfish aggregate at Port Augusta as part of a spawning event is important for the management of the recreational fishery because if this area is the spawning ground for a single stock of kingfish, that stock may be susceptible to localised depletion. This is particularly significant in light of our data that suggests that the population of kingfish in Spencer Gulf is small.

The recapture rate of large kingfish in northern Spencer Gulf by recreational anglers was high (18.4%), and most likely reflects fishing pressure in this region combined with fish remaining in the area for an extended period (i.e. five months). Recapture rates for kingfish tagged on the east coast of Australia and for Samson fish tagged in Western Australia are considerably lower at 8.0% and 1.8 %, respectively (Gillanders et al. 2001, A Rowland pers. comm.). Promotion of tag and release fishing by recreational anglers could help mitigate the possibility of local depletion in northern Spencer Gulf.

In this study, the majority of fish were tagged by experienced taggers. Recaptures of tagged fish in our study indicate that individuals can survive the capture, tag and release procedure. Certainly, an individual fish captured by line, tagged and subsequently netted in a school demonstrates that individuals are capable of locating and rejoining a school. Gillanders et al. (2001) found that more experienced taggers have better recovery rates. Therefore, it may be necessary to educate inexperienced taggers about appropriate tagging and handling methods for large kingfish to reduce tag-associated mortality. This would also ensure that tags are positioned and inserted correctly, minimising tag loss.

Identifying the nature of wild Seriola migrations will assist in the understanding of potential parasite interactions between wild and farmed fish. In the northern hemisphere, there is intense debate about the effects of parasitic crustaceans (= sea lice) from farms on wild salmonids. Some scientists have linked increased parasite

loads with declines in wild fish stocks (e.g. Butler 2002; Krkošek et al. 2005; Tully et al. 1993) while others argue that over-fishing, habitat loss and climate change are responsible for the declines (Noakes et al. 2000). We have identified that kingfish may remain in waters of northern Spencer Gulf for an extended period, which is important when considering the potential expansion of the kingfish sea-cage aquaculture industry in Spencer Gulf. A recapture of a large kingfish tagged in Port Augusta and recaptured at Point Lowly indicates that large wild fish migrate past kingfish sea-cage farms in Fitzgerald Bay (2 km north of Point Lowly). Further recapture data may indicate whether kingfish migration patterns create opportunities for substantial interactions between wild and farmed kingfish and when critical exposure windows for pathogens and their transmission between wild and farmed stocks occur.

3.4. Research development and interest in the community

The South Australian fishing community expressed an overwhelming interest in our tagging programme. In northern Spencer Gulf, retired kingfish netters and recreational fishers volunteered their time to locate fish and informed us when kingfish were sighted. Other recreational fishers heard about the project and made contact, inquiring about participation. In one instance, a fisher provided a spreadsheet containing detailed records for over 40 large kingfish captured near Fitzgerald Bay in northern Spencer Gulf over the past six years. Anglers‟ pursuit to participate in this programme in order to understand more about the nature of kingfish migrations is a testament to an emerging tag and release ethos for this species in the recreational fishing community. Indeed, we are aware of several fishers who target Seriola spp. and now practice and promote tag and release after being involved in this work. Prior to this tag programme, only 57 kingfish and 1 Samson fish had ever been tagged and released in South Australian waters (since 1991). A further 31 kingfish and 14 Samson fish were tagged by recreational fishers operating independently of us (using Saftag/NSW Fisheries tags) during our tagging programme.

BENEFITS

Parasite risk analyses provide a disciplined and consistent approach for the calculation of the relative level of risk associated with individual parasite species. Our assessment can be used as a model to assess the risk posed by parasites found in local fish populations. It is important to note that the predictive value of risk assessment is limited without supporting likelihood data. The parasite species determined to pose the greatest threat to the sustainability and profitability of the kingfish industry in Australia are now identified. Suitable control measures may be put in place to help alleviate parasite levels in the event of infection or outbreak.

Alizarin Red S at 250 ppm was the best chemical marker for otoliths of kingfish. This chemical dye is inexpensive and easily detectable using UV filters on a light microscope. This chemical marker could be developed to mark batches of kingfish produced at kingfish hatcheries and enable scientists to distinguish between naturally occurring and kingfish that have escaped from aquaculture.

High recapture rates of large wild kingfish in Port Augusta (18.4%) suggest that a limited number of individuals congregate in this area. At present, this population is under significant recreational pressure. A limited number of individuals are also collected for brood stock to supply kingfish hatcheries in Port Augusta and Arno Bay. Limiting the number of large individuals removed from this fishery may ensure the seasonal return of this population to the area for future supply.

FURTHER DEVELOPMENT

Results from this project have been widely disseminated throughout the Australian kingfish aquaculture industry through meetings and workshops (Appendix 3). Presentations were also delivered at the Australian Society for Parasitology and Australasian Aquaculture conferences (Appendix 3). This will result in a better understanding of parasite collection methods, identification and potential interaction between wild and farmed fish. We have also developed potential techniques to discriminate between wild and farmed fish. This research could be used to attempt an industry-wide approach to parasite and stock management. The results can be exploited commercially by aquaculture companies if they choose to adopt any of the suggested strategies.

More specifically, the risk analysis informed a list of chemicals proposed for minor use permit (MUP) development by the Veterinary Medicines in Aquaculture Working Group (VMA-WG) of Aquatic Animal Health Committee (AAHC). This involves development of applications for MUPs and their submission to the APVMA. One priority chemical identified by the risk analysis (peroxide) has had an MUP approved after submission of an application through the VMA-WG. The risk assessment has also assisted in the planning of further research projects under the proposed Seafood Co-operative Research Corporation (CRC).

PLANNED OUTCOMES

1.1. Improved prevention and management of parasites

We identified standard sampling methods that can be used for ongoing assessment of parasite prevalence and intensity in wild and farmed kingfish. This has allowed for increased sensitivity in parasite sampling and a standard and consistent approach to parasitological effort. Sampling of parasite fauna of wild and farmed kingfish should be incorporated into an ongoing surveillance and monitoring program for effective parasite management, risk identification and impact assessment at farm locations (McVicar 1997). Information on the parasite assemblages of wild and farmed Seriola spp. could result in improved recognition of disease and prevention and management of parasite infections in kingfish aquaculture. This will enable proactive rather than reactive parasite management and prevention of serious outbreaks.

Further taxonomic descriptions were beyond the scope of this research project. It is likely that many more new parasite species infect wild Seriola spp. in Australian waters. Additional sampling and collection of parasite specimens from Samson fish and kingfish will aid in the description of new species where only a few specimens were recovered in the present study. Voucher specimens of the parasite species documented here have been deposited in the South Australian Museum (see Table 3) and can be accessed for further study.

Accurate assessment of risk is dependent upon current information available on parasite presence and distribution. As new information on parasite distributions in wild fish populations becomes available, assessments of the likelihood of parasite establishment and proliferation in sea-cage aquaculture may change.

Aquaculture husbandry practices can also have an impact on the presence or absence of parasite species and their relative infection prevalence and intensity on a farm. The effects of parasite species estimated by this study to pose a potential risk could be alleviated by a variety of management practices. Regular net changes may reduce the accumulation of monogenean eggs and potential intermediate hosts of myxozoans and

sanguinicolids. Stocking farming areas with single age classes of fish and adopting more systematic fallowing has been shown to reduce crustacean infections in farmed salmon (Bron et al. 1993). Feeding only with extruded pellets will lower the likelihood of cestode and anisakid nematode establishment by eliminating consumption of infective stages in intermediate hosts (Ogawa 1996). Spatial segregation of sea-cages, lowered stocking densities, fallowing and strategically timed treatments across entire farm leases may lessen the impact of parasites in aquaculture (e.g. Mooney et al. 2006). Prevention measures may include establishment of zones with movement restrictions of fish between these zones.

1.2. New zone investigation In 2006, 1 500 tonnes of farmed S. lalandi was produced annually in Australia, with a plan to increase production to 10 000 tonnes by 2012 (Anon. 2006). The expansion of this industry will require identification of new sites for sea-cage aquaculture. Our results may help define decision criteria for aquaculture allocation to facilitate ecologically sustainable development of the sea-cage aquaculture industry.

1.2.1. Sources of infection One of the most important discoveries from this research was the recovery of three species of Paradeontacylix from wild Seriola species in Australian waters. The association of this genus with mass mortality and the absence of mitigation methods have serious implications for the future development of kingfish sea-cage aquaculture. Specifically, P. godfreyi presents a moderate likelihood of establishment and proliferation in kingfish sea-cage aquaculture in Spencer Gulf, South Australia. It is essential to identify potential intermediate host species for Paradeontacylix species so that sea-cages containing Seriola species can be positioned to keep fish and invertebrate intermediate hosts spatially segregated. Alternatively, if the intermediate host was found to inhabit the sea-cage netting, procedures could be put in place for more regular net changes or appropriate de-fouling of the nets.

It appears that an appropriate intermediate host for Paradeontacylix may be absent from S. lalandi farming areas in Spencer Gulf, South Australia because adult parasites have not been detected in farmed fish. Identifying the species of blood fluke infecting farmed S. lalandi in New Zealand and examination of potential intermediate hosts near S. lalandi sea-cages in New Zealand may allow us to draw comparisons or assess

potential intermediate host species for blood flukes in Australia. Identification of the intermediate host(s) would help to determine suitable sea-cage sites for S. lalandi away from potential infection sources as the industry expands.

1.2.2. Escaped fish

Effective management of sea-caged fish may be jeopardised by re-infections of parasites from escaped fish. Accidental release of farmed fish can occur through mishandling, faulty equipment, vandalism, weather and predator damage. Our knowledge of the biology of escaped fish is very limited. Important factors that influence the behaviour of escaped fish may include their degree of domestication. It is unknown whether farmed fish are able to feed independently or recognise predators after escaping. Escaped fish may stay close to their sea-cage immediately following an escape event, allowing a high proportion to be recaptured.

Quantifying fish lost during an escape event is technically difficult and once fish have escaped, it may be equally difficult to determine their existence in wild fish populations. Few reliable methods are available to distinguish natural fish from farmed fish and while subjective judgments may use fish appearance, size, and even parasite size and morphology to determine fish origin, these characteristics do not allow definitive determination and may not be replicable (Nordhagen et al. 2000). Distinguishing fish using molecular tools is expensive and may not be valid if brood stock or farmed fish are sourced from the wild. Marking otoliths has emerged as a potential tool that could be used to discriminate between escaped and naturally occurring fish. Without an understanding of the biology of escaped fish, the health and parasite risks associated with interactions between escaped and farmed fish cannot be estimated or managed.

1.2.3. Wild fish migration This research provided data on migrations of large, wild kingfish in northern Spencer Gulf. Our results indicate that wild kingfish migrate south from Port Augusta, past kingfish sea-cages in Fitzgerald Bay in summer. There may be enhanced interaction between wild and farmed fish in this period. Currently, summer is the period in which

farmed fish require more frequent treatment for monogenean parasites. This is because parasite eggs hatch more quickly in warmer sea temperatures. However, increased interaction between farmed and wild fish in warmer months may also contribute to seasonally enhanced infection levels.

Our discovery that large wild kingfish remain in, or return to, Port Augusta between May and October indicates that an expanding aquaculture industry in this region may experience heightened interactions between wild and farmed fish.

1.2.4. Stock structure Describing the stock structure of kingfish in Australian waters allows for the selection of suitable locations for kingfish sea-cage aquaculture where parasites of potential problems may be absent. We found evidence that wild S. lalandi in the Tasman Sea comprise a single stock (Hutson et al. in press). This was supported by previous genetic analysis of fish sampled from Australasia (Nugroho et al. 2001) and studies based on conventional tagging (Gillanders et al. 2001). Mackenzie (2002) suggests that a comparison of entire parasite communities may be an efficient approach for distinguishing populations of large pelagic fish species. More recently, parasite genotypes have been used to identify source populations of migratory fish (Criscione et al. 2006). Research into parasite genotypes of key parasite species would be useful to assess the degree of mixing in S. lalandi populations in Australia. Tissues from wild S. lalandi captured from NSW, Victoria and South Australia during this study were provided to Prof Steve Donnellan (Evolutionary Biology Unit, The South Australian Museum, South Australia) for inclusion in a fish DNA database. This source may be useful to supplement future research into kingfish stock discrimination throughout southern Australia. Using a combination of multiple stock identification methods including parasites, otolith chemistry and conventional tagging may be the most powerful method to enable effective discrimination of different stock.

CONCLUSIONS

Objective 1. Potential for parasite interactions between wild and farmed kingfish Knowledge of parasite assemblages of wild and farmed kingfish enables proactive rather than reactive parasite management. Procedures and control measures can reduce the likelihood of establishment and proliferation of harmful parasite species and may prevent serious outbreaks in sea-cage aquaculture. Continued sampling of parasite fauna from wild and farmed fish will identify parasites that pose potential risks and the most appropriate parasite management practices to put in place. This will also enable assessment of farm locations and improve the welfare of both farmed and wild fish.

Objective 2. Assessment of migratory behaviour of wild kingfish Integration of our tagging programme with the fishing community enabled us to tag an unprecedented number of wild kingfish and Samson fish in South Australia. We believe this work has helped to promote sustainable recreational fishing practices for Seriola spp. in South Australia and expect that this will be reflected in further tag and recapture records at more localities throughout South Australia. Multiple recapture data would be especially informative to determine seasonal movements of Seriola spp. Collaborating with fishers is the only method of ensuring that this information is obtained.

Objective 3. Discrimination of farmed and wild kingfish We determined that Alizarin Red S (ARS) at 250 ppm was the best chemical marker for otoliths of kingfish. ARS overcomes some the limitations of other dyes - it is inexpensive and easily detectable using UV filters on a light microscope. We also experienced minimal associated fish mortality with this chemical treatment and no cessation of growth. This chemical marker could be developed to mark batches of kingfish produced at the two kingfish hatcheries currently operating in South Australia and enable scientists to distinguish between naturally occurring and escaped kingfish. Marked fish would need to be raised for longer periods of time than they were in the current study with non-marked fish to ensure there are no negative effects on growth and whether there are any related food safety issues. It would also be

beneficial to combine naturally occurring signatures with other identification approaches (e.g. artificial elemental signatures) to accurately distinguish aquaculture from wild-caught fish.

REFERENCES

Adjei EL, Barnes A, Lester RJG (1986) A method for estimating possible parasite- related host mortality illustrated using data from Callitetrarhynchus gracilis (Cestoda: Trypanorhyncha) in lizardfish (Saurida spp.). Parasitology 92, 227- 243.

Anon. (2006) Kingfish industry heads for production boom. In. ([WWW document]. URL http://www.abc.net.au/rural/news/content/2006/s1571458.htm)

Australian Quarantine and Inspection Service. (1999) Final reports of import risk analyses on non-viable salmonid products, non-viable marine finfish products and live ornamental finfish and adoption of new policies. AusInfo, Canberra.

Ausvet Animal Health Services, 2002 (WWW document). URL http://www.ausvet.com.au/ (accessed on 29 June 2006).

Bashey F (2004) A comparison of the suitablity of alizarin red s and calcein for inducing a nonlethally detectable mark in juvenile guppies. Transactions of the American Fisheries Society 133, 1516-1523.

Beckman DW, Schulz RG (1996) A simple method for marking fish otoliths with alizarin compounds. Transactions of the American Fisheries Society 125, 146- 149.

Begg GA, Cappo M, Cameron DS, Boyle S, Sellin MJ (1998) Stock discrimination of school mackerel, Scomberomorus queenslandicus, and spotted mackerel, Scomberomorus munroi, in coastal waters of eastern Australia by analysis of minor trace elements in whole otoliths. Fishery Bulletin 96, 653-666.

Beltran RR, Champigneulle A, Vincent G (1995) Mass-marking of bone tissue of Coregonus lavaretus L. and its potential application to monitoring the spatio- temporal distribution of larvae, fry and juveniles of lacustrine fishes. Hydrobiologia 300/301, 399-407.

Bjørn PA, Finstad B, Kristofferson R (2001) Salmon lice infection of wild sea trout and Arctic char in marine and freshwaters: the effects of salmon farms. Aquaculture Research 32, 947-962.

Bray RA, Cribb TH, Barker SC (1993a) The Hemiuroidea (Digenea) of pomacentrid fishes (Perciformes) from Heron Island, Queensland, Australia. Systematic Parasitology 24, 159-184.

Bray RA, Cribb TH, Barker SC (1993b) Hemiuridae (Digenea) from marine fishes of the Great Barrier Reef, Queensland, Australia. Systematic Parasitology 25, 37- 62.

Bron JE, Sommerville C, Wooten R, Rae GH (1993) Fallowing of marine Atlantic salmon, Salmo salar L. farms as a method for the control of sea lice, Lepeoptherius salmonis (Kroyer, 1837). Journal of Fish Diseases 16, 487-493.

Brothers EB (1990) Otolith marking. American Fisheries society symposium 7, 183- 202.

Bullard SA, Overstreet RM (2002) Potential pathological effects of blood flukes (Digenea: Sanguinicolidae) on pen-reared marine fishes. Proceedings of the Gulf and Caribbean Fisheries Institute 53, 10-25.

Butler JRA (2002) Wild salmonids and sea louse infestations on the west coast of Scotland: sources of infection and implications for the management of marine salmon farms. Pest Management Science 58, 595-608.

Campana SE (1999) Chemistry and composition of fish otoliths: pathways, mechanisms and applications. Marine Ecology Progress Series 188, 263-297.

Chambers CB, Ernst I (2005) Dispersal of the skin fluke Benedenia seriolae (Monogenea: Capsalidae) by tidal currents and implications for sea-cage farming of Seriola spp. Aquaculture 250, 60-69.

Costa G, Pontes T, Rego AA (2004) Prevalence, intensity and abundance of Rhadinorhynchus pristis (Acanthocephala, Rhadinorhynchidae) in chub mackerel Scomber japonicus (Pisces, Scombridae) from Madeira Island. Acta Parasitologica 49, 41-44.

Costello MJ (1993) Review of methods to control sea lice (Caligidae: Crustacea) infestation on salmon (Salmo salar) farms. In 'Pathogens of wild and farmed fish: sea lice'. (Eds GA Boxshall and D Defaye) pp. 219-252. (Ellis Horwood: Chichester)

Crespo S, Grau A, Padrós F (1992) Sanguinicoliasis in the cultured amberjack Seriola dumerili Risso, from the Spanish Mediterranean area. Bulletin of the European Association of Fish Pathologists 12, 157-159.

Criscione CD, Cooper B, Blouin MS (2006) Parasite genotypes identify source populations of migratory fish more accurately than fish genotypes. Ecology 87, 823-828.

Deveney MR, Bayly TJ, Johnston CJ, Nowak BF (2005) A parasite survey of farmed southern bluefin tuna, Thunnus maccoyii (Castelnau). Journal of Fish Diseases 28, 279-284.

Dezfuli BS, Simoni E, Rossi R, Manera M (2000) Rodlet cells and other inflammatory cells of Phoxinus phoxinus infected with Raphidascaris acus (Nematoda). Diseases of Aquatic Organisms 43, 61-69.

Diggles B, Hutson KS (2005) Diseases of kingfish (Seriola lalandi) in Australasia. In 'Aquaculture Health International'. pp. 12–14.

Egusa S (1983) Disease problems in Japanese yellowtail, Seriola quinqueradiata, culture: a review. Rapports et proces-verbaux des Reunions conseil International pour l'exploration de la mer 182, 10-18.

Elsdon TE, Gillanders BM (2003) Relationship between water and otolith chemical concentrations in juvenile black bream Acanthopagrus butcheri. Marine Ecology Progress Series 260, 263-272.

Ernst I, Whittington ID, Corneillie S, Talbot C (2002) Monogenean parasites in sea- cage aquaculture. Austasia Aquaculture, February/March, 46-48.

Ernst I, Whittington ID, Corneillie S, Talbot C (2005) Effects of temperature, salinity, desiccation and chemical treatments on egg embryonation and hatching success of Benedenia seriolae (Monogenea: Capsalidae), a parasite of farmed Seriola spp. Journal of Fish Diseases 28, 157-164.

Finstad B, Bjørn PA, Grimnes A, Hvidsten NA (2000) Laboratory and field investigations of salmon lice (Lepeoptheirus salmonis (Krøyer)) infestation on Atlantic salmon (Salmo salar L.) post-smolts. Aquaculture Research 31, 795- 803.

Fletcher WJ, Chesson J, Fisher M, Sainsbury KJ, Hundloe TJ (2004) 'National ESD reporting framework: the 'how to' guide for aquaculture.' FRDC, Version 1.1, Canberra, Australia.

Fowler AJ, Ham JM, Jennings (2003) 'Discriminating between cultured and wild kingfish (Seriola lalandi) in South Australia.' South Australian Research and Development Institute Aquatic Sciences Publication, RD03/0519.

Francis RICC, Paul LJ, Mulligan KP (1992) Ageing of adult snapper (Pagrus auratus) from otolith annual ring counts: validatoin by tagging and oxytetracycline injections. Australian Journal of Marine and Freshwater Research 43, 1069-1089.

Fujita S, Yoda M, Ugajin I (1968) Control of an ectoparasitic copepod, Caligus spinosus Yamaguti, on the cultured adult yellowtail. Fish Pathology 2, 122-127.

Gillanders BM (2002) Temporal and spatial variability in elemental composition of otoliths: implications for determining stock identity and connectivity of populations. Canadian Journal of Fisheries and Aquatic Sciences 59, 669-679.

Gillanders BM, Ferrell DJ, Andrew NL (2001) Estimates of movement and life- history parameters of yellowtail kingfish (Seriola lalandi): how useful are data from a cooperative tagging programme? Marine and Freshwater Research 52, 179-192.

Gillanders BM, Joyce TC (2005) Distinguishing aquaculture and wild yellowtail kingfish via natural elemental signatures in otoliths. Marine and Freshwater Research 56, 693-704.

Hernaman V, Munday PL, Schlappy ML (2000) Validation of otolith growth- increment periodicity in tropical gobies. Marine Biology 137, 715-726.

Ho J, Kim I, Cruz-Lacierda ER, Nagasawa K (2004) Sea lice (Copepoda, Caligidae) parasitic on marine cultured and wild fishes of the Philippines. Journal Fish Society Taiwan 31, 235-249.

Ho J, Nagasawa K, Kim I, Ogawa K (2001) Occurrence of Caligus lalandei Barnard, 1948 (Copepoda, Siphonostomatoida) on amberjacks (Seriola spp.) in the Western North Pacific. Zoological Science 18, 423-431.

Hutson KS, Ernst I, Mooney AJ, Whittington ID (2007) Metazoan parasite assemblages of wild yellowtail kingfish (Seriola lalandi) from eastern and southern Australia. Parasitology International.

Hutson KS, Whittington ID (2006) Paradeontacylix godfreyi n. sp. (Digenea: Sanguinicolidae) from the heart of wild Seriola lalandi (Perciformes: Carangidae) in southern Australia. Zootaxa 1151, 55-68.

Johnson SC, Treasurer JW, Bravo S, Nagasawa K, Kabata Z (2004) A review of the impact of parasitic copepods on marine aquaculture. Zoological Studies 43, 229- 243.

Jonsson B (1997) A review of ecological and behavioral interactions between cultured and wild Atlantic salmon. ICES Journal of Marine Science 54, 1031-1039.

Krkošek M, Lerwis MA, Volpe JP (2005) Transmission dynamics of parasitic sea lice from farm to wild salmon. Proceedings of the Royal Society B. 272, 689-696.

Lagardere F, K. T, M.L.B. A (2000) Feasibility of otolith markings in large juvenile turbot, Scophthalmus maximus, using immersion in alizarin-red S solutions. ICES Journal of Marine Science 57, 1175-1181.

Lebedev BI (1967) Australorhynchus tetramorphacanthus gen et sp. n. found in fish of the New Zealand-Australian shelf (Acanthocephala, Rhadinorhynchidae Travassos, 1923) (in Russian). Zoologichessky Zhurnal 46, 279-282.

Leips J, Baril CT, Rodd FH, Reznick DN, Bashey F, Visser GJ, Travis J (2001) The suitablity of calcein to mark Poeciliid fish and a new method of detection. Transactions of the American Fisheries Society 130, 501-507.

Lester RJG (1982) Unicapsula seriolae n. sp. (Myxosporea, Multivalvulida) from Australian yellowtail kingfish Seriola lalandi. Journal of Protozoology 29, 584- 587.

Mackenzie K (2002) Parasites as biological tags in population studies of marine organisms: an update. Parasitology 124, 153-163.

Maeno Y, Sorimachi M, Ogawa K, Kearn GC (1995) Myxobolus spirosulcatus n. sp. (Myxosporea, Bivalvulida) infecting the bile duct of the yellowtail Seriola quinqueradiata from Japan. Systematic Parasitology 31, 189-193.

Manera M, Dezfuli BS (2003) Lernanthropus kroyeri infections in farmed sea bass Dicentrarchus labrax: pathological features. Diseases of Aquatic Organisms 57, 177-180.

Mansell B, Powell MD, Ernst I, Nowak BF (2005) Effects of the gill monogenean Zeuxapta seriolae (Meserve, 1938) and treatment with hydrogen peroxide on pathophysiology of kingfish, Seriola lalandi Valenciennes, 1833. Journal of Fish Diseases 28, 253-262.

McGlennon D (1997) 'Report to the scalefish management committee on yellowtail kingfish (Seriola lalandi).' South Australian Research and Development Institute, Adelaide.

McVicar AH (1997) Disease and parasitic implications of the coexistence of wild and cultured Atlantic salmon populations. ICES Journal of Marine Science 54, 1093- 1103.

Milton DA, Chenery SR (2001) SOurces and uptake of trace elements in otoliths of juvenile barramundi (Lates calcarifer). Journal of Experimental Marine Biology and Ecology 264, 47-65.

Mooney AJ, Ernst I, Whittington ID (2006) An egg-laying rhythm in Zeuxapta seriolae (Monogenea: Heteraxinidae), a gill parasite of yellowtail kingfish (Seriola lalandi). Aquaculture 253, 10-16.

Moran JDW, Whitaker DJ, Kent ML (1999) A review of the myxosporean genus Kudoa Meglitsch, 1947, and its impact on the international aquaculture industry and commercial fisheries. Aquaculture 172, 163-196.

Noakes DJ, Beamish RJ, Kent ML (2000) On the decline of Pacific salmon and speculative links to salmon farming in British Columbia. Aquaculture 183, 363- 386.

Noga E (2000) 'Fish disease: diagnosis and treatment.' (Iowa State University: Ames)

Nordhagen JR, Heuch PA, Schram TA (2000) Size as indicator of origin of salmon lice Lepeophtheirus salmonis (Copepoda: Caligidae). Contributions to Zoology 69, 99-108.

Nugroho E, Ferrell DJ, Smith P, Taniguchi N (2001) Genetic divergence of kingfish from Japan, Australia and New Zealand inferred by microsatellite DNA and mitochondrial DNA control region markers. Fisheries Science 67, 843-850.

Ogawa K (1996) Marine parasitology with special reference to Japanese fisheries and mariculture. Veterinary Parasitology 64, 95-105.

Ogawa K, Fukudome M (1994) Mass mortality caused by blood fluke (Paradeontacylix) among amberjack (Seriola dumerili) imported to Japan. Fish Pathology 29, 265-269.

Ogawa K, Yokoyama H (1998) Parasitic diseases of cultured marine fish in Japan. Fish Pathology 33, 303-309.

Rach JJ, Gaikowshi MP, Ramsay RT (2000) Efficacy of hydrogen peroxide to control parasitic infestations on hatchery-reared fish. Journal of Aquatic Animal Health 12, 267-247.

Roberts RJ (2001) 'Fish Pathology.' (W.B. Saunders: London)

Roff JC, Hopcroft RR (1986) High precision microcomputer based measuring system for ecological research. Canadian Journal of Fisheries and Aquatic Sciences 43, 2044-2048.

Rohde K (1976) Marine parasitology in Australia. Search 7, 477-482.

Rohde K (1978) Monogenean gill parasites of the kingfish Seriola grandis Castlenau (Carangidae) from the Great Barrier Reef. Publications of Seto Marine Biological Laboratory 24, 369-376.

Rohde K (1981) Ultrastructure of the buccal organs and associated structures of Zeuxapta seriolae (Meserve, 1938) Price, 1962 and Paramicrocotyloides reticularis Rohde, 1978 (Monogenea, Polyopisthocotylea). Zoologischer Anzeiger 206, 279-291.

Rowland A, Mackie M, Gill H, Bevan A (2006) A tagging study on Samson fish (Seriola hippos) in Western Australian waters: movement and post-release survival. In 'Australian Society for Fish Biology Conference'. September, Hobart, Tasmania

Secor DH, White MG, Dean JM (1991) Immersion marking of larval and juvenile hatchery-produced striped bass with oxytetracycline. Transactions of the American Fisheries Society 120, 261-266.

Sheppard M (2004) 'A photographic guide to diseases of yellowtail (Seriola) fish.' (Quadra Printers: British Columbia, Canada)

Smith A, Pepperell J, Diplock J, Dixon P (1991) 'Study suggests NSW kingfish one stock.' Australian Fisheries.

Tang D, Kalman JE (2005) A new Caligiform copepod (Siphonostomatoida: Dissonidae) parasitic on Seriola hippos from Western Australian waters, with new records and morphological variation for Dissonus nudiventris and Dissonus similis, and an updated key to the species of Dissonus. Journal of Parasitology 91, 427-437.

Thomas LM, Holt SA, Arnold CR (1995) Chemical marking techniques for larval and juvenile red drum (Sciaenops ocellatus) otoliths using different fluorescent markers. Recent developments in fish otolith research.

Tully O, Poole WR, Whelan KF, Merigoux S (1993) Parameters and possible causes of epizootics of Lepeophtheirus salmonis Krøyer infesting sea trout (Salmo

trutta L.) off the west coast of Ireland. In 'Pathogens of wild and farmed fish: Sea Lice'. (G Boxshall and D Defaye, editors) pp. 202-213. (Ellis Horwood: Chichester) van der Walt B, Faragher RA (2003) Otolith marking of rainbow trout fry by immersion in low concentrations of alizarin complexone. North American Journal of fisheries Management 23, 141-148.

Weber M, Junge R (2000) Identification and treatment of Moniliformis clarki (Acanthocephala) in cotton-topped tamarins (Saguinus oedipus). Journal of Zoo and Wildlife Medicine 31, 503-507.

Whittington ID, Ernst I, Corneillie S, Talbot C (2001) Sushi, fish and parasites. In 'Australasian Science'. pp. 33-36.

Yokoyama H, Fukuda Y (2001) Ceratomyxa seriolae n. sp. and C. buri n. sp. (Myxozoa: Myxosporea) from the gall-bladder of cultured yellowtail Seriola quinqueradiata. Systematic Parasitology 48, 125-130.

APPENDIX 1: INTELLECTUAL PROPERTY

The intellectual property and valuable information arising from this report are: 1. Copyright of this report

APPENDIX 2: STAFF

Principal Investigator: Colin Johnston1,2

Co-investigators: Ingo Ernst3,4, Ian D. Whittington3,5, Bronwyn M. Gillanders6, Kate S. Hutson3 and Clinton B. Chambers3,7

PhD student: Kate S. Hutson3

Honours student: Tanya C. Joyce6

Research Staff: Justin E. Rowntree6 and Bradley P. Smith3

------

1. Primary Industry and Resources South Australia, Aquaculture, GPO Box 1625, Adelaide, South Austrlaia, 5001, Australia 2. Present address: Biosecurity New Zealand, Ministry of Agriculture and Forestry, PO Box 2526, Wellington, New Zealand 3. Marine Parasitology Laboratory, School of Earth and Environmental Sciences, DX 650 418, The University of Adelaide, South Australia 5005, Australia 4. Present address: Aquatic Animal Health Unit, Australian Government Department of Agriculture, Fisheries and Forestry, GPO Box 858 Canberra, ACT 2601, Australia 5. Monogenean Research Laboratory, Parasitology Section, The South Australian Museum, North Terrace, Adelaide, South Australia 5000, Australia 6. Southern Seas Ecology Laboratories, School of Earth and Environmental Sciences, The University of Adelaide, South Australia 5005, Australia 7. Present address: Woodside Energy Ltd. Woodside Plaza 240 St Georges Terrace, Perth Western Australia 6000, Australia

Email addresses:

Colin Johnston: [email protected] Ingo Ernst: [email protected] Ian D. Whittington: [email protected] Bronwyn M. Gillanders: [email protected] Kate S. Hutson: [email protected] Clinton B. Chambers: [email protected] Tanya C. Joyce: [email protected] Justin Rowntree: [email protected] Bradley P. Smith: [email protected]

APPENDIX 3: PUBLICATIONS ARISING FROM THIS REPORT

1. Joyce, T.C. (2003) Distinguishing aquaculture and wild yellowtail kingfish via natural and aritifical elemental signatures in otolith. Honours thesis, School of Earth and Environmental Sciences, The University of Adelaide, South Australia Australia.

2. Diggles, B. and Hutson, K.S. (2005) Diseases of kingfish (Seriola lalandi) in Australasia. Aquaculture Health International. VIP Publications Ltd, Auckland Issue 3, Nov 2005.

3. Gillanders, B.M. and Joyce, T.C. (2005) Distinguishing aquaculture and wild yellowtail kingfish via natural elemental signatures in otoliths. Marine and Freshwater Research 56, 693- 704.

4. Hutson K.S. and Whittington I.D. (2006) Paradeontacylix godfreyi n. sp. (Digenea: Sanguinicolidae) from the heart of wild Seriola lalandi (Perciformes: Carangidae) in southern Australia, Zootaxa 1151, 55-68.

5. Hutson, K.S. and Tang, D. (2007) Naricolax hoi n. sp. (Poecilostomatoida: Bomolochidae) from Arius maculatus (Siluriformes: Ariidae) off Taiwan and redescription of N. chrysophryenus (Roubal, Armitage and Rohde, 1983) from a new host, Seriola lalandi (Perciformes: Carangidae), in Australian waters. Systematic Parasitology 68, 97-113.

6. Hutson, K.S., Ernst, I., Mooney, A.J. and Whittington, I.D. (2007) Metazoan parasite assemblages of wild Seriola lalandi (Perciformes: Carangidae) from eastern and southern Australia. Parasitology International 56, 95-105.

7. Hutson, K.S., Ernst, I. and Whittington, I.D. (2007) Risk assessment for parasites of Seriola lalandi (Carangidae) in South Australian sea cage aquaculture. Aquaculture 271, 85-99.

8. Hutson, K.S., Smith, B.P., Godfrey, R.T., Whittington, I.D., Chambers, C.B., Ernst, I. and Gillanders, B.M. (2007) A tagging study on yellowtail kingfish (Seriola lalandi) and Samson fish (S. hippos) in South Australian waters. Transactions of the Royal Society of South Australia 131, 128-134.

9. Hutson, K.S. (2007) Parasite interactions between wild and farmed yellowtail kingfish (Seriola lalandi) in southern Australia. PhD Thesis, School of Earth and Environmental Sciences, The University of Adelaide, South Australia Australia.

APPENDIX 4: PRESENTATIONS ARISING FROM THIS REPORT

1. Hutson, K. S. Invited Speaker. „Parasites of wild kingfish (Seriola lalandi)‟. Marine Life Society of South Australia (MLSSA), Adelaide, Australia. 15 September 2004.

2. Hutson, K. S., Ernst, I., and Whittington, I. D. „Parasite interactions between wild and farmed kingfish (Seriola lalandi).‟ Australian Society for Parasitology, Annual Meeting, Fremantle, Western Australia. 28 September 2004.

3-6. Hutson, K.S. „Potential for parasite interactions between wild and farmed fish, discrimination of farmed and wild fish and assessment of migratory behaviour‟. Innovative Solutions for Aquaculture, Adelaide and Port Lincoln, Australia. 29 June 2005, 9 September 2005 and 30 August 2006.

7. Hutson, K.S. and Smith, B.P. Invited Speakers. „Why tag fish?‟ Junior Field Naturalists Society, Adelaide, Australia. 4 May 2006.

8. Hutson, K.S. „Parasite risk assessment for kingfish aquaculture in South Australia‟. Australasian Aquaculture Conference, Adelaide, Australia. 30 August 2006.

9. Hutson, K.S. „Why little guys matter! A new species of blood fluke infecting wild yellowtail kingfish (Seriola lalandi) in southern Australia‟. Royal Society of South Australia, Adelaide, Australia. 10 August 2006.

10. Smith B. P. Invited Speaker. „How to tag and release fish.‟ Science Alive, National Science Week, South Australia, Australia. 13 August 2006.

APPENDIX 5: MEDIA ARISING FROM THIS REPORT

1. Mensforth, S. „Parasitically Speaking: An Interview with Kate Hutson’. South Australian Angler Magazine. December 2004/ January 2005 Issue No 160.

2. Australian Broadcasting Commission (ABC) rural radio „Parasites of wild kingfish’ Port Lincoln, 13 September 2005.

3. Hunt S. ‘Going Pelagic.’ Blue Water: Boats and Sportsfishing Magazine September/October 2005.

4. Postcards, Channel 9 South Australia. „Tagging wild kingfish in Port Augusta.‟ October 2 2005.

5. FAB: Fishing and Boating Channel 7 Network South Australia, Australia. „Tagging wild kingfish in Port Augusta.‟ 12 November 2005.

6. King, B. (2005). „Great catch‟ The Sunday Mail. 20 November 2005.

7. Salkow H. „Brad’s catch of the day: yellowtail kingfish, and research!‟ Adelaidean News from the University of Adelaide Volume 14, No. 9 2005 http://www.adelaide.edu.au/adelaidean/issues/7901/news7906.html

8. Gill, T. „Tough job, but Brad had to do it!‟ Southern Fisheries Magazine Lane Print Group, Adelaide, Summer 2005/2006,.

9. ARC/NHMRC Network for Parasitology Network newsletter. „Network Travel Award for Marine Parasitologist, Kate Hutson’ April 11 2006. www.parasite.org.au/arcnet/Newsletter/Newsletter_110406.pdf

10. Anon. (2006). „Kingfish tagging helps determine potential for parasite interactions.‟ Fisheries and Research Development Corporation, R and D News, Volume 14, No. 3, 2006.

11. Crawford, J. (2007) „Studying Seriola.’ Fishing World Magazine, July 2007.