SANDFISH HATCHERY TECHNIQUES. By Natacha Agudo PROVINCE DES ÎLES LOYAUTÉ

Transcription

1 SANDFISH HATCHERY TECHNIQUES By Nataca Agudo PROVINCE DES ÎLES LOYAUTÉ

2 SANDFISH HATCHERY TECHNIQUES By Nataca Agudo

3 Copyrigt Australian Centre for International Agricultural Researc (ACIAR), te Secretariat of te Pacific Community (SPC) and te WorldFis Center, 2006 All rigts for commercial / for profit reproduction or translation, in any form, reserved. ACIAR, SPC and te WorldFis Center autorise te partial reproduction or translation of tis material for scientific, educational or researc purposes, provided tat ACIAR, SPC, te WorldFis Center and te source document are properly acknowledged. Permission to reproduce te document and/or translate in wole, in any form, weter for commercial / for profit or non-profit purposes, must be requested in writing. Original ACIAR, SPC and te WorldFis Center artwork may not be altered or separately publised witout permission. Secretariat of te Pacific Community Cataloguing-in-publication data Sandfis atcery tecniques / by Nataca Agudo 1. Holoturians Spawning Handbooks, manuals, etc. I. Agudo, Nataca II. Title AACR2 ISBN ACIAR, GPO Box 1571, Canberra, ACT 2601, Australia SPC, B.P. D5, Noumea Cedex, New Caledonia WorldFis Center, c/spc, B.P. D5, Noumea Cedex, New Caledonia Cover design and layout by Muriel Borderie Potos by Nataca Agudo, Caty Hair and Steve Purcell Printed at SPC Headquarters, Noumea, New Caledonia ii

4 ABOUT THIS MANUAL Sandfis is arguably te most commercially valuable of te tropical species of sea cucumber tat are processed into bêce-de-mer. It is widely distributed trougout te Indo-Pacific, occurring in sallow insore areas were it is easily accessible to coastal fisers. A-grade bêce-de-mer processed from sandfis commands some of te igest prices on te international market. But tese same attributes also make it vulnerable to overexploitation. Sadly, tis as appened in most places were it occurs. Wile sandfis was an important component of bêce-de-mer fiseries 20 to 30 years ago, its contribution to bêce-de-mer exports is now relatively small, even trivial. Not surprisingly, tere is widespread interest in restoring te production of sandfis, especially were it promises to deliver benefits to coastal fising communities wit few oter options for earning livelioods. Altoug improved management of capture fiseries, troug measures designed to safeguard te remnant spawning adults, will always be key to restoring production, aquaculture as te potential to elp restore production of tis valuable species in tree ways: troug production and release of cultured juveniles in restocking programmes to increase te number of spawners, but only were suc releases are predicted to add value to oter forms of management; troug put and take sea rancing operations, were cultured juveniles are placed in te wild to be regatered at a larger size wit no intention of allowing tem to spawn; troug farming cultured juveniles in earten ponds and sea pens. Tis manual is designed to elp government agencies and members of te private sector interested in implementing any of tese ways of increasing production of sandfis by outlining te basic metods for spawning and rearing juvenile sandfis. It builds on te pioneering work done in 1988 at te Tuticorin Researc Centre of CMFRI (Central Marine Fiseries Researc Institute) in India and is based largely on metods developed and applied by te WorldFis Center (formerly ICLARM) in Solomon Islands, Vietnam and New Caledonia. Te information in te manual will enable atceries to produce sandfis suitable for release in te wild in relatively large numbers (tens of tousands) regularly. However, it does not pretend to be fully compreensive. Rater, it is a reflection of current knowledge. We ope tat it will soon be made out of date by tose of you wo apply and improve te metods described ere. Figure 1. Bêce-de-mer. iii

5 ACKNOWLEDGEMENTS Te preparation of tis manual was made possible troug dedicated researc on various aspects of te culture and ecology of sandfis by two groups of people: firstly, tose wo contributed to basic knowledge on spawning and larval rearing, and te ecology of sandfis Stepen Battaglene, Cantal Conand, Jean-François Hamel, D.B. James, Claude Massin, Annie Mercier, Andrew Morgan, Rayner Pitt, Cris Ramofafia and J. Evizel Seymour; and secondly, my colleagues at te Worldfis Center, staff at te Nortern Fiseries Centre, Queensland Department of Primary Industries and Fiseries, and trainees from Conservation International, Papua New Guinea (Priscilla Eka, Pamela Mua), wo elped me to refine and develop te basic metods. Te work in New Caledonia arose from te early efforts by Debora Gardner, and Eric Danty to establis te WorldFis Center atcery at Saint Vincent, and to acieve our first successes tere. For teir commitment and perseverance I am grateful. I would also like to tank Warwick Nas, Joann Bell, Steve Purcell, Caty Hair, Gale Semmens and Ricard Knuckey for teir encouragement, comments on te content and editing, and Steve Purcell and Caty Hair for teir potos. Tis manual was funded by te Australian Centre for International Agricultural Researc (ACIAR), te Secretariat of te Pacific Community (SPC) and te WorldFis Center. iv

6 CONTENTS Basic biology of sandfis... 7 How to identify sandfis...7 More about te biology of sandfis...8 Were do sandfis occur?...8 Broodstock...9 Wen sould broodstock be collected?...9 How many broodstock are needed? Wat size sould tey be?...9 Transport of broodstock to te atcery...9 Maintaining broodstock in tanks, sea pens and earten ponds Ways to increase spawning success...11 Spawning...13 Preparing broodstock for spawning...13 Inducing sandfis to spawn...13 Observing te beaviour of broodstock...14 Steps to take wen spawning fails...15 Te first female as just spawned! How sould fertilisation be managed?...15 Collection of eggs from te spawning tank...16 Estimating egg density...17 Larval rearing...19 Transferring fertilised eggs to larval tanks...19 Life cycle of sandfis...19 How sould larvae be reared?...22 Te first doliolaria larvae appear! Wat do tey need to settle? Feeding pentactula larvae and juveniles...26 Review of daily tasks during larval rearing...27 Nursery Culture of early juveniles...29 Detacing and counting juveniles before transfer to nursery tanks...30 Te two stages of te nursery pase...31 Review of tasks during te nursery pase...32 Grow-out of juveniles...33 In net pens in ponds...33 In ponds...34 In sea pens...35 Problems and possible solutions...36 Promising applications for atcery-reared sandfis...37 Annex 1: Algal culture...39 Maintenance of algal cultures...39 Preparation and inoculation of algal cultures...40 Furter reading v

7

8 BASIC BIOLOGY OF SANDFISH How to identify sandfis Te body of te sandfis is elongated, cylindrical and stout. Te dorsal body surface is relatively smoot and as small papillae (i.e. sensory tube feet) wit black dots; colour varies from grey to black wit dark transverse wrinkles. Te ventral surface of te body is flattened and is generally witis in colour. Te mout is on te ventral surface at te anterior end of te body. It is oval in sape and as 20 sort peltate tentacles. Te anus is located dorsally at te posterior end of te body. Sandfis are Ecinoderms, related to starfises and sea urcins. Te precise taxonomy of sandfis is: Pylum Ecinodermata Class Holoturoidea (wit tube feet) Order Aspidocirotida (wit tentacles peltate) Family Holoturiidae (wit body usually circular and gonads single) Genus Holoturia (Metriatyla) Rowe, 1969 Holoturia (Metriatyla) scabra Jaeger, 1833 Figure 2. Sandfis. Adult size range: cm lengt Adult weigt range: g BASIC BIOLOGY OF SANDFISH

9 More about te biology of sandfis Sandfis ave te same general anatomy as oter sea cucumbers. Te gonads (ovaries or testes) lie in one tuft and open dorsally at te anterior end of te body troug a single gonopore (i.e. genital orifice). Te digestive system is composed of a mout, oesopagus, stomac, intestine, cloaca and anus. Respiratory trees, wic sandfis use to obtain oxygen, lie in te posterior of te body and open to te cloaca. Te body wall tat is processed into bêce-de-mer accounts for about 56% of total weigt. Sandfis move wit te elp of tube feet densely distributed on te ventral face, and troug muscular action of te body wall. Sandfis feed on detritus, i.e. organic matter in te mud or sand. Tey appear to feed continuously using te peltate tentacles surrounding te mout to place sediment into te mout. Sandfis are usually observed partially buried in sediment. Te daily burrowing cycle varies according to environmental conditions. Te growt rate of sandfis depends on environmental conditions and te time of year. At medium size, sandfis grow on average 0.5 cm per mont, corresponding to 14 g per mont. Under good conditions tey grow to a size of 300 g in one year. We still do not know ow long sandfis live, but it may be around 10 years. Sea cucumbers ave tiny calcareous plates called spicules in teir skin. Microscopic examination of spicules is used to distinguis species. Sandfis ave many spicules in te sape of tables and knobbed buttons. Sandfis can be sexually mature at a size as small as 200 g. Tere is no apparent relationsip between fecundity (egg production) and body size. Like oter sea cucumbers, sandfis can regenerate some of teir organs. After spending long periods out of water, or being affected by te use of cemicals, being andled during collection and transport, or wen stressed by predators, sandfis may eviscerate teir internal organs. Regeneration of internal organs occurs witin 2 monts. Sandfis and oter tropical sea cucumbers can produce numerous toxins from teir skin and viscera. Tese toxins inflict distress, loss of equilibrium and deat in fis, but do not affect umans. Were do sandfis occur? Sandfis are found in many countries in te Indo- Pacific, from east Africa to te eastern Pacific. Tey are usually found between te latitudes of 30 N and 30 S. Te preferred abitats of sandfis are sallow tropical waters, usually less tan 20 m deep, suc as seltered areas wit ig levels of nutrients, including muddy substrata and seagrass beds. Tey can tolerate reduced salinity (20 ppt) for sort periods and so are sometimes found in brackis water. Figure 4. Wild sandfis in a seagrass abitat. Figure 3. Geograpical distribution of sandfis.

10 BROODSTOCK Wen sould broodstock be collected? Broodstock sould be collected during te reproductive season so tat animals are ready for immediate spawning. Te spawning season for sandfis varies among countries. In countries close to te Equator, sandfis spawn trougout te year. As te latitude approaces 25, spawning is restricted to a sort summer period of 3 monts. In some countries, tere can be two spawning periods eac year. It is important to determine wen tese periods occur. Transport of broodstock to te atcery At sea, collected broodstock sould be kept in seawater in insulated containers. Aeration sould be provided if te animals are to be eld tis way for more tan 2 ours. Preferably, te animals sould be left in te transport containers to defecate before tey are transferred to plastic bags. To transfer te broodstock from te boat to a veicle for transport to te atcery, animals sould be cleaned gently and packed individually in oxygen-filled bags wit 1 L of seawater. Te bags sould be placed in insulated containers during transport. Te containers sould be protected against direct sunligt to maintain te temperature witin te range of C. Figure 5. Collection of broodstock. How many broodstock are needed? Wat size sould tey be? Batces of individuals are usually required to induce a small proportion of animals to spawn. Ideally, te average weigt of broodstock sould be around 500 g. However, te size of broodstock may be smaller in some places, e.g g in Vietnam. Broodstock sould be undamaged, wit no visible skin lesions. Skin appearance sould be smoot and siny, wit a tin, transparent mucous layer. Animals sould react by moving wen you touc and disturb tem. It is impossible to distinguis males and females externally. Sex can only be determined by biopsy or dissection of animals to examine te gonads, or by observation wen tey spawn. Figure 6. Broodstock packed individually in plastic bags in an insulated container. Sandfis can tolerate low dissolved oxygen and ig temperatures in static water (up to 30 C) for long periods (over 80 ours) witout eviscerating. However, it is preferable to maintain tem in optimal conditions to avoid stress, wic can induce premature spawning. Te transport of individual sandfis in damp teatowels for many ours as been practised wit varying success. Avoid Sudden temperature socks and olding for long periods out of water, wic can result in evisceration Damaging te skin of te animals andle tem gently Socks during road transport BROODSTOCK

11 Maintaining broodstock in tanks, sea pens and earten ponds In sea pens Broodstock tat are not yet mature can be kept in captivity until tey ripen. Similarly, following spawning, broodstock can be maintained near te atcery for future use. Tese animals can be eld in tanks, sea pens or earten ponds. In tanks Figure 8. Sea pen. 10 Figure 7. Broodstock maintained in tank wit sand. Broodstock tanks sould ave a flat bottom wit a volume of 1000 to 4000 L and be placed outdoors in sade. Te tanks sould contain a cm layer of sand or mud. Feed for broodstock sould be added at te rate of 50 g/day. Suitable feed ingredients are: prawn ead waste, soya bean powder, rice bran and seagrass powder. Care sould be taken to ensure tat broodstock do not lose weigt. Broodstock sould be stocked at a density of animals per 1000 L tank in static aerated water. Excange te water eac day. Broodstock eld at low density in tanks wit continuously flowing seawater, fed powdered dried algal preparations and ground-up prawn pellets, can often be spawned more tan once. Sea pens for olding broodstock sould be located close to te atcery so tey can be monitored easily and regularly. Sea pens sould be around 800 m 2 and te stocking density of te broodstock sould be <200 g/m 2. Additional feed is not needed. Survival in sea pens is often very ig, but growt rates may be lower tan in ponds depending on local environmental conditions. In eartern ponds Figure 9. Earten ponds. Earten ponds, typical of tose used for maintaining srimp broodstock (450 to 1500 m 2 ), ave proved suitable for olding sandfis broodstock. Pond sediments sould be friable, sandy-mud witout large rocks. Water depts of <1 m are best. Ponds sould be filled at least 2 weeks before transfer of broodstock to ensure natural plankton development. Te stocking density of te broodstock sould be <250 g/m². Tere sould be daily water excange or continuous flow. It sould not be necessary to add any feed. Water quality parameters (temperature, dissolved oxygen and salinity) sould be measured regularly, and daily if possible.

12 Possible problems: Heavy rain can lead to stratification of water in te pond. Stratification can be detected by te presence of a tin layer of low-salinity water at te surface. Te consequences of stratification can include an increase in temperature, fall in dissolved oxygen especially in te bottom layers, and development of anaerobic areas in te sediment. Te combination of tese extreme conditions can be dangerous for sandfis, wic are bentic and slow moving. It can lead to total loss of broodstock in a few days. How to overcome tese problems Hig temperatures and low dissolved oxygen Heavy rain Regularly ceck water temperature and dissolved oxygen (especially in te early morning). Increase water flow, preferably continuously 24 ours per day. Set up paddle weels or oter pond aeration equipment; installation sould provide aeration on te bottom were te animals are located. Remove te surface layer of freswater by adjusting te eigt of te outlet pipe. If a cyclone is predicted, transfer te animals from ponds to indoor tanks (were possible). Summary Batces of individuals (average weigt 500 g) are needed to induce spawning. Wen relying on wild broodstock, ripe and ealty broodstock must be collected during te spawning season. Transport wild broodstock individually in oxygen-filled bags wit seawater, using insulated containers at C. Avoid canges in temperature and oter socks during transport. Figure 10. Feeding broodstock. Ways to increase spawning success Animals eld in ponds or tanks for several monts to 2 years are easier to spawn tan broodstock maintained in sea pens or animals taken directly from te wild. Anoter advantage of olding broodstock in ponds is tat tey generally spawn earlier in te season tan wild individuals, presumably because of te iger temperatures in ponds. Note, owever, tat olding broodstock in ponds or tanks for several monts before spawning is not essential. In Solomon Islands, for example, newly collected sandfis were often induced to spawn. Te advantages and disadvantages of keeping broodstock in captivity, or collecting individuals from te wild, sould be evaluated for eac location. Broodstock can be conditioned by keeping tem in: (a) tanks (15 30 animals/1000 L) wit a sand or mud substratum, wit flowtroug seawater and a supply of food; (b) sea pens of 800 m 2 at densities of <200 g/m 2 ; and (c) earten ponds at densities of <250 g/m 2 wit continuous water excange. Ponds used to old broodstock require management to avoid ig temperatures and low dissolved oxygen caused by stratification due to eavy rain. Salinity must be maintained witin te range of ppt. Te advantages and disadvantages of maintaining broodstock, or collecting tem directly from te wild, sould be evaluated for eac location. BROODSTOCK 11

13 12 Figure 11. Broodstock in a clean spawning tank before induction (30 45 animals are recommended).

14 SPAWNING Preparing broodstock for spawning Inducing sandfis to spawn 1. Use a bare, flat-bottomed tank up to 2 m 3. Provide cover and eaters to maintain constant water temperature at nigt. 2. Clean te tank and disinfect wit clorine (sodium ypoclorite). Fill wit 1-µm filtered and UV-sterilised seawater at ambient temperature (<30 C), to a eigt of cm. Aerate te water moderately. 3. Gently clean te animals and place tem in te tank. Figure 13. Cold sock treatment. 4. Sipon te tank bottom, to remove any sediment and faeces, until spawning starts. 5. Wen spawning is complete, return broodstock to tanks, sea pens or ponds. Also return broodstock if tey fail to spawn. 6. Use a different group of animals for eac spawning. Notes Stress caused by collection and transport is often sufficient to induce spontaneous spawning. Spontaneous spawning usually occurs during te afternoon, evening and/or nigt on te day broodstock are collected. Spawning as often been observed just prior to te full moon and new moon, but can also be induced at oter times. Figure 12. Close-up of broodstock in spawning tank. Termal stimulation Raise water temperatures by 3 5 C for 1 our, eiter by adding warmed seawater to te spawning tank or using aquarium eaters. Keep water temperatures witin te range of C. Stir te water to maintain uniform temperature trougout te tank. If te ambient water temperature is >30 C, give a cold sock treatment for 1 our before te eat sock. To do tis, reduce te level of water in te spawning tank and add sealed plastic bags containing ice to quickly lower te water temperature by 5 C below ambient. Ten apply eat sock as above. After termal stimulation, replace water wit new water at ambient temperature, keeping te animals covered. Gonad extraction metod Dissect a few animals to extract gonads from 1 to 2 ripe males. Store te gonads at 5 C to extend te viability of te sperm. Use te sperm as a spawning stimulant by adding blended fres male gametes to te spawning tank at ambient water temperature. Water pressure Leave broodstock to dry in te tank in te sade for about alf an our before subjecting tem to a powerful jet of seawater for a few minutes. Return broodstock to te spawning tank at ambient water temperature. SPAWNING 13

15 Possible problem: After two days of spawning attempts, broodstock can lose teir mucus and te skin may sow wite lesions due to stress and andling. Avoid andling te animals too muc. Observing te beaviour of broodstock Figure 14. Dry treatment. Dry treatment Leave animals completely dry, or in 2 cm of seawater in te tank for minutes. Keep tem in te sade. Refill tank wit water at ambient temperature. Figure 16. Pre-spawning beaviour. Te beaviour of sandfis often indicates tat spawning is imminent. Pre-spawning beaviours include: rolling movements rytmic contractions lifting and swaying of te front end of te body climbing te tank walls Figure 15. Spirulina bat. Food stimulants Add dried algae (Spirulina at a rate of 30 g per L, or Algamac 2000 at a concentration of 0.1 g/l) for 1 our. Stir te water. After 1 our, remove as muc waste from te tank as possible and replace water wit new water at ambient temperature. Combined treatments Often a combination of treatments is needed to induce spawning. Te best combinations are given below: Figure 17. Animal stands erect ready to spawn. 14 Treatment combinations A B C 1. Dry treatment 1. Hot sock treatment 1. Dry treatment 2. Cold sock treatment 2. Spirulina bat 2. Hot sock treatment 3. Hot sock treatment 3. Spirulina bat

16 Males usually spawn before females, wic start to release eggs 1 our (sometimes sooner or later) after te first male releases sperm. Spawning males are erect and sway from side to side, releasing a continuous stream of sperm. Males spawn for several minutes to ours, even wen tey are disturbed. Spawning females erect teir body before releasing eggs in a sort powerful spurt from te bulging gonopore (i.e. genital orifice). Females can spawn 2 3 times over a period of an our or more, but often stop spawning if disturbed. Record all information about spawning trials, including te unsuccessful attempts. Tis elps to improve future spawning success. Collect te following information: total number of males and females tat spawned spawning time for eac individual wit comments observations of egg development (% regular and irregular sape, % fertilised, diameter) individual broodstock weigt Reduce te risk of losing eggs from spontaneous or delayed spawnings at nigt by: 1. Maintaining broodstock in a spawning tank wit a flow-troug system. 2. Installing a second tank in series wit an internal 100-µm sieve at te outlet pipe, wit moderate aeration, to retain eggs in suspension overnigt until collection te following morning. Provide te tanks wit air diffusers, eaters and covers to maintain constant water temperature overnigt. Te first female as just spawned! How sould fertilisation be managed? Natural fecundity ranges from 9 to 17 million eggs per female, but induced females generally release a fraction of tis, i.e. 1 to 2 million, altoug some individuals can sometimes release up to 4 to 6 million eggs. Figure 18. A spawning sandfis. Steps to take wen spawning fails If broodstock do not respond during spawning induction, try different metods of stimulation. Wen broodstock demonstrate pre-spawning beaviour but do not spawn, biopsy or dissect a few animals and examine te gonads under a microscope to determine if tey are ripe. Ripe ovaries look translucent and ripe testes are milky wite. Commence spawning induction later at nigt. Sandfis usually spawn at nigt in te wild. Management of males Note tat too muc sperm in te spawning tank causes polyspermy (i.e. multiple fertilisation). Polyspermy can reduce te rate of fertilisation and cause damage to egg development and induce larval deformities. If te water in te spawning tank becomes cloudy due to excess sperm, reduce te amount of sperm by siponing and adding new water, or use a flowtroug system. To prevent excess sperm in te spawning tank, remove most males and transfer tem to smaller containers sortly after tey begin releasing sperm. Often tey will continue releasing sperm. Leave one or two males spawning in te tank until te first female releases eggs. Retain te more vigorous spawners. Record te time tat males start releasing sperm, and ow many males spawn. Sperm stays active for several ours. SPAWNING 15

17 Management of females Record te time tat females begin spawning. Tis will be elpful in estimating te progress of egg stages. Increase aeration to moderate levels to keep eggs in suspension. Count and record te number of egg releases per female. Females usually spawn 2 3 times. Once females finis spawning (usually evident wen tey stop moving), remove all broodstock (males and females). Return te animals to teir tank, sea pen or pond. Separation of males into a separate tank once tey begin to spawn, and later addition of controlled amounts of sperm to te females tank, is a way to avoid polyspermy wile maintaining a larger number of faters. Avoid Disturbing or moving females wen removing surplus males so tat spawning of eggs is not interrupted. Observation of egg development After te first eggs are released, sample tem from te water column in a small beaker. Measure egg diameter, using a micrometer eyepiece placed in te lens of a microscope. Record te measurements of te eggs and oter observations suc as: egg stages and size, percentage of regular round eggs, and fertilisation rate. Repeat tese observations every minutes. Fertilised eggs ave a swollen membrane. If many sperm continue to cluster around te egg (i.e. polyspermy), irregular egg development will occur. Figure 20. A fertilised egg, two cell, and four cell stages. Collection of eggs from te spawning tank Use a small beaker to collect a sample of eggs from te water column to estimate te fertilisation rate and percentage of eggs wit advanced cell division. Ensure tat te fertilisation rate is ig and te majority of eggs are at te advanced cell division stage. Tis is at least 1 our after fertilisation. Wait at least 1 our after fertilisation before collecting all te eggs. Recently sed eggs are wite, sperical and visible to te naked eye. Te diameter of te eggs ranges from 80 to 200 µm, but tis varies widely witin te geograpic range of sandfis. Spermatozoa are not visible to te naked eye; tey appear as small, active dark dots clustered togeter around te eggs. Figure 19. Fertilised egg. Figure 21. Egg collector. 16

18 Sipon te eggs slowly from te tank into a µm sieve placed in a bowl (Fig. 22). Make sure te water level is above te mes of te sieve so tat te eggs are not squased onto te mes. Introduce a gentle flow of filtered seawater, at ambient temperature, in te bowl. Rinse eggs to remove excess sperm and dirt. Maintain eggs in suspension in te sieve. Be patient! Collecting eggs takes time. Estimating egg density Transfer te collected eggs regularly and carefully, using beakers, into clean 10 L buckets, until te buckets are filled. Stir te water in te buckets gently to distribute te eggs uniformly. For eac bucket, take tree 1-ml subsamples. Estimate te egg density for eac sample, using a counting cell (e.g. Sedgewick-Rafter camber) under a microscope. Calculate te average density for eac bucket. Estimate te fertilisation rate. Record all data. Transfer and distribute eggs from te buckets to te larval tanks quickly after determining te total number of eggs needed for eac tank. Use egg densities from all buckets to estimate te total number of eggs spawned. Avoid Hig densities of eggs in te sieve and buckets as tis can damage eggs. Average egg density in te buckets sould be <200 eggs/ml, i.e. <2 million eggs per 10 L bucket. Figure 22. Metod of egg collection. Useful tips Carry out spawning in sallow water, and use several sipons wit sieves in bowls to reduce te time needed to collect eggs. Tis requires more atcery staff for a sorter period. Start by siponing eggs from te water column; te eggs in te water column are cleaner (i.e. not surrounded by faeces or sediment). Distribute tese batces of eggs in te larval tanks. If more eggs are required, ten collect tem from te bottom of te tank. Figure 23. Transfer of eggs into a larval tank. SPAWNING 17

19 Summary Batces of clean broodstock are placed in a spawning tank, filled wit 1-µm filtered and UV-sterilised seawater. Termal socks, extracts of male gonads, water pressure, dry treatment, and food stimulants used alone or in combination can induce spawning. Pre-spawning beaviours include rolling movements, rytmic contractions, and lifting and swaying of te anterior end of te body. Males usually spawn first and females usually start one our after te first male. It is important to record spawning data, i.e. induction metod used, number of males and females, spawning time, and observations of eggs. Females usually release eggs 2 to 3 times. Moderate aeration maintains te eggs in suspension. Samples of eggs are regularly taken from te water column to examine te stage of egg development. After >1 our post fertilisation, te eggs can be siponed gently from te water column into a µm sieve placed in a bowl. Flow-troug seawater sould be used to rinse te eggs. Te eggs are transferred to buckets for counting before being placed in te larval tanks. Figure 24. Observation of eggs under a microscope. 18

20 LARVAL REARING Transferring fertilised eggs to larval tanks Avoid Large differences in temperature (over 1 2 C) and salinity of water between buckets and larval tanks as tis can affect survival of eggs. Hig levels of aeration during transfer as tis can trap te eggs and larvae in a strong current, wic can trow tem against te tank walls, damaging or killing tem. Hig initial egg density. Discard excess eggs. Lower egg densities are preferable because tey reduce te risk of total mortality of larvae during te first days. Life cycle of sandfis Figure 25. Larval tank. Prepare cylindrical tanks (up to 2 m 3 ) wit conical bottoms and central drains. Set up 100-µm mes outlet screens. Was te tanks wit clorine (sodium ypoclorite), ten rinse wit freswater. Fill te tanks wit 1-µm filtered and UV-sterilised seawater. Ensure te water temperature is witin te range of 26 to 30 C, and salinity is between 32 and 36 ppt. Install two central air diffusers in eac tank to insure medium aeration and gentle water circulation. Use two air diffusers as a precaution against failure. Immerse aquarium eaters wit termostats to maintain a constant water temperature in te larval tanks. Set up lids or covers for te nigt to retain te eat, if necessary. Maintain a minimum of 12 ours continuous artificial illumination per 24 ours, wit 1 2 fluorescent tubes (400 lux) per tank. Alternatively, expose larval tanks to natural potoperiod and dayligt. Pour eggs carefully into te larval tanks, using buckets or beakers, to acieve a density of 0.3 to 1 egg per ml. Gastrula Figure 26. Life cycle of cultured sandfis. Te larval development of sandfis consists of te auricularia (feeding stage) larvae transforming into non-feeding doliolaria larvae before settling as pentactula larvae. Figure 26 illustrates te life cycle of cultured sandfis from spawning to post-settlement. LARVAL REARING 19

21 Stage Fertilised egg Blastula Gastrula Auricularia larvae early mid late Doliolaria larvae Pentactula larvae Juvenile Time after fertilisation 0 40 min to 3 ours 24 ours 2 days 4 days 5 6 days 10 days days 15 days Figure 27. Gastrula stage. Size range: µm Auricularia larvae Main features: Transparent slipper-saped larvae wit ciliated bands (for locomotion) A single pre-oral anterior lobe and anal posterior lobe Digestive tract complete: mout, oesopagus and stomac Slow moving continuous activity Pelagic and actively feeding on microalgae Duration of tis stage: 8 days Size range: µm Figure 29. Mid auricularia larva. Size range: 853 µm 1.1 mm Diameter of yaline speres: µm Figure 28. Early auricularia larva. Figure 30. Late auricularia larva. 20

22 Doliolaria larvae Main features: Dark-brown, barrel-saped larvae wit 5 ciliated bands around body Rapid canges occur inside te body and all adult features begin to form Larvae wit 5 yaline speres on eac side Diameter of yaline speres: µm Sort transitional pase wit decreasing size before metamorposis and settlement Fast moving Duration of tis stage: 2 3 days Pentactula larvae Main features: Dark tubular-saped larvae wit 5 tentacles at te anterior end and a single posterior foot (for locomotion) Rapid and differential growt Moving and crawling over te edge and bottom of tank, and settlement surfaces (e.g. diatom plates) Duration of tis stage: variable Bentic, crawling and feeding larvae (feed on bentic diatoms) Size range: µm Pelagic and non-feeding larvae Size range: µm Figure 32. Pentactula larva. Juveniles Figure 31. Doliolaria larva. Main features: Same sape as adult but wit two long tube feet at te posterior end for early juveniles Slow moving and strongly attaced to settlement substrata Growt to 4 5 mm in 1 week Settled and feeding stage (feed on bentic algae and detritus) Average initial size: 1 mm LARVAL REARING Figure 33. Juvenile. 21

23 How sould larvae be reared? Figure 34. Seawater filtration system. Seawater Seawater sould be sand filtered, ten passed troug 1-µm filter bags or cartridges and finally sterilised by UV. Seawater parameters sould be maintained as follows: Temperature Oxygen (DO) Salinity ph C 5 6 ppm ppt Ammonia mg/m 3 Cleaning tanks Sipon te tank base and yellow patces of dead larvae daily for te first four days of larval rearing. Healty larvae stay in te water column, wereas deformed or dead larvae are found in te lower water column or settled on te bottom. Sipon any pink patces resulting from te development of bacteria. Dead larvae, faeces from larvae and overfeeding produce bacteria during te advanced stages of larval rearing. Illumination Place a lid or a cover on top of te larval tanks for te first two days to keep eggs and early larvae in darkness. 22

24 Water cange Add etylenediaminetetraacetic acid (EDTA) (5 g/m 3 ) wen larval tanks are first filled. EDTA is a cemical used to bind wit eavy metals naturally present in seawater; ig concentrations of eavy metals can be armful for larvae. EDTA renders eavy metals armless. Do not cange te water in te larval tanks until day 2 (i.e. two days after fertilisation). Tere are tree protocols for canging water: Protocol 1: Partial water cange From Day 2 to pentactula stage Cange 30% of te water using a 100-µm mes outlet screen inside te larval tanks eac day. Use a gentle water flow (maximum 2 L/min). Add EDTA after eac water cange at a rate of 5 g/m 3 of te added water until pentactula stage. Protocol 2: Complete water cange Day 2 Day 4 Day 6 During and after late auricularia stage Complete water cange (100%) every second day until late auricularia stage. Drain te tank completely troug a 100-µm sieve immersed in a bowl, at a maximum flow rate of 5 L/min. Transfer larvae periodically from sieve to aerated containers, using beakers. Clean te empty tank. Fill it wit 1-µm filtered and UVsterilised seawater. Add EDTA at a rate of 5 g/m 3. Transfer and stock larvae at a density of larvae/ml. Cange water daily using a flow-troug system at a flow rate of 200 ml/min wit a 100-µm mes outlet screen inside te larval tanks. Protocol 3: Partial water cange wit te antibiotic Erytromycin Day 2 Day 4 Day 6 Day 8 Cange 30% of te water using a 100-µm mes outlet screen inside te larval tanks. Use a gentle water flow (maximum 2 L/min). Add EDTA after eac water cange at a rate of 5 g/m 3 of te added water until pentactula stage. Add Erytromycin* at a rate of 2 g/m 3 after eac water cange. From Day 10 Carry out a daily water cange (30%). * Handle te antibiotic carefully. Use protection for te ands and face (no direct inalation). Erytromycin is used to prevent bacterial infection but is not always successful. Avoid Introducing undesirable organisms, suc as copepods and ciliates, during water canges, and from algal cultures. Practise water cange wit 1-µm filtered and UV-sterilised seawater. Use ealty algal cultures witout ciliates. Copepods can be removed by cemical treatment wit te insecticide Dipterex (common name, Triclorfon) at 1 3 ppm for 1 3 ours followed by rapid dilution (50 to 100% water cange); it is effective in killing te swimming stages of copepods but not te eggs. Contamination from one tank to anoter by torougly rinsing all materials wit freswater before and after use and storing tem in containers wit clorinated water. LARVAL REARING 23

25 Feeding Figure 35. Feeding auricularia larvae. Start feeding at day 2. Increase te quantity of microalgae gradually from to cells/ml. Continue feeding as long as auricularia larvae are still present in te water column. Examine te gut content of larvae under a microscope and estimate te residual algae in te water to adjust te amount of food. Wellfed larvae ave guts tat are brown or golden in colour. Provide food twice a day after te water cange. Feeding rates for microalgae commonly used for larval rearing: Hatcing day* Larval stage Feeding rate (cells/ml) 2 Early auricularia Mid auricularia Mid and late auricularia Late auricularia For auricularia larvae, te main microalgae used are: Caetoceros muelleri, C. calcitrans, Isocrysis aff. galbana, Rodomonas salina and Tetraselmis sp. Availability of algal species may vary sligtly between locations. A mixture of algae is better tan using a single species for rearing larvae. C. muelleri and R. salina, given in equal parts, are optimal for larval culture of sandfis. Also, Isocrysis aff. galbana given for te first days and ten mixed wit Caetoceros sp. four or five days later is an adequate diet. *Day 0 is fertilisation. Avoid Hig algal concentration (> cells/ml). Tis can inibit growt and development of larvae and decrease te survival rate. In cases of overfeeding or algal blooms in ig ligt areas, reduce te amount of feed given, and use a ig rate of water cange. 24

26 Te first doliolaria larvae appear! Wat do tey need to settle? Doliolaria larvae look for a favourable substratum to settle on and metamorpose into pentactula larvae. If suitable conditions are not found, doliolaria larvae continue swimming for a long time, delaying settlement. Suitable settlement surfaces must be provided. Covering larval tanks elps acieve settlement because doliolaria larvae are attracted to ligt and will oterwise aggregate near te surface. Metods used to induce settlement of doliolaria larvae include te addition of: Algamac 2000 at a concentration of g/m 3 /day. Seagrass leaves, wic promote a distinctive biofouling layer. Suitable settlement surfaces include: plastic seets (PVC, polytene or polypropylene), fibreglass plates, mes screens, and roug surface tiles suspended in te water. Tere are four ways to prepare te settlement surfaces: Immerse tem in cultures of diatoms (Nitzscia sp., Navicula sp., or Platymonas sp.) for a few days. Add extracts of filtered seaweed Sargassum sp. or seagrass (Talassia empricii, Enalus acoroides) over a period of 4 5 days to form a fine coating on te settlement surfaces. Paint te surfaces wit Spirulina (1 2 g dry powder/ m 2 ), and ten leave tem to air dry before use. Immerse te surfaces for 4 10 days in outdoor tanks under partial sade (50 75%) in running 1-µm filtered seawater to promote natural conditioning wit diatoms. Settlement surfaces sould be transferred into te larval tanks wen te first doliolaria larva is observed. Possible problems Te biofilm on settlement surfaces comes off easily after 4 5 days in te larval tanks, in sady conditions. Settlement surfaces conditioned in unfiltered seawater can introduce predators to te larval tanks. Avoid contaminating larval tanks wit unwanted organisms, suc as copepods and protozoa. Figure 36. Settlement surface structure conditioned wit cultured diatoms. LARVAL REARING 25

27 Feeding pentactula larvae and juveniles Wen conditions are suitable for settlement, doliolaria larvae usually disappear from te water column in around tree days. At tis stage, tey are settled and begin to metamorpose into pentactula larvae. Pentactula larvae need food: fres and dried algae are an important source of food for pentactula larvae, and for juveniles up to at least 50 mm in lengt. Feed wit cultured diatoms (Nitzscia sp., Navicula sp.) daily from te doliolaria stage in larval tanks, and for te first mont in nursery tanks. Add sodium metasilicate (5 g/m 3 ) and fertiliser (7 g/m 3 ) once a week to promote growt of diatoms in te tanks. Supplement feeding wit commercial dried algae (Algamac 2000, Spirulina). Daily feeding rates are summarised in te table below. Hatcing day Stage Algamac 2000 Spirulina 10 Doliolaria 12 Pentactula After 12 Pentactula and juveniles After 20 Juveniles 0.5 g/m g/m g/m 3 After 30 Juveniles Up to 1 g/m 3 Up to 1 g/m 3 Oter possible feeds: A fine paste of sieved (40 80 µm) seaweeds (Sargassum sp., Halimeda sp.) and seagrass (e.g. Syngodium isoetifolium). A fine-grade srimp starter pellet at a daily rate of g/m 3 in te nursery tanks. Figure 37. Early juveniles in larval tank. 26

28 Review of daily tasks during larval rearing Record temperature, oxygen and salinity twice a day, in te morning and afternoon. For te first 3 4 days, purge te tank bottom troug te central drain pipe. Turn off te aeration for a few minutes. Was te air diffusers wit freswater. Gently sipon te bottom of tanks, including any yellow or pink patces, into a 100-µm sieve. Examine te residual dirt under a microscope to ceck for dead larvae. Turn te aeration back on. Estimate larval density in te tanks: Take a sample of te water column wit a 250 ml beaker. Pipette a 1 ml aliquot and place on a counting camber (e.g. Sedgewick- Rafter camber). Count te larvae in tree samples of 1 ml, Alternatively, take a sample of te water column in a test tube and count te larvae. Do tis tree times. Estimate average larval density per ml. Observe larval development: take a sample of larvae from te water column by gently submersing a small 100-µm sieve. Pipette a few of tem and transfer into a small beaker. Transfer 1 ml of te sample to a counting camber. Examine te larvae under a microscope. Formalin is usually added to fix te larvae. Record te lengts of 10 larvae per tank and te proportion of normal and abnormal larvae. Apply te water cange protocol. On days wen te water is canged, place an air diffuser below te 100-µm immersed sieve to avoid aggregation of larvae on te sieve. Wen te water cange is finised, take a sample of te water column in a small beaker or test tube. Count te concentration of algae under a microscope. Add microalgae to te tanks to acieve te desired feeding rate. Remove te sieve at te outlet. Rinse it wit clorinated freswater. Figure 38. Recording water parameters Summary Maintain constant water temperature and salinity between buckets and larval tanks wen transferring eggs. Early sandfis larvae, called auricularia larvae, are motile and planktonic. After rearing days, auricularia larvae metamorpose into active swimming doliolaria larvae, wic settle on te walls and floor of te tank and on oter settlement surfaces. Doliolaria larvae metamorpose into bentic, slow-moving pentactula larvae. Juvenile sea cucumbers, saped like adults, appear from day 15. Larval tanks sould be cylindrical (up to 2 m 3 ) wit conical bottoms and supplied wit moderate aeration. Seawater must be 1-µm filtered and UV-sterilised. Larval tanks sould be illuminated for 12 ours a day, or ave natural potoperiod. Lids or covers are used for te first two rearing days and during te doliolaria stage. Water cange and feeding start at day 2. Te water cange can be partial or complete. Cultured microalgae are added at te rate of cells/ml, increasing to cells/ml as long as auricularia larvae are still present ins te water column. After rearing days, doliolaria larvae appear and look for suitable settlement surfaces before metamorposing into bentic pentactula larvae. Stimulants and conditioned settlement surfaces encourage settlement. Pentactula larvae and juveniles feed mainly on diatoms (Nitzscia sp., Navicula sp.) and commercial dried algae (Algamac 2000, Spirulina). LARVAL REARING 27

29 28 Figure 39. Preparation of nursery tank includes inoculation wit diatoms and insertion of settlement surfaces (plates).

30 NURSERY Culture of early juveniles At days old, juveniles are transferred from larval tanks into nursery tanks. Nursery tanks are raceways, bigger pools or large tanks of 6 to 10 m 3, usually made of fibreglass, flexible PVC-clot liner or concrete. Te water in te nursery tanks sould be at least 60 cm deep (maximum 1 m). Nursery tanks need to be conditioned prior to te transfer to ensure food is available for te juveniles. Procedure for conditioning nursery tanks: Clean all tank surfaces. Install te aeration system and bare, clean settlement surfaces. Fill te tank wit 1-µm filtered and UV-sterilised seawater, fully immersing te settlement surfaces (i.e. a dept of cm). Inoculate te water wit fres diatom cultures at a rate of 6 7% of te total volume of water in te tank. Add sodium metasilicate (5 g/m 3 ) and a general fertiliser (7 g/m 3 ). Switc on te ligt. Turn off te water flow for te first 3 4 days to allow a diatom coating to develop on te plates or oter settlement surfaces and tank walls. Maintain moderate aeration and mix te water daily. Keep te water temperature constant and warm (26 28 C). After conditioning, te nursery tanks are ready to receive pentactula larvae and early juveniles. nursery Figure 40. Nursery tanks. 29

31 Detacing and counting juveniles before transfer to nursery tanks Pentactula larvae and early juveniles are caracterised by a very wide size range and are difficult to detac. Te transfer procedure involves detacing juveniles from te larval tanks, estimating teir abundance, and ten putting tem into te nursery tanks. Figure 41. Detacment by draining. First, remove te settlement surfaces and attaced animals from te larval tanks wile olding a plastic tray underneat to catc any juveniles tat fall off te plates. Transfer te plates and any loose juveniles to te nursery tanks. Second, remove te juveniles from te walls of te larval tanks. Tere are two metods for detacing juveniles from tank walls. 1) Drain te larval tank completely troug te bottom pipe and by siponing te floor on to mes sieves (0.3 to 1 mm) placed in bowls. Use a gentle jet of seawater to detac any remaining animals. 2) Add potassium cloride (KCl) to te water using a 1% concentration. Leave for 10 minutes and ten replace wit normal seawater. Te juveniles will detac rapidly wen te new seawater is added. Drain te tank onto mes sieves placed in bowls. Use a spray of 1% KCl to detac any remaining animals. Estimate abundance by direct counts of juveniles on plates and sieves. 1) For plates, count all juveniles on several individual plates, selected randomly. 2) For sieves, count only a alf or quarter of te total surface were tere are ig densities of juveniles. Use abundance estimates to calculate te survival from te egg to pentactula stage for eac larval tank. Survival is usu ally variable and rarely iger tan 1 2%. Use abundance estimates to calculate te initial density of juveniles placed in nursery tanks. Possible problem Te wide variation in te size distribution of juveniles can lead to brased estimates of abundance; tousands of pentactula larvae are too small to count witout arming tem. Estimates of abundance become more reliable as te juveniles grow larger. Figure 42. Detacment by siponing. 30

32 Te two stages of te nursery pase Figure 43. Nursery tank wit settlement surfaces. First nursery pase (first mont) Te early juvenile stage (<5 mm lengt) is a vulnerable and critical pase. Substantial mortality can be expected during te week following te transfer of juveniles, due to andling and ig density. Juveniles are maintained in bare tanks until tey reac a size of about 1 g. Juveniles sould grow to mm (0.3 1g) in 30 days. However, ig densities can affect growt and survival. To avoid suc problems, initial density sould be between 500 and 700 juveniles per m 2 of tank floor. At a density of 500 juveniles per m 2, te survival rate can be up to 50%, and te growt rate can be mm/day after one mont. Second nursery pase At a size of 20 mm or 1 g, juveniles are transferred to grow on sand because tey can now ingest large amounts of sediment. A tin layer of sand (3 5 mm) is evenly distributed on te bottom of nursery tanks. Te sand sould be cleaned and ten enriced wit mud or food (e.g. dried algae). After 1 2 monts of grow-out on sand, juveniles need to be reared at optimum densities of juveniles/ m 2. Removal of large juveniles will an increase in te growt of smaller individuals. After 2 monts, te growt rate is 0.5 mm/day, but will slow down if densities reac 225 g/m 2. Te survival rate of juveniles usually exceeds 50% wen tey are >20 mm. nursery 31

33 Review of tasks during te nursery pase Daily tasks Record temperature, oxygen and salinity twice a day, in te morning and afternoon. Was te screen placed at te tank outlet. Cange te water, using filtered seawater (1 µm for te first two monts, µm later), wit constant flow-troug of 6 L/min ( %). In case of restricted seawater supply, flow-troug can be reduced to 3 L/min (40 50%) every nigt. Hig flow rates are generally better tan low rates. Stop te water flow for a few ours for feeding. Feed wit diatom cultures, dried algae, seaweed paste and/or fine-grade starter srimp pellet. Weekly tasks For te first weeks, leave animals in sade; tey avoid brigt conditions and prefer sady sides of settlement surfaces. Add diatom cultures during te first mont. Add sodium metasilicate and general fertiliser. Mix te water. Invert all settlement surfaces. Diatoms grow in abundance on te upper side but not on te lower side. Pentactula larvae and juveniles stay at te bottom. Take a sample of 30 juveniles per nursery tank to estimate te growt rate. Measure total and individual weigts and calculate te average. Examine te pysical condition of a few animals under a microscope. Montly tasks Estimate te number of juveniles per nursery tank to grade densities. Sample over 5% of te tank area, including te settlement surfaces, using small quadrats (20 cm x 20 cm). In cases of ig density, collect juveniles wit a sipon and and nets. Distribute surplus juveniles to a new nursery tank. Possible problem Unfiltered seawater can result in better growt tan filtered seawater, but predators suc as copepods can quickly contaminate te tank. Copepods compete for food wit te juveniles. Figure 44. Copepod contamination. Summary Juveniles can be transferred to nursery tanks after days. Nursery tanks are raceways, pools or tanks of 6 to 10 m 3. Tey must be conditioned before transferring te animals. Conditioning consists of developing a diatom coating on te tank walls and settlement surfaces. Water excange is flow-troug or partial, wit filtered seawater. Transfer of juveniles from larval tanks to nursery tanks requires detacing, counting and grading te animals. Spraying te walls of te tank wit seawater after complete draining by siponing, or cemical detacment using potassium cloride, are two metods of detacment. Tere are two nursery pases. During te first pase, te juveniles are eld in bare tanks. After about one mont, wen tey are 20 mm or 1 g, tey are transferred to tanks wit sand substrata (second nursery). Grading consists of reducing te density of juveniles over time. Initial densities range from 500 to 700 juveniles per m 2 in bare tanks; optimum densities vary from 100 to 300 juveniles per m 2 in tanks wit sand (second nursery). 32

34 GROW OUT OF JUVENILES Te following results are from experiments in Vietnam and New Caledonia. Results can be expected to vary from place to place. Figure 45. Hapas (1 m 2 ). Figure 46. Bag nets (4 m 2 ). In net pens in ponds In New Caledonia, to overcome space limitations in nursery tanks, juveniles mm long (0.5 1 g), but as small as 5 mm were transferred to apas (1 m 2 net pens wit fine mes) in earten ponds at densities of 150 juveniles/m 2. Survival averaged 97% over 23 days and growt rate was iger (0.1 g/day) in apas wit artificial seagrass. Larger juveniles (1 2 g) can be placed in bag nets (4 m 2 net pens wit coarse mes) in earten ponds at densities of 150 juveniles/m 2. Feeding was not necessary in ponds wit good natural productivity. Wen productivity was low, good growt was obtained by adding ground srimp pellets (cicken manure or Sargassum sp. can also be used, but growt is not as good). Addition of sandy-mud substrate to te bag nets did not improve growt or survival. Growt rate averaged g/day over 3 weeks. Weigt range Net pens Net mes g Hapa 1 m 2, 1 m deep µm 1 2 g Bag net 1 4 m 2, 1 m deep 1 mm Possible problems To reduce risks of low oxygen levels, ig temperature and anaerobic sediment, raise te floor of te apas and bag nets off te substrata. Measure te temperature, oxygen and salinity of te pond water regularly. Some netting materials are too abrasive for sandfis juveniles, or too fragile to be durable. Tentex is a suitable net material. Nets become dirty after several weeks in ponds. Brus nets regularly to allow free water flow. GROW OUT OF JUVENILES 33

35 Growt rate (g/day) Figure 47. Sandfis juveniles grown in an earten pond. In ponds In Vietnam, large sandfis ( g) stocked in ponds ad growt rates of g/day. Tese growt rates varied inversely wit stocking density in te range g/m 2. Survival was ig (88 97%) until te start of te wet season wen massive mortality occurred due to stratification and letally low salinity. In New Caledonia, juveniles of 1 g ave been reared in earten ponds at a density of 1.4 juveniles/m 2. Te growt rate observed over 1 year averaged 0.8 g/day (24 g/mont). Possible problems Quality and structure of sediment in ponds, and seawater supply, will affect growt and survival. Poor environmental conditions can lead to excessive growt of filamentous algae and anaerobic sediment. Stratification of water from eavy rain leads to low oxygen levels, ig temperatures and inadequate salinity (below 20 ppt). Poor environmental conditions in ponds can lead to development of skin lesions Mean water temperature ( C) Grow out period (monts) Figure 48. Growt rate of juveniles of 1 g reared in earten ponds in New Caledonia, and mean temperature of pond water.

36 In sea pens Figure 49. Sea pen of 500 m 2. Various sea pens and cages ave been used to grow out sandfis. Tese include 25 m 2 pens made from bamboo screens or palmira rafters erected in sallow water wit 4 mm mes; 0.6 m 2 rectangular iron cages wit 2 mm mes; 2 m 2 velon screen cages wit 7 mm mes; and 2 m 2 netlon cages wit 5 mm mes. In general, growt and survival of sandfis in sea pens is muc lower tan in earten ponds. In Vietnam, juveniles stocked at ig density (500 g/m 2 ) in sea pens survived well (98%) but did not grow. Reducing te density to 390 g/m 2 Figure 50. Juvenile sandfis (2.5 to 12 g). resulted in ig survival (90%) and growt of 1.7 g/day. Juveniles wit an average weigt of 84 g stocked at 0.73 juveniles/m 2 grew at a rate of 1.05 g/day over 5 monts wen reared in a 2000 m 2 sea pen. Te sea pen ad 8 mm mes and was placed in water m deep wit a silty-sand substratum and broken coral. In New Caledonia, survival 1 20 g juveniles placed in 500 m 2 sea pens in sallow seagrass beds at densities of 4 juveniles/m 2 was 5 9 % after 18 monts. Summary Juveniles as small as 5 mm can be transferred to apas (net pens wit fine mes) in earten ponds. At 1 g, juveniles can be reared in large bag nets wit coarse mes in earten ponds. Hapas and bag nets sould be kept off te bottom of te pond to reduce risks of unfavourable environmental conditions. Large juveniles (>1 g) can be released directly into earten ponds. Juveniles can be stocked in sea pens. Less management is required, but growt and survival are far lower tan in ponds. GROW OUT OF JUVENILES 35

37 PROBLEMS AND POSSIBLE SOLUTIONS PROBLEMS AND POSSIBLE SOLUTIONS Obtaining eggs: In many countries, eggs can only be obtained during a relatively sort spawning season. Tis limits te scope for farming and sea rancing sandfis. Metods need to be developed to extend te spawning season by a few monts. Size of rearing tanks: Small larval rearing tanks require considerable labour and are prone to variation in temperature. Use of larger larval tanks tat require only partial water canges eac day will improve te efficiency of larval rearing and lead to iger larval survival troug reduced temperature-induced stress. Use of dayligt instead of artificial ligt is also an advantage. Production of large volumes of algae: Algal production can be increased by using large outdoor cultures. However, keeping larger cultures free of contamination by copepods requires systems for adequate filtration, UV treatment and sterilisation by clorination-declorination. Limited space for nursery tanks: As larger nursery tanks are required for te first nursery stage, tis can often result in substantial costs. However, space can be saved by using additional conditioned plates and by grading to remove larger animals for te second nursery pase. For te second nursery, bag nets in ponds or seabed cages can be used to replace nursery tanks. 36 Figure 51. Earten pond suitable for installation of apas and bag nets.

38 PROMISING APPLICATIONS FOR HATCHERY-REARED SANDFISH Provided metods can be developed for releasing cultured sandfis in te wild tat ensure a ig proportion of tem survive, atcery-reared juveniles could be used for restocking, stock enancement and sea rancing programmes. Restocking involves releasing cultured juveniles to restore te spawning biomass of a severely depleted fisery to a level were it can once again provide regular, substantial yields. Stock enancement is designed to increase te productivity of a fisery tat is still operating reasonably well by overcoming sortfalls in te natural supply of juveniles. Sea rancing differs from restocking and stock enancement in tat tere is no intention of allowing te released animals to augment spawning biomass or particular age-classes witin te population. Instead, animals are placed in te wild to be gatered at a larger size in put and take operations. Te availability of cultured juvenile sandfis also provides potential for farming tis species in earten ponds or sea pens. However, sandfis sould not be reared in ponds togeter wit srimp because srimp prey on sandfis. Te following questions need to be answered to fully assess te potential for pond farming of sandfis: Wat is te optimal density for stocking sandfis into ponds including tose previously used to farm srimp? Do sandfis ave a beneficial effect on te sediments of recently arvested and filled srimp ponds? Do sandfis need to be tinned out, or does food ave to be added, later in te production cycle to maintain good growt rates? If enricment of sediments is needed, wat is te best metod? Are some sediments better tan oters for rearing sandfis? PROMISING APPLICATIONS FOR HATCHERY-REARED SANDFISH Figure 52. Hatcery-reared sandfis. 37

39 38 Figure 53. Large volumes of cultured microalgae in 150 L bags in an indoor culture room wit artificial ligt.

40 ANNEX 1: ALGAL CULTURE Maintenance of algal cultures Purcase of stock algal cultures Hig quality algae to start new cultures are available from te following organisations: CSIRO Marine and Atmosperic Researc, GPO Box 1538, Hobart, Tasmania, 7001 Australia Culture Collection of Algae and Protozoa, SAMS Researc Services Ltd, Dunstaffnage Marine Laboratory, Oban, Argyll PA37 1QA, Scotland Plymout Culture Collection, Marine Biological Association, Citadel Hill, Plymout, PL 2PB, U.K. Maintenance of stock cultures Store te stock cultures in Erlenmeyer flasks (250 ml) in sterile and static conditions, close to artificial ligt in an air-conditioned room (20 24 C). Keep two series of stock cultures of eac species to reduce te risks of contamination and complete loss of algae. Swirl te stock cultures by and every day to maintain te algae in suspension. Sub-culture te stock cultures once a week (or no longer tan 2 weeks). Conditions required for te algal room (volumes of algae <500 L) Avoid contamination by cultures of oter organisms, i.e., ave a room exclusively for algal culture. Maintain air temperature between 20 and 24 C by air conditioning. Provide ligt by fluorescent tubes ( cool wite or dayligt ) 14 ours per day. Provide aeration for large flasks, carboys and bags. Use a laminar flow or simple glass cabinet for te sterile transfer of algae, if possible. Conditions required for large volumes of algae (500 L) For indoor cultures, provide artificial ligt and continuous mixing of water. For outdoor cultures wit natural dayligt, provide continuous mixing of water and protection against rain. Possible problem Ciliates could be introduced troug te air line due to ig levels of moisture. To avoid tis contamination, place a filter of µm in te air line before it enters te culture. ANNEX 1: ALGAL CULTURE Figure 54. Stock cultures of algae in Erlenmeyer flasks. 39

41 Preparation and inoculation of algal cultures Preparation or purcase of culture media Te most common medium used in te culture of microscopic algae is Guillard s f/2 medium. However, wen using tis medium te quantity of all ingredients sould be doubled; f/2 medium becomes f medium. Tis f medium is more appropriate for culturing Rodomonas salina. Prepare te f/2 (Guillard s) medium (nitrate, pospate, trace metal mix, ferric citrate, vitamins). A solution of sodium metasilicate is added only for cultures of diatoms. Use te f medium for Rodomonas salina culture. Alternatively, te media can be purcased from: Culture Collection of Algae and Protozoa, SAMS Researc Services Ltd, Dunstaffnage Marine Laboratory, Oban, Argyll PA37 1QA, Scotland. Store te media in a dark, clean place in a fridge. Preparation of flasks and 10-L carboys Was te culture vessels wit detergent, rinse and leave to dry. Fill flasks and carboys wit µm filtered and UV-sterilised seawater at low salinity (25 30 ppt). Add te culture medium. Add te glass tubes (aeration line), cotton and foil caps. Autoclave at 121 C for 20 minutes. Store flasks and carboys for at least 24 ours before use to overcome te temporary rise in ph. Inoculate wit algal cultures. Figure 55. Carboys (20 L). 40

42 Preparation of 20-L carboys, bags and tanks Was wit diluted clorine (10%), rinse and leave to dry. Fill wit 1-µm filtered and UV-sterilised seawater. Add clorine (12.5% active) at a concentration of 0.2 ml/l for a minimum of 4 ours, or overnigt for up to 24 ours witout aeration. Turn on te aeration for minutes to remove residual clorine. Add sodium tiosulpate (stock solution at 250 g/l) at a rate of 0.2 ml/l for 1 our to neutralise residual clorine. Conduct a clorine test. After several minutes, add agricultural fertiliser suc as Aquasol (2.5 g/100 L). For diatom cultures, also add sodium metasilicate (3.75 g/100 L). Inoculate wit algal cultures. Sodium metasilicate is added for diatom cultures suc as Caetoceros sp., Navicula sp. and Nitzscia sp.; diatoms use te silicate for production of an external sell. ANNEX 1: ALGAL CULTURE Figure 56. Algal culture in a 500 L tank. 41