Part 6
Hatchery operation: culture of spat in remote setting site, in the hatchery and in nurseries

6.1 INTRODUCTION

The word "spat" is an old English term applied to the early juvenile stage of bivalve development and is perhaps the most commonly used term applied to juveniles in hatcheries. It relates to bivalve larvae that have set and undergone metamorphosis.

Another frequently used term for early juveniles is "seed" and this word is used to describe juvenile products supplied by hatcheries to shellfish farmers.

The extent to which hatcheries are involved in growing spat beyond pediveliger stage larvae varies considerably and relates to the preferences of the growout industry. Provision of eyed, pediveliger Pacific oyster larvae for remote setting at oyster farms is common practice on the Pacific coast of North America. Hatcheries provide the mature larvae and the farmers themselves set them and grow the spat for seeding oyster beds or in suspended culture. Details of the methodology are given in 6.2.

In other parts of the world hatcheries set the larvae and grow the spat to a size that growers are comfortable to handle and grow. This may be when spat are 1 to 2 mm in shell length or often larger. The size that spat is supplied is largely dictated by the requirements and maturity of the growout industry. Hatcheries would prefer to deliver them at the smallest size possible because the economic implications of growing them further within closely controlled conditions are significant. It takes only a relatively small tank volume and a comparatively small quantity of algae to grow larvae and set a million spat but once they are set costs associated with growing them escalate rapidly.

Consider the requirements of 1 million oyster spat. At 1 mm shell length, individual live weight (shell and body) is approximately 0.3 mg. Clam and scallop spat are about 30% lighter than oyster spat for a given shell length within the size range grown in hatcheries. The biomass (total live weight) of one million oyster spat is therefore 0.3 kg. Growth rate of spat in closed seawater systems (systems without continuous water exchange) is biomass dependent. To ensure commercially acceptable (not maximum) growth rates, spat need to be grown at a maximum of 200 g live weight biomass per 1 000 l (0.2 kg per m³). This is the biomass at the beginning of a weekly period irrespective of the size spat are and allows for significant growth during the course of the week. Biomass is reduced at the end of a weekly period by distributing the spat at 0.2 kg per m³ in a greater tank volume - either more tanks of the same size or larger tank systems.

Growth rate decreases significantly as stock density per unit volume increases. At 0.4 kg per m³, for example, newly metamorphosed Manila clam spat will grow to only about 0.5 mm in a 6-week period compared to a mean shell length of 1.4 mm at 0.2 kg per m³. This is at the same temperature and with food ration calculated on the basis of biomass (section 6.4). It is not important to know the numbers of spat at this stage. Total live weight biomass is the criterion upon which food ration is based, i.e. the weight of shell, body and water contained between the shells. Section 6.3.5. describes protocols for grading and estimating seed.

Returning to the example, one million 0.3 mg - 300 g in total - oyster spat will need a minimum culture tank volume of treated and heated seawater of 1 500 l. By the time they reach 5 mm shell length individual live weight has risen to approximately 32 mg. The biomass of one million 32 mg spat has increased to 32 kg and the volume of treated and heated water required to grow them is now 160 000 l (Table 14). Food requirements increase proportionately (section 6.4). For example, 1 million 0.3 mg spat require 17 g dry weight of algae per day, which is equivalent to 85 700 million cells of Tetraselmis suecica, or 85.7 l of harvested culture at 1 million cells per ml. At 5 mm shell length, food requirement for the same number of spat has risen to 9 130 l Tetraselmis at the same harvest cell density (Table 14). The 4 mm increase in shell length is associated with more than a 100-fold increase in biomass and the same increase in food is required. Clearly, there is a limit to the size hatcheries can grow the spat in terms of spatial requirements to accommodate them, the need to treat and heat seawater and the volumes of food required to feed them.

Table 14: Tank water volume and daily food requirements for bivalve spat of different sizes when grown at a biomass of 200 g live weight per 1 000 l (0.2 kg per m³). Data are for oysters but relate to other bivalves where clam and scallop spat are approximately 70% of the weight of oyster spat for a given shell length.

* Daily food requirement calculated as l of Tetraselmis at 1 x 106 cells per ml

Various solutions and approaches are adopted to overcome cost limitations to growing spat within the hatchery. These are described in section 6.3. Most commonly, spat are grown in closely controlled conditions to a size at which they will be retained by either a 1 or 1.5 mm mesh screen at 2 to 3 mm shell length. They are then transferred to outdoor nursery systems, which may be part of the hatchery operation or belong to a farmer or group of farmers. Or such nurseries may be part of a vertically integrated company operating a hatchery and producing seed for its own growout requirements. Outdoor nurseries are designed to protect small spat from predators while growing them at high density to a size at which they can be transferred to sea-based growout. Key features of outdoor nurseries are that they operate on the flow-through principle, utilizing natural phytoplankton productivity to provide the food supply (section 6.6). They may be land or sea-based and if they are located on land the source of seawater may be from artificially dug or natural ponds that can be emptied and re-filled from the sea. Measures are often taken to enhance algal productivity of the ponds by the application of fertilizers (see 3.4.6).

The following section deals with the special case of procedures for setting mature larvae at remote sites and growing them from the time they set to the time they begin growout to market size. Subsequent sections follow the various methods in common usage to grow recently set spat to suitable sizes within a hatchery until they are sold directly to farmers or transferred to land- or sea-based nursery systems.

6.2 REMOTE SETTING

The technique by which eyed larvae are supplied by hatcheries to farmers who set and grow them on the Pacific coast of North America is described here. This is a special case and its use is commercially confined to the Pacific oyster, Crassostrea gigas, although it is equally as applicable to other oyster species in other parts of the world.

6.2.1 Background

On the Pacific coast of North America most oyster production is from intertidal bottom culture and more recently floating culture. Originally juvenile oysters were imported annually from Japan and spread on a grower’s lease for growout. Juvenile oysters were set on bivalve shell, usually old scallop shell, but this supply of seed ended when it became too expensive. Breeding areas were located along the Pacific coast and used to augment seed imported from Japan and eventually to replace it. Oyster larvae were generally set on bivalve shell, mostly old oyster shell, and allowed to grow on the shell in breeding areas until the juveniles attained a shell length of about 1 cm at which time the cultch with attached juvenile oysters was transported to a grower’s facility. In intertidal bottom culture the seed was spread either directly in growout areas or held on seed ground for upwards of a year and then spread in growout areas. In floating culture the cultch with juveniles could be strung on wires or ropes and suspended from floats or longlines. The method was generally effective for reliably supplying growers with their seed requirements, but there were disadvantages with the system. The main disadvantage was that failures or insufficient breeding occurred in breeding areas in some years. Consequently growers did not have sufficient seed for growout operations. Cost was another problem. Shell is bulky and heavy and it was expensive to move large quantities of juveniles attached to oyster shell. Another disadvantage was that the seed could generally only be moved during the cooler, wetter months, October and November, and this was frequently inconvenient for a grower who wanted the seed at other times, particularly during spring and early summer. It was also impossible to select for a particular strain or race of oyster in natural breeding areas.

Studies showed that mature Pacific oyster larvae with well developed eye spots could be held out of water in a damp but cool condition (5-10°C) for upwards of a week. Thus it became possible to ship mature Pacific oyster larvae considerable distances, literally anywhere in the world. A grower could purchase mature oyster larvae from a hatchery whenever it was convenient, have them shipped to his/her facility and set them on the preferred type of cultch used in their growout operation. The disadvantages of previous techniques including reliability of a seed supply, cost of handling bulky cultch with attached juveniles and not being able to obtain seed when desired, could be averted. Further, a grower did not have the expense nor time consuming effort to build and operate a bivalve hatchery. The method, now widely used by growers along the North American Pacific coast, provides a convenient and efficient way to ensure a reliable and plentiful supply of oyster seed for culture operations.

6.2.2 Preparing larvae for shipment

The method, developed in the 1980’s, has been refined over the years and is straightforward. It provides good results if proper procedures are followed. Oyster larvae are produced in a hatchery and a grower will make arrangements with the hatchery to deliver the amount of larvae required at his/her facility at convenient times. Larvae are filtered on screens at the hatchery and placed in a piece of nylon mesh to form a bundle that is kept damp, a bundle about 5 cm in diameter contains about 2 million mature oyster larvae (Figure 89). The bundle of larvae is placed in a chilled styrofoam container with ice packs to maintain a temperature of 5-10°C. The container with larvae is then shipped to the grower.

Figure 89: Receiving a consignment of eyed Pacific oyster larvae wrapped in nylon mesh at a remote setting site in British Columbia, Canada.

6.2.3 Preparations at the remote site

An important consideration for a grower is site selection for the remote setting operation. Of primary concern is water quality and the criteria used in selecting a site for a hatchery apply equally to a remote setting operation. Areas with known sources of pollution need to be avoided. Salinity must be within an accepted range (greater than 20 PSU for Pacific oysters), the water should be well oxygenated and the temperature close to 20°C or above during summer months to eliminate the need to heat water. Water should be pumped from at least 2 m below the surface to reduce variations in salinity in areas of high rainfall. The water should be phytoplankton rich so that it can be used as a source of food for juveniles and reduce the need for adding food. The site must have electrical power, sufficient space for tanks and other parts of the operation, good communications so that larvae can be readily received, and the site should be close to the intertidal beach area where juveniles will be transferred and held after they are removed from setting tanks.

Tanks are constructed at a grower facility for setting the larvae. There are no set dimensions for the tanks, it depends partly on the type of cultch used, size of the operations, methods used to handle juveniles and individual preference (Figures 85 and 90). Cultch used on the Pacific coast is either old bivalve shell - mostly oyster shell - or grooved plastic pipes of about 2 cm in diameter. Bivalve shell is placed in plastic mesh ("vexar") bags that are 1 to 2 m in length and 50-70 cm in diameter. Each bag holds 100 to 200 pieces of shell. Grooved plastic pipe is usually cut in 2 m lengths. Smaller tanks may measure 1.5 x 2.5 x 2.5 m in size but may be much larger and hold 40 000 l.

Figure 90: Setting tanks at a site in British Columbia, Canada. Note the loose cultch and vexar bags filled with cultch stacked on the bank behind the tanks. Refer also to Fig. 85, Part 5.4.3.2.

The tanks can be constructed of a variety of materials, including concrete, fibreglass, or fibreglass coated wood. Regardless of the material used, tanks should be well leached before use. In temperate areas the walls of fibreglass tanks are generally insulated with styrofoam to help maintain water temperature. In some instances the tanks are also fitted with a lid to further improve insulation. A 2-cm plastic pipe with holes drilled at regular intervals is placed around the inner circumference of the bottom of the tank and serves as an air line. The water may need to be heated at certain times of the year in temperate regions. A hot water line can be run from the main facility or individual electrical heaters placed in each tank. Tanks should be constructed so they can be readily cleaned and be fitted with drain valves.

The first step when a remote set is planned is to add the cultch to the tanks so that they are as completely filled with settlement material as possible. Vexar bags with bivalve shell are stacked one on top of the other or plastic pipes bundled together in modules. The cultch, whether plastic pipes or old bivalve shell, is generally not conditioned in seawater for a sufficient period of time to acquire a biofilm. Plastic pipes are well leached before use. Shell is generally air dried and exposed to the elements for at least six months before use and then it is washed so that the surfaces are clean.

The amount of cultch required depends on size of the tanks. Generally, between 16 and 20 vexar bags of cultch will occupy about 1 m³. The tanks are filled with seawater filtered to about 50 µm either through a sand filter or by individual filter bags on each of the tanks. Seawater is heated to the desired temperature which is generally between 20 and 25ºC for Pacific oysters.

6.2.4 Receiving the eyed larvae

Mature larvae are shipped from the hatchery to the remote setting site. Two million mature Pacific oyster larvae form a ball about 5 cm in diameter when wrapped in mesh (Figure 89). Once received, they are placed in a plastic bucket with 10 l of water at 20 to 25°C and allowed to acclimate for fifteen to thirty minutes. The contents of the bucket are then poured into the tank. The number of larvae added per tank depends on size of the tank and the amount of cultch but as a "rule of thumb" 1 300 to 2 200 larvae are added per 2 m length of plastic pipe and about 100 larvae are added per piece of shell cultch. Air is turned on for about thirty minutes to ensure thorough mixing of larvae in the tank and is then turned off to allow the larvae to attach to the cultch. If pipe cultch is used, only half the larvae may be added originally and after one day the pipe modules are flipped and the remainder of the larvae are added. This helps create an even set on all surfaces of the pipe.

6.2.5 Setting the larvae and growing the spat

Oyster larvae attach to the cultch and metamorphose into spat usually within 24 hours from the time larvae were added to the tanks. Some setting can occur on the bottom and lower part of the sides of the tank but this can be avoided by painting these parts of the tank with liquefied wax (paraffin). Loose shell can also be scattered over the bottom of the tank to catch larvae that may settle there.

Once larvae have metamorphosed to spat they must be fed. When remote setting began, hatcheries supplying eyed larvae often also supplied algal paste for use as food. The algal paste was algae grown in a hatchery and centrifuged down to form a disc of concentrated algae about 12 cm in diameter and 3 cm in thickness. A portion of the paste was broken off, placed in a bucket with seawater, stirred briskly to break up the clumps and then added to the tanks. Air was turned on to ensure adequate mixing of the food in the tanks. Species used to make the algal paste were the same as those grown in the hatchery to rear larvae. Algal paste is still used by some growers but it is not as common as was the case previously. Most hatcheries now require their entire algal production for their own use and have none to ship to remote setting sites. There are companies that grow algae for sale as a concentrated slurry and this can be used as food. Many growers now culture their own algal food using standard methods as described previously. Species used vary from site to site but are the same as those used in hatcheries to feed larvae.

Water in the tanks is not exchanged for the first two or three days after setting but after that a slow flow-through of coarsely filtered seawater is begun. The objective is to acclimate the spat to local environmental conditions and also provide additional natural food. If algal food is added to the tanks, the water flow from the open environment is turned off for a short period so that as little as possible of the added food is lost.

The length of time spat are held in the tanks is variable. In early spring and late fall it may be upwards of a month but in summer if can be as brief as one week. It also depends on the schedule used at a grower’s facility as the following example illustrates.

The spat are usually 2 to 3 mm in size when they are transferred to growout. Bags of cultch with spat are placed in the mid- to lower intertidal zone on pallets to keep the cultch out of the substrate and reduce mortalities. In summer, transfer from the tanks to growout generally occurs in early morning or late evening when temperatures are lower. The time taken for the transfer should be kept to a minimum to reduce stress and mortalities. Bags can be stacked to a height of 2-3 m, depending on tidal range. Tarpaulin covers are placed over the bags to keep the spat out of direct sunlight and to reduce settlement of fouling organisms. Bags with spat are left in the intertidal area for varying periods of time and then the cultch with spat is spread either on good growing ground or is strung on ropes or wires for floating culture.

As with hatchery operations, it is important that growers keep accurate records of each set. With acquired experience they can determine optimum conditions to maximise the production of seed from larvae.

The remote setting concept was developed and perfected as a relatively inexpensive way to produce Pacific oyster seed but it could be used for clams, scallops and mussels. To date it has not been used widely for species that do not attach firmly to cultch as do the oysters.

The technology has opened new opportunities for bivalve culture worldwide. If a grower wishes to culture a species of bivalve and cannot obtain sufficient seed from local natural sources or prefers to use hatchery seed, he no longer needs to build an expensive hatchery. Arrangements can be made to produce larvae at any hatchery and ship them to the grower’s site. It is important to realize that the hatchery can be located anywhere in the world since larvae can be shipped great distances and arrive in a healthy state. Hence large, efficient hatcheries can be located at ideal sites rather than at locations that may be politically expedient but are not ideally suited for the purpose.

A distinct advantage in shipping mature larvae rather than juveniles is that larvae are grown in water that is finely filtered and may also have been sterilized with either UV-light or ozone. The danger of spreading diseases or parasites is much reduced compared to shipping juveniles that are generally grown to the desired size in the sea and may have acquired local diseases or parasites.

6.3 METHODS FOR GROWING SMALL SPAT

6.3.1 Introduction

Constraints to growing spat to a large size under closely controlled conditions in a hatchery have been dealt with in general terms in 6.1. Space, the supply of treated, heated water and the large volumes of cultured algae required are major cost considerations. Hatchery managers will know the production cost factors that need to be taken into account when fixing the price of seed. Prices will increase exponentially as size in terms of mean shell length increases and a point will be reached when growers are no longer prepared to pay for spat in the larger size categories. In developed countries with mature industries this point is generally reached when spat are 3 to 4 mm and very often when they are somewhat smaller.

Methods commonly used to handle and grow newly set scallop and clam spat were introduced in 5.4.3.2. Procedures for oysters are different but before describing what these differences are it is relevant to begin with the various tank system options for this part of the hatchery process, beginning with those used for spat set on cultch.

6.3.2 Growing systems for spat set on cultch

Tanks systems - essentially similar to those described for remote setting in the previous section - are commonly used in the hatchery for the initial stages of the growth of oyster, scallop and mussel spat set on cultch (Figure 91). They may be closed systems, i.e. with a static volume of water changed two or three times each week, or open systems operated on flow-through, depending on the extent to which the water needs heating. Very often they will be a combination of the two with aeration to mix and circulate the water and the daily food ration throughout the tank volume. Food will be added continuously in the case of flow-through. Oyster spat may spend as little as a week in these systems whereas the slower growing scallops and mussels will stay longer before transfer to the sea.

Sand-filtered water or water filtered to about 20 µm particle size is usually used at this stage so that spat can benefit from the diversity of naturally occurring algal species in the water in addition to the added ration of cultured food. Feeding is not usually closely controlled in terms of species composition of the diet and ration. Enough food will be added to the tanks to colour the water sufficiently. If the algae is grazed quickly then more will be added. If the water is heated then spat will be gradually acclimated to ambient sea temperature before leaving the hatchery.

Figure 91: Simple tank systems are used for growing spat set on cultch. They are either closed or flow-through systems or a combination of both. A - growing tanks mainly used for scallop spat set on cultch at a British Columbia hatchery. B - note that the lined plywood tanks are situated outdoors and are roofed over to shade the water surface. C - scallop spat may be set on filamentous cultch packed in onion bags, initially in tanks of the type shown in A and B at the hatchery site. D - detail of spat set on the filamentous material after a period of floating culture in the sea. E - growing mangrove oyster spat set on strings of oyster shell cultch at a Cuban hatchery. F - when the spat are 2 or 3 mm in size the strings of cultch are hung from mangrove poles on rafts located in productive waters.

6.3.3 Growing systems for unattached spat

Unattached spat (i.e. spat grown free of cultch - "cultchless") are grown in large volume tanks equipped for recirculation - often with a gradual continuous exchange of water - or they are grown in open, flow-through systems. Whichever method is used depends on species and the size of spat. Smaller spat may be grown in recirculation systems until 1 or 2 mm in size and then be transferred to flow-through to grow to 3 or 4 mm before being sold or transferred to an outdoor nursery.

The spat growing area of a hatchery may contain a number of different growing systems for spat of different sizes and species. Most commonly the systems will utilize oblong concrete, fibreglass or lined or epoxy painted plywood tanks to use space as efficiently as possible. The large tanks that act as the reservoirs have drains plumbed directly into the hatchery’s main drain since large volumes of water will be discharged periodically.

Hatchery managers have their own preferences as to the best way to handle spat of the species they produce according to cost factors and what suits the particular requirements of the local industry. As with larval culture many different approaches are taken but there are a number of common factors that apply in the basic methodology.

Oysters are completely sedentary and clam and mussel spat are mostly so once they have settled and completed metamorphosis - scallop spat are the exception. Scallop spat retain the ability to detach their byssus attachment and briefly swim in the water column to find a different location to attach. Food needs to be carried to spat of any species in the water currents. How to manage them in a convenient way and the manner in which the water - as the carrier of food - is delivered to the spat become important considerations.

Figure 92: A closed tank system designed for holding scallop spat in cylinders with a downwelling flow of water. A - spat holding cylinders are held in shallow troughs (t) stacked one above the other. B - water flows into each cylinder (cy) through a flexible tube connected to the supply line (sl). C - water returns to the reservoir tank (r) via a stand-pipe (sp) fitted in each trough which maintain the water depth in the troughs. Water is pumped from the reservoir back to the troughs. Systems of this type are also suitable for clam spat.

Spat are almost always contained in mesh based trays or cylinders in a holding tank which, if not of sufficient volume itself, is connected with a large-volume reservoir tank. Containment of spat in trays or cylinders facilitates ease of management in cleaning and grading the animals. Water with algal food is circulated by electric pump or air-lift from the reservoir to the holding tank, passed by the spat and then returns to the reservoir. Examples suitable for growing scallop and clam spat have previously been given in Figures 87 and 88. Figure 92 shows the delivery of water to each of the cylinders in the holding tank by a flexible hose attached to nipples in the delivery pipe. Water flows in from above the water surface within the cylinder at a controlled rate, downwards passed the spat and out through the mesh base of the cylinder to be returned to the reservoir by a stand-pipe or an overflow which maintains the water level constant in the holding tank. This flow pattern is called downwelling. The other approach used for oysters and clams is to reverse the direction of flow so that it enters at the base of the cylinder (or tray), passes upwards through the bed of spat and is discharged at the top, from wich it flows back into the reservoir. This is referred to as an upwelling circulation. Both of these principles are illustrated in Figure 93.

Figure 93:

A - diagram illustrating the difference in flow circulation in upwelling and downwelling spat systems. Arrows show the direction of water flow. Upwelling systems are used for oyster spat from set size upwards and for fully metamorphosed clams. Downwelling systems are used for clam pediveligers (until they have completely lost the ability to swim) and for scallops from the pediveliger stage onwards. Only rarely are upwellers used for scallops and then at a much lower biomass per unit area than for oysters and clams.

B - diagram of an upwelling tank system showing the reservoir (r) from which water is pumped (p) to a holding tank (ht) maintained at a constant water level (head) by an overflow pipe (of) through which excess water is discharged back to the reservoir. The holding tank contains a number of tall, narrow cylinders (c) with mesh bases in which the spat are held as a fluidized bed (fb). Holes are drilled in the holding tank below water level to take flexible tubes (ft) that interconnect with the cylinders. Thus, there is a head difference between the water level in the holding tank and the water level that can be maintained within the cylinders. Water flows through the mesh bases of the cylinders, up through the bed of spat and then back to the reservoir through the flexible tubes. The extent to which the spat bed is fluidized (i.e. the spat lifted by the flow) can be changed by altering flow rate.

It is quite common to use inverted, plastic soft drinks bottles of 1 to 3 l volume as upwelling cylinders. Instead of a mesh screen containing the spat, a ball or large marble is placed inside to cover the opening of the neck. This serves as a non-return valve. The flow of water from the bottom keeps the juveniles suspended in the water column inside the cylinder but if water pressure is lost the ball or marble seals off the neck so that no juveniles are lost. Discharge water from a series of upwelling bottles is passed over a screen to collect any juveniles that may accidentally float away.

6.3.4 Operation of closed upwelling systems

Upwelling is particularly useful in the culture of post-settlement oysters. Small spat are amenable to stocking in depth at high density, i.e. layered one above the other. The same applies to clam spat once they approach 0.5 mm in size. Holding the small oysters in this way with a sufficient flow of water to fluidize the "bed" of spat prevents adjacent spat from fusing together to form clusters as they grow. Cluster formation can be a problem in Crassostrea species if the spat are not kept moving - for example, if they are grown in trays with a downwelling flow. This habit is more pronounced at the high water temperature, which is generally between 22 and 25ºC for oyster spat growth. An upwelling flow is also more efficient in keeping the spat free of faecal deposits than is downwelling, where faeces tends to accumulate on and around the spat. This can result in blockage of the mesh, which is less of a problem in upwelling containers.

Figure 94: A and B - Closed upwelling systems in use for the growing of small oyster spat. The total volume of each tank unit is approximately 3 m³ and the spat holding tanks hold 10 cylinders each stocked with 60 g live weight of spat at the beginning of a weekly period. The flexible outflow tubes from each cylinder are fitted with an adjustable clamp to permit individual control of flow rate. B - water is lifted from the reservoir to the holding tank by air-lift (al). This is a 5 cm diameter pipe with an airline fitted into the base. The flow of air into the bottom of the pipe lifts a sufficient volume of water to operate the system without the need for an electric pump.

Upwelling containers (referred to as cylinders or tubes) can be of varying diameter and they are made from sections of PVC or acrylic pipe fitted with nylon mesh bases of different aperture size for the range of sizes of spat grown. They do not need to be transparent as in Figure 94, but transparency is an advantage in gauging the flow rate required to fluidize the biomass (bed) of spat contained. The flow rate needed to fluidize, i.e. lift and circulate, the bed will depend on the size/weight of the spat and on the diameter of the pipe section. The larger the spat are the more flow will be needed to fluidize the bed. Lower flows are required in narrower cylinders. Typically, a flow of 1 or 2 l per minute through 5 or 10 cm diameter cylinders will fluidize a bed of 1 to 3 mm oyster spat. A flow of between 25 and 40 ml per minute per g of spat is ideal. Beds of clam spat, which embyss together, will not fluidize. Nevertheless the method works just as well as it does with oysters. Possibly the effect of spat being aggregated together is advantageous in that it simulates conditions when they are buried in substrate. Beds of clam spat - tightly embyssed together - when held in downwelling conditions tend to act as sediment traps and meshes soon become blocked.

The numbers of spat that can be held in an upwelling tank system depends on their size/weight (Table 14). Take, for example, the system shown in Figure 94 in which the combined volume of each reservoir and spat holding tank unit is close to 3 000 l. There are 16 similar units in the hatchery. Each unit is suitable for growing a live weight biomass of 600 g. Assuming the spat to be grown are 2 mm shell length, reference to Table 14 indicates that 272 700 spat of this size will make up the initial biomass. The holding tank shown in Figure 94 contains 10, 10 cm diameter cylinders. At the beginning of a 7-day period, each cylinder is stocked with 600/10 = 60 g of spat. These will have been graded using a 1.5 mm mesh screen over another screen of 1 mm aperture upon which they will have been retained (2mm spat will not be retained by a 1.5 mm mesh). In this context, the relevance of accurately knowing the numbers of spat stocked in a unit is superseded by the importance of knowing their biomass. For further explanation see section 6.3.5.

Seawater to fill the tank units is filtered and heated to larval culture standards for spat in their first week after set. After that time, they are filled with either sand-filtered or 10 or 20 µm cartridge filtered water and the temperature is decreased by 1 or 2ºC per week to start acclimation to prevailing conditions in the nursery or sea.

At the end of the 7-day period, during which time the tank volume will have been changed twice and the spat and system cleaned at each water change, the spat are graded and redistributed again. The 600 g biomass at the beginning of the week will have doubled or even trebled for oysters by the end of the 7-day period and so will need to be redistributed between two or three 3 000 l units to grow for a further week. Spat will not have grown evenly in size during the previous week. By grading through a stack of screens production from the tank unit can be size fractionated (section 6.3.6). The growing process functions more efficiently if spat of the different size fractions (grades) are grown in separate tank units so that those in any one unit are of the same size grade.

6.3.5 Operation of closed downwelling systems

Downwelling tank systems without continuous water exchange are dealt with following the same procedures as described above. The only major difference is that the biomass of spat is spread over a much greater surface area than in upwelling systems because the juveniles - most commonly scallops - are sensitive to overcrowding. Thus, they are maintained with sufficient spatial separation to permit growth as a single layer so that individuals are not in immediate contact with adjacent spat.

Methods to maintain spatial separation vary from hatchery to hatchery and where the spat have been set on cultch procedures outlined in section 6.2.2 apply. If not set on cultch but on the mesh bases of cylinders or trays as shown in Figure 88 (section 5.4) and Figure 92 then system design and operational management details are different. The holding tanks supplied from the reservoir will need to have a large enough area to fit the numbers of trays or cylinders required to hold the biomass of spat appropriate to the total tank unit volume. For this reason, holding tanks of the kind shown in Figure 92 are shallow and are often stacked one above the other.

As in closed upwelling systems, water quality is maintained by complete water changes twice or three times each week. The trays or cylinders containing the spat are removed and each is sprayed with a seawater jet to dislodge and remove detritus adhering to the spat and to the mesh of the containers. The reservoir and holding tanks are cleaned and refilled before returning the spat containers. Seawater used may be finely or coarsely filtered depending on the size of spat. It is usually filtered to 1 to 2 µm for early-stage spat and is sand-filtered for larger spat about to be transferred to the sea. Spat are gradually acclimated to ambient sea temperature before they are transferred.

Scallop spat are not as amenable to removal from the containers for grading and size determination. Their shells are more fragile and care needs to be exercised not to damage their byssus gland or displace their shell valves and damage the resilium during removal. Gentle water jets may be used but it is more appropriate to count them if necessary in situ. This can be done as shown in Figure 88B by using a plastic sheet marked with a grid (1 cm squares) under the mesh base of a random selection of trays or cylinders. Means calculated from counting the numbers per cm² in random squares over 10% of the surface area of a selection of containers multiplied by the total area occupied by spat will give a good approximation of total numbers. Small samples can be removed to weigh and measure to track growth and biomass.

6.3.6 Grading and estimating spat

Mechanical graders are available from specialist equipment suppliers and they are applicable when millions of seed are being handled on a routine basis. Otherwise hand graders are used in the majority of cases. These can easily be made as a series of large diameter (>30 cm) fibreglass or PVC pipe sections with nylon or plastic meshes of various aperture sizes fixed to one cut face.

Spat grading is best done in water. The grading screens, each marked with the size of mesh, should fit comfortably within a plastic tray fitted with a stopper or drain valve at one end. The tray is part-filled with seawater when in use. Small numbers of spat are added to a screen of a mesh size slightly smaller than the largest individuals. The sieve is then shaken from side to side and up and down in the water until no more spat escape through the mesh (Figure 95). More spat are added periodically until all have been graded through that screen. Those retained in the sieve will need to be removed from time to time to maintain efficiency of the process. They are transferred to a screen of known (tare) weight with the same mesh size and are left dry to await estimation. The tray is then emptied and the smaller seed recovered for further grading. The procedure is then repeated with a screen of the next smaller mesh aperture, and so on.

Figure 95: Grading of spat with hand-held sieves (screens) in shallow tanks. The grading screen is rotated from side to side and up and down in the tank until all spat smaller than can be retained on the mesh fall through and collect at the base of the tank. Once grading with a particular screen is complete, the tank is drained into a receiving screen of appropriate mesh size, marked with the grade size of the spat. In this example, spat smaller than 4 mm mesh retention size will be collected in a 1 mm receiving container (of small enough mesh size to collect all that remain). The process continues with grading screens of decreasing mesh size until all spat have been fractionated into the various grade sizes.

Once separated by grade, the next task is to determine the biomass of spat in each grade. Screens containing the different grades need to be allowed to thoroughly drain until the contained spat are "damp dry." Draining can be accelerated by dabbing the bottom mesh of the screens with dry clothes or paper towelling until excess water has been removed. The screens are then weighed and the weight of the screen itself subtracted to provide the weight of spat it contains. This is the biomass of spat of that particular grade.

At the same time, numbers of spat and a check on survival can be made. Numbers can be estimated either by weight or volumetrically. The former method requires accurate balances while the latter can be achieved with simple apparatus, e.g. small plastic containers of between 1 and 5 ml volume to hold sub-samples. This method will be described.

From the screen containing the largest sized seed, fill three sub-sample containers to the brim. Empty one into a shallow white tray containing a little seawater. A Petri dish marked with a grid, observed under a low power microscope is useful for counting very small spat. Count the total number of seed in the sub-sample. If there is no dark spot within the shells (the digestive system) or if the shells are gaping open, place to one side. Record the total number of seed and the number dead. Repeat for the second and third sub-sample. Determine the total volume of spat of that grade by transferring them to graduated containers and reading the volume they occupy. From this information, the total number alive and percentage mortality can be calculated as follows:

Numbers in the other grade fractions are estimated in the same way. Smaller volume sub-samples will be necessary for the smallest seed sizes.

Estimating numbers by weight follows the same basic method with small sub-samples taken for accurate weighing from the bulk of spat in the particular grade(s). Numbers in the weighed sub-samples are counted. Once the total weight of spat in the grade has been determined, total numbers can be calculated as above. Spat of the various clam species are more difficult to grade than oysters because of their habit of attaching to one another and to the meshes and internal surfaces of the containers in which they are grown by byssus threads. Nevertheless they are handled in similar manner using water jets from a pressurized hose to separate them during the grading process.

6.3.7 Operating systems on flow-through

Tank systems of the various kinds described above are often operated with a partial exchange of water each day or on a continuous open flow. Partial or total flow-through systems are used for growing larger spat when maintaining a temperature higher than the ambient is not an important consideration, i.e. when ambient water temperature is sufficiently high to support good growth. Two advantages of flow-through are a) the increased biomass of spat that can be held and grown in the spat holding tanks and b) spat can benefit from the natural or enhanced productivity of the exchange seawater. The diversity of algal species in the generally coarsely filtered exchange water more closely resembles natural conditions as spat are gradually acclimated in preparation for transfer to growout.

Figure 96: Upwelling tank units for larger size spat operating on flow-through. A, B and C - a system for growing clam spat at high density. This system is in circuit with a 90 m³ outdoor concrete reservoir in which naturally occurring algae is "bloomed" by adding nutrients. Note the central collecting channel which carries the upwelling water flow from the cylinders back to the reservoir. This is a pumped system. D, E and F - a system for growing scallop spat at lower density. This unit is plumbed directly into the hatchery’s main seawater supply and is fed continuously from a reservoir containing diluted algal paste, which can be seen in D. Otherwise, the configuration of the system is similar to A with cylinders either side of a central water collecting channel.

It was pointed out in section 6.2.3 that the optimum biomass for growing spat in closed systems is 200 g per m³ of the total volume of the reservoir and spat holding tank combined. Consider the example of a 3 000 l tank unit given in section 6.2.4 in which 600 g live weight biomass of spat will grow at a satisfactory rate. When total tank volume is completely exchanged in the course of a 24-h period then the biomass can be approximately doubled. At this rate of water exchange - equivalent to 125 l per hour - and assuming that cultured algae is the principal food source, little food will be lost especially if it is added directly to the spat holding tank of the unit. The food ration will need to be doubled because the biomass of spat has been increased by a factor of two. At the higher spat density and the increased quantity of food, tank fouling with faeces and detritus is greater and needs to be taken into account during routine husbandry. The tank units may need to be drained and cleaned three times a week instead of twice.

Spat holding tanks operating on total flow-through are usually configured differently. Instead of being in circuit with an adjacent reservoir tank they are stand-alone units, each being plumbed directly to the seawater supply (Figure 96D). They may be located inside the hatchery or outdoors. Many hatcheries operating flow-through units plumb a water supply from outdoor shallow ponds or very large volume tanks located adjacent to the hatchery premises. They are used to bloom algae. Furthermore, the temperature of the water in these ponds will be elevated above ambient sea temperature by solar heat gain for much of the year, particularly in temperate latitudes (see section 6.6). Seawater discharged from the spat holding tanks is returned to the ponds. In this way algae is conserved.

In essence, flow-through units are little different to the concept of nursery culture, which is described in greater detail in section 6.6. Hatchery-based flow-through units tend to be used for spat in the smaller size grades and many hatcheries will also have a nursery for the further growth of larger seed in close proximity. Staff can thereby manage the entire production process from egg to larger seed with the hatchery infrastructure of equipment, laboratory space, etc., available in support.

6.4 DIETS AND FOOD RATIONS FOR SMALL SPAT

6.4.1 Species composition of diets

Foods suitable for growing small spat in closely controlled conditions within hatcheries are the same as those used in larval culture (section 5.1). When spat are in their first week after settlement they are usually fed the same diet they were fed before settlement occurred. As they increase in size it may not be possible to produce sufficiently large quantities of some of the more delicate and difficult to grow algae. Diets for larger spat tend to be made up of the hardier species such as Tetraselmis sp. and the diatoms Chaetoceros muelleri, Thalassiosira weissflogii and Skeletonema costatum.

The highly unsaturated fatty acid (HUFA) DHA (22:6n3) does not appear to be as important in the development of spat as it is during larval development so that Isochrysis galbana and species with a similar HUFA profile - while useful as minor components of the diet - are not essential. Typically, diets will be approximately a 50:50 ratio of a Tetraselmis species and one of the above named diatoms. Part of the ration may be in the form of an algal paste rather than freshly grown live algae (Figure 97). Some products support satisfactory growth rates. The suggested reading list at the end of this section includes papers on recent research with a range of non-living foodstuffs.

Figure 97: An example of a proprietary algal paste product suitable as a partial or complete replacement for hatchery grown live algae in the culture of bivalve spat. Packs of Tetraselmis and Thalassiosira contain the equivalent of 3 600 l at 410 cells per µl and 1 800 l at 2 600 cells per µl respectively. When refrigerated, shelf life is 12 to 14 weeks. A range of useful species is available.

6.4.2 Calculating food ration

Ration is calculated on the basis of the biomass of spat held in a tank unit whether it is a closed downwelling or upwelling system or a system operated with partial water exchange. Spat of most bivalves have similar requirements in terms of the quantity of food required per unit biomass. Thus, a ration calculated for a given biomass of oyster spat will be equally as suitable for the same biomass of clams and mussels although growth responses may be very different. For example, clams will initially grow more slowly than oysters even in the best possible conditions. Scallops are again the exception and respond best to lower rations per unit biomass.

Ration in terms of the dry weight of algae required is calculated from the equation:

F = (SxR)/7

where, F = the dry weight of algae per day (mg); R = ration as dry weight of algae (mg) per mg live weight of spat per week and S = the live weight of spat (mg) at the beginning of each week.

A worked example is given below together with an extension of this equation to calculate the volume of harvested algae required for the daily ration.

A diet made up of Tetraselmis and Skeletonema in a 50:50 ration by dry weight will be 57.2 l of the former at 1 500 cells per µl and 76.5 l of Skeletonema at a harvest cell density of 7 000 cells per µl. The dry weight of one million cells of Skeletonema is 0.032 mg.

A biomass of 600 g of oyster or clam spat will need to be grown in a 3 000 l volume. Adding the above ration will result in an initial algal cell density within the system of 57 cells equivalent in size to Tetraselmis per µl (57 000 cells per ml). This algal cell density is too high to support optimum growth if it is delivered as a single batch feed. The optimum food cell density in this respect is 10 000 cells per ml. The solution is to add (10/57x114.3) l = 20 l of food as a batch feed and the remainder by drip feed or dosing pump over the 24-h period.

A ration of 0.4 mg algae per mg live weight of spat per week is towards the upper limit for spat of warm water scallops, such as Argopecten species, which are grown at the same temperature as the oysters and warm water clams (i.e. 23+2ºC). Ration needs to be reduced for cold water scallop species.

Calculations given in the example above apply equally to systems operated with a partial daily water exchange. Ration is calculated for the biomass of spat held and not the water volume in which they are grown.

When spat systems are operated on flow-through and food supply is from a nutrient enhanced pond or tank it is not possible to accurately assess the species composition of the food supply or the ration that needs to be provided. It will vary from day to day according to the state of the bloom. An experienced technician will be able to judge whether the pond water will need to be diluted with non-bloomed seawater in order to keep the daily ration within reasonable bounds. Excessive pseudofaecal production by spat indicates that the food supply is too high.

6.5 GROWTH AND SURVIVAL

Assuming spat are being cultured at reasonable density, their growth rate is largely influenced by the quality of food given in terms of the nutritional value of the component species of the diet, the ration of food provided and water temperature. Other factors play a part, such as salinity and genetics, but their effects are relatively minor in comparison. The effects of the biomass of spat per unit volume of the system in which they are grown have already been discussed. A density of 200 g live weight per m³ of tank volume represents a good compromise between density for maximum growth, which occurs at below 25% of that biomass, and economic considerations such as the space needed to accommodate tanks and the volumes of heated, treated seawater required.

6.5.1 Variability in spat growth between species

The different bivalve species commonly grown in hatcheries have widely differing growth rates when grown at reasonable densities on an adequate diet and ration and at close to the optimum temperature. Oyster spat grow considerably more quickly to a saleable seed size than do spat of the various commercial clams and scallops. Cold water scallops grow more slowly than warmer water species. Partly this is related to the larger larval size of oysters at settlement and partly the fact that there is no lag phase while metamorphosis takes place.

Figure 98: Comparison of the growth of Pacific oyster, Manila clam and calico scallop spat in similar conditions. Growth is shown as mean shell length (shell height in the case of calico scallop spat) at the beginning and end of a 7-day period

A comparison is given of the growth of Pacific oyster, Manila clam and calico scallop spat in Figure 98. This contrasts growth of the three species from settlement size by comparing mean shell length at the beginning of a 7-day period with mean shell length 7 days later. Spat of the three species were grown at pilot-scale in systems of the types described previously at commercial densities and with adequate diets, rations and at 23+1ºC. In this graph, the more steeply the growth curves are inclined to the left, the faster is the growth rate. Manila clams are in fact growing faster than Pacific oyster spat but they are starting from a smaller size. At the end of a 3-week period from settlement, Pacific oyster spat will grow to approximately 3.4 mm mean shell length compared with Manila clam spat which will reach 1.14 mm. This is mean shell length and the distribution about the mean is much greater in clam spat than it is in oysters. Calico scallops grow more slowly with an equally as large-size distribution about the mean. After 5 weeks growth they will reach approximately 1.5 mm mean shell height (where height is almost the same as length at this stage). Manila clam spat will exceed this size in 4 weeks (1.6 mm).

Cold water scallops such as the Japanese scallop, Patinopecten yessoensis, will take 4 or 5 months to reach 5 mm shell height even when grown in ideal conditions.

6.5.2 Effect of ration on growth

The ration given in sections 6.3 and 6.4 for the purposes of explanation of spat culture methodology is 0.4 mg dry weight of algae per mg live weight of spat per week (R 0.4). It has proved to be a practical ration in hatcheries because it is not excessive in terms of algal food production requirements and it is adequate in providing satisfactory growth rates of most species. Better growth rates can be achieved by feeding higher rations. As an example, the growth of Pacific oyster spat is given in Figure 99 when experimentally fed rations ranging from R 0.1 to R 1.0 at a mean temperature of 24ºC. The graph shows growth in a 7-day period for spat of different mean live weights at the beginning of the week. Clearly, growth continues to increase when spat are provided with higher rations than R 0.4. Spat of 2 mg at the beginning of a week will reach almost 7 mg by the end of the week when fed R 0.5 and 9 mg when fed R 1.0.

Figure 99: The relationship between food ration and growth for Pacific oyster spat.

Figure 100: Comparison of the growth of European flat oyster and Pacific oyster spat at 24ºC when fed various rations of a mixed diet of Isochrysis and Tetraselmis.

Among the oysters cultured in hatcheries, the various Crassostrea species respond very similarly in terms of growth rates for given rations. Spat of the European flat oyster, Ostrea edulis, do not grow quite as rapidly when provided the same conditions. Comparative growth in live weight of European and Pacific oysters is shown as the growth coefficient G₇ when fed rations ranging from R 0.1 to R 0.5 at 24ºC in Figure 100. G₇ is calculated from the following equation:

G₇ = ln wt₇ - ln wt₁

where wt₇ is the mean live weight of spat at the end of a 7-day period and wt₁ is the mean live weight at the beginning of the period (ln denotes natural logarithm).

The size spat will grow to at the end of a week when they begin the week at a specified size can be calculated from the equation. Growth coefficients are marked on the graph for both species when fed the same rations per unit live weight biomass. What this means for spat of both species when they begin a week at 2 mg mean live weight is shown in Table 15. Spat at least doubled their weight by the end of the week on all rations and Pacific oyster spat more than trebled their weight on rations R 0.4 and R 0.5.

Table 15: Mean live weight of Ostrea edulis and Crassostrea gigas spat at the end of a 7-day period when beginning the week at an initial mean live weight of 2 mg and fed rations ranging from R 0.2 to R 0.5 at 24ºC. Ration is as dry weight of algae (mg) per mg live weight of spat per week. The diet was Isochrysis and Tetraselmis in a 50:50 ratio by dry weight.

6.5.3 Combined effects of ration and temperature

The effects of growing, for example, European flat oyster spat on different rations each at a range of temperatures are shown in Table 16. These data were calculated from similar growth curves to those shown in Figure 100 and apply to spat beginning a weekly growth period at 2 mg mean live weight.

Table 16: The combined effects of temperature and food ration on Ostrea edulis spat beginning a weekly growth period at 2 mg mean live weight. Rations provided are lower than in Table 15 and range from R 0.05 to R 0.2. The diet supplied was Isochrysis. ND - no data.

The lowest ration tested (R 0.05) was still adequate to support growth at the highest temperature although growth rate at this ration was declining rapidly as temperature increased. Food supply must be sufficient to support metabolism, the rate of which increases as the temperature rises, with energy remaining for growth. Low food rations at high temperatures result in spat that may increase in size in terms of shell growth but at the expense of the soft body. Spat that leave the hatchery in poor condition are more likely to die during early growout. Much information exists in the literature and the reader is directed to the suggested reading at the end of Part 6 to pursue this topic further.

6.5.4 Survival

The percentage of spat that will survive to be sold is extremely variable between species, within and between years and between hatcheries. As a general rule spat are not as vulnerable as larvae to pathogenic micro-organisms but, occasionally, abnormal rates of mortality will occur in smaller-size spat coincident with mass larval mortalities.

The survival of oysters is usually in the region of 50 to 70% from set to 2 to 4 mm shell length. For clams and scallops it may be in the 10 to 20% range (Figure 101). Much of the mortality takes place in the first week following settlement in oysters and during the first two weeks for clams and four weeks for scallops. Many larvae that set fail to survive metamorphosis, presumably because they have insufficient food reserves to complete this critical life history stage. Early mortality does not appear to be as much of a problem with the oysters which set and complete metamorphosis within a day or two. However, it has frequently been observed within hatcheries that a higher than average set does not necessarily mean that more viable spat will be obtained. Conditions may favour a good set but they do not necessarily improve the levels of reserves in larvae that may not be fitted for survival through metamorphosis.

When spat are set on cultch, survival is dependent on the density of set. This applies mostly to oysters that cement themselves to the substrate. Clams and scallops are able to change their position relative to their neighbours if overcrowding occurs. Where density of set is intense in oysters the stronger will overgrow the weaker which will inevitably die.

Mortalities will occur if oyster spat are grown at too high a biomass per unit volume in closed systems. The first symptoms are when the shells of the spat gradually or suddenly turn pale in colouration. If they are not reduced in density at this time then the calcium carbonate crystals in the shell will dissolve. This only happens when biomass grossly exceeds the recommended or if a water change has been missed. A check of the water contained in the tank system with a pH meter will show that the pH level has dropped sharply. It normally decreases between water changes from pH 8.2 to pH 7.6 or thereabouts, but if for the reasons mentioned above husbandry has been neglected, it may drop to below pH 7.0. The reason is partly the build-up in CO₂ in the system from the respiration of the biomass of spat and the numerous bacteria in the water. The only remedy if the problem is recognized soon enough is to change the water and reduce the biomass of spat.

Figure 101: The survival (blue line) and growth (orange line) of calico scallop, Argopecten gibbus, spat during a 6-week period post-settlement. Estimates were made of survival at 2-weekly intervals.

6.5.5 Hatchery production

Before considering the nursery culture of spat output from the hatchery it is relevant to consider the process of hatchery production as an entity. When designing a new hatchery the various parts of the operation need to be assessed in relation to expectations in terms of the targeted output of seed. For example, the larval facility may be capable of setting 100 million larvae per year, therefore the capacity to grow spat needs to be equally matched to handle that production to whatever size the market requires. Likewise, the algal unit needs to be designed to reliably produce the daily volume of the required food species to feed the broodstock and the maximum numbers of larvae and spat at the various stages in development which will be in production at any point in time. These factors will vary from hatchery to hatchery according to the species to be cultured and the anticipated volume of sales.

Figure 102: A summary flow diagram of the various aspects of hatchery production showing the temperature range and the daily food requirement per unit number of animals at each stage. This diagram is applicable to most warm water bivalve species.

As a general guide, a summary of the various facets of culture operation and requirements in terms of the temperature of the water and daily food rations is given in Figure 102. Also shown is a range in days that each stage in the production cycle will take which encompasses most warm water bivalve species. Food requirements have been calculated for the average size of larvae and spat that will be in culture on any given day when the hatchery is functioning at maximum capacity. It is assumed that spat will be grown to 3 mm shell length before sale or transfer to a nursery

6.6 NURSERY CULTURE

Bivalve nurseries serve as an interface between hatcheries and the growout phase, i.e. the culture of bivalves in suspension or off-bottom in the sea. They are costefficient systems that eliminate the necessity of growing very small seed in fine-mesh nets such as Pearl nets, whose meshes readily clog with floating seaweed, sediment and the settlement of fouling organisms. The purpose of nurseries is to rapidly grow small seed at low cost to a size suitable for transfer to growout trays, bags, or nets with mesh apertures of 7 to 12 mm. Larger mesh size growout trays are not as prone to rapid clogging and require less maintenance.

Nursery systems evolved in Europe and the USA in the 1970s and early 1980s as a natural adjunct to hatcheries. They can be regarded either as the final stage in hatchery production or the first stage in growout. The most efficient nurseries stock seed at high density in upwelling containers. Others may consist of floating or submerged tray units in productive waters with or without an element of forced as against passive flow, but these systems are more akin to growout and will not be considered here.

Nursery spat holding containers may be mounted on rafts or barges moored in productive estuaries or saltwater lagoons. Others are placed in troughs adjacent to or on upwelling rafts floating in natural or artificially constructed seawater ponds (Figure 103). Primary production can be enhanced in ponds and lagoons as already explained by the application of natural or artificial fertilizers to encourage blooms of algae, usually of naturally occurring species. In this respect, they are more amenable to management than sea-based nursery systems because the quantity and to some extent the quality of the available food supply can be manipulated and controlled.

Figure 103: A - a land-based nursery with food supplied by a pair of blooming ponds that are filled and fertilized at different times to promote a succession of blooms. Food is controlled by allowing a flow of water from the most productive pond - Pond 2 in the diagram - into the stock pond from which the troughs holding containers of spat are supplied. B - a floating barge or raft nursery that may be moored in a productive estuary or in a large coastal lagoon or system of ponds. Small floating nurseries may be powered by a low-head propeller (axial flow) pump and larger versions by a paddle wheel, both of which drain water from the discharge channel generating upwelling through the mesh bases of the spat holding containers.

6.6.1 Land-based nurseries

Land-based pond nurseries are generally located on low-lying land close to the sea. The ponds are flooded at high tide via a sluice, through a culvert with flap valves that opens to the sea, or by low-head pumping. They can be drained by gravity at low tide (see Figure 103). A land-based nursery system usually comprises a number of shallow, large surface area ponds or tanks interconnected by channels or pipes with sluices or valves. Most of the ponds are used to bloom naturally occurring microalgae species present in the water at the time of filling. Blooming can be controlled and enhanced by the application of agricultural-grade nitrogen and phosphorus fertilizers and a soluble form of silica (section 3.4.6) although reliance on the natural fertility of the water is the more common approach. These algal ponds are used in rotation to supply "bloomed" water to a pond adjacent to the upwelling seed containment unit. Excess water from the pond drains back to the sea and in many cases there is a regular or continuous partial replacement of the water direct from the sea to control the food density and to flush out waste and metabolites. Water is pumped from the supply pond to the upwelling unit, which operates according to the same basic principle as upwellers in the hatchery. Alternatively, if the upwelling unit is a floating structure, water flow is generated by propeller pumps or paddle wheels. Examples of land-based nurseries are shown in Figures 104 and 106.

Figure 104: Examples of land-based nurseries. A and B - concrete spat holding tanks containing upwelling cylinders of spat (Tinamenor S.A., Pesues, Spain). Water is pumped from ponds into the tanks and is discharged into a drainage trough at the base of the tanks. C and D - a upwelling nursery system supplied from a nutrient enriched 450 m³ concrete tank at the Fisheries Laboratory, Conwy, Wales, UK. Water is delivered to the spat holding unit (D) by a high capacity submersible pump. E and F - the progenitor of most European bivalve nurseries developed by Seasalter Shellfish at Reculver, Kent, England.

The biomass of spat stocked in a land-based nursery is dependent upon the productivity of the ponds or tanks and this can be influenced by such factors as temperature and salinity as well as nutrient levels. Shallow pond systems of large surface area and volume act as heat sinks and will gain temperature from solar irradiation. They will often be at a significantly higher temperature than the adjacent seawater, which is beneficial for the growth of warm water species but requires careful management since blooms may be sudden and short-lived (Figure 105). There is always the risk that excessive blooming of algae will result in oxygen depletion of the pond water. Algae, which normally output oxygen as a by-product of photosynthesis, switch to a net uptake of oxygen for respiration during the hours of darkness when unable to photosynthesise. During intense blooming, sufficient oxygen is withdrawn from the water by the algae that the level of oxygen saturation can drop to as little as 20% over the course of a few hours, usually reaching a low point in the early hours of the morning. This can give rise to unexpected mass mortalities of the small bivalves. It is a wise precaution to have oxygen monitoring equipment connected with an alarm installed in the system. Careful management is exercised to control blooming by water exchange between ponds - assuming that there is more than one - and by diluting blooms with water drawn directly from the sea. If the sea is at a lower temperature than the ponds then it will have a higher oxygen content. Aeration equipment is often used to help maintain oxygen levels in pond systems.

Figure 105: Data from a land-based nursery pond system in Nova Scotia, Canada, operated from early May to the end of October: A - the temperature advantage of the ponds over ambient sea temperature; B - mean weekly temperature of the pond system; C - mean weekly suspended particulate matter (as thousands of particles per ml) in the size range 2.5 to 5.0 µm diameter (green histograms) and 5.0 to 10.0 µm (brown histograms). Particulates were determined using a Coulter Counter. Samples were examined by microscopy to ascertain that the particles were mainly of algal origin.

Salinity in the ponds can be lowered by heavy rainfall and by unexpected sources such as freshwater seepage through the ground or by springs or streams that may be seasonal in nature. As in site selection for hatcheries, careful research needs to be undertaken before committing to the development of a nursery at an unknown location.

Determining the biomass of spat that can be held in a pond system is largely a matter of trial and error. A general rule is that 1 hectare surface area of shallow pond will support the production of between 1 and 3 tonnes biomass of seed, depending on levels of algal productivity, over the course of a growing season. This represents the maximum sustainable biomass that can be maintained with careful management. The areas covered by many European nurseries can be measured in the tens of hectares. Spat are managed in much the same way as in hatcheries. They are regularly graded and redistributed so that any spat container will hold spat of a particular grade. Grading is usually accomplished with mechanical graders (Figure 106). Management also

Figure 106: Examples of raft or barge-type nurseries: A to C - raft floating in a man-made pond connected to a large network of blooming ponds with interconnecting channels (Tinamenor S.A., Pesues, Spain); B - detail of the raft showing the cylindrical spat holding cylinders and the lifting device; C - the same raft with detail of the paddle wheel which drives water from the raft’s discharge channel into the pond on the other side of the dam. Clam spat are being handgraded on the work platform. D - a mechanical seed grader (right) as part of an oyster hatchery/ nursery operation in Atlantic Canada. E - a barge operating on the same upwelling principle but in an estuary in Prince Edward Island, Canada. F - loading the base of a spat container with small oysters from an insulated "cooler" in which they were transported from the hatchery. In this example the stainless steel base is detachable from the fibreglass body of the container. involves controlling the blooming of algae and this requires regular observation on some parameter or parameters connected with algal production i.e. determinations of suspended particulate material, either as numbers per unit volume (Figure 105C) or as weight per unit volume, chlorophyll determination, or by microscopy. A reference to methodologies can be found in the suggested reading list at the end of Part 6 (Strickland and Parsons, 1968).

While it is generally possible to raise primary production in ponds to levels significantly higher than those prevailing in the sea it cannot always be guaranteed that the types of algae growing are of the size, digestibility, and nutritional value appropriate to the seed in culture. On occasion it may be necessary to alter the mix of fertilizers being used and "spike" a pond with a sufficient quantity of cultured algae to promote a bloom of the required composition (see section 3.4.6).

6.6.2 Barge-type nurseries

Water flow in barge-type nurseries is generated by low-head (axial flow), propeller pumps, or electrically driven paddle wheels mounted in channels which receive the discharge from the upwelling containers (Figures 103 and 106). The pumps or paddle wheels force water out of the channel(s) back to the surrounding water. This causes a head difference between the level of the surrounding seawater and the lower water level in the discharge channel, with the result that water flows through the mesh bases of the upwelling containers from the outside. The water passes through the bed of contained spat and is discharged into the channel from which it is driven back to the sea or pond.

Regardless of the technology used, careful management must be exercised to match the total biomass of seed held in the unit with the continuity, quantity, and quality of available food. This is dependent on whether the barge is moored within a pond system (Figure 106A - C) or it is floating in an unmanaged saltwater lagoon or in an estuary (Figure 106E - F). The operator can choose between producing large numbers of small seed grown to a moderate size or a smaller number of seed grown to a larger size. Assuming the case of a barge moored in a productive estuary, a flow of between 10 and 20 l per minute per kg of spat should carry a sufficient supply of food to the animals. Each spat holding container (1 m² base area), of which there may be up to 32 in a unit, will hold up to 120 kg of seed at the maximum biomass loading, requiring a flow per container of >1 200 l per minute. The total flow per 32-container unit will therefore be in excess of 38 400 l per minute (38.4 m³ per minute). A paddle wheel is more energy efficient in inducing a flow of this magnitude than is an axial flow, propeller pump. Driving a paddle wheel with an electric motor connected via a gearbox provides scope to vary the total flow according to the size of spat and the total biomass held. Lower flow rates per unit biomass than those quoted above may be appropriate in a managed land-based pond system where levels of algal productivity are higher.

Nurseries of the various types described above are in common usage in Europe and North America as part of well-established regional shellfish industries. There are, however, occasions when smaller nurseries are applicable, for example when a new industry is in the early stages of development or as part of a small owner-operated, vertically integrated business. Small, floating nursery units can be home-built or purchased direct from manufacturers without major financial investment (Figure 107). The operating principle is exactly the same as the large-scale commercial units. They are generally powered by an axial flow pump of about 1 m³ per minute capacity.

Nursery systems as shown in Figures 104 and 106 require access to a reliable supply of electricity. If power is not available at a remote site or on a barge floating in a tidal estuary then tidal power can be harnessed to operate an upwelling system. The principle is known as "FLUPSY" - floating upwelling system - and is illustrated in Figure 108. FLUPSYS require a tidal flow of at least 50 to 100 cm per second to function efficiently.

Figure 107: A small, commercially manufactured upwelling nursery powered by an axial flow pump in use at Harwen Oyster Farm, Port Medway, Nova Scotia, Canada. Information on this and similar types using solar power to provide electricity for the pump is available on the internet. The operation of this nursery is exactly the same as previously described.

Figure 108: Tidal powered, floating upwelling systems - "FLUPSYS": A - a small experimental unit showing the various components. The unit floats on the water surface buoyed by styrofoam filled flotation pipes (f). It swings about a single point mooring (m - one of two mooring brackets) to face the direction of the tide so that water is forced into the throat (t) of the device and up through the spat container (sc). The spat container has a mesh base and may contain a bed of spat or a stack of trays. The forced water flow is discharged at the rear of the spat container with provision to prevent the escape of spat. B - a commercial application of the principle where several large "FLUPSYS" are mounted in a raft.

Land-based nurseries have advantages compared with sea-based systems. They function at higher temperatures during the growing season and food supply can be manipulated. The disadvantage is that they are less stable than sea conditions and can be prone to eutrophication if not properly managed. The concept of managed productive seawater pond systems offers much potential for development beyond the application to the bivalve seed nursery. In the foreseeable future, fertilized natural- or artificial-pond systems or coastal inlets enclosed by dams with sluices could be effectively used for the semi-natural production of bivalve seed, in this way bypassing the need for hatcheries. This approach has been used successfully by Atlantic Shellfish Ltd in Ireland and by various companies in Norway.

6.7 SUGGESTED READING

Baldwin, R.B., Mook, W., Hadley, N.H., Rhoses, F.J. & DeVoe, M.R. 1995. Construction and operations manual for a tidal-powered upwelling nursery system. Nat. Coastal Res. Devel. Institute. NOAA contract AQ66.90-5228-03

Bayes, J.C. 1979. How to rear oysters. p 7 - 13. In: Proc. 10th Annual Shellfish Conf. The Shellfish Assoc. of Great Britain, London: 104 pp.

Bayes, J.C. 1981. Forced upwelling systems for oysters and clams using impounded water systems. p 73 - 83. In: Claus, C. De Pauw N. & Jaspers, E. (eds), Nursery culturing of bivalve molluscs. EMS Spec. Pub. No. 7. European Mariculture Society, Bredene, Belgium: 394 pp.

Bourne, N., Hodgson, C.A. & Whyte, J.N.C. 1989. A Manual for Scallop Culture in British Columbia. Canadian Tech. Rep. Fish and Aquatic Sciences, No. 1694: 215 pp.

Castagna, M. & Manzi, J.J. 1989. Clam culture in North America: hatchery production of nursery stock clams. p 111 - 125. In: Manzi, J.J. & Castagna, M. (eds) Clam Mariculture in North America. Developments in Aquaculture and Fisheries Science, 19. Elsevier, Amsterdam, Oxford and New York.

Couturier, C., Dabinett, P. & Lanteigne, M. 1995. Scallop culture in Atlantic Canada. p 297 - 340. In: Boghen, A.D. (ed) Cold-Water Aquaculture in Atlantic Canada. The Canadian Institute for Research on Regional Development, Moncton, Canada: 672 pp.

Ito, H. 1991. Scallop culture in Japan. In: Shumway, S.E. (Ed), Scallops: biology, ecology and aquaculture. Elsevier. Developments in Aquaculture Fish. Sci., 21: 1017 - 1055

Helm, M.M., 1990. Moderna proettazione e gestione di schidutitoi per molluschi bivalvi e nuovi sviluppi (Hatchery design and general principles of operation and development). p 65 - 87. In: Alessandra, G. (ed) Tapes philippinarum: Biologia e Sperimentazione. Ente Svillupo Agricolo Veneto, Venice, Italy: 299 pp. (Italian and English text)

Helm, M.M. 1990. Managing Production Costs - Molluscan Shellfish Culture. p 143 - 149. Congress Proceedings, Aquaculture International, September 4-7, 1990, Vancouver, BC, Canada: 480 pp.

Helm, M.M. 1991. Development of industrial scale hatchery production of seed of the mangrove oyster, Crassostrea rhizophorae, in Cuba. Food and Agriculture Organization of the United Nations. FAO: TCP/CUB/8958: 46 pp.

Helm, M.M. 1992. Operation of a managed, shore-based nursery for the early growth of hatchery produced European oyster seed. Department of Aquaculture and Fisheries, Halifax, Nova Scotia, March 1992: 35 pp.

Helm, M.M. 1994. Towards reliable bivalve seed supply in Nova Scotia. Bull. Aquacul. Assoc. Canada 94 (4): 9 - 14

Kennedy, V.S., Newell, R.I.E. & Eble, A.F. (eds). 1996. The eastern oyster, Crassostrea virginica. Maryland Sea Grant, Univ. Maryland, College Park, USA: 734 pp.

King, J.J. 1986. Juvenile feeding ontogeny of the geoduck Panope abrupta (Bivalvia: Saxidacea), and comparative ontogeny and evolution of feeding in bivalves. MSc thesis. Univ. Victoria, Victoria, BC, Canada: 281 pp.

Kingzett, B.C. 1993. Ontogeny of suspension feeding in post-metamorphic Japanese scallops, Patinopecten yessoensis (Jay). MSc Thesis. Simon Fraser Univ. Vancouver, Canada: 117 pp.

Kraeuter, J.N. & Castagna, M. (eds). 2001. The biology of the hard clam. Elsevier, Devel. Aquaculture Fish. Sci. 31. 751 pp.

Laing, I. 1987. The use of artificial diets in rearing bivalve spat. Aquaculture, 65: 243 - 249

Laing, I. & Millican, P.F. 1986. Relative growth and growth efficiency of Ostrea edulis L. spat fed various algal diets. Aquaculture, 54: 245 - 262

Laing, I., Utting, S.D. & Kilada, R.W.S. 1987. Interactive effect of diet and temperature on the growth of juvenile clams. Aquaculture, 113: 23 - 38

Laing, I. & Psimopoulous, A. 1998. Hatchery culture of king scallop (Pecten maximus) spat with cultured and bloomed algal diets. Aquaculture, 169: 55 - 68

Langton, R.W. & McKay, G.U. 1976. Growth of Crassostrea gigas (Thunberg) spat under different feeding regimes in a hatchery. Aquaculture, 7: 225 - 233

Mann, R. & Glomb, S.J. 1978. The effect of temperature on growth and ammonia excretion of the Manila clam, Tapes japonica. Estuarine and Coastal Mar. Sci., 6: 335 - 339

Manzi, J.J. & Castagna, M. 1989. Clam mariculture in North America. Elsevier, Devel. Aquaculture and Fish. Sci., 19: 461 pp.

O’Foighil, D., Kingzett, B., O’Foighil, G.O. & Bourne, N. 1990. Growth and survival of juveniles scallops, Pateinopecten yessoensis, in nursery culture. J. Shellfish Res. 9 (1). p. 135 - 144

Quayle, D.B. 1988a. Pacific oyster culture in British Columbia. Can. Bull. Fish. Aquat. Sci. 218. 241 pp.

Quayle, D.B. 1988b. Natural molluscan seed production. In: E. Uribe (ed). Produccion de larvas y juveniles de especies marinas. Universidad del Norte, Coquimbo, Chile: 45 - 49

RaLonde, R. 1999. Final report of the Kachemak Bay shellfish nursery culture project 1997-98. Alaska Dep. Fish and Game: 53 pp.

Reid, R.G.B., McMahon, R.F., O’Foighil, D. & Finnigan, R. 1992. Anterior inhalent currents and pedal feeding in bivalves. Veliger 35 (2): 93 - 104

Rosenthal, H., Allen, J.H., Helm, M.M. & McInerney-Northcott, M. 1995. Aquaculture Technology: Its Application, Development, and Transfer. p 393 - 450. In: Boghen, A.D. (ed) Cold-Water Aquaculture in Atlantic Canada. The Canadian Institute for Research on Regional Development, Moncton, Canada: 672 pp.

Spencer, B.E. 1988. Growth and filtration of juvenile oysters in experimental outdoor pumped upwelling systems. Aquaculture, 75: 139 - 158

Spencer, B.E. & Hepper, B.T. 1981. Tide-powered upwelling systems for growing nursery-size bivalves in the sea. p 283 - 309. In: Claus, C., De Pauw, N. & Jaspers, E. (eds), Nursery culturing of bivalve molluscs. EMS Spec. Pub. No. 7. European Mariculture Society, Bredene, Belgium: 394 pp.

Spencer, B.E., Akester, M.J. & Mayer, I. 1986. Growth and survival of seed oysters in outdoor pumped upwelling systems supplied with fertilized sea water. Aquaculture, 55: 173 - 189

Strickland, J.D.H. & Parsons, T.R. 1968. A practical handbook of seawater analysis. Bull. Fish. Res. Board of Canada. 167: 1 - 311

Taguchi, K. 1978. A manual of scallop culture methodology and management. Fish. Mar. Ser. (Can). Transl. Ser. 4198: 146 pp.

Urban, E.R., Pruder, G.D. & Langdon, C.J. 1983. Effect of ration on growth and growth efficiency of juveniles of Crassostrea virginica (Gmelin). J. Shellfish Res., 3: 51 - 57

Utting, S.D. 1986. A preliminary study on growth of Crassostrea gigas larvae and spat in relation to dietary protein. Aquaculture, 56: 123 - 138

Utting, S.D. 1988. The growth and survival of hatchery-reared Ostrea edulis L. spat in relation to environmental conditions at the on-growing site. Aquaculture, 69: 27 - 38

Utting, S.D. & Spencer, B.E. 1991. The hatchery culture of bivalve mollusc larvae and juveniles. Lab. Leafl., MAFF Fish. Res., Lowestoft, No 68: 31 pp.

Ventilla, R.F. 1982. The scallop industry in Japan. Adv. Mar. Biol., 20: 309 - 382

Ver, L.M.B. & Wang, J.K. 1995. Design criteria of a fluidized bed oyster nursery. Aquacultural Engin., 14 (3): 229 - 249

Walne, P.R. 1972. The influence of current speed, body size and water temperature in the filtration rate of five species of bivalves. J. Mar. Biol. Assoc. UK, 52: 343 - 374

Warfel, P.E. 2002. Growth and economic advantages of distributed powered upwellers: creating a new aquaculture niche. Presented, Nat. Shellfisheries Assoc. Meeting, Mystic, Conn. 2002.

Whitney, L.F. & Zahradnik, J.W. 1970. Fluidization of juvenile oysters: a progress report. Proc. Nat. Shell. Assoc., 60: 11

Whyte, J.N.C., Bourne, N. & Hodgson, C.A. 1987. Assessment of biochemical composition and energy reserves in larvae of the scallop Patinopecten yessoensis. J. Exp. Mar. Biol. Ecol. 113: 113 - 124

Williams, P. 1981. Offshore nursery culture using the upwelling principle. p 311 - 315. In: Claus, C., De Pauw, N. & Jaspers, E. (eds), Nursery culturing of bivalve molluscs. EMS Spec. Pub. No. 7. European Mariculture Society, Bredene, Belgium: 394 pp.

Wisely, B., Holliday, J.E. & MacDonald, R.E. 1982. Heating an aquaculture pond with a solar blanket. Aquaculture, 26: 385 - 387