**3. Spat collection**

#### **3.1 Barnacle life cycle**

Knowledge about the barnacle life cycle is essential in order to design culture strategies. Identification of the embryonic and larval development particularities - when, where and in what magnitude they occur – will permit the elaboration of production technologies that are suitably adapted to these characteristics.

The life cycle of barnacles that are cultured, or in the process of being cultured, such as the "giant barnacle" and the "craca", presents characteristics that facilitate semi-intensive cultures. The fact that fecundation is internal and that the embryonic stages are incubated, determines that mortality during the early stages of development is lower than in many invertebrate species, such as mussels and clams, where fecundation is external and the

competent larvae, as well as ecological and functional aspects of juveniles during growth. From experiences undertaken to date, the main difficulties affecting culture are the following: (a) juveniles cannot be obtained from the wild, or can only be obtained in very limited quantities. It may also be possible to procure juveniles through larval cultures in hatcheries; however this increases costs considerably. The limited amount of spat obtained from the wild can be associated with various aspects, such as: - low supply of competent *cyprid* larvae; - substrates that are inadequate for the exploratory and attaching behaviour of the competent *cyprid* larvae; - lack of synchronization between period of maximum competent larvae quantities and the installation of artificial substrates (spat collectors) in the water column; - high spatial and temporal variation in spat collection, associated with climatic, oceanographic and topographic factors (Goldberg, 1984; Lagos et al., 2008; Andrade et al., 2011). (b) high levels of unpredictability in competent larvae supplies, that operate on a macro and meso-scale in the ocean. Consideration must be given to the fact that barnacles can form metapopulations and larval dispersion can be very wide (Lagos et al., 2005); (c) problems during growth related to the presence of predators or species that compete for the substrate or food (Pham & Silva, 2010); (d) density-dependent effects associated with mass recruitment (Hills & Thomason, 2003); (e) heavy weight of the shell, that occurs mainly in balanomorphs species. This also generates

On the other hand, the main advantages associated with the biological characteristics of the barnacle species are: (a) possibility of obtaining spat from the wild; (b) gregarious larval settlement, which, if suitable substrates are available, can ensure an adequate supply of spat; (c) internal embryonic development in the case of balanomorphs barnacles, that limits early mortality and d) filter feeding, that permits suspended cultures without provision of

Future projections for barnacle culture are aided by the wide-ranging and diversified literature available on larval development in various species; specifically, on aspects such as: duration, effects of environmental factors and feeding, as well as ecological, physiological and behavioural factors associated with larval settlement and recruitment (Walker, 1995; Jenkins et al., 2000; Dionisio et al., 2007; Tremblay et al., 2007; Pineda et al., 2009). Although this information is mainly related to species that are not commercially important, it can be

Knowledge about the barnacle life cycle is essential in order to design culture strategies. Identification of the embryonic and larval development particularities - when, where and in what magnitude they occur – will permit the elaboration of production technologies that are

The life cycle of barnacles that are cultured, or in the process of being cultured, such as the "giant barnacle" and the "craca", presents characteristics that facilitate semi-intensive cultures. The fact that fecundation is internal and that the embryonic stages are incubated, determines that mortality during the early stages of development is lower than in many invertebrate species, such as mussels and clams, where fecundation is external and the

low harvest yields.

exogenous feed.

used to optimize barnacle culture.

suitably adapted to these characteristics.

**3. Spat collection 3.1 Barnacle life cycle**  entire ontogenetic development occurs in the water column. Nevertheless, during planktotrophic larval development, mortality occurs according to the duration of this phase. Similarly, larvae experience wide dispersion that determines high variability in the quantity of competent larvae, both spatially and temporally. This aspect is important when determining the location of the culture centre, in terms of provision of spat from the wild. On the other hand, the high selectivity of the competent larvae, conditions the choice of artificial substrates,that must satisfy the biological and physical-chemical requirements necessary for larval settlement. The *cyprid* larva possesses a complex sensorial apparatus associated with the need to evaluate the substrate where larval settlement will occur. This stage is crucial to larval effectiveness and determines the subsequent viability of specimens.

#### **3.2 Barnacle spat collection from the wild. Culture technologies: The cases of**  *Austromegabalanus psittacus* **(Chile) and** *Megabalanus azoricus* **(Portugal)**

Technologies for obtaining spat from the wild have only been developed for two species of balanomorphs. In the case of the "giant barnacle", *Austromegabalanus psittacus*, it is possible to obtain quantities of spat that can sustain the development of commercial cultures.

The "giant barnacle" is bathymetrically distributed in the shallow subtidal zone (5-7 m depth), although, occasionally, it can also be found in the lower intertidal zone. Furthermore, it is an encrusting species, with a considerable presence associated with the fouling of ships, culture systems and pier support structures (Brattström, 1990; López et al., 2007a). In cultures, it has been possible to take advantage of its high spatial variability and the wide range of substrate-types on which larval settlement develops. Different types of collectors have been used as spat catchment from the wild, including artificial substrates that differ in texture, size, form, colour and buoyancy. In this species, materials tested included 900 cm2 rectangular plates of expanded polystyrene, polythene and synthetic felts ("bidin"), treated with tar. In addition, polythene tubular substrates measuring 70 to 100 cm (equivalent to 2,200 to 3,000 cm2 useable attachment surface) were tested (Fig. 3). Each collector system is composed of three substrate units, joined by polypropylene cables that are placed vertically in the water column from surface levels down to 8 m depth, suspended from floating systems (long-lines or culture rafts) in shallow, wave protected or semiprotected bays .

According to results, artificial spat collectors with the following characteristics ensure high spat collection levels: they should be innocuous (do not affect settled organisms), with rough textures, dark colour and low buoyancy. Spat preference for these characteristics is explained due to the sensorial characteristics of the competent larvae at the moment of attachment to the substrate. These possess negative phototaxism (preference for low luminosity), positive geotaxism (preference towards the bottom) and rugofilic behaviour (preference for rough substrates).

Spat density levels have varied according to type of collectors, obtaining values ranging from 0.01 individuals/cm2 up to 0.2 individuals/cm2, with substrate coverage of close to 100%.

In the Azores, spat collection and growth experiments in *Megabalanus azoricus* were initially conducted on PVC tubes with recruitment densities reaching up to 800 barnacles/ m2\*month

Fig. 3. "Giant barnacle", *Austromegabalanus psittaccus*, spat collection from the wild*,*  southern*,* Chile.

during late spring/early summer (Pham et al., 2011). Recent experiments showed that the choice of materials and their orientation are important factors to be considered for optimizing recruitment values and future spat collector designs (Pham et al., unpublished data). Densities on material placed horizontally, whatever its category, are higher than when material is placed vertically. This difference is particularly pronounced for the most efficient material tested, PVC plates. PVC plates should be considered for the design of spat collection structures.

The private sector has tested a pilot structure for the production of barnacles (Pham & Silva, 2010). The structure tested was designed for both phases of the culture cycle: (a) spat collection and (b) ongrowing. The system was capable of holding 12 PVC tubes within a single structuring frame and was placed at 8 m depths over a period of 1.5 years. Results were satisfactory, as more than 9,000 barnacles of commercial size were produced with low mortality rates. Work is being currently being conducted to improve the design and efficiency of the system, based on recent results.

### **3.3 Spatial and temporal variability**

Spat collection from the wild can vary in spatial on different scales (between latitudes, between locations , between depths, between types of substrate) and temporal terms (interannual, seasonal, monthly), given that natural populations of barnacles function as metapopulations, that is, there is a continuous exchange of larvae between geographically adjacent populations by means of passive transport by currents. Thus, abundance of

Fig. 3. "Giant barnacle", *Austromegabalanus psittaccus*, spat collection from the wild*,* 

during late spring/early summer (Pham et al., 2011). Recent experiments showed that the choice of materials and their orientation are important factors to be considered for optimizing recruitment values and future spat collector designs (Pham et al., unpublished data). Densities on material placed horizontally, whatever its category, are higher than when material is placed vertically. This difference is particularly pronounced for the most efficient material tested, PVC plates. PVC plates should be considered for the design of spat

The private sector has tested a pilot structure for the production of barnacles (Pham & Silva, 2010). The structure tested was designed for both phases of the culture cycle: (a) spat collection and (b) ongrowing. The system was capable of holding 12 PVC tubes within a single structuring frame and was placed at 8 m depths over a period of 1.5 years. Results were satisfactory, as more than 9,000 barnacles of commercial size were produced with low mortality rates. Work is being currently being conducted to improve the design and

Spat collection from the wild can vary in spatial on different scales (between latitudes, between locations , between depths, between types of substrate) and temporal terms (interannual, seasonal, monthly), given that natural populations of barnacles function as metapopulations, that is, there is a continuous exchange of larvae between geographically adjacent populations by means of passive transport by currents. Thus, abundance of

southern*,* Chile.

collection structures.

efficiency of the system, based on recent results.

**3.3 Spatial and temporal variability** 

individuals at a given site does not depend solely on the stock of breeders from that location, but also on that of more distant zones. Local oceanographic factors exert an influence, as do variables operating on a meso and macro-scale. In the case of the "giant barnacle", spat collection has been evaluated using artificial collectors in different sites along the Chilean coast, and at different depths and periods of the year. There is a temporal pattern in larval settlement, given that spat capture in artificial collectors is greater in the spring-summer period, principally between October and January, observing a second settlement period in the autumn, during the months of March and April. This is associated with the fact that gonadal maturity reaches its highest values when water temperature and food supply increases. Latitudinal differences have been evaluated in three sites along the Chilean coastline: North Totoralillo (29o29'S; 71o42'W), Tongoy Bay (30o36'S; 71o37'W) and Metri Bay (41o36'S; 72o42'W); the former two are located in the northern zone of Chile and the third, in the south of the country. All three sites are characterized as being shallow, wave-protected or semi-protected bays. There are marked climatological and oceanographic differences along the length of the Chilean coast, principally in temperature, which is higher in the northern zone. After 9 months, density of specimens recruited (number of specimens/cm2) in the collectors differed between sites, with significantly higher densities in the locations of North Totoralillo than in Tongoy Bay and Metri Bay. Thus variability in spat collection does not have a latitudinal pattern, but rather varies from location to location. In addition, differences have been reported between types of collectors, spat collection being greater in the tubular plastic collectors than in the plates, which could be associated with the different surface/volume relationship.

Bathymetric variability also exists between collectors located on the surface and those located at greater depths. Cultures are more effective at greater depths, although no variations where recorded as at depths of over 4 m; this could be associated with the positive geotactic and negative phototrophic behaviour of the competent larva.

In some locations pronounced inter-annual variations in spat collection of spat from the wild have been verified. Oceanographic aperiodic events, such as ENSO (El Niño Southern Oscillation) and salinity variation on a local scale, could account for these results.

Spatial-temporal variability in spat collection from the wild limits the development of industrial cultures. As a result, it may be advisable to develop mass cultures of the "giant barnacle" in locations that are particularly apt for spat collection, and to establish other sites for the growth process. This practice has been used for many years in mussel cultures (*Mytilus chilensis)* in southern Chile.

Recent records indicate that the probability of local "giant barnacle" presence is associated with variables that are related to productivity (e.g. nitrate concentration, phosphate). This may be related with processes of coastal upwelling,that modify the physical-chemical regime of the coastal ocean. Furthermore, chlorophyll concentration and its variability are important variables that account for the presence of juveniles and adults, probably related to the trophic relationships of these filtering organisms.

In *M. azoricus*, results collectec monthly from settlement panels located in two sites off Faial Island, over a period of 2.5 years, revealed that recruitment takes place all-year round, with a peak in the summer, from early June until the end of September, coinciding with an increase in water temperature. This same study showed that newly settled barnacles (less than 15 day old) have a base diameter ranging between 0.79 and 2.80 mm and after two months can reach up to 12 mm. This large size and rapid initial growth increases the competitive capacity of *M. azoricus* recruits in relation to other barnacle species settling on the plates (*Chtamalus montagui* and *Balanus trigonus*).

#### **3.4 Optimization models for spat collection from the wild**

Dynamic models have been applied to "giant barnacle" spat collection from artificial substrates in the wild (Andrade et al., 2011). These models enable us to establish the variables involved in the spat collection process, the relationships between them and their relative effect on spat harvest. The dynamic hypothesis proposes that the number of "giant barnacle" spat obtained from the wild is influenced by competent larval abundance (*cyprid* density) over time, substrate availability and mortality after larval settlement. The number of factors affecting their spatial-temporal variability, illustrates the complexity of the spat collection process and the difficulties that exist with respect to predicting results. Simulation tests to establish consistency of the model were undertaken, using empirical background information obtained from semi-industrial cultures in southern Chile. In these cultures, average spat density in artificial collectors fluctuated between 0.1 and 3 spat/cm2, while early mortality was below 90%. If artificial collectors of 10,000 cm2, suspended from 100 m long lines are used, between 8,000 and 9,000 "giant barnacle" spat will be required to produce a harvest of 1 gross ton. Synchronization between *cyprid* abundance levels and installation of artificial substrates in the water is critical to achieve maximum collector efficiency. A period of less than 1 week out of synchronization, produces significant losses (60-70%) in spat production. When deployment of collectors and maximum quantity of competent larvae are synchronized, an increase of almost double the number of spat can be obtained. Quantity of competent larvae, as well as substrate quantity and quality, are key factors to achieving adequate spat provision from the wild, enabling the development of commercial cultures.

#### **3.5 Spat's production in hatcheries**

To date, barnacle larval cultures have not been undertaken on a commercial scale; however considerable experience has been gained with regard to the larval development of different species, studying, not only the effect of environmental and feeding factors on survival and duration of the entire development process, but also the variables that influence larval settlement. Larval development, and the factors that influence it, has been studied in various species of lepadomorphs and balanomorphs (Miller et al., 1989; Molares et al., 1994; Yan & Chan, 2001; Dionisio et al., 2007; Li et al., 2011).

Spat production in hatcheries possesses various advantages over obtaining spat by collection from the wild. Production can be programmed over time, and is not dependent on the spatial-temporal variability of the larval pool in the environment. The energetic quality of the *cyprid* larvae can also be improved, that will be reflected in the percentage of larval settlement, size and growth rate of juveniles (Thiyagarajan et al., 2002; Thiyagarajan et al., 2003b; Desai & Anil, 2004). Post-settlement mortality can be, for the most part, reduced or completely eliminated.

Experiences related to the production of balanomorph barnacle larvae in the laboratory, date back considerably (Knight-Jones, 1953; Rittschof et al., 1984; Brown and Roughgarden,

than 15 day old) have a base diameter ranging between 0.79 and 2.80 mm and after two months can reach up to 12 mm. This large size and rapid initial growth increases the competitive capacity of *M. azoricus* recruits in relation to other barnacle species settling on

Dynamic models have been applied to "giant barnacle" spat collection from artificial substrates in the wild (Andrade et al., 2011). These models enable us to establish the variables involved in the spat collection process, the relationships between them and their relative effect on spat harvest. The dynamic hypothesis proposes that the number of "giant barnacle" spat obtained from the wild is influenced by competent larval abundance (*cyprid* density) over time, substrate availability and mortality after larval settlement. The number of factors affecting their spatial-temporal variability, illustrates the complexity of the spat collection process and the difficulties that exist with respect to predicting results. Simulation tests to establish consistency of the model were undertaken, using empirical background information obtained from semi-industrial cultures in southern Chile. In these cultures, average spat density in artificial collectors fluctuated between 0.1 and 3 spat/cm2, while early mortality was below 90%. If artificial collectors of 10,000 cm2, suspended from 100 m long lines are used, between 8,000 and 9,000 "giant barnacle" spat will be required to produce a harvest of 1 gross ton. Synchronization between *cyprid* abundance levels and installation of artificial substrates in the water is critical to achieve maximum collector efficiency. A period of less than 1 week out of synchronization, produces significant losses (60-70%) in spat production. When deployment of collectors and maximum quantity of competent larvae are synchronized, an increase of almost double the number of spat can be obtained. Quantity of competent larvae, as well as substrate quantity and quality, are key factors to achieving adequate spat provision from the wild, enabling the development of

To date, barnacle larval cultures have not been undertaken on a commercial scale; however considerable experience has been gained with regard to the larval development of different species, studying, not only the effect of environmental and feeding factors on survival and duration of the entire development process, but also the variables that influence larval settlement. Larval development, and the factors that influence it, has been studied in various species of lepadomorphs and balanomorphs (Miller et al., 1989; Molares et al., 1994; Yan &

Spat production in hatcheries possesses various advantages over obtaining spat by collection from the wild. Production can be programmed over time, and is not dependent on the spatial-temporal variability of the larval pool in the environment. The energetic quality of the *cyprid* larvae can also be improved, that will be reflected in the percentage of larval settlement, size and growth rate of juveniles (Thiyagarajan et al., 2002; Thiyagarajan et al., 2003b; Desai & Anil, 2004). Post-settlement mortality can be, for the most part, reduced or

Experiences related to the production of balanomorph barnacle larvae in the laboratory, date back considerably (Knight-Jones, 1953; Rittschof et al., 1984; Brown and Roughgarden,

the plates (*Chtamalus montagui* and *Balanus trigonus*).

commercial cultures.

completely eliminated.

**3.5 Spat's production in hatcheries** 

Chan, 2001; Dionisio et al., 2007; Li et al., 2011).

**3.4 Optimization models for spat collection from the wild** 

1985). *Cyprid* have been obtained in various species and larval settlement trials undertaken (Yan, 2003; Desai and Anil, 2004; Anil et al., 2010; Li et al., 2011).

In species such as *Balanus amphitrite*, trials have been carried out with different diets and temperatures, establishing development periods that can fluctuate between 4 and 18 days (Qiu and Quian, 1999), which indicates the influence of these factors on culture success. Microalgae densities of the genera *Chaetoceros sp*, *Skeletonema* and *Isochrysis*, vary from 1x105 to 10x105 cells/ml. In this species, survival has exceeded 90% (Qiu and Quian, 1999; Mishra et al., 2001). Percentage of larval settlement can vary from 10% to 80%, depending on the culture conditions, the substrate type and the use of inductors (Rittschoff et al., 1984; Mishra and Kitamura, 2000; Mishra et al., 2001; Thiyagarajan et al., 2003b).

With regard to larval settlement induction, a great variety of chemical signals have been described as fundamental to trigger the larval settlement response. Natural inductors have been characterized, based on studies of larval settlement response as a consequence of natural substrates and have been associated, mainly, with three types of sources: conspecific individuals, prey species and microbial films (Rodríguez et al., 1992). Different kinds of microbial biofilms have been described as important inductors of larval settlement in benthonic marine invertebrates. Induction has been observed both by diatom and cyanobacteria biofilms (Akashige et al., 1982), as well as by bacterial biofilms (Kirchman et al., 1982; González et al., 1987). Induction by bacteria has been widely studied; the positive effect of these types of biofilm has been observed in barnacles (Maki et al., 1988, Maki et al., 1990). The settlement response associated with these biofilms would be generated, apparently, by the presence of polysaccharides or extracellular glycoproteins attached to the bacterial wall (Kirchman et al., 1982, Hadfield, 1986), or, alternatively, soluble compounds released by the biofilms (Bonar et al., 1990). An increase in the settlement inductive potential of older bacterial biofilms was a recurring factor observed (Maki et al., 1989, Pearce and Scheibling, 1990, Mataxas and Saunders, 2009). Similarly, the contrary effect has been described (Maki et al., 1989).

In synthesis, sufficient knowledge is available on which to base the design of barnacle hatcheries, although the transition still has to be made from the experimental to commercial level. No information is available with regard to the costs that this would imply. As commercial cultures are implemented, the development of spat production in hatcheries will become a necessity, in order to improve predictability and quality in spat production.
