Cage and pen culture, like other methods of rearing fish, may be conveniently classified as extensive, semi-intensive, or intensive on the basis of feeding. Extensive culture relies solely on naturally available foods such as plankton, detritus, benthos and drift, and no supplementary feeding is given. Semi-intensive culture involves the addition of low protein (<10%) feedstuffs, usually compounded from locally available plants or agricultural byproducts to supplement the intake of natural food, whereas in intensive culture operations, fish rely almost exclusively on an external supply of high protein (>20%) food, usually based on fish meal.
There are several factors which demographically restrict the range of species grown and the methods employed. The first constraint is geographic. Primary production, which governs all successive energy transactions in the aquatic food web (Barnes, 1980), has been shown to be correlated with latitude (Brylinsky, 1980). Data derived from the summary report of the 13-year International Biological Programme (IBP) illustrates this (Le Cren and Lowe-McConnell, 1980). Between temperate (23°–67°) and tropical (23°N–23°S) zones, there is a considerable increase in the range of production values (Fig. 3) and thus tropical water bodies offer better opportunities for extensive and semi-intensive cage and pen culture.
In Europe and North America, there are few extensive operations. In the Federal Republic of Germany there is some extensive production of carps in earth ponds (Bohl, 1982). However, extensive cage culture in Europe is largely restricted to the rearing of juvenile planktivorous stages of fishes, using illumination to attract zooplankton (Bronisz, 1979; Uryn, 1979; Jager and Kiwus, 1980). In USA, recent experiments in the extensive culture of bighead carp in cages have proved disappointing, with slow growth and low survival, and thus poor economic prospects (Engle, 1982).
Extensive and semi-intensive methods are only suitable for fish which are planktivorous, or which feed on benthos, detritus or drift, and are not suitable for fish with high protein requirements or which do not have the anatomical, physiological or behavioural adaptations to deal with these types of food. Carnivorous species, such as the salmonids and many of the catfishes (e.g. Ictalurus punctatus, Pangasius sutchi) cannot be successfully grown without recourse to intensive methods, using largely fish protein based diets (see Cowey, 1979, for review). Although all of the tilapias have comparatively low protein requirements and many therefore appear suitable for extensive cage culture, this is not so. All tilapias possess both jaw teeth and pharyngeal teeth, and these vary in size, structure and mobility (Trewavas, 1982), thus influencing the type of diet and particle size they can deal with. Microphagous species, such as O. niloticus, O. mossambicus and O. aureus grow better in extensive culture than do the macrophagous species, T. zilli and T. rendalli (Coche, 1982; Pullin, in press).
There are major differences between pens and cages, and between lotic and lentic sites in terms of availability and types of natural feeds. Fish grown in pens have access to benthic organisms, and there is some evidence that certain species grow better in pens than in cages. It is probably for these reasons as well as because of their size that there is no intensive culture in pens. In both temperate and tropical waters, primary production is generally lower in lentic than in lotic sites (Fig. 3) and energy inputs are dominated by allochthonous (external) rather than autochthonous (internal) inputs (Minshall, 1967; Knoppel, 1970; Fisher and Likens, 1973; Dudgeon, 1982). Autochthonous production in flowing waters is primarily by attached plants - macrophytes, periphyton - with little contribution from the small plankton community, and organic matter is processed by the detrital and benthic micro- and macrofloral and faunal communities (Fahy, 1972; Dela Cruz and Post, 1977; Blackburn and Petr, 1979; Dudgeon 1982a). There are thus few plankton-feeding fishes in most running waters, and cage culture of such fishes without supplementary feeding is likely to be impractical. Recent experiments in Tengi River, Malaysia, with extensively cultured bighead carp have confirmed this (Othman, et al, in press). Cages were stocked with 25.3g fish, at a stocking rate of 15 fish m-3. During the 2 month trial 95% of the carp died, and the average weight of the survivors was 19.5g.
Under some circumstances, however, planktivorous fishes may be grown in running water. At the head waters of the Bicol River in the Philippines, where it flows out of Lake Buhi, the plankton discharged from the lake is sufficient to permit the rearing of caged O. niloticus without supplementary feeding (Job Bisuña, pers. comm.) (Fig. 4).
Small rivers enriched with some organic material will have larger benthic populations and carry more detritus and insect life in the drift, than in polluted streams. Such sites are probably best for extensive/ semi-intensive culture of omnivores such as carps and catfishes, as is practiced in Indonesia and Thailand (Vass and Sachlan, 1957; Ling, 1977). Heavily polluted sites, however, are not suitable due to low O2 levels which can retard growth and cause fish kills.
In fast-flowing rivers, however, intensive or semi-intensive fish culture is not advisable due to excessive loss of feed. Although losses can be reduced by using a feeding ring (Coche, 1979), slow-flowing lowland or delta sites are preferable.
There are also economic and technical considerations which greatly influence the extent and methods of cage and pen culture practiced in inland waters in different regions of the world. Whilst the reasons for intensification are clear - increased production per unit water use, reduced labour costs, etc. - the use of intensive enclosure culture is only feasible if the fish being cultured realise a sufficiently high price to generate a profit when harvested. According to recent data published by ADCP (ADCP, 1983), feed represents 40–60% of the total operating costs in intensive aquaculture. However, in order for the venture to be profitable, ADCP recommend that feed costs do not exceed 20% of the farm gate value of the fish. This is the case in Western Europe and North America where the intensive culture of carnivorous salmonids and catfishes is feasible due to the high market prices these fish demand. However, this is not the case in the tropics. Although intensive feeds for fishes such as the tilapias and Indian major carps have been developed (Jauncey and Ross, 1982; ADCP, 1983), there has to date been little commercial interest. In 1982 in Laguna Province in the Philippines, for example, the price paid by retailers for tilapia was between 7.55 and 11.50 per kilo, depending on size of fish and season (Aragon et al, 1983), whereas in Bicol Region, the price varied between 3.75 and 5.40 (Escover et al, 1983). According to ADCP guidelines, therefore, intensive feedstuff prices must remain between 1520 and 2300 (US$100 – US$150) per tonne in Laguna, and 750 and 1080 (US$50 – US$72) per tonne in Bicol, which is well below the production costs of US$320 per tonne estimated by ADCP for a 27% protein diet suitable for intensive culture of tilapia (ADCP, 1983). Similar prices for tilapias and carps occur in many other tropical countries, and therefore intensive feeds are not yet a widely available option. Exceptions to this can be found in some countries in Southeast Asia. For example, a wide range of tilapia feeds of variable quality is available in Taiwan, where nearly all tilapia culture is intensive (R.S.V. Pullin, pers. comm.).
Technical problems of feed manufacture and storage can hamper the development of intensive cage and pen culture which has been proven economically viable, and this is particularly true in the tropics. For example, expansion of the intensive cage culture of rainbow trout in Bolivia faces problems due to the poor quality of commercially available diets (Beveridge, 1983). Contamination of feeds by aflatoxin-producing species of Aspergillus has also been reported as causing problems in tilapia farms in Africa and Southeast Asia (Roberts, 1983; Olufemi et al, 1983).
In many areas of the world, fry and fingerling production is the main technical problem yet to be overcome. The culture of many species, such as the milkfish, Chanos chanos, in the Philippines, Taiwan and Indonesia is still dependent upon the seasonal collection of fry from the wild, despite successful spawning in laboratory conditions (Liao and Chen, 1979; Lam, 1982; PCARRD, 1982). Over-exploitation of wild fry in some areas has led to shortages, high prices, and put a brake on the growth of the industry.
As fish culture industries expand, output from hatcheries must keep pace with the demand from producers. In the Philippines, a tremendous increase in interest in the cage culture of tilapias over the past five years has resulted in a greatly increased demand for fingerlings (Guerrero, 1982). Although hatchery production is keeping pace, concern is growing about the huge volume of fingerlings being produced from backyard hatcheries (Pullin, in press.) Farmers buying slow-growing fry put their businesses at risk, and BFAR are currently trying to minimise the risk of this happening by supplying commercial hatcheries with good-quality broodstock from their hatchery at CLSU, Nueva Ecija (Broussard et al, 1983).
Most of the above problems are common to both land-based (e.g. ponds, tanks, raceways) and water-based aquaculture. However, there are several problems which are peculiar to cage and pen culture, and which have caused the collapse or arrested the development of the industry. In several instances in the Philippines cages and pens have been established in highly eutrophic lakes, where regular fish kills occur through deoxygenation of the water following the collapse of algal blooms, and the subsequent decomposition of the algae (Barica, 1976; PCARRD, 1981). In Laguna de Bay, regular fish kills have occurred almost every year since the early 1970s, the worst being in 1975 when 5 × 106 milkfish were killed. By 1981, 73% of fishpens in the lake had experienced fish kills (PCARRD, 1981).
Toxic industrial pollution may also cause problems. In November 1983, 150 million worth of milkfish and tilapia in 30 ha of fish pens and cages in the western part of Laguna de Bay were killed by the appearance of “masamang tubig”, or highly polluted water (Source: Bulletin Today, Nov. 10 1983). The polluted water was described as black and oily in appearance, and blamed by local fishermen on industrial sources.
Other problems experienced by the cage and pen industry include damage during storms etc. and theft and vandalism (Coche, 1979, 1982; PCARRD, 1981). In the Philippines in July 1983, Typhoon Bebeng devastated large numbers of lake-based fish farms in Bicol, Laguna and Rizal Provinces in South and Central Luzon, and many of the operators have been unable to rebuild their farms since, due to the prohibitive costs involved.
Theft and vandalism have been cited by cage tilapia producers in the Philippines as the major problem they encounter (Escover and Claveria, 1983). Although farms are unlikely to close for these reasons development may be restricted, since operators are often unwilling to site their enclosures far from their homes, and viability may also be affected through increased expenditure on security.
Intensive cage and pen culture of fishes is largely restricted to temperate, developed regions, where luxury carnivorous species are grown on expensive, high-protein feeds compounded from fish meal. Intensive feeds are not essential in tropical fish culture, since many of the commercially important species such as the tilapias, carps and milkfish feed readily on natural macrophyte, plankton and detrital production. Supplementary feeds derived from low-cost, low-protein agricultural by-products or wastes, are widely used in order to improve production. There are some technical problems which retard the development of intensive feeds. However, even in countries where complete diets suitable for intensive feeding have been formulated, such diets are generally not in common use since they are too expensive (Guerrero, 1982). Most of the popularly cultured species have low retail prices, with small profit margins for the producers.
One of the few exceptions seems to be Taiwan, where a large number of commercial brands of feed for tilapia and milkfish are available (R.S.V. Pullin and J. Kuo, pers comm.). However, Taiwan is a sub-tropical country with a limited season for the growth of tropical species, and limited land and water resources for aquaculture development. Thus intensive fish rearing makes economic sense. Also, middle-class consumers (of which there are a large number) are willing to pay more for fish which are intensively reared and which do not have the muddy taste often associated with fish reared in extensive or semi-intensive earth pond and lake conditions (J. Kuo, pers. comm.). Intensive cage and pen culture may also become viable in the tropics if the farming of high-priced, carnivorous species such as Marble-headed goby (Oxyeolotris marmorata) develops, or if intensive and semi-intensive/extensive culture are practiced at the same site (see Section 5 below).
The supply of quality fry to certain sectors of the industry, as well as adverse weather conditions, theft, vandalism and pollution, also take their toll and affect the development and viability of cage and pen culture in different parts of the world. Lentic systems seem to offer the best potential for enclosure culture. Organically enriched streams and slow-flowing stretches of rivers offer some potential for semi-intensive culture, although low plankton concentration precludes most forms of extensive culture, and the comparatively high flow rates, with their associated feed losses, make intensive cage or pen culture impractical.