Previous Page Table of Contents Next Page



Fisheries Biology Unit
University of Singapore
Republic of Singapore


The method of prawn culture in Singapore differs from the one used in the Philippines mainly in that young prawns are not stocked in the pond by operators but are brought in by tidal flow. The method appears simple, yet many operators have failed to achieve success because they are not sufficiently experienced in the selection of the site, and in the construction, operation and maintenance of the pond. In this paper these methods, which are largely empirical, are described and an attempt made to explain the basic principles involved.



La méthode d'élevage des crevettes à Singapour se distingue de celle utilisée aux Philippines surtout par le fait que les jeunes crevettes ne sont pas introduites dans l'étang par les exploitants mais sont apportées par le flux de la marée. Cette méthode paraît simple, et pourtant de nombreux exploitants ont échoué faute d'une expérience suffisante dans le choix de l'emplacement ainsi que dans la construction, l'exploitation et l'entretien de l'étang. On s'attache ici à décrire les méthodes utilisées, qui sont en grande partie empiriques, et on s'efforce d'expliquer les principes fondamentaux à respecter.



El método del cultivo de camarones en Singapur se diferencia principalmente del empleado en Filipinas en que los cultivadores no sueltan los camarones jóvenes en el estanque sino que éstos llegan a él con la marea. Aunque el procedimiento parece sencillo, sin embargo muchos productores no han podido aplicarlo por carecer de suficiente experiencia en la elección del lugar y en la construcción, funcionamiento y conservación del estanque. En este estudio se describen tales métodos, que en gran parte son empíricos, tratando de explicar los principios básicos pertinentes.


The method of prawn culture in Singapore originated more than 30 years ago and has been improved over the years as a result of experience. It differs from the method used in the Philippines, as described by Villadolid and Villaluz (1951), mainly in that the prawn ponds in Singapore are not stocked with young prawns by the operators. It is believed that young prawns enter the ponds with the tides. This method is extremely simple and appears to take advantage of the peculiar aspects of the biology of prawns belonging to the family Penaeidae. The local fishermen construct earth bunds to form ponds which receive the incoming tidal currents through sluice gates twice a day and let off the excess water with the receding tide, keeping the water level at about 2 ft (61 cm).

This method of prawn culture has been described by Burdon (1950), Tham (1955) and Hall (1962). The first two of these reports are rather general and many details in respect of site selection, construction, operation and maintenance have not been mentioned. During the last 30 years the author had the opportunity of discussing this method of prawn culture with many operators with extensive experience. These operators are normally rather uncommunicative regarding the details of their operations which are considered occupational secrets. Some of the practices detailed in this paper are used by all the local operators while a few others are used only by some and no single operator employs all of them. These details are, however, vitally important to the success of the project. The purpose of this paper is to draw attention to the details of the method used and to review the probable basis for such practices.


Although, generally speaking, all the brackish-water swamps along the coast of Singapore and the southern coast of Johore (the southernmost state in West Malaysia) are suitable for the cultivation of penaeid prawns, the cultivator normally chooses the site with care, as the production of the pond as well as the cost of construction and maintenance depend on the following factors:

  1. The presence of the commercially important species of prawns in the area

  2. The tidal range at the site

  3. The location of the site in relation to the coast

  4. The depth of the swamp

  5. The suitability of the bottom soil

  6. The size of the swamp

  7. The presence of large fresh-water streams draining into the swamp

  8. The salinity of the sea water draining into the swamp

The penaeid prawns which may be caught locally in quantity and which have a high market value are the following:

  1. Penaeus indicus H. Milne Edwards, 1837

  2. Penaeus merguiensis de Man, 1888

  3. Metapenaeus ensis (de Haan, 1844) = M. mastersii (Haswell, 1879)

  4. Metapenaeus burkenroadi Kubo, 1954

  5. Metapenaeus brevicornis (H. Milne Edwards, 1837)

The larger prawns have a higher market value, those of about 4 cm carapace length (measured from the inside of eye socket to the posterior end of carapace) fetching the maximum. Where the sizes are equal the members of the genus Penaeus have a higher value. Among these the banana prawn P. merguiensis has the highest market value, followed by the Indian prawn P. indicus, the giant tiger prawn P. monodon and and the green tiger prawn P. semisulcatus. Among the members of the genus Metapenaeus, the yellow prawn M. brevicornis commands the highest price in the local markets. In the selection of a site, the prawn pond operators attempt to assess if the most valued species are present in sufficiently large numbers to make the project economically viable. Cast nets and filter nets are operated in the proposed swamp site and if the catches yield a high percentage of carideans as compared to penaeids the site is not considered a good one for prawn culture.

The tidal range at the site is considered important by the pond operators because they prefer to harvest their ponds daily if possible in order to have fresh prawns for sale every day. The local consumers pay a higher price for freshly caught prawns. They do not realise, however, that the productivity of the pond depends on the volume of tidal flow since in this method of prawn culture the young prawns as well as their food are brought into the pond by the daily tidal flow. But even in Singapore, where there are two high waters every day and the height of high water at spring tide is 11 ft (3.35 m) or a little more, they can only harvest on about 22 days in a month. During the height of the neap tides there is no harvesting. Generally, the tidal range is always considered together with the height of the swamp site. The two should be such that harvesting could be carried out on about 20 to 22 days in a month. If the level of the swamp is too high, no water will enter the pond even during high water at neap tide. This results in a lower number of operational days in the month. For example, if the swamp is 4 ft (1.2 m) above the level of lowest low water at spring tide, there is at least a difference of 2 to 3 ft (0.6 to 0.9 m) between the level of the swamp and the lowest high water at neap tide, in Singapore. This range is considered suitable. If the height of the swamp is much more than 4 ft (1.2 m) there will be too many days in the month when the pond remains without sea water and this is considered undesirable. Where the high water level at spring tide is unduly high, e.g. more than 14 ft (4.3 m), this is also considered unsuitable because it would involve building higher and stronger bunds, resulting in high cost of construction and maintenance.

Another important consideration is the location of the site in relation to the coast. The site should be near the coast but not so near as to cause the damage to bunds by continuous wave action. It should not be too far inland since there is then a danger of flooding with rain water during the monsoon season. Usually a site along the coast is chosen with a fringe of at least 50 ft (15.2 m) between the beach and the prawn pond, where the mangrove trees serve as protection against wave action.

The next point to be taken into account is the texture of the bottom soil at the site. It should be hard clay and there should be not more than 2 in (5.1 cm) of silt above the clay. It has been found that when the pond bottom is heavily silted the prawn catch declines, as the author observed in the prawn pond owned and operated by the Singapore Fisheries Department from 1954 to 1958. The silt on the bottom of this pond in 1955 was 2 in (5.1 cm), but by 1958 it was nearly 10 in (25.4 cm). The catch dropped from 12,800 lb (5810 kg) in 1955 to 6,000 lb (2720 kg) in 1958. It is still not clear why the catch should drop with the increase in silt level. It has been shown by Dall (1958) that M. bennettae buries itself in mud during the daylight hours; so any increase in silt level should not drive it away. It may, however, be that when the mud is more than 2 in (5.1 cm) deep anaerobic conditions are set up in the mud layer, leading to the production of poisonous gases (such as hydrogen sulphide) which then slowly diffuse through the mud thus deteriorating the habitat of the prawn. Hard clay is preferred also for the construction of the earth bunds that enclose the prawn pond. Bunds built from soft silty material will not last and sink so fast that they cannot be consolidated.

The size of the swamp is also an important consideration, because ponds of less than 20 acres (8.1 ha) are not considered economical to maintain. At least two workers are required even for a pond of 10 acres (4 ha), to open and close the sluice gates four times a day. For large ponds of more than 40 acres (16.2 ha) three or four workers may be required. The smallest size of pond which is economical to maintain under local conditions is 30 acres (12.1 ha) because the income derived from it is sufficient to pay two workers and other operational expenditure and still leave a sizeable profit for the owner of the pond.

Large fresh-water streams draining into the swamp site are not favoured but where other conditions are favourable, steps might be taken to divert the stream if it is not too costly an operation. The salinity of a prawn pond can be lowered considerably by monsoon rainfall if the precipitation on the catchment is drained into it. It is believed that this lowering of salinity will drive out the penaeid prawns. There is also the danger of flooding of the pond during heavy rainfall causing breaches in the bunds.

The salinity of the sea water draining into the swamp is also important. Generally it should not be lower than 18 at any time during the year. A salinity range of between 24 and 30 is considered most suitable. The lower limit should never be below 15 .


For the project to be successful the construction of the pond has to be carefully planned with the following points in mind:

  1. The bunds must be strong enough to withstand the difference in pressure between the water levels inside and outside the pond at any state of the tide.

  2. They must be high enough to ensure that water will not flow over the top and should be wide enough for operators to walk on the bund itself.

  3. There must be sufficient number of sluice gates to allow the inflowing tidal current to get into the pond quickly, so that unduly high pressure will not be built up on the outside of the bunds, and also to allow the outflowing water to leave the pond fast enough to obviate any unduly high pressure being built up against the bunds on the inside of the pond.

  4. The sluice gates must be sited in such a way that the inflowing tide will flow directly through them without generating eddies and whirlpools outside the gates. It is believed that this may scare the young prawns from entering the pond.

  5. The channels bringing the sea water into the pond must be straight or only gently curved and they must be kept clear, as plants growing in the channels will vibrate with the fast flowing current and it is believed that they will deter the young prawns from entering the pond.

  6. The bottom of the pond must be cleared of mangrove stumps and levelled, as it has been found that removal of the stumps and levelling invariably increased the yield. It may be that retention of these stumps provides a suitable habitat for crabs and other animals which are predatory.

  7. Channels must be provided on the bottom of the pond to disperse the inflowing water to every part of the pond. These channels should radiate from the sluice gates to the farthest corners of the pond. This arrangement seems to improve the yield, and it is possible that it helps to disperse the young prawns evenly so that they are less vulnerable to predation and enjoy a higher level of food availability. These channels are planned before the construction of the pond, and the clay for the building of the bunds is taken from them, thus fulfulling two objectives.

  8. The bottom of the pond is graded in such a way that the part nearer to the sluice gates is deeper. This will enable the operators to concentrate the prawns prior to catching.

  9. To facilitate operation, the sluice gates are built in batteries wherever possible with concrete foundation and sides, and wooden shutters.

  10. The bunds adjacent to the sluice gates are reinforced by wooden piles as otherwise the daily flow of water past the gates will weaken them.

  11. In the construction of the bunds the clay is removed from the pond bottom in slabs of about 1 ft (30 cm) in length, 6 in (15.2 cm) wide and 4 in (10.2 cm) thick. These slabs are laid side by side to form the bund and each layer is baked in the sun before the next layer is put on. This reduces the sinking rate of the bunds. Even a well-baked and well-built bund has a sinking rate of 1 ft (30 cm) per year. Badly constructed bunds may sink as much as 2 to 3 ft (61 to 91.4 cm) per year. The bunds are built with hand labour which is costly. The author has tried using dredges and earth-movers to build these bunds. Contrary to expectation it is neither cheaper, nor faster, because the bunds have to be built much wider both at the top and bottom and also higher to allow for a greater sinking rate. On account of the soft nature of the terrain and the weight of the machinery, the shifting of the machines consumes a lot of time which adds to the cost of construction.


The sluice gates are opened when the incoming tide reaches a height of 2 ft (61 cm) above the floor of the sluice gates. This ensures a rush of inflowing water which prevents the prawns swimming out. When the tide has risen to the highest point in the daytime a wire screen of ½ in (1.3 cm) mesh is placed across the gate to prevent escape of any prawn with the receding tide. During the daytime the prawns are ‘adsorbed’ on the bottom of the pond and the above operation is not to catch prawns but to reduce the water in the pond to such a level that prawn catching can begin at dusk. But if the ebb tide begins at night or at dusk when the prawns will be swimming about, a fishing net of rami netting (usually about 30 ft (9.1 m) long with meshes varying from 5/6 in (2.1 cm) bar at the mouth to 3/8 in (0.95 cm) bar at the cod-end) -with mouth large enough to be fitted to a wooden frame of the same size as the cross section of the gate - has to be fixed in place in the slots on the sides of the gate, just before the tide ebbs. The receding tide then flows through the filter net so that all prawns and fish going through the sluice gate are caught. This procedure serves not only to catch the prawns but also to reduce the level of water in the pond thus equalising the pressure on the bunds both inside and outside. Before the net is hauled up, the water flow through the gate is stopped by operating a wooden windlass which lowers the sluice gate. A wooden trough-like raceway is also provided on the outer side of the sluice so that the net will not be torn when the current rushes through it. The net frame is then lifted and the catch released on the sorting floor by pulling on the rope securing the cod-end of the net.

Damage to nets have to be repaired every day after washing in fresh water and drying in the sun. Once a fortnight they are treated with mangrove bark extract and dried in the sun, after which they are steamed and dried again. The bunds have to be maintained, i.e. breaches and leaks are repaired and they are topped up at regular intervals to make up for the sinking rate. The crustacean known locally as the mud lobster, Thalassina anomala (Herbst), builds burrows in the bunds, weakening them and causing leaks. They can be spotted by the funnel-shaped external opening of their burrows and destroyed in any of the following ways:

  1. By placing a piece of calcium carbide, about 1 in (2.5 cm) in diameter in the funnel of the burrow. The acetylene gas given off on contact with water kills the mud lobster.

  2. By pouring quicklime (calcium oxide) into the funnel. The heat generated on contact with the moist animal will kill it, or

  3. By pouring a solution of DDT in kerosene into the funnel. This will also kill the mud lobster.

Since the pond is filled twice a day with sea water by the tides, it is obvious that a large number of fish species will get into the pond. Many of these species are predatory to all stages of the penaeid prawns and so production will gradually drop if they are not killed. A list of these species is given by Hall (1962). It has been found by experimentation that ‘teaseed cake’ is a very efficacious poison as it kills the fish quickly but does not appear to harm the prawns. The active principle in ‘teaseed cake’ appears to be saponin. In the prawn pond owned and operated by the Singapore Fisheries Department from 1954 to 1958, it was found by a series of experiments that the use of this fish poison once every four months will raise the yield of prawns to at least ⅓ ton per acre (837 kg/ha) per annum. This has now become standard practice in Singapore.

It is also customary to dry the pond bottom in the sun for three or four days before refilling with water. This has the effect of hardening the pond bottom and perhaps also hastening the mineralisation of the organic debris in the soil. It may also help to fertilize the pond by breaking down the hard frustules of the diatoms on the pond bottom enabling the cell contents to run out.


The method of prawn culture as practised in Singapore consists of provision of rearing ponds for young penaeid prawns which are brought in with every incoming tide. It is certain that very few of the penaeid prawns caught in the pond, entered the pond as large individuals, i.e., of total length greater than 3 cm, because repeated filtration of the inflowing tide with fine nets never produced more than 20 individuals. But filtration of the subsequent outflowing tide always produced more than a hundred individuals, sometimes over a thousand. Hall (1962) working on the plankton catches of a prawn pond in Singapore produced evidence to support the hypothesis that the penaeid prawns entered the pond as larvae or post-larvae and grew in the ponds until they were harvested. This is supported by the work of Dakin (1938), Anderson, Lindner and King (1949), Morris and Bennett (1951) and Racok (1959) which indicate that, apart from one or two species of the genus Metapenaeus, the penaeid prawns breed fairly far out at sea. The larval stages are then brought into the littoral and estuarine areas by the tides. There is thus ample evidence to support the view that the penaeid prawns cultivated in the Singapore prawn ponds were hatched at sea, were brought into the prawn ponds by the tides and grew up in the ponds until harvested. Whether these larval stages are carried towards the shore by tidal currents or whether there is also a directed movement resulting from an urge of the larvae to seek an environment of lower salinity is, however, not clear.

From the study of catches of the Sondong push nets operated off the south coast of Singapore by Tham (1955), it appears that many larval stages grow up in the littoral areas without entering the estuarine mangrove areas. It is thus highly probable that the larval stages are just carried into the estuarine areas by tides and are indifferent to the slightly lower salinity. Racek (1957) came to a somewhat similar conclusion when he stated that probably only a small portion of the penaeid larvae and post-larvae would ever reach the estuaries whereas the rest might spend their whole life cycle at sea. On the other hand, the predominance of certain species, e.g. Penaeus indicus, Metapenaeus ensis, M. brevicornis and M. burkenroadi, in the prawn pond catches in Singapore does not rule out the possibility that these species have a preference, during their juvenile stages, for slightly lower salinity. Other species, such as Parapenaeopsis tenella, species of Trachypenaeus Alcock and Metapenaeopsis barbata which are rather common in Singapore Straits, are rare in the prawn pond catches.

The question of salinity tolerance, or preference, of penaeid prawns, however, deserves further consideration. The prawn pond operators in Singapore have experienced that with heavy drainage of rain water into their ponds, the prawn catches are significantly lowered. According to Hall (1962) the salinity in the pond he studied dropped from a normal range of 24 to 31 to as low as 4.11 during periods of heavy rainfall. Certain prawn ponds along the north coast of Singapore have much wider variations in salinity, as the prawn pond studied by Cho (1965) which showed a range of 15 to 27 . With heavy rainfall the drop in salinity in such a pond might be even more marked. Racek (1957) states that the mass migration of prawns of all age groups, from the estuaries to the ocean during extensive river floods is a result of a rapid fall in salinity. This supports the observation of the local prawn pond operators.

One point in the construction of the pond which merits discussion is the belief that the young prawns may be frightened by noise due to eddies or vibrations of weeds and plants in the intake channels. This belief is based on the common knowledge among local fishermen that, when agitated by sounds such as those caused by rowing with oars, penaeid prawns will jump out of the water into their boats. The practice of keeping the channels leading from the sea to the sluice gates, free of weeds and obstruction is known to improve prawn pond catches. The other point relates to the construction of channels radiating from the sluice gates to the farthest corners of the pond. This may be expected to have two effects, viz: (i) the even distribution of the young penaeids as well as food and (ii) the avoidance of turbulence in the pond. Turbulence in a heavily silted pond may be expected to stir up undesirable gases such as hydrogen sulphide which have been formed under anaerobic conditions. This would creat an environment unconducive to the young penaeids, thus driving them out of the pond and lowering the production.


Anderson, W.W., M.J. Lindner and J.E. King, 1949 Early stages in the life history of the common shrimp, Penaeus setiferus (Linnaeus). Biol.Bull.mar.biol.Lab.Woods Hole, 96:168–72

Burdon, T.W., 1950 Annual report of Fisheries Department, Singapore, 1949. Singapore, Govt. Print. Off.

Cho, Y.W., 1965 A preliminary study of the ecological aspects of a prawn pond in Singapore. Fisheries Diploma Project Report. Singapore, University of Singapore, Fisheries Biology Unit, 44 p.

Dakin, W.J., 1938 The habits and life-history of a penaeid prawn Penaeus plebejus Hesse. Proc.zool.Soc.Lond., 108(2):163–83

Dall, W., 1958 Observations on the biology of the greentail prawn Metapenaeus mastersii (Haswell) (Crustacea, Decapoda:Penaeidae). Aust.J.mar.Freshwat.Res., 9(1):111–34

Hall, D.N.F., 1962 Observations on the taxonomy and biology of some Indo-West-Pacific Penaeidae (Crustacea, Decapoda). Fishery Publs.colon.Off., (17):229 p.

Morris, M.C. and I. Bennett, 1951 The life-history of a penaeid prawn (Metapenaeus) breeding in a coastal lake (Tuggerah, N.S.W.). Proc.Linn.Soc.N.S.W., 76(5–6):164–82

Racek, A.A., 1957 Penaeid prawn fisheries of Australia with special reference to New South Wales. Res.Bull.St.Fish.N.S.W., (3):19 p. also issued as Proc.Indo-Pacif. Fish. Coun., 6(3):347–59,1956

Racek, A.A., 1959 Prawn investigations in Eastern Australia. Res.Bull.St.Fish.N.S.W., (6):1–57

Tham, A.K., 1955 The shrimp industry of Singapore. Proc.Indo-Pacif.Fish.Coun., 5(2):145–55

Villadolid, D.V., and D.K. Villaluz, 1951 The cultivation of sugpo Penaeus monodon Fabr. in the Philippines. Philipp.J.Fish., 1:1–3

Previous Page Top of Page Next Page