Previous page Top of Page Next Page


Kenneth McAndrew

CAGES Project, CARE-Bangladesh
65, Road No. 7/A
Dhanmondi, Dhaka-1209, Bangladesh

McAndrew, K. 2002. Risks to small-scale cage farmers in Bangladesh, with emphasis on fish health experiences of the CARE-CAGES Project. p. 215-223. In: J.R. Arthur, M.J. Phillips, R.P. Subasinghe, M.B. Reantaso and I.H. MacRae. (eds.) Primary Aquatic Animal Health Care in Rural, Small-scale, Aquaculture Development. FAO Fish. Tech. Pap. No. 406.


The CARE-CAGES Project (Cage Aquaculture for Greater Economic Security) was initiated in September 1995, began cage-aquaculture activities in 1996, and is now working with approximately 4,200 households, through 45 partner non-governmental organisations (NGOs). The only pathogens that have been observed in caged fish are the parasitic isopod Alitropus typus (Milne-Edwards, 1840), and epizootic ulcerative syndrome (EUS). EUS effects mostly Java barb (Barbodes gonionotus) during the winter season, resulting in chronic low-level mortalities. Although found sporadically in other sites, A. typus only significantly effects cage-farming operations in one river in the Jessore region. It is suspected that the large amount of aquatic vegetation present in this river is a pre-requisite for large parasite numbers. This parasite makes cage culture impossible due to 100% fish mortalities in late spring. When the monsoon begins, parasite numbers reduce, and with lower numbers, there is milder damage and no mortalities. Other pathogens may be present, however, due to lack of appropriate resources, these have not been observed. Fish health concerns for CARE-CAGES largely relate to providing quality fingerlings and feeds, good site selection and cage management. A breakdown of mortalities for the CARE-CAGES project for 1998 and 1999 is provided, and key issues relating to fish health highlighted. Total losses of stocked fish were reduced from 36% in 1998 to 22% in 1999. All categories (stocking mortalities, mortalities during culture, escapees, poaching and other) were reduced, and this is due principally to greater farmer experience in this new culture system.


Cage culture has a limited history in Bangladesh. CARE-CAGES (Cage Aquaculture for Greater Economic Security) is the first project in Bangladesh to focus exclusively on cage aquaculture. Established in September 1995, the project has attempted to develop a range of systems that offer resource-poor farmers an opportunity for successful cage culture, within their social, economic and institutional environment. The project is due to finish in March 2001 and, as shown in Table 1, has expanded each year since start-up and is working with approximately 4,200 households in 2000. The CAGES Project is currently delivering its outputs to farmers through 45 small-partner non-governmental organisations (NGOs) in seven districts: Barisal, Comilla, Jessore, Dhaka, Natore, Sylhet and the Chittagong Hill Tracts.

Table 1. Expansion of the CARE-CAGES Project, showing numbers of households and cages operated by these households.

There has been a trend in the project to move towards smaller (1 m3) floating cages (Kabir and Huque 2000). It has been found that ease of management and the lower investment required with these smaller cages are the key reasons for changing from larger 8 m3 cages. A full review of cage types used in the project is provided by Kabir and Huque (2000).

It has become clear over the past few years that some species thrive in cages better than others. The major carps for example, though popular with the people and capable of obtaining a high market price, do not generally grow well in cages. Species farmed by CAGES beneficiaries in 1998 are shown in Table 2.

Table 2. Species farmed by cage farmers in the CARE-CAGES Project in 1998.

Species cultured in 1999 were similar to those cultured in 1998, with tilapia remaining the most popular species. In 1999, however, less grass carp were farmed, due to poor results relating to under-feeding by farmers. Percentages for sutchi catfish and freshwater prawn (Macrobrachium spp.) have increased from 1998, with farmers encouraged by their high survival and good market price. In 1999, due to high returns, rapid return on investment, and less risk for these resource-poor groups, some farmers changed from table-fish production to nursery operations in fine-mesh hapas (McAndrew et al. 2000).


Only two pathogens have been identified that cause disease and economic loss to farmers. It is, however, accepted that there may be more that are presently unrecorded, due to difficulties of implementing a detailed fish health-monitoring system for these widely dispersed, poor farmer groups. The two pathogens are the parasitic isopod Alitropus typus (Milne-Edwards, 1840) and epizootic ulcerative syndrome (EUS).

Epizootic Ulcerative Syndrome

EUS almost exclusively affects Barbodes gonionotus, although Pangasius spp. may also be affected. It is estimated that around 3% of all Barbodes cages are affected (CAGES regional Technical Officers' estimation). The disease occurs between November and January, when water temperatures fall. Farmers are well aware of the disease, and recognise the sluggish movement of fish and reduced appetite that are often the first clinical signs of infection. No treatment is recommended, as it has been advised that interventions may increase stress levels and result in higher losses. Farmers remove any diseased fish immediately, which are usually consumed by the household, although it is reported that approximately half the market price can be obtained for infected fish at the local market. Farmers are also advised to harvest all fish in an infected cage soon after the first fish are seen to be diseased, as experience has shown that few fish recover from the disease over the winter period.

Parasitic Isopod

In the Nabagonga River in the Jessore region, there are large numbers of Alitropus typus. Although reported by CAGES Technical Officers in other rivers in Bangladesh, the parasite numbers in these rivers do not appear to pose a serious threat to fish health. It is hypothesised that the reason for such large numbers in this particular river is the large amount of aquatic vegetation present. Farmers with cages on the river reported that fish culture was impossible during early spring, as the high numbers of parasites led to all fish being attacked by large numbers of the isopods, resulting in anaemia and death within 24 hr. Local fishermen also reported that, when the monsoons arrived, the parasite disappeared.

Investigation of the biology of the isopod; its impact on cage farmers, local fisher groups and other water users; and development of appropriate alleviation strategies for farmers were carried out in 1999 (Bulcock 1999). This work was financed through the DFID-funded Research Project "Improved Management of Small-scale Cage Aquaculture in Asia," the Bangladesh component of which is supporting the CARE-CAGES project.


Fish loss is inevitable in cage culture. In this report losses to the farmer are split into five different categories. These are:

  • stocking mortalities - when fish die within five days of stocking;
  • culture mortalities - when fish die after five days of stocking but before harvest;
  • escapees - all fish that are lost accidentally throughout the culture period;
  • poaching/vandalism - all fish lost due to deliberate human theft or damage; and
  • other - any other losses not covered above.

    There were a total of 1,072,426 fingerlings stocked in cages in 1999. Of these, 22% were not sold by beneficiaries at harvest. A breakdown of losses is shown in Table 3, and compared to 1998.

    In all categories there are less losses in 1999 then in 1998. The main reason for this is greater experience of the farmers and NGOs in cage culture, and the improved information and training that was provided by CAGES staff based on their previous experience. This overall reduction may be also be due to the increased numbers of 1 m3 cages, which have previously been shown to have lower mortalities during production (CARE unpubl. data). A particularly large fall was seen in escapees, which is due to better cage management, as well as improved site selection.

    Table 3. Reasons for fish loss in cages in 1999 compared with 1998.

    Of the losses detailed in Table 3, only some of the stocking and culture mortalities could be due to preventable fish health problems. Stocking mortalities are often beyond the control of the farmer, with poor quality and badly transported seed the major problem, although poor management practices when handling and acclimatising the fry may also result in severe losses. Escapees are a result of poor management and, in some cases, inappropriate site selection, while poaching and theft are a result of social problems and inadequate security.

    Although farmers have not identified disease to be a major constraint to cage-culture operations, good fish health is essential for adequate growth and survival, which is necessary if adequate income is to be obtained from the cage operation. In the case of Bangladesh, as elsewhere, health of caged fish depends on seed transportation and quality, adequate nutrition, appropriate site selection and good management practices. Each of these will be summarised, with examples of experiences from the Bangladesh CARE-CAGES Project.

    Seed Quality

    Transportation and stocking

    Mortalities due to poor transportation and stocking have been the most common cause of losses in the CAGES Project each year since inception, although each year the number of mortalities has declined. The project encourages farmers to buy from reputable fry traders and to obtain fry from as close to the farm as possible. It has been found that tilapia suffer especially high mortalities, perhaps due to their dorsal spines physically damaging other fingerlings, or to the lack of knowledge concerning transport of this relatively new aquaculture species in Bangladesh.

    As would be expected, there is a strong relationship between transportation time for fingerlings and stocking mortalities. The CAGES Project is attempting to overcome these problems with hapa breeding of tilapia, over wintering of fry and the development of local linkages between cage farmers and reputable seed suppliers (hatcheries, nurseries and fry traders).

    In addition to a supply of fry that are healthy at the time of arrival, adequate preparations for stocking of the cage are also essential. Fingerlings that are being placed in a cage are stressed, both by the change in water chemistry (water in the cage compared to water in the fry-carrying container) and by being placed in a cage. It is important that fingerlings are acclimatised slowly to the water in the cage to reduce stress. Fingerlings that have never been enclosed by a net may be panicked by the experience and attempt to escape through the netting. It is obvious that netting must be sufficiently small to prevent fry escaping or being trapped or "gilled." Such trapping of the fry is frequently observed, especially with less-experienced farmers. Where resources allow, the acclimation of fingerlings in a fine-mesh hapa is recommended.

    Nets must be conditioned by placing them in the water for up to a week before the introduction of fry. This allows periphyton to become established on the net, reducing the chance of physical damage to fingerlings when they are first placed in the new environment of a cage.

    Stocking mortalities, although still high, have been falling every year, and with greater farmer experience, availability of early quality fingerlings from hapa breeding and over-wintering, and local linkages with seed suppliers, it is expected that these will be reduced further in 2000.

    Genetic quality of seed

    The seed obtained must be capable of reasonable growth, and this is not the case with all seed currently used by the CAGES-Project beneficiaries. Common carp (Cyprinus carpio) have consistently shown very poor growth, or even sometimes weight loss in cages (CARE unpubl. data) and are no longer recommended by CAGES staff as a suitable culture species. Problems also exist with some strains of local tilapia, with early maturation and poor growth being found by some farmers. This may be the result of hybridisation of Oreochromis mossambicus with the stocked O. niloticus, resulting in a reduction in growth potential of the resultant fry. In other locations, trials have shown that some strains of local tilapia have similar growth to GIFT tilapia (Genetically Improved Farm Tilapia), recently imported into Bangladesh (CARE unpubl. data). The GIFT strain has the advantage, however, of consistently good growth potential, while local strains vary widely in their growth potential.

    Farmers invariably prefer wild seed to hatchery-produced seed, reporting that wild seed is hardier and also has the obvious advantage of being free of charge in certain locations. However, wild fingerlings must be carefully collected, handled and transported to cages. Careful handling of fry and careful stocking are also essential for hatchery-produced fingerlings. In the Barisal region, the success or failure of climbing perch (Anabas testudineus) was largely determined by the amount of care taken during seed collection (CARE unpubl. data).

    Size and timing of seed

    For successful cage culture, farmers require large fingerlings early in the season in order to maximise the growing season, and hence, increase returns. In the CAGES Project, several strategies are being tested to achieve this aim. Hapa breeding of tilapia is being attempted by most of the 44 NGOs involved in the project. Success in 1999 was mixed, but with better planning, it is expected that hapa breeding and nursing will provide a locally available source of high-quality seed for cage operators in 2000. Hapa over-wintering of several species was undertaken in 1998/99 with good results. Over-wintered fingerlings produce a seasonal "growth spurt" in spring, with superior results to same-season fry (CARE unpubl. data). Most of the project's 44 NGOs are attempting over-wintering in the 1999/2000 winter season, and so far, results appear encouraging, with few mortalities observed.

    Fish Nutrition

    Inadequate feeding is the major reason for poor growth of fish amongst project beneficiaries (CARE unpubl. data). Except in the case of certain species in eutrophic ponds, caged fish require supplementary feeds. Especially with inexperienced farmers, the amount of feed provided to fish has often been insufficient to allow reasonable growth. Lack of feed is the major reason for inadequate growth of grass carp, resulting in poor returns and a reduction in the number of farmers culturing this species in 1999.

    Poor growth and poor fish health also result from poor quality of feed. A complete pelleted diet is neither available nor affordable for CAGES beneficiaries; feeds, therefore, vary greatly in quality. Farmers are often reluctant to spend relatively large amounts of money on fish protein, which results in nutritional problems; this is especially true of high-valued piscivorous fish. Fisher groups attempted to culture several of these species (Channa marulius, C. striata and Wallago attu), however, all ended in failure (CARE unpubl. data). CAGES staff hypothesised that these fisher groups would be able to supply "trash fish" to these fish, however, as every fish caught has a market value, fishers were reluctant to feed them to the caged fish. This resulted in inadequate fish nutrition, poor fish health, poor growth and subsequent failure of the cage operation.

    Fish are usually fed a "ball-form" feed, the composition of which varies according to the species under culture and the local resources available. The feed usually contains, rice bran, oil cake and fishmeal, and often, many other ingredients (e.g., kitchen wastes, flour, rice water, snails, mussels). Farmers have also obtained good results by including cheap "opportunistic" ingredients, such as broken bread and damaged biscuits fed to tilapia, banana leaves and trunks fed to grass carp, and viscera from dead cows fed to sutchi catfish. The key is to obtain low-priced feeds that satisfy the nutritional requirements of the species being cultured. For the more herbivorous species, culturing in green water helps reduce the effects of inadequate nutrition in supplemental feeds.

    Site Selection

    Good site selection is essential for the success of any cage-culture operation. Poor site selection greatly increases the risk to the cage-culture operation. A complicating factor in Bangladesh is that water levels vary substantially during the season. This results in most cages being moved during the production cycle. Therefore, cage operators often have to select two sites, one for during the monsoon and one for when water levels are lower.

    Some areas of Bangladesh (Sylhet, Barisal) are prone to flash floods. During this time, water currents and turbidity are high, which results in problems with feeding, physical damage to cages, and possible gill damage to fish from suspended solids.

    Feed loss has been significantly reduced by the use of fine-mesh panels on the bottom of the cage and 10 cm up the sides from the bottom of the cage. This mesh is now used in all project areas, provided turbidity is not a major problem. In areas with high turbidity, the fine mesh becomes clogged, reducing water flow and placing structural stress on the cage (Kabir and Huque 2000).

    In one site at Barisal, a bamboo barrier was constructed around the cage to protect it from physical damage and to produce a microclimate of calmer water that is more conducive to fish growth. Although the cage of sutchi catfish survived the flash floods, it was not economically successful.

    Poor site selection usually results in economic loss, with fish unable to thrive. In areas prone to flash floods, rivers and canals are not recommended for cage culture. Ponds are more suitable, as they are not subjected to the same conditions, and only in severe flooding are they subjected to currents. However, not all potential cage farmers have access to ponds, as many beneficiaries do not own any land. This means that poorer groups, who are less able to risk investing in cage culture, are placed in a higher risk category due to inferior site availability. This is a constraint to entry of poorer groups in some areas.

    Jute retting is also practised in many areas of Bangladesh, especially in the Jessore and Dhaka regions. After harvesting in August, jute is placed in water bodies until the flesh of the plant has been removed by bacterial decomposition. As a result, a black, anaerobic sludge forms. This process results in water quality deterioration, making fish culture impossible in waters in which jute retting is carried out. Unfortunately, jute farmers seldom give warning of their activities, and many cages have suffered 100% mortality as a result of such practices.

    Site selection is critical to the participation of woman in cage culture (Brugere et al. 2000). In Bangladesh, it is difficult for women to move a long distance from their homesteads, both for reasons of security and in order for them to carry out their many other tasks. If a woman is responsible for feeding the fish (which they often are), cages must be located close to the homestead, or fish will not be adequately maintained. This is a major issue when the project is attempting to involve women in the activity.

    The distance from the homestead is also critical in reducing poaching, which is a major cause of fish loss (Table 2). Remote cages are at much greater risk from poaching than those close to the home. Many farmer groups have formed cage clusters, making it possible to have a guard on the cages every night, with each farmer carrying out the duty in turn.

    Management Practices

    Early failure of the project's beneficiaries was due, in part, to the rapid move to the new practice of cage culture from the more traditional and better-understood pond aquaculture system. Cage culture differs from pond-based culture in that management must be pro-active. There are activities that must be carried out at regular intervals in order to obtain any outputs from the cage (e.g., cage construction, stocking, feeding, net cleaning, net checking, removing mortalities). Many of these are complex activities that have to be learned and understood by beneficiaries.

    Although all beneficiaries were trained in all the practices necessary for successful cage culture, they still lacked experience. Beneficiaries in their second and third year are continuing to improve their management practices, and to fine-tune their production system for local conditions. This is leading to improved health, growth and economic returns from the system. Some of the key management practices include:

    Net cleaning

    Maintaining clean nets is essential for good water exchange. Brushing of nets is usually required once a week, and care must be taken not to injure fish or damage the netting, especially the easily damaged fine-mesh (hapa) netting. Rubbing the netting together with the hands is the most effective method of cleaning hapas. Some fish, especially Morulius calbasu and tilapia, actively clean periphyton from the net walls, maintaining good water exchange without the need for net cleaning. This provides supplemental feed for the fish, although nets are only cleaned effectively if there is also a relatively high density (usually a stocked pond) of fish around the cage.

    Removal of mortalities

    Dead and dying fish must be removed as soon as possible to prevent the spread of disease, and also, to prevent nets being cut by crabs. In many areas in which the project is working, most escapees are the result of holes cut by crabs that are attracted to the dead fish at the bottom of the cages. In areas where crab cutting of nets is a serious problem, inverted funnels are placed on all cage legs to prevent crabs reaching the netting. This has been found to be highly effective in reducing the problem (D. Griffiths pers. comm.).

    Re-location of cages

    Moving cages during the monsoon is a difficult and high-risk activity. This is one of the reasons why most beneficiaries have changed to smaller, floating cages with a rigid frame, which are much easier to move.

    Examination of fish

    Fish are best examined during twice-daily feeding, where cage operators can observe their approximate number and general health. Beneficiaries are advised to avoid completely removing fish from the water. This often occurs when visitors come to the cages and the farmer wishes to show off his/her fish. Fish deaths, or more commonly reduced feeding (and hence growth), occur after such events.


    At present, fish diseases are not a major problem for farmers in the CAGES project, but fish health issues do affect the success of many cage operators. Good fish health is essential for adequate fish growth and the economic viability of the fish cage. With the relatively low density of cages present in water bodies of Bangladesh at present, diseases are rarely noted by farmers or CARE extension staff as a constraint to successful cage culture; however, with expansion in the number and density of cages, it is possible that fish diseases may become a constraint on the system in the future. This will be continually monitored by extension workers, along with all other aspects of farmers' constraints and opportunities. Fish health in cage aquaculture in Bangladesh is, at this time, determined by appropriate fish seed quality, cage-site selection, fish nutrition and cage management.


    Brugere, C., McAndrew, K. and Bulcock, P. 2000. Social impacts of cage aquaculture on communities in Bangladesh. Aquacult. Asia, 3: 29-33.

    Bulcock, P. 1999. Investigations into the biology and ecology of an unidentified isopod affecting the CARE-CAGES aquaculture programme, Bangladesh. M.Sc. Thesis, Institute of Aquaculture, University of Stirling, UK., 78 p.

    Kabir, A.K.M.N., and S.M.Z Huque. 2000. Cage design in CARE-CAGES project, Bangladesh. p. 125-131. In: I.C. Liao and C.K. Lin. (eds.) Cage Aquaculture in Asia. Proceedings of the First International Symposium on Cage Aquaculture in Asia. Asian Fisheries Society, Manila, and World Aquaculture Society, Southeast Asian Chapter, Bangkok.

    McAndrew, K.I., D. Little and M.C.M. Beveridge. 2000. Entry points and low risk strategies appropriate for the resource-poor to participate in cage aquaculture. Experiences from the CARE-CAGES project, Bangladesh. p. 225-231. In: I.C. Liao and C.K. Lin. (eds.) Cage Aquaculture in Asia. Proceedings of the First International Symposium on Cage Aquaculture in Asia. Asian Fisheries Society, Manila, and World Aquaculture Society, Southeast Asian Chapter, Bangkok.

    Previous page Top of Page Next Page