Field Working Paper 10
Description and Assessment of Fish Farms in Ghana

based on the work of

M.M.J. Vincke (FAO Senior Aquaculturist)
L.K.A. Awity (Regional Fisheries Officer)

Rome, 1990

1. Introduction.

The mission visited 24 fish farms, of which 14 are privately owned, 2 are government farms and 8 are owned and operated by the following parastatal organisations: Volta River Authority (VRA), the Irrigation Development Authority (IDA), through the Irrigation Company of Upper Region (ICOUR), and the Upper Region Agriculture Development programme (URADEP). Additional information was obtained for other fish farms which the mission was unable to visit because of limitations on time.

2. Private Fish Farms.

2.1. Fish farm at Logba-Adiveme, Hohoe, Volta Region. Owner -Mr. K. Amedume.

Mr. Amedume is a retired bank manager and now farms 88 hectares of land, 12 hectares of which are for oil palm (Elaeis guineensis). The land is on a 55 year lease.

The fish farm was set up in 1982 at Logba-Adiveme near Hohoe in Volta Region. A dam was constructed across a small river, immediately below a swamp area. When full, the reservoir has a surface area of approximately 6 hectares. At the time of visiting the farm, which was at the end of the dry season (4 April, 1990), the surface area of the reservoir was about 2 hectares.

The reservoir was stocked in 1982 using tilapia from Lake Volta. The first harvest was in 1985 and this was achieved by draining the reservoir completely. The pond bottom was also cleaned during this operation. The total harvest was four pickup loads, estimated at 3 tonnes, and this was sold at c.200 per kilo.

In 1986, Mr. Amedume sold fish for about c.350 per kilo and for about c.200 per kilo in both 1987 and 1988.

Until 1986, the farmer had no harvesting nets and used the Fisheries Department's beach seine for c.3,000 per day. He now has his own seine of 220m. in length 2.1m. deep and with a stretched mesh of 6.4cm. The completed net cost c. 450,000 in 1986.

After the first harvest of the reservoir, more ponds were constructed downstream of the dam. At present, there are five ponds with a total area of about four hectares. Ponds were excavated using a bulldozer at a daily hire rate of c. 30,000 per day. Each pond has inlet and outlet valves and small lateral spillways on the dikes. The pond bottoms have yet to be levelled. Up to the present time, the owner estimates that he has invested between 20 and 25 million cedis in his fish farming operation.

In 1989, only three ponds were stocked, these with various tilapia species and some Clarias gariepinus. These ponds have not yet been harvested.

Pig manure is applied, although this is done irregularly and not at standard rates. Oil palm kernel cake is fed to the fish. The farmer intends to raise ducks and a duck pen has been constructed near pond No. 1. Markets for ducks are available locally.

According to Mr. Amedume, the main constraints to fish farming are:

  1. lack of credit for pond construction
  2. lack of technical information on fish culture and fish farming systems
  3. high costs of equipment such as nets

2.2. Fish Farm at Adiveme, Hohoe, Volta Region.

The owner of the farm, who was not present at the time of the farm visit by the mission (4 April, 1990), is an employee of Ghana Airways.

The farm has four poorly designed ponds of which one of 1000 m.sq. was dry. According to the watchman interviewed, the other three (3,150, 2,800 and 2,600 m.sq.) have been stocked but it was not possible to obtain information about species stocked, stocking densities, feeding or fertilisation of the ponds.

The lower part of the farm is a swamp with an area of about five hectares and it appears that the farm layout does not take advantage of the topography to ensure a year-round water supply. Seven geese were present at the time of the visit and the owner is reported to have had 20 ducks on his ponds.

2.3. Aprappong Fish Farm at Mankessim Irrigation Project, Central Region.

Owner: Mr. Willis Bell.

The ponds are situated about four kilometres from Mankessim, immediately downstream of the Mankessim dam. The owner of the farm has several irrigated rice fields, which are occasionally flooded, and seven ponds of a total area of about 2.8 ha.. Five ponds are functional and two are abandoned. Water supply to the ponds is from the reservoir. One of the five ponds is used for spawning and is stocked with a number of large tilapias (T. zillii, Sarotherodon galilaeus and o. niloticus). The four other ponds are used as grow-out ponds.

During the mission's visit (5 April, 1990), Mr. Willis Bells was absent and information was obtained from his foreman, the only full-time employee on the farm. Other labourers are employed on a contract basis for special jobs. The ponds have no monks and are harvested using the farmer's own beach seines.

According to the foreman, the ponds and ricefields are stocked according to the availability of fingerlings. A 1.6 acre (6,400m. sq.) pond is stocked with about 500 fingerlings - i.e. 780 fingerlings/ha.. Fish are fed rice bran at a rate of 15–20 kg per day per pond. Ponds are not fertilised.

Harvesting is carried out once a year. Fish for human consumption is sold at between c.150 and c.200 per kg, with the best prices being achieved in February and March. Fingerlings are also sold to other fish farmers.

In the rainy season, the ponds and rice fields are sometimes flooded for periods of between 2/3 days and one month.

Rice-cum-fish culture is practised. Rice fields are seeded by broadcasting and are stocked three weeks later with 8 to 10 cm tilapia fingerlings at a density of about 700/ha. The water depth in the rice fields is between 15 and 45 cm..

Fish in these rice fields are fed rice bran at the same rate as those in the ponds and rice and fish are harvested at the same time - i.e. about 120 days after seeding. A 6,400 rice field examined by the Mission was, for example, seeded on 8 January, 1990 and due to be harvested in May,1990. The rice variety used was IRRI 132/72.

Occasionally, rice fields are restocked with fish after the harvest.

Information received for yields from a 6,400 m.sq. rice field was as follows:

  • rice: 1,200 – 2,500 kg of paddy
  • fish: 90 – 120 kg of fresh fish with average weight of between 60 and 100g.

2.4. Fish farm at Aiyinasi, Western Region.

Owner : Chief Kansa Nana Kwadjwo Ahale II.

The Chief started fish farming in 1983 and constructed his first pond using manual labour ( five boys for two weeks = 10 man days.). The pond is circular in shape and has an area of 78m. sq.. Water supply is from groundwater.

The pond was stocked with 200 tilapia fingerlings purchased from river fishermen. Wheat bran is distributed, at a rate of half a bucket each feed, twice per day. Wheat bran is delivered from Takoradi at a cost (delivered) of c.750 per bag of 25–35 kg.

Fertilisation of the pond has been started recently using poultry manure from the farmer's own layers, of which he has 3,000, and applied at a rate of approximately 2kg. every two weeks.

Since the beginning of 1989, the pond has been harvested once every two months using a cast net. Income since January 1989 has been c.72,000 and , at each harvest, sufficient fish to feed a household of 12 adults and children has been retained.

In 1989, the owner constructed a dam and stocked the reservoir with between 400 and 500 catfishes purchased at the riverside. These fish are fed chopped paw-paw (Casica papaya) and cocoyam leaves (Colocasia esculenta), mixed with wheat bran, twice each day. This reservoir has not yet been harvested.

Another pond of about 200m. sq. has been constructed recently using hired labour at a cost of c.40,000. This pond has not yet been stocked.

According to the Chief's son, at least 20 other villagers, both from within his own and neighbouring chiefs' communities, have shown interest in fish farming.

2.5 Freedom Farms, Koforidua, Eastern Region.

Owner: Mr. J.K. Osei.

The owner runs a shop and raises 1,000 chickens and 25 pigs near to his ponds. He started fish farming in 1982 by building a concrete pond of 475m.sq. in a waterlogged area of his farm. The bottom of his pond was covered with mud and then stocked with Clarias gariepinus fingerlings purchased from Accra. At his first harvest, the farmer found more catfish than had been originally stocked and decided to construct more ponds. The concrete pond was manured regularly with chicken and pig manure and, at the time of the mission's visit (9 April,1990), the pond water was dark green in colour.

Mr. Osei, using hired labour, constructed four other ponds of 374, 95, 525 and 612m. sq. in the same waterlogged area. The ponds are supplied by groundwater.

Two ponds totalling 1137m.sq. were constructed in three months by 15 labourers at an estimated cost of approximately c.600 per man/day.

Three earthen ponds were stocked with a mixture of T. zilli, T. galilaea, Heterotis niloticus, Clarias gariepinus and a small quantity of Heterobranchus bidorsalis. Pond number four (612m.sq.) is stocked with O. niloticus, which is preferred for its faster growth and reduced problems of stunting compared to T. zilli.

Fish are fed wheat bran, costing c.750 per 50kg., and cassava leaves. The ponds are fertilized irregularly with pig and chicken manure from the farm. The ponds are not drainable and are harvested partially using a cast net and a beach seine hired from the Fisheries Department. There is no harvesting schedule but fish are sold during periods of low supply on the market at c.250 per kg..

According to Mr. Osei, customer preferences are firstly for tilapia, followed by catfish and then heterotis. Mr. Osei is keen to expand and is satisfied with the results so far. He sees his main problems as being:

  • lack of credit for pond construction

  • lack of supply of Clarias gariepinus, Heterobranchus bidorsalis and other fingerlings.

Poaching is not seen as a problem.

2.6. Farm at Koforidua, Eastern Region.

Owner: Mr. Doe.

Mr. Doe is a timber merchant and farmer. He has 16 acres of palm oil plantation and 37 acres of coconut palm. Having started fish farming in 1983, he now has 6 ponds with a total area of 3,600m. sq.. All his ponds are drainable but , since their construction in 1983/84, they have only been drained twice for repairs to dikes and levelling of pond bottoms.

The ponds have been stocked with different tilapia species, some catfish and some Heterotis niloticus. The fish are fed twice a day, in the morning and evening, using a mixture of wheat bran and chopped leaves at a rate of one five litre container-full per pond - i.e approximately 10 kgs per pond per day.

Ponds are sometimes fertilised using ammonium sulphate which costs c.3,500 per 50 kg bag. The ponds are harvested when there is demand for fish and harvesting is done using the beach seine of the Fisheries Department. Women buyers are reportedly keen to buy because of the relatively low price of the fish (c.250 per kg.). The fish are subsequently sold with “banku” and “kenkey”, which are local maize products.

The farmer is satisfied with the returns from sales , although no records of inputs and outputs are kept. Poaching is not seen as a problem.

2.7. Fish Farm at Koforidua, Eastern Region.

Owner: Mr. Ampadu.

Mr. Ampadu is a member of the Ghana Fish Farmers' Association and has one pond of about 3,000 m.sq. He also has a herd of 500 cattle, a piggery and grows pepper, cowpeas and oil palm.

The farmer set up his farm in 1985 after visiting fish ponds near Akosombo. He aims to sell fish when there are low supplies of marine fish on the market.

The ponds are stocked with different species of tilapia, catfish and Heteroris niloticus which nest and reproduce naturally. The farmer does not fertilise the pond but feeds pawpaw and cassava leaves and oil palm nuts which are too small to be pressed for oil.

The ponds have been drained only twice and sales from fish harvested reached c.200,000 and c.300,000. The farmer considers fish farming to be very lucrative and is keen to implement an expansion programme when funds are available. Drawings of a modern farm were presented during the mission's visit.

The main problems are seen as being:

  • poaching by villagers

  • lack of credit for the construction of new ponds

  • lack of technical advice on fish farming

2.8. Fish farm at Koforidua, Eastern Region.

Owner: Mr. J.K. Watey.

Mr. Watey is the owner of two ponds, with a total area of 2,400m.sq., and one concrete pond of 96m.sq. which has been abandoned. He also has 2,000 oil palm trees and a small factory for oil extraction. He is a member of the Ghana Fish Farmers' Association.

The farm was constructed in 1980, using a D7 bulldozer. The farmer did not recall the cost of construction. The ponds were first stocked with a variety of fingerlings from the Ashiaman area. These died because they were brackishwater species. He subsequently restocked with different tilapia species, some Heterotis niloticus and a few tigerfish, all supplied from Lake Volta.

The farm has a constant water supply (from the Ghana Water and Sewerage Corporation) and all ponds are drainable, but only to a depth of 60 cm. No feed is given, but each pond receives one shovel-full of poultry manure each day.

The ponds were harvested in August 1989 and produced about 45 kg of fish each. Harvesting is now carried out every six months. Tilapia are the consumers' first preference, followed by heterotis niloticus. The fish reproduce naturally in the ponds.

Mr. Watey is keen to expand the farm and prepared drawings for new ponds in 1981. The main problems are seen as :

  • lack of technical advice

  • lack of funds for pond construction

  • lack of nets for harvesting

  • lack of fingerlings

2.9. Fish Farm, Koforidua, Eastern Region.

Owner: Mr. Tweneboah.

The farmer, who is a member of the Ghana Fish Farmers' Association, is the owner of a small oil extraction factory and has one fish pond close to his workshop. The pond, which is not drainable, was constructed in 1983/84 using hired labour. (216 man/days).

The pond was stocked with catfish and different tilapia species from the Densu river, which is also the source of Koforidua's water supply.

Fish are fed regularly with food remnants, cassava and papaya leaves and wheat bran. Palm kernel waste is also being tested as fish feed. One “minibag” (30 kg. approx.) of poultry manure is broadcast on the water surface each week.

Harvesting is carried out using a Fisheries Department beach seine. In 1989, the harvest from the pond was sold for a total of between c.6,000 and c.7,000. Fish are sold to market women for subsequent sale in both fresh and salted forms.

2.10. Osepo Fish Farm at Bebu, Kumasi, Ashanti Region.

Owner: Mr. Osei Poku.

The farmer is a retired journalist, has a concession for small-scale gold extraction on his land and also has plans for a tourist development. He started fish farming in 1983, stocking his ponds with a mixture of species including Sarotherodon galilaeus, O. niloticus, T. zilli, T. discolor, T. busumana, Heterotis niloticus, Clarias gariepinis and Ophiocephalus spp. In 1988, two ponds were stocked with adult and juvenile freshwater prawns (Macrobrachium vollenhoveni) collected from Ada.

Some of the tilapias seen during the visit of the mission (12 April, 1990) were hybrids with large red spots, and occasionally black spots, on their flanks. Very few are completely red and all are marketed along with other tilapias at the same price and without any significant problems.

Osepu Farms has a total of 12 ponds, 10 of which are stocked and in regular use. None of the ponds have monks. Two ponds have severe seepage problems and do not hold water. The total surface area of the 10 useable ponds is about 4 hectares.

The land is family land belonging to Mr. Osei Poku. A farm manager is employed full-time.

Fish are fed wheat bran, costing c.800 per 25–30 kg bag, and spent grain (draff from brewery waste) from breweries in Kumasi. Spent grain is bought at c.700 per truck-load but this is soon to be increased to c.1,000. Two bags of wheat bran are applied to four ponds each week.

The ponds are fertilised with chicken droppings. These are supplied free of charge although the farmer incurs a cost of c.6,000 on transportation from the poultry farm. Harvesting is carried out using the Fisheries Department beach seine. All 10 ponds stocked so far were first harvested in 1988. A total of 186 kg were produced. In 1989, six ponds were harvested for a total of 93 kg of fish. The selling price in 1989 was c.300 per kg., but the current selling price is c.400 per kg.. Despite the low yields obtained, the farmer is happy with results so far. Fish are sold from a pick-up truck to surrounding villages.

In August 1988, Mr. Osei Poku collected 800 adult and juvenile freshwater prawns (Macrobrachium vollenhoveni) at Ada in the Volta delta. They were transported by car to the farm in plastic bags. About 200 died during the transfer and the surviving 600 were stocked in two ponds, 300 prawns in each. The prawns were fed wheat bran and brewery waste. No sampling was carried out but the manager reported seeing prawns around the edge of the ponds every morning.

During the mission's visit and for the first time since stocking in August 1988, the two ponds were harvested using beach seines of 2cm. and 5cm bar mesh size. After three hauls in each pond ( two with 2cm. mesh net and one with 5mm mesh net) , no prawns were harvested. It appears that the prawns have disappeared from the ponds.

2.11. Darko Farms, Kumasi, Ashanti Region.

Owner: Darko Farms Ltd..

The company's main involvement is in intensive poultry production. The farm currently has 40,000 layers. According to the farm manager, Mr. Adade, the first of two ponds was constructed in 1984. A bulldozer was used for pond excavation at a total cost of c.120,000 (c.30,000 per day, four days for 1,500m. sq. pond area). A second pond of 1,400m. sq. was constructed in 1985 using hand labour. Both ponds are built on an area of land which was previously waterlogged. Plans exist to construct a third pond in August 1990. the manager intends to utilise a specialised hand-digging team which is now available in the area.

Fingerlings of different tilapias were obtained from three sources: a nearby river, Accra and the Owabi Water Works. About 3,000 fingerlings were stocked in Pond No. 1 (1,500m.sq.). The second pond was stocked solely with O. niloticus.

Fish are fed a mixture of wheat bran and fish meal, prepared on the farm from poultry feed ingredients at a cost of c.1,500 per bag of 30kg. Chicken manure from the farm is not used at present because it is mixed with wood shavings. However, Mr. Adade plans to begin using droppings from younger chickens which is not mixed with wood shavings.

The first year's production from Pond No. 1, which was harvested with assistance from the Fisheries Department, was 1,000 kg of large tilapia. This was sold at c.400 per kg.. Women buyers come to the farm gate to buy and local demand cannot be satisfied.

During the visit of the mission, sampling of one pond was carried out using a beach seine. About 25 kg of fish were harvested with the largest fish weighing 250 grams.

2.12. Vanassah Farm, Kumasi, Ashanti Region.

The owner of this farm has constructed two concrete ponds of about 300m.sq. each, close to his intensive poultry unit at Afari, near Kumasi. The water depth is about 0.6 m. and the water colour was dark green due to intensive manuring with chicken droppings. The cost of construction of these two ponds is not recorded.

The ponds were stocked with an unknown number of tilapia and Chrysicthys spp. The latter were stocked on 30 November, 1989. Average weight of the fingerlings was 25g.. During the mission's visit (12 April, 1990) a sample was taken and an average weight of 57g. was recorded., giving a weight gain of 32g. for 133 days or daily increase of 0.24g.. This is a very low weight gain in comparison with other species such as Clarias gariepinus and O. niloticus. Fisheries Department report that good results were obtained from this farm when Heterobranchus were raised in the same ponds along with tilapia.

2.13. Fish farm at Navrongo/Tono, Upper East Region.

Owner: Mr. Philip Abayori.

Mr. Abayori was Second National Best Farmer of Ghana in 1989 and is president of the local fishermen's association which has a membership of 200.

He has two newly constructed ponds, one of 630m.sq. and one of 564m. sq.. Water is provided from an irrigation canal. There are no monks in the ponds. At present, a piggery with five boxes is being constructed close to the ponds and the farmer intends to begin integrated pig-fish culture. The piggery has been located on land close to but at a lower level than the ponds. Manure has, therefore, to be transported to the ponds.

The ponds have been stocked with O. niloticus supplied from the ICOUR fish farm at Tono. No harvesting has taken place as yet.

The farmer's main activity is irrigated rice farming but he is keen to expand his fish farming operation and has plans to build a further 10 ponds. The main problems for fish farming are seen as being:

  • lack of technical advice
  • lack of published technical information
  • lack of credit.

2.14. Pacific Farms Ltd., Lashibi, Tema, Greater Accra.

Owner: Mr J. Bonney.

This farm was visited by a mission member (Dr. J Kapetsky) in August, 1989, and in 1988 during the preparatory assistance phase of the UNDP/FAO project “Integrated Approach to Aquaculture Development in Africa”, RAF/87/077.

The fish farm is situated at Lashibi, 3 km. from Tema and 20 km from Accra. Nine ponds of between 0.11 and 1.32 ha (0.25 and 3 acres) each were constructed in 1980 using a bulldozer. The total area of the ponds is 4.5 ha. and the total cost in 1980 was c. 220,000, bulldozer rental at that time being c.3,000 per 8-hour day. The corresponding cost in 1988 was c.40,000. (Satia and Vincke, 19891).

The other holdings on the farm are 8.8 ha (20) of irrigated rice fields, 2 pigsties which used to hold about 500 pigs, but now hold only 130; one pigsty is not being used. The two pigsties are linked by a canal to a 0.55 ha (1.25 acres) pond. The farm also has 17 cattle, a number of free range ducks and 1500 chickens.

The ponds are rain fed with some pumping, have no monks, no water inlets and cannot be drained. They are situated in an area where they are likely to benefit from floods.

Since their construction, the ponds have been stocked with tilapias (O. niloticus, Sarotherodon galilaeus, T. zilli) and Clarias spp. Gymnarchus, Ophiocephalus spp., Heterotis niloticus, Chrysichthys spp. and the freshwater prawn, Macrobrachium vollenhoveni (caught in the Volta Lake), have also been stocked at various times (Satia and Vincke, 1988). According to the owner, the M. vollenhoveni reproduced naturally in the ponds but the whole stock disappeared during the drought of 1983. They were fed irregularly with fresh tuna offal from Tema.

The ponds are fertilised once every 2–3 months with chicken manure but quantities applied are not known. Feeding is irregular and ingredients used include rice bran, brewery waste (c.300/tonne), maize (c.100–120 per kg) and dried anchovy (c.6,000 per 50 kg.). Harvesting of the ponds is also irregular and carried out according to demand. Ponds are skimmed out using a beach seine with a mesh of about 3 cm.

In 1988, the bulk of production was O. niloticus ranging from between 20 and 130 g.. Tilapia were sold in wooden crates containing about 30 kg at c. 3,500 per crate. Species other than tilapia were sold at prices which varied according to species and size.

In 1988 the farm's other products were sold at the following prices:

- Ducks: C. 600–800 per bird
- Poultry: C. 500–600 per kg (live wt.)
- Eggs: C. 23 each
- Pigs: C. 280–350 per kg (live wt.)

1 Satia B. & M.M.J. Vincke; Aquaculture Sector Survey. Ghana fAO; 1989.

The fish ponds are operated using mainly family labour with some help provided by 8 labourers whose main work is other farm activities. According to Mr. Bonney, the income from his 4.5 ha fish farm in 1987 was c. 400,000. On an average price of c. 200 per kg., the annual income indicates a production of 2,000 kg from the 4.5 ha -i.e. an average of only 400–500 kg/ha/year. However, Fisheries Department estimates that some ponds are producing about 3t\ha\year.

2.15. Comment on Private Fish Farms Visited.

  1. The farmers do not keep records of their activities. In most cases, they are not aware of the numbers of fingerlings stocked, the surface area of their ponds, the amount of feed or fertilisers used or the weight of fish harvested.

  2. Most farmers do not fertilise their ponds.

  3. Very few farmers have clear ideas about the ingredients which would be suitable for fish feeds.

  4. Most of the ponds are groundwater fed and are not drainable. Some fish always remain in the ponds after harvesting.

  5. Almost all the fish farmers express satisfaction with the yields obtained so far despite the fact that they are low in comparison with other African countries. Farmers' assessments are normally based on personal feelings alone as they do not generally keep records of inputs and outputs.

  6. All fish farmers in Ghana could at least double their current production purely through better management and slight increases in the present level of inputs.

  7. According to the fish farmers, the main constraints, in order of importance, are as follows:

    • lack of credit for pond construction
    • lack of technical information on fish culture and fish farming systems
    • lack of fingerlings of Clarias gariepinis
    • high cost of fish farming equipment, especially nets
    • poaching in ponds.

3. Government Fish Farms.

3.1. Afife Fish Ponds.

This fish farm is situated in the Afife Irrigation Project, is owned by the Irrigation Development Authority of the Ministry of Agriculture and is managed by the Fisheries Department. The farm was built at the same time as the Afife Dam in 1983. The dam was constructed, with technical assistance from China, to provide water to irrigate 800 ha for rice culture.

There are 10 ponds with a total area of about 5 ha (6 production ponds of 6,000m.sq. each, 2 stocking ponds of 1,100m.sq. each and 2 spawning ponds of 1,800 m.sq. each). The ponds are fed with water from a concrete diversion canal and are equipped with inlets and monks.

These ponds were handed over to the Fisheries Department under VORADEP by IDA towards the end of the VORADEP programme. Therefore, no major activity could be carried out and the only activity has been some pond maintenance work carried out in 1988 when the Fisheries Department employed two labourers. At the time of the mission's visit, only three ponds were well maintained and being used. The remaining ponds had weeds growing on the pond bottoms and several of the dikes were eroded.

The local District Assembly has expressed interest in managing the ponds.

The following operations are recommended for the rehabilitation of those ponds which have not been maintained:

  • cleaning of pond bottoms, including uprooting of weeds. Estimated cost is c.666,500, based on: total area 40,000m.sq. at 30m.sq. per man/day 1333 man/days at c.500 per day

  • repair of dikes. Estimated cost is c.875,000, based on: 3,500m. cube. of earth to move at 2 m.cube per man/day 1,750 man/days at c.500 per man/day

  • cutting grass on dikes. Estimated cost is c.25,000, based on: 15,000m.sq. at 300m.sq. per man/day 50 man/days at c500 per man/day

  • construction of office/store room. Estimated cost is c.502,400, based on: m.x 4m., 32m.sq. at c.15,700 per m.sq.

  • maintenance and repair of inlets and outlets. Estimated cost is c.10,000, based on:20 man/days at c.500 per man/day

Total estimated cost of rehabilitation using the methods outlined above is, therefore, c.2,078,900. It should be noted that these costs are indicative only and are based on a brief site visit. Detailed and comprehensive costing should be carried out by an engineer prior to any decision to invest. If it transpires that mechanical rather than manual methods are required, the costs will, of course, be substantially higher.

Once completely rehabilitated and functional, the Afife Fish Farm should be managed by a full-time, competent technician living on the site. The farm could become a very good demonstration facility for fingerling production, production of food fish and a training centre for technicians and fish farmers.

3.2. Libga fish ponds, Northern Region.

The Libga fish ponds were constructed in 1970. They are situated downstream of the Libga reservoir which stores water for the irrigation of 40 ha of rice. The location is 25 kms north of Tamale in Northern Region. The original purposes of the ponds included fingerling production, early rearing and experimental work.

There are seven ponds of which five are 250m.sq. each and two 2,000 m.sq. each. The latter were intended for hatchery and rearing purposes. Total surface area is 4250m.sq.. From the beginning, there has been a seepage problem and the ponds did not hold water after 1975. In an attempt at overcoming this problem, a layer of clay was compacted on the pond bottoms. However, this was subsequently washed out during draining operations. In 1987, a company then involved in the construction of Bontanga reservoir was commissioned to repair the ponds. They considered that the importation of clay was too expensive and attempted to seal the ponds with plastic sheeting. This, however, has proved to be ineffective and, at the time of the mission's visit (14 April, 1990), the plastic sheeting had disappeared and all seven ponds were without water. The water supply canal, water intakes and monks are all in good condition.

According to Broughton Associates Ltd (1980), Libga was one of the main fish culture stations in the Northern Region and was used for supplying tilapia fingerlings for stocking other ponds, dug-outs and reservoirs in the area. Lack of transport prevented fisheries personnel from properly carrying out programmes. The station was supervised by a technical officer and records were kept of research work on growth characteristics. Silver carp (Hypophthalmichthys molitrix) imported from Israel was at that time stocked in one pond. The bulk subsequently disappeared but five fish did attain a size of two kgs. by the second year (personal comm., Asafo, 1988). Species present at Libga in 1980 were: O. niloticus, Sarotherodon galilaeus, T. zilli, Lates niloticus, Auchenoglanis occidentalis, Heterotis niloticus, Labeo coubie and the silver carp.

The mission was requested by the Fisheries Department to assess the ponds and to propose, if possible, any action which could be taken to reduce or stop seepage in the ponds. Two previous attempts (introduction of clay and covering with plastic sheeting) to seal the ponds have failed. Reducing seepage in ponds after their construction is often very difficult and expensive but, in this case, a final attempt could be made with the use of manures.

Sealing of ponds with manures has been attributed to physical blockage of soil pores, with secondary biological clogging occurring later due to slime-forming micro-organisms. Two types of manure are recommended for testing: chicken manure and pig manure. Two tests should be carried out, one with fresh chicken manure and one with fresh pig manure and these should be applied in different ponds of 250m.sq. each.

  1. Chicken manure trial. The pond should be flooded by a single fill and as rapidly as possible by diverting all the available water from the supply canal. A water depth of one metre at the monk is required and, once filled, the loss of water in the pond should be measured each day at the same time. A staff gauge, graduated in millimetres and placed close to the monk, should be used. Loss of water, which will be high in the first week, has to be measured every day by taking a gauge reading. A water depth of one metre at the monk has to be maintained by diverting sufficient water into the pond. This water flow should be stopped at 1700 hours each day and the gauge read at 0800 hours the following morning. After this, the water flow should be restarted to reestablish a depth of one metre at the monk. In this way, the rate of seepage before treatment with manure can be established.

    After one week of such measurement, chicken manure should be applied. This should be fresh (wet), should include waste feed and wood shavings and should be broadcast over the pond surface at a rate of 1000 kg/ha/week - i.e. 25kg/week for the 250m.sq. pond. This application should be carried out each week for a period of two months. The necessary amount of chicken manure, 200 – 250 kgs., should be stored close to the pond at the beginning of the experiment.

    Water should be added to the pond each morning to reestablish a depth of one metre at the monk. If the treatment is successful, water loss will be reduced to normal rates of between 7 and 10 mm. per day.

  2. Pig manure trial. A trial, following a parallel programme to that described above, but with the use of pig manure, should be conducted in another 250m.sq. pond. The pig manure should be applied at a rate of 20,000 kgs/ha/week (2kg/m.sq./week). That is, for the 250m.sq. pond, 500 kgs of pig manure should be applied each week. The total quantity required for the two month trial period will be between 4,000 and 4,500 kgs and this should be stocked close to the pond at the start of the experiment. The manure could be transported in 200 litre petrol drums, opened at one end and placed in the upright position.

The same procedures as those recommended for the chicken manure trial should be strictly followed and similar results could be achieved. If, however, the use of manure does not result in significant reduction in the rates of water loss, the ponds should be definitely considered as unsuitable for fish farming and abandoned.

3.3. Bontanga fish ponds, Northern Region.

The proposed Bontanga fish farm, which was intended to be part of an irrigation project situated downstream of the Bontanga dam, about 25 kms. west of Tamale, has never been completed. Ponds were to be constructed in an area where clay was extracted for dam wall construction, but this was not carried out (Personal comm., Asafo, 1990).

3.4. Golinga fish ponds, Northern Region.

The Golinga Irrigation scheme is situated 20 km west of Tamale. Three ponds, each of 2,400m.sq., were built downstream of the dam in 1975. They are managed by the Fisheries Department. Two labourers are employed. At the time of the mission's visit (14 April, 1990), two ponds were operational and one had been abandoned because of heavy seepage.

The two operational ponds are stocked with O. niloticus and T. galilaea. Rice bran is used for feeding and daily records of feed distributed, and of Secchi disc readings, are kept. The farm produces about 10,000 fingerlings for sale to fish farmers and for stocking of reservoirs.

As the dam capacity is limited, there is competition for use of water between irrigated crops and the fish ponds. There is also a need for a small building to store feed and farm equipment.

3.5. Nasia fish farm, Northern Region.

The Nasia ponds were constructed in 1973 as part of the Nasia irrigation scheme. The scheme is situated 80 km north of Tamale on the main road to Bolgatanga.

There are 11 ponds and these were estimated by the Mission (15 April, 1990) to have a total area of approximately 16,000 m.sq.. Water has to be pumped from Nasia river at a point two kms. away from the crop land and fish ponds. The pumping system became defective soon after the completion of the ponds which have never been effectively used. If water does become available, the ponds could be rehabilitated and used for fingerling production and commercial fish farming.

3.6. Vea fish ponds, Upper East Region.

These ponds are owned by ICOUR and the Fisheries Department (see section 4.2) and are located 6 km west of Bolgatanga. The Fisheries Department, however, is responsible for the management of 4 rearing and 4 spawning ponds. They were constructed in 1980.

The rearing ponds have a total surface area of 12,350 m sq and the spawning ponds a total of 1,800 m sq. All the ponds are well constructed but poorly managed. Water supply is by gravity from the Vea reservoir.

One pond has two pigsties but these have level floors and pig manure has to be transported to the pond by wheelbarrow. The pigs are fed “pito” waste - i.e. dried sorghum waste produced after the preparation of sorghum beer or “pito”. The pig-cum-fish culture is considered by the Department to be experimental. However, there was no information available from farm staff or from the Regional Fisheries Office on stocking rates, numbers of pigs during the culture period or the rate of manuring.

The spawning ponds are reported to be producing approximately 24,000 O. niloticus fingerlings every 4 months. The broodstock and fingerlings are not fed. The reason for this, according to the Regional Fisheries Officer, is that feeding of fish is a practice which farmers are not prepared to do and it would, therefore, be inappropriate to do so on the department station. The fingerlings are sold to private fish farmers.

The Fisheries Department also owns a fisheries training centre and this is located on the edge of the Vea reservoir. The building, which was constructed in 1980, consists of :

  1. one office, equipped with a desk and chair, two cupboards and three shelves

  2. one other office, equipped with a desk and chair, two cupboards and one filing cabinet

  3. one wet laboratory with five tables, each for six persons, five wash stands, five cupboards, four laboratory tables and several chairs

  4. one room currently used for the storage of feed and fertilisers for the fish farm

In addition to the above, there are four toilets, a water tank, one large concrete platform for drying fish and a watchman's house. The building is not supplied with electricity, although it is wired for that purpose. According to information given, the building has not yet been used and the offices and laboratories are not occupied. For a possible use of this facility, see Section 5.2.

3.7. Nafkolga fish ponds, Upper East Region.

The Nafkolga (or Nafkuliga) dam was built in 1959 by the Irrigation Department of the Ministry of Agriculture. The dam is situated 30 kms south-west of Bawku. Water from the reservoir is used for irrigation. Six ponds, each of 8,400 m sq (total 50,400 m sq), were constructed immediately downstream of the dam in 1960. They have been used at various times for reproduction of O. niloticus. However, they have had to be abandoned several times to repair leaks and their use has, therefore, not been continuous.

There are three intake valves, two for irrigation and one for water supply to the ponds. The latter has not been operational since 1986. During the mission's visit (16 April, 1990), no repair works to the valve were in progress and all the ponds were dry.

3.8 Kpando Fishery Complex, Volta Region.

As requested by the Fisheries Department, the mission visited this complex to assess its suitability for development as an aquaculture training establishment. The mission concluded that this facility would not be suitable for aquaculture training but that its present use as a centre for training in capture fisheries could be significantly enhanced by rehabilitation.

3.9. Comment on Government Fish Farms.

According to published information, there are 15 government fish farms managed by the Fisheries Department. (Pedini, 1987; Balarin, 1988; Satia and Vincke, 1988; Fisheries Department, 1988). In fact, one of these farms, which is 16 ha. in size and is located on an irrigation project at Asutsuare, has been listed as “under construction” since 1986. This farm is adjacent to a proposed joint Hungarian/ADB aquaculture project. In 1986 site selection was carried out and ponds designed but there has as yet been no construction work undertaken. There are, therefore, only 14 government fish farms in existence.

Of these 14 farms, only four are operational and the remainder are either not functioning or abandoned. There are a variety of reasons for this but the main one is wrong site selection, reportedly arising from a refusal by the management authorities (IDA) to site the ponds in suitable areas. Lack of water and loss of water through seepage are some of the problems associated with these farms.

Most of the farms are poorly managed, there is a lack of technicians and farm managers and the work which is carried out is by labourers who do not have well-defined work programmes. Limited funding, lack of transport and fuel shortages prevent fisheries personnel from carrying out their programmes. Fisheries Department also lacks sufficient personnel with sufficient knowledge about fish culture techniques and practices.

4. Parastatal fish farms.

4.1 Tono Aquaculture Project, Upper East Region.

The Tono irrigation scheme, which is situated 10 kms. north of Navrongo, is one of three medium-sized irrigation projects in Northern Ghana. Water from the Tono Reservoir (1540 ha at high water level) irrigates 2,480 ha of land and an area of over 60 ha has been allocated for aquaculture development.

IDA constructed the existing ponds in 1985 and ICOUR manage the water supply. At Tono, ICOUR has a well-managed and welldesigned fish farming research station with a total of 9 ponds, each of 0.1 ha. and with a total surface area of 9,000 m sq.. There are an additional 6 spawning ponds, each of 50 m sq, although these are dry and are no longer used. Sex reversal of O. niloticus through hormone is carried out in two concrete tanks of about 6 cubic metres.

Two ponds are stocked with different species of river fishes, such as Labeo coubie, for observations on growth rates in ponds. Another two ponds, each of about 1,000 m sq, are being used as breeding ponds and are stocked with 3,000 females and 200 males of O. niloticus per pond. These are seined every two weeks (two hauls per pond) and between 2,000 and 3,000 fingerlings per pond, with an average weight of 10g., are produced. The fingerlings are then stocked in four 1,000 m sq ponds and are distributed to fish farmers from these. Current selling price is c.5 per fingerling.

One pond has been stocked with 3,000 all-male O. niloticus for monosex culture as a demonstration for farmers. All fish on the farm are fed a mixture of rice bran, pito waste and ground, sundried O. niloticus females which have been discarded during hand sexing of fingerlings.

The ICOUR Fisheries Unit has also constructed eight ponds which are to be leased to fish farmers. During the visit of the mission (17 April, 1990) these 6,000 m sq ponds had been stocked with O. niloticus fingerlings. One pond had been allocated for rice cultivation but it is intended to cultivate fish in this pond after the rice is harvested. Four other ponds were empty. The current ICOUR charge for irrigation is c. 20,000 per ha. per year.

In future years, the Fisheries Unit intends to concentrate on the development of sex-reversal (all-male O. niloticus) and rice-cum-fish culture.

4.2 Vea Fish Farm, Upper East Region.

Fisheries Department owns and runs four grow-out ponds and four hatchery ponds at Vea, while ICOUR is responsible for the remaining 16 grow-out ponds. (See Section 3.6). The ponds were constructed in 1980 and, of the 16 ponds under ICOUR's management, all are being leased to farmers for rice cultivation.

The mission interviewed a fish farmer who in 1988 rented one and subsequently two ponds from ICOUR. These were stocked with O. niloticus fingerlings provided by the ICOUR Fisheries Unit. Fertiliser and feed were also supplied by the Unit and assistance in harvesting and transport of fish to Bolgatanga was provided at the end of a six month growing period. The farmer himself sold the fish (mainly tilapia and a few Clarias gariepinis) to both fish mammies and consumers.

Total income from the first pond (1,200 m sq) for the first harvest was c. 19,000 and the selling price was c. 150 per kg.. This suggests a harvest of about 127 kg or about 2,100 kg/ha/year. Costs were reported to have been c. 3,000 for pond rental and c. 3,000 for feed. When the farmer operated the two ponds, his income came to c.20,000 per pond but expenses had increased because of increased fertiliser prices.

4.3 Kpong Farms, Eastern Region.

Kpong Farms is a subsidiary of the Volta River Authority. It is located at Akuse near the Kpong Dam. The company's main activity in the area is the development of irrigated agriculture which now focuses on the cultivation of rice. The farm has a small fish farm unit based on the sewage treatment plant for the town waste water. Effluents are diverted in two parallel series of three 0.4 ha. ponds. Water passes through the ponds and arrives at a single pond (Pond No.1) of 1.2 ha. There are, thus, a total of seven ponds, none of which are drainable. They all have steep slopes and are very deep. The total area is 3.6 ha.

In 1986, the ponds were stocked with unknown quantities of the following species: O. niloticus (main species), Sarotherodon galilaeus, T. zilli and Hemichromis fasciatus (Ofori, pers. comm., 1988). Some Peking ducks were stocked in Pond No.1 (Satia and Vincke, 1989; see footnote 1).

The ponds are irregularly fished using seine nets and records have been kept since November, 1987. The harvests are not recorded pond by pond or species by species, but only by series of hauls in different ponds as shown in the following table:

No. of Ponds fished Total Area of ponds (ha) No. of hauls Total catch (kg) Ave catch per haul (kg)
Nov,87.. 2 0.8 5 184.5 36.9
Dec,87.. 1 0.4 4 567.0 142.0
Jan,88.. 3 2.0 8 823.0 103.0
Feb,88.. 3 1.2 9 944.0 105.0
Mar,88.. 2 0.8 11 1.275.5 115.9

Source: Satia and Vincke, 1989.

The estimated average size of the different species at harvest is as follows (Ofori, pers. comm., 1988):

O. niloticus 200g.
Sarotherodon galilaeus 150g.
T. zilli 50–70g.
H. fasciatus 100g.

Recent fish harvests from the Kpong ponds are given in the following table:

Month 1988 (kg) 1989 (kg)
Jan. 823 41
Feb. 944 343
Mar. 1,276 503
Apr. n/a 300
May n/a 64
June 322 288
July 124 80
Aug. 173 72
Sept. 0 -
Oct. 0 -
Nov. 0 -
Dec. 36 -
Total(kg) 655 1691

The above levels of production from 3.6 ha. indicate yields of between 182 kg (1988) and 470 kg (1989) per ha/year.

4.4 Institute of Renewable Natural Resources Fish Farm, Kumasi, Ashanti Region.

The Institute of Renewable Natural Resources (IRNR) of the University of Science and Technology (UST) at Kumasi has an experimental fish farm close to the campus. The farm consists of six fry production ponds, each of 200 m sq, five production ponds, each of 300 m sq, and two holding ponds, each of 1,000 m sq. The holding ponds have broodstock of male and female Heterobranchus longifilis. The total surface area of the 13 ponds is 4,700 m sq (0.47 ha.). There is also a small hatchery made from 200 litre petrol drums, although this is not yet operational, and a building with office, laboratory and store-room.

The main interest of the IRNR research staff is in induced spawning and fingerling production of the catfish, Heterobranchus longifilis, and sex reversal of O. niloticus. There is no research on Macrobrachium vollenhoveni. Unfortunately, because of a lack of access to scientific aquaculture publications and lack of contact with other research institutes in Africa, the research staff of IRNR were not aware that successful induced spawning of Heterobranchus had been achieved 3–4 years ago in Cote d'Ivoire and, more recently, in Nigeria.

The Institute has regular contacts with the Fisheries Department and cooperates with IAB, although not with other universities. The staff want to be involved in the training of technicians and farmers and in the provision of extension services but funding for such activities and qualified personnel are lacking. The Institute offers training in aquaculture at the BSc. level but staff complain that trainees are not absorbed by the Fisheries Department.

The main problems of IRNR, according to discussions held with the staff on 10 April, 1990, are the following:

  1. lack of travel grants to visit aquaculture establishments in other areas

  2. lack of a coordinating mechanism for effective cooperation between the Fisheries Department, other institutes and universities involved in aquaculture development, training and research. It is felt that regular, official meetings would make it possible to establish research priorities, would help avoid duplication of effort and would improve the chances of successful implementation of valid research projects

  3. lack of facilities for practical training of students

  4. lack of exposure of lecturers to basic, practical techniques

  5. lack of equipment, such as fish transport tanks, boxes and laboratory glassware.

4.5 Aquaculture Research and Development Centre, Akosombo, Eastern Region.

The Institute of Aquatic Biology (IAB) has an Aquaculture Research and Development Centre (ARDEC) at Akosombo. Construction of the centre started in 1979 and has yet to be completed. ARDEC has the following facilities: 20 earthen ponds of 0.2 ha each, four ponds of 0.7 ha each, 14 concrete circular tanks, 30 rectangular concrete tanks and an office-store-laboratory complex. Water is supplied to the ponds and tanks by pumping from the Volta River.

The research programme is mainly concerned with Tilapia genetics, the effects of insecticides used for the eradication of Onchocerciasis (river blindness) on fish populations, different methods of fry production and sex reversal using hormones.

Funding for the research programme is one of the main problems of IAB and ARDEC.


The ICOUR managed fingerling production centres at Vea and Tono are supporting small-scale fish farmers by providing them with O. niloticus for stocking ponds and, in some cases, for rice-cum-fish culture in irrigated ricefields.

The ponds constructed by ICOUR and rented to farmers are used most of the time only for rice cultivation, rather than for fish or fish-cum-rice culture. In irrigated rice cultivation (two crops per year), profit is about c. 193,000 per ha/year compared with profits estimated at about c. 300,000 per ha/year for rice-cum-fish culture. Profit from O. niloticus culture, using rice bran only as feed, is estimated at c. 85,000 per ha./year. Tilapia farming integrated with pig farming gives an estimated profit of c. 2,400,000 per ha/year (sales of pigs on the hoof included).

The Kpong fish farm, which belongs to VRA, is not adequately managed and could produce more than is now the case. The maximum yield from the 3.6ha. of ponds is not more than 470 kgs per ha./year.

The ARDEC of the Institute of Aquatic Biology at Akosombo, which has been under construction since 1979, is still not fully operational and, during the mission's visit in April 1990, only three ponds were in use. The research staff are still living in Accra and not at the farm site at Akosombo.

The IRNR of the University of Science and Technology at Kumasi has relatively good facilities which could be used for training and research.

5. Recommendations.

The following recommendations are aimed at the enhancement of private fish farms, government fish farms and parastatal organisations involved in fish farming and they are based on field observations, interviews of fish farmers and discussions with staff of institutions involved in aquaculture.

5.1 Private Fish Farms.

It is recommended that private fish farmers, as a matter of urgency, be provided with adequate technical advice concerning the following:

  • site selection before pond construction
  • choice of species for pond culture
  • stocking densities
  • culture systems, particularly for integrated fish culture, and animal husbandry
  • feeding of fish using locally available agro-industrial by-products
  • choice of available feedstuffs
  • manuring of ponds
  • duration of grow-out period
  • use of syphons for pond draining
  • harvesting
  • marketing, particularly in relation to problems which may arise when supplies become heavy

The above advice is principally related to extension support. It is, therefore, strongly recommended that:

  • existing fish farmer extension be incorporated into the Agriculture Extension Service, as envisaged by the Fisheries Department

  • extension staff be trained and be provided with adequate funding for the transport and equipment needed for an efficient service to fish farmers

  • the extension service provide the fingerlings required by fish farmers in their areas

  • the Ghana Fish Farmers' Association play a more prominent role in the promotion of the industry

  • the Association be given more institutional support to help increase its level of activity

5.2 Government Fish Farms.

Of the 14 government fish farms, only four are actually operational and these are not managed as they should be (See Section 3.8). It is, therefore, recommended that:

  • the Afife Fish Farm be rehabilitated to become an efficient fingerling production centre, a demonstration farm and a training centre for technicians and fish farmers (see Section 3.1)

  • the Libga fish ponds be restored using simple techniques (see Section 3.2). In the event of failure to seal the ponds, they should be abandoned

  • if water becomes available, the Nasia fish farm be rehabilitated (see Section 3.5)

  • a fishery training centre be established at the Vea, using the existing un-utilised facility, by undertaking repairs and re-arranging the present accommodation to create a lecture room and dormitory

  • training be provided at this centre for technicians, extension staff and farmers

  • the associated practical training be provided at the nearby Vea ponds (see Section 3.6)

  • the intake valve at Nafkolga be repaired and the ponds be rehabilitated to create a fry production facility for stocking of both ponds and dug-outs in the Bawku area

  • technicians be trained in efficient fish farm management

5.3 Parastatal Organisations.

As the likely profits from fish farming in the ponds at Vea and Tono are greater than what is being achieved from the current use for rice cultivation, ICOUR's Fisheries Division should review their policy relating to the lease of the ponds.

Similarly, at Kpong fish ponds, a revised plan of production should be introduced with the aim of achieving yields of at least 4 to 5 tons/ha/year (See Section 4.3)

Funding for the implementation of research projects which have a direct impact on aquaculture development should be made available to the Institute of Renewable Natural Resources at Kumasi.

The ARDEC facilities at Akosombo should be made fully operational and the research staff outposted to Akosombo (see Section 4.5 and 4.6).

A coordination committee, headed by the Fisheries Department, should be established to coordinate research activities and to ensure that the research programmes address the real needs of aquaculture development in Ghana. It is recommended that the following applied research projects, in the following order of priority, be undertaken to support aquaculture development in Ghana:

Feeding trials under local conditions

  • testing in ponds of simple feedstuffs (2 3 months)
  • testing optimum mixtures of feedstuffs (2 6 months)
  • testing of manures (2 6 months)
  • testing of inorganic fertilisers (2 6 months)

Artificial reproduction of Clarias gariepinis

  • construction of hatchery

  • induced spawning trials with broodstock from ponds

Freshwater Prawn, Macrobrachium vollenhoveni

  • survey of biology in natural waters

  • grow-out trials of wild stock in ponds and reservoirs

  • natural reproduction in ponds

  • artificial reproduction in hatchery


  • survey of physio-chemical conditions (ecological aspects)

  • socio-economics

  • improvement of catches (Acadja/brush parks)

  • potential for marine shrimp culture

Experimental work on the use of Azolla.

  • testing of Azolla in rice fields stocked with fish

  • testing of Azolla with pond fish culture

Finally, it is recommended that there be a project established with the objective of upgrading the capability of IRNR to provide graduate and post-graduate level education in aquaculture. The main inputs of the project would be the provision of expert staff, in the form of one coordinator for a period of two years supported by short-term consultancies in the areas identified as being of priority for research (see above). The expert staff would provide instruction to IRNR lecturers and researchers over a two year period.

The experts would also establish applied research programmes in Ghana with IRNR staff and links with appropriate university departments and institutions in other countries could be developed so that the Institute's staff capabilities and qualifications (e.g. through higher degrees) could be further upgraded.

Enter supporting content here

Okurantsir Amanfo