FI: DP/IVC/71/526/6 1 April 1974

RAPPORT TECHNIQUE No. 19

by

Z.H. Shehadeh
Fishery Resources Officer (Aquaculture)

This report is one of a series of reports prepared during the course of the UNDP project identified on the title page. The conclusions and recommendations given in the report are those considered appropriate at the time of its preparation. They may be modified in the light of further knowledge gained at subsequent stages of the project.

The designations employed and the presentation of the material in this document do not imply the expression of any opinion whatsoever on the part of the United Nations or the Food and Agriculture Organization of the United Nations concerning the legal or constitutional status of any country, territory or sea area, or concerning the delimitation of frontiers.

ACKNOWLEDGEMENTS

The Food and Agriculture Organization is greatly indebted to the following organizations and individuals who assisted in the implementation of the project by providing information, advice and facilities:

Dr. A. Coche - Project Manager
Dr. A. Fanny - Project Co-Manager and Personnel
Mr. V.G. Frank - Fishery Extension Officer
Mr. S. Krzelj - Limnologist
Mr. R.G. Lefebvre - Naval Architect
Dr. P. Planquette - Director of the CTFT Fish Culture Station, Bouake
Mr. F.C. Roest - Fishery Biologist

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS
Rome 1974

Hyperlinks to non-FAO Internet sites do not imply any official endorsement of or responsibility for the opinions, ideas, data or products presented at these locations, or guarantee the validity of the information provided. The sole purpose of links to non-FAO sites is to indicate further information available on related topics.

This electronic document has been scanned using optical character recognition (OCR) software. FAO declines all responsibility for any discrepancies that may exist between the present document and its original printed version.

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

1. INTRODUCTION

1.1 Terms of Reference
1.2 Background Information

2. OBSERVATIONS AND ACCOMPLISHMENTS

2.1 Availability of Species Suitable for Cage Culture
2.2 Availability of Suitable Sites

2.2.1 Cage Culture
2.2.2 Enclosures

2.3 Feed

2.3.1 Availability of feed and feed components
2.3.2 Preparation of feed for cage-rearing tests

2.4 Cage Materials and Cage Construction

3. RECOMMENDATIONS

LIST OF TABLES

1. Ingredients available for preparation of fish feed in Bouake/Abidjan
2. Composition of ingredients available for preparation of fish feed in Ivory Coast
3. Chemical analysis of poultry feed prepared by Cocoservice, Bingerville
4. Cost of materials suitable for cage construction available in Abidjan/Bouake

LIST OF FIGURES

1. Geographic location of Lake Kossou
2. Detail of Lake Kossou
3. Dissolved oxygen and water temperature: Dam Station
4. Dissolved oxygen and water temperature: Konankuikro Station
5. Dissolved oxygen and water temperature: Station 0313
6. Fluctuations in air/water temperatures over a 24-hour period during “Harmattan”
7. Effect of turbine operation on dissolved oxygen and water temperature
8. Galvanized wire/mahogany frame cage
9a. Modified galvanized wire/mahogany frame cage
9b. Detail of feeding trough and styrofoam box
10. Double net/pine wood frame cage

1. INTRODUCTION

1.1 TERMS OF REFERENCE

The Government of the Republic of the Ivory Coast, assisted by the United Nations Development Programme and the Food and Agriculture Organization of the United Nations are engaged in a project whose main purpose is to assist the Government in developing and exploiting the fishery to be created in Lake Kossou, located in the Central Province.

The project became operational on 18 January 1972 and, as part of the project operation, FAO assigned a Fishery Resources Officer, (Aquaculture) from 18 February to 1 March 1974 with the following terms of reference “to investigate the availability of seed resources of selected fish species considered suitable for cage culture; determine the availability of feed or feed components; and initiate activity in feasible areas.

1.2 BACKGROUND INFORMATION

The UNDP/FAO Lake Kossou Fishery Development Project, which became operational on 18 January 1972, was aimed to develop a commercial fishery on the lake and to train villagers in fishing techniques as a new socio-economic occupation. The creation of the Lake necessitated the displacement of about 75 000 villagers predominantly of an agricultural occupation, and the inundation of 1 600 km² of agricultural land.

It has been estimated that the required annual supply of fish in the Bandama valley will amount to about 35 950 t by 1975 and 48 965 t by 1980¹. The yield from the lake fishery on the other hand, is expected to stabilize at a maximum of about 9 000 tons per year.

¹ Commercialisation du Poisson; report of H. Renson, FAO Fish Marketing Expert, July 1972.

In November 1973, at the request of the AVB, Dr. T. Scudder, Senior UNDP Consultant, conducted a socio-economic study in the Bandama Valley. In his recommendations for the future development of the area he stressed the importance of the development, intensification and diversification of the lake fishery and in particular the urgent necessity to investigate the feasibility of controlled intensive fish culture in cages and enclosures.

In view of these facts and as one of the objectives of the project is to conduct a preliminary study of the feasibility of aquaculture in the lake, the present mission was undertaken.

Lake Kossou, formed by damming of the White Bandama River, lies in the centre of the country between 5° and 6° longitude and 7° and 8° latitude (figure 1). The level of the water at present (February 1974) stands at 188 m above sea level; maximum water depth is 33 m. On complete filling the lake will have a water level at 204 m, maximum depth of 49 m and total surface area of 1 600 km ².

Meteorological conditions in the lake area are characterized by a wet season during July to September (May to September in the northern parts of the lake) with a total annual precipitation of about 120 cm. Mean monthly air temperatures fluctuate between 25° and 29° C with minimum and maximum values ranging between 15° and 38° C during the dry season and 18° and 32° C during the wet season respectively. The south-southwestern winds prevail except during the period December to February when desert winds (“Harmattan”) blow from the north-northeast. Wind velocity varies between 4 and 14 km/hour.

The lake undergoes two mixing periods each year - beginning of the wet season and again during “Harmattan”. Due to the great amount of bottom organic matter still undergoing decomposition since the inundation of the valley, water mixing brings about rich plankton blooms, oxygen depletion and occasional fish kills. The lake area most affected is shown in Figure 2 and at times extends toward the dam. Similar, but more restricted, mixing and fish kills occur in the area adjacent to the dam whenever water is drawn through the turbines.

The lake stratifies during the dry season excluding “Harmattan”; the depth of the thermocline at the present water level varies between five and ten meters with a maximum temperature gradient of 3° C. Surface water temperature and dissolved oxygen values in the main channel of the lake (depth of 25–33 m) are 27–28° C and 3–4 mg/l respectively during the wet season and 29–30°C and 4–6 mg/l during the dry season.

2. OBSERVATIONS AND ACCOMPLISHMENTS

2.1 AVAILABILITY OF SPECIES SUITABLE FOR CAGE-CULTURE

Fish species to be cultured in cages should have the following attributes:

Adaptability to high density stocking (e.g. 200–300/m³)

Rapid growth rate

When expensive, nutritionally complete artificial feed is used exclusively, fish must accept and convert the feed to flesh at an acceptable rate. Under such conditions feed can account for as much as 60–70 percent of production costs; a small difference in conversion rate would, consequently have a significant effect on the margin of profit.

Satisfactory market value, especially when expensive feeds are used.

As in the case of any cultured species, there should also be a readily available source of fingerlings for stocking, or methods for the artificial breeding and rearing of fingerlings of the chosen species must already exist.

A review of the species composition of fish populations in the lake revealed some which have been commercially reared in captivity. These are: the cichlids - Tilapia nilotica, T. galilaea, T. melanopleura, T. zilli; the clariids - Clarias lazera, C. senegalensis and Heterobranchus longifilis; and the centropomid, Lates niloticus. Some comparative information on the suitability of these species for cage culture in the lake is given below. The prices noted apply for fish sold at the lake; prices are at times 100–150 percent and 300–400 percent higher in Bouake (150 km from Mossou) and Abidjan (400 km from Kossou) respectively.

1. Clariidae

   1. C. lazera and C. senegalensis

      Local name (s): Gué blé, Djomété

      Price: 250 Fr. CFA/kg for 250 g fish
      300 Fr. CFA/kg for 1 kg fish

      Major breeding period July to December; breeders can be collected from the lake by longline fishing; the feasibility of collecting young from the lake in great numbers not as yet determined; bred in captivity by fluctuating pond water level or injection of hormones (DOCA); mass rearing of juveniles from eggs possible with some problems yet to be resolved. C. senegalensis not yet bred in captivity but will probably respond to treatment as in the case of C. lazera.

      Both species are omnivorous and C. lazera is known to accept artificial feeds with feed conversion rates of 1.0–1.2 in ponds (30 percent protein feed) and a growth rate of 300–400 g/3 months (Bangui, Central African Republic).

      As yet not cultured in cages and reaction to high density culture not known; clariids possess accessory respiratory organs for aerial breathing and are particularly suited to dense culture from the viewpoint of oxygen requirements.

   2. H. longifilis

      Local name, market value and breeding season as for Clarias sp.

      Does not breed in captivity; artificial breeding not yet attempted

      As yet not cultured in cages and reaction to high density stocking not known

      Attains a weight of 300–400 g/3 months in ponds when stocked at the rate of 1 000/ha in conjunction with tilapia.

2. Cichlidae (Tilapia sp.)

   Price: 200 Fr. CFA/kg for 250 g fish
   100 Fr. CFA/kg for fish under 250 g

   Breed several times per year in ponds and the lake; collection of large numbers from the lake possible

   Accept artificial feed with good conversion rates; growth rate is low (100 g/3 months)

   Amenable to dense stocking (T. nilotica grown in cages at stocking rate of 288/m³ in the United States).

3. Centropomidae

   1. Lates niloticus (local names: Ngouin, Dro)

      Price: 200 Fr. CFA/kg for 250 g fish

      Breeds in ponds; mass rearing of young not attempted

      Carnivorous; acceptance of artificial feed not known - will probably require feed with very high protein content

      As yet not cultured in cages and reaction to high density stocking not known.

It would appear, from consideration of acceptance of artificial feed, growth rates, market value and hardiness, that clariids are the best candidates for cage culture provided high density stocking will be possible. The possession by these species of accessory respiratory organs for aerial breathing simplifies the transport of great numbers of fingerlings and minimizes the importance of dissolved oxygen for their culture. However, although adult breeders of Clarias and Heterobranchus can be obtained from the lake, production of young through artificial breeding cannot be carried out at this time in so far as existing government fish culture stations are not operational. The CTFT¹ Fish Culture Station at Bouake is the only functioning facility; tanks and pond facilities, however, are occupied by on-going research on culture of Lates - Tilapia combinations and Auchenoglanis sp. Adult breeders of C. lazera have been imported to the station from Bangui and bred in ponds with the assistance of J. Micha, FAO Fish Culture Expert (Bangui). In addition, little or no information is to hand on the availability and distribution of clariid juveniles in natural waters. The absence of flood plains in the country, where these fishes are usually collected in great numbers in other parts of Africa, adds to the difficulty of obtaining adequate numbers for culture.

¹ Centre Technique Forestier Tropical.

These problems were discussed during a meeting held in Abidjan and attended by personnel from the Fisheries Department, CTFT, the Institute of Tropical Ecology, the Project Manager and the consultant. It was agreed that CTFT would commence work on the breeding and rearing of Clarias lazera and/or C. senegalensis on completion of Auchenoglanis sp. tests. Although the need for a survey of natural seed resources was agreed upon, it was not possible to schedule this work due to lack of adequate funds.

There also exists the possibility of cage rearing of phytophagous tilapia species such as T. melanopleura and T. zilli. Juveniles can be easily obtained from breeding ponds and suitable vegetation such as leaves of banana, manioc, sweet potato, etc. are readily available in the lake area. Conversion rates of feed that can be expected from such leafy material will be in the order of 5–7:1 with a growth rate of about 150–200 g/year. Whether dense stocking and cheap feed would compensate for the low growth rate and market value to give acceptable economic returns should be the subject of field tests.

2.2 AVAILABILITY OF SUITABLE SITES

2.2.1 Cage culture

The lake, by virtue of its numerous lateral extensions, offers a great number of bays of varying depths sheltered by low hills or dense forest. Submerged and semi-submerged trees are a constant characteristic of these sites. Many of the bays offer good sites for cage culture from the point of view of water depth, circulation and shelter. The main channel of the lake along the old river bed, which has been cleared for navigation, is characterized by a relatively strong current throughout the year and rough waters during the wet season. The placement of cages in this area is not advisable as fish would continuously swim against the current with undesirable expenditure of energy, servicing of cages would be difficult and artificial feed, other than floating pellets, would tend to be lost from the cages. Suitable locations exist, however, around bends in the lake channel where the current is moderate and calmer conditions exist in the wet season.

In order to determine the suitability of some of the above sites for cage culture two stations were chosen before the arrival of the consultant and records made of water temperature and dissolved oxygen during the period July 1973 to January 1974. In addition data collected during the course of an entire year (1973) were examined for one station in the main lake channel. The location of the stations are shown in Figure 2; water conditions were as follows:

1. Dam station (Figure 3)

   Location: northwest corner of dam in a small bay

   Water depth at station: 4 m

   Station sheltered from SW wind and exposed to NE wind

   Current: negligible.

   1. Dissolved oxygen

      Surface: about 2 mg/l (25 percent saturation) during rainy season, 4–6 mg/l (50–80 percent saturation during dry season, 1–6 mg/l (13–78 percent saturation) during “Harmattan”

      1 m depth: as for surface water during wet season; 3.5 mg/l (45 percent saturation) during season; 0–1 mg/l (0–13 percent saturation) during “Harmattan”.

   2. Temperature

      Surface at 27–30° C; bottom one to two degrees lower; water temperature one to two degrees lower during wet season and “Harmattan”.

2. Konankuikro station (figure 4)

   Location: long, narrow sheltered bay on eastern shore of lake opposite dam

   Water depth at station: 4 m

   Station protected from NE wind; exposed to SW wind

   Current: negligible.

   1. Dissolved oxygen

      Surface: stable at 3 kg/l (40 percent saturation) during wet season; 3–6 mg/1 (40–70 percent saturation) during dry season; 3–4 mg/l (45–48 percent saturation) during “Harmattan”

      1 m depth: as for surface during wet season; decreased from 3 mg/l (58 percent saturation) to 1 mg/l (14 percent saturation) between wet season and “Harmattan”; 2–4 mg/l (25–50 percent saturation) during “Harmattan”

      Bottom: 1–4 mg/l (25–50 percent saturation) during most of year; 0–2 mg/l 0–25 percent saturation) late part of wet season and dry season.

   2. Temperature

      Surface at 28–29° C during wet and dry season; about 27° C during “Harmattan”; maximum difference of 1° C between surface and bottom water.

3. Station 0313 (Figure 5)

   Location: main lake channel over old river bed

   Water depth; 25 m

   Station exposed to NE and SW winds

   Current: relatively strong.

   1. Dissolved oxygen

      Surface: 6 mg/l (76–78 percent saturation) during dry season; 4 mg/l (52 percent saturation) during wet season; 8–10 mg/l (100–125 percent saturation) during the following dry period; 5–6 mg/l (64–78 percent saturation) during “Harmattan”

      1–3 m depth: significant difference from surface values during period between end of wet season and “Harmattan” with values of 5–7 mg/l (64–95 percent saturation) at 1 m and 3–4.5 mg/l (40–60 percent saturation) at 3 m

   2. Temperature

      Water column to 5 m depth at 28–30° C with lowest values during “Harmattan” and wet season and highest during dry season.

Seasonal effects are evident at station 0313 and to a lesser extent at the Dam station -lower oxygen and temperature values and smaller variations in these parameters with depth during the mixing periods (wet season and “Harmattan”). This effect is less evident from Konankuikro station data due to the sheltered nature of this site.

Dissolved oxygen values, with the exception of station 0313, are rather low and, in the absence of moderate currents at the sites, are unsuitable for cage culture of most fish especially during “Harmattan” and the wet season. This, however, may not apply to fish with accessory aerial respiratory organs such as the clariids. Oxygen content at station 0313 is excellent throughout the year but a strong current and rough waters during the wet season rule against this site. Cages should therefore be placed over deeper water (minimum of 7–10 m depth) and floated closer to the inlets of selected sheltered bays.

Temperature fluctuations in the upper 2 m of water where fish cages would be floated, are negligible and do not pose any problems. Wide temperature fluctuations, however, do occur between day and night during “Harmattan”; continuous temperature recordings were therefore taken during a 24-hour period at the two shallow stations during this season. The results are shown in Figure 6 - a drop of 14° C in air temperature was noted between 14.00 and 04.00 hours (from 32° to 18° C); corresponding changes in water temperature at 1 and 2 m depths, however, were in the order of 4°C (from 29 to 25° C) at the exposed dam station and 1° C (from 28 to 27° C) at the sheltered Konankuikro station.

In order to assess the effect of turbine function on water conditions near the dam, continuous oxygen/temperature records were taken at a station near the project marina over 7 m of water during a 24-hour period when water was being drawn through the turbines. The results are plotted in Figure 7; mixing is indicated by the convergence of oxygen and temperature values. Oxygen values dropped below 0.5 mg/l and water temperature to 26° C. Since the turbines are activated without notice, it was not possible to obtain baseline measurements. Two weeks later, however, oxygen and water temperature values at the same station, in the upper three metres of water, ranged between 4 and 5 mg/l and 27–29° C respectively. The area near the dam would therefore pose problems during turbine operation for species other than air-breathers and may even stress the latter when confined in cages.

In considering the above data it should be noted that the low oxygen values recorded in the upper 1–2 m of water in bays and the marked decreases observed during “Harmattan” and the wet season are characteristic of new man-made lakes and are due to the great amount of submerged organic matter still undergoing decomposition. A marked improvement in oxygen content will occur when the lake stabilizes.

2.2.2 Enclosures

Fish culture in enclosed bays is similar to pond culture inasmuch as nutritionally complete artificial feed is not an absolute requirement, culture of mixed species is possible and comparable yields can be obtained. An efficient and inexpensive barrier permitting water exchange is needed to retain stocked fish and prevent the entry of young of undesirable species especially carnivores. Exchange of water across the barrier is desirable for maintenance of proper water quality but makes water fertilization less effective. Supplementary feeding with available agricultural by-products such as cotton-seed cake, brewery waste, wheat bran, etc. will be required.

Despite the availability of suitable sites, the culture of fish in enclosures, in Lake Kossou, would be premature at this time. The Lake is not as yet completely filled and water level is expected to rise an additional 16–20 m. In addition, bays chosen for this purpose will have to first be cleared of submerged and semi-submerged trees to facilitate harvest and general management.

2.3 FEED

In open pond culture fish are able to forage on natural food which, although limited in quantity, may be very important in terms of a balanced diet. Fish confined in cages must depend entirely on the artificial feed supplied and the latter must therefore be nutritionally complete. It should also contain greater levels of high quality protein and more vitamins than those generally used in pond culture. Such feeds are expensive and unless feed conversion is good and fish survival is high, cage culture is not economically feasible. The feed must be presented in the form of the water-stable pellets (i.e. that do not disintegrate quickly in water). These can be either of the floating or sinking type. In the former case the feed is limited to the cage by either fine-mesh screens placed around the cage and extending a few centimetres above and below the water level or a cube is built into the cage lid such that its open lower end is submerged and pellets are delivered into the open top. Sinking pellets are either dropped free in the cage or placed on an underwater platform; demand feeders are often used with sinking pellets.

2.3.1 Availability of feed and feed components

A preliminary survey of the availability of feed components suitable for formulation of fish feed was conducted with the assistance of Dr. A. Coche, Dr. A. Fanny and Mr. V. Frank. The information given in Table 1 is by no means comprehensive and a more extensive survey over a longer period of time would probably yield additional feed components. The composition of available by-products is given in Table 2.

Intensive animal husbandry based on artificial feed is a recent development in the Ivory Coast. It is anticipated that 5 000 tons of poultry (meat and eggs), 2 500 tons of pork and 2 000 tons of beef will be produced in the country in 1975¹, with an estimated consumption of 3 950 tons of cotton-seed cake, 1 650 tons of coconut cake (copra), 1 650 tons of palm-nut cake, 6 220 tons of molasses and 3 150 tons of wheat bran. Fish culture is not a developed industry in the country; it does not contribute to the protein supply or compete for available feed and feed components.

Much of the locally available feed components is exported at present; for example about 65 percent of the expected production of cotton seed and 90 percent of the copra will be exported in 1975¹. The result is a shortage of certain by-products on the local market which brings about increases in prices and necessitates changes in the composition of animal feeds. This situation, however, is mainly due to the fledgling nature of intensive animal husbandry in the country - the lack of an appreciable local market resulting in the export of these by-products. It is expected, however, that exports will decline as the local market expands; it is estimated for example that, given the present rate of growth of the local market, export of cotton-seed cake and molasses will stop and that of copra will decline to about 50 percent of production by 1980. In addition production of presently available components will increase (Table 1) and new products such as soybean cake and silkworm grubs, etc., will become available as a result of agricultural growth and the establishment of new agriculture-based industries.

¹ Estimates obtained from the Ministry of Planning, Republic of Ivory Coast.

Feed components of animal origin are in limited supply. Artisanal fish meal is produced by a local concern (REAL) in Abidjan but total production is limited with present stocks at 5 tons. This is mainly due to the absence of an adequate local market. The company has the raw material to increase production significantly but investment in expansion and new equipment is awaiting the development of a good local or export market. The high price of the product, 120 Fr. CFA/kg for 60 percent protein-content meal, probably contributes to the limited nature of the local market. Blood meal is not commercially available as it is discarded at slaughter houses; bone meal is not available and it was not possible to determine the availability of slaughter-house wastes.

Visits were made to two major feed producers located between Abidjan and Bingerville. Cocoservice, a private poultry farm is probably the largest producer of poultry feed in the country with a current production of 900 tons/month and a total capacity of 2 300 tons/month. Feed components are stock-piled whenever available to avoid high prices and the need to alter feed composition during shortage periods. Powdered feed of 25 percent protein content is marketed at 58 Fr. CFA/kg; its composition is shown in Table 3. The manager expressed willingness to prepare feed of 30 percent protein composition for the project at a price of about 60 Fr. CFA/kg for powdered feed and to pellet this if provided with a binder at a total retail price of 63–65 Fr. CFA/kg for quantities less than 500 kg/month. The feed would have the following approximate composition:

Humidity: 7.4 percent
Fat: 6–7 percent
Total protein: 30 percent
Cellulose: 10–12 percent
Nitrogen-free extract: 35–38 percent
Calcium: 0.25–0.50 percent
Phosphorous: 0.40–0.50 percent

Ingredients would include fish meal, peanut cake, wheat bran, brewery waste, white corn, dry alfalfa meal and vitamin mix (Table 3).

The Centre Avicole de Bingerville, a government poultry station attached to the Ministry of Animal Production, produces a total of 205 tons/month of mostly poultry feed. Due to the absence of adequate stock piles of feed components and the resulting high price of these components during shortages, the feed is marketed at a loss at 58 Fr. CFA/kg (1 US $ = 250 Fr. CFA. - Feb: 1974).

2.3.2 Preparation of feed for cage-rearing tests

In order to test the suitability of feed marketed by Cocoservice for feeding fish in cages, 50 kg of 25 percent protein feed were purchased and attempts were made to bind the powder into water-stable pellets to permit feed conversion tests in cages. The objective was attained by preparing a mixture of 80 percent dry powder feed and 20 percent whole fish (wet weight), obtained from the project's esperimental fishing efforts, to which water was added at the rate of 50–60 percent of the total weight of the mixture (i.e. 1 000 g dry feed + 200 g whole fish + 500–600 cc water). The ingredients were mixed thoroughly and put through a meat grinder with 3 mm diameter grill perforations. The resulting threads were then either sun dried or placed in a Fisher oven at 60°C for about six hours.

A total of 285 Tilapia nilotica, obtained from the CTFT Fish Culture Station in Bouake, were placed in a one cubic metre cage floated near station 0313 and later moved to a more suitable location near the marina station. A total of 50 fish were lost due to transportation damage but no further losses were noted until the departure of the consultant three weeks later. Final stocking rate was 235 fish/m³ or 9.4 kg per cubic metre (average weight of individual fish = 40 g). Fish were fed the prepared pellets at the rate of 6 percent the body weight per day (total of about 500 g/day). It should be noted here that the amount of fish added as a binder did not affect the composition of the feed significantly due to its high water content (80 percent). The pellets were water-stable and taken readily by the tilapia: feed was initially offered in a bucket suspended from the cage lid and later placed on a fixed submerged feeding trough. The tests were in progress at the time of the consultant's departure.

Similar tests will be made with Clarias lazera juveniles (10 g weight) when these become available from CTFT in May to June, 1974. Two additional cages were built for this purpose.

2.4 CAGE MATERIALS AND CAGE CONSTRUCTION

Three cages were constructed during the consultant's mission to estimate construction costs and to undertake cage rearing tests. Materials used were either available at project headquarters or purchased from Bouake and Abidjan. A list of materials available for construction of cages, together with local prices are given in Table 4.

A 1.3 m³ galvanized wire/mahogany frame cage was constructed for Tilapia nilotica rearing tests; cage specifications are given in Figure 8. Floatation was effected by means of two 1 m lengths of PVC pipe sealed with wooden plugs and four 24 cm diameter styrofoam spheres; this did not prove to be suitable as it was not possible to fix the spheres firmly to the cage frame at a given level underwater and the cage floated with the lid only about 5 cm above water level. In addition, the galvanized wire/mahogany frame lid proved to be too heavy to be handled by one man during feeding of the fish. Feeding from a bucket suspended from the cage lid was not satisfactory as a small number of fish tended to monopolize the bucket and swimming movements dispersed the pellets with considerable loss of feed. The cage was therefore modified to float 25 cm above water level by means of styrofoam-block floats (figures 9a and b) and a shallow feeding trough fixed 30 cm below water level. The cage lid was rebuilt with 5 mm mahogany plywood and a small feeding hatch built into it to obviate the need to lift the entire lid during feeding.

The galvanized wire/mahogany frame cage proved to be expensive (12 000 Fr. CFA) and too heavy to be managed by one operator. A double-net cage was therefore constructed using the remaining float material; cage specifications are given in Figure 10. Light pinewood with several coats of marine paint was used for the frame; the cage weighed 10 kg and was floated with two 1 metre lengths of PVC pipe fixed 25 cm below the frame. A V-shaped feeding trough with a flat bottom fixed to the frame was suspended 30 cm below water level. Total cost of materials for the cage was 1 500 Fr. CFA/m³. The net was kept open by placing a 6 mm iron rod, with the same dimensions as the top frame, on the bottom. Cage cost can be further reduced by using empty plastic jugs or bottles for floatation and woven palm leaves instead of plywood for top cover.

3. RECOMMENDATIONS

There are major gaps in the information available on which to base a judgement. The present preliminary study established the availability of suitable sites and feed components but the economic feasibility of cage culture will depend on feed conversion values and maximum possible stocking rates of fish. This information can only be derived from properly designed and conducted field tests.

A major problem is the availability of juvenile fish for stocking in cages. It is clear from the discussion in section 2.1 that Clarias spp. are the most likely candidates in terms of market value, hardiness, acceptance of artificial feed, growth rate and feasibility of artificial breeding. Juveniles, however, are only available in small numbers from the CTFT Fish Culture Station during summer. An adequate natural supply of fingerlings is yet to be demonstrated. Since these fish are capable of attaining market size of 300–400 g in 3–4 months, a dependable supply of fingerlings throughout the year would permit marketing of up to three crops per year. This will be possible only if controlled breeding and rearing of these species is practised on an adequate scale in the country.

It is therefore recommended that an initial one to two year programme be initiated to:

1. investigate the availability of natural seed resources;

2. establish a breeding-rearing programme for Clarias sp. and ensure a reliable supply of fingerlings;

3. determine maximum stocking densities in cages;

4. determine conversion rates of feed at various levels of protein content and with different components;

5. prepare suitable pelleted feed with the assistance of a nutrition consultant, and

6. train local personnel in cage culture methods.

Minimum staff requirements will consist of one fish culture expert, one associate expert and six months of consultant time.

Preliminary work along these lines but on a small scale can be initiated under the present project, IVC/526, provided an associate expert and three to six man months of consultant time are provided. Additional cages will also be needed for this purpose.

¹ 1 US$ = 250 Fr. CFA (February, 1974)

¹ From “Valeurs bromatologiques de 150 aliments de l'ouest Afrique; B. Mongodin et R. Rivière. Institut d'élevage et de médecine Véterinaire des Pays Tropicaux, 1965.

² Nitrogen-free extract.

¹ Feed composition: white corn, peanut cake, wheat bran, fish meal, alfalfa meal (dry), vitamin mix; exact composition not available.

² Composition of vitamin mix: (per 100 kg of feed)

Fig. 1 GEOGRAPHIC LOCATION OF LAKE KOSSOU

Fig. 2 DETAIL OF LAKE KOSSOU. Scale: 1/300 000.

Fig. 3 DISSOLVED OXYGEN AND WATER TEMPERATURE: DAM STATION 1973 – 1974; SAMPLES TAKEN AT 11 00 HOURS.

Fig. 4. DISSOLVED OXYGEN AND WATER TEMPERATURES: KONANKUIKRO STATION 1973 – 1974; ALL SAMPLES TAKEN AT 11 00 HOURS

Fig. 5 DISSOLVED OXYGEN AND WATER TEMPERATURE: STATION 0313, 1973; POINTS REPRESENT AVERAGE WEEKLY VALUES; ALL SAMPLES TAKEN AT 11 00 HOURS

Fig. 6 FLUCTUATIONS IN AIR/WATER TEMPERATURES OVER A 24 HOUR PERIOD DURING “HARMATTAN”

Fig. 7. EFFECT OF TURBINE OPERATION ON DISSOLVED OXYGEN AND WATER TEMPERATURE: MARINA STATION

Fig. 8. GALVANIZED WIRE/MAHOGANY FRAME CAGE: DRAWING NOT TO SCALE

Fig. 9a MODIFIED GALVANIZED WIRE/MAHOGANY FRAME CAGE: DRAWING NOT TO SCALE.

Fig. 9b DETAIL OF FEEDING TROUGH AND STYROFOAM BOX: DRAWING NOT TO SCALE.

Fig. 10 DOUBLE NET/PINE WOOD FRAME CAGE: DRAWING NOT TO SCALE.