Aquaculture feeding strategies were assessed on the basis of 22 field visits to the Project's four fry production centres at Kibos, Borabu, Rongo and Chwele, and to 18 private fish farms (for list see Annex 2) within the Lake Basin Region. During each field visit the farm manager/farmer was interviewed and a questionnaire completed (see Annex 3).
At present a semi-intensive pond feeding strategy is employed for the production of Tilapia (Oreochromis niloticus) within the Project's four fry production centres; fish stocked at low densities within diammonium phosphate (DAP) fertilized earthen ponds and fed a commercial dairy feed mash or an in-house produced dry fish pellet. Fry production is achieved through the natural spawning of brood fish (ca. 300–500g body weight) in broodstock ponds and the periodic harvesting of market size fingerlings (ca. 20–50g body weight) for distribution to the farmers in the Lake Basin Region. It is estimated that approximately three tonnes of fingerlings were produced from the four production centres during 1990; equivalent to about 100,000 fingerlings of 30g body weight. However, despite these generalizations, all centres display different fish, pond, fertilizer and feed management practices, including;
Fish production reported within grow-out ponds in the centres during 1990 ranged from 1 to 7.8 tonnes/ha/year (mean 2.9 tonnes/ha/yr) over a six to eight month culture cycle; the reported food conversion ratios ranging from 2–2.4 for dairy feed mash and 1.6 in one instance for the in-house produced Tilapia pellet. At present DAP granules and dairy feed mash is administered in dry form by hand application onto the pond surface within all centres.
Although the highest recorded grow-out fish production was reported with male fingerlings using a diary feed supplementary diet feeding strategy (equivalent to a net fish production of 7.8 tonnes/ha/year), the preliminary results obtained in one centre with the use of a dry in-house produced fish pellet have been encouraging (ie. on the basis of the low food conversion ratio obtained). Manufactured since August 1990, the pellet is composed of a mixture of rice bran (70%), dried “Omena” fish (Rastrineobola argentea; 24%) and sorghum (6%). The pellets are prepared by first hand grinding individual ingredients (so as to pass through a 0.5–1mm sieve), mixing (by hand), and wet extrusion through a hand operated kitchen meat mincer with a 2–3mm die plate. The resultant moist pellets (containing ca. 80% moisture) are then sun dried for one to two days (depending on the weather) and stored in plastic lined sacks within a general purpose metal store house at ambient temperature until fed. For fingerlings, the sun-dried pellets are ground by hand to a smaller particle size; the resulting crumble containing a high proportion of ‘fines’. Apart from the loss of ‘fines’ during hand feeding and the difficulty of sun drying pellets during the rainy season, mould infestations are common after the prolonged storage of feeds on the farm. The results of the in-house feeding trials with this diet have not been published or analyzed from an economic viewpoint (at the time this consultancy was undertaken).
The feeding regimes employed by private fish farmers within the Lake Basin Region are related to the aim of the feeding activity (ie. fish production for home consumption and/or cash sale), labour inputs (ie. part-time or full-time farming activity), the financial status of the farmer, and the local availability of fertilizers and feeds. At present the Tilapia farmer in the Lake Basin Region employs a part-time semi-intensive pond feeding strategy; fish growth being dependent upon a combination of natural pond food organisms (resulting from the fertilization of ponds with animal manures or more rarely chemical fertilizers) and the use of external supplementary feed inputs. For example, the following fish, fertilizer and feed management practices were observed within the 18 private farms visited;
The calculated mean net fish production within the farms visited ranged from 0.9 to 11.1 tonnes/ha/year (mean 4.1 tonnes/ha/yr, n = 15). The highest individual farm production was recorded for the Joachim Miyoyo Farm (Siaya District); a 100m2 pond stocked with 210 O. niloticus fingerlings (mixed sexes) producing a net fish yield of 72kg over a 7.75 month production period. The feeding strategy employed on this farm included the weekly application of fresh cow manure and rumen contents as fertilizers within a compost crib, the daily feeding of chopped green leaves (paw paw, sweet potato, cabbage) and purchased mill sweepings, supplemented on a weekly basis with collected live termites. It is also of interest to note that between 1986 and 1990 the Kasigunga Women's Group Farm (Siaya District) achieved a net fish production ranging between 2.1 to 6.6 tonnes/ha/year (mean 4.0 tonnes/ha/yr, n = 7 production cycles within 100–200m2 ponds, 1–2 fish/m2) at no raw material input cost to the group; ponds fertilized with cow and pig manure on a daily basis and fish fed chopped green leaves and spoiled fruit.
The highest recorded production for a farm using an in-house produced feed mash was 6.6 tonnes/ha/year for the Bomwancha Women's Group Farm (Kisii District), where fish were fed with a maize flour: omena mash (1:1, w/w) in addition to the use of cow dung/poultry manure pond fertilization and the feeding of chopped green leaves. Although no private farms currently use pelleted supplementary diets, two feeding trials have recently been conducted using the In-house produced fish pellets from the Kibos Fry Production Centre; net fish production ranging from 2.6 tonnes/ha/year (Weaver Bird Women's Group Farm, Kisumu District; initial density male-1/m2, pellet apparent food conversion ratio 6.2 and fish survival 67.5% after 204 day culture period) to 4.9 tonnes/ha/year (Sakagwa Primary School Fish Pond, Kisii District; initial density male-2/m2, pellet apparent food conversion ratio 1.77 and fish survival 65% after 182 days). By contrast, a net fish production of 3.3 tonnes/ha/year was recorded the year previously at the Weaver Bird Women's Group Farm using 100% rice bran mash as the sole supplementary feed input; initial density male-1/m2, rice bran apparent food conversion ratio 4.65 and fish survival 75.6% after 162 days. On the basis of reported feed costs alone (ie. Tilapia pellet Ksh 11/kg and rice bran Ksh 2.1/kg), the cost of producing a 1kg increase in fish biomass would be Ksh 9.76 using rice bran and Ksh 19.47–68.31 using the in-house produced Tilapia pellet. The average market cost of Tilapia reported by the 18 farms visited was Ksh 38/kg. Clearly, on the basis of these feeding trials it would be much more economical for the farmer to use rice bran.
Although the average fish production within the 18 private farms visited was generally high, it is difficult to extrapolate these findings to the other farmers in the Lake Basin Region. At present there are over 4000 fish farms in the Region, representing a total area of about 80 hectares of pond water surface (ie. 6000 ponds of 130m2 average size). For example, from a sample of 205 and 254 private ponds harvested in 1988 and 1989, the average fish production was 1.23 tonnes/ha/year and 1.36 tonnes/ha/year, respectively (Table 1). By contrast, 1990 production statistics indicate that total fish production within the Lake Basin Region (not including the Vihiga and Migori Districts) was only 25.1 tonnes from a total of 6127 ponds (Table 1). Assuming an average pond size of 130m2 this production would be equivalent to an annual fish production of 0.31 tonnes/ha/year. Clearly, an in-depth survey of the status of the private fish pond operators within the LBDA Region must be undertaken before an accurate assessment of fish production within the sector can be made.
On the basis of discussions held with the Project's District Field Coordinators it is estimated that on average only 50% of the farmers in the Region use regular (ie. daily) fertilizer and feed inputs.
