Abstract:

Bottom trawls, gill nets and other sampling devices were used to study the distribution and abundance of Nile perch (Lates niloticus (L.)) and Nile tilapia (Oreochromis niloticus (L.)) in the Uganda portion of Lake Victoria from 1999 to 2001. A total of 17 fish taxa (belonging to 14 genera) were encountered in the Uganda portion of Lake Victoria in this study. This reflects a decline in fish species diversity over the last 30 years when compared with results from similar studies in 1969/71. Tilapia rendalli was encountered for the first time since it was introduced in 1950s. The fish catch was dominated by Lates niloticus (86.5% by weight), followed by Oreochromis niloticus (9.8%). L. niloticus was found in all areas surveyed while O. niloticus was restricted to waters <20 m deep. The highest catches (13.7±1.1 t km-2) of L niloticus in bottom trawls were obtained in Zone I - between Tanzania/Uganda border and Bukakata - and Zone III - between Kiyindi and Uganda/Kenya border - had the lowest concentrations of fish (8.3±0.6 t km'2). This trend was also reported during the period 1993/98. Length frequency data for L. niloticus suggest that all size groups were found in all areas surveyed during bottom trawling. Juveniles of L. niloticus of less than 6 cm TL were found in all areas surveyed. This may suggest that breeding of L. niloticus takes place throughout the lake. O. niloticus during the bottom trawl surveys in 1999/2000 were restricted to water <20 m deep. The bulk of (93% by weight and 40% by numbers) O. niloticus was mature (>22 cm TL). This was partly because trawling covered only waters 4-60 m deep leaving the waters <4 m deep, a potential area for the small, young fish of the species. In the monofilament gill net surveys, L. niloticus dominated the catches in the open deep water while O. niloticus dominated in the shallow inshore areas. In the open deep-water areas, the nets set at the bottom did not catch fish; this was attributed to low oxygen levels. More male than female L. niloticus was caught during both the bottom trawl and gill net during the survey period. The fish stock densities (CPUA) in the bottom trawls have continued to decline from 36.2 t km'2 for all fish taxa during 1969/1971 to 8.8 t km'2 for L. niloticus in the present study in the 4-30 m deep waters. Fish stock densities of O. niloticus were higher in 4-10 m deep water than in waters 10-20 m deep. The biomass of 121 000 t of L niloticus and 12 000 t of O. niloticus were estimated in the 4-40 m water depth zone. These estimates increased from 117 053 t in 1999 to 121 930 t in 2000 for L. niloticus and from 7316 t in 1999 to 17 070 t for O. niloticus. Such increases may partly have been due to the El-Nino rains of 1998/1999, the reduction in fishing pressure resulting from the ban of export of fish and fish products from the Uganda portion of Lake Victoria to the European Union during 1999. Juvenile fish measuring <50 cm TL (size at first maturity for male L. niloticus') constituted 72.5% of the catch by weight while 27.5% were mature fish. Accordingly, about 32 800 tonnes of the 119 491 tonnes estimated for L. niloticus in the 4-40 m water depth zones during the survey period 1999-2000 were mature fish. The bulk of the immature L. niloticus were caught in waters <20 m deep (63.36 % by numbers) and in Zone III where they (immature fish) contributed 99.02% (by numbers) during 1999-2000. The abundance of the juvenile fish of various sizes promises good future recruitment to the population while the relatively small numbers of large mature fish raises concern on the future of the fishery. In this study the asymptotic lengths (L-) for L. niloticus and O. niloticus were estimated as 256 cm and 75 cm TL, respectively, while the respective growth rates for the two species were 0.29 yr'1 and 0.40 yr'1. The size at first maturity for male and female L. niloticus were 50 cm and 64 cm TL, respectively, while the size at first maturity for O. niloticus were 21.5 cm and 22.5 cm TL for male and female, respectively. The high mortality rates for both L niloticus (Z=1.91) and O. niloticus (Z=1.65) increased from 1999 to 2000 and varied between zones. During this study, recruitment in the fisheries of L. niloticus and O. niloticus occurred throughout the year but with two peak periods. Peak recruitment for both L. niloticus and O. niloticus occurred during the rainy season (March-May and August-November). The diet of L. niloticus varied with its size. Caridina nilotica formed the most important food item in juvenile fish up to 40 cm TL. Fish and fish remains were observed from fish specimens as young as 6.2 cm TL. The occurrence of fish in the diet of L. niloticus increased with increasing size of the fish specimens, becoming dominant (67.4% occurrence) in fish of 40 cm TL and contributing about 100% of the diet in fish of 90 cm TL. During this study the water temperature varied in the different habitats and seasons but generally ranged from 28.1°C at the surface to 24°C at the bottom. Dissolved oxygen, a requirement for all living things, varied with water depth, season and habitats. The varied distribution of the fish during this study was considered to have been influenced by the observed distribution of oxygen. The high catches of fish in the 20-30 m depth area in Zone I is likely to be due to the presence of sufficient oxygen in the area. The low abundance of fish in waters >25 m deep are partly due to insufficient or no oxygen.