Two major reservoirs have been installed in the southeast of the country: Mwadingusha on the Lufira River and Nzilo on the Lualaba River. Their positions are presented in Figure 7.1. Other names for the Mwadingusha Reservoir are Tshangalele or Lufira Reservoir.
The Lufira River originates near Lubumbashi, close to the border with Zambia. The river has a length of 630 km and is one of the major tributaries of the Lualaba River. The dam at Mwadingusha was closed in 1947, its main purpose was power generation (Ruwet, 1961; Damas, 1961; Vanden Bossche and Bernacsek, 1990a).
No data available.
In 1925 a power station was constructed in the Upper Katanga. The falls in the Lufira River at Mwadingusha were 113 m high and very well suited for the purpose of power generation. A dike, 8 m in height, was sufficient to create a reservoir of 33 km3. The dike was raised three times: in 1934, 1938 and 1947, to 10, 12 and 13 m respectively (Damas, 1961).
The altitude of the lake at Full Supply Level is 1 105.75 m or at watermark 210; its maximum depth is 14 m with a mean of 2.6 m. The maximum length and width are 25 and 24 km respectively. Its maximum volume is 1 063 km3, the annual variation in water level is 1 to 3 m. Available physico-chemical data are summarized in Table 7.1 (Ruwet, 1961; Vanden Bossche and Bernacsek, 1990a). Available figures on the surface area are not consistent. Ruwet (1961) mentioned a total surface of 410 km2, Damas (1961) produced a figure of 450 km2 and Vanden Bossche and Bernacsek (1990a) stated it to be 446 km2.
Water passages reportedly got blocked by mats of floating plants, which then had to be removed by hand or even by means of explosives. Due to the variations in water level plants settled easily in the exposed parts. When the water level rises plants start rotting and the content of O2 diminishes and production of H2S starts, which then affects the pH. The organic load of the water was observed to be high.
The gillnet was the main fishing gear in 1959–1960 according to Ruwet (1961), who led a study mission organized by the Foundation of the University of Liège (Belgium) for Scientific Research in Congo and in Rwanda-Burundi (FULREAC). The net measured 100 m in length and 2 m in height. There was a minimum mesh size of 5 cm knot-to-knot. Ruwet quoted Magis (in Goorts et al., 1961) who observed that the 5 cm meshed net was the most appropriate; however, he stated that 30 to 40% of the nets used in the commercial fishery had mesh smaller than the legal minimum size. Professional fishermen drive the fish actively into their nets by beating the water surface.
The second type of gear is the ‘mutobi’, a kind of scoop net. The net has the shape of a bag, with mesh of 2 to 4 cm, and is fixed on two sticks. It is used in deeper water. Other gear are the bottom longline and the trap to catch catfish (Clarias gariepinus); the longline is usually set close to floating mats of aquatic plants, while the traps are set with their openings in such a way that catfish migrating upstream are trapped (Ruwet, 1961). Ruwet (1961) quoted Goorts et al. (1961) who stated that the authorities had encouraged fishermen from the Lualaba and Luapula Rivers to start fishing in the reservoir. They estimated the number of fishermen to be between 1 500 and 2 000, producing some 4 000 t of fish annually. Vanden Bossche and Bernacsek (1990a) quoted Konare (1984) who estimated the number of fishermen at 1 410 and the number of boats at 1 583.
Commercial fish species are Tilapia (Oreochromis) macrochir, T. melanopleura, Serranochromis kafuensis and Clarias gariepinus (Ruwet, 1961; Konare, 1984).
7.2.5.1 Annual yield
Annual production figures are available since 1953, although the time series is not complete (Table 7.2). The highest production was recorded in 1955 when the annual yield totalled 7 991 t. In that particular year, due to a very bad rainy season, 95% of the remaining part of the lake was covered with aquatic plants (Damas, 1961). In 1956/1957 the lake filled up to its maximum level and then much-reduced production figures were obtained: 674 and 1 359 t for 1956 and 1957 respectively.
The most recent production figure available is of the order of 3 106 t in 1983, reported by Konare (1984). Out of 3 106 t about 1 500 t would have been consumed by the fisherfolk. In case 1 410 fishermen produce 1 500 t per year, each fisherman would, together with his family, consume over 1 t annually. Even if a family consists of 10 persons, the consumption per caput per year would be 100 kg. This amount seems rather high.
Fish yield of the Mwadingusha Reservoir was estimated at 127 kg/ha by Welcomme (1972). Marshall (1984a) attempted to relate the yield to the morphometry of the reservoir; he obtained a model for a number of large reservoirs but the Mwadingusha data did not fit the model. This phenomenon was probably due to the fact that Mwadingusha Reservoir is very shallow (approximately 2.5 m) which suggests that catchment and climatic influences will have a much greater effect on smaller lakes. Konare (1984) estimated the potential annual yield at 4 460 t, which was based on a figure of 100 kg/ha. Existing models to predict potential yields were revised by Crul (1992). Application of his models results in an annual potential yield of 2 278 t.
7.2.5.2 Catch rates
No data available.
7.2.5.3 Fishing effort
The only figure available is 1 410 for the number of fishermen in 1983 (Konare, 1984).
7.2.5.4 Mesh selectivity
Ruwet (1961) collected data on species composition for different meshed mutobi nets. Principal species in the catches is Tilapia macrochir. Data are presented in Table 7.3.
7.2.5.5 Species composition
Ruwet (1961) reported interesting species composition data for various biotopes. His data are presented in Table 7.4. It is quite clear that Tilapia macrochir and T. melanopleura are the dominating species, accounting for 83 to 98% of the catches. T. macrochir is the most abundant species in open water catches, while T. melanopleura is the dominating species in the catches on the grassy plains. Serranochromis and Clarias species are relatively more abundant around the floating aquatic plants and along the grassy edges of channels than in the other biotopes.
Ruwet also collected species composition data for gillnets at two different sites during various months of the year (Table 7.5).
The species composition of the catches from the entire lake over the period 1957–1960 show an interesting picture: the contribution of T. macrochir remains more or less stable, however, that of T. melanopleura increases explosively, while the contribution of Clarias drops tremendously.
Table 7.6 represents the data collected by Ruwet.
7.2.5.6 Small clupeids
No data available. The lake is rather shallow and therefore it is doubted whether Mwadingusha Lake is suitable for small clupeids.
The FULREAC mission found that the Mwadingusha Lake supports an important fishery of tilapias; over 95% of the catches consist of T. macrochir and T. melanopleura. Therefore it was concluded that rational exploitation of the lake could be ascertained by regulation of the fishery, as far as fishing gears and fishing grounds are concerned, and of the water level fluctuations.
The rational exploitation of cichlids should protect juvenile and immature fish. It was observed that gillnets with mesh of 5 cm (knot-to-knot) respect the juvenile tilapias satisfactorily and that the use of smaller mesh should be prohibited (Goorts et al., 1961).
The mutobi scoop net used on the Lufira River should not be allowed on the lake. Along the river the net is used to catch Barbus paludinosus; however, if this gear is operated on the lake, it will definitely exploit the juvenile tilapias. Ruwet (1961) expects the gear to be used when the water level in the lake is low.
Theoretically, spawning areas could be protected by a permanent closure of the fishery or by a temporary closure during the peak spawning activities. However, for socio-economic reasons a permanent closure could not be justified, and although fishing took place in important spawning areas of T. melanopleura, this species managed to maintain itself; therefore temporary closures are not justified either (Ruwet, 1961).
Water level fluctuations should be controlled to prevent destruction of the spawning grounds by the following measures:
A temporary closure of the fishery should be imposed when the water drops to a level below that considered critical for the fish stocks. In such a case the following measures should be taken: closure of the gillnet fishery for Tilapia during the spawning season when the water no longer reaches the watermark 208. Bottom longlines may then still be used on the lake and the mutobi scoop net upriver. When the water level drops under the watermark 207 the use of all gear except the longline should be stopped immediately. In case such closures occur, financial compensation should be given to the fishermen.
Damas (1961) (as quoted in Ruwet, 1961) reported that attempts were made to introduce T. zillii into the reservoir and that the lake was stocked with T. melanopleura regularly, mainly to contribute to the control of water plant growth.
