Previous Page Table of Contents Next Page



Mambona Wa Bazolana
Research Centre of Natural Science
Uvira Station


Michael Fryd
FAO-IFIP Project


The two endemic clupeids of Lake Tanganyika, Stolothrissa tanganicae and Limnothrissa miodon, dominate the pelagic catches in the northern part of the lake in most years. Length-frequency data were obtained from samples collected each month from commercial fisheries in the Zaïre and Burundi sectors of the lake. The ELEFAN programme was used for the estimation of growth parameters (L∞, K and Φ'), mortality rates (Z, M and F) as well as the exploitation rate (E) and the recruitment pattern. The results from the two sides of the lake were compared with each other and with data obtained from other studies.


Les deux clupéidés endémiques du lac Tanganyika, Stolothrissa tanganicae et Limnothrissa miodon, dominent la plupart du tempes les captures de poissons pélagiques effectuées dans la partie nord du lac. Des données concernant les fréquences de longueur ont été obtenues à partir d'échantillons recueillis chaque mois dans les pêcheries commerciales des secteurs zaïrois et burundais du lac. Le programme ELEFAN a été utilisé pour estimer les paramètres de croissance (l∞, k et Φ), les taux de mortalité (Z, M et F), ainsi que le taux d'exploitation (E) et le schéma de recrutement. Les résultats des deux côtés du lac ont été comparés entre et aux ainsi qu'avec les données tirées d'autres études.


The population of small pelagic fish in Lake Tanganyika consists of two endemic clupeids, Limnothrissa miodon and Stolothrissa tanganicae, which are called “Ndakala” in Zaïre and Burundi. They are an important component of the artisanal and industrial fisheries in the countries sharing the lake. The two species, of which Stolothrissa is usually the most abundant, have made up about 80% of the total catch in the northern part of the lake for some years. They are fast-growing fish with short life-spans and high rates of reproduction which are features typically associated with r-selected species (Marshall, 1987). Their populations fluctuate in abundance over time, even in the absence of fishing, since they respond to short-term changes in the environment that affect recruitment and feeding (Coulter, 1988). There has recently been a decrease in clupeid catches in the northern part of the lake (Enoki et al., 1987; Mambona, 1987; Shikahara and Mambona, 1988).

This decrease in catch, and especially in catch per unit effort (CPUE) indicates that there is a need for proper management of the clupeid fishery. This requires a knowledge of their population dynamics but relatively few investigations have been made on the fish in Lake Tanganyika. This paper presents some population parameters of the clupeid stock in the northern waters of the lake in three successive years. The values obtained on the two sides of the lake are compared with each other and with those obtained elsewhere.


During 1988–1990 samples of both clupeids were taken from commercial fisheries in Zaïre by the Fisheries Research Centre at Uvira and in Burundi by the Department of Fisheries. The samples from Zaïre were taken from the artisanal lift net fisheries operating in the north-western part of the lake just offshore from Uvira. Fish were collected twice a week from the landing beaches and every month about 1300 Stolothrissa and 600 Limnothrissa were measured and the results were combined into 5 mm length classes.

The Burundi samples came from the semi-industrial purse-seine fishery which covers a large area of northern Lake Tanganyika. Fish were collected at the central fish market on three occasions before and three after the full moon. About 80000 Stolothrissa and 3500 Limnothrissa were measured every month. The fish were measured to within 1 mm but, for practical reasons, they were combined into 2 mm size classes when the data were being processed.

The length-frequency distributions of the samples from Zaïre and Burundi were entered into a computer and processed by the ELEFAN (Electronic Length Frequency Analysis) programme developed by Pauly and David (1981). The ELEFAN I procedure was used for the estimation of the growth parameters L∞ and K which are used in fitting the von Bertalanffy growth equation

Lt = L∞ [1 - exp(-K(t - to))]

where L(t) is the length of the fish at time t; to is the “age” of the fish when L is equal to zero; L∞ is the asymptotic length and K is the rate at which the Lt)approaches L∞. The resultant growth curves were compared by Munro's index of growth performance (Φ') which is calculated as follows

Φ' = In K + 2 In L∞

where K is an annual rate and L∞ is expressed in centimetres (Sparre et al., 1989)1. Once these parameters had been estimated the ELEFAN II procedure was used to determine mean length (L), total mortality (Z), natural mortality (M) and the exploitation rate (E).

1 Editor's note. Munro and Pauly (1983) and Pauly and Munro (1984) used log10 when they first proposed this index. Most of the subsequent literature uses log10 rather than In as proposed by Sparre et al (1989). Values calculated with In are 2302585 times higher than those calculated with log10. Another source of confusion is that the asymptotic length must be expressed in centimetres, although most people use millimetres when working with vary small fish.


The maximum size of Stolothrissa in the samples was 1225 mm for fish from Uvira compared to 99.0 mm for those from Burundi. The values of K varied slightly and were close to those obtained by other authors (Table 1). There was rather more variation in the L∞ values; in 1988 the value for the Uvira fish was higher than that of the Burundi ones but in 1989 and 1990 they were very similar. The values of Z obtained from the catch curve analyses (ELEFAN II) varied considerably but were within the same order of magnitude as those published elsewhere (Table 2). The growth curves show considerable variation with the smallest fish (Burundi 1988) being some 20% smaller than the largest ones (Uvira 1989) at the age of one year (Fig. 1). The patterns of recruitment suggest that recruitment is continuous with several peaks during the year (Fig. 2).

The maximum size of Limnothrissa in the samples was 990 mm for fish from Uvira compared to 1450 mm for those from Burundi. There was little variation in the K values which were similar to those recorded by other authors (Table 1). They were higher than those recorded in Lake Kivu fish but much lower than those from Lakes Kariba and Cahora Bassa (Table 1). The estimates of L∞ were consistently about 15% larger in the Burundi fish. Other estimates from Lake Tanganyika varied considerably but the L∞ in Lake Kariba was much lower (Table 1). The K values were relatively similar in samples from Uvira and Burundi and there was little difference between them and those estimates made earlier in Lake Tanganyika or Lake Kivu. However, the estimates of K were much higher in the man-made lakes (Kariba and Cahora Bassa). The rates of total mortality (Z) were much higher in the Uvira samples than they were in the ones from Burundi, presumably because the rates of fishing mortality (F) were also much higher (Table 2). The exploitation rate (E) was between two and three times higher at Uvira.

The growth curves of Limnothrissa were also very variable and the smallest fish (Uvira 1988) were only 70% of the length of the largest (Burundi 1989) at the age of one year (Fig. 3). The pattern of recruitment indicated that there were sometimes two major peaks of recruitment during the year (Fig. 4).


The two clupeid species in Lake Tanganyika are short-lived with life-spans of one year (Stolothrissa) and two years (Limnothrissa) and high mortality rates. These characteristics are common in clupeid species. There were some differences between the fish from Uvira and Burundi and they may be a result of the fact that different length-classes were used on each side of the lake.

The lack of variation in their growth parameters of Stolothrissa suggests that they live in a relatively stable environment. The larger variations in Limnothrissa indicate that it might live in a more variable environment. This is to be expected because it lives closer to the shore where it is more likely to be influenced by run-off from the rivers (Lindqvist & Mikkola, 1989). The mortality and exploitation rates are high on both sides but generally rather higher in the Uvira samples. Both species are heavily-exploited although Stolothrissa is usually the most abundant of the two species in the catches (Coulter, 1981; Roest, 1988). Limnothrissa is more common in samples taken at Uvira since it lives closer inshore (Coulter, 1988) and the Uvira fishery operates nearer to the shore than does the Burundi fishery.

The recruitment pattern provided by the ELEFAN programme is influenced by the growth parameters. This makes it difficult to obtain a precise estimate and the data for Stolothrissa and Limnothrissa obtained in this study are difficult to interpret.


We would like to thank the Département des Eaux, Pêches et Pisciculture, Bujumbura for providing the data from the Burundian side of the lake. We would also like to thank the staff of the CSRN and the Japanese Cooperation for help in collecting data from the Zaïrean side. Piero Mannini of the IFIP project gave constructive help and criticism of the manuscript.


Coulter, G.W., 1988. Production dynamics in Lake Tanganyika. CIFA Occas. Pap. 15: 18–24.

Enoki, A., W.B. Mambona and M.S. Nukirania, 1987. General survey of fisheries in the north-western part of Lake Tanganyika. Ecol. Limnol. Lake Tanganyika, 4:98–101.

Lindqvist, O.V. and H. Mikkola, 1989. Lake Tanganyika: review of limnology, stock assessment, biology of fishes and fisheries. GCP/RAF/229/FIN, FAO Rome, 51pp.

Mambona, W.B., 1987. Statistical analysis of fisheries production in the north-western area of Lake Tanganyika. Ecol. Limnol. Lake Tanganyika, 4: 102–4.

Marshall, B.E., 1987. Growth and mortality of the introduced Lake Tanganyika clupeid, Limnothrissa miodon, in Lake Kariba. J. Fish Biol., 31: 603–15.

Mannini, P., 1990. Parametres de la population de Limnothrissa miodon du lac Kivu (1980–1989). Document de travail PNUD/FAO RWA/87/012/DOC/TR/32, 41pp.

Mannini, P., 1991. Aspects de la dynamique et de l'amenagement du stock de Limnothrissa miodon du lac Kivu. Document de travail PNUD/FAO RWA/87/012/DOC/TR/43, 25pp.

Munro, J.L. and D. Pauly, 1883. A simple method for comparing the growth of fishes and invertebrates. Fishbyte, 1(1): 5–6.

Munyandorero, J., 1989. Estimation des parametres demographiques des Clupeidae exploites dans les eaux burundaises du lac Tanganyika. Univ. Bretagne Occidentale, Fac. Sci. Tech. du Brest, 30pp.

Pauly, D. and N. David, 1981. ELEFAN I, a basic program for the objective extraction of growth parameters from length-frequency data. Meeresforschung, 28: 205–11.

Pauly, D. and J.L. Munro, 1984. Once more on the comparison of growth in fish and invertebrates. Fishbyte, 2(1): 21.

Pearce, M.J., 1988. Some effects of Lates spp. on pelagic and demersal fish in Zambian waters of Lake Tanganyika. CIFA Occas. Pap. 15: 69–87.

Roest, F.C., 1978. Stolothrissa tanganicae: population dynamics, biomass evaluation and life history in the Burundi waters of Lake Tanganyika. CIFA Tech. Pap. 5: 42–62.

Shikahara, K. and W. Mambona, 1988. Collections of fisheries statistics for population study on sardines in Lake Tanganyika. Ecol. Limnol. Lake Tanganyika, 6: 25–7.

Sparre, P., E. Ursin and S.C. Venema, 1989. Introduction to tropical fish stock assessment, Part 1: Manual. FAO Fish. Tech. Pap. 306/1, 337 pp.

Table 1. Growth parameters of Stolothrissa tanganicae and Limnothrissa miodon in various African lakes and reservoirs.

SpeciesLocalityL∞ (mm)K (yr-1)Φ'Source
Stolothrissa:Uvira 1988105.22.335.551
Burundi 198887.82.346.681
Uvira 1989107.52.585.701
Burundi 1989104.02.285.511
Uvira 199098.02.655.541
Burundi 199092.02.295.271
Tanganyika     90.0 FL2.525.322
Tanganyika     94.0 FL2.525.413
Limnothrissa:Uvira 1988118.01.105.031
Burundi 1988141.01.045.331
Uvira 1989126.01.225.271
Burundi 1989148.21.195.571
Uvira 1990126.01.005.071
Burundi 1990141.01.295.551
Kivu     145.0 FL1.206.557
Cahora Bassa70.05.405.588

Sources: (1) this study;
(2) Chapman and van Well (1978);
(3) Roest (1978);
(4) Pearce (1988);
(5) Munyandorero (1989);
(6) Mannini (1990);
(7) Mannini (1991);
(8) Marshall (1987).

Table 2. Rates of mortality
(Z = total mortality; M = natural mortality and F = fishing mortality) and exploitation of Stolothrissa tanganicae and Limnothrissa miodon in various African lakes and reservoirs.

Solothrissa:Uvira 19887.603.943.650.481
Burundi 19883.452.191.260.371
Uvira 198911.51  4.197.320.621
Burundi 198910.04  3.916.130.611
Uvira 19906.454.382.070.321
Burundi 19907.042.134.910.701
Tanganyika5.20   2
Tanganyika5.50   3
Limnothrissa:Uvira 19886.392.344.230.641
Burundi 19882.742.150.590.221
Uvira 19897.142.464.680.661
Burundi 19893.242.310.930.291
Uvira 19906.622.164.460.671
Burundi 19902.492.530.960.271
Tanganyika 19804.44   4
Tanganyika 19819.77   4
Tanganyika 19826.67   4
Kivu 19806.62   5
Kivu 19837.64   5
Kivu 198611.68     5
Kivu 19899.62   5
Kariba 19786.48   6
Kariba 19797.56   6
Kariba 19808.28   6
Kariba 19818.16   6
Kariba 198213.78     6

Sources: (1) this study;
(2) Chapman and van Well (1978);
(3) Roest (1978);
(4) Munyandorero (1989);
(5) Mannini (1990);
(6) Marshall (1987).

Figure 1

Figure 1. The growth of Stolothrissa tanganicae at Uvira (U) and in the Burundi waters (B) of Lake Tanganyika from 1988–1990.

Figure 2

Figure 2. Patterns of recruitment in the samples of Stolothrissa tanganicae collected at Uvira and in Burundi, 1988–1990.

Figure 3

Figure 3. The growth of Limnothrissa miodon at Uvira (U) and in the Burundi waters (B) of Lake Tanganyika from 1988–1990.

Figure 4

Figure 4. Patterns of recruitment in the samples of Limnothrissa miodon collected at Uvira and in Burundi, 1988–1990.

Previous Page Top of Page Next Page