Since the creation in 1958 of Lake Kariba, situated on the Zambezi River and shared between Zambia and Zimbabwe, substantial changes in both its fisheries and in the fish communities have been observed. Although probably one of the best studied fresh water systems in Africa (Karenge and Kolding, 1995a), the sustainable exploitation levels of its fish communities are still largely unknown (Anon., 1992; Kolding, 1994). Fear of overfishing, or at least indications of fully exploited resources, has repeatedly been expressed (Marshall, 1981; Marshall, Junor and Langerman, 1982; Kenmuir, 1982; Marshall 1985, Machena and Mabaye, 1987; Marshall and Langermann, 1988; Moyo, 1990; Scholz, 1993; Sanyanga, Machena and Kautsky, 1995; Sanyanga, 1995; Machena and Kwaramba, 1997), whereas other studies have contested these views (NORAD, 1985; Ramberg et al., 1987; Marshall, 1992; Machena, Kolding and Sanyanga, 1993; Kolding, 1994; Karenge and Kolding, 1995a). Most of the attempts to calculate sustainable yields (Marshall, Junor and Langerman, 1982; Marshall, 1985; Moyo, 1986, 1990) are from classic fisheries stock assessment models based on catch and effort variables, with the underlying basic assumption of ecological stability and constant regenerative capacity. Most of the results from these analyses have proven of dubious value (see Box 5.2 in Volume I).
Lake Kariba is not a stable system as most other small or medium sized lakes in Africa. Karenge and Kolding (1995b) showed that the environment, in terms of the changing hydrological regime, explains a large proportion of the variability in catch rates (CPUE). They concluded that Lake Kariba was an allothropic riverine lake where productivity was largely driven by the nutrient pulses carried by the annual floods. The question is therefore how much of the observed changes can be attributed to fishing activities and how much is due to natural environmental fluctuations. Another important management issue, particularly on the Zambian side, is the high fishing pressure and changing fishing pattern in terms of increased use of small mesh sizes and customary use of illegal fishing methods such as drive fishing (Kutumpula). It is widely believed that such uncontrolled development is a potential sign of overfishing and poses a threat to the biodiversity (FAO, 1992; Lowe-McConnell, 1994; Pitcher, 1995). On the other hand, in highly variable systems, susceptibility to increased fishing effort is thought to be low, while resilience is high and recovery potential is rapid.
Lake Kariba is a man-made grand-scale ecological laboratory with a unique chance to observe and monitor the intricate pathways of natural succession under exploitation, and a relatively good catch and effort monitoring scheme has been in place since its creation. In addition, the inshore fisheries of Zambia and Zimbabwe have evolved differently and have been subject to different types of management regimes (see Bourdillon, Cheater and Murphree, 1984; Malasha, 2003; Jul-Larsen, 2003 and Overå, 2003 for a detailed historical account and analysis). Since the overall fishing effort, catch rates and fishing patterns on the two sides of the lake are very different due to different management and enforcement, a comparative study may shed some light on the impact that these different fisheries have on each side of the same ecosystem.
After a brief description of the physical and biological environment, and the different management regulations in place, we will use the long-term time series of commercial and experimental catch rates available to describe and discuss the observed development and changes on both sides of the lake. From this comparison we will evaluate the impact of fishing on the fish stocks relative to their natural succession and fluctuations in the environment.
There are two distinct fisheries in Lake Kariba: the low cost, non-mechanized, multispecies, inshore artisanal fishery, and the highly mechanized, capital intensive, semi-industrial single-species offshore fishery on the introduced pelagic clupeid Kapenta (Limnothrissa miodon). The biological, technical, and socio-economic interactions between these two fisheries are so small (Karenge and Kolding, 1995a; Bourdillon, Cheater and Murphree, 1985) that the fisheries must be treated separately. As the particular problems of the industrial Kapenta fishery are not the main focus of this study, it will be described and analysed only cursorily.
