Relatively little is known about the fishery potential of small water bodies in the SADC countries. Even basic information is largely unavailable, including their location, sizes and distribution, water retention capability (i.e., do they dry out or not?), ownership and present usage. This situation has partly come about because the data are held by different Departments or Ministries. Furthermore, the data that exist are often collected unsystematically or inadequately. For example, in Lesotho, the Soil and Water Conservation Office of the Ministry of Agriculture registered small water bodies in 1961, but only noted them by name and index number, without recording any physical data.
One reason for this state of affairs is that their fishery potential is generally unrecognized. Very little is known about their biological productivity or the biology of their fish species. Consequently, fish production from small water bodies has been regarded as a subsistence activity with only marginal economic value. Since they are unlikely to make a significant contribution to national economies in terms of export and foreign exchange earning, they have not been accorded a high priority.
Because of financial problems and other difficulties, Fisheries Departments have not been able to give much attention to small fisheries. Small water bodies that might support fisheries are seldom under the control of Fishery Departments, as they are privately-owned or run by another government department and used for some other purpose. Their multipurpose usage was illustrated by a survey in Lesotho, where it was found that 80% of them were used to water livestock, 64% supplied water to households, 25% were used for irrigation, 25% were stocked with carp and 10% supported angling (FAO/ALCOM project, SWB/LES survey, 1993).
In order to meet the needs of the rural population and to realize the full potential of small water bodies, it will be necessary to know their number, extent and characteristics. In view of the current lack of such data, the preparation of an inventory should be given a high priority. This inventory would include data on the type of water bodies that exist in an area, their surface areas, permanence and location. This information is of fundamental importance to their management, and the preparation of a data base on small water bodies in the SADC countries is one of the principal objective of the FAO/ALCOM project, as discussed earlier, in Section 5.
Much of the information needed for an inventory can be obtained from maps, especially if digitization is possible. This method can give basic information on numbers and areas of water bodies, but is not reliable. In some cases, reservoirs can completely or partly disappear, especially in areas where siltation is a serious problem. In Lesotho, for example, 13 out of 25 dams in the Leribe district that were identified from a map were found on inspection to be completely silted. Five reservoirs that are shown on maps of the Maseru and Barea Districts no longer exist because of siltation or because their walls have been breached (C. Tilquin, ALCOM, Lesotho, personal communication).
Remote sensing by satellites can provide a quick and reasonably accurate assessment of water resources in an area. The potential of GIS for the management of small water bodies has been discussed by Kapetsky (1993) and it may become a useful tool in future. The visual interpretation of Landsat TM images (30 m resolution) at a scale of 1:250 000 has been carried out in Ghana (Kapetsky, 1991), where field verification has shown that water bodies of less than 1–2 ha in size were not adequately recorded. These were generally temporary waters in low-lying areas, and which were of limited fishery potential, and so need not, perhaps, be recorded in detail. Remote sensing has not so far been widely used in southern Africa, except for some work in Zimbabwe (Kapetsky, 1987).
Neither mapping nor satellite imagery can be relied upon if the data are not checked by verification in the field. This requires, as the first step, the collection of as much of the available documentation on small water bodies as is available. This is frequently held by various agencies, such as Fisheries, Agriculture, Water or Forestry Departments, which may not cooperate very closely with each other. The second step requires field visits to verify the data and to obtain more information from local people. Whilst field work will ultimately give the most reliable data, it is very expensive and time consuming, and some kind of subsampling will be necessary. A good example of this approach can be found in Kenmuir (1981) who attempted to assess the productive potential of small dams in a commercial farming area of Zimbabwe.
The seasonality of many small water bodies is an important consideration in their management. Because they are frequently built in semi-arid areas which are subject to frequent droughts, most of the smaller ones are impermanent. Serious droughts can affect even the largest reservoirs, such as Lake Mutirikwe (formerly Lake Kyle) in Zimbabwe, which, when full, has a surface area of 91.1 km2 but which was effectively empty by the end of 1992. It is known that at least 750 Zimbabwean reservoirs dried out completely during the 1991/92 drought. Evaporation, which is especially high during periods of drought, may also have a severe effect on small dams, as in Botswana, where evaporation can reach 2 000 mm/yr (U. Nermark, ALCOM, Botswana, personal communication).
These changes in water level may have a serious impact on fish populations, which are obviously destroyed in those waters that dry out completely. The extent of seasonal changes should therefore be included in any inventory that may be developed and appropriate management strategies developed. These could include restocking with fingerlings after dry periods; such a programme was implemented in Zimbabwe during 1993 following the drought of 1992/93 (C. Nugent, FAO, Zimbabwe, personal communication; van den Mheen, 1994).
Water bodies can be classified according to their size, since this will have a major effect on their potential yield. In general, small water bodies are more productive (per unit area) than large ones but their total production may be too low to sustain any significant fisheries. This means that a greater management input is likely to be given to the larger ones, which can support permanent fishers. On the other hand, smaller ones might be more easily managed and controlled, so making it possible for their full production potential to be realized. Since it is not possible to design specific management policies for every reservoir, general principles have to be used, and size is one factor that could assist in their formulation. However, so little is currently known about the relative productivity of large and small dams in the region that assessments of productivity based on their size need to be treated with some caution.
Small water bodies can also be classified according to their use, which will, in turn, determine the management approach that can be applied. For example, fish production in many of them could be enhanced by the addition of manure or compost, but this would be inappropriate in a dam whose main purpose is to provide drinking water. Some indication of the use to which reservoirs are put is therefore an important aspect of their classification.
It is generally assumed that almost any water body will support fish populations that can be exploited. This is not always the case, since some of them, especially highly seasonal ones, may have only a few small fish species or none at all. Dams are generally closed for the first time during the dry season and, if the river bed is dry, their fish populations may be greatly depleted and some species may be absent altogether (e.g., Chimutsi Dam, Zimbabwe, which apparently has no Clarias gariepinus). Natural water bodies usually have well-established fish populations adapted to their environment.
Determining the state of fish stocks in small water bodies should therefore be given a high priority. It would be valuable to know which cannot support any fish, which can do so only at certain times of the year, and which can do so permanently. The nature of the fish populations could also be taken into account, from which management strategies, such as stocking with certain species, could be developed.
Once something is known about the size, usage and fish stocks of small water bodies, appropriate fishing strategies could be devised. Highly seasonal waters, for example, could be fished most intensively while the water level is falling, in order to capture as much fish as possible before they are lost through natural mortality. Fishing could then be restricted during the rains, when it is filling and the surviving fish are reproducing. Permanent waters, by contrast, can be fished continuously, but at a lower level of intensity, which means that the fishers have to be subject to a greater degree of control by the Fisheries Department. Finally, some waters are suitable for interventions such as cage culture or fertilization, and in which case they are managed more like aquaculture systems than capture fisheries.
The question of which physical, chemical and biological variables needed to be measured was considered at the Technical Consultation on Enhancement of Fisheries in Small Water Bodies (Harare, Zimbabwe, 25 to 29 January 1993). It was concluded that it would be desirable to collect as much detailed information as possible from as many water bodies as possible, but it was clear that this would not be possible with the resources that were available in the region. Consequently, the parameters were ranked in three levels of priority, based on the cost and difficulty of collecting them. These are summarized below. Note that an asterisk (*) indicates those variables that change extensively and for which the sampling frequency will determine the precision of the estimates. They should not be sampled less often than once every season.
Priority 1: Relatively easy to obtain and should be available for all small water bodies.
Altitude and latitude, which determine water temperature.
Catchment area, from which information on inflows and outflows can be obtained.
Geology and soil types in the catchment area, which affect water quality and productivity.
Annual and monthly rainfall, related to hydrology and biological productivity.
River systems, related to the species composition of the fish and productivity.
Capacity, to determine residence time, which affects biological processes and may be a better means of estimating productivity than surface area.
Mean depth, related to productivity.
Maximum depth, related to stratification and the probability of an anaerobic hypolimnion.
Surface area *, related to the total yield, with some indication of variability also required.
Priority 2: Rather more difficult to obtain, but should be collected wherever possible.
Water and air temperature*, maximum, minimum and average.
Shoreline length, influences littoral processes and allochthonous inputs.
Hydrology*, water residence time affects productivity.
Drawdown pattern*, affects productivity, may cause fish kills.
Secchi disk transparency*, measures turbidity and may be related to primary productivity.
pH*, can limit productivity at extreme values.
Conductivity*, related to nutrient concentrations.
Alkalinity*, influences productivity.
Chlorophyll a*, determines primary productivity, large-scale evaluation possible with remote sensing.
Gross primary production*, measures autochthonous energy fixation, is closely correlated with fish production.
Macrophyte cover*, also measures autochthonous energy fixation and influences biological processes.
Species composition, fish species assemblage influences overall yield.
Previous introductions of fish, determines the composition of the species assemblage.
Land use in the catchment area, related to allochthonous inputs.
Number of fishers and gear in use*, measures fishing intensity, catch and catch per unit effort.
Number of households and livestock using reservoir, influences water quality and nutrient supplies.
Priority 3: Can only be obtained from selected reservoirs as part of major research programme.
Water chemistry*, ionic and nutrient status determines biological productivity.
Nitrogen and phosphorus loadings*, influence productivity.
Fish biomass*, useful data for yield prediction.
Biological fish production*, important for understanding ecosystem and predicting yield.
Fishery catch and effort*, important data for stock assessment.
Experimental catches*, provide important data for stock assessment.
