Previous Page Table of Contents Next Page

Other resources

Economic geology and minerals

Southern Africa is very rich in mineral and mining products, including (in order of importance): gold, diamonds, coal, platinum, iron, copper, limestone, nickel and chromium (Figure 29). In addition, quarried stone is processed for road and other construction, and sand and gravel are extracted from riverbeds and other sources. The richest concentration of minerals, in particular extensive gold fields, in South Africa is found along the Witwatersrand around Johannesburg in Gauteng Province, which is at the southern divide of the Limpopo River Basin. Botswana is especially rich in diamonds, and also has coal, copper and nickel. The main mineral reserves of Zimbabwe in the Limpopo River Basin include gold (southeast of Bulawayo), coal, asbestos, limestone, iron and emeralds. These resources are of great economic importance to local and national economies, especially in Botswana and South Africa.

Location and type of all mining operations located in the Limpopo and Elephants River Basins

Source: MMSD Southern Africa Working Group (2001).

The mining of minerals has significant ecological consequences. Until recently, there was a general lack of rehabilitation and ecological protective measures at mining sites. Although substantial environmental improvements have been made in the last few years, many land and water areas remain heavily polluted. Some of the most common environmental effects of mining are:

Fish resources

The Limpopo River has few fish species compared with other rivers in Africa. This is primarily because of the harsh environment with its wide variations in temperature, prolonged dry periods, and highly variable river levels. There are greater fish populations in the more permanent tributaries and in the many dams built within the catchment, especially in South Africa.

The lower zone of the Limpopo River system is important to Mozambique as its flows contribute to the productivity of the coastal brackish water area, where fish and shrimp production is significant. The fisheries of the Limpopo River make very little contribution to the economy and nutrition of the people of Botswana and Zimbabwe at present. Some Tilapia species have been introduced in Zimbabwe from the Zambezi River system.

There are at least 30 fish species inhabiting the Limpopo River (Box 12). Fish species, such as cyprinids (Schilbe spp.), catfish (Clarias spp.), substrate-brooding tilapias (Tilapia spp.), mouth-brooding tilapias (Oreochromis spp.), the introduced trout (Salmo trutta) and several brackish-water species in the lower reaches of the river in Mozambique, can be a source of food and income for the people living near the rivers and dams. These same species are suitable for aquaculture wherever soil conditions permit and where water is available for a substantial part of the year. The same indigenous and introduced species can be stocked in dams and reservoirs in order to enhance fish production. However, the abundance and catch magnitudes have yet to be determined although some estimates have been made of fish populations in some dams. Possibilities exist for fish farming in all four countries, and increasing fish production and supply should be studied and ascertained. The major problem of managing fish in available dams will always be one of re-stocking when water is plentiful after a long period of drought.

BOX 12

Fish species of the Limpopo River

Ambassis spp., possibly three species in lower reaches;

Amphlius natalensis, in coastal areas;

Amphlius uranoscopus, in the lower reaches of the river;

Aplocheilichthys johnstoni; A. katangae; a variety of exotic poeciliids;

Austroglanis sclateri, translocated through the Orange-Vaal water transfer schemes;

Chetia flaviventris;

Chiloglanis pretoriae, C. paratus, C. swierstrai;

Clarias gariepinus, C. ngamensis, C. theodorae;

Gambusia, an exotic species;

Glossogobius callidus, G. giurus in the lower reaches;

Lepomis macrochirus, an exotic species;

Micropterus spp., an introduced bass;

Mugilidae spp. (mullet), numerous species in coastal areas;

Nothobranchius orthonotus, N. rachovii, N. furzeri in coastal areas;

Oncorhynchus mykiss, introduced exotic;

Oreochromis macrochir, translocated into parts of Zimbabwe from the Zambezi River system;

Oreochromis mossambicus;

Oreochromis niloticus, an exotic species with a barred tail;

Oreochromis placidus, probably in lower river area;

Perca fluviatilis, an exotic species;

Psedocrenilabrus philander, a small cichlid;

Salmo trutta, brown trout;

Schilbe intermedius;

Serranochromis meridianus, possible in lower reaches of the river;

Serranochromis thumbergi, introduced in parts of Zimbabwe from the Zambezi River system;

Synodontis zambezensis; an introduced exotic species;

Tilapia sparrmanii, T. rendalli, substrate-spawning tilapias.

Source: Bell-Cross and Minshull (1988); Lévêque, Bruton and Ssentongo (1988).

Land degradation and desertification

Land degradation and desertification are related terms or processes. The accepted UNCED definition is that desertification is land degradation in arid, semi-arid and dry subhumid areas resulting from climatic variation and human activities.

Land degradation threatens economic and physical survival (UNEP, 1999), and could lead to household and national food insecurity in many countries, including the southern African region. Crop yields could be reduced by 50 percent within 40 years if degradation continued at present rates. In South Africa, it is estimated that about 400 million m3 of soil is lost annually. Key issues to be addressed throughout the region include declining soil fertility, escalating soil erosion, agrochemical pollution and desertification.

Main types of erosion and land degradation

Land degradation is a composite term, loosely defined as a sustained loss in the quality and the productive capacity of the land. As land degradation progresses, efforts by land users to secure a living become increasingly precarious and uneconomic. The most common indication of land degradation is soil erosion. Subtler but equally important factors include reduction in vegetation cover and changes in vegetation species composition. Stocking and Murnaghan (2000) provide sound practical guidelines on land degradation and its assessment in the field, viewing degradation from the perspective of the land user. Figure 30 illustrates how components of land degradation interlink with many other components that influence the quality and productivity of land, including how it is used or misused.

Identification of the direct and indirect causes of land degradation is essential, as any remedial measures designed to rehabilitate land must tackle the root causes of the problem in order for the reversal of land degradation processes to be successful.

The land degradation wall

Source: Stocking and Murnaghan (2000).

A clear distinction between causes, mechanisms and impacts of land degradation is often lacking in degradation studies (Mainguet, 1991), and it is difficult to distinguish between human and natural influences as both may occur simultaneously. Drought is often quoted as a direct cause of degradation, but it is also seen as the catalyst for other processes that lead to degradation.

Physical factors

Physical factors always play a role in degradation processes, but their role is less crucial than assumed in erosion hazard mapping. Nevertheless, the topography (slope in particular), the properties of the soil and underlying rock, the vegetation, and climate characteristics (rainfall in particular) are important factors in the acceleration of humaninduced erosion.

Increased runoff and accelerated erosion relate strongly to poor surface conditions, including surface crusts, lack of vegetation, and compaction (decreased infiltration). In addition, the properties of soil horizons and other underlying materials play a major role, especially with respect to gully erosion. Porous materials such as weathered rock (saprolite) and soils with high hydrologic conductivity are conducive to gully erosion as relatively large and rapid subsurface flows may occur, causing collapse of the gully head or sides. Soils with weak structure and friable consistency are vulnerable to erosion, e.g. soils high in illitic or mixed-layer clays. The same applies to soils with high sodicity, inducing a high clay dispersion rate (Solonetz and Planosols).

Climate variability has a profound accelerating effect on erosion and land degradation. Extreme rainfall events aggravate the condition of already degraded land through increased runoff and flooding. Lack of rainfall and resulting drought accelerate desertification processes. Drought acts as a strong catalyst in the initial and progressive degradation of land.

The cause of accelerated erosion is mostly a complex of several factors, as the two main spheres of influence - physical and human - are interrelated and interactive. There is always an element of human influence involved, related to the management of the resources, which may be aggravated by the conditions and characteristics of the environment (climate, soils, geology and landscape). Stable landscapes resistant to erosion normally show less erosion compared with vulnerable environments. Similarly, areas with reliable rainfall in general show less degradation than areas with frequent drought.

Most surveys and studies of land degradation conclude that the primary causes are related to land use, management and socio-economic attitudes. Increase in population is cited as the single most important cause of degradation. However, with land becoming scarce, communities may become more aware of the necessity to improve land management and conservation.

Assessment of erosion and land degradation in the Limpopo River Basin

The Global Assessment of Soil Degradation (GLASOD) by the UNEP is considered the first global assessment of the geographical distribution of human-induced soil degradation (Oldeman, Hakkeling and Sombroek, 1990). Soil degradation is described in terms of: type of erosion and deterioration; cause; degree; rate; and relative extent. The overall status or severity of soil degradation is indicated by a combination of its degree and extent, represented on maps by five classes of severity: none, slight, moderate, high and extreme.

Erosion severity in the Limpopo River Basin

The main result of the GLASOD project is a map of the world (scale: 1:5 million) showing the occurrence of human-induced actual soil and land degradation. The occurrence of the several general areas of degradation in the Limpopo River Basin can be distinguished on the map as follows (Figure 31):

Apart from the generalized pattern showing several unnaturally shaped units, the validity of the information may be questioned in a number of areas. The strong contrast between the southern part of the Limpopo River Basin and the adjacent Vaal River Basin does not seem realistic. The differences within Botswana do not appear to reflect original information or present status adequately. A substantial part of the South African lowveldt strip has no apparent erosion, notably the area covering Kruger National Park. This same unit should not include Swaziland, where some of the most degraded areas in southern Africa occur.

Hakkeling (1989) described the GLASOD results in more detail for the southern and eastern African regions covering Botswana, Mozambique and Zimbabwe, but not South Africa. This regional map shows a pattern of erosion and degradation in places quite different from that on the integrated global map. In general, the regional map provides a more accurate overview and the country descriptions below draw heavily on this information.


The information from the GLASOD regional map (Hakkeling, 1989) indicates that almost all units fall into the overall class of high degradation, except for one area with slight degradation, because of dominant sandy soils. The units most affected by water erosion - both sheet and gully erosion - are found in the northern and southern parts of the Limpopo River Basin, caused mainly by overexploitation of vegetation and intensive cropping. Wind erosion is the dominant type of degradation in the middle units, caused primarily by overgrazing. Up to 50 percent of all units are considered to have recovered by natural stabilization from earlier degradation.

Reports from the first half of the twentieth century already expressed concern about the state of the land resources. Studies from the second half of the twentieth century indicated that the problem was severe, and that numerous examples of extreme overgrazing could be found. However, it is also claimed that such conclusions are based on flawed data, that the perception of range condition and management is biased in favour of western models, and that most of the examples of degradation are temporary conditions that are natural to a variable savannah ecosystem.

Evidence is based mainly on examples of: bush encroachment; decrease in general grass density and in numbers of the more nutritious and palatable species; and of an increase in patches of bare soil. The consensus is that rangeland degradation is occurring, but there is disagreement as to its extent, severity and reversibility. Abel and Blaikie (1989) argue that degradation should not be measured by using indicators such as short-term vegetation changes. Instead, changes in the environment should be judged according to their degree of irreversibility over longer periods. The assessment of detrimental changes must also take into account estimates of secondary production interest of the users, for which reason the productivity of communal grazing land may be higher than that of commercial ranches (De Ridder and Wagenaar, 1984).

The description of erosion and land degradation within the Limpopo River Basin (hardveldt) of Botswana is confined mostly to local observations, and although overviews have been produced (Arntzen and Veenendaal, 1986; Dahlberg, 1994), a comprehensive inventory is lacking. Ringrose and Matheson (1986) reported an increase in desertification manifested as decreased vegetation, increased erosion and reduction of soil water retention as a result of loss of soil organic matter induced by overgrazing and fuelwood collection. In this regard, the BRIMP also generates various degradation maps and datasets, based on ground monitoring and the interpretation of Landsat images specifically aimed at addressing desertification questions.

Occurrence of severe sheet and gully erosion is reported from several sites in east Botswana near Serowe and Kalamare, but erosion is generally estimated to be slight or moderate. Most of the erosion is associated with sloping land, including the footslopes of hills. The flat and slightly undulating parts of the plains show less evidence. Wind erosion is reported from areas with a bare surface, especially from fallow arable land, but the severity is difficult to estimate. This effect is most pronounced after periods of drought and reduction of the protective vegetation cover.


The information from the GLASOD regional map indicates that the northernmost units of the Limpopo River Basin have no livestock because of tsetse fly, hence, no erosion is described. Most other units have the overall class of slight degradation, except for two units with moderate degradation. High degradation is reported in the Changane Valley owing to crusting and sealing. Most of the degradation is caused by wind erosion - with some nutrient losses - but a variety of causes have been observed: overgrazing in the Lebombo Hills, intensive cropping along the coast, and salinization in the irrigated areas near Xai-Xai in the Limpopo River floodplain. Moreover, natural stabilization seems a common process in Mozambique.

Assessment of erosion risk in Mozambique was first undertaken on a national scale by FAO (1985) when Reddy and Mussage compiled a first approximation of an erosive capacity index. The low rainfall areas of Gaza Province were classified as a low erosion-hazard zone. High erosion-risk zones included the coastal areas of Gaza Province. Population density in the coastal belt has raised concern about dune vegetation, mangroves and coral reefs. While none of these ecotypes could yet be considered critically threatened in Mozambique, local areas of degradation have been identified and the government is anxious to take remedial action before the problems become more severe.

The various studies and surveys reviewed indicate that major problems in the development of the Limpopo Valley soils are related to salinity and alkalinity. These problems arise from the geological structure, which has very saline marine sediments under the recent alluvial sediments (primary salinization). Furthermore, mismanagement of the irrigation water supplies and of the drainage systems has raised groundwater tables and led to the efflorescence of salts at the ground surface (secondary salinization).

South Africa

The GLASOD map of Africa (Oldeman, Hakkeling and Sombroek, 1990) generally indicates less soil degradation in the Limpopo River Basin as compared with the rest of South Africa. Most of Limpopo Province shows none to slight degradation, with moderate degradation in the eastern parts, all caused by water erosion, and primarily caused by overgrazing. Moderate to high degradation resulting from pollution and acidification occurs in the south of the catchment (Pretoria-Johannesburg-Witbank area). The most severe soil degradation caused by overgrazing occurs in the former homeland areas in the central Limpopo Province.

Hoffman and Todd (1999) have provided an overview of erosion and land degradation in South Africa, in a first-phase review for the development of South Africa’s national action programme to combat desertification. A number of general conclusions from this report were presented, which are also relevant to the Limpopo River Basin. The focus of land degradation has historically been on the degradation of vegetation, in particular of the rangelands. The study suggests that equal attention should be paid to degradation of vegetation, soil and water resources.

Earlier inventories of soil and land degradation have concentrated on erosion hazard with relatively little attention given to the influence of land use practices. The present view is that both land use and land tenure exercises have a significant influence on land degradation.

Hoffman and Todd (1999) state that soil and vegetation degradation is perceived as being significantly greater in communal areas as compared with commercial areas, by at least a factor of two. However, specific forms of vegetation degradation are more of a problem in commercial areas, such as change in species composition, alien plant invasion, and encroachment of indigenous woody species. It is suggested that areas with steep slopes, low rainfall and higher temperatures are significantly more eroded. Climate change in the last century may have had an impact on the intensity of erosion, but this needs to be further studied, in particular in order to define the relationship between drought and erosion.

The government policy of land allocation has had a major effect on land use and land degradation. In the commercial farming areas, this has been partially conducive to sustainable land use, but not so in the communal areas, where practices of crop production, communal grazing and use of the vegetation have led to accelerated erosion and degradation. The understanding in South Africa of communal land use and production systems is poorly developed, as is its relationship with land degradation (Hoffman and Todd, 1999).

In an overview on soil degradation in South Africa, a report by FAO (1998a) states that data on degradation are incomplete and fragmented, with little information on spatial distribution and distinction between natural and human-induced erosion. Relevant land degradation information is available from a number of other sources (Newby and Wessels, 1997; Van Zyl, 1997; Scotney, 1995; Laker, 1994).

Nationally, it is estimated that water erosion affects 6.1 million ha, or more than 40 percent of the total of 14.6 million ha of cultivated soil in South Africa. Of this, 15 percent is seriously affected, 37 percent moderately affected and the remaining 48 percent slightly affected. Laker (1994) estimates that about 20 percent of all topsoil has been lost since the beginning of the twentieth century. Average annual soil loss is estimated at 2.5 tonnes/ha, equal to about 300 million tons. Wind erosion affects even more cultivated land, an estimated 10.9 million ha, or more than 70 percent. Of the affected cultivated land, 7 percent is seriously eroded, 29 percent moderately and 64 percent slightly. Compaction affects about 2 million ha, or about 15 percent of all cultivated land. Surface sealing is also a widespread and a serious problem. It is estimated that a total of 15 percent or more of the 1.2 million ha of irrigated land is moderately to severely affected by salinization and/or waterlogging. In addition, large areas are affected by pollution, acidification and fertility losses. Of the national grazing land about 3 million ha are rendered worthless for grazing as a result of encroachment by undesirable species.

Figure 32 shows the occurrence of soil and vegetation degradation combined in a single index per magisterial district. The provinces within the Limpopo River Basin are described as belonging to the most eroded and degraded parts of the country (which contradicts the GLASOD result). However, these estimates are based on perceptions from agricultural extension and resource conservation officers, and not on actual observations.

Land degradation in South Africa

Source: Hoffman and Todd (1999).

BOX 13

Status of soil health in Mpumalanga Province

In a study on land quality indicators, Mpumalanga Province was stratified according to vegetation biome, maize, other crops and minesoils. Within each stratification, data were collected for the indicators set out in the table below (median topsoil values are shown):





Other crops

Mine cover soil

Organic C (%)






Total N (%)






C/N ratio












EC (mS/m)












P (mg/kg)






Median values did not reflect any serious soil ill health, the climate context being taken into account. However, lower quartile pH values tended to fall below 5.5, the threshold for acid saturation problems. Atmospheric deposition of acidifying agents from coal-fired power stations might have played a role in low topsoil pH values under grassland.

Source: Nell et al. (2000).

According to Newby and Wessels (1997), the estimated extent of degradation of vegetative cover is 25 percent for North West Province (poor to very poor condition), 87 percent for Limpopo Province (poor to critical condition) and 50 percent for Mpumalanga Province (poor to critical condition). Serious wind and water erosion is reported from North West Province, respectively 53 and 36 percent of all area affected. The mining industry - predominantly coal mining - has occupied and severely polluted 50 percent of all high potential arable land in Mpumalanga. However, this statement is contrary to the experience and results of other researchers (e.g. Box 13).

Wessels et al. (2001) mapped the conservation status of natural vegetation and soils in Mpumalanga Province, including the present land cover, rangeland condition, alien vegetation and bush encroachment. Table 14 summarizes some of their main findings. One-third or more of both the grassland and savannah biomes in the province is stated to be in poor to very poor condition. Visible soil erosion damage is stated to occur over less than one-third of the province.

Soil and vegetation degradation in Mpumalanga



Percentage of biome in Mpumalanga



Good to very good



Poor to very poor


Soil erosion

Light to considerable


Soil erosion damage

None visible




Poor to very poor


Bush encroachment



Bush encroachment



Soil erosion damage

None visible


Source: Wessels et al. (2001).


Information from the GLASOD regional map indicates overall strong degradation in the communal areas in the northern and western part of the Limpopo River Basin adjacent to Botswana, with severe sheet and gully erosion caused by deforestation and intensive cropping. No erosion is described for the areas with commercial ranching as the dominant land use, and for national parks. The communal grazing area in the east has moderate soil degradation. As in Botswana, most units are considered to have stabilized in a natural way from earlier degradation.

The history of land degradation in Zimbabwe and the different views about it are quite similar to Botswana. Official interventions based on reports citing the occurrence of severe soil erosion have largely failed (Scoones, 1989a). The increase in crop and animal production in the 1980s is considered sufficient evidence that there is no substantial decline in the natural resource base (Scoones, 1992a). However, this may be a preliminary conclusion as the 1990s have shown a general decline.

Campbell et al. (1990) claim that there is sufficient evidence to indicate that cattle are linked to major environmental changes, but they also conclude that the issue of whether communal grazing practices cause degradation cannot be answered without additional data, in particular a spatial analysis.

Scoones (1989b, 1992b, 1993) finds that disregard of the heterogeneity of the environment is one of the main reasons for conflicting perceptions of environmental change. Areas will differ on the extent of degradation and the effects of the changes on secondary production. Scoones found that in most years the farmers are avoiding irreversible damage, in terms of productivity, by using strategies of herd mobility and local ecological knowledge. In drought years, migration between different savannah types is employed on a regional scale. Certain elements of the landscape or ecosystem, such as riverine strips, vleis (wet bottomlands) and drainage lines, are vital in providing fodder in critical periods.

The extent of soil erosion in Zimbabwe was mapped nationally using air photography (Whitlow, 1988). Only 5 percent of the country was badly eroded, in comparison with communal areas where badly eroded land comprised about 10 percent of the land area. Little difference was observed between cropland and grazing land. This is somewhat surprising as most arable land in Zimbabwe seems adequately protected by contour ridges. In other countries in southern Africa, erosion is always reported to be excessively higher on communal grazing land, e.g. in Swaziland (Remmelzwaal and McDermott, 1997).

Elwell (1980) calculated annual soil losses on cropland caused by sheet erosion followed by national surveys. Grohs (1994) re-interpreted Elwell’s results according to administrative areas. The highest soil losses of more than 100 tonnes/ha were recorded in the north and east of the country. The lowest values of less than 5 tonnes/ha were found in the semi-arid south, within the Limpopo River Basin. On the basis of these findings, an average annual soil loss of about 55 million tonnes was calculated.

Previous Page Top of Page Next Page