Chapter 3
GRASSLANDS OF SOUTH AFRICA

Anthony R. Palmer and Andrew M. Ainslie

SUMMARY

South Africa is subtropical, with temperatures modified by altitude. The interior, where the bulk of grasslands are found, is semi -arid to arid, with rainfall decreasing westwards. The south and southwest have winter rainfall; the eastern Cape is bimodal; and Kwa-Zulu Natal has summer rainfall. Grassland is mainly in the central, high regions: sour-veldt occurs under high-rainfall on acid soils, and sweet -veldt on fertile soils in semi-arid zones. Savannah occurs in the north and east; arid savannah extends to the Kalahari. The Nama-karoo, a vast area of steppe in the centre and west, is mostly used for sheep and goats. There are three categories of land tenure: 70 percent is freehold and managed commercially; 14 percent is communally managed without clear individual boundaries and managed for subsistence; 16 percent is reserves or freehold industrial and urban. South Africa is multi-ethnic with a majority indigenous population and a minority of descendants of colonists who own much of the commercial farm land. Natural pasture is the main feed source for grazing livestock. Production systems in communal areas, based on pastoralism and agropastoralism, are subsistence-based and labour intensive; cropland is allocated to households, grazing areas are shared by a community. Commercial areas are fenced ranches and further subdivided into paddocks; rotational grazing is normally practised. Stock rearing is very ancient; cattle predominate but sheep and goats are very important. In subsistence systems, traditional breeds predominate; in commercial farming, exotic and locally-created improved breeds prevail. Sheep are mainly commercial, and goats are for subsistence. Cattle predominate in the east, and sheep in the drier west and southeast. Goats are widely distributed. The region is home to large numbers of grazing and other wildlife, which are common on large-scale ranches and are increasing in importance as a managed resource. Low profits from domestic stock have led to an increase in game farming and ecotourism. Much of the better-watered grassland has been converted to crops; in communal areas this gives a patchwork with thicket. Fire and browsing has reduced woody vegetation, but bush encroachment remains a problem. Sown pasture is not of major importance, except on dairy farms. Over-seeding of degraded range is of limited use. Strategies for maintaining pastoral production include rotation, resting, bush control and provision of winter pasture in cool areas.

Introduction

The Republic of South Africa is situated at the southern tip of Africa. It is bordered to the north by Namibia, Botswana, Zimbabwe and Mozambique; in the west by the Atlantic Ocean; and in the south and east by the Indian Ocean (Figure 3.1). The total land area is 1 223 201 km² (excluding Lesotho and Swaziland). The enclaves of Lesotho and Swaziland are sovereign states. South Africa's population is estimated at 40.6 million (Stats SA, 1996), of which approximately 46 percent is rural and 54 percent is urban. Agriculture accounts for 3.2 percent of GDP and 7 percent (R 14.57 billion in 2000^[6]) of exports and supports, directly or indirectly, 15 percent of the population (National Department of Agriculture and Land Affairs, 2001).

In its position at the southern end of the African continent, South Africa is the gateway to the subcontinent, providing and maintaining ports and road, rail and telecommunication links between southern Africa and the rest of the world. With a long history of trade and scientific exchange with Europe and North America, South Africa has developed opportunities for marketing its agricultural products within these economies. Many of these products (beef, mutton, fleece and hides) have been derived directly from grasslands. The national science programmes and those associated with resource management and agriculture have been linked through government and tertiary education initiatives to explore trends in resource condition and production. These initiatives have focused on three primary research areas: (1) describing the biodiversity of rangelands and their associated resources; (2) developing an understanding of the impact of herbivory on the resource; and (3) developing methods for improving production. In addition, research programmes have endeavoured to understand the relationships between herbivory by domestic livestock and the sustained use of the resource for agricultural production.

South Africa has a unique combination of natural resources, climatic environments and ethnic groups, making it an interesting and challenging country. Grasslands are a major component of the natural vegetation, with the biome comprising some 295 233 km² of the central regions of the country, and adjoining and extending into most of the major biomes (forest, savannah, thicket, Nama-karoo) in the region. This interface between grasslands and other biomes contributes substantially to their floristic and faunal diversity and to the important role they play in the agricultural economy. The grasslands of South Africa are also the home to most of the human population, with the mining and other industrial complexes of Gauteng (formerly the Witwatersrand) being located on the high-veldt grasslands. This proximity to large human populations and their associated markets, as well the climatic environment, which favours commercial, rainfed agriculture, has had a large impact on the native grasslands. Millions of hectares have been ploughed and converted into dryland cultivation for the production of maize, oilseed, millet and other commercial rainfed crops. Commercial ranching of cattle and sheep for the markets in Gauteng, Mpumalanga and the Free State has placed pressure on the grasslands, resulting in changes in species composition and production potential. However these trends are not ubiquitous, and millions of hectares of native grassland still occur.

Figure 3.1 - The Republic of South Africa, showing its position in southern Africa.

Grasslands are also the most important resource available to the graziers in developing regions of South Africa. The former homelands of Transkei, Ciskei and KwaZulu Natal, situated on the eastern seaboard, are predominantly grassland. The inhabitants of these regions are dependant upon this resource for the production of meat, milk, hides and fleeces, and for the provision of draught power, as well as other traditional uses of livestock. Although these products are not produced in conventional commercial systems, they contribute substantially to the economy and food security of these regions. In this chapter, we will introduce the role of grasslands in this economy based on communal land tenure systems.

The grasslands (see Figure 3.2) adjoin a number of other economically important biomes (savannah, thicket and Nama-karoo) and grassland patches are found within these biomes. It is important to include these biomes in this chapter, as their ecology is strongly interlinked, and to consider the role that all grasslands and their associated biota play in the economy of South Africa. In this chapter, grasslands in South Africa will have a wider definition than at the biome level.

Figure 3.2 - The extent of grasslands in South Africa.

The grassland resources of South Africa have been extensively reported, with four important publications appearing recently (Cowling, Richardson and Pierce, 1997; Dean and Milton 1999; Tainton, 1999, 2000). These provide exhaustive information on types of range land resources; their general ecology, including history, biodiversity, species composition and associated environmental conditions; dynamics; productivity; and land-use and management options available to their peoples.

In addition, information applicable to the management of grasslands in southern Africa is provided in the approximately 980 research publications that have appeared since 1966 in the African Journal of Range and Forage Science and its predecessors. Other peer-reviewed scientific journals that provide exhaustive information on the natural resources of South Africa include the South African Journal of Botany, South African Journal of Science, Memoirs of the Botanical Survey of South Africa, South African Journal of Wildlife Research, and Bothalia. Researchers are strongly encouraged to publish in the wider international literature and many important research articles appear in peer-reviewed journals published elsewhere. This chapter does not attempt to synthesize or review all this available information, but provides a brief summary of the current status of our understanding of southern African grassland ecosystems.

Physical features

The Great Escarpment and the Drakensberg mountains provide the physical barriers that largely determine the climate and vegetation of much of the livestock growing regions of South Africa. The combination of moderate to high rainfall and high elevation associated with these features means that the largest area of native grasslands occurs here. In geological time, several phases of uplifting, erosion and deposition have created complex landforms determined by the underlying geology. The country has five main physiographical regions at differing elevations (Figure 3.3).

• The southwestern fold mountains, which influence the climate and vegetation patterns of the southern Cape.

• The coastal plain, which extends from the Namibian border on the west, all along the coast to southern Mozambique on the east. This narrow plain between the Ocean and the Great Escarpment is the region with the most fertile soils, moderate to high rainfall, and where most intensive livestock production occurs.

• The Great Escarpment, which forms the major barrier to moisture reaching the interior, together with the central high-veldt, contain most of the high elevation grasslands. The major urban, mining and agricultural activities take place in the central high-veldt, which lies at 1 600-1 700 m above sea level.

• The great Karoo basin lies at 1 400-1 600 m and contains the steppe-type vegetation associated with fertile aridosols of a semi -arid region.

• The Kalahari region, bordering on Namibia and Botswana, also represents a very important extensive livestock producing area. The region is the southern part of the continental-scale basin, which is covered by sands of varying depth (sometimes >200 m). Deep boring technology has enabled commercial graziers to become permanently established in the region and to optimize livestock production off arid grasslands. The vegetation is an arid savannah (Plate 3.1), with a carrying capacity of 30 to 40 ha per Livestock Unit (LSU).

Figure 3.3
Elevational map (in metres) for South Africa.

SOURCE: Dent, Lynch and Schulze, 1987.

Plate 3.1
Kalahari: The arid savannah occurs in the northwestern portions of South Africa and southern Botswana, and is associated with the sands of the Kalahari system. The vegetation comprises a woody layer of mainly single-stemmed deciduous shrubs, and a ground layer of grasses and forbs.

A.R. PALMER

South Africa is thus characterized by a high interior plateau, surrounded on three sides by the Great Escarpment and the Drakensberg mountains, which provide the physical barriers that largely determine the climate and vegetation. The plateau is intruded by several mountain massifs, with the highlands of Lesotho exceeding 3 000 m in places. The northern and western sections of the plateau contain two large basins, namely the Kalahari and the Transvaal Bushveldt (Partridge, 1997). Adjacent to the Great Escarpment lies a coastal plinth that varies in width from 50 to 200 km. This plinth is incised by deep riverine gorges.

Climate

Rainfall

With a mean annual rainfall of approximately 450 mm, South Africa is regarded as semi- arid. There is wide regional variation in annual rainfall (Figure 3.4), from <50 mm in the Richtersveldt on the border with Namibia, to >3 000 mm in the mountains of the south western Cape. However, only 28 percent of the country receives more than 600 mm (Table 3.1).

The uncertainty of the rainfall is best expressed by the coefficient of variation in annual rainfall (Figure 3.5). The low rainfall regions have the highest coefficient of variation and drought is common. Annual rainfall distribution is skewed such that there are more below-average than above-average rainfall years, and the median is more meaningful than the mean. The high seasonal variations are accompanied by high spatial variability, and the annual potential evapotranspiration (PET) may exceed annual precipitation by ratios of up to 20:1, hence drought conditions are a common phenomenon (Schulze, 1997). The declaration of drought status to a magisterial district has historically been used by the Department of Agriculture and Land Affairs to intervene in exceptional circumstances to assist land users. Since 1994, this intervention has been discouraged; instead, graziers are encouraged to plan their production system within the long-term expectations of their farms.

Figure 3.4
The median annual rainfall for South Africa.

SOURCE: Dent, Lynch and Schulze, 1987.

TABLE 3.1
Annual rainfall distribution and climatic classification in South Africa

Figure 3.5
The coefficient of variation in annual rainfall for South Africa. Derived from the long-term rainfall records (50 years or more data) from 1015 stations.

Seasonality of rainfall

There are three major zones within the country, namely the winter rainfall region of the western, southwestern and southern Cape; the bimodal rainfall region of the Eastern Cape; and the strong summer seasonality of the central high-veldt and KwaZulu Natal. The regions with strong summer seasonality are strongly influenced by the inter-tropical convergence, which moves southwards during the southern hemisphere summer. The season of rainfall in the southwestern and southern coastal regions is influenced by the frontal systems developing in the southern Oceans. These frontal systems bring cool, moist air during the winter season (June-August) and promote the development of sclerophyllous and succulent floras. In general, the natural vegetation of these regions is less useful for livestock production. Because of the varying rainfall seasonality, growing periods vary throughout the country. In the north, east and along the coastal belt, summer seasonality encourages C4 grass production and the main focus is on cattle and sheep production. In the semi -arid central and western regions, C3 grasses and shrubs predominate, and this favours sheep and goat production.

Temperatures

Temperatures in South Africa are strongly determined by elevation and distance from the sea. The high elevation (1 500-1 700 m) inland regions experience a warm summer (January) with mean daily maximum temperatures of 26-28°C and cool winter (July) mean daily minima of 0-2°C), with frost during the coolest months (Schulze 1997). These conditions favour the development and maintenance of grassland. This region experiences occasional snow. The warm Mozambique current on the east coast plays a strong role in ameliorating temperatures along the coastal zone between East London and Mozambique. The northern parts of the coastal zone experience warm winter daily minima (8-10°C) and warm summer maxima (32°C) and the climate is strongly subtropical. The vast interior, represented by the Kalahari basin and the Nama-karoo, experiences a more extreme climate, with low winter mean daily minima (0-2°C) and high mean daily summer maxima (32-34°C). The southern and southwestern coastal zone experiences moderate winter mean daily minima (6-8°C) as a result of the circumpolar Westerlies that bring moist, cold air from the southern Oceans during June, July and August (winter). The temperatures on the west coast, from Cape Town to Port Nolloth, are influenced by the cold Benguela current. This arid region experiences July mean daily minima of 6-8°C, but little or no frost, and is able to support a rich succulent flora. The cold ocean current favours the development of fog during the winter months, bringing cold, moist air onto the coastal plain.

Soils

The relatively young South Africa n geology gives rise to soils of high nutrient status. The Nama-karoo biome of the central regions comprise predominantly mudstones and sandstones of the Karoo Supergroup, which give rise to shallow (<30 cm) aridosols, typically with a calcareous hardpan layer in the profile. During the Jurassic age, these sedimentary rocks were intruded by dolerites, which criss-cross the landscape in characteristic dykes. The dolerites contain plagioclase, which give rise to soils of high clay content, and these features contain many grasses and associated phreatophytic woody shrubs and represent refugia for many desirable (to the herbivore) plant species. The dolerite sills and dykes provide summer grazing, whereas the nutrient-rich, calcareous plains provide abundant, high quality winter forage.

The savannahs of the Mpumalanga Lowveldt are associated with the gabbros and granites of the Bushveldt igneous complex. The latter give rise to sandy soils of moderate nutrient status. The gabbros give rise to a nutrient-rich Mispah rock complex.

In geological time, several phases of uplifting, erosion and deposition have created complex landforms determined by the underlying geology. The Cape Fold Mountains and the Lesotho Highlands are the largest surfaces that intrude above the African plane. The Cape Fold Mountains are siliceous rocks, giving rise to immature, litholic soils. The Lesotho Highlands, in contrast, are basaltic, giving rise to mollisols (Partridge, 1997). The grasslands of the high-veldt are associated with soils of basalt and andesitic origin, with high nutrient status.

People

Hominids have occupied southern Africa for three million years (Volman, 1984), and although the ancestral forms used fire (Thackeray et al., 1990), there are no artefacts that enable us to quantify the extent of their impact. They undoubtedly burnt large tracts of land in the interior, thereby promoting the development of grasslands, and early hominids must be regarded as a significant agent in the evolution of grasslands in South Africa.

Contemporary South Africa is a multicultural nation, with many ethnic groups and colonial nations represented in its populations. It is this wide variation in the origins of its people that make understanding the management of its natural resources so challenging. The remaining San people of the southern Kalahari represent the oldest traditional users of natural vegetation for survival. San people are still able to subsist as hunter-gatherers in the most arid regions of the country, providing some evidence of how it is possible to sustain small human populations in this region. San exhibit a strong understanding of resource limitations and probably follow the principles embodied in the disequilibrium theory (Ellis and Swift, 1988) the closest of all southern African people. Until the end of the nineteenth century, San also survived in the mountainous regions of the Drakensberg and along the Great Escarpment. The evidence of their history is found in the numerous rock paintings and other artefacts that occur in caves along the Great Escarpment. They hunted on the grasslands of the mountainous interior.

The Nguni people of the eastern seaboard are graziers with a long (>10 000 years) history of maintaining domestic livestock. These people comprise the Seswati, AmaZulu and AmaXhosa nations, and occupy the leasehold lands in the former homelands of Gazankulu, KwaZulu Natal, Transkei and Ciskei. The society is organized around a village, comprising dwelling units, cultivated lands and grazing lands. Their early cattle were of Bos indicus stock and this line is being developed and protected in recent years with the establishment of an Nguni studbook. Situated on the eastern escarpment and in the Drakensberg is the mountain kingdom of Lesotho, the home of the Basotho people. Lesotho falls entirely within the grassland biome and the Basotho people are cattle and sheep farmers, depending largely on the natural grassland for production. Almost all of Lesotho is communally managed and the challenges to managing the grasslands sustainably remain the same as those of communal rangeland in South Africa.

Europeans of Dutch descent first arrived in South Africa in 1652, and settled initially at the supply station in Cape Town. These settlers were joined by French Huguenots, who brought with them a knowledge of viticulture and animal husbandry (mainly sheep). Descendants of the early Dutch settlers began moving into the interior of the country with the abolition of slavery, and developed the extensive cattle and sheep farming enterprises that currently occupy land in the Kalahari, central Free State and the North West Province. It was only in 1820 that settlers of British origin arrived and settled on the eastern seaboard. They developed mixed -farming operations in the Eastern Cape and Kwa-Zulu Natal, including cattle and wool-sheep enterprises.

Livestock

South Africa's national commercial cattle herd is estimated to number 13.8 million, including not only various international dairy and beef-cattle breeds, but also indigenous breeds such as the Afrikaner (or Afrikander). Locally developed breeds include the Drakensberger and Bonsmara. These breeds are systematically and scientifically improved through breeding programmes, performance testing and the evaluation of functional efficiency. Almost 590 000 t of beef was produced in 2000. Owing to relatively low carrying capacity on the natural pastures, extensive cattle ranching is practised in the lower rainfall regions.

In addition to the cattle, in 1999 there were about 25.8 million sheep and 6.3 million goats in the country, in addition to smaller numbers of pigs, poultry and farmed ostriches. The numbers of cattle and small stock fluctuate in response to high and low rainfall years. The 1999 census data shows the distribution between the freehold and communal sectors (Table 3.2). Beef production is the most important livestock-related activity, followed by small-stock (sheep and goat) production. The combined livestock sector contributes 75 percent of total agricultural output (National Department of Agriculture, 1999). Livestock numbers and production for the period 1995-2003 are shown in Table 3.3).

TABLE 3.2
National livestock census 1999.

SOURCE: National Department of Agriculture, 1999.

TABLE 3.3
Production (×1000 t) in the period 1995-2003 of beef and veal; chicken; mutton and lamb; goat; game; wool; and milk.

SOURCE: FAO database 2004.

The grasslands support a high proportion (70-80 percent) of the total sheep and wool produced. The main breeds of sheep are fine-wool Merino, the South African mutton Merino, Dohne Merino, Dormer, Dorper (the last-named two are locally developed breeds) and the Karakul. The Nama-karoo, a steppe like vegetation of the central and western regions, supports both sheep and goat enterprises. The Karakul industry is limited to the dry northwestern regions of Northern Cape Province.

Wildlife

South Africa possesses a rich and diverse wildlife resource, with many unique and interesting mammals, birds, reptiles and amphibians, providing a wide range of products, including tourism opportunities, meat, hides, curios, recreation and trophy hunting. There are 338 large and small mammal species (Smithers, 1983) and 920 bird species (Maclean, 1993). Four families (Elephantidae, Equidae, Bovidae and Suidae) contribute most of the large mammal taxa and represent the largest biomass of primary consumers. During the last thirty years, the large mammal fauna has begun to make a significant contribution to the economy of rangeland through increasing ecotourism and the development of private game farms and nature reserves. About 10 percent of the country is designated as National Parks and formal conservation areas, but a considerable proportion of the wildlife exists outside formally proclaimed conservation areas. Many livestock farmers derive some or all of their income from hunting or ecotourism. In 1997, approximately 8 000 private game ranches, covering some 15 million hectares, had been established (Grossman, Holden and Collinson, 1999). This figure has continued to increase rapidly since then, with many farms being enclosed by game-proof fencing. The issuing of a certificate of adequate enclosure by the various provincial nature conservation authorities permits the landowner to exercise rights over the wild herbivores that would otherwise only exist during the so-called “hunting season” in May-July each year. Individual landowners are now able to capture, transport, hunt and introduce any wild animal for which a permit has been provided and for which there is a certificate of adequate enclosure. Although this has had some positive consequences for the protection of certain rare species (e.g. mountain zebra, blesbok), where the last remaining populations were on private land, it has had some negative impacts. These include the large-scale introduction of common and freely available native species (e.g. impala, nyala, warthog) to regions where there is no record of their historical occurrence. The full consequences of these introductions on other species (e.g. nyala on kudu; impala on bushbuck) are not properly understood and further research is required. Conservation agencies have themselves been guilty of transgressions of this nature, re-introducing species such as warthog in the Eastern Cape Province, which have proliferated and are now regarded as a problem animal by graziers.

The natural flora, comprising some 24 000 taxa, is one of the richest floras in the world and creates many opportunities for developing the ecotourism industry. Regions of particular significance include Namaqualand, which attracts visitors to its unique floral displays during September of each year. The diverse Cape floral kingdom, with its estimated 8 000 taxa and associated avifauna, also provides the visitor with a glimpse of unique evolutionary forces driving speciation in the region. The southern Cape and its “garden route”, with a high structural diversity (Afromontane forests, coastal thicket, lowland fynbos and mountain fynbos), attracts many international visitors.

Land tenure

There are four broad categories of land use in South Africa that are relevant to agricultural production, representing various land tenure regimes. Approximately 70 percent of the country is “commercial” farmland under freehold tenure, 14 percent is state land that is communally managed, 10 percent is formally conserved by the State as National and other parks, and the remaining 6 percent is freehold land used for mining, urban and industrial development. The communal areas are situated mainly in the former homelands of Transkei, Ciskei, Bophutatswana, Lebowa, Kwa-Zulu, Venda and Gazankulu in the north and east of the country, while the commercial areas occupy most of the western, central and southern regions.

There are two widely disparate types of land tenure systems (Table 3.4). On the freehold farms there are clear boundaries, exclusive rights for the individual properties, and commercial farming objectives. These landowners are able to trade with their properties and use their title as collateral security against loans. In contrast, in the communal areas, there are often unclear boundaries, generally open access rights to grazing areas and farmers are subsistence oriented. Here, land tenure issues considerably hamper the introduction and adoption of improved management practices.

TABLE 3.4
A comparison between communal and freehold tenure systems in a similar area (approximately 15 000 ha) of the Peddie district, Eastern Cape, South Africa.

SOURCE: Palmer, Novellie and Lloyd, 1999.

Freehold and commercial sector

The commercial farming sector is well developed, capital-intensive and largely export oriented. Commercial-area livestock production accounts for 75 percent of national agricultural output and comes from 52 percent of the farming and grazing land (Table 3.5). The freehold area in the rural Western Cape, with its associated cropping economy, comprises 53 072 land parcels with an average size of 243 ha. In the Eastern Cape, where land parcels can be regarded as individual ranches or enterprises, there are 37 823 land parcels with an average size of 451 ha. There are approximately 50 000 large-scale commercial farmers, who are predominantly, but not exclusively, drawn from the white population. In 2000, they exported products worth about R 16 billion, or nearly 10 percent of South Africa's total exports.

Cattle are predominant in the eastern parts of the country where the rangelands generally have a higher carrying capacity. Beef-cattle ranching is the largest contributor to commercial farming income and the major breeds are Brahman, Afrikaner and Simmentaler. Sheep are largely concentrated in the drier west, and also in the southeast. Goats are more widely distributed and the main breeds are the Boergoat and the Angora. Grazing livestock are raised under extensive ranching conditions, relying on natural pasture, occasionally supplemented by protein and mineral licks. Ostriches are farmed in the southern parts of the country, and also utilize natural vegetation, supplemented by fodders and concentrates.

The commercial areas are divided into fenced ranches (farms) and then further subdivided into a number of paddocks (camps). Rotational grazing is normally practised. Compared with the communal areas, stocking rates tend to be more conservative and are adjusted by the rancher to track production.

Communal and subsistence sector

The communal areas occupy about 13 percent of the total farming area of South Africa and hold approximately 52 percent of the total cattle population, 72 percent of the goats and 17 percent of the sheep (see Table 3.2). They differ markedly from the freehold areas in their production systems, objectives and property rights; only the cropping areas are normally allocated to individual households, while the grazing areas tend to be shared by members of a community. The communal sector has a substantially higher human population per unit area than the freehold sector, and has suffered from lower levels of state intervention. Investments in infrastructure (access roads, fences, water provision, power supply, dipping facilities) have not kept pace with those in freehold areas, where regional authorities have orchestrated the maintenance of roads and fences. The production systems in the communal areas are based on pastoralism and agropastoralism, and the majority of households are subsistence-based and labour intensive, with limited use of technology and external inputs. The outputs and objectives of livestock ownership are more diverse than in commercial livestock production, and include draught power, milk, dung, meat, cash income and capital storage, as well as socio cultural factors. The combination of objectives tends to be met by a policy of herd maximization rather than turnover; hence even the large herd owners tend to sell only to meet cash needs, leading to higher stocking rates than in the freehold system. The mean land parcel size (612 ha) in the former homelands of Ciskei and Transkei is greater than that of the freehold areas of the Western (243 ha) or Eastern Cape (451 ha), reflecting the free-ranging nature of livestock.

TABLE 3.5
Land areas (million hectares) of the major land use types in South Africa.

NOTE: N/A = not available.
SOURCE: Development Bank of Southern Africa, 1991.

Communal area livestock production contributes 5-6 percent of formal agricultural output and is mainly confined to the eastern and northern part of the country. However, herd sizes vary considerably between and within regions, and livestock ownership is strongly skewed, with a small number of people owning large herds and the majority owning few animals or none at all.

Stock numbers tend to be unevenly distributed across the landscape in communal areas. There is a tendency for high concentrations of people and livestock near to permanent water, while other areas remain potentially underutilized due to a lack of water. In the rugged terrain of Ciskei, Transkei and Kwa-Zulu Natal, livestock spend the longest part of the day on the inter-fluvial ridges. Animal numbers tend to be geared more to the quantity of reliable water than to the reliable quantity of forage, hence drought effects tend to be more severe in communal than in commercial areas.

Mixed livestock ownership is more common in communal than freehold areas. Cattle are the generally preferred livestock species, but economic and ecological conditions often limit the possibilities for cattle ownership. Ownership of livestock is skewed, with 5 percent of residents owning 10 or more cattle in rural villages in the former Ciskei (Ainslie et al., 1997), while 67 percent of households own no cattle. In the case of sheep, 7 percent of households own 10 or more, with 82 percent owning none. For goats, 18 percent of households own 10 or more, while 43 percent own none.

Cattle, sheep and goats are herded during the cropping season in cropping areas, and where there are predator or theft risks in other areas, but herding tends to be relaxed during the dry season, during which animals have access to crop residues. In the northern communal areas, many larger herdowners have “cattle posts” away from the village and crop lands and maintain most of their animals there, keeping only the milk and draught animals at the village during the wet season. Pigs and poultry in the communal areas are generally free ranging and scavenging, although some owners practise housing and feeding.

The exclusion of fire from the savannah regions under communal management has encouraged bush encroachment. In the semi -arid regions of Mpumalanga, the Northern Province and the North West province, fire has generally been excluded. Cutting large trees for fuel or building material has resulted in coppice growth (sprouting) and has stimulated shrubbiness. Consequently, large areas of the medium-rainfall savannahs have become severely bush infested, to the detriment of the grazing potential for cattle and sheep. In the subhumid communal areas of Kwa-Zulu Natal and the Transkei, fire is used to stimulate grass production during the early summer, and this maintains a grassland state along the coastal region (Shackleton, 1991).

Authorities responsible for management

The National Department of Agriculture within the Ministry of Agriculture and Land Affairs is the key institution dealing with forage resources. The National Department of Agriculture is divided into five directorates, one of which deals directly with rangeland and pasture resources. The Land and Resource Management Directorate is responsible for the implementation of the Conservation of Agricultural Resources, Act No. 43 of 1984. This act empowers the head of the Directorate to intervene when the grassland resources of the country are perceived to be threatened by herbivory, alien infestation or cultivation. In addition, each of the nine provinces has a division or directorate that provides research and management advice on rangeland and pasture resources. These sections provide support to extension services and planners, establish standards, develop capacity, and conduct research appropriate to the needs of that province.

Market systems

Marketing of grassland products is conducted through a commodity-based marketing system. Since 1994, the so-called Control Boards of the single-channel marketing system have been disbanded, and a free market system prevails. Each commodity has had to develop its own competitive marketing framework. For example, wool is marketed through numerous brokers, including Cape Mohair & Wool and BKB. Brokers are able to buy direct from the producer and offer the product for sale at auction. Generally, auction prices are determined by the international wool price and local markets have little influence. Negative changes in the exchange rate (Rand against US$) advantages those farmers who produce export-quality wool. In 2000, the greasy wool clip was 52 671 t and South Africa produced 25 percent of Africa's wool crop. Approximately 80 percent of the South African wool crop is processed locally to “tops” level, making it suitable for export to the European market. Beef and mutton marketing has also recently been released from the controlled marketing environment of the previous regime. In 2000, South Africa's mutton production amounted to 114 000 t, most of which was consumed locally.

Landforms and agro-ecological zones

The country has five main physiographic regions, as discussed earlier.

Based on bioclimatic and growth form information, Rutherford and Westfall (1986) defined six biomes in South Africa. An improvement has been suggested by Low and Rebelo (1996), who further subdivided the savannah biome to include the category “Thicket ”, which occurs predominantly in the river valleys of the eastern and southeastern coastal region (Figure 3.6).

Biomes

The areas of the various biomes are given in Table 3.6.

Figure 3.6
The biomes of South Africa.

SOURCE: After Rutherford and Westfall, 1986; Low and Rebelo, 1996.

TABLE 3.6
Area occupied by each of the biomes in South Africa (excluding Lesotho and Swaziland).

Grassland

The grassland biome is situated mainly in the central, high lying regions of South Africa (Figure 3.6) (O'Connor and Bredenkamp, 1997). The biome spans a precipitation gradient from ca. 400 to >1 200 mm/yr, a temperature gradient from frost-free to snow-bound in winter, ranges in altitude from sea level to >3 300 metres and occurs on a spectrum of soil types, from humic clays to poorly structured sands (O'Connor and Bredenkamp, 1997). Although the general structure is fairly uniform, there is a wide range in floristic composition, associated environmental variables, dynamics and management options. There is a strong dominance of hemicryptophytes of the Poaceae. Standing biomass is moisture dependant and decreases with the rainfall gradient. Herbivory from domestic and wild herbivores has a decisive impact on standing biomass and species composition.

The biome was originally defined on climatic factors and is limited to summer and strong summer rainfall areas, with a summer aridity index between 2.0 and 3.9 (Rutherford and Westfall, 1986). Frost is common, occurring for 30-180 days/yr. The most common soil in the biome, accounting for 50 percent of the area, is the red-yellow-grey latosol plinthic catena. This is followed by black and red clays and solonetzic soils, freely drained latosols and black clays (Rutherford and Westfall, 1986).

Acocks (1953, 1988) defined thirteen pure grassland types and six “false” or anthropogenically-induced grasslands, ranging from the so-called “sweet ” grasslands of the semi -arid regions of the Eastern Cape, to the “sour ” grasslands of the high-rainfall regions of the Drakensberg. There are now six recognizable grassland floristic regions (O'Connor and Bredenkamp, 1997), reflecting a topo-moisture gradient from the dry western region to the eastern mountains and escarpment (Table 3.7). Following the completion of a revised vegetation map of South Africa (National Botanical Institute, 2004), sixty-seven grassland units have been described on the basis of floristic and climatic uniqueness, including, for example, the Bedford Dry Grassland (Plate 3.2).

TABLE 3.7
Regions within the grassland biome.

SOURCE: O'Connor and Bredenkamp, 1997.

Plate 3.2
The Bedford Dry Grassland Unit.

A.R. PALMER

The concepts “sweet ” and “sour ” refer to the palatability of the grasses, dwarf shrubs and trees to domestic livestock. Although difficult to define in a strict scientific sense, these terms have retained their use throughout the farming community, being applied to both individual species and to components of the landscape. Sweet-veldt usually occurs on high nutrient status soils under arid and semi -arid conditions. These soils are generally derived from the shales, mudstones and sandstones of the Karoo Supergroup. Sour-veldt is associated with acid soils of quartzite and andesitic origin, and occurs in higher (>600 mm) precipitation and high elevation (>1 400 mm) areas. Ellery, Scholes and Scholes (1995) have suggested that the concept is driven by the C:N ratios of the grasses and that the sweet-veldt has a lower C:N ratio than sour-veldt.

Savannah

Vegetation dynamics in the African savannah are driven by a number of variables, including rainfall amount, rainfall uncertainty, frost, fire, herbivory, ambient CO₂ levels and soil moisture. Depending on the seasonal environmental conditions and management history, a grassland at the boundary of the savannah biome can change from a monolithic physiognomy, to one dominated by shrubs and trees. O'Connor and Bredenkamp (1997) summarize five hypotheses to account for the possible exclusion of woody elements from grasslands. In this dynamic environment, where the grasslands abut the savannah, it is necessary to provide some information on the savannah biome.

Plate 3.3
Arid grasslands of southern Africa occur in southern Namibia and the northwestern portions of South Africa. Dominant genera include Stipagrostis, Eragrostis and Enneapogon.

A.R. PALMER

The savannah biome comprises the northern and eastern portions of South Africa, with the arid savannah extending into the southern Kalahari. The savannah biome is the region where large portions of the national beef production occur under extensive rangeland conditions. The flora comprises a woody layer (mainly single-stemmed, seasonally deciduous, trees and shrubs), with a ground layer of grasses and forbs. The standing biomass of shrubs and trees is 16-20 t/ha (Rutherford, 1982). The dominant grasses are C4 and form the important production component for domestic livestock. A strong summer seasonality in the rainfall encourages woody shrub production. There is strong evidence of woody shrub encroachment throughout this and other biomes (Hoffman and O'Connor, 1999). A number of explanations have been suggested for the increase in woody shrub biomass, including (1) a reduction in fire frequency (Trollope, 1980); (2) the removal of grass biomass by domestic herbivory, with the resultant success of woody shrubs (du Toit, 1967); and (3) the C₃ shrubs having a competitive advantage over C₄ grasses under elevated CO₂ conditions (Bond and van Wilgen, 1996). Graziers attempt to control the woody encroachment using a number of approaches, including clear felling; burning followed by intensive browsing by goats; and chemical control. The last-named seems to be the favoured approach, with an estimated R 10 million spent annually on herbicides. The biome is utilized by both commercial and communal graziers. In general, the woody encroachment problems are more severe in land under communal tenure, although multiple use ensures that wood is used for fuel, construction and traditional purposes.

Nama-karoo

The Nama-karoo biome covers much of the central and western regions of the country. The biome is dominated by a steppe-type vegetation, comprising a mixture of shrubs, dwarf shrubs and annual and perennial grasses. The biome is associated with the moderate rainfall regions (250-450 mm per annum) and is suited to commercial sheep and goat production. The summer seasonality of the rainfall in the eastern parts of the biome means that there is often abundant grass production during the growing season. Graziers attempt to optimize production by sparing or resting grassy dwarf shrubland in the wet season. Herbivory by domestic livestock during the growing season has been shown to reduce grass cover and promote the growth of larger shrubs (species of Rhus, Acacia and Euclea) and dwarf shrubs. In the winter months, the dwarf shrubs maintain their crude protein at around 8 percent, providing excellent forage. The nutrient-rich substrata provided by the mudstones, sandstones and dolerites mean that this production can be considered sustainable. There were earlier suggestions that large-scale structural transformations were taking place in this biome (Acocks, 1964), with the dwarf shrubs supposedly spreading into the adjoining grasslands of central Free State. This process has not continued in the manner envisaged and the relatively high rainfall of the 1990s has promoted grass production in the eastern portions of the biome. In the western portions of the biome, there is alarming evidence of woody encroachment, with two species in particular (Acacia mellifera and Rhigozum trichotomum) increasing in density and cover in regions with a long history of domestic herbivory. Production of mutton and fibre continues to thrive in the Nama-karoo. During the recent past there has also been an increase in the area of land set aside for informal conservation, with many farmers capitalizing on the unique landscapes and indigenous fauna of the biome to develop ecotourism operations. Important indigenous herbivores, which contribute to red meat production, include springbok, blesbok, kudu, gemsbok and wildebeest.

Thicket

The thicket biome occurs in the drainage lines and ridges of the southeastern coastal region and inland to the Great Escarpment. The thicket comprises a dense cover of succulent shrubs, woody shrubs and small trees, with a height of 1.5-3.0 m. The woody shrubs are multistemmed, seasonally-deciduous, C3 plants. In the xeric portions of the thicket biome (300-450 mm/yr precipitation), there is a large component of crassulacean acid metabolism (CAM)-type leaf succulents (e.g. Portulacaria afra), stem succulents (Euphorbia spp.) and many species of small succulent shrubs (e.g. species of Aloe and Crassula). Important grass species that contribute significantly to cattle production include Panicum maximum, P. deustum, Digitaria eriantha and Setaria sphacelata. Although the thicket biome does not contain extensive grasslands, clearing of thicket is carried out by both freehold and communal graziers to promote grasslands. These grasslands provide high quality, year-round grazing, as the shale-derived soils associated with thicket are rich in nutrients.

Two important thicket types are recognized: succulent thicket and mesic thicket. Succulent thicket occurs under semi -arid conditions (300-450 mm/ yr) and is dominated by leaf- and stem-succulent shrubs (Portulacaria afra, Euphorbia bothae, E. ledienii, E. coerulescens, E. tetragona, E. triangularis, numerous Crassula spp. and Aloe spp.). Mesic thicket contains fewer succulents and occurs under higher rainfall (>450 mm/yr) conditions. The cover by woody shrubs is usually continuous, but they seldom exceed 4 m in height.

Short succulent thicket occurs in the low elevation (300-350 m) inland river valleys of the Great Fish, Bushmans, Sundays and Gamtoos rivers. Here the Ecca series mudstones provide a nutrient-rich substratum. The soils are shallow, comprising arid lithosols derived from the basement rock. The valley climate is hot and dry, with extremely high summer maximum temperatures (45°C) and low winter minimum temperatures. In the Great Fish river valley, the structure and composition of the “pristine” form of short succulent thicket is dominated by Portulacaria afra, Rhigozum obovatum and Euphorbia bothae, in small (3-4 m diameter) clumps. The vegetation in the clumps may contain emergant shrubs (e.g. Boscia oleoides), but they are usually <1.5 m in height. The clumps are interspersed by karroid shrubs (Becium burchellianum, Walafrida geniculata and Pentzia incana) and grasses (Cymbopogon plurinodis, Aristida congesta and Eustachys mutica). Further west, Euphorbia bothae is replaced by either E. ledienii (Sundays River) or E. coerulescens (Jansenville and the Noorsveldt).

Degradation of the succulent thicket type takes a number of forms, including a decline in the abundance of Portulacaria afra, which is susceptible to excessive browsing by goats and cattle. There are some alien species present (e.g. Opuntia ficus-indica or O. aurantiaca), but there remains a structure akin to the short succulent thicket. Clumps are still an important feature, but the vegetation has been opened up by livestock. In the severely degraded state, clumps no longer exist. Portulacaria afra and other palatable species (e.g. Rhigozum obovatum and Boscia oleoides) have been removed, exposing the Ecca shales.

The structure and composition of the “pristine” form of the medium succulent thicket comprises Portulacaria afra, Rhigozum obovatum, Ptaeroxylon obliquum and Cussonia spicata in larger (10-12 m diameter) clumps, often associated with the nests or mounds of the wood-eating harvester termite (Microhodotermes viator). The vegetation in the clumps may be up to 2.5 m in height, with inter-clump areas covered by dwarf shrubs (Becium burchellianum, Walafrida geniculata and Pentzia incana) and grasses (Cymbopogon plurinodis, Aristida congesta and Eustachys mutica) with Nama-karoo affinity.

Tall succulent thicket is associated with the steep slopes of the incised river valleys. Emerging tree succulents, which are a distinctive feature of this form, include Euphorbia triangularis and E. tetragona. These shrubs, which can reach 8-10 m in height, are eaten by domestic herbivores only when short. When the endangered black rhinoceros were recently re-introduced to this region, adult animals knocked the tall succulents down to eat the new leaves at the tops of the shrubs.

The mesic thicket type contains far fewer leaf and stem succulents and is composed primarily of multistemmed woody shrubs such as Scutia myrtina, Olea europea var. africana, Rhus longispina, R. incisa and R. undulata. Succulent taxa include numerous species of the genera Aloe, Euphorbia, Gasteria and Crassula, although these growth forms do not represent a major component of the standing biomass.

Succulent karoo

The succulent karoo biome occurs in the winter rainfall regions of the southern and southwestern portions of South Africa. The flora of the biome comprises mainly shrubs (0.5-1.5 m) and dwarf shrubs (<0.5 m), with succulent leaves and stems. The climate of the region is arid to semi -arid (100-350 mm/yr), with a strong winter seasonality. The succulent karoo is well known for its high floristic diversity, part of which is a function of its proximity to the adjacent floristically-rich fynbos biome, but also the unique climatic environment (low, seasonal (winter) rainfall with relatively low uncertainty (coefficient of variation = 30-45 percent) (Cowling and Hilton-Taylor, 1999). Some of the areas with high floristic diversity, such as Richtersveldt and Namaqualand, receive a large portion of their precipitation in the form of coastal advective fog during the coolest months of the year. There are many species in two succulent families (Crassulaceae and Mesembryanthecaceae), numerous endemic taxa (e.g. several species of the genus Pachypodium) and distinctive growth forms (leaf and stem succulents). This diversity has made the biome ideal for the development of an ecotourism industry that promotes the unique floristic character of the region. The arid conditions mean that the region is most suited to extensive livestock production and the flora of the biome has been subjected to herbivory from domestic goats and ostriches. These herbivores form the main suite of animals responsible for much of the direct impact on the vegetation of the biome. In recent times there have been known changes in the species composition, with some landscapes currently dominated by species unpalatable (to domestic livestock), such as Pteronia incana and P. pallens. The productivity of the biome has been significantly affected by these changes and many graziers now depend on irrigated pastures to sustain livestock production.

Plate 3.4
Succulent karoo occurs in the winter-rainfall region of the southern and southwestern portions of South Africa. The vegetation comprises shrubs (0.5-1.5 m) and dwarf shrubs (<0.5 m) with succulent stems and leaves.

A.R. PALMER

Fynbos

The fynbos biome occurs in the winter rainfall regions of the southern and southwestern portions of South Africa, being associated with the moderate to high rainfall (450-1 000 mm) region. The vegetation of the biome is dominated by sclerophyllous shrubs and trees, with rich floristic diversity, but has little or no forage value. Much of the natural vegetation has been cleared to enable wheat, oats, rye, barley, canola and lupin production. The crop residues provide large areas of post-harvesting stubble for sheep production during the dry summer months. Within the biome, irrigated pastures are also major contributors to sheep production.

Forest

Forest occurs in patches along the southern coastal zone, in the cooler southern facing slopes of the Great Escarpment and in the high-rainfall regions of the Drakensberg. Forest is not significant in livestock production in South Africa.

Pastoral and agricultural systems

The main forage resource for livestock in South Africa is rangeland grazing, with 68.6 percent of the area being used for livestock grazing and a further 9.6 percent used by wild herbivores. In the higher-rainfall zones, crop residues are a very important feed supplement in both freehold and communal areas during the dry season, when range grazing is scarce. In freehold areas, many livestock farmers plant fodder species as dryland pasture. Irrigated fodder production is important in freehold areas, but varies from season to season, as cash crops are more favoured. In 1980, 468 000 ha were under irrigated cultivation with alfalfa, but this had declined to 214 000 ha in 1987. In times of drought, South Africa imports maize from the international market. There are some zero-grazing dairy operations near large cities, and three large commercial feedlots are found on the high-veldt.

Veldt grazing

The principal agro-ecological units within South Africa are illustrated in the generalized image of the Acocks (1953) map of the Veldt Types of South Africa (Figure 3.7). Acocks (1953) provided a unique perspective on the classification and distribution of the agro-economic divisions of vegetation in South Africa. This map serves to illustrate the broad floristic diversity of the vegetation and continues to remain an important classification for graziers. There are 70 Veldt Types, with a primary focus on those types most useful for livestock production. As this diversity is reflected in composition, structure, phenology and production, it is extremely difficult to provide broad generalizations concerning the management options for each veldt type, most of which have received some research attention, with the mesic grass -veldt - with its higher production potential and greater economic importance - being given the greatest attention. The research focus has been on supporting government intervention in three major areas.

Figure 3.7
Veldt types of South Africa, as described in Acocks, 1953.

The first area receiving government research support is the estimation of sustainable production (carrying capacity), which was deemed important, as government attempted to enforce restrictions on the numbers of livestock on freehold properties. Grazing trials (mainly on-station), attempted to determine sustainable production levels, using a number of ecological and animal performance indices. Ecological indices that were measured to assess livestock impact on the rangeland included plant species composition, plant vigour and biomass production. In general, on-station trials did not permit the application of extreme treatments that would be appropriate to test the ecosystem. Researchers were reluctant to be perceived to be degrading a state-owned resource and trials were frequently terminated within the time-frames of system run-down. Conclusions for each veldt type vary enormously, but we would like to elaborate on those delivered for the semi -arid grasslands of the Eastern Cape.

It is well recognized that rainfall is the primary determinant of forage production and a number of production model s have been developed for predicting the aboveground primary production in natural rangeland in southern Africa. Coe, Cumming and Phillipson (1976) demonstrate a linear relationship between annual rainfall and primary production for conservation areas in southern Africa. These predictions are regarded as conservative by commercial graziers, many of whom suggest that production for livestock can be optimized by rotational grazing (Danckwerts and Teague, 1989). In an effort to assess the sustainable production of grasslands in the Eastern Cape Province, a grazing trial was established on a freehold ranch (the so-called Kroomie Trial) to test the impact of animal type (cattle or sheep), number (light, moderate or heavy stocking rate) and duration (rotation versus continuous) on rangeland condition and animal production. Preliminary results suggest that continuous grazing under moderate stocking rate (that recommended by the National Department of Agriculture) yields the best livestock mass gain. However, in the Kroomie trial, no significant changes in species composition are obvious and the duration of the trial (10 years) is insufficient to make conclusive assertions regarding system run-down. Even in situations like this, where the questions have been clearly defined and the treatments meticulously applied, no clear answers to “sustainable” production levels are available. Using SPUR2 (Wight and Skiles, 1987; Hanson et al 1994), Palmer, Ainslie and Hoffman (1999) simulated a 50-year beef operation under continuous grazing for a site receiving 500 mm/yr (similar in elevation and rainfall to the conditions at Kroomie). The “recommended stocking rate”, determined by the National Department of Agriculture and Land Affairs, was 6 ha/LSU. When running another simulation at 4 ha/LSU, with ambient CO₂ at 330 ppm, the system remained sustainable (Figure 3.8). Only when the stocking rate was increased dramatically, doubled to 2 ha/LSU, did the system run down within the 50-year simulation period (Figure 3.9). These results suggest that the recommended stocking rates for grassland systems are well below those that are likely to lead to system run-down. In the communal areas of the Eastern Cape, livestock numbers at the district level reflect the fact that stocking rates are substantially higher than those applied by freehold graziers and recommended by the Department of Agriculture. This presents a problem for administrators, who are unable to reduce livestock numbers in areas where graziers are unresponsive to regulation.

Figure 3.8
Simulated carrying capacity for Grahamstown using SPUR2, with stocking rate set at that recommended by the Department of Agriculture (4 ha/LSU) under a continuous grazing system.

Figure 3.9
Simulated carrying capacity for Grahamstown using SPUR2, with stocking rate set at 2 ha/LSU under a continuous grazing system. System shows signs of run-down after 300 months.

Figure 3.10
Rangeland production in South Africa using the model of Le Houérou, Bingham and Skerbek (1988) and median annual rainfall (Dent, Lynch and Schulze, 1987).

Production relationships can be simplified to straightforward expressions of kilograms of annual dry matter production of forage per millimetre of annual rainfall (Le Houérou, 1984). An aboveground biomass production model based on the concept of rain-use-efficiency has been developed (Palmer, 1998) and applied to rangeland. The resultant map for commercial production is shown as Figure 3.10. Production may be converted to carrying capacity for cattle by assuming a daily requirement of 11.25 kg DM/LSU and a use factor of 0.4 (Le Houérou, pers. comm.). The use factor may decline to 0.2 in mesic, “sour ” grasslands with high C:N ratios.

The second focus of research to receive substantial government funding in support of intervention policies was assessment of grazing systems. During 1950-1990 it was expedient for government to provide support for fencing, water points and stock management infrastructure. Field trials were designed to assess the advantages of rotational versus continuous grazing. Rotational grazing requires that the pasture allocated to a group or groups of animals be subdivided into one enclosure more than the number of groups (Booysen, 1967). According to Tainton, Aucamp and Danckwerts (1999), the primary objectives of rotational grazing are to:

• control the frequency at which plants are grazed by controlling the frequency with which each camp in the system is grazed;

• control the intensity at which plants are grazed by controlling the number of animals that graze each camp and their period of occupation; and

• reduce the extent to which veldt is selectively grazed by confining a relatively large number of animals to a small proportion of the veldt so as to offer them little opportunity to select.

The published results of many grazing trials (Department of Agriculture [South Africa], 1951) suggest that while animal performance in continuous grazing systems was superior to that of various rotational grazing ones, continuous grazing was condemned. After the contribution of Booysen (1969), who maintained that retaining sufficient active green biomass was essential to optimize regrowth, further research continued into the advantages of variations in rotational grazing. The concepts of Booysen (1969) are encompassed in the term High Performance Grazing (HPG), with the alternative, intensive-use, approach being High Utilization Grazing (HUG) (Tainton, Aucamp and Danckwerts, 1999). The HPG approach is thought to perform better in the semi -arid grasslands and savannahs, whereas HUG is more appropriate in the fire -induced grasslands of the humid regions. The inappropriate application of HUG, encompassed by the protagonists of Holistic Resource Management (Savory, 1988), to the more fragile, semi-arid systems, has been controversial and is discouraged.

The third area receiving research attention funded by the Department of Agriculture was veldt condition assessment. In this work, floristic composition was regarded as an indicator of the impact of management and stocking rates. Gradient studies, which explored changes in species composition along grazing gradients, were popular. Within this research area, it was difficult to attribute the floristic variation along gradients directly to herbivory. Differences in species composition were often a consequence of enrichment, trampling and associated changes in soil structure and chemistry.

Legume and fodder introduction

A number of subtropical pasture legumes and fodders have been screened at sites with from 100-700 mm annual rainfall. Range reinforcement is done on a large scale in commercial dairy regions. Favoured grasses include Pennisetum clandestinum (Kikuyu grass), Panicum maximum and Digitaria eriantha, while legumes such as silver leaf Desmodium are over-sown into natural vegetation.

Foggage (Plate 3.5) is important in commercial beef and dairy production systems in South Africa. Graziers use a wide range of commercially available grasses and legumes. The performance of growing beef steers grazing foggaged dryland Kikuyu grass pastures and given limited access (3 hours daily) to leucaena (Leucaena leucocephala cv. Cunningham) was better than that of steers grazing only Kikuyu foggage during autumn and early winter (Zacharias, Clayton and Tainton, 1991). Animals grazing leucaena performed better and gained 24.8 kg per animal more, over 90 days, than those on Kikuyu alone. There is concern about the risk of leucaena becoming invasive in the humid coast and its use, and that of other potentially aggressive species (e.g. Lespedeza sericea), has been discouraged until further evaluation has been carried out.

Plate 3.5
Foggage Kikuyu for winter grazing.

S.G. REYNOLDS

Investigations to determine whether frosted Kikuyu provides better quality foggage than natural pasturage in the sour-veldt area during the winter months revealed that this grass had a crude protein content of 8-10 percent in winter. The performance of animals grazing such frosted Kikuyu was highly satisfactory (Rethman and Gouws, 1973). Sheep performance and patterns of herbage utilization were determined in two grazing trials involving different amounts and quality of Kikuyu foggage. In two grazing trials involving different quantity and quality of Kikuyu foggage, wether lambs maintained live mass in one, whereas dry ewes and wether lambs both lost 8-10 percent of their initial mass in the other. This suggests that Kikuyu foggage alone does not provide a viable source of fodder. Grazing capacity was proportional to the yield of foggage and some 50 percent of the total herbage was utilized. The estimates of quality indicated that a higher level of utilization would have resulted in poorer sheep performance (Barnes and Dempsey, 1993).

Dryland fodder

Dryland fodder production is only possible in the higher-rainfall regions of the country. The principal form of dryland fodder is cereal crop residues, which make an important contribution to livestock diets in communal areas during the dry season. Some communal area farmers collect and store at least part of their residues to feed to selected animals, such as milch cows and draught oxen, but most is utilized in situ.

Cultivation of rainfed crops in South Africa is widespread in both freehold and communal land use systems. The most significant commercial grain producing areas are the maize triangle of the central high-veldt, the wheat growing region of the southwestern Cape and the maize growing regions of central Kwa-Zulu Natal. Maize is widely preferred as the staple food in the communal areas, but millet and sorghum are more reliable crops apart from in the highest-rainfall zones. National cereal production (roughly 80 percent maize, 16 percent wheat and 4 percent others, including millet and sorghum) fluctuates considerably from year to year according to rainfall. Production has varied from a low of 5 044 000 t in the drought year of 1991/92 to a record high of 15 966 000 t in 1993/94 (Table 3.8). It is difficult to assess what proportion crop residues contribute to national production as no research has been published, but it is thought to be considerable in areas of commercial rainfed cultivation (>600 mm mean annual rainfall).

In drier central and western zones, farmers commonly have small areas of drought -tolerant fodders (e.g. Agave americana, Opuntia spp. or Atriplex nummalaria) as a drought reserve.

Irrigated fodder

Irrigation has two main functions in the humid summer rainfall regions. In winter it is used for temperate pasture species such as ryegrass and in summer it is used to supplement rainfall. In winter, the temperate species are completely dependant upon irrigation for survival and it can only be justified in intensive production systems such as dairy or the production of fat lambs.

TABLE 3.8
Commercial field crop production for South Africa from 1992 to 2000 (×1000 t).

SOURCE: FAO database.

Plate 3.6
Dairy cows on irrigated ryegrass pastures (Lolium multiflorum) near Fort Nottingham, KwaZulu-Natal.

Dryland pasture production may be improved by irrigation. Where irrigation is available, despite the relative advantages of using water for other more lucrative crops, some farmers may choose to grow irrigated pasture. Alfalfa (Medicago sativa) is the main purpose-grown irrigated fodder, and is grown throughout the country. In 1980, 468 000 ha were under irrigated alfalfa, but this declined to 214 000 ha in 1987. This may show a preference amongst farmers with water rights to grow cash crops. New legislation (Water Act of 1998), which separates water rights from property rights and increases the cost of abstracting water from rivers, will reinforce this trend. In high-performance production systems (e.g. dairy - Plate 3.6), Kikuyu grass (Pennisetum clandestinum), cocksfoot (Dactylis glomerata), tall fescue (Festuca arundinacea) and ryegrasses (Lolium multiflorum and L. perenne) are cultivated. Some other legumes (Trifolium pratense and T. repens) respond well to irrigation. Many other species and numerous cultivars are available commercially (Bartholomew, 2000).

Exceptional circumstances fodder

In times of drought, South Africa has provided fodder at subsidized rates to farmers. According to the new drought policy (National Department of Agriculture, 1995), the fodder subsidies have been terminated in order to encourage farmers to build up their own forage reserves and to discourage them from retaining excessive stock numbers. Nonetheless, it is likely that some commercial farmers, and probably the government, will continue to import fodder in extreme drought conditions. In the arid and semi -arid regions, farmers are encouraged to plant suitable drought-tolerant fodder crops (Table 3.9). Since 1994, there have been no magisterial districts declared as drought stricken, and so these new policies have not been tested.

TABLE 3.9
Plants used for fodder during exceptional circumstances.

Constraints to pasture and fodder production and improvement

Low and uncertain rainfall throughout most of the country are the main constraints to the productivity of natural pastures and to the establishment of exotic pasture crops. Concern about exotics becoming problematic limits the introduction and testing of hardy species considered suited to the environmental and utilization rigours of the communal areas (e.g. Leucaena spp. and Lespedeza sericea). The availability and price of seeds for fodder or for pasture improvement are major constraints to communal area farmers. Considerable portions of the savannah vegetation on the freehold farms are severely bush infested, but the cost of thinning or clearing generally outweighs the benefits in terms of increased carrying capacity. Open access to grazing, at least within communities, in the communal areas necessitates broad collective agreement and cooperation in any pasture improvement venture - something most communities, socially fragmented as they are, seem unableto attain. Traditionally, communal area farmers do not retain exclusive use of their unfenced croplands for their own livestock after harvest, which blocks opportunities and incentives for undersowing or alley cropping.

Commercial ranchers find it increasingly difficult to maintain production in the western and central regions, which receive low and uncertain rainfall and where there are increases in undesirable woody species. Low profit margins and higher production costs discourage many landowners from maintaining commercial herds. There has been a decline in sheep and wool production from the Nama-karoo region (Dean and MacDonald, 1994), which has been attributed to a decline in resource condition. There appears to be an increase in the number of uninhabited freehold farms in the arid and semi -arid regions, suggesting that farms are being abandoned or managed as larger units. Reflecting de-agrariani-zation trends throughout the developed world, South Africa n rangelands under freehold tenure are becoming depopulated. The children of freehold farmers do not regard farming as an exciting career option, and leave the farm for training in more lucrative career paths. When freehold land is converted from livestock ranching to game farming, the staff complement required to manage the farm is reduced and labour is encouraged to move to the smaller towns, with better education, health and municipal services.

In contrast to freehold land, the population in communal areas has increased since the initiation of re-location policies by the previous regime. This trend has continued since the advent of the new democracy, with rural land being used for the construction of dwellings, roads, clinics, schools and stores. Many residents have access to a piece of cultivated dryland, either close to their dwelling or at an allocation some distance away, but within the administrative region of the village. This allocation is generally used for maize, millet or cash crops, and is seldom planted to pasture crops. Crop residues may be available to livestock at the end of the harvest. Access to irrigation is limited to a very few villagers, who are usually part of government schemes.

Evolution of grasslands over the last 40 years

There is clear evidence of changes in the structure and specific composition of grasslands in southern Africa in the recent past. In a comprehensive review of the impact of recent human occupation of the eastern seaboard, Hoffman (1997) reports that “crop farmers first entered southern Africa along the northeastern coastal margins”, where they practised slash-and-burn agriculture and small-stock farming. At first, only the vegetation around the coastal forest margins was cleared. These early farmers moved westwards, clearing cropland in woodland and forest, creating a mosaic of open grassland and thicket patches. The expansion of grassland brought about a shift in herd composition and there was an increase in cattle -based economy. In the communal rangelands of KwaZulu Natal, Transkei and Ciskei, grasslands are maintained by the continued removal of shrubs and trees for firewood, annual burning (Plate 3.7) and the use of goats to control woody encroachment. However, the encroachment of woody plants into grassland remains a constant threat elsewhere. During a project to re-photograph, from the same viewpoint, historical photographs taken in natural rangeland, M.T. Hoffman of the University of Cape Town (pers. comm.) has shown that woody encroachment is a feature of almost every photograph that has been re-taken. Published examples of this are available in Hoffman (1997) and Hoffman and O'Connor (1999). Urbanization and cultivation has played an important role in transforming grasslands. Rural villages and abandoned cultivated land have replaced natural grassland in the former homelands of Transkei, Ciskei, Qwa-Qwa, Venda, Lebowa, Bophutatswana and Kwa-Zulu. In rural villages, free-ranging livestock use the entire landscape without restriction, concentrating nutrients around the homesteads and kraal where they are held overnight. The areas near homesteads have a lower standing biomass, but are extremely photosynthetically active and provide short, nutritious grazing during the growing season. However photosynthetic activity in other components of the landscape and in abandoned cultivated land is low, suggesting suboptimal production of natural rangeland (Palmer et al., 2001).

Plate 3.7
Grass regrowth after the annual burn in Kwa-Zulu Natal.

Research

Range and botanical sciences in South Africa have a very active research community, with funding and leadership in a number of ministries. The primary research agency for rangeland is the Agricultural Research Council (ARC), which supports two institutes dealing with understanding the processes (Range and Forage Institute - ARC-RFI) and condition assessment (Institute for Soil Climate and Water - ARC-ISCW) of rangeland. These institutes receive most of their core funding (approximately R 45 million in 2001) from the Ministry of Arts, Culture, Science and Technology. They also receive direct, project-orientated funding from the National Department of Agriculture and Land Affairs. Research direction is driven largely by the needs of directorates within the Department, which at present has five primary programmes: (1) Monitoring; (2) Problem organisms; (3) Rangeland resources; (4) Geographical Information Systems (GIS); and (5) Decision Support Systems (DSS). In the monitoring programme, research efforts of ARC-RFI and ARC-ISCW are directed towards resource evaluation using remote sensing and GIS modelling techniques. There is a strong emphasis on using satellite imagery to explore the extent of and trends in soil erosion, bush encroachment and rangeland degradation. The directorate has supported the calibration of NOAA AVHRR data for use in assessing trends since the launch of these satellites in 1980, and is now funding research into using high resolution infrared instruments (digital cameras and other high resolution sensors) to assess range and landscape degradation at the farm and village scale. The problem organism division continues to show interest in brown locusts, quelea control, woody weed encroachment (especially alien taxa such as Prosopis spp.) and alien weed control. The list of alien weeds is extensive, with some 197 taxa being listed as declared weeds and invader plants (Henderson, 2001) and a further 60 taxa being considered as proposed weeds (Henderson, 2001) and a threat to range and water resources. The rangeland resources division funds projects that assess the impact of different grazing management approaches (e.g. continuous versus rotational grazing) on rangeland. This is being carried out in a network of replicated grazing trials throughout the country.

The Grassland Society of Southern Africa (GSSA) is the professional organization representing the discipline. GSSA maintains a full-time secretariat for its members, organizes annual congresses at localities around the subcontinent and publishes a peer-reviewed journal (African Journal of Range and Forage Science). The journal has been published since 1966, and about a thousand peer-reviewed articles have appeared.

Botanical research relating to rangeland is conducted by the National Botanical Institute (NBI) of the Ministry of Environmental Affairs and Tourism. NBI has focused on exploring the natural patterns and processes driving vegetation status in the arid and semi -arid regions of the Nama-karoo and succulent karoo biomes.

Management of grasslands

In response to a decline in profit margins and negative sentiments associated with domestic livestock production, there has been a marked increase in game farming and ecotourism on commercial ranching areas. This is manifested in the large numbers of game-proof fences erected on farm boundaries and the removal of internal fences and stock watering points. This change has an impact on the management of rangeland, as livestock can no longer be manipulated and it is more difficult to apply rotational grazing. In commercial farming operations, fire is used on many high-elevation rangelands to provide grazing during the early growing season. Fire is used primarily by commercial ranchers to remove low quality material remaining after winter and to encourage a flush of short green grass in spring (see Plate 3.7).

Development of techniques for the rehabilitation of grasslands

The processes of degradation in arid and semi -arid rangelands are poorly understood and only recently have some researchers provided a conceptual framework for exploring rehabilitation within the context of an understanding of landscape function. Ludwig et al. (1997) suggest that the landscape comprises a series of inter-connected patches, with resource control being a key feature in the maintenance of the integrity of the landscape. Nutrients and moisture move from one patch to another, mainly though water flow patterns. The obstructions or patches in the landscape prevent nutrients from being lost, with run-on areas acting as sites of nutrient and moisture accumulation. Run-on areas are connected to one another by runoff zones or fetches. In a degrading landscape, it is suggested that these runoff areas increase in size and the run-on areas can no longer capture and accumulate nutrients, and so nutrients are lost to rivers and transported away from the landscape. In order to quantify these processes and relate them to South Africa n landscapes, landscape function analysis (LFA) (Ludwig et al., 1997) was applied to two contrasting land use types (Palmer et al., 2001). The results showed that landscape with a long history of communal management has surface accumulations of C and N, which may not have been processed efficiently. Although Palmer et al. (2001) did not investigate nutrient loss, there were significant differences between land-scape-scale organizations, with communal landscape having a lower landscape functionality index than the freehold grassland. There was greater patchiness in the communal landscape, with longer fetches than the landscape with a long history of commercial management. In accordance with Ludwig et al. (1997), it is recommended that rehabilitation of degraded rangelands in South Africa should strive to reduce the extent of runoff zones and increase the resource control exercised by patches.

Spatially explicit diversity indices (moving standard deviation index) have been applied to near-infrared (NIR) imagery (Landsat TM and SPOT) (Tanser and Palmer, 1999) recorded over land with different management histories and condition classes. Degraded grasslands, located in areas with a long history of communal management, had higher spatial diversity of selected growth indices than healthy grassland. Grasslands, savannahs and thicket with a high standing biomass and a long history of conservative land management, showed low spatial diversity indices. Rehabilitation techniques should attempt to reduce landscape photosynthetic heterogeneity.

Over-seeding with commercially available seeds has long been regarded as a solution to rehabilitation of degraded rangeland. In the thicket biome, re-veg-etation of former cultivated lands has been successful when lime-coated seed (a mixture of seven local species, including Panicum maximum, Cenchrus ciliaris and Eragrostis curvula) was broadcast over the cultivated land and a long (three-year) rest applied. However, there has been limited success reported elsewhere, where commercial seeds generally require irrigation after planting and any early effect is usually discounted within a few years.

Using the principles embodied in the LFA theory, Van Rooyen (2000) has shown that it is possible to rehabilitate degraded biospheres in the southern Kalahari using brush packing with Rhigozum trichotomum. In this highly mobile sand-dune environment, brush packing results in sand stabilization and enables seedling establishment.

Sustainable management of the environment and maintenance of biodiversity

South Africa has ratified the United Nations Conventions that strive to maintain bio diversity and improve sustainable management of rangelands, specifically the Convention on Biological Diversity (CBD) and the United Nations Convention to Combat Desertification (UNCCD). These conventions are administered by the Ministry of Environmental Affairs and Tourism, which supports the assessment of the state of resources. The most recent product of this programme has been an assessment of degradation and desertification at a national scale (Hoffman and Ashwell, 2001), which defines the nature and extent of rangeland transformation.

Seed production

There is formal certification of pasture and fodder seed in South Africa. South African seed merchants produce approximately 200 t of seed per annum for sale locally and for export. With the long-term goal of preserving germplasm (in most cases, as seed) of the entire South African flora, the ARC-Plant Genetic Resources Division in Pretoria focuses at present on preservation of seeds of plant species of economic importance. A wide variety of South African pasture grasses are included in the current accessions, such as species of the genera Anthephora, Brachiaria, Cenchrus, Cynodon, Panicum, Pennisetum, Setaria and Stipagrostis.

Recommendations and lessons learned concerning sustainable grassland management

Many of the wild herbivores in South Africa (e.g. blesbok, bontebok, black wildebeest, blue wildebeest, springbok, Burchell's zebra and red hartebeest) create enriched patches which provide highly nutritious fodder throughout the year (Palmer, Novellie and Lloyd, 1999). These patches form a relatively small portion of the landscape (Lutge, Hardy and Hatch, 1996) and researchers (Booysen, 1969) recognized that the remainder of the rangeland was not used. It was suggested that in order to prevent this “area-selective grazing ” in livestock production systems, ranchers should fence numerous paddocks and rotate the domestic animals in a system that maximized the use of the aboveground primary production. This principle was built into the legislation to permit government intervention in the primarily white -owned ranches. This intervention, which provided subsidies to farmers for fencing, dam construction and the erection of water points, encouraged ranchers to develop their farms. In the process, the government provided the rancher with the tools to efficiently remove large quantities of aboveground standing biomass. This approach may have been useful in the grasslands where research had been carried out to show that grass species composition could be changed by applying rotational grazing. However, this has had some very serious consequences in the thicket biome, where succulent shrubs such as Portulacaria afra have been completely removed from the landscape and do not regenerate after severe defoliation (Stuart-Hill and Aucamp, 1993). In the savannah biome, there is clear evidence of continuing woody encroachment by a wide range of shrub species, including Acacia karroo, A. mellifera, Dichrostachys cinerea, Rhus undulata and Rhigozum trichotomum. Researchers suggest that this is largely due to the removal of competitive grass species by herbivory, although alternative hypotheses are suggested.

Maintenance of production and productivity

Following on the thinking of J.C. Smuts, who spear-headed rangeland conservation initiatives in South Africa, Davies (1968) suggested that a holistic approach is needed whereby plant, soil and animal influences are studied as controllable parts of the environment. Savory (1988) further embellished on this approach, coining the term Holistic Resource Management, which embodied principles that were contrary to the conventional thinking amongst range scientists (Clayton and Hurt, 1998). Savory (1988) encouraged ranchers to employ non-selective grazing by maintaining large, mixed species - herds that would intensively use a restricted area until the standing biomass had been severely reduced, thereby eliminating competition among species. This area would then receive a long rest, whereupon herbivores would return. Numerous commercial graziers in South Africa have been encouraged to follow this thinking, with no formal scientific debate being entertained on the success of the approach. Vorster (1999) provides a convincing argument for discouraging ranchers from pursuing this approach without careful assessment of the rangeland resource.

Strategies for maintaining and optimizing aboveground grass production include rotational grazing and resting; control of woody encroachment; provision of winter pastures in the cool, mesic regions where forage quality declines in winter; and supplementary feeding on cultivated lands. In beef production systems, livestock are finished in feedlots where maize provides the major feed element. In these systems, production costs are determined by the international maize price.

Priorities for the development of programmes and research

The National Department of Agriculture, in cooperation with ARC-RFI, ARC-ISCW and CSIRO-Environmentek, is the key institution dealing with forage resources. The Directorate of Land and Resource Management in the National Department of Agriculture and Land Affairs is responsible for the implementation of the Conservation of Agricultural Resources, Act No. 43 of 1984. This act empowers the Directorate to intervene when the agricultural resources of the country are threatened by soil erosion, alien infestation or woody encroachment. Prior to 1994, this act was used to subsidize the provision of fencing; the erection of new water provision points; the purchase and transport of supplementary fodder during exceptional circumstances; the construction of soil erosion works; the clearing of all weeds (alien and indigenous); and to support rangeland research. Since 1994, intervention from the Directorate has concentrated on supporting research in the focus areas mentioned above and to intervene at community level though the Landcare programme. In addition, each of the nine Provinces has a section that deals with rangeland and pasture research. These sections conduct research appropriate to the needs of that Province. South Africa's National Agricultural Policy states that the main objective is improvement of research in natural resource management (National Department of Agriculture, 1995). On a project basis, pasture science-related programmes deal with the characterization of rangeland, production modelling, rangeland reclamation, agroforestry and rangeland management systems. Examples of individual projects related to rangeland and pasture science can be found on the ARC Web site.

The National Department of Education maintains eleven agricultural colleges, and carries out topic-oriented, formal training courses. All courses are certified by one of the tertiary training institutions.

References

Acocks, J.P.H. 1953. Veld types of South Africa. Memoir of the Botanical Survey of South Africa, No. 28.

Acocks, J.P.H. 1964. Karoo vegetation in relation to the development of deserts. pp. 100-112, in: D.H.S. Davis (ed). Ecological Studies of Southern Africa. The Hague, The Netherlands: W. Junk.

Acocks, J.P.H. 1988. Veld types of South Africa. 3rd edition. Memoirs of the Botanical Survey of South Africa, 57: 1-146.

Ainslie, A., Cinderby, S., Petse, T., Ntshona, Z., Bradley, P.N., Deshingkar, P. & Fakir, S. 1997. Rural livelihoods and local level natural resource management in Peddie district. Stockholm Environment Institute Technical Report.

Barnes, D.L. & Dempsey, C.P. 1993. Grazing trials with sheep on kikuyu (Pennisetum clandestinum Chiov.) foggage in the eastern Transvaal high-veld. African Journal of Range and Forage Science, 10: 66-71.

Bartholomew, P.E. 2000. Establishment of pastures. In: N.M. Tainton (ed). Pasture Management in South Africa. Pietermaritzburg, South Africa: Natal University Press.

Bond, W.J. & van Wilgen, B. 1996. Fire and Plants. London, UK: Chapman and Hall.

Booysen, P. DeV. 1967. Grazing and grazing management terminology in southern Africa. Proceedings of the Grassland Society of Southern Africa, 2: 45-57

Booysen, P. DeV. 1969. An evaluation of the fundamentals of grazing management systems. Proceedings of the Grassland Society of Southern Africa, 4: 84-91

Coe, M.J., Cumming, D.M. & Phillipson, J. 1976. Biomass production of large African herbivores in relation to rainfall and primary production. Oecologia, 22: 341-354.

Cowling, R.M. & Hilton-Taylor, C. 1997. Plant biogeography, endemism and diversity. pp. 42-56, in: Dean and Milton, 1999, q.v.

Cowling, R.M., Richardson, D.M. & Pierce, S.M. (eds). 1997. Vegetation of South Africa. Cambridge, UK: Cambridge University Press.

Danckwerts, J.E. & Teague, W.R. 1989. Veld management in the Eastern Cape. Government Printer, Pretoria, South Africa.

Davies, W. 1968. Pasture problems in South Africa. Proceedings of the Grassland Society of Southern Africa, 3: 135-140.

Dean, W.R.J. & Macdonald, I.A.W. 1994. Historical changes in stocking rates of domestic livestock as a measure of semi -arid and arid rangeland degradation in the Cape Province, South Africa. Journal of Arid Environments, 26: 281-298.

Dean, W.R.J. & Milton, S.J. (eds). 1999. The karoo. Ecological patterns and processes. Cambridge, UK: Cambridge University Press.

Dent, M., Lynch, S.D. & Schulze, R.E. 1987. Mapping mean annual and other rainfall statistics over southern Africa. Water Research Commission., Pretoria, South Africa. Report No. 109/1/89

Department of Agriculture [South Africa]. 1951. Pasture research in South Africa. Progress Report. Division of Agricultural Education & Extension, Pretoria, South Africa.

Development Bank of Southern Africa. 1991. Annual Report. DBSA, Midrand, South Africa.

Du Toit, P.F. 1967. Bush encroachment with specific reference to Acacia karroo encroachment. Proceedings of the Grassland Society of Southern Africa, 2: 119-126.

Ellery, W.N., Scholes, R.J. & Scholes, M.C. 1995. The distribution of sweet-veld and sourveld in South Africa's grassland biome in relation to environmental factors. African Journal of Range and Forage Science, 12: 38-45.

Ellis, J.E. & Swift, D.M. 1988. Stability of African pastoral ecosystems: alternate paradigms and implications for development. Journal of Range Management, 41: 450-459.

FAO. 1973. Soil map of the world. Paris, France: UNESCO.

FAO. 2001. Data downloaded from FAOSTAT, the FAO online statistical database. FAO Rome, Italy. See: http://faostat.fao.org

Grossman, D., Holden, P.L. & Collinson, R.F.H. 1999. Veld management on the game ranch. In: N.M. Tainton (ed). Veld Management in South Africa. Pietermaritzburg, South Africa: University of Natal Press.

Hanson, J.D., Baker, B.B. & Bourdon, R.M. 1994. SPUR2. Documentation and user guide. GPSR [Great Plains System Research] Technical Report, No. 1. Fort Collins, Colorado, USA.

Henderson, L. 2001. Alien weeds and invasive plants. Plant Protection Research Institute Handbook, No. 12. Plant Protection Research Institute, Pretoria, South Africa.

Hoffman, M.T. 1997. Human impacts on vegetation. pp. 507-534, in: Cowling, Richardson & Pierce, 1997. q.v.

Hoffman, M.T. & O'Connor, T.G. 1999. Vegetation change over 40 years in the Weene/Muden area, KwaZulu-Natal: evidence from photo-panoramas. African Journal of Range & Forage Science, 16: 71-88.

Hoffman, M.T. & Ashwell, A. 2001. Nature divided. Land degradation in South Africa. Cape Town, South Africa: University of Cape Town Press.

Le Houérou, H.N. 1984. Rain use efficiency: a unifying concept in land use ecology. Journal of Arid Environments, 7: 213-247.

Le Houérou, H.N., Bingham, R.L. & Skerbek, W. 1988. Relationship between the variability of primary production and variability of annual precipitation in world arid lands. Journal of Arid Environments, 15: 1-18.

Low, A.B. & Rebelo, A.G. 1996. Vegetation of South Africa, Lesotho and Swaziland. Pretoria, South Africa: Department of Environmental Affairs and Tourism.

Ludwig, J., Tongway, D., Freudenberger, D., Noble, J. & Hodgkinson, K. 1997. Landscape Ecology. Function and Management. Principles from Australia's Rangelands. CSIRO, Canberra, Australia. 158 p.

Lutge, B.U., Hardy, M.B. & Hatch, G.P. 1996. Plant and sward response to patch grazing in the Highland Sourveld. African Journal of Range and Forage Science, 13: 94-99.

Maclean, G.L. 1993. Roberts' Birds of southern Africa. 6th ed. Cape Town, South Africa: John Voelcker Bird Book Fund.

National Botanical Institute. 2004. Vegetation Map of South Africa, Lesotho and Swaziland. Beta Version 4.0. National Botanical Institute, Cape Town, South Africa.

National Department of Agriculture. 1995. White Paper on Agriculture. Government Printer, Pretoria, South Africa.

National Department of Agriculture and Land Affairs. 1999. Annual Report.

National Department of Agriculture and Land Affairs. 2001. Annual Report.

O'Connor, T.G. & Bredenkamp, G.J. 1997. Grassland. In: Cowling, Richardson & Pierce, 1997, q.v.

Palmer, A.R. 1998. Grazing Capacity Information System (GCIS). Instruction Manual. ARC-Range & Forage Institute, Grahamstown, South Africa.

Palmer, A.R., Ainslie, A.M. & Hoffman, M.T. 1999. Sustainability of commercial and communal rangeland systems in southern Africa. pp. 1020-1022, in: Proceedings of the 6th International Rangeland Congress. 17-23 July 1999, Townsville, Australia.

Palmer, A.R., Novellie, P.A. & Lloyd, J.W. 1999. Community patterns and dynamics. pp. 208-223, in: Dean & Milton, 1999. q.v.

Palmer, A.R., Killer, F.J., Avis, A.M. & Tongway, D. 2001. Defining function in rangelands of the Peddie District, Eastern Cape, using Landscape Function Analysis. African Journal of Range & Forage Science, 18: 53-58.

Partridge, T.C. 1997. Evolution of landscapes. pp. 5-20, In: Cowling, Richardson & Pierce, 1997, q.v.

Rethman, N.F.G. & Gouws, C.I. 1973. Foggage value of Kikuyu (Pennisetum clandestinum Hochst, ex Chiov.). Proceedings of the Grassland Society of Southern Africa, 8: 101-105.

Rutherford, M.C. 1982. Above-ground biomass categories of the woody plants in the Burkea africana-Ochna pulchra savanna. Bothalia, 14: 131-138.

Rutherford, M.C. & Westfall, R.H. 1986. The Biomes of Southern Africa - an objective categorization. Memoirs of the Botanical Survey of South Africa, 54: 1-98.

Savory, A. 1988. Holistic Resource Management. Covelo, California, USA: Island Press.

Shackleton, C.M. 1991. Seasonal changes in above-ground standing crop in three coastal grassland communities in Transkei. Journal of the Grassland Society of Southern Africa, 8: 22-28.

Schulze, R.E. 1997. South Africa n Atlas of Agrohydrology and -climatology. Water Research Commission, Pretoria, South Africa. Report TT82/96.

Smithers, R.H.N. 1983. The mammals of the southern African subregion. Pretoria, South Africa: University of Pretoria.

Stats SA. 1996. The People of South Africa. Population Census 1996. National Census. South African Statistical Services.

Stuart-Hill, G.C. & Aucamp, A.J. 1993. Carrying capacity of the succulent valley bush-veld of the eastern Cape. African Journal of Range and Forage Science, 10: 1-10.

Tainton, N.M. (ed). 1999. Veld management in South Africa. University of Natal Press, Pietermaritzburg, South Africa. 472pp.

Tainton, N.M. (ed). 2000. Pasture management in South Africa. University of Natal Press, Pietermaritzburg, South Africa. 355pp.

Tainton, N.M., Aucamp A. J. & Danckwerts J. E. 1999. Principles of managing veld. Grazing programmes. pp. 169-180, in: Tainton, N.M. (ed). Veld management in South Africa. Pietermaritzburg, South Africa: University of Natal Press,.

Tanser, F.C. & Palmer, A.R. 1999. The application of a remotely-sensed diversity index to monitor degradation patterns in a semi -arid, heterogeneous, South African landscape. Journal of Arid Environments, 43(4): 477-484.

Thackeray, A.I., Deacon, J., Hall, S., Humphreys, A.J.B., Morris, A.G., Malherbe, V.C. & Catchpole, R.M. 1990. The early history of southern Africa to AD 1500. Cape Town, South Africa: The South African Archaeological Society.

Trollope, W.S.W. 1980. Controlling bush encroachment with fire in the savanna areas of South Africa. Proceedings of the Grassland Society of Southern Africa, 15: 173-177.

Wight, J.R. & Skiles, J.W. (eds). 1987. SPUR: Simulation of production and utilization of rangelands. Documentation and user guide. US Department of Agriculture, Agricultural Research Service, ARS 63. 372 p.

Van Rooyen, A.F. 2000. Rangeland degradation in the southern Kalahari. Unpublished PhD thesis. Department of Range and Forage Science, University of Natal, Pietermaritzburg, South Africa.

Volman, T.P. 1984. Early prehistory of southern Africa. pp. 169-220, in: R.G. Klein (ed). Southern African Prehistory and Palaeoenvironments. Rotterdam, The Netherlands: AA Balkema.

Vorster, M. 1999. Veld condition assessment. In: Tainton, 1999. q.v.

Zacharias, P.J.K., Clayton, J. & Tainton, N.M. 1991. Leucaena leucocephala as a quality supplement to Pennisetum clandestinum foggage: A preliminary study. Journal of the Grassland Society of Southern Africa, 8: 59-62.