S27.2: Collision of first world and third world conservation problems: The IBA example

Avian Demography Unit, University of Cape Town, Rondebosch 7701, Cape Town, South Africa, fax 27 21 650 3726, e-mail kbarnes@maths.uct.ac.za

For over 100000 years, nomadic subsistence dwellers and Bantu-speaking pastoralists and cultivators were the only inhabitants of South Africa (Maylam 1986). Apart from the introduction of moderate levels of grazing and agriculture, their impacts on their environment were negligible and wildlife continued to thrive. The discovery of a sea route to India via the Cape of Good Hope by Portuguese sailors in the late 15th century brought Europeans to the region for the first time (Worden 1994). Change was rapid over the next 500 years. Post industrial-revolution development proceeded rapidly, leading to widespread alteration of the natural environment. The last two hundred years have brought dramatic and intense changes in land-use and resource utilisation.

Apartheid policies have exacerbated inequalities and compounded the environmental problems facing contemporary South Africa (Davenport 1991). The disparity and juxtaposition of first and third world environmental issues facing the country are directly attributable to the controversial homelands policy of the Apartheid government, which forced over 67% of the population onto 14% of the land (Central Statistics Office 1994 pers. comm.). These areas experienced intense subsistence agriculture and pastoralism, resulting in overgrazing and trampling, soil erosion, vegetation loss and unsustainable farming practices. Meanwhile, in adjacent industrialised areas, pressing environmental issues included large-scale habitat destruction for massive stands of monospecific cropland, urbanisation, industrial pollution, river and wetland degradation and acid rain from coal power stations. Rapid population growth and a huge demand for job creation, fresh water, food, raw materials and land further compound these problems.

Very few nations are posed with such a diversity and complexity of environmental issues. This paper reviews avifaunal conservation in South Africa, with particular reference to BirdLife International’s Important Bird Area (IBA) Programme, and poses the questions: Does the protected areas network achieve the goal of adequately protecting South Africa’s avifauna? If it does not, which areas require additional conservation action? Given the diversity of environmental problems facing the country, how can the IBA programme work towards effectively protecting biodiversity in the New South Africa?

The function of the Important Bird Areas (IBA) programme is to identify and protect a network of sites, at a biogeographic scale, critical for the long-term viability of naturally occurring bird populations, across the range of those bird species for which a site-based approach is appropriate.

Purely numerical analyses could not be used for IBA site selection for two primary reasons: 1) complete inventories of bird species are unavailable for most sites within South Africa, including most protected areas and 2) where basic inventories are available, virtually no information on the status of viable populations at each site exists. Instead, emphasis was placed on criteria that were regarded as scientifically measurable but easily and widely applicable.

Sites were selected according to four categories and their associated criteria defined by BirdLife International (Fishpool 1996). It was recommended that they are followed as closely as possible but, because definitions of this sort cannot cover all possibilities, they are not inflexible rules. The need for scientific objectivity and standardisation was balanced by common sense and the practical objectives of the exercise. Each site and its avifauna were judged on merit and an even-handed approach was essential for the credibility of the programme.

A site could be designated as an IBA if it satisfied any of the criteria for any of the following categories.

This category refers to species classified as globally or nationally threatened with extinction, conservation dependent or data deficient according to the new IUCN criteria for threatened status (sensu Collar et al. 1994; Barnes in press). The last two categories of species, although not strictly threatened, are considered to be of sufficient conservation concern to merit the identification of Important Bird Areas.

In general, the regular presence of a critical or endangered species, irrespective of population size, at a site was regarded as sufficient to propose the site as an IBA. For vulnerable and near-threatened species, the site needed to hold more than a defined threshold to qualify; minimum figures for vulnerable species were 10 pairs/30 individuals for passerines and non-passerines and for near-threatened species, 10 pairs/30 individuals for non-passerines and 30 pairs/90 individuals for passerines. These figures may not be appropriate for all species in these threat categories and common sense dominated in the application of this criterion. Provision is also made for the inclusion in this category of species of near-threatened (NT) status, as defined and listed in Birds to Watch 2 (BTW2) Collar et al. 1994).

This category is for species of Endemic Bird Areas (EBAs). EBAs are defined as places where two or more species of restricted-range, i.e. with world distributions of less than 50000 km², occur together (Stattersfield et al. 1998). Also included here are species of Secondary Areas. A Secondary Area (SA) supports one or more restricted-range species, but does not qualify as an EBA because less than two species are entirely confined to it (Stattersfield et al. 1998). Typical SAs include single restricted-range species which do not overlap in distribution with any other such species, and places where there are widely disjunct records of one or more restricted-range species, which are clearly geographically separate from any of the EBAs.

For many EBAs, which hold a large number of restricted-range species, it was necessary that a network of sites were chosen, to protect adequately all relevant species. In cases where data on bird distribution within the EBA were insufficient, the network will ensure that sufficient key habitat types are included.

This category applies to groups of species with largely shared distributions of greater than 50000 km², which occur mostly or wholly within all or part of a particular biome and are, therefore, of global importance (Fishpool 1996). Many of these assemblages occur in places where delimiting IBAs is particularly difficult.

A biome may be defined as a major regional ecological community characterised by distinctive life forms and principal plant species. No system of global biome classification has been found which can be adequately used as a basis for generating bird species lists. This has necessitated a regional approach and resulted in inter-regional differences between the classification systems employed but, as far as possible, the overall scale - the ‘depth’ of treatment - at which biome divisions are recognised is comparable.

More than one habitat type, and therefore bird community, often occurs within a given biome and this is reflected by the set of sites identified. For most biomes, a few large sites were selected that reflect the distribution of the biome across the country, rather than many small ones confined to only a part of it. This ensured that a greater number of species were represented per site and took into account their geographical distribution. The existing protected area network was a pragmatic starting point for choosing sites under this criterion.

This category applies to those species that are (perceived to be) vulnerable by congregation at valuable or sensitive sites when breeding or wintering or while on passage. The term waterbird is used here in the same sense as the Ramsar Convention uses waterfowl and covers the list of families as more precisely defined by Wetlands International (WI), formerly the International Waterfowl and Wetlands Research Bureau (Rose & Scott 1994). Congregatory non-waterbird species (A4iia) includes both terrestrial species and those families of seabird not covered by Rose & Scott (1994) including Spheniscidae, Diomedeidae, Procellaridae, Hydrobatidae, Pelecanoididae, Phaethontidae, Sulidae, Fregatidae, Chionididae and Stercoraridae. The threshold for criterion (A4ia) is 1% of the biogeographic population of a congregatory waterbird species. Relevant flyway populations are combined to produce biogeographic population estimates. WI’s ‘Biogeographic’ populations are approximately the equivalent of BirdLife’s IBA regions. There are however differences in some areas due to the geopolitical basis of the IBA coverage. This results in some complications in setting thresholds, as does the way some flyway populations are shared between regions in either breeding or wintering ranges. Threshold figures are set for all congregatory waterbird species, including those for which none is currently recognised by Ramsar. WI have collaborated in generating numeric thresholds from range estimates and from unpublished population data.

Once the network of IBAs had been selected, each IBA was designated as protected, partially protected or unprotected. This analysis was performed at a provincial scale, to determine the geographical distribution of the unprotected IBAs in South Africa - those most in need of conservation attention. Based on the primary habitat(s) present at the unprotected sites, they were lumped into one of the following general habitat categories: fynbos, Karoo, grassland (terrestrial or vlei), forest, wetland, savanna / woodland and other. The total and average area of each of the unprotected IBA categories was also calculated.

IBAs are distributed very unevenly among South Africa’s provinces. Together, two of South Africa’s nine provinces, KwaZulu/Natal and the Western Cape, hold nearly 50% of the country’s sites (Table 1). This is not surprising, however, as these two provinces also hold an immense diversity of habitats, and a concomitantly large proportion of the species that were used for site selection.

The IBA network should be considered as the minimum essential set of sites that would ensure the survival of their component species, across their ranges, should remaining habitat elsewhere be lost through anthropogenic, or other, modification (Fishpool 1996). The existing protected area network was a pragmatic starting point for selecting IBAs, and protected areas were chosen preferentially to unprotected areas where they fulfilled similar IBA roles. It was thus surprising that within the IBA network, 25% of the sites were shown to receive no protection whatsoever, and a further 24% were only partially protected (Table 1). Alarming proportions of these unprotected sites are found on private and unzoned government land. Given the number of sites in the IBA network lacking legislative protection, how adequate is the protected areas network for bird conservation?

Siegfried (1989) concluded that, in general, the South African mainland avifauna appeared to be relatively well represented within protected areas. All but 18 of c. 600 indigenous breeding species were recorded in one or more of the region’s publicly owned nature reserves. Although Siegfried (1989) conceded that this was a somewhat unexpected and satisfactory result, one major anomaly should be noted. Due to their enhanced dispersal ability, birds are likely to show much greater incidences of vagrancy or occasional occurrence within the reserve network.

As a consequence, the proportion of birds that have viable populations within South Africa's protected areas is likely to be much lower then the 97% of the species Siegfried (1989) recorded within it. Not surprisingly, Siegfried (1992) later considered most of South Africa's endemic birds to be inadequately represented within the network (populations <1000 individuals).

Most of the unprotected IBAs lie in Mpumalanga, Free State and the Northern Cape (Table 1). The conservation of the majority of these critical sites remains the responsibility of private landowners within these provinces, suggesting that South Africa’s state-owned protected areas network is poorly developed. Similar conclusions have been reached for other taxa. Cowling & Hilton-Taylor (1994) concluded that the protected area network is not optimally located with regard to the region’s hot spots of plant diversity and endemism.

Closer analysis of the physical nature of the unprotected IBAs shows that 15 sites (>50%) are in the grasslands, and seven of these hold seeps, bogs, marshes or other wetlands within the grassland matrix (Table 2). This suggests that the highest priority conservation areas for birds (and possibly other taxa) in South Africa lie in the grasslands, particularly in the sensitive wetland systems.

The Atlas of Southern African Birds (Harrison et al. 1997), has also indicated that the species endemic to the grasslands are, collectively, the most threatened avifaunal group in South Africa and that some (e.g. Rudd’s Lark Heteromirafra ruddi, Botha’s Lark Spizocorys fringillaris and Yellow-breasted Pipit Hemimacronyx chloris) face imminent extinction if degradation and destruction of natural grasslands continues at the present rate (Allan et al. 1997; Stattersfield et al. 1998). These species typically show disrupted patterns of distribution. Their ranges do not correspond to the former extent of the grasslands and also tend to be fragmented (e.g. White-bellied Korhaan Eupodotis cafra); these species now persist as several isolated meta-populations, probably with little gene flow (Allan 1997). The plight of the grassland birds highlights the urgent need for implementation of conservation policies and management action for their benefit (Tarboton 1997a, b). What is the potential for reserve establishment?

In 1997, South Africa had some 741 publicly owned protected areas, covering 74 956 km², accounting for 6.14% of the country’s land surface area (Avian Demography Unit unpublished data). Some 42 189 km², a disproportionate 64% of this, is in the savanna. There are many significant gaps in the existing protected area network. It is undoubted that the grassveld, Succulent Karoo, Nama-Karoo and lowland fynbos (particularly renosterveld) vegetation types are inadequately protected in South Africa (Table 3).

Nearly a decade ago, Siegfried (1989) suggested that additional land be acquired for state-owned reserves in the grassland, Karoo and lowland fynbos biomes. Despite these recommendations, the last decade has seen very little land added to the formal conservation network (Table 3). The situation has improved marginally in the Succulent Karoo, where the proportion of conserved land has been elevated from 0.79% to 2.82% in the last eight years (Table 3). The conservation status remains unchanged for other vegetation types, differing minimally from the pattern presented historically (e.g. Greyling & Huntley 1984; Siegfried 1989) and it is alarming that still only 2.52% of the grasslands, 2.82% of the Succulent Karoo and 0.57% of the Nama-Karoo are currently protected in conservation areas. Moreover, most of the existing protected areas in the inadequately protected vegetation types are small, often isolated from one another, and separated by large tracts of mostly transformed land (Siegfried 1989). Aggravating this situation is the fact that protected areas have been managed as islands of biodiversity rather then as part of a holistic land-use policy. Also of concern is that the existing network has arisen through a largely ad hoc process, rather than being part of a deliberate conservation strategy.

Given that the establishment of new state-owned protected areas is unlikely in the near future, given current priorities for spending state money, it seems appropriate to examine the apparent threats to those habitats presently under-represented in the protected area network, and to seek strategies for the long-term conservation of their ecosystem functioning and biota. An examination of the South African grasslands biome follows.

South Africa is experiencing two major environmental problems that are characteristic of third world countries. An explosive population growth rate (2.2% p.a.) and a concomitant urbanisation phenomenon, which has still to manifest its full impact on our society and environment (Central Statistics Office 1998, pers. comm.). Several major urban centres are located within the grassland biome, including the Gauteng megalopolis, Bloemfontein and Pietermaritzburg. This biome, which encompasses less than 25% of the country, will hold more than 60% of its population in 22 years time. It is estimated that by 2020, South Africa's population will have increased by a factor of 1.65; during the same time period the population of Gauteng will triple, to reach some 30.5 million people (Central Statistics Office 1998 pers. comm.).

The pressures of combating environmental problems in urban environments, such as water pollution, air pollution, soil erosion and increasing quantities of sewage and urban and toxic waste, will increase severely. Failure to take cognisance of these will result in the deterioration of the South African environment beyond recovery, to the stage where it cannot sustain a vigorous economy, nor provide its population with any semblance of a quality-lifestyle.

Simultaneously, the grasslands are burdened by intense, first world environmental pressures such as commercial afforestation, which is probably the most critical threat (Armstrong et al. 1994). It is likely that afforestation will dramatically alter vast areas of grassland habitat in the next 20 years (Stattersfield 1998). In addition, the concentration of large numbers of trees has profound consequences for the hydrology of the area, with the drying up of vleis, streams and wetlands being a common result (Taylor 1994).

Most of the areas identified as suitable for commercial afforestation in South Africa lie within the grasslands and, more particularly, within the high-rainfall Sour grasslands (Tarboton 1997a, b). There is conclusive evidence that commercial afforestation has already had a major impact on grassland birds, and the potential for further negative impacts on endemic and threatened species is serious (Allan et al. 1997).

The cumulative effects of habitat fragmentation, removing patches of land from pastoral use and changing burning regimes impact bird diversity in a far greater way than just transforming areas of grassland to trees (Allan et al. 1997).

Although cultivated crops such as wheat and maize are also responsible for habitat destruction, they do not transform the grassland system in a comparable fashion. Many of the threatened grassland species are able to survive in crop growing areas, especially where crops are grown in tandem with pastoral farming and where areas of natural veld are maintained for grazing (Tarboton 1997b).

Although afforestation is the overriding threat, several others compound the conservation problems the grasslands face. The biome also holds over 75% of the country’s coal generated power stations. The associated acid rain deposits some million tonnes of toxic sulphur dioxide on the surrounding crops, grasslands and wetlands each year (Tarboton 1997b). Many of the remaining grasslands are subject to overgrazing and uncontrolled burning, or are being destroyed by open-cast coal mining.

Despite the severity and multiple sources of threat, it is unlikely that grassland reserves will be created in the short term. Should any money become available, grassland wetlands are the highest priority candidates for reserve creation, for the following reasons. They are among the most threatened systems in the country and are of extreme importance for biodiversity conservation. They are vital for water purification and supply to agriculture and urban centres, and they are small discrete systems that occupy minute areas compared to other IBAs (Table 2).

Custodianship of species and habitats that remain unprotected in reserves rests in the hands of private landowners. Providing incentives for landowners to manage parts of the grassland on their farms for particular species or communities of birds can be a very effective conservation strategy (Tarboton 1997b). For many threatened species, effective conservation is not about establishing reserves for them, but about ensuring that deleterious land-use practices affecting them are minimised or prevented. Much of South Africa’s remaining natural grassland is farmland used for stock production. Under natural conditions burning and grazing would have driven the functioning of these grasslands. Although domestic stock has replaced wild ungulates, and humans control the fire regime, the processes remain unchanged (Tarboton 1997b). Grassland birds can and do continue to survive and, in many species, to thrive, where extensive stock farming is practised (Tarboton 1997a). The mosaic created by varying stocking rates and hence grazing pressure and varying burning regimes provides areas of suitable habitat for a wide spectrum of species.

To prevent habitat loss from grasslands does not necessarily require adding land to the existing reserve network, but it does require the continued practice of extensive pastoral farming in the grasslands. Growing timber offers a greater short-term financial yield than the pastoral economy (Allan et al. 1997). It is this that threatens the future of the grasslands. It should be recognised in South Africa, as it has in Europe, that pastoral farming practices are crucial to the survival of much of the continent’s remaining wildlife. If Africa is to maintain its staggering levels of biodiversity, in situ management of populations in multipurpose matrices is essential. Biodiversity conservation must exist alongside other activities, such as the production of potable water, crop or livestock farming, urbanisation and even traditionally destructive practices such as mining.

In order to conserve species on private land, research into their habitat requirements is an essential first step in towards holistic conservation of species, habitats and landscapes. Long-term, effective conservation will be reliant on effective research and properly implemented management on private land.

This work was supported by BirdLife International and WWF - South Africa, to whom I am particularly grateful. Leon Bennun commented on a draft manuscript and many other people have been instrumental in the success of the IBA programme, and although they are too numerous to list, I extend my utmost gratitude for their input.

Allan, D.G. 1997. Whitebellied Korhaan Eupodotis cafra. In: Harrison, J.A., Allan, D.G., Underhill, L.G., Herremans, M., Tree, A.J., Parker, V. & Brown, C.J. (eds) The atlas of southern African birds: Vol. 1: Non-passerines. Johannesburg; BirdLife South Africa: 356-357.

Allan, D.G., Harrison, J.A., Navarro, R.A., van Wilgen, B.W. & Thompson, M.W. 1997. The impact of commercial afforestation on bird populations in Mpumalanga province, South Africa - Insights from bird-atlas data. Biological Conservation 79: 173 - 185.

Armstrong, A.J., van Hensbergen, H.J. & Geertsems, H. 1994. Evaluation of afforestable montane grasslands for wildlife conservation in the north-eastern Cape, South Africa, Part 1. Methods. South African Forestry Journal 171: 7-20.

Barnes, K.N. (ed.) in press. Important Bird Areas of southern Africa. Johannesburg; BirdLife South Africa.

Collar, N.J., Crosby, M.J. & Stattersfield, A.J. 1994. Birds to watch 2. The world list of threatened birds. Cambridge; BirdLife International.

Cowling R.M. & Hilton-Taylor, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. Strelitzia 1:31-52.

Davenport, R. 1991. South Africa - a modern history. London; MacMillan

Fishpool, L.D.C. 1996. Important Bird Areas in Africa: IBA Criteria - Categories and thresholds. Cambridge; BirdLife International.

Geldenhuys, C.J. & MacDevette, D.R. 1989. Conservation status of coastal and montane evergreen forest. In: Huntley, B.J. (ed.) Biotic Diversity in Southern Africa: Concepts and Conservation. Cape Town; Oxford University Press.

Greyling, T. & Huntley, B.J. 1984. Directory of southern African conservation areas. South African National Scientific Programmes Report 98: pp 309.

Harrison, J.A., Allan, D.G., Underhill, L.G., Herremans, M., Tree, A.J., Parker, V. & Brown, C.J. (eds). 1997. The atlas of southern African birds: Vol. 1 & 2: Johannesburg; BirdLife South Africa.

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

Maylam, P. 1986. A history of the African people of South Africa: From the Early Iron Age to the 1970s. Cape Town; David Phillip.

Rose, P. M. & Scott, D. A. 1994. Waterfowl Population Estimates. Slimbridge, U.K.; International Waterfowl and Wetlands Research Bureau. IWRB Special Publication 29.

Siegfried, W.R. 1989. Preservation of species in southern African nature reserves. In: Huntley, B.J. (ed.) Biotic Diversity in Southern Africa: Concepts and Conservation. Cape Town; Oxford University Press.

Siegfried, W.R. 1992. Conservation status of the South African endemic avifauna. South African Journal of Wildlife Research 22: 61-64.

Stattersfield, A.J., Crosby, M.J., Long, A.J. & Wege, D.C. 1998. Endemic Bird Areas of the World - Priorities for Biodiversity Conservation. Cambridge; BirdLife International.

Taylor, B.P. 1994. The biology, ecology and conservation of four flufftail species, Sarothrura (Aves:Rallidae). PhD Thesis. University of Natal, Pietermaritzburg, South Africa.

Tarboton. W.R. 1997a. Whither grasslands? Africa – Birds and Birding 2(2): 49-53.

Tarboton. W.R. 1997b. Grasslands – The way forward. Africa – Birds and Birding 2(3): 41-44.

Worden, N. 1994. The making of modern South Africa. Cape Town; Juta.