Environment

7.11 Woodlands project - good progress being made with phytoremediation project

Good progress is being made with the Mine Woodlands and Sustainable Vegetation of Slimes Dam projects being conducted by the University of the Witwatersrand, Johannesburg (Wits) and AngloGold Ashanti South Africa. This research programme, for which the initial work began in 1996, combines ecological engineering with a phytoremediation approach to reduce environmental impact and liability.

AngloGold Ashanti has championed this programme since its inception, and between 1996 and 2005, contributed a total of R9.1 million ($1.42 million) for slimes dam slope reduction, the planting of trials, infrastructure and R&D. (See box on project partners).

The programme is overseen by Wits, under the leadership of Isabel Weiersbye of the School of Animal, Plant and Environmental Sciences. According to Weiersbye, "This is one of the largest R&D programmes into gold mine pollution containment ever undertaken in terms of the extent of affected land and water to be rehabilitated, the diversity of landscapes and species that need to be taken into account, and the requirement for sustainable livelihoods across an entire region after mining."

Background

South Africa produces around 450 million tonnes of waste annually, with 70% of this generated by the mining industry. Gold mines on the Witwatersrand Basin alone produce 105 million tonnes per annum (23% of the total) with about 200,000 tonnes of waste generated for every tonne of gold produced. Much of this waste is deposited into tailings dams, of which there are more than 270 on the Witwatersrand Basin, covering some 400 km². These dams are all unlined and many are unvegetated, and can be a source of extensive dust, as well as soil and water pollution.

Environmental planning of waste disposal by the South African gold mining industry in the past, although legal and recommended by government at the time, has since been proven environmentally unsound. Among common practice was the location of unlined tailings dams on natural pans, wetlands, water courses and catchment areas, and the disposal of mine process water into pans and unlined evaporation dams. Consequently there has been poor containment of pollution. Research has shown that erosion from steep slopes of tailings dams in South Africa can reach 500 tonnes per hectare annually, compared to losses of 15 tonnes per hectare annually from agricultural land¹. In addition, around 400Ml of water passes through the major South African gold mines every day: Of this, 260Ml goes into groundwater and 130Ml to surface water². Approximately 600,000 tonnes of salts are discharged into the Vaal and Orange rivers each year by the gold mines³. It has been estimated that there is more than 6,000 km² of polluted soil in the Witwatersrand Basin, a consequence of gold mining, and more than 30,000km² of land overlying polluted groundwater⁴.

The production of mining waste on such a large-scale waste has serious consequences for the environment. It causes dysfunctional hydrology, as well as acidification and salinisation of soils, groundwater and surface water bodies, resulting in breakdowns in nutrient cycling and environmental degradation. This can lead to losses in biodiversity and ecosystem services, and, therefore, both tailings and contaminated water can be expected to eventually contribute to negative health impacts in humans if mitigation measures are not put in place.

Added to this is the historical exploitation of South Africa's indigenous woodlands to meet demands associated with mining in the last century. By 1917, 250,000 tonnes per year of indigenous timber was being consumed by the Rand gold mines alone. The exotic forestry industry in South Africa now meets the demands of the mining and other industries, and this helped to prevent the destruction of the remnant indigenous timber, albeit at the cost of extensive afforestation of virgin lands. Currently, AngloGold Ashanti requires about 70,000 tonnes of timber per year, which is approximately 2% of the total South African timber market.

Given the improved state of environmental knowledge, and the changes in legislative emphasis (see legislative background), there is now potential for significant long-term liability on the part of land owners and users. The AngloGold Ashanti-Wits programme is pioneering methods to prevent pollution, remediate polluted soils and water, and convert tailings dams to safe and sustainable land uses.

History of programme

The current programme began in 1996 when Anglo American's Gold and Uranium Division initiated research with Professor Ed Witkowski and Weiersbye at Wits to determine the feasibility of an ecological engineering and phytoremediation approach to the rehabilitation of tailings dams. The project's aims were to determine which indigenous plants and micro-organisms were naturally colonising tailings dams and polluted soils, and then to develop 'designer' ecosystems from this naturally selected flora to remediate polluted soils, rehabilitate tailings and even produce novel crops on tailings dams.

The mining industry's approach to reducing erosion and dust from tailings dams largely involves the planting and irrigation of pasture grasses. Pasture grassing costs on average R90,000 per hectare for slopes and R30,000 per hectare for tops, utilises up to three times annual rainfall for irrigation, and prevents surface erosion (i.e. dust control) for less than 10 years. Although grassing prevents dust emissions for some years, it is rarely effective in containing erosion in the longer term, and does not prevent seepage and water pollution, or achieve ecological sustainability^5,6. Since the pasture-grassing approach has proven unsustainable for mine closure purposes, there is a need to determine what type of vegetation would prevent pollution emissions from tailings dams, and so lead to rehabilitation.

Surveys of tailings dams and polluted soils were carried out between 1996 and 2002, to characterise biological and chemical status in relation to region, climate, geochemistry and vegetation practices. Constraints to current vegetation practices on tailings dams were identified, as well as the practices and flora that have a higher probability of contributing to sustainable rehabilitation. Key findings of the project were that over a century of pollution from gold mining has resulted in the development of a tolerant flora, and that pasture grasses are not the most suitable form of vegetation⁷. In addition, the steep slope angle of tailings dams was found to be more detrimental to vegetation persistence than the actual chemistry of slimes. Most naturally-colonising plant species are indigenous small bushes, shrubs and even trees, and dependent on symbioses with soil microbes such as mycorrhizal fungi and nitrogen-fixing bacteria ^8,9. There was a need to lower tailings dam slopes, reduce dependency on irrigation and fertilisers, increase biodiversity and use vegetation structure to provide better erosion control and prevent seepage.

Tolerant plants and micro-organisms were isolated and screened, and bulk production commenced. Slimes dam trials were initiated in 1998 with the grassing of a tailings dam slope that had been reduced from 30 to 18°, and the establishment of 'pseudo'-savanna, tree-trials and refuse trials on the top of tailings dams at Welkom.

By 2005, the studies had demonstrated that:

• reducing the slopes of tailings dams from about 30º to less than 16º reduces the cost of grassing, significantly improves the longevity of grassing and erosion control, and significantly reduces irrigation requirements;
• the use of compost, domestic kitchen waste, sewage sludge, garden refuse, clays, gravels and various rubbles on the tops of tailings dams instead of the traditional lime, fertilizer and irrigation approach prevents erosion, kick-starts nutrient-cycling and results in vegetation establishment at a cost of less than R2,000 hectare;
• the effects of acid mine drainage and polluted groundwater on plant growth and reproduction extend well beyond tailings dams, resulting in strong selection pressures and the development of tolerant plant 'land races' and micro-organisms ¹⁰;
• there is DNA evidence for the evolution of genetically distinct and tolerant varieties of some plants on old tailings-polluted soils¹¹;
• the growth in acidic tailings of tolerant plants from polluted soils is significantly better than that of plants from normal soils;
• a range of indigenous species (trees, shrubs, forbs and grasses) from polluted sites were found to be extremely tolerant of growth in acid and saline slimes, and could maintain productivity and viable seed production equivalent to the same species in unpolluted veld ^12,13;
• the use of 'designer' combinations of tolerant plants, together with nitrogen-fixing bacteria and arbuscular mycorrhizal fungi, also isolated from slimes-polluted soils, greatly improves the condition of vegetation on tailings dams ^14,15; and
• some indigenous plant species have high commercial potential and low radiological risk when grown on slimes dams .

Between 2001 and mid-2004, approximately 30,000 indigenous trees and shrubs were planted on the top of a tailing dam at Welkom. These trials, and a tailings dam that has been wooded since the 1960s, are being assessed for rehabilitation processes, the impact of vegetation on slimes dam water balances, and economic potential. In 2005, planting will start on the top of AngloGold Ashanti's Old North Tailings dam at West Wits operations. The findings of the project have also been incorporated into AngloGold Ashanti's closure plan for the Brakpan tailings dam (at 560 hectares, the largest gold tailings dam in the world).

The Mine Woodlands Research Project (2001-2008) was initiated to assess the feasibility of encapsulating tailings dams within a system of woodlands, to prevent water-borne pollution and dust emissions ^16,17. The project also assesses the ability of other vegetation types (wetlands, grasslands, savannas and riparian woodlands) to naturally attenuate pollution.

The three primary objectives of the Mine Woodlands Project are to:

• prevent pollution at the source by planting vegetation on the tops of tailings dams that will evaporate all incoming rain water, and abstract water from within the slimes dam itself;
• decontaminate and rehabilitate polluted soil and groundwater, and foster ecosystem processes; and
• provide a sustainable solution, (i.e. an environmentally, socially and economically acceptable solution), to the problem of tailings dam waste through the use of vegetation. This includes the transfer of technology to small businesses, such as community-based nurseries and planters. (See case study: Community nurseries - based solution.)

Trees and shrubs have several advantages over grasses for the purposes of pollution control. Many trees are evergreen, in contrast to grasses, which are dormant in winter,. The more extensive root systems of trees can therefore abstract seepage all year-round, and from greater depths. The fine root and leaf litter of trees is more effective than that of grasses in fostering the formation of top soil, and fine roots can also take-up or immobilise some pollutants. Woodlands can remove organics, nitrates, phosphates, sulphates, various heavy metals and radionuclides from soil and groundwater, and internationally, phytoremediation is gaining acceptance as the technology of choice for landfills, mining waste and contaminated land.

The planning of land use on mines is essential in order to foster existing ecosystem services, and ensure that tree planting is strategic. Thus the role of existing natural vegetation in pollutant control is being characterised at AngloGold Ashanti's Vaal River and West Wits operations, and on the Free State mines. Reedbeds, wetlands, grasslands, savanna, indigenous woodlands and exotic plantations are being assessed ¹⁸. Networks of trees are also being used in combination with remote sensing to identify the extent, history and sources of groundwater contamination - the chemistry of tree leaves and annual growth rings in wood is a reflection of groundwater quality. Optimisation of the remote sensing techniques for tree canopies will allow for the more precise siting of pollution control measures at vastly reduced cost, and have international applications¹⁹.

Click on image to enlarge (90KB)

Between 2002 and 2008, a series of woodland trials are being planted on and around two tailings dam complexes at AngloGold Ashanti's Vaal River and West Wits operations. A few hundred indigenous tree and shrub species are being assessed, as well as 12 exotic tree species, provided permission for the latter is granted by government. In the 2004/5 season, 67,000 trees were planted on polluted soils and seepage from tailings dams at the West Wits operations and almost 250,000 at the Vaal River operations. The aims of the trials are, to determine which are the most suitable tree species and soil micro-organisms to use on each site, which are best at decontaminating and rehabilitating soils and groundwater, how much water the trees use, where the best place to plant them would be, how the tree plantings will affect the local ecology, geohydrology and chemistry, how to optimise tree growth for both pollution control and economic return from timber, biofuels, medicinals and other chemicals, and whether such woodlands can be sustainable on tailings dams and polluted soils, and thus contribute to job creation.

In the intensive 'site-species' matching trials, approximately 30 tree species and land-races are planted in a 5 hectare area each year. The aim of these trials is to optimise the technology for local conditions, and determine which trees are more suited to each set of conditions in terms of:

• the size and growth rate of the tree - a fast growing tree will consume more water, whereas a large canopy will create a dust windbreak and the leaves will generate compost;
• the amount of water a tree consumes in relation to pollutants, identify which species immobilise or remove the most pollutants for the least amount of water consumed;
• which species are good growers, easy to maintain and regenerate naturally. Such a selection would remediate the soil and produce compost and help to establish ecosystem services;
• which species and woodland products are radiologically safe, i.e. do not contain radionuclides or heavy metals, as these could enter the food chain. In some cases, species that are found to take up radionuclides and metals are being tested for their ability to 'mine', or clean, polluted soils and water. Litter from these 'accumulating' plants would be disposed of safely in a hazardous waste disposal site; and
• which species have commercial potential when grown on slimes dams and polluted lands, i.e. from timber, biofuels, pharmaceutical chemicals, fibres and other natural products. Secondary industries could then be set up to harvest, process and market 'rehabilitation plant products'.

The impacts of large blocks of trees on local geohydrology and ecosystem processes, are assessed in the parallel woodland 'block' trials, where large areas of the best-performing trees are being planted on polluted soil and seepage at the West Wits and Vaal River operations.

The planting and care of woodlands is labour intensive. Between 2000 and February 2005, approximately 300 people have been trained in site preparation, and the planting and care of trees. In 2003, Wits entered into a partnership with DWAF's Directorate of Participatory Forestry. The aim of the partnership is to empower community nurseries so that they can acquire the niche market for 'designer' plants for gold mine rehabilitation. Technology transfer to community-based nurseries started in 2003 and 2004. AngloGold Ashanti is fostering the nurseries through business skills, and the provision of guaranteed orders for plants.

As results of this study are also to be used as an approach for mine closure, every effort is made to ensure that the technology being developed is in line with government requirements. South Africa is a water-poor country with strict environmental legislation. The disturbance of virgin land, changes in land-use, and the planting of exotic trees, or any trees for commercial purposes or water abstraction, is subject to government legislation. The programme therefore focuses mainly on the use of indigenous trees, and plantings on historically disturbed or polluted soils and tailings dams, with government approvals being sought for the trial planting of any exotic trees. Scientific workshops for government and mining are held regularly, and in 2004, Government formed the Woodlands Authorisation Forum with AngloGold Ashanti and Wits to ensure the solutions being developed are in accordance with current and planned legislation.