There is growing public concern about the availability of fresh water in the world. Demand for such water is projected to outstrip supply by a staggering 40% by 2030, and an estimated half of the world’s population is likely to live in areas of high water stress by the same year (CDP Water Disclosure South Africa Report 2011). Consequently, impacts on water increasingly present risks to business from supply and demand management, changing regulatory regimes, licence to operate and reputational damage from misuse, or perceived misuse, of  this shared, life-sustaining resource.


Water is required to mine, concentrate, smelt and refine our base and precious metals. If access to water is restricted and if we have a negative impact on water resources, then our ability to produce is directly affected.

For us, the threat of water scarcity is very real, given that more than 90% of our operations are located in South Africa, a country that is water-stressed. However, sufficient water has been secured to ensure the continuation of our business.

Water strategy

The Anglo American water strategy and policy coincide with our aim to demonstrate leadership in water management within the areas in which we operate. This strategy is premised on being a “responsible water steward”, i.e. on maximising the value from water resources while seeking to achieve no long-term net harm in places where we operate. It sets out a three-stage journey that is being phased in over a ten-year period, to 2020. It includes a commitment to make our operations water-resilient, invest in water treatment and relevant technology innovation, build water infrastructure for mutual benefit and proactively partner with key stakeholders. Our implementation of this strategy is being realised through our initiatives in improving operational excellence, investing in technology, and engaging and partnering with our stakeholders.

We seek to:

  • develop new water resources and secure alternative water resources for mutual benefit
  • identify and secure post-consumer domestic effluent for use as industrial grade water
  • use water resources efficiently by adopting our waste-hierarchy principles of reduce, reuse and recycle
  • achieve water targets
  • protect water-quality resources and manage water quality within our operations


Through these strategies we continually strive to manage our water in a manner that does not compete with other sectors for the same water resource and to maintain the environmental reserve. Over time, this approach should prevent material impacts on the environment, downstream ecosystems and food security. In respect of reduction of water demand, a key feature of our water strategy is the drive towards “zero-potable-water” use in process operations (excluding domestic-use demand).

table116 Water supply

As part of our water-supply strategy, we have designed water-supply scenarios for the next 20 years based on the latest production predictions. 
To ensure the long-term security of water availability for our operations, other users and surrounding communities, we have also developed a bulk water strategy and infrastructure plan to protect, manage and maintain the water supply.

In 2011, we reported on the Olifants River Water Resources Development Project, which includes the construction of the De Hoop Dam and associated distribution components. The latest progress can be seen in the case study.

Water supply to Rustenburg

Water supply to Rustenburg is a concern because of a continued increase in the demand for potable water in the area by other users. This means that we have to reduce our potable water demand. As a result, we signed an off-take agreement with the Rustenburg Local Municipality to use 15 Ml/d of treated sewerage effluent from its sewage treatment plant. However, inconsistent water quality and supply are limiting the optimal use of this water. Several options to improve the situation were explored and a R15-million water-treatment plant was commissioned in November 2011, to improve the quality of the treated sewage water introduced into our water-reticulation system. The objective is to reduce potable water consumption by replacing potable water with upgraded treated sewage water. Since we commissioned the plant, the substitution of treated sewage water for potable water has resulted in the average conservation of 2 Ml/d of potable water. The plan for 2013 is to install additional pipelines to promote the use of a further 3 Ml of treated sewage water and simultaneously offset the use of an equivalent volume of potable water.

In addition, and in conjunction with other stakeholders, we initiated a pre-feasibility study to increase the supply of water to Rustenburg. This would involve 100 Ml/d originating from Hartbeespoort Dam, of which 50% would be provided to the municipality. The Department of Water Affairs has since taken over the pre-feasibility study and increased the scope of water supply to the greater Bojanala District area. We and other mines participated in the study as stakeholders through the Western Limb Producers’ Forum (WLPF), an association of mines in the province of North West. This study was completed in 2012. The parties will review the findings in 2013 and agree on which projects need to be developed for implementation.

Furthermore, WLPF has embarked on a programme to explore the opportunities for water-conservation and water-demand management between the various mines. We are participating in the study, which commenced in 2012 and will be completed during 2013.

Finally negotiations are at an advanced stage to secure a further 2 Ml/d of treated sewage water from the Northam Sewage Works, currently under construction, for use at Union Mine. The final contract to use the water will be concluded in the first quarter of 2013.



De Hoop Dam

A key pillar of Amplats’ water strategy is to manage water-supply issues by building partnerships. In 2011, we reported on the Olifants River Water Resources Development Project, which includes the construction of the De Hoop Dam and associated distribution components. Through the Joint Water Forum (JWF), we are actively involved in this public-private partnership initiative that aims to meet some of the mining sector’s water requirements and provide water to local people, agriculture and industry.

The De Hoop Dam, which started to store water in 2012, is now near completion. Owing to its size, it may take four to eight years to reach full capacity.

The Department of Water Affairs (DWA) has begun to construct the first phase of the pipeline from the dam towards Steelpoort. This phase will provide water up to the Sekhukhune take-off point, and our target date for delivery of water into the area is 2015.

Because of various investigations carried out by the DWA – the Yield Analysis of 2010 and the Olifants River Reconciliation Strategy of 2012 − it has become necessary to redevelop the bulk distribution system. The DWA has accepted the new layout proposed by the JWA.

The scheme’s progress is now envisaged as follows:

  • The De Hoop Dam will be completed by the DWA in 2013.
  • A pipeline from the dam will link into the existing southern extension pipeline of the Lebalelo Water Users’ Association (LWUA).
  • The upgrading of the pipeline between Steelpoort and Modikwa will be delayed and the LWUA will upgrade this pipeline when necessary.
  • The building of the other pipelines (phases 2e, 2f) is to be excluded from the project.
  • The Phase 2b pipeline, from the Flag Boshielo Dam to Mogalakwena Mine, will be completed by the end of 2017.

The development of new resources to augment the scheme, to ensure water surety for the mines and the communities in the area, is currently under investigation. The main source under investigation is the treatment of acid mine drainage from coal mines in the Emalahleni area.

Water resources classification

To support the consistent interpretation and reporting of water resources, we apply the following classification system:

  • New water consumed is separated into water used for primary activities and water used for non-primary activities. Primary activities include all water used to produce our products, from mining to refining. It excludes domestic use (e.g. at houses within Company-managed villages) and recreational use (e.g. soccer fields, golf courses, swimming pools, etc), which are classified as water used for non-primary activities. Internally recycled water is also excluded from the water used for primary activities as this water is accounted for when it enters the system.
  • Potable water is sourced from water utilities such as Rand Water, Magalies Water, Lepelle Northern Water and Rustenburg Municipality. These water utilities source their water from various dams, but in no instances do our abstractions account for more than 3% of the average annual yield of these water bodies.
  • Non-potable water use at managed operations is low and comes from the pipeline of the Lebalelo Water Users’ Association, in the Eastern Limb, which abstracts water from the Olifants River based on an approved allocation.
  • Surface water is consumed by Unki Platinum Mine (Unki) in Zimbabwe from the Lucilia Poort Dam. Water abstracted from the open pits at Mogalakwena Mine is also classified as surface water consumption.
  • Treated sewage effluent is classified as waste or second-class water and is sourced from municipal sewage plants to supply process water to Mogalakwena Mine and our operations in the Rustenburg mining area.
  • The groundwater parameter includes groundwater from boreholes used for primary and non-primary activities, as well as fissure water ingress from underground operations where this can be measured or estimated.
  • Rainfall harvested is defined as rain water collected or harvested by the operation, regardless of collected quality, that is used or stored for use in the operation. It is a predicted number from a rationalised water balance and reporting model developed for each of our operations. Rainfall harvested for use finds its origins in contained dirty-water circuits, open dams, water tanks and tailings facilities.
  • Discharge to surface water is defined as the total volume of water discharged from our operations to a receiving environment, such as rivers, during the reporting period regardless of the quality. Included is the excess water dewatered from mines and not used for primary or non-primary activities, for example at Dishaba Mine.

No water source, ecosystem (such as a Ramsar-listed wetland) or habitat is materially affected by our extraction and use of water.

Water consumption

In 2012, we consumed 34.9 million m3 of new water, against a total usage of 36.3 million m3 in 2011, a decrease of 4%. This also shows a 14% reduction against the 2012 water consumption target of 39.9 million m3. The main factors contributing to the savings were the successful implementation of two water-saving projects and reduced production as a result of the labour unrest. Water used for primary activities decreased by 8%, to 
28.8 million m3, while water used for non-primary activities increased by 21% to 6.2 million m3. The increase in water used for non-primary activities is a result of the improvement in water metering at the Union Mine.

Potable water

Potable water used for primary and non-primary activities decreased by 3% to 18.4 million m3 during 2012, compared with 19.0 million m3 during 2011. The decrease in potable water consumption was influenced mainly by the consistent use of the treated sewage water at Rustenburg operations to offset the use of potable water. We remain committed to striving towards the zero use of potable water in our operations.


Groundwater consumption increased by 2%, from 4.3 million m3 in 2011 to 4.4 million m3 in 2012.

Surface water

Surface water consumption decreased by 2%, from 1.54 million m3 in 2011 to 1.50 million m3 in 2012. A reduction in water use at our Mogalakwena Mine influenced surface- water consumption.

Stormwater management

table117 Rainfall harvested at operations decreased by 10%, from 0.038 million m3 in 2011 to 0.034 million m3 in 2012. Since 2011 there has been increased focus on the separation of clean and dirty water at our operations, which allows clean water to by-pass the operation and enter the natural environment. There is continued awareness of and focused compliance with Government Notice 704 (dealing with regulations on the use of water in mining), and our operations are at different stages of implementing and upgrading stormwater-management plans and systems.

Recycled water

Water recycled from internal sewage plants, tailings return-water dams, mine service water and other internal water sources, such as pollution control and stormwater dams, is not included in water used for primary or non-primary activities. Total recycled water use increased by 5% to 53.7 million m3 in 2012 (from 51.3 million m3 in 2011), with the ratio of recycled to new water consumed standing at 1.54. The overall emphasis remains on optimising the use of recycled water, and improving our water-monitoring and water-measuring initiatives.

Water use and intensity targets

Our new water intensity target for 2012 was 10.6 m3 per refined ounce of platinum group metals (PGMs) and gold from managed operations, and was calculated using a projected production of 3.9 million ounces of PGMs and gold from our managed operations.

Actual new water intensity per refined ounce of PGMs and gold from managed operations was 10.5 m3 in 2012 compared to the target of 10.6 m3 per refined ounce of PGMs and gold, a 1% improvement. However, a 3% increase in the intensity was observed from the 2011 intensity of 10.3 m3 per refined ounce of PGMs and gold. This intensity was calculated using the actual production of 3.32 million ounces of PGMs and gold in 2012 from managed operations, a 6% decrease in production compared to the 2011 production of 3.54 million ounces. When compared to the 2012 forecast production of 3.93 million ounces from managed operations, the decrease in production was 16%. The poor intensity observed was a result of the lower production output and operational disruptions during the industrial strike action.



While the new water consumption and intensity trends for our mining, smelting and refining operations showed an upward trend, our concentrator operations showed 
a consistent decrease in water consumption and intensity.

Water used for primary activities per refined ounce of PGMs and gold from managed operations improved by 2%, from 8.8 m3 in 2011 to 8.7 m3 in 2012. The potable water-use intensity per refined ounce of PGMs and gold from managed operations increased by 4% to 5.5 m3 (compared with 5.4 m3 in 2011).

An important focus by Anglo American in 2012 was on embedding operational water targets through the implementation of our water-efficiency target tool (WETT). The tool forecasts the projected business-as-usual (BAU) water demand of individual operations and establishes a register of water-saving projects, linking the two in order to deliver future performance targets. During 2012, we aligned to the WETT programme.

Consequently, our 2013 new water target is calculated to be 33.1 million m3 (or to achieve a 5% reduction in our 2012 water consumption of 34.9 million m3). The water consumption intensity for 2013 is expected to increase to 18.7 m3/ounce PGM and gold as a result of a lower production forecast.


Discharge to surface water

Total excess water discharged decreased by 57%, from 1.76 million m3 in 2011 to 0.77 million m3 in 2012. The average discharge for 2012 was 2 Ml/d (compared with 4.8 Ml/d in 2011). A contributing factor to reduced discharge was our water-management programmes and in particular the implementation of our integrated water and waste management plans (IWWMPs) at several operations.

The salt load of discharges to surface water was 25,670 tonnes of total dissolved solids and 7,461 tonnes of sulphates.

Some 68% of the total authorised discharge is from the Amandelbult mining right area, where excess water pumped from the Dishaba Mine is discharged into the Crocodile River as per section 21 permit requirements. To manage the excess water ingress, the mine continues with measures to reduce groundwater ingress and explore opportunities to reuse excess water. The excess water discharge from Amandelbult has been reduced by 80%, from 2.5 million m3 in 2009 to 0.52 million m3 in 2012. Despite the current efforts, however, the mine still discharged an average of 1.4 million litres per day during 2012 (2.2 million litres per day in 2011).

Chemical analysis and bio-monitoring surveys are conducted in the Crocodile River and the Bierspruit to determine any possible decline in the biotic integrity of the receiving water bodies (the last survey was conducted in November 2012). While there is a marginal deterioration in the chemical quality of the Crocodile River, it appears that the biotic integrity of this river is unaffected by any potential impacts originating between upstream and downstream sites. The biotic integrity of the Bierspruit may be affected during the dry season by its reduced dilution capacity, but survey results indicate that toxicity risks associated with the river have not increased as a result of the discharge.

The balance of the excess water discharge occurred as the result of spills owing to high rainfall or accidental discharges from various operations. These are reported as incidents and investigated and managed through our environmental management system.

Surface-water and groundwater quality around our operations

table120 Surface water and groundwater are monitored at all our mines and process plants, both upstream and downstream of operations, as well as inside and outside the mining areas in the catchments where we operate (see the case study on the IDDS water monitoring tool). Groundwater-monitoring results are used to model groundwater flows and contaminant plumes, if any, and surface-water and groundwater monitoring results are compared with various regulatory standards. Bio-monitoring of surface water bodies is also conducted. 
The tailings return-water dams at all operations continue to provide habitat for fish, birds and plant life. The quality of groundwater is affected at all mining operations, mainly as the result of seepage from the tailings storage facilities (TSFs). The impact is, however, localised in all instances and no external groundwater users are affected. Seepage from the TSFs contributes mainly to an increase in salinity of localised groundwater bodies.

In 2011, we reported on the abstraction and treatment of impacted groundwater using desalination technology at Rustenburg Base Metal Refiners (RBMR) as a pilot project. 
The abstraction and treatment of groundwater have not progressed as well as planned owing to technical problems experienced with the desalination plant. An upgrade of the desalination is planned for completion in July 2013. Thereafter the project will recommence.

At Rustenburg, surface-water quality, notably at Klipfontein Spruit, Klipgat Spruit, Paardekraal Spruit and the Hex River, is affected by the legacy of mining and process activities as well as by non-mining-related activities. Guideline values for electrical conductivity, sodium, chlorides, sulphates, calcium, magnesium, nitrites, nitrates and phosphate are exceeded for a section of the Hex River. This is the result of both industrial activities and non-industrial activities (i.e. agriculture). The Hex River has been shown to have good assimilative capability, with the result that the exceeded parameters are assimilated and water entering the Bospoort Dam complies with the Class I SANS 241:2006 drinking water standard.

Localised groundwater aquifers in the greater Rustenburg mining right area have shown evidence of groundwater quality deterioration. Such deterioration in groundwater quality could be compounded by future process-plant projects planned on our mining right area. 
The Rustenburg operations have embarked on a capital project, the Water Action Plan (WAP), to explore potential interventions to reduce and, where appropriate, mitigate impacts on surface and groundwater quality in the mining right area. While an integrated water and waste management plan is being implemented for Rustenburg operations, it is largely driven by legal compliance, whereas the WAP has a more overarching focus. The WAP is being developed in two phases:

  1. Phase 1 aims to be proactive by going beyond compliance in setting water targets and securing water. Furthermore, Phase 1 is intended to mitigate both surface and groundwater regional quality impacts and liabilities.
  2. Phase 2 aims to align the Rustenburg section with water security, water efficiency and stakeholder partnerships. The premise is to optimise the usage of internal water sources at the Rustenburg operations in order to reduce reliance on purchased potable water.

A series of concept studies, to the value of R6.5 million and supporting Phase 1, was initiated in 2012 and will be completed during 2013. The Phase 2 concept studies will 
begin in 2013.


Edward Masoba taking surface water samples at the Hex River, Rustenburg



Pipeline from De Hoop Dam

At Twickenham, shaft dewatering has led to a reduction in water yield from a surface spring on which the community is reliant for water. The mine is now supplying water to 
the community to address the loss. Twickenham has a positive water balance. The mine has evaluated water-treatment scenarios in order to manage the excess water and concomitant nitrogen loads. A waste-water treatment plant based on biological nutrient removal technology is being constructed to deal with water quality, and will be commissioned during 2013.

At Mogalakwena Mine, the total dissolved solids and the sulphate levels in the vicinity of the tailings dam increased slightly in the groundwater from last year, as was predicted in the recalibrated contaminant model for the operation. The current action plan calls for the monitoring and updating of the model as data becomes available. There are no users of this water.

Mogalakwena also completed the development of an integrated water balance during 2012, using advanced simulation tools. The mine has used the water balance to evaluate scenarios to further optimise its water-demand and reticulation management.

Acid rock drainage

Acid-base accounting to determine acid rock drainage and hazardous heavy metal leachate potential on both the Merensky and the UG2 tailings has indicated that such tailings have a negligible potential to generate acid or to mobilise metals. Although acid production and metal mobilisation do not occur, the sulphide content is sufficient to produce some soluble sulphates under oxidising conditions. This may increase the sulphate concentration in water that comes into contact with the tailings should there not be sufficient buffering capacity.


The total rock broken at managed operations was 3.7% less in 2012 than in 2011, mostly as the result of decreased production at most of the mines owing to labour unrest. There was a marginal decrease of 3.1% in tonnes ore milled from managed operations between 2011 and 2012.

Key bulk consumable materials required for mining and processing include liquid fuels, coal, grease and lubricants. Lower production contributed to a reduction in the use of liquid fuels, with a small decrease of 6.8% in the usage of diesel, petrol and paraffin. The consumption of other consumable materials, such as lubricants and grease, also decreased; with grease consumption down by 29%.

LPG use, however, increased by 37%, mainly at the Waterval Smelter and the ACP. The main reasons for this were flash dryers that were started more often during 2012, more intensive utilisation of the ladle and launder burner, increased pre-heating requirements of the main sulphuric acid vessel and a leaking heat exchanger.

Other key materials used include wood (for underground support), chemicals and packaging. Although different types of packaging materials are used, their volumes are minimal and therefore not material. Amplats does not currently use waste, processed or unprocessed, from external sources.



Willie van Loggerenberg at the Phoenix Slurry Pump, Amandelbult Concentrator

Slurry refers to any mixture consisting of a liquid mixed with spent tailings particles. Many industrial and mining processes operate on the basis of a wetted product to facilitate chemical reactions. Slurries are a convenient method of handling bulk materials.

Many types of slurry are abrasive, and some could be described as liquid sandpaper! Although the effects of their abrasive nature can be reduced by decreasing their flow velocity, many slurries have a minimum flow velocity that has to be adhered to in order to prevent their solid particles from being deposited. Slurries almost always result in abrasion in the pipelines and pumps used to move them along.

Moreover, the pumping of slurry is energy intensive, wastes water and requires intensive maintenance. Firstly, because tailings slurries are usually quite a lot denser than water, they require much higher pressure to pump them up elevations. Booster pumps are often required. Secondly, the centrifugal slurry pumps traditionally employed in the pumping of slurry have a number of serious limitations.

All centrifugal pumps require a seal where the pump shaft enters the pump casing.  
In the case of a water pump, this is achieved rather simply through the use of a “gland” (a mechanical seal). In the case of a slurry pump, maintaining a seal is far more complex because the slurry erodes the gland. The problem is overcome by injecting water at high pressure into the pump to keep the slurry away from the gland. However, this requires large quantities of clean (often potable) water at pressures higher than the pump-discharge pressure. Significant quantities of water are wasted, and additional power is required first to pump the water into the seal and then to pump the additional water through the slurry pipeline.

In response to all these drawbacks, International Slurry Pumping Services (ISPS) has developed a very different type of slurry pump. The Phoenix, as it is called, is able to pump high volumes of slurry at high pressure, with high overall system efficiency and low maintenance requirements. It does this by using an efficient water pump and then transferring the energy to the slurry using a bladder and vessel arrangement. Two vessels are utilised to achieve a constant pressure and flow rate in the pumping of water and the discharge of slurry. The system is operated in a manner that ensures smooth switching from vessel to vessel. The features developed for the Phoenix have been designed to greatly reduce the heavy wear and tear and the intensive maintenance of existing slurries pump systems.

Amplats has been testing the new technology through the pilot installation of a Phoenix-type pump at its Amandelbult Concentrator to validate the benefits. The system, which eliminates the need for a V-belt drive (and the energy losses associated with this), has undoubtedly questioned established ideas regarding the pumping of slurries in our business. The system has a design flow rate of 350 m3/h at 16 bar, and its overall efficiency is 74% compared with the previous system’s 54%. This, together with its use of a switchgear, means that the need for booster pumps in the pumping of slurries simply falls away.

Greater efficiency, reduced set-up and running costs, and less time and money spent on maintenance – the new system seems to have it all.



Pollution control dams at Twickenham Mine

In 2012, Amplats embarked on the development of an integrated data-display system (IDDS) that would make it possible for us to collate all the water-quality data for our operations into a single database. The water-quality data covers the various water sources that Amplats interacts with, such as groundwater, surface water, potable receiving water bodies such as rivers, process water such as return-water dams, stormwater dams and pollution-control dams, mine service water, treated sewage water and

cooling water.

The water-quality analysis for Amplats is undertaken by external laboratory service providers, which supply feedback data to Amplats on a regular basis. They are Aquatico, AGES Lab, GCS Labs, Capricorn Veterinary Services and the University of Zimbabwe. All the water quality data they provide is uploaded into the IDDS.

Until quite recently, the detailed water-quality data gathered painstakingly over the years was available mostly in the form of individual hard-copy reports at our operations, with each report covering one or two aspects only. Few people were able to make sense of or even access the wealth of information they contained.

We have now captured most of the old data, which dates as far back as 1997 for some of our operations. In future, all water-quality data will be uploaded into the database.

The use of the visual representation of data in countering this type of limitation − and in making data accessible − has been growing in popularity, sophistication and ease of use. Amplats’ IDDS works as a geospatial display of our operations; and its fit-for-purpose reporting functionality provides comparisons of water-quality data and trends analyses on demand.

The result is that Amplats has been visualising monthly results for all its operations on a common interface. Attributive information about a particular aspect or topic for a particular month is being accessed merely by clicking on it. It is now possible to compare data and results against many different standards, not just one or two as in the old reports. If a new standard is introduced, this is easily incorporated into the system. Moreover, data is processed and integrated in a way that affords technicians and managers new and valuable insights into aspects that were previously not noticeable.

Being able to integrate data from multiple sources onto a common visual interface reduces the time required to access critical information, and thus leads to faster, better-informed decisions. The new system also saves money by using technologies currently in place at the suppliers (e.g. Microsoft SQL databases, IE web browsers), and increases the accuracy of the data entering the database. (The database administrators and also the software service provider (Geosemantic) analyse the data for errors before it enters the Anglo American database.)

The IDDS will be further developed as we proceed with customised reporting for each operation. Amplats envisages adding databases for other environmental data categories, for example air quality, biodiversity and land management.