AN ENVIRONMENTAL MANAGEMENT PLAN FOR THE MERRIESPRUIT SLIMES DAM DISASTER AREA
by
THEUNIS JOHANNES DUVENHAGE 9605046
MINI-THESIS
submitted in partial fulfilment for the degree
MASTER IN NATURAL SCIENCES
in
GEOGRAPHY AND ENVIRONMENTAL MANAGEMENT
in the
FACULTY OF NATURAL SCIENCES
at the
RAND AFRIKAANS UNIVERSITY
STUDY LEADER: Dr. J.M. Meeuwis
JUNE 1998 The Nation is in the position of a man, bequeathed a fortune, has gone on spending it recklessly, never taking the trouble to ask the amount of his inheritance, or how long it is likely to last.
National Conservation Commission (1908)
Democracy is not a matter of sentiment, but foresight. Any system that doesn't take the long run into account will burn itself out in the short run.
Charles Yost, The Age of Triumph and Frustration Acknowledgements
My thanks to
my parents for supporting me through the initial part of the study; my wife for helping me seeing it through, Frik Senekal for the printing and design, Annette, Desire and Jacqueline for the typing, Maria de Wet and Virginia Publicity Association for some of the photos, Rob Gillmore at Harmony Gold Mine who took the time to supply me with crucial information.
This work is dedicated to the seventeen people who died in the disaster and all the people who helped during the disaster. CONTENTS
Page
TABLE OF CONTENTS LIST OF FIGURES iii LIST OF TABLES ABSTRACT vi OPSOMMING vii
SECTION 1
INTRODUCTION 1
1.1 Preamble 1 1.2 Problem Statement 2 1.3 Aims and Objective 3 1.4 Methodology 4 1.5 Studies Completed or in Progress 5
SECTION 2
ENVIRONMENTAL MANAGEMENT PLAN 8
2.1 What is an Environmental Management Plan? 8 2.2 The EMP Process 9
SECTION 3
PRE-DISASTER AREA 11
3.1 Locality 11 3.2 Climate 12 3.3 Topography 15 3.4 Geology 16 3.5 Soils 17 3.6 Surface water 18 3.7 Sub-surface water 19 3.8 Pans and wetlands 20 3.9 Biotic component 21 3.9.1 Plants 21 3.9.2 Animals 22 11
3.10 Human aspects 23 3.10.1 Infrastructure 23 3.10.2 Land use 23
SECTION 4
POST-DISASTER AREA 25
4.1 Surface water management 25 4.1.1Surface water quality 25 4. I.2Pre-disaster water quality 25 4.2 Ground water management 29 4.3 Storm water 30 4.4 Dust 32 4.5 Waste management 32 4.6 Aesthetics and land use 33
SECTION 5
MANAGEMENT PLAN
5.1 Management goals 34 5.2 Project actions 34 5.3 Surface water management 34 5.4 Ground water 35 5.5 Storm water control 37 5.6 Waste management 38 5.7 Dust 39 5.8 Aesthetics and social-economic implications 40 5.9 Rehabilitation 42
SECTION 6
CONCLUSION 43
REFERENCES 46
APPENDIX In
LIST OF FIGURES
Figure 1.1: The northern wall of the Merriespruit Tailings dam (Dam 33) after the disaster. 1
Figure 1.2: The Lower Convent dam after the disaster. 2
Figure 3.1: Locality map. 11
Figure 3.2: The study area. 12
Figure 3.3: Evaporation and precipitation. 12
Figure 3.4: Mean annual rainfall. 13
Figure 3.5: Mean monthly rainfall. 13
Figure 3.6: Monthly evaporation. 14
Figure 3.7: Area north of the Convent dam. 16
Figure 3.8: Runoff distribution. 17
Figure 3.9: Geological section through the study area. 17
Figure 3.10: Soil structure. 18
Figure 3.11: Harmony Wetland and Bird Sanctuary (Convent dam ) prior to the disaster. 20
Figure 3.12: Natural vegetation of the area. 21 iv
Figure 4.1: Positions of boreholes. 26
Figure 4.2: Conductivity and sulphates taken at Linabo Bridge between 1987-1993. 27
Figure 4.3: Conductivity and sulphates taken at Linabo Bridge between 1994-1995. 27
Figure 4.4: Conductivity and sulphates taken at Mostert Canal between 1987-1993. 28
Figure 4.5: Conductivity and sulphates taken at Mostert Canal between 1994-1995. 28
Figure 4.6: Analytical results from borehole sampling. 30
Figure 4.7: Visual pollution in the area. 33 V
LIST OF TABLES
Table 3.1: Maximum rainfall intensities per month. 14
Table 3.2: Temperature statistics for the Virginia area. 14
Table 3.3: Predicted flood peaks and volumes for the Sand River. 19
Table 3.4: Population distribution. 23
Table 4.1: Summary of the conductivity and sulphates at Linabo Bridge. 26
Table 4.2: Summary of the conductivity and sulphates east of the Mostert Canal. 26
Table 4.3: Summary of the conductivity and sulphates east of the Mostert Canal after the disaster. 29 vi ABSTRACT
The Merriespruit Tailings dam disaster killed seventeen (17) people and covered a part of Virginia with approximately 2.5 million cubic metres of tailings, causing such an emotional uproar that all resources were focused on repairing the dam and addressing some of the social issues. Little attention was given to the environment. The identified need in this study was therefore to investigate the consequences of the disaster on the environment, a need which derives from the uniqueness of this particular disaster and its consequences.
The Department of Minerals and Energy require the submission of an Environmental Management Program Report (EMPR) on all prospecting and mining operations. It is clear that, in the compilation of such an EMPR, Harmony Gold Mine neglected to establish a Management Plan to regulate the physical impact of the disaster on the environment, mainly because no attention was given to disasters in the Aide-Memoir.
A Management Plan was established by adapting existing formats of management plans to the uniqueness of this disaster. By following the procedure stipulated in the Management Plan it can be ensured that Environmental Management requirements will be effectively integrated into either the project management actions and contracts or operational systems and processes for the following issues:
Water management Storm water control Waste management Dust Aesthetics and socio-economic implications Rehabilitation of the area.
The investigation showed that the disaster exerted a definite negative influence on the environment, which can be managed by taking preventative measures stipulated in the Management Plan. However, one of the main issues identified in this study is that storm water management has been problematic for a period of time. It is therefore noted that some attention should be given to establishing a wetland system to contain the storm water runoff. Although this study does not focus on the socio-economic impacts in detail, it is recommended that these impacts are considered as it is evidently problematic.
The primary aim of this study was to compile an EMP in order to manage, and possibly mitigate, the physical impact of the disaster on the immediate environment, an aim which clearly was accomplished. Harmony Gold Mine can benefit from the compilation of this EMT, as management goals were set and feasible means of achieving them were specified. vii
OPSOMMING
Na die slikdamramp by Merriespruit waartydens 'n deel van Virginia met ongeveer 2,5 miljoen kubieke meter silk bedek en sewentien (17) mense se dood veroorsaak het, was daar groot emosionele reaksie. Dit het tot gevolg gehad dat alle mannekrag by die herstel van die damwal aangewend moes word sodat min aandag gegee is aan die impak van die ramp op die omgewing. Die doel van die studie is om die gevolge van die ramp op die omgewing te ondersoek en 'n bstuursplan op te stel.
Die Departement van Mineraal- en Energiesake vereis dat 'n myn 'n Omgewingsbestuurs- plan (OBP) indien, om alle aktiwiteite tydens die mynproses aan te spreek. 'n OBP mask egter the voorsiening vir gebeurtenisse soos rampe nie, en daarom het Harmony Goud Myn the 'n bestuursplan vir die ramp in werking gestel the.
'n Bestuursplan is ontwikkel deur bestaande formate aan te pas. Die voorgestelde aksie in die bestuursplan sal verseker dat die moontlike impak beperk sal word. Daar is spesifiek na die volgende areas gekyk:
Waterbestuur Stormwater kontrole Afvalbeheer Stof Estetika en sosio-ekonomiese impak Rehabilitasie
Die ondersoek het getoon dat daar 'n geringe mate van invloed op die omgewing was weens die ramp, maar dat dit met voorkomende maatreels beheer kan word. Alhoewel dit the 'n direkte oorsaak van die ramp was the, blyk dit dat die stormwater probleme oplewer. Daar is voorgestel dat die stormwater na 'n vleigebied gedreineer word.
Die doelwitte van hierdie stude, naamlik om 'n bestuursplan te ontwikkel en moontlik die impak van die ramp op die bepaalde omgewing reg te stel is bereik. Harmony Goud Myn kan dan ook direkte voordeel trek uit die versiag aangesien die beste oplossing vir die probleem deur die studie aangeteken is. SECTION 1
INTRODUCTION
1.1 Preamble
On the night of 22 February 1994, the 29-year old Merriespruit Tailings' dam (Dam 33) northern dam wall gave way, unleashing a wave of mud which covered nearly 2 km 2 of the Virginia area, killing seventeen (17) people and leaving hundreds homeless, with damages estimated in the region of R100 million. The dam lost 25% of its 10 million capacity through a 50 m gash in the dam wall, when it failed. This was only the second disaster of its kind in South Africa2, but the first with major damage to private property and with the loss of lives. The identified need in this study is therefore to investigate the consequences of the disaster on the environment, a need which derives from the uniqueness of this particular disaster and its consequences. It can also be postulated, however, that information generated by this investigation could be of significant benefit to the general field of Environmental Management.
Figure 1.1: The northern wall of the Merriespruit Tailings dam (Dam 33) after the disaster ( Photo: Courtesy of the Virginia Publicity Association).
I Seeing that the mining industry uses the term 'tailings' rather that the colloquial 'slimes', the former term will be utilised for the purpose of this study. 2 Two small failures occurred in 1937 at Simmer & Jack and at Grootvlei in 1956. In 1974 the Tailings dam at the Bafokeng mine near Rustenburg gave way and flooded the mine. The slime covered a area of 0,8 x 4km, 10m deep. The damage was confined to the mine property and no lives were lost (Jones & Pearce, 1996). 2
1.2 Problem Statement
Wells, J. D. et al in Fuggle and Rabie (1992) note that although legislation exists, it does not cater for the modern trend toward the holistic consideration of the environment, nor does it stress the need for environmental management to be integrated into everyday mining management. In recognising this fact, the Department of Minerals and Energy require the submission of an Environmental Management Program Report (EMPR 3) on all prospecting and mining operations. The day to day operations and the compilation of this EMPR are regulated by an Aide-Memoire that was published by the Government Mining Engineer (Department of Minerals and Energy, 1992). It is clear that, in the compilation of such an EMPR, Harmony Gold Mine neglected to establish a Management Plan to regulate the physical impact of such a disaster on the environment (Unpublished EMPR, Harmony Gold Mine). According to Hugo (1992) and the Minerals Act (Act 50 of 1991), the holder of the mineral rights has the right to enter upon land and disturb the surface thereof in order to search for and mine a mineral. Such right must in all respects be exercised in a responsible manner and the mining entrepreneur has a grave responsibility to manage the effects of the mining activity on the environment in such a way as to mitigate the negative impact whilst maximising the positive features (Hugo, 1992).
Figure 1.2: The Lower Convent dam after the disaster ( Photo: Courtesy of the Virginia Publicity Association).
In the mining industry an Environmental Management Program(EMP) needs to be compiled in order to address the influence the mine has on the environment. The program regulates the mining procedures from the inception to decommissioning and closure. In other sectors the same type of program is referred to as an Environmental Management Plan (EMP). In this study the term EMP is used to correspond with the second example, and should not be misinterpreted as an EMP that regulates mining activities which necessarily includes the operation of a mine (Prinsloo, I997). 3 In correlation with the statement above, can it be said that the Harmony Gold Mine acted in a responsible manner toward the environment? It has already been proven that the mine acted negligently in maintaining and operating the Tailings dam as such 4. The scope of this paper is therefore to address the question of formulating an Environmental Management Plan (EMP), through a consideration of both the pre-disaster and the post- disaster environments. In order to draw an informed conclusion, it is essential to evaluate the biophysical environment, as is detailed in the heading 1.4 'methodology', so as to determine possible negative effects the disaster had on the environment and the best way of addressing it. It is clear that, although the mine paid a measure of attention to the disaster's economic and social impact, the impact on the environment was and is neglected. It is certain that, due to the disaster, some forms of pollution occurred, although some intermediate actions were taken to prevent further pollution of the area. Being intermediate at the time, no such rehabilitative actions are exercised at present. The primary aim of this study is therefore to compile an EMP in order to identify, manage and possibly mitigate the physical impact of the disaster on the immediate environment. No comprehensive study that focuses on the Merriespruit Tailings dam disaster environment and its management has been conducted to date. Other studies done on the disaster are mostly of a specialised nature with emphasis on the reasons why the dam gave way, not the 3-fold focus of this study as stipulated under 1.3. Social consequences have been considered yet no focus has been placed on strategies that could be utilised in order to manage the environmental impact of the disaster. In summary, can Harmony Gold Mine benefit from the compilation of an EMP, what management goals should be set and how could these goals be achieved? What makes this study unique, and thus essential, is that this disaster, like any other, is a single occurrence with a major impact on the environment, with results and conclusive deductions that simply cannot be obtained under any other circumstances. It is therefore clear that the need for this investigation and subsequent EMP are beyond question.
1.3 Aims and Objective
The aims of this paper are 2-fold: to investigate the physical area (pre-disaster) to investigate the post-disaster area (the effects on the natural environment)
The objective is to establish an Environmental Management Plan (EMP) that will assist with the rehabilitation of the Merriespruit area. The plan will identify the problems, and compose management goals and actions to achieve the goals.
4 Harmony Gold Mine was found guilty in the regional Court of Virginia and fined R120 000. The construction company Fraser Alexander was also found guilty and fined RI50 00. The court found that these companies contravened section 37 of the Minerals Act, being negligent in the operation of the Tailings dam (Jones & Pearce,1996). 4 1.4 Methodology
It is thus apparent that an EMP should be drawn up as a form of regulatory measure, and to identify harmful impacts caused by the disaster on the environment and human well- being. An EMP is a ways and means to prevent and limit harmful impacts. An EMP is a combination of a plan / a scope of work / specifications to manage (through mitigation or avoidance) environmental impacts associated with a particular project or operation / process. Its aim is to ensure that the following are in place: responsibilities key performance indicators implementable and measurable specifications channels of reporting a monitoring schedule.
Since this is all very generalised, there is a definite need for focus. For the purposes of this study ( the Merriespruit Tailings dam disaster ) it was necessary to determine what influence an unpredicted event might have had on the area. EMP's make provision for most actions except events such as the Tailings dam disaster. In order to compile a comprehensive plan on the possible implications of the incident as was stipulated, it will be necessary to obtain a holistic insight into the area. In order to accomplish the objective, the ESCOM guidelines (Lucas, 1997), on compiling an EMP will be adapted for the Merriespruit disaster area. The Aide-Memoire that was published by the Government Mining Engineer (Department of Minerals and Energy, 1992) will also be used to structure the process. Boonzaaier & Van der Walt (1997), as well as Freer (1997)and Lucas (1997), noted that there is not one format for an EMP, but that the latter should be individualised for each study. Although the above-mentioned guidelines were used, some adaptations were made due to the study's uniqueness. As part of the investigation a basic Environmental Impact Assessment of the post- disaster area, that will be compared to the pre-disaster environment, will be done in accordance with guidelines provided by the Department of Environmental Affairs and Tourism (The Integrated Environmental Management Procedure, vol. 1-6). The guidelines on rehabilitation of trailing dams as prescribed by the Chamber of Mines, in conjunction with the departments of Minerals and Energy and of Water Affairs and Forestry, will also be utilised. The Environmental Management Programme will be developed and implemented through the following steps: identification of the environmental elements or issues to be addressed and managed (some form of assessment is to be undertaken) translation of the above-mentioned issues into specific management objectives or goals conceptualisation of the Project or Operational procedures, processes and systems of which the EMP is to form an integral part definition of what is required to achieve the management objectives or goals i.e. in 5 terms of facilities, procedures, sittings, permits etc. linkage of the above requirements to the Project or Operational processes, systems and structures i.e. integration of the required environmental programmes into the Project or Operational processes and systems ensuring that there are mechanisms to deal with the unexpected environmental problems that may arise.
Seeing that the accident has already occurred, the EMP will be handled as though it were for an existing operation or plant. The use of a Life-cycle Assessment (LCA) or an Environmental Risk Assessment (ERA) will therefore not be necessary. The physical environment and its composition will be investigated and will form the basis from which deductions can be drawn. The resulting identified problems will be translated into the following specific management goals or objectives: water management storm water control waste management dust aesthetics and socio-economic implications rehabilitation of the area
All the above-mentioned goals will be addressed in the Management Plan, with suggested mitigating measures to be performed. By following the procedure outlined above it can be ensured that environmental management requirements will be effectively integrated into either the project management actions and contracts or operational systems and processes. As preamble to the physical study it will be necessary to establish what studies have been done in the area, to determine if these studies are informative and what their relevance to this study are or might be.
1.5 Studies Completed or in Progress
Cozby (1985) states the following: " Before any research project is conducted, the investigator must have a thorough knowledge of earlier research findings. Even if the basic idea has been formulated, a review of past studies will aid the researcher to clarify his idea and to design the study" (p 20). Although much work has been done on the effect of mining activities on the environment, little work has been done in the Free State Goldfields. The studies that have been done mainly are the EMP's for the different mines in the area. Existing examples include Hodgson (1987) who touched on the subject of water quality in the Free State Goldfields. The study indicates that the pollution of ground water is restricted to the immediate vicinity of the pollution sources. Water originating from Virginia and Welkom seems to pollute the Sand River as it flows through the Goldfields. Additionally, James and Mrost (1965) state that pyrite oxidation in Tailings dams is confined to a surface layer of about 2 m in depth and that oxidation is limited to the depth to which oxygen can gain access. These results were verified by Mrost and Lloyd (1970). According to Marsden (1986), the resultant release of sulphate, due to the oxidation of pyrite, is negligible in old tailings deposits, mainly due to the fact that oxidation does not proceed below a depth of 2 to 3 m in slime and 10m in sand deposits. Several authors have documented the effect of acid mine drainage on the water in rivers and streams (Dyer, 1977 and Hoehn & Sizemore, 1977). They deduce that the low pH, high total dissolved solids and high concentrations of heavy metals, cause considerable damage to ecosystems, while Jones et al (1988) investigated the present contribution of mine deposits to the pollution load in the Vaal Barrage. They concluded that the contribution of Tailings dams to the salt load are negligible and only sand dumps are considered to contribute to the total salt load. Cogho et al (1992) investigated the development of techniques for the evaluation and effective management of surface and ground water contamination in the Free State Goldfields. They concluded that the evaporation areas pose the largest pollution potential in the study area, due to the large quantities of unpotable water which are stored in them daily. The Water Research Commission completed a study on the use of vegetation in the amelioration of the water quality impacts of mining. It showed a correlation between plant species used and water quality. The importance of this study is found in the fact that the tailings deposited during the spill might have had an influence on the soil and water quality. These factors might influence the rate and type of plants to be used in the rehabilitation. The National Energy Council has compiled a land-soil classification system for rehabilitated land. The purpose of this system is to establish a classification system to be introduced in the rehabilitation environment and which is to be used as a reference work during the rehabilitation phase. Both the South African Bureau of Standards, and the Council for Scientific and Industrial Research, undertook major studies on the design of tailings dams in the mining industry. National standards for the construction maintenance and decomposition of tailings dams resulted from this study. A complete Strategic Communication Plan was drawn up under the guidance of the Department of Social Development by the Pronam Partnership. The study addressed the social and economic issues following the disaster, and ways to alleviate the problems. Of more particular relevance to this study is the fact that the Environmental Management Program Report (EMPR) for Harmony Gold Mine is still under evaluation. As has already been mentioned (see section 1.2), this report compiled in relation to guidelines of the Aide-Memoire, revealed that the Merriespruit disaster area has not been addressed as a factor in the EMPR. This EMPR draft also formed the background to be used in compiling data. It formed the basis of the assessment, as specialists were involved in compiling data for the report. As part of the EMPR, an extensive study on the pollution and quality of ground water for the Virginia area was compiled by Rand Mines 7 Environmental Division, with constant monitoring of surface and ground water still in progress. The physical aspects of disasters, specifically those pertaining to the Merriespruit Tailings dam disaster management, have not yet been addressed. Other studies, although generalised and focusing on water quality, Acid Mine Drainage (AMD) and other relevant factors, need to be utilised in order to address the resulting effects. Weather patterns and geological strata can influence the drainage patterns directly, the Acid Mine Drainage indirectly, and eventually even the biotic life forms. Although some parts of the present study might seem to be irrelevant at first glance, from the perspective of an Environmental Manager the influences could indirectly contribute to the total outcomes. In conjunction with this, the effectiveness of alternative mitigating measures completed should also be assessed and possibly addressed. All the usual factors that are coherent to mining, and especially Tailings dams, are of importance to this study. Features that make the Merriespruit Tailings dam disaster unique are the fact that it had a sudden high quantitative impact and not the normal slow cumulative impact, while the physical environment makes it different from other studies. 8
SECTION 2
ENVIRONMENTAL MANAGEMENT PLAN
2.1 What is an Environmental Management Plan (EMP)? The aim of an EMP is to describe how negative environmental impacts should be managed and monitored, how positive impacts must be maximised and how negative areas should be rehabilitated (Preston et al, 1993). An EMP is a document that should be generated from the outcome of some form of environmental assessment. For a proposed new development, this could be an Environmental Impact Assessment (EIA). In this case the EMP will be to mitigate or avoid predicted impacts. For an existing operation, process or plant it could be from an Environmental Risk Assignment, and Environmental Management System (EMS) audit, or a Life Cycle Assessment (LCA). In the case of an existing operation, the EMP would set out specifications to either mitigate or eliminate those identified impacts as occurring or the identified risk areas and operations. In the case of this study the aim is to set out specifications mitigating the impacts that have occurred and to achieve specified environmental performance goals and objectives. The EMP is there to formulate actions to be taken and standards to be met in order to avoid, control, reduce and/or remediate adverse environmental impacts so as to conform to ETA findings and recommendations, Life Cycle Assessment evaluations, legislation obligations, permit requirements, license conditions and an organisational policies and standards. The question can be posed as to wether the specifications set out in an EMP should be prescriptive or not. The answer is not very clear cut although the author feels that a measure of prescriptiveness should be build into the EMP. Each requirement or specification of the EMP must be measurable or quantifiable. The EMP should have an objective, goal or target, a responsible person, adequate resources and a time frame. To implement the EMP, through the management thereof, it must be measurable. To audit and ensure compliance, it must be measured against something quantifiable. The test then should be to ensure that each requirement of the EMP can be quantified. Although it sounds logical, it is not so easily achieved. It is necessary to look at specific management tools that can be used to undertake environmental assessments for use in the development and integration of environmental management plans and programmes. 9 2.2 The EMP Process
The EMP should not be seen as purely beneficial to the Environmental Practitioner but also as a management tool for Project and Operational staff to meet their environmental responsibilities and targets. When the EMP is compiled and incorporated into the existing Project or Operational Management systems, structures, processes and terminology, the general process for the development and integration of the EMP should be as follows:
1 Identify Environmental Management Requirements, this will include identifying the environmental elements or issues that need to be addressed, managed or controlled. These issues can be determined through undertaking some form of assessment. For new development projects it could be an Environmental Impact Assessment (EIA). For an existing operation or plant it could be from a Life-cycle Assessment (LCA), an Environmental Risk Assessment (ERA) or Audit. These issues could include: surface water management storm water control ground water management air pollution dust noise waste management protection of fauna and flora aesthetics rehabilitation.
2 Management Goals
After determining the environmental issues, these should then be translated into specific management goals or objectives.
3 Project Actions
Each of the required goals or objectives that are to be achieved need to be actioned. It must be established what action is required to achieve the set goals and objectives. The project action/s could be one or more of the following: construction of a facility (water treatment plant) development of specific operational procedures for the carrying out of certain activities (to preserve archaeological sites) specific siting of plant or activities policy, formulation with respective authorities (not internal organisational policy) 10 specialist studies may be required monitoring requirements liaison with the authorities in connection with information on the project contract conditions for specific requirements for the contractor to achieve or adhere to permits, where specific permits may be required and thereafter adhered to.
By following the procedure outlined above it can be ensured that environmental management requirements are effectively integrated into either the project management actions and contracts or operational systems and processes. In order for an Environmental Practitioner to manage this process it is recommended that a matrix be structured whereby the environmental management programmes or requirements are linked to either the project or operational activities in terms of project or operational segments or actions. 11
SECTION 3
PRE-DISASTER AREA
In the problem statement it was already noted that the study of the biophysical environment is essential in order to draw an informed conclusion. It is necessary to understand the biophysical environment in order to establish a possible Management Plan as well as the necessary rehabilitation strategies at the end of the study.
3.1 Locality
The discovery of gold in the Free State Goldfields during 1946 led to the establishment of the town Virginias .which is located about 290 km south-west of Johannesburg, on the banks of the Sand River. The Sand River, that runs through the town, is one of the major tributaries of the Vet River, that in turn, is one of the Vaal River's tributaries. South of Ventersburg the river is dammed by the Allemanskraal dam, and a weir was also erected in the river east of the study area. These factors contribute to the fact that the river is prone to accumulate quantities of pollution in drier periods (Cogho et al, 1990). This can be problematic as Virginia's water supply is drawn from the river (Loubser, 1996) and pollution will have a direct impact on the population. As can be seen in figure 3.2 the drainage from the study area could affect the river, via storm water drainage from the area.
C: Land C=I Hater bodies Ice field ne• Coontry boundaries /v Coastline O ettjor oily ne Perennial rivers Intermittent rivers N Major Roads "-e Roads ♦i-t- Major railroads Major trails ❑ City + Airport
Figure 3.1: Locality.map (Source: http:// www.esri.com ).
Virginia can be found on the topographic sheet 2826BB, the area of interest is centred at 28°06 93'S and 26°51 20'E. !
(t-
:f • '•
• • 12 The Merriespruit tailings dam (Dam 33) was built in 1956, south-west of the central business district of Virginia6. When the northern wall gave way the area north of the dam was covered with tailings. The boundaries of the study area are marked out on figure 3.2.
From figure 3.2 it is clear that the area is divided into an urban component and a natural component. Focus will be on the natural component, as most of the pressing issues in the urban area have already been addressed during the cleaning operations.
The proximity of the Sand River, together with its use as a potable water source, makes water quality and water management a main concern in this study. Climate influences the rehabilitation methods and management strategies, and it is therefore important to comprehend the overall climatic composition. The information in this section relates to the methods and materials used in section seven, the Management Plan (Coppin & Richards, 1990, Gardiner, 1991 and Joughin et al, 1997).
3.2 Climate
This area is a semi-arid region with an annual rainfall of between 400 and 600 mm, according to the Koppen classification. It is an area with a chronic water shortage , i.e. the annual potential to evaporate water exceeds the precipitation, as depicted in figure 3.3. This factor needs to be taken into consideration during the rehabilitation phase as it will determine the method used in re-establishing fauna, as well as storm water management.
MEAN ANNUAL RAINFALL AND PRECIPITATION
Rainfall 0 a — Evaporation YEARS
Figure 3.3: Evaporation and precipitation (Data Courtesy of Glen Agricultural College).
Local thunderstorms and showers are responsible for most of the precipitation during the summer, from October to March and peaking in January. It is therefore necessary to look at the months with peak rainfall and compare this to low rainfall intensities, so to
6 According to Loots in Jones & Pearce (1996), this was done in direct contrast to current construction practices.
13 establish if the spill had any long term effect on the water quality ( Brocksen & Wisniewski, 1988 and Corbitt, 1989).
The mean monthly and annual rainfall is depicted in figures 3.4 and 3.5. Rainfall intensities depicted in table 3.1 will be relevant in determining the storm water management.' The reason for noting this data is to determine if the rainfall had an influence on the water quality, as dilution (by means of rainfall) is the best way of dissipating pollution spills (Best, 1987 and Hammer, 1989).
MEAN ANNUAL RAINFALL 1987-1995
800 700 26O0
02 503 01937 400 SI 1998 a. 330 01989 uh 200 01993 g 103 0 1991 0 01992 1 01993 YEARS 01934 01995
Figure 3.4: Mean annual rainfall (Data Courtesy of Glen Agricultural College).
AVERAGE RAINFALL 1987-1995 IN (MM) 0 JAN 13 FEB 200.0 0 MRT 150.0 O APR 0 MAY 100.0 O JUN 0 JUL sao O AUG
n1 0 SEPT 0.0 run n hiondif Irk 1119 [[k 1997 1988 1989 1990 1991 1992 1993 1994 1995 0 OCT YEARS 0 NOV CI DEC
Figure 3.5: Mean monthly rainfall (Data Courtesy of Glen Agricultural College).
'The information was obtained from the Glen Agricultural College and was measured at the Sandvet weather station. It includes the period January 1987 to December 1995.
14
Table 3.1: Maximum rainfall intensities per month (Data Courtesy of Glen Agricultural College).
Temperature has a direct influence on the rate of evaporation and the oxygen demand in the water (Hammer, 1989 and Kiely,1997). Temperature statistics for the area are included in Table 3.2. Temperatures show large daily and seasonal variations. The Management Plan and method of rehabilitation should be structured to take temperature influences into consideration as noted by Rawlings (1989) and Coppin & Richards (1990).
v ..rv flati
0',04.4444 klicandai
Table 3.2: Temperature statistics for the Virginia area (Data Courtesy of Glen Agricultural College).
AVERAGE EVAPORATION 1987-1995 IN (MM) O JAN 0 FEB 12.0 O MRT 2 10.0 O APR g 8.0 MAY 6.0 O JUN qa 4.0 O JUL O AUG 2.0 SEPT 0.0 O OCT 0 NOV 0 DEC
Figure 3.6: Monthly evaporation (Data Courtesy of Glen Agricultural College).
Mean temperatures reach a maximum in December/ January and minimum in June/ July (see table 3.2). The period during which frost can be expected lasts about 100 days (June to August). As early as 1971, Mostert et al noted that climatic factors, in conjunction with the biotic diversity, will have a direct impact on the type and method of plant 15 growth, a fact that was re-affirmed by Coppin & Richards (1990). Mean monthly evaporation for the area is depicted by figure 3.6 and covers the period January 1987 to December 1995.
Wind can also be an influential factor during rehabilitation, as strong and continuous wind will iterate the need for wind breaks to protect the area under rehabilitation (Coppin & Richards, 1990 and Gardiner, 1991). The winds in the region are usually north-easterly and north-westerly, reaching their maximum speed in the afternoon. During thunderstorms, strong and gusty south westerly winds prevail. Nevertheless, the duration of these winds is very short.
In the dry seasons (July to September) dust storms are a common occurrence. A major contributor to dust in the atmosphere is exposed surfaces on tailing dams. The small particle size of the tailings are easily eroded by aeolian action (Best, 1987, Fuggle & Rabie, 1992 and the Department of Minerals and Energy, 1992). The spill mainly consists of tailings, thus the same problem noted by Best (1987) will be evident in the study area.
The wind direction can aid in the spreading of plant seeds and thus limit the manual seeding process ( Coppin & Richards, 1990 and Van Niekerk, 1998). In conclusion it can be noted that the actions taken in the Management Plan could be regulated by extreme weather conditions as these are not an exception but rather a norm and should thus play an important role in the planning process.
3.3 Topography
The area forms part of the Highveld region (Cooks, 1989) and has an average elevation of about 1360 m above mean sea level. Most of the area has a gentle undulating surface and prominent landmarks are rarely visible. Natural pan formation is a standard feature of this area (Le Roux,1978, De Bruyn, 1971 and Marshall, 1987).
The Sand River meanders and flows in an alluvial flat almost throughout its entire length. The valley of the Sand River has an elevation of 1 300 m in the east and 1 257m in the west (Coetzee, 1960). The total depth of the river valley is about thirty metres from the top of the cliffs to the river bed.
The watershed between the two major drainage regions in the area is very flat, wide and difficult to pin-point exactly ( Cogho et al, 1992). The areas south, east and north-east of the Sand River are fairly well drained, whereas the areas north and north-west of the Sand River are relatively poorly drained, due to the absence of small streams and vleis (Hudgson, 1987 and Cogho et al , 1992). The area of study is part of a well drained area in the sense that all the runoff south of the river drains into the study area (see figure 3.2), an area of approximately 41 km 2 (Steffen, Roberts and Kirsten, 1993). 16
Figure 3.7: Area north of the Convent dam (Photo: TJ Duvenhage).
3.4 Geology
The geology of the area has a direct influence on the sub-surface water, as will be discussed later in this section, as well as the surface water (Hodgson, 1987, Cogho et al, 1992 and Van Rooyen, 1994). It is therefore necessary to place the geological formations and features in perspective. A detailed account of geological formations present in the Free State Goldfields will justify a study on its own, thus only the prevailing geological formations will be discussed here.
The area under investigation is extensively covered by soils and sands of aeolian origin. Outcrops are therefore limited to surface limestone, river terrace gravel, dolerite, kimberlite and sediments of the lower Beaufort group, the upper Ecca Group and lava sediments of the Ventersdorp Supergroup (King, 1969, Beneke, 1993 and Harmony Gold Mine, 1996). In the vicinity of Harmony No 3 shaft, north of the Sand River a major fault known as the De Bron fault occurs, but is of little relevance to the study area.
To illustrate the geology a section was drawn from the Tailings Dam (Dam 33) through the study area to the Sand River (see figure 3.2). The sub-surface geology was shown down to the top of the Witwatersrand Supergroup. A second section was drawn along each line with a ten times vertical exaggeration showing the relationship between the dykes, sills, waterways and tailings dams.
This section is situated to the west of the De Bron fault and shows a relatively thick it1'3,
:4..
' IC
Os Section A - A 17 This section is situated to the west of the De Bron fault and shows a relatively thick package of sediments and igneous rock overlying the Witwatersrand Supergroup. The Karoo Supergroup has a uniform thickness of approximately 330 m, while the base of the Dwyka Formation may be up to 580 m below surface due to sill intrusion. The Platberg Group in this section is very thick, up to 700 m, and occurs mainly to the west of the De Bron fault. The Klipriviersberg Group lava are found underlying the Platberg Group, but are totally eliminated in some areas by pre-Platberg Group erosion. The VS1 forms the upper sediments of the Witwatersrand Supergroup.
The exposed rocks are fine grained, and are of the Volksrust formation and is limited in the area. To the western side, intrusions of blue grey dolerite are found. These are weathered to round boulders and are covered by a layer of yellowish gravel.
In that part of the area which is underlain by the Beaufort sediments, fragments and small outcrops of surface limestone appear frequently. Where the area is underlain by the Ecca sediments, outcrops of surface limestone also occur, but are not nearly as abundant as those found on the Beaufort sediments.
3.5 Soils
Soil is the basic growth medium of plants, so to know the soil is essential for the rehabilitation of an area (Down & Stocks, 1977 and Coppin & Richards, 1990). The exact soil forms were not identified as some subjectivity is present in such an identification process and the general characteristics of the soils are of more importance to this study. The area under investigation is extensively covered by soils of aeolian origin. Soils are mainly deep weathered paleosols with underlying calcrete duricrusts.
A - Horizon: Ortic- Mispah B - Horizon: Apedale- unspecified- Hutton
As most of the soils are of a clay origin, the pH is more acidic than that of sandy soils. A long strip of alluvium, varying from 400 m to 1.6 km in width, occurs along the Sand River. The alluvium may attain a thickness of up to 20 m.
Scattered pebbles, probably derived from river terrace gravel, have been observed in the area (Coetzee, 1960). These gravel deposits are generally found near the Sand River on a surface which is much higher than the top of the alluvium in the valley along the river. These deposits can reach a thickness of 18 m and are normally covered by red, sandy soil, averaging 1 m in thickness. The deposits are cross-bedded and different types of aggregate tend to wedge out over short distances. The deposits are unconsolidated and consist of brownish clay, yellowish silt, coarse greenish sand and lenses of pebbles. 18
Figure 3.10: Soil structure (Photo: TJ Duvenhage).
3.5 Surface water
The Water Act (Act 54 of 1956) draws a distinction between public and private water streams (Fuggle and Rabie, 1992). Private water is water that naturally rises, falls, drains or is led onto land but which cannot be used for irrigation purposes. Although several provisions of the Act regulate the use of private water, in principle the owner on whose land it is found enjoys sole and exclusive use and enjoyment of it. It was agreed, between Harmony Gold Mine and the Department of Water Affairs, that the Mine would be responsible for the Sand River up to where the lower Convent dam overflow spruit enters the River, just below the new Welkom/Theunissen road bridge. This also constitutes the point of discharge from the study area (see figure 3.2).
The Sand River originates in the Eastern Free State, near Senekal. Most of the catchment area is south of the river, with the northern water very close to the river. The town, and informal settlements occur south of the river, a major factor to be kept in mind, seeing that all surface storm water on the south side of the river drains into the Convent dam's located in the study area (figure 3.10), and eventually into the Sand River (Gillmore, 1997 & Loubser, 1996). The mean annual runoff from the catchment upstream is 86,63 million m3 ( Steffen, Roberts and Kirsten, 1993). This poses a problem due to the fact that during, and directly after, a storm the runoff peaks increase drastically and subside just as quickly, so creating flood problems for these Convent dams and leading to spillage of retained polluted waters.
19
;1;llood pEiths . t:IVolathes
bst Table 3.3: Predicted flood peaks and volumes for the Sand River (Steffen, Roberts and Kirsten, 1993).
Only the regional maximum flood peak (4030 m3/s) is available from Allemanskraal dam (station no C4N001). However, due to the catchment area, this will increase through the section affected by the mine. The main contributing factors are the runoff from the urban area, and the water pumped from underground by the Mine which is released into the Convent dams (Loubser, 1996 and Gillmore, 1997). The volume of water for the 1:100 year storm is 327,54 million m3 ( Department of Water Affairs).
An additional factor that needs to be considered is the weir in the river. Although this does not form part of the study area, the river above the weir has water in it all year round, while below the weir, surface water only flows after substantial rainfall. This factor can influence the measurements as lower water levels will lead to higher concentrations of pollutants. The two Convent dams' water levels fluctuate during dry seasons, but are constantly topped up with water that has been used in the mining processes.
3.7 Sub-surface water
Leachage and seepage can pollute the sub-surface water. It is therefore important to investigate the influence of the spill on the sub-surface water. In the Free State Goldfields, very large volumes of ground water are encountered during deep-level mining (Bekker, 1986). According to Cogho et al (1991) and Van Rooyen (1994) there are two major aquifers in the area, a shallow one usually 30-40 m below the surface but not more than 3000 m in the Ecca shales, Beaufort sandstone and shales of the Karoo Sequence, and a deep one, 150 -1800 m below surface, occurring in the fractured and faulted rocks of the Witwatersrand and Ventersdorp groups.
The original occurrence of artesian water intersections and the subsequent self-evident development of two distinctly separate water tables within the Welkom Goldfields have to be based on the existence of a persistent impervious layer (Belcker,1986, Cogho et al, 1991 and Van Rooyen, 1994). Taking into account the fact that the early artesian intersections were always associated with pre-Karoo rocks and the fact that the layers of the Ecca Group represent the only persistent stratigraphic zone in the required position, it has to be accepted that this group represents the impervious layer responsible for the development of the artesian conditions. It hardly needs to be emphasised that the relatively undisturbed, impervious, and compounded nature of the shales of the Ecca Group result in this zone being an ideal barrier between the two water systems. 20
The only possible alternative would be the Venterdorp Supergroup, but this possibility may be eliminated without reservation, on the grounds that many artesian intersections actually occurred within this zone and that layers of this group are, by far, too limited in their aerial distribution to explain the observed facts (Bekker, 1986).
3.8 Pans and wetlands
Pans commonly occur in the Free State Goldfields (De Bruiyn, 1971, le Roux, 1978, Hodgson, 1987, Marshall, 1987 and Cogho et al, 1992), their relevance related to storm water management and surface water quality. Pans occur extensively to the west of Virginia in the Ecca sediments, that is within the study area. The fact that these sediments are softer than the Beaufort sediments, and therefore weathered quicker, seems to be the main reason (Marshall, 1987 and Hodgson, 1987).
Figure 3.11: Harmony Wetland and Bird Sanctuary (Convent dam ) prior to the disaster (Photo: M de Wet).
De Bruiyn (1971) classified pans into two major groups, namely lime pans and clay pans, noting that salt pans are a major occurrence to the west and south-west. The type of pan will have a direct influence on the water quality. Origins of pans were intensively studied by authors such as De Bruiyn (1971), Le Roux (1978) and Marshall (1987) and will therefore not be discussed in detail. Due to mining activities, artificial wetlands or retention ponds are created. Because of the nature of the water that is pumped into these holding pans, these could be classified 2t artificial wetlands within the mines boundaries:
area north-west of no 33 dam area North-west side of no 32A and 32B dam area between Convent and Lower Convent dam
Most of the spill from the disaster was deposited in the area north-west of the dam and in the Convent dam. Most of the water in the area drains into this artificial pan and then into the adjacent area that is prone to pan formation. This factor should be kept in mind when drawing up the proposed Management Plan
3.9 Biotic component
3.9.1 Plants
Rutherford and Westfall (1986) class this area under grassveld common of areas underlayed by the Beaufort and Ecca groups. These areas are also well-known for the commercial cultivation of maize. According to Mostert et al (1971) and Mocks (1988), the site is situated in the area classified as sour grassveld (Cymbopogon - Themeda veld). This veld type occurs between the Vaal and the Orange rivers, as far south as Bloemfontein, mostly on sandy soils.
Figure 3.12: Natural vegetation of the area (Photo: M de Wet).
There are 15 species of plants that generally occur in the area and the total species number 150 (Mostert et al, 1971). The Karoo invasion is well under way, as patches of Karoo type vegetation are developing in certain areas. In the study area, the invasion of Karoo thorn bushes (Ziziphus mucronata and Acacia spp.) is due to the fact that these 22 areas are being overgrazed by cattle. Along the flood plains of the Sand River, thorn trees (Acacia karoo) are quite abundant.
Species of general occurrence in this veld type are: Themeda triandra, Aristida congesta, Eragrotis lehmanniana, Eragrotis suerba, Cynodon dactylon, Setaria flabellataand many more. The importance of Aristida congesta subsp. congesta, Eragrostis lehmanniana and Tragus koelerioides, shows the more arid nature of this veld. Open grasslands are found in and around the town. The vlei and pan habitats are found on the outskirts of town, surrounding bulrush and reed beds, sewage disposal work pans and three small man-made dams in the municipal game reserve on the south bank of the river.
The area has a great number of habitats ranging from riverine bush to vlei and open grassland. Riverine bush consists of indigenous Ziziphus mucronate (Buffalo thorns), Acacia karoo (Sweet thorn), Celtis africana (White Stinkwood), Combretum erythrophyllum (River Bush Willow) and exotics such as Populus spp. and Salix spp.
Salix mucronata spp. and a few Rhus species , with Olea europea subsp. africana (wild olive) on the higher cliffs overlooking the river. The base cover consists of herbaceous plants with dense scrub, largely Acacia and Asparagus spp.
Two species, the Celtis africana (White Stinkwood) and the Olea europea subs. africana (Wild olive) are protected by law in the Free State.
There are signs of thorn and karoo invasion in this dry Cymbopogon - Themeda veld. Exotic species such as Popuhts deltoides (Poplar), Gleditscia triacanthos (Honey locust), Salix babylonica (Weeping Willow), Fraxinus americana (American Ash) and Ulmus parvifolia (Chinese elm) also exist along the river.
The vegetation should be considered when the area is rehabilitated as this will influence the fauna of the area. Special attention should be given to the two protected species.
3.9.2 Animals
Seeing that the study area is situated mainly in a semi urban area it has limited animal life. However, the area is directly connected to the river and is thus still accessible to some wildlife, a fact to be considered in the proposed Management Plan. Bird life constitutes the bulk of animal species in the area. Species of bird life that have been identified include Squacco Heron (Ardeola ralloides), Lesser Gallinule (Porphyrula alleni), Helmeted Guineafowl (Numida meleagris), Orange River Francolin (Francolinus levaillantoides), Marsh Sandpiper (Tringa stagnatilis) and the Yellowbilled Duck (Anas undulata). Common butterflies are the Citrus Swallowtail (Papilio demodocus), Yellow Pansy (Precis hierta cebrene), African Monarch (Danaus chrysippus) and African Common White (Belenois creona severina).
Commonly identified animals are slender mongoose (Galerella sanguinea), yellow mongoose (Cynictis penicillata), cape clawless otter (Anoyx capensis), scrub hare (Lepers saxatilis), hedgehogs (Erinaceus frontalis), black backed jackal (Canis mesomelas) and 23 saxatilis), hedgehogs (Erinaceus frontalis), black backed jackal (Canis mesomelas) and porcupine (Hystrix afriaeaustralis).
The sungazer (ou yolk) has been classified as an endangered species in the Free State. Specimens have not been spotted in the area recently. The interaction between wildlife and the human component should be a serious consideration during actioning of the Management Plan .
3.10 Human aspects
3.10.1 Infrastructure
Nearly half the area consists of an urban component, including a population in the region of 58 000. Virginia constitutes 13 500 and Meloding an estimated 45 000, with a growth of 4%. Table 3.4 depicts the population distribution in the area.
.` `COLOURED ASIAN'- WIHTE :4 .5" 2 =Wan Maratiri Table 3.4: Population distribution (Loubser, 1996).
Harmony, Beatrix, Oryx, and H.J. Joel Mines play an important role in the development of the town's economy, as these are the major employers in the area. Most of the houses in the Merriespruit residential area belong to Harmony Gold Mine and the rest is divided between the other three mines and some private owners. Agriculture is the second biggest economic force in the area, with water being extracted from the river for irrigation and cattle watering purposes. No formal industrial sector exists in the study area, except for the Merriespruit no 1 shaft. However, the industrial sector of the greater Virginia forms a part of the greater drainage area. The industrial area mainly consists of brickworks, precast products, transport depots, scrap yards and an oil recycling depot. The runoff from these petro-chemical products can be problematic (Livingston in Hammer, 1989) and should be kept in mind when the actions for the Management Plan are developed.
Unemployment is estimated at 2080, a figure which represents only the number of people registered at the Unemployment Commissions Labour Office. The actual figures are estimated to be much higher, a major burden on the infrastructure and the eventual upkeep of the town (Cogue in Ramphele, 1991) due to an increase in pollution and runoff pollution (Miller et al, 1994).
3.10.2 Land use
The existing land use (figure 3.2) consists of
• mining structure i.e. tailings dams, evaporation areas, waste rock dumps 24
mining infra-structure i.e. hostels, workshops, plants and shaft areas additional land use for residential areas and light industries connected to the mine petrochemical recycling facility residential structures farming Correctional Services detention facility which constitutes the remaining land-use in the area.
The water authority is the Goldfields Water Board. The water consumption for Virginia and Meloding is'/- 349 megalitres per month. This will increase with the growth in population, placing a greater burden on the river during dry periods.
Owing to the fact that mining is the major activity, headgears, production plants, waste rock dumps and tailings dams are visible. During the windy months, August to October, dust from the tailings is evident and creates visual pollution.
These factors should form part of the overall planning process in establishing an EMP.
25
SECTION 4
POST-DISASTER AREA
The second aim as reflected in the problem statement is the investigation of the post- disaster area. The date used was obtained from studies done by Harmony Gold Mine, and as was mentioned in the previous section, the Mine is responsible for the Sand River as it flows through the town and therefore sampling points exist in the river. Relevant pre-disaster data on water quality that reflects on the influence of the disaster does not exist. An existing water monitoring point below the point where the study area discharges into the Sand River was used as a testing point. A reference point further upstream (Linabo Bridge) was used to draw a comparison between the water quality upstream and that at the discharge point. The purpose of this study is not to determine the quantitative influence of the spill on the area, but to determine if it had a qualitative influence on the area and how to manage any subsequent impacts.
4.1 Surface water management
4.1.1 Surface water quality
Sampling is done by Harmony Gold Mine and the Department of Water Affairs on a weekly basis. The following chemicals have been sampled for calcium, chlorine, sodium, sulphate and electrical conductivity. Due to the mining activity in this area of the Sand River, according to the CS1R (1991) and Kiely (1997), the presence of the above- mentioned chemicals can be ascribed to the following:
Calcium addition of lime in reduction process Na and CI pumping of deep ground water to surface Sulphate oxidation of pyrite in the slimes dams, rock dumps and oxidation of pyrite in older stopes underground (see Attachment A).
In the analysis of the water quality at the sampling positions, only electrical conductivity (EC) and sulphates are considered. The reason for this is that Na an CI are well represented by EC measurements and sulphates, which are not common in natural water, and is an indication of water seepage from mines.
4.1.2 Pre-disaster water quality
Figure 4.2 depicts the conductivity and sulphates taken at the Linabo Bridge sampling point and figure 4.4 depicts the conductivity and sulphates taken east of the Mosterd Canal (see figure 4.1). The figures indicate the quarterly sampling results for a 7 year period (1987 - 1993) before the disaster, in which a characterisation of the Sand River is made. The conductivity and sulphates taken at the Linabo Bridge sampling point and %Mt 4'0 rc.:1 `4N, 4: ge •
\No,
"Or." ratii7A4) * • te• 1 4'k!
••• 26 east of the Mosterd Canal taken after the disaster (1994 -1995) are depicted in figures 4.3 + 4.5.
The conductivity and sulphate concentration characteristic of the river can be summarised as follows.
- - -- Table 4.1: Summary of the conductivity and sulphates at Linabo Bridge (Harmony Gold Mine).
The peak of 110 ppm of sulphates measured during the 4th quarter of 1988 could be ascribed to the high rainfall and flooding which occurred in 1988. Referring back to figure 3.5, a total of 750 mm of rain was recorded for 1988. High sulphate content was also measured during 4th Quarter of 1989 and the 2nd Quarter of 1992.
As noted in Attachment A, sulphates are indicative of mining operations. As no mining operations exist upstream of the Linabo Bridge, and any flooding of Harmony Gold Mine operating area would only affect the river downstream of the Linabo Bridge, the source of this sulphates is unknown. The average sulphate concentration of 25 ppm in the river seems rather high. However, the most recent readings show that very little or no sulphate is present in the incoming water.
Post-disaster readings are depicted in figure 4.3 while the high sulphate content in the 1st Quarter of 1995 can be ascribed to high runoff due to high rainfall during that time. No significant increase in the sulphate can be detected following the disaster.
The conductivity and sulphate concentration characteristic of the river east of Mosterd Canal can be summarised as follows:
Table 4.2: Summary of the conductivity and sulphates east of the Mostert Canal (Harmony Gold Mine). 27
WATER QUALITY IN SAND RIVER AT LINABO BRIDGE
o Conductivity (mSm) 0 Sulphates (ppm)
YEARS FROM 1987-1993
Figure 4.2: Conductivity and sulphates taken Linabo Bridge for 1987-1993 ( Harmony Gold Mine ).
WATER QUALITY IN SAND RIVER AT LINABO BRIDGE
70
50
50
33
20 it 10 0 Conductivity (mSm) 0 -4 f 1 I > C.) Z CC CC z Ea Sulphates (ppm) C.) 0 < 12. w 0 < CO YEARS FROM 1994-1995
Figure 4.3: Conductivity and sulphates taken Linabo Bridge for 1994-1995 ( Harmony Gold Mine ).
28
WATER QUALITY IN SAND RIVER AT MOSTERT CANAL
T30 soo
150
100
SO
0 14111411141! El Conductivity (mSm)
4s To ti CI Sulphates (ppm) — 9 — — 9 — — 0 0 CO a 4./
YEARS FROM 1987-1993
Figure 4.4: Conductivity and sulphates taken Mostert Canal for 1987-1993 ( Harmony Gold Mine ).
WATER QUALITY IN SAND RIVER AT MOSTERT CANAL
1 -4- CI Conductivity (mSm) CB Sulphates (ppm) 4 YEARS FROM 1994-1995
Figure 4.5: Conductivity and sulphates taken Mostert Canal for 1994-1995 ( Harmony Gold Mine ). 29 Considering the historical data (as depicted on figure 4.4) and the more recent data . (depicted in figure 4.5), the conductivity and sulphate concentration of the river can be summarised as follows:
Table 4.3: Summary of the conductivity and sulphates east of Mostert Canal after the disaster (Harmony Gold Mine).
No marked increase can be seen in sulphates and conductivity after the disaster. What is relevant, however, is that there is a marked increase in both conductivity and sulphate from the reference point to the discharge site. From figures 4.4 and 4.5 it can be deduced that there is a marked decrease in conductivity from 1994 to 1997, indicating that the spill had some impact on the water quality but has been decreasing over time. Sulphates and conductivity are up to a hundred times more abundant at the discharge point than at the reference point. Some form of seepage occurs from the mining site to the river which can either be a result of storm water runoff or ground water pollution.
4.2 Ground water management
Mining has a negative impact on ground water (Attachment A), thus making the monitoring of ground water essential. Boreholes exist around the Merriespruit area and due to the cost of drilling new boreholes, the existing boreholes were used to establish and monitor the quality of water in the area. An electro-magnetic survey was done by Harmony Gold Mine as part of their EMPR. This data was used to determine if the ground water has been contaminated. The positions of these boreholes are indicated on figure 4.1.
A total of 30 boreholes, in and around the Harmony Gold Mine complex, were analysed for ground water quantity. The borehole depth ranged from 20 m to 70 m, with the water level (before pumping) ranging from 0,2 m to 10 m below surface. The siting of the existing boreholes allows the ground water quality to be classified into 2 distinct types, namely:
polluted/contaminated ground water
natural ground water.
A total of 6 boreholes sites are situated in the study site close to known pollution sources such as slimes dams, return water dams and evaporation areas. The following figure lists the analytical results from these boreholes: 30
GROUND WATER QUALITY
Conductivity (mSm) R3 Sulphates (mgA) 0 Chlorine (mgA) D Sodium (mgA) B/1180 BM 1 BB221 SB 1 MS4 MS 3 li Calcium (mgfi) BIRCH BOREHOLES
Figure 4.6: Analytical results from borehole sampling (Van Rooyen, 1994).
In order to establish the extent to which ground water contamination may have affected the water users in the immediate surrounding area of the mine boundary, a ground water use and water analysis survey of water users within 500 m of the mine boundary, has been conducted by Van Rooyen (1994). According to the survey the extent of the ground water contamination makes the feasibility of cleaning the entire area impossible. The possibility of a waste load allocation must be pursued by the mine to gain maximum cost benefit from remedial measures proposed. Van Rooyen (1994) also established that the Convent dam spruit area is one of the areas that needs attention, a fact that is even more relevant after the disaster, as this study was conducted prior to the spill.
4.3 Storm water
Storm water runoff is one of the major diffuse pollution sources. It is, however, very difficult to characterise storm water runoff due to widely varying contaminant concentrations ( Wanielistra et al, 1977, Duda, 1982 and Simpson, 1986). This, together with large fluctuations in runoff volume and the large number of discharge points affects limited treatment of these effluents which often contain toxic and refractory compounds (Simpson, 1986 and Meyer, 1985). The pollutants present in storm water runoff include plant material, debris, plastics, oxygen-demanding substances, suspended solids, nutrients, heavy metals, pathogens and toxic organic compounds such as pesticides and 31 petroleum hydrocarbons (Bradford, 1977, Hunter et al, 1979, Brown et al, 1985, Schmidt & Spencer, 1986)..
Storm water runoff can be divided into two broad categories namely urban runoff (formal and informal residential developments, industrial and commercial) and rural runoff (agricultural, forest etc.). The quality and quantity of storm water runoff is, to a large extent, determined by the catchment and the rainfall characteristics (see figure 3.10).
After a storm the first flush effect, which is evident as a peak of highest pollutant concentrations at the beginning of a storm event, is the result of accumulated materials being washed from the catchment surface. The rural storm water runoff is mainly from agricultural catchments and is found to contain a high suspended solids load, high iron and manganese concentrations, as well as high nutrients and pesticide concentrations. Copper, lead, zinc and petroleum hydrocarbons are predominant in runoff from urban catchments (Helsel et al, 1979, Duda, 1982, Hoffman, 1986 and Moore et al, 1988). The first flush effect seems to increase in frequency and intensity as the degree of urbanisation increases (Simpson, 1986, Brocksen & Wisniewski, 1988 and Gardiner, 1991). Of particular concern is the rapid development of the informal settlements bordering on the study area.
A further contributor to the storm water pollution is Acid Mine Drainage(AMD) 8. This increases flocculation of silt and clay and can increase the rate of precipitation, resulting in water of low turbitity. It might increase the rate of decomposition of clay minerals, feldspars and carbonates (Corbitt, 1989, Peavy et al, 1985, Rawlings, 1989 and Hammer, 1989). The magnitude and duration of acidic drainage that might occur from the tailings deposits depend on the mass of sulphite present in the dump and on the management activities employed to reduce infiltration. The tailings dam (Dam 33) in the study area is likely to generate acidic seepage or leachate because disposal of alkaline tailings has stopped. No data on the quantity of seepage or leachate from the dams is available at present.
As noted previously, pollution from storm water runoff mainly derives from diffuse sources, thus no data exists on the water quality and deductions should be made keeping the information in section 4.1 in mind. Exact data on the quality and floodpeaks is not available. There is, however, consensus that storm water runoff poses a problem in the area as all the storm water drains into the Convent dams (Loubser, 1996 and Gillmore, 1997).
According to Corbitt (1989) runoff quantities can be calculated with the formula Q = CIA (where Q = the runoff quantity, m 3/s, C = locality coefficient, I = intensity of the rainfall, mm/h and A = catchment area, km 2). Kiely (1997) proposes the following rational method to be used in urban sewer design whereby Cv = volumetric runoff coefficient and CR = routing coefficient, thus the formula becomes more site specific. The runoff for the site can be calculated using the following data:
8 Acid Mine Drainage and the influence of mining are discussed in detail in Attachment A 32 Cv = 0.75 (Kiely, 1997) CR = 1.3 (Kiely, 1997) I = 134mm/h (Table 3.1) A= 41 km2 (Section 3.5)
Thus Q, = 0,278 CV CR IA (Kiely, 1997) = 0,278x0.75x1.3x 0,037 x 41 000 m3/s =411 m3/s = 1,5x10 6 1/h
In the space of an hour the runoff from the study area will be 1,5x10 61 yet the capacity
of the Lower Convent dam is 57 x10 6 / and the Convent dam is 223 x106. 1 The dams are used to contain contaminated mine water for the mine's water circuit and are usually more than 70 percent full. It is thus clear that the holding capacity of the dams is insufficient to retain the runoff as well as the mine water. Inevitably, the dams will overflow, discharging the polluted water into the natural system and eventually into the river.
4.4 Dust
A dust survey was conducted, as part of Harmony's EMP, using CIP 10 gravimetric samplers. The samplers were placed in strategic residences and were allowed to sample for a period of 480 minutes. The mass of dust assessed was not excessive and, after analysis, indicated an air quality index of between 0,1679 and 0,1971. This is not excessive.
During the initial cleaning phase after the spill some dust was evident due to the mechanical nature of the cleaning process. The full sampling results could not be obtained for this period.
During the windy months high ambient dust load is a norm in the Goldfields area mainly as result of aeolian erosion of the tailings dams. The spill that consists of tailings is of a small particle nature and is very susceptible to aeolian erosion. No monitoring has been done during times when the wind blows more frequently as indicated in Section 3. No other air pollution is evident in the area.
4.5 Waste management
The tailings from the spill that has been recovered were deposited in the area between dams 33 and 32 (see figure 3.2). This area is seen as a natural wetland area, as noted in section 3.8. Data on waste management is not available and qualitative assessments were made in relation to the original natural environment. Some areas are still covered by tailings from the spill. 33
4.6 Aesthetics and land use
The change in topography regarding tailings dams and waste rock dumps that exist in the area, poses a permanent visual impact. The damage to the existing infrastructure has been addressed to a certain extent but is still not aesthetically acceptable as some wins still form part of the area (see figure 4.7).
The agricultural potential of the area is restricted to grazing, due to the soil formation. The soils are mainly of the Mispah forms and have moderate mechanical limitations for agriculture. As the area is situated in a tributary, any form of agrithltural activity should be limited. This potential is further limited by the thorn bush (Acacia kazoo) invasion in the area. Most of the area is overgrazed and removal of the thorn bushes is a positive option. According to the capability classes proposed by the Chamber of Mines of SA (Chamber of Mines, 1981) where an area is divided into arable, grazing wetland and wilderness, the only option left would be to develop the area into a wetland.
Figure 4.7: Visual pollution in the area (Photo: M de Wet).
34
SECTION 5
MANAGEMENT PLAN
5.1 Management goals
In the aims and objectives of this study, it was stated that an EMP with management goals would be established. The main aim is to rehabilitate the area as closely as possible to the original state by addressing water management (surface and ground water), storm water management, waste management, dust, fauna and flora, aesthetics and socio-economics.
5.2 Project actions
Each of the goals or objectives is to be actioned. It must thus be established what action is required to achieve the set goals and objectives. The project actions could be one or more of the following:
construction of a facility (wetland) development of specific operational procedures for carrying out certain activities (to remove tailings and redeposit it) specific siting of activities policy setting with respective authorities (not internal organisational policy) specialist studies may be required monitoring requirements liaison with the authorities in connection with information on the project.
5.3 Surface water management
Objective
Minimising the risk of discharging contaminated water, so as not to have a significant impact on the fitness of water for use downstream.
Action
Water and water management are strictly regulated issues in South Africa (Fuggle & Rabie, 1992). However, better planning before the establishment of a mine is required to prevent possible pollution by e.g. evaporation areas situated on a shallow dolerite sill, as well where large quantities of water are disposed in unconsolidated deposits. Prevention of such negative effects can reduce the pollution hazard considerably (Cogho et al, 1992). Measures have been taken by the Mine to keep the water clean. In the vicinity of the Tailings Dam itself; all storm water is diverted from sites where mining activities take place and all rainfall (up to and including the 100 year storm) that falls on the site is contained. The water can only be discharged if it meets statutory requirements. Water budgeting at mines is problematic, since in dry periods, the mine uses all excess water, but 35 during wet periods rain water can cause problems as shown in section 4.3. It is clearly indicated, that even with moderate rainfall the Convent dams' capacity is exceeded. Kelly (1988) notes that this can be avoided if the mine does better planning.
Exceeding the storage capacity, the uncontrolled discharge is deposited into the Sand River. The existing water course and plant life restricts discharge of excessive pollutants into the Sand River. However, the natural environment will inevitably reach a saturation point. Biotic control through the establishment of a wetland will prevent this while further monitoring should also be undertaken to ensure that limits are not exceeded.
In the study area, the existing Convent dam system receives most of the waste water in urban storm runoff and mine water south of the Sand River. The water in the Convent dams is low in pH and high in salt content and needs to be contained in the dams for the natural processes to take place. Expanding the existing wetland area and introducing controlled biotic processes (Hammer, 1989) can increase the evaporation ponds and create a much-needed aesthetically pleasing area. This area was originally set aside in the early nineties as a bird sanctuary under the guidance of Mr John Wesson, the then Head of Parks and Recreation of Virginia (Muniviro, 1994). Natural pan formation is an acceptable occurrence in the Free State Goldfields (see section 3.8), thus a wetland will fit into the natural topography.
Surface water sampling positions for which existing water quality is known should be selected in order to determine a background reference for water quality. Water quality compliance limits, control limits and action limits 8 for the selected positions should be set for a period of three years and the results should be reported to the Regional Director of Water Affairs as well as Minerals and Energy. Guidelines for water quality and vegetation monitoring are currently being developed by the Department of Water Affairs.
5.4 Ground water
Objective
Maintaining the existing ground water quality. Improving existing ground water quality.
Action
Cogho et al (1990) and Van Rooyen (1994) have concluded that the ground water levels closely follow the surface topography, thus implying that the water levels are topographically controlled and that vertical leakage to the deeper aquifer is negligible. Furthermore, south and north-east of the Sand River, the slope of the water levels is locally towards the Sand River. This corresponds very well with the surface water drainage in the vicinity of the Sand River.
Van Rooyen (1994) noted that in the western and northern sides of the area, where the
8 See Attachment B for the compliance, control and action limits given by the Department of Water Affairs. 36 Van Rooyen (1994) noted that in the western and northern sides of the area, where the drainage of surface water is very poor due to the absence of well-defined watercourses, the regional ground water gradient is toward the west. Therefore, it seems to correlate well with the drainage of the surface water. It can, therefore, be concluded that although the local ground water movement is toward the Sand River, the general direction of ground water movement is toward the west.
The evaporation areas (the Convent dams) pose the largest water pollution potential in the area due to the large quantities of unpotable water which are stored in them daily. The residual impacts are those associated with land occupied by evaporation facilities (see figure 3.2) and overflows from these cause local ground water contamination. The impact of dewatering the mine should be limited to an area already contaminated. Cogho et al (1992) suggested the storage of excess water in old mines that are not in use. This option is not viable due to the fact that most of the mines in the area are connected, posing a potentially hazardous situation.
The impermeable nature of the shallow Karoo aquifer prevents that ground water, beyond 1 km from a pollution source, has been affected by seepage from these evaporation areas (Van Rooyen, 1994). However, overflows from the evaporation facilities caused by rainfall pose a pollution threat in this area. Van Rooyen (1994) suggested that, in the areas where the ground water is contaminated, trees with a high evapo-transpiration rate, such as bluegums, should be planted to reduce ground water flow. As shown in studies (Duvenhage et al, 1992) and emphasised by the Department of Water Affairs, the use of exotic trees tend to cause problems along watercourses and in terms of the overall availability of ground water in an area.
A constructed biotic wetland would be a more viable alternative. As the discharges of the redundant sewerage work's aeration dams are deposited in the same stream, a constant water level can be attained even in dry periods. This water that is already at the disposal of the mine authorities will not force the mine to buy extra water from the Goldfields Water Board.
The current legislation concerning mine, municipal and industrial effluent standards seem to be sufficient to control water pollution. The legal implementation of the standards might, however, be lacking. One issue that might be problematic is that according to the Water Act (Act 54 of 1956), wetlands will normally only contain private water. This means that the sole and exclusive rights to this water belong to the owner. However, in terms of section 59(2) the Minister may declare land on either side of a channel or stream, or any other area situated in the catchment of such a stream, to be a catchment control area, thus limiting the owner's exclusive use of land on which such a wetland occurs.
Water control facilities that are already in place should be maintained and monitored so as to continue achieving the objective.
9 In 1958 the Merriespruit No 1 Shaft was flooded when an aquifer was opened. Currently all the shafts belong to Harmony Gold Mine and are interlinked (Gillmore, 1996). 37
5.5 Storm water control
Objective
Minimising the risk of discharging contaminated water into the river, to prevent a significant impact on the water quality downstream. Minimising the risk of discharging contaminated water and maintaining water control structures and vegetation for a period agreed upon.
Action
Containing storm water (up to and including the 1:100 year storm event) that falls on the site within the mine water circuits. These circuits include tailings delivery, the pool on top of the tailings dam, penstock water, underdrain discharges, return water trenches, pumps and pipe lines and the paddocks. In times of heavy rainfall, excess water may have to be stored on top of the dams and slowly let off at a rate that the return water pumps can manage. Only excess water that meets statutory demands should be discharged. Vegetation should be established on the tailings dam itself
The storm water runoff from the residential area seems to be a major problem since high rainfall intensities led to difficulty in containing water in the Convent dams due to high peak flows (see section 4.3). The catchment area above the Convent dam is +/-41square kilometres. The 1:50 year storm water, from the area would be +/-90 cubic metres/second. Diverting this magnitude of storm water would be impossible because it meets up with the trench below dam 32 A, B, C that carries water, via the Bluegumhoek trench, from Virginia shaft and plant area to the Convent dam.
An alternative suggested by SRK Consulting Engineers was the construction of a canal to divert the runoff from the Meloding area directly into the Sand River. This option might create temporary work, but the problem is that urban storm water runoff is more toxic and detrimental to water quality due to oil and petroleum hydrocarbons, than the possible increase of salt contents due to mining operations.
According to Gilbert (1989) an alternative to channelisation is 'natural' river engineering whereby wildlife conservation and natural beauty are enhanced. These techniques involve the construction of low flood banks set back from the river, which allow for overflow onto washlands used for recreation or nature conservation and retention ponds where pollutants can be filtered by the natural vegetation. The processes of bioleaching by means of bacterium such as Thiobacillus ferrooxidans can also take place in these ponds (Hammer, 1989). Flood peaks can be controlled and would be a much better option than a concrete channel.
Not only are T. ferrooxidans able to fix CO2, it can fix its own nitrogen from the atmosphere. This means this organism is able to grow in a totally inorganic environment such as that found in mining. Air provides the carbon source (CO2), nitrogen source (N2) and electron acceptor (02), while water is required as the medium of growth and the ore 38 (e.g. pyrite FeS) provides the energy source and trace elements (Rawlings, 1989). These bacteria are easily washed away in streams, thus pond systems together with surface wetting agents such as detergents, enhance the process (Kelly, 1988). An artificial wetland that will intercept urban runoff which contains soluble detergents will be beneficial to the process.
Furthermore, storm water needs to be controlled on the rock dumps to prevent sulphate pollution in the area. Trenches separating storm and polluted water would have to be maintained until seepage from polluted areas and tailings dams become insignificant. The principle is to keep contaminated and non-contaminated water separate. The primary option is recycling of all contaminated water (or effluent) away from the Convent dams, by maintaining dam level controls and monitoring seepage, which can be done through the construction of diversion berms and containment paddocks.
The Convent dam area is in the process of being hosed out so as to reclaim gold washed into the Convent dam. A dredge is used to enlarge the capacity of the dam but this process has been stopped due to cost implications (Gillmore, 1997). The suggestion of constructing a wetland area north of the lower Convent dam made in section 5.4 might be a solution. An extra dam can assist with retaining water-holding capacity, as water can be pumped to the wetland that can be used as a temporary holding facility.
The water quality should be monitored for a period agreed on, the results to be reported to the Regional Director of Water Affairs as well as Minerals and Energy.
The existing main sewer line between Virginia and the new Sewage Works seems to cause a'major problem as areas exist where overflow from the line has spilled into the natural environment. The pipeline and its flow should be controlled and monitored in the area. This monitoring needs to be done by the Municipality to the satisfaction of the Department of Water Affairs.
5.6 Waste management
Objective
Removing the waste that was deposited in the area and finding a suitable way of redepositing it.
Action
The waste was mostly removed mechanically and dumped in the area between tailings dams 33 and 32, in the wetland area (see section 3.8 and figure 3.2). This was done as an intermediate measure after the disaster but this deposit is not satisfactory as it covers a big area with little control. The tailings should be deposited in Dam 22 that is being reclaimed, so that the treated tailings can be pumped onto an existing tailings dam that has better control measures. The area covered by the tailings spill should be rehabilitated to a natural grassveld (see section 3.9). Some of the spill is still evident in the area and, seeing 39 that it will be difficult to remove these leftovers mechanically, it should be recovered manually. Depending on the grade and quality, the tailings could be recycled through the gold plant to extract any gold to be pumped into an existing tailings dam. Alternatively it can be deposited on Dam 22 that is being reclaimed.
Seeing that the pH of the tailings is low and that South African soils are by nature acidic (Taljaard, 1997), lime application would be necessary. Sampling positions should be identified, established and monitored for a period of three years. The results must be reported to the Regional Director of Agriculture as well as Minerals and Energy.
5.7 Dust
Objective
Reducing the risk of air pollution caused by fugitive dust.
Action
During the mechanical cleaning operations dust is a problem and no samples were taken during the previous cleaning-up operations. Dust samples are to be taken by the Mine and Department of Health during removal of the temporary waste dump between dams 33 and 32. The prevailing wind direction is in an easterly direction and the fact that the waste dump is situated away from a residential area should not pose a problem. Road traffic should be restricted to defined access ways and further spillage of material during the cleaning operations should be prevented. Access ways should be watered if necessary.
Some of the tailings from the spill are still evident in the area and should be cleaned. Sampling should be done during the cleaning of areas that are still covered by the tailings. The amount of tailing still left is minimal, thus it would not pose a major dust problem. The wind direction relevant to the area that is being cleaned should be taken into consideration, as some of the sites are situated near residential areas. These sites to be cleaned are situated south and west of the residential areas. Some sites might need to be cleaned manually as they are situated in the existing stream bed and are not easily reachable by means of machinery.
Vegetating the exposed slopes can control air pollution due to dust from existing tailings dams. An extensive research program in this regard is being done under the guidance of the Chamber of Mines. The import of exotic species during this phase can lead to invasion of the natural watercourses so limiting the availability of water and should thus be stringently controlled.
Air pollution caused by fugitive dust could be minimised by watering exposed surfaces where possible and restoring plant species to their original state. The damaged area should be vegetated with natural grass species to cover exposed areas and soil pH should be monitored. Sampling positions should be identified and monitored for a period of three years. The results are to be reported to the Regional Director of Agriculture as well as 40 Minerals and Energy. The Department of Health should also monitor the area periodically during this time.
5.8 Aesthetics and socio-economic implications
Objective
Rehabilitating the area to its former natural state and creating an area that is aesthetically pleasing.
Action
It was mentioned in the introduction that the overall appearance of the town has been greatly affected. One of the most prominent factors mentioned in the Strategic Communication Plan (Pronam,1994) is the negative impact of the disaster on the public image of the town and the Mine. One of the major concerns from the community was the possibility of a re-occurrence of the disaster although major resources such as economics and man power have endeavoured to repair the dam wall, so making it safe. In the process, some essential factors have been overlooked. As this phase is nearing completion, more attention will hopefully be given to ecological and aesthetic aspects since the environment still seems to be dismal and tends to create a morbid, gloomy atmosphere.
The overall mental health in the town does not seem to be very good (Pronam, 1994). The town has the dubious honour of the highest per capita juvenile suicide in the country. Alcohol misuse is of the highest in the country and can perhaps be ascribed to the lack of facilities and the high risk element that is experienced during work hours. The Strategic Communication Plan drawn up in March 1994 noted the fact. One of the aims to counteract this is to enhance the appearance of the town since creating an aesthetically pleasing environment leads to a community with a definite sense of belonging (Bellah et al 1985).
This need was recognised in the early fifties by the then chairman of the Anglo American Corporation, Sir Harry Oppenheimer (Pim, 1969). This led to designed parks and the aesthetically pleasing work done by the first Landscape Architect in South Africa, Joane Pim. This is also recognised by other major companies, resulting in the inception of landscaped office parks.
Although the Strategic Communication Plan (Pronam,1994) mentioned that Business, Town Council and Mine companies are positive towards nature conservation, it does not seem as though any real effort towards instituting a nature conservation ethic is executed. This is evident in the open areas under Council and Mine management that are unkempt and while the major owner of property in the area is the Mining Companies, the overall appearance of the properties is one of untidiness and neglect. The mine properties are unkempt, the most probable reason being that money is spent on production and not on enhancing the aesthetic appearance of the environment. This is contradictory to the beliefs 41 expressed by authors such as Pim (1969), McHarg (1969), Bentley et al (1985) and Trancik (1988).
Additionally, this created a negative attitude especially on the part of developers in the town. The Convent dam area was developed as a bird sanctuary prior to the disaster. Although it has been cleaned out, no real effort in terms of redeveloping the area has been made.
Bentley et al (1985) suggests a few guidelines for successfully redeveloping an urban area, including:
permeability - accessibility variety - uses of a site robustness legibility visual appropriateness richness of variety personalisation.
The vegetation of the slopes of the tailing dams is a major step that can be taken by the mine to enhance the appearance and prevent major dust pollution of the area. According to Gilbert (1989) a 'natural' river environment, through which wildlife conservation and natural beauty are enhanced, can make a positive contribution to the experience of the area. Although the Sand River runs through the town, accessibility to the river is very restricted. The creation of a wetland park with braai facilities should therefore be considered (Muniviro, 1994). As the town's people already use the Convent dams as fishing spots this attribute can be developed further to create an established and well- maintained park.
The mining companies posses the necessary infrastructure to create a positive image for the town as a whole. The disaster ironically evoked sympathy and distributed knowledge about the town, which could in itself be used to project a positive image of reconstruction. As reconstruction seems to be the buzzword in South Africa, it can be utilised to the advantage of the towns public image. The negative aspects of the disaster received extensive media coverage yet the same media can be utilised to create a positive image of reconstruction in the area.
The construction of an artificial wetland can assist with the creation of job opportunities, firstly in the construction phase and then in harvesting reeds from reed beds that can be used in basket weaving (Wood & Hensman in Hammer, 1989). 42
5.9 Rehabilitation
Objective
Re-establishing the natural vegetation in the area.
Action
During rehabilitation the principle should be to restore the land-use capacity. The land- use mainly consists of grazing, wetland and natural vegetation.
Due to the high compaction of the area during the cleaning-up process, the soil should be ripped with a single tooth ripper to a depth of at least 500mm. The surface area should be left uneven to assure water accumulation and retention during rainfall. The tailings deposits are low in pH and high in sulphates, some having already leached into the soil. Most soils in South Africa are acidic (Taljaard, 1997), and this specific topsoil is very sterile so requiring a great deal of fertiliser. Two tons of lime, 200kg LAN and 800kg of 2:3:2 should be added to one hectare. The usual procedure is sowing the area with grass seed harvested from adjacent areas. In this case, however, the area adjacent to the natural vegetation can be left after soil preparation since wind will help with the establishment of vegetation. The area south of the residential area (see figure 3.2) should be sowed with a mixture of harvested seed.
Maintenance of the area will be limited to protecting the area from fire in the first two or three years so ensuring the establishment of the vegetation. The vegetation should be monitored and bare patches sowed during that stage. Within the space of five years no extra maintenance would be necessary.
After establishing the vegetation, regular monitoring needs to be undertaken to determine the progress in the vegetation. The soils should be monitored constantly to ensure that the pH is not too low. This might require some maintenance in the sense of adding lime to the area. The micro-elements in the soil should be monitored as the low pH leads to the dissociation of some micro-elements. The soils might have different rates of nitrogen, potassium and phosphorus and should be monitored constantly for the first few years. The ideal is to bring the phosphorus levels to 30 parts per million and the potassium to 120 million parts per million. Fertilising should not be too heavy as the area needs to become self-sustainable in the long run.
Establishing a higher vegetation density will aid both water retention and storm water management. The species Celtis and Olea are protected by law in the Free State and special attention should be devoted to assuring that these species are not disturbed or destroyed during the rehabilitation. 43
SECTION 6
CONCLUSION
Mining impacts cannot be eradicated, nor is it desirable to do so since it is through conflicting situations that humanity finds new meaning for its existence. There is no specific human ecology. There is only one ecology, and humans are part of it, most often as destroyer yet ultimately as victim. History has not recorded one advanced society reversing itself; thus it will be foolhardy to assume that a new primitivism is our answer. We should rather use present technology and experience to halt further destruction of the ecosystem, as most of the damage done is irreversible. We should strive to contain it, rather than remaining and acting ignorant and furthering the process of destruction (Iltis, 1970).
This report is summarised in the previous sentence in that, although the mine paid due attention to the disaster's economic and social impact, the impact odthe environment was and is neglected. Some intermediate actions were taken to prevent further pollution of the area. Being intermediate at the time, no such rehabilitative actions are exercised at present. The primary aim of this study is therefore to compile an EMP in order to manage and possibly mitigate the physical impact of the disaster on the immediate environment.
The scope of this paper is to contemplate the notion of formulating an Environmental Management Plan (EMP), through considering both the pre-disaster and the post-disaster environments of the Merriespruit Disaster area. The aims are 2-fold:
to investigate the physical area (pre-disaster) to investigate the post-disaster area (the effects on the natural environment) with the objective of establishing an EMP that will assist with the rehabilitation of the Merriespruit area. The plan identifies problems and establishes management goals and remedial action plans necessary to achieve the set goals.
The physical environment and its composition were subsequently investigated and formed the basis from which deductions were drawn. The resulting identified problems were translated into the following specific management goals or objectives:
Water management
In any mining environment water management is problematic. In the present context some form of pollution occurred mainly from the evaporation areas. The answer to the problem lies in containing the water and making use of natural systems to aid the purification process. The responsibility for the actions should be taken by the Mine. 44 Storm water control
In a mining environment storm water poses some problems and in the context of this study even more so, as most of the urban runoff also enters the evaporation system. Containing the water in a wetland and making use of natural systems to aid the purification process will alleviate the problem. The responsibility for the actions should be taken by the Mine and the Municipality.
Waste management
Water management , might be problematic in a mining environment, waste even more so. In this study area, pollution occurred mainly because of the tailings in the area and those that were removed. The temporary dump created for the tailings should be monitored and relocated in the most economical way. The Mine should take responsibility for these actions.
Dust
Dust pollution does not seem to be problematic. The Mine should, however, pay more attention to vegetating the tailings dams and establishing natural vegetation in the area.
Aesthetics and socio-economic implications
Although this study did not focus on the social impacts of the disaster, there is a major problem in this area which could be addressed by enhancing the environment. Further studies should be undertaken by the Mine. Co-operation between the Mines, the Municipality and the residents could help to address these issues.
Rehabilitation of the area
The rehabilitation of the area seems to take place slowly as most of it is due to the natural processes re-establishing themselves. Some extra work should be done by the Mine to help speed up the process of establishing natural vegetation in the area.
In summary, Harmony Gold Mine can benefit from the compilation of this EMP, as management goals were set and feasible means of achieving them were specified. This study is unique, and thus essential, as this disaster, like any other, is a single occurrence with a major impact on the environment and with results and conclusive deductions that simply cannot be obtained under any other circumstances. By following the procedure stipulated in the Management Plan it can be ensured that Environmental Management requirements will be effectively integrated into either the project management actions and contracts or operational systems and processes. 45
"Henceforth, the laws to govern us must be the laws of ecology, not the laws of a self- destructive laissez-faire economics. And what the laws of ecology say is that we, we fancy apes, are forever related to, forever responsible for this clean air, for this green flower decked, and fragile earth" (Otis, 1970). 46
REFERENCES
Acocks, J.H.B. 1988: Veld Types of South Africa. (3rd ed.) Pretoria: Botanical Research Institute.
Alexander, C. 1965: A City is not a Tree. Architectural Forum 122, April 1965, 56-62.
Atchia, M & S. Tropp, 1995: Environmental Management: Issues and Solutions. New York: John Willey & Sons.
Ballot, J.S. 1991: Challenges for Environmental Management in South Africa. Pretoria: CSIR.
Barnard, J.J. 1990: Die Effek van Besoedeling vanaf die Mynbedryf soos gesien deur die Department van Waterwese. Report for the Council of the Environment. Pretoria: Council of the Environment.
Beale, J.G. 1980: The Manager and the Environment. Oxford: Pergamon.
Bekker, E.H. 1986: Fossil water and the associated gas mixture within and below the Ventersdorp Supergroup in the Harmony mine area, Welkom Goldfields. Mineral Deposits of South Africa. Johannesburg: Chamber of Mines
Bellah, R.N., R. Madsen, W.M. Sullivan, A Swidler & S.M. Tipton; 1985: Habits of the Heart: Individualism and Commitment in American Life. Berkley: University of California Press.
Bentley, I., A. Alcock, P. Murrain, S. McGlynn & G. Smith, 1985: Responsive Environments: A Manual for Designers. London: Architectural Press.
Berger, J. J. (Ed), 1990: Environmental Restoration: Science and Strategies for Restoring the Earth. Washington: Island Press.
Best, H.J. 1987: Management and industrial waste management: state of the art. Proceedings of the International Conference on Mining and Industrial Waste Management. Johannesburg: SA Institute of Civil Engineers.
Boonzaaier, J.M. & I.J. van der Walt, 1997: EMP a futile exercise? Mining World, Jan, 33.
Bradford, W.L. 1977: Urban stormwater pollutant loading: a statistical summary through 1972. Journal of Water Pollution Control Federation. 49(4).
Brocksen R.W. & J. Wisniewski (Ed), 1988: Restoration of Aquatic and Terrestrial Systems. Dordrecht: Kluwer.
BroWn, R.C., R.H. Pierse, & S.A. Rice, 1985: Hydrocarbon contamination in sediments from urban stormwater run-off. Maritime Pollution Bulletin. 16(6). 47
Burcell, R.W. & D. Listokin, 1975: The Environmental Impact Handbook. New Jersey: Rutgers.
Bruton, M..N. & F.W. Gess (Ed), Towards An Environmental Plan For The Eastern Cape. Port Elizabeth: University of Port Elizabeth.
Clark, M. & J. Herington, 1988: The Role of Environmental Impact Assessment in the Planning Process. London: Mansell.
Clark, W.C. 1989: Managing Planet Earth. Scientific American, 261. Sept, 46-54.
Cock, J & E. Koch (Ed), 1991: Going Green: People, Politics and the Environment in South Africa. Cape Town: Oxford University Press.
Coetzee, C.B. 1960: The Geology of the Orange Free State Goldfields. Geology Survey Memoir, 49. Pretoria: Government Printer.
Cogho, V.E., L.J. van Niekerk, H.P.J. Pretorius & F.D.I . Hodgson, 1992: The Development of Techniques for the Evaluation and Effective Management of the Surface and Ground water Contamination in the Orange Free State. Water Research Council Report No K5/224. Pretoria: Water Research Commission.
Coppin, N.J. & I.G. Richards, 1990: Use of Vegetation in Civil Engineering. London: Butterworths.
Cooper, E. & J.A. Palmer, 1992: The Environment in Question: Ethics and Global Issues. London: Routledge.
Corbitt, R.A. (Ed), 1989: Standard Handbook of Environmental Engineering. New York: MacGraw-Hill.
Council for the Environment, 1989: a) Integrated Environmental Management in South Africa. Pretoria: Government Printer.
Council for the Environment, 1989: b) A Framework for Harmony between Development and the Environment. Pretoria: Government Printer
Cozby, P.C. 1985: Methods in Behavioural Research (3 ed.) Palo Alto: Mayfield.
Croal, P. 1995: Training seminar: Scoping. Environmental Assessment and Compliance Division. Canadian International Development Agency.
De Bruiyn, H. 1971: 'n Geologiese Studie van die Panne in die Westelike Oranje Vrystaat. University of the Orange Free State, Unpublished M.Sc. thesis.
Department of Environmental Affairs, 1991: a) Integrated Environmental Management: Guideline series no.l. Pretoria: Government Printer. 48
Department of Environmental Affairs, 1991: b) CSIR Program for the Environment. Pretoria: Government Printer.
Department Minerals and Energy, 1992: Aide-Memoire. Pretoria: Government Printer..
Down, C.G. & J. Stocks, 1977: Environmental Impact of Mining. London: Applied Science Publishers.
Dud; A.M. 1982: Municipal point source and agricultural non-point source contribution to coastal eutrophication. Water Resources Bulletin. 18(3).
Duvenhage, T.J., S. Glen, D.F. Theron, P. J. Theron & J.N. van Nieuwenhuizen, 1992: Sabie Catchment. Report to Water Research Commission. University of Pretoria. (Unpublished).
Dyer, R. 1977: The effects of mine acid on the Pond River watershed in western Kentucky. Water Resources Bulletin, 13(5), 1070-1075.
ESRI, 1998 (http://nutria.esri.com/cgi-bin/wb?c=26.488804,-28.015559&p=4&m=0&cd=0)..
Freer, J.S. 1993: Integrated Environmental Management in the Mining Industry. The Journal of the Institute of Mining and Metalluty, 93(1), Jan, 21-24.
Fuggle R.F. & M.A. Rabie, 1992: Environmental Management in South Africa. Cape Town: Juta.
Funke, J.W. 1990: The Water Requirements and Pollution Potential of South African Gold and Uranium mines. Water Research Commission Report KV 9/90. Pretoria: Water Research Commission.
Gardiner, J.L. (Ed), 1991: River Projects and Conservation: A Manual for Holistic Appraisal. Chichester: John Wiley & Sons.
Gilbert O.L. 1989: The Ecology of Urban Habitats. London: Chapman and Hall.
Gillmore, R. 1997: Personal Corrununication.
Glasson, I, R., Therivel & A. Chadwich, 1994: Introduction to Environmental Impact Assessment. London: UCL.
Hammer, D.A. (Ed), 1989: Constructed Wetlands For Wastewater Treatment: Municipal, Industrial and Agricultural. Michigan: Lewis. •
Harmony Gold Mine, 1996: Environmental Management Program Report (Unpublished)
Harmse, J.T. Riglyne vir die Skryf van Referate in die Departement Geografie. RA. U. Internal publication Department Geography, Rand Afrikaans University. 49
Helsel, D.R., J.I. Kim, T.J. Grizzard, C.W. Randall & R.C. Hoehn, 1979: Land-use influences on metals in storm drainage. Journal of Water Pollution Control Federation. 51 (4).
Hodgson, F.D.I. 1987: Report on the water quality in the Welkom Area. Report to Free State Consolidated Gold Mines (Operations) Ltd Pretoria: Department of Water Affairs.
Hoehn, R.C. & D.R. Sizemore, 1977: Acid mine drainage (AMD) and its impact on a small Virginia stream. Water Resources Bulletin, 13(1), 153-159.
Hoffman, E.J., 1986: Urban run-off pollutant input to Narragansett Bay: Comparison to point sources. Perspectives on non-point source pollution. Proceedings of a National Conference, Kanses City. Missouri.
Hoffman, S. & V. Houseman. 1990: Update on regulations for mine waste management under RCRA. Mining Engineering, 42, Nov, 1242-4.
Holmes, G., B.R. Singh & L. Theodore, 1993: Handbook of Environmental Management and Technology. New York: John Wiley & Sons.
Hugo, M.L. & A.T. Viljoen, 1992: Hulpbronbewaring: 'ri Ekologiese Perspektief. Pretoria: University of Pretoria.
Hunter, J.V., T. Sabatino, R. Comperts & M.J. Mackenzie, 1979: Contribution of urban run- off to hydrocarbon pollution. Journal of Water Pollution Control Federation. 51( 8).
Institute of Mining and Metallurgy. 1992: Minerals, Metals and the Environment. London: Elsevier.
Iltis, H.H., 1970: The Paradox of Human Ecology. Bio Science. 20(14).
James, A.L. & M. Mrost, 1965: Control of acidity of tailing dams and dumps as a precursor to stabilisation by vegetation. Journal of South African Institute of Mineral and Metallurgy, 70. 488-495.
Jones, B & J. Pearce, 1996: White-collar killers walk free. Mail & Guardian. March (http://www.mg.co.za/mg/news/96mar/22mar-slimesverdict.html)
Joughin, J., A. Smithen & S. Posnik, 1997: The Environmental Management Puzzle. Mining World May, 28-29.
Kelly, M. 1988: Mining and the Freshwater Environment. London: Elsevier.
Kempster, P.L., W.H.J. Hattingh & H.R. Van Vliet, 1982: Summarised Water Quality Criteria. Hydrological Research Institute. Technical Report No TR 108. Pretoria: Department of Environmental Affairs.
Kiely, G. 1997: Environmental Engineering. London: McGraw-Hill 50
King, L. C. 1963: South African Scenery: A Textbook of Geomorphology. Edinburgh: Oliver & Boyd.
Le Roux, J.S. 1978: The origin and distribution of pans in the O.F.S. S.A. Geograaf 6, 167- 176.
Loubser, J.T. 1996: Personal Communication.
Lucas, D. 1997: Development and Integration of Environmental Management Plans (EMP 's). (An unpublished paper presented to RAU students). Johannesburg: Escom.
Maberly, A. & R Goss, 1990: Mammals of Southern Africa. Johannesburg: Jonathan Ball.
MacVicar, C.N. (Ed), 1991: Grondklassifikasie: 'n Taksonomiese Sisteem vir Suid Afrika. Pretoria: Department of Agriculture.
Marinov, U. & D. Sandler, 1993: The Status of Environmental Management in Israel. Environmental Science & Technology, 27, July, 1256-62.
Marsden, D.D. 1986: The current limited impact of Witwatersrand gold-mine residues on water pollution in the Vaal River system. Journal of South African Mineral and Metallurgy. 86 (12). 481-504.
Marshall, T.R. 1987: Morphotecnic analysis of the Wesselsbron panveld. South African Journal of Geology. 90 (3), 209-218.
McHarg, I. 1969: Design with Nature. New York: Natural History Press.
Meyer, J.L. 1985: A detention basin/artificial wetland treatment system to renovate stormwater run-off from urban, highway and industrial areas. Wetlands. 5.
Miller S.E., C.W. Shinn & W.R. Bentley, 1994: Rural Resource Management for the Long Term. Arkansas: Iowa State University Press.
Moore, L.W., H. Matheny, T. Tyree, D. Sabatini & S.J. Klaine, 1988: Agricultural run-off modelling in a small West Tennessee watershed. Journal of Water Pollution Control Federation. 60 (2).
Mostert, J.W.C., B.R. Roberts, C.F. Heslinga & P.G.F. Coetzee, 1971: Veldbestuur in die OVS-Streek. Pretoria: Government Printer.
Motloch, J.L. 1991: Introduction to Landscape Design. New York: Van Norstrand Reinhold.
Mrost, M. & P.J. Lloyd, 1970: Bacterial oxidation of Witwatersrand slimes. LA.E.A Symposium on the Recovery of Uranium from its ores and other sources held in Sao Paulo. Vienna: International Atomic Energy Agency.
Muniviro, 1994: Virginia bewaar die vlei. Muniviro 11 (15). 51
Naveh, Z. & A.S. Lieberman, 1990: Landscape Ecology: Theory and Application. New York: Springer-Verlag.
Naiman R.J. (Ed), 1992: Watershed Management: Balancing Sustainability and Environmental Change. New York: Springer-Verlag.
O'Riordan, T. & R.D. Hey, 1976: Environmental Impact Assessment. London: Saxon
O'Sullivan, D. 1991: Science needs for setting environment policy listed. Chemical & Engineering News.69 (9) Dec, 6.
Park, C.C. 1980: Ecology and Environmental Management. London: Butterworths.
Peavy, H.S., D.R. Rowe & G Tchobanoglous, 1985: Environmental Engineering. New York: McGraw-Hill.
Petulla, J.M. 1987: Environmental management in Industry. Journal of Professional Issues in Engineering. 113, April, 167-83.
Pickering , K.T. & L.A. Owen, 1994: An Introduction to Global Environmental Issues. London: Routledge.
Pim, J. 1969: Beauty is Necessary: Preservation or Creation of the Landscape. Cape Town: Purnell.
Preston, G. 1993: Integrated Environmental Management: Will it work? African Environment and Wildlife. 1(1). 31-35.
Prinsloo, J.H.B. 1997: Environmental Management Plans vs. Rehabilitation. Unpublished lecture, presented at Research Institute for Reclamation Ecology. Potchefstroom: Potchefstroom University for CHE.
Pronam Partnership, 1994: Strategiese Kommunikasieplan vir die Merriespniitramp. Bloemfontein.
Ramphele, M. (Ed), 1991: Restoring the Land, Environment and Change in Post-Apartheid South Africa. London: Panos.
Rau, J.G. & D.C. Wooten, 1980: Environmental Impact Analysis Handbook. New York: McGraw-Hill.
Rawicz, M. 1991: A Guideline in Corporate Environmental Management for Directors and Managers.
Rawlings, D.E. 1989: Microbes, Mining and the Growth of Knowledge. Cape Town: University of Cape Town.
Richter, C. 1993: Rehabilitation and the Minerals Act 1991. South African Mining, Coal, Gold & Base Minerals. Feb. 25-28. 52
Rutherford, M.C. & R.H. Westfall, 1986: Biomes of Southern Africa: An Objective Categorisation. Pretoria: Botanical Research Institute.
Sadar, M.H. 1996: Environmental Impact Assessment. Ottawa: Carlton University Press.
Samiullah, Y. 1990: Prediction of the Environmental Fate of Chemicals. London: Elsevier,
Schmidt, S.D. & D.R. Spencer, 1986: The magnetite of improper waste discharges in an urban Stormwater system. Journal of Water Pollution Control Federation. 58 (7).
Simmons, I.G. 1991: Earth, Air and Water: Resources and Environment in the late 20th Century. London: Edward Arnold.
Simonds, J. 1983: Landscape Architecture: A Manual for Site Planning and Design. New York: McGraw-Hill.
Simpson, D.E. 1986: A Study of Runoff Pollution from an Urban Catchment. M Sc (Ing) Thesis, Department of Civil Engineering, University of Natal.
Sinclair, I., P. Hockey & W. Tarboton. 1993: Sasol Birds of Southern Africa. Cape Town: Struik.
Smit, G.J. 1993: Navorsing: Riglyne vir Beplanning en Dokumentasie. Pretoria: Southern.
Spalding, H. 1995: Evaluating Environmental Impact Assessment: Approaches, Lessons and Prospects. Environment Year. 2, 63-74.
Steffen, Robertson and Kirsten, 1993: Preliminary study of pollution of the Sand River. Report No 192062/2.
Steffen, Robertson and Kirsten, 1994: Rossway Quarry. Environmental Management Programme Report. Report 208705.
Steffen, Robertson and Kirsten, 1996: West Driefontein Mine: Environmental Management Programme Report. Report No196369/E.
Taljaard, C. 1997: Grondsuurheid 'n Nasionale Ramp. Mielies/Maize. July, 49-50.
Terblance, J.C. 1996: Personal communication
Thomas, B.A. 1970: Planning of Gold Mines to Minimise Water Pollution Problems at Closure. Convention: Water for the future.
Tietenberg, T. 1994: Environmental Economics and Policy. New York: HarperCollins.
Trancik, R. 1997: Finding Lost Space: Theories of Urban Design. New York: Van Nostrand Reinhold. 53
Tricart, J. & C. KiewietdeJonge, 1992: Ecography and Rural Management: Contribution to the International Geosphere- Biosphere Program. New York: Longman.
Van Niekerk, H. 1998: Personal Communication.
Van Rooyen, K.C. 1994: Groundwater Investigation for Harmony Gold Mine. (Unpublished Internal Report).
Van Zyl, W.J.S. 1997: EMP's-a quantum leap ahead. Mining World, Feb, 32-34,.
Wali, M.K. (Ed), 1992: Ecosystem Rehabilitation: Preamble to Sustainable Development. Vol. 2: Ecosystem Analysis and Synthesis. The Hague: SPB Academic Publishing.
Wanielistra, M.P., Y.A. Yousef & W.M. McLellon, 1977: Non-point source effects on water quality. Journal of Water Pollution Control Federation. 49(3).
Wenmonth, B.A. 1994: Quality systems and environmental management. Journal of Environmental Health.57. Nov, 15-20.
World Bank 1995: Environmental Assessment: Challenges and Good Practice. Environmental Management Series, Paper No. 018. Washington D.C.: World Bank ATTATCHMENT A
THE MINING ENVIRONMENT
1 Mining
Generally, deep underground mining methods have little or no direct effect on the environment. Indirectly, environmental impacts are associated with mine residue deposits and acid mine drainage due to the fact that huge amounts of water (155 000 m 3/ day) are pumped to surface to the ensure safe mining activities.
Unlike other minerals, gold is a stable substance with no environmental impacts (Wells et aL in Fuggle & Rabie, 1992). However, it occurs in only minute quantities in its ore bodies and thus has to be extracted from them in sophisticated plants with the use of complex metallurgical processes.
Once the few grains from each ton mined have been extracted, the rest of the material is a waste product which creates very large residue deposits on the surface. Pyrite is associated with all but alluvial gold-bearing ores (Burke, 1999). The pyrite is not normally extracted with the gold and therefore ends up in the residue deposits where it constitutes a significant source of water pollution (Cogho et al., 1992). If iron pyrites, FeS2, are exposed to air and water, they oxidise to form sulphuric acid, iron oxides and hydroxides (yellow boy), which causes the pH to drop to about 4. The oxidation reaction is accelerated and extended by bacteria, the most important being Thiobacillus ferroxidans. According to authors such as Rawlings(1989) and Marsden (1989) the pH of the solution can drop as low as 1.5.
The gold present in the ore is of such small quantities that that the ore has to be ground to the consistency of face powder to make extraction possible (Metcalf; 1996). Thus the slime going to the residue deposits is extremely fine and when it dries can cause severe dust pollution.
Cyanide used in the gold-extraction process is treated with extreme caution at the reduction works. Fortunately cyanide entrained in the slime which is pumped to the tailing dams, is broken down by sunlight into non-toxic compounds and therefore does not cause an impact (Metcalf; 1996).
1.1 Waste Management
The mining sector is the single largest generator and accumulator of solid wastes in South Africa. Mine tailings and coal discard accounted for 74 per cent of the total waste stream in 1990 (CSIR, 1991). The total area of land covered by mine residue was 10 700 ha in 1981, most of which is in the Transvaal and Free State (Best, 1987). The total mass of 2
mineral ore removed by mining operations is estimated to be 294.35 million tons per annum, 40,8 per cent of which arises from the gold and uranium sector (CSIR, 1991).
1.1.1 Rock Dumps
The acid produced reacts with bases in the country rock or residue deposits to form salts and to mobilise heavy metals that may be contained in the rock residue. During this reaction the acidity is often neutralised. The resulting drainage, referred to as acid rock drainage (ARD), contains elevated levels of salts (primarily calcium and magnesium sulphates) and metals (mainly iron, manganese and aluminium) (Marsden, 1989).
The main types of pollution resulting from residue deposits are water and air pollution. Visual pollution and change of land-use can, however also be significant.
1.1.2 Tailing Dams
The mineral extraction processes from past mining activities were not as effective as modem methods. Tailings generated still contain payable values of minerals. Dump reclamation refers to the reprocessing of the dumps. Typically, the material in the old dump is monitored (squirted with high pressure jet of water which erodes the dump material away) into a sluice. The sluice gravitates the dump material to a low point where it is collected and pumped to the treatment plant. The main environmental protection activities during reclamation are to keep stormwater away from the working areas, to prevent rainwater that has fallen on the site from leaving it in an uncontrolled fashion and to prevent dust pollution during dry, windy conditions.
Once the whole dump has been reclaimed down to the original soil level under the dump, reclamation stops and rehabilitation of the site begins. The options available for different- land-uses on these sites are varied. In many cases the top part of the soil profile at these sites are contaminated with acid water seepage from the dump. This has to be ameliorated with agricultural lime. Radon seepage from the dump is found commonly in the soils but the concentration is usually below dangerous levels after rehabilitation (Fuggle & Rabie, 1992). If buildings are constructed on these sites, special ventilation may be required to prevent high levels of radon gas accumulating within the closed building.
The main residual impact of reclaimed precious- metal dumps is not at the site of the reclaimed dump but rather the new slimes dam which has to be build to accommodate the same volume of material that was in the original dump.
Detailed design and construction of tailing dams are discussed in the appendices. One of the main problems common to tailing dams other than acid drainage, is air pollution by dust particles. Many methods have been tried, establishing a vegetation cover on the dump or slimes dam is the most practical and cost effective long-term solution (Fuggle & 3
Rabie 1992). Steep slopes, an acid and salty growing medium in gold dumps, high phreatic water table emerging on the side, a low nutrient content in almost all deposits and seasonal fires burning the established vegetation are the main challenges, all of which have been overcome to greater or lesser extent in the mining industry's almost 40 years of experience.
A very effective temporary measure to control dust on fine-grained, flat surfaces, such as the top of a gold slimes dam, is to ridge plough them with a potato ridger. The low-level wind turbulence induced by the ridge causes dust to be lifted from the crests of the ridges and to be deposited immediately in the adjacent valleys.
1.1.3 Waste Water
Mentioned earlier the Free State Goldfields produce great amounts of underground water that are pumped to surface. According to Authors such as Wells et al. in Fuggle & Rabie (1992) and Cogho et al. (1992) the average conductivity of the water is about 450 mSncl ( A total dissolved solids content of about 3 000 mg/1), the main salt being sodium chloride (NaCI). Due to this fact of the high salt content the water, it can not be used for drinking or irrigation and although some is used for gold processing, the bulk is therefore discharged to evaporation dams.
This water is at present disposed of in evaporation areas resulting in ground water pollution in many of the areas The volume is estimated at between 80 to 100 Mid.
In regard to mine water treatment, while SO +/- is the main or predominant contaminant of mine service water in most mining areas, high sodium and chloride concentrations are the main constituents of the underground water on the Evander and Free State gold fields. This aspect is referred to by Funke (1990). This is a brackish water with low sulphate content but high levels of sodium and chloride.
ATTATCHMENT B
WATER QUALITY MANAGEMENT POLICIES AND STRATEGIES IN THE RSA, DEPARTMENT OF WATER AFFAIRS AND FORESTRY, APRIL 1991
GENERAL AND SPECIAL EFFLUENT STANDARDS
GOVERNMENT GAZETTE 18 MAY 1984 NO 9225
REGULATION No. 991 18 MAY 1984
REQUIREMENTS FOR THE PURIFICATION OF WASTE WATER OR EFFLUENT
By virtue of the powers vested in me by section 21(1)(a) of the Water Act, 1956 (Act 54 of 1956) I, Sarel Antoine Strydom Hayward, in my capacity as Minister of Environment Affairs and Fisheries, hereby prescribe the following requirements for the purification of waste water or effluent produced by or resulting from the use of water for industrial purposes.
1 SPECIAL STANDARD
Quality standards for waste water or effluent arising in the Catchment area draining water to any river specified in Schedule 1 or a tributary thereof at any place between the source thereof and the point mentioned in the Schedule, in so far as such Catchment area is situated within the territory of the Republic of South Africa.
1.1 Colour, Odour or Taste
The waste water or effluent shall not contain any substance in a concentration capable of producing any colour, odour of taste.
1.2 ti
Shall be between 5,5 and 7,5.
1.3 Dissolved Oxygen
Shall be at least 75 per cent saturation
1.4 Typical (Faecal) Coli
The waste water or effluent shall contain no typical (faecal) coli per 100 millilitres.
5
1.5 Temperature
Shall be a maximum of 25 C.
1.6 Chemical Oxygen Demand
Not to exceed 30 milligrams per litre after applying the chloride correction.
1.7 Oxygen Absorbed
The oxygen absorbed from acid N/80 potassium permanganate in 4 hours at 27° C shall not exceed 5 milligrams per litre.
1.8 Conductivity
1.8.1 Not be to increased by more than 15 per cent above that of the intake water.
1.8.2 The conductivity of any water, waste water or effluent seeping or draining from any area referred to in section 21(6) of the aforementioned Water Act shall not exceed 250 milli-Siemens per metre (determined at 25 C).
1.9 Suspended Solids
Not to exceed 10 milligrams per litre.
1.10 Sodium Content
Not to be increased by more than 50 milligrams per litre above that of the intake water.
1.11 Soap, Oil or Grease
None.
1.12 Other Constituents
1.12.1 Constituents:
1.12.2 The waste water or effluent shall contain no other constituents in concentrations which are poisonous or injurious to trout or other fish or other forms of aquatic life. 2 SPECIAL STANDARD FOR PHOSPHATE
Waste water or effluent arising in the Catchment area within which water is drained to any river specified in Schedule II or a tributary thereof at any place between the source within the territory of the Republic of South Africa shall not contain soluble orthophosphate (as P) in a higher concentration that 1,0 Milligrams per litre.
3 GENERAL STANDARD
Quality standards for waste water or effluent arising in any area other than an area in which the SPECIAL STANDARD is applicable, as described in paragraph 1.
3.1 Colour. Odour or Taste
The waste water or effluent shall not contain any substance in a concentration capable of producing any colour, odour or taste.
3.2 pH
Shall be between 5,5 and 9,5.
3.3 Dissolved Oxygen
Shall be at least 75 per cent saturation
3.4 Typical (Faecal) Coli
The waste water or effluent shall not contain any typical (faecal) coli per 100 millilitres.
3.5 Temperature
Shall be a maximum of 35 C.
3.6 Chemical Oxygen Demand
Not to exceed 75 milligrams per litre after applying the chloride correction.
3.7 Oxygen Absorbed
The oxygen absorbed from acid N/80 potassium permanganate in 4 hours at 27° C shall not exceed 10 milligrams per litre.
7
3.8 Conductivity
3.8.1 Not to be increased by more than 75 milli-Siemens (determined at 25) above that of the intake water.
3.8.2 The conductivity of any water, waste water or effluent seeping or draining from any area referred to in section 21(6) of the aforementioned Water Act shall not exceed 250 milli-Siemens per metre (determined at 25° C).
3.9 Suspended Soils
Not to exceed 25 milligrams per litre.
3.10 Sodium Content
Not to be increased by more than 90 milligrams per litre above that of the intake water.
3.11 Soap, Oil or Grease
Not to exceed 2,5 milligrams per litre.
3.12 Other Constituents
3.12.1 Constituents
3.12.2 The sum of the concentrations of the following metals shall not exceed 1 mg/1: Cadmium (as Cd), chromium (as Cr), copper (as Cu), mercury (as Hg and lead (as Pb).
3.12.3 The waste water or effluent shall contain no other constituents in concentrations which are poisonous or injurious to humans, animals, fish other that trout, or other forms of aquatic life, or which are deleterious to agricultural use.
4 METHODS OF TESTING
All tests shall be carried out in accordance with methods prescribed by and obtainable from the South African Bureau of Standards, referred to in the Standards Act, No. 30 of 1982, as listed in Schedule III. NOTE
Further information and elucidation may be obtained from the Director General: Environment Affairs, Private Bag X313, Pretoria, 0001.
Government Notices R. 553 of 5 April 1962, It 969 of 22 June 1962 and R. 1567 of 1 August 1980 are hereby withdrawn.
NOTE
In addition to the gazetted requirements for the purification of waste-water or effluent, there are informal guidelines which are of considerable value in assisting decision-making regarding effluent disposal.
Lucher, J.A., ed. 1984, Water Quality Criteria for the Coastal Zone, South African National Scientific Programme, Report No. 94, Dec. 1984.
In this document water quality criteria reflecting the needs of the South African coastal zone are presented for beneficial uses encompassing maintenance or ecosystems, recreation, ocean migration, edible and non-edible resources, desalination, mineral exploitation, industrial purposes and miscellaneous uses.
Kempster, D.L. et al., 1982, Summarized Water Quality Criteria, Department of Environmental Affairs Hydrological Research Institute, Technical Report TRIOS.
The purpose of this report is to provide an overall picture of the present world opinion (1980) as regards the desirable limits for water quality constituents.
Drinking water guidelines: Guidelines with respect to drinking water are given in the South African Bureau of Standards publication, SABS 241-984.