THE EFFECT OF THE CHEMICAL PROPERTIES OF TAILINGS AND WATER APPLICATION ON THE ESTABLISHMENT OF A VEGETATIVE COVER ON GOLD TAILINGS DAMS
Report No 899/1/04
FEBRUARY 2004

EXECUTIVE SUMMARY

The core economy of South Africa has developed around the mining industry but the finite quantity of an ore body places a limit on the life of a mine. As the viability of mines begins to decline, mine closure becomes more and more a reality, as do the liabilities associated with mine closure. One of the liabilities is the rehabilitation of areas disturbed by mining or associated activities. Regulatory standards and guidelines for the rehabilitation of tailings dams are becoming increasingly prescriptive as regards procedural and technical requirements. Sustainability is a closure requirement, however, no guidelines for the assessment of the sustainability and/or rehabilitation success are provided by mining legislation in South Africa.

Although the successful rehabilitation of gold tailings dams goes back as far as the early 1960's there is still uncertainty with regards to the sustainability of the vegetation cover established using different rehabilitation methods. Uncertainty with regard to the sustainability of re-vegetated systems stems from:

• The concept of sustainability is as difficult to define as it is to put into practice and qualifiers for rehabilitation methods such as "ecological", "natural recovery", and "self-maintenance", are often used interchangeably with sustainability, causing even more confusion
• Over emphasis of certain vegetation attributes such as diversity of species, grazing potential etc. while ignoring the function that the vegetation cover has to full-fill (the main function of a vegetative cover on gold tailings is to reduce air pollution (dust) and to reduce water pollution (sediment and salt load).
• Lack of knowledge with regard to vegetation succession on gold tailings dams

Objectives of the study

The primary objective of this study was:

• To determine which commonly used rehabilitation methods for gold tailings produced the most sustainable vegetation cover at the end of the rehabilitation period

• To quantify the water needed for the different rehabilitation methods

Methodology

A literature survey had been conducted with regard to- the properties of gold tailings, the environmental impacts associated with tailings dams, rehabilitation methods, methods to evaluate rehabilitation projects, sustainability of vegetated areas and regulations regarding closure.

The following rehabilitation methods most commonly used in South Africa for re-vegetating tailings dams, were selected for evaluation:

• Leaching method
• Revised leaching with lime filled trenches on contour
• Elementary amelioration
• Minimum cultivation with topsoil, and
• Chemical amelioration to eliminate the need for leaching.

Plots on the south-facing slope of the old President Steyn number 6 tailings dam at Welkom, were vegetated in accordance with the method applicable to a particular rehabilitation method. For each of the methods the following was recorded:

• Quantity of irrigation water used
• Type and amount of ameliorants
• Type and amount of fertilisers
• Seed cocktail (species and amount of seed)

The following were monitored for each method over the period November 1998 to March 2002:

• Chemical attributes of the tailings
• Species abundance, basal cover, and root development
• Soil microbial activity

To assess the "tailings quality" and the "vegetation quality" obtained in this study it was decided to compare them with the "tailings quality" and "vegetation quality" of tailings dams rehabilitated in the past, using different methods, which currently sustain a good vegetation cover.

The following tailings dams were selected representing the different methods of vegetation establishment on gold tailings dams:

1. Leaching method (selected by Mr Cook and Mr Dawson)
• 1L1 East Champ D'or (vegetated in the early 1960's)
• 2L3 Consolidated Main Reef (Fleurhof- vegetated late 1960's)
• 3L44 City Deep (Wemmer Pan-vegetated in the late 1970's)
• 4L48 East Rand Proprietary Mines (ERPM Dam 1-vegetated in the mid 1990's)
• 4L50 ERPM Dam 4-vegetated in the mid 1990's
• 2L24 Durban Roodepoort Deep -mid 1990's
2. Chemical amelioration (selected by Mr van Deventer)
• Stilfontein no 5 -vegetated in the mid 1990's
• Saaiplaas -vegetated in 1996
• PS 5- vegetated in 1999
• Oryx -vegetated in 1997
• 2L24 DRD-vegetated 1999
3. Others
• PS 3-vegetated 1996 (revised leaching)
• Beatrix -vegetated mid 1990s (topsoil)

Discussion and conclusions

Criteria used to evaluate the sustainability of a vegetation cover include, diversity of species, dynamic stability, and productivity. The productivity of plants depends largely on climate and the fertility of the growing medium. Tailings fertility is related to:

1. The physical properties of the tailings which influence the amount of water that can be held in the profile and the ease with which roots can penetrate, and
2. The tailings chemistry (pH, EC, and nutrients).

The dynamic nature of re-vegetated systems requires data not only on the initial conditions but also over time to monitor the system's evolution.

The amelioration of tailings, in accordance to the different rehabilitation methods, has changed the following chemical attributes for the 0-30cm layer significantly, pH, K, Mg, Ca, Al, CEC, and P. Liming had the largest effect on the chemical attributes of the tailings as it caused an increase in pH, Ca, and Mg. The increase in pH in turn caused Al to precipitate and the CEC to increase. The measured active and reserve acidity (H+ + Al3+) before amelioration indicate that 3 tons lime per ha was needed to neutralise it. But the results obtained indicate that almost 13 tons of lime per hectare was needed to neutralise the active and reserve acidity. This large discrepancy can be explained by the presence of iron sulphate minerals in the tailings that also contribute to reserve acidity. The standard agricultural laboratory tests to determine reserve acidity must be modified for material containing iron sulphate minerals to be more applicable for rehabilitation purposes.

After amelioration, a comparison between the leaching methods indicates that none of the chemical attributes differs significantly, while a comparison between the chemical amelioration methods indicate that only the EC and K differ significantly. When the leaching methods are compared with the chemical amelioration methods the chemical attributes that differ significantly were pH, Al, and CEC. However, all methods had after amelioration produced a growing medium that is suitable to sustain a vegetation cover. Six weeks after amelioration none of the chemical attributes within the 30-60 cm layer had changed significantly for any of the rehabilitation methods.

The application of compost had increased the water holding capacity of the 0-30 cm layer significantly, and had an additional effect on the CEC. The compost had however, no effect on the EC of the tailings, which is in contradiction with laboratory studies. For this study it was found that the EC of the saturated paste was not a good indicator of the salt content of the tailings. This could be explained by the formation of the ion pair CaSO40, which is important in solutions containing more than 1000 mg/l sulphate, (on average more than 2000 mg/l sulphate for this specific tailings) as more than half the dissolved calcium could be present in the form of the CaSO40 ion pair. This neutral ion pair had no influence on the electrical conductivity of the saturated paste, and the net result of the ion pair is that electrical conductivity measurements underestimate the ion concentration and total dissolved solids.

The phosphorus status of the tailings after amelioration is on the high side if compared against the phosphorus level of the adjacent natural veld. Potassium did not increase in relation to applied fertilisers because it either was leached from the tailings or become fixed through the precipitation of jarosite.

The change in chemical attributes in the period after amelioration to the end of the study period was very similar between the different rehabilitation methods although the magnitude of change differed. At the end of the study period the only chemical attribute for the 0-30 cm layer that differed significantly between the leaching methods was Ca. Between the chemical amelioration methods the drip irrigation method differs in EC, Ca, and Cl from the other irrigation methods. This difference can be explained in terms of the amount of possible leaching that could take place, between the different irrigation systems. The chemical attributes that differ at the end of the study period between the leaching and chemical amelioration methods were mainly EC, Ca, and Cl. The chemical amelioration methods had at the end of the study period, on average, a higher salt content compared to the leaching methods.

The chemical constraints that the tailings of the experimental site imposed on vegetation establishment were not very severe because the elementary amelioration method that was neither a leaching method nor a chemical amelioration method was just as successful for establishing vegetation and there were no extreme differences in chemical attributes relative to the other rehabilitation methods.

A comparison between the chemical attributes of tailings dams rehabilitated in the past with that of this study indicate that the chemical conditions of the experimental site had at present a better overall fertility. The only negative elements were Na and Cl, which were higher for the experimental site. If the fertility can be maintained then all the rehabilitation methods were successful in turning the tailings into a suitable growing medium to sustain a good vegetation cover. However, it is foreseen that the tailings of the chemical amelioration plots will become more saline in future although the pH will remain more or less neutral. On the other hand the tailings of the leaching methods will become more acidic but the salinity would be relatively constant.

Cone penetration studies conducted during the study period indicate that plant roots can exploit the upper 60 cm of the tailings once the chemical constraints that limit root development were removed.

The initial seeding involved the sowing of a mixture of 16 species for the leaching methods and the sowing of a mixture of 8 species for the chemical amelioration methods (five species were common to both methods). At the end of the study period 8 of the sown species remained of a total number of 39 species that have been identified on the site at the end of the study period. However, the three dominant plant species, in order of dominance, at the end of the study period were Dactylis glomerata, Medicago sativa, and Eragrostis curvula for the leaching methods and Medicago sativa, Eragrostis curvula and Dactylis glomerata for the chemical amelioration methods.

The results obtained from tailings dams rehabilitated in the past reveals that in the early stages of succession (up to 5 years after re-vegetation) the plant community is dominated by Cloris gayana, at a later stage Eragrostis curvula become the dominant species, to be follow by a Hyparrhenia hirta dominated community which can be regarded as the secondary succession climax on tailings dams. In this study Chloris gayana dominate only in the first year after re-vegetation. The fast disappearance of Chloris gayana was attributed to the fact that the experimental site was located on the south facing slopes of the tailings dam (Chloris gayana still dominated the north facing perimeter wall of the experimental plots at the end of the study period). Although Eragrostis curvula was not the dominant specie at the end of the study period, this specie had steadily increased over the study period to be the subdominant specie for all chemical ameliorated methods. At the end of the study period the Hyparrhenia hirta occurring on the experimental plots were those established by hand planting.

It is anticipated that in the near future the vegetation will have to adapt to changes in the chemical conditions of the tailings, which is expected to become more saline for the chemical ameliorated plots while the leaching plots will become more acidic. The neutral pH and higher salinity conditions for the chemical amelioration plots is more compatible with chemical conditions occurring in soils of dry climates and therefore natural colonising of the plots by grass species from the surrounding veld is foreseen to be favoured towards the chemically amelioration plots.

At this stage (three years after rehabilitation) no conclusive statement can be made regarding the sustainability of the vegetation cover for any of the rehabilitation methods. However, results from tailings dams rehabilitated according to the different methods in the past indicate that stable plant communities can be obtained. The leaching method had the longest track record with stable Hyperhenia hirta communities in excess of 40 years old on certain tailings dams.

Water is used in rehabilitation methods to establish vegetation and in the leaching methods also for reclamation (leaching salts and acids out of the root zone). In this study the water inputs in terms of rainfall and water applied through irrigation were recorded. The exact amounts of water that leached through the root zone could only be estimated roughly.

Where a micro-jet irrigation system was used, about twice as much water was used in the leaching method compared to the chemical amelioration method. The amount of water applied for a chemical amelioration method, under sprinkler irrigation, was about the same as that of a leaching method under a micro-jet irrigation system. The least amount of water was applied under a drip irrigation system, i.e. three times less than for a leaching method under micro-jet irrigation.

The experimental site was unfortunately on a tailings dam where the conditions were favourable for the establishment of vegetation. Furthermore the experimental site was on a south-facing slope, which is wetter than a north, west, or east, facing slope. The beneficial role of water in terms of vegetation establishment or as a reclamation agent was therefore not highlighted in this study.

Further studies and recommendations

During this study some of the problems that become apparent were:

• The sustainability of re-vegetated tailings dams, even with a proven stable plant community in excess of 40 years, are questioned, only due to a lack of diversity in the plant community
• The degradation of vegetated tailings dams is driven by a combination of forces such as lack of funds, low biological productivity, lack of knowledge, lack in ability or desire to maintain the biological productivity, and poorly defined or inadequate tenure systems. Unfortunately, any vegetation failure, irrespective of the reason for failure, is used as an example to point out that re-vegetation efforts are unsustainable.

The problem experienced over the years is that once vegetation has been established the mining companies viewed it as a walk away solution and see little value in maintenance and monitoring as part of the rehabilitation process. Hopefully this attitude will change, as the application for closure of a tailings dam has to be substantiated with data to prove that the rehabilitation measures taken are effective.

Future research needs to address biological productivity/diversity, and tenure systems. Tenure systems (land management) should be evaluated and tested against the five pillars of sustainability, namely, productivity/services, security/resilience/risk, protection, viability, and acceptability.

Recommendations with regard to future research are:

1. Models: If the ACRU-model and the Salmine-model, or any other water balance and geochemical model, can be modified and integrated then it can become an important tool in the rehabilitation of tailings dams for:
• Estimate the water needs for leaching and vegetation establishment
• Estimate the amount of erosion (sediment yield) under different plant covers and or erosion control measures
• Estimate the effect of different rainfall scenarios on vegetation performance
• Estimate the runoff from vegetated tailings for determining the size of paddock structures and return water structures
• Estimate the chemical attributes of the tailings for different rehabilitation methods and climatic scenarios
• Evaluating "what if " -scenarios, and
• Risk assessments.

 

2. Monitoring: Monitoring is an important aspect but is a relative new endeavour (at least for tailings material) and the necessary indicators, minimum data sets, procedures, thresholds, and standards have still to be developed.
    
   Re-vegetation monitoring on the other hand comes a long way but the exact vegetation parameters and goals (ecological condition assessment, pro-active management systems, evaluation of the rehabilitation process, etc) have yet to be defined.
    
   Monitoring of environmental impacts should also be addressed.
    
   
3. Other research:
• Methods to effectively build up potassium to favourable levels in tailings should be investigated
• Soil organic matter dynamics in tailings and their relationship with other chemical/physical attributes should be investigated
• Methods to address the physical limitations of tailings need urgent attention
• The fate of additional lime to counteract potential acidity should be investigated over a long period
• Vegetation succession or state and transition models for rehabilitated gold tailings dams have to be developed because such models are essential for land management of re-vegetated systems and to make more conclusive statements regarding sustainability of the vegetation cover, when application for closure is made.
• A thorough understanding of the chemistry and mineralogy of iron-bearing minerals such as pyrite, jarosite, schwermannite, ferrihydrite, and goethite, is essential to understand the chemical behaviour of gold tailings.
• Certain standard methods of soil chemical analysis should be modified to ensure that the results are more applicable for the chemical amelioration of gold tailings.
• A search for other plant species adaptable to gold tailings conditions is required to ensure a higher plant diversity

Recommendations

• Every rehabilitation project should be monitored, preferable by tertiary and research institutions, to ensure that a better understanding of re-vegetated ecosystems can be gained, so that a closure application can be substantiated with relevant data
• Environmental scientists, researchers, rehabilitation specialists, the mining industry and the regulators should work together to develop a framework for assessing sustainability as well as for the sustainable management of re-vegetated tailings dams.