i 
 
 
A comparative evaluation of two 
rehabilitation options employed to 
rehabilitate industrially generated 
elemental seepage on a coal mine  
 
 
MC Visagie 
orcid.org/0000-0002-0460-5184 
 
 
 
Dissertation submitted in fulfilment of the requirements for the 
Masters degree in Environmental Sciences at the North-West 
University 
 
 
Supervisor:        Prof MS Maboeta 
 
Graduation May 2018 
21145482 
 
 
 
 ii 
 
ABSTRACT 
Mining has a very long history in South Africa and together with the mining activities 
comes environmental degradation. The mining sector makes use of a number of 
different procedures and products that causes this degradation of the environment. One 
of the products harming the environment which are critical for mining activities to 
commence is explosives. The environment is protected by law in South Africa; it is 
required that the environment must be rehabilitated after the mining of an area has 
stopped. Similar environmental laws are also applicable to other industries that can 
cause harm to the environment.  
Land in South Africa, and all over the world, is an important asset for ecosystems and 
people alike. This makes the rehabilitation of degraded land and ecosystems of the 
utmost importance. In most cases where land was degraded by human activities, 
human intervention is needed to rehabilitate the area. This is mostly to speed up the 
process and to make sure future harm to the environment is minimalised. Scientific 
research is undertaken in order to set up guidelines and procedures for the 
rehabilitation of degraded sites. The aim of this research is to ensure that each site, 
with its site specific problems, has the chance to be rehabilitated to the best possible 
state. 
Environmental impacts created by human activities were not always a concern and in 
numerous historical cases the environment was degraded without any thought of the 
implications. As an example, the effluent from industries was often not disposed of in an 
environmentally friendly way; this is discussed further in this study.  
In this study, an explosives manufacturing company (EMC) leased land from a coal 
mining company for a period of twelve years. The explosives manufacturing plant was 
erected on top of a watershed within the borders of the coal mining site. Effluent from 
the manufacturing plant was disposed of by means of a “soak-away” which entails 
dumping the effluent and wash water into the soil where the contaminants penetrates 
the soil and soaks away. This resulted in a series of problems with the soil fertility and 
caused the vegetation on the site to deteriorate and eventually die off completely which 
left the area denuded from any vegetation in certain areas.  
 iii 
 
After an assessment of the area, a company specializing in the rehabilitation of 
degraded land drafted a rehabilitation plan. This plan is explained and examined in this 
study. The rehabilitation and monitoring was done by the same company that compiled 
the remediation plan and the site was deemed to be successfully rehabilitated in 2008. 
The success of the rehabilitation however was short lived (approximately 2 years), 
since the site again showed signs of denudation in 2010. Intervention was needed once 
again and a site visit was conducted in 2010 (by a different rehabilitation company) after 
which a different rehabilitation plan was drawn up and implemented.  
This study entails the evaluation of the different rehabilitation procedures implemented 
by the two companies in question and explains the different views of each. The main 
objective of each of the rehabilitation companies was to rehabilitate the site to the 
predetermined state were neither the coal mine nor the EMC faces any liability in terms 
of environmental law. In the case of the first company, this was attempted by removing 
the contaminants from the growth medium after which the soil was ameliorated with 
different biological inoculants and chemical ameliorants. The site was then reseeded for 
vegetation establishment with the long term goal of creating a self-driven vegetation 
cover. 
The site was rehabilitated and monitored after the second rehabilitation attempt and all 
the findings indicate that the rehabilitation of the site was successful. The second plan 
entailed removing the contaminants which the first attempt failed to remove due to a 
possible lack of understanding regarding the underlying geology of the site. The second 
procedure also used better amelioration techniques that provided a better result. 
 
Key words: Rehabilitation, denuded, contamination, drainage, amelioration, vegetation. 
 
 
 
ACKNOWLEDGEMENTS 
1 Thessalonians 5:18 
 iv 
 
“Give thanks in every circumstance, for this is God’s will for you in Christ Jesus.” 
Firstly I would like to thank my Heavenly Father for giving me the opportunity, strength 
and an able mind to complete this degree and study. 
I would like to sincerely thank the following people for their contributions and support 
that made this study possible: 
Prof Mark Maboeta, for his guidance and insight in the study. Without his support and 
supervision this would not have been possible. 
Dr. Leon van Rensburg for the opportunity of coming aboard the project and also his 
guidance and insight. 
David Hamman for his help and insight on the project. 
The staff of Eco Analytica, especially Mrs. Yvonne Visagie for the analysis done for the 
project. 
Dr. Jaco Visagie for his support and help with the statistical analysis during late nights 
and early mornings. 
My parents, Kobus and Yvonne Visagie for their love, support and constant nagging to 
complete the study. 
Salomé van Heerden for the support and having the power to make or break while 
choosing to make. 
My brothers, Jaco and Johan for their support. 
Jaco and Jeanette Visagie for the editing. 
Kobus and Nikki van Heerden for their support and prayers. 
Music café as well as Impala Hotel Potchefstroom for creating a place to blow off 
steam. 
LIST OF ABBREVIATIONS: 
Ave. - Average 
 v 
 
CEC - Cation Exchange Capacity 
DTEC - Department of Tourism, Environment and Conservation 
DWAF - Department of Water Affairs and Forestry 
DME - Department of Minerals and Energy 
EC - Electrical Conductivity 
EMC - explosives manufacturing company 
MPRDA - Mineral and Petroleum Resources Development act 
NEMA - National Environmental Management Act 
PCA - principle component analysis 
PSD - particle size distribution 
RC1 - first rehabilitation company 
SABS - South African Bureau of Standards 
SD - Standard deviation 
Tot. - Total 
  
 vi 
 
LIST OF TABLES 
Table 4.1: Macro-elements, measured with the 1:2 water extraction procedure, present 
in the soil (mmol l-1) at the sampled areas during 2010. 
Table 4.2: Micro-elements (µmol l-1) and other data (pH, mS/cm and ppm), measured 
with the 1:2 water extraction procedure, present in the soil at the sampled areas during 
2010. 
Table 4.3: Nutrient status (mg/kg), pH and EC (mS/m) of the soil measured with the 
Ammonium acetate extraction analysis at the sampling areas during 2010. 
Table 4.4: Exchangeable cations (cmol(+)/kg), measured with the ammonium acetate 
extraction analysis, present in the soil at the sampling areas during 2010.  
Table 4.5: Particle size distribution (PSD) of the soil determined in 2010. 
Table 4.6: Key to sample identities as used in the statistical ordination of the data. 
Table 4.7: Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2010. 
Table 4.8: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm, ppm and 
%), measured with the 1:2 water extraction procedure, present in the soil at the 
sampled areas during 2010. 
Table 4.9: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the soil 
at the sampling points, by utilising the ammonium acetate extraction analysis during 
2010. 
Table 4.10: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at the 
sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2010. 
Table 4.11: Mean macro-elements, measured by utilizing the 1:2 water extraction 
procedure, present in the soil (mmol l-1 ± SD) at the sampled areas during 2011. 
 vii 
 
Table 4.12: Mean micro-elements (µmol l-1 ± SD) and other data (pH and mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2011. 
Table 4.13: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the soil 
at the sampling points, by utilising the ammonium acetate extraction analysis during 
2010. 
Table 4.14: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at the 
sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2011. 
Table 4.15: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2011 samples. 
Table 4.16: Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2013. 
Table 4.17: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2013. 
Table 4.18: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the soil 
at the sampling points, by utilising the ammonium acetate extraction analysis during 
2013. 
Table 4.19: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at the 
sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2013. 
Table 4.20: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2013 samples. 
Table 4.21: Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2016. 
Table 4.22: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2016. 
 viii 
 
Table 4.23: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the soil 
at the sampling points, by utilising the ammonium acetate extraction analysis during 
2016. 
Table 4.24: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at the 
sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2016. 
Table 4.25: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2011 samples. 
Table 4.26: Mean macro-elements (mg/l ± SD) present in the water samples taken from 
the collection pond in 2013. 
Table 4.27: Mean micro-elements (mg/l ± SD) present in the water samples taken from 
the collection pond in 2013. 
Table 4.28: P-values associated with t-tests in the 1:2 water extraction procedure 
(macro-elements) data for the denuded area between 2011 and 2016. 
Table 4.29: P-values associated with t-tests in the 1:2 water extraction procedure 
(micro-elements and other data) data for the denuded area between 2011 and 2016. 
Table 4.30: P-values associated with t-tests in the ammonium acetate (Exchangeable 
cations) data for the denuded area between 2011 and 2016. 
Table 4.31: Mean grass species (amount ± SD) found over all five transects. 
Table 4.32: Mean of each species (amount ± SD) found between the different 
quadrants on the five transects. 
Table 4.33: Average maximum temperatures (̊C) measured in Emalahleni (weather 
station 0515320 8) between – 2010 and 2016. 
Table 4.34: Average (Ave./day) and total (Tot.) monthly rainfall (mm) measured in 
Emalahleni (weather station 0515320 8) between 2010 and 2016. 
 
 
 ix 
 
LIST OF FIGURES 
Figure 2.1: A geographical illustration of the location of the study area in South Africa. 
Figure2.2: A site map illustrating the different sections of the study area. 
Figure 2.3: An illustration of the physical rehabilitation at the prills silo excavation by 
RC1. 
Figure 2.4: The first rehabilitation company’s understanding of underlying geology. 
Figure 2.5: Physical rehabilitation measures illustrated on the site map with the RC1 
drainage channels in green and contour berms in orange. 
Figure 3.1: Cross section display of where the first five samples were taken during the 
initial site visit in 2010.  
Figure 3.2: Site plan displaying the position of the cut off drains installed within the 
denuded area (area 1). 
Figure 3.3: Cross section of the main cut off drain and illustrations of the imbedding. 
Figure 3.4: Cross section of the side cut off drains and illustrations of the imbedding. 
Figure 3.5: Cross section of the evaporation pond where the drainage system empties 
into. 
Figure 3.6: A photograph displaying the lying of geofabric during construction of the 
drainage system. 
Figure 3.7: Imbedding of geopipe during construction of the drainage system. 
Figure 3.8: Backfilling of trenches after installation of drainage system. 
Figure 3.9: Completed evaporation pond with drainage pipes entering on the right. 
Figure 4.1: Visible Eutrophication due to the N leaching at the natural stream to the 
south of the denuded area. The stream is around 1.5m wide. 
Figure 4.2: Vegetation producing anthocyanin due to salt and Mn load adjacent to the 
denuded area (east). 
 x 
 
Figure 4.3: Visible signs of surface runoff and erosion on the denuded area. 
Figure 4.4: Very shallow water table on the denuded area close to the ¼ denuded 
sampling point due to saturation of the soil. (The size of the hole is 30 x 30 x 30cm). 
Figure 4.5: N pollution on surface at the natural stream down slope from the denuded 
area. 
Figure 4.6: Display of data ordination (PCA). 
Figure 4.7: Map overview of the rehabilitated area (the white lines indicate the transects 
numbered 1 to 5 from North to South and the red blocks indicate the quadrants 
numbered from 1 to 5 from the East to the West). 
Figure 5.1: New understanding of the underlying geology with the sub-surface laterite 
indicated by the black line. 
Figure 5.2: Photographs taken before the implementation of the rehabilitation procedure 
(on the left) and 1 month after seeding (on the right). 
Figure 5.3: Photos taken 3 months after seeding on the previously denuded area. 
 
 
 
 
 
 
 
 
 
 xi 
 
Table of contents 
ABSTRACT…………………………………………………………………….……………..i 
ACKNOWLEDGEMENTS…………………………………………………………...……..iii 
LIST OF TABLES……………………………………………………………………...……iv 
LIST OF FIGURES…………………………………………………………...…………….vii 
CHAPTER 1: INTRODUCTION………………………………………………………...….1 
1.1 Background…………………………………………………………………………...…1 
1.1.1 Mining in South Africa………………………………………………………….…….1 
1.1.2 Environmental impact and legislation of mining in South Africa…………......….2 
1.2 Background of the specific study area………………………………………..………4 
1.3 Perspective and outline of dissertation……………………………………………….5 
1.3.1 Problem statement…………………………………………………………..………..5 
1.3.2 Aim and objectives…………………………………………………………...……….6 
1.4 Outline of dissertation……………………………………………………………..……6 
CHAPTER 2: OVERVIEW OF THE PREVIOUS REHABILITATION ATTEMPT……..8 
2.1 Study area: Site description……………………………………………………………8 
2.2 First rehabilitation attempt……………………………………………….……………10 
2.2.1 Historical overview of the first rehabilitation procedure…………………………11 
2.2.2 First attempt to rehabilitate the old soak-away…………………………….….…12 
2.2.3 The first attempt to rehabilitate the excavation of the prills silo……………...…14 
2.2.4 The first rehabilitation attempt of the denuded area………………………..……16 
2.2.5 The first rehabilitation attempt of the plant footprint……………………………..20 
2.2.6 Issuing of the completion report by RC1…………………………………….……21 
 xii 
 
CHAPTER 3: MATERIALS AND METHODS……………………………..….…………23 
3.1 Soil analysis procedures used…………………………………………..…….....…..23 
3.2 Rehabilitation procedure implemented on the site…………………..…………….23 
3.3 Sampling of the site…………………………………………………..……….………26 
3.4 Soil analysis procedure used…………………………………………………….…..26 
3.5 Sequence of implementation of the rehabilitation procedure…………………….28 
3.6 Implementation procedure of soil amelioration……………………………..…...…34 
3.7 Vegetation monitoring……………………………………………………………...…34 
3.8 Data analysis………………………………………………………………..…………34 
CHAPTER 4: RESULTS AND DISCUSSION………………………………………..…36 
4.1 Soil and water data obtained from 2010, 2011, 2013 and 2016……………...….36 
4.1.1 Results and discussion of the data obtained in 2010………………………...…36 
4.2 Addressing the problems on the site…………………………………………….…..43 
4.3 Data ordination (2010 analysis)…………………………………………………...…44 
4.4 Results and discussion of the data obtained in 2010 throughout the growth medium 
(2010)…………………………………………………………………………..…47 
4.5 Summary results and discussion of soil analysis (2011)………………………….51 
4.6 Summary results and discussion of soil analysis (2013)……………………….…57 
4.7 Summary of the results obtained in the 2016 data…………………………...……63 
4.8 Summary results and discussion of water analysis (2013)…………………...…..68 
4.9 Overall discussion of data obtained in 2011, 2013 and 2016……………...……..68 
4.10 Changes in values between 2011 and 2016……………………………….……..71 
4.11 Vegetation survey (monitoring of the site)……………………………………...…72 
4.11.1 Mean transect data………………………………………………………………..73 
 xiii 
 
4.11.2 Mean quadrant data……………………………………………………………….74 
4.12 Weather data of the area………………………………………………………..…..74 
CHAPTER 5: CONCLUTIONS AND RECOMMENDATIONS…………………...……78 
5.1 Conclusions………………………………………………………………………..…..78 
5.2 Initial recommendations for area 2 in 2010…………………………………………82 
5.3 Initial recommendations for area 1 in 2010…………………………………...…….83 
5.4 General recommendations…………………………………………………….……..83 
5.5 Concluding remarks………………………………………………………………...…84 
CHAPTER 6: REFERENCES……………………………………………...……………..85 
 
 1 
 
CHAPTER 1: INTRODUCTION 
1.1  Background 
1.1.1 Mining in South Africa 
Mining has a long history in South Africa and for a very long time mining has had an 
enormous impact on the country’s economy. Over time, the mining sector has grown 
larger than the agricultural sector in South Africa (in financial terms) and became the 
main developer of infrastructure and the main contributor to the economic growth of 
the country (Ngcofe & Cole, 2014). 
Mining in South Africa goes back as far as the 1800s when copper mines were in 
operation in the Namaqualand region where the town of Springbok is today 
(Cairncross, 2006). Gold mining also started around this time in the Witwatersrand 
area in 1867 and later in the East Rand and West Rand located in the Gauteng 
province. The gold mining industry further expanded to the Klerksdorp area in the 
North West province and also to Welkom in the Free State province (Reichardt, 
2012). 
Diamonds were discovered in the Kimberley area in 1867 and so we can say that the 
mining industry was prosperous in South Africa during the 1800’s. Sarsby & 
Meggyes (2010) stated that 55 different minerals where sourced in 2005 from 1113 
operating mines in South Africa. The majority of these minerals were exported. The 
mining sector contributed around 8% of the entire country’s gross domestic product 
during the first decade of the 21st century. In 2008 the world production of coal was 
5845 mega ton (Mt) of which South Africa produced 236 Mt. The major mining 
companies responsible for this production were BHP Billiton and Anglo Coal 
(www.mbendi.com). 
As mentioned above, the South African mining sector is comprised of several 
different elements. The second largest element, measured in monetary value, is 
coal. Coal comprises 6.1% of the total merchandise exports of South Africa 
(www.universalcoal.com). Coal is used by Eskom in South Africa to generate 
electricity and thus plays a major role in the economy both directly and indirectly. 
(Eberhard, 2011). 
 2 
 
There is a price associated with the economic growth that mining, as the largest 
single industry in South Africa, brings to the country. Over the years mining has 
produced a legacy of wastes and contamination that negatively impacted the 
environment (McDonald, 2004). 
1.1.2 Environmental impact and legislation of mining in South Africa 
It is inevitable that ecosystems and landscapes will be degraded and polluted with 
the mining of these much needed natural resources (Bradshaw & Chadwick, 1980; 
McDonald, 2004). In South Africa the Mineral and Petroleum Resources 
Development act (MPRDA, 2002) states in section 38 (1) (d) that: “The holder of a 
reconnaissance permission, prospecting right, mining right, mining permit or 
retention permit must as far as it is reasonably practicable, rehabilitate the 
environment affected by the prospecting or mining operations to its natural or 
predetermined state or to a land use which conforms to the generally accepted 
principle of sustainable development.” 
Other legislation regarding the rehabilitation of degraded land is also in place that 
forces mining and other companies to restore these sites. This includes the 
Constitution of South Africa (1996), the National Environmental Air Quality Act 39 of 
1994, the National Environmental Management (NEMA) Act 28 of 1998 as well as 
the National Environmental Biodiversity Act 10 of 2004. All of this legislation was set 
in place in order to protect the environment from industries that cause it harm. The 
aim of this legislation is the prevention of environmental pollution and rehabilitation of 
the affected areas.  
The Constitution of South Africa (Act 108 of 1996) states in section 24, “everyone 
has the right to an environment that is not harmful to their health or well-being”. It 
also states that “everyone has the right to have the environment protected, through 
reasonable and other legislative measures that prevent pollution and ecological 
degradation, promote conservation and secure ecologically sustainable development 
and use of natural resources while promoting justifiable economic and social 
development”.  
 
 3 
 
Soil can be degraded by toxic industrial substances. This leaves environmental 
problems that have a large cost implication in order to correct. Soil is an important 
resource and it can be seen as re-useable but with regard to the scale of human 
lifetimes soil cannot be seen as a renewable resource (Brady & Weil, 2008). 
It is also stated in NEMA, Act 107 of 1998, that degradation and pollution of the 
environment must be rehabilitated. The National Department of Tourism, 
Environment and Conservation (DTEC) and the Department of Minerals and Energy 
(DME) plans and budgets yearly for ownerless mines in order to rehabilitate them 
and minimize the environmental risk (Nel, 2006). 
As seen above, the government of South Africa considers the mitigation of disturbed 
land very important. Thus, it will be a high priority for all companies in the affected 
sectors to comply with the legislation in order to protect the environment for 
generations to come as well as for business reasons, for example to obtain a mine 
closure certificate. 
The mining sector uses a number of procedures that can create a negative 
environmental impact. One example of such a procedure is the production and 
misuse of explosives (Folchi, 2003) which create a cost effective way of clearing the 
unwanted material in order to get to the valuable and desirable resources. Even in 
the early developmental stages of mining explosives were used for blasting open the 
shafts to recover the ore beneath the ground. 
The explosives used in the mining sector can become pollutants when released into 
the environment. Pollutants are defined by Newman (2015) as “A substance that 
occurs in the environment at least in part as a result of man’s activities, and which 
has a deleterious effect on living organisms”  
 
 
 
 
 4 
 
Pollution is known to migrate from one medium to another. One example is pollution 
in the air that pollutes the soil and water when precipitation takes place. Pollution in 
the soil can enter groundwater systems and cause contamination. We also see two 
different ways of how pollutants enter the environment. The first is known as point 
source pollution and the other nonpoint pollution. With point source pollution the 
pollutant is discharged from an identifiable source a nonpoint is where the pollution 
discharges from different points apart from each other and this makes it more difficult 
to control (Spellman, 2010). 
As mentioned previously we see that one of the possible environmental pollutants 
from the South African mining industry can be explosives. This phenomenon is seen 
all over the world where the contaminants are released into soil and water. 
Explosives are known to be less biodegradable than other nitrates, which means 
under natural circumstances it will remain in the environment as a pollutant for a very 
long time (Spain, 2000). 
1.2   Background of the specific study area 
In this study we find that an explosives manufacturing company (EMC) leased a 
piece of land from a coal mining company on a mining site. The area was polluted by 
effluent from the explosives plant and this resulted in a complex set of legal issues. It 
was found that the EMC was liable for the rehabilitation of the site, and so the 
rehabilitation process was started. 
The explosives company called upon an established rehabilitation company. This 
company performed a site assessment and a rehabilitation plan was drawn up for 
the specific site that was contaminated by the explosives effluent. 
The rehabilitation plan was approved and executed where after the rehabilitation was 
considered to be successful.  However, this success was short-lived (approximately 
2 years) and the problem started to reappear.  
 
 
 
 5 
 
1.3   Perspective and outline of dissertation 
1.3.1 Problem statement 
The use of explosives in mining is a much needed commodity. However, if 
explosives are released into the environment it can have a negative impact. 
Explosives contaminate the soil and water that it comes into contact with and 
therefor causes a problem that must be addressed.  
As was mentioned before, this study is concerned with an EMC that leased land on a 
coal mining site. The EMC erected a manufacturing plant on the site at the top of 
watershed with the natural relief sloping to the south. An emulsion based explosives 
manufacturing plant started operating on the site in 1978 and was decommissioned 
in 1990. During the operation of the plant the spillages, effluents and plant and 
vehicle washings were disposed of by means of a soak-away to the south of the 
plant. The old soak-away was backfilled in 1991 and seeded to establish vegetation 
(Anonymous, 2003:1). 
Down-slope, to the south of the plant, vegetation started to die off to such an extent 
that an area of 8.5 ha was completely denuded of all vegetation. The denudation of 
the area was caused by pollutants emanating from the explosives manufacturing 
plant and not the mining activities. This assumption can be made because of the fact 
that the closest mining related activities are more than a kilometre away and the 
mining commenced down slope from this specific denuded area.  
Because the area is situated within the borders of a mining site, it is required to be 
rehabilitated in order for the mine to obtain a closure certificate. A closure certificate 
is needed by all mines when mining is no longer commencing. This document 
certifies that the mining company is no longer liable for the safety and environmental 
responsibilities and this means the mined land is environmentally rehabilitated to the 
prescribed state (MPRDA, 2002). If a mine which is not in use any more cannot 
obtain a closure certificate it has a large negative financial impact to the mining 
company.  
 
 
 6 
 
This site under consideration in this study is required to be rehabilitated to a state 
which can be utilised as grazing. This will ensure that there is no liability remaining to 
the EMC or the coal mining company. The rehabilitation must be successful to a 
level that is acceptable to the relevant authorities in order for the mine to obtain mine 
closure (Anonymous, 2003:4). 
1.3.2 Aim and objectives  
The aim of this study was a critical analysis of a rehabilitation procedure 
recommended and implemented on the specific site. Further, to recommend and 
implement an alternative rehabilitation procedure in order to successfully rehabilitate 
the same site after failure in the first rehabilitation attempt.  
The objectives of this study included the following:  
 Critically evaluating the rehabilitation procedure initially used to rehabilitate 
the specific site. 
 Implementing a different and more effective rehabilitation procedure on the 
same site. 
 Assessing the differences between the two procedures and identifying the 
shortcomings of the procedures. 
 Successfully completing the rehabilitation of the site to a state where it can be 
utilised as grazing with no liability to the explosives manufacturing company or 
the coal mining company. 
 
1.4  Outline of dissertation 
Chapter 1 serves as the introduction to the study. This chapter includes some 
background of the South African mining industry with regard to the history of mining 
in South Africa. The chapter also gives more information on the environmental 
impact and legislation of mining in a South African context. This chapter furthermore 
gives insight on the perspective and outline of the study with regard to the problem 
statement and the aim and objectives. 
 
 7 
 
Chapter 2 provides an overview of the previous rehabilitation attempt and gives more 
information on the specific study area with regard to the previous rehabilitation 
procedure and how it was implemented. Some methodology about the procedure is 
also included. 
Chapter 3 contains the materials and methods used in the study to obtain the data 
needed. This chapter also explains the procedures implemented in order to reach 
the end result of the rehabilitation. 
Chapter 4 shows the results from the data obtained in this study and a discussion 
thereof. The aim of this discussion is to highlight the changes in the measurements 
over time. This chapter also contains a summary of applications implemented in 
order to correct the deficiencies, key ratios and excess elements. 
Chapter 5 provides conclusions and recommendations with regard to the specific 
study. 
Chapter 6 shows the references used in the study. 
 
 
 
 
 
 
 
 
 
 
 8 
 
CHAPTER 2: OVERVIEW OF THE PREVIOUS 
REHABILITATION ATTEMPT 
2.1  Study area: Site description 
The specific site for this study is situated on a coal mine to the south of Emalahleni in 
the Mpumalanga province of South Africa in the grassland biome. To be more 
specific, the eastern Highveld grassland biome (Mucina et al, 2014). According to 
Thompson & Turk (2007) a biome is defined as “a community of plants living in a 
large geographical area characterized by a particular climate”.  
 
Figure 2.1: A geographical illustration of the location of the study area in South 
Africa. 
The Highveld grassland bioregion is the largest grassland area in South Africa and it 
contains the highest number of vegetation types of all the grasslands found in South 
Africa. This specific site is located in the higher rainfall parts of the South African 
Highveld (Mucina et al., 2006). 
According to climate-data.org the average annual rainfall of the Emalahleni area is 
around 693mm and most of the precipitation comes in the summer months. The 
climate is classified as warm and temperate.  
 9 
 
To the east of the mining site we find where the old explosives manufacturing plant 
was situated on top of a watershed with the denuded area downslope to the south of 
the plant as illustrated in figure 2.2. The site is divided into different sections (Figure 
2.2) namely:  
i) The plant or plant footprint: This is where the plant was situated when it was still in 
production. 
ii) The prills silo: Situated to the west of the plant on the area where the storing 
facilities was situated before decommission of the plant 
iii) The soak away: Placed just south of the plant where disposal of effluent and wash 
water took place 
iv) The denuded area: To the South of the plant below the farmed field. This area is 
where the contaminants affected the soil and caused the denudation of vegetation. 
 
Figure2.2: A site map illustrating the different sections of the study area. 
 
 
 
 10 
 
2.2  First rehabilitation attempt 
The following section is a summary of different anonymous and confidential reports 
(Report No. 1-6) of the first rehabilitation effort at the study site. The reports were 
written between 2003 and 2008, starting with a site assessment report and ending 
with a completion report in 2008. The completion report stating that the site had been 
rehabilitated successfully. 
The first report (Report no. 1), issued in September 2003, was entitled “Conceptual 
rehabilitation plan”. In this report the first rehabilitation company (RC1) stated the site 
history and the plan that was devised to rehabilitate the problem areas. Together 
with the first report, RC1 issued report no. 2 also in September 2003 which was 
entitled “Methodology statement”. This report gave an overview of what they needed 
to do on the different areas of the site in order to rehabilitate it. This was conditional 
upon the approval of the rehabilitation plan by the client, namely the EMC. The 
methodology statement included pricing for amelioration of the growth medium and 
pricing for all the different areas.  
The third report (report no. 3) was issued in October 2003 and was entitled “Scope of 
work for the remediation of the site”. This report contained a detailed rehabilitation 
plan for all the areas on the site accept for the plant footprint, namely the denuded 
area, the old soak-away, and prills silo. This report explained what needed to be 
done in terms of amelioration as well as soil preparation and also the reasons why 
each process should have been implemented.  
The fourth report (report no. 4, September 2004), was named “Addendum to scope 
of work for the rehabilitation of the site – Remediation of the plant footprint”. The 
report contained the same information as the scope of work report for the 
remediation of the site, but only with regard to the plant footprint (after the plant has 
been demolished). Report no. 5 was issued during February 2005 and this report 
was entitled “Environmental rehabilitation of plant – methodology statement”, and the 
report contained a detailed plan for demolishing the physical structure of the 
decommissioned manufacturing plant. 
 
 
 11 
 
The final report (Report no. 6) was issued in October 2008 and the title of this report 
is “Completion report for rehabilitation of the site”. This report stated the rehabilitation 
history that was done on the site, the monitoring process and the conclusion that the 
site was successfully rehabilitated to the predetermined standards. 
2.2.1 Historical overview of the first rehabilitation procedure 
In 1996 RC1 conducted a site visit and evaluated the area. They came to the 
conclusion, by means of soil analysis, that the denudation of the area was caused by 
the excessively high levels of nitrogen salts and sulphate in some areas of the site. 
Another site visit was conducted by the same company in 1997. There was little 
change in the results of the soil analysis between the two visits. Thereafter, in 1999, 
more soil samples were taken and most of the sampling points showed a decrease 
in contamination in terms of nitrogen salts and sulphate, but the levels of acidification 
showed an increase on the entire area. They stated that the reason for this could be 
the horizontal movement of the pollutants and the pollutants accumulated in some 
poorly drained areas and that it would take many years for the site to be cleaned up 
under natural conditions.  
During 2002 RC1 undertook monitoring of the site and found that the levels of 
contamination had decreased from 1999 but the amount of nitrate and ammonia 
were still high enough to maintain the salinity of the soil and to prevent plant growth. 
It was reported that: “Information recently discovered has indicated that the denuded 
areas are not a result of toxicity of the chemicals that are present in the soils, rather 
it is their contribution to the total salinity of the soil which impacts on the osmotic 
potential of the soil.” This implies that the moisture gradient is created so that water 
moves from the plant to the soil. Because of this, the plant cannot retain water. RC1 
took the Electrical Conductivity (EC) value of below 400mS/m as a target value for 
the rehabilitation.  
 
 
 
 
 12 
 
2.2.2 First attempt to rehabilitate the old soak-away 
The old soak-away (Figure 1.2) was backfilled and seeded in 1991. The seeding was 
not completely successful because the vegetation that was established was patchy. 
Monitoring on the area was done by RC1 and they found that the salinity decreased 
significantly, but some hot spots were still present where the EC was higher than the 
target value of 400mS/m. The proposed remedy was to mix the soil in order to get a 
better overall salinity. This was done by ripping the soil at a depth of 500mm and 
thereby mixing the more contaminated soil with the soil that shows a lower EC value. 
The ripping was also done in order to create a more hospitable microbial 
environment by means of better aeration in the soil. Their hope was that this would 
create a more suitable substrate for plant growth and also increase the permeability 
of the soil which would help with the leaching of the contaminants.  
In addition to the above, compost was added (500 m3 per ha) before the ripping of 
the soil in order to enhance the microbial environment further by adding a food 
source to for the soil microbes. Studies have shown that compost increases the 
microbial biomass (Saison et al., 2006) and this was one of the desirable outcomes 
that the company wanted to achieve. Compost also enhances permeability for 
leaching of contaminants and serves as a long term source of plant nutrients in the 
soil. The area was then seeded with a number of grass species. The application of 
50kg seeds per ha was used with the following species selected: Eragrostis curvula 
(Making up 40% of seed mix); Eragrostis tef (Tef); Digetaria eriantha (Smuts Finger 
grass); and Chloris gayana (Rhodes grass). It was recommended that the grass 
should be cut or burned in order to prevent the contaminants from re-entering the 
soil when the grass dies and decomposes.  
In the above-mentioned, the microbes are seen as important because they were 
added to the soil for the purpose of de-nitrification and the plants also play a role in 
the uptake of the contaminants according to RC1. The area was seeded with the 
maximum historical amount of summer rainfall. Denitrification is a dissimilatory 
reduction process that is done by the metabolism of essential aerobic bacteria, of 
nitrogen oxides namely nitrate (NO3
-) and nitrite (NO2
-) to a gas stage (nitric oxide, 
NO3 and nitrous oxide), (Knowles, 1982). 
 13 
 
The hand over criteria for the rehabilitation process that was identified by RC1 for the 
old soak-away was determined to be average soil salinity with a value of less than 
400mS/m in the upper 300mm of the soil. Furthermore, the summer basal cover of 
vegetation should have been at least 60% of the total area. 
The old soak-away was monitored by RC1 by measuring the residual salt 
concentration in the growth medium and also the success of vegetation. The residual 
nitrogen concentrations were measured in the soil six months after the rehabilitation 
procedure took place. This was done by determining the EC by means of field tests. 
The tests were done on a 5 by 5m grid across the old soak-away and if the EC 
exceeded 400mS/m soil samples were collected to determine the cause of the 
salinity. Monitoring continued until no area on the old soak-away exceeded 400mS/m 
where after the rehabilitation of the soak away was seen as successful in terms of 
the identified salinity levels. 
The vegetation monitoring was done by using the line transect method and three 
transects were used to monitor the success of the vegetation. The areas where 
plants intersected the line where recorded in addition to the bare patches. This was 
done in order to determine the proportion of bare areas and the proportion of areas 
covered with vegetation. This process took place at the end of the first growing 
season, in March 2004 to be precise.  After that, the monitoring was repeated after 
every summer and every winter until 2008.  
At the time of the report in 2003, RC1 stated that the maintenance would be largely 
dependent on the result of the monitoring that was required. If they found that the 
hotspots that were present at the time of the report were still present in 2005 they 
would have ripped the area and more compost would have been added (this was 
most likely the case). Field trials were also conducted as monitoring in order to see if 
the added microbes were successful in the denitrification of the growth medium. 
After the first five years of growth, monitoring would have determined whether it 
would be necessary to re-seed the old soak-away area after September 2004. If this 
was the case, a higher proportion of Eragrostis curvula would have been used in the 
seed mix in order to establish a community with more climax grasses.  
 
 14 
 
2.2.3 The first attempt to rehabilitate the excavation of the prills silo 
The first rehabilitation company evaluated the area and found that the base of the 
excavation was contaminated. Because of the pollution, it was unlikely that the area 
could support plant growth. The contamination present was localised to a hotspot 
next to the sump of the prills silo where the contaminants discharged. The area was 
rehabilitated by making use of a procedure that entailed adding freeze-dried bacterial 
cultures to the soil. This was done in order to get a more concentrated microbial 
population in the soil. A carbon rich source (compost) was added to the soil in order 
for the carbon to act as an energy source for the microbes when they metabolise the 
nitrates for de-nitrification. The excavation was then backfilled with stockpiled soil 
provided by the mine with two layers of compacted laterite half way up the 
excavation. The excavation was then filled with 800mm of topsoil on top of the 
compacted layers which serves as a growth medium.  
The methodology statement that RC1 provided stated the following to describe the 
work to be undertaken: 
Firstly, 100mm of clean topsoil was placed in the base of the excavation where after 
500m3 of compost was placed and spread over the area. The next step was to place 
200mm of clean topsoil on top of that. Thereafter 500mm of un-compacted laterite 
was placed upon this topsoil. Two layers of laterite were placed and compacted in 
200mm loose layers (compacted to MOD ASSHTO 95%) on top of the un-
compacted laterite. The excavation was then backfilled with un-compacted laterite up 
to 750mm from the surface and filled with 800mm of clean topsoil before the area 
was seeded with the following species in the seed mix: Eragrostis curvula (making 
up 40% of the total seed mix); Eragrostis tef (Tef); Digetaria eriantha (Smuts Finger 
Grass); Chloris gayana (Rhodes grass). Seed mix was to be applied at 50kg per ha.  
This rehabilitation method described by RC1 was implemented in order to separate 
the contaminated area and the pollution from the topsoil. The freeze-dried bacteria 
were for de-nitrification (turning the nitrate pollutants into nitrogen gas as discussed 
previously). The compost that was added was used as a source of carbon and the 
compacted layers of laterite were used to stop water from reaching the contaminated 
zone and also to stop the contaminants from reaching the growth medium.  
 15 
 
 
 
Figure 2.3: An illustration of the physical rehabilitation at the prills silo excavation by 
RC1. 
The handover criteria selected by RC1 for the excavation at the prills silo was only to 
establish a basal vegetation cover of at least 60%. The reason why this was the only 
hand over criteria for this area is that the pollution was only present below the 
surface and the top soil added would not have a high EC value.  
Vegetation monitoring was done after completion of the mentioned procedures by 
means of the line transect method. Two transects were used to monitor the success 
with the same proportional bare patch to vegetation method described under the old 
soak-away monitoring. This was also undertaken at the end of the first growing 
season in March 2004 and thereafter at the end of each summer and winter until 
2008.  
The monitoring required at the time of the report would have determined if the re-
seeding of the area would be necessary after the first year of growth in September 
2004. If re-seeding were necessary, a higher proportion of Eragrostis curvula would 
have been included into the seed mix.  
Topsoil 100 mm 
Compost 500 
m3/ha 
Topsoil 200 mm 
Backfill– Mine Overburden** 
Compacted Laterite* 
Laterite
* 
Topsoil 
Lateritic overburden from 
excavation* 
*   Laterite will be sourced from that existing around the pit 
** Mine overburden generated during the construction of the refuge 
bays 
 16 
 
2.2.4 The first rehabilitation attempt of the denuded area 
RC1 evaluated the denuded area and stated that, from their historical knowledge of 
the practices in the area, the main sources of pollution were the old soak-away and 
the excavation at the prills silo. The company constructed a conceptual model of the 
area that shows the laterite layer below the surface and the area where the pollution 
daylights. Based on this conceptual model of the underlying geology they assumed 
that the pollution entered the soil trough shallow seepage, mostly in the area where 
the old soak-away was situated, and moved below the soil surface on the laterite 
layer under the farmers field to the point where it daylights at the top of the denuded 
area.  
 
 
Figure 2.4: The first rehabilitation company’s understanding of underlying geology. 
 
 
 
 17 
 
With the above mentioned perception of the underlying geology thought to be 
present in the area RC1 constructed two cut off trenches (1000mm in width at a 
depth of 2000mm) on the farmers field in order to intercept the pollution between the 
source (old soak-away) and the denuded area. This was done to reduce the ongoing 
pollution of the denuded area and these two connected trenches drained into a 
holding tank (10m3). The local farmer emptied the tank once every week and used 
the content as fertiliser.  
The scope of work on the denuded area entailed that two trenches were excavated; 
2000mm deep and 1000mm wide with a trench length of 750m. Geofabric was laid in 
the excavation and extended 1000mm up each side of the excavation. On top of 
that, a 100mm layer of 19mm washed stone was placed in the Geofabric and 65mm 
of Geopipe was laid on the stone. The pipe was then covered with stone with an 
average cover of 100mm and the overlap of Geofabric was used to enclose the 
stone fill. Thereafter, 200mm of double washed filter sand was placed on the stone 
fill and the trenches were backfilled with the remaining topsoil to surface level. Both 
drains were connected with a main drain that was discharged into the holding tank, 
which was buried 5000mm deep and the excavation was backfilled. 
Other rehabilitation procedures were also implemented on the denuded area at the 
same time. These procedures entailed ripping the area to a depth of 500mm to 
enhance the permeability of the soil for better leaching conditions and the ripping 
against the contour created an irregular surface that reduced surface runoff (less 
erosion) and increased the amount of water that infiltrated the soil. This also created 
a physically suitable growth medium for plants to establish.  
After the area was ripped, compost was added together with a freeze-dried bacteria 
culture to the denuded area. Contour berms of 300mm in height were also 
constructed at regular intervals over the denuded area. This was done in order to 
reduce surface runoff and increase the amount of water that infiltrated the soil. 
Thereafter, the area was seeded with a number of pasture species namely 
Eragrostis curvula (Making up 40% of the total seed mix); Eragrostis tef (Tef); 
Digetaria eriantha (Smuts Finger Grass); Chloris gayana (Rhodes grass). Seed mix 
was to be applied at 50kg per ha. 
 
 18 
 
There was no hand over criteria identified for the farmer’s field because the 
infrastructure was put in place in order to stop the sources of contamination rather 
than treating the contaminated area. This area was considered to be successfully 
rehabilitated when the vegetation was established on the denuded area. This drain 
was decommissioned when the water drained into the tank had a nitrogen 
concentration of less than 30mg/l. The denuded area was required to have a soil 
salinity of less than 400mS/m and a basal vegetation cover of 60% or more.  
The water in the holding tank (placed in the farmer’s field, upslope from the denuded 
area and downslope from the old soak away) was tested twice a year to determine 
the amount of ammonia and nitrogen in solution. This was done in order to 
determine when the residual contamination in the subsurface no longer presented a 
risk of causing contamination on the denuded area.  
The residual nitrogen concentrations and the success of the vegetation were 
monitored in this area. RC1 monitored the nitrogen concentrations in the medium by 
soil sampling six months after the rehabilitation procedure was implemented. The 
monitoring consisted of field tests to determine the EC of the medium and the tests 
were conducted on a 5m by 5m grid across the denuded area. Where the EC 
exceeded 400mS/m, soil samples were taken to determine the cause of the high 
salinity. Monitoring continued every 6 months until no areas showed an EC value of 
higher than 400mS/m. Vegetation was monitored once again by the line transect 
method discussed previously. Three transects were used in this case. The 
monitoring started in March 2004 and was repeated twice a year thereafter. The 
monitoring of the area continued until 2008.  
Once again, the ongoing maintenance was dependant on the monitoring results 
obtained from the area. If hotspots were still found in 2005 the area would have been 
ripped again and more compost would have been added. Field trials were done to 
determine whether de-nitrification of the medium by means of the added microbes 
was successful and the vegetation monitoring would have determined whether it was 
necessary to re-seed the area after the first year in September 2004. If this was the 
case a higher proportion of Eragrostis curvula would have been included into the 
seed mix.  
 
 19 
 
 
Figure 2.5: Physical rehabilitation measures illustrated on the site map with the RC1 
drainage channels in green and contour berms in orange. 
 
 
 
 
 
 
 
 
 
 
 
 
 20 
 
2.2.5 The first rehabilitation attempt of the plant footprint 
In 2003, when the rehabilitation procedures were implemented on the other areas, 
the plant infrastructure was still present on the site. The plant was decommissioned 
and removed in 2003 and the rehabilitation procedure was started. RC1 stated that 
the plant footprint would have the same characteristics of pollution as the other areas 
because of spills and accidents that might have occurred during the time of 
operation.  
The methodology used for the disposal of the plant started with the disposal of the 
concrete from the plant that was excavated and moved to a waste disposal site. 
Thereafter, the excavation was backfilled with the laterite provided from the mine 
stockpile. 
The plant footprint was rehabilitated thereafter by ripping the area to a depth of 
500mm before 500m3 of compost were placed on the area. The area was seeded for 
revegetating the area with a seed mixture containing Eragrostis curvula (making up 
40% of the seed mix); Eragrostis tef (Tef); Digetaria eriantha (Smuts Finger Grass); 
Chloris gayana (Rhodes grass). Seed mix was to be applied at 50kg per ha. The 
above-mentioned procedures (ripping and adding of compost) were done for the 
same reasons as mentioned previously with the other areas.  
The hand over criteria selected was the same as with the other areas (i.e. a soil 
salinity value of less than 400mS/m and at least 60% basal cover of vegetation).  
Residual nitrogen concentrations in the growth medium were monitored here as well 
as the success of vegetation. Six months after the completion of the rehabilitation 
work monitoring started and concentrations of residual nitrogen was measured. The 
monitoring consisted of field tests to determine the EC on a 5m X 5m grid across the 
area of the plant footprint. Where the EC exceeded 400mS/m samples were taken to 
determine the cause of the salinity and monitoring continued until there were no 
areas with an EC higher than 400mS/m.  
The vegetation was monitored by the line transact method as described previously 
and this monitoring continued until 2008. Once again the maintenance depended on 
the monitoring at the time. If they found that the hotspots were still present in 2005 
the area would have been ripped and compost would have been added again.  
 21 
 
Field trials were commissioned in order to determine whether the microbes were 
successful in the de-nitrification in the soil and monitoring of vegetation would have 
determined if re-seeding would be needed after the first year of growth. If this was 
the case a higher proportion of Eragrostis curvula would have been added to the 
seed mix.  
2.2.6 Issuing of the completion report by RC1 
In 2003, RC1 started with monitoring, this showed an increase in biomass and an 
increase in the overall soil quality for plant growth. Monitoring continued and in 2008 
the RC1 issued a completion report stating that the rehabilitation of the site was 
successful. The reason that they saw the rehabilitation as successful was because of 
the better soil quality and the removal of all sources of pollution. They also stated 
that the remaining pollution present in the soil would have been removed by means 
of the cut off drains, de-nitrification from the soil microbes, leaching and plant uptake.  
Between 2003 and 2007 the EMC contracted a local farmer to maintain the site after 
the rehabilitation procedures had been implemented. In January 2008 the 
rehabilitation of the site was complete, but before that ongoing maintenance was 
implemented on an annual basis. The maintenance entailed the establishment of 
Phragmites reeds on the wet patches of the denuded area. Ploughing was done on 
the denuded area to a depth of 200mm in order to loosen the growth medium and 
reseeding was done on the area with the seed mix already discussed. The erosion 
control berms were repaired when needed. A section of the Geopipe was replaced 
because of damage done to it in a veld fire. Controlled burning of the veld was done 
to remove the weeds that had accumulated on the site. 
During the ongoing monitoring RC1 found that despite of cut off drains that were 
installed down slope of the plant in the farmers field there were still groundwater day-
lighting in the denuded area. They stated that the cut off drains only isolated 
approximately 25% of the groundwater catchment. This lead to wet patches on the 
denuded area and RC1 was of the opinion that these saturated patches were 
independent of the plant activities and that they will remain there and have an impact 
on the vegetation.  
 
 22 
 
During the first growth season from 2003 to 2004 the vegetation was established 
with a basal cover of over 60% and the predominant species was Eragrostis tef. In 
the next growth season from 2004 to 2005 the basal cover still exceeded 60% but 
some wet patches appeared on the denuded area and these patches had 0 to 20% 
basal cover, depending on the specific area considered within the denuded area. 
These wet patches occupied ±2% of the entire denuded area at that time, and the 
other 98% had a change in species dominance from Eragrostis tef to pioneer 
species such as Bidens pilos and Tagetes minuta. 
In 2005 to 2006 the wet patches increased in size and a fire changed the 
predominant species again to Eragrostis curvula and Chloris gayana. In late 2005 
Phragmites reeds were planted on the wet patches with only partial success. From 
2006 to 2008 the predominant species remained Eragrostis curvula and Chloris 
gayana and almost no vegetation was visible on the wet areas which covered ±8% of 
the total area at that time.  
The soil was also monitored; samples were analysed from the bottom of the slope on 
the denuded area, the middle of this slope, the top of this slope and approximately 
5m upslope from the start of the denuded area. The samples showed that the pH 
has increased overall and that there were no areas with a pH value lower than 6. 
This indicates a substantial increase, since the pH value was around 4.5 in 2002. 
The EC values declined dramatically but the EC was higher on the higher parts of 
the slope. The nitrates and ammonia in the soil has also decreased over time. 
However, the ammonia levels decreased at a faster rate, which was an indication 
that nitrification (ammonia to nitrates) was proceeding faster than the de-nitrification. 
RC1 stated that this was due to the annual ripping of the soil that created oxidizing 
conditions and that de-nitrification will become more rapid because of the fact that 
the ripping of the soil was stopped at that time.  
After the completion report was issued by RC1 the site degraded again and this 
shows that the first rehabilitation procedures were not as successful as it seemed at 
that time.  
 
 
 23 
 
CHAPTER 3: MATERIALS AND METHODS 
3.1  Rehabilitation of the site 
In 2010 the first site visit was conducted. This was done after being called in by the 
EMC because of doubts as to the success if the first rehabilitation attempt. After the 
site visit, it was found that the rehabilitation done by the RC1 was not successful. 
This conclusion was based on the fact that the denuded area was showing signs of 
denudation again and that the pollutants on the site would take a very long time to 
remove under natural conditions. At that time, it was clear that the surface and 
subsurface water that enters the denuded area of the site needed to be reduced 
because the water carried the remaining pollution on the site to the denuded area. 
Together with the reduction of the water, it was required to utilize pH adjusting, 
chemical and organic material for amelioration in order to create better conditions to 
sustain plant establishment and plant growth in the soil. 
The rehabilitation procedure implemented took place in a phased approach. This 
decision was taken in order to make the necessary detail of information available 
and to ensure that the rehabilitation was done in a cost effective manner. Phase 1 
was the investigation, analysis, planning and recommendations of what was required 
for successful rehabilitation. The construction of the subsoil drains and the initial in 
situ treatment of the soil were done during Phase 2. The treatment of the soil and re-
vegetation of the site was finalised during Phase 3. This took place after the 
subsurface drains removed enough contaminants and the analyses showed the 
pollution has decreased to acceptable values. 
3.2  Rehabilitation procedure implemented on the site 
During the first site visit some relevant information was obtained from the mine 
personnel and the local farmer leasing the area at that time. The location of the 
borehole and some other information were collected from the people mentioned 
above. Soil samples were taken of the growth medium and were analysed by the 
Eco Analytica  laboratory in Potchefstroom. 
 
 
 24 
 
Five samples were collected on the historical sample sites (one on each). These soil 
samples were taken from the plant footprint, the old soak-away, the footprint at the 
prills silo, ¼ of the way down slope on the denuded area and ¾ of the way down 
slope on the denuded area. The rationale behind this was to determine the overall 
state of the site in terms of soil chemistry and pollutant levels in order to get a better 
view of what the problems were that needed to be addressed. The reports from RC1 
were made available in order to see what has been done and to make more 
informed recommendations. Figure 3.1 is an illustration of the sampled area and 
where the samples have been taken. 
 
Figure 3.1: Cross section display of where the first five samples were taken during 
the initial site visit in 2010.  
3.3  Sampling of the site 
Soil samples were taken at an approximate depth of 20cm by making use of a 
ground auger. Each sample weighed around 500g. Three soil samples were 
collected at each of the five sampling points on the site during later site visits, but 
with the initial site visit in 2010 only one sample was taken from each sampling point. 
Three water samples of 250ml were taken from the collection pond at a depth of 
10cm in bottles provided by the laboratory. 
 25 
 
3.4  Soil analysis procedures used 
The 1:2 water extraction procedure (as described in detail by Van Rensburg & 
Pistorius, 1998) was used to determine the water-soluble basic cation friction in the 
growth medium as well as the trace elements present. The water-soluble basic 
cation friction entails Ca2+, Mg2+, K+ and Na+ and the trace elements entails Fe, Mn, 
Cu and Zn. The quantification was done by means of atomic absorption 
spectrophotometry with the Spectr. AA – 250 Varian, Australia (Ramiriz-Munoz, 
1968). The 2:1 water extraction procedure is described by Rhodes (1982). 
P-Bray 1 analysis was done (as described by Bray &Kurtz, 1945) to determine the P 
concentration. This was done by using a solution of 0.03 mol l-1 ammonium floride in 
0.025 mol l-1 HCl (Bray 1 solution) with a soil to extractant ratio of 1:7.5g ml-1. 
The anions, phosphates, nitrates, sulphates and chloride were determined by using 
an Ion Chromatograph (Metrohm 761, from Switzerland) and the ammonium 
concentrations were determined by using an ammonium selective electrode as 
described by Banwart et al. (1972). The bicarbonate HCO3
- in the growth medium 
was determined by the potentiometric titration method with a pH end-point of 4.5 and 
using a standard 0.005M HCl solution as described by Skougstad et al. (1979). 
Boron concentrations were determined by the azomethine-H method (Barrett, 1978). 
The extractible bases were determined by using the Spectr. AA – 250 (Varian, from 
Australia) following extraction with ammonium acetate solution as described by 
Thomas (1982). Base saturation was calculated by determining the ratio between 
total exchangeable cations which are Ca, Mg, K and Na to the exchangeable 
capacity of the growth medium (Buys, 1986) 
The CEC (cation exchange capacity) was determined by the stepwise replacement 
of the cations from exchange sites by adding sodium acetate and thereafter 
ammonium acetate. The suspension was placed in leach tubes and leached with 
ammonium acetate in order to replace the Na2+ on the exchange complex with 
ammonium and the Na concentration of the leachate, which was determined by the 
AA – 250 (Varian, from Australia), was then used to calculate the CEC. 
 
 26 
 
Both the pH (H2O) and the pH (KCl) were determined by adding 50ml of deionised 
water or KCl to 20g of the media. The suspension pH was determined by means of a 
calibrated pH meter (Radiometer pHM 80, Copenhagen). The supernatant for 
measurement of the EC was prepared by adding 50ml of deionised water to 50g of 
the growth medium and then determined with a WTW LF92 conductivity meter. The 
available phosphate was determined by the adding of 75ml p – Bray solution to 10g 
of the growth medium and the phosphate concentration was analysed with a 
Continuous Flow Analyses System (Skalar, Netherlands) at 340 to 650nm in a 15mm 
tubular flow cell. 
The organic carbon content was determined by using the Walkley-Black method as 
described by Walkley (1935). As described by Rhodes (1982), the exchangeable 
acidity is quantified by applying KCl to the filtrate and titrating the product with 
NaOH. 
The PSD (Particle Size Distribution) was determined by the percentages sand, silt 
and clay method. This was analysed by separating the sand friction from the 
percentage silt and clay and, thereafter, the clay percentage was quantified by 
means of the Hydrometer as described by Gee and Bauder (1986). 
 
 
 
 
 
 
 
 
 
 
 27 
 
3.5  Sequence of implementation of the rehabilitation procedure 
The rehabilitation procedures discussed were implemented in the following order: 
The first procedure implemented was the installation of the subsurface drains where 
after the area 1 and area 2 (See figure 3.2 below) was ripped to a depth of 500mm. 
The third step was the formation, repair and cleaning of the drain along the road to 
prevent surface water flowing over the rehabilitated area. After this, an application of 
30 tons of dolomitic lime per hectare was spread on area 1 before the rainy season. 
The next step entailed the forming of contouring on area 1 to channel the surface 
water to the drain along the road. An application of 20 tons dolomitic lime per hectare 
was spread on area 2 before the rainy season with an application of 30 tons well-
cured organic material on area 2 and 300kg super phosphate per hectare. After this, 
area 2 was seeded with seed mixture as indicated before. The monitor and 
addressing of area 1 was done in the next phase during the winter months. 
 
Figure 3.2: Site plan displaying the position of the cut off drains installed within the 
denuded area (area 1).  
 28 
 
 
Figure 3.3: Cross section of the main cut off drain and illustrations of the imbedding. 
 
Figure 3.4: Cross section of the side cut off drains and illustrations of the imbedding. 
 29 
 
 
Figure 3.5: Cross section of the evaporation pond where the drainage system 
empties into. 
Specifications on lying and bedding of Geopipe: 
The purpose of the bedding was to provide a surface on which the Geopipe can be 
placed in order to separate the pipe from the surrounding rocks and clay material. 
Geopipe is a flexible pipe which does not have a large load bearing capacity. 
Therefore the quality of the backfill would have determined the strength of the pipe. 
With a reasonable material, the bedding thickness needed to be 100mm but when 
using softer or saturated material the bedding thickness should be between 150 and 
200mm according to the SABS 1200 standards. 
 
 
 
 
 
 30 
 
The proper compaction of the bedding material was essential to ensure that the 
gradient of the pipe could be maintained. The side fill was also very important and 
special care was taken with the placement of the haunch areas under the Geopipe. If 
the backfill material was not placed under the pipe, voids would have formed and the 
flexible pipe would have deformed into the voids. The compaction of the material 
was done at 90% Mod AASHTO. Fortunately, granular soils are easily graded and it 
provided uniform support to the Geopipe. The granular soil accommodates the 
haunch support material, which was compacted next to and above the Geopipe after 
it was laid. 
Because the Geopipe is flexible it can carry a greater load. This is because of the 
fact that the pipe can deform horizontally and vertically without breaking. If the trench 
was too wide or to deep it may have caused instability and this could create 
problems with the laying of the Geopipe and also backfilling. If the trench collapses 
before proper bedding took place the pipe would have deformed. This is largely 
dependent on the soil type and the depth of the trench and in some cases trench 
support systems might be needed, however, this was not the case on this site. 
Deformation of Geopipe is not seen as a problem but rather a requirement. The 
correct placement of bedding and padding should be implemented so that the 
deformation of the pipe does not exceed the limits. The backfilling of the trench was 
done in a symmetric manner in order to prevent the lateral movement of the pipe as 
this can cause problems with alignment. No dumping of soil in an indiscriminate 
manner to backfill the trench was done because this would have a negative influence 
on the integrity of the pipe. Furthermore, rocks pressing against the pipe should be 
avoided.  
A layer of selected aggregate was placed above the pipe for protection against rocks 
and to provide support for backfilling. Hand compaction was used on the first 300mm 
of medium above the pipe. The selected backfill specifications (SABS 1200: 
Earthworks) divided the material into two classes. Class 1 is a clean, crushed rock or 
stone aggregate in the range between 9 and 38mm with little or no fines (particles 
smaller than 0.075mm). Class 2 is coarse grained, granular, free-draining, non-
cohesive soils graded between 0.6 and 38mm with less than 5% fines.   
 
 31 
 
 
Figure 3.6: A photograph displaying the lying of Geofabric during construction of the 
drainage system. 
 
Figure 3.7: Imbedding of Geopipe during construction of the drainage system. 
 32 
 
 
Figure 3.8: Backfilling of trenches after installation of drainage system. 
 
Figure 3.9: Completed evaporation pond with drainage pipes entering on the right. 
 33 
 
3.6  Implementation procedure of soil amelioration 
The implementation of the rehabilitation procedure was done in October 2015 to 
utilise the maximum amount of summer rainfall. The procedure used entailed the 
broadcasting 15 tons of lime/ha, there after the broadcasting 1.7 tons of rock 
phosphate per ha. Potassium and zinc fertilizers were broadcasted at a rate of 350 
kg/ha. The applied ameliorants were incorporated into the soil by means of a chisel 
plough and roller. Contouring of the erosion affected areas and controlling 
preferential flow patterns was implemented. Microbial inoculation at a rate of 4L/ha 
and application of the humic acid at a rate of 25kg/ha with fulvic acid at a rate of 
7.5kg/ha were done by using a tractor and agricultural sprayer. After this the site was 
seeded by broadcasting the seeds (25kg/ha) and rolled by a rolling implement and 
tractor. Monitoring of the site was the next step. The indigenous seed mixture that 
was used contained Cynodon dactylon, Pennisetum clandestinum, Eragrostis 
curvula, Eragrostis tef, Digitaria eriantha and Sorghum bicolor. 
3.7   Vegetation monitoring  
The vegetation establishment survey was conducted by means of the wheel point 
method as explained by Tidmarsh and Havenga (1955). The previously denuded 
area was divided into 5 transects on which a wheel that marks each meter with a 
clicking sound was implemented to determine species on the transects. The species 
within 10cm from the mark where identified and recorded on each of the 5 transects.  
Quadrants where then spaced evenly on each transect with a size of 1m2 each. 
Within each of the quadrants, all the plants were identified and counted. 
3.8  Data analysis 
The data obtained from the soil sampling were analysed with the multivariate 
analysis method referred to as principle component analysis (PCA) in CANOCO 
version 4 (Ter Braak & Smilauer, 1998). Data analyses entailed nonparametric 
methods because the data were not normally distributed and the data was highly 
skewed due to the high null values and the data showed heterogeneous variances 
among groups. 
 34 
 
Standard deviations and means were calculated by using Microsoft Office Excel 
2010 together with the Student’s one sided t-tests at a nominal significant level of 5% 
for equality of means. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 35 
 
CHAPTER 4: RESULTS AND DISCUSSION 
4.1  Soil and water data obtained from 2010, 2011, 2013 and 2016  
4.1.1 Results and discussion of the data obtained in 2010 
During the first site visit in 2010, only one sample was taken on each of the 
designated sample areas. This was done in order to get an overview of the area and 
the problems that was present at that time (as described by Conklin, 2004). The 
information obtained from the resulting analyses was used to set up a conceptual 
rehabilitation plan. The reason for the plan being conceptual was to address the 
problems at hand. It was not possible at that time to plan for the future with precision 
with regard to the soil chemical amelioration that would be needed.  
The results of the 1:2 water extraction procedure provide a good indication of water 
soluble elemental concentrations as well as other elements that are available for 
plant uptake. The elemental concentrations seen in this procedure can also be seen 
as the elements that can be transported by water and leach into the environment. 
Table 4.1: Macro-elements, measured with the 1:2 water extraction procedure, 
present in the soil (mmol l-1) at the sampled areas during 2010. 
Macro-elements 
Sample ID Ca Mg K Na PO4 SO4 NO4 NH4 Cl HCO3 
Millimol per litre 
Sample 1 0.14 0.88 0.42 0.10 <0.01 1.17 0.17 0.02 0.10 0.00 
Sample 2 0.87 0.65 1.18 0.20 <0.01 0.51 3.54 0.21 0.13 0.00 
Sample 3 7.86 10.84 1.64 3.25 0.05 0.41 78.49 38.34 1.42 0.00 
Sample 4 0.27 0.13 0.35 0.07 0.01 0.34 0.14 0.05 0.34 0.10 
Sample 5 5.35 3.04 0.93 3.53 0.02 0.96 19.77 1.03 0.64 0.00 
 
 36 
 
Table 4.2: Micro-elements (µmol l-1) and other data (pH, mS/cm and ppm), measured 
with the 1:2 water extraction procedure, present in the soil at the sampled areas 
during 2010. 
Micro-elements and other data P-Bray 1 
Sample ID Fe Mn Cu Zn B pH EC ppm 
Micromol per litre   mS/cm 
Sample 1 4.10 1.10 0.10 0.25 <1 4.46 0.26 12.62 
Sample 2 31.91 11.52 0.20 0.73 12 4.32 0.47 57.23 
Sample 3 1.28 499.76 0.10 5.74 4 4.32 8.20 78.60 
Sample 4 11.03 0.60 0.81 1.26 <1 5.17 0.13 26.32 
Sample 5 3.08 164.22 0.35 2.86 <1 4.15 2.26 13.00 
 
Tables 4.1 and 4.2 above shows the results of the samples and the following can be 
inferred: 
Sample 1 was taken on the plant footprint and the macro-elemental concentrations 
revealed slightly elevated SO4 and no HCO3. This indicated that the sampling site 
was completely devoid from any buffering capacity. These problems can easily be 
addressed by applying the correct amount of dolomitic lime and by doing so the Ca 
deficiency would also be corrected. By applying a fertilizer with a NO3 basis, the P 
and K concentrations will address the macro-elemental concerns. The dolomitic lime 
will also correct the low pH of the area and this will address the slightly elevated Mn 
concentration. 
Sample 2 was collected half way down the slope from sample 1 and this area 
showed only sparse vegetation. The analyses showed that the area was slightly 
nutrient poor and elevated K and NO3 concentrations were seen here. The norm 
values for vegetation establishment for a 1:2 water extraction are Ca: 1.8, Mg: 0.6, K: 
0.8, Na: <1.0, SO4:  0.9, NO3: 2.1, Cl: <1.0 and HCO3: 0.8 mmol l
-1. (Bergmann, 
1992) 
 37 
 
Sample 3 was taken on the slope that was completely devoid of any vegetation 
(denuded area). This area on the site showed clear signs of leaching and 
eutrophication. Eutrophication is the process by which a water body becomes 
increasingly rich in aquatic plant life because of plant nutrient pollution in the water 
body (Ansari, 2011). The analyses of the sample showed clear signs of amelioration 
that was done on the site previously because of the particularly elevated Ca and Mg 
concentrations in the growth medium. The N concentration in this area is 
phytotoxically high (both NO3 and NH4) and together with that the Na and Cl 
concentrations exceeds the adsorption capacity of this area completely. 
Furthermore, with the low pH value of the area and the depleted buffering capacity 
the Mn, and to some extent the Zn, in the area was at phytotoxically high 
concentrations (Bergmann, 1992). 
Sample 4 was taken on the opposite bank of the slope and does not show chemical 
traces from the mining activity but the area is slightly nutrient poor. This problem can 
easily be addressed by applying the correct amount and combination of inorganic 
and organic fertilizer.  
Sample 5 was taken at a pond in the natural stream at the bottom of the slope and 
showed a combination of chemical residue from the mining activities and the 
pollutants (fertilizer/explosives) leaching from the source at the top of the slope. 
Elevated Na, NO3, NH4 Mn and Zn concentrations were seen in the analyses 
together with a low pH and a high EC value of 2.26mS/cm which should be 1.8 to 
2.0mS/cm for plant establishment and growth. (Bergmann, 1992) 
 
 38 
 
 
Figure 4.1: Visible Eutrophication due to the N leaching at the natural stream to the 
south of the denuded area. The stream is around 1.5m wide. 
 
Figure 4.2: Vegetation producing anthocyanin due to salt and Mn load adjacent to 
the denuded area (east). 
 
 
 
 
 39 
 
Table 4.3: Nutrient status (mg/kg), pH and EC (mS/m) of the soil measured with the 
Ammonium acetate extraction analysis at the sampling areas during 2010. 
Nutrient Status 
Sample ID Ca Mg K Na P pH(H2O) pH(KCl) EC 
mS/m 
     mg/kg 
Sample 1 275.0 146.0 104.5 23.5 12.6 5.21 4.34 31 
Sample 2 417.5 141.5 134.5 65.5 57.2 4.78 4.23 53 
Sample 3 1051.0 635.0 159.5 191.0 78.6 4.52 4.33 266 
Sample 4 455.5 206.0 174.5 68.0 26.3 6.04 5.09 45 
Sample 5 446.0 227.0 92.5 148.5 13.0 4.48 4.22 392 
 
Table 4.4: Exchangeable cations (cmol(+)/kg), measured with the ammonium 
acetate extraction analysis, present in the soil at the sampling areas during 2010.  
Exchangeable cations  
Sample ID Ca Mg K Na CEC S-
Value 
Base 
saturation 
(%) 
pH(H2O) pH(KCl) 
(cmol(+)/kg) 
Sample 1 1.37 1.20 0.27 0.10 5.96 2.94 49.43 5.21 4.34 
Sample 2 2.08 1.16 0.34 0.28 8.26 3.88 46.94 4.78 4.23 
Sample 3 5.24 5.23 0.41 0.83 13.71 11.71 85.40 4.52 4.33 
Sample 4 2.27 1.70 0.45 0.30 6.22 4.71 75.78 6.04 5.09 
Sample 5 2.23 1.87 0.24 0.65 9.94 4.98 50.07 4.48 4.22 
 
 
 40 
 
The ammonium acetate extraction procedure gives us an indication of the total 
elemental concentrations in the growth medium. The exchangeable cation ratios 
should be Ca 65 : Mg 25 : K 8 : Na 2 as norm values (MVSA, 1997) but in this case it 
was 46 : 41 : 9 : 3, 54 : 30 : 9 : 7, 45 : 46 : 4 : 7, 48 : 36 : 10 : 6 and 45 : 38 : 5 : 13 
respectively for the five different samples. These results showed that the medium 
had a significant Ca requirement and that the Na concentration was far too high. This 
is an important factor to take into account for vegetation establishment as the Na 
inhibits K adsorption in a competitive manner. If this problem is not addressed the 
vegetation will be prone to stress from drought and appear yellow, which is a 
symptom of N or P deficiency (Bergmann, 1992). 
The pH(KCl) analyses showed that the pH is below 5.5 and this can result in 
potential micro-elemental toxicity to plants (Sparks, 2003). The cation exchange 
capacity (CEC) fell in the 0 to 5 cmol(+) kg-1 range for most samples, which is a very 
low value. This can be addressed by adding organic material to the medium 
(Bjorklund & Mello, 2012). CEC is seen as the sum of all the exchangeable cations 
that can be absorbed by the soil (Brady & Weil, 2008). The percentage base 
saturation value of sample 3 was close to 100%, this should not be the case under 
normal circumstances. Sample 3 seemed to be an exception to the rule because of 
the higher clay content present as seen in Table 4.5.  
Table 4.5: Particle size distribution (PSD) of the soil determined in 2010. 
Sample > 2mm 
% 
Sand Silt Clay 
% < 2mm 
1 0.4 82.0 2.5 15.4 
2 0.8 86.6 4.7 8.6 
3 0.2 86.9 4.6 8.4 
4 10.3 82.5 5.3 12.2 
5 17.9 77.5 6.0 16.5 
 
 
 
 41 
 
In Table 4.5 we see that the particle size distribution of all samples are dominated by 
the sand friction and is, therefore, characterized by the low nutrient status and the 
limited water holding capacity. This can be seen in the CEC values of the samples. 
 
Figure 4.3: Visible signs of surface runoff and erosion on the denuded area. 
 
Figure 4.4: Very shallow water table on the denuded area close to the ¼ denuded 
sampling point due to saturation of the soil. (The size of the hole is 30 x 30 x 30cm). 
 42 
 
 
Figure 4.5: N pollution on surface at the natural stream down slope from the 
denuded area. 
By looking at the soil sample analysis it is clear that all sources of pollutants were not 
removed and that the remaining pollutants will not be taken up by plants, removed by 
de-nitrification or removed by the drainage trenches installed by RC1 in the near 
future. As a result, further rehabilitation is required. 
4.2  Addressing the problems on the site 
The site was divided into two areas for the rehabilitation of the site viz. Area 1; the 
denuded area down slope from where the plant was situated (south of the plant) and 
Area 2; the combination of the other sites mentioned previously (plant footprint, prills 
silo and the old soak-away). 
Samples were taken on each sampling point at different depths to determine the 
depth to where contaminants were present in the soil. It was believed that 
contaminants were rising through the medium by capillary action. Samples were 
taken at the surface, 0.5m, 1m, 1.5m and 2m. 
 
 
 
 43 
 
The historical sample sites were used again in the second rehabilitation project in 
order to have a reference from the data collected by RC1. These sampling points 
provide a reference for monitoring from 2002 when RC1 implemented their 
rehabilitation procedures up until now. For sampling area 1 the practical sites one 
quarter down the slope into the denuded area (with sample identification 1/4 
denuded) and three quarters down the slope into the denuded area (with sample 
identification 3/4 denuded) were used because these sites are situated inside the 
worst affected area on the site. For sampling area 2 the sites at the prills silo, old 
soak-away and plant footprint was used.  
After the site was visited in 2010 and information was collected using the means 
mentioned above, soil and water data was collected in 2011, 2013 and 2016 
respectively. This was done to determine if the subsurface drains were working 
properly and to do possible adjustments to the amelioration plan for the site. 
As mentioned previously, the site was divided into five different areas with regard to 
sampling points. The plant footprint, the prills silo and the old soak-away, all in area 
1, as already discussed, were sampled in order to determine if the contamination has 
stopped in those areas and to revegetate these specific areas. The other two 
sampling points were taken one quarter way down the denuded area and three 
quarters down the slope on the denuded area. These samples are called ¼ denuded 
and ¾ denuded respectively.  
4.3  Data ordination (2010 analysis) 
The data obtained from the 2010 soil sampling were analysed with the multivariate 
analysis method principle component analysis (PCA) in CANOCO version 4 (Ter 
Braak & Smilauer, 1998). Data analyses entailed nonparametric methods because 
the data were not normally distributed and the data was highly skewed due to the 
high null values and the data showed heterogeneous variances among groups. 
 
 
 
 
 44 
 
Table 4.6: Key to sample identities as used in the statistical ordination of the data. 
Full site name  Abbreviated name 
Prills 0.0m Pr_0.0 
Prills 0.5m Pr_0.5 
Prills 1.0m Pr_1.0 
Prills 1.5m Pr_1.5 
Prills 2.0m Pr_2.0 
Soak-away 0.0m S_Aw_0.0 
Soak-away 0.5m S_Aw_0.5 
Soak-away 1.0m S_Aw_1.0 
Soak-away 1.5m S_Aw_1.5 
Soak-away 2.0m S_Aw_2.0 
Plant 0.0m Pl_0.0 
Plant 0.5m Pl_0.5 
Plant 1.0m Pl_1.0 
Plant 1.5m Pl_1.5 
Plant 2.0m Pl_2.0 
¼ Denuded 0.0m 1_4D_0.0 
¼ Denuded 0.5m 1_4D_0.5 
¼ Denuded 1.0m 1_4D_1.0 
¼ Denuded 1.5m 1_4D_1.5 
¼ Denuded 2.0m 1_4D_2.0 
¾ Denuded 0.0m 3_4D_0.0 
¾ Denuded 0.5m 3_4D_0.5 
¾ Denuded 1.0m 3_4D_1.0 
¾ Denuded 1.5m 3_4D_1.5 
¾ Denuded 2.0m 3_4D_2.0 
 
 45 
 
Of all the ordination data analyses conducted, it was decided that the ordination 
based on the exchangeable cations should be selected as being the most 
representative of the dataset. Statistical analysis based on the groupings were 
obtained from area 1 (as the 1_4D sample series), whilst the rest of the samples 
were designated as area 2 (Pl_1.0 and Pl_1.5 was grouped into group 2 as well as 
the remaining members of that group). 
 
Figure 4.6: Display of data ordination (PCA). 
Data ordination (PCA) based on the exchangeable cations from the ammonium 
acetate extraction procedure. 
 
 
 46 
 
4.4   Results and discussion of the data obtained in 2010 throughout the 
growth medium (2010) 
The results of the 1:2 water extraction procedure that provides an indication of the 
elemental concentrations that is soluble in water, and for that reason, available for 
plant uptake and leachable, are the following: 
The area 1 samples showed extremely high concentrations in terms of Ca, Mg, K, 
Na, NO3 and NH4 (Bergmann, 1992). 
Sampling area 1 was completely devoid from any buffering capacity as represented 
by the HCO3 concentration. 
This was a clear indication that the acidification process was still active at that time 
and that the contaminants were still present in this area. The norm values of these 
elemental concentrations are as follows: Ca: 1.8, Mg: 0.6, K: 0.8, Na: <1.0, NO3: 2.0 
and NH4: 0.4 mmol l
-1. 
Table 4.7: Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2010. 
Macro- elements 
Sample 
ID 
Ca Mg K Na PO4 SO4 NO3 NH4 
mmol l
-1
 ± SD 
Area 1 9.50±0.
94 
6.98±1.41 2.79±0.62 3.06±0.63 0.02±0.01 0.88±0.09 56.58±9.21 18.90±4.61 
Area 2 0.57±0.
54 
0.35±0.38 0.22±0.13 0.16±0.24 0.01±0.00 0.39±0.18 4.5±7.94 2.19±4.12 
 
 
 
 
 
 47 
 
Table 4.8: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm, ppm 
and %), measured with the 1:2 water extraction procedure, present in the soil at the 
sampled areas during 2010. 
Micro-elements 
Sample ID Fe Mn Cu Zn B 
Micromol per litre 
Area 1 1.48±10.28 211.94±45.43 0.42±0.17 6.65±2.09 3.40±4.69 
Area 2 6.13±9.01 30.86±48.98 0.17±0.12 0.80±0.57 0.72±1.50 
 
Other data pH EC P-Bray1 C N 
 mS/cm ppm % % 
Area 1 4.18±0.13 5.87±0.92 32.38±24.13 1.47±0.57 0.01±0.01 
Area 2 4.42±0.78 0.54±0.77 15.94±13.89 1.19±0.27 0.04±0.20 
 
The samples of sample area 2 were slightly nutrient poor but still showed signs of 
elevated NO3 and NH4. In these samples there were trace amounts of HCO3 present 
in the soil and this indicated that the acidification process had almost reached its end 
and that the leaching of contaminants had decreased in these areas (Brady & Weil, 
2008). 
By looking at the micro-elements from the 1:2 water extraction we see that the area 
1 samples showed extremely elevated Mn and Zn concentrations and Mn is 
phototoxic to plants (Fernando & Lynch, 2015). The norm value for Mn concentration 
in the soil for a 1:2 water extraction that is acceptable for plant growth is 0.9 µmol l-1. 
In the area 2 samples the Mn was also very high but in terms of the micro-elemental 
concentrations it was the only element that warrants some concern.  
 
 
 48 
 
In the table for other data it is clear that the pH values of both area 1 and area 2 
samples are very low. The elevated EC values showed in the area 1 samples were 
much higher than that of the area 2 samples (Van Rensburg & Maboeta, 2004). The 
norm range value for the EC is between 0.5 and 0.8mS cm-1 for the 1:2 water 
extraction.  
The ammonium acetate extraction procedure gives a good indication of the total 
elemental concentrations in the medium. The results of this procedure revealed that 
the exchangeable cation ratios deviate from the norm values to a large degree. The 
norm percentages of the exchangeable cation ratios (Ca:Mg:K:Na) are: 65:25:8:2. In 
group 1 the ratios were 48:36:6:10 and 62:17:9:10 for area 2. This indicated that the 
Ca in area 1 is significantly lower than the norm value and the Na concentration was 
significantly higher than the norm value.  
Table 4.9: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the 
soil at the sampling points, by utilising the ammonium acetate extraction analysis 
during 2010. 
Nutrient status 
Sample ID Ca Mg K Na P pH(H2O) pH(KCl) EC 
mg/kg  mS/m 
Area 1 537.00 
±114.05 
242.70 
±43.97 
132.80 
±32.27 
121.20 
±10.81 
32.38 
 
±24.13 
4.76 
 
±0.25 
4.32 
 
±0.21 
1773.40 
 
±201.16 
Area 2 285.48 
±210.16 
48.33 
±26.88 
82.10 
±37.80 
53.95 
±25.25 
15.94 
 
±13.89 
4.96 
 
±0.84 
4.25 
 
±0.80 
150.35 
 
±208.50 
 
 
 
 
 
 49 
 
Table 4.10: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at 
the sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2010. 
Exchangeable cations 
Sample 
ID 
Ca Mg K Na CEC S-
Value 
Base 
saturation 
pH(H2O) pH(KCl) 
cmol(+)/kg %  
Area 1 2.68 
±0.57 
2.00 
±0.36 
0.34 
±0.08 
0.53 
±0.05 
9.75 
±1.05 
5.54 
 
±1.01 
56.94 
 
±8.54 
4.76 
 
±0.25 
4.32 
 
±0.21 
Area 2 1.42 
±1.05 
0.40 
±0.22 
0.21 
±0.10 
0.23 
±0.11 
4.22 
±1.01 
2.27 
 
±1.10 
52.58 
 
±16.85 
4.96 
 
±0.84 
4.25 
 
±0.80 
  
The average pH(KCl) of the area 1 and the area 2 samples were well below 5.5 and 
this could cause problems with potential heavy metal toxicity and result in the 
observed elevated micro-elemental concentrations. The average cation exchange 
capacity (CEC) for both area 1 and area 2 samples were in the range between 0 to 
10 cmol(+) kg-1. These values were very low and indicated that the medium was in 
need of organic material. The base saturation of sample areas 1 and 2 were both 
significantly lower than 100%. This was due to the low CEC values of both sample 
areas, and also due to the low pH values present in the medium of both groupings at 
that time. 
The physical soil analyses of the medium showed that the particle size distribution 
(PSD) was ±80% for the sand friction, ±10% for the clay friction and ±6% for the silt 
friction. This indicated that the soil composition was adequate for the rehabilitation 
procedures prescribed and the rehabilitation commenced without any special 
requirements. The permeability of the soil was deemed adequate for the functioning 
of subsurface drains.  
 
 
 
 50 
 
4.5   Summary results and discussion of soil analysis (2011) 
Tables 4.11 to 4.14 reflect the average of the results obtained of the soil analysis 
conducted in May 2011: 
Table 4.11: Mean macro-elements, measured by utilizing the 1:2 water extraction 
procedure, present in the soil (mmol l-1 ± SD) at the sampled areas during 2011. 
Macro-elements 
Sample 
ID 
Ca Mg K Na PO4 SO4 NO3 NH4 Cl HCO3 
Millimol per litre 
¼ 
Denuded 
10.14 
±0.32 
7.17 
±1.93 
3.04 
±0.71 
3.29 
±0.75 
0.02 
±0.01 
0.91 
±0.11 
61.18 
±9.51 
21.18 
±4.79 
0.41 
±0.06 
0.00 
±0.00 
¾ 
Denuded 
1.38 
±0.13 
1.02 
±0.33 
0.26 
±0.08 
0.54 
±0.15 
0.01 
±0.00 
0.20 
±0.04 
16.71 
±2.60 
8.24 
±1.06 
0.08 
±0.04 
0.00 
±0.00 
Plant 
 
0.55 
±0.42 
0.12 
±0.03 
0.21 
±0.15 
0.03 
±0.01 
0.01 
±0.00 
0.61 
±0.20 
0.09 
±0.06 
0.02 
±0.01 
0.06 
±0.05 
0.23 
±0.40 
Prills 
 
0.12 
±0.10 
0.12 
±0.08 
0.18 
±0.10 
0.02 
±0.00 
0.01 
±0.01 
0.28 
±0.20 
0.09 
±0.05 
0.02 
±0.00 
0.08 
±0.06 
0.01 
±0.01 
Soak-
away 
0.18 
±0.03 
0.15 
±0.07 
0.20 
±0.09 
0.03 
±0.01 
0.01 
±0.00 
0.40 
±0.06 
0.04 
±0.02 
0.02 
±0.00 
0.06 
±0.02 
0.00 
±0.00 
 
 
 
 
 
 
 
 51 
 
Table 4.12: Mean micro-elements (µmol l-1 ± SD) and other data (pH and mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2011. 
Micro-elements and other data 
Sample 
ID 
Fe Mn Cu Zn B pH EC 
Micromol per liter  mS/cm 
¼ 
Denuded 
1.53±1.19 240.20±33.56 0.44±0.19 7.37±2.56 5.56±5.14 4.27±0.03 6.33±0.95 
¾ 
Denuded 
4.75±0.77 110.41±19.80 0.12±0.08 1.59±0.20 1.28±1.17 3.87±0.11 1.72±0.27 
Plant 
 
4.59±2.46 2.27±1.93 0.14±0.04 0.23±0.22 0.57±0.49 5.38±1.83 0.16±0.08 
Prills 
 
21.93±16.43 1.99±0.97 0.27±0.13 0.64±0.54 0.00±0.00 4.56±0.32 0.08±0.04 
Soak-away 0.74±0.63 8.50±1.89 0.14±0.07 0.79±0.16 0.00±0.00 4.22±0.07 0.09±0.01 
 
 
 
 
 
 
 
 
 
 
 
 52 
 
Table 4.13: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the 
soil at the sampling points, by utilising the ammonium acetate extraction analysis 
during 2010. 
Nutrient status 
Sample 
ID 
Ca Mg K Na P pH(H2O) pH(KCl) EC 
mg/kg  mS/m 
¼ 
Denuded 
561.67 
±151.93 
235.33 
±60.37 
136.33 
±44.64 
119.50 
±14.76 
47.16 
±18.60 
4.88 
±0.27 
5.84 
±0.48 
1809.67 
±267.91 
¾ 
Denuded 
168.33 
±18.43 
73.50 
±11.79 
28.50 
±6.08 
59.17 
±26.99 
23.92 
±18.48 
4.39 
±0.15 
5.36 
±1.23 
476.67 
±132.25 
Plant 
 
555.67 
±430.47 
53.00 
±3.04 
92.33 
±20.25 
56.67 
±15.88 
24.23 
±11.61 
5.78 
±1.34 
5.92 
±1.21 
55.67 
±34.93 
Prills 
 
256.50 
±42.44 
37.67 
±8.52 
121.50  
± 21.98 
27.50 
±17.59 
30.49 
±20.32 
5.24 
±0.23 
5.50 
±0.10 
38.67 
±5.77 
Soak-
away 
141.67 
±16.62 
11.67 
±2.84 
87.83 
±25.26 
47.00 
±16.93 
9.33 
±3.78 
4.32 
±0.09 
4.24 
±0.00 
28.67 
± 2.31 
 
 
 
 
 
 
 
 
 
 53 
 
Table 4.14: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at 
the sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2011. 
Exchangeable cations 
Sample 
ID 
Ca Mg K Na CEC S-
value 
Base 
saturation 
% 
pH 
(H2O) 
pH 
(KCl) 
cmol(+)/kg  
¼ 
Denuded 
2.80 
±0.76 
1.94 
±0.50 
0.35 
±0.11 
0.52 
±0.06 
9.42 
±1.26 
5.61 
±1.42 
59.35 
±10.49 
4.88 
±0.27 
5.84 
±0.48 
¾ 
Denuded 
0.84 
±0.09 
0.60 
±0.10 
0.07 
±0.02 
0.25 
±0.12 
3.74 
±0.56 
1.77 
±0.20 
47.58 
±1.85 
4.39 
±0.15 
5.36 
±1.23 
Plant 
 
2.77 
±2.15 
0.44 
±0.03 
0.24 
±0.05 
0.25 
±0.07 
5.18 
±0.49 
3.69 
±2.08 
70.46 
±36.83 
5.78 
±1.34 
5.92 
±1.21 
Prills 
 
1.28 
±0.21 
0.31 
±0.07 
0.31 
±0.06 
0.12 
±0.08 
4.41 
±0.68 
2.02 
±0.39 
45.84 
±4.37 
5.24 
±0.23 
5.50 
±0.10 
Soak-
away 
 
0.71 
±0.08 
0.10 
±0.02 
0.23 
±0.06 
0.20 
±0.07 
2.71 
±0.20 
1.23 
±0.19 
45.31 
±4.16 
4.32 
±0.09 
4.24 
±0.00 
 
 
 
 
 
 
 
 54 
 
As mentioned in the previous chapter, the 1:2 water extract shows the elemental 
concentrations of in the soil that is soluble in water and thus available for plant 
uptake. From the 2011 data it is evident that the K concentrations exceeded the 
norm value of 0.84mmol l-1 at the ¼ denuded area sampling site, but in contrast the 
other four sampling sites showed low K concentrations. The Na norm value of <2 
mmol l-1 was also exceeded at the ¼ denuded area sampling site. At each of the 
sampling sites the PO4 concentrations were found to be below the norm value of 
0.03 mmol l-1. Potentially, phytotoxic values of 61.18 mmol l-1 and 16.71 mmol l-1 at 
the ¼ denuded and ¾ denuded sampling sites, respectively, for both  NO3 (norm 
value is 2.1 mmol l-1) and NH4 (norm value is 0.03 mmol l
-1) characterized both these 
sampling sites in terms of the data collected during 2011. In the three remaining 
sampling sites, the NO3 values are below the concentrations required for optimal 
plant establishment and growth. At each of the sampling points the HCO3 values 
indicated that the soil was devoid of any buffering capacity at that time (2011) with 
values well below the norm value of 0.8 mmol l-1 typifying the sites. 
With regard to the micro elemental concentrations as reflected by the 1:2 water 
extract data, the Mn concentrations stood out as a possible element of concern with 
elevated average levels of 240.20 μmol l-1 and 110.41 μmol l-1 respectively, at the ¼ 
denuded and ¾ denuded sampling points. This was substantially higher than the 
norm value of 1 μmol l-1 to a maximum value of 3 μmol l-1. This was significant as Mn 
in these concentrations is potentially phytotoxic to plants (Bergmann, 1992). The Zn 
concentration typifying the ¼ denuded area was also elevated above the norm value 
(1 μmol l-1 to a maximum value of 3 μmol l-1) to a level of 7.37 μmol l-1. The pH of all 
the sampling points were lower than the norm of 6.3 with the EC values at the ¼ 
denuded area as well as at the ¾ denuded area that showed very high values.  
 
 
 
 
 
 55 
 
Analysis of the results from the ammonium acetate extraction procedure was done 
by the calculation of the exchangeable cation ratios. This provides us an indication of 
the ratios between the different cations in the ratio of Ca:Mg:K:Na. The norm ratio 
values for this analysis is 65:25:8:2 and is calculated by using the following equation: 
100 𝐶𝑎
𝑆
:
100 𝑀𝑔
𝑆
:
100 𝐾
𝑆
:
100 𝑁𝑎
𝑆
. 
 
The S-value is indicated for each sample and is calculated as follows: 
𝑆 − 𝑣𝑎𝑙𝑢𝑒 =  𝐶𝑎 +  𝑀𝑔 +  𝐾 +  𝑁𝑎, 
(MVSA, 1997). 
Table 4.15: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2011 samples. 
Sample 
ID 
Ca Mg K Na 
(cmol(+)/kg ± SD) 
¼ 
Denuded 
49.91±2.84 34.58±2.35 6.24±0.65 9.27±0.87 
¾ 
Denuded 
47.46±2.21 33.90±3.70 3.95±1.30 14.12±5.03 
Plant 
 
75.07±10.96 11.92±4.26 6.50±3.99 6.78 ± 3.53 
Prills 
 
63.37±1.49 15.35±2.23 15.35±3.35 5.94 ± 2.60 
Soak-
away 
57.72±4.30 8.13±6.50 18.70±3.99 16.26±2.95 
Norm 
Values 
65 25 8 2 
 
 
 56 
 
In comparison to the norm ratio it was clear that the Ca percentages were low at the 
¼ denuded, ¾ denuded and the soak-away sample points while the Na percentages 
were much higher than the norm value of 2%. The pH(KCl) was also above 5.5, 
which can result in micro-elemental toxicities in plants. This is due to the fact that the 
micro-elements become soluble under these conditions (FERTASA, 2016). The low 
CEC value is also indicative of a limited binding capacity on the exchangeable matrix 
that can potentially be problematic for different physical soil properties. 
4.6   Summary results and discussion of soil analysis (2013) 
Table 4.16:  Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2013. 
Macro-elements 
Sample 
ID 
Ca Mg K Na PO4 SO4 NO3 NH4 Cl HCO3 
Millimol per litre 
¼ 
Denuded 
3.76 
±1.91 
0.44 
±0.38 
0.05 
±0.04 
0.29 
±0.12 
0.01 
±0.01 
0.22 
±0.11 
7.89 
±5.00 
0.04 
±0.03 
0.07 
±0.06 
0.47 
±0.64 
¾ 
Denuded 
13.02 
±11.95 
7.08 
±7.47 
1.08 
±1.03 
2.06 
±1.96 
0.02 
±0.02 
0.60 
±0.44 
41.52 
±40.50 
0.20 
±0.19 
0.38 
±0.34 
0.38 
±0.28 
Plant 
 
2.32 
±1.313 
0.44 
±0.26 
0.38 
±0.20 
0.45 
±0.30 
0.00 
±0.00 
0.95 
±0.09 
4.84 
±3.39 
0.12 
±0.19 
0.09 
±0.06 
0.00 
±0.00 
Prills 
 
0.09 
±0.03 
0.07 
±0.03 
0.35 
±0.16 
0.07 
±0.02 
0.00 
±0.00 
0.32 
±0.05 
0.16 
±0.11 
0.13 
±0.09 
0.08 
±0.04 
0.05 
±0.00 
Soak-
away 
0.15 
±0.08 
0.07 
±0.03 
0.20 
±0.18 
0.07 
±0.01 
0.01 
±0.01 
0.31 
±0.07 
0.11 
±0.10 
0.08 
±0.12 
0.07 
±0.06 
0.05 
±0.00 
 
 
 
 57 
 
Table 4.17: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2013. 
Micro-elements and other data 
Sample 
ID 
Fe Mn Cu Zn B pH EC 
Micromol per liter  mS/cm 
¼ 
Denude
d 
15.42±15.87 50.13±72.64 0.31±0.12 0.35±0.37 16.83±9.98 5.87±1.33 0.89±0.47 
¾ 
Denude
d 
15.55±23.38 44.10±71.19 0.62±0.11 0.70±0.13 19.10±8.63 6.11±0.94 4.35±4.17 
Plant 
 
0.78±0.39 154.98±122.24 0.91±0.40 9.30±4.50 2.97±2.73 4.07±0.04 0.68±0.35 
Prills 
 
25.28±10.41 1.08±0.70 0.06±0.03 0.28±0.12 9.56±9.21 4.75±0.08 0.09±0.03 
Soak-
away 
34.53±6.65 1.65±0.82 0.42±0.03 0.93±0.18 10.01±6.19 4.88±0.03 0.09±0.03 
 
 
 
 
 
 
 
 
 58 
 
Table 4.18: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the 
soil at the sampling points, by utilising the ammonium acetate extraction analysis 
during 2013. 
Nutrient status 
Sample 
ID 
Ca Mg K Na P pH(H2O) pH(KCl) EC 
mg/kg  mS/m 
¼ 
Denuded 
1214.33 
±985.37 
63.33 
±37.33 
22.33 
±6.75 
9.50 
±1.32 
50.43 
±7.43 
5.81 
±1.70 
5.48 
±1.86 
178.33 
±92.65 
¾ 
Denuded 
1327.00 
±694.82 
266.83 
±164.61 
97.33 
±51.81 
48.83 
±38.76 
62.75 
±38.76 
5.73 
±0.70 
5.62 
±1.39 
1114.00 
±934.46 
Plant 
 
147.17 
±61.92 
13.83 
±9.00 
53.17 
±12.39 
11.83 
±8.13 
46.66 
±17.22 
3.96 
±0.14 
3.75 
±0.10 
123.67 
±47.54 
Prills 
 
146.67 
±25.18 
31.67 
±5.53 
115.00 
±3.00 
3.17 
±1.89  
25.58 
±16.83 
4.93 
±0.01 
4.02 
±0.07 
17.67 
±2.52 
Soak-
away 
302.67 
±45.62 
40.83 
±3.25 
91.33 
±28.11 
1.67 
±1.26 
72.08 
±31.32 
5.04 
±0.10 
4.06 
±0.10 
22.33 
±7.02 
 
 
 
 
 
 
 
 
 
 59 
 
 
Table 4.19: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at 
the sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2013. 
Exchangeable cations 
Sample 
ID 
Ca Mg K Na CEC S-
value 
Base 
saturation 
% 
pH 
(H2O) 
pH 
(KCl) 
cmol(+)/kg  
¼ 
Denuded 
6.06 
±4.37 
0.52 
±1.30 
0.06 
±0.11 
0.04 
±0.14 
13.38 
±2.24 
6.68 
±5.84 
48.40 
±49.58 
5.81 
±0.85 
5.48 
±1.57 
¾ 
Denuded 
6.62 
±0.83 
2.20 
±0.03 
0.25 
±0.13 
0.21 
±0.04 
13.34 
±1.42 
9.28 
±0.70 
74.33 
±8.02 
5.73 
±0.49 
5.62 
±0.16 
Plant 
 
0.73 
±0.23 
0.11 
±0.03 
0.14 
±0.07 
0.05 
±0.01 
11.62 
±1.03 
1.04 
±0.21 
8.94 
±2.36 
3.96 
±0.10 
3.75 
±0.15 
Prills 
 
0.73 
±2.71 
0.26 
±0.32 
0.29 
±0.04 
0.01 
±0.02 
12.41 
±2.21 
1.03 
±3.01 
4.93 
±18.65 
4.93 
±1.07 
4.02 
±0.95 
Soak-
away 
1.51 
±5.57 
0.34 
±1.58 
0.23 
±0.15 
0.01 
±0.19 
12.56 
±1.10 
2.09 
±6.39 
5.04 
±48.66 
5.04 
±1.38 
4.06 
±1.77 
 
 
 
 
 
 
 
 60 
 
As was the case in 2011, the 1:2 water extract data for 2013 still reflected potential  
K deficiencies at the majority of the sampling points. Furthermore, it became clear 
that the K decreased dramatically from the toxic levels from 2011 at the ¼ denuded 
sample point. The same held true for the Na concentrations, but the Na 
concentration increases at the ¾ sampling point over the same period. The NO3 
values decreased dramatically at the ¼ denuded sampling point but was still at very 
high levels at that time (2013). An increase in the NO3 concentration became 
apparent in the ¾ denuded samples down slope and a potentially problematic 
increase was noted at the plant sampling point in terms of the NO3 value. NH4 
concentrations decreased from phytotoxic values in 2011 to acceptable values within 
the norm value range for this sampling period. There was a slight increase in the 
HCO3 values of almost all sampling sites but all was still critically low in 2013. 
The micro-elemental and other data (1:2 water extract) still mirrored the concerns in 
terms of high Mn values although it decreased from 2011 in the ¼ denuded sample 
point and the ¾ denuded sample point. The plant sampling point revealed a dramatic 
increase in Mn concentrations in the 2013 data as well as the Zn concentration at the 
plant sampling site. In terms of the pH values (water extractable) an increase was 
seen at the ¼ denuded area and the ¾ denuded area to an acceptable value but a 
decrease was seen at the plant area while the other two sampling sites (prills and 
soak-away) remained almost unchanged. 
 
 
 
 
 
 
 
 
 61 
 
Table 4.20: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2013 samples. 
 Ca Mg K Na 
¼ 
Denuded 
90.72±2.66 7.78±1.74 0.90±0.48 0.60±0.71 
¾ 
Denuded 
71.34±1.39 23.71±5.19 2.70±4.02 2.27±0.37 
Plant 
 
70.20±2.00 10.58±3.39 13.47±2.69 4.81±1.20 
Prills 
 
56.16±0.13 20.00±0.05 22.31±0.01 0.77±0.01 
Soak-
away 
72.25±0.23 16.27±0.03 11.01±0.07 0.48±0.01 
Norm 
Values 
65 25 8 2 
 
In general the Ca percentages showed an increase from the 2011 analysis. We see 
a dramatic decrease in Na percentages, however, these percentages was still high 
at the ¾ denuded area and the plant sampling point. The CEC values revealed an 
increase, but the pH values had a negative change and decreased further over this 
period. 
 
 
 
 
 
 
 
 
 62 
 
4.7   Summary of the results obtained in the 2016 data 
Table 4.21: Mean macro-elements (mmol l-1 ± SD), measured with the 1:2 water 
extraction procedure, present in the soil at the sampled areas during 2016. 
1:2 Water extract 
Macro-elements 
Sample 
ID 
Ca Mg K Na PO4 SO4 NO3 NH4 Cl HCO3 
Millimol per litre 
¼ 
Denuded 
2.08 
±0.24 
0.35 
±0.05 
1.40 
±0.16 
0.12 
±0.03 
0.04 
±0.01 
0.29 
±0.16 
4.33 
±0.92 
0.04 
±0.03 
0.07 
±0.06 
0.73 
±0.40 
¾ 
Denuded 
13.94 
±1.06 
7.31 
±2.42 
2.30 
±0.41 
2.80 
±0.86 
0.03 
±0.01 
1.35 
±0.22 
44.13 
±6.33 
0.35 
±0.09 
0.48 
±0.09 
0.60 
±0.13 
Plant 
 
2.58 
±0.42 
0.37 
±0.09 
0.70 
±0.06 
0.50 
±0.04 
0.03 
±0.01 
0.77 
±0.12 
5.19 
±0.83 
0.18 
±0.12 
0.08 
±0.02 
0.45 
±0.05 
Prills 
 
0.55 
±0.21 
0.16 
±0.05 
0.42 
±0.10 
0.16 
±0.04 
0.02 
±0.01 
0.28 
±0.08 
0.89 
±0.33 
0.13 
±0.09 
0.13 
±0.08 
0.50 
±0.05 
Soak-
away 
0.50 
±0.07 
0.14 
±0.04 
0.35 
±0.08 
0.12 
±0.06 
0.01 
±0.01 
0.29 
±0.09 
0.55 
±0.07 
0.08 
±0.05 
0.09 
±0.04 
0.63 
±0.10 
 
 
 
 
 
 
 
 
 63 
 
Table 4.22: Mean micro-elements (µmol l-1 ± SD) and other data (pH, mS/cm), 
measured with the 1:2 water extraction procedure, present in the soil at the sampled 
areas during 2016. 
Micro-elements and other data 
Sample 
ID 
Fe Mn Cu Zn B pH EC 
Micromol per litre  mS/cm 
¼ 
Denuded 
11.17±6.30 17.41±13.94 0.31±0.12 0.30±0.33 11.58±4.16 5.96±0.33 0.65±0.07 
¾ 
Denuded 
5.95±5.71 23.43±28.49 0.55±0.03 0.60±0.15 12.33±3.06 6.50±0.42 4.80±0.67 
Plant 
 
0.78±0.39 37.11±18.64 0.77±0.35 7.82±3.87 3.32±1.53 6.04±0.06 0.74±0.10 
Prills 
 
21.95±8.98 1.05±0.75 0.06±0.03 0.26±0.15 7.00±2.00 6.61±0.23 0.22±0.04 
Soak-
away 
27.81±7.69 1.29±0.20 0.38±0.07 0.87±0.26 9.69±6.74 6.27±0.57 0.19±0.03 
 
 
 
 
 
 
 
 
 
 
 64 
 
Table 4.23: Mean nutrient status (mg kg-1 ± SD), pH and EC (mS m-1) present in the 
soil at the sampling points, by utilising the ammonium acetate extraction analysis 
during 2016. 
Nutrient status  
Sample 
ID 
Ca Mg K Na P pH(H2O) pH(KCl) EC 
mg/kg  mS/m 
¼ 
Denuded 
494.00 
±5.27 
70.83 
±2.08 
179.33 
±2.02 
3.33 
±0.29 
56.28 
±0.34 
5.85 
±0.07 
5.21 
±0.05 
61.33 
±1.53 
¾ 
Denuded 
1340.50 
±10.48 
85.67 
±2.47 
241.50 
±4.58 
5.33 
±1.04 
28.27 
±0.17 
7.56 
±0.06 
7.13 
±0.25 
63.00 
±14.80 
Plant 
 
997.17 
±3.55 
140.83 
±2.08 
188.33 
±2.47 
4.50 
±0.50 
61.00 
±0.17 
6.53 
±0.05 
5.95 
±0.09 
48.00 
±3.00 
Prills 
 
3296.00 
±5.89 
149.33 
±1.61 
158.17 
±3.01 
11.67 
±1.04 
26.68 
±0.21 
8.19 
±0.04 
8.03 
±0.07 
52.67 
±1.15 
Soak-
away 
3408.00 
±7.09 
329.33 
±2.02 
241.67 
±3.88 
23.33 
±1.04 
35.32 
±0.20 
8.07 
±0.06 
7.97 
±0.11 
169.33 
±3.06 
 
 
 
 
 
 
 
 
 65 
 
Table 4.24: Mean exchangeable cations (cmol(+)/kg ± SD) (% and pH) in the soil at 
the sampling points, measured by utilizing the ammonium acetate extraction analysis 
during 2016. 
Exchangeable cations 
Sample 
ID 
Ca Mg K Na CEC S-
value 
Base 
saturation 
% 
pH 
(H2O) 
pH 
(KCl) 
cmol(+)/kg   
¼ 
Denuded 
2.47 
±0.03 
0.58 
±0.02 
0.46 
±0.01 
0.01 
±0.00 
8.93 
±0.09 
3.52 
±0.03 
39.46 
±0.11 
5.85 
±0.07 
5.21 
±0.05 
¾ 
Denuded 
6.69 
±0.05 
0.71 
±0.02 
0.62 
±0.01 
0.02 
±0.00 
7.11 
±0.12 
8.04 
±0.08 
113.05 
±2.06 
7.56 
±0.06 
7.13 
±0.25 
Plant 
 
4.98 
±0.02 
1.16 
±0.02 
0.48 
±0.01 
0.02 
±0.00 
10.75 
±0.15 
6.64 
±0.02 
61.73 
±0.82 
6.53 
±0.05 
5.95 
±0.09 
Prills 
 
16.45 
±0.01 
1.23 
±0.01 
0.41± 
0.01 
0.05 
±0.00 
13.08 
±0.07 
18.13 
±0.05 
138.62 
±0.46 
8.19 
±0.04 
8.03 
±0.07 
Soak-
away 
17.01 
±0.02 
2.71 
±0.02 
0.62 
±0.01 
0.10 
±0.00 
14.69 
±0.06 
20.44 
±0.04 
139.10 
±0.31 
8.07 
±0.06 
7.97 
±0.11 
 
In 2016 the 1:2 water extract data indicated an increase in K at each of the sampling 
points. At the ¼ denuded sampling point the value increased to a acceptable 
concentration, but the ¾ denuded sampling point still showed excess K while the 
other 3 samples was still slightly below the norm of 0.84 mmol l-1. The Na at the ¾ 
denuded sample point still exceeded the norm value and even showed a slight 
increase. The NO3 values all showed an increase from 2013 to 2016 accept for the 
¼ denuded area in which the value decreased. The NO3 values still exceeded the 
norm value of 2.1 mmol l-1 at the ¼ denuded, ¾ denuded and plant sampling points 
with the ¾ denuded point at an extremely high value of 44.13 mmol l-1 on average. 
 66 
 
The HCO3 values increased from almost 0.00 mmol l
-1 to values with a minimum of 
0.45 which is much closer to the norm value of 0.83 mmol l-1. 
The micro-elements and other data of the 1:2 water extract revealed another 
decrease in the Mn concentrations from 2013 to 2016. However, these values were 
still much higher than the norm in the ¼ denuded area, the ¾ denuded area as well 
as the plant sampling point. The Zn concentration decreased but was also still high 
at the plant sampling point. All the pH values showed an increase to a value within 
the acceptable norm value with the EC value of 4.80 mS/cm (which is still high) 
present at the ¾ denuded sample point.  
Table 4.25: Mean exchangeable cation ratios (cmol(+)/kg ± SD) from the ammonium 
acetate extraction procedure in the 2011 samples. 
 Ca Mg K Na 
¼ 
Denuded 
70.17±0.52 16.48±0.45 13.07±0.04 0.29±0.03 
¾ 
Denuded 
83.21±0.20 8.83±0.19 7.72±0.11 0.25±0.05 
Plant 
 
75.00±0.22 17.47±0.23 7.23±0.11 0.31±0.03 
Prills 
 
90.74±0.11 6.79±0.05 2.27±0.04 0.28±0.02 
Soak-
away 
83.22±0.06 13.26±0.09 3.04±0.04 0.49±0.02 
Norm 
Values 
65 25 8 2 
 
From these ratios we can see that, from 2013 to 2016, the Na percentages were 
reduced to acceptable levels for plant growth and a similar conclusion can be drawn 
regarding the rising Ca percentages. The pH values also showed an increase to 
more favourable conditions. 
 
 67 
 
4.8   Summary results and discussion of water analysis (2013) 
Table 4.26: Mean macro-elements (mg/l ± SD) present in the water samples taken 
from the collection pond in 2013. 
Macro-elements 
Sample 
ID 
Ca Mg K Na PO4 SO4 NO3 NH4 Cl HCO3 
Milligram per litre 
Pond 
water 
272.93 
±6.17 
125.82 
±3.04 
35.97 
±1.17 
60.08 
±1.31 
<0.01 
±0 
74.78 
±1.30 
1960.15 
±53.06 
100.83 
±3.40 
14.03 
±1.61 
1.02 
±0.35 
 
The target water quality range for this analysis is specified by DWAF (Department of 
Water Affairs and Forestry of South Africa) (1996). Here we saw the drainage 
system that had been constructed was working effectively because of the high 
concentrations of NO3 and NH4 that was leaching from the soil. The norm value for 
NO3 is <26.56 mg/l
 and in this case it was elevated to an average value of 1960.15 
mg/l. The NH4 norm value is <1.29 mg/l and in this sample an average value of 5.59 
mg/l was seen. Ca and Mg has norm values of <32 mg/l and <30mg/l respectively 
and it was much higher in this case.  
Table 4.27: Mean micro-elements (mg/l ± SD) present in the water samples taken 
from the collection pond in 2013. 
Micro-elements and other data 
Sample 
ID 
Fe Mn Cu Zn B pH EC 
Milligram per litre  mS/cm 
Pond 
Water 
0.42±0.02 12.46±0.43 0.03±0.01 0.14±0.00 0.03±0.00 4.65±0.04 3.37±0.09 
 
 
 
 68 
 
The DWAF norm value for Fe is <0.1 mg/l and in this sample the value measured 
was 0.42 mg/l on average. The Mn should be <0.05 mg/l but here we see an 
elevated average value of 12.46 mg/l. The pH of the water was low with a value of 
4.65 on average while the norm is 6 or higher. The EC value of 3.37 mS/cm is also 
elevated when compared to the norm of <0.7 mS/cm. 
4.9   Overall discussion of data obtained in 2011, 2013 and 2016 
The samples taken from the site showed an undesirable acidity and, in some cases, 
the pH(H2O) values were significantly below the desired margins of 6.5-7 with values 
of below 4 in 2011 and 2013 respectively. The acidity became more suitable for 
vegetation establishment in 2016 but amelioration of the growth medium was still 
needed in order to fully correct the pH and the nutrient imbalances. Together with the 
low pH values a very low buffer capacity, or even a complete lack of buffer capacity 
was recorded in the soil in 2011. The buffer capacity was still low in 2013, but in 
2016 the bicarbonate levels increased to values closer to the norm value of 
0.83mmol l-1. The acidic conditions in the soil created an environment where metals 
like Mn and Fe were soluble and they were present in phytotoxic concentrations. 
This was seen in the water analysis data from the samples that were collected from 
the collection pond in 2013. The pH levels needed to be corrected in in order to 
prevent the liberation of metals and keep these contaminants from entering the 
natural stream downslope of the area. In order to achieve this, a lime application of 
15 tons per hectare was recommended.  
The results seen in the plant, prills and soak-away areas were difficult to monitor. 
This is due to the fact that, after the rehabilitation procedures of RC1, these areas 
were used by a local farmer to produce maize. The drainage system that was 
implemented was effective in reducing the undesirably high levels of inorganic solids 
in the ¼ denuded area to acceptable levels that made a revegetation program 
possible. In contrast, the plant, prills and soak-away areas showed signs of increase 
in these inorganic solids, but this could be accounted for by the land use for maize 
production.  
 
 
 69 
 
Nutrient imbalances and deficiencies have been identified by means of soil analysis. 
This analysis includes the following: Ca shortages, Ca: Mg imbalances, K shortages, 
P shortages, Zn shortages, excessive Na and an organic matter deficiency. The Ca 
shortage and the Ca: Mg imbalance was corrected by the lime application while the 
problems associated with K, P and Zn were corrected by the application of 
potassium sulphate, soft rock phosphate and zinc sulphate. The lime application had 
another function, that of displacing the excessive Na from the colloidal fraction in the 
soil to be leached as Na2SO
4. The organic matter fraction was increased by applying 
both humic and fulvic acids that is water soluble and also degradable by 
microorganisms. The humic and fulvic acids applied to the soil also ads carbon to the 
growth medium which is a source of energy for soil microbes and also a vital element 
with regard to soil health (Waring et al., 2014; Brady & Weil, 2008). 
The gully erosion on the site was caused by a lack of vegetation (Montgomery, 2007) 
together with poor soil structure (Brady & Weil, 2008). This problem was addressed 
by tillage of the soil against the natural gradient, construction of contour berms and 
vegetation establishment. The vegetation establishment on the site was not only 
done to prevent soil erosion, but also to restore the natural succession of vegetation 
with regard to the predetermined specifications needed to rehabilitate the site.  
The microbial activities in the soil (bacteria and fungi) play an integral role in soil 
health, plant health and also the utilization of nutrients by plants. The microbial 
inoculation of the soil plays a large role in the success of the rehabilitation of the site 
for the previously mentioned reasons. With the nutrients present in the soil together 
with the low pH values we saw the growth medium as an abiotic environment, this 
situation needed to be corrected (Miransari, 2012). The micro-organisms that were 
introduced into the soil are beneficial to the health of the soil and the plants 
established in it (Waring et al., 2014). 
 
 
 
 
 70 
 
4.10  Changes in values between 2011 and 2016 
Table 4.28: P-values associated with t-tests in the 1:2 water extraction procedure 
(macro-elements) data for the denuded area between 2011 and 2016. 
Sample K Na PO4 NO3 NH4 HCO3 
¼ 
Denuded 
0.003 0.000 0.011 0.000 0.000 0.044 
¾ 
Denuded 
0.006 0.022 0.032 0.006 0.000 0.008 
 
Table 4.29: P-values associated with t-tests in the 1:2 water extraction procedure 
(micro-elements and other data) data for the denuded area between 2011 and 2016. 
Sample Mn Zn EC 
¼ Denuded 0.000 0.001 0.000 
¾ Denuded 0.010 0.000 0.005 
 
Table 4.30: P-values associated with t-tests in the ammonium acetate 
(Exchangeable cations) data for the denuded area between 2011 and 2016. 
Sample Ca CEC pH(H2O) pH(KCl) 
¼ Denuded 0.224 0.077 0.000 0.005 
¾ Denuded 0.000 0.001 0.000 0.029 
 
With Student’s one sided t-test for equality of means at a nominal significant level of 
5%, all the p-values below 0.05 indicate statistically significant differences between 
the sample means. Virtually all of the tests reject the null hypothesis of equality of 
means. As a result we infer that there are substantial differences between the results 
of measurements between 2011 and 2016. This indicates that the rehabilitation 
program has resulted in statistically significant changes in the composition of the soil. 
This, in turn, provides strong statistical evidence to support the claim that the 
rehabilitation program has improved the quality of the soil. 
 71 
 
4.11 Vegetation survey (monitoring of the site) 
Due to the fact that the sample sites at the plant footprint, prills silo and old soak-
away were used for crop production by a local farmer; the area was not seeded with 
the prescribed seed mix, however, the amelioration was implemented. The 
vegetation assessment was done in September 2016. This assessment included 
only the previously denuded Transects, which were numbered from 1 to 5, starting 
from the farmers field (up-slope) and moving down to the natural stream. Figure 4.5 
indicates the transects and each quadrant on them: 
 
Figure 4.7: Map overview of the rehabilitated area (the white lines indicate the 
transects numbered 1 to 5 from North to South and the red blocks indicate the 
quadrants numbered from 1 to 5 from the East to the West). 
 
 
 
 
 
 72 
 
4.11.1  Mean transect data 
The average of the various pasture species and composition of each transect are 
provided in Table 4.28. The results show germination of all the different seeds that 
were introduced. Only species that have grown to reproductive maturity are included.  
Table 4.31: Mean grass species (amount ± SD) found over all five transects. 
Species Average amount 
Cynodon dactylon 
4.40 ± 2.97 
Pennisetum clandestinum  
2.80 ± 3.11 
Eragrostis curvula 
15.60 ± 2.30 
Eragrostis tef 
4.00 ± 1.58 
Digitaria eriantha 
22.60 ± 3.29 
Sorghum bicolor 
4.20 ± 3.11 
Weeds 
1.80 ± 1.30 
 
 
 
 
 
 
 
 
 
 
 
 73 
 
4.11.2   Mean quadrant data 
Table 4.32: Mean of each species (amount ± SD) found between the different 
quadrants on the five transects. 
Species Quadrant 1 Quadrant 2 Quadrant 3 Quadrant 4 Quadrant 5 
Cynodon 
dactylon 
1.33 ± 0.58 3.67 ± 4.04 1.25 ± 0.50 2.00 ± 1.83 1.20 ± 1.30 
Pennisetum 
clandestinum  
1.00 ± 1.00 5.00 ± 8.66 2.25 ± 2.63 0.50 ± 0.58  0.40 ± 0.55 
Eragrostis 
curvula 
4.00 ± 2.00 3.67 ± 1.53 5.00 ± 3.92 6.25 ± 1.71 7.80 ± 4.60 
Eragrostis  
tef 
2.00 ± 2.65 5.33 ± 1.15 7.25 ± 4.65 7.25 ± 2.99 3.40 ± 1.14 
Digitaria 
eriantha 
7.00 ± 2.65 11.67 ± 9.02 6.75 ± 2.22 15.75 ± 3.40 8.40 ± 2.61 
Sorghum 
bicolor 
1.33 ± 1.15 1.33 ± 1.53 0.75 ± 0.96 1.25 ± 1.26 1.40 ± 0.55 
Weeds 
1.67 ± 1.15 1.00 ± 1.73 1.00 ± 0.82 2.00 ± 1.63 2.40 ± 1.14 
 
4.12 Weather data of the area 
During the rehabilitation of the site between 2010 and 2016 South Africa had a few 
years of severe drought with higher than normal temperatures. The following 
weather data was compiled by and received from the South African Weather 
Service. The data was extracted on 6 November 2017. 
 
 
 74 
 
Table 4.33: Average maximum temperatures (̊C) measured in Emalahleni (weather 
station 0515320 8) between – 2010 and 2016. 
Year 
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 
2010 24.8 
±2.21 
26.2 
±2.14 
25.7 
±2.17 
21.4 
±3.80 
20.8 
±2.78 
18.1 
±3.13 
17.9 
±2.49 
21.7 
±3.37 
25.9 
±2.75 
27.1 
±3.17 
25.3 
±3.66 
25.4 
±2.99 
2011 23.9 
±2.50 
N/A N/A N/A N/A N/A N/A N/A N/A 28.7 
±3.57 
27.9 
±3.52 
27.9 
±3.40 
2012 28.3 
±2.64 
29.8 
±1.41 
28.5 
±2.91 
24.6 
±2.13 
24.7 
±2.79 
19.9 
±2.88 
20.9 
±1.97 
22.5 
±5.21 
22.4 
±4.58 
24.4 
±3.74 
25.7 
±3.96 
26.3 
±2.84 
2013 27.3 
±3.36 
29.3 
±2.38 
27.0 
±2.78 
23.4 
±3.84 
22.6 
±1.53 
21.3 
±2.27 
20.3 
±2.35 
20.6 
±3.31 
26.1 
±4.90 
26.0 
±4.13 
28.3 
±2.73 
28.6 
±3.60 
2014 31.2 
±2.37 
30.5 
±2.00 
26.9 
±3.05 
26.1 
±1.98 
26.2 
±2.39 
23.4 
±2.47 
22.3 
±2.81 
24.4 
±3.02 
29.6 
±3.31 
29.4 
±3.93 
29.0 
±1.36 
N/A 
2015 N/A 28.3 
±N/A 
27.2 
±1.86 
24.0 
±2.59 
24.6 
±1.93 
18.8 
±2.36 
19.6 
±2.45 
24.5 
±3.25 
25.2 
±4.81 
28.7 
±3.08 
28.0 
±4.18 
29.3 
±2.79 
2016 27.8 
±3.09 
29.5 
±2.81 
26.8 
±3.51 
25.6 
±3.00 
20.9 
±1.77 
19.4 
±3.12 
18.9 
±2.29 
23.3 
±3.08 
22.6 
±7.51 
27.6 
±4.36 
26.0 
±3.27 
27.5 
±3.03 
 
During March and July in 2010, the averages are omitted due to unreliable and 
missing daily data. In 2011 the data for February to September is missing and the 
data for January are deemed unreliable due to missing daily data. In 2013 the 
average maximum temperatures for January and May considered to be unreliable 
due to missing daily data. In 2014 the average maximum temperatures for July, 
November and December contains missing daily data. In 2015 the average 
maximum temperatures for January and February is considered unreliable due to 
missing daily data. In 2016 the average maximum temperatures for September also 
includes missing daily data. 
 
 
 
 
 
 
 
 
 75 
 
Table 4.34: Average (Ave./day) and total (Tot.) monthly rainfall (mm) measured in 
Emalahleni (weather station 0515320 8) between 2010 and 2016. 
 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 
2010 
(Tot.) 
(mm) 
114.8 52.4 67.2 114.0 38.8 0.0 0.2 0.0 0.0 63.4 48.0 259.0 
2010 
Ave./day 
(mm) 
7.7 
±5.34 
5.8 
±6.78 
6.7 
±14.61 
8.1 
±10.27 
6.5 
±6.71 
N/A 0.2 
±N/A 
N/A N/A 6.3 
±7.15 
4.4 
±6.88 
15.2 
±17.95 
2011 
(Tot.) 
(mm) 
172.2 N/A N/A N/A N/A N/A N/A N/A N/A 31.4 31.0 121.6 
2011 
Ave./day 
(mm) 
14.4 
±15.55 
N/A N/A N/A N/A N/A N/A N/A N/A 5.2 
±5.56 
2.8 
±3.48 
6.1 
±9.56 
2012 
(Tot.) 
(mm) 
92.6 39.8 66.6 10.8 0.0 0.8 0.0 0.6 84.8 94.8 92.4 98.6 
2012 
Ave./day 
(mm) 
5.1 
±8.81 
3.3 
±5.89 
6.7 
±8.80  
1.8 
±2.11 
N/A 0.4 
±2.28 
N/A 0.6 
±N/A 
12.1 
±22.04 
10.5 
±13.08 
9.2 
±10.12 
5.2 
±7.58 
2013 
(Tot.) 
(mm) 
75.4 29.4 101.0 79.8 6.4 0.0 0.0 3.2 6 107.6 180.8 115.6 
2013 
Ave./day 
(mm) 
6.3 
±6.50 
3.3 
±3.66 
10.1 
±15.58 
10.0 
±9.97 
1.3 
±2.09 
N/A N/A 1.1 
±1.33 
2.0 
±2.16 
9.8 
±8.83 
10.6 
±17.57 
5.5 
±6.55 
2014 
(Tot.) 
(mm) 
90.6  113.0 221.4 22.8 0.4 9.4 0.0 0.4 1.2 71.0 160.2 197.8 
2014 
Ave./day 
(mm) 
6.0 
±7.49 
9.4 
±14.11 
10.5 
±16.85 
2.3 
±2.77 
0.2 
±0.00 
4.7 
±6.36 
N/A 0.4 
±N/A 
0.4 
±0.20 
7.1 
±8.08 
7.6 
±8.07 
9.4 
±12.11 
2015 
(Tot.) 
(mm) 
65 21.8 30.6 29.4 0.0 0.8 1.0 0.4 46.2 34.2 54.2 135.2 
2015 
Ave./day 
(mm) 
4.3 
±3.96 
2.0 
±1.66 
2.6 
±3.89 
2.9 
±4.02 
N/A 0.8 
±N/A 
0.3 
±0.23 
0.4 
±N/A 
9.2 
±10.14 
3.8 
±3.43 
6.8 
±6.18 
9.0 
±10.33 
2016 
(Tot.) 
(mm) 
85.4 40.8 170.2 5.8 29.0 12.0 3.0 0.6 6.4 84.0 224.6 81.8 
2016 
Ave./day 
(mm) 
5.0 
±4.57 
2.7 
±3.63 
17.0 
±15.24 
1.5 
±2.43 
14.5 
±15.13 
2.0 
±3.24 
0.8 
±0.85 
0.6 
±N/A 
3.2 
±3.96 
12.0 
±6.39 
14.0 
±14.69 
5.5 
±8.56 
 
 76 
 
In 2010 the total and averages for January, September and December are 
considered unreliable due to missing daily values. In 2011 only the total and 
averages of December are reliable due to missing daily values. In 2012 the total and 
average of September are unreliable due to missing daily values. In 2013 the totals 
and averages of January, March, May and August are unreliable due to missing daily 
values. In 2014 the total and average of October are unreliable for the same reason. 
In 2015 the totals and averages of February, June and August are unreliable due to 
missing daily values. In 2016 the total and average of September are also deemed 
unreliable for this reason. 
For the area of Emalahleni that receives a historical average precipitation of 693mm 
annually, we see according to the data provided by the South African Weather 
services that in 2010 a total of 278.8mm, 121.6mm in 2011, 497mm in 2012, 
519.2mm in 2013, 817.2mm in 2014, 395.8mm in 2015 and 737.2mm in 2016 of 
precipitation was recorded. Even though the data contains a large number of missing 
values, the reported values are very low for 2010, 2011, 2012 and 2014 due to 
drought. 
 
 
 
 
 
 
 
 
 
 
 
 77 
 
CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS 
5.1  Conclusions 
When considering the rehabilitation procedures implemented by RC1, it becomes 
clear that their main focus was on the EC of the growth medium. Clearly the 
pollutants that were present in the soil were considered to be of secondary 
importance. The first rehabilitation company stated in a report that the primary 
problem is the salinity, which is caused by the pollutants and that the pollutants did 
not constitute a severe problem in and of itself. Further investigations showed that 
this was not the case and, therefore, RC1 underestimated the degree and extent of 
the pollution on the site.  
Throughout the reports issued by RC1, it is evident that their efforts were mainly 
focused on the EC. This is clear because of the fact that that the EC forms the main 
focus of the monitoring and hand over criteria that were formulated by RC1. The 
results of their growth medium analysis are not seen in any of these reports because 
they focussed more on the physical aspects of the rehabilitation than on the soil 
chemical aspects. The only chemical analysis reported by RC1 were performed in 
order to determine the levels of the EC, pH, NO3, NH4 and SO4 in the soil and water. 
However, these results were not made available in the issued reports. 
RC1 stated in their completion report that all sources of contaminants were removed. 
Later analyses clearly demonstrated that this was not the case. It is clear from the 
soil analysis performed in 2010 that the source of contaminants was not removed, 
due to the high nitrate levels seen after the site had been rehabilitated by RC1. 
Some soil amelioration was attempted on the site by RC1, but this proved to be 
inadequate.  
RC1 constructed various drainage trenches; however, these were not placed in the 
optimal location. The only apparent reason for this is a lack of understanding of the 
underlying geology. The monitoring done by RC1 was inadequate, since they 
deemed the site as being successfully rehabilitated before it returned to its denuded 
state without any human intervention. 
 
 78 
 
One of the reports drafted by RC1 contained a figure that was used in order to 
explain the subsurface geology on the site. When inspecting this figure closely, 
several misconceptions are observed. In actuality, the subsurface laterite does not 
slope in the same direction as the slope on the surface (to the south) but the 
underground geology slopes to the opposite side (as is seen in Figure 5.1). The 
drainage trenches were constructed by RC1 in such a way that it would only stop the 
shallow seepage from the plant, old soak-away and prills silo (which can be seen as 
the source of the pollutants). This strategy did not prove to be successful because 
the contaminants leached deep into the soil over a very long period. When 
conducting the first rehabilitation, RC1 attempted to stop the water on the surface 
and shallow seepage from moving over the denuded area by means of the drainage 
trenches and contour berms. However, shallow seepage from the plant, prills and 
soak-away did not carry substantial concentrations of the pollutants. The contour 
berms constructed by RC1 were constructed with the soil on the site. The soil 
contained a high salt concentration.  This soil is also dispersive, in other words, there 
is a substantial probability that the berms were flushed away by the first rain (Bell & 
Maud, 1994). 
The intensive focus on a physical rehabilitation procedure (mainly the leaching of 
contaminants and the removal of all water from entering the denuded area), coupled 
with the neglect of the ignored chemical imbalances conspired to ensure that the first 
rehabilitation of the site proved to be unsuccessful. As a specific example of the 
problems caused in this regard, consider the bicarbonate (HCO3) concentration in 
the soil. The level of HCO3 was never taken into account by RC1. As a result, the 
soils lacking buffering capacity were not taken into consideration. The low 
bicarbonate was the cause of the low pH values, which, in turn, caused a problem as 
is discussed below (Van Rensburg & Maboeta, 2004).  
 
 
 
 
 79 
 
No nutrients are mentioned in the reports of RC1, and no fertilizers were applied on 
the site in order to create a favourable chemical environment for vegetation 
establishment and growth. Micro-elements weren’t taken into account and with the 
low pH levels, this may cause problems with vegetation establishment due to the 
toxicity of these micro-elements for plants (Sparks, 2003). Furthermore, neither the 
CEC nor base saturation was studied by RC1. A further oversight was that the 
exchangeable cation percentages in the soil were not taken into account. This 
proved to be a significant problem for vegetation establishment.  
The second rehabilitation attempt was substantially aided by a better understanding 
of the underlying geology. This rehabilitation approach showed results that were 
more successful because the drainage channels were placed in the optimal area for 
the removal of the contaminants from the soil. After the first rehabilitation attempt, 
these were still entering the denuded area. With the improved soil chemical 
rehabilitation procedures implemented, the result proved more successful. This 
positive change was affected by creating a viable growth medium for vegetation 
establishment and growth. 
 
Figure 5.1: New understanding of the underlying geology with the sub-surface 
laterite indicated by the black line. 
 
 
 80 
 
Revegetation of the denuded area: 
After the successful revegetation of the site in October 2015, a vegetation survey 
was conducted, in May 2016, in order to determine the success of the rehabilitation. 
The findings were that all the plant species introduced to the site was not only 
present, but has reached maturity and was in a reproductive state. With the drought 
of 2016 in mind, the vegetation cover seen on the site in such a short period of time 
was impressive. 
Soil quality is reflected from the physical, chemical and biological properties of the 
soil. The second rehabilitation approach ensured that each of these properties were 
addressed in order to better the soil quality on the site (Brady & Weil, 2008).  
 
The soil prior to rehabilitation showed substantial areas, completely denuded of all 
vegetation. This was especially noticeable after large amounts of rain that caused 
contaminants to enter these areas. The soil was an abiotic medium with 
characteristics like poor or no structure and with erosion problems all over the 
surface. The little vegetation found on some spots of the site during this time was 
clearly struggling for survival and could be described as sparse. Figure 5.2 displays 
images captured before and after the 
rehabilitation procedure were implemented. 
 
 
Figure 5.2: Photographs taken before the implementation of the rehabilitation 
procedure (on the left) and 1 month after seeding (on the right). 
 81 
 
One month after seeding the area the site changed significantly. Broad leaf weeds 
were present throughout the site, however, the intended species were also present 
in large numbers. Weed control was implemented, but this step will have to be 
repeated at a later time. In Figure 5.3 the transformation is seen from a site denuded 
from all vegetation to a vegetated landscape. 
 
 
 
Figure 5.3: Photos taken 3 months after seeding on the previously denuded area. 
After 3 months the vegetation cover showed a significant increase in planted pasture 
species in spite of the drought that caused lower than average rainfall in the region 
and higher than average daily temperatures.  
 
 
 
 82 
 
5.2  Initial recommendations for area 2 in 2010 
Because the acidification had already stopped in this area in 2010 (because of 
residual HCO3 present, slightly higher average pH and low average EC value) the 
area could have been ameliorated and rehabilitated. The amelioration and 
rehabilitation procedures entailed ripping the area to a depth of 500mm, against the 
natural gradient, and applying 20 tons of dolomitic lime per ha. This was done prior 
to the rainy season and after the first large amount of rain. At that time, it was 
recommended that the area should receive 30 tons of well-cured organic material 
per ha, 300kg of super phosphate and no K. Following this step, the area was 
seeded with 25kg per ha of an indigenous seed mixture.  
5.3  Initial recommendations for area 1 in 2010 
The group 1 samples (taken in the denuded area at two sampling points, namely ¼ 
denuded and ¾ denuded) indicated that it would not be wise to implement any of the 
procedures that were used on area 2 (a combination of the plant, prills and soak-
away) on the denuded area. Investigations into this area indicated that these steps 
would likely not be cost-effective or successful because of the ongoing pollution and 
acidification in this area observed at that time. The pollutants in the soil must was 
required to be removed first and the alkalization process of the soil needed to be 
accelerated. This was done by the construction of drainage channels that were dug 
into the denuded area at a depth of 1 - 2m and starting 5 - 10m up slope from where 
the denuded area is found. Thirty tons of dolomitic lime per ha was then ripped into 
the growth medium at a depth of 500mm in the areas between the channels. This 
area did not receive any further attention before the end of the rainy season. Water 
samples were taken in order to determine whether or not the drains were succeeding 
to remove the pollutants. Additionally, soil sampling was done to determine the 
changes in quality of the growth medium. 
5.4  General recommendations 
It is good practice to keep monitoring the site in future in order to determine the long-
term success of the rehabilitation. From their reports, RC1 believed the site to be 
successfully rehabilitated, but shortly thereafter problems common to many 
rehabilitation problems started to reoccur. The rehabilitation will only truly by 
successful when it is not only stable, but can be considered to be a functional 
ecosystem (Thompson & Thompson, 2004). 
 83 
 
The drainage system has not been removed to date and the liner of the collection 
pond was damaged by some severe wind storms. It is recommended that it should 
be repaired. The reason for not removing these structures at this time is to use the 
water in the collection pond for analysis if any problems occur in the future. This 
should simplify further monitoring.  
The site has been barren for more than 10 years and during the time of the 
rehabilitation and even after reseeding the previously denuded area weeds 
appeared. Weed treatment may be required in the future, however, it is 
recommended that this be determined based on the monitoring of the site.  
Because this is a rehabilitated pasture and cannot yet be seen as a natural pasture, 
more fertilizer applications and other soil amelioration may be needed in future. 
Again, the need for this will be determined by monitoring and sampling the site in 
future. 
5.5. Concluding remarks 
In line with the objectives of this study, the first rehabilitation attempt implemented by 
RC1 was critically evaluated. This provided information regarding, and insight into, 
the problems that made it possible to learn from previous mistakes and compile a 
more efficient rehabilitation procedure. A different and more effective rehabilitation 
procedure was implemented on the site. This rehabilitation procedure proved to be 
successful. The rehabilitation of the site is now complete and the site is considered 
to be successfully rehabilitated to a state where it can be utilized as grazing. 
However, until the coal mine on which the site is situated obtains a closure 
certificate, the mining company will still be responsible for the site from a legal point 
of view. The EMC will not have any liability after monitoring is complete and the site 
is self-sufficient from an ecological perspective. 
 
 
 
 
 84 
 
CHAPTER 6: REFERENCES 
Anonymous, Confidential report no.1. September 2003, Conceptual rehabilitation 
plan. 
Anonymous, Confidential report no.2. September 2003, Methodology statement. 
Anonymous, Confidential report no.3. October 2003, Scope of work for the 
remediation of the site. 
Anonymous, Confidential report no.4. September 2004, Addendum to scope of work 
for the rehabilitation of the site – Remediation of the plant footprint. 
Anonymous, Confidential report no.5. February 2005, Environmental rehabilitation of 
plant – methodology statement. 
Anonymous, Confidential report no.6. October 2008, Completion report for 
rehabilitation of the site. 
Ansari, A. A. (2011). Eutrophication causes, consequences and control. Dordrecht, 
Springer. http://dx.doi.org/10.1007/978-90-481-9625-8. 
Banwart, W.L., Brember, J.M. and Tabatabai, M.A., 1972. Determination of 
ammonium in soil extracts and water samples by an ammonia electrode. Comm. Soil 
Sci. Plant Anal. 3:449. 
Barrett, J. 1978. Vogels textbook of quantitative organic analysis. pp. 435-436. 
Bell, F.G. and Maud, R.R. 1994. Dispersive soils: a review from a South African 
perspective. Quarterly Journal of Engineering Geology and Hydrogeology. 27, 195-
210. 
Bergmann, W., 1992. Nutritional disorders of plants. Gustav Fisher Verlag Jena, 
Stuttgart. 
Bjorklund, P.A. and Mello, F.V. 2012. Soil Organic Matter: Ecology, Environmental 
Impact and Management. Nova Science Publishers, Inc. 
Bradshaw, A.D., Chadwick, M.J. The Restoration of Land. University of California 
Press. Berkeley and Los Angeles. Blackwell Scientific Publications. 
 85 
 
Brady, N. C. & Weil, R. R., 2008. The Nature and Properties of Soils. 14th Edition. 
Pearson Prentice Hall. Upper Saddle River, New Jersey Columbus, Ohio. 
Bray, R. H., & Kurtz, L. T., 1945. Determination of total, organic, and available forms 
of phosphorus in soils. Soil Science. Volume 59, issue 1, p39 – 46. 
Buys, A.J. 1986. MVSA bemestingshandleiding. Hennopsmeer: Die 
Misstofvereniging van Suid-Afrika. 
Cairncross, B. 2006. History if the Okiep Copper district. Johannesburg: Rand 
Afrikaans University. 
Conklin, A. R. 2004. Field Sampling: Principles and Practices in Environmental 
Analysis. CRC Press. 
Department of Water Affair and Forestry (DWAF), 1996. South African Water Quality 
Guidelines. Volume 4. Agricultural use: Irrigation. The Government printer, Pretoria. 
Eberhard, A. 2011. The future of South African coal: market, investment, and policy 
challenges. Working Paper #100 January 2011, Program on Energy and Sustainable 
Development Encina Hall East, Room E415 Stanford University Stanford, CA 94305-
6055. 
Fernando, D. R. and Lynch, J. P. 2015. Manganese phytotoxicity: new light on an old 
problem. Annals of Botany, Volume 116, Issue 3, 1 September 2015, Pages 313–
319. 
FERTASA. 2016. Fertilizer association of Southern Africa. Bemestingshandleiding. 
8th edition. Lynwoodrif, South Africa. 
Folchi, R. 2003. Environmental Impact Statement for Mining with Explosives: A 
QUANTITATIVE METHOD. I.S.E.E. 29th Annual Conference on Explosives and 
Blasting Technique, Nashville, Tennessee, U.S.A. February 2-5, 2003. 
Gee, G.W., and J.W. Bauder. 1986. Particle-size Analysis. P. 383 – 411. In A.L. 
Page (ed.). Methods of soil analysis, Part1, Physical and mineralogical methods. 
Second Edition. 
 
 86 
 
Knowles, R. (1982) Denitrification. Microbiological reviews, 46 (1), Available at:  
Luncedo Ngcofe; Doug I.Cole, Council for Geoscience, Cape Town, South Africa 
The distribution of the economic mineral resource potential in the Western Cape 
Province South African Journal of Science On-line version ISSN 1996-7489 
Print version. ISSN 0038-2353 S. Afr. j. sci. vol.110 n.1-2 Pretoria Jan. 2014. 
Miransari, M. 2012, Soil Microbes and Environmental Health, Nova Science 
Publishers, Inc., Hauppauge. Available from: ProQuest Ebook Central. [13 
November 2017]. 
Misstofvereniging van Suid Afrika (MVSA)., 1997. Bemesting handleiding. Vierde 
hersiende uitgawe. FSSA-MVSA. 
Montgomery, D.R. 2007, Dirt: The Erosion of Civilizations, University of California 
Press, Berkeley. Available from: ProQuest Ebook Central. [13 November 2017]. 
Mucina, L. & Rutherford, M.C., eds. 2006. The vegetation of South Africa, Lesotho 
and Swaziland. Pretoria: South African National Biodiversity Institute. (Memoirs of 
the Botanical Survey of South Africa, Stelitzia 19.) 
Nel, J. 2006. The current risk of asbestos exposure to the citizens of the Prieska 
area, Northern Cape Province in South Africa. (Mini-dissertation). Potchefstroom: 
North-West University. 
Newman, M.C. 2015. Fundamentals of ecotoxicology: The science of pollution. 4th 
edition. Boca Raton, F1 : CRC Press. 
Ramiriz-Munoz, J. 1968. Atomic absorption spectroscopy and analysis by atomic 
absorption flame photometry. American Elsevier Publishing Co., New York. 
Reichardt, M.A. 2012. History of mine wastes rehabilitation techniques in South 
Africa: a multi-disciplinary overview of mine waste rehabilitation and the non-
scientific drivers for its implementation 1950s - 1980s. Johannesburg: University of 
the Witwatersrand. (Thesis - PhD.) 
Rhoads, J.D. 1982. Soluble salts, in A.L. Page, RH Miller and D.R. Keeney, Methods 
of soil analysis part 2: chemical and microbiological properties, Agronomy, 9 (2), 
American Society of Agronomy, Madison, Wisconsin, USA. 167-178 pp. 
 87 
 
Saison, C., Degrange, V., Oliver, R., Millard, P., Commeaux, C., Montange, D., Le 
Roux, X. 2005. Alteration and resilience of the soil microbial community following 
compost amendment: effects of compost level and compost-borne microbial 
community. Environmental Microbiology. Volume 8, Issue 2 
February 2006. Pages 247–257. 
Sarsby, R.W. & Meggyes, T. 2010. Construction for a Sustainable Environment. 
Taylor and Francis group, Boca Raton, CRC Press.  
Skougstd, M.W., Fishman, M.I., Friedman, L.E., Erdman, D.E. & Duncan, S.S., 1979. 
Methods for determination of inorganic substances in water and fluvial sediments. In 
Techniques of water resources investigation of the United States geological survey. 
U.S. Geological Survey. Wahington D.C. Book 5, Chapter AI. 
South Africa. 1996. Constitution of South Africa, Act 108 of 1996. Pretoria: 
Government Printer. 
South Africa. 1998a. National Environmental Management Act 107 of 1998. Pretoria: 
Government Printer. 
South Africa. 2002. Mineral and Petroleum Resources Development Act 2002. 
Pretoria: Government Printer. 
South Africa. 2004a. National Environmental Management Air Quality Act 39 of 
2004. Pretoria: Government Printer. 
South Africa. 2004b. National Environmental Management Biodiversity Act 10 of 
2004. Pretoria: Government Printer. 
South African Weather Service, data 2010 – 2016, weather station [0515320 8] - 
WITBANK  -25.8320 29.1920 1555 m  2016   08:00 (Extracted 2017/11/06 15:10). 
Spain, J.C., Hughes, J.B. and Knackmuss, H. 2000. Biodegradation of Nitroaromatic 
Compounds and Explosives. Lewis Publishers, Boca Raton. 
Sparks, D. (2003) Environmental soil chemistry. 2nd ed. Amsterdam: Academic 
Press. 
 88 
 
Spellman F.R. 2010. The science of environmental pollution, 2nd Edition, CRC Press 
Taylor & Francis Group, 6000 Broken Sound Parkway NW, Suite 300, Boca Raton, 
FL 33487-2742. 
Ter Braak, C.F. and Šmilauer, P.: 1998, CANOCO Reference Manual for user’s 
guide to CANOCO Community Ordination (ver. 4), Microcomputer Power, Ithaca, 
NY, USA. 
Thomas, G.W., 1982. Exchangeable acidity. In A.L. Page (ed.). Methods o soil 
analysis. Part 2, 159-161. Am. Soc, Agron. Madison, Wis. 
Thompson, G. & Turk, J. 2007. Earth Science and the Environment. 4th Edition. 
Thompson Higher Education. Belmont, USA. 
Thompson, S. A. and Thompson, G. G., 2004. Adequacy of rehabilitation monitoring 
practices in the Western Australian mining industry. Ecological management & 
restoration, vol 5, no 1, April 2004, p.30 – 33. 
Tidmarsh, C.E.M.; Havenga, C.M. 1955. The wheel-point method of survey and 
measurement of semi-open grasslands and Karoo vegetation in South Africa. 
Grootfontein Coll. Agric., Union of S. Africa. Mem. 29 Bot. Surv. S. Afr. 1955 pp.53 
pp. ref.bibl. 11. 
Van Rensburg, L. & Pistorius, L., 1998 An investigation into the problems associated 
with revegetating chrysotile tailings. South African journal of Plant and Soil 15, 130 – 
140. 
Van Rensburg, L. and Maboeta, M. S. 2004. Rehabilitation of co-disposed kimberlite 
tailings: Quantifying physical and chemical properties from a rehabilitation 
perspective. South African Journal of Plants and Soil, 21:3, 133-138. 
Walkley, A. 1935. An examination of methods for determining organic carbon and 
nitrogen in soils. Journal of Agricultural Science 25: 598-609.McDonald, D.A. 2004. 
Environmental Justice in South Africa. Juta and Company Ltd. 
 
 
 89 
 
Waring C., Stockmann U., Malone B.P., Whelan B., McBratney A.B. (2014) Is 
Percent ‘Projected Natural Vegetation Soil Carbon’ a Useful Indicator of Soil 
Condition?. In: Hartemink A., McSweeney K. (eds) Soil Carbon. Progress in Soil 
Science. Springer, Cham.