_~ ' •. O- _ - ._- _j
''>:ijinl)l\''ii;;._.JJG~~~_1'r ~,,_~
HIERDIE EKSn ..if;'L\/\H ;::'f~5'(-~~~D~E!,~'~ University Free State
GEEN Ol\lSTANDIGUEOE UIT DlE I
~l II~~~~~~~~~I~II~~34300004920298
M
~.~.~~~TE",~~~~~~?:,~~\:~:.~:p".~~J Universiteit Vrystaat~

lHE ~~fllJlIEINCE OIF flOOD~tNG O~
lJl!Nl[J)IERGRO!UJ[I\ij[C) OAL M~NES
by
Nicolaas Lessing van Zyl
Thesis submitted in fulfilment
of the requirements for the degree
Magister Scientiae
in the Faculty of Natural and Agricultural Sciences
(Institute for Groundwater Studies)
at the
University of the Free State.
Supervisor: Dr. Danie Vermeulen
November 2011
DECLARATION
November 2011
I, Nicolaas Lessing van Zyl, declare that the dissertation herby submitted by me for the
Masters of Science degree at the University of the Free State, is my own independent work
and has not previously been submitted by me at another University/Faculty. I further cede
copyright of the thesis in favor of the University of the Free State.
Nico van Zyl
The influence of flooding on underground coal mines Page ii
ACKNOWLEDGEMENTS
The contributions and support of the following persons and institutions towards this
investigation and report are gratefully appreciated and acknowledged:
• First of all I would like to thank our Lord Jesus Christ for this opportunity and for
being my Hope and Salvation always.
• My supervisor, Dr. Danie Vermeulen, for his guidance and motivation throughout the
project and also for providing me projects to complete the dissertation.
• Eelco Lukas for the computer software and programs that I used in this project.
• My family for their support and understanding.
• My wife, Riette van Zyl, whom I love, for her motivation.
The influence of flooding on underground coal mines Page iii
Table of Contents
DECLARATION ii
ACKNOWLEDGEMENTS iii
List of figures viii
List of tables xiii
CHAPTER 1 1
INTRODUCTION 1
1.1 Introduction to the study 1
1.2 The History of coal mining in South Africa 3
1.3 Background to the research 3
1.4 Geographic information system (GIS) .4
1.5 The Scope of the investigation .4
1.6 Previous mine flooding case studies 5
1.7 Thesis layout 5
CHAPTER 2 7
DEWATERING 7
2.1 Introduction 7
2.2 Coal mining 8
2.3 The Phase approach in mine dewatering 11
2.3.1 Different methods of removal 11
2.3. 1. 1 Simple drainage 13
2.3. 1.2 Boreholes and well points 14
CHAPTER 3 24
USUTU COLLIERY, ERMELO: CASE STUDY 24
3.1 Background 24
3.2 Location 26
3.3 Topography and drainage 29
3.4 Land use 32
3.5 Rainfall and c1imate 32
3.6 Recharge using Chloride method 33
3.7 Hydrogeology 35
3.8 Geology 39
The influence of flooding on underground coal mines Pageiv
3.8.1 Vryheid Formation 41
3.8.2 Quality of the coal 42
3.9 Hydrocensus 42
3.10 Boreholes under investigation .45
3.10.1 Usutu A 45
3.10.2 Usutu B 45
3.10.3 Usutu D 45
3.10.4 Usutu E 45
3.10.5 Usutu F 45
3.10.6 Usutu 1 45
3.11 Water levels 46
3.12 Mining methods 48
3.12.1 Bord-and-pillar extraction .48
3.13 Mine layout 48
3.13.1 Roof thicknesses 49
3.13.2 Floor contours 51
3.14 Ventilation seals 53
3.15 Faults 54
3.16 High extraction zones 55
3.17 Water balance 58
3.18 Mine Interflow 67
3.18.1 Mooiplaats 67
3.18.2 Vunene Mining 69
3.19 Quality of the water 70
3.19.1 Regional waterquality 78
3.20 Conclusions and recommendations 80
CHAPTER 4 82
KILBARCHAN MINE, NEWCASTLE: CASE STUDY 82
4.1 Introduction 82
4.2 Locality 83
4.3 Topography 85
4.4 Land use 85
4.5 Rainfall and climate 86
4.6 Mine details 86
4.7 Geology 90
4.8 General Hydrogeology 91
The influence of flooding on underground coal mines Page v
4.9 Recharge 92
4.10 Monitoring 93
4.11 Borehole information 95
4.11.1 Detail of boreholes inside the mine 96
4.11.2 Details of surface water samples 96
4.12 Water levels 96
4.13 Mining methods 101
4.14 Water Balance 102
4.14.1 Pumping 102
4.14.2 Water quality 107
4.14.3 Recharge 109
4.15 Chemical analyses 111
4.15.1 Chemistry of the boreholes 111
4.15.3 Opencast sampling points (Decant) 118
4.15.2 Chemistry of the surface water 121
4.16 Conclusions and recommendations 127
CHAPTER 5 129
CONCEPTUAL COMPARISON 129
5.1 Introduction 129
5.2 Mining area 130
5.3 Depth of mining 130
5.4 Water levels 131
5.5 Rainfall 131
5.6 Recharge 132
5.7 Pumping 132
5.8 Decant. 133
5.9 Sampling 133
5.10 Residence time 133
5.11 Water quality 134
5.12 Recommendations 135
5.13 Comparative flow diagram 137
BIBLIOGRAPHY 138
APPENDIX A: Mine water classification 143
APPENDIX B: Water standards 145
APPENDIX C: Example of maps and subsidence 146
APPENDIX D: Water qualities 148
The influence of flooding on underground coal mines Page vi
APPENDIX E: Cl values of boreholes 151
APPENDIX F: Pictures of the Usutu boreholes 152
APPENDIX G: Pictures of the Kilbarchan boreholes 155
APPENDIX H: Pictures of the opencast boreholes at Kilbarchan 161
The influence of flooding on underground coal mines Page vii
List of figures
Figure 1: Klipspruit mine pit, where water is seeping into the pit. 8
Figure 2 : Plates 1, 2, 3 and 4 area of sump pumping from Grootgeluk coal mine in the
North- west of South Africa 9
Figure 3: Effects of dewatering around a pit 10
Figure 4: The different methods in mine dewatering summarized 12
Figure 5: Simple drainage from excavation 14
Figure 6: Groundwater lowering through wells 16
Figure 7: Coal mine dewatering by surrounding boreholes 16
Figure 8: Design of well point system to dewater upper sediments 17
Figure 9: Isometric view of a two stage well-point system to dewater an elongated open pit.
............................................................................................................................................ 17
Figure 10: Plan view of a room and pillar mine showing seepages 18
Figure 11: Site dewatering, using a two-stage dewatering system with wells to lower the
water table beneath the excavation 20
Figure 12: Types of open cast mining in advanced dewatering. From Clarke 1995 with
permission of lEA Coal research 21
Figure 13: Channel dewatering 23
Figure 14: Map of the Mpumalanga coal fields focusing on Ermelo 25
Figure 15: Location of Usutu colliery in South Africa 27
Figure 16: The location of Usutu colliery in Google earth image 28
Figure 17 : Location of the coal seams at Usutu colliery near Camden power station 28
Figure 18: Surface contours of the mine area 29
Figure 19: 3D visualization of the surface contours with projected underground and
opencast. 30
Figure 20: Rivers and dams in the area of the mine 31
The influence of flooding on underground coal mines Page viii
Figure 21: Regional surface contours of the area around the mine with rivers and dams 31
Figure 22: Rainfall graph for Ermelo 33
Figure 23: Recharge according to Vegter (1995) 34
Figure 24: Porosity and bulk density variations in shales of the Karoo Basin 38
Figure 25: Source areas for the southern and western Ecca Formations (B) and the northern
Pietermaritzburg, Vryheid and Volksrust Formations (C). Depositional environment of the
Ecca Group in the southern Karoo .41
Figure 26: Location of the boreholes in the hydrocensus for the investigation 43
Figure 27: Water level time graph of a few boreholes measured since 2009 46
Figure 28: Proportional distribution of the water levels last measured 47
Figure 29: Layout of the B- and C-seams at Usutu colliery, together with Vunene opencast.
............................................................................................................................................ 49
Figure 30: Roof thickness of the C-seam (northern mine) displayed in mamsl. 50
Figure 31: Roof thickness of the B-seam (southern Mine) displayed in mamsl 50
Figure 32: The combined floor contours of both the Band C-seam (illustrated in mamsi) 51
Figure 33: Cross section of the floor contours together with the surface topography 51
Figure 34: The combined roof contours of both the Band C-seam (illustrated in mamsi) 52
Figure 35: Cross section is shown of the roof contours 52
Figure 36: 3D combined surface and seam elevation with boreholes into the underground. 53
Figure 37: Layout of the Band C-seams illustrating the ventilation walls, as well as the
position of the adjacent Mooiplaats Mine 54
Figure 38: Faults and dykes identified inside the mine 55
Figure 39: Different methods of mining 57
Figure 40: High extraction (stooping) example on an old mine map 57
Figure 41: High extraction areas identified in the mine 58
Figure 42: Position of the Vunene opencast pits in relation to the C-seam 60
Figure 43: Water bodies in the C-seam 61
Figure 44: Stage curve for the C-seam 62
The influence of flooding on underground coal mines Page ix
Figure 45: Water bodies in the B-seam 63
Figure 46: Stage curve for the B-seam 63
Figure 47: Photograph of decanting borehole 64
Figure 48: Position of the decant borehole in the far south of the B-seam 65
Figure 49: Water level elevations (mamsi) of Usutu A and B (with the blue cross-section line
intersecting the two boreholes ) 66
Figure 50: Cross-section of boreholes Usutu A and B 66
Figure 51: Planned future mining and mining adjacent to Usutu Colliery 67
Figure 52: Photograph of the extraction wells pumping water from Usutu South to Mooiplaats
Colliery 68
Figure 53: EC profiling of the pumping borehole 68
Figure 54: Time series of the EC quality over time of water being pumped to Mooiplaats
Colliery 69
Figure 55: Expanded Durov diagram of the Usutu boreholes 74
Figure 56: EC profiling for Usutu A. 75
Figure 57: EC profiling for Usutu F 75
Figure 58: Stiff diagrams of the mine boreholes 76
Figure 59: Time graph of the EC for the mine boreholes 77
Figure 60: Time graph of sulphate, calcium and magnesium of the mine boreholes 77
Figure 61: Time graph of sodium and chloride of the mine boreholes 78
Figure 62: EC of all the boreholes measured during the hydrocensus in 2011 79
Figure 63: Stiff diagrams of the top mine samples and the farm samples 79
Figure 64: Picture of the Kwazulu-Natal area 83
Figure 65: Location of Kilbarchan Mine in South Africa 84
Figure 66: The location of Kilbarchan mine in Google earth image (Source: Google earth). 84
Figure 67: Location of Kilbarchan colliery in the investigation 85
Figure 68: Rainfall graph for Newcastle 86
Figure 69: Extraction percentages are 50% at a height of 3, 5 m 87
The influence of flooding on underground coal mines Page x
Figure 70: Layout of Kilbarchan together with the opencast areas 88
Figure 71: Kilbarchan coal floor contours (mamsi) 89
Figure 72: Depth of mining at Kilbarchan 90
Figure 73: Position of the mine boreholes at Kilbarchan 94
Figure 74: Surface streams monitoring positions at Kilbarchan colliery 95
Figure 75: Water level time graph of the three boreholes measured since 2003 97
Figure 76: Water level elevation time graph of the three boreholes measured since 2003 98
Figure 77: Cross section of boreholes VOID BH and KW1-99 (with the red line intersecting
the two boreholes 99
Figure 78: Cross-section of boreholes VOID BH and KW1-99 100
Figure 79: Cross-section of boreholes VOID BH and KW1-99 (with the red line intersecting
the two bore holes ) 100
Figure 80: Cross-section of boreholes KW1-98 and KW1-99 101
Figure 81: High extraction areas together with fly-ash identified in the mine 102
Figure 82: Pumping from Kilbarchan to Roy Point through transfer pipelines 103
Figure 83: Picture of the mobile pump 104
Figure 84: Position of 01, where decant is taking place 105
Figure 85: Picture of the place where the water is decanting in opencast 1B 106
Figure 86: Picture of the V-notch of water decanting out of opencast 1B 107
Figure 87 : Time graph of the EC and Sulphate over time at 01 108
Figure 88 : Time graph of the Na, Mg and Ca over time 108
Figure 89: EC profiling for KW1/98 111
Figure 90: EC profiling for KW1/99 112
Figure 91: EC profiling for VOID BH 112
Figure 92: Expanded Durov diagram of the Kilbarchan boreholes 114
Figure 93: Stiff diagrams of the mine boreholes 115
Figure 94: Time graph of the EC for the mine boreholes 116
Figure 95: Time graph of sulphate, calcium and magnesium of the mine boreholes 116
The influence of flooding on underground coal mines Page xi
Figure 96: Time graph of sodium and chloride of the mine boreholes 117
Figure 97: Time graph of the pH of the mine boreholes 117
Figure 98: Opencast sampling points 118
Figure 99: EC profiling for BH 30 119
Figure 100: EC profiling for BH 26 119
Figure 101: Time graph of EC, sulphate calcium and magnesium for the decant point and the
transfer lines 120
Figure 102: Stiff diagrams of the decant and transfer lines sampling points 121
Figure 103: Photograph of KMH3 122
Figure 104: Position of the samples of the surface water 122
Figure 105: Proportional distribution of EC values of the surface samples and discard
boreholes 124
Figure 106: Stiff diagrams of the surface water and the discard dump 125
Figure 107: Time graph of the EC for the surface samples 126
Figure 108: Time graph of the EC for the discard dump boreholes 126
Figure 109: Comparative flow diagram 137
The influence of flooding on underground coal mines Page xii
List of tables
Table 1: Comparison of different dewatering methods 13
Table 2: Water recharges characteristics for opencast mining 35
Table 3: Chloride method for recharge 35
Table 4: Information of the boreholes received during the hydrocensus .44
Table 5: The sample depth of the boreholes .45
Table 6: Anticipated recharge to bord-and-pillar mining in the Mpumalanga area 59
Table 7: Properties of the seams 60
Table 8: EC quality over time of water being pumped to Mooiplaats Colliery 69
Table 10: Chemistry of the boreholes 72
Table 11: Results of the chemical analyses for the boreholes sampled during the
hydrocensus February 2011 73
Table 12: Chloride method for recharge 93
Table 13: Information of the boreholes during Apri12011 95
Table 14: Sample depth and depth of boreholes in the mine 96
Table 15: Different areas of the Kilbarchan mine 101
Table 16: Transfer pipelines volume of water released per annum 103
Table 17: EC of the transfer pipelines over time 103
Table 18: Volume water pumped by mobile pump 104
Table 19: Approximate decant volumes of the opencast areas 106
Table 20: Rainfall for the Kilbarchan underground area (excluding stooping and ash filling) .
.......................................................................................................................................... 109
Table 21: Rainfall for the stooped areas at Kilbarchan 109
Table 22: Rainfall for the opencast areas 109
Table 23: Recharge table for Total Kilbarchan 109
The influence of flooding on underground coal mines Page xiii
Table 24: Different scenarios regarding recharge percentages 110
Table 25: Results of the chemical analyses for the boreholes sampled during the last site
visit April 2011 113
Table 26: Water quality of the decant point and the transfer lines 120
Table 27: Results of the chemical analyses of the surface water sampled during the last site
visit April 2011 123
The influence of flooding on underground coal mines Page xiv
CHAPTER 1
INTRODUCTION
1.1 Introduction to the study
Predicting flooding in coal mines is highly uncertain. The prediction of mine water quality
and mine water stratification is even more challenging. Closed or orphaned mine sites
cover approximately 240 000 km2 of the earth's surface and can be a hazard to both
humans and the environment according to Wolkersdorfer (2008).
Balkau (1999) states that among the outstanding environmental problems confronting the
mining industry, that of abandoned mine sites has been practically slow to tackle. Such
publications are categorized as "unused", "closed", "abandoned", or "orphaned" mines.
Mining is no longer taking place, but someone has to take care of the legacy.
Acidic water discharging from underground mines in sulphur-rich coal deposits is usually
acidic, and may be distinguished into two classes (Chen and Soulsby, 1999)
• Above surface drainage
• Below-surface drainage elevation
Above-drainage mines often remain largely dry after closure and may continue to
discharge acidic water for long periods according to Chen and Soulsby (1999). In
contrast, below-drainage mines tend to partially or completely fill with water (flood) to high
levels, according to Burke and Younger (2000).
Such flooded mines tend to have restricted oxygen infiltration in comparison with free-
draining above-drainage mines. This fact has been long recognized to influence the
chemistry of mine discharge, motivating early efforts to reduce acid discharges from
mines by bulk heading and flooding mines and reducing the oxygen supply, according to
Donovan, Leavitt & Werner (1978). Some of these 'flooding' efforts have reported some
long-term success. In a study done by Stoertz, Hughes, Wanner and Farley (2001) a
change inferred in pH from 2.7 to 5.3 and the conductivity changed from 2 700 to 600
usiernens/crn for the discharging Ohio coalmine over a 20-year timeframe.
It is well known that flooding improve water chemistry, but the precise timeframe and
controlling factors, such as acid-neutralizing capacity, by which such changes might occur
in specific mines, have yet to be identified. An understanding of the details by which
The influence of flooding on underground coal mines Page 1
changes in chemistry would occur following mine flooding, would be very relevant to the
long-term control of mine-water acidity (Younger, 2000).
Mines fill up with water after closure. As a result, hydraulic gradients develop between
them and different hydraulic water pressures are exerted onto peripheral areas or
compartments within mines. This results in water flow between mines, or onto the
surface. This flow is referred to as intermine flow. Intermine flow as a concept includes
the quantity and quality of the water (Grobbelaar, et aI, 2004).
Information available in the South African coalmining industry states that mines fill up with
water and decant after closure. This usually occurs within 10 years. At the more isolated
collieries, rebound of the water level may take up to 50 years. Apart from the fact that
mine water is saline, low pH-values may also be encountered (Grobbelaar et al., 2004).
Mine water has historically been pumped from active mine workings to allow unhindered
coal production. Almost no consideration has been given to the best management
strategy for water while mining. Yet, this is simple: Mine from deep areas to shallow areas
and leave water behind in the mined-out workings. This strategy has, for the past few
years, been applied in several of the larger collieries with significant success. The
advantage of this mining sequence does not only lie in managing water volumes, but also
in water quality management. Mined-out areas are flooded, thus excluding oxygen.
Furthermore, the natural alkalinity of water is not flushed from the rock. This counteracts
acidification in these mines (Grobbelaar; et al., 2004).
Flooding of open cut mines can be a very real problem if a mine is located in a valley or in
the path of a stream or a river with a significant upstream catchment. Depending on how
quickly it occurs and how severe it is, flooding can cause a variety of problems such as
loss of life or injury, damage to machinery and infrastructure, and far more likely, loss of
access to the pit due to water and silt and the subsequent loss of production. All of these
scenarios are highly undesirable to mine operators (Bedient, Rifai & NewelI, 1994).
The main reasons for mine flooding according to Wolkersdorfer (2008) are the following:
e The mine is no longer economical.
• All the raw material has been exploited.
• Accident, war, or political reasons.
C) Geotechnical stability of the abandoned mine workings.
Cl Prevention of disulphide oxidation.
Cl Safety reasons.
The influence of flooding on underground coal mines Page2
The Mpumalanga province and the KwaZulu-Natal province are the focus areas in this
thesis, as both case studies are located in these provinces. Both the mines are coalmines
which have been flooded for approximately 20 years.
1.2 The History of coal mining in South Africa
The history of coal mining in South Africa is closely linked with the economic development
of the country. Commercial coal mining started out in the Eastern Cape near Molteno in
the year 1864. The discovery of diamonds in the late 1870s led to expansion of the mines
in order to meet the growing demand for coal. Commercial coal mining in KwaZulu-Natal
and on the Witwatersrand commenced in the late 1880s following the discovery of gold
on the Witwatersrand in 1886. In 1879 coal mining commenced in the Vereeniging area
and in 1895 in the Witbank area to supply both the Kimberly mines and those on the
Witwatersrand. South Africa began a period of major economic development after World
War II. New goldfields were discovered and developed in the Welkom, Klerksdorp and
Evander areas; a local steel industry was established with mills being built at Pretoria,
Newcastle and Vanderbijlpark; an oil-from-coal industry was established, initially at
Sasolburg and later at Secunda; mining of iron, manganese, chromium, vanadium,
platinum and various other commodities commenced and expanded; and power stations
were erected on the coalfields to supply energy to these developing industries and to the
growing urban population in the country. In addition to meeting local needs, coal mining
companies began to develop an export market, making South Africa a major international
supplier of coal (Mccarthy; Pretoruis, 2009).
1.3 Background to the research
In the Ermelo and Newcastle area, a situation occurs where the mines have been flooded
to prevent the water quality from degrading. Managing the mine after closure and
protecting the environment after production has ceased are important considerations.
After this opencast mining was introduced to increase the life of a mine, for example
Vunene opencast mining at Usutu colliery. This study was done to determine the
influence of flooding and the potential chance of the decanting of polluted water, looking if
stratification occurs. A further aim was to compare end examine the quality of the water
over time and to establish whether it was influenced by the quantity of the mine water.
The challenge of underground coal mines is the management of the mine water following
the closure of the mines after they were flooded. Developing a cost-effective and
sustainable mine water management program is of great value. The water qualities
expected and the volumes may be investigated using a program called WACMAN.
Plotting the water quality in Stiff and expanded Durov diagrams may reveal a trend over
The influence of flooding on underground coal mines Page 3
time in the water quality. This research comprises case studies of a shallow- and a deep
underground mine where the mine was flooded after operation stopped.
• The main aim of the proposed work is to determine what influence flooding has
comparing the two case studies.
• The two studies are very different from each other. Usutu mine is located at a flat
area and is deep, while Kilbarchan mine is located on a steep area and is quite
shallow.
• How this may influence the water quality looking at something like residence time.
• The quantity of water and quality of water after mine closure.
• The study aims to find answers to the following questions:
• Flooding: What influences do stooping, flow zones, faults and recharge have on
the effects of flooding?
• Mine water quality: Will flushing take place? Does the water quality deteriorates or
improve or stay the same? What is the quality at different depths? What
influences the residence time?
• Mine water quantity: What is the influx into the mine? What area the pumping
volumes? Will decanting take place?
1.4 Geographic information system (GIS)
Grobbelaar et al (2004) give a good discussion on the GIS used in this study. The WISH
(Windows Interpretation System for Hydro-geologists) software package was selected for
the visualization and interpretation of the data. The reasons for this are as follows:
• WISH is easy to use, sponsored in part of its development by the Water Research
Commission (WRC), and available from the Institute for Groundwater Studies
(IGS).
• It consists of a map drafting and display facility. Maps may be integrated from
other applications that are commonly in use at the collieries.
• Datasets from Microsoft Excel may be superimposed onto the maps. With regard
to relevant data processing, the processing power of WISH is unsurpassed by
other software.
1.5 The Scope of the investigation
The purpose of the study is to investigate the effects of flooding on the groundwater
resources at Usutu colliery, Ermelo and Kilbarchan mine, Newcastle. The activities
carried out to achieve the overall objective of the study include the following:
o Physiographical description and overview of the regional geology, including
the structure of the study area.
The influence of flooding on underground coal mines Page 4
• Summary on dewatering of mines.
• The mine layout and mining.
• Overview of the mining methods used.
• Geology.
• Hydrogeology.
• Data collection, analysis and interpretation.
• Water quality.
• Water quantity.
• Calculation of recharge by various methods.
• Analysis of spatial and temporal variation of groundwater levels and qualities.
• Calculation of the water influx into the mine.
• The effects of high extraction on the groundwater resource.
• Available management strategies.
1.6 Previous mine flooding case studies
Currently there is no case study known where the stratification of a mine was predicted
precisely and where remediation methods based on stratification predictions were
successful. The most comprehensive study of a single mine water stratification so far was
constructed at the Niedersclema/Alberoda/Germany uranium mine done by
Wolkersdorfer, (2008). Between June 1992 and December 1994, a total of 115 depth
profiles in seven underground shafts were measured and the stratification was observed
by Wolkersdorfer, (2008).
The conceptual model for an individual mine or interconnected block is based on a
general model. The model is based on the general development of mining in the UK and
assumes four basic depth controlled mining units, A-D, which mayor may not be
interconnected. Water inflows into all the units can then be put into three basic categories
(Whitworth, 2002).
1.7 Thesis layout
Chapter 1: Introduction.
Chapter 2: The effects of dewatering are discussed in detail as well as methods of
dewatering used in previous studies.
Chapter 3: The Usutu case study is discussed.
Chapter 4: The Kilbarchan case study is discussed.
The influence of flooding on underground coal mines Page 5
Chapter 5: Gives a comparison of the two case studies looking at differences
encountered when considering the influences of flooding on underground mines.
The influence of flooding on underground coal mines Page 6
CHAPTER 2
DEWATERING
2.1 Introduction
Dewatering is an effect of the flooding of a mine. In this chapter dewatering will be
discussed, looking at different aspects of dewatering and their effects.
The three most costly expenses to a mine according to Morton, 2009 that can save mines
millions of capital is:
• Compressed air
• Water control (mine dewatering)
• Labor
According to Morton (2009) dewatering means the removal of water by lowering the water
table from a high-wall or underground mine. The problem in South Africa is that mine
dewatering is only looked at when it becomes a problem for the mine. The low rainfall and
low-yielding aquifers have meant that the control of mine water inflows was covered in
the design of the mine. Designs for different resources of coal, gold and diamonds do not
differ much.
Essential to any operating mine, according to Result and Vermilion (2007), is the
dewatering process whereby water is expelled from the mine. The efficiency and
effectiveness of this process is directly proportional to the operating cost of the mine. It is
vital that the engineers and supervisors overseeing the dewatering process should know
the parameters within which the dewatering system operates.
A functional specification of the dewatering system involves the water mapping of the
entire dewatering process and calculating water flows in and out (water balance) of
pumping stations. Consequently as the mine expands its production, the dewatering
system finds itself in unknown territory (Result and Vermilion, 2007).
In the majority of surface mines, groundwater will generally not be encountered below 50-
150 metres. The amount of groundwater present, the rate at which it will flow through the
rock, the effect it may have on stability and the influence it will have on the economical
development of the pit, depend on many factors (Connelly and Gibson, 1985).
The most important of these factors, according to Brawne (1982), are the topography of
the area, precipitation and temperature variation, the permeability of the rock mass and
The influence of flooding on underground coal mines Page 7
overburdened soil, and the fragmentation and orientation of structural discontinuities in
the rock.
Water influences the mine in many ways, according to Connelly & Gibson (1985):
• Inflows may flood the mine and hold up production.
• Water flowing in affects the cost of drilling new boreholes.
• Slopes becomes unstable
• Equipment is damaged
• The mine has to decide where to go with the water after it has been abstracted
Carter (1992) states that groundwater is by far the greatest natural cause of problems in
civil engineering. Very good ground investigations will do much to stop unexpected
problems.
Figure 1: Klipspruit mine pit, where water is seeping into the pit.
2.2 Coal mining
The coal is usually found in fractured or secondary aquifers. The coal can also
sometimes act as an aquifer. So the hydrogeology can be called a layered system. This
means that the different layers need to be dewatered individually. The structures can be
dewatered by using geophysics to site anomalies and these structures (Morton, 2009).
The influence of flooding on underground coal mines Page 8
Morton (2009) states if water gets in contact with exposed coal not on virgin coal that is
not exposed the quality of the water degrades in pH, TDS and colour. This requires that
the water should be tested and treated before pumping it into streams and lakes. In open
pit mines, the main problem is storm water control. The clean and dirty water needs to be
managed. Figure 2 shows plates 1, 2, 3 and 4 in the area of sump pumping from
Grootgeluk coal mine in the North West of South Africa.
Figure 2 : Plates 1, 2, 3 and 4 area of sump pumping from Grootgeluk coal mine in the
North- west of South Africa (Morton, 2009).
Thomas (2002) states that open pits need to be dewatered for the mine to operate
successfully and functionally. The basic idea is to stop water from flowing into the pit in
order to a maintain slope stability and protect water for abstraction outside the mine area.
The hydrogeology and geology of the mine site plays the most important role in choosing
the dewatering method.
The influence of flooding on underground coal mines Page 9
The standard process is usually that the water level is excavated by the mine. This then
requires that the water table be lowered to avoid flooding. Installing a pump in the sump
at the pit bottom usually solves this problem. However, this method does not always work
for every situation.
According to the Minerals Council of Australia (1997), dewatering is commonly carried out
to lower the water table by pumping water out of the aquifer and away from the mine. A
series of bores or spear points may be positioned in areas of good hydraulic connectivity
to allow pumping at a sufficient rate to draw down the aquifer. Drawdown of the water
table reduces the flow through the area of groundwater near the mine pits. Ideally, the
water table should be drawn down below the floor of the pit so that groundwater inflows
are eliminated altogether. Figure 3 indicates the effect produced by dewatering.
According to Libicki (1993) other methods are used when there are geological
discontinuities (faults, folding, etc).
Future
mine pit
, ., .~ .
Ii'litiál·positi()n of'the'
. .water table
.Long-te.n:n drawdowD. of the
'Watertable Cromdewatering
Figure 3: Effects of dewatering around a pit (Minerals Council of Australia, 1997).
Brawne (1982) states that based on field investigations, a design can be prepared for the
control of groundwater in the slope and in the pit. Methods of control include the use of
horizontal drains, blasted toe drains, the construction of adits or drainage tunnels and
pumping from wells in or outside of the pit. Recent research indicates that subsurface
drainage can be augmented by applying a vacuum or by selective blasting.
Instrumentation should be installed to monitor the groundwater changes created by
The influence of flooding on underground coal mines Page 10
drainage. Typical case histories are described that indicate the approach used to
evaluate groundwater conditions.
Morton (2009) explains that the use of gravity is always the first option when pumping is
done from sumps and collector dams. When pore pressures are high and gravity
drainage is insufficient, then active pumping from specific permeable horizons of
structures is used to supplement the drawdown created by the passive drainage. Where
groundwater flow is predominantly vertical, horizontal drainage is most effective, in for
example, drainage galleries. When the dominant flow is horizontal, flow vertical methods
of dewatering, such as vertical pit perimeter boreholes, are more effective. As the
direction of flow and mine development change with time, the methodology can be
adapted to suit the new conditions.
2.3 The Phase approach in mine dewatering
This approach covers phases to go about tackling a dewatering problem according to
Morton & van Miekerk (1993).
• Phase 1: Desk study and borehole senses
This phase consists of all the information one can get from the mine without incurring any
expenses. Information consists of the regional groundwater level of the area, borehole
information, where water strikes occurred, maps of the mine etc.
The borehole census is there to determine the regional groundwater level of the area.
This can be done by doing a hydrocensus.
• Phase 2: Impact of mining on the groundwater
• Phase 3: How to remove/reduce the hazard.
The hazard may be removed either by handling or diverting the flow, depending on where
the water is coming from. All this can be accomplished through experience using trail
dewatering and computer modelling. Dewatering is the removal of groundwater from an
area through the lowering of the water table.
2.3.1 Different methods of removal
The different methods of mine dewatering are summarized in Figure 4. There are a large
number of dewatering methods to choose from, but one has to find the method that best
suits the situation.
• Well points
• Deep boreholes
• Dewatering galleries
The influence of flooding on underground coal mines Page 11
• Drains
• Sump pumping
• A combination of some of the above.
Pol t . zometers
o measure pore pressu e
Vertical pumped Tension era
borehOlê Oraln or trench{2m)
Dverslolb'l m
3" F U____..r
lo
Sump
Zone 01 retaxation
Figure 4: The different methods in mine dewatering summarized (Morton, 2009).
Table 1 shows a comparison between a few dewatering methods constructed by
Hustrulid (2000), giving the advantages, disadvantages and the best application for the
'method.
The influence of flooding on underground coal mines Page 12
Table 1: Comparison of different dewatering methods (Hustrulid, 2000).
Method Advantages Disadvantages Best application
Usually have a long life of mine; Impacts on centre of pit may Where flow to pit is lateral
have large space for drilling; can be be limited; must usually be for small pit; where
Perimeter wells installed proir to mining; can deeper ( hence higher cost) ; hydraulic conductivity is
intercept lateral inflow; logistiacally might be in less permeable primarily horizontal
simple rock
Creates most drawdown in pit; Hard to mine aroun; short life; Compartimentilized rock
located in zone of potensial large drilling logistics; need to mass; very asymmetric
hydraul ic conductivity (often the deliver water and power to and pits; mine in which large
In-pit wells core zone); relatively shallow; can from well; potential drilling penmanent benches are
intercept vertical flow through problems; can be installed established early
bottom of pit only after mining commences
Increase slope stability; can drain/ Winter freezing; Deep pits; permeability
In-pit depressurize through targeted drains/depressurize only rock masses; higly
horizontal/drain structures; passive dewatering; no limited area; can only be anisotropic rock masses;
holes specail location needed; installed after mining begins; to breach groundwater
inexpensive water delivery "dams"(e.g. gouge zones)
Dewaters from below the mine; can High cost of excavation; large Long mine life; good
intercept structures at optimum lead time for construction tunnelling conditions;
angles; can handle large qauntiities where construction can be
Drainage galleries of water done for dual purpoese
(e.g. exploration or high
grading)
Carter (1992) states that the excavation of the water level in dewatering falls into two
categories:
• Pumping methods
• Exclusion methods
Pumping methods include:
• Simple pumping
• Well pointing
• Specialized well pointing
• Filter wells
• Electro-osmosis
These methods are best used for sands. Gravels are generally too permeable and clays
are just ignored.
2.3.1.1 Simple drainage
Carter (1992) describes surface water inflows that are intercepted by surface drainage
ditches as shown in Figure 5. This is a very simple method.
The influence of flooding on underground coal mines Page 13
Ditch to intercept
surface water
-----_ ...Original water level , ...
Drawdown ,. "
,," Sand
Garland Clay
drain pipe
surrounded
by gravel
Side support
as necessary Base of excavation
Sump pumping at
lowest point of
excavation
Figure 5: Simple drainage from excavation (Carter, 1992).
2.3.1.2 Boreholes and well points.
Morton et aI., (1993) explain that interference pumping through boreholes causes a
deeper cone of depression between the boreholes. The more boreholes there are the
more effective the dewatering will be. Example: Letlhakane diamond mine, Northern
Botswana.
Well points are used in unconsolidated sediments. This method dewaters shallow
aquifers; it is usually used in combination with deep boreholes to dewater multi layer
aquifers. Example: Vaal River Channel, Northern Free State.
According to Libicki (1993) drainage wells between 20 and 400 metres drilled from the
ground surface. The depth of the boreholes varies from 20 m- 400 m, with drilling
diameters from 350 to 1200 mm. The yield varies from 1.6 I/s to 166 I/s in extreme cases,
but in average cases the yield varies between 3.3 I/s - 25 I/s. These boreholes are sited
in the area where the water flows into the mine. This is done to intercept the water before
it flows into the pit. The boreholes are usually fitted with submersible pumps (Brawne,
1982). Where very heavy seepage is expected, pumping from deep wells located around
the periphery of the pit may prove practical and economical. Facilities of this type have
been installed in excess of 125 metres with success. Where the groundwater flow is large
and when the influence on stability of pore water pressures and seepage pressures is
significant, the pumping system must be designed with reasonable over capacity. If one
pumping unit becomes inoperative, there is sufficient excess of pumping capacity to
The influence of flooding on underground coal mines Page 14
prevent the development of local areas of high water pressure which might cause
instability.
In addition to drainage control within the pit itself, the control of surface drainage outside
the pit boundaries is necessary to ensure that surface water does not flow into the pit.
Besides the extra pumping capacity required, water flowing into the pit percolates into
surface fractures and openings. This aggravates rock falls and the occurrence of local
slides between benches. It is only desirable not only to determine the influence of
groundwater on stability but also to determine whether drainage of the pit slopes will
allow an increase in the overall slope angle.
For the same safety factor, reducing pore pressures by 6 to 10 metres will usually allow
an increase in slope angle approximating 3 to 6 degrees. An evaluation can be made of
the cost of drainage versus the economical benefit to be gained by the increase of the
slope angle that the drainage will allow. In order to evaluate the effectiveness of drainage,
it is necessary to install piezometers at key points in and around the pit to measure cleft
water pressures. It will normally be adequate to read instrumentation on a monthly basis,
with more frequent readings during the spring runoff period, following heavy rains and
during the late winter period.
As the open pit deepens the probability of high pore water pressures developing in the
base area of the pit increases. These pressures could conceivably become sufficiently
large to cause a blow up of the base of the pit. This probability increases where the
bedrock structure is horizontal or where significant horizontal tectonic stress exists in the
rock. To reduce the water pressures in the base of the pit, pressure relief wells should be
considered. The design of drainage control in open pit mines should always be preceded
by a moderately detailed field permeability testing program me, unless extensive previous
experience at the site is available (Brawne, 1982).
Morton (2009) explains that in some areas the sediments in the area overlying the source
have high clay content and depressurization is required through well points. Figure 8 and
Figure 9 show a cross-section of a Well-point dewatering system, and an isometric view
of a multi-layered system.
Well points are used in combination with good storm water control and sump pumping.
Mining up gradient is the best, where possible at a slope of >1.5 degrees to let the water
drain down the walls, towards pumps or a decline. Figure 10 shows a plan view of a room
and pillar mine. In this example the main haulages are used to drain water towards the
shaft but there are also areas of ponding (shown in blue) that are undrained. The mine
The influence of flooding on underground coal mines Page 15
also has an area of subsidence where there is stream flow capture, which is then drained
to the main haulage. This is a problem when the water in the mine becomes dirty.
Carter (1992) shows in Figure 6 the lowering of the original water level under the
excavation. This method is used in shallow excavations and is relatively cheap. The
chemistry of the water needs to be known to prevent clogging of the equipment.
Water
Well or w~IIRojn!
pumped
away Original groundwater revel
'-" -- ....... - --- -' ...._----/'-
" I I Excavation /
" t , I I /1
\ :: :: / Sand
\ I I I I I
'II _---- .._. _ II1
"1I v/ ......\... .........'~..(I II
LJ Lowered groundwaterr level lJ
Figure 6: Groundwater lowering through wells (Carter, 1992).
Boreholes can be used to dewater a mine. By drilling a borehole every 25 m; each
pumping 0.1 I/sec; 100 m from mine shown in Figure 7.
II
II
II
II 1111 IIII II
II
II II
Figure 7: Coal mine dewatering by surrounding boreholes (source: UFS mine water
course, 2008).
The influence of flooding on underground coal mines Page 16
VACUUM PUMP
DIRECTION OF
GROUND WATER
FLOW
"'
Figure 8: Design of well point system to dewater upper sediments (Morton, 2009).
c
.
.I
11
] 'c
uring 'lag mpmg
Figure 9: Isometric view of a two stage well-point system to dewater an elongated open
pit (Morton, 2009).
The influence of flooding on underground coal mines Page 17
TO SHAFT N
A
D.wUnmmed coalter
Drainage along
side drains
'00
Figure 10: Plan view of a room and pillar mine showing seepages (Morton, 2009).
2.3.1.3 Electro-osmosis
According to Carter (1992) electro-osmosis is a rarely used method. It is occasionally
used to dewater very fine grain soils such as silts. An electric current is induced through
the soil which causes water to move from the positive anode to a negative cathode.
• Anodes are metal stakes driven into the ground
• Cathodes consists of well points
2.3.1.4 Horizontal drains
Brawne (1982) describes a technique which may be utilized to improve the stability of the
rock slopes, namely to install horizontal drains, a technique which is commonly used to
stabilize earth landslide. Holes 5 to 8 cm in diameter are drilled near the toe of the slope
on about a 5 per cent grade for a distance of 50 to 100 metres into the slope. If the holes
cave, a perforated drain must be installed. To reduce drilling time it is common to fan 3 to
5 holes from one drill location.
Groundwater flows into the drain holes, lowers the groundwater level and improves
stability. During the cold winter weather in northern climates it may be necessary to
protect the outlets of the drains from freezing and to collect the water with a frost free
collector system. In the winter months in northern climates, it is common for the pit slopes
to freeze so that seepage does not exit from the slopes. As a result high pore water
The influence of flooding on underground coal mines Page 18
pressures frequently develop. This factor appears to account for the fact that many
failures occur in the February to April period in Canada (Brawne, 1982).
An alternative to horizontal drainage is to minimize the buildup of pore water pressures in
the slopes to blast the entire lower bench 10 to 13 metres wide around the toe of the
slope in the open pit and not to - excavate this blasted toe during the winter months. This
area will have high permeability and will act as a large drain in allowing water to seep
from the slope. Water from this area must be collected in one or more sump areas and
pumped from the pit (Brawne, 1982).
Horizontal drains can be a very effective method if used as a depressurizing method,
together with the other system set in place to dewater. These drains are drilled in the
benches of a mine pit. Drains are set to intercept the anticipated inflow of water or where
water is already flowing into the mine. The length of these drains is 150 m with a diameter
of 100 mm. When working in sandy layers, a PVC pipe can be used to filter (Libicki,
1993).
2.3.1.5 Needle filters
These drains are 50 mm pipes 10 m length in the soil to dewater, a group of 20-30 pieces
2-4 m apart. All these connected to one pump allow one to lower the water level by 6-7
m. These are extra if it is necessary to dewater more (along the roads). The ditches
inside the pit only take up the rainwater and water from the slopes. These ditches are dug
on the slopes to provide a sort of permanent structure. The water is then fed to pumping
stations that can be moved. This method is used with great effect in Poland (Libicki,
1993).
2.3.1.6 Interconnected wells
Interconnected wells are used to dewater the open pit under the excavation. Figure 11
shows a two-stage dewatering scheme to lower the water level below the two levels of
excavation (Price, 1996).
The influence of flooding on underground coal mines Page 19
SUC1ion
Well - pipe
points __ _' Natural
LO'll'ercd ,""- grO\Jn(; level
watsr table
.-.':
I
Well painis
Base ol
excavatlon
S:rtu ratert
lone
Figure 11: Site dewatering, using a two-stage dewatering system with wells to lower the
water table beneath the excavation (Michael Price, 1996),
2.3.1.7 The gravity wells
Gravity wells drain water from an upper aquifer to a lower aquifer below the level of the
pit bottom, The rate of pumping is maintained steadily to form a cone of depression; this
level is monitored by taking water levels, The cone of depression will influence the areas
outside the mine site, Figure 12 shows the dewatering of open pit mines using several
methods of excavation, The wells are installed through the overburden, coals and footwall
sequence (Clarke, 1995),
The influence of flooding on underground coal mines Page 20
(al 80,,-< ...t ~nd 'eWC41 (up-dlp) minIng
(b) ShaJIqw dowo-dilp mln'-'u
~c) Open pil mln."'Q
d~octiol'lof working -
(dl Up-dip mining, pumps in$l4fled In btlCkllned bOI-<:1Il area
_'T_ drawdoWl1lvater level
- - - final h~hwall
Figure 12: Types of open cast mining in advanced dewatering. From Clarke 1995 with
permission of lEA Coal research.
2.3.1.8 Cuf-offwells
This is one of the best methods in open pit protection against water that flows into the
mine, especially overburden. There are different types of cut-off wells. The easiest one to
make is the dug one. This is a ditch dug by a special excavator of 0.4 - 0.7 m. The cut-off
wells are leak-proof, because of a special sealing substance. One disadvantage of this
method is that the range of depth is only 70 m (Libicki, 1993)
Grouting is another method. Boreholes are drilled and special sealing substances are
injected into the borehole. The advantage is that the depth of investigation can be at 300
The influence of flooding on underground coal mines Page 21
m. A disadvantage is that it is extremely complicated. Cut-off wells are best used in areas
with high permeability and where the surface water is recharging the aquifer (rivers or
lakes). Another advantage of this method is that it does not involve a cone of depression.
From an environmental point of view this is ideal, because streams will not be affected
and shallow wells are safe (Libicki, 1993).
2.3.1.9 Underground galleries
Together with gravity flow, filters are also used for drainage. This method is effective in
disturbed aquifers and shallow low yielding wells. The reason is that the sediment in the
water does not affect them as much as it affects submersible pumps. This method was
used in the 1950s and 1960s. It is much scarcer today and is used less and less. The
method is mostly used in old mine operations, because of the cost of labour (Libicki,
1993)
2.3.1.10 Pumping stations
These pumping stations are equipped with pumps and sump pumps. They are placed at
the lowest point in an open pit to pump out the water from inflows and rainfall. Other
actions include the sealing of river beds so that the surface water does not come into
contact with the cone of depression (Libicki, 1993).
2.3.1.11 Drainage adits
Brawne (1982) show that in some instances it may be practicable to construct an adit
under the ore body and use it as a drainage gallery from which water is pumped or
drained by gravity. For large volumes of water or for deep pits, drainage galleries at more
than one elevation may be required. To increase the effectiveness of the drainage
gallery, drill holes can be drilled in a fan pattem outward from the adit to increase the
effective drainage diameter. Drainage adits have been used at Marcopper and Atlas in
the Philippines, Similkamene Mining in Canada, Anamax Twin Buttes in the U.S.A. and
the Deye Mine in China. It is recommended that the drains or adit be placed under a
partial or complete vacuum.
Recent research at Gibralter Mine, Canada, showed a dramatic reduction in pore water
pressure when the vacuum was applied. Drainage galleries may be particularly adaptable
where open pits are located on steep mountain side slopes so that the edit may be
drained by gravity.
2.3.1.12 Channel dewatering
The Minerals Council of Australia (1997) found that groundwater may also be intercepted
outside the pit if the topography, groundwater regime and mine plan allow this. A channel
may be constructed to lower the water table and drain the water to downstream
The influence of flooding on underground coal mines Page 22
catchments. However, lowering of the water table in this manner is generally less
effective because of the reliance on steady gravity drainage. Figure 13 shows the method
of channel dewatering. When groundwater flows are not highly significant, the water is
often intercepted in the pit, collected in a sump and pumped to a retention dam for
treatment or storage as required.
Each method of managing groundwater inflows will have different environmental impacts.
These will need to be evaluated prior to implementing a control technology. Issues such
as volume of flows, water quality and the effect on other users of the groundwater,
surface drainage systems and receiving water bodies should be addressed.
Minor
seepage into Channel
pit
Mine Pit
Mine Pit
Seepage into pit
caused by channel dewatering
Figure 13: Channel dewatering (Minerals Council of Australia, 1997).
The influence of flooding on underground coal mines Page 23
CHAPTER 3
USUTU COLLIERY, ERMELO: CASE STUDY
3.1 Background
Usutu Colliery is the first case study that will be investigated with regard to the influence
of flooding on a mine where the mine has been flooded for 20 years. Chanzo Investment
Holdings (Pty) is the current owners of Usutu Colliery. Vunene Mining currently operated
an opencast mine above the old Usutu Colliery, and also plans to mine underground in
future. To come to a conclusion the mining depths of the coal seams must be observed
underneath the opencast
Opencast areas have been mined at points where the underground seams are in danger
of being mined into. This could have an influence on the recharge in the underground
areas of the mine, thus influencing the quantity of the mine water. It is therefore
important for Eskom, the owners of Usutu, to understand the geohydrology of the mine
for future liabilities during mine closure.
Usutu is applying for partial closure (Usutu East and South), and also needs to know
what the current groundwater status in both the north and south mines are. The coal mine
has been flooded (recharged) with water since production was stopped between 1987
and 1990.
In 2002 there were ten operating collieries in the Ermelo coalfield, most of which were
small to medium-sized. Mining in this coalfield has been dormant for some time with most
mines closed with reserves. Of the total saleable production of 222 551 Mt in 2001, the
Ermelo coalfield contributed about 7.2 million tons. Most of the high-grade steam coal
produced by Xstrata Coal SA in the Ermelo Coalfield is destined for export (Jeffrey, 2005).
Figure 14 shows a map of the coal fields in Mpumalanga.
The influence of flooding on underground coal mines Page 24
-2850000
-2880000
-2910000
-2940000
-2970000
-30000 o 30000 60000 90000 120000 150000
Figure 14: Map of the Mpumalanga coal fields focusing on Ermelo (source: UFS mine
water course, 2008).
In the past, the now closed Ermelo mines and Usutu colliery supplied Eskom's Camden
power station, with defunct Majuba colliery supplying the Majuba power station. Camden
was brought back on-stream by the end of 2004 and managed by a black empowerment
consortium operating Golang colliery, incorporating Golfview colliery and the former
Usutu Colliery (Jeffrey, 2005).
The following methodologies were used to sample the boreholes:
• Measuring water levels
An electronic dip meter was used for this operation to determine the depth of the water
level below the collar of the borehole. It is important always to measure this from the
collar of the casing, thus ensuring uniform measurement.
• Water sampling
Sampling included either a sophisticated pressurized depth sample or a flow-through
bailer (depending on the depth of the sample).The bailer was cleaned with de-ionised
The influence of flooding on underground coal mines Page 25
water before each sample was taken. The samples were stored in 500 ml plastic bottles
and transported to the IGS laboratory for analysis.
It is essential that samples should always be taken at exactly the same depth in order to
obtain a uniform true estimation of the water quality.
Chemical analysis for macro- and micro-parameters as specified by the contract
was performed by the laboratory at the IGS.
• Inorganic parameters:
pH, EC, Ca, Mg, Na, K, p-Alk, m-Alk, Cl, S04, N03 and P04. Si, AI, Fe, Mn and B
E.C profiling was also performed on a number of mine boreholes to determine
whether any stratification occurs.
3.2 location
Usutu Colliery is situated 8 km outside the town of Ermelo in Mpumalanga, close to
Camden power station on the N2 road to Piet Retief. Figure 15 shows the location of the
mine by indicating the coal seams of the mine. Above the coal seams normal grasslands
exist. Maize, cattle, potatoes, beans, wool, pigs, sunflower seeds, lucerne and sorghum
are the main farming produce of this area. Anthracite, coal and torbanite mining is
practised here.
The influence of flooding on underground coal mines Page 26
Ohngs ad. Sabl Sand a ure ReserveLimpopo
Pi . m's Rest •• Graskop
Hazyview.
Lydenburg. Sabie.
Groblersdal V\iI"'ule R•Iver
Dulls KMIA Airport +•lroom N4
NH Schoeman
•s oof 'elspruil
Belfast. _. Waterval Boven Barberten
Middelburg • Machadodorp •VlMbank
•Badplaas
•Carolina
~N12 LochielMpumalanga •
•Chris sic smeer Swaziland
•BelhalSecunda • •Ermelo• Usutu colliery
N2
N3 ..Standerton Piel Retief
Free State KwaZulu-Natal @ www.places.co.za
Figure 15: Location of Usutu colliery in South Africa (www.places.co.za).
The Google image in Figure 16 shows the coal seams with the roads superimposed on
them. There is still farming activity taking place around the closed mine. The map in
Figure 17 indicates the mine in relation to the power station. The direction of the town of
Ermelo is shown with an arrow. The location of a mine that has been flooded can have n
massive effect on flooding. Different locations mean different rainfalls, topography etc.
The influence of flooding on underground coal mines Page 27
-293500
-294000
-294500
Roads
-295000 Outline a-Seam
Outline C-S •• m
C-L •• g-L •• r
D Googl. Earth
6000
-100000 -95000 -90000 -85000 -80000 -75000
Figure 16: The location of Usutu colliery in Google earth image (Googie earth).
-293500
-294000
-294500
~LEGEND
Outline B-St.am
Outline C-Seoilm
C.L.... g-L.I.'
- i/ 3000 i 4600 6000
-100000 -95000 -90000 -85000 -80000 -75000
Figure 17 : Location of the coal seams at Usutu colliery near Camden power station.
The influence of flooding on underground coal mines Page 28
3.3 Topography and drainage
Ermelo is situated in the upper reaches of the Vaal River, about 1 720 m above sea level.
The town lies in the midst of a varying topography, superb layout and vivid green flora.
Together with its high rainfall Ermelo is unmistakably known as the garden town of
Mpumalanga. The topography is of a gentle, rolling nature. Steeper slopes are present at
sandstone outcrops. Studies by the mining industry indicate that surface run-off for this
area is in the order of 6 to 10% of the annual rainfall of 710 mm, with 8% as an average
in a study by Grobbelaar et al., (2004).
The surface drainage system is obviously important in intermine flow management
explained by Grobbelaar et al., (2004), because topographically low areas would be the
areas where decanting from mines is expected. The most vulnerable areas would be
areas where connections between mines and the surface occur, and where these
coincide with a surface low.
The regional surface contours of the mine itself are illustrated in Figure 18. The area
topography slopes mostly away from the mining area towards the south with an average
3.2%, as the mining area is situated on a topographic high in the northern part at 1 750
mamsl decreasing towards the southern part to 1 625 mamsl. The unit for the map is
mamsl
·293400
-293700 ,...
-294000
-294300
-294600
Oullln. C-S .. m
-294900
C-LAoIg-L •• r
Ou1line 8-Seam
Surf & DC FI•• ,
-99000 -96 00 -93000 -90000 -87000 -84000 -81000 -78000
Figure 18: Surface contours of the mine area.
The influence of flooding on underground coal mines Page 29
In Figure 19 the surface topography is displayed in 3D, with the topographic high to the
north of the mining area clearly visible. The high towards the north-west of the Southern
Mine is also visible (pink colour towards the green). The 3D image helps us get an idea of
the topography in the way we see with our eyes when looking at hills or mountains.
,
~---
---~~-.....
Figure 19: 3D visualization of the surface contours with projected underground and
opencast.
Figure 20 shows the rivers and dams in the area together with the coal seams. A non-
perennial stream system exists in the area and runoff accumulating in the surface can
persist for several weeks until water has evaporated or infiltrated into the ground. Water
that is polluted needs to the treated before it can be released into a river system.
The influence of flooding on underground coal mines Page 30
-293500
-294000
-294500
-100000 -95000 -90000 -85000 -80000 -75000
Figure 20: Rivers and dams in the area of the mine.
In Figure 21 the regional surface contours of the area around the mine are displayed in
3D with the rivers and dams superimposed on that. According to this picture the local
drainage pattern is towards the south-east.
Figure 21: Regional surface contours of the area around the mine with rivers and dams.
The influence of flooding on underground coal mines Page 31
3.4 Land use
In Mpumalanga most land which is currently not mined is used for commercial grain and
livestock farming and, where possible, farmers make use of mine water to irrigate their
crops. In Mpumalanga, mining still has a future of more than 20 years, and in some areas
up to 50 years, if one takes into account the life of existing coal mines, the mining of new
coal reserves and the mining of new minerals. The energy industry in the area is also
growing, with suppliers intending to increase the number of electrical power stations in
the area. Most former mining land has been rehabilitated and then converted into farming
land, with both crop and livestock farming succeeding. Most farms in the vicinity of former
mining land use water from the closed mine for irrigation (Nthabiseng, Molapo &
Chunderdoojh, 2006). In the hydrosensus done some of these farms around the closed
Usutu mine were investigated. The tap water in some houses had an unpleasant smell.
Some of the farmer's wives complained that clothing after washing were yellowish in
colour. After looking at the quality of the water no problems were detected. The regional
water quality was the same as the "top" sampled samples.
3.5 Rainfall and climate
The highest temperatures in this region according to WeatherSA are from December to
March (24 to 25°C). The coldest months in winter are June and July (16°C). Most of the
rainfall occurs in the summer months from November to January and this can also be
seen in the water levels. Rainfall from April to September is low.
The annual rainfall (MAP) for the area is 705 mm (SA Weather Service - Rainfall stations:
Ermelo airport no. 0442841 8; 04428128; 0480170 4; 0479870 X; 10491078). Figure 22
shows the rainfall from 1960-2011.
The influence of flooding on underground coal mines Page 32
Annual Total Rainfall
Rainlal [nvnJ
1100.----------------------------,
1000
900
800 I.Rainfall
700
600
500
Time
Figure 22: Rainfall graph for Ermelo.
3.6 Recharge using Chloride method
The assumption necessary for the successful application of the chloride method to
determine recharge is that there is no source of chloride in the soil water or groundwater
other than that from precipitation. Chloride levels are low in the system. Steady-state
conditions are maintained with respect to long-term precipitation and chloride
concentration in the case of the unsaturated zone. However, this assumption may be
invalidated if the flow through the unsaturated zone is along preferred pathways (Van
Tonder and Xu, 2000). According to Vegter (1995) groundwater recharge can be read off
a map indicated in Figure 23
The influence of flooding on underground coal mines Page 33
Groundwater Recharge (Vegter 1995)
Recharge (mm/yr)
c::J 0.1
c::J 3
8
12
"20
c::J32
c::J45
CJ65
"95
"135
"180
"220
50~0 ~-0-~~50~0 ~~10-00 Kilom-eters--~s
Figure 23: Recharge according to Vegter (1995).
Rainfall that infiltrates into the weathered rock soon reaches an impermeable layer of
shale or dolerite undemeath the weathered zone. The movement of groundwater on top
of this layer is lateral and in the direction of the surface slope. The groundwater
reappears on the surface at fountains where the flow paths are obstructed by a barrier,
such as a dolerite dyke, paleo-topographic highs in the bedrock, or where the surface
topography cuts into the groundwater level at streams. It is suggested that less than 60%
of the water recharged to the weathered zone, eventually emanates in streams. The rest
of the water is evapotranspirated or drained by some other means (Hodgson, Vermeulen,
Cruywagen & de Necker (2007).
In areas of extensive underground high extraction, it can be safely assumed that all
recharged water will migrate downwards to enter into the collapsed mine workings. Under
undisturbed conditions, 3% of the annual recharge would be an acceptable average
value. Under disturbed conditions above long wall panels, recharge is likely to be greater
and 5% of the annual rainfall would be a good first estimate (Hodgson et al., 2007).
Water in operating opencast pits is derived from various sources. Table 2 provides a
breakdown of these sources as a function of the average annual rainfall or the total
ingress of water into a pit. (Hodgson & Krantz, 1998)
The influence of flooding on underground coal mines Page 34
Table 2: Water recharges characteristics for opencast mining (Hodgson et al., 1998).
Sourceswhich contribute water Water sources into opencast pits Suggested average values
Rainonto ramps and voids 20-100%of rainfaII 70%of rainfall
Rainonto unrehabilitated spoils (run-off and seepage) 30- 80%of rainfall 60%of rainfall
Rainonto levelled spoils (run-off) 3 - 7%of rainfall 5%of rainfall
Rainonto levelled spoils (seepage) 15- 30%of rainfaII 20%of rainfall
Rainonto rehabilitated spoils (run-off) 5-15% of rainfall 10%of rainfall
Rainonto rehabilitated spoils (seepage) 5-10% of rainfall 8%of rainfall
Surface run-off from pit surroundings into pits 5 -15% of total pit water 6%of total pit water
Groundwaterseepage 2- 15%of total pit water 10%of tata I pit wate r
General Equation: R = (P Cl, + D)/CI
[R = recharge (mm/a); P = mean annual precipitation (mm/a); Cl, = chloride in rain (mg/I);
D = dry chloride deposition (mg/m2/a); Cl, = Clsw = chloride concentration (mg/I) in soil
water below active root zone in unsaturated zone OR Cl, = Clgw = chloride concentration
(mg/I) of groundwater where for many boreholes the Clgw = harmonic mean of the Cl
content in the boreholes]
The regional recharge was calculated as 5.7% by using the chloride method. The chloride
values measured at boreholes in the mine area are displayed in Table 3. See Appendix E
for the chloride values of all the boreholes.
Table 3: Chloride method for recharge.
HARMEAN (mg/I) CI- rainwater (mg/I) Recharge (%)
17.54 1 5.7
3.7 Hydrogeology
3.7.1 Pre-mining groundwater occurrence
Three distinct superimposed groundwater systems are present within the Oliphants
catchment. They can be classified as the upper weathered Ecca aquifer, the fractured
aquifers within the unweathered Ecca sediments and the aquifer below the Ecca
sediments (Hodgson et al., 2007).
3.7.2 The Ecca weathered aquifer
The Ecca sediments are weathered to depths of 5 to 12 m below the surface throughout
the Mpumalanga area. The upper aquifer, typically perched, is associated with this
The influence of flooding on underground coal mines Page 35
weathered zone and water is often found within a few metres below surface. This aquifer
is recharged by rainfall. The percentage recharge to this aquifer is estimated to be in the
order of 1 - 3% of the annual rainfall in other parts of the country (Hodgson et al., 2007).
Observed flow within the Mpumalanga Area confirmed isolated occurrences of recharge
values as high as 15% of the annual rainfall. It should, however, be realized that within a
weathered system, such as the Ecca sediments, highly variable recharge values can be
found from one area to the next. This is due to variations in the composition of the
weathered sediments, which range from coarse-grained sand to fine clays (Hodgson et
al.,2007).
The aquifer within the weathered zone is generally low yielding (range 100 - 2000 Uh),
because of its insignificant thickness. Few farmers therefore tap this aquifer by borehole.
Wells or trenches, dug into this upper aquifer, are often sufficient to secure a constant
water supply of excellent quality (Hodgson et al., 2007).
The excellent quality of this water can be attributed to the many years of dynamic
groundwater flow through the weathered sediments. Leachable salts in this zone were
washed from the system long ago and it is only the slow decomposition of clay particles
which presently releases some salt into the water (Hodgson et al., 2007).
3.7.3 The fractured Ecca aquifers
The pores within the Ecca sediments are too well-cemented to allow any significant
permeation of water. All groundwater movement is therefore along secondary structures,
such as fractures, cracks and joints. These structures are better developed in competent
rocks such as sandstone, hence the better water-yielding properties of the latter rock type
(Hodgson et al., 2007).
It should, however, be emphasized that not all of the secondary structures are water-
bearing. Many of these structures are closed because of compressional forces that act in
the earth's crust. The chances of intersecting a water-bearing fracture by drilling decrease
rapidly with depth. At depths deeper than 30 m, water-bearing fractures with significant
yield were observed in opencast coal-mines to be spaced at 100 m or greater. Scientific
siting of water-supply boreholes is necessary to intersect these fractures. The conclusion
is drawn that boreholes have insufficient yields for organized irrigation. This is confirmed
by a survey of the catchment, during which no irrigation from this aquifer could be found
(Hodgson et al., 2007).
Coal seams are often fractured and have some hydraulic conductivity. Underlying the
coal is the Dwyka tillite. It is impermeable to groundwater flow due to its massive nature
The influence of flooding on underground coal mines Page 36
and fine matrix, forming a hydraulic barrier between Pre-Karoo aquifers and those high
up in the succession (Hodgson et aI., 2007).
In terms of water quality, the fractured Karoo aquifer always contains higher salt loads
than the upper weathered aquifer. This is ascribed to the longer residence time of the
water in the fractured aquifer ((Hodgson et aI., 2007).
Although the sulphate, magnesium and calcium concentrations in the Ecca fractured
aquifer are higher than those in the weathered zone, they are well within expected limits.
The occasional high chloride and sodium levels are from boreholes in areas where salt
naturally accumulates on the surface, such as at pans and some of the fountains
((Hodgson et al., 2007).
3.7.4 Pre-Karoo aquifers
In only a few instances, drilling has intersected basement rocks underneath the Karoo
Super group. Very few, if any, of the farmers tap water from the aquifer beneath the
Dwyka formation. Reasons for this are (Hodgson et aI., 2007):
• The great depth.
• The low-yielding character of the fractures.
• Inferior water quality, with high levels of fluoride, associated with granitic
rocks.
• Low recharges characteristics of this aquifer, because of the overlying
impermeable Dwyka tiIIite.
In the southern portion of the catchment, dewatering of this aquifer has occurred to some
extent, because of the pumping in the Evander Goldfields. This does not impact on the
;,1
Ecca Aquifer due to the presence of the (Hodgson et aI., 2007).
3.7.5 Pre-mining surface hydrology
Annual rainfall in the areas where underground high extraction is done ranges from 650 -
760 mm per annum. Surface slopes are gentle. Run-off is in the order of 8(Surface water
in streams gains from groundwater seepage. Seepage into streams is mainly from the
weathered aquifer. This is generally not sufficient to cause significant flow in the streams
(Hodgson et al., 2007).
The influence of flooding on underground coal mines Page 37
3.7.6 Coal Seam (mine)
This system is mined out, with a much higher transmissivity value than the layers above
and below it.
• The transmissivity (T-value) of the mined coal seam (Alfred and Gus Seams were
mined as a single seam) is very high (in the order of thousands) and the storativity
(S) is also very high (65% in the mined out section as opposed to approximately
0.1% in typical Karoo aquifers).
• Once the mine has filled up with water, a horizontal piesometric level will occur
(this piesometric level is also horizontal during the filling up process). If the
piesometric level intersects the surface, decant could take place at the point of
intersection if there is a link between this position and the mine (e.g. a borehole).
The rate at which the piesometric level rises is dependent on the amount of influx from
the top layers (or along subsidence areas) and the amount of lateral groundwater outflow
The Ecca Group consists mainly of shales, with thicknesses varying from 1 500 m in the
south, to 600 m in the north. Since the shales are very dense, they are often overlooked
as significant sources of groundwater. However, as illustrated in Figure 24, their
porosities tend to decrease from -0.10% north of latitude 28°S, to < 0.02% in the
southern and southeastern parts of the basin, while their bulk densities increase from -2
000 to > 2 650 kg.m-3 (Woodford & Chavallier, 2002).
PoroolIy 22 24 26 28 30
'Mr
» lO S 2400
>2· '0 ,.,.00·2&00
"2800· 2650
2 "2MO·,eeo
"2MO 2670
<, '2870· '880» '880
30
J2
IS
Figure 24: Porosity and bulk density variations in shales of the Karoo Basin (Woodford &
Chavallier,2002).
The influence of flooding on underground coal mines Page 38
The possibility exists that economically viable aquifers may exist in the northern parts of
the Basin underlaid by the Ecca shale. It is therefore rather surprising to find that there
are areas, even in the central parts, where large quantities of water are pumped daily
from the Ecca formations. One should thus not neglect the Ecca rocks as possible
sources of groundwater, especially the deltaic sandstone facies. Rowsell and De Swardt
(1976) report that the permeability's of these sandstones are usually very low (Woodford
& Chavallier, 2002).
The main reason for this is that the sandstones are usually poorly sorted, and that their
primary porosities have been lowered considerably by diagenesis. The deltaic
sandstones represent a facies of the Ecca sediments in which one would expect to find
high-yielding boreholes. Unfortunately, Rowsell and De Swardt (1976) have found that
the permeability of these sandstones is also usually very low. However, the Vryheid
Formation sandstones in KwaZulu-Natal (west of Pietermaritzburg) appear to be more
permeable, with a median borehole yield of 0.33 fis and 62% yielding greater than 1 fis
(KwaZulu-Natal project, 1995, unit 8) (Woodford & Chavallier ,2002).
3.8 Geology
The Geology of Usutu mine lies within the Karoo group, Ecca subgroup in the Vryheid
formation (Jeffrey, 2005).
The Permian-aged Ecca Group comprises a total of 16 formations, reflecting the lateral
facies changes that characterize this succession. Except for the fairly extensive Prince
Albert and WhitehilI formations, the individual formations can be grouped into three
geographical zones, the southern, western - northwestern and northeastern. The basal
sediments in the southern, western and north-western zones (Prince Albert and WhitehilI
formations) of the basin will be described first, followed by the southern Collingham,
Vischkuil, Laingsburg, Ripon, Fort Brown and Waterford formations. The remaining
western and northwestern sediments of the Tierberg, Skoorsteenberg, Kookfontein and
Waterford Formations and the north-eastern Pietermaritzburg, Vryheid and Volksrust
Formations will then be considered. In addition, a relatively small area along the eastern
flank of the Basin, between the southern and north-eastern outcrop areas, contains 600 -
1 000 m of undifferentiated Ecca mudrock, which has not yet been studied in detail
(Woodford, Chevallier, 2002).
According to Grobbelaar et al., (2004) the sediments of the coal-bearing Ecca Group of
the Karoo Sequence were deposited on an undulating pre-Karoo floor, which had a
significant influence on the nature, distribution and thickness of many of the sedimentary
formations, including the coal seams. Post-Karoo erosion has removed large parts of the
The influence of flooding on underground coal mines Page 39
stratigraphic column, including substantial volumes of coal over wide areas. The Karoo
Super Group comprises the Ecca Group and the Dwyka Formation. A general thickening
occurs from north to south. The Ecca sediments consist predominantly of sandstone,
siltstone, shale and coal.
Furthermore, Grobbelaar et al., (2004) explain that Ecca sediments overlie the Dwyka
group (loosely referred to as the Dwyka tillite). This formation consists of a proper tillite,
siltstone and sometimes a thin shale development. The upper portion of the Dwyka
sediments may have been reworked. The Dwyka sediments are underlaid by a variety of
rock types, such as the Bushveld Complex in the north, Witwatersrand Super group in the
south, Waterberg Super group in the north-west and Transvaal Super group to the west.
Technically, the Karoo sediments are practically undisturbed. Faults are rare, except for
displacement along dolerite ring structures. Fractures are common in competent rocks
such as sandstone and coal.
The Vryheid Formation thins towards the north, west and south from a maximum of
approximately 500 m in the Vryheid-Nongoma area. The uneven pre-Karoo topography in
the vicinity of the northern and north-western margins of the basin, directly on pre-Karoo
rocks or the Dwyka Group, gives rise to marked variations in thickness. In these areas,
the Vryheid Formation pinches out against numerous local basement highs. Thinning and
pinch-out towards the southwest and south are due to a facies gradation of its lower and
upper parts into shales of the Pietermaritzburg and Volksrust Formations respectively
(Woodford & Chavallier, 2002).
The Vryheid Formation comprises mud rock, rhythmite, siltstone and fine- to coarse-
grained sandstone (pebbly in places). The Formation contains up to five (mineable) coal
seams. The different lithofacies are mainly arranged in upward-coarsening deltaic cycles
(up to 80 m thick in the south-east). Linear coastline cycles are, however, fairly common,
particularly in the thin north-western part where they constitute the entire Vryheid in
places (see Figure 25).
A relatively thin fluvial interval (60 m thick) which grades distally into deltaic deposits
towards the south-west and south occurs approximately in the middle of the formation in
the east and north-east. Fining-upward fluvial cycles, of which up to six are present in the
east, are typically sheet-like in geometry, although some form valley-fill deposits. They
comprise coarse-grained to pebbly, immature sandstones with an abrupt upward
transition into fine-grained sediments and coal seams (Woodford & Chavallier, 2002).
The influence of flooding on underground coal mines Page 40
1· AlJu'ILaI P In
2· Der...1~a~n
3 . Delta Fronl
~ - P,odeltil
5 - Conltnental SJope
6 - Basln Pla n
Congbmerale
o Sandstone
m• MudrodcBasemen.
(C)
Q... 5.O...O. km
:::::::::::::::~!~~!~(k~:::::::::::::==
~ SOUTHERN ~ MAGMA TIC ARC ~
o 500 km
'"-----'
__________ =: _ :AGPiEP_F+OLl.D BIC!.T."-""811·1"..!...5.9~_k"__-'m-_=-··=:_=:-:'=-- _
~ SOUTHERN ~ MAGMATIC ARC ~
(A)
Figure 25: Source areas for the southern and western Ecca Formations (B) and the
northern Pietermaritzburg, Vryheid and Volksrust Formations (C). Depositional
environment of the Ecca Group in the southern Karoo (Woodford & Chavallier, 2002).
Ermelo 0 to 100 m geology shown below:
3.8.1 Vryheid Formation
• E Seam (0 to 3 m),
• D Seam (0.6 m),
• C Lower Seam (1.5 m, sandstone partings in upper section),
• C Upper Seam (well developed, 0.7 to 4 m, sandstone, siltstone or mudstone
partings split seam into 2 to 3 plies, devolatilized I destroyed by dolerite over large
areas)
• B Lower Seam, B Upper Seam (may coalesce in south, 0 to 3 m), A (isolated
outliers, 1 m),
The influence of flooding on underground coal mines Page 41
• A Seam (0 t01.5 m, mainly removed by erosion). Dip gently southwest, minor
folding; dykes (2 to 5 m) common, up to 8 sills (10 to 250 m) transgress and uplift
the seam.
3.8.2 Quality of the coal
The Ermelo coalfield's E Seam is of reasonable quality but the economic potential of the
seam decreases southwards as it becomes torbanitic and/or shaly whereas in other
areas it might be too thin to be viable for mining.
The D Seam is of good quality and has no clastic partings but has a high proportion of
vitrain with minor durain bands. The C Lower Seam is the most important seam as it is
the main source of export coal. The C Upper Seam is generally of poorer quality, has no
in-seam partings and may be torbanitic in the upper part; however, the lower part of the
seam is usually of good quality, making it the main target for mining. It is typically mined
to supplement the C Lower. The B-seams are low quality, dull coal that contains fewer
vitrain bands compared with the lower portion of the C Upper Seam (Jeffrey, 2005).
The coal seams at Kilbarchan were mined through the method of Bord-and-pillar
extraction. Information sources vary, but seem to suggest that the Alfred and Gus Seams
were mined as a single seam. The seams are separated by a very thin sandstone parting,
forming a composite coal seam in some areas (Hodgson, 2006).
3.9 Hydrocensus
The hydrosensus were performed on the boreholes on the mine and the surrounding
farms in the area (Figure 26).The hydrocensus gives one a very good idea of the quality
and water levels of the area before starting an investigation.
The influence of flooding on underground coal mines Page 42
-293500 .. , ~.~-t- ~.
.' .
. . : \:'. . . - •..•' ,:.. t.,' i
- • "0-. r ~ !
..... l !, ~. 't
r, I
-294000
-294500
-295000
D~h L.lyer
OutIIn. C·Sum
C·L.. g·L.. ,
Outline a-Seam
\ll'GS2630CA.TIF
·15000 1500 3000 4600 0000
-295500~r- .- r- ~ ~ ~ r
-100000 -95000 -90000 -85000 -80000 -75000
Figure 26: Location of the boreholes in the hydrocensus for the investigation.
The information of all the boreholes in the hydrosensus is shown in Figure 26. Information
on the date, water level and comments on the person's name or anything observed
during the hydrosensus. As previously discussed many of these farmers complained
about the colour and odour of the water at their farms.
The influence of flooding on underground coal mines Page 43
Table 4: Information of the boreholes received during the hydrocensus
Name Date Water Level (mbgl) Comment
Usutu F top 2011/02/15 21.6 Running water
Usutu A top 2011/02/15 3.3 Running water
Usutu Atop 2009/07/03 6.2
Usutu B top 2011/02/15 12 Running water
Usutu B top 2009/07/03 24.5
Usutu D 2011/02/15 18 Running water
Usutu D 2009/07/03 26.3
Usutu E top 2011/02/15 28 Running water
Usutu E top 2009/07/03 34.8
Usutu I 2011/02/15 49.7 Running water
Farm1 2011/02/15 3.6 Vlakfontein
Farm2 2011/02/15 4.2 P.Cilliers-Roodewal
Farm3 2011/02/15 3.98 ROSENDAL
Farm4 2011/02/15 4.02 ROSENDAL
Farm5 2011/02/15 4.07 ROSENDAL
Farm6 2011/02/15 3 Water smells like swael S. Cloete 0835661929
Farm7 2011/02/15 20.39 Saambou
Farm8 2011/02/15 6.26 Jan Theron 1666-ERMELO-Jan Hendriksfontein
Farm9 2011/02/15 9.77 Jan Theron 1666-ERMELO-Jan Hendriksfontein
Farm10 2011/02/15 19.5 Jan Theron 1666-ERMELO-Jan Hendriksfontein
Farm11 2011/02/15 4.08 Jan Theron 1666-ERMELO-Jan Hendriksfontein
Farm12 2011/02/15 30.98 Kleinveld
Farm13 2011/02/15 8.2 van Outshoorn stream
Farm14 2011/02/15 20.12 van Outshoorn stream
Farm15 2011/02/15 2.25 van der Merwe
Farm16 2011/02/15 3.3 Bethesda
Farm17 2011/02/15 10.6 Trucks
Farm18 2011/02/15 3.9 Guesthouse
Farm19 2011/02/15 4 Panelbeaters
Farm20 2011/02/15 4.26 Panelbeaters
Farm21 2011/02/15 2.1 Panelbeaters
Table 4 show information from the hydrosensus done at Usutu mine. The date, water
level and comments are recorded during the hydrosensus
It is important to know whether the boreholes are drilled into the mine. 3D models
together with the coal seams will tell whether this occurs. Water quality also comes into
account, to know whether the qualities inside the mine are really inside the mine, but is it
clear that the quality inside the mine is poorer than the quality not inside the mine.
The influence of flooding on underground coal mines Page 44
Table 5: The sample depth of the boreholes.
Site name Sample depth (mbgl) Depth of borehole
Usutu A top 5 90
Usutu A bottom 89 90
Usutu B top 15 112
Usutu B bottom 111 112
Usutu D 20
Usutu E top 30 50
Usutu E bottom 49 50
Usutu F top 23 26
Usutu F bottom 25 26
Usutu I 48 46
3.10 Boreholes under investigation
In general it was difficult to locate the boreholes, with the grass being long and the
boreholes not well marked.
3.10.1 UsutuA
• This borehole is 90 m deep and the water level is at 3.3 m.
• The borehole is not locked, as the photograph below shows.
3.10.2 Usutu B
• This borehole is 112 m deep and the water level is at 12 m.
• The sound of running water can be heard at the borehole.
3.10.3 Usutu 0
• The water level in Usutu D was 18 m.
• The sound of running water can be heard at the borehole.
• The borehole is not locked, as the photograph below shows.
3.10.4 Usutu E
• This borehole is 50 m deep and the water level is at 28 m.
• The sound of running water can be heard at the borehole.
• The borehole is not locked, as the photograph below shows.
3.10.5 Usutu F
• This borehole is 26 m deep and the water level is at 21.6 m.
• The sound of running water can be heard at the borehole.
• The borehole is not locked, as the photograph below shows
3.10.6 Usutu I
• This borehole is 48 m and the water level is at 46.7 m.
• The sound of running water can be heard at the borehole.
The influence of flooding on underground coal mines Page 45
• The borehole is not locked as the photograph below shows.
• It is difficult to locate.
3.11 Water levels
The water levels in the mine were measured in July 2009 and again during the
hydrocensus. These levels and all the others from the farm boreholes are presented in
the time graph in Figure 27. There has been a definite rise in the water levels since 2009,
but this may be attributed to the exceptional rainy season during 2010/2011. These levels
should be monitored over time to draw meaningful conclusions as only 4 boreholes water
levels were measured previously in 2009 as to mismanagement. These levels should be
monitored over time to draw meaningful conclusions. Figure 28 illustrates the proportional
water level distribution for February 2011.
Water Level Depth
Water Level Depth [ml
• Fannl
0.0.-----------------------------, ... Fannl0
+ Fannll
Fann12
• Fann13
• Fann14
• ... Fann1510.0 •, + Fann16
... Fann17
-I- Fann18
• -I- Fann19+ FarnQ
20.0 •• , FarnQO... FarnQl
• Fann3
-I-Fann4
+ FaIlTÓ
30.0 + Fann6
• Fann7
• FannB
• Fann9
L8J1uAtlp
40.0 , ll!JtuBklp
• ll!JtuD
-I- ll!JtuEklp
t- ll!JtuFtop
+ll!Jtul
50+.-0~_____,_____,-r__.____r_-y---y----,---,----r-___._---,-----r-----,-----r-----,_____,,__.•---.---,..---r-'
7/2009 1/2010 7/2010 1/2011
Time
Figure 27: Water level time graph of a few boreholes measured since 2009.
The proportional distribution of the water levels in Figure 28 clearly indicates that
extraction occurs in some farming boreholes, namely the farm boreholes north of the
mine. The average regional water level is currently less than 4 m. Boreholes displayed as
blue dots.
The influence of flooding on underground coal mines Page 46
The water levels of the boreholes inside the mine differ between the two mines. The
water levels in the southern mine are shallower than those in the northern mine.
However, the water levels differ substantially in the different boreholes, most likely the
result of compartmentalization due to the ventilation walls. This makes interpretation of
the mine water levels very difficult, and a number of additional boreholes have to be
drilled to obtain a clear indication of the exact water levels in the mine.
Once the mine has filled up with water, the piesometric level of the mine will rise with the
storage coefficient value of the mine (and not the specific yield) as conditions have
changed from unconfined to confined. The flux from the overlying aquifers into the mine
aquifer will decrease as the two water levels approach each other.
What is clear from the water levels is that the aquifers above the mining areas are not
fully recharged yet, as these pressure levels differ from the regional water table aquifer
water levels.
. , .: .t- ," .. ''~. .
' ...... ~
-293
-294
-294
"l.~-;';I .. Y ••• r
'.' . ;..,..' <: ..
," j:;;: 'S ':'~~<'::~';.'
-2950 LEGEND .";~.
_ Outline C-Seam .' .ï>: .~l ..
C-Laag-Laer
Outline B-Seam
OIC Outlne - B-Seam
Vl.l3S2630CA .TIF
-295 -150m 1500 3000 4500 6000
-100000 -95000 -SUIJO -85000 -80000 -75000
Figure 28: Proportional distribution of the water levels last measured.
The influence of flooding on underground coal mines Page 47
3.12 Mining methods
According to Grobbelaar et al., (2004) coal extraction has been ongoing in the
Mpumalanga coalfields for more than 100 years. At first, mining was mainly confined to
the west of Witbank. Many of these mines have closed down and are presently a major
source of pollution. This poor quality should serve as a warning of what would happen in
the rest of the mining area, if intermine flow commenced there. Through the years, mining
extended from its original position to the south and east. Many new collieries commenced
with mining, particularly during the past 30 years. Since 1970, mining has increasingly
been mechanized. All modern coal-mining methods are currently employed. These are:
• Bord-and-pillar extraction.
• Secondary mining through pillar extraction. This is commonly referred to
as stooping.
• Underground high extraction through long wall and short wall methods.
• Auger mining is presently considered on a limited scale for the extraction
of thin coal seams.
• Opencast mining, using truck and shovel or dragline methods.
Initial underground mining was relatively shallow, in the range of 10 - 50 m below surface.
Mining was mainly through underground methods. Access to the underground workings
was commonly through inclined shafts.
3.12.1 Bord-and-pillar extraction
Bord-and-pillar extraction has been the primary method of mining throughout the
Mpumalanga coalfields. The reasons for this are Grobbelaar et al., (2004):
It allows access to the coal in a structured and organized way.
It can be manoeuvred around geological or coal quality constraints.
The extraction rate is reasonably high, ranging from 70% in the shallow mines to
50% in deeper areas.
3.13 Mine layout
Grobbelaar; et aI, 2004 state as follows: Water in mined-out areas will flow along the coal
floor and accumulate in low-lying areas. This is a simplistic view of the situation. In reality,
water would accumulate in many isolated areas, where it would dam against barriers of
coal or dolerite dykes left
The layout plan for Usutu colliery gives the B-seam; C-seams (see Figure 29 ). Mining
activity ceased in the late 1980s and then the mine was flooded. Usutu has mined two
seams underground namely, the Band C-seam respectively; with an average mining
The influence of flooding on underground coal mines Page 48
height of 1.6 m. Currently Vunene Mining has an opencast mine above the C-seam of the
northern mine. This may result in cracks and water moving into the underground.
~S2630CA.TIF
·295500
-100000 -95000 -90000 -85000 -80000 -75000
Figure 29: Layout of the B- and C-seams at Usutu colliery, together with Vunene
opencast.
3.13.1 Roof thicknesses
The combined C-seam varied in depth from very shallow (1.4 m) to nearly 80 m in the
north, the shallowest part of the roof as illustrated in Figure 30. The southern mine, where
the B-seam was mined, varied between 75 m and 160 m, as illustrated in Figure 31.
The combined floor contours for both seams are illustrated in Figure 32. From this the
difference in depth between the two seams is very clear, with the B-seam much deeper
than the C-seam.
The influence of flooding on underground coal mines Page 49
t '.'. -ti. -
• '\ .._!.
III
. ,. - l ::;
.....-. . '/';' .:'
~..,,:>~'" -:-:.C
• .l ,~~, -,
,·'-1
-29450
"<,"" t ,IJ+":":-\;~"' ...rI ,iJ",
OIC Vunene > ''';.~', _~>..;.,:.,.~: ' -:, )1 '1
•.~ ..,........ I ._;
Dc-seem Roof Thickness LowesP!aint
.' '... 1, "'1" ,.,i •. ,,,/' ,:,',' ,'_,_'.\
'.: ~.•_~~~~:/.. rÓ.«:
c-seem Roof Thickness
-295000 VItOS263OCA,TIF
-100000 -95000 -90000 -85000 ,80000 -75[0)
Figure 30: Roof thickness of the C-seam (northern mine) displayed in mamsl.
-293500 c_ • -'~. I /'
.... ~~~'l ",.
"lj/
: ·
--~~":_..J. ---::',1."" /t ...•.~-
I
-294000
-294500
LEGEND
8 Seam EASTRoofThlckness
Dc-seam RoofThlckness Lowest Point
-295000
WGS2630CA.TIF
-100000 -95000 -90000 -85000 -80000 -75000
Figure 31: Roof thickness of the B-seam (southern Mine) displayed in mamsl.
The influence of flooding on underground coal mines Page 50
3.13.2 Floor contours
-293700
-294000
1051),0
-294300 10<10.0
1530.0
1520.0
-294600
LEGEND
_ Outline C.S'Jm
C.L~ ..g·L.. ,
Outlinl B-Sl:lm 1-J-294900 -100CO 1000 2000 3000 4000
-96000 -93000 -90000 -87000 -84000 -81000
Figure 32: The combined floor contours of both the Band C-seam (illustrated in mamsi).
Cross Section
Figure 33: Cross section of the floor contours together with the surface topography
Shows a cross section of the floor contours together with the surface topography. The
section line is indicated with a red line in the previous figure:,.._~~ _
uv -
The influence of floodin
-293700
-294000
-294300
-294600
C-Laag-Laer
Outlin. B·Sum
-294900 ·10000 1DOO 2000 3000 4000
-99000 -96000 -93000 -90000 -87000 -84000 ·81000 -78000
Figure 34: The combined roof contours of both the Band C-seam (illustrated in mamsi).
Cross Section
Figure 35: Cross section is shown of the roof contours.
In the cross section is shown of the roof contours. The blue indicates the surface
topography. The section line is indicated in the previous figure by a red line.
In Figure 36 the surface in relation to the two seam elevations is illustrated in 3D. From
this the surface elevation to the north is clear, and the difference in depth for the two
seams in relation to the surface is also illustrated. Also illustrated is the position of the
Usutu boreholes that are drilled into the mine.
The influence of flooding on underground coal mines Page 52
------ -
Figure 36: 3D combined surface and seam elevation with boreholes into the
underground.
3.14 Ventilation seals
Barrier pillars between mines or compartments should be designed for the specific
purpose of keeping water from transgressing through it; or to release water at a specific
rate through controlled flow mechanisms. Only one of the mines has ever done
permeability testing in a barrier pillar. It is essential that barrier pillars be tested and that
monitoring systems be installed before intermine flow takes place in an uncontrolled
fashion. Issues such as liability and accountability can only be addressed once this
information is available Grobbelaar et al., (2004).
In order for a mine to function properly, ventilation walls are needed to channel the fresh
air away from those parts of the mine that were finished and directed the flow to the front
of the mining operations. In the olden days these walls were built very solid and sealed
corridors completely from bottom to roof.
The walls are so strong that they will function as "Iow pressure" seals resulting in a
compartmentalized underground, withstanding huge pressures created by the recharged
groundwater. These seal positions are important in the recharge and management
calculations for a mine as well as for the water balance and mine dewatering. The seals
in Usutu (also digitized from the old maps) are illustrated in Figure 37.
The influence of flooding on underground coal mines Page 53
Also important is the barrier between different mines, which will determine mine interflaw.
The position of the Mooiplaats Colliery to the south of the southem mine is also illustrated
in this figure. The shortest distance of the pillar between the collieries is 250 m.
,.
" '..~"
..-I
-293500 - . , .... , . J',.. ~:.--'" "
. .: ,~ .
-294000
.". J • ,I ". '
-294500 ,.. ,: l} '.~-
',:, . II...... f·....' -. ; r.. I :~.:
l· I I ' "<::.:':' .. /'/I' ,
" I':", ..1.,........ ,/,I" : ..:~y~:.
;};.. "'_ ..::/1
.: -_~,1_.: :-:--" ': ~~.,>.,~.'.)'~'", / .-;,';l.;~
-295000 -....._. t -t'" .T-(,',-- _.
, I I --. Jo" ,/ _' '. :
I ,_.-150(0 1500 3000':4500 6000
-100000 -95000 -90000 -85000 -80000 -75000
Figure 37: Layout of the Band C-seams illustrating the ventilation walls, as well as the
position of the adjacent Mooiplaats Mine.
3.15 Faults
According to Grobbelaar et aI., (2004) dolerite intrusions in the form of dykes and sills are
present in the Ecca Group. The sills usually precede the dykes, with the latter being
emplaced during a later period of tensional forces within the earth's crust
The faults and dykes in the area were mapped using maps of the mine plans provided by
Usutu (Figure 38) there are a few long faults cutting through the mine. This may have an
influence on the water flow inside the mine. It can easily be seen that where a fault
occurred the mining stopped and just continued on the other side of the fault. The faults
and dykes also play a role in compartmenting the mine as it normally also displaced the
coal seams vertically. The water will move along these faults in the underground.
The influence of flooding on underground coal mines Page 54
-293500
-294000
-294500
. .'. I
I . •••. ' ..1
shafts & Inclines "J I "\' ,.r'-
Faults
Outline c-seam >L!/<;~,;
c-Laap-Laar
I
Outline 8·Seam
-295000
WOS2630CA.TIF
'-- _J( . '. I~
-100000 -95000 -90000 -85000 -80000 -75000
Figure 38: Faults and dykes identified inside the mine.
3.16 High extraction zones
According to Hodgson (2007) underground high extraction of coal has been performed for
many years in South Africa. The impact that this has on groundwater and surface water
resources has been investigated in most instances, and this information is available as
part of their EMPR application.
Two mining methods for the underground high extraction of coal are generally used in
South Africa. These are:
• Pillar extraction, also referred to as stooping.
• Long wall mining, with a short wall variation.
To date, stooping has been performed at many collieries. The extent of this may vary
from experimentation in a single area to stooping over most of the mine.
Long wall and short wall mining, in view of its significant initial capital outlay has only
been implemented at a few collieries. These are the Eskom tied mines, such as Coal
braak, Matla, New Denmark and Arnot; the Sasol mines at Sasolburg and Secunda and
Durnacol in KwaZulu-Natal (Hodgson, 2007).
The influence of flooding on underground coal mines Page 55
Much of this information is of a historic kind, unable to predict with confidence the final
geohydrological and hydrochemical outcome of such systems. This complicates matters
when applying for closure of the mine, because of uncertainties associated with future
water quantities and qualities.
Stooping is the process whereby pillars from bord-and-pillar mining areas are extracted.
Continuous mining are generally used for this extraction process, cutting sections into
pillars until most or all of the pillars have been removed from specific areas (Hodgson,
2007).
Pillar extraction usually commences at the furthest point, retreating to the entrance of an
area. The roof in areas where pillars have been extracted is left to cave in, which it
usually does. This caving often causes subsidence on the surface. Subsidence cracks
may penetrate to the surface, allowing groundwater and surface water to drain into the
collapsed mine workings (Hodgson, 2007).
The amount of subsidence at the surface and the severity of cracks depend on (Hodgson,
2007):
• The depth of mining and the nature of the overlying rock.
• The coal seam thickness mined.
• The area stooped and panel geometry.
The degree to which pillars have been left in the stooped area, thus still supports the coal
roof.
• The following interpretation is relevant to water flow (Hodgson, 2007):
• The intensity of stooping varies tremendously from area to area.
• The stooping method applied also varies, removing anything from a small portion
of a pillar to removing all of it.
Information on whether or not the roof in stooped areas has collapsed and, if so, whether
the impact extends to surface, is not available.
The influence of flooding on underground coal mines Page 56
Bord.-and-iPillar- Pos:t-andi-Stall
(a)
Stoo p-anldl-Room
Figure 39: Different methods of mining (Hodgson, 2007).
Figure 40: High extraction (stooping) example on an old mine map.
High extraction areas were also identified in the mine from the old maps of Usutu shown
in Figure 40. High extraction was only done in the B-seam; the combined total high
extraction percentage was 14% of total the mining area (stooping in the central area is
22% and in the southern area 13%). These areas are illustrated in Figure 41.
The influence of flooding on underground coal mines Page 57
These areas may result in subsidence, increasing the recharge factor into the mine. The
position of the shafts may also influence recharge if not properly sealed, and may also act
as decant points if it is lower than the lowest point of mining) .
... :, _ ...,;; r
/ .- - ,,'
-293500
-294000
-294500
Outline C-Seam
c-Laau-Laer
Outline 8-Seam
8 Seam Stooping
-295000 WGS2630CA.TIF t.
'\ I
-100000 -95000 -90000 -85000 -80000 -75000
Figure 41: High extraction areas identified in the mine.
3.17 Water balance
The minimization of water volumes in mines leads to high salt concentrations, but smaller
volumes of water that need to be handled. Up to now, minimization strategies have not
been applied to a significant extent as a management tool. This is largely due to the vast
scale of coal mining operations. It is almost impossible to change a mining strategy, for
instance, from underground high extraction to bord-and-pillar, once mining has
commenced. Other water minimization strategies include the artificial flooding of closed
mines, improved rehabilitation of surface and the planting of trees. The conclusion is that
water influx minimization should be considered carefully against the option of active
flushing according, to Grobbelaar et aI., (2004).
The WISHNVACCMAN software was used to calculate the amount of water stored in the
defunct mine. The software calculates the volume of water in the underground taking into
The influence of flooding on underground coal mines Page 58
consideration the current water level elevations and the geometry of the workings,
including mining height, extraction factors and the position of the ventilation walls.
Due to the many compartments in the underground and only three water levels in the
workings, a few compartments show up as being empty. It is highly unlikely that the
compartments are empty but without water level information a water volume for a
compartment cannot be calculated. This makes the management of a flooded mine with
ventilation walls very difficult.
Mine compartments will fill themselves with water. The speed at which this happens is
defined by the recharge rate which is depending on the geology, the depth, the type of
mining and whether some form of high-extraction was performed.
Table 6: Anticipated recharge to bord-and-pillar mining in the Mpumalanga area
(Hodgson et al., 2007).
Description Recharge as a % of annual rainfall
Influx into bord and pillar,deep mining 1--4
Influx into bord and pillar,shallow mining 4--9
Influx into bord and pillar mining with stooping 4--12
Stooping has been done in some parts of the B-seam at Usutu colliery. Stooping, the
process of removing pillars from the workings causes the overlying strata to collapse.
Cracks will develop. If these cracks intersect with the surface, rainfall water and run-off
may flow down directly into the mining cavity. From studies done at collieries where
stooping was done (Hodgson et al., 2007), it has been found that normally a recharge
rate of 4 - 12% should be applied, depending on the depth of mining and ratio of
stooping.
According to the data the B-seam is completely filled with water. This has happened over
the last twenty years (this is the number of years since mining has ceased in the B-
seam). From this can be concluded that there must be a recharge of at least 5%.
Although the C-seam is shallower, no stooping has been done and the recharge is also
considered to be 5%. The Ermelo region receives, according to Weather SA, an average
rainfall of 705 mm per annum.
The influence of flooding on underground coal mines Page S9
Table 7: Properties of the seams.
Seam Avg Depth Area not stooped Area stooped Recharge factor
m ha ha %
CSeam West 44 1560 0 5
C seam Central 125 108 0 0.1
B seam Central 97 333 96.5 5
B seam East 107 2125 315 5
The total water make for the B-seam is 2 687 m3/day and for the C-Seam 1 612 m3/day.
After the opencast mining activities ended there will be a rehabilitated open cast on top of
the old workings (Figure 42). From the hydrology it is known that the average
evapotranspiration is about 80% in terms of the rainfall, leaving 20% for recharge on a
not fully rehabilitated pit. A pit that is badly rehabilitated and where all rainfall would run
off towards the spoils may expect a maximum recharge of 20%. A very well
rehabilitated pit with free draining from the spoils can expect a recharge of 10%.
·293500
-294000
-294500
LEGEND
_ Outline C-Seam
C-La·u-Laer
OlC Outlin. - B-S•• m
-295000
WO S2030 CA_ TI r
-100000 -95000 -90000 -85000 -80000 -75000
Figure 42: Position of the Vunene opencast pits in relation to the C-seam.
The pit is currently in use and is not rehabilitated. A recharge of 20% is assumed.
Only pit A and B are situated on top of the underground; the combined size of the two pits
is almost 59 ha (588 000 m2). The thickness of the layer between the C-seam
The influence of flooding on underground coal mines Page 60
underground and the floor of the opencast (B-seam) vary between 10 and 1.6 metres.
With such a thin layer between the opencast and the underground cracks are imminent.
All the water that recharges into the opencast will flow through the cracks into the
underground. The amount of water anticipated from the two opencast pits is 588
000*0.705/365*0.20 = 227 m3/day, resulting in a substantial higher water for the C-seam
of 1612+227-22.7 = 1816 m3/day.
The total water in the C-seam is currently 19.1 Mm3. Some areas show no water (Figure
43), not because there is no water but no information is available on the existence of
water!
WL 1624.0....YQ1: 492 994
~I
..54)))
Figure 43: Water bodies in the C-seam.
A stage curve depicts the relationship between a stage height (the actual water level
elevation) and the volume of water inside the underground.
This assumes that the water table is horizontal and continues over the entire mine. In
Figure 44 the stage curve for the C-seam is illustrated, indicating that for a water level of
1647 mamsl (Usutu E water level elevation) the total water volume will be 17300 ML.
The influence of flooding on underground coal mines Page 61
Stage Curve - Complete C-Seam
Heigtt [mj Confined by 1.6 m (65% void space)
1660,0 -,-----------------------------,
1650,0 Dalun 1 7,0
1640,0
1630,0
1620,0
1610,0
1600,0
1590,0
Voh.me [MIJ
Figure 44: Stage curve for the C-seam.
For the B-Seam the total water volume is 22.88 Mm3. The water bodies are illustrated in
Figure 45. The dry areas are not necessarily dry; there is just no information available
The influence of flooding on underground coal mines Page 62
"8000 ... 000 "'000 "2000
Figure 45: Water bodies in the B-seam.
Stage Curve - Complete B-Seam
Height [mj Confined by 1.6 m (65% void space)
Voltme [MIJ
Figure 46: Stage curve for the B-seam.
In Figure 46 the stage curve for the B-seam is illustrated, indicating that the total water
volume will be nearly 25 000 ML.
The influence of flooding on underground coal mines Page 63
Decant:
It appears that there is a minimum stratification in the water columns in the two mines. As
there are a number of ventilation seals in the mine, that reduce any change of pressure
that can "push" the mine cavity water up to a decant level.
In 2010 there was a borehole decanting water at a supposed rate of 7 I/s (Figure 47).
This borehole is situated in the far south of the B-seam (Figure 48) in a very isolated
panel which (according to mine personnel) was completely sealed off during mining. One
possible reason for this sudden decanting may be the collapse of a ventilation wall.
Figure 47: Photograph of decanting borehole.
The influence of flooding on underground coal mines Page 64
..~~~1'.')
.> ",- . 11
-294500
. :'. '.._: I". ·.f··- '. ;/
... ; ,J " c·';:.V. l·"
.:~- w -'./
, .: Data Layer ." '-:.,' «, : ,1',./.,: .. '..'•
" •• \ ~_•I• -
_ Outline C-Seam r, '~'.. " ,I •- • •
-295000 -.-.....~.. ", ,,':,- ..._, ,C-Leeg-Laer . " ;..:r·:,,' .,'. '_ '.. l- :L ~..:.-,".Outline B-Seam !,' ,:'-.: .~": .
WGS263OCA.TIF J
-15000 1500 3000 4500 6000
-295500~ .- -r- -.- -'- ---'---'
-100000 -95000 -90000 -85000 -80000 -75000
Figure 48: Position of the decant borehole in the far south of the B-seam.
The water level elevations of the two boreholes in the B-seam are illustrated against the
background of the surface contours in Figure 49. A section was drawn through the mine
on a line as indicated in Figure 49. The section (as illustrated in Figure 50) intersects both
boreholes. The section illustrates the boreholes, their water level elevations, the position
of the mining cavity and the surface in relation to each other. From this two things are
clear:
• The water levels (not pressure levels) follow the topography (which is an accepted
fact in geohydrology).
• At no point the mining cavity is close to the surface. Therefore no pressure can be
created from the water in the mining cavity for mine water to decant. It is therefore
unlikely that the borehole that decanted resulted from pressure that has been
created from inside the mining cavity. The break of a ventilation seal, creating
sudden pressure, may thus be a possibility.
The influence of flooding on underground coal mines Page 65
-293500 1800
1775
1750
1725
1700
-294000 1675
11650
1625
1600
1575
-294500
1550
-295000
LEGEND
o Outline 8-Seam
D Surf s oe Floor
-295500
-95000 -85000 -80000 -75000
Figure 49: Water level elevations (mamsi) of Usutu A and B (with the blue cross-section
line intersecting the two boreholes).
The general flow direction is from right to left (north-east to south-west) following the
topography shown in Figure 50
150QL-----------------------------
o Cross Section 10000
Figure 50: Cross-section of boreholes Usutu A and B.
The influence of flooding on underground coal mines Page 66
3.18 Mine Interflow
Several mining operations are currently being conducted in the vicinity of Usutu Colliery.
The next heading will deal with Mooiplaats as shown in Figure 51
-2935000
-2940000
-2945000
AI Futu .. UIG Mning
_ Die Outline· S-Seam
I CJ Outline S-Seam
I CJ Outline C·Seam
.::;C, J C-Laag-Laer
-2950000 L J l'uboiplaas
-100000 -95000 -90000 -85000 -60000 -75000
Figure 51: Planned future mining and mining adjacent to Usutu Colliery.
3.18.1 Mooiplaats
Mooiplaats Colliery operates to the south of the South Mine (B-seam). The two collieries
are divided by a regional dyke. The barrier (pillar) between the two collieries is 250 m at
the closest point between them. The level of mining also differs (according to Mooiplaats
geohydrologists). It is highly unlikely that any mine interflow will thus occur between these
mines, as a pillar of that width will result in an impregnable layer (which will even increase
with the presence of the regional dyke that acts as a buffer).
Currently water is being pumped from the Usutu B-seam to provide Mooiplaats Colliery
with water (Figure 52). Recharge into the Usutu South Colliery is 2 687 m3/day; therefore
pumping must be less than this to ensure that the water level does not drop below the
roof of the mining cavity. Currently only 0.07 Mm3 - 0.14 Mm3 is used on average daily,
according to figures obtained from the Mooiplaats geohydrologist.
The influence of flooding on underground coal mines Page 67
Figure 52: Photograph of the extraction wells pumping water from Usutu South to
Mooiplaats Colliery.
During an investigation, including EC profiling of the borehole (Figure 53), water could be
heard running down the borehole. This is aquifer water draining directly into the mining
cavity. These boreholes are situated on a topographic high and water from the high
ground thus drains into the mine.
Borehole Log - Pompstasie BH
Depth (mj Lol»lIty - x -87Q21 00 v- 25145210.00 Z: 1157300
30
40
50
.0
70
80
00
'00
Figure 53: EC profiling of the pumping borehole.
The influence of flooding on underground coal mines Page 68
The quality of the water being pumped is displayed in Table 8. From this it is clear that
there is an improvement of the water quality. The measured quality corresponds with the
profiling done during the field visit. Flushing is an accepted practice to improve the quality
of mine water if the pumping rate (and thus the water levels) is managed properly,
keeping the water level above the mining cavity.lf the mine stays flooded then no
additional oxygen ingress and pyrite oxidation will occur.
Table 8: EC quality over time of water being pumped to Mooiplaats Colliery.
The time series for the above tabel is shown in Figure 54
Electrical conductivity
EC
Time
Figure 54: Time series of the EC quality over time of water being pumped to Mooiplaats
Colliery
3.18.2 Vunene Mining
Underground:
The block in the north-east of the main dicates the planned future underground mining of
Usutu. If this underground mining operation uses the central shaft of Usutu Colliery for
entrance, it means that the mine has to be kept dry. This may result in a constant ingress
The influence of flooding on underground coal mines Page 69
of oxygen into the old mine workings, which may enhance acid rock drainage. No detailed
data is available as yet.
Opencast:
The thickness of the layer between the C-seam underground (mined out) and the floor of
the opencast (B-seam), currently mined by Vunene Mining, varies between 10 and 1.6
metres. With such a thin layer between the opencast and the underground cracks are
imminent. All the water that recharges into the opencast will flow through the cracks into
the underground. The amount of water anticipated from the two opencast pits is
588000*0.780/365*0.18 = 226 m3/day, resulting in a substantial higher water make for the
C-seam of 941 rrr'/day, There is no water quality available due to the current mining
operations, but such ingress through the opencast will result in poor quality water seeping
into the underground workings, deteriorating the water currently in these workings.
Hodgson et aI., (2003) indicate that sulphate generation in the flooded underground is
less than 2 kg/ha/day [depending on the depth of mining and flooding rate and that of
opencast mining 7 kg/ha/d).
3.19 Quality of the water
At most of the larger mines, the opportunity exists for mine water of different qualities to
be mixed, thus improving the overall water quality. Typical benefits of doing this would lie
in pH adjustment and iron precipitation. For the latter, retention of the mine water in a
surface holding facility where aeration is possible, is necessary. Such a facility could also
be used for the quick release of the water during flood discharge. Very few other
chemical benefits would be forthcoming from mine water mixing, because most of the
constituents are under saturated in this water (Grobbelaar et aI., 2004)
Water quality assessment is used to evaluate hydro-geochemical site conditions and the
current state of aquifers. Sulphate, TDS and pH are the essential chemical components
used for first-order pollution categorization in South African coal mines.
Mining at this locality has been in progress since the mid-eighties. Through accurate
monitoring of water from different collapsed panels, it should have been possible to plot
sodium and sulphate concentrations to determine:
• The possible lifespan of sodium neutralisation.
• The possible escalation of sulphate production due to the intervention of bacterial
oxidation as the pH of the water drops.
Unfortunately, this kind of information is not available, since the importance of detailed
and complete data sets is only realized by the mining community nowadays. Stream
The influence of flooding on underground coal mines Page 70
water quality in the coalfields deteriorated over the past 20 years, due to seepage and the
discharge of mine water (Grobbelaar; et a', 2004)
Table 9 shows the chemistry data for the boreholes over time 2009-2011. All these
boreholes were sampled at the top of the water column. During the hydrocensus the mine
boreholes were sampled at the top as well as at the bottom in the mine.
The criteria used for inorganic sampling is the SANS 241 :2006, and for organic analysis
the USEPA Standards. Inorganic water samples are classified as:
• Class 1- acceptable (Not color coded)
• Class II - allowable (color coded
• Above - not allowable (color coded red)
The influence of flooding on underground coal mines Page 71
Table 9: Chemistry of the boreholes
The influence of flooding on underground coal mines Page 72
Table 10: Results of the chemical analyses for the boreholes sampled during the
hydrocensus February 2011.
It is clear from this table that there is a definite difference in water quality of the top and
bottom samples of those boreholes drilled into the mining cavity.
• Usutu 0 is not drilled into the mine, as is clear from the water quality.
• Usutu A indicates high sodium and alkalinity with sulphate values slightly elevated
but still within the drinking water standards. It appears as though the formation
has an influence on the water quality in this borehole, even at the top (see also
the Stiff diagram in Figure 58).
• Usutu B also differs top and bottom. Again the sodium is elevated in the bottom
sample, with the difference that the sulphate is elevated and not the alkalinity; the
coal seam has a definite influence on the sample in the mining cavity.
• Usutu E shows no influence from mining activities; it is not clear whether this
borehole is drilled into the mining cavity or perhaps into a pillar.
• Usutu F and Usutu I show typical coal water reactions, with high sulphate, calcites
and magnesium (see also the Stiff diagram in Figure 58). The mining depth in this
area is shallow compared with the B-seam in the southern mine. It is likely that
there is less buffering material, hence the acid rock drainage (ARD) reactions and
the subsequent decrease in pH.
The influence of flooding on underground coal mines Page 73
The expanded Durov diagram clearly indicates that Usutu F and I are mine water, while
Usutu B top, Usutu E top and Usutu D are unpolluted (see Appendix for explanation of
the Durov interpretation). Usutu A and Usutu E bottom are coal mine water.
Expanded Durov Diagram
Mij
Na+K l•o U;utu IU;utu F Iq>
• U;l.(u F bátan
• U;utu Elq>
• U;utu Ebátan
• U;l.(uD
• U;l.(u Blq>
•o U;utu B bátanU;utu A Iq>
S04 • U;utu A bátan
Cl
Figure 55: Expanded Durov diagram of the Usutu boreholes.
Two boreholes were profiled for EC. Usutu F (Figure 57) clearly illustrates the polluted
water in the mining cavity compared with higher up in the water column. Usutu A also
differs in depth, but the change happens halfway down the water column, strengthening
the statement that the sodium bicarbonate water is geology-related.
The influence of flooding on underground coal mines Page 74
Borehole Log - Usutu A top
Depth(ml Locality - X: -87788.00 Y: 21147098.00 Z: 1641.00
'"
10
20
30
40
50
60
70
60
go
Figure 56: EC profiling for Usutu A
Borehole Log - Usutu F top
Depth (m) Locality - X: -89914.00 Y: 2941048.00 Z: 1661.00
EC Tomp
20.5 r200--------,------.,.-----""l800 1700 1712
21.0
21.5
22.0
22.5
23.0
23.5
24.0
24.5
25.0
25.5
Figure 57: EC profiling for Usutu F.
Another interesting and conclusive way of comparing the different waters that emanate
from the coalfields is using the so-called Stiff diagram as explained by Grobbelaar et a.,
(2004). This allows comparison between six components for each site. The six
The influence of flooding on underground coal mines Page 75
components are usually calcium, magnesium, sodium + potassium, alkalinity, sulphate
and chloride + nitrate. Each of these is plotted on horizontal axes, with cations to the left
and anions to the right. The extremities of the points are connected and the inside is
coloured, thus creating unique shapes that represent the overall chemistries for each site.
While these diagrams may appear to be highly varied in shape, this is exactly the
purpose of this plot. It characterizes water with a unique signature, clearly showing the
dominant constituents at each of the sites.
The Stiff diagrams of the boreholes (which compare the water qualities in equivalents)
also illustrate in Figure 58 that the bottom samples are more polluted than the top
samples, with Usutu I and to a lesser degree Usutu F (bottom) very typical coal mine
water.
STIFF Diagrams
Usutu A bottom Usutu Atop Usutu B bottom Usutu B top
20110217-12tOO Z)11l217·12HX) 20,.:»217- 1300
~-~ Cl N... K Cl N....K Cl N.++< "
,." ca 'Ik ca Alk ca
Mg V SOl Mg SO< Mg SO< Mg so<
50 meq/l 50 50 meqn 50 50 meq/l 50 50 meq/l 50
Usutu D Usutu E bottom Usutu E top Usutu F bottom
20110217- 'aDO 3:)11:1217·12'0:) 3:l11)217·12tOO
NI'" Cl N.+K Cl tiI+K Cl NI++< "
.." ca ..Ik ca
Mg SOl Mg SO< Mg SO< Mg so<
50 meq/l 50 50 50 50 meq/l 50 50 meq/l 50
Usutu Ftop Usutu I
20110217-12tOO 2l1m17·12tOO
NI'" Cl NI+K Cl
'" ca 'Ik
Mg SO< Mg SO<
50 meq/l 50 50 50
Figure 58: Stiff diagrams of the mine boreholes.
The time graph of the mine boreholes indicates that the water quality for the top
boreholes did not deteriorate since 2009. The time graphs of all the macro-elements
follow the same trend. Only Usutu F shows a different value, but this may be because the
borehole was sampled at a different depth during the hydrocensus (the water in this
borehole is only 6 m deep).
The influence of flooding on underground coal mines Page 76
Figure 59 Usutu A shown in red shows typical mine water development where it
increases, stabilizes and then decreases. The bottom sampled samples shows higher EC
values than the top sampled. This is due to stratification and when sampling at the bottom
the water sampled is located inside the mine.
Electrical conductivity
EC
350.----------------------------------------------------------,
•
300
__ lBJtu A top
250
__._ lBJtu B top
___ lBJtu E top
lBJtu F top
200
e-. l.BJtu A bottom
.•. ...... lBJtu B bottom___ lBJtu E bottom150
+- lBJtu F bottom
___ lBJtu I
Time
Figure 59: Time graph of the EC for the mine boreholes.
2000
Sulphate as 504 ·
1500
1000
500 •
----== ~ ---- · __ l..B.Jtu A bottom
Calcium as Ca -.- l...I!lJtuAtop
.. _._ l.BJtu B bottom300 l.BJtu B top
- --+- l.BJtu 0200 • • l.BJtu E bottom100 _- .._ - ___ .. __._ l.BJtu E top......._ U:utu F bottomI --+- l.BJtu F top--. l...BJtul
Magn.. umas fv\l
100 •
50 /·
. ~ ..o
71200; 112010 712010 112011
Time
Figure 60: Time graph of sulphate, calcium and magnesium of the mine boreholes.
The influence of flooding on underground coal mines Page 77
70.0,---------------------------------,
Chloride as Cl
60.0
50.0
40.0
30.0
II- U;Uu AIq>
.- U;Uu Blq>
_..,. U;Uu D
U;Uu Elq>
.. U;Uu Flq>
--- U;Uu I
-.- U;Uu Abcitcrn
_..,. LtUu Bbdlcrn
+- LtUu Ebdlcrn
, U;Uu F bdlcrn
400
,
• •
112010 712010 1/2011
Time
Figure 61: Time graph of sodium and chloride of the mine boreholes.
3.19.1 Regional water quality
The EC of only two farm boreholes is available, namely Farm 2 and Farm 6. These are
depicted in Figure 62. From this it is clear that the regional aquifer water quality is similar
to that of the "top" samples in the mine boreholes, indicating aquifer water for most areas
above the mine; the exceptions are Usutu A, which has completely different sodium-
bicarbonate water, and Usutu F which has only a few metres of water above the mining
cavity. This is also clearly shown in the Stiff diagrams of the top mine samples and the
farm samples (Figure 63). Usutu A top shows typical mine water characteristics with high
levels of Na, K and Cl. Usutu F top shows high levels of pollution with high S04 values.
The influence of flooding on underground coal mines Page 78
~i'«:' ..' .
,:;;;'r./, .... ... ;....J ~
-293500 J. :<... ,
'Jj " ~ . .'
-294000
'~.':., .' _;;:< .,_:',
,t..i:'~:..-~.,.\"',~~'/;'":"';:.::\,\',":)~,_-
-294500
" ...... : .... I.... :" ve, • 1\:
"I. ~ ... J \t
::-}iJ;?~~:.:~~· :~;.<. )~
-295000 "-:" " . ~"'..- ....... '-~.' ',' ~ ,' vc'.LEGEND . ~'.-! ' " tI. .: .....OutIIn" C·S .. mc-t, .. g·l...Outline B·Seam ~ _]
WGS2630CA.TIF ·15000 1500 3000 41500 6000
-295500
-100000 -95000 -90000 -85000 -80000 -75000
Figure 62: EC of all the boreholes measured during the hydrocensus in 2011,
STIFF Diagrams
Farm2 FarmS UsubJ A Iq!
20110211·12tOO 2I'J110217-12t'OO 2011Q217·12tO)
NI'K a No'" a NI'" Cl
C> AI< C> V AI< C> [> AI<~ ~ SOl ~ SOl ~ ~ SOl
25 meqn 25 25 meqn 25 25 meqll 25
UsubJ B top UsubJ D UsubJEIq!
NI'" 20110217·12tOQ a 2011Q211·12tOJ NI'" 20110217·12tOONo'" Cl a
C> AI< C> AI< C> AI<
~ ~ SOl ~ ~ SOl ~ \V SOl
25 meqn 25 25 meq/l 25 25 meqn 25
UsubJFIq!
NI'" 20110217'12tOO a
C>
~ lli AI<SOl
25 meqn 25
Figure 63: Stiff diagrams of the top mine samples and the farm samples.
The influence of flooding on underground coal mines Page 79
3.20 Conclusions and recommendations
The following conclusions are drawn from this case study:
• The area topography slopes mostly away from the mining area to the south with
an average slope of 3.2%, as the mining area is situated on a topographic high in
the northern part.
• Non-perennial rivers exist in the area.
Water levels and quality:
• The average regional water level is currently less than 4 m. Some of the farm
boreholes to the north of the mine are being pumped, resulting in water levels of
deeper than 20 m. In the hydrocensus it was noted that usually one borehole was
being pumped from at a farm.
• The water levels of the boreholes inside the mine differ for the two mines. The
water levels in the southern mine is shallower than those in the northern mine.
However, the water levels differ substantially in the different boreholes, most likely
the result of compartmentalization due to the walls. This makes interpretation of
the mine water levels very difficult, and a number of additional boreholes have to
be drilled to get a clear indication of the exact water levels in the mine.
• The regional aquifer water quality is similar to that of the "top" samples in the mine
boreholes, indicating aquifer water for most areas above the mine.
• The bottom samples are more polluted than the top samples, with Usutu I and to a
lesser degree Usutu F (bottom) very typical coal mine water. The mine boreholes
indicate that the water quality for the top boreholes did not deteriorate since 2009.
It also appears that there is a minimum stratification in the water columns in the
two mines. As there are a number of ventilation seals in the mine, they reduce any
change of pressure that can "push" the mine cavity water up to a decant level. At
no point is the mining cavity close to the surface. Therefore no pressure can be
created from the water in the mining cavity for mine water to decant. It is therefore
unlikely that the borehole that decanted resulted from pressure that has been
created from inside the mining cavity.
• The regional recharge was calculated as 5.7% by using the Chloride method.
Recharge into the mine was calculated as >5%.
The influence of flooding on underground coal mines Page 80
Mining data:
• The average mining height of 1.6 m.
• The combined C-seam varied in depth from very shallow (5 m) to nearly 80 m in
the north. The southern mine, where the B-seam was mined, varied between 75 m
and 160 m.
• Also important is the barrier between different mines, which will determine mine
interflow. The shortest distance of the pillar between the collieries is 250 m. The
faults and dykes also play a role in compartmentalizing the mine as it normally
also displaced the coal seams vertically.
• High extraction areas were also identified from the old maps. These areas may
result in subsidence, increasing the recharge factor into the mine
• The total water make for the B-seam is 2 687 m3/day and for the C-seam 1 612
m3/day. The total water in the C-seam is currently 19.1 Mm3. For the B-seam the
total water volume is 22.88 Mm3.
Recommendations
• It is recommended that the monitoring programme in the mine be executed
properly with water being sampled in the mining cavity and higher up in the water
column very quarter that is 4 times a year of all the Usutu boreholes. More
boreholes into the mining cavity will also contribute to a better understanding of
the current water volumes.
• It is further concluded that flooding is also influenced by factors occurring in and
around the mine. Thus in looking at the influence of flooding the bigger picture is
important. This makes the management of a flooded mine very complex. The
compartmentalization of the mine makes it even more difficult to interpret the data.
• The final conclusion is that if flooding is well managed it is a very successful
operation in managing a mine through preventing acid generation. This is done by
depleting the oxygen component in acid generation.
The influence of flooding on underground coal mines Page 81
CHAPTER 4
KILBARCHAN MINE, NEWCASTLE: CASE STUDY
4.1 Introduction
Kilbarchan is applying for partial closure and also needs to know what the current
groundwater status of the mine water quality is. The coal mine has been flooded
(recharged) with water as the production stopped between 1990 and 1992.
The coalmines in KwaZulu-Natal are another example of mining impact on water
resources. The coal-mines are situated in a topographically dissected area, where coal
seams outcrop above the valley bottoms. In most of the collieries, high extraction mining
methods were used: bord-and-pillar followed by pillar extraction and long wall mining (at
one mine). The result of the high extraction mining at all of these mines has been the
collapse of the overlying strata
Rainwater easily moves through the mining-induced fractures in the collapsed strata. The
water is contaminated (acidified) through the oxidation of pyrite in the rocks and loose
material left in the mines. Because of the limited water retention in the mine openings,
most of them decant water from sub-outcrop areas or shafts, into the valleys below. Coal
discard dumps have been left on the surface at the coalmines in Natal, in different states
of rehabilitation, adding an additional source of water pollution (Zeelie & Hodgson, n.d.).
As a result of this, the Department of Water Affairs and Forestry (DWA&F) had to review
their catchment water quality standards for some of the catchments. Sulphate
concentrations of up to 1 000 mg/L in the Black-Mfolozi and the Nkongolana Rivers are
allowed (Zeelie & Hodgson, n.d.).
Some of the larger coalmines have closed or are in the process of applying for closure
certificates. Many other mines have simply been abandoned and very little information is
available regarding conditions in the mine or their possible impact on the water resources
(Zeelie & Hodgson, n.d.).
The influence of flooding on underground coal mines Page 82
Figure 64: Picture of the Kwazulu-Natal area (Zeelie & Hodgson, n.d.).
4.2 Locality
Kilbarchan colliery is located 10 km south of Newcastle in KwaZulu-Natal. It comprises
two underground sections, called Ray Point and Kilbarchan. Figure 67 shows the location
of Kilbarchan colliery with the two sections Kilbarchan and Roy Point. The Ingagane River
runs along the eastern side of the colliery. The mine supplied coal to the Ingagane power
station, to the east of the mine. To the south of Kilbarchan lays Natal Cambrian Colliery,
which has also closed down.
The influence of flooding on underground coal mines Page 83
• Kosl Bay
Mpumalanga Ndumu Game Reserve Tembe Elephant Reserve
ltala ature Resl'rve • uk. Sibaya
Mkuzl Game Rp\l!rve
PAutpl l&r u•rg odwanaBay
Phlnda Game Reserve
Free State CapeVldal
.Ounooe
lundl St Lucia Wetlands Park
11 .Ror $O"•~ • " StLucla3b3na~o
.udy.mith J
.,Rich d. Bay
KwaZulu-Natal
EShc-,,.. • M Unzlnl
NZ
Drakensbet toOl Iviet:
·Zmkwan
NOl1l"Oh Road S ana '. ~lleld Se eh
B I~.~n • Wartburg I~Oa"k:;'.~ock
, HOWl: p':ler~~IF'JL _ .t!~a_ \[0 _
Lesotho : • Umdloli Sead> :• 8 ver ....... • U hlamgaRoe •uno rbOrg • ...-'"""".0
Hamma~~~& .. ,~~~~l~~_~-_'-ff.b:~~-_-.:.:_-_•~
I iAmaflZlm10lJ ,
I
: Indian Ocean
Eastern ,I 4 PennnglonKoltltad Cape : I~·opo• I, Hibb deneN
I •
I Pori ShEpotone I
~ _~ ~~r~~e J
Eastern Cape
• Port Edw rd @ www.places.co.za
Figure 65: Location of Kilbarchan Mine in South Africa (www.places.co.za).
Figure 66: The location of Kilbarchan mine in Google earth image (Source: Google earth).
The influence of flooding on underground coal mines Page 84
4.3 Topography
Significant topographic differences are present in the KwaZulu-Natal coalfields. Coal
deposits have, in many instances, only been preserved within the mountains. Outcrops of
coal are therefore plentiful along many of the mountain slopes. In these instances, the
depth of the coal seams below surface typically ranges from 0 m to more than 100 m
below surface. Access to these coal deposits has been gained by tunneling into the
mountainside (Hodgson et al., 2007).
-307600
-307800
'~.. .
. "'"
,/
~.'.... I" t~·h '~•._~'J
f -. 6'·
~ _d -,
.' 1,
Kilbarchan : ..:;'1 <, .{:~
Opencast ;:~·..~. ~~'1:!:3:;Jt~·~~~~;;é$~~~./. 150O:~2250'-3ci:~~j;;~. ~:
90000 92000 94000 96000 98000
Figure 67: Location of Kilbarchan colliery in the investigation.
4.4 Land use
Most undermined land has been sold to commercial farmers or has been donated to local
communities for carrying out farming activities. Former mining communities in KwaZulu-
Natal have benefited from timber plantations facilitated by the closing mines through
seasonal jobs to maintain plantations, as well as income generated from the sale of
mature timber to milling companies. Taking into consideration land-use limitations
imposed by previous mining activities, district and local municipalities of KwaZulu Natal
are of the view that large-scale economic revival of former mining areas should include:
tourism, through the marketing of former Zulu-English battlefields; the small-scale mining
The influence of flooding on underground coal mines Page 85
of previously mined areas; and further promotion of commercial agricultural activities in
former mining areas (Nthabiseng, Molapo & Chunderdoojh, 2006)
4.5 Rainfall and climate
The annual rainfall (MAP) of the area is 864 mm (rainfall stations: Kilbarchan offices and
Ray Point. Figure 68 shows a graph of the average rainfall for Kilbarchan area. The
average temperature is high for most of the year, but the highest in December and
January. High rainfall occurs from November to February and a low rainfall in the winter
months from May-July.
Annual Total Rainfall
Rainfall [mm)
1500
1000 I I
I · ~lbatices
500
o J .1, J .1 J .1
2004 2006 2008 2010
Time
Figure 68: Rainfall graph for Newcastle.
4.6 Mine details
Mining activity ceased in 1992. Bord-and-pillar mining followed by stooping has been the
main mining method. Extraction percentages are 50% at a height of 3.5 m (Figure 69).
Floor contours are undulating and several low lying areas exist (Hodgson, 2006). The
mining outline of the two mining areas Kilbarchan and Ray Point is shown in Figure 70.
The influence of flooding on underground coal mines Page 86
-3076000
I,"
;
-3078000
-3080000 ",
-3082000
-3084000
90000 92000 94000 96000 98000
Figure 69: Extraction percentages are 50% at a height of 3, 5 m.
Opencast mining was performed to extend the life of the mine where the depth to the coal
seam was less than 20 m. Five opencast sections have been mined. The average depth
of mining was 11 m for opencast 1 and 15 m for the other areas. Of the opencast pits
only 1A and 1B connect to the underground workings. These connections were sealed off
with concrete plugs. In a previous report (Hodgson, 2006) it was indicated that the seals
were destroyed by blasting in 1992. Thus water moved from the opencast into the
underground workings. This will influence the water quality with the water seeping
through, because the water from the opencast is rich in oxygen. This means oxygen will
move into the system.
The influence of flooding on underground coal mines Page 87
-3076000
-3078000
. '-,
." -. 7' t '( "1"
r '.
~ -(.)
-3080000 --, I
, '
-3082000
-3084000
90000 92000 94000 96000 98000
Figure 70: Layout of Kilbarchan together with the opencast areas.
Coal floor contours dip to the central part of Kilbarchan Underground, with a low spot at 1
144 mamsl.
In an update of the water balance done by Hodgson (2006) problems in the compilation
of coal floor and roof contours were plentiful. Not only does the Kilbarchan section date
back to the fifties, but the maps are in a coordinate system that is no longer in use.
Converting the large maps into digital format presented a major challenge, hence only the
outlines of the mines were digitized by hand (by Jones and Wagener). Scanning and
vectorising of the detailed mine plans proved to be too costly.
Looking at the coal floor and surface contours we can clearly see that the mining depth
was shallow. The depth of mining is in the range of 0 to 200m; in the eastern part the
The influence of flooding on underground coal mines Page 88
mining is much shallower than in the western part shown in Figure 72. In the eastern part
the depth is between 0 to 120m and in the western part 120 to 200m.
'.. -c:~~·"II~ 1'",
1196.0 t .r
-3076000
-3078000
-3080000
-3082000
-3084000
90000 92000 94000 96000 98000
Figure 71: Kilbarchan coal floor contours (mamsi).
Figure 72 shows the depth of mining at Kilbarchan. The depth of mining is important,
because it plays a role in the residence time of the water flow.
The influence of flooding on underground coal mines Page 89
. ~"'}
..~•.I "_
-3076000
-3078000
-3080000
-3082000 '
-3084000
~' LEGEND : ... :.~.:
·~N .lOlb..-cT<n dop ti ofmln ~ - " i
-3086000~ -r .- ~r- -. ,-_r
90000 92000 94000 96000 98000
Figure 72: Depth of mining at Kilbarchan.
4.7 Geology
The Kilbarchan Colliery is situated within the Klip River Coal Field, comprising rocks of
the coal bearing Vryheid Formation forming part of the Ecca Group of the Karoo
Supergroup. Two economic coal seams can be distinguished (Bottom & Top Seams)
within the Klip River Coal Field and is separated by 0.3 to 15 m of coarse-grained, pebbly
sandstone which fines upwards to carbonaceous shale. The Bottom Seam has a
thickness of 1.3 m in the north of the coal field, decreasing to 50 cm towards the southern
area. The Top Seam has a thickness of 3.5 m in the north, decreasing to less than 0.5 m
in the central area and then increasing to 1.5 m in the south (Wilson & Anhaeusser,
1998).
The influence of flooding on underground coal mines Page 90
4.8 General Hydrogeology
Three groundwater systems are normally present in the mining areas of the Ecca
formations. These are the shallow weathered aquifer, the intermediate unweathered
fractured Karoo, and the coal seams (mining area).
The weathered groundwater system:
The top 5 - 15 m in the area consists of soil and weathered rock. The upper aquifer is
associated with this weathered horizon. In boreholes, water may often be found at this
horizon. This aquifer is recharged by rainfall.
Rainfall that infiltrates into the weathered rock reaches impermeable layers of solid rock
underneath the weathered zone. Movement of groundwater on top of the solid rock is
lateral and in the direction of the surface slope. This water reappears on surface at
fountains, where the flow paths are obstructed by barriers such as dolerite dykes, paleo-
topographic highs in the bedrock, or where the surface topography cuts into the
groundwater level at streams.
The fractured groundwater system:
The grains in the fresh rock below the weathered zone are too well cemented to allow
any significant flow of water. All groundwater movement is therefore along secondary
structures, such as fractures, cracks and joints in the rock. These structures are best
developed in sandstone, hence the better water-yielding properties of the latter rock type.
Dolerite sills and dykes are generally impermeable to water movement, except in a
weathered state. In terms of water quality, the fractured aquifer always contains higher
salt loads than the upper weathered aquifer. The higher salt concentrations are attributed
to a longer contact time between the water and the rock.
Coal Seam (mine):
This system is mined out, with a much higher transmissivity value than the layers above
and below it.
• The transmissivity (T)-value of the mined coal seam is very high (in the order of
thousands) and the storativity (S) is also very high (50% in the mined out section
as opposed to approximately 0.1% in typical Karoo aquifers).
• Once the mine has filled up with water, a horizontal piesometric level will occur
(this piesometric level is also horizontal during the filling up process). If the
piesometric level intersects the surface, decant could take place at the point of
intersection if there is a link between this position and the mine (e.g. a borehole ).
The influence of flooding on underground coal mines Page 91
4.9 Recharge
The assumption necessary for the successful application of the chloride method to
determine recharge is that there is no source of chloride in the soil water or groundwater
other than that from precipitation. Chloride is conservative in the system. Steady-state
conditions are maintained with respect to long-term precipitation and chloride
concentration in that precipitation, and in the case of the unsaturated zone. However, this
assumption may be invalidated if the flow through the unsaturated zone is along preferred
pathways (Van Tonder & Xu, 2000).
In areas of extensive underground high extraction, it can be safely assumed that all
recharged water will migrate downwards to enter into the collapsed mine workings. Under
undisturbed conditions, 3% of the annual recharge would be an acceptable average
value. Under disturbed conditions above long wall panels, recharge is likely to be greater
and 5% of the annual rainfall would be a good first estimate (Hodgson, 2006).
General Equation: R = (P Clp + D)/CI
[R = recharge (mm/a); P = mean annual precipitation (mm/a); Cl, = chloride in rain (mg/I);
D = dry chloride deposition (mg/m2/a); Cl, = Clsw = chloride concentration (mg/I) in soil
water below active root zone in unsaturated zone OR Cl, = Clgw = chloride concentration
(mg/I) of groundwater where for many boreholes the Clgw = harmonic mean of the Cl
content in the boreholes]
The regional recharge was calculated as 4% using the chloride method. The data from
the chloride values from the boreholes is displayed in Table 11. Different recharge
methods were used to compare with the Chloride method this included maps from
Vegter, ACRU recharge by Roland Schulze, Geology, harvest potential and soil and
vegetation maps. The average recharge percentage compared well with the findings in
the chloride method of 4%. The EARTH model could be used for a single borehole with
monthly rainfall and water levels used as parameters this also compared well with the
recharge percentage.
Table 11 Different recharge methods were used to compare with the Chloride method this
included maps from Vegter, ACRU recharge by Roland Schulze, Geology, harvest
potential and soil and vegetation maps. The average recharge percentage compared well
with the findings in the chloride method of 4%. The EARTH model could be used for a
The influence of flooding on underground coal mines Page 92
single borehole with monthly rainfall and water levels used as parameters this also
compared well with the recharge percentage.
Table 11: Chloride method for recharge.
HARMEAN (mg/I) CL- rainwater(mg/I) Recharge (%)
25.56 1 4
The actual mine recharge was calculated using the water balance of the Kilbarchan mine.
It was calculated at 11.3 %.
4.10 Monitoring
Kilbarchan colliery is currently experiencing problems regarding the handling of excess
water. The underground and opencast workings have been filling up with water since
mine closure in 1992. Active water management is required to prevent mine water from
spilling onto the surface.
A study by Hodgson (2006) in updating the water balance shows Kilbarchan at the water
levels of 1 160,1 175, 1 183 and 1 190 mamsl (maximum fill level), representing:
• 1992 (1 160 mamsi) - When mining stopped.
• 1999 (1 175 mamsi) - Additional monitoring holes drilled.
• June 2002 (1 183 mamsi) - Water level before the sudden rise of 3.5 m.
• June 2004 (1 990 mamsi) - Mine full and decanting starts.
The conclusion is that, by 1992 when mining ceased, only about 20% of Kilbarchan was
flooded. It took 12 years for the mine to fill.
Currently only three boreholes and a few surface water samples are being used as
monitoring for Kilbarchan. The boreholes are KW1/98, KW1/99 and the VOID BH. The
water levels of these boreholes are very well monitored, almost weekly. The surface
water sampling positions with the position of the boreholes in the area are shown in
Figure 73.
The influence of flooding on underground coal mines Page 93
4:..'. '"- .. ., ..:--','". '~.' .~~.(.• • "'1 """_ .._.~: ,:,,,~ .t:k: t. .::~i '.).:_._..'.3076000 ..~, ._='~",J." II
\ .
i' .:
I..'
3078000
; .... h·P I f.! I: .. ,i.~R
r .
3080000 ~.. :JI
• I
1 ,
3082000
3084000
3086000 188Ed Ed
-7500 750 1500 2250 3000
90000 92000 94000 96000 98000
Figure 73: Position of the mine boreholes at Kilbarchan
The influence of flooding on underground coal mines Page 94
-3075000
-3077500
-3080000
-3082500
-3085000
90000 92500 95000 97500
Figure 74: Surface streams monitoring positions at Kilbarchan colliery.
4.11 Borehole information
The information of the water levels after the previous site visit in April 2011 is shown in
Table 12
Table 12: Information of the boreholes during April 2011.
Site Name Water Level (m below surface)
KW1/98 21.16
KW1/99 7.49
VOID BH 16.50
Table 13 provides a summary of the information for each borehole that was sampled in
the mine during April 2011. Sampling was done at two different depths at each borehole.
The influence of flooding on underground coal mines Page 95
Table 13: Sample depth and depth of boreholes in the mine.
Site Name Borehole depth Sampling depth (T) Sampling depth (B)
KW1/98 45.00 22 44
KW1/99 21.00 8 20
VOID BH 30.00 18 29
4.11.1 Detail of boreholes inside the mine
4.11.1.1 KW1/98
• This borehole is 45 m deep. The water level is 21.16 m deep.
• Borehole is drilled into the mine.
• Water levels are very well monitored.
4.11.1.2 KW1/99
• This borehole is 21 m deep. The water level is 7.49 m deep.
• Borehole is drilled into the mine.
• Water levels are very well monitored.
4.11.1.3 VOID BH
8 This borehole is 30 m deep. The water level is at 16.5 m.
• No past chemistry data is available.
• First time sampled at previous visit April 2011.
• Water in the borehole is already acidic.
4.11.2 Details of surface water samples
4.11.2.1Ingagane River upstream
• Water in the river is unpolluted.
4.11.2.2Ingagane River downstream
o Water in the river is unpolluted.
4.11.2.3 Greenwich stream area 1B
• Sampled
4.11.2.4 Greenwich upstream of dump
• Monitoring water that is upstream of the dump.
4.12 Water levels
As discussed previously only three boreholes water levels are currently being monitored
weekly. The three boreholes water levels KW1/98, KW1/99 (these boreholes are drilled
The influence of flooding on underground coal mines Page 96
into the mine) and the VOID BH are shown in Figure 75. Over time the boreholes have
the same trend. The variations in water levels are only fully understood when looking at
the water elevation figure. Over time the boreholes have shown the same trend.
The variations in water levels are fully understood when looking at the water elevation
figure. Water level depth can easily fool people in this case. The water levels seem to
differ in depth, but they actually lie on a flat line when plotted in metres above mean sea
level (mamsi).
Water Level Depth
Water Level Depth [ml
5.0 -,-------------------------------,
10.0
-- KWlI98(1)
, -..- KWlI99(1)
15.0 -+- VOCB11)
20.0
2004 2006 2008 2010
Time
Figure 75: Water level time graph of the three boreholes measured since 2003.
• KW1/98 has the deepest water level over time, 20-25 m over the past 8 years.
• KW1/99 has the shallowest water levels over the same amount of time, 5-10 m
• VOID BH is in the middle, 15-20 m over time.
The fact that the water level elevations of the different boreholes are all the same shown
in Figure 76, indicate that the water level is flat throughout the underground and
connected opencast mine. Due to decant through the opencast, these mine water levels
will not rise in the underground mine above the decant level.
The influence of flooding on underground coal mines Page 97
Water Level Elevation
Water Level Elevation [mj
1192.00,-----------------------------,
1191.00
1190.00
1189.00
1188.00
1187.00
1186.0+-0~~_~---,--~-~___,______r-~---,--~-.___---,------r-~___._----r-..,....---'
2002 2004 2006 2008 2010
Time
Figure 76: Water level elevation time graph of the three boreholes measured since 2003.
One of the most striking features (Figure 76) is the jump in water levels on a number of
occasions in all the boreholes. In the 2006 report by Hodgson it was explained as a
possible collapse of underground walls. Currently the explanation is that it is due to
recharge and the resulted pumping from the underground, resulting in the 3 m variation in
water level.
Coal Seam (mine)
This system is mined out, with a much higher transmissivity value than the layers above
and below it.
• The transmissivity (T-value) of the mined coal seam is very high (in the order of
thousands) and the storativity (S) is also very high (65% in the mined out section
as opposed to approximately 0.1% in typical Karoo aquifers).
• Once the mine has filled up with water, a horizontal piesometric level will occur
(this piesometric level is also horizontal during the filling up process). If the
piesometric level intersects the surface, decant could take place at the point of
intersection if there is a link between this position and the mine (e.g. a borehole).
The rate at which the piesometric level rises is dependent on the amount of influx from
the top layers (or along subsidence areas) and the amount of lateral groundwater outflow.
The influence of flooding on underground coal mines Page 98
Once the mine has filled up with water, the piesometric level of the mine will rise with the
storage coefficient value of the mine (and not the specific yield) as conditions have
changed from unconfined to confined. The flux from the overlying aquifers into the mine
aquifer will decrease as the two water levels approach each other.
What is clear from the water levels is that the aquifers above the mining areas are not
fully recharged yet, as these pressure levels differ from the regional water table aquifer
water levels.
Two cross-sections (Figure 77 and Figure 79 indicates the position of the sections) were
drawn through the three monitoring boreholes to indicate the water levels in relation to
the surface and the mining void (Figure 78 and Figure 80). The water elevation indicated
by the blue line. From this it is clear that boreholes KW1/99 is the closest to surface, but
that all the water level elevations are the same.
-307600 I1420.0 ~ _.
1400.0
1380.0
I-
1360.0
13'10.0
-307800 1320.0
1300.0
1280.0
1260.0
- 12'10.01220.0-308000 I- 1200.0
1180.0
1160.0
~
... 11'10.0
1120.0
-308200 I1100.01080.0
-308400
LEGEND
N Cross section
_ .I(jlbarchan outline
N .Surt.c. contours from OlM
188R R
-308600 ·5000 500 1000 1500 2000
88000 90000 92000 94000 96000 98000
Figure 77: Cross section of boreholes VOID BH and KW1-99 (with the red line
intersecting the two boreholes
The influence of flooding on underground coal mines Page 99
'''''
VOIDB"
,:DJ
".. KW1.99
WL elevation 1190
'"'' WL elevation 1190
,...
Kilbare.han e.oal
,,"" noor
lnl5
l11D
11]]
Cross SectkJn
Figure 78: Cross-section of boreholes VOID BH and KW1-99.
~ .
-307600 r- ne.
1420.0
f-
1400.0
1380.0
1360.0
1340.0
-307800 I- 1320.0
1300.0
1280.0
1200.0
- 1240.0-308000 1220.01200.0
1180.0
,.., 1100.0
1140.0
I-
1120.0
-308200 I1100.01080.0
-308400
LEGEND
cross section
.I(jlbarchan oUlline
.Surfac. contours from OlM
11:11:11==1 ~ I
-308600 ·5000 500 1000 1500 2000
88000 90000 92000 94000 96000 98000
Figure 79: Cross-section of boreholes VOID BH and KW1-99 (with the red line
intersecting the two boreholes ).
The influence of flooding on underground coal mines Page 100
Surface contours
KW1-99
11!1l Wl elevation 1190
WI.. elevation 1190
Kilbarchan floor
11!1l
contours
117°L_ ....... - .... ---------------,,-.
lleD+----------------------------------
o U(I)Cross Sectioo
Figure 80: Cross-section of boreholes KW1-98 and KW1-99.
4.13 Mining methods
The colliery was mined by bord and pillar method and stooping was done in certain
areas.
Table 14: Different areas of the Kilbarchan mine (Hodgson, 2006).
Decription Area (ha)
Kilbarchan underground area 126.2
Stooped areas 11.84
Oencast areas interconnected with the under round 2.1
Ash-filled areas 10.1
Fly ash from Ingagane power station was introduced into the mine at different locations
(Figure 81); the fly ash needs to be brought into consideration in the volume calculations
of the area. Fly ash fills the voids and limited water gets recharged.
The influence of flooding on underground coal mines Page 101
-3076000
I / -
I,".'' "".'.y'1 ",
. .r ..:IJ.
-3078000
-3080000
-3082000
-3084000
90000 92000 94000 96000 98000
Figure 81: High extraction areas together with fly-ash identified in the mine (Hodgson,
2006)
4.14 Water Balance
4.14.1 Pumping
There are two transfer pipelines positioned at Roy Point (Figure 82). With these transfer
pipelines water is currently being pumped from Kilbarchan to Roy Point, where water is
stored for releases. This is done to create storage capacity at Kilbarchan in order to avoid
uncontrolled mine decants.
The influence of flooding on underground coal mines Page 102
:);./'.~;"~':,';:.r.~v, .~.,.-,~- .'~')
-307600 , .~~tr~t';,.- '"
-307800
-308000
-308200
-308400
-308600
90000 92000 94000 96000 98000
Figure 82: Pumping from Kilbarchan to Roy Point through transfer pipelines.
Table 15: Transfer pipelines volume of water released per annum.
Pumping [m3/day] Pumping [m3/a]
2055 750000
The EC over time of the transfer pipeline does not improve as in the case at Usutu. At
Usutu pumping improved the EC over time. The main reason for this is the fact that too
much pumping is taking place at Kilbarchan.
Table 16: EC of the transfer pipelines over time.
Date 2005-02-17 2006-02-27 2007-03-23 2008-02-08 2009-01-20 2010-01-19 2011-02-16
EC 647.0 719.0 733.0 538.0 670.0 624.0 637.0
Water is also actively released from the Kilbarchan opencast areas via active pumping
during the controlled / licensed releases by a big mobile pump.
The influence of flooding on underground coal mines Page 103
Table 17 shows the volume of water released in the different seasons. The average
pumping is 300 ML/a. This is equal to 300 ML*1000/365 = 822 m3/d. This corresponds to
300 000 rrr'/annum and 10 lIs.
Figure 83: Picture of the mobile pump.
Table 17: Volume water pumped by mobile pump.
Season Big Mobile Pump Deutch (MI)
2004/2005 0
2005/2006 0
2006/2007 0
2007/2008 375
2008/2009 307
2009/2010 292.384
2010/2011 168
4.14.2 Decant
Water is decanting at opencast area 1A and opencast area 1B. For the past 3 years
water has not been decanting from opencast 2; thus it is not used in the calculations. The
water decants from the underground at the opencast areas 1A and 1B which is
connected to the underground.
The influence of flooding on underground coal mines Page 104
The decant point is illustrated in Figure 84, and the v-notch where flow volumes are
measured, in Figure 85.
. 'V". ~ •• ~:
-3076000
-3078000
-3080000
-3082000
-3084000
-3086000
90000 92000 94000 96000 98000
Figure 84: Position of D1, where decant is taking place.
The influence of flooding on underground coal mines Page 105
Figure 85: Picture of the place where the water is decanting in opencast 1B.
Water decanting from the opencast areas according to data from Kilbarchan is measured
at a total of 3.5 I/s as shown in Table 18.
Table 18: Approximate decant volumes of the opencast areas.
Opencasts Decant [lIs] Decant [m3/day] Decant [m3/a]
lA 1 86 31536
lB 2.5 216 78840
Opencast area 2 0 0 0
Area 2: No decanting has taken place for the past 3 years.
1B: Average decanting is measured at 2.5 I/s = 216 m3/Day*365 =78840 m3/year
1A: Average decanting measured at 1 lis = 86 m3/Day*365= 31 536 m3/year
Total decant for all the opencast areas are 110 376 m3/year.
The influence of flooding on underground coal mines Page 106
Figure 86: Picture of the V-notch of water decanting out of opencast 1B.
4.14.2 Water quality
The water decanting out of D1 has been decanting since 2004. Water qualities are not
improving over time at D1. The water decanting out of opencast area 1B is of a poor
quality. The quality of this water has been measured over time. Since 2004 until now the
quality remained the same, as indicated in the time graphs of electrical conductivity and
sulphate (Figure 87) and calcium, magnesium and sodium (Figure 88).
The background colors in the figures represent standards:
The criteria used for inorganic sampling is the SANS 241 :2006, and for organic analysis
the USEPA Standards. Inorganic water samples are classified as:
• Class 1- acceptable (color coded green)
• Class II - allowable (color coded )
• Above - not allowable (color coded red)
The influence of flooding on underground coal mines Page 107
1500~----------------------------------------------------------,
Electricalconductivity - SANS241:2005 (-. -.150.370)
1000
•
500
EJ
Sulphate as S04 - SANS241:2005 (-. -. 400. 600)
8000
6000
t
4000
2000
2006 2008 2010
Time
Figure 87 : Time graph of the EC and Sulphate over time at 01.
2000r-------------------------------------------------------------,
Sodiumas Na - SANS241:2005(-. -. 200.400)
1500
1000
o~======~~==========~======~==~======~
Magnesiumas Mg SP; 241:2005(-.-, 70.100)
Calciums Ca - SANS241:2005(-, -,150,300)
500
400
300
200
100
2006 2008 2010
Time
Figure 88 : Time graph of the Na, Mg and Ca over time,
The influence of flooding on underground coal mines Page 108
4.14.3 Recharge
Recharge rate which depends on the geology, the depth, the type of mining and whether
some form of high-extraction was performed. Stooping has been done in some parts of
the B-seam at Usutu colliery. Stooping, the process of removing pillars from the workings
causes the overlying strata to collapse. Cracks will develop. If these cracks intersect
with the surface rainfall water, and run-off may flow down directly into the mining cavity.
From studies done at collieries where stooping was done (Hodgson et al., 2007), it was
found that normally a recharge rate of 4 - 12% should be applied, depending on depth of
mining and ratio of stooping. The Newcastle area receives, according to Weather SA, an
average rainfall of 864 mm per annum
Table 19: Rainfall for the Kilbarchan underground area (excluding stooping and ash
filling).
Area Area Rainfall Rainfall
[m2] [m/a] [m3/a]
Underground 10424711 0.864 9006950
Table 20: Rainfall for the stooped areas at Kilbarchan.
Area Area Rainfall Rainfall
[m2] [m/a] [m3/a]
Stooped areas 1183775 0.864 1022782
Table 21: Rainfall for the opencast areas.
Opencasts Area Rainfall Rainfall
[m2] [m/a] [m3/a]
1A+IB+ Area 2 319582 0.864 276119
Table 22: Recharge table for Total Kilbarchan.
Total Rainfall: Underground (B&P) Stooped areas Opencasts Total
(m3/a) 9006950 1022782 276119 10305851
Total discharge: Total decant Mobile pump Transfer Pipelines Total
(m3/a) 110376 300000 750000 1160376
Recharge% 11.3
Show the recharge percentage taking into consideration the stooped areas, underground
areas and opencasts. The value of the recharge is compared to the output of the system
(decant, mobile pump and transfer pipelines).
The influence of flooding on underground coal mines Page 109
A total of 10305851 m3 of rainfall is deposited onto the mining area of Kilbarchan
annually. Total extraction (decant plus pumping) of the colliery is 116376 m3/a of water
(or 3200 rrr'/d), The mobile pump pumps out 300000m3/a water per annum on average.
The transfer pipelines pumps 750000 m3/a water to Roy Point, and decant is estimated
(from the V-notch flow gauge) as 110376 m3/a (Vermeulen, P.D, 2011).
Recharge (%) = (Decant + Pumping)/Rainfall
The final conclusion is a recharge of 11.3 % in Kilbarchan.
Assumptions made when calculating the recharge is a closed system. No outside
influences other than stated.
Significant external sources of mine water do not exist at the Kilbarchan Colliery. Water
levels in Natal Cambrian Colliery, south of Kilbarchan, are 7 - 9 m below the level in
Kilbarchan and can therefore not impact Kilbarchan (Hodgson, 2006).
A few scenarios in Table 23 were set up in changing the respective percentage of
recharge at opencast and stooped areas. The effect on the water balance can be seen in
the following tables. The influence different scenarios have on the underground is
dependent on the size of the opencast and stooped areas. In this case stooped areas will
have the biggest influence on the recharge in the underground followed by the opencast
areas. The main purpose of the scenarios is to show how the recharge percentage is
influenced by the different areas. Stooping, opencast en underground plays a vital role in
understanding the recharge percentage.
Table 23: Different scenarios regarding recharge percentages.
Recharge Underground Stooped Opencast
[%] [%] [%]
10.24 12 10
10.1 12 15.00
9.95 12 20.00
Recharge Underground Stooped Opencast
[%] [%] [%]
10.778 4.00 20
10.36 8.00 20
9.95 12.00 20
Recharge Underground Stooped Opencast
[%] [%] [%]
10.912 4.00 15
10.496 8.00 15
10.09 12.00 15
The influence of flooding on underground coal mines Page 110
4.15 Chemical analyses
4.15.1 Chemistry of the bore holes
All three boreholes were profiled for EG. The profiling is done with an TLG meter
(Temperature, water level and conductivity) the meter consists of a probe at the end the
tape. This is lowered into the boreholes and readings are taken every 5m. This way the
bore can be profiled with EG values.
• KW1/98 clearly illustrates the polluted water in the mining cavity compared with
higher up in the water column. At about 42 m the EG value increases rapidly
(Figure 89).
• KW1/99 also differs in depth, but there is a slight change halfway down the water
column. At 19 m the EG increases, indicating the mining cavity (Figure 90).
• VOID BH is not drilled into the mine, but increases in EG at 22 m and then returns
at 28 m. This may indicate a fracture where the water is seeping through into the
borehole (Figure 91).
Borehole Log - KW1/98(T)
Depth (m( Locality- X: 93671.00 Y: 3084170.00 Z: 1211.00
'" ",.,20 "" 0
25
30
35
40
)
45 ~
Figure 89: EG profiling for KW1/98.
The influence of flooding on underground coal mines Page 111
Borehole Log - KW1/99(T)
Depth {ml Locality - X: 94359.00 Y: 3083504.00 Z: 1197.50
r\o'~---.----~----~----~~'r·----~--~~----.---~--~
10
11
12
13
14
15
16
17
18
19
20
21
22
Figure 90: EC profiling for KW1/99.
Borehole Log - VOIDBH(T)
Ilepth[m[ locality- X: ;.4282.00 Y: 3083OQ1 00 1:. 1206.14
T....
16 ".
17
16
,g
20
21
22
23
24
25
26
27
28
Figure 91: EC profiling for VOID BH.
The table below shows the chemistry data for the boreholes last measured in April 2011
(Table 24). All these boreholes were sampled at the top of the water column. During the
last visit to the mine the three monitoring boreholes were sampled at the top as well as at
the bottom in the mine.
The influence of flooding on underground coal mines Page 112
The criteria used for inorganic sampling is the SANS 241 :2006, and for organic analysis
the USEPA Standards. Inorganic water samples are classified as:
• Class I - acceptable (not color coded)
• Class II - allowable (color coded
• Above - not allowable (color coded red)
See appendix B for class ranges.
Table 24: Results of the chemical analyses for the boreholes sampled during the last site
visit April 2011.
From this table it is clear that there is a definite difference in the water quality of the top
and bottom samples of those boreholes drilled into the mining cavity.
• KW1/99 differs from top to bottom. The chemistry of the top sampled water shows
no problem according to the standards, but the bottom sample chemistry shows
high electrical conductivity (EC) together with high sulphate. High levels of
magnesium and sodium typical of coalmine waters were also found.
• KW1/98 was the same as KW1/99, with the bottom differing from the top, with
exactly the same result with regard to the chemistry.
• VOID BH is not drilled into the mine. We have no other chemistry data on this
borehole. It seems that the water is already acidic with a pH of 2.84. There are
very high sulphate values together with high levels of the metals (AI, Fe and Mn),
as well as high values of calcium, magnesium and sodium.
The expanded Durov diagram (Figure 92) clearly indicates that VOID BH top and bottom
and KW1/99 top are mine water, while KW1/98 top is unpolluted. (See Appendix A for
The influence of flooding on underground coal mines Page 113
explanation of the Durov interpretation) KW1/98 bottom and KW1/99 bottom are coal
mine water.
Expanded DllOV Diagram
"11
Na+!(
I~
I~
IIWII~
IIWII~B)
OIWllIl8(1)
IIWIIIl8(B)
504
Cl
Figure 92: Expanded Durov diagram of the Kilbarchan boreholes.
The Stiff diagrams of the boreholes (Figure 93) (which compare the water qualities in
equivalents) also illustrate that the bottom samples are more polluted than the top
samples. This is clearly visible in the Stiff diagrams, but to a lesser extent in the VOID
BH. The VOID BH seems to be the most polluted BH, as the stiff diagram shows with high
sulphate values. The stiff diagrams show that the bottom samples show evidence of mine
water mine while the top sampled samples are clean.
The influence of flooding on underground coal mines Page 114
STIFF Diagrams
KW1/98(B) KW1/98(T) KW1/99(B)
20110lOS ·131'05 2()110509·131'O5 20110lOS ·131'05
Na+!( a Na+!( a Na+!( a
Ca Alk Ol Alk Ca Alk
Mg 504 Mg 504 Mg 504
75 meQA 75 75 meQA 75 75 meQA 75
KW1/99(T) VOIDBH(B) VOIDB~T)
20110lOS ·131'05 2()110509· 131'05 20110lOS ·131'05
Na+!( a Na+!( a
Ca Alk o Alk
Mg 504 Mg 504
75 meQA 75 75 meQA 75 75 meQA 75
Figure 93: Stiff diagrams of the mine boreholes.
The time graph of the mine boreholes show that the water quality for the top boreholes
did not deteriorate since 1999, except the VOID BH. As already stated, the bottom
samples show high EC values (Figure 94). The high EC values indicate the water in the
mine is still polluted compared to the top sampled.
The criteria used for inorganic sampling is the SANS 241 :2006, and for organic analysis
the USEPA Standards. Inorganic water samples are classified as:
• Class I - acceptable (not color coded)
• Class II - allowable (color coded
• Above - not allowable (color coded red)
See appendix B for class ranges.
The influence of flooding on underground coal mines Page 115
Electrical conductivity
EC SANS 2412005 (-, -,150,370)
BOO
•
600
•
· 4 KW1/96(B)-.- KW1/96(T)
-e- KW1/99(B)
· KW1/99(T)400 • • VOIDBt-t;B)
• VOIDBt-t;T)
•
200 ·h
.v._f#...• ........
~ •
...~W ...--- ,.-
o
1998 2002 2006 2010
Time
Figure 94: Time graph of the EC for the mine boreholes.
400
Magnesium as Mg - SANS 241 :2005 (-. -, 70. 100)
300
•
200 •.. ··~
100 •
_...,._,
o W ·
Caldum as Ca - SAr-s 241:2005 (-. -.150.300)
400 • ... KW1/9B(B)
... KW1/98(T)
300 .l , -+ KW1/99(B)KW1/99(T)200 ....- .. VOIDBH;B)-~U ,. • .... VOIDBH;T)100 ...,. A'
Sulphate as S04 .~ ANS 241:2005 (-.·.400.600)
4000
3000 •
2000
I
1000 ,1
o . .1:. :A'
'998 2002 2006 2010
Time
Figure 95: Time graph of sulphate, calcium and magnesium of the mine boreholes.
The influence of flooding on underground coal mines Page 116
600
Chlaideas Cl-SANS241:200(5-,-,200,600)
500
400 .
300
•
200
1\
100 JJ\. · -- KW1/98(B)-.. KW1198(T)
0 .... KW1/99(B)
KW1/99(T)
SodiumasNa - SANS241:200(5-,-,10,20) • VOIDBI-'(B)
• VOIDBI-'(T)
1000
••
500 ·
Time
Figure 96: Time graph of sodium and chloride of the mine boreholes.
pH
The pH of the top sampled boreholes (Figure 97) KW1/98 and KW1/99 is still acceptable,
but shows a slight decrease from 2006 till now, while it appears as if the VOID BH top
sample is already acidic with, a pH of 2.84.
pH
pH SANS 241 :2005 (4.0,50,9.5, 10)
10.0,------------------------------,
8.0
,._ KW1/9B(B)
I -- KW1191!(T)
-- KW1I9!i(B)
KW119!i(1)
6.0 • VOCBl-(B)
,._ VOCBl-{T)
2002 2006 2010
Time
Figure 97: Time graph of the pH of the mine boreholes.
The influence of flooding on underground coal mines Page 117
4.15.3 Opencast sampling points (Decant)
Water decants from position D1 (Figure 98). D2 ceased decanting since pumping started
a few years ago. Borehole 26 and 30 are also drilled into the opencast, but is not part of
the monitoring program. For this investigation they were profiled (Figure 99 and Figure
100). These profiles illustrate the polluted nature of these boreholes with BH26 indicating
an EC of 7000mS/m at 7 m depth and BH30 an EC of 3000 mS/m at a depth of 15 m.
-307800~--~----~----~----~----~----~----~----~----~~
-307900
,Slangdraai
-308000
'2opencast
-308100
-308200
Kilbarchan
-308300
BH30
-308400 Dump 01
Kilbarchan
Opencast
-308500 IR R I I-1000 0 1000 2000 3000 4000
90000 91000 92000 93000 94000 95000 96000 97000 98000
Figure 98: Opencast sampling points.
The influence of flooding on underground coal mines Page 118
Borehole Log - BH30
Depth (m( Locallty- X: 94795.00 Y: 3083962.00 Z:-1.00
10
15
20
Figure 99: EC profiling for BH 30.
Borehole Log - BH26
Depth (m) Loceltty- x: g4554.00 Y: 3084221.00 Z: ·1.00
EC
10
11
Figure 100: EC profiling for BH 26.
The water quality from the decant point as well as the water pumped with the transfer
pipelines (directly out of the mining void) is illustrated in Table 25
The influence of flooding on underground coal mines Page 119
These qualities indicate a very high sulphate (higher than the average of 2500mg/1 in
most of the mines in South Africa)and sodium character, with high calcium and
magnesium values, implying a very dynamic system with a serious ARD problem. No
buffer material (alkalinity) is present in the water from the opencast decant. The time
graphs (Figure 101) of EC, sulphate, calcium and magnesium indicates that the water did
not improve or deteriorate since 2004.
Table 25: Water quality of the decant point and the transfer lines.
8000
eeeo
I
4000
2000
'000 •
500
0_
Time
Figure 101: Time graph of EC, sulphate calcium and magnesium for the decant point and
the transfer lines.
The influence of flooding on underground coal mines Page 120
The Stiff diagrams in Figure 102 clearly illustrate the sodium sulphate character of the
water originating from the opencast (D1) and the underground at Kilbarchan (transfer
pipe lines) show mine water where sulphate has formed in the mine due to oxygen inside
the mine.
STIFF Diagrams
D1 KILBARCHAN TRANSFER PIPELINE 1
20110811 - OOttlO 20110509 - 13h05
Na+K Na+K Cl
Ca Ca Alk
504
100 meql1 100 100 meql1 100
KILBARCHAN TRANSFER PIPELINE 2
20110509 -13h05
Na+K Cl
Ca Alk
504
100 meq/l 100
Figure 102: Stiff diagrams of the decant and transfer lines sampling points.
4.15.2 Chemistry of the surface water
In Figure 103 the position of all the surface sampling points sampled on a regular basis,
together with shallow boreholes around the discard dump, is illustrated. In Figure 103 one
of these discard boreholes is displayed.
The influence of flooding on underground coal mines Page 121
Figure 103: Photograph of KMH3.
-307600
-307800
-308000
QUARRY S EAM EXTENSION U/S •
·KMH7
-308200
Kilbarchan
AREA1 B
-308400 LEGEND
Kilbarchan
-308600 IH 81 I I I
-1000 0 1000 2000 3000 4000
90000 92000 94000 96000 98000
Figure 104: Position of the samples of the surface water.
The chemistry of the surface water samples and the boreholes monitoring the dump is
shown in Table 26
The influence of flooding on underground coal mines Page 122
The criteria used for inorganic sampling is the SANS 241 :2006, and for organic analysis
the USEPA Standards. Inorganic water samples are classified as:
• Class I - acceptable (calor coded green)
• Class II - allowable (calor coded vst
• Above - not allowable (calor coded red)
Table 26: Results of the chemical analyses of the surface water sampled during the last
site visit April 2011.
• From this data it is clear that the discard dump has an influence on the
surrounding groundwater quality.
• The Greenwich stream extension downstream from the decant point
indicates pollution. This is a result of the water decanting from the
opencast.
• It is not clear if the affected water at the Village stream V-notch is a result
of water flowing from the mine, as this is also down-gradient from the
Ngagane power station.
The influence of flooding on underground coal mines Page 123
The proportional size distribution of the electrical conductivity values of all these samples
are displayed in Figure 105 the boreholes shown with circles and the surface water with
squares, and the Stiff diagrams of the last values in Figure 106. The diagram of the
discard boreholes clearly indicates the sodium-sulphate character of these waters, with
elevated calcium and magnesium as well.
-3075000
-3077500
-3080000
-3082500
-3085000 LEGEND
I<Ilb
8",,",,,
Op.1"IC!S
fI.O\'PoIn I B B I I I I
·7:50 0 7:50 1:500 2250 3000
90000 92500 95000 97500
Figure 105: Proportional distribution of EC values of the surface samples and discard
boreholes.
The influence of flooding on underground coal mines Page 124
STIFF Diagrams
GOLFDAMDtS GREANWlCH STREAM AREA 1 B GREANWlCH UtS OF DUM' INGAGANE RI'v£R utS
2011C!i09- 13"«; 20110406 - 1:t66 20010104- (X)1OO 2011Cfj()9· 1:ta)
Na+K a Na+K Cl Na+K a Na+K a
Ca Alk Ca Alk Ca Alk Ca Alk
Mg SOl Mg SOl Mg SOl Mg SOl
100 meq/l 100 100 rneq/l 100 100 meqn 100 100 meq/l 100
KMH1 KMH2 KMH3 KMH7
2011000g· 13115 201105a> - 1:ta; 2011CM9- 1:ta; 2011Cfj()9-1:1'1OO
Na+K a Na+K Cl Na+K a Na+K
Ca l Alk Ca Alk Ca Alk CaMg SOl Mg ~ SOl Mg ~ SOl Mg ~
100 meq/l 100 100 meqn 100 100 meqn 100 100 meq/l 100
UARRY STREAM EXTENSION UtS VILLAGE STREAM V-NOTCH
2011C!i09- 1:ta; 2011Q5aJ - 1:J-a;
Na+K a Na+K Cl
Ca Alk Ca Alk
Mg SOl Mg t SOl
100 meq/l 100 100 meqn 100
Figure 106: Stiff diagrams of the surface water and the discard dump,
The time graph of the surface water samples indicates that the two elevated samples
discussed above (Greenwich stream extension and Village stream V-notch) have very
erratic values and are definitely influenced by surface dilution (e.q. rainfall events), but
the overall trend indicates the elevated values (S04) in relation to the other surface
samples,
The EC values of the surface samples are in the standards for water quality show in
Figure 107_ The time series show that the surface waters are not polluted. The manholes
around the discard dump shows EC values shown in Figure 108 that are not acceptable.
The influence of flooding on underground coal mines Page 125
Electrical conductivity
EC SANS 241:2005 (-, -, 150, 370)
1000,----------------------------------------------------------------,
.. OOFOIMDS
... <J£IM\IOismE.'M1fEA 1
+ <J£IM\IOi USCl' D.M'
INCl'.ClOI£~DS
• 1~R\ffiUS
.. "'_"sr~EXT9S
... "u..cEsmE.'M v-NOTOi
2013
1irre
Figure 107: Time graph of the EC for the surface samples.
Electrical conductivity
EC SANS 241:200(5-.-,150.370)
700
•
600
500
.. KIVH1
400 ... KIVH2
... K1VH3
KIVH7
300
200
100
0
1997 2001 2005
Time
Figure 108: Time graph of the EC for the discard dump boreholes.
The influence of flooding on underground coal mines Page 126
4.16 Conclusions and recommendations
The following is concluded from this study:
• The fact that the water level elevations of the different boreholes are all in the
same order indicate that the water level is flat throughout the underground and
connected opencast mine. Due to decant through the opencast, these mine water
levels will not rise in the underground mine above the decant level. The variation
in water level elevation is due to recharge and the resulted pumping from the
underground, resulting in the 3 m variation in water level.
• Water is being pumped on a regular basis from Kilbarchan to Roy Point, and also
controlled release pumping at the opencast.
• The mobile pump pumps out approximately 300000m3/a of water per annum on
average and the transfer pipelines pumps approx. 750000m3/a. Decant of the
opencast pits is at approximately 110000m3/a of water.
• Taking the annual rainfall into account, the conclusion is that recharge of 11.3 %
occurs into Kilbarchan.
• It is clear that there is a definite difference in water quality of the top and bottom
samples of those boreholes drilled into the mining cavity, indicating stratification.
• VOID BH is drilled into the opencast void. The water is already acidic with a high
sulphate value this results in a pH of 2.84, most probably due to the flow of water
through the void, ingressing oxygen. High sulphate values together with high
levels of AI, Fe and Mn, as well as calcium, magnesium and sodium are present.
The water quality for the top boreholes did not deteriorate since 1999, except the
VOID BH. The bottom samples show high EC values. This indicates that
sampling is done at the incorrect depth, or that sampling should be done at
least at different depths to have a clear indication of the mine water quality
as well.
• Water decants from the opencast. Two boreholes BH26 and BH30 are also drilled
into the opencast, but is not part of the monitoring program. Profiling illustrates the
polluted nature of these boreholes with BH26 indicating an EC of 7000mS/m at 7
m depth and BH30 an EC of 3000 mS/m at a depth of 15 m.
The influence off/ooding on underground coal mines Page 127
• The water quality decanting indicates a very high sulphate (higher than the
average of 2500mg/1 in most of the mines in South Africa)and sodium character,
with high calcium and magnesium values, implying a very dynamic system. No
buffer material (alkalinity) is present in the water from the opencast decant.
• The discard dump has an influence on the surrounding groundwater quality. The
Greenwich stream extension downstream from the decant point indicates
pollution. This is a result of the water decanting from the opencast. It is not clear if
the affected water at the Village stream V-notch is a result of water flowing from
the mine, as this is also down-gradient from the Ngagane power station.
Greenwich stream extension and Village stream V-notch have very high S04
values and are definitely influenced by surface dilution (e.g. rainfall events), but
the overall trend indicates the elevated values in relation to the other surface
samples.
Recommendations
• It is recommended that the monitoring programme in the mine be executed
properly with water being sampled in the mining cavity and higher up in the water
column every quarter that is 4 times a year of all the Kilbarchan sites. More
boreholes into the mining cavity will also contribute to a better understanding of
the current water volumes. As only three boreholes are currently used for
monitoring
• The water balance of the system needs be better understood to reduce the risk of
oxygen flushing through the system. The water balanced needs to be balanced.
•
The influence of flooding on underground coal mines Page 128
CHAPTER 5
CONCEPTUAL COMPARISON
5.1 Introduction
This chapter will focus on the differences between the two different collieries, Usutu and
Kilbarchan. A comparison of the two studies will give a good indication of the influence of
flooding at the different sites. This will show whether the depth of mining, topography, mining
methods, water levels, exposure to oxygen, rainfall, recharge, residence time and pumping
influence the effects of flooding at coal mines. The conclusion is that these factors do play a
vital role in the management of flooded coal mines. There is a difference between the shallow
underground mine and the deep underground mine. These differences will be discussed in the
sections to follow.
The aim is to know what to expect when managing a deep underground or shallow underground
colliery which have been flooded. Both mines have been flooded for more than 20 years. The
effect of flooding is influenced by a number of factors
The differences between the two are with regard to:
• Water quality
• Topography
• Depth of mining
• Mining methods
• Rainfall
• Exposure to oxygen
• Residence time
• Decanting
• Recharge
• Pumping
• Management
Comparisons between the two studies include:
• Geology
The influence of flooding on underground coal mines Page 129
• Mineralogy
• High extraction
• Coalmining
• Both fully flooded with water
• Production stopped
• Underground
5.2 Mining area
In general an underground coalmine that is mined by bord-and-pillar methods at a depth of 100
m below surface can be flooded with water after mining operations have ceased, without any
danger to the adjacent environment. The mine is too deep to allow the ingress of oxygen into
the workings and also too deep to allow water to flow from the mine after it has filled up (Zeelie
& Hodgson, n.d).
The mining area between the two studies differs. At Kilbarchan very shallow mining took place
in the mountains. Water has a shorter residence time in the system as the gradient is steeper.
At Usutu the area is flatter and much deeper, water stays in the system much longer and the
sensitivity to pollution is greater than at Kilbarchan.
The topography of Kilbarchan mine are more vivid than the flat surfaces at Usutu mine. At Usutu
the topography is of a gentle rolling nature. Steeper slopes are present at sandstone outcrops.
Significant topographic differences are present in the KwaZulu-Natal Coal-fields.
In contrast, in the case of the high underground extraction of coal, numerous cracks are created
from the mining level all the way to the surface. Ingress of fresh water containing oxygen and
also ingress of air are possible. The probability of the mine water becoming acidic over a period
of time is therefore significantly greater for high underground extraction mining than for bord-
and-pillar mining (Zeelie & Hodgson, n.d).
5.3 Depth of mining
Usutu mine is a deeper mine than Kilbarchan mine. The depth of water flow plays a role in the
residence time of water. Therefore Kilbarchan mine will have a short residence time and a
greater risk of pollution. This is one of the main reasons why the water quality at Kilbarchan is
more polluted.
The influence of flooding on underground coal mines Page 130
The rainfall of the two areas differs. There is about a 100 mm difference in the annual
precipitation. Recharge will be greater at the shallow Kilbarchan mine 11.3 %, but by managing
the releases the water balance could be solved. It is important to keep the mine flooded to stop
oxygen interacting with water and sulphur rich minerals (pyrite) to cause acid mine drainage
(AMD), but also to avoid decants. This is done by the water level remaining above the mining
cavity.
Mining is taking place at a shallower depth at Kilbarchan. Water from rainfall is recharged and
moves into the surface; this water is rich in oxygen. At Usutu mine the mining is deep and the
oxygen-rich rainwater takes a long time to reach this level. Recharge differs significantly in both
studies. The deeper Usutu mine is calculated at 5.7% and the shallower Kilbarchan mine at
11.3%. Shallow mines seem to be more susceptible to pollution than deeper mines. Based on
the evidence that the mine is shallower, topographically more up and down and a shorter
residence time.
5.4 Water levels
There are big differences in depth of the water levels at both areas. At Usutu mine the reason
for the difference in water levels was the ventilation seals in the underground. On old maps
ventilation walls show that the mine has been compartmentalized; water becomes trapped
inside a compartment, in a sense creating its own water level. The measured water levels differ
from one another. A different problem was encountered at Kilbarchan. The water levels also
differed in depth, but the water level elevation are the same. It means that the overlying aquifer
is not fully recharged as yet, because of decant. Mining is taking place at a hilly area very
shallow, together with opencast areas. As the mine fills up with water the unconfined water level
and the mine water level is moving towards each other. Decant then takes place before the
aquifer gets fully recharged. Interpreting these water levels is a very difficult exercise. Water will
sometimes decant at a borehole, caused by a break in these seals at Usutu. This is interpreted
as the water level rising, which is not really the case. Finally it is concluded that when looking at
water levels at flooded mines, water level elevation should be used rather than water level depth
as it can be misleading.
5.5 Rainfall
Rainfall causes an increase in the recharge at a mine site. Kilbarchan is located in a higher
rainfall area than Usutu. Kilbarchan is located in the KwaZulu-Natal province and Usutu in the
The influence of flooding on underground coal mines Page 131
province of Mpumalanga. Kilbarchan - 864 mm/a and Usutu - 710 mm/a. The rainfall at the
different sites show that kilbarchan has an 20% higher rainfall than Usutu.
5.6 Recharge
The tempo at which this happens is defined by the recharge rate which depends on the geology,
the depth, the type of mining and whether some form of high-extraction was performed. Under
undisturbed conditions, 3% of the annual recharge would be an acceptable average value.
Taking the annual rainfall into account, the conclusion is that recharge of 11.3 % occurs into
Kilbarchan and 5.7 % at Usutu colliery. This difference in recharge is due to the opencast areas
and stooped areas. At Kilbarchan colliery opencast mining took place, because of the shallow
depth of the coal seams. The recharge at opencast areas is between 15-20 % and at stooped
areas between 8-12 %.The depths of mining also influence the recharge, recharge is calculated
easier at the shallow Kilbarchan mine than at Deeper Usutu. This changes the recharge in the
underground as it influenced by these activities.
This work compared well with work previously done in this field of study in a WRC report.
A summary of the percentage influx to be expected for the various mining methods, as
determined during this investigation, is as follows done by Hodgson & Vermeulen (2007):
• Deep bord-and-pillar with no subsidence 1-4% of the rainfall.
• Shallow bord-and-pillar 3 -10% of the rainfall.
• Stooping 4 -12% of the rainfall.
• Long wall 6 -15% of the rainfall.
• Opencast 14 - 20% of the rainfall
5.7 Pumping
Pumping is probably the main difference between these two sites. If the pumping is not
managed properly, the mine water quality will not improve. This is clearly seen in the water level
elevation, where water is being pumped more than is necessary for long periods at Kilbarchan.
The shallowness of the mine causes oxygen to infiltrate into the underground workings as the
water level drops from pumping. The EC over time at Kilbarchan is not improving because of the
excessive pumping.
Pumping also takes place at Usutu, but it is managed well. The mine is kept full and the water
levels stay steady over time. This prevents oxygen from infiltrating into the underground. The
The influence of flooding on underground coal mines Page 132
EC over time has improved. Finally, in pumping a flooded mine management is vitally important.
Care should be given at shallow underground mines to an understanding of the water balance.
Flushing is an accepted practice to improve the quality of mine water if the pumping rate (and
thus the water levels) is managed properly, keeping the water level above the mining cavity. If
the mine stays flooded then no additional oxygen ingress and pyrite oxidation will occur. This is
seen at the pumping station BH at Usutu. A large proportion of the collieries will eventually
decant. The flushing of mines is inevitable once they are flooded. This could improve the water
quality in the mine, but reduces the available base potential of the mine as a whole. Plans to
manage this must be in place.
5.8 Decant
Decant is taking place at the shallow underground Kilbarchan colliery. The water is decanting
from opencasts at 2.5 I/s. The EC of the decant water over time is not improving, measured at
637 mg/I. Decant then takes place in the opencast before the aquifer gets fully recharged,
because of the gradient at which mining is taking place. This is an important concept in
understanding the reason for decant. At the deep underground mine Usutu no decanting is
taking place. The area is flat and the water levels are monitored to prevent the water level to
drop under the mining cavity, thus preventing oxygen from moving into the system.
5.9 Sampling
Sampling is such an important management tool at flooded coal mines. Sampling at the different
sites was done at the incorrectly depths, it should be done at different depths to have a clear
indication of the mine water quality. The water needs to be sampled at the bottom of the
boreholes sampling inside the mine, sampling was done at the top of the boreholes. This is not
a representable sample of the mine water At Usutu the water were sampled just beneath the
water level, results of the hydrocensus concluded that the water quality sampled was exactly the
same as the farmer's boreholes around the mine. This means that the water of the regional
aquifer was sampled and not the actual mine water inside the mine. The sampling for this study
was done at the top and the bottom of the boreholes inside the mining cavity. Results indicated
a substantial difference in qualities at depth. If you can't measure it you can't manage it.
5.10 Residence time
Groundwater sensitivity to pollution is best understood in relation to travel time, which is the
approximate time that elapses from when a drop of water infiltrates the land surface until it
enters an aquifer or reaches a specific target such as a spring. This is also called residence
The influence of flooding on underground coal mines Page 133
time. The pollution sensitivity of an aquifer is assumed to be inversely proportional to the time of
travel. In addition, contaminants are assumed to travel at the same rate as water. Very high
sensitivity indicates that water moving downward from the surface may reach the groundwater
system within hours to months. In these areas, there is little time to respond to and prevent
aquifer contamination. Conversely, low sensitivity indicates there is time for a surface
contamination source to be investigated, and possibly corrected, before serious ground-water
pollution develops (Alexander &Alexander, 1989)
The residence time of the two sites differs. Kilbarchan has a short residence time, because of
the shallow depth and excessive pumping. This makes it more sensitive to contamination. On
the other hand, Usutu has a longer residence time, because of the deeper depth and controlled
pumping. The sensitivity to pollution is less the longer the residence time. The opencast area 1A
and 1B that is connected to the underground causes more oxygen rich rainwater to enter into
the workings
The residence time is influenced by the depth of mining:
• The shallower, the shorter the residence time and the more chance for a poor water
quality. Higher sensitivity.
• The deeper, the longer the residence time and a better chance for good water quality.
Lower sensitivity.
5.11 Water quality
The S04 values / sulphate generation comparison compared shows a very large difference in
the values. The S04 values at Usutu range between 300 and 900 mg/I, while at Kilbarchan the
range is between 1500 and 3000 mg/1.The reason for this is that Kilbarchan is more exposed to
oxygen, because of the shallow depth and excessive uncontrolled pumping. Sulphate
generation occurs when sulphate minerals (pyrite) combine with water and oxygen. At Usutu
VOID BH is drilled into the opencast void. The water is already acidic with a pH of 2.84, most
probably due to the flow of water through the void, ingressing oxygen. This result in very high
sulphate values together with high levels of AI, Fe and Mn, as well as calcium, magnesium and
sodium are present. The two Durov figures clearly show the differences in water quality between
the two sites. Firstly Kilbarchan samples plot more in the 4 and 5 fields. The Usutu samples
plots in the 2 and 3 fields (See Appendix). All the samples that have data were used to draw up
an Expanded Durov diagram. This can also be used to see the differences in water quality.
The influence of flooding on underground coal mines Page 134
5.12 Recommendations
The following actions are recommended:
• Much can be done to improve mine-water planning at all collieries. It is essential that as
much mine water as possible be kept underground during and after mining. Mining
should be done from low-lying areas to higher ground.
• Compartments should be created that could fill up with water while mining continues in
other sections of the mine. This should ensure the minimization of water treatment.
• The recommendation from these results is to flood the mines as quickly as possible. This
can be achieved by the selection of mining method, the mining sequence and flooding.
Mining methods and mining geometry play a decisive role in the amount of water to be
managed during and after mining.
• Mine scheduling is one of the most important water management tools to be considered
from the outset of mining, throughout the life of the mine. The filling up of mines with
water is inevitable after closure. Seepage of water away from the mines will occur once
they approach the full mark.
• EC profiles should be done at depth, to see whether stratification keeps on occurring.
Sampling for water quality at the "top" and "bottom" of a borehole. This means sampling
just below the water level and in the mining cavity. This will give us a better
understanding of the qualities inside the mine itself.
• Water pumped from the Kilbarchan mine should be treated. The water quality of the
water pumped to the transfer pipelines at Roy point is below standard.
• Water levels, pumping yields, discharge, rainfall and chemistries should be monitored as
frequently as possible. Weekly water levels and quarterly water samples should be
taken. Anomalies should be identified and checked as soon as they occur. The
information should be interpreted on an ongoing basis.
The influence of flooding on underground coal mines Page 135
BIBLIOGRAPHY
Alexander, S.C., and Alexander, E.C., Jr. (1989). Residence times of Minnesota ground waters:
Minnesota Academy of Sciences Journal, 55(1 ):48-52.
Aryafar, A, Doulati, F, Ardejani, Singh, Rand Shokri, B.J. (2007). Prediction of groundwater
inflow and height of the seepage face in the deep mine pit using numerical finite element model
and analytical solutions. In: IMWA Symposium 2007: Water in Mining, 27th - 31st May 2007,
Cagliari, Italy.
Balkau, F. (1999). Abandoned mine sites In: Natural Resources Management Unit Division,
Protection of the environment and Natural Resources: Report on the International Round Table
on Mining and the Environment (Berlin 11 Declaration). pp 40-43; Kohn (Carl Duisberg,
GeselIschaft).
Bedient, P.B., Rifai, H.S. & NewelI, C.J. (1994). Groundwater Contamination; Transport and
Remediation. 541 pp, Prentice Hall, New Jersey. Cited in Minerals Council of Australia (1997).
Chapter 8: Open Cut Mines, In: Mine site Water Management Handbook, edited by Minerals
Council of Australia, First Edition.
Brawne, C.O. (1982). Control of groundwater in surface mining. International Journal of Mine
Water, 1:1-16.
Carter, P. (1992). Chapter 11 Dewatering. In: Drainage design, by Smart, P. and Herbertson,
J.G, http://bestcoaltrad ing.blogspot.com/20 10/09/mine-dewatering. html (accessed September
2010).
Chen, M, Soulsby, C and Younger, P, L. (1999). Modeling the evolution of mine water pollution
at Polkemmet Colliery, Almond catchment, Scotland, The Ouarterly Journal of Engineering
Geology, 32(4):351-362. Cited in: J.J Donovan, B R Leavitt and E Werner. (2003). Long-Term
Changes in Water Chemistry as a Result of Mine Flooding in Closed Mines of the Pittsburgh
Coal Basin, USA In: 6th ICARD, Cairns, OLD, 12 -18 July 2003, pp 869-874.
Connelly, R.J, Gibson, R. (1985). Dewatering of The Open Pits at Lethakane and Orapa
Diamond Mines, Botswana. International Journal of mine water, 4:pp 25-41, Madrid, Spain.
The influence of flooding on underground coal mines Page 138
5.13 Comparative flow diagram
Recharge 5.7 % Recharge 11.3 %
D Uneventopography
pography Deep Underground Both flooded for Shallow Underground I'"
more than 20 years
Chemistry
J 1 j \ 1
Moderate levels of
EC value S04 values
pumping; well between 300- Last EC value High levels ofsured at 5 S04 values
managed; low
mg/l 900 mg!l; long
measured at pumping, poorly
oxygen ingress rp.sirlp.nr:p. timp. 637 mg!1 managed; high
between 1500-
oxygen ingress 3000 mg!l; short
Water that is pumped Better sampling
should be treated; more procedures should
Figure 109: Comparative flow diagram. controlled pumping be set in place
should take place
The influence of flooding on underground coal mines Page 137
BIBLIOGRAPHY
Alexander, S.C., and Alexander, E.C., Jr. (1989). Residence times of Minnesota ground waters:
Minnesota Academy of Sciences Journal, 55(1 ):48-52.
Aryafar, A, Doulati, F, Ardejani, Singh, Rand Shokri, B.J. (2007). Prediction of groundwater
inflow and height of the seepage face in the deep mine pit using numerical finite element model
and analytical solutions. In: IMWA Symposium 2007: Water in Mining, 27th - 31st May 2007,
Cagliari, Italy.
Balkau, F. (1999). Abandoned mine sites In: Natural Resources Management Unit Division,
Protection of the environment and Natural Resources: Report on the International Round Table
on Mining and the Environment (Berlin 11 Declaration). pp 40-43; Kohn (Carl Duisberg,
Gesellschaft).
Bedient, P.B., Rifai, H.S. & NewelI, C.J. (1994). Groundwater Contamination; Transport and
Remediation.541 pp, Prentice Hall, New Jersey. Cited in Minerals Council of Australia (1997).
Chapter 8: Open Cut Mines, In: Mine site Water Management Handbook, edited by Minerals
Council of Australia, First Edition.
Brawne, C.O. (1982). Control of groundwater in surface mining. International Journal of Mine
Water, 1:1-16.
Carter, P. (1992). Chapter 11 Dewatering. In: Drainage design, by Smart, P. and Herbertson,
J.G, http://bestcoaltrading.blogspot.com/2010/09/mine-dewatering.html(accessed September
2010).
Chen, M, Soulsby, C and Younger, P, L. (1999). Modeling the evolution of mine water pollution
at Polkemmet Colliery, Almond catchment, Scotland, The Ouarterly Journal of Engineering
Geology, 32(4):351-362. Cited in: J.J Donovan, B R leavitt and E Werner. (2003). long-Term
Changes in Water Chemistry as a Result of Mine Flooding in Closed Mines of the Pittsburgh
Coal Basin, USA In: 6th ICARD, Cairns, OLD, 12 - 18 July 2003, pp 869-874.
Connelly, R.J, Gibson, R. (1985). Dewatering of The Open Pits at lethakane and Orapa
Diamond Mines, Botswana. International Journal of mine water, 4:pp 25-41, Madrid, Spain.
The influence of flooding on underground coal mines Page 138
Connelly, R.J. (1982). A Case History of The Open Pits mine dewatering at Letlhakana Mines in
Groundwater 82. Groundwater division of the Geological Society of South Africa, University of
the Witwatersrand.
Donovan, J.J, B. Leavitt, Rand Werner, E. (2003). Long-Term Changes in Water Chemistry as
a Result of Mine Flooding in Closed Mines of the Pittsburgh Coal Basin, USA In: 6th ICARD,
Cairns, QLD, 12 - 18 July 2003, pp 869-874.
Fuji, S. (2002), Modern Trends in Computer-Integrated Technologies in Minerals Industry, IV
International Symposium on the Occasion of the 40 Years of Ex is tence of the Department for
Machinery in Mining, Proceedings, Mechanization and Automation in Mining and Energy
Sources, Faculty of Mining and Geology, University of Belgrade, Belgrade, 2002.
Fuji, S. et al (2004), Adapt able Computer-Supported Monitoring-Management Model of the
System for Dewatering of the "Drmno" Coal Open Pit Mine (Phase I) (in Serbian), Project 102-
154A, Internal Report, Faculty of Mining and Geology, University of Belgrade, Belgrade, Serbia
and Montenegro.
Grobbelaar, R. Usher, B, Cruywagen, L-M, de Necker, E. Hodgson, F.D.1. (2004). Long Term
Impact of Intermine Flow from collieries in the Mpumalanga coalfields, WRC Report No
1056/1/04. Institute for Groundwater Studies, University of the Orange Free State,
Bloemfontein.
Hodgson, F.D.I, Vermeulen, P.D, Cruywagen, L-M and de Necker, E. (2007). Investigation of
water decant from the underground collieries in Mpumalanga WRC Report No: 1263/1/07,
University of the Free State, Bloemfontein.
Hodgson, F.D.1. (2006). Kilbarchan Colliery: Update of the Water Balance for Kilbarchan and
Roypoint Sections, Institute for Groundwater Studies, Report no: 2006/02/Kilbarchan, University
of the Free State, Bloemfontein.
Hodgson, F.D.1. and Krantz, R.M. (1998). Groundwater Quality Deterioration in the Olifants
River Catchment above the Loskop Dam with Specialized Investigation in the Witbank Dam
Sub-Catchment. Report to the WRC, Report291/1/98, Water Research Commission, Pretoria.
Hodgson, F.D.I., Cruywagen, L-M and de Necker, E. (1997). Chapter 4 Coal mining in outcrop
areas of KwaZulu-Natal. In: Coal Mining in the Vryheid Area - Geohydrology and
Hydrochemistry, Institute for Groundwater Studies, University of the Free State, Bloemfontein.
The influence of flooding on underground coal mines Page 139
Hoek, E. and Pentz, DL (1968). The Stability of Open Pit Mines Two-dimensional Flow Net
IIlustrfftiog Seepage Forces in a Slope, Imperial College Rock Mechanics, Research Report no
5. Cited in: Brawne. c.a. (1982). Control of groundwater in surface mining. International
Journal of Mine Water, 1: 1-16.
Hustrulid, W. A. (2000). Slope stability in surface mining. Hydrogeological model approach for
dewatering mining, SRK Consulting. Citing Internet Resources.
http://www.srk.com/files/File/papers/Numerical%20Groundwater10WIM_Chapter06_17.pdf.
Jeffrey L.S. (2005). Characterization of the Coal Resources of South Africa. Journal of the
South African Institute of Mining and Metallurgy, 95-102.
Libicki, J. (1993). Proposal of Criteria for The Selection of Dewatering Methods in Surface
Mining. International Mine Water Symposium, Zambia, pp53-332. Poltegor Engineering ltd.
Mccarthy T. S. and Pretoruis, K. (2009). Coal Mining on the Highveld and lts implications for
Future Water Quality in the Vaal River System, In: Abstracts of the International Mine Water
Conference, Pretoria, South Africa, 19 _ 23 October 2009, pp 56-65.
Minerals Council of Australia. (1997). Chapter 8: Open Cut Mines, In: Mine Site Water
Management Handbook, edited by Minerals Council of Australia.vol 1.
Morton, K.L and van Miekerk F.A. (1993). A Phase Approach to Mine Dewatering. Mine Water
and the Environment, 12: 27-34.
Morton, KL (2009). Comparison of Designs for the Dewatering of Coal, Gold and Diamonds.
KLM Consulting Services. Lanseria, Gauteng, South Africa.
Naise, E.J. (1993). Dewatering Concepts at Zambian Copper Belt Mines, Mine water and the
Environment, 11(3):35-45, September 1993.
Nthabiseng, M, Molapo, M and Chunderdoojh, B. (2006). Task 7.8.5 Socio-economically
Sustainable Communities Post-mine Closure, Phase I: Situational Analysis, Coaltech 2020.
Parra, G, Merz, S.K. (2009). Modeling Dewatering Effects on Groundwater Dependent
Ecosystems at Raymond Terrace, Newcastle. Presented at the Intemational Association of
Hydrogeologists, Groundwater in the Sydney Basin Symposium, August 4-5.
The influence of flooding on underground coal mines Page 140
Rasui, M.G. and P.J. Vermeulen. (2007). Improvement Strategies for Mine Dewatering Process.
Proceedings of the 4th WSEAS International Conference on Fluid Mechanics, Gold Coast,
Queensland, Australia, January 17-19: pp137-142. Faculty of Sciences, Engineering and
Health, Central Queensland University, Rockhampton, Queensland 4702, Australia.
SA Weather Service - Rainfall Figures.
Stoertz, M.W., Hughes, M. L., Wanner, N. S. and Farley, M .E. (2001). Long-term Water Quality
trends at a Sealed Partially Flooded Underground Mine. Environmental and Engineering
Geoscience, 7(1):51-65. Cited in: Donovan J.J, Leavitt B.R. and Werner. E. (2003). Long-Term
Changes in Water Chemistry as a Result of Mine Flooding in Closed Mines of the Pittsburgh
Coal Basin, USA. In: 6th ICARD, Cairns, QLD, 12 -18 July 2003, pp 869-874.
Stoertz, M. W. Hughes, M. L, Wanner N. S., and Farley, M. E. (2001). Long-term water quality
trends at a sealed partially flooded underground mine. Environmental & Engineering
Geoscience; February 2001 ; 7(1): 51-65.
Thomas, L. (2002). Coal Geology. West Sussex, England. Wiley, J & Sons LTD, The Artruim,
Southern Gate, Chichester, West Sussex, P019 8SQ, England.
Van Tonder, G. and Xu, Y. (2000). A Guide for the Estimation of Groundwater Recharge in
South Africa. Project conducted for DWAF under the supervision of Eddie van Wyk, June 2000.
Visit-South-Africa. (2011). Citing Internet Resources. http://www.visit-south
africa.co.za/Ermelo.html (accessed 27 June 2011).
WeatherSA (2011). Citing Internet Resources. http://www.weatherSA.com/we
atheraverages.aspx?q=Ermelo,%20South%20Africa (accessed 19th July 2011).
White, S.P., Creighton, AL, Bixley, P.F, Kissling, W.M. (2003). Modeling the Dewatering and
Depressurization of the Lihir Open Pit Gold Mine. Lawrence Berkley National Laboratory,
Berkeley, California, May 12-14.
Whitworth, K.R. (2002). The Monitoring and Modeling of Mine Water Recovery in UK Coalfields.
IMC Consulting Engineers, PO Box 18, Hutkwaite, Sutton-in-Ashfield, Nottinghamshire, NG17
2NS, UK.
Wilson, M.G.C. & Anhaeusser,C.R., Editors (1998). The Mineral Resources of South Africa.
Handbook, Council for Geoscience.
The influence of flooding on underground coal mines Page 141
Wolkersdorfer, C. (2008). Water Management at Abandoned Flooded Underground Mines,
Series: Mining and the Environment, pp456, Freiberg.
Woodford, A.C and Chevallier, L. (2002). Hydrogeology of the Main Karoo Basin: Current
Knowledge and Future Research Needs, WRC Report No. TT 179/02, Gezina, Pretoria.
Younger, P. L., (2000). Predicting Temporal Changes in total Iron Concentrations in
Groundwaters Flowing from Abandoned Deep Mines. A First Approximation. Journal of
Contaminant Hydrology, 44(1 ):47-69. Cited in: J.J Donovan, B. R. leavitt and E. Wemer. 2003.
long-Term Changes in Water Chemistry as a Result of Mine Flooding in Closed Mines of the
Pittsburgh Coal Basin. USA. In: 6th ICARD, Caims, OLD, 12 - 18 July, pp869-874.
Zeelie, S. and Hodgson, F.D.I. (n.d.). Risk Assessment for Groundwater Pollution from Mines in
South Africa. In: Post Mining Impacts. Institute for Groundwater Studies, University of the Free
State, Bloemfontein.
The influence of flooding on underground coal mines Page 142
APPENDIX A: Mine water classification
E.p. Durov Dlogrom A
<.1-/\VD.-~2 3
4 6
'O<'_;<>A7__ _f-5_t- ___,8 9
• Field 1: Fresh, very clean recently recharged groundvvater with HC03- and C03
dominated ions.
• Field 2: Field 2 represents fresh, clean, relatively young groundwater that nas started
to undergo Mg ion-exchange, often found in dolomitic terrain.
• Field 3: This field indicates fresh, clean, relatively young groundwater that has
undergone Na ion-exchange (sometimes in Na-neh granites or other felsic rocks), or
because of contamination effects from a source rich in Na.
• Field 4: Fre sh , recently recharged groundwater with HC03_ and C03 dominated ions
that has been in contact with a source of S04 contamination, or that has moved
through 504 enriched bedrock.
• Field 5: Groundwater that is usually a mix of different types - either clean water from
Fields 1 and 2 that has undergone 504 and NaCI mixing/contamination, or old
stagnant NaCI dominated water that has mixed with clean water.
• Field 6: Grounm'Vater from Field 5 that has been in contact with a source rich in Na,
or old stagnant NaCI dominated water that resides in Na-rich host rock / material.
• Field 7: Water rarely plots in this field that indicates N03 or Cl enrichment, or
dissolution.
• Field 8: Groundwater that is usually a mix of different types - either clean water from
Fields 1 and 2 that has undergone 504, but especially Cl mixing / contamination, or
old stagnant NaCI dominated water that has mixed with water richer in Mg.
• Field 9: Very old, stagnant water that has reached the end of the geohydrological
cycle (deserts, salty pans, etc.) or water that has moved a long time and/or distance
through the aquifer and has undergone significant ion-exchange.
Figure A: Interpretation of the Expanded Durov.
The influence of flooding on underground coal mines Page 143
I'!fl' (....... ';llel CUZ,.aU\lItg:
:'iOIli Co, ,,:'i~~I!;- l:'iw "MK C'W"'u.'l\'mIPW-UInI
8(}~. T.AI!';'. lOr;.. (1 und loc;. S04 bntt>rmtr.·•."
Ntl-+K T. Alk.
Plotting procedure on the Piper Diagram Examples of typical plouing positions
nn the Piper Diagram, for water from
various environments
Examples of typical plotting positions on the
Expanded Durov Diagram, for water from
various environments
Na
weste wmer dn.chaTj;e
l'npolhned lJnrollu1cd In'i!;Dtian return no ....
ILnd::rsroulIJ coal mille. ..
VIIU;UhUIIlClolrudi,.\
1.l1Ilt' .remmcnl Gold uune water
"rlltldwnler ~"erQD.lion,
Domeauc "'II~It"dumrs
Se1dllm
Natural ... Ime ~OIlcr
found
Deep mine werer
Cl
Figure B: Piper interpretation and classes in the expanded durov,
The influence of flooding on underground coal mines Page 144
APPENDIX B: Water standards
Figure C: Water standards
The influence of flooding on underground coal mines Page 145
APPENDIX C: Example of maps and subsidence
c:
VI
~ I c~:
,.
E, !! lP
' !.~ii ij!p .
~
I
11 .. ~ll..
>
~ :.~~~, ;::{~ ! 'jI,B ~~ r I~
~
m ~r-
~
~ ~m
*.... ""C;IF"'" 0:0~ i I ?Z tgmE :;>:I.--=-- •h ""
Figure D: Example of the maps that had to be digitize
The influence of flooding on underground coal mines Page 146
Figure F: Surface subsidence and prior rehabilitation above a stooped area
The influence of flooding on underground coal mines Page 147
APPENDIX D: Water qualities
Sulphate as S04
S04 SA Drinking Water Standard (-, -, 400, 600)
7000
6000
5000
.
4000 . H~ - .3000 - -
2000 .
. ~ - - -
1000 _- - .. .
o "IIf '" .
-1000
BHD1 BHR2 GREANWICHU... KMH2 KM-I7 KW1199(T) VOIDBH(B)
BHD2 BHR4(2) INGIlGANERIV... KMH3 KM-I8 KW2198 VOIDBH(T)
BHD4(3) BHR4(5) KILBARCHANT... KM-I4 KW1198(B) OPENCASTAA...
BHD4(6) GOLFDAMDIS KILBAACHANT... KMH5 KW1198(T) QUARRYSTRE...
BHR1 GREANWICHS... KM-I1 KM-I6 KW1199(B) \I1LLAGESTRE...
• Last Value 0 2 x Std Dev. I Min & Max.
Figure G: Kilbarchan standard deviation graph for Sulphate
Sulphate as S04
S04
2000.------------------------------------------------------------------,
1500
1000
500
o
Farm2 Usutu Atop Usutu E bottom Usutu I W13 W2
Farme Usutu Bbottom Usutu E top W1 W14 we
PumpstatIon BH Usutu B top Usutu Fbottom W11 W15 W7
Usutu AboUom Usutu 0 Usutu F top W12 W17
• Last Value 0 2 x Std Dev. I Mn & Max.
Figure G: Usutu standard deviation graph for Sulphate
The influence of flooding on underground coal mines Page 148
Piper Diagram
.~ I<M-fl' ~..... • I<M-fl• KM<5
•.Ot<KMMI<1M1-f-lG
Cations Anions
• KJ..BORCHIINlRI'NS'ffi 1'II'B.N02
• KJ..BORCHIINlRI'NS'ffi PIFR.N:,
~ ..•~II'G'G'I-ERMR LI'S.S • rnEI'NV\Oi Ll'SCF [)..M'I • rnE2I'NV)\Oi STR6AM I'PEA' B.f • OOFDJlMOS/ .EI-fi,
• !Hlot;6)
•o !EHHlol2t;3)
.... !Hl,
'_Calcium- -Chloride_'
Figure G: Kilbarchan piper
Piper Diagram
o W7
Ows
OWl
• o W17: o W15
• W14
0 • W13
•• • W12• OW11• .W1• ·0 • Uoliul
• • UoliuFtop
Cations • Anions • Uoliu Fbottomo• Uoliu EtopUoliuEbottom
• UoliuD
• UoliuBtop
• Uoliu B bottom
• UoliuAtop
.. Q • Uoliu A bcttom
••• -=>~ o• F\npstatlon BHFem15
• Fsmi2
.-Calclum- -Chlorids_
Figure H: Usutu piper
The influence of flooding on underground coal mines Page 149
S.A.R. Diagram
4 IOCB1TJ
D• I\.O1CUB..'1CBE)SIRE'M,",NOr~
• ~SlRE'MEXT"e<9o.1US
54 • CHNCASf -'AEA2 933'i'(E
Very
High o• ~KW>II8
O~B)
O·~~B)oq,f18tq,t-!7.14.4_-i,6.,
o• >tMMi2i3
• >Mi,
• ~1RI'N9'ERAFl3JN02
• ~1RI'N9'ERAFB.N;,
.~RI\ffiUS
• ~U'SCFDl.M"
10 25 75 225
Salinity Hazard - Electrical Conductivity [m Slm]
C1 C2 C4
V.Hlgh
Figure I: Kilbarchan
S.A.R. Diagram
4
oOW7ewWs2
o W17
S4 o W15
Very
High • W14
• W13
•o W12W11
.W1
• UsLtul
• UsLtuFtop
• UsLtu F bottom
1•0 UsLtuEtopUsLtu E bottom
• UsLtuD
• UsLtu Btop
• UsLtu B bottom
• UsLtuAtop
• UsLtu A bottom
•o FU1l>statlon BHFarrr6
25 75 225
Salinity Hazard - Electrical Conductivity [m Slm] • F..m2
C1 C3 C4
Low V.Hlgh
Figure J: Usutu.
The influence of flooding on underground coal mines Page 150
APPENDIX E: Cl values of boreholes
Cl values I
---_._- -_.--. - ._. -_ .._-_._ - ...
11 63 36 20
.. - ....-. - .
35.26 58 45.0 31
9 28.68~ 14 34--
10 14.89 17 20
---
2_.- 5.60 15 23.0 13.0_ .. _.~-. - ~
21.56 17 34.0 15
- -
13 14 41 15
- - -,-
15 19 40 25
. ........ ,' ..- ..
20 15 21.0 20
..-._". __ ._-._-_ ..•- .
26.41 I 14 21
5.12 i 9.18 22.0
--f-f--19 ~-·I--l~.--~~-~.1-1.---- - 54 62.38...... ,. "_
---342.839- ! - 2199:0 : 11-- --.- --_' .---2-- 36.656 ,
i
-1--2~- 27 55~.~.. ..-. ~ C',', ,. __ ,._,_.
9 14 I 59.0
•• _ ' __ Mo' _., •• ,_.
HARMEAN , 17
of. .-. __ .•• _ •• .-_.._543__01988-, .
Cl rainwater I 1
.% recharge : 5.7
The influence of flooding on underground coal mines Page 151
APPENDIX F: Pictures of the Usutu boreholes
Usutu A
The influence of flooding on underground coal mines Page 152
Usutu B
Usutu 0
The influence of flooding on underground coal mines Page 153
Usutu E
Usutu F
The influence of flooding on underground coal mines Page 154
Usutu I
APPENDIX G: Pictures of the Kilbarchan boreholes
The influence of flooding on underground coal mines Page 155
KW1/98.
KW1/99
The influence of flooding on underground coal mines Page 156
VOID BH
Ingagane river upstream
The influence of flooding on underground coal mines Page 157
Ingagane river downstream
Greenwich stream area
The influence of flooding on underground coal mines Page 158
Greenwich upstream of dump
Quarry stream extension upstream
The influence of flooding on underground coal mines Page 159
Village stream V-notch
Golf dam downstream.
The influence of flooding on underground coal mines Page 160
APPENDIX H: Pictures of the opencast boreholes at
Kilbarchan
BH26
BH30
The influence of flooding on underground coal mines Page 161
SUMMARY lOPSOMMING
The purpose of the study is to investigate the influence of flooding on underground coal mines.
Two case studies were investigated the shallow underground Klibarchan coal mine and the
deep underground Usutu mine. Kilbarchan colliery is located 10 km south of Newcastle in
KwaZulu/Natal. It comprises two underground sections, called Roy Point and Kilbarchan. Usutu
colliery is situated just 8 km outside the town of Ermelo in Mpumalanga, close to Camden power
station on the N2 road to Piet Retief. The geology of both studies lies within the Karoo Group,
Ecca subgroup in the Vryheid formation. Higher precipitation at Usutu and Kilbarchan occurs in
the summer months, while Kilbarchan has a higher annual rainfall of 864 mm/a compared to
Usutu's 705 mm/a.
The water levels at both mines yielded interesting findings. Usutu mine is compartmentalized
with walls in the underground. These walls are so strong that they function as "Iow pressure"
seals resulting in compartmentalized underground, withstanding the huge pressures created by
the recharged groundwater. This causes water levels to differ in the underground. Water levels
at Kilbarchan mine vary in depth, but when plotted in metres above mean sea level (mamsi)
they plot in a straight line. Regional recharge at Usutu was calculated as 5.7 % and 11.3 % at
Kilbarchan. Recharge is influenced by what type of mining activity was practised in that specific
area. It was concluded that recharge on opencast is between 15 to 20%, the stooped area
between 10-15% and in an underground shallow mine it could be as high as 10%.
Mining activity ceased in 1992 at Kilbarchan. Pumping is a common practice at flooded
underground mines, because the mine needs to be filled with water on an ongoing basis. This
prevents sulphate generation and the water quality from deteriorating. Pumping at Usutu is well
managed and flushing started to occur in the underground with the electric conductivity
improving over time. Pumping at Kilbarchan is poorly managed and over pumped. The electric
conductivity over time, is not improving indicating that oxygen infiltrates the system when too
much pumping occurs. Bord-and-pillar mining followed by stooping has been the main mining
method. At Usutu mining activity ceased in the late 1980 and then the mine was flooded.
It is finally there is concluded that an underground should be flooded as quickly as possible and
then managed well. Shallow underground mines have a higher potential of contamination,
because of a shorter residence time. The depth of mining, topography, mining methods, water
levels, exposure to oxygen, rainfall, recharge, residence time and pumping have an influence on
the effects of a flooded coal mine.
The influence of flooding on underground coal mines Page 162
OPSOMMING
Die doel van die studie is om ondersoek intestelop die invloed wat n gevloede myn het op die
ondergrond. Die twee studies wat ondersoek was is die vlak ondergrondse myn Kilbarchan en
die diep ondergrondse myn Usutu. Kilbarchan myn is geleë 10 km suid van Newcastle in
KwaZulu-Natal. Dit word opgedeel in twee ondergrond seksies, naamlik Roy Point en
Kilbarchan. Usutu myn is geleë 8 km buite die dorp van Ermelo, naby die Camden kragstasie op
die N2 pad na Piet Retief. Die gelogie van beide die studies val binne die Karoo Groep, Ecca
subgroup in die Vryheid formasie. Hoe reënfal vind plaas by Usutu met n somer reënval, terwyl
Kilbarchan n hoer jaarlikse reenval het van 864 mm/a in vergelyking met die 705 mm/a van
Usutu.
Water vlakke van beide studies het interresante bevindinge opgelewer. Usutu myn is
gekomparlimentaliseerd met mure in die ondergrond. Hierdie mure is so sterk dat hulle as lae
druk seels funksioneer en so die ondergrond comparlimentaliseer, die mure kan hoe druk van
aanvulling weerstaan. Dit veroorsaak dat watervlakke verskil in die ondergrond. Water vlakke
by Kilbarchan mag verkil in diepte, maar as dit in meter bo seespiëel getrek word le dit op n
reguit lyn. Die regionale aanvulling was bereken as 5.7 % by Usutu en 11.3 % by Kilbarchan.
Aanvulling word beinvloed deur waste tipe myn aktiviteit in daardie area plaasgevind het. Die
gevolgtrekking is dat aanvulling op oopgroef areas tussen 15-20 % is en hoë ekstraksie areas
tot 10 % kan wees.
Myn aktiwiteit is gestaak in 1992 by Kilbarchan en in 1980 by Usutu. Om water tepomp by
ondergrondse gevloede myne is baie algemeen, want n myn moet gevloed bly die myn goed the
bestuur. Dit voorkom sulfaat generasie en die kwaliteit van die water om te verswak. By Usutu
word die myn goed bestuur en daar het reeds spoeling in die ondergrond begin plaasvind deur
dat die EC oor tyd verbeter. By Kilbarchan word die myn teveel gepomp and swak bestuur. Die
EC oor tyd verbeter nie, die rede hiervoor is dat suurstof deur die myn getrek word oor daar
teveel gepomp word.
Laastens is daar opgesom dat n ondergrondse myn so vinnig as moontlik gevloed moet word
and bestuur word. Vlak ondergrondse myne het n hoër potensiële risiko vir contaminasie, oor dit
'n korter akkomodasie tyd het. Die myn diepte,topografie, myn metode,water vlakke,
blootstelling tot suurstof, reënval, aanvulling, akkomodasie tyd en die pomp het n invloed op die
effek van n gevloede steenkool myn.
~ uv • UIFS
lllOEMFONTEIN
The influence of floodina on underaround coal mines IB!ElLiOTIEI;_K_ • I!..!~.~RV Paae 163
KEYWORDS
• Flooding
• Decant
• Pumping
• Recharge
• Rainfall
• Sampling
• Chemistry
• Depth
• Opencast
• Mining
The influence of flooding on underground coal mines Page 164