Insights from Selected Case Studies

Water Research Commission 40 Year Celebration Conference 31 August – 1 September 2011; Emperor‟s Palace, Kempton Park, Johannesburg (South Africa) BLUE vs. TRUE GOLD Impacts of deep level gold mining on water resources in South Africa – insights from selected case studies

Frank Winde NWU Potchefstroom Campus Mine Water Research Group Contents

(1) Introduction

(2) Au mining impacts on water resources: 3 x case studies

(A) Dewatering of karst aquifers

(B) Uranium pollution

(3) The future? Largest urban agglomeration in Africa: - triggered by Au rush 125 years ago, today: - 25% population SA - 50% of energy consumption in Africa - 70% GDP SA Ferreira Camp (1886) ~400 x diggers - 70 km from nearest major river: strongly negative water balance  imports from Lesotho

Johannesburg (2011) ~4 million residents

50 Total surface runoff ) 45 40 km³/a 35 Economically 30 exploitable run off 25 demand ( demand 20

15 total

water water 10 5

1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020

125 years of Au production: >6 bn t of tailings covering ~400 km²

Total since 1886: 42,000 t 1970: Peak of SA gold production (989 t) = 68% of world production 17m all gold ever poured:

127 000 t

Au 17m 33%: SA

worldwide more steel is poured in 1 hour …1700

1400 Au-prize [$/oz] 1300 ‚Sunset industry‘? SA: 35.000t Au still available 1200  15 kt accessible with current technology  20 kt ultra deep mining needed 1100

1000

900

800

700

600

price [US$/ ounce] [US$/ price -

500 Au Au

400 WDL

300

200 m 4300 >

100

2011

1900 1870 1880 1890 1910 1920 1930 1940 1950

2010

1850 1860 1960 1990 2000

1970

1980

1800 1810 1820 1830 1840 1. Background: Au-mining and dolomite

Case study 1: Dewatering (FWR) GAUTENG

FAR WEST RAND WEST RAND NORTH WEST PROVICE CENTRAL RAND JHB Case study 3: KLERKSDORP Flooding/ AMD (CR) Case study 2: EAST RAND U pollution (WFS) MPUMALANGA

EVANDER FREE STATE

FREE STATE 100 km (Draper report) Zuurbekom pumping station  continuous pumping since 1899

EVANDER Goldfield Mine lease area Case study 1:

Dewatering of karst aquifers

(Far West Rand) Case study 1: Dewatering

Dolomitic compartments  highly karstified ~365 km²

JHB Mooi River Mooi

Ohz. c VP c - 5 x longest caves in SA Bank c - 2 x strongest ~800 km² springs - storage capacity dewatered GMB compartments

> Vaal Dam ####### ## ## # # # # # # ### Boskop Dam #

# ## Potchefstroom ### # Apocalypse cave ## # ### # # Case study 1: Dewatering of karst aquifers - consequences Case study 1: Dewatering - altering recharge rates Case study 1: Dewatering – tailings injection into karst aquifers

Dolomitic outcrop area of Sinkhole dewatered compartments

Wetland Slimes dam Stream

Khutsong

Winde (2006) Case study 1: Rewatering – decant of AMD

flooded mine voids 30 Ml/d active mine voids

~130 Ml/d

GMB

GMB as potential outflow point Total mine void volume FWR: est. 600 Mio. m³ (prior to deepening projects) Case study 2: Uranium contamination - Wonderfonteinspruit (2) Case study 2: U pollution - history

DRD Rand Uranium poss. re-open GM Tailings 1st U-pilot plant 1949 Blyvooruitzicht GM First U-plant SA: WR Cons. 1952

JHB GFL Mintails Tailings + by-prod .+ U-plant Tailings WFS

Mooi River NUFCOR GMB Dominion (240.000t) New U-mine Simmer & Jack N Potchefstroom Re-opened GM Chem Wes Tailings Buffelsfontein West Wits Re-opened GM Tailings + by-prod .+ U-plant Vaal River Vaal Reefs 10 x GM produce U for Tailings + by-prod .+ U-plant 5 x U-plants in FWR/ WR (2) Case study 2: U pollution- sources

West Rand (mined out)

Slimes dams in WFS catchment: Mass: ~1.5 bn t Surface area covered: ~50 km² U-contents: ~150,000 t of U

slimes dams

GMB Far West Rand (most mines still operational)

(DWAF 2006) (Coetze 2003) (2) Case study 2: U pollution – media reports (2) Case study 2: U pollution - studies Selected studies (16 out of 35) 1967 – CoM/ DWAF – Onderstepoort study on health impacts (confidential) 1991 – Deelkraal GM – U-levels in urine of miners 1991 – Pulles (COMRO) – Health hazard of radionuclides in GM (confidential) 1992 – Slabbert (Dfnt. GM) – Radioactivity concentration in mine effluents 1996 – IWQS (DWAF) – Screening on water radioactivity in mining regions 1997 – IWQS (DWAF) - Radioactivity Monitoring in WFS 1998 … ongoing: DWA - Mooi Rivier radioactivity surveillance programme 2000 – Wade et al. (CSIR for WRC) – Radioactivity in sediments of the WFS 2000 - Nel (Potchefstroom municipality) – U-monitoring programme 2002 – CGS (for DWAF) – Radioactivity in sediments of AC Dam 2003 – Metago (for WRC) – U-pollution by filling sinkholes with tailings 2004 - Blyvooruitzicht Gold Mine U-monitoring programme 2004 – Coetzee, Winde, Wade (WRC 1214) - U-pollution risks 2005 – DWAF – U-contamination of mine canal sediments 2007 – NNR (Brenk report) - radioactive dose calculation WFS 2007 – Sediment analysis by GFL Ltd. (WAG)

2008-2011: 10 x more studies added 1 Site-nr. (Tab. 1a/b) G: 100 (1) Source (A...K): av U (n) max.U [µg/l] U-conc. stream water: 1997 - 2008 4 (C2H152) A: 156 (25) 406 Slimes dam B: 319 (19) 705 Rock dumps - 12 x studies/ monitoring programms 5 Mine lease boundary - ~3400 x U-conc. data for water C: 29 (69) 198 Outcrop of dolomite 6 (C2H153) A: 20 (25) 99 Dewatered dolomite B: 8 (22) 23 C: 13 (66) 78

36 8 A: 0,8 (29) 9,1 10 (Att. Dam) C: 17 (69) 543

C: 16 (64) 289 (23)76 B: 0,8 (2) 1,1 (17)100

(2) 69 J: 14 (1)

60 50

37 67 Harmony GM

A: B: J:

J: 3,0 (1) 22 16 Doornkop (2)48

C: 62 (67) 759 sectrion

(77) 124(77) (1)

46 18 (1m – out)

24 J: 44

37 A: 31 (31) 58 17 (1m – in)

(31) 160(31) 106(55) (1)

(1) B: 26 (21) 59

D: D: K:

26 A: 38 (25) 71

(C2H069)

57 85 79 J: 19 (1) 23 23 B: 44 (23) 76

K: 31 (1)

31 A: B: J: K:

(305) 566(305) Harmony GM (3)84 (4) 94 (4) Kloof GM

Cooke

(AC dam) (AC 71 71 65 (Venterspost)

55 sectrion

32 B: D: J:

(31) 7,0 (31) 20 (20)

(16) 43 (16)

(C2H013)

0,9 1,6 14

34 A: E: B:

(25) 104(25) 52 (22)

(105) 172(105) Simmer & Jack

U-mine

19 19 22

35 A: B: D:

Driefontein Cons. GM South Deep GM 33 Doornfontein/ J: 105 (1) Blyvooruitzicht GM K: 99 (1)

39 Elandsrand A: 0,5 (31 2,1 B: 2,3 (16) 24 GM Deelkraal GM 40 (Boskop dam) A: 2,7 (31) 9,5 B: 11 (26) 142 N

43 (Potch dam) A: 4,2 (23) 48 10km B: 14 (12) 57 out of map Decanting mine 15 West Rand void water 6679 (2) Case study 2: U pollution - Wonderfonteinspruit 1219 793 8 319

Bovenste Oog +104% perennial perennial

- 50

Mooi River Mooi 0,8 C2H152

non 993

1m pipe 1m Outflow Dam Don 15 Outflow C2H025

C2H172 +130% Harry„s dam Harry„s 0% Outflow 1607 80 871 34 15

C2H080 +286%

Gauging stationGauging upstr. A Dam ofCoetzee C2H080 100 159 C2H175 1358 C2H157 62 Klerkskraal +69% Cooke #1 dam C2H172 143 C2H154 effluent 175

C2H069 26 3196 50 C2H069 +38% 1099 292 20 C2H171 -32% 35 C2H063 +17 425 57 86 32 -66% Venterspost (Kloof) 25 GM

C2H061 Driefontein GM Fissure water dirt road road culvets dirt C2H161 at Muiskraal WFS 13-fold increase 190 95 270 C2H160 C2H156 of U-load 1342 Mooi River Mooi - 33% +1260% - 40% Blyvooruitzicht GM -52% 40 18 C2H013 5 At Doornfnt # 3 shaft 35 13 14 Upper Turffontein 148 105 N Eye Not to scale

51 Gerhard 43 2,3 Minnebron Eye Receiving stream +330% C2H011C2H011 Discharge from active gold mines into the WFS 200 Dolomtic karst springs 11 C2H162 Inflowbridge Boskop dam 803 C2H069 DWAF code for sampling points +308% 3-fold increase C2H174 Boskop dam Ml/d average flow [Megalitre per day] of U-load ppb average U-conc. in water [µg/l] (italic- sources other than DWAF) 10 51 Potch dam 14 +240% 793 U-load [kg/a] (1998-2008) +104 To Potch water supply U-load difference [kg/a] (since 1997), (+ load increased; - load decreased) (Winde 2009) to Vaal River Case study 2: U-pollution

~3.5 t/a

TFT eyes

? Mooi Rivier Mooi

GMB eye

Natural dissolution of dolomite results in abundance of: Boskop dam 2+ 2+ - Ca , Mg , HCO3 each ion found to reduce U- removal effiency of standard potabilization processes to 60% Potch (Baeza, 2008 water works ) i.e. In dolomitic water U standard is more likely to slip potabilization: chlorination, through treatment flocculation, plants filtration 2003: Kettle scale 21 vs 1 ppm Case study 2: U pollution – health effects

- 1995: Dr de Villiers (Stellenbosch Univ.) high number of leukaemia patients from area around Poffadder (N-Cape) - blood samples of 418 people from 120 localities (>16 a of age, 1993) - water quality: 126 boreholes, AEC data (early 1980s) (2) Case study 2: Uranium pollution – health effects U-levels Lymphocythe count in 126 boreholes (418 residents from 120 localities)

Reddish colours: Reddish colours: 30-100 % of sampled U: >70 – 300 µg/l population with abnormally high lymphocyte count

WFS: >300 (upper) … 80 µg/l (lower) Mine water: 200 … >3000 µg/l 500 000 grid cells correllated r = 0,5731

 3 x follow-up studies confirmed high U-levels and domestic use of water Case study 3

Flooding of mine voids (AMD) (2) Case study 3: Flooding of mine voids

flooded mine voids 15 Ml/d active mine voids (2002)

60 Ml/d (2012)

~130 Ml/d (2050?) GMB

Total mine void volume FWR: est. 600 Mio. m³ (prior to deepening projects) (2) Case study 3: Flooding of mine voids (2) Case study 3: Flooding of mine voids

2007

‚You‘ magazine (March 2011)

(Water Wheel 2011) (3) (3) Case study 3: Safety margin to PBL: GRC # W Findings:

106 m Std. Std. Bank (1) No flooding risk for CBD for risk floodingNo (1)

90 m CPBL 106 m (Std. Bank Std. Bank Gold Reef City # Mine voidflooding FMWL FMWL (decant): mamsl 1614 City Deep 4# (market) Shafts Shafts hydraulically connected various at deeper levels

CD 4 # (market) ABSA tower E ) and 90 m (ABSA)

S&J Howard # S&J: Howard #

ERPM-SWV #

SWV ERPM-Central # N

Cinderella West #

Cinderella East #

Central

Cinderella W (1613.7 mamsl) S Cinderella E E (2) Case study 3: Flooding of mine voids

(1) No flooding risk for CBD (single basin, many open shafts)

(2) No sinkhole risk for CBD (no dolomite)

(3) Ingress sources mainly from surface (50-60%), many managable

(4) Lower than predicted decant volume (35 vs. 60 Ml/d)  pumping = ingress = decant  overestimation  - service water + natural cut off of ingress sources

(5) Primary pollution sources mainly on surface not underground  plastic stope closure (>70% void volume reduction)  neutral water decant WB

(6) Lesser pollution impact on streams than predicted

(7) Slower rise than predicted, less depended on rainfall (37 vs. 57 cm/d)

Mine water level at 5 x monitoring shafts (23 July 2009 - 24 March 2011) 450

av. rate of rise (2.7.10-24.3.11): av. rate of rise (2.7.10-24.3.11): Enough time to consider 0.37 m/d 500 0.55 m/d

550 alternative to very 600 expensive pump- and DRD 6#

650 GRC 14# Elevation CD 4# WL of Central sub-basin (at GRC #) Howard # treat option currently

700 reaches WL in DRD sub-basin and stops SW Vert # [m below [m collar each of shaft] 2nd inflow from DRD. WL at DRD starts to rapid rise along with WL in Central sub-basin. 750 rise 1st proposed  sustainable rapid 800 rise stagnation

continous rise low-cost, low-energy

6. Jul09 6. Jul10 1.

3. Mrz 10 Mrz 3. 10 Okt 9. 11Apr 7.

2. Mai10 2.

2. Jan10 2.

3. Nov09 3. Dez10 8. 11Feb 6.

4. Sep09 4.

26. Jul26.09 Jul21.10

14. Okt 09 Okt 14. 10 Mrz 23. 10Apr12. 10 Okt 29. 11 Mrz 18. 11Apr27.

22. Mai22.10

16. Jun16.09 Jan22.10 Jun11.10 Jan17.11

15. Aug15.09 Nov23.09 Dez13.09 10Feb11. Aug10.10 Aug30.10 Nov18.10 Dez28.10 11Feb26.

24. Sep24.09 Sep19.10 Date of measurement solution possible (3) The future? Karst aquifers in the Far West Rand (3) The future: Post-mining opportunities (FWR) Underground pump-storage schemes

ESKOM principle of pumping scheme at Sterkfontein dam: day-night tariff differences  Use existing underground infrastructure such as shafts, reservoirs, pumping chamber, pipelines, turbines, etc.  no electricity losses through long-distance transport  cost neutral pumping of water for irrigation/ utility

Old shaft

Day Karst cavity Night

Construction of underground Turbine hydropower pumping scheme in karst area on the island of Java (Indonesia)

Driefontein GM Pump

Kloof GM

(3) The future? Waste land transformation: wind energy Height above surface above Height

Wind speed

20...>50m (3) The future: Post-mining opportunities (FWR)

Underground hydro storage of wind energy

Use wind when it is vailabe to pump water up and let it run down if energy is required

Old shaft

wind Karst cavity

Turbine

Pump (3) The future: Post-mining opportunities (FWR) Waste land transformation: Solar-thermal up-wind power generation

Prototype: Prevents wind erosion of tailings Manzanares, Spain and rainfall infiltration (no seepage) Chimney - generates energy for sustained draft pumping of plumes where necessary

Turbine Chimney -1km high - 200m diameter

Transparant dome (foil or glas)

Greenhouse effect

Slimes dam pump - high albedo GW purification - surface area SD (SA): ~400km²

Groundwater pollution plume Thank you!

Photo: DWAF advertisement in “The Star”