Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. An Assessment of the Revegetation Potential of Base-Metal Tailings from the Tui Mine, Te Aroha, New Zealand. A thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Soil Science at Massey University William John Morrell 1997 11 Abstract The overall objective of this study was to investigate the revegetation potential of abandoned base-metal (Cu-Pb-Zn) tailings at the Tui mine site near Te Aroha, New Zealand. An estimated 100,000 m3 of sulphide-rich tailings are the legacy of a once prosperous mining venture conducted at the site between 1967 and 1974 by the now defunct Norpac Mining Ltd. The oxidation of remnant sulphides, which constitute as much as 15% of the tailings by weight, has prevented plants from colonising the tailings for more than 20 years and resulted in the formation of Acid Mine Drainage (AMD) which continues to degrade ground and stream waters in the vicinity of the dam. This study focused on characterising the physical and chemical properties of the tailings in terms of their plant growth potential using a variety of techniques including; Scanning Electron Microscopy (SEM), Energy Dispersive X-ray Spectrometry (EDS), Flame Atomic Absorption Spectrometry (FAAS), X-ray diffraction (XRD) as well as field based observations and standard laboratory analyses. A long term incubation experiment (carried out over a 1 � year period) and two plant growth trials were also conducted to investigate the ability of liming materials and/or organic waste to ameliorate the tailings in order to provide a suitable growing medium for plants. The research indicated that, whilst the tailings do not exhibit any major physical limitations to plant growth, chemically the tailings are an extremely hostile plant growth medium. The surface tailings exhibited variable but generally very low pH (2.76 - 3.85) and high concentrations of potentially phytotoxic elements including As (254 mglkg), Cu (26-991 mglkg) , Pb (1503-27416 mglkg) and Zn (123-2333 mglkg). The high availability of these and other metals, including Al and Fe, under the acidic conditions prevailing in the tailings, were identified as the primary factors currently inhibiting plant growth. The distribution of heavy metals, sulphate and total SUlphur with depth indicated that the surface tailings (0-200 mm) have been extensively weathered since their deposition and contain relatively low concentrations of most heavy metals compared to subsurface III tailings. An assessment of the Acid Generating Potential (AGP) of the tailings, using both static and kinetic tests, similarly indicated that the surface tailings have a comparatively low ability to generate acidity by sulphide oxidation. The application of lime at a rate of about 50 Mg CaC03/ha is calculated to theoretically prevent the surface tailings from reacidifying. Below 200 mm depth, however, the AGP is appreciably higher (>140 Mg CaC03/ha) and concentrations of both total and labile (O. 1M HCl extractable) Cu, Fe and Zn were found to increase substantially, reflecting an increase in the abundance of chalcopyrite (CuFeS2), pyrite (FeS2) and sphalerite (ZnS) and sulphates with depth. The presence of high concentrations of acid-generating sulphide minerals (primarily pyrite) at shallow depths has important implications in that revegetation of the tailings should be based on techniques that minimise the exposure of the largely unweathered, sulphide-rich, subsurface tailings. The results obtained from the plant growth trials indicated that, on tailings treatments sown with metal-tolerant varieties of Festuca rubra or Agrostis capillaris, satisfactory cover was achieved upon the addition of lime at a rate of 16.5 Mg/ha or composted sewage sludge at rates >220 Mglha. These metal-tolerant plants were found to out yield their non-metal-tolerant counterparts on the limed treatments by as much as 4 and 1 0 times, respectively. Although vegetation was successfully established on Tui tailings treated with lime, dry matter yields were relatively low compared to treatments receiving high rates of sludge. The growth of all plant taxa was found to significantly decrease where very high rates of lime (112 Mg/ha) were added due to pH-induced nutrient deficiencies. Results obtained from both the plant growth trials and the lime incubation experiment indicated that the application of low rates of lime and/or sludge (8.25 and 1 1 0 Mglha respectively) were ineffective at creating a suitable plant growth medium and, in fact, exacerbated growing conditions within a few weeks of application by increasing the availability of labile (O. lM HCl extractable) metals. IV The findings of this study indicated that the use of amendments may provide a relatively inexpensive way of facilitating the establishment of plants on the tailings at the Tui mine site. It is envisaged that, at the very least, a vegetative cover will improve the aesthetic appearance of the site and at best reduce AMD by creating an oxygen­ depleting, organic-rich cover which may ultimately facilitate the establishment of native species from the adjacent forest. v Acknowledgments I would like to thank sincerely my chief supervisor Dr. Bob B. Stewart and his colleagues and my co-supervisors Professor Paul E.H. Gregg, Dr. Nanthi S . Bolan and Dr. Dave J. Home. Their interest, encouragement and advice during this study were invaluable. To the technicians and staff of the' Soils Department' at Massey University who helped me along on my 'tremulous' journey, thank you kindly. Particular thanks to 'Russo' Wallace for helping out with some of the ' less than exciting' aspects of practical research. I would also like to acknowledge the assistance of the 'Keith Williamson Electron Microscope Unit' operated by The Horticulture and Food Research Institute of New Zealand Limited, Palmerston North. Particular thanks to Doug Hopcroft and Raymond Bennett for providing technical assistance with microscope work. Much needed funding for this research project was kindly received from a variety of scholarships including the Quinphos PhD. Study Award, the Leonard Condell Farming Trust Scholarship, the C. V Fife Memorial Scholarship and the Macmillan Brown Agricultural Research Scholarship. I appreciate the contributions made by the 'tax payers' of New Zealand (including my uncles!) who largely facilitated this study by funding the University'S Vice-chancellors PhD Study Award and the Graduate Research Fund. To my friends, near and far, thank you for your friendship, encouragement and words of wisdom over the years. You have made the experience worthwhile. A special thanks to Shane for putting up with my office antics over the years and for sharing more than the odd joke, and to Shiva for his friendship and patient help with some of the 'stats'. To my family thank you for your love, support and understanding during my 'student years'. To Joan, my mother, thank you for you unbridled faith and support. And finally to Helen Leslie, my companion and confidant, who has born the brunt of my writing days (but knew the right time to escape), thank you for your inspiration. VI CONTENTS Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V Table of Contents . . . . . . . . . . . . . , .... '" ...................... ,. ... ...... ...... . . .. .. ... .. ... .. VI List of Tables . ................................................................................ IX List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XIV List of Photographic Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , xx List of Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XXlll CHAPTER! Thesis Introduction 1 .0 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 . 1 General Obj ectives of Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1 .2 Specific Objectives of Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 . 3 Implementation of Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . "' . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1 .4 Thesis Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 CHAPTER 2 Tui Mine Site - Study Area 2 .0 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2 . 1 LocationlPhysiography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Vegetation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2 .3 Climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 .4 Site Hydrology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 2 .5 Geology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.5 . 1 Tectonic Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 9 2 .5 .2 Regional Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2 .5 .3 Local Geological Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2 .6 Mineralogy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2 .6 . 1 Origin of Sulphide Mineralisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.7 Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.8 The History of Mining at Te Aroha . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2 .9 Tui Mine Politics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 3 .0 Chapter 2 Summary & Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 va CHAPTER 3 The Reclamation of Mine Wastes 3.0 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 34 3. 1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2 Mine Site Degradation & Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.2. 1 Australian Mine Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2. 1 . 1 Rum Jungle, Northern Territory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2. 1 .2 Captains Flat, New South Wales . . . . . . '" . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . 38 3.2. 1 . 3 Mount Isa & Broken Hill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2.2 New Zealand Mine Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.2.2. 1 Martha Hill Mine, Waihi . . . . . . . . . . . . . . . . , . . . . . . . .. . . . . . . . ' " . . . . . . . . . . . . 42 (i) Tailings Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 (ii) Tailings Reclamation Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 (iii) Waste Rock Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3.2.2.2 Golden Cross Mine, Waihi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.2.2.3 Macraes Flat Mine, Eastern Otago . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.2.2.4 Alluvial Gold Mine Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.2.2.5 Titanomagnetite Mining, Waipipi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,. 52 3.2.2.6 Aggregate Mine Reclamation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. .. 53 3.2.2.7 Coal Mine Reclamation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . 54 3.2.2.8 Tui Mine, Te Aroha. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 (i) Past Reclamation EffortslResearch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 62 3.3 3.3. 1 Sulphide Oxidation & Acid Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . 66 Pyrite Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68 Sulphide Oxidation in situ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1 Factors Affecting Sulphide Oxidation Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 3.3.2 3.3.3 3.4 3.4. 1 Acid Mine Drainage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 74 3.4. 1 . 1 (i) (ii) (iii) 3.4. 1 .2 (i) (ii) (iii) 3.4. 1 .3 3.4. 2 3.4.2. 1 Acid Mine Drainage Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' " . . . . . . . . .... 76 Acid Base Accounting (ABA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 The Development of ABA . . . . . . . . . ........................ ... ... ... .... 76 ABA Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. 77 Limitations of ABA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , '" . . . . . . . . . . . . 79 Hydrogen Peroxide Oxidation (HPO) . . . . . . . . . . . . . . . . . . '" . . . . . . . . . '" 8 1 HPO Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8 1 Limitations ofHPO Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1 Application of HPO Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Alternative Prediction Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Metal Quantification in Mine Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Non-Sulphide Fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' " . . . . . . . . . . . . . . . . . Extractants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chelating Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dilute Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . Total Metal Extractants . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . 3.4.2.2 (i) (ii) 3.4.2.3 3.5 Acid Mine Drainage Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 86 86 87 90 92 94 94 95 96 97 3.5. 1 3.5.2 3.5.2. 1 3.5.2.2 Active Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Passive Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anoxic Limestone Drains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aerobic & Anaerobic Wetlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vlll 3.6 3.6.1 3.6.2 3.6.3 3.6.4 3.6.4.1 3.6.4.2 3.6.4.3 3.7 3.7.1 3.7.1.1 3.7.1.2 3.7.1.3 3.7.1.4 3.7.1.5 3.7.1.6 3.7.2 3.7.2.l 3.7.2.2 3.7.3 Preventing Acid Mine Drainage. . . . . .. .. . . . . . . ... . . . . . . . .. . . . . . . . . . . . . ............... 99 Subaqueous & Subterranean Deposition . . . . . . .. . .. . . . . . . . . . . . .. . . . . . . . . ... .. 101 Bactericides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 pH Modification . . . ... . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. .. . . . . . ... . . .. . . . . . . . . . . . . .. , 104 Dry Cover Barrier Methods . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 105 Synthetic Barriers . . . .. . . . . . . . . . . . . . . . ... . . . . . ... . . . ... . . . . . . . . . . . . . . . . . . .. 106 Geological Barriers ... . . . . . . . . . . . . . ... . . . . . . . .. . . .... . . . ... . . . . .. . . . .. . ... 106 Organic Barriers . . . . . . ... . . . .. . . . . . . . . . . . .... . . . . . . . ..... . .... . . . . . . . . . . . . 109 Revegetative Reclamation . . . . . . . . . . .... . . . . . .. . . . . . . . . . . . . ...... . . . . .... . . . . . . . . . . . . . . I I I 3.7.3.1 3.7.4 3.7.4.1 (i) (ii) (iii) (iv) 3.7.4.2 (i) (ii) (iii) 3.7.4.3 3.7.4.4 3.7.4.5 (i) (ii) (iii) (iv) Potential Beneficial Effects of Revegetation . .. . .. . . . ... . . . . . . .. . . . . .. . . . . . . . 112 Stabilises Against Erosion . .... . . . . . . . . .. . . . . .. . .... . . . ... . ... . . . . . .. ... , 112 Increases CEC & Specific Adsorption . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. 114 Reduces Oxygen Ingress . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115 Decreases Leaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 115 Increases Plant Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 116 Improves S ite Aesthetics . . . . . . ... . . . . . . . . . .... . .. . .. . . ... '" ........... , 116 Potential Negative Effects of Revegetation . . . . ... . . . .. .. . . ... . ..... . ........ 117 Reductive Dissolution . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . .... . . . . ..... . 117 Increased Oxygen & Water Ingress . . . . . . . . . . . . .. . . . . . .. . .. . . . . . . . . . .. , 118 Lime Application . . . . . . .. . . . . . . . . . . . . . .. . . . . .. . . .... . . . . . ... .. . . . .. . . . . . . . ... . . . . 118 Acid Neutralisation . . . . . . . . . . ..... . .. .. . . . .. . .. . . . . . . . . . . . . ... . . . . .. . . . . . . 119 Organic Waste Application . .. . . .. . . .. . . . . . . . . . . . . . . . ... . . . . . ... .. . . .. . . . . . . . . . . 120 Beneficial Chemical Effects . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . .. .. . . . . . . . . . . 122 Nutrient Addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . .. . . . . .. . . . . . . 122 Increased CEC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 123 Moderated pH . . . . . . . . . . . . . . . .. . . . . . . . . . . . ... . . . .. . . . . . . . ... . . . . . . . . . . .. . . 123 Heavy Metal Attenuation .. . . . . . . . .. . . . . . . . . .. . . . .. . . . . . .. . .. . . . . . . . . .. . . 124 Beneficial Physical Effects . . . . . . . . . . . . .... . ......... . . . . . .. ... . . . . .. . . .. 125 Bulk Density . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . ... .... . .. . . . .... . . . . . . . . . . . . 126 Aggregation & Porosity . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .... . . . . .. . . . .. . 126 Hydraulic Conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Biological Effects . . . .. . . . ...... . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . .. . . . .. . 127 Effects of S ludge Addition on Plant Yield . . . . . . . . . . . . . . . . . . . . . . . . . .. 129 Problems Associated with Sludge Application . . . . .. .. . . . . . . . . . . . . .. 129 Heavv Metal Contamination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . .. .. . . . .... 130 Peslicides . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 132 Nutrient Imrnobilisation . . . . . . ... . . . . . . .. . . . .. . . . . ... . . . . . . . ... . . . . . . . . . . 132 Application Rates & Frequency . . . .. . . . . . . . . . . . . . . . . ... . . . . . . . . . .. . . . . . 133 Metal Tolerant Plant Species . . . . . . . .. . . . . . . . . . ... . .. . . . . . . . . . . ..... . ... . . . . . . . . 134 Limitations of Metal Tolerance . . . . . . . . . . ... . . . . . .. . .. . . . . . . . . . . . . . . .. . . 136 Mechanisms of Metal Tolerance . . . . . . . . . ... . . . .. . . . . . . . . . . . . . . . . ... . . . 137 3.7.5 3.7.5.1 3.7.5.2 3.8 Chapter 3 Summary & Conclusions . . . .. .... . . . . . . . . . . . . . . . ... .. ... . . ... . .. . . . . . . . . . 139 4.0 4.1 4.1.1 4.1.1.1 4.1.1.2 4.1.1.3 4.1.1.4 4.1.1.5 4.1.2 4.1.2.1 4.1.2.2 4.1.2.3 4.1.2.4 4.1.2.5 4.1.3 4.2 4.2.0 4.2.1 4.2.1.1 4.2.2 4.2.2.1 4.2.2.2 4.2.3 4.2.3.1 4.2.3.2 4.2.4 4.2.5 4.2.6 4.3 4.3.0 4.3.1 4.3.1.1 4.3.1.2 (i) (ii) 4.3.2 4.3.3 4.3.3.1 4.3.3.2 4.3.3.3 4.3.3.4 4.3.3.5 4.3.4 4.3.4.1 4.3.4.2 lX Chapter 4 Physical & Chemical Characterisation Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Preliminary Characterisation . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . ... . . . ... . . . . . . . , 143 Methodology . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 143 pH (water) . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 pH (H202) . . . . . . ... . . . . ... . . . .. . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 145 Total Sulphur . .... . . . . . . . .... . . . . . . . . . . . . ... . . . . . . . . . . .. . . . . . . . . . . . . . . . . . ,. 145 Phosphate Extractable Sulphate . ... . . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Heavy Metal Concentrations .. . . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . " 146 Results . . . . . . . .. . . . .. . . ... . . . . . . . . . .. . . . ... . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . , 146 Metal Concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 146 Total Sulphur & Sulphate Concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 150 Plasma Emission Spectrometry . . . . . . . . . . . . . . . . . . . .. . . . . . . . .... . . . . . . ... 150 pH (water) . . . . . . . . . . . . .. . . . . . . . . . ...... . .. . . .. . . . . . . . . . . ..... . .. . . . . . . . . . . . 152 Acid Generating Potential . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Preliminary Characterisation Conclusions . . . . .... . . . . . . .. . . . . . . . . . . . . . . . . . ... 155 Physical Characterisation . . . . . . . . .. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . ... 157 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . .. . . . . . .. . . . . . . . . '" 157 Piezometer Installation . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , ....... 157 Piezometer Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. 158 General Description . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . " 160 Tailings Grain size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 163 General Discussion . . . . .. . ... . . . . . ... . . . . . . . . . . . . . . . . .. . . .. . . . ... . . . . . . . " 165 Penetration Resistance . . . . . . . . . . . . . .. . ... .. . . . . . . . . . . . . . . , ., .................. ,. 167 Methodology . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 167 Penetration Resistance Results . . . . . , .................................... 168 Infiltration Capacity .. . . . . . . . . . .. . . . . ... . . . . . . . . . . .. . .. . . . . . ... . . . . .. . . . . . . . . . . .. 172 Bulk Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 174 Conclusions on Physical Characterisation . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 175 Chemical Characterisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . ... . . . . 177 Introduction . . . . , ................................................................. 177 Sampling Locations! Procedures . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Grid Sampling at Sites 1 to 12 .......................................... 177 Incremental Sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 50-mm Incremental Samples . . . . ... . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . ... . 178 100-mm Incremental Samples . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . 178 Methodology . . . . . . ... . . . . . . . . . . . . . . . . . . . . . ... . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . 178 Extractant Selection . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. 180 Extractant Comparison . . . . . . . .. . . . . . . . . . . . . . . . ...... . . . . . . ... . . . . . . . . . . . . . . . . . . . 181 Validation of Analytical Results . . . . . . . . . . .. . . . . ... . . . . . . . . . . . . . . . . . . . . . 181 Assessment of 25% Nitric Acid Extractant . . . . . . . .. . . . . . .... . . . . .. . . , 182 Hydrogen Peroxide Oxidation . . . . . . . .. . . . .. . . . . .. , ..................... 183 Discussion on Extractants . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 Metal Analysis of Incremental Samples . . . . . . . . . . . . . . . . . . . . . . . . . . '" ......... 187 'Total' Extractable . . . . . . . . . . . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . 187 0. 1MHCI Extractable . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . 188 4.3.5 4.3.5.1 4.3.5.2 4.3.5.3 4.3.6 4.3.6.1 4.3.6.2 4.3.6.3 4.3.6.4 4.3.6.5 4.3.7 4.4 4.4.0 4.4.1 4.4.2 4.4.2.1 4.4.2.2 4.4.3 4.4.3.1 4.4.3.2 4.4.4 4.5 5.0 5.1 5.1.1 5.1.2 5.1.3 5.2 5.2.1 5.2.2 5.2.3 5.2.4 5.3 5.3.1 5.3.2 5.3.3 5.3.4 5.3.4.1 5.3.5 5.4 x Metal Analysis of Grid Samples . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. .. . . . . . . . . . 194 'Total' Extractable . . . . . ... . . . . .. . . . . .. . . . .. . . . . . . . . . . . . . . .. .. . . . . . . . . .. . . . 194 O. lMHCl Extractable . . . . . . .. . . . . . . . . . . . . , ............................... 197 Discussion on Metal Concentrations . . . . . . . . . . . . . . . . ... . . . . .. . . . . . .. . . . 198 Sulphur & pH Characterisation . . . . . . . . . . . . . . .. . . . . . . . . . . . '" .................. 208 Total Sulphur . . . . . . . ... . . . . .. . . . . . .. . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 Phosphate Extractable Sulphate . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . .. . . . 208 pH Variation with Depth . . . . .. ... .. . . . .... . . . . .. . .. . . . . . .... . . . .. . . . .. . . 209 pH/Sulphate Correlation . . . . . . . . . .. . . . . . . . . . . ... . . . . . . ... . . . . .. . .. .. . . . .. 212 Spatial pH Variation . . . . . . . . . .. .. .. . . . . . . . . . . . . . . . . . . . . . ... . . . .. . . . . . . . . .. 216 Conclusions on Chemical Characterisation . . . . . .. . . . . . . . . . . . . . .... . . . . . . .. . . 218 Acid Generation Potential . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 Introduction . . . . ... . ... . . . . . . . . . . . . . . . . . . . . .... . . . .. . . .. . . . . . . . . . . . .. . .. . . . . . .. . . . , 220 TAGP Method: Theory & Methodology . ... . . . . . . . . . . . . .. . . .. . . . . .. . .. . .. . . . . 221 Acid Generating Potential Results . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . .... . ... .. . . . . 224 100-rum Incremental Samples . . . .. . . . . . . . . . . . . ... . . ... . . . . .. . . . . . . . . . . . 227 Comparison of AGP Results . . . . . . . . . . ..... . . . . . . . . . . . . .. . ... . . . . . .. . . . . 230 Neutralisation PotentiaL . . . .. . . . . .. . . . . . . . . . . . . . . . . .. .. . . .. " .. , ............... , 231 Neutralising Potential Assessment. . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. 232 Neutralising Potential Results . . . .. . . . , . , ............................... 234 Conclusions on AGP Characterisation . . . . . . . . . .. . . . .. .. . . . . . . . .. . .. . . . . . . . . . 235 Chapter 4 Summary & Conclusions . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . ..... . ... .. . . ... . .. 236 Chapter 5 Mineralogical Analysis & Scanning Electron Microscopy Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 X-Ray Diffraction Analysis . . . . . . . . . . . . . . . . . . . . . ... . . . . . ... . . . . . . . . . . . . . . . . .. . . . . . . . . . . 238 Aims of X-ray Diffraction Study . . . . . . . . . . . . . . . . . .. . .. . . . . .. . . . . . ... . . . . . . . . . . 239 Preliminary X-ray Diffraction Method . . . . .. . ... . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 239 Preliminary X-ray Diffraction Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 240 Sodium Polytungstate Separation . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . " 241 Sodium Polytungstate Separation Method . . .. . . . . . . . . . . . . . . . .. . . . ... . . . . . . " 241 X-ray Diffraction Method . . . .. . . . . . . . . . . . . . . . . . . . ... . . . ... . . . . . . ... . . . . . . .. . . . . 243 Sodium Polytungstate Separation Results . . . . .. . . . . . .. . . . . . .. . . . . . . . . . . . . . . . . 243 X-ray Diffraction Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 245 Scanning Electron Microscopy & Energy Dispersive X-ray Spectrometry . . ... 247 Aims of SEM & EDS Study . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . 251 Sample Preparation . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . .. . . . . . . . . . . .. . .. ... . . . . . . ... 251 Scanning Electron Microscopy Method ... . . . . . . . . . . . . . . . . . . . . .. .. . . . . .. . . . . . 253 Results & Discussion . . . . . . . .. . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . 253 Scanning Electron Microscopy . . . . . . . . . . . . . . . . . . . . . . . . ... . . .. .. . . . . . . . . 253 Low Magnification Energy Dispersive X-ray Spectrometry . . . .. . . . . . . . . . . 262 Chapter 5 Summary & Conclusions . . . . . . . .. . . . . . .. . . . .. . . . . . . . " .................... 265 Xl Chapter 6 Lime Incubation Experiment 6.0 Introduction . . . .. . . . . . . .. . . . .. . .. .. . .. . . . . .. . .. . . . . . .. .. . . . .. . . . . . . .. . . . . .. . . . . .. . . . . . . ... 268 6.1 Aims & Hypotheses . . . . . . .. . . . . . . . . . . . .. .. . . . . . . . . . . . . . .. . . . . . . . . .. . . . .. . . . . . . . .. . .. ... 269 6.2 Experimental Design . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . .. . . . . . . . .. . . . . . . , ................ 269 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.3.4. 1 6.3 .4.2 6.3.4.3 Methodology . . . . . . .. .. . . . ... . . . . . . ... . . . . . . . . . .. . . . .. . ... . . . . . . . . . . . . . .... . . . . . . . . .. . . . . 270 Lime Application Rates .. . . .. . . .. . . .. . . . . . . . . . . . . . . . . . . . . ..... . . . . .. . . . . . . .. . .. . 270 Water Regime .... . .. .. ... . . . ... . .. .. . .. . ... . . .. .. . . . . . . . .. . . . ..... . . . .. .. . . ... . . . 272 Incubation Environment .. ... . . . . . ... . . . . . . .... . . .. . .. . . . . . .... . . . . . ... .. . . . . . . . 274 Monitoring of the Treatments . .. .... . .. . . . . . . . . . . . . . ..... . . .. . .... . . . ... . . .. . . . 274 pH ... . . . ... . .. . .. . .. . .. . . . ... . .. .. . . ... .. . .. . .. .. . .. . . . . . . . .. . . . . . . . . . . . . . .. . 274 Metals . . . .. . . . . . . .. . .. . . . . .. .. . .. ... .. . . . . . . . . . . . . . . . . ... .. . . .. . .. . .. . .. . . . . 275 Sulphate . . . . . . .. . . .. . . . . . .. . ... . ........ . . . . . . . ... ... . .. . . . . . . . . . . .. . . . . . . . . 275 6.4 6.4.1 6.4.1.1 6.4.1.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 (i) (ii) (iii) (iv) 6.4.2 6.4.2.1 6.4.2.2 (i) (ii) (iii) (iv) (v) Variation in pH & Metal Levels with Time . . . . .. .. . . . . . . . . . . . .. . . . . . . . . . . .. . 278 pH . .. . ... . . . . . . . . . . . . . . . . . .. . . . . . . . . .. .. . . .. .. . . . . . . . .. . . . ... . . . . . .. . . . . . . . .. 278 Metal Concentrations . . . .. .... . .. . . . . . . . . . . . . . .. . . . . ...... . . . . . . . . . . . . . . . 282 Cu ... . .. . ... . . . . . . . ... . . . . . . . .. . . ... . . . .. .. . . . .. . . . . . . . . . . . . . . .. . . . . . . .. . . . . . 282 Fe .. . . . .. . . . . . . .. . . . .. . .... . .. . . . . . .. . .. . . . .. . .. . . . . .. . . . . .. . . . . . . .. . .. . . . . . . 282 Zn . . . . .... . . . . . . . . . .. . . . .. . ... . . . . . .. . . . . . .. . . . . . . . . . . .. . . . . . .. ..... . . . . . . . . . 283 Pb . . . . .. . . . . .. . . ...... . .. . . . . . . .. . .. . . . . . .. . . . . . . . . .. . . . . . . . . . . . . . .. . . .. . . . . . 283 Variation in pH, Sulphate & Metal Levels with Depth & Liming Rate . . 292 pH . . .. . . . .. . . . .. . . . . . . . ... .. . .. . . . . . . . . .. . . . . . . . . . . . . .. . .. . . . . .. . ...... . . ... 292 Metal Concentrations . . . .. . .. . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 295 Cu . . . .. . . . . .. . . .. . .. . . . ..... . .. . . . . . . . . . . .. . ... . . . . . . . . .. . . . . . . . . .. . . .. . .. . . . 295 Fe . . . . . . . . . . .. . . . . . . . . .. . . . .. . . . .. . . . . . . . . . . . .. . . . . . .. . . . . . . . . . . . . . ..... . . . . . 296 Zn . . . . . .... . .. . . . . . . .. . . . .. . . . .. . .. . . . . . . .. . . .. . ... .. . . . . . . . .. . . . . .. . . . . . . . . . 296 Pb . . .. . ..... . ... . . .. . . . .. . . . ... . . . .... . . .. . . . . . . . .. .. . .... . . .. . . . . . . . . . .. . . . . 296 Sulphate . . . .. . . .. . . .. . .... . . .. . .. . . . . .. . . . . . . .. . . . . . . . . . . .... . . . . . .... . .. . . . 297 Parameter Correlation Analysis . . . .. . .. . . . . . . . . . . . . .... . .. .. . ... . . . . . . ... . . . . . . 303 Discussion .. . . . . . . . . . .... . . ... . ... . . .. . . . . . ... . . . . .. . . . .. . . . . . . . ... . . . . ... . . . . . . . . 306 6.4.3 6.4.4 6.5 Chapter 6 Summary & Conclusions . . . . ... .. . . . ... . . . . .. .. . . ... . .. . . . . . . . . .. . .. . . . . .. 308 7.0 7.1 7.1.1 7.1.1.1 (i) (ii) (ii) 7.1.2 7.1.3 Chapter 7 Plant Growth Trials Introduction . . .... . . . . . . . . . .. . . . . . .. . . .. . . . .. . .. . . . . . ... . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . 309 Materials . . . . . . . ...... . .. . . . . . .. . . ...... . . .. . .. . . . . . .. . . . . . . .. . . . .. . . . . . . .. . .. .. .. . . . . . . . . 311 Tailings Selection .. .. . .. . ... . . . . . . . . .. . . . .. .. .. . .. . . . . . . .. . . . . . . . ... . . . .. . . . . . . . 311 Characterisation of Glasshouse Trial Tailings . .. . . . .. . .. . . . . . . . . .. . . . 311 ICP-AES analysis . . .. . . . . . . . ... . . . . . . .. . . . . . . . . .. . . . ... . . ... .... , ......... 312 Acid Generating Potential ... . . .... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . 314 pH buffering capacity . . ... . .. .. . . .. ... . .. . ... . . . . . . . . .. .. . . . ... . . . . . . . . . , 315 Lime (Calcium Carbonate) .. . . . .. . . '" ......................................... 315 Composted Sewage Sludge . .. . . . . .. .. . . . . . . . '" .................. '" ........... 317 xii 7.2 7.2.1 7.2.2 7.3 7.3.1 7.3.2 Trial Methodologies . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 Water & Fertiliser Regime ...................................................... 318 Glasshouse Environment .... .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 319 Plant Species Selection ................................................................ 319 (i) (ii) (iii) Seed Germination TriaL . . . . . . . . . . . . .... . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 Trial Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . .. . . . . . 320 Germination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 Lead & Zinc Tolerance .................................................. 321 Copper Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Discussion of Germination Results . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 7.3.3 7.4 7.5 Sowing Rates & Dates ................................................................. 323 Monitoring of the Growth Medium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 324 pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 Electrical Conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326 Metal Concentrations ............................................................ 328 Seed Germination ............................................................... 329 7.5.1 7.5.2 7.5.3 7.5.4 7.6 7.6.1 7.6.2 7.6.3 7.6.3.1 7.6.3.2 7.6.3.3 Trial One - The Effects of Lime Application Rate & Plant Metal Tolerance . . . 332 7.6.4 7.6.5 7.6.5.1 7.6.5.2 (i) (ii) (iii) (iv) 7.6.5.3 (i) (ii) (iii) (iv) Aim . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . 332 Trial Design . . . . . .. ... . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 332 Trial Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 333 Determination of Liming Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 333 Lime Incorporation & Pot Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Sowing ..................................................................... 334 Hypotheses ....................................................................... 334 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Germination Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335 Yield Response ......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 336 Nil Lime Treatment . . . . . . . . . . . . .... . . .. . . . . . . . . . . . . . . . . . . .. . .... . . . . . . . . . 336 Yz pH-6 Treatment . . . . . .. . . . . . . . . .. . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 343 1 pH-6 Treatment . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 TAGP Treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 Plant Nutrition .. . . . .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360 Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . 361 Nitrogen . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . , ......... 363 Calcium & Magnesium .................................................. 364 Potassium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . ... 365 Trial One Conclusions . . . . . . . . . . . , .............................................. 365 7.6.6 7.7 7.7.1 7.7.2 Trial Two - Organic Matter Amendment . . . . . . . . . . . . . . . . . . . . . . . . . , .................. 368 7.7.3 7.7.3.1 7.7.3.2 7.7.3.3 7.7.4 7.7.5 7.7.5.1 7.7.5.2 Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . 368 Trial Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 368 Trial Methodology . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369 Sludge Addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 369 Liming Requirements & Addition . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 371 Sowing . . .. . .. ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 Hypothesis . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371 Germination Results . . . . . . . . . . . . . . . . . . .... . . . . . . . .. . . . . . . . .. . . . . . . . . . . . . . . . . · · · · 372 Germination Trends on Non-Limed Treatments . . . . . . . . . . ... . .. . . . . . 372 Germination Trends on Limed Treatments . . . .... . . . . . . . . .. . . . . . . . . . . 373 7.7.6 7 .7 .6 . 1 (i) (ii) (iii) 7 .7 .6.2 (i) (ii) (iii) Xlll Dry Matter yields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 Non-Limed Treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375 pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Extractable Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380 Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382 Limed Treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390 pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 1 Extractable Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 Nutrients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399 Trial Two Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 7 .7 .7 7 .8 Chapter 7 Summary & Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 CHAPTER 8 Thesis Summary & Conclusions 8.0 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 8 . 1 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 8.2 Characterisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404 8 .3 Plant Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406 8 .4 Overall Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408 8 .5 Scope for Future Work at the Tui Mine Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 10 Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448 Table 2.1 Table 2.2 Table 3.1 Table 3.2 Table 3.3 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 IX List of Tables Typical water quality of the tailings leachates, the Tui stream both below and above (control) its confluence with the tailings leachates, the northern and southern branches of the Tunakohoia stream below the mine site and World Health Organisation (WHO) Guideline values for drinking water .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Water quality parameters for the Tui Stream and background values for six Coromandel streams (adapted from Livingston, 1987)....... 16 Kinetic Tests for the Prediction of Acid Mine Drainage (adapted from Ferguson , 1986 cited in Ritcey, 1989).................................. 84 Maximum recommended cumulative loading rates and maximum permitted total concentrations of heavy metals in sludge amended soils. Bracketed data represents the maximum recommended annual loading rate of heavy metal ................................................ 132 Selected grasses that have ecotypes exhibiting tolerance to metals found in non-ferrous metal mine spoils (Gemmell, 1977; Eslick, 1995) ... . . .. , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Selected properties of materials from the Tui mine site based on average values from duplicate analysis. See Appendix 4.l(C) for calculation of estimated sulphide percentages. ND = metal concentrations were not determined ................................ '" 147 Elemental concentrations of the Tui Mine tailings at preliminary sampling site S2 as determined by ICP-AES [mg/kg). Bracketed data represent Cu, Pb and Zn levels determined by F AAS and S concentrations determined by sodium bicarbonate fusion ............ 151 pH data and Acid Generating Potential (AGP) estimates [Mg CaC03/ 1000 Mg material] of the Tui dam materials based on (1) hydrogen ion liberation and (2) sulphur analysis ................... , ................... 154 Piezometric data for the Tui tailings dam. Readings are in mm and refer to the distance of the water-table below ground level at the site. 'Dry' indicates that the water-tables lay below 1000 mm ............ 159 Particle size determination of site 7 incremental samples. Analyses are presented as a percent of oven-dry sample. Bracketed data = grain-size range [mm] based on USDA classification ............................. 164 Penetration Resistance (PR) data from Sites 1 to 9. 0) = gravimetric water content; S = significant (P<0.05); PS = possibly significant (PO.l) .................................... 171 Table 4.7 Table 4.8 Table 4.9 Table 4.10 Table 4.1 1 Table 4.12 Table 4.13 Table 4.14 Table 4. 15 Table 4 . 16 Table 4.1 7 Table 4.18 x Average initial and final surface infiltration rates at grid sampling sites 7 and 9 based. The measurements were conducted in the field using ring infiltrometers. The standard deviation values presented are based on six replicate measurements made at each site . . . . . . . . . . . . . . . . . . . . . . . 172 Average bulk density values [Mglm3] of the Tui tailings at various depths. Values and sample standard deviations are based on 5 replicate cores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Flame type and lamp current settings generally used for atomic absorption spectrometry in the present study . . . . . . . . . . . . .. . . . . . . . . . . . . ' 180 Comparison of average F AAS values with CANMET Ore standard MP-la and CAAS Sulphide Ore-l rock standard . . . .. . . . . . . . . .. . . . . . . 182 Relationship between total (HFIHN03) metal concentrations versus 'pseudo total' (25% HNO}) ............................................... 183 Percentage total Cu, Fe, Pb and Zn extracted by Hydrogen Peroxide procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 186 Percentage of total metal extracted by O.lM HC} in site 7 incremental samples and grid samples . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 Average and range of Cu, Fe, Pb and Zn [mglkg] in grid samples extracted by 0.1M HCl and/or 25% v/v HNO}. Log-normally distributed metal concentrations are presented with standard deviation ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . 194 Student's I-test and F -test statistical data for grid sampling sites 1 to 12. HoI is null hypothesis that there is no difference in the variance of surface and subsurface samples and Ho2 is the null hypothesis that there is no difference between the means of the data sets (0-100 mm and 100-200 mm samples) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . 195 Relationship between 0-100 and 100-200 mm tailings samples at grid sampling sites 1 to Where Ho is the null hypothesis that there is no relationship between metal concentrations in the respective layers. (r = Pearson correlation coefficient; degrees of freedom = 10) ... ......... 197 Average pH values and range for grid sampling sites 1 to 12. pH values of the grid samples and bulk samples (bracketed data) are also included for sites 7 and 9 .................... ............................... 216 Pearson's correlation coefficient (r), Student's t-test and F-test results based on pH data for 0-100 mm and 1 00-200 mm depths (grid sampling sites 1 to 12) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . 217 Table 4.19 Table 4.20 Table 4.21 Table 4.22 Table 4.23 Table 5 .1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Xl Acid potentially released from the oxidation of 1.0 mole of chalcopyrite and pyrite (Bruynesteyn & Hackl, 1982) ................. 223 AGP data [Mg CaC0311 000 Mg of tailings] based on the assessment of sites 1-12. Presented data is based on 48 analyses conducted at sampling sites 1 to 12 ...................................................... 224 Average chalcopyrite and pyrite concentrations in the surface 200 mm of the Tui Mine tailings. t Refers to the percentage of the tailings T AGP attributable to pyrite oxidation. Presented data is based on 48 analyses conducted at sampling sites 1 to 12 ........................... 227 Sulphide content and metal ratios of tailings from Site 7 estimated from Fe and eu extractions ............................................... 229 Neutralising Potential of sampling site 7 tailings. The results are the averages of 4 replicate analyses conducted on each sample .......... 234 Total elemental concentrations of selected metals and minerals detected by XRD in unseparated Tui mine tailings. The minerals are listed in alphabetical order and abbreviated as follows ............... 241 Heavy mineral fraction (HMF) and XRD identified minerals. Minerals are listed alphabetically and abbreviated as follows ................... 244 Solubility and density of selected sulphate minerals (Aylward & Findlay, 1974) ............................................................... 246 Tailings Samples Examined by SEM & EDS ........................... 252 EDS analysis ofTui tailings particles .................................... 263 Galvanic series of some common sulphide minerals (adapted from Natarajan, 1990) ............................................................ 263 Design Treatments for Lime Incubation Trial .......................... 270 pH buffering capacity, TAGP (theoretically acid generating potential) upper and lower limits and various liming rates based on the 1 pH-6 buffering rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 Percentage water content (oven dry basis) of the amended tailings at 0.05 bar pressure. 70% (0.05 bar) values and the approximate water content used in the Lime Incubation Trial are also provided ......... 273 Sampling times for Lime Incubation Trial .............................. 275 Table 6.5 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 7.9 Table 7.10 Table 7.1 1 XII S ignificance of Pearson's Correlation Coefficient (bracketed data) of measured parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. 3 04 Elemental concentrations within the Trial Tailings as determined by ICP-AES. Bracketed data represent 25% Nitric acid extractable Cu and Fe concentrations determined by FAAS . . . . . . . . . . . . . . . . . , . . . . . . . . . . 3 1 3 Summary of the properties of the North Shore Sewage Sludge (Limpus-O'Reilly, 1 997) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 7 Percentage water content (oven dry basis) of the amended tailings at 0.05 bar pressure. 80% values and the approximate water content . . 3 1 8 Description of plant Taxa used in Glasshouse Trials . . . . . . . . . . . . . . . . . . 323 Seed viability and sowing rates for each of the plant taxa. Seeds per pot = 1 00/ proportion viable seed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 Sowing, termination and sampling dates for Glasshouse Trials . * The fertility measurements include Olsen P, CEC, exchangeable Ca, K, Mg and Na. t Metals include O.lM HC} and 0.02M CaCh extractable As, Cd, Cu, Fe, Pb, Mn and Zn levels and 1M KCl and O.02M CaCh extractable Al concentrations . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 The pH and electrical conductivity (EC) of the tailings treated with different amendments in the presence of plants (sown pots) and in the absence of plants (control pots). The average pH values of the 1 60 sown pots are also presented along with standard deviation values. Average pH calculations: n = 24 and n = 8 for glasshouse Trials 1 and 2 respectively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 27 Design Treatments for Trial One . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333 Soil fertility parameters of Trial One control pots determined at sowing (Week 3) and at the end of the growing period (Week 1 1). Medium values for New Zealand soils are based on the data of Blakemore et al. (1987) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340 Nutrient content and estimated total nutrient uptake by herbage from selected treatments. sat. = satisfactory; v.1ow = very low; nd = values not determined due to insufficient dry matter product . . . . . . . . . . . . . . . 3 6 1 Trial Two design and amendment rates. Liming rates: nil = nil lime treatment, 1 pH 6 = amount of lime required to raise the pH of the sludge amended tailings to approximately 6 .0 . Fr = Festuca rubra (var. Merlin), Lc = Lotus corniculatus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 Table 7.12 Table 7.13 Table 7.14 Table 7.15 X1l1 Lime and sludge application rates for Trial Two treatments. The area rate assumes amendment incorporation to 200 mm depth and a tailings' bulk density of 1 .3 Mglm3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '" 3 7 I Soil fertility parameters of Trial Two non-limed control pots determined at sowing (Week 3) and at the end of the growing period (Week 1 1). Medium values for New Zealand soils are based on the data of Blakemore et al. ( 1987). *Values are expressed as mg/kg. t Values expressed as meql l OOg .nd = not determined . . . . . . . . . . . . . . . . . 3 8 3 Average pH of sown pots at week 1 1 (termination) of trial . . . . . . . . '" 39 1 Soil fertility parameters of limed control pots determined at sowing (Week 3) and at the end of the growing period (Week 1 1) . Medium values for New Zealand soils are based on the data of Blakemore et al. ( 1987) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '" . . . . . . . . .. . . . . . . . . . . . '" . . . . . . . . . 400 Figure 2.1 Figure 3.1 Figure 3.2 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.1 1 Figure 4.12 XIV List of Figures Location of the Tui mine site, Te Aroha, New Zealand......... ...... 8 Economic importance of various mined materials in New Zealand in 1995 (source: Ministry of Commerce, 1996) ........................... 41 Infiltration rate, runoff and seepage for (a) bare tailings and (b) capped tailings. Material depths:- Soil layer = 1.0 m, Coarse layer = 0.3 m; Barrier layer = 1.0 m. (adapted from Bi Tech Publication, 1989 cited in Sengupta, 1993) ......................................................... 110 Schematic plan view of the Tui tailings dam indicating location sites of the preliminary sampling and piezometer installation. Grid sampling and piezometer sites coincide ................................. 144 (a) 'Total' copper and (b) zinc versus depth at each of the 3 preliminary sampling sites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 148 (a) 'Total' lead and (b) pH versus depth at each of the 3 preliminary sampling sites ............................................................... 149 Cross-sectional diagram of the Tui mine tailings profile at grid sampling site 7 ........................................................... '" 162 Tailings penetration resistance versus depth at specified grid sampling sites ........................................................................... 169 Tailings penetration resistance versus depth at grid sampling site 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 Schematic diagram explaining infiltration rate variability at grid sampling sites 7 and 9 ..................................................... 173 25% Nitric acid and H202 extractable (a) copper and (b) iron versus Hydrofluoric acid extractable concentrations ...... '" .............. , ... 184 25% Nitric acid and H202 extractable (a) lead and (b) zinc versus Hydrofluoric acid extractable concentrations .......................... 185 (a) 25% Nitric acid extractable Fe and Pb and (b) O.lM HCI extractable As, Fe and Pb versus depth ................................. 189 (a) 25% Nitric acid extractable Cu and Zn and (b) O.lM HCI extractable Cu and Zn versus depth ..................................... 190 (a) O.1MHCI extractable Cu versus 'Total' Cu and (b) Log O.1M Hel extractable Fe versus 'Total' Fe .......................................... 192 Figure 4.13 Figure 4.14 Figure 4.15 Figure 4.16 Figure 4.17 Figure 4.18 Figure 4.19 Figure 4.20 Figure 4.21 Figure 4.22 Figure 4.23 Figure 4.24 Figure 4.25 Figure 4.26 Figure 4.27 ----------- ---- xv (a) Log O.lM HCl extractable Pb versus Log 'Total' Pb and (b) 0.1M HCl extractable Zn versus Log 'Total' Zn ........................ ' " ... 1 9 3 (a) Log O.lMHCI extractable Cu versus 'Total' Cu and (b) Log O.lM HCI extractable Fe versus 'Total' Fe. Separate correlations are shown for individual depths (0� 1 00 mm and 1 00-200 nun) ................... 1 99 (a) O.lM HC} extractable Cu versus 'Total' Cu and (b) O.lM HCI extractable Fe versus 'Total' Fe. Separate correlations are shown for individual depths ( 0- 1 00 nun and 1 00-200 nun) ....................... 200 O.lM HCI extractable Cu concentrations [mglkgJ in (a) 0- 1 00 nun zone and (b) 1 0 0-200 nun zone ......... ' " .... ... .... . ........ .......... 201 O.lM HCl extractable Fe concentrations [mg/kg] in (a) 0- 1 00 nun zone and (b) 1 00-200 nun zone ... ' " .................................... 202 'Total' extractable Cu concentrations [mglkg] in (a) 0- 1 00 mm zone and (b) 1 00-200 nun zone ...................................... .......... 203 'Total' extractable Fe concentrations [mg/kg] in (a) 0- 1 00 mm zone and (b) 1 00-200 nun zone ... ' " .................................... ' " ... 204 'Total' extractable Zn concentrations [mglkg] in (a) 0- 1 00 mm zone and (b) 1 00-200 nun zone ..................... ' " ........................ 205 'Total' extractable Pb concentrations [mgikg] in (a) 0- 1 00 mm zone and (b) 1 00-200 mm zone ................ ....... . .... .. .. ................. 206 pH levels in (a) 0- 1 00 nun zone and (b) 1 00-200 nun zone . ........ 207 Total extractable sulphur concentrations versus depth at (a) grid sampling site 7 and (b) grid sampling site 9 ............................ 21 0 Phosphate extractable sulphate concentrations versus depth at (a) grid sampling site 7 and (b) grid sampling site 9 ............................. 2 1 1 Mean pH versus depth at grid sampling sites 7 and 9 .................. 2 1 3 Phosphate extractable sulphate concentrations and pH versus depth at (a) grid sampling site 7 and (b) grid sampling site 9 .................. 2 14 Log phosphate extractable sulphate concentrations versus pH at (a) grid sampling site 7 and (b) grid sampling site 9. Correlation was conducted both with and without the uppermost 3 surface samples (circled) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1 5 Figure 4.28 Figure 4.29 Figure 4.30 Figure 5.1 Figure 5.2 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Figure 6.7 Figure 6.8 xvi Tailings TAGP [MgCaC031l 000 Mg of tailings] in a (a) 0- 1 00 mm zone and (b) 1 00-200 mm zone. Based on tailings Cu and Fe content and assumes complete hydrolysis of ferric iron to ferrihydrite . . . . . . .225 Liming requirements of the Tui tailings versus depth at site 7 as determined by: 48 hour pH buffering incubation ( 1 pH-6), hydrogen peroxide oxidation at 1 50 microns (HP- I SO) and 2000 microns (HP- 2000), stoichiometry based on sulphur analysis (Sulphide-S) and Fe/Cu analysis (TAGP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Binuclear Fe Complex (source Cotton & Wilkinson, 1 972) . . . . . . . . . . 233 (a) XRD trace of 0- 1 00 mm tailings sample and (b) XRD trace of 500- 600 mm tailings sample containing detectable pyrite (P) as well as kaolinite (K) and quartz (Q) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Graph of heavy mineral distribution [%] versus depth [mm] at site 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 (a) pH buffering curves for tailings at various sampling depths and (b) Relationship between profile depth and the amount of lime required to attain a pH of 6. This figure clearly demonstrates that the pH buffering capacity of the tailings tended to increase with depth . . . . . . . . . . . . . . . . . 276 (a) Graph comparing Theoretical Acid Generating Potential (TAGP) and the rate of lime required to amend the pH of the tailings to 6 .0 ( 1 pH-6 rate). The 8 pH-6 rate represents highest rate of lime used in trial. (b) Acid generated by the incubation samples over 76 week period as calculated by hydrogen ion liberation . . . . . . . . . . . . . . . . . . . . . . . 277 pH versus time in (a) 0- 1 00 mm sample, (b) 1 00-200 mm sample and (c) 200-300 mm sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280 pH versus time in (a) 300-400 mm sample, (b) 400-500 rum sample and (c) 500-600 rum sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1 O . lMHCI extractable Cu versus time (a) 0- 1 00 mm sample, (b) 1 00- 200 rnm sample and (c) 200-300 rnm sample . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 O . 1MHCI extractable Cu versus time (a) 300-400 rnm sample, (b) 400-500 mm sample and (c) 500-600 rum sample . . . . . . . . . . . . . . . . . . . . . . 285 O. 1MHCl extractable Fe versus time (a) 0- 1 00 mm sample, (b) 1 00- 200 rnm sample and (c) 200-300 rum sample . . . . . . . . . . . . . . . . . . . . . . . . . . .286 O. l M HCI extractable Fe versus time (a) 3 00-400 rnm sample, (b) 400-500 mm sample and (c) 500-600 rum sample . . . . . . . . . . . . . . . . . . . . . 287 Figure 6.9 Figure 6.10 Figure 6.1 1 Figure 6.12 Figure 6.13 Figure 6.14 Figure 6.15 Figure 6.1 6 Figure 6.1 7 Figure 6.18 Figure 7.1 Figure 7.2 XVll O . 1MHCI extractable Zn versus time (a) 0- 1 00 nun sample, (b) 1 00- 200 mm sample and (c) 200-300 mm sample . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 88 0 . 1MHCl extractable Zn versus time (a) 300-400 mm sample, (b) 400-500 nun sample and (c) 500-600 mm sample . . . . . . . . . . . . . . . . . . . . . 289 O . l M HCI extractable Pb versus time (a) 0- 1 00 nun sample, (b) 1 00- 200 mm sample and (c) 200-300 mm sample . . , ' " . . . . . . . . . . . . . . . . . . . 290 O. lMHCl extractable Pb versus time (a) 3 00-400 nun sample, (b) 400-500 nun sample and (c) 500-600 mm sample . . . . . . . . . . . . . . . . . . . . . 29 1 Variation of pH and O. lM HCl extractable metal concentrations with depth and liming rate at week zero. LSD = Depth/Rate Interaction (5%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298 Variation of pH, sulphate and 0 . 1M HCl extractable metal concentrations with depth and liming rate at week 1 6. LSD = Depth/Rate Interaction (5%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .299 Variation of pH and O. l M HCI extractable metal concentrations with depth and liming rate at week 32. LSD = DepthlRate Interaction (5%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 00 Variation of pH and O . l M HCI extractable metal concentrations with depth and liming rate at week 48. LSD = DepthlRate Interaction (5%) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 0 1 (a) Cumulative percent versus particle size of the six 1 00-mm incremental tailings samples obtained from grid sampling site 7. (b) Correlation of hydrogen ion liberation versus silt from control treatment samples . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' " . . . . . . . . . . . . . . . . . . . 302 (a) Correlation of O . l M HCI extractable sulphate versus Fe and (b) Correlation ofPb versus sulphate (week 1 6 data) . . . . . . . . . . . . . . . . . . . . . . 305 (a) The pH buffering capacity of the tailings amended with four rates of composted sewage sludge. The parallel nature of the best fit curves indicates that the pH buffering capacity of the tailings was not markedly improved by the addition of sludge in the short-term. (b) Relationship between the rate of sludge application and the amount of lime required to attain a pH of 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1 6 The changes in electrical conductivity of various glasshouse trial treatments. (a) Trial One: limed treatments. (b) Trial Two: non­ limed/sludge amended treatments and (c) Trial Two: limed ( 1 pH-6 rate)/sludge amended treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 0 Figure 7.3 Figure 7.4 Figure 7.5 Figure 7.6 Figure 7.7 Figure 7.8 Figure 7.9 Figure 7.10 Figure 7.1 1 Figure 7.12 Figure 7.13 Figure 7.14 XV1l1 The changes in sulphate levels within various glasshouse trial treatments. (a) Trial One: limed treatments. (b) Trial Two: non­ limed/sludge amended treatments and (c) Trial Two: limed ( 1 pH-6 rate )/sludge amended treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 1 Percentage germination at various periods after sowing at (a) 1 pH-6 liming rates and (b) TAGP liming rates. All plant taxa failed to germinate on the nil and 12 pH-6 1ime treatments . . . . . . . . . . . . . . . . . . . . . . 3 37 Changing pH of non-sown ' control pots' treated with various rates of lime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339 Changes in metal concentrations as extracted by O . I M HCl Ce), 0 .02M CaCh (0) and 1M KCl (+) for nil lime treatment during the growing period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346 Changes in metal concentrations as extracted by O . l M HCI Ce), 0 .02M CaCh (0) and 1M KCl (+) for 12 pH-6 lime treated tailings during the growing period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 Dry matter yields of different plant taxa at various liming rates . LSD1 = least significant difference for non-leguminous plants (5%); LSD2 = least significant difference for L. corniculatus only C 5%) . . . . . . . . . . . . 349 Changes in metal concentrations as extracted by O . l M HCI Ce), 0 .02M CaCh (0) and lM KCl (+) for 1 pH-6 1ime treated tailings during the growing period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 Changes in metal concentrations as extracted by O . I M HCl Ce), 0 .02M CaCh (0) and 1M KCl (+) for TAGP lime treated tailings during the growing period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 54 Concentrations of various metals extracte.d by 0 .02M CaCh at different liming rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355 Concentrations of various metals extracted by O . l M HCl and 1M KCl (aluminium only) at different l iming rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356 Relationships between 1M KCI and 0.02M CaCh extractable AI and between O . 1M HCl and 0 .02M CaCl2 extractable Cu, Fe, Pb and Zn [mglkg] . The 3 data points for each treatment originate from analysis of unsown control pots conducted at week 3 (sowing), week 7 and week 1 1 (termination) ofthe trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 Percentage germination at various periods after sowing in (a) non­ limed, sludge amended tailings and (b) sludge amended tailing limed to pH 6.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 74 Figure 7.15 Figure 7.16 Figure 7.17 Figure 7.18 Figure 7.19 Figure 7.20 Figure 7.21 Figure 7.22 Figure 7.23 Figure 7.24 Figure 7.25 Figure 7.26 Figure 7.27 Figure 7.28 XiX Dry matter yields of Festuca rubra CMerlin) and Lotus corniculatus in (a) non-limed tailings and (b) tailings limed initially to pH 6.0 . . . . . . 376 Changing pH of sludge treated tailings (a) without lime addition and (b) initial limed to pH 6 .0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Changes in metal concentrations as extracted by O . l M HCI ce), 0.02M CaCb Co) and 1M KCI C +) for nil treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ' " 384 Changes in metal concentrations as extracted by O . l M HCl ce), 0.02M CaCh Co) and 1M KCI (+) for 1 1 0 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385 Changes in metal concentrations as extracted by 0 . 1 M HCI (e) , 0.02M CaCh Co) and 1M KCI (+) for 220 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386 Changes in metal concentrations as extracted by O. I M HCI (e), 0.02M CaCh (0) and 1M KCI (+) for 440 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387 Concentrations of various metals extracted by 0 .02M CaCh at various rates of sludge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . " . . . . . . . . . . . 388 Concentrations of various metals extracted by O . IM HCl at various rates of sludge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 Changes in metal concentrations as extracted by O . l M HCl (e) , O.02M CaCh (0) and 1M KCl C+) for limed, nil treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393 Changes in metal concentrations as extracted by O . I M HCl ce), 0.02M CaCh Co) and 1M KCl C +) for limed, 1 10 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394 Changes in metal concentrations as extracted by O . l M HCI (e), O.02M CaCh Co) and 1M KCl C+) for limed, 220 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . ' " . . . . . . . . . . . . . . . . 395 Changes in metal concentrations as extracted by O . 1 M HCI ce), O.02M CaC12 (0) and 1M KCI (+) for limed, 440 Mglha sludge treated tailings during the growth period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 Concentrations of various metals extracted by O .02M CaCh at various rates of sludge treatments which have been limed to pH 6 .0 . . . . . . . . . 397 Concentrations of various metals extracted by O . I M HCl at various rates of sludge treatments which have been limed to pH 6 .0 . . . . . . . . . 398 Plate 2.1 Plate 2.2 Plate 2.3 Plate 3.1 Plate 3.2 Plate 3.3 Plate 3.4 Plate 4. 1 Plate 4.2 Plate 5.1 Plate 5.2 Plate 5.3 - -------- ---- List of Photographic Plates xx Panoramic view of the Tui tailings dam and Hauraki Plains (August 1 993) . . . . . . ' " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 0 The Tui tailings dam and surrounding vegetation. The mine adits are located in dense bush in the centre-top and left-hand-side of the photograph (August 1 993). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 Silt trap situated below the front (western) face of the Tui tailings dam (August 1 993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 7 Metal contaminated drainage emanating from an adit at the Tui mine site. Precipitated ferric iron compounds are seen coating the rocks and mining debris (August 1 993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1 Rock and mining debris choking the upper reaches o f the northern Tunakohoia Stream above the tailings dam (August 1 993) . . . . . . . . . . . 6 1 Surface erosion of Tui tailings embankment following intense rainfall (August 1 993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 View of the front (western face) of the Tui tailings dam which is covered with a 1 5-20 em layer of what is now effectively 'crushed quartz' (August 1 993) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Tui mine tailings exposed in cross-section at grid sampling site 7. This photograph highlights the layered nature of the tailings . . . . . . . . . . . . . 1 60 Debris and degradation at the Tui mine site. The foundations of the Tui ore mill and processing site are visible in the top right hand comer of this photograph. Debris from this site including wood and angular pebbles were encountered at grid sampling site 1 (August 1 993) . . . 1 65 A photo-mosaic of six microphotographs correlating to each of the sampling depths from 0-600 mm. Particles within the top 200 mm of the profile generally exhibit a craggy and irregular morphology (C). The occurrence of cube-shaped pyrite minerals (P) and angular grains increases with depth (Magnification 1 80 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 A SEM image showing three cuboidal pyrite grains (P) isolated in the 500-600 mm zone. Their well defined cube shape indicates that they are resistant to the micro-environmental weathering conditions prevailing at this depth (Magnification 470 x) . . . . . . . . . . . . . . . . . . . . . . . . . 255 A magnified view of central pyrite cube in Plate 5 .2 confirms that the surface is largely unweathered and that it is partially coated with plates of kaolinite (Magnification 4700 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 Plate 5.4. Plate 5.5 Plate 5.6 Plate 5.7 Plate 5.8 Plate 5.9 Plate 5.10 Plate 5.1 1 Plate 6.1 Plate 7.1 Plate 7.2 XXI Surface etching of a pyrite grain (centre) indicates that limited biochemical oxidation is occurring at 300-400 nun depth (Magnification 480 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256 A magnified view of the pyrite grain in Plate 5 .4 indicates the presence of several small hexagonal etch holes (E). Like the smaller holes, the larger non-symmetrical etch pit (N) is thought to have resulted from biological and/or chemical oxidation of the pyrite along a chemically active zone (Magnification 4800 x) . . . . . . . . . . . . . . . . . . . . . . 256 An 'aggregate grain' comprised of sub-cuboidal crystallites isolated in the 0- 1 00 nun depth zone. These distinctively shaped grains are more prevalent in the surface 200 mm of the profile (Magnification 470 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 A magnified view of the sub-cuboidal crystallites constituting the aggregate grain in Plate 5 .6 (Magnification 9200 x) . . . . . . . . . . . . . . . . . . . 259 An 'aggregate grain' isolated in the 0- 1 00 mm depth zone exhibiting mamillary growth habit (Magnification 460 x) . . . . . . . . . . . . . . . . . . . . . . . . . 260 SEM image of a quartz particle 'contaminating' the heavy mineral fraction of the tailings. Quartz particles such as these constitute > 90% of the tailings and are characterised by conchoidal fracturing and angular morphologies (Magnification 360 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 A high resolution image of plate-like kaolinite located on the surface of a mineral grain in the 500-600 nun sample (Magnification 4700 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1 SEM microphotograph of the surface of a sulphide mineral isolated in the Tui ore sample (Magnification 9500 x) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Lime Incubation Trial. Polyethylene bags containing lime amended tailings (January 1 995) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Trial One 4 weeks after sowing. Comparative growth of the various taxa on uncontaminated soil is seen in the 4 plastic trays in the upper left hand comer of the photograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334 This photograph which was taken about 8 weeks after sowing compares the growth of two cultivars of Agrostis capillaris (tenuis) on tailings amended with a rate of lime equivalent to 1 6 .5 Mglha. The metal-tolerant cultivar (Parys Mountain) located on the right-hand-side of the late was found to out-yield the non -metal-tolerant cultivar (Egmont) by a factor of more than 1 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 Plate 7.3 Plate 7.4 Plate 7.5 Plate 7.6 Plate 7.7 Plate 7.8 Plate 7.9 - ----- -- - XXll This photograph compares the growth of two varieties of Festuca rubra gwwn on tailings amended with 1 6. 5 Mg CaC03, 8 weeks after sowing. The metal-tolerant variety (Merlin) on the right-hand-side of the plate was found to produce about 4 times more dry matter than the non-metal tolerant cultivar (Lobi) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 The Lotus corniculatus seen growing in this photograph (pots 53-5 5) are about 4 weeks old and are clearly performing poorly. Grown in tailings limed at a rate of 1 6.5 Mg/ha the plants showed signs of chlorosis and a generally stunted yellow appearance, symptoms consistent with N deficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352 Trial Two: 4 weeks after sowing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 Trial Two: 8 weeks after sowing and immediately prior to herbage harvest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370 This photograph taken 4 weeks after sowing compares the growth of Frubra (Merlin) grown on tailings amended with 4 different rates of sludge; nil, 1 1 0, 220 and 440 Mg/ha. Sludge rate increases from left to right . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 This photograph was taken 4 weeks after sowing and compares the growth of Festuca rubra (Merlin) on tailings amended with 4 rates of sludge (nil, 1 1 0, 220 & 440 Mglha; left to right) and 4 rates of lime ( 1 6 .5 , 1 0. 1 , 8 .5 & 5 . 1 Mg CaC03/ha; left to right) . . . . . . . . . . . . . . . . . . . . 392 This photograph was taken 4 weeks after sowing and compares the growth of Lotus corniculatus on tailings amended with 4 rates of sludge (nil, 1 1 0 , 220 & 440 Mglha; left to right) and 4 rates of lime ( 1 6.5, 1 0. 1 , 8 .5 & 5 . 1 Mg CaC03/ha; left to right) . . . . . . . . . . . . . . . . . . . 392 xxiii Appendices Appendix 4(A) Total Sulphur Determination by Sodium Bicarbonate Fusion . . . . . . . 448 Appendix 4(B) Calcium Phosphate Extractable Sulphate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449 Appendix 4(C) Acid Generating Potential (AGP) Based on Sulphur Analysis . . . . . . . 449 Appendix 4(D) HFIHN03 Total Metal Extraction Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . .450 Appendix 4(E) 25% (v/v) Pseudo-Total Metal Extraction Procedure . . . . . . . . . . . . . . . . . . 45 1 Appendix 4(F) O. lMHCI Metal Extraction Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1 Appendix 4(G) 30% H202 Metal Extraction Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1 Appendix 4(H) Theoretical Acid Generating Potential (TAGP) Method Working Examples . . . . . . . , . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . 452 Appendix 4(1) Hydrogen Peroxide Oxidation AGP Method . . . . . . . . . . . . . . . . . . . . . . . . . . . 456 Appendix 4(J) Neutralising Potential Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457 Appendix 5(A) EDS spectrum for a pyrite cube such as those visible in Plate 5 .2 . The 3 major peaks are associated with the S and Fe constituents of iron pyrite. Very little ' contamination' is evident . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 Appendix 5(B) EDS spectrum for an 'aggregate grain' such as that seen in Plate 5 .6 . The major peaks are associated with AI , Fe, S i , S and possibly Pb . Trace quantities of Cu were also identified in the sample . . . . . . . . . . . . 460 Appendix 5(C) EDS spectrum for the quartz grain shown in Plate 9. The major peak is that of Si, however, minor quantities of AI, Fe and Sn were also detected in the sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 1 Appendix 5(D) This EDS spectrum identifies the presence of K in the tailings probably in the form of mica. The occurrence of a large Fe peak and relatively small S peak indicate that the Fe is probably in the form of oxides or alumino-silicate clays as opposed to sulphides . . . . . . . . . . . . . 462 Appendix 5(E) Low magnification ( 1 80 x) EDS spectrum for the 0- 1 00 mm tailings sample. A significant amount of Ba was identified in 3 EDS taken in the surface horizon. Pb concentrations also tended to be higher relative to the 500-600 mm sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463 xxiv Appendix 5(F) Low magnification ( 1 80 x) EDS spectrum for the 500-600 mm tailings sample. The 3 EDS samples showed little or no Ba and unlike the surface sample Cu, W and Zn were detected in all 3 low magnification EDS analyses at this depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464 Appendix 6(A) Liming Requirement Based on pH Buffering Procedure . . . . . . . . . . . . . . 465 Appendix 6(B) Acid Generation Estimates Based on Hydrogen Ion Liberation . . . . . . 466 Appendix 7(A) Seed Germination Trial Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467 Appendix 7(B) Trial Two: Organic Matter Rate Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . 468 CHAPTER ! Thesis Introduction 1 .0 Introduction '1 walk in the way of Nature, till anon I shall fall and be at rest, yield up my breath to that element from which I draw it day by day, and sinking into that same earth that gave my father his seed, my mother her blood, and my nurse her milk; that earth that has given me food and drink for many a year, and borne with me while I trampled her underfoot and abused her at will. ' 1 Perhaps never in the history of humanity have the words of the Roman philosopher Marcus Aurelius 1 ( 1 2 1 - 1 80 AD) been so poignant. Environmental degradation of soils resulting from phenomena such as mining, deforestation, intensive chemical-based agriculture, salinisation, over irrigation and desertification is occurring worldwide at a catastrophic rate (Hillel, 1 992). This degradation is ultimately being driven by a continuous and accelerating increase in population and resource exploitation. Thus, at the very time the human population is proliferating, it is becoming increasingly evident that our poor treatment of the Earth and its soil resources is diminishing its capacity to support us (Hillel, 1 992). Although on a global scale environmental degradation resulting from both active and abandoned mine sites is small, it is an extremely serious problem facing many regions around the world and its localised impact on communities and the environment can be devastating. Bradshaw & Chadwick ( 1 980) estimated that about 0.2% of the landmass in the US and 0 .4% of the landmass in Great Britain has been seriously impacted by past and present mining operations. In Canada, ever-expanding tailings dams cover more than 50,000 ha of land (Ritcey, 1 989). Compared to many countries New Zealand has been little impacted by mining activity. Whilst mining was a mainstay of the New Zealand economy during the early days of European settlement, it is estimated that less than 0 .02% of the country' s land area has been disturbed by mining. Most of this 1 Cited in Bradshaw & Chadwick ( 1 980). 2 degradation was caused by 'eady' alluvial gold mining operations in the South Island on land generally hosting soils of moderate to low versatility (Fricker, 1 986). Over the past few decades, the introduction of modem mining techniques and high­ capacity processing plants has seen a dramatic increase in the potential for mining operations to damage the environment. Rising gold and oil prices during the 1 970' s and 1 980's resulted in a worldwide resurgence in mining activity for both coal and gold. Many of the larger current mining operations involve the daily displacement of many thousands of tonnes of rock and the generation of huge volumes of wastewater and tailings. Fortunately attitudes towards mining and mining practices have also changed substantially over the same time period (Ritcey, 1 989). As awareness of the potential impacts of mining has grown, so too has the pressure from the public and regulatory agencies on mining companies to minimise environmental impact. Much of the impetus for new mining legislature has stemmed from peoples' realisation that the world' s resources are finite and that soil is not a readily renewable commodity but rather a complex biological ecosystem and ultimately a precious resource (Bradshaw & Chadwick, 1 980). New legislation and cooperative efforts between industry, regulatory authorities and research organisations have resulted in considerable progress being made towards creating a more ' environmentally responsible' global mining industry. Mining legislation in most countries now requires mining companies to implement measures before, during and after operations to safeguard the environment. Mining practices, for example, now commonly try to ensure that the exposure of sulphide bearing wastes is minimised, fugitive dust, noise pollution and silting of water ways is controlled and that the burial of soil materials and the disruption of communities, natural ecosystems, water courses and aquifers is avoided where ever possible. Thus, nowadays most mining operations (particularly in the First W orId) are conducted in an environmentally and sociably 'responsible' manner. The same, however, can not be said of the majority mining operations of yesteryear or of some operations currently being conducted in Third W orId countries. ----�� ---- - -- 3 Many mining operations conducted prior to the early to mid 1 970's were often carried out with little or no regard for potential environmental impact. The lax attitudes towards rectifying environmental damage or reclaiming mine sites were driven, at times, by greed and complacency but more often as a result of lax governmental legislation which was based on ignorance of the processes and potential problems associated with mining. The Tui base-metal mine site near Te Aroha, New Zealand provides a graphic example of the potential hazards that are associated with many metalliferous and coal mining operations and forms the focus of this thesis . Major mining operations were commenced at the site during the 1 960's at a time when legislation in New Zealand governing mining was antiquated and primarily aimed at the promotion of such activities rather than at environmental protection (Fyson, 1 99 1 ). Few environmental constraints were thus in place to curb the activity of the mining company operating at Te Aroha or to protect the integrity of the surrounding aquatic and terrestrial ecosystems. Whilst the Tui mine site exhibits a unique character in terms of its mining history, operational practices, geology, climate, hydrology, topographical setting, flora and fauna, many of the problems encountered at the site are common to abandoned mine sites where sulphidic minerals have been mined and/or exposed during operations. The most serious problems at the Tui mine site stem from the exposure of finely ground tailings which contain high concentrations of remnant sulphide minerals capable of generating acidity and liberating base-metals. Not only has sulphide oxidation acidified the tailings and prevented vegetation from colonising the tailings dam, it has also resulted in the degradation of surface and ground water via the generation of severe Acid Mine Drainage (AMD). The Tui mine site is without question a 'thorn in the side' of the New Zealand mining industry. It is continually cited by environmentalists and anti-mining groups alike as a graphic example of why the mining of 'non-essential' commodities should be discontinued in the Coromandel and elsewhere within New Zealand. Few people will argue, however, that the notoriety of the Tui mine truly reflects the impact that the site has had on the environment. When looked at in context of the country as a whole the effects of the Tui mine are really very localised and effect few people. 4 In spite of its comparatively small regional impact, it is clear that something must be done to reclaim the Tui mine site. Drainage from the site continues to pollute local streams and at least ethically impact the lives ofTe Aroha people. Whilst some consider the tailings dam to be a curiosity, most consider it to be an 'eyesore' in the middle of scenic native bush; an unsightly reminder that mining should never have been conducted in such a fragile (and in many senses hostile) environmental setting. It is evident that after more than 20 years mother nature and time have made little headway in ameliorating environmental degradation at the Tui mine site. Unless geological time is left to play its role, human intervention will be necessary to remedy or at least 'kick start' recolonisation of vegetation at the site. The science of land reclamation in New Zealand is still very much in its infancy. Whilst many researchers have gained considerable expertise with respect to implementing reclamation strategies for current, well-planned mining operations, expertise and research into the reclamation of highly degraded abandoned mine sites is relatively scarce. Sites where the processes of sulphide oxidation and acid mine drainage are well established are notoriously difficult to reclaim. This is particularly the case where few resources are available for the implementation of tried and proven (but often prohibitively expensive) reclamation strategies. This study was seen as a 'golden' opportunity to increase the reclamation knowledge base within New Zealand and to further global understanding of mine reclru'1lation in different and unique environmental settings . 1 . 1 General Objectives of Study Prior to this study, no comprehensive examination had been conducted to investigate the option of revegetating the Tui mine tailings. It was envisaged that this study would:- • Determine the potential for establishing a vegetative cover on the Tui tailings. • Provide groups potentially interested in reclaiming the Tui mine site, or similar sites, with useful data that would form the basis of well informed and practical reclamation strategies. • Elucidate the processes and problems associated with the revegetation of polysulphidic mining wastes . 1.2 Specific Objectives of Study Specifically this study was aimed at- 5 • Characterising the physical and chemical properties of the Tui tailing to identifY what limitations are currently prohibiting plants from colonising the tailings. • Assessing the total and labile metal concentrations of the tailings and their Acid Generating Potential (AGP) to identifY tailings material most suited for use as a growing medium for plants. • IdentifYing the primary mineral constituents of the Tui tailings and the depth and degree to which the tailings have oxidised. • Investigating the feasibility of establishing plants in Tui tailings amended with lime and/or organic wastes. • IdentifYing ' optimal' amendment rates for preventing reacidification of the tailings and facilitating satisfactory plant growth. • Assessing the performance of several plant species, including a legume and two commercially cultivated metal-tolerant grass varieties, to identifY potentially suitable taxa for revegetating the tailings. 1 .3 Implementation of Objectives The objectives of this study were implemented by the combined use of:- • Field and laboratory-based analysis and experimentation. • Plant growth trials conducted on disturbed tailings samples In the controlled environment of a glasshouse. • A lime incubation trial designed to monitor the effectiveness of lime amendments at preventing reacidification. 1 .4 Thesis Layout This thesis comprises eight chapters including the present introductory chapter. The following chapter (Chapter 2) provides a description of the information relevant to the experimental site such as its history, geological setting and climate. Chapter 3 reviews literature pertaining to land reclamation and outlines techniques and cases studies of mine reclamation practices commonly used both here in New Zealand and abroad. Chapter 4 reports on the results of physical and chemical characterisation of the tailings 6 and discusses factors pertaining to the tailings' plant growth characteristics. Chapter 5 presents the results of XRD and SEM work undertaken to elucidate the mineralogy of the tailings and morphological and geochemical trends with depth. Chapter 6 reports the results of a lime incubation experiment designed to assess the reactivity of tailings treated with various rates of lime in terms of their potential to generate acidity. The penultimate chapter (Chapter 7) reports on the results of two plant growth experiments that were conducted simultaneously to investigate the efficacy of lime and organic amendments to ameliorate the tailings plant growth properties. Finally, Chapter 8 summarises the findings and conclusions of the research presented in this thesis and discusses the opportunity and potential direction for further research at the Tui mine site in particular and at other sites similarly impacted by mining activity and geochemical weathering processes. 7 CHAPTER 2 Tui Mine Site - Study Area 2.0 Introduction The aim of this chapter is to describe the study area in terms of its location, vegetation, climate, hydrology, geology, mineralogy, soils and history. Previous research investigating the impact of the Tui mining operation on surface and ground water quality is summarised and discussed in section 2.4. Further discussion of the Tui mine site' s impact on the environment is presented in Chapter 3 (section 3 .2.2.8). This chapter discusses each of the physical characteristics of the study area and concludes with a brief section discussing the past laws under which the Tui Mine operated, together with a brief overview of the political issues currently pertaining to the site including ownership and responsibility. 2.1 Location \ Physiography The Tui mine site (NZMS 260 T 1 35 1 5055) is located on the NW flank of Mount Te Aroha, approximately 3 km NE of the township of Te Aroha (Figure 2 . 1 ) .This small rural township is nestled at the base of Mt. Te Aroha which forms a part of the Kaimai ranges and dominates the landscape, towering some 952 m above the Hauraki plains. Te Aroha is located approximately 1 1 0 km SE of Auckland city and 25 km from the coast on the southern periphery of the Hauraki Goldfield, an area that was extensively prospected and mined during the late 1 9th and 20th Centuries (Main, 1 979). Te Aroha township (population c. 3 ,500) acts as a service centre for the Thames Valley region whose primary industries include high production dairying, maize, pip and berry fruit production (McLauchlan, 1 989). The physiography of the Hauraki Goldfield is characterised by the generally rugged and steeply-dipping topography of the Coromandel and Kaimai Ranges. These Ranges are dissected by a multitude of streams and creeks, including the Tui and Tunakohoia streams that flank the Tui mine site. Both of these streams have been severely impacted by past mining activity at the site (Tay, 1 980; Pang, 1 992) . The Tui and Tunakohoia River 37 0 33' 1 km Railway Road \ \ \ \ \ \ \ \ \ , \ Tui Stream Te Aroha Tui Mine -¢:J Site Mt. Te Aroha 952 m Figure 2.1 Location of the Tui mine site, Te Aroha, New Zealand. 00 9 streams are tributaries of the Waihou River, a slowly flowing river that meanders its way north into Thames Estuary along the eastern margin of the Hauraki Plains. The low-lying Hauraki Plains form a part of a large tectonic feature, about 20 to 30 km wide and over 220 km long, called the Hauraki Depression (Hochstein & Nixon, 1979). North of Waitoa the Hauraki Plains generally stand only a few metres (m) above sea­ level. Before extensive flood control measures were implemented the Plains were regularly inundated with flood waters from local streams, the Waihou and Piako rivers, and occasionally by tidal waters. South of Waitoa the surface elevation of the land gradually increases to a few tens of metres and the alluvial surface becomes increasingly dissected and eroded (Hochstein & Nixon, 1 979). The northern tract of the Hauraki Depression is occupied by the Firth of Thames, a shallow marine body typically less than 35 m deep. The Tui Mine consists of a series of underground shafts and adits which lie at altitudes of between 500 and 600 m. Although the land adjacent to the mine adits is degraded, the most obtrusive feature of the site and the focus of this study is a rudimentary dam containing an estimated 1 00,000 m3 of base-metal tailings (Plate 2 . 1 ). The tailings dam is located at an altitude of approximately 350 m on a gently sloping area of land that was previously known as 'Fern Flats ' (Cochrane, 1 969). Although the mine tailings have largely de-watered, saturated zones still persist adjacent to the confining valley slopes where the water table remains at, or close to, the tailings surface throughout much of the year! . 2.2 Vegetation Due to the varying topography and the effects of introduced fauna, past mining and logging operations, considerable variety is seen in the age and type of vegetation on Mt. Te Aroha (Bollard, 1 99 1 ). While a number of vegetation types, including exotic forestry, pasture and even small areas of virgin forest can be found on the mountain' s slopes, the majority of the land above 200 m altitude, including that surrounding the tailings dam, is covered with dense, secondary native bush. 1 See discussion in Chapter 4 (section 4 .2. l . l ) 1 0 I I Hardwood specIes, including Kamahi (Weinmannia racemosa), Rata (Metrosideros robusta), Tawa (Beilschmiedia tawa) and Tawari (lxerba brexioides) dominate the upper canopy, while smaller trees including Five-finger (Pseudopanax arboreus), Hangehange (Geniostoma rupestre), Mahoe (Melicytus ramiflorus), Mamaku (Cyathea medullaris), Pate (SchejJlera digitata), Pigeonwood (Hedycarya arborea) and large­ leafed coprosmas make up the under-story. Epiphytic Astelia species and climbers such as keke, creeping rata, mangemange and supplejack are also present. The ground cover is dominated by mosses and ferns because of the moist conditions prevailing on the Mountain (Bollard, 1 99 1 ). At higher altitudes the canopy more commonly consists of Tawari, Tawherowhero (Quintinia serrata), Toatoa (Phyllocladus glaucus), Kamahi and Dracophyllum species, with the podocarps Miro (Prumnopitys ferruginea) and Totara (Podocarpus totara) as emergents (Salmon, 1 980; Bollard, 1 99 1 ). Toetoe (Cortaderia spp.) is one of the most prevalent plant species found growing on the abandoned roadways and processing areas of the Tui mine site. The site ' s moist conditions and low fertility suits the growth of most of toetoe species which are often found colonising roadsides and stream banks throughout New Zealand (Pollock, 1 986) . Small clumps of Toetoe have also established on the tailings dam at sites where decaying vegetation (tree trunks/limbs) has provided a suitable micro-environment for its growth. Small trees and shrubs including Manuka (Leptospermum scoparium), Puka (Griselinia lucida ) and Toro (Myrsine salicina) are similarly seen growing in organic debris at a few locations upon the tailings (Plate 2.2). 2.3 Climate Although there is no specific climatic data for the Tui mine site, reliable meteorological data has been collected at Te Aroha township since 1 888 (NZMS, 1 983) . Te Aroha lies 3 km SW of the Tui mine site at an altitude of 1 8 m (approximately 330 m below the Tui tailings dam). Mean annual rainfall at the town is 1 459 rum and is spread over an average of 1 35 rain-days each year (NZMS, 1 983) . Rainfall is distributed relatively evenly throughout the year, although an average of 79 mm more rainfall is received during July than January. Heavy deluges of rain frequently occur throughout the year. 12 -- Plate 2.2 The Tui tail ings dam and surrounding vegetation. The mine adits are located in dense bush in the top centre and left-hand-side of the photograph. (August 1 993). 1 3 The mean annual air temperature at Te Aroha is 14.5 °C. Mean monthly temperatures vary from 1 9.5 °C in January to 9.3 °C in July. Temperature extremes of 35.0 °C and - 6. 1 °C have been recorded in the months of January and August respectively. The town receives an average of 20 ground frosts and 1 8 days of heavy winter fogs each year. Occasional snowfalls occur on the upper slopes of Mt. Te Aroha which are subjected to fog on an average of 1 90 days per annum (Bollard, 1991 ). Average relative humidity at the township varies from around 74% in January to 86% in May (NZMS, 1 983) . The prevailing wind in the area is a westerly which at times reaches gale force along the ridges of the Kaimai Ranges. 2.4 Site Hydrology The impact of mining activity on the ground and surface water quality at the Tui mine site has been well studied over the past two decades. Scientific interest in the quality of the Tui and Tunakohoia Streams stemmed largely from the fact that the streams formed an integral part of Te Aroha's tOVV11 water supply between 1 968 and 1 979 (Fyson, 1 99 1 ). Distinctive iron-staining and colloidal deposits in the Tui Stream can be traced from where the tailings discharge enters the stream through to its confluence with the Waihou river (Tay, 1 980). Concentrations of Fe, Mn and Cd in water collected from the former intake screen of the Tui Stream were found to exceed World Health Organisation (WHO) limits for drinking water by 56, 35 and 1 .5 times respectively (Tay, 1 980; World Health Organisation, 1 984a, 1 984b). The pH of the water was found to vary 2 to 3 .5 pH units below the recommended WHO levels. Although town water is no longer obtained from the Tui and Tunakohoia streams, there is concern that shallow ground water used for domestic water supply in the vicinity of the tailings darn may be contaminated with heavy metals (Pang, 1 995). Several studies have been undertaken at the Tui mine site investigating various aspects of the site ' s hydrology and the impact of past mining activity on water quality (Tonkin & Taylor, 1 974; Ward et ai., 1 976; Tay, 1 980; Pang, 1 992). The impact of the Tui mining operation on the aquatic environment was also investigated as a part of two comprehensive studies conducted in the late 1 980's by the now defunct 'Water and Soil Directorate' of the Ministry of Works and Development (Smith, 1 986; Livingston, 1 987). Research indicates that 'background' concentrations of Cd, Cu, Pb and Mn in 14 ground and stream water in the vicinity of the mine site are naturally elevated due to base-metal mineralisation in the country rock (Pang, 1 995). Even water taken from stretches of the Tui Stream not effected by mining and from the south branch of the Tunakohoia stream was found to contain concentrations of Cd and Pb exceeding WHO guidelines for drinking water (Table 2. 1 ) . The effect of N orpac' s past mining operation has been to exacerbate the problem of heavy metal contamination of surface and groundwater in the vicinity of the Tui Mine. This is evidenced by the extremely poor quality of water in the north branch of the Tunakohoia (northern Tunakohoia) Stream and in the Tui Stream downstream of its confluence with the tailings stream (Table 2 . 1 , Figure 2 . 1 ) . The mine adits and the tailings impoundment are the two primary sources of contaminated water in the study area. The surface water draining from the abandoned mine adits is intercepted largely by the northern Tunakohoia Stream. Although drainage from the mine adits is generally alkaline (PH 7.4 to 7.7), the water contains high concentrations of Cd, Mn, Pb and Zn (Pang, 1 995). Whilst this mine drainage does not exhibit acidic pH, it has appreciable mineral acidity associated with dissolved metals and is potentially as hazardous as the acidic drainages emanating from the tailings (Chapter 3 .4). Groundwater flow modelling suggests that the northern Tunakohoia Stream may also receive drainage from the tailings dam, however, the majority of this highly contaminated water is believed to flow into the Tui Stream and into the groundwater via fractured andesitic bedrock (Pang, 1 992, 1 995). Elevated levels of heavy-metals in the northern Tunakohoia Stream has rendered the stream largely devoid of life . Dilution by 'uncontaminated' water of the south branch of the Stream, however, largely mitigates these adverse environmental effects downstream of the confluence (Pang, 1 995). The Tui Stream is substantially more acidic and generally hosts higher concentrations of Fe and Mn and lower concentrations of Cu, Pb and Zn than the northern Tunakohoia Stream (Table 2 . 1 ; Livingston, 1 987; Pang, 1 995). This trend may reflect the predominance of acid generating iron pyrite (FeS2) in the Tui tailings. Arsenic levels in the Tui Stream were found to be lower than in many other streams contaminated by Saruple pH As Cd Cu Fe Mn Pb Zn Reference Location [/lg/I] [rug/I] [rug/I] [rug/I] [rug/I] [rug/I] [rug/I] Tailings Leachates 2.7 - 4 .0 - 0 .35 - 0 .75 0 .84 - 9. 1 5 737 - 906 98 - 1 02 0 .80 - 2 .68 224 - 302 Tay ( 1 980) 2 .6 - 3 .2 0 .2 1 - 0 .56 0 .07 - 3 . 1 3 1 1 8 - 3 1 1 26 .5 - 39 .0 1 . 30 - 2 .45 70 .0 - 1 1 3 Pang ( 1 992) Tui Stream 4.5 - 5 . 1 - 0 .02 - 0 .03 0 .04 -0.27 3 . 7 - 1 8 . 3 1 .74 - 5 . 1 - 1 1 . 0 Tay ( 1 980) 6.7 - < 0 .05 < 0.02 0 .55 0 .42 < 0 . 1 0 1 .0 1 Pang ( 1 992) 5 .9 - 6 .4 1 . 0 - 1 . 5 ::; 0 .02 0 .04 - 0.23 - - 0 .02 -0.06 2.08 - 1 7. 0 Livingston ( 1 987) � Tui Stream (control) 7. 1 -7 . 8 - ::; 0 .0 1 ::; 0 .03 <0.03 - 0 . 1 5 <0.0 1 0 .02 - 0 .50 0.02 Tay ( 1 980) 6 .5 - < 0 .02 < 0 .0 1 0 . 1 0 < 0.05 < 0. 1 0 0 . 1 0 Pang ( 1 992) Nth. Tunakohoia 7.4 - 7 .7 - 0 . 0 1 - 0. 1 3 0 . 0 1 -0.05 <0 .04 - 0 . 1 0 <0 .0 1 - 0 .74 0 .04 - <0. 1 0 1 . 59 - 1 6. 2 Tay ( 1980) 7 .5 - <0.05 - 0 . 1 4 <0.02 - 0 . 1 1 0 .20-4.05 <0.09 - 1 . 1 0 0 . 1 1 - 0 .5 1 2 .30 - 1 6 . 0 Pang ( 1 992) Sth. Tunakohoia 6.7 - 7 . 5 - 0 .0 1 - 0 .02 0 . 0 1 - <0.03 0 .03 - 0 .06 0 .0 1 - 0 .05 <0.03 - <0. 1 0 0 .0 1 - 0 . 50 Tay ( 1 980) 6 .9 - < 0 .05 0 .02 0 . 1 0 < 0 .09 < 0 . 1 0 < 0 .09 Pang ( 1 992) WHO 6.5 - 8 . 5 50t 0.005t 1 . 0* 0 . 3* 0. 1 * 0 .05t 5 .0* (WHO 1 984a1b) Guidelines Table 2.1 Typical water quality of the tailings leachates, the Tui Stream both below and above (control) its confluence with the tailings leachates, the northern and southern branches of the Tunakohoia Stream below the mine site and World Health Organisation (WHO) Guideline values for drinking water. � Unfiltered water samples were used in this study. t Health guidelines for inorganic constituents. * Aesthetic guidelines for chemical constituents that may affect the quality of drinking water though seldom found in toxic concentrations. 16 mining activity in the Coromandel region (Livingston, 1 987). Research indicates that the high concentrations of metals in the lower reaches of the Tui Stream originate from acid mine drainage (AMD) emanating from the Tui tailings dam (Tay, 1 980). An overview of the generation of acid from sulphide oxidation and the formation of contaminated mine drainages and AMD is discussed in Chapter 3 (sections 3 .3 & 3 .4). Due to its topographically low position, the tailings dam and pond is thought to be a local discharge area of up-gradient groundwater. Groundwater flow modelling and water level contours suggest that about 200 to 450 m3 of leachates percolate into the groundwater from the tailings dam each day (Pang, 1 995). Leachates generated by groundwater saturation of the tailings and by rainfall infiltration collect in a silt trap to form a small pond below the tailings dam (Plate 2 .3) . The water draining from the pond flows into a small stream containing diverted spring and surface water from the tailings and adj acent land. The resulting stream has approximately a third of the flow of the Tui Stream and intersects the larger stream about 300 m downstream of the tailings dam (Tay, 1 980). The inflow of this highly contaminated water into the Tui Stream not only increases the water's metal content but also substantially elevates the stream's acidity, sulphate content and total hardness, whilst reducing its alkalinity (Table 2.2). A large difference between total hardness and alkalinity is often indicative of AMD­ impacted streams (Livingston, 1 987; Ritcey, 1 989). This difference arises due to the consumption of the water's alkaline anions (eg. carbonate and bicarbonate) by introduced acid. The alkalinity is reduced relative to the total hardness (calcium and magnesium ion content) by an amount equivalent to the quantity of acid neutralised (Livingston, 1 987). Table 2.2 Water quality parameters for the Tui Stream, and background values for six Coromandel streams. (Adapted from Livingston, 1 987). Total Hardness Alkalinity Sulphate (g/m3 CaC03) (g/m3 CaC03) (g/m3 S04) Tui Stream 69.8 1 . 1 90.2 Background 1 6.9 1 4.6 6.4 1 7 Plate 2.3 Silt trap situated below the front (western) face of the Tui tail ings dam (August 1 993). 1 8 Biologically, the Tui Stream is one o f the most degraded streams in the Coromandel Region as indicated by the extremely low diversity and numbers of macro-invertebrates in its waters (Livingston, 1 987). A survey conducted on the stream in the early 1 980's found it to be devoid of fresh water fish (Penny, 1 983) . Although the survey was able to identifY past mining activity as the general causative factor in biological impoverishment of the stream, it was unable to identifY the specific factor or factors responsible. It was recognised that it may have resulted from one of several chemically interdependent parameters including toxic metal concentrations, low pH conditions and the deposition of excess fine sediments and/or precipitates . Some correlative evidence did indicate, however, that high metal concentrations in the stream (particularly Zn) may be causing toxicity (Livingston, 1 987) . Unless control measures are undertaken to mitigate sulphide oxidation at the Tui mine site, AMD is likely to continue impacting adjacent streams and groundwater in the vicinity of the dam for many hundreds of years. Pang ( 1 995) estimates that more than 1 6 tonnes of various heavy metals will b e released from the tailings into groundwater and streams over next few millennia. The release of such large quantities of metals, including Cd, Cu and Pb into the environment is of great concern due to the potential health risks these metals pose to humans and due to the adverse impact they are currently having and will continue to have on the aquatic system. 2.5 Geology The Tui Mine is located within the Hauraki Goldfield, a region that hosts some 47 identified epithermal gold-silver and base-metal deposits (Christie & Robinson, 1 992) . The goldfield is approximately 200 km long and up to 40 km wide, extending from Great Barrier Island in the north down to Te Puke and across to Te Aroha in the south­ west (Brathwaite et al. , 1 987) . The region hosts two dominant types of mineralisation, disseminated 'porphyry' type and ' epithermal ' vein type (Livingston, 1 987). Although both kinds of mineralisation originate from hydrothermal activity, the majority of the economic deposits in the region (including the Tui Mine) are epithermal. This type of mineralisation is characterised by discrete, well-defined deposits that often produce high grade are. The Tui 's Champion vein reportedly yielded about 28 g of gold per ton of ore 1 9 (Robinson, 1 974), a value typical of the region's epithermal gold deposits that usually yield between 3g and 30g of gold per m3 (Livingston, 1 9 87). The potential economic importance of the Tui mineral deposits, which lie on the south­ western margin of the Haurah Goldfield, was realised soon after their discovery in 1 884. Henderson & Bartrurn ( 1 9 1 3) were among the first to publish information relating to the Tui mine site ' s geology and history. Rising base-metal prices in the early 1 960' s prompted renewed interest in the area, resulting in the initiation o f several studies that investigated the geology and mineralogy of the Tui deposits. Research includes that of Schofield ( 1 967), Cochrane ( 1 969), Weissberg & Wodzicki ( 1 970), Wodzicki & Weissberg ( 1 970), Adams et ai., ( 1 974) and Robinson ( 1 974) . A useful review of the mine's early history and geology is included in Williams ( 1 965). Research investigating the geology and mineralisation of the wider region includes that of Boyle ( 1 979), Hochstein & Nixon, ( 1 979), Brathwaite et at. , ( 1 987), and Christie & Robinson, ( 1 992). Brathwaite & Christie ( 1 996) have recently completed a geological report and map of the Waihi area which summarise the region'S stratigraphy, geological structure and economic geology. 2.5.1 Tectonic Setting The igneous rocks of the Hauraki regIOn were produced during the Miocene and Pleistocene by intrusive and volcanic activity occurring in a NNW trending continental margin arc. This arc is believed to have been associated with a westward dipping subduction zone that once existed east of the Coromandel peninsula (Skinner, 1 986). Cole & Lewis ( 1 98 1 ) suggest that the Hauraki arc may have evolved into the presently active Taupo-Hikurangi subduction system. The main phase of andesitic volcanism in the Coromandel Peninsula is thought to have occurred between 1 0 and 1 5 m.y. ago (Livingston, 1 987), contemporaneously with the intrusion of high level porphyry plutons and with minor eruptive activity at the southern end of the Northland Peninsula (Cole & Lewis, 1 98 1 ) . The change to rhyolitic volcanism that took place approximately 6 m.y. ago in the Hauraki Volcanic Region is apparently related to tensional faulting in a marginal basin (Cole & Lewis, 1 98 1 ) . 20 2.5.2 Regional Setting The Hauraki Goldfield coincides with a continental margin volcanic zone (Coromandel Volcanic Zone) that consists of thin alternating beds of complexly folded Mesozoic greywackes and argillite, unconformably overlain by Miocene to early Quaternary sub­ aerial, andesite-dacite-rhyolite volcanics and volcaniclastics (Brathwaite et al., 1 987) . Jurassic greywacke and argillites constitute the basement rocks of the whole region and have been folded approximately along the axis of the peninsula (Robinson, 1 974). Extensive block faulting is evident in the form of normal faults within the basement rocks along NW to NNW and NE, ENE trends that gives rise to horst and graben structures (Brathwaite et al., 1 987). The whole region has been tilted to the SE, exposing the Jurassic basement rocks in areas north of Thames (Hochstein & Nixon, 1 979). The overlying volcanic rocks, which range in composition from andesites to rhyolites with rare basalts, become progressively thicker and younger to the south where they merge with the Quaternary volcanic deposits of the Taupo Volcanic Zone (Wodzicki & Weissberg, 1 970). In order of decreasing age the more important volcanic units of the Hauraki Goldfield include; the Beesons Island Volcanics (dominantly andesitic Miocene rocks, c. 1 6 m.y.); the Whitianga Group and Minden Rhyolite (rhyolitic rocks of Pliocene age, c. 3-5 m.y.) and the Late Pliocene Omahia Andesites (Adams et al., 1 974). Although regional propylitic2 alteration of all rock types is ubiquitous, mineralisation in the Hauraki Goldfield is confined largely to the propylitised andesites of the Beesons Island Volcanics (Robinson, 1 974) . Far fewer vein deposits are found within the hydrothermally altered basement rocks and rhyolitic rocks of Pliocene age and mineralisation is absent in the Omahia Andesites (Adams et ai., 1 974). Immediately to the east of the Kaimai and Coromandel Ranges lies the Hauraki Depression which is a large tectonic feature about 20-30 km wide and at least 220 km long. The Depression is bounded on the western and southern sides by the Hunua and Hapuakohe Ranges respectively. Within the Hauraki Depression the basement 2 Propylitic alteration refers to weak and often pervasive alteration of rocks caused by hot meteoric waters. It is characterised by a slight reduction in rock strength and by the development of secondary minerals such as chlorite, calcite and pyrite. 2 1 Greywacke rocks that form the surrounding ranges are overlain by Quaternary and Tertiary sediments with a combined thickness of up to 3 km (Hochstein & Nixon, 1 979). Based on the striking similarities of the Hauraki Depression to other well studied continental rifts, Hochstein & Nixon (1 979) postulate that the southern part of the Hauraki Depression is an active rift formed after the collision of the Indian and Pacific Plates. Active faulting and shallow earthquake activity in the Southern part of the Depression support this theory. 2.5.3 Local Geological Setting The Tui Mine occurs in extensively hydrothermally-altered volcanic rocks. The most common rock type is massive andesite, although in places the lodes also dissect andesitic breccia and tuffaceous lithologies (Cochrane, 1 969; Wodzicki & Weissberg, 1 970) . K-Ar dating indicates that the andesitic host rocks are of Miocene age and were erupted approximately 1 6.2 m.y. ago (Adams et al. , 1 974). Schofield ( 1 967) mapped the country rocks of the Tui Mine collectively as the Te Aroha andesites, a sub-unit of the Beesons Island Volcanics (Schofield, 1 967; Robinson, 1 974). In locations where the andesite is relatively unaltered the rock consists of phenocrysts of plagioclase, augite and hypersthene in a fined-grained crystalline ground mass with a pilotaxitic texture. The ground mass contains moderate quantities of microscopic quartz (Wodzicki & Weissberg, 1 970). The andesite is overlain by a 3m bed of fine-grained, moderately well layered tuff that has been so strongly hydrothermally altered that no primary minerals remain. Overlying the tuff is an 8 m bed of rhyolitic or dacitic ignimbrite. This deposit contains numerous fragments of argillite, volcanic rocks and flattened pumice, and again is highly altered (Wodzicki & Weissberg, 1 970) . Above the ignimbrite lies a quartz-bearing volcanic rock, thought to be a dacite or quartz andesite. To the south of the Ruakaka lode, a strongly altered quartz-bearing volcanic breccia probably of dacitic or quartz andesite origin, mantles the underlying units (Wodzicki & Weissberg, 1 970). 22 2.6 Mineralogy of the Tui Ore The Tui Mine deposits are typical of the epithermal base-metal ore deposits that occur as lodes in fissures and faults in hydrothermally altered Tertiary andesitic and pyroclastic rocks (Boyle, 1 979). The Tui Mine consists of two major mineralised quartz reefs (the Champion and Ruakaka) which are up to 2 m and 8 m in width respectively (Robinson, 1 974). The reefs intersect in the upper levels of the mine but diverge at depth due to them dipping steeply in opposite directions. The average strike and dip of the Ruakaka and Champion reefs are N 80° E, 68° S and N 40° E, 64° NW respectively (Wodzicki & Weissberg, 1 970) . Surface outcropping of the reefs is extensive with the strikes of both reefs mapped for distances of over 500 m (Williams, 1 965). The reefs comprise a quartz-cemented wall-rock breccia containing a discontinuous zone of brecciated quartz cemented by sulphide minerals and a younger generation of quartz (Robinson, 1 974).The minerals constituting these lodes were deposited in fault controlled, open space filling type environments in which fissures formed and remained open for sufficient lengths of time to enable the formation of coarse-grained sulphide minerals (Wodzicki & Weissberg, 1 970). The ore at the Tui Mine is thus characterised by lenticular masses of sulphide minerals that are interspersed between barren portions of quartz and remnant andesite breccia (Robinson, 1 974). The ore (defined as sulphide mineralisation with total Cu + Pb + Zn greater than 5%) is generally coarse-grained, with 3 cm crystals of sphalerite and galena being common (Wodzicki & Weissberg, 1 970) . The sulphide rich zone that forms the ore bodies varies in thickness from approximately 0.05 m to exceptionally greater than 0.9 m and has an average width in the vicinity of 0.5 m (Cochrane, 1 969). The ore body in the Ruakaka lode is generally wider and more continuous than that of the Champion lode. Wodzicki & Weissberg ( 1 970) suggest that this may be due to the Ruakaka lode' s stronger normal faulting component and generally steeper dip. These factors are thought to have influenced the size and persistence of the fissures that were ultimately filled with epigenetic mineralisation (Wodzicki & Weissberg, 1 970) . 23 The Tui ore is essentially an aggregate of sphalerite (ZnS), galena (PbS), chalcopyrite (CuFeS2) and pyrite (FeS2) with subordinate marcasite (FeS2), and tennantite ((Cu, Fe) I 2As4S I 3) in quartz gangue (Williams, 1 965). Appreciable amounts of tetrahedrite ((Cu, Fe) 1 2Sb4S 13) and trace quantities of cinnabar (HgS) also occur in the ore (Wodzicki & Weissberg, 1 970). Late stage alteration minerals include chalcocite (CU2S), covellite (CuS), anglesite (PbS04) and cerussite (PbC03), but these generally occur in minor quantities only (Cochrane, 1 969). Sphalerite is the most abundant of the sulphide minerals in the deposits followed by galena, chalcopyrite and pyrite . Although pyrite is not the most abundant sulphide, unlike the other sulphide minerals that are restricted to the lodes, pyrite is also found widely disseminated in the altered wall-rocks (Wodzicki & Weissberg, 1 970) . The ore yielded an approximate grade of 1 6.7% zinc, 7 .0% lead and 0 .62% copper in addition to significant quantities of gold and silver (Wodzicki & Weissberg, 1 970). Quartz (Si02) is by far the most abundant gangue mineral in the lodes, however, well crystallised kaolinite (AhSi20s(OH)4) is also common throughout the ore body. Carbonate minerals are a minor component of the ore with only small quantities of calcite (CaC03) and siderite (FeC03) present (Cochrane, 1 969). In all, over 70 species of minerals have been identified in the ore and host rocks of the Tui Mine, making it one of New Zealand's "most prolific mineral localities" (Courtney et at., 1 990). Despite considerable variations in metal ratios within the different lodes, W odzicki & Weissberg ( 1 970) noted that textural relations between the minerals were remarkably constant, with both vein systems exhibiting the same minerals and paragenetic sequences. Mineralisation is believed to have occurred during two distinct phases. The first phase resulted in the formation of barren quartz veins varying in thickness from about 1 .5 to 5 m. These quartz veins were deposited along previously active fault planes and contained numerous angular shaped, highly altered wall rock inclusions. Renewed movement along the fault planes caused brecciation of the quartz and the rock wall inclusions, forming a porous zone containing numerous cavities. It was within this zone that the minerals of the second phase of mineralisation were deposited to ultimately form the Champion and Ruakaka lodes . Included in the minerals are galena, sphalerite, 24 tennantite and cinnabar as well as at least two generations of quartz, pyrite and chalcopyrite (Wodzicki & Weissberg, 1 970). 2.6.1 Origin of Sulphide Mineralisation Research indicates that the zoned alteration pattern In Tui Mine 's host rocks is genetically related to ore emplacement, as evidenced by the increasing concentrations of As Cu, Hg, Pb, Sb and Zn in the country rock with increasing proximity to the lode (Weissberg & Wodzicki, 1 970) . It is believed that the ore metals were added to the lodes and wall rocks by hydrothermal solutions ascending along the initially porous brecciated zones within the quartz veins (Cochrane, 1 969; Weissberg & Wodzicki, 1 970) . K-Ar dating of the hydrothermally altered andesites associated with ore formation indicates that mineralisation occurred at least 2 .6 m.y. but less than 7.0 m.y ago, contemporaneously with the Pliocene volcanism of the region. Adams et al., ( 1 974) suggest that the alteration most probably occurred between 2 .6 and 4.0 m.y. ago, with the date of 7 .0 m.y. being erroneously high due to the introduction of excess radiogenic argon during alteration. The Tui ore's relatively high base-metal sulphide content (> 1 0%) is believed to result f{,'m its deposition in a slightly deeper and higher temperature environment than usual (Brathwaite et ai., 1 987). Sulphur isotope studies indicate that the major sulphide minerals were deposited at temperatures of between 250 and 400 °C (Robinson, 1 974; Christie & Robinson, 1 992) . The exact origin of the mineralised solutions is not known. Cochrane ( 1 969) suggests that they may be related to the loss of volatiles during the late stages of solidification of a plutonic stock below Mt. Te Aroha. Similarly Weissberg & Wodzicki ( 1 970) suggest that the ore metals may have originated from some deep seated magmatic process, or alternatively from trace-metal depleted volcanic rock or basement greywacke at depth. Based on isotope data, Robinson ( 1 974) suggests that the Pb in the deposits originated from basement greywackes. There are two schools of thought as to the origin of sulphur in the deposits. Christie & Robinson ( 1 992) believe that the sulphur was derived from Tertiary volcanic magmas with slight crustal contamination, however this contrasts with previous interpretations that postulated that the sulphur was 25 hydrothermally leached from greywacke basement rocks (Robinson, 1 974; Robinson & Christie, 1 980). These differing theories suggest that the ongm of the hydrothermal solutions is complex. The water, trace elements, sulphur and ore metals forming the mineralised solutions may each have originated from a different source. The deposits of the Tui Mine may well represent an ancient geothermal system similar to the deeper parts of those presently active in the Taupo Volcanic Zone (Weissberg & Wodzicki, 1 970; Brathwaite & Blattner, 1 995). 2.7 Soils According to the New Zealand Genetic soil classification system the soils in the area of the Tui mine site are classified as Steepland soils related to brown granular clays (Taylor & Pohlen, 1 962; Wilson, 1 980). The soils belong to the Aroha steepland soil series and have formed under mull-forming dicotylous-podocarp forest. Near Te Aroha, the soils have developed on weakly argillised and propylised andesitic parent material at relatively high altitudes ( 1 20 m - 750 m) and under high rainfall (c.2000 mmlyr). Whilst the soil depth varies considerably with the steepness of the terrain, the soils typically consist of shallow forest litter (3 cm) and shallow topsoil ( 1 5 cm) horizons overlying 30 em of subsoil . The topsoils are usually dark brown in colour and exhibit clay loam texture and moderately to weakly developed medium nut and crumb structure. The subsoils typically exhibit clay loam to sandy clay loam textures and weakly developed medium blocky and nut structures. The soils are generally well to excessively drained and have low nutrient status (Wilson, 1 980). Their greatest limitations are their low nutrient status, steep slope and associated severe erosion risk. They are poorly suited to pastures and are generally used for timber production or catchment protection under native or exotic forest (Gibbs, 1 980; Wilson, 1 980). Under the recently developed New Zealand Soil Classification the soils developed upon relatively stable sites, with deeper profiles and strongly developed granular structures, would probably be classified as Acidic Orthic Granular Soils, whilst the shallower soils (profiles depth < 75 em) that have developed on steeper slopes, probably equate to Acidic Orthic Brown Soils (Hewitt, 1 993). 26 2.8 The History of mining at Te Aroha Mining has played an important role in New Zealand's development as a nation, as it not only provided valuable overseas revenue during the initial phases of European colonisation, but also helped to establish the pattern of settlement in New Zealand (Williams, 1 985) . The township of Te Aroha is one of many such settlements which sprang into existence largely as a result of mining activity in the Coromandel. Reports of gold in the Te Aroha region date back as far as 1838, when George White of Whitianga reportedly discovered auriferous ore in the ranges 'opposite' Te Aroha. It is believed that White was referring to the Hangawera Ranges. It was not until 1879, however, that any ' substantial ' discovery of ore was to be made. A prospector by the name of Hone Werahiko found a small 'rich' reef in a gully, immediately to the left of Bald Spur. The ore-bearing reef was found "just behind the present township, at a height of nearly 1000 ft [305m] above-sea-Ievel "(Henderson & Bartrum, 1 9 1 3). This discovery led to a gold rush to the locality but the district was virtually deserted within 6 months, as no workable claims were established (Downey, 1 935) . The most promising of the Te Aroha deposits was the Champion lode discovered by A. C. Comes in 1884. The lode lay in "inaccessible country near the Tui creek on the Karangahake range" (Salmon, 1 963). The Champion Mine was set up at the site and after several name changes, this mine was to eventually become known as the Tui Mine. The lode was rich in Pb, and was originally mined with the intentions of supplying Pb flux for smelting gold ore from the Waiorongomai Valley, approximately 7 km to the west. In the early 1880's, about 30 companies were formed on the Waiorongomai Valley goldfield and a large battery was built on the flats, at the entrance of the Valley, to crush ore from the different claims (Galvin, 1 906). The battery was commissioned by Messrs. Firth and Clark, and was described as being "one of the finest stamping-batteries in the colony" (Henderson & Bartrum, 1 9 1 3) . To enable the various claim-owners to supply ore to the plant, the Piako County Council constructed a 3 mile tramway at a cost of £ 1 9,000. In 1888 a £20,000 concentrating and smelting plant was erected. This plant was widely regarded as the "most up to date reduction works in the A ustralian 27 Colonies" (Henderson & Bartrum, 1 9 1 3) . Approximately one ton of Pb flux was required for every ton of Waiorongomai ore concentrate smelted and much of the substantial investment made in Waiorongomai Valley was based on the assumption that the Tui Mine could supply sufficient Pb for this purpose (Downey, 1 935) . Unfortunately it was soon discovered that the Tui ore contained as much Zn as it did Pb, rendering it "quite unsuitable" for use as a flux (Downey, 1 935) . The whole mining operation in the Waiorongomai Valley was described by Downey ( 1 93 5) as being "one long chapter of disaster for the many companies that ventured their capital into it. " The main reason for the large financial losses were due to the fact that many of the owners "had not taken the precaution to see that the underground appearances of the mine warranted the outlay" (Downey 1 935) . Although initial assays of the Tui and Waiorongomai ore proved highly satisfactory, the ore whilst valuable "was so complex that by the recognised method of treatment - crushing and amalgamation - very little of its valuable contents could be recovered " (Galvin, 1 906). The refractory nature of the Tui ore did not deter mine development and by 1889 three adit levels had been constructed and communication made between them on the main lode. An aerial tramway was under construction to take the ore 1 20 chain (= 2.5 km) down the mountain to the flats below (Carter & Pharo, 1 965). The earliest known owners of the Tui Mine were the 'Thames Lead and Silver Company' , the mine, however, changed ownership many times during its early days. When it was realised that the Tui ore was unsuitable as a source of Pb for the smelters, attention turned to the mine as a potential source of the noble metals, gold and silver. In 1888 1 1 tons of Tui ore were sent to England realising £ 6 1 0s per ton. This was intended to be the first batch of some 200 tons of ore bound for England, however, no further ore was sent due to the poor yields of bullion being recovered from the ore (Downey, 1 935) . By 1893 the Company operating the mine was facing financial difficulty, because they were still unable to find an economical means of processing the ore. As the Waiorongomai smelter had only managed to extract approximately 30% of the bullion from the Tui ore, so it was decided to wait for work to be completed on a new 28 processmg plant, based on the more economical 'cyanogen process ' . This process proved unsuccessful due to the Tui ore' s high chalcopyrite content which resulted in the excessive consumption of cyanide (Galvin 1 906). In 1896 the New Zealand Syndicate bought the lease and then in 1897 the mine changed hands once again with the Tui Gold-Mining Company acquiring the lease. By this stage five adit levels had been opened on, or near, the Champion lode and had been driven in for distances between 9 and 83 m (Downey 1 935) . The Tui Gold-Mining Company apparently undertook little work to further develop the mine. In 1899 a further attempt to process the ore, using the thermo-hyperphoric method3 developed by the Rev. Joseph Campbell, also failed to economically extract the precious metals (Henderson & Bartrum, 1 9 1 3) . When this process failed, the claim was abandoned until 1902 when a local syndicate took up the lease. The mine changed ownership several times after 1902, but little work was to be carried out until about 1960, when interest was once again sparked in the Tui Mine by steadily rising base­ metal prices. The potential of the Tui Mine as a source of Cu, Pb and Zn, rather than as a source of gold and silver was mentioned by Baker in 1930 (Hodder & Scadden, 1 98 1 ) but it would be more than 30 years until the mine would reopen. In 1 962 South Pacific Mines Ltd. acquired the lease to prospect and evaluate the area as a base-metal proposition. They subsequently reopened the lower 3 levels of the old workings and prospected the lodes using 'horizontal drives' . In 1966 Norpac Mining Ltd. was formed as a subsidiary of the Cable Price Downer Group, acquiring the assets of South Pacific Mines Ltd. (Cochrane, 1 969). In the same year Norpac was granted two mining licenses and developmental work was started on the mine, including the driving of a new low level cross-cut to provide access to the Champion and Ruakaka reefs (Bollard, 1 99 1 ). By the end of 1967 Norpac Mining Ltd. had opened a number of stopes to supply ore to a mill and flotation plant that was built adjacent to the then proposed tailings dam. The 3 The thenno-hyperphoric method refers to a process where the ore was heated to a temperature of 1 200 OF in the presence of water gas to create a free-mill ing are (Henderson & Bartrum, 1 9 1 3). 29 destination of the ore concentrate was the Mitsui Mining and Smelting Company of Japan. The ore concentrate was shipped out of New Zealand via the Port of Tauranga. Mining at Tui was undertaken by ' shrinkage stoping' , a method of underground mining where the roof of the mine is selectively caved in and used to work on as a floor, allowing progress to be made along the lode (Hodder & Scadden, 1 98 1 ). By January 1969 ore was being extracted and processed at the Tui Mine at the rate of 1 00 tons/day. After processing, 1 00 tons of ore would yield about 1 0 tons of concentrate, with a market value at the time of approximately $200(NZ) /ton (Cochrane, 1 969). During its heyday the mine employed 64 people and pumped some $250,000 in salaries into the local economy, annually. In addition to the Pb-Cu-Zn concentrate produced each year, small quantities of Cd, Ag and Au were also produced (McGrath, 1 973) . A total of 1 63 ,000 tonnes of ore was extracted from the Champion and Ruakaka reefs, yielding 1 3 , 1 59 tonnes of Zn concentrate, 7,755 tonnes of copper/lead concentrate, 2,220 02 (63 kg) of gold and 98,000 02 (2778 kg) of silver (Bollard, 1 99 1 ). There is some variation between production figures among different sources (McOnie, 1 987; Bates, 1 989). In 1971 mining operations were marred by the slumping of several thousand tonnes of mine waste into an adjacent stream following a rainstorm in which approximately 230 rnrn of rain fell in 24 hours (Fyson, 1 99 1 ). After several years of 'successful ' operation, falling Cu pnces and growmg environmental concerns about alleged high mercury levels in the Tui ore, prompted the Mitsui company of Japan to relinquish their contract when it expired in 1 972 (McGrath, 1 973). The last shipment of ore concentrate (3000 tons) was exported to Japan in July 1 973, after which time the ore was stock piled while new customers were desperately sought. Despite 1970 estimates indicating that some 250,000 tories of ore remained (sufficient to keep the mill operational until 1978) and grants totaling $4 1 8,000 were made available to help establish new buyers, none were found. By November 1973 only 1 4 people remained on the payroll and a receiver/manager was appointed for the Norpac Company (McGrath, 1 973). Finally in 1974 New Zealand' s only base-metal mine closed and all of its assets were sold. During its operation the mine accounted for 80% of all 30 New Zealand' s mineral exports and earned a total of $4 million in foreign exchange (McGrath, 1 973) . Mineral Resources (NZ) Limited purchased some of Norpac' s assets from the receiver in 1974. These included the Tui plant and mining licenses, however, exploration was not resumed in the area until 1983. A base metal and gold prospecting program was initiated that included "grid ding, soil sampling, geological mapping and channel sampling . . . " (Bollard, 1 99 1 ). In 1987 Mineral Resources subsidiary, Tasman Gold Development Limited, applied for a prospecting license for a 1 96 Ha site on the Northern slopes of Mt. Te Aroha, an area which included the old Tui Mine (the Tui Tailings Dam lay immediately to the east of the application area). Tasman Gold's primary mining interest in the site was as a source of gold, however i t also intended to continue the evaluation of the area as a potential source of Ag; eu, Pb and Zn. More than 30 official objections were lodged against Tasman Gold's application. The major points of concern raised by the obj ectors related to the detrimental effects on the environment caused by Norpac' s past activities and that further mining activity in the area would cause even more damage, primarily due to the presence of sulphides in the ore (Bollard, 1 99 1 ). Another major area of concern was the steep topography and land instability, especially when adverse weather conditions prevailed. Other objections were based on the ecological, historical and cultural value of the land, with particular reference to past Maori occupation (Bollard, 1 99 1 ) . In an "historic" ruling in July 1991 the Planning Tribunal quashed the intentions of the Mining Company by turning down its application. Upon deliberation of its verdict the Tribunal admitted, that based upon the evidence submitted during the hearing, carefully controlled prospecting " . . . would not be likely to have any major additional impact on the environment, nor place the Te Aroha town water supply . . . at risk:' (Bollard, 1 99 1 ) . The Tribunal, however, also held the view that " a commercially feasible mining operation would not be possible without grave risk of further unacceptable environmental impact" . Since the prospecting was precursory to mining, the Tribunal rejected the application because they believed that the environmental risk was 3 1 unacceptable, a risk it saw against an historical background of "a most regrettable, albeit instructive, legacy of environmental degradation". The decision was based on the AMOCO Minerals case in 1 982 where it was ruled that "it may be that the environmental importance of certain land is so high that it can be said immediately that, whatever the value of the mineral deposits in it might be , the land should never be disturbed by mining operations and that therefore not even a prospecting license should be issued' (Bollard, 1 99 1 ). In its ruling the Tribunal noted among other things; ( 1 ) the steepness of the area, (2) the potential for storm-induced debris movement, land slips and erosion, and (3) the potential risk of heavy metal contamination to Braeside Aquaria's water supply (Orsman, 1 99 1 ). Tasman Gold Development Ltd. had spent $802,000 in preliminary prospecting of the area prior to the Tribunals ruling (Bollard, 1 99 1 ) . An appeal was subsequently lodged against the Tribunals decision by the applicant, however, it was overruled and to date no further prospecting has taken place on Mt. Te Aroha. 2.9 Tui Mine Politics The Matamata-Piako District Council and the Department of Conservation (DOC) are currently the "reluctant" owners of the land upon which the tailings dam lies (WRCM, 1 99 1 ) . Any responsibility for the problems associated with the abandoned mine appears to rest with the land owners despite the fact that all revenues derived from the royalties went to the owner of the resource at that time, the Crown (Fyson, 1 99 1 ). Although most governments have vindicated themselves from any kind of responsibility for cleaning up the Tui mine site, in 1981 the National Government conceded somewhat by making an ex gratia payment of $25 ,000 to the Hauraki Catchment Board and $2,500 to the Te Aroha Golf club due to damages arising from the Tui tailings (New Zealand Herald, 1 98 1 ) . At the time many people believed the Government was trying to buy its way out of a "political storm" (Wallace, 1 98 1 ) . The reluctance of the Crown to take responsibility for reclaiming the Tui mine site stems largely from the liberal legislation that existed when the mine was operational. The Tui mining operation was established under the rules of the 'Wardens Court' , a system that in the mid 1 960 ' s saw its role mainly to promote mining. The Wardens 32 Court system predated the Water and Soil Conservation Act of 1 967 and the Mining Act of 1 97 1 , two acts which may have helped to prevent the environmental degradation that was to mar the operation. The one relevant piece of legislature that should have worked to protect the environment seems to have been totally ignored by Norpac (Fyson, 1 99 1 ). This was the Te Aroha Borough Water Supply Empowering Act of 1 962. This Act stated that . . . . . "no mining privileges shall be granted over, and no mine waste shall be discharged into the water supply streams or their tributaries" (Fyson, 1 99 1 ). At the hearings preceding the initiation of mining, the Hauraki Catchment Board made submissions and objected to the inadequacy of environmental controls being implemented by the mining company. Its views were overruled due primarily to the support the operation had from both the local community and the Government (Hansen, 1 989). The aforementioned acts have all been superseded by the Resource Management Act (RMA) and Crown Minerals Act (CMA) which were both enacted in 1 99 1 . The RMA has essentially tightened environmental laws on mining activity, laws that some considered to be already overzealous. Palmer ( 1 982) for example, expressed the view that . . . . " Considered as a land use, mining is given a prominence which it probably doesn 't deserve, and the applicants for mining licences are subjected to more vigorous procedures generally than other persons utilising land or resources ". Despite what some may consider to be tortuous legislation it is evident that the RMA will ensure that environmental degradation on a scale such as that seen at Te Aroha does not reoccur in New Zealand. The RMA has a new focus on the stewardship of the Country's physical and natural resources and is designed in part to safeguard "the life supporting capacity of air, water, soils and ecosystems" (RMA, section 5(2)b) . Therefore activities such as mining which may conflict with the RMA's objectives must now comply with much more stringent environmental controls. Safeguards have also been set up in the CMA to protect the environment and the rights of the landowners from the potentially adverse effects of mining associated activities. Recently the Tui mine site has been at the centre of renewed interest, as regional and local authorities and mining companies have been considering options for reclaiming 33 the site. A background paper was prepared by the Waikato t�egional Council in 1 99 1 which briefly discussed the history of the mine, responsibility for the site and possible remedial options (WRCM, 1 99 1 ). The Waikato Regional Council is currently setting up a technical group to look at cleaning up the Tui mine site. The technical group is likely to include representatives from the Matamata-Piako District Council and Department of Conservation (the joint landowners), the Waikato Regional Council, New Zealand's Mineral Industry, the Ministry of Commerce, the Te Aroha Mining Information Group and Tangata Whenua. The group 's main task will be to develop terms of reference and to explore options for the environmental stabilisation of the Tui Mine and process site (EWM, 1 995). Much of the impetus for the setting up of the working group has arisen from the interest that Waihi' s two currently operational gold mining companies (Coeur Gold NZ Ltd. and Waihi Gold Mining Company Ltd.) have shown in helping reclaim the site (Waihi Gazette, 1 995). These companies are undoubtedly keen to see the site rehabilitated as the continued bad press that the Tui Mine generates sheds ' contaminated' light on current mining operations within the region. 3.0 Chapter 2 Summary & Conclusions • Based on information presented in this Chapter it is clear that the Tui mine site is located within a unique and, in many respects, fragile environment. The surrounding terrain is steep and rainfall at the site is high and frequently intense. • The land and bush encompassing the Tui mine site forms an important wildlife and recreational area and is rapidly regenerating into an important tract of native forest. • The Tui mine tailings not only aesthetically mar the landscape but also form a source of highly contaminated water which continues to degrade stream, surface and ground water in the vicinity of the dam. • The Tui Mine has had a long and ' turbulent' history which appears set to continue until steps are taken to reclaim the site. Whilst responsibility for the reclamation of the Tui mine site currently lies with the Department of Conservation and the Matamata-Piako District Council (who own the site), many believe that the Crown should front the cost of a comprehensive reclamation program, as they set the conditions under which the mine operated and were the original recipients of royalties during the Tui Mine's operational years. 34 CHAPTER 3 The Reclamation of Mine Wastes 3.0 Introduction This chapter reviews the problems and practices of land reclamation paying particular attention to the reclamation of sulphide-bearing metalliferous tailings and waste rocks. A cursory review of the reclamation of waste materials generated at aggregate, alluvial gold, coal and titanomagnetite (irons and) mine sites is also given, however, as it is recognised that many of the reclamation techniques used to reclaim such materials are potentially useful for the rehabilitation of sulphide-bearing materials. The process of sulphide oxidation and its role in generation of metal contaminated mine drainage and Acid Mine Drainage (AMD) are reviewed, as is the means of predicting and controlling the Acid Generating Potential (AGP) of mine wastes. Particular reference is also given to lime (CaC03), composted sewage sludge and metal-tolerant plant species in terms of their ability to aid in the revegetation of sulphide-bearing mine wastes. 3.1 Definitions In this thesis the terms reclamation and rehabilitation have been used interchangeably. It is recognised, however, that some researchers distinguish between the terms. McQueen ( 1 982) for example, uses rehabilitation to describe the establishment of conditions for a 'new' specified, often productive land use which is substantially different from the original land use (eg. native forest� mine site � pasture), whilst he uses reclamation to describe the restoration of derelict land to a use and vegetative state similar to that existing prior to land dereliction (eg. pasture� mine site � pasture). In the context of this thesis revegetation is used to describe a specific form of reclamation which aims at establishing a vegetative cover on land that may or may not have been previously vegetated. The term restoration, which is commonly encountered in literature, refers to returning land to its previous topography and vegetation cover. This type ofrecIamation is often impractical and/or prohibitively costly (Fricker, 1 986). 3 5 The term tailings i s mostly used in this study to describe geological materials that have been finely crushed « 2 .0 mm) and from which metallic compounds have been extracted for the purposes of refining. In this review no attempt has been made, however, to differentiate between milled tailings and tailings derived from dredging operations which are not intentionally crushed during processing and can range in size from clay and silt-sized particles to boulders (>256 mm). 3.2 Mine Site Degradation & Reclamation Environmental degradation resulting from both active and abandoned mine sites is a serious problem in many regions throughout the world (Bradshaw & Chadwick, 1 980; Sengupta, 1 993). Although some degradation is obviously anticipated with an industry which is responsible for the mining and processing of an estimated 24 billion tonnes of minerals each year, many of the problems associated with degraded mine sites are the result of liberal laws which in the past placed few environmental controls on mining practices (Miller, 1 992). Continuing environmental damage arising from polluted waters and dispersal of contaminated solid wastes is a common feature of many abandoned mines in North America, Europe, Australia and numerous other countries including New Zealand (Ritcey, 1 989; Johnson et al., 1 994). In the US alone, 557,650 hard rock mines have been catalogued and the clean up of these mine sites is expected to cost between $32 and $7 1 .5 b illion dollars (IALR Newsletter, 1 993). Reclamation of one of the most degraded sites (the Summitville Goldmine in Colorado) is expected to cost over $ 1 00 million US dollars alone (Skow, 1 993). The clean up of this site will involve the neutralisation and relocation of 6'l2 million tonnes of acid generating waste to prevent further contamination of the local river system by cyanide and metal contaminated water. Mine site reclamation programs of even greater magnitude are currently being conducted in Germany in the Saxon Erzgebirge region. Uranium mining conducted between 1 946 and 1 989 has seriously degraded the environment and clean up of the region is expected to take more than 1 0 years and to cost between 1 0 and 1 5 billion German marks (Keller, 1 993) . 36 Metalliferous wastes, consisting of mineralised ore materials brought to the surface during mining activities and containing elevated concentrations of one or more heavy metal elements, are notoriously difficult to reclaim (Williamson & Johnson, 1 98 1 ). The majority of highly contaminated mine sites contain such material and many are plagued with problems specifically associated with the oxidation of remnant sulphide minerals in mine tailings and waste rocks. 3.2.1 Australian Mine Sites Examples of severe environmental degradation resulting from metalliferous mining operations are not confined to the Northern Hemisphere. Degraded mine sites are found in several Pacific Island countries including, Papua New Guinea (PNG), Solomon Islands, New Caledonia and Fij i (Harris , 1 997). PNG is the most heavily impacted of these countries and hosts the well known Bougainville and Ok Tedi copper mines. Some of the most severe environmental degradation in PNG has arisen from the use of rivers as disposal systems for tailings and contaminated mine water. In Australia, the Broken Hill and Captains Flat mine sites in New South Wales and the Mount Isa and Rum Jungle mine sites in Queensland and the Northern Territory respectively, are classic examples of degraded mine sites. A brief overview of the environmental degradation and reclamation measures that has or is being employed at these 4 sites is given below. 3.2.1 .1 Rum Jungle, Northern Territory Prior to reclamation the Rum Jungle mine site, situated 64 kilometres south of Darwin, was one of the most degraded mining sites in Australia (Bell et al. , 1 989). Between 1 983 and 1 986 an estimated $AI 8.6 million dollars was spent rehabilitating an experimental heap leach pile, a 32 ha tailings dam, 3 waste dumps and 4 million m3 of highly contaminated water (Bennett et al., 1 987; Lowson et al. , 1 987; Bell et al. 1 989). Rum Jungle was mined for Cu and U238 during the peak of the cold war ( 1 954 - 1 964) and was finally abandoned in 1 97 1 . At that time in Australia little or no legislation existed requiring the reclamation of mine sites, thus many mine sites were effectively left to the elements at the end of mining operations (Bastias, 1 987; Ryan, 1 987). 37 The majority of environmental problems that arose at Rum Jungle were related to the presence and oxidation of pyritic minerals within the heap leach pile and waste rock dumps (Harries & Ritchie, 1 987). Acid mine drainage emanating from these sulphide­ bearing materials seriously contaminated the eastern branch of the Finniss River with acid and a variety of metals including AI, Cu, Co, Fe, Mn and Zn (Bennett et aI, 1 987). S iltation and metal contamination of the river system also arose due to the erosion of the tailings dump by monsoonal rains and flood waters (Bennett et ai., 1 987; Lowson et aI., 1 987). The average metal contents of these tailings were; 6000 mglkg Cu, 1 000 mglkg Pb, 50 mglkg Zn, 470 mglkg U and 26000 Bqlkg Ra (Lowson et al. , 1 987). To prevent erosion of the tailings into the river-system and to inhibit sulphide oxidation, the tailings were excavated and dumped into one of several large water-filled, open-cut pits on the site (Bell et aI., 1 989). The land on which the tailings were originally deposited was treated with 8 tonnes of hydrated lime per hectare and covered with 300 mm of topsoil. The soil cover was then fertilised and sown with a mixture of tropical grasses, native shrubs and trees (Bennet et al., 1 987; Bell et al., 1 989). Reclamation of the overburden wastes dumps at Rum Jungle involved recontouring the heaps and capping them with clay and soil materials to reduce the infiltration of water and to facilitate plant establishment. Considerable research was conducted investigating the effectiveness of the vegetated covers at reducing water infiltration, oxygen ingress and ultimately sulphide oxidation within the tailings (Bennett et al., 1 987; Harries & Ritchie, 1 987, 1 988). Although the primary purpose of the compacted clay cap was to abate the generation of AMD by restricting the ingression of water and oxygen into the sulphide-bearing wastes, the cap also helped to prevent contamination of the overlying soils by reducing capillary rise of acid and metal contaminated pore waters. The clay/soil barriers used on gently sloping areas of the Rum Jungle dumps consisted of a ;::: 225 mm layer of compacted clay, followed by a minimum of 250 mm of clayey sand (plant root zone), which was inturn overlain by 1 50 mm of gravelly sand (erosion control layer). - - ��-� ----- 3 8 Whilst the results of reclamation work conducted at Rum Jungle appear very encouraging to date, only time will reveal how effective Australia's costliest rehabilitation program has been at curbing problems associated with sulphide oxidation. 3.2.1 .2 Captains Flat, New South Wales In terms of mineralogy, climate and history the Captains Flat Mine is similar to the Tui Mine. Captains Flat Mine is a former Au, Cu, Pb and Zn mine located 50 kilometres upstream of Canberra on the Molongolo River, in New South Wales. Mining was carried out at this site between 1 874 and 1 962 (Craze, 1 977; Bell et aZ., 1 989). In contrast to Rum Jungle, rainfall at the site (740 mm/yr) is relatively well distributed throughout the year and mean monthly temperatures range from 20°C in summer to 6°C in winter. Despite being appreciably drier the climatic conditions at Captains Flat mine site are comparable to those at the Tui mine site in New Zealand (Chapter 2, section 2.3) . In a process similar to that which was used at the Tui mine site, metals were extracted from the crushed ore using froth flotation methods. The tailings were separated into fine and coarse fractions and disposed of in rudimentary earth dams and dumps respectively. By the time the mine closed, an estimated 1 5 ha of land was effectively covered by tailings (Bell et aZ. , 1 989). Not only were the tailings unsightly, they were also a source of air and water pollution. The tailings exhibited similar characteristics to the Tui tailings in that they had a pH of less than 3 .0 and contained high concentrations of sulphidic Cu, Pb and Zn. The tailings also generated AMD which effectively sterilised the Molongolo river for many kilometres downstream and even impacted water quality within Canberra lakes. After several unsuccessful attempts (c. l 96 1 ) to chemically stabilise the tailings dams (which on several occasions partially collapsed into the Molongolo River), a concerted campaign was finally mounted in 1 974 to investigate full reclamation of the site (Anon., 1 974). Initial field trials found, that despite the application of high rates of lime (40 - 50 t/ha) and compound fertilisers, vegetation failed to establish satisfactorily on the amended tailings (Craze, 1 977, 1 979). Following extensive laboratory, glasshouse and field trials, a decision was made to recontour the tailings dumps and to cap them with a clay/soil 39 cover capable of supporting plant growth. To reduce erOSIOn and to facilitate revegetation and adequate drainage, the dumps were recontoured to produce terraced slopes with grades of between 5% and 33% (Bell et ai., 1 989). The terraced slopes were then covered with 220 rom of compacted clay to inhibit air and water ingression into the tailings. This barrier was in turn covered with 450 rom of shale rock designed to inhibit capillary rise of contaminated water from the underlying clay cap and tailings into the plant root zone and to facilitate lateral water drainage. A 300 mm layer of locally sourced soil material was placed at the surface of the dam to host vegetation. The reconstituted soil was subsequently fertilised and sown with a grass/legume mix which included common pastoral species such as Agrostis capillaris and Trifolium repens. Secale cereale was also included as a nursery crop. After sowing, the soil surface was covered with a bitumen-based straw mulch and sprinkler irrigation was installed to ensure that water was non-limiting. Maintenance of the site included fertiliser dressing and the periodic slashing of pasture to encourage tillering. The reclamation of the site is estimated to have cost in the vicinity of $3 .5 million but has thus far shown very promising results (Bell et ai., 1 989). 3.2.1 .3 Mount Isa & Broken Hill The climatic conditions prevailing at the Mount Isa and Broken Hill mine sites contrast with those at Rum Jungle and are significantly drier than those at the Captains Flat. Reclamation efforts at both sites are primarily governed by the arid climates in which evaporation greatly exceeds rainfall throughout most of the year. Summer temperatures at both sites commonly exceed 30° C and the availability of water during the height of summer i s clearly an important factor limiting the establishment of plants on less than optimal growing media. Although a variety of wastes are present at Mount Isa and Broken Hill, the reclamation of sulphidic base-metal tailings at each site has been the focus of considerable effort over the past few decades. Tailings at both sites are characterised by fine grain-size, low penneabiiity, high bulk density, excessive salinity and low levels of nitrogen and phosphorus. Attempts to reclaim them have been aimed at controlling wind blown dust and to improve the appearance of the sites (Hunter & Whiteman, 1 975; Bell et al., 1 989). Tailings at Broken Hill exhibited neutral to slightly alkaline pH due to the 40 addition of alkaline material during processing and to the predominance of what Bell et al. ( 1 989) described as ' inactive sulphides ' . The pH of the Mount Isa tailings was similarly near neutral. Dolomite present in the tailings is believed to actively neutralise acid generated from the oxidation of pyrite . Whilst research into the reclamation of tailings dams first began at Mount Isa in the 1 960' s, the site is still very much operational and the deposition of tailings into several active dams continues. It is estimated that one of the dams currently in use will upon completion early next century, cover an area of 1 400 ha and contain some 1 00 x 1 06 m3 of tailings (Bell et al., 1 989). Trials used to assess vegetation establishment techniques at Mt. Isa and Broken Hill mine sites have met with mixed success over the years. Vegetation was established on about 30 ha of tailings at Mt Isa , however, during the early 1 980 ' s via the use of trickle irrigation, pH amendment (fly ash) and the emplacement of a 0 .3 to 1 .0 m 'rock' cap over the tailings. A variety of exotic and native shrubs and grass species were then used to revegetate the material. Further discussion of this and other trials conducted at the two sites is outlined by Bell et al. ( 1 989) in a useful review paper of mine tailings reclamation in Australia. 3 .2.2 New Zealand Mine Sites As briefly discussed in Chapter 2 (section 2 .9), up until 1 99 1 the 1 97 1 Mining \vas the principal Act governing the extraction of Crown-owned minerals and the reclamation of land disturbed by mining in New Zealand. Prior to 1 97 1 there were few environmental restrictions placed on mining companies and providing the criteria of the Water and Soil Conservation Act of 1 967 were met, mining was carried out with little regard for the environment or the future use of the land. Thus, in terms of environmental protection, the Mining Act was an important piece of legislation in that it aimed to reduce environmental degradation associated with mining and forced mining companies to take some responsibility for the reclamation of mined land. The passing of the Crown Minerals Act and Resource Management (RMA) in 199 1 effectively tightened environmental controls governing mining activity in New Zealand 4 1 and was designed to minimise the likelihood of large scale environmental degradation resulting from mining activity in the future. The RMA requires mine practitioners to ' avoid, mitigate or remedy' any adverse effects of activities on the environment. Environmental damage is an inevitable consequence of many mining operations but has the greatest potential with large scale open-cast mining and/or with metalliferous mines which involve the extraction and processing of highly sulphidic ore. $ ] 6 ] Million $ 427 Million $ 28] Million Non-Metallic Minerals $ 4 1 Million Figure 3.1 Economic importance of various mined materials in New Zealand in 1 995. (source: Ministry of Commerce, 1 996) As previously discussed in Chapter 2 (section 2 .8) the mining of metals, particularly gold and silver, has been an important industry in New Zealand. During the second half of the 1 9th Century gold was one of this country' s mqjor export earners and although gold/silver mining is no longer a mainstay of the New Zealand economy, this industry is currently an important source of overseas revenues. According to the New Zealand Annual Mining Review 1995 over 1 2. 1 tonnes of gold, 27.8 tonnes of silver and over 2.3 million tonnes of magnetite were produced in New Zealand (Ministry of Commerce, 1 996). The combined worth of these metals was estimated at over $ 28 1 million (Figure 3 . 1 ). 42 The following section outlines the operation and reclamation strategies and measures employed at several major mine sites in New Zealand. In the case of alluvial gold mining, aggregate mining and coal mining, an overview of these industries is given in terms of their nature and the means by which land disturbed by these activities is generally reclaimed. An overview of the environmental problems associated with the Tui mine site is included, as is a summary of the limited research that has been conducted investigating possible reclamation strategies. 3.2.2.1 Martha Hill Mine, Waihi Epithermal gold deposits were first discovered at Martha Hill in 1 878 and by 1 9 1 2 the mine was reported to be the 'eleventh most productive mine in the world in terms of bullion won, and the second . . . in terms of greatness and stability' (Moore e\- �\ . s 1 996). Production continued at the mine until 1 952 when it was forced to close down due primarily to declining gold prices. A dramatic increase in gold prices in the 1 970s renewed interest in Martha Hill and mining finally resumed at the site in 1 988 . Annual gold and silver production levels are currently in the vicinity of 75000 oz (2 1 00 kg) and 590000 oz ( 1 6700 kg) respectively (Moore et ;:1.1 " 1 996). The open pit at Martha Hill is situated near the centre of Waihi township, an historical mining centre and service town for surrounding agricultural and horticultural enterprises. The present day mining operation involves conveying andesitic overburden and ore from the pit to a waste disposal area and processing plant situated 2 .3 kilometres away which currently covers a ground surface area of 1 70 ha. Oxidised waste rock from the pit has been used to engineer a tailings dam embankment within which unoxidised (pyritic) waste rock has been buried in sealed 'cells' (Gregg et al. , 1 995). Finely ground tailings consisting of sand, silt and clay sized particles from the treatment plant are transported as an aqueous slurry to the tailings impoundment where they are generally deposited sub-aerially via a series of spigots. The construction of a second, smaller tailings impoundment, to be situated immediately to the south-east of the existing dam, is likely to commence within the next few years. The close proximity of the gold mine to Waihi township and stringent environmental controls outlined in the mining licence ensured that the planning and implementation of 43 reclamation of disturbed land was given high priority from the onset of mining. Gregg & Stewart ( 199 1 ) perceived reclamation of the waste disposal area to be of particular importance, not only for aesthetic reasons but also because of its situation on productive farmland, formerly used for dairying. In preparation for a mining licence Waihi Mining Company Ltd. was required to prove that the tailings and waste rock could be restored to levels of production at least equal to that of soils present at the site. Baseline studies identified that the existing productivity of soils to be impacted by mining ranged between 1 2000 and 1 6000 kg dry-matter per year (Mason et ai., 1 995) . A preliminary study was undertaken to investigate the physical and chemical properties of the tailings and wastes materials (both oxidised and unoxidised sulphide bearing rock) in terms of their plant growth properties (Widdowson et at. , 1 984) . The tailings were identified as comprising poorly aggregated, predominantly silt-sized particles with good moisture holding capacity but poor aeration and drainage. Chemically the tailings were found to be deficient in N, P and K and to exhibit low CEC and high pH (�9) . Despite these limitations, glasshouse trials demonstrated that once nutrient deficiencies were corrected, pasture plants could be successfully grown in tailings materiaL Similarly, the oxidised waste rock was found to show good potential as a plant growth medium provided it was adequately limed and fertilised. The unoxidised (pyritic) waste rock in contrast, was identified as an unsuitable plant growth medium due to its excessive soluble aluminium and salt content and high iron sulphide content (Widdowson et ai., 1 984) . The initial engineering design proposed to strip between 1 00 and 1 50 mm of topsoil only from the tailings dam site prior to mining for resurfacing the bund and tailings pond (Gregg et a!. , 1 995). A surrogate subsoil was to be used on the tailings dam embankment and comprised of waste rock only. To investigate what depth of topsoil (if any) was required and whether oxidised waste was a suitable subsoil 3 field trials were set-up in 1 985 (Gregg et al., 1 990; Home et at., 1 990) . The growth potential of tailings and waste rock amended with 4 depths of soil (0, 50, 1 50 & 250 mm) were assessed and compared to pasture production from both disturbed (topsoil removed and replaced) and undisturbed soil plots located next to the proposed waste disposal site (Gregg et al., 1 995). 44 (i) Tailings Trials Pasture plots ( 1 0 m2) established on tailings amended with the varying depths of topsoil and 1 50 kg Plha, 1 60 kg Klha and 1 80 kg S/ha applied as potassic superphosphate in the first year, showed comparable levels of production to non-disturbed soils within 2 years of sowing (Gregg et oZ. , 1 990). On average, yield differences between topsoil treatments for the first 5 years of the field trial were small, with the nil soil treatment performing well. Dry matter yields were, however, found to decrease significantly as the harvesting frequency of the trials was reduced. In 1 992 a study was carried out to investigate the rate of soil formation and nutrient status of the nil soil treated tailings (Gregg et oZ. , 1 995) . After 6 years of growth the pH of the tailings was found to have decreased from 9.2 to 6 . 0 and carbon and nitrogen levels had increased about 6 and 7 fold respectively. Mason et of. ( I 993) estimated that if C and N increases remained linear it would take about 24 years for C and N levels to reach those of the control soil (an undisturbed soil at the mine site used to gauge productivity). Additional field trials were set-up in 1 992 to investigate the amendment of tailings with locally sourced municipal compost. Pasture was established on sixteen 2 x 4 m plots, 8 of which were treated with 50 rum layer of compost. In a separate trial of similar design a variety of native plant species were trialed including, kohuhu (Pittosporum tenuifolium), kanuka (Leptospermum ericoides), cabbage tree (Cordyline australis) and flax (Phormium tenax) (Mason, 1 996). Mason ( 1 996) once again found that the tailings exhibited few limitations to plant growth, however, compost addition was found to significantly improve pasture yields by improving P status. During this trial significantly higher yields of pasture ( 1 4000 kg cf 6000-7000 kg dry-matterlha) were obtained off the compost amended tailings compared to tailings amended with a similar depth of soil in the previously outlined trial of Gregg et oZ., ( 1 990). These differences were attributed to the superior P status of the compost amended treatment and differences in seasonal rainfall variation (Mason, 1 996). 45 In the case of native plants, compost was found to improve growth in the first 6 months, but beyond 6 months no significant yield differences were observed between the nil compost and compost treated plots (Mason, 1 996). In the presence of slow release fertilisers growth rates of the various native species were all very promising, suggesting that reclamation of tailings with native vegetation is a feasible long term option (Mason, 1 996). (ii) Tailings Reclamation Strategies Reclamation strategies for reclaiming the surface of the existing tailings dam, upon decommissioning of the Waihi mine, include using the area as a waste rock disposal area, leaving the area flooded, capping the entire dam surface with a dry cover such as clay or compost, or using a combined water/cap cover (Mason, 1 996). Whilst none of these options have been ruled out it is probable that reclamation of the dam will involve the later option (combined water/cap covers). Such a cover is likely to involve the capping of higher strength (coarser) tailings near the impoundment's perimeter with a composite waste-rock and soil cover and maintaining a water body over very finely ground tailings which take considerable time to consolidate and thus to develop shear strength. This type of cover would potentially improve stability of the dam crest, reduce soil material requirements for capping purposes, accommodate consolidation of the tailings and potentially provide an aesthetically attractive wetland and recreational area. Predictive modelling suggests that the reclamation of capped areas to pasture would be less costly and would require less maintenance than reclamation to native plants (Mason, 1 996). It highly unlikely that the surface of the dams will be sealed in an attempt to prevent water and air infiltration into the tailings. Such measures are likely to be prohibitively expensive and unnecessary given the low toxicity of the tailings. The Waihi tailings differ markedly to the Tui tailings, in that they contain relatively low concentrations of acid generating pyrite as indicated by their low total sulphur content « 0.001 5%). In addition, the Waihi tailings generally contain comparatively low concentrations of most heavy metals and exhibit alkaline pH (Widdowson et aI. , 1 984; Gregg et al. , 1 995) . 46 (iii) Waste Rock Trials In addition to investigating the growth response of pasture species grown on oxidised waste rock amended with 4 depths of soil, the underlying waste-rock on half of the sixteen 30 m2 plots was modified with lime (6 Mg/ha) and superphosphate ( 1 Mg/ha) . A full discussion of the trial which yielded some interesting results is detailed in Gregg et al. ( 1 990 & 1 995) . In short, modification of the waste rock with lime and superphosphate substantially increased yield by almost 4000 kg of dry matter per hectare in the first year of the trial . It was assumed that the lime and superphosphate treatment reduced phytotoxic levels of soluble aluminium in the surrogate waste subsoil, although improved P status may have also encouraged plant growth. Surprisingly, where no topsoil was used, the highest level of pasture production occurred due to what was assumed to be a more favourable P status. Less weeds, however, and lower populations of soil borne pathogens may have also aided plant growth on this treatment. By the third year of the trial no differences in pasture production between the modified and unmodified waste treatments were evident (Gregg et at., 1 990, 1 995). Research conducted by Gurung et al. ( 1 996) indicates that the reduced yield differences between the modified and unmodified waste rock arose due to the progressive leaching of organic ligands (produced by the decomposition of organic matter) which complexed toxic Al within the waste rock. The results of initial trial work carried out at the mine site indicated that both waste rock and overlying soil were well suited to plant growth providing they were adequately limed and fertilised. Although the waste rock was found to have limited ability to store water (Horne et aI., 1 990), this was not anticipated to be a major l imitation to plant growth due to the site's high and regular rainfall (c.2500 mmlyr). Gregg et al., ( 1 990), however, considered that 'high' inputs of fertiliser and well managed grazing of the restored soils were imperative to avoid pasture regression, soil compaction and erosion. They further suggested that in the first 3 to 5 years of pasture production, grazing use of the land should be dictated by pasture availability and soil condition and not by animal feed requirements. 47 Progressive reclamation of the bund area has been commenced as per a condition of the mining licence and to date (July 1 997), over twenty hectares of the waste bund have been successfully revegetated using methods developed from the various trial work. The reinstatement of pasture on the bund currently involves ripping and amending the surface waste rock with lime (6000 kg!ha) and superphosphate ( 1 000 kg/ha) prior to the addition of 1 00 mm of previously stripped and stockpiled topsoil. The soil cover is then topdressed with 1 400 kg/ha of 5% potassic superphosphate (and limed if required) and sown with ryegrass and clover species at a rate of 40 kglha. The use of low ground pressure machines and a subsoiler in the placement of subsoils ensures compaction is minimised and root penetration and drainage between the horizons is improved (Lapwood, 1 99 1 ). Ongoing research at Martha Hill goldmine investigating pit wall revegetation and reclamation options for the tailings disposal area, combined with stringent management practices and environmental controls is likely to ensure the long-term success of reclamation efforts at this site. 3.2.2.2 Golden Cross Mine, Waihi Golden Cross is a combined open-pitltmderground hardrock gold mme situated 8 kilometres NW of Waihi . Whilst small areas of land at the mine site have been revegetated with exotic grasses and native plant species, large scale reclamation of the site will not be undertaken until the end of the mining proj ect (Bokich, 1 995). Upon closure of the mine (originally projected to be in the year 200 1 ) it is anticipated that most disturbed areas will be revegetated with native plant species (Bokich, 1 995; Raea, 1 997) . The general aim of reclamation at the site will be to stabilise soils and to prevent the generation of AMD from mine workings and sulphide-bearing wastes (Bokich, 1995). The tailings embankment at Golden Cross is currently being constructed from overburden waste rock from the open pit. Highly sulphidic wastes are either blended with benign waste rock and compacted to a permeability of 1 x 1 0-8 rnJs or encapsulated within compacted fine grained wastes to help prevent oxidation (Bokich, 1 995, Weston, 1 995). Prior to deposition within the impoundment, the tailings produced by the 48 goldmining venture are treated to remove cyanide and limed to pH 9.5 to encourage the formation of highly insoluble metal complexes (Bokich, 1 995). Upon closure of the mine the company intends to rehabilitate the tailings pond to wetlands which will maintain a reducing environment (Bokich, 1 995). Recently, reclamation efforts at Golden Cross have been drawn into the media spotlight following revelations that the tailings dam, which contains over 3 .0 million tonnes of tailings, is situated on unstable terrain (Raea, 1 997). Land immediately adjacent to the tailings dam has begun to slowly move due to the activity of a deep seated fault within underlying strata (Crampton, 1 996; Mare, 1 997; Raea, 1 997). Coeur Gold New Zealand Limitedl , who manage and own 80% of the mine, first publicly announced that there were potential problems with the dam's stability in November 1 995 (Weston, 1 995; Crampton, 1 996). Since that time the company has reputedly spent $20 million dollars in attempts to stabilise the site (Mare, 1 997; Raea, 1 997). Remedial works undertaken at Golden Cross include the excavation of a 1 km drainage tunnel and 1 00 horizontal boreholes to dewater failing strata, the construction of rock buttresses and ongoing reduction of water levels within the tailings impoundment (Crampton, 1 996; Mare, 1 997). Upon notification of the stability problem, the Waikato Regional Council ordered Coeur Gold to investigate the feasibility of relocating the dam. It is estimated that if such measures were required it would cost the company more than $ 1 00 million (Raea, 1 997). To date environmental bonds totaling $ I 2 . 1 million have been signed by Coeur Gold New Zealand Limited to ensure that the company does not ' abandon' the site. Obviously any efforts at rehabilitating the tailings dam are likely to take a back seat to insuring its long term stability. 3.2.2.3 Macraes Flat Mine, Eastern Otago The Macraes Flat Mine is New Zealand's largest and currently most profitable goldmine. It is located approximately 60 kilometres due north of Dunedin on the eastern edge of the Otago Goldfield. In 1 996 the mine produced 1 38,464 oz (3925 kg) of gold from chlorite schist-hosted quartz lodes (Gregg, 1 997). The majority of ore mined at I Coeur Gold NZ Ltd. are a fully owned subsidiary of Idaho based Coeur d' Alene Mines Corporation. 49 Macraes contains appreciable quantities of sulphide minerals including pyrite, arsenopyrite and minor chalcopyrite. Gold is recovered from the ore using cyanide and activated carbon processing. The finely ground tailings are pumped into a dam adjacent to the processing plant where they are left to dewater. As with the Golden Cross dam, the primary tailings dam at Macraes Flat is engineered to withstand a Maximum Credible Earthquake for the region (Crampton, 1 993; Gregg, 1 997). The objectives of reclamation at Macraes is to progressively rehabilitate the mined land to a state compatible with the landscape and the environment that existed prior to mining activity (Gregg, 1 997). The shallow soils, low rainfall and temperature extremes at the site present challenges to the mining company in terms of meeting these goals and rehabilitating the site. The revegetation of the tailings dames) upon completion of mining activity is likely to involve the 'replacement' of topsoil that has been excavated and stockpiled from stripping areas. Whilst the arid conditions at the site preclude the likely production of large volumes of AMD, sulphide bearing wastes will have to be managed appropriately and capped where necessary to minimise sulphide oxidation. Despite ongoing expansion of the site, significant areas have already been revegetated using pastoral plant species consistent with that of land surrounding the 400 ha licensed area. Trials are currently being carried out to identify the most appropriate combination of vegetation to be used for reclamation. Additional trials are planned to investigate the suitability of various native shrub and tree species for revegetating steeper slopes at the site (Gregg, 1 997). 3.2.2.4 Alluvial Gold Mine Reclamation It is estimated that there are about 60 active alluvial mining licences working gold­ bearing gravels and sands in New Zealand (Davidson, 1 997). The majority of these placer deposits occur in fluvioglacial gravels, uplifted beach deposits and in Holocene river gravels and beach sands in Westland and Otago (Christie & Brathwaite, 1 997). The Grey river dredge operated by Birchfield Minerals Ltd. is currently the largest alluvial goldmine in the country employing 24 full-time staff (Davidson, 1 997). Other relatively large operations include Arahura Valley, Jones Flat, Mikonui and Nokomai (Mineral Resources, 1 997). Alluvial gold production from these and numerous smaller operations 50 currently accounts for about 25% of the total gold produced in New Zealand. Conservative estimates suggest that the alluvial gold resource in Westland is over 3 00 tonnes and about 1 00 tonnes in Otago (Davidson, 1 997). As previously discussed, up until the introduction of the Mining Act in 1 97 1 , mining legislation in New Zealand poorly protected the environment against large scale degradation associated with alluvial gold and other forms of mining. The Kanieri Dredge, for example, described by Fricker ( 1 986) as the ' last of the leviathans' , operated for 44 years, during which time the dredge won over 420,000 oz (c. 1 2,000 kg) of gold from alluvial gravel deposits in the Kanieri and Taramakau river systems (Gregg et al., in press). For 39 years the company operating the dredge was not obliged to reclaim land disturbed by the mining operations. Clauses in the mining legislation of the time provided a 'no reclamation' option in which companies could pay a flat fee of $75 per hectare in lieu of reclamation (Gregg et a!. , in press). As a consequence of this legislation large tracts of land were left highly degraded by the Kanieri and other dredging operations in Westland and Otago. In Westland, herring bone heaps of coarse tailings interspersed with dredge ponds disrupted natural water courses and replaced natural river flats and terraces, many of which supported po do carp forests prior to mining (Gregg et ai., in press). About 3 000 ha of land is estimated to have been highly degraded by alluvial dredge operations in Westland and Central Otago over the last 1 00 years or so (Fricker, 1 986). The manner in which the Kanieri and other early dredging operations processed alluvial gravels often resulted in the burial of fine-grained tailings and elevation of mined land (often by as much 4 to 6 m) due to swelling arising from the poor mixing of tailings fractions upon redeposition (Gregg et al., in press). Such operations effectively reduced ground water levels and resulted in the deposition of coarse tailings with low water holding capacity and nutrient status at the ground surface. Despite Westland's high rainfall these factors made the material prone to drought and inhibited natural revegetation of the tailings by most plant species for periods in excess of 20 to 80 years (Ross & Mew, 1 99 1 ; Gregg et al., in press). 5 1 Changes in the legislation in 1 97 1 forced the mining companies to begin rehabilitating disturbed land to a state at least as productive as its pre-mine state. The legislation change prompted a surge in reclamation research associated with the alluvial gold­ mining industry. Much of the research conducted at West Coast alluvial mine sites concentrated on the reclamation of tailings to exotic forestry andlor pasture. White pine (Pinus radiata) and Douglas fir (Pseudotsuga menziesii) were two exotic tree species that were trialed along with a variety of legume and grass species (Keating & Buckley, 1 98 1 ; Fitzgerald, 1 98 1 ; Ross, 1 988; Frazer & Keoghan, 1 990; Metcalfe & Godfrey, 1 990; Morton & Harrison, 1 990; Ross & Mew, 1 99 1 ; Gregg et a!. , in press) . Useful literature pertaining to the reclamation of alluvial gold tailings to forestry and pasture in New Zealand are given by Ross & Mew ( 1 990), Mew & Ross ( 1 99 1 ), and to indigenous vegetation by Norton ( 1 99 1 ) and Mew & Ross ( 1 992a). It is generally considered that thorough site assessment, early planning and careful consideration of post mining vegetation are of paramount importance in ensuring that problems associated with reclamation are minimised. Research also indicates that revegetation is usually best achieved by redepositing stripped topsoil onto mixed tailings substrates. Furthennore, the contouring of soils and the replacement of debris (eg. logs and stumps) to provide micro-sites for establishing seedlings, has also been shown to aid the establishment of trees and shrubs. In the case of pasture, well managed grazing, fertiliser topdressing, weed and pest control are considered to be critical post reclamation practices (Ross & Mew, 1 990; Mew & Ross, 1 99 1 ) . To date, there has been mixed success establishing vegetation on 'abandoned' alluvial gold dredge tailings. Whilst the productivity of land at some sites surpasses pre-mine levels, other attempts at revegetation have been unsatisfactory (Ross & Mew, 1 990; Mew & Ross, 1 99 1 ). Reclamation efforts on recent tailings derived from the Grey River Dredge are reportedly "gratifYing" in that the quality of established pasture has exceeded expectations (Davidson, 1 997) . Clearly continuing research will ensure that future alluvial mining will have far less impact on the environment than have past operations. It is l ikely that research into the reclamation of tailings to indigenous vegetation will be of high priority in Westland, particularly in areas administered by the Department of Conservation (Norton, 1 99 1 ) . 52 3 .2.2.5 Titanomagnetite Mining, Waipipi The mining of titanomagnetite minerals from coastal sands in Taranaki region commenced in 1 97 1 . Waipipi Ironsands Limited mined more than 800 ha of land over a period of almost 2 decades . The mining operation was carried out on 2 distinct physiographic units, (i) coastal sand dunes with weakly developed, shallow (60 mm) topsoils and (ii) sand flats characterised by deeper topsoils and gleyed subsoils hosting iron pans and concretions (Gregg et al., in press) . The carrying capacity of the 2 units before mining ranged from 3 to 8 sulha on moderately to steeply sloping sand dunes, to 1 2 su/ha on the sand flats (Gregg et al. , in press). Reclamation of mined land to pasture was generally conducted using the following technique. Prior to mining 0.3 to 0.5 m of soil material was bulldozed to the sides of the dredge's path where it was stockpiled for up to 4 months. After the removal of the titanomagnetite by wet concentration from the crude sands, the tailings2 were pumped back into the dredge pond and left to settle and dewater before being recontoured by bulldozers. The soil material was then respread over the levelled tailings and sown with a 5 0 : 5 0 nursery crop of barley (Hordeum vulgare) and oats (Avena sativa) at a rate of 1 1 5 kg/ha (Gregg et al. , in press). This vegetative cover was designed to rapidly stabilise the erosion-prone tailings. When the cereal crop reached a height of about 0.25 m it was over-SOvvTI by 40 kg of a grass/clover seed mix using a direct drill seeder. Included in the mix were white clover (Trifolium IYpens), taU fescue (Festuca arundinacea), cocksfoot (Dactylis glomerata), canary grass (Phalaris tuberosa) and type specific rhizobium. An NPKS ( 1 2 : 1 0: 1 0: 8) compound fertiliser was applied at a rate of 1 5 0 kg/ha at each sowing. The reclaimed land was generally fit for grazing within 1 8 months of mining (Gregg et al. , in press) . During operations at Waipipi some 800 ha of land was rehabilitated to farmland which was generally more versatile and productive than the original farmland (Taylor & Walker, 1 987; Gregg & Stewart, 1 99 1 ; Gregg et al., in press). The success of reclamation at the site is attributable to the area' s well distributed rainfall (c .950 mm1yr) and to the company's commitment to reclamation research from the onset of operations. 2 Gangue minerals within the tailings were primarily augite, hornblende and plagioclase feldspars with minor hypersthene, olivine, quartz, gamet, apatite and zircon (Gow, 1 987) 53 Several research projects were conducted at Waipipi including trials that investigated topsoil/subsoil stripping, fertiliser application, crop harvest, the performance of lucerne (Medicago sativa) and kikuyu (Pennisetum clandestine), conservation tillage trial, rotational grazing and weed control and animal nutrition (Bruere, 1 98 1 , Gregg, 1 983; Gregg & Stewart, 1 99 1 ; Gregg et at., in press). The results of this research provided the company with valuable information and enabled it to deal with problems encountered during the course of the operations. Problems associated with excessive drainage and stock copper deficiencies for example, were identified and remedied by changes in practice (Gregg et al., in press). Recontouring (flattening) of the land and the sowing of improved pasture species were primary factors which enabled the land to be used more intensively. The stock carrying capacity of much of the mined land was doubled as a result of the reclamation techniques employed at Waipipi (Gregg et al., in press) . Mining operations ceased at Waipipi in c. 1 988, however, irons and is currently being mined at Taharoa and at Waikato North Head on the west coast of the North Island. Over 2 million tonnes of magnetite with an estimated value of $23 million dollars were produced from these sites in 1 995 (Ministry of Commerce, 1 996). 3.2.2.6 Aggregate Mine Reclamation In terms of both volume and monetary worth, aggregate mining is the largest extractive industry in New Zealand. In 1 995, the mining of aggregate material (chiefly sand and rock used for building and roading) accounted for about 47% of mineral production and was valued at over $426 million (Figure 3 . 1 ). Some 23 million tonnes of aggregate were mined from a variety of sources including, the greywacke rocks of the ranges in both the North and South Islands, river gravels particularly in the south of the North Island and throughout the South Island and volcanic rocks (mainly basalt and andesite), particularly in the north of the North Island and near Christchurch and Dunedin (Simcock, 1 995; Ministrv of Commerce, 1 995). Given the variable lithology, climate and topography of land hosting aggregate resources it follows that the scale and extractive nature of aggregate mining varies considerably 54 about the c ountry. Whereas hardrock quarries are often located in steep terrain that host shallow soils and low quality pasture and/or scrub, some of New Zealand ' s most fertile and versatil e soils mantle equally important reserves of alluvial aggregate (McQueen, 1 9 83; S imcock, 1 99 3 ; Simcock & Ross, 1 995). To date there have been only 3 m ajor studies conducted in New Zealand investigating the reclamation of aggregate mines (McQueen, 1 983; Duck & Burton, 1 985 ; Simcock, 1 993 ). This is a surprising fact given the potential impact and economic importance of the industry in New Zealand. The studies all identified that aggregate mining could seriously impact the long-tenn productivity of land and that soil/substrate compaction, topsoil dilution and inadequate topsoil depth were common problems mitigating against successful revegetation. S imcock ( 1 993), however, found that acceptable levels of vegetative production (comparabl e to pre-mining levels) could be achieved by stripping and replacing topsoil and subsoils separately using a hydraulic excavator to minimise compaction. The study highlighted that different reclamation techniques were required for different soils. Simcock & Ross ( 1 995) suggest that the segregation and ' burial ' of poorly drained material and the use of progressive reclamation are also key factors in maximising p l ant growth and minimising soil structural degradation associated with organic matter depletion. 3.2.2.7 Coal Mine Reclamation In tenns of volume extracted, the coal mining industry is the second largest mining industry in New Zealand. The combined production of New Zealand ' s coal mines in 1 995 was 3 ,445 ,03 8 tonnes of mainly bituminous and sub-bituminous coal valued at about $ 1 6 1 m illion (Ministry of Commerce, 1 996). Even by international standards New Zealand's recoverable coal resources (8600 million tonnes) are substantial in size. The vast lignite deposits of Otago and Southland make up the bulk of these reserves (7055 million tonnes) and are New Zealand ' s largest known energy resource (Ministry of Commerce, 1 995). Waikato in the North Island and Westland in the S outh Island are New Zealand' s two primary coal mining districts. Bituminous coal reserves are concentrated in the West Coast ' s Buller and Grey coalfields, whereas the nation ' s largest sub-bituminous 5 5 resources occur in the Waikato region. Among the largest producing coal mines in New Zealand are Stockton, Webb, Wairaki and New Vale in the South Island and the Huntly East, Maori Farm, Waipuna and Kopuku mines in the North Island. The combined production from these mines in 1 994 was over 2 million tonnes of coal (Ministry of Commerce, 1 995). New mines recently commissioned include the Mount Davy mine near Greymouth and the Kopako III mine near Maramarua, 30 kilometres east of Huntly (Mineral Resources, 1 997). Coal was first commercially mined in New Zealand in the 1 830 ' s and remained the Country' s primary energy source for the next 1 00 years. The OPEC oil crises of 1 973 and 1 974 renewed interest in New Zealand' s coal resources. Detailed surveys of New Zealand's coalfields were conducted between 1 975 and 1 987 and led, ultimately, to an increase in coal mining activity (Ministry of Commerce, 1 995). Traditionally, most coal in New Zealand has been mined using open-cast methods and currently only about 1 5% of coal is extracted via underground mining methods (Ministry of Commerce, 1 995). Open-cast mining generally has far greater impact on the aquatic and terrestrial environment than underground mining and can potentially lead to major environmental problems, particularly where overburden contains high concentrations of pyrite . An increase in open-cast mining at Ohai in Southland c. 1 975 led to severe stream pollution from poorly managed and unvegetated pyritic overburden heaps and from uncontrolled discharges of acidic and metal contaminated waters. Strong public concern led to the reclamation of 90 ha of land at Ohai at a cost of $50001ha to control water quality at the site (Fricker, 1 986) . Similar environmental problems arose in the Waikato coalfield as mining activity increased in response to rising coal prices in the late 1 970s. This prompted the industry in this region to more thoroughly investigate reclamation options and practices. Early coal mining operations on the West Coast have often been poorly regulated and managed. Fricker ( 1 986) reported that as recently as 1 986 some mines were discharging acidic mine wastes and/or effluent into local streams and using overburden disposal methods known to be detrimental to the environment. Extensive trial work and increased awareness of the West Coast's fragile ecosystems has seen greater efforts being made by mining practitioners to adequately reclaim mine 56 sites and to minimise the discharge of AMD from sulphide-bearing overburden (Mew & Ross, 1 992b; Davis et al., 1 995; Ross et al. , 1 995; Simcock & Ross, 1 995). One of the more innovative studies presently being conducted is a collaborative research project between Manaaki Whenua-Landcare Research, Forest Research Institute and Dunollie Coal Mines Limited at Giles Creek Coal mine (Davis et al., 1 995 ; Ross et al. , 1 995; Simcock & Ross, 1 995). The research focuses on developing techniques of establishing native vegetation on mine wastes, an area of reclamation that is considered to be poorly developed compared to reclamation using exotic forestry and pasture (Gregg et aI. , 1 990; Davis et al. , 1 995 ; Ross & Mew, 1 99 1 ; Simcock & Ross, 1 995). Giles Creek is situated in Westland, 1 2 kilometres north-west of Reefton and is a relatively small sub-bituminous coal mine with reserves estimated at 1 0 million tonnes (Mineral Resources, 1 997). The mine is located in beech-podocarp forest which has been selectively logged for rimu and miro. Rainfall at the site is about 2900 mm/yr. Initial reclamation research began at the mine site in 1 990 when trials were established on a capped overburden dump (Davis et al. , 1 995). Despite good control of competing vegetation, native seedling mortality was high on the trial plots and plant growth rates were low. The poor performance of most native species was attributed to inadequate drainage caused by the fine-grained nature of the capping material . The trial highlighted that good drainage was a key factor governing the establishment of native woody species in a high rainfall environment. The results of the trial also demonstrated the necessity to provide adequate animal control to prevent excessive browsing. Good fencing and the use of low palatability species are important methods to prevent browsing (Davis et ai. , 1 995 ; Ross et at., 1 995) . Additional field trials were set up in 1 992 at the site to study the methods and effects of soil replacement on the survival and growth of a variety of native tree and shrub species grown on overburden. The trial highlighted the need to strip and replace adequately drained soils , to minimise compaction and to ensure surrogate soils are adequately fertilised. Nitrogen was considered to be the most severe nutrient deficiency affecting plant growth (Davis et al., 1 995). 57 In response to trial results the company has begun salvaging tree stumps with attached soil material and replacing them on recontoured spoil to produce ' instant forest' . The placement of stumps and forest debris is believed to "kick starf' regeneration of native forest by supplying plant material, nutrients and shelter for establishing plants. This type of revegetation technique is well suited to sites containing existing native forest on terrain that allows heavy machinery access (Simcock & Ross, 1 995) . S imcock & Ross ( 1 995) and Davis et al., ( 1 995) provide a useful review outlining the methods of restoring New Zealand native forests after mining and specifically cite the Giles Creek mine as an example of an integrated approach to site reclamation. Although the reclamation research at Giles Creek is perhaps the most innovative coal mine reclamation research conducted in New Zealand to date, several other notable studies have been conducted. These include the studies of McQueen & Ross, ( 1 982), Widdowson et ai., ( 1 982), Ross & Widdowson ( 1 985) and Widdowson & McQueen ( 1 990). Informal trials have also been conducted by Solid Energy (formerly Coal Corporation) at mine sites around Huntly in the Waikato. Common factors identified in these studies shown to improve plant establishment were the need to:- (i) strip and replace topsoils and well drained subsoils where at all possible, (ii) minimise and/or ameliorate compaction and (iii) assess and rectify nutrient deficiencies. Clearly the appropriate handling and disposal of acid generating overburden is of paramount importance at sites where these materials occur. 3.2.2.8 Tui Mine, Te Aroha The Tui Mine is a graphic example of a base-metal mine that was very poorly managed in terms of environmental impact. As discussed in Chapter 2 (section 2 .8), the Tui Mine was first opened in 1 884 but the bulk of the mine' s production occurred during the late 1 960's and early 1 970 's . By the time the Mining Act ( 1 97 1 ) was passed considerable environmental damage had occurred at the Tui mine site and the scene was set for continued degradation. As outlined in Chapter 2 (section 2 .8), the mining operation at Tui was carried-ovl1 underground and involved driving adits (horizontal drives) into the host rock. Upon interception of the quartz reefs hosting the sulphide ore, mining was redirected along the 5 8 reefs and high grade ore was selectively removed via adits which were developed on 5 levels and ranged in length from about 9 to 4 1 2 m (Carter & Pharo, 1 965) . The are was transported from the mine adits to a flotation mill where it was first pulverised and then passed through flotation cells to physically concentrate the are. The flotation process involved feeding finely-ground are into a series of separation tanks containing water and a variety of oil reagents. The oil reagents were formulated to selectively coat one or more sulphide mineral components of the are. Aeration of the mixture resulted in the formation of a metal-bearing 'oil froth' at the surface of the tanks which was removed to yield the metal-concentrate (Blaiklock, 1 98 1 ; Brady & Humiston, 1 982). The 'differential' flotation process used at the Tui mine site produced 2 concentrates, a zinc and cadmium concentrate and a lead and copper concentrate which also contained gold and silver (Norpac Mining Limited, undated). Waste water produced from the are processing was first decanted and then discharged "into a stream [the Tui Stream] specified by the local Borough Council" (Norpac Mining Limited, undated). Tailings from the mill are believed to have been deposited into the tailings dam as a thick slurry from a point source(s) immediately beneath the flotation milL Photographs taken during the mine' s operation indicate that tailings were redistributed and levelled with a bulldozer (Cochrane, 1 969). During the initial stages of the mine' s development the tailings were contained by a 5 m clay bund. This bund is situated beneath the front (western) face of the dam and is now covered by over 7 m of tailings (Tonkin & Taylor, 1 974). Thus the bulk of the tailings are therefore not structurally contained by a purpose-built retaining wall such as that constructed at the Waihi waste disposal area. The ' on site' problems associated with the Tui Mine are numerous and include:- a) Visual impact - aesthetically the unvegetated tailings dam, disused roads and abandoned plant degrade the landscape. The negative visual impact of the these structures and particularly the tailings dam, contrasts strongly with the scenic native bush surrounding the mine site (Plate 2. 1 ) . 59 b) Water Contamination - acid and/or heavy metal contaminated leachates emanating from the tailings dam and abandoned mine adits have severely polluted the Tui and the northern Tunakohoia streams (Plate 3 . 1 ). Pollution of the streams is most severe near the site and is exacerbated by the precipitation of ferric oxy-hydroxide which acts as a scavenger of heavy metals and contaminates the stream sediments (Ward et aI. , 1 976; Hendy, 1 98 1 ; Pang, 1 992) . c) Stream Choking - debris in the form of tailings, waste rock and timber derived from the mining operation choke natural stream valleys and inhibit water flow (Plate 3 .2). A severe rainstorm in 1 97 1 reportedly caused the collapse of "several thousand tons" of material from waste rock dumps situated at the 2 lower levels of the mine. Debris from these dumps was carried into the northern Tunakohoia Stream, blocking culverts (WRCM, 1 99 1 ). Mass movement and debris flows are known to be a serious hazard on the steep slopes of Mount Te Aroha particularly during periods of prolonged heavy rainfall (Arand, 1 986). d) Soil Contamination - the studies of Ward et al. ( 1 976, 1 977) indicates that heavy metal (Cd, Zn, Cu, Pb) contamination of the surrounding area occurred via the airborne dispersal of dust particles derived from treatment plant during operations. It is highly probable that wind erosion of the tailings continues to contaminate land adjacent to the dam. e) Tailings Instability/Erosion - whilst an engineering study conducted in 1 974 suggests that "there is an adequate factor against an overall failure of the tailings dam", its structural integrity is considered to be dependent on the tailings remaining largely unsaturated (Tonkin & Taylor, 1 974). Past events suggest that small scale slumping and erosion pose the greatest threat to the short-term stability of the tailings (Plate 3 .3 ) . Rainstorms in 1 97 1 , 1 974 and 1 976 reportedly caused severe rill and gully erosion of the tailings and led to surface flooding of the dam (Hansen, 1 989; WRCM, 1 99 1 ). Some have also voiced concerns about the possibility of the northern Tunakohoia Stream eroding laterally and undermining the long term stability of the tailings dam. 60 The maj or ' off site' problem associated with the Tui mine site is that of heavy metal contamination of water and sediments down stream of the mine site. As discussed in Chapter 2 (section 2 .4) several studies have investigated the impact of mining operations local streams and ground water (Ward et al. , 1 976; Tay, 1 980; Hendy, 1 98 1 ; Smith, 1 986; Livingston, 1 987; Pang, 1 992, 1 995). Whilst water quality improves with distance from the mine site as a result of ferric oxy-hydroxide precipitation and dilution by groundwater and tributaries, research indicates that the quality of water in the northern Tunakohoia and Tui streams is far from optimal at the streams' confluences with the Waihou River (Figure 2. 1 ). Of the two streams, the Tui Stream has been the most seriously impacted by past mining activity due to the inflow of AMD from the tailings dam (Table 2. 1 ). As discussed in section 3 .4, AMD is a serious problem that impacts many active and abandoned mine sites around the world where sulphide-bearing wastes are disturbed and exposed to oxidising conditions. The paucity of acid neutralising carbonate minerals in the finely­ ground Tui tailings, combined with the tailings' comparatively high sulphide content and the site's high rainfall and warm climate, has led to the oxidation of substantial quantities of acid generating sulphide minerals and ultimately to the production of AMD. Water percolating through the tailings thus acts as a conduit of acid, sulphate and heavy metals into the Tui Stream (Ward et ai., 1 976; Tay, 1 980; Hendy, 1 98 1 ; Pang, 1 992). The capacity of the Tui mine tailings to continue to liberate AMD is considered to be the primary problem at the mine site (Miller, 1 995, pers. comm. ; Morrell et at., 1 995; EWM, 1 995). Plate 3. 1 Metal contaminated drainage emanating from an adit at the Tui mine site. Precipitated ferric iron compounds are seen coating the rocks and mining debris (August 1 993). Plate 3.2 Rock and mining debris choking the upper reaches of the northern Tunakohoia Stream above the Tui tailings dam (August 1 993). 6 1 62 (i) Past Reclamation EffortslResearch To date no comprehensive attempts have been made to reclaim the Tui mine site or to remedy the impact of AMD and metal contaminated mine drainage on ground and stream water. Furthermore, surprisingly little practical research has been conducted to investigate potential methods of reclaiming the site. One of the few known studies to have specifically investigated a means of rehabilitating the site was conducted in 1 975 (by the now defunct Hauraki Catchment Board) when field trials were set up to determine if pasture grass species could be established on the tailings dam (WRCM, 1 99 1 ). Unfortunately and somewhat ironically, these trials (which were in part designed to investigate the ability of vegetation to stabilise tailings) were prematurely terminated by a rainstorm in September 1 976 which is believed to have effectively washed the trials away. Documents detailing the location, design and results of these trials are not held by the Matamata-Piako District Council which raises the question of whether in fact they actually existed. Based on the limited information that is available, it appears that the trials were established on the tailings embankment and that vegetation was only ' successfully' established on those plots which received very 'heavy' applications of lime (Hansen, 1 989; WRCM, 1 99 1 ) . Following the 1 976 rainstorm which destroyed the aforementioned trials and caused substantial erosion of the tailings embankment, the former Hauraki Catchment Board undertook measures to stabilise the tailings embankments and to "revamp" the drainage system to reduce the risk of blockage and saturation of the tailings (Hansen, 1 989; WRCM, 1 99 1 ). Amongst other measures implemented, the face of the dam (Plate 3 .4) was ' stabilised' by covering it with a 1 5 - 20 cm layer of "crushed quartz" (Hansen, 1 989) . The placement of this cover3 was primarily designed to reduce rill and gully erosion of the tailings (Plate 3 .3 ) . The capacity of the external drainage system which diverts catchment water around the tailings dam was also increased. Stabilisation and drainage work carried out at the site is believed to have cost the Hauraki Catchment Board in the order of $75,000 ( 1 989 estimate; Hansen, 1 989). As mentioned in Chapter 2 (section 2.9), the National Government made a $25,000 ex gratia payment to the Hauraki Catchment Board in 1 98 1 to offset the costs of stabilising the dam. The current 3 As discussed in Chapter 4 (section 4. 1 ), this material is thought to have originally had an appreciable sulphide content. 63 owners of the land upon which the Tui tailings dam is located, the Matamata-Piako District Council, are responsible for the ongoing maintenance of the site which costs and estimated $2,500 per annum ( 1 993 estimate; Tui Meeting, 1 993). The Waikato Regional Council and the former Hauraki Catchment Board have made preliminary investigations into the costs of possible options for reclaiming the Tui mine site . Amongst the options initially investigated was the relocation of the tailings dam. In 1 979 it was estimated this would cost in the vicinity of $200,000 however, no alternative location was suggested (WRCM, 1 99 1 ). This option is highly unlikely to ever be implemented unless a 1 00,000 m3 sinkhole conveniently forms near the mine site (a somewhat highly improbable scenario given the non-calcareous l ithology of the . f ) regIOn . . Options considered for mitigating contamination of the Tui and northern Tunakohoia Streams by heavy metal-contaminated drainage, included intercepting drainage from each adit and piping it to a central treatment facility along with the tailings' leachates. Carter ( 1 98 1 ) suggested that the yearly running cost of such a facility would be between $600,000 and $ 1 ,000,000. Another option considered to prevent contamination of the northern Tunakohoia Stream by mine adit drainage was to seal and flood the underground mine workings (Fyson, 1 99 1 ; WRCM, 1 99 1 ) . This option was recognised as requiring a large scale geotechnical investigation and approval from the Ministry of Commerce to effectively seal off an important mineral resource. Reclamation of the site with vegetation has long been considered a viable option. With the exception of the present study and the field trials of 1 975, however, the only other known studies to have investigated the properties of the tailings in tenns of their plant growth p otential, were small-scale glasshouse trials conducted at Massey University in 1 992 and 1 996. The first of these studies investigated the growth of radish (Raphanus sativus) on Tui tailings treated with varying rates of lime, whereas the second trial investigated the potential of revegetating the tailings with a species of Thlaspi known to hyperaccumulate zinc. Satisfactory plant cover was achieved in both trials providing high rates of lime were used to amend the tailings extremely low pH. 64 Plate 3.3 Surface erosion of Tui tailings embankment following intense rainfall (August \ 993). Plate 3.4 View of the front (western) face of the Tui tailings dam which is covered with a \ 5-20 cm layer of crushed sulphide-bearing quartz (August 1 993). 65 As later discussed in section 3 .7, it is recognised that the establishment of vegetation directly on the tailings may do little to prevent the generation of AMD from the tailings. It is perceived, however, that direct vegetation of the tailings may be a cost effective method of reducing the negative visual impact of the tailings dam and may ultimately inhibit the oxidation of sulphide minerals in and below the root zone of the tailings and thus curb AMD. Pang ( 1 992) made some tentative suggestions on how to reclaim the tailings dam to prevent AMD. It was suggested that AMD could be controlled by reducing the ground water level in the tailings and by capping the surface of the dam with a combined clay and soil cover. Although the idea of capping the tailings dam is sound, the logic that covering the tailings with a compacted 1 m clay cap may be a "very economic" means of preventing AMD is clearly flawed. Such a cover would require the purchase and transportation of an estimated 1 62,000 m3 of clay to the site. Pang further suggests that if "rehabilitation" of the site is required, the clay cap should be covered with a sand drainage layer and 2 m of topsoil sown with metal-tolerant plant species. Why the use of metal-tolerant plant species is advocated on the proposed cover is unclear, as such a cover would clearly host a large variety of both native and exotic plants. The construction of the proposed cover would obviously be extremely costly and superfluous to needs. It is estimated that the purchase of the topsoil alone (excluding transportation costs) would exceed $3 million (based on conservation price of $ 1 0/m3 and a surface area of 1 .62 ha). Pang's ( 1 992) suggestion that an essential means of controlling leachates from the dam would be to "create and maintain an unsaturated zone for the tailings deposits by lowering the ground water table" is also somewhat flawed. Whilst the reduction of groundwater levels in the dam would serve to further stabilise the tailings, this action may potentially expose more sulphide minerals to oxidising conditions and could in fact increase the contaminant load of drainage waters that would inevitably find their way through the proposed 'dewatering' system. As discussed in section 3 .6. 1 maintaining a water cover is a potentially very useful means of inhibiting the formation of AMD. 66 Based on the numerous problems associated with the Tui mine site and the anticipated cost of reclaiming the site, it becomes apparent why environmentalists and anti-mining groups alike continually cite Tui as a worst-case mining scenario and as a reason why further hardrock gold/metal mining should not be conducted in Coromandel. Others, however, believe that the reputation of the Tui mine site far outweighs its size and impact on the environment and that much of the anti-mining sentiment in the Coromandel is unjust in that modem-day hardrock goldmining operations are incomparable to that conducted at the Tui Mine during the 1 960's and 1 970's and that current mining regulations are effective. One thing that is certain is that until the site is 'successfully reclaimed' it will continue to kindle many a heated debate on the pros and cons of the mining industry in New Zealand. 3.3 Sulphide Oxidation & Acid Generation One factor that is common to many of the highly degraded mmmg sites found throughout the world is the acidification of sulphide bearing mine wastes and the associated production of metal contaminated mine drainages (Ritcey, 1 989). Although sulphide oxidation and the production of metal contaminated drainage have traditionally been associated with the coal mining industry, these processes can potentially impact any site where large volumes of sulphide-bearing materials are exposed to the atmosphere (Skousen, 1 995). Problems arising from sulphide oxidation are also encountered in areas which host acid sulphate soils (Dent, 1 986). The natural occurrence of high concentrations of pyrite within the profile of these soils renders them susceptible to acidification when artificially drained or during periods of drought. Acidification occurs when oxygen diffuses into the nonnally anaerobic subsoil, resulting in the oxidation of pyrite and the concurrent production of acid. Acid sulphate soils are characterised by low pH, nutrient deficiencies, low base status and high salinity and elevated soluble sulphate, AI, Fe and Mn concentrations (Dent, 1 986). Physical problems associated with these soils include water stress resulting from the inhibition of root development and poor drainage due to the precipitation of colloidal Fe3+ oxides (ochre) in pore spaces and subsurface drains (Dent, 1 986). After an extended period of drought, drainage flushed from acid sulphate soils is characteristically very acidic (PH 3.2 to 3.8) and loaded with dissolved 67 sulphates, Fe2+ and Ae+. Large volumes of this drainage can seriously impact the aquatic environment and in extreme cases result in wholesale annihilation of fresh water fish popUlations (Creagh, 1 992) . In order to effectively manage wastes from geological material containing sulphide minerals it is important to understand the reactions that influence acid generation. A substantial amount of research has been published relating to the processes and consequences of pyrite oxidation and methods o f controlling it (Singer & Stumm, 1 970; Caruccio et al., 1 977; Stumm & Morgan, 1 98 1 ; Costigan et al., 1 98 1 ; Sullivan & Yelton, 1 988 , Ritcey, 1 98 9) . Useful reviews on AMD control and treatment are also included in Sengupta ( 1 993) and Skousen & Ziemkiewicz ( 1 995) . Although much of the research outlined in this l iterature is focused on the coal mining industry, many of the findings can be directly applied to the reclamation of hard rock gold and base-metal mine sites such as Tui. It is important to note that not all hazardous drainages emanating from mine sites exhibit acid pH. Many drainage waters including those emanating from the TUl mine adits contain alkalinity predominantly in the form of dissolved carbonate (C03 "') or bicarbonate (HC03-). The presence of alkalinity does not preclude the water containing mineral acidity in the form of dissolved ferrous iron (Fe2+) and manganese (Mn2+). S ome initially alkaline drainages that contain very high concentrations of these metals are net acid producers. Such drainages can seriously impact the aquatic environment via direct metal toxicity and/or as they acidifY due to the oxidation and hydrolysis of dissolved metal cations (Hedin et ai. , 1 994). Base-metals, including Cu, Pb and Zn commonly occur as sulphide deposits. Thes� deposits are usually poly-sulphidic and contain appreciable quantities of two or more heavy metals that may or may not be of economic significance. Metalloids such as arsenic (As) and antimony (Sb) are found in some sulphide deposits as substituted ions in minerals such as arsenopyrite (FeAsS). Mining wastes originating from poly­ sulphidic ore are often particularly difficult to reclaim. Acidification and dissolution of heavy metal-bearing minerals resulting from sulphide oxidation tend to severely inhibit 68 plant growth and contaminate surface and ground water with these potentially toxic metals. The AGP and reactivity of sulphide minerals varies considerably and relates to several factors, including galvanic interactions and the crystal structure and composition of the mineral (Kwong, 1 995). Not all sulphide minerals, however, oxidise to produce acid. The oxidation of bornite (CuSFeS4) for example, results in the net consumption of acid and other non-ferrous sulphides, such as galena (PbS) and sphalerite (ZnS), undergo oxidation without either producing or consuming acid4 (Bruynesteyn & Hackl, 1 982; Kwong, 1 995). Several iron-bearing sulphide minerals such as pyrite, pyrrhotite, marcasite and chalcopyrite do, however, have the potential to liberate appreciable quantities of acid. Of these minerals pyrite (FeS2) is the most ubiquitous sulphide mineral and is usually the primary source of acidity in mine spoils and AMD (Arora et al., 1 979). This unstable, high energy compound is the most abundant sulphide mineral in the Tui tailings. 3.3.1 Pyrite Oxidation It is widely accepted that, in an oxidising environment, pyrite reacts with oxygen and water to form a series of soluble hydrous Fe and metal sulphates that may in turn hydrolyse to produce highly acidic Fe2+ and sulphate enriched drainages. Subsequent oxidation of Fe 2+ to Fe 3+ and its complexation with water forms additional acidity and a series of ferrous and ferric oxy-hydroxides (Caruccio et al., 1 98 8). The oxidation of pyrite is a complex process which includes several types of oxidation-reduction reactions, hydrolysis, complex ion formation, solubility controls and kinetic effects (Nordstrom, 1 982). Singer & Stumm ( 1 970) interpret the oxidation of pyrite as an electrochemical corrosion reaction where the reduction of H+, H20 or O2 is accompanied by the oxidation of pyrite. 4 See discussion in section 3.3 .2 . 69 The overall oxidation reaction of pyrite is exothermic and is often summarised by reaction ( 1 ) below. This incongruent equation indicates that the ultimate products of the oxidation of pyrite are sUlphuric acid and ferrihydrite (Backes et al. , 1 986). Ferrihydrite or ferric oxy-hydroxide is a highly insoluble, amorphous iron hydroxide gel that is often referred to as ochre or 'yellow boy' (Tucker et al., 1 98 7) . It forms from the hydrolysis of Fe3+ and is a tell-tale sign of AMD when found lining stream or river beds. Reaction ( 1 ) indicates that under ideal conditions each mole o f pyrite will oxidise to liberate 1 mole of ferrihydrite and 2 moles of acid. ( 1 ) This overall equation simplifies the oxidation of pyrite, however, which in reality i s not a simple one-step reaction but a series of generally aerobic reactions involving a number of intermediates (Backes et at., 1 986). Once initiated, the reactions that oxidise pyrite and result in the net production of acid are believed to develop syngenetic interactions that accelerate the oxidation process (Chamber of Mines of South Africa, 1 98 1 ; Caruccio et al., 1 988). These reactions are summarised below (Stumm & Morgan, 1 980). FeS2 + 7/2 02 + H2O � Fe 2+ + 2 sol + 2 H+ (2) Fe2+ + 1 14 O2 + H+ � Fe 3+ + 1 /2 H2O (3) Fe3+ + 3 H2O � Fe (OH)3 + 3H+ (4) FeS2 + 1 4 Fe3+ + 8 H2O � 1 5 Fe 2+ +2 soi- + 1 6 H+ (5) Reactions (2) and (5) indicate that pyrite is initially oxidised by oxygen and then by Fe3+. The oxidation of pyrite is initiated in reaction (2) where oxygen is seen to react with pyrite to release Fe2+ . This reaction is purely chemical and highly dependent on pH, taking place only at an appreciable rate above pH 5 (Singer & Stumm, 1 970). The overall rate of pyrite oxidation is relatively slow above pH 3 .5 because the maj ority of ferric ions (Fe3+) are removed from solution as the result of Fe(OH)3 precipitation (reaction 4). The formation of ferrihydrite renders the Fe3+ unavailable for direct oxidation of pyrite, a reaction which occurs at a much faster rate than pyrite oxidation 70 which involves oxygen only (Dent, 1 986). As the acidity of the system increases, however, Fe3+ become increasingly soluble and begins to act as an oxidising agent (reaction 5). Below pH 3 .0, the maj ority of Fe3+ are dissociated from water and are available to rapidly oxidise pyrite (Nordstrom, 1 982). This reaction readily proceeds under anoxic conditions and can lead to the acidification of sealed pyritic spoils (Wiggering, 1 993). Whereas the ultimate driving force of pyrite oxidation is atmospheric oxygen, the fundamental mechanism and the maj or rate determining step is not thought to involve oxygen at all (Nordstrom, 1 982) . The oxidation of Fe2+ to Fe3+ as described in reaction (3) is widely believed to be the rate limiting step in the oxidation of pyrite. This is due to pyrite being able to reduce Fe3+ to Fe2+ faster than Fe2+ can be regenerated to F e3+ by oxygen (Nordstrom, 1 982). In nature, reaction (3) is usually catalysed by the bacterium Thiobacillus ferrooxidans, which can dramatically increase the overall oxidation rate of pyrite (Zaj ic , 1 969). T ferrooxidans is estimated to increase the oxidation rate of Fe2+ by 5 to 6 orders of magnitude, making the oxidation rate of Fe2+ in the presence of this bacterium, comparable to, or slightly higher than the pyrite oxidation rate by Fe3+ (Lacey & Lawson, 1 970; Singer & Stumm, 1 970; Nordstrom, 1 982). Where the pH of the material has been amended to between 6 and 7, however, the abiotic oxidation of Fe2+, although slow, will be the dominant oxidation pathway and the diffusion of atmospheric oxygen into material will determine the rate of reaction (Van Breeman, 1 973 ; Backes et al., 1 986: Caruccio et al., 1 988). The bacterial catalysis of reaction (3) i s extremely important in that it initiates a cyclical system in which Fe2+ ions released from the initiator reaction (reaction 2) are oxidised to Fe3+ ions which then oxidise pyrite (reaction 5) to liberate more Fe2+ ions (Singer & Stumm, 1 970; Backes et al., 1 986). Research suggests that iron-oxidising bacterial popUlations are responsible for about Yz to % of the acid produced during pyrite oxidation (Zaj ic, 1 968; Caruccio et al. , 1 988) . Whilst a variety of bacteria are known to play a role in the biochemical oxidation of sulphide minerals, the acidophilic bacterium Thiobacillus ferrooxidans is considered to be by far the most important organism involved (Ritcey, 1 989; Ehrlich, 1 995). 7 1 T ferrooxidans i s classified as a chemo-autotrophic bacteria due to its ability to aerobically oxidise Fe2+ and sulphur compounds as a sole source of energy for growth and carbon dioxide fixation (Zajic, 1 969; Brown et al. 1 990). Although the bacterium obtains its oxygen and carbon dioxide from the air, it requires an aquatic environment, ammonia and small quantities of phosphate, calcium and magnesium for growth. The ability of T ferrooxidans to oxidise many other sulphide minerals including chalcopyrite, sphalerite galena, millerite and molybdenite, has resulted in its widespread use in the commercial bioleaching and biobenefaction of low-grade sulphidic ores (Zajic, 1 968; Ehrlich, 1 995) . Optimum pH levels for the growth of Tferrooxidans and hence the catalytic oxidation of sulphide minerals are between 2.4 and 3 .6 (Caruccio et aI. , 1 98 8) . Below pH 2.0 and above pH 3 .6 biotic pyrite oxidation rates decrease dramatically as either the bacterial population is displaced and/or the solubility of Fe3+ decreases (Dent, 1 986). Ritcey ( 1 989) suggests that the optimum temperature for growth of T ferrooxidans is between 3 5 and 45°C. To summarise, acid generation is generally most severe in highly pyritic materials where the acid produced can quickly neutralise carbonates, overcome pH buffering systems and lower the pH sufficiently to solubilise Fe3+ and promote the activity of iron­ oxidising bacteria (chiefly Thiobacilli spp.) which in turn accelerate the oxidation of pyrite by Fe3+ (Backes et al., 1 986). Maintaining the pH of an acid generating material above pH 4.0 is believed to substantially reduce its ability to produce acid as the activity of Thiobacilli is greatly reduced and Fe3+ becomes insoluble and unavailable to oxidise pyrite. Although acid is produced in reactions (2), (4) and (5), the majority of acid forms as a result of the hydrolysis of soluble hydrous metal sulphates formed during the oxidation reactions (reaction 5). Further acid is produced as Fe3+ complexes with water to form ferrihydrite or ferric oxy-hydroxide (reaction 4) (Caruccio et al., 1 988) . 3.3.2 Sulphide Oxidation in situ The above discussion is somewhat simplified in that reactions ( 1 ) to (5) assume that complete hydrolysis of Fe3+ occurs, resulting in the net production of 2 moles of sulphuric acid per mole of pyrite. In an acidic environment, however, complete hydrolysis of the Fe3+ is unlikely to occur and the amount of acid generated in the field 72 situation is likely to be significantly less. Bruynesteyn & Hackl ( 1 982) suggest that Fe is likely to be precipitated as basic ferric sulphate or as jarosite-type minerals5 . If it is assumed that all Fe is precipitated as hydronium-jarosite [H30.Fe3(S04)2(OH)6], Bruynesteyn & Hackl ( 1 9 82) suggest that the oxidation of pyrite would be better summarised by reaction (6) below. In this reaction every mole of pyrite is oxidised to produce 1 .33 moles of sulphuric acid. Bruynesteyn & Hackl ( 1 982) believe that, in a natural situation, the amount of acid produced will depend upon a combination of equations ( 1 ) and (6) and will thus vary between 1 .33 and 2 .0 moles of sUlphuric acid per mole of pyrite. Similarly, the oxidation of 1 .0 mole of chalcopyrite is thought to produce between 0.33 and 1 .0 mole of acid (reactions 7 & 8 respectively). Although jarosite weathering via hydrolysis will generate acidity research outlined by Sobek et al. ( 1 987) found that the effective acidity produced from jarosite hydrolysis is not large compared to that produced by pyrite oxidation (Sobek et al. , 1 98 7) As previously mentioned, other non-ferrous sulphides, such as galena and sphalerite undergo oxidation. Acid, however, is neither produced nor consumed and the reaction can be represented by reaction (9). Although acid is not produced by the oxidative dissolution of these minerals, the resulting sulphate salts have an acidic pH which may mobilise the metals that they contain (Bruynesteyn & Hackl, 1 982) . Dissolved sulphate, Pb2+ and Zn2+ liberated during the reaction can contaminate the soil solution and 5 A.Fe3(S04)z(OH)6 where A can equal H30+, NH/, K+, Na+ etc. Jarosite commonly occurs as earthy filling of voids or as mottles in the soil matrix and is only fonned in acidic (PH 2 to 4), oxidising (Eh > 400 mY) environments (Sobek et al. , 1987). From a plant growth perspective the fonnation of potassium jarosite [KFe3(S04)z(OH)6J is significant in that it tends to precipitate K and thus reduce the availability of this macronutrient to plants (Barnhisel et al., 1982). 73 drainage waters. Fortunately the mobility of Pb2+ is comparatively very low due to its tendency to react with sulphate to form highly insoluble anglesite (PbS04) under oxidising conditions. MS + 2 O2 � MS04 (where M = Zn, Ni, Pb & Co etc . ) (9) Pyrite is thus seen to be the most significant source of acid and can theoretically produce as much as 2 moles of acid per 1 mole of pyrite. The amount of acid generated in reality, however, is likely to be significantly less due to incomplete SUlphur oxidation. A variety of oxidation products can potentially form in pyrite-bearing materials ranging from insoluble iron hydroxides such as ferrihydrite [Fe(OH)3J to highly soluble iron sulphate hydrates including, melanterite (FeS04.7H20), rozenite (FeS04.4H20) and szomolnokite (FeS04.H20) (Nordstrom, 1 982). The more soluble sulphates generally form during dry periods (evapotranspiration > rainfall) which invoke capillary rise, the subsequent concentration of ions and the precipitation of efflorescent salts (Nordstrom, 1 982; Sobek et ai., 1 987) . Goethite (FeOOH) and haematite (Fe203) in contrast are the most stable of the oxidation products and form from the dehydration of ferrihydrite and/or jarosite-type minerals. 3.3.3 Factors Affecting Sulphide Oxidation Rates As alluded to in the previous discussion, there are a multitude of factors that affect the reaction rate of sulphide minerals. These include; crystal structure and adsorbed impurities (Pugh et ai., 1 984; Caruccio et ai., 1 988, 1 995; Keller & Murr, 1 982; Kwong, 1 995), surface area (Singer & Stumm, 1 970; Caruccio, 1 973; Pugh et ai., 1 984; Arora et ai. , 1 979), temperature (Caruccio et ai., 1 988; Scharer et ai., 1 995), oxygen concentrations (Pugh et ai., 1 984; Erickson et at., 1 985 ; Wiggering, 1 993 ; Scharer et ai., 1 995; Lin, 1 997), water partial pressure (Caruccio et at., 1 988), solution pH (Costigan et ai., 1 984; Gemmell, 1 98 1 ; Backes et ai., 1 986),jlushing frequencies (Geidel, 1 979; Caruccio et ai., 1 988), galvanic interactions (Natarajan, 1 990; Tuovinen, 1 990, Kwong, 1 995) iron-oxidising bacteria (Zaj ic, 1 969; Brown et at., 1 990; Ehrlich, 1 995) and inhibitors (Baldi & Olson, 1 987; Ritcey, 1 989) and of course time. Caruccio et ai. ( 1 988) provide a useful review of the various factors governing the oxidation rate 74 of pyrite with an emphasis on the role of this mineral in the production of acid from colliery spoils. 3.4 Acid Mine Drainage Based upon the preceding discussion it is evident that solution pH is one of the most important factors controlling the rate of sulphide oxidation as it not only governs the solubility/availability of Fe3+ but also the activity of iron-oxidising bacteria responsible for the biocatalysis of sulphide minerals. The presence of reactive (acid generating) sulphides, iron-oxidising bacteria, water and oxygen and low concentrations of acid neutral ising carbonate minerals can ultimately lead to the release of acid mine drainage (AMD) which is a solution characterised by low pH and high concentrations of sulphate, Fe and other metals (Caruccio, 1 975). Some researcher define AMD even further as , . . . water with a pH below 2.3, anion concentrations exceeding 1 0,000 mg/l and acidities averaging 5,000 mg/l as CaC03' (Caruccio et al., 1 98 1 cited in Tucker et al. , 1 987). Acidity in AMD is comprised of hydrogen ion acidity (a measure of free H+ ions), organic acidity associated with dissolved organic compounds and mineral acidity which arises from the presence of dissolved metals commonly AI, Fe and Mn in the water (Hedin et at., 1 994). These metals generally account for most of the mineral acidity found in coal spoil, however, drainage from base-metal mines such as the Tui Mine, typically contains significant amounts of Cd, Cu, Pb and Zn which can contribute significantly to mineral acidity (Pang, 1 992; Skousen, 1 995). The drainage quality of water emanating from sulphide bearing material is largely dependent on the ratio of pyrite to carbonate. In general, sulphide rich and carbonate poor materials are expected to produce acidic drainage if exposed to oxidising conditions (Skousen, 1 995). The presence or absence of calcareous material in pyrite­ bearing strata is thus generally regarded as the most critical element in determining drainage quality (Caruccio et at., 1 988). The pH of the effluent produced from the waste materials plays an important role in solution quality. Not only does the pH control the solubility of the various metals but in the case of the sUlphide wastes, a low pH encourages biological breakdown of metal sulphides, thus increasing the rate of 75 solubilisation of corresponding metal sulphides as well as production of ferric sulphate lixiviant (Bruynesteyn & Hackl, 1 982). In the case of acid sulphate soils, while most of the Fe mobilised by the oxidation of pyrite remains in the soil profile only a small fraction of the sulphate is retained, usually as jarosite or as basic Al sulphate (Dent, 1 986). Most soluble sulphur is lost to drainage although some diffuses down to the reduced substratum where it may re-precipitated as sulphides (Dent, 1 986). Gypsum and soluble sulphates often accumulate at locations where evapotranspiration exceeds rainfall (Sobek et al. , 1 987). The concentration of metals and metalloids in mine drainages depends on a multitude of factors including Eh, pH, water composition (eg. high Fe3+ levels may result in co­ precipitation and/or adsorption of heavy metals), flow rates, sediment grain-size and the type and extent of mining operations (Livingston, 1 987). Highly toxic AMDs are often generated from base metal mines containing metals and metalloids such as As, Cd, Cu, Hg and Pb. These potentially toxic metals may be liberated into the aquatic system and food chain as a result of sulphide oxidation, acidification and dissolution processes occurring in host materials. Whilst it is beyond the scope of this study to further reView the production, geochemistry and impact of AMD, it must be stressed that it is a widespread phenomenon that affects numerous active and abandoned mine sites around the world. In the US alone it is estimated that AMD adversely affects over 6,500 km of rivers and streams and 730 km2 of lakes and reserves (Kleinmarm, 1 99 1 ) . Sites impacted by AMD are found in a diverse range of countries including Wales (Fuge et al., 1 99 1 ), England (Davies & Ballinger, 1 990), Spain (Perez & Calvo de Anta, 1 992), Greece (Kelepertsis & Bibou, 1 99 1 ), USA (Kleinmarm, 1 99 1 ; Skousen, 1 995) Canada (Ritcey, 1 989), Australia (Harries & Ritchie, 1 988; Bell et al. , 1 989) and of course New Zealand (Tay, 1 980; Hendy, 1 98 1 ; Pang, 1 992, 1 995). A variety of useful literature pertaining to AMD exists (Singer & Stumm, 1 970; Helz et al. , 1 987; Ritcey, 1 989; Skousen et al. , 1 990; Kleinmann, 1 99 1 ; Sengupta, 1 993 ; Skousen & Ziernkiewicz, 1 995). 76 3.4.1 Acid Mine Drainage Prediction Several approaches have been developed to predict mine drainage quality including:- (i) comparison with neighbouring mines, (ii) geological and paleoenvironmental mapping, (iii) geochemical static tests, (iv) geochemical kinetic tests and (v) mathematical modelling (Ritcey, 1 989; Sengupta, 1 993). Whilst all of these methods are potentially useful, the foHowing review focuses on the two most commonly used methods of assessing the Net Acid Generating Potential (NAGP) and liming requirements of sulphide-bearing mine wastes in Australasia (Acid Base Accounting and Hydrogen Peroxide Oxidation). A cursory discussion of alternative kinetic tests is given. 3.4. 1 . 1 Acid Base Accounting (ABA) Acid base accounting is a static test that provides a theoretical estimate of the NAGP of a geological material, based on the difference between the acid-forming and acid­ neutralising capacities of the material (Sobek et al. , 1 987; Miller et al., 1 994). ABA was the first technology available to assess the chemical quality of in situ overburden. This prompted the development of pre-mine planning for reclamation and selective handling procedures of acid generating material, practices which are now commonplace and usually required by law (Skousen et al. , 1 995). (i) The Development of ABA The ideas that initiated the development of ABA originated from research conducted in the early twentieth century investigating the composition and relative reactivity of elemental compounds in the Earth's crust (Skousen et al. , 1 995). The method was chiefly pioneered and trialed in the coal fields of eastern USA, with reports acknowledging the effects of spoil quality on plant growth dating back to the 1 940' s (Skousen et al. , 1 995) . Early ABA systems focused on the categorisation of mine spoil into classes based on rock composition and acidity. Knabe ( 1 964) was amongst the first researchers to document the importance of the relative proportions of acid to non-acid generating minerals in mine spoil. He developed a method called a 'base-acid balance' which estimated the theoretical quantity of acidity or alkalinity that a mine spoil could produce upon weathering (Skousen et al., 1 995). The early methods, although useful for 77 detennining lime and fertiliser application requirements and selecting suitable plant species for revegetation purposes, fell short of being able to accurately estimate the potential of undisturbed overburden to generate acid (Skousen et al. , 1 995). Thus, suitable, non-acid generating material was not always accessible for revegetation purposes because spoil management couldn't be accurately planned prior to mining. Much of the subsequent ABA research was conducted by researchers at West Virginia University, where a system for balancing the acid and alkaline production potential of in situ overburdens was developed (Smith et al., 1 974; Smith & Sobek, 1 978; Sobek et al., 1 978). Soon after the development of such a system, practitioners began to log SUlphur profiles and the acid-neutralising capacities of materials in stratigraphic sections overlaying coal seams prior to mining. This enabled the net acidity or alkalinity for each geologic unit to be calculated and an estimate of the material ' s NAGP to be made (Skousen et al., 1 995). Whilst a variety of factors are known to influence the acid producing potential of geological materials, ABA was developed on the premise that the mineral composition of a material largely dictates its NAGP (Skousen et al., 1 995). (ii) ABA Methodology The ABA procedure firstly involves estimating the Acid Generating Potential (AGP)6 of a material. For coal overburden this is usually done by detennining the sulphidic sulphur content of a material. The AGP of a material is then estimated stoichiometrically by assuming that two moles of CaC03 will be required to neutralise the acidity potentially produced from the oxidation of 1 .0 mole of pyrite as per reactions ( 1 ) and (2). FeS2 + 1 5/4 O2 + 7/2 H20 H2S04 + CaC03 Fe (OH)3 + 2 H2S04 CaS04 + H20 + CO2 ( 1 ) (2) Based on the stoichiometry of reaction (2) it holds that each mole of pyrite contains 2 moles of sulphur which if oxidised will require 2 moles of CaC03 to neutralise it. Using the molecular weights of sulphur (�32) and calcium carbonate (�100) it is possible to calculate the theoretical quantity of CaC03 required to neutralise the acidity produced 6 Also commonly referred to as Maximum Potential Acidity (MPA) or Acid Producing Potential (APP). 78 from 1 000 Mg of material containing 0. 1 % ( 1 000 kg) of sulphidic sulphur (Ss) as follows; Since 32 g Ss 1 000 kg Ss 1 00 g CaC03 ( 1 00/32) * 1 000 CaC03 3 1 25 kg CaC03 The material 's AGP is generally presented as calcium carbonate equivalents [Mg CaC0311 000 Mg material] . Based on the stoichiometry of the oxidation of pyrite and its reaction with calcium carbonate, 1 000 Mg of material containing 0 . 1 % sulphur (all as pyrite) theoretically yields an amount of sulphuric acid that would require 3 . 1 25 Mg of CaC03 to neutralise it (Smith et al., 1 974; Skousen et al., 1 995). The second stage of the ABA method involves determining the material ' s neutralising potential (NP). The base content of a material (commonly present as carbonates or exchangeable cations on clays) significantly influences the material ' s AGP (Smith & Sobek, 1 978; Geidel, 1 979; Skousen et al. , 1 995). In a material containing alkaline components and pyrite the soluble bases must generally be sufficiently active to neutralise the acid at a rate equal to or exceeding the rate of acid production to prevent the material 's acidification under oxidising conditions (Skousen et al. , 1 995). The most common means of assessmg a material 's acid neutralising potential (Neutralisation Potential, NP) is to chemically digest a pulverised sample in an acidic solution of known concentration (0. 1 0M HCI is often used) . The solution is then back­ titrated to a predetermined endpoint with a standard base (NaOH) to determine the amount of acid consumed by the material. The base account is calculated from the weight of sample digested, the volume and normality of the acid and the amount of titrant used (Sobek et al.J 1 978). The NP of a material is again converted to calcium carbonate equivalents. Using ABA the NAGP of a material (and thus the quality of drainage likely to emanate from the material) is predicted by subtracting the material 's AGP from the material 's NP after conversion to CaC03 equivalents. A negative NAGP value indicates that the material is capable of producing acid mine drainage, whereas a positive value --------- --- 79 conversely indicates that the material is likely to produce circumneutral or alkaline drainage (Evans & Skousen, 1 995). (iii) Limitations of ABA The major limitations associated with traditional ABA are that:- (i) all sulphur is generally assumed to be present as pyrite, (ii) it is assumed that all of the acid consuming minerals are available for acid neutralisation, (iii) results are difficult to interpret if AGP and NP are near or in balance and (iv) the reaction kinetics of the system are not considered (Miller & Murray, 1 988 ; Ritcey, 1 989). Since ABA has largely been developed by the coal mining industry for the assessment of colliery spoil, it is based on the assumption that all acidity originates from the oxidation of pyrite (Skousen et ai., 1 995). Although this assumption may hold true for the majority of colliery spoils, it may overestimate the AGP of materials containing appreciable amounts of non-acid generating sulphide minerals, particularly if the AGP of the material is estimated from sulphur analysis of the material. Overestimation is likely to stem from the inability of standard sequential sulphur extractions to adequately differentiate between acid generating and non-acid generating sulphide sulphur. The pyrite content of polysulphidic materials may therefore be better predicted by determining pyritic Fe content. Allowances must, however, be made for materials containing substantial quantities of non-pyritic but acid generating minerals such as covellite (CuS) or chalcopyrite (CuFeS2). Determination of pyritic Fe alone would fail to take into account the acidity that oxidation of these sulphide minerals would generate. It was thus postulated that without significant modification, traditional ABA methods would overestimate the AGP of the Tui mine tailings due to the occurrence of relatively high concentrations of sulphide minerals such as galena (PbS) and sphalerite (ZnS) which are considered to be non-acid generating (Bruynesteyn & Hackl, 1 982; Kwong, 1 995). Research by Cravotta et al. , ( 1 990) and Brady et al., ( 1 994), however, suggests that traditional ABA methods may in fact substantially underestimate the liming 80 requirements of acid generating materials. The standard ABA method outlined above presumes 2 moles of H+ may be neutralised by 1 mole of CaC03. The stoichiometry of this reaction thus assumes that all CO2 will exsolve. In reality, however, some CO2 will dissolve forming a weak acid that reacts with carbonate minerals. Assuming no exsolution of CO2, 1 mole of FeS2 will produce 4 moles of H+, which may be neutralised by 4 moles of CaC03. On a weight basis, 1 g of pyritic sulfur will thus be neutralised by 6.25 g of CaC03 . Considering this equivalent weight, the AGP as CaC03 deficiency per 1 ,000 Mg of overburden should be computed by multiplying total sulfur, in weight percent by 62.5 instead of the factor that has been previously used (3 1 .25) . Cravotta et ai. , ( 1 990) concede that actual acidity may be less than the computed AGP because not all CO2 dissolves and not all sulfur generates acidity. The methods commonly used to assess the NP of mine wastes have also drawn some criticism in recent years. Lapakko ( 1 994) suggests that digestion methods used to assess NP of wastes may overestimate the NP present as calcium carbonate and magnesium carbonate. The overestimation is attributed to the fact that whilst at low pH (during the digestion) the dissolution of acid consuming minerals (eg. plagioclase, pyroxene and olivine) present in the host rock neutralise acid, under field conditions the rate of acid neutralisation by such minerals is often too slow to neutralise generated acid. The presence of high concentrations of siderite (FeC03) may further contribute to excessive NP estimates due to inadequate quantification of Fe2+ acidity. Errors in NP estimation may result in the erroneous classification of acid generating material as non acid­ generating (Lapakko, 1 994) . Despite the apparent simplicity of ABA this method cannot be used to definitively predict the quality of drainage emanating from mine wastes at any given time in the future. Acid generating processes and thus the production of acid mine drainage are time-dependent and are functions of numerous interactive factors such as mineralogy, grain-size, site hydrology and climate. For this reason ABA can only be used as a qualitative prediction method to indicate whether or not materials have the potential to generate net acidity in the future (Sengupta, 1 993). ABA does, however, provide a useful guide for estimating the liming requirements of acid generating materials where -- - - ---- -------- -- 8 1 pH amendment IS the intended means of controlling the generation of acid mme drainage. 3.4.1 .2 Hydrogen Peroxide Oxidation (HPO) The ability of hydrogen peroxide (H202) to rapidly oxidise pyritic material has long been recognised and several procedures to determine the potential acidity of pyritic material are based on this reaction (Sobek et al. , 1 978; O' Shay et al. , 1 990; Finkelman & Giffin, 1 986; Finkelman, 1 987; Miller et al. , 1 994) . Although several HPO methods have been devised, they are all basically variations of one another, with many of the more recent variants being based largely on the method of Smith et al. ( 1 974). (i) HPO Methodology The principal of these methods is to use H202 to rapidly oxidise pyrite which results in the subsequent production and liberation of hydrogen ions (H+) and metal sulphates into solution. A semi-quantitative estimate of the AGP of the material can then be made by quantifying either the titratable acidity or the amount of sulphate (S04 =) produced during the oxidation reaction. Qualitative HPO methods also exist and are commonly applied in the field as a means of rapidly assessing the ability of a material to produce acidity. Although the exact procedures of these qualitative tests tend to vary widely, they basically involve the oxidation of a set amount of material with a specific volume of dilute H202. The oxidation of the pyrite results in the liberation of H+ and their concentration is measured indirectly using a standard soil pH meter. The degree of acidification (PH) indicates the material 's ability to produce acid. Many of the methods use an arbitrary solution value of pH � 3 .0 to indicate if the material is likely to acidify and potentially generate AMD (Blakemore et al., 1 987). (ii) Limitations of HPO Methods Several previous studies have indicated that the HPO methods used for predication of AGP of colliery spoils are unreliable (Dacey & Colbourne, 1 979; Sobek et al. , 1 987; Caruccio et al. , 1 988). Research reported by Dacey & Colbourne ( 1 979) assessed six methods for the determination of iron pyrites in coal mine spoil, including one HPO --- ------------ --- 82 method that quantified S04 = liberation. One of the disadvantages recognised with the BPO method studied was that when it was used to assess spoils containing carbonate minerals, erroneously low pyrite content determinations were made. Even when the samples were pre-treated with BCI to prevent the catalytic decomposition of B202 by carbonates, considerable scatter in the results was found. Although the method was found to be capable of distinguishing between samples of low, medium and high pyrite content it was not considered to be reliable (Dacey & Colbourne, 1 979). More recently, a comparative study was made between Acid-Base Accounting and BPO prediction methods (Ammons & Shelton, 1 988). In this study nearly one hundred samples of overburden of varying sulphur content (0.006- 1 .328 %) were analysed to compare these procedures. Results indicated that for unweathered samples there was a positive correlation between the two methods, however, the relationship was found to be less pronounced for weathered overburden samples. The coefficients of determination (r2) for the unweathered and weathered samples were 0.89 and 0.7 1 respectively. The authors suggested that the BPO procedure was best used to predict the NAGP of fresh, unweathered samples (Ammons & Shelton, 1 988). Caruccio et al. ( 1 988) also report that BPO methods tend to give highly variable results and may grossly overestimate the potential acidity of samples high in organic sulphur. Sobek et al.( 1 987) suggest that BPO methods can be used successfully in some overburden samples providing the unit mineralogy is "simple", unstabilised B202 is used and excess B202 is destroyed prior to titration. The authors did not recommend the procedure for the routine analysis of all overburden. Finkelman & Giffin ( 1 986) similarly reported that the use of B202 to rapidly oxidise pyrite has met with "decidedly mixed results". (iii) Application of HPO Methods Despite the fact that there is much debate about the ability of BPO methods to accurately predict the AGP of pyritic material, development of the method is continuing. An BPO method is currently being used to identify potentially acid forming materials at mine sites in Australia, New Zealand, Papua New Guinea and Indonesia (Miller et al., 1 994). Research is being conducted to evaluate this same method to see if the kinetics of acid 83 generation and neutralisation can be predicted. The potential of HPO methods to indicate pyrite activity was recognised in earlier research (Dacey & Colbourne, 1 979; Finkelman & Giffin, 1 986). Preliminary investigations by Miller et al. ( 1 994) indicate that a relationship may exist between the kinetics of the HPO method and the lag period of acid generation observed in the field. The paper suggests that if a relationship is established it may be possible to provide a time factor for use in waste management design (Miller et al. , 1 994). Most of the HPO methods developed to date (with the notable exception of that of Miller et at. , 1 994) have been designed almost exclusively for assessing the AGP of colliery spoil. When used to assess the AGP of poly-sulphidic mine tailings such as the Tui tailings, semi-quantitative HPO methods based on S04 = determination may tend to overestimate the material ' s AGP. These methods assume that all the SO/ produced by the oxidation reaction results from the oxidation of pyrite. In the case of poly-sulphidic tailings a significant proportion of the S04 = may in fact be produced by the oxidation of non-pyritic sulphides such as galena (PbS) or sphalerite (ZnS). Bruynesteyn & Hackl ( 1 982) and Kwong ( 1 995) report that whilst these non-pyritic minerals undergo biochemical oxidation, acid is neither consumed nor produced during the solubilisation reaction. Therefore in the case of a material which contains appreciable quantities of non-ferrous sulphides, the assumption that all SO/ is derived from the oxidation of pyrite is erroneous and will result in the overestimation of the materials pyrite content and thus its AGP. It is postulated that the HPO method of Miller et al. ( 1 994) which is based on the quantification of titratable acidity (as opposed to SO/ content) may also overestimate the AGP of poly-sulphidic tailings that contain significant quantities of non-pyritic sulphide minerals. Although these minerals reportedly react with H202 to release hydrogen ions (Finkelman & Giffin, 1 986) their biochemical oxidation does not result in the net production of acid (Bruynesteyn & Hackl, 1 982; Kwong, 1 995). The extent to which the method of Miller et al., ( 1 994) would overestimate the AGP of poly-sulphidic tailings would clearly depend on the relative proportions of pyritic to non-pyritic sulphides in the material. 84 3.4. 1 .3 Alternative Prediction Methods Over the past few decades a variety of alternative procedures have been developed to characterise mine wastes. Caruccio ( 1 975), for example, discusses the combined use of optical mineralogy and sUlphur analysis to predict the reactivity and net acid generating of pyrite bearing mine wastes. Pulford & Duncan ( 1 978) outline a very different method to qualitatively assess the degree of weathering and the reactivity of coal wastes based on plotting pH against acetic acid extractable (0 . 5 M) sulphate levels. Kinetic AMD prediction tests differ from the aforementioned static tests in that they involve the weathering of mine material under field or controlled laboratory conditions and are designed to provide information relating to the reaction rates and temporal variation of acid generation. Due to their nature they are run over much longer periods (generally several weeks) and require a greater commitment of resources. They are often used to confirm the findings of static tests and to select and optimise appropriate drainage control techniques based on the predicted severity and likely duration of AMD (Sengupta, 1 993 ; Miller et al., 1 994). There are numerous types of kinetic tests; the most commonly encountered are summarised in Table 3 . 1 . Kinetic Test Table 3.1 Kinetic Tests for the Prediction of Acid Mine Drainage (adapted from Ferguson , 1 986 cited in Ritcey, 1 989). Principles Designed to simulate the weathering of References Soxhlet Reactors geological material. Uses water or dilute acid Sullivan & Sobek ( 1 982) to repeatedly leach material in a sealed reaction vessel. Leachates are analysed. Bacterial Sample is placed within reaction vessel and Bruynesteyn & Hackl Incubations/ inoculated with Thiobacillus ferrooxidans. ( 1 982) Shake Flasks Samples incubated for several weeks and leachates are analysed periodically. Humidity Cells Sample is placed in a sealed container Caruccio ( 1 977) through which humid air and then water is Sobek et al. ( 1 978) passed. Leachates are analysed Sullivan & Sobek ( 1 982) Columns Material is placed in a column and leached Sturey et al. ( 1 982) periodically with distilled water to simulate in situ leaching by rainfall. Resulting leachates are analysed. Test Plots Bulk sample of mine waste is placed on Eger & Lapakko ( 1 98 1 ) impervious surface in field. Leached precipitation is collected in a sump and monitored over long term (> I year) . ------- 85 Review of the literature clearly indicates that currently there is no universally accepted approach to predicting the net acid generating potential of sulphide-bearing mine wastes. Sengupta ( 1 993) suggests that accurate prediction of long term drainage quality is only possible through the application of quantitative mathematical models that can reliably extrapolate the results of kinetic tests. 3.4.2 Metal Quantification in Mine Materials Based on the perceived limitations of most standard static tests it was decided to approach the prediction of the Tui tailings' AGP using a less conventional approach and to compare the results of the devised method (Theoretical AGP) with those obtained using the HPO method of Miller et al. ( 1 994). The TAGP method is outlined fully in Chapter 4 (section 4.4) but basically involved the quantification of sulphidic Cu and Fe from which the tailings' chalcopyrite and pyrite content (and ultimately TAGP) was estimated. The TAGP technique is similar in principal to pre-existing methods that have been developed for determining the pyrite content of colliery spoil via Fe analysis? The T AGP method used in this study required the quantification of two forms of Cu and Fe in the tailings; acid generating (sulphidic) and non-acid generating (non-sulphidic) Cu and Fe. One of the primary objectives born in mind during the selection of suitable extractants for the determination of these fractions was to use extractants that would also provide useful information relating to the 'plant available' and total metal content of the tailings. In addition the speed, simplicity and versatility of the �xtractants and/or extraction procedures was also considered. The following literature review was undertaken to identify potentially useful extractants available for the quantification of sulphidic and non-sulphidic metals in the Tui tailings and other mining wastes. 7 Discussions of AGP prediction using Fe analysis are included in Burns ( 1 970), Dacey & Colbourne ( 1 979) and Sobek et al. ( 1 987). 86 3.4.2.1 Non-Sulphide Fraction Although a great number of secondary iron-bearing minerals such as rozenite (FeS04.4H20), ferrohexahydrite (FeS04 .6H20) and melanterite (FeS04.7H20) can potentially form in the Tui tailings as a result of sulphide oxidation, these minerals are highly soluble and thus generally unstable and prone to leaching in acidic conditions. The majority of non-sulphidic iron in the tailings was therefore believed to be associated with relatively insoluble secondary iron phases including amorphous ferrihydrite (Fe(OH)3), goethite (FeOOH) and sulphates such as jarosite (KFe3(S04)2(OH)6) which are stable at low pH (Sullivan et ai. , 1 988). The mobility of non-sulphidic Cu in the tailings was expected to be relatively high due to the lack of both organic matter and high CEC clays. Like many of the hydrated Fe sulphate minerals, the majority of the copper sulphate minerals such as chalcanthite (CuS04.5H20), antlerite (CU3S04(OH)4 and brochantite (Cll4S04(OHkH20) are highly soluble. The primary form of non-sulphidic copper in the tailings was thus predicted to be present as adsorbed Cu2+ on hydrous iron/manganese oxides and aluminosilicate gels which are known to scavenge copper and other transition metal ions released from dissolution of sulphides (Ritcey, 1 989; Walder & Chavez, 1 995). For the purposes of this study the non-sulphidic fraction of the tailings was considered to broadly correlate to labile Cu and Fe forms in the tailings and was thus considered to be best estimated by using extractants suited to extracting 'plant available' (labile) Fe and Cu. 3.4.2.2 Extractants Although commonly requested by regulatory authorities, total elemental analysis of mining wastes is generally of little use in assessing the potential toxicity of the material to plants and the environment in general (Berg, 1 978; Safaya et ai., 1 987). Plant available trace element concentrations are thus generally estimated using one of what Brooks ( 1 987) describes as a "bewildering" range of soil extraction procedures that have been developed (eg. Lindsay & Norvell, 1 969, 1 978; Baker & Amacher, 1 982; Jopony & Young, 1 993, Luo & Rimmer, 1 995). Despite being the focus of much research over the past few decades, there is still little consensus as to the extractant and procedure 87 which should be universally adopted (Brooks, 1 987). Bioassay procedures, although generally considered to offer the most reliable data pertaining to the availability of various elements to plants are generally time consuming and expensive (Chadwick, 1 973 ; Brooks, 1 987; Luo & Rimmer, 1 995). Many of the commonly employed extraction methods have been developed to detect deficiencies rather than toxicities and are therefore of dubious value when it comes to estimating the bioavailability of 'micronutrients' in mining wastes which may be present in concentrations 1 000-fold greater than those found in 'natural ' soils (Berg, 1 978). Although the complexity of the extractants and procedures varies substantially, the two most commonly used types of soil extractants are weak inorganic acids and chelating agents. (i) Chelating Agents The use of a chelating agents as a potential extractant to assess the concentrations of 'non-sulphidic' Cu and Fe and other cations in the Tui mine tailings was carefully considered. Although extraction procedures involving chelating agents such as EDTA (ethylene-diaminetetraaceticacid), DTP A (diethylene-triaminepentaaceticacid) and the Mehlich reagents were first developed to identify trace-element nutrient deficiencies in near-neutral and calcareous soils used for cropping (Lindsay & Norvell, 1 969; 1 978; Mehlich, 1 984), they are being increasingly used to assess toxicities in metal contaminated soils and mining wastes (Gough et al., 1 980; Haq et al., 1 980; Severson & Gough, 1 983 ; Norvell, 1 984; Ma & Lindsay, 1 993 ; Hetrick et al., 1 994; Shetty et al., 1 994). Since plants exhibit considerable genetic variation in their ability to tolerate non­ essential metals such as Cd and Pb the need to quantify the bioavailability of such elements is now recognised as being extremely important (Salisbury & Ross, 1 992) . Although a number of chelating agents can effectively complex Cu and Zn in soils, DTPA is one of the most commonly used owing to its ability to effectively chelate Fe and Mn (Lindsay & Norvell, 1 978 ; Neuman et al., 1 987). The extraction method of Lindsay & Norvell ( 1 969) has been successful with acid and reduced soils far beyond the original scope of the application, due mainly to the versatility of DIP A, which 88 provides effective chelation of a wide spectrum of metal cations, including the highly insoluble ferric ion (Norvell, 1 984). Chelating agents function as metal extractants by complexing free metal ions in solution. This reduces the activity of free metal ions in solution and in response metal ions desorb from soil surfaces or dissolve from labile phases to replenish the solution. The amount of chelated metals that are extracted from soils depends largely on two factors, ( 1 ) the activity of the metal ions in the soil and (2) the ability of the soil to replenish those ions (Lindsay & Norvell, 1 978). Research has shown that environmental factors such as length of shaking, pH, temperature and DTP A concentration significantly influence the amount of metal cations extracted by DTP A (Lindsay & Norvell, 1 978; Joshi et at., 1 983; Norvell, 1 984). Lindsay & Norvell ( 1 978) found that by increasing the concentration of the DTP A extractant a greater amount of metal could be extracted from soils, due to more stress being placed on the labile metal phases. Increasing pH was also found to dramatically decrease the concentrations of Mn and Fe extracted. The use of DTP A to assess the plant availability of metals in acidic and metal contaminated soils has, however, been called into question in several studies (Dudas & Pawluk, 1 977; Gough et at., 1 980; Haq et at., 1 980; Norvell, 1 984). Research indicates that for strongly acid and metal enriched soils, the DTP A extraction procedure of Lindsay & Norvell ( 1 969) may be unsuitable becau�e it involves buffering the soil at pH 7.3 and the addition of CaCh to preclude the dissolution of CaC03 and the release of occluded nutrients (Lindsay & Norvell, 1 978; Gough et at., 1 980; Norvell, 1 984). Norvell ( 1 984) suggests that by doing so, the extraction of metals from acid soils may be retarded or otherwise altered to the extent that estimates of plant-availability are distorted. It is also suggested that the pH -buffering capacity of the buffer used in the procedure (O. 1M triethanolamine) is not adequate to maintain a relatively constant pH when strongly acid soils are extracted due to the fact that these soils often contain sufficient titratable acidity to neutralise most or all of the remaining conjugate base of the buffer (Norvell, 1 984). 89 Additionally, Norvell ( 1 984) found that the concentrations of DTPA and extractant to soil ratio of 2: 1 recommended by Lindsay & Norvell ( 1 969) is not always sufficient to extract all of the plant available micronutrients in metal rich soils. Adjustments in the concentration of the agent, the number of extractions and the soil to extractant ratio may be required so as to facilitate comparison with other soils. Lindsay & Norvell ( 1 978) state that in the vicinity of 600 ppm of Cu, Fe, Mn and/or Zn would be required to saturate the amount of DTP A used in soil tests. Although the chelating capacity is generally sufficient when dealing with uncontaminated soils, it is suggested that the Tui tailings could potentially saturate the DTP A exchange sites. Research undertaken by Joshi et ai. , ( 1 983) investigating the influence of several soil parameters on DTP A extractable micronutrients found that CaC03 content, the reduction-oxidation status and soil pH exert maj or influences on the availability of Fe and Mn in soils. Lindsay & Norvell ( 1 978) suggested that if DTPA is to be used to extract Zn in soils with pH below 7 .0, inclusion of pH in the regression equation should be considered due to the solubility of Zn being strongly pH dependent. While investigating the utility of EDT A and DTP A in range of diverse soils, Gough et ai. ( 1 980) also found a strong relationship between soil pH and metal concentrations in plants. Based on this assessment of previous research it was decided not to proceed with the use of a chelating agent such as DTP A to assess the bioavailability of metals in the Tui Mine tailings. Although DTP A has great potential as an extractant for Zn and Cd on mined land, its suitability for such purposes has not been adequately assessed and the influence of variables such as pH, CEC and organic matter content is not fully understood (Haq et ai. , 1 980; Safaya et ai., 1 987). Research by Norvell ( 1 984) found that the use of chelating agents was limited when extracting metal-rich soils at low pH. Considering the fact that the Tui tailings are both acidic and metal-rich, the assessment of the bioavailability of the metals in the tailings using DTP A would have involved considerable experimentation beyond the scope of this study. 90 (ii) Dilute Acids Varying concentrations of a wide range of acid extractants, including acetic (CH3COOH), nitric (HN03), sulphuric (H2S04) and hydrochloric (HCI), have been used to estimate the availability of plant micronutrients such as Cu, Fe and Zn in soils (Dudas & Pawluk, 1 977; Lagerwerff, 1 97 1 ; Haq et al., 1 980; Baker & Amacher, 1 982; Haynes & Swift, 1 985a, 1 985b) . Amongst these extractants, dilute HCI and CH3COOH are two of the most commonly used. The results obtained through the use of CH3COOH-based extractants in the determination of ' soil ' metal concentrations are like that for majority of acid extractants, highly variable and in many cases contradictory. Although some research suggests the use of dilute CH3COOH (commonly O.SM) as an extractant results in an overestimation of plant available concentrations of soil elements, other research outlined by Brooks (1 987) demonstrated that bioassay data for several elements agreed well with values obtained from CH3COOH extraction. Haq et al., ( 1 980) also found dilute CH3COOH to be a useful extractant. Their study looked at nine different extractants in an attempt to find a suitable one for estimating the plant-available Zn, Cd, Ni and Cu content metals in metal-contaminated soils. Although none of the extractants investigated proved suitable for all four metals, 0 .5 N acetic acid was found to be the best extractant for estimating plant available Cd and Ni, with only pH being required as an independent variable (Haq et al. , 1 980). Acetic acid has also been used as a sulphate extractant in a study that assessed the potential acidity of colliery spoils (Pulford & Duncan, 1 978). The authors stated that although 0 .5M acetic is unlikely to extract sulphate present as jarosite or other basic iron and aluminium sulphates, it does extract the more soluble forms. The inability of acetic acid to solubilise some forms of sulphate is attributable to it being a relatively weak acid that does not fully dissociate in aqueous solution, as indicated by its small acid dissociation constant (Ka = 1 .8 x 1 0.5). Using Ka it is possible to calculate the hydrogen ion concentration and hence pH of a O .SM solution of acetic acid as follows; Ka = 1 . 8 x 1 0 -s = [H +] [CH300-] [CH3COOH] 9 1 If we let x equals the number of moles of acetic acid that dissociates and assume it is very small compared to the initial concentration of acetic acid (0.5M) then the above equation can be simplified to; 1 . 8 x 1 0 -s 0.50 x 3 .0 X 1 0-3 pH - log [W] pH 2.52 Unlike acetic acid, hydrochloric acid (HCI) is a strong acid that becomes fully ionised in solution (eg. HCI � H+ + el} The full dissociation of this solvent results in the acidity of dilute Hel being far greater than that of dilute acetic acid solutions at comparable concentrations. The pH of a solution of O . lM Hel for example has a pH of 1 .0 . Dilute HCI has been extensively used as an extracting agent to investigate the plant­ availability of heavy metals in soils and uptake by plants (Nelson et al., 1 959; Lagerwerff & Specht, 1 970; Kinehiro & Sherman, 1 967; Lagerwerff, 1 971 ; Takij ima & Kasumi, 1 973 ; John, 1 974; Dundas & Pawluk, 1 977; Norvell, 1 984; Haynes & Swift, 1 985a; 1 985b). Neuman et al., ( 1 987) state that for Cu, more extraction and calibration work has been conducted with He1 than with any other acid. Lagerwerff ( 1 97 1 ) who used a solution of 1M HCI to investigate the uptake of cadmium, lead and zinc from soils was prompted to use this extractant for two reasons :­ ( 1 ) as a fully dissociated acid lacking oxidising qualities it provided H+ for exchange and limited dissolution in a non-destructive manner, (2) 1 M HCI is a strong comp1exing agent for certain transition-group metals such as Cd and Pb. Lagerwerff ( 1 97 1 ) suggested that the extractable ion concentrations determined by 1M Hel provided useful reference levels to which the amounts of ions absorbed by plants could be related. The acid extractant, however, was found to demonstrate little relationship between non­ polluted soils and plant uptake. The failure of this extractant to adequately predict metal uptake is presumed to reflect that at a 1M concentration HCI is likely to have solubilised 92 some non-labile forms of metal such as that occluded in oxides and secondary clay minerals (Ure, 1 995). In most investigations where HCI is used to assess the bioavailability of metals, it is generally used in weaker concentrations. Extractant solution concentrations of O . lM HCI have long been commonly used to assess the bioavailability of various metal cations in soil media (Viets & Boawn, 1 965; Kanehiro & Sherman, 1 967; Wear & Evans, 1 968 ; Norvell, 1 984; Haynes & Swift, 1 985a, 1 985b ) . Norvell ( 1 984) compared the ability of O . lM HCI to extract AI, Fe, Mn, Zn, Cu, Cd and Ni from diverse soil materials with 5 chelating agents. He observed that O. I M HCI was the most severe extractant, particularly for AI, Fe and Mn which are often present in part as acid soluble oxides. For this reason it is generally accepted that dilute HCI is a relatively harsh solvent that liberates most forms of labile metals that are available for plant uptake. Based on previous research (Norvell, 1 984; Haynes and Swift, 1 985a & 1 985b) and the fact that the Tui mine tailings contain negligible organic matter, it was predicted that O . lM HCI would not only provide useful information relating to the availability of metal to plants but would also satisfactorily estimate non-sulphidic metal concentrations in the tailings. It was postulated that unlike dilute acetic acid solutions (Pulford & Duncan, 1 978), O. l M HCI would be sufficiently harsh to render most forms of Cu and Fe c-f sulphate in the tailings soluble without significantly affecting the stability(sulphide minerals in the tailings. 3.4.2.3 Total Metal Extractants Techniques used to measure the total heavy metal contents of soils may be broadly divided into two categories :- ( 1 ) single-element methods such as flame atomic absorption spectrometry (F AAS) or (2) multi-element methods such as inductively coupled plasma-atomic emission spectrometry (ICP-AES) or X-ray fluorescence spectrometry (XRFS) CUre, 1 995). Some multi-element analyses were conducted on the Tui tailings using ICP-AES, however, most elemental analyses were conducted using F AAS as these methods were simple and accurate. Although dilution arising from sample preparation limits the ability of atomic absorption spectrometry to detect some ------------------------- 93 trace elements in non-contaminated soils, the tailings were well suited to this technique due to their metal-rich nature. A variety of extractants are used to estimate the total levels of heavy metals in contaminated soils and mining wastes. These include aqua regia8 (Van Loon & Lichwa, 1 973 ; Haq et ai., 1 980; Smith, 1 994), concentrated nitric acid (Smith & Bradshaw, 1 979; Clevenger, 1 990; Whitely & Williams, 1 993), multi-acid extractants (Ure, 1 995) and hydrofluoric acid based extractants (Ward et ai., 1 977; Safaya et ai. , 1 987, Neuman et ai. , 1 987). Because most soils contain substantial amounts of silicates, total dissolution generally reqUIres the use of an HF -based extractant which is capable of attacking silicate minerals (Ure, 1 995). Concentrated acids such as nitric (HN03) and/or perchloric acid (HCl04) are often used in combination with HF as a pre-treatment to oxidise organic matter and sulphide minerals (Lim & Jackson, 1 982; Safaya et ai. , 1 987; Ure, 1 995; Kucha et ai., 1 996). Where the vast proportion of metals are not bound up in the silicate lattice of the material, drastic attack with HF acid may not be necessary and pseudo-total analyses can be carried out using strong acids such as HCl, HN03, HCl04 either singularly or in combination (Ure, 1 995). Although these acids do not dissolve silica or silicates completely, they are sufficiently vigorous to dissolve non silicate-bound heavy metals, in the form of sulphide minerals for example (Smith & Bradshaw, 1 979). Sulphuric acid, in contrast, is poorly suited to F AAS analysis because of its complex interference effects. Losses of Pb as insoluble PbS04 may also occur and other elements may be lost by occlusion in precipitated CaS04 in calcium-rich soils (Ure, 1 995). 8 A mixture of concentrated nitric and hydrochloric acids ( 1 :3 v/v). ------- - --- 94 Aqua regia is probably the most widely used extractant adopted for pseudo-total soil analysis CUre, 1 995). The extractant generally removes between 70 and 90% of heavy metals in soils. Research outlined by Ure ( 1 995), however, indicates that strong acids such as concentrated HCI, HN03 are generally equally effective as aqua regia at extracting similar levels of Cd, Cu, Mn, Ni, Pb and Zn from metal contaminated soils. Smith & Bradshaw ( 1 979) found that the use of 25% (v/v) HN03 in a quick extraction procedure extracted about 90% of the metal from sulphidic mining wastes. 3.5 Acid Mine Drainage Treatment Clearly the most effective way of preventing Acid Mine Drainage (AMD) is to ensure acid generating materials are completely isolated from oxidising conditions and iron­ oxidising bacteria. Unfortunately, this approach is not alway possible and has certainly not always been used in past mining operations. Thus drainage water derived from mining sites often requires treatment prior to discharge to reduce its impact on the aquatic environment. The long term cost of treating AMD can be extremely expensive (Skousen et al., 1 990; Hellier et al., 1 994; Skousen & Larew, 1 994) and is a cost too often burdened on the ratepayers of historic mining communities. Fortunately a variety of effective technologies has been developed to treat AMD. These include neutralisation, reverse osmosis, low pressure ultra-filtration, bacterial facilitated sulphate reduction, evaporation, co-disposal with domestic wastes, anoxic limestone drains (ALDs), anaerobic (organic substrate) wetlands and aerobic wetlands (Bell et al., 1 982; Hedin et al., 1 994). Research into the use of ion exchange resins and natural and synthetic zeolites in the treatment of AMD also shows promising results and signal that these materials are likely to be increasingly used in the removal of metals from contaminated mine drainages (Schultze et al., 1 994; Riveros, 1 995). 3.5.1 Active Treatment Neutralisation is the most widely used method of active AMD treatment. In principle it is a simple technique that involves the addition of a base to the AMD to reduce acidity, increase alkalinity and precipitate Fe, Al and various heavy metals (Bell et al. , 1 982). In practice, however, the great variability in the composition of AMD and the different effluent quality requirements often precludes the application of simple neutralisation 95 techniques (Bell et al. , 1 982; Skousen et al. , 1 990). In the past, attempts to improve water quality were often rudimentary and involved diverting AMD firstly into settling ponds and then through porous limestone barriers. In the United Kingdom this method was found to be "notoriously unsuccessful" as the limestone beds were often rapidly clogged with reaction products such as Al(OH)3, FeOOH, Fe(OH)3, rendering them ineffective (Palmer, 1 99 1 ) . Most modern neutralisation techniques typically involve a several step process in which a variety of reagents including, alkali reagents flocculants and oxidants may be added to both aerated and standing 'water' (Skousen et al. , 1 990; Murdock et al. , 1 995). Many different base additives have been used in the treatment of AMD but the most commonly used alkalies are hydrated lime (Ca(OH)2), limestone (CaC03), sodium carbonate (Na2C03) and sodium hydroxide (NaOH) (Bell et al., 1 982; Skousen et al. , 1 990). Each chemical has its advantages and disadvantages and the selection of a suitable chemical or combination of chemicals depends on the geochemistry of the AMD being treated. Where Fe2+ is present in high concentrations for example, oxidising agents such as hydrogen peroxide, potassium permanganate and calcium hypochlorite may be used to speed the oxidation of Fe2+ to the less soluble Fe3+ form (Skousen et al. , 1 990). Whilst neutralisation methods are effective at ameliorating water quality, they can substantially increase the hardness of effluents and result in the production of large quantities of potentially toxic sludge which is characteristically difficult to 'handle' and dewater. Chemical costs are ongoing and can be particularly high if the AMD contains high concentrations of Fe2+ and Mn2+ (Evangelou, 1 984; Skousen et al. , 1 990). 3.5.2 Passive Treatment The treatment of contaminated mme drainage usmg paSSIve treatment systems IS currently the focus of considerable research and shows great promise for long-term, low­ cost treatment of contaminated water from degraded mine sites (Brodie & Hedin, 1 994; Hedin & Watzlaf, 1 994; Hedin et al., 1 994; Kepler & McCleary, 1 994; Skousen & Ziernkiewicz, 1995). Passive treatment technology has been chiefly developed by staff at the US Bureau of Mines and to date has been used primarily in the treatment of AMD 96 originating from colliery spoil in the US coal fields. The technology does, however, show promise for use in the treatment of AMD from metalliferous mines (Eger et al. , 1 994) There are currently three principal types of passive treatment systems; aerobic and anaerobic wetlands and anoxic limestone drains. These systems exploit natural chemical and biological processes that ameliorate water contamination. The primary advantages of these systems are that in theory they generally require fewer chemical reagents and less operational attention and thus provide a more cost-effective, long term means of treating metal contaminated mine drainage (Hedin et al., 1 994; Hellier et al., 1 994) . 3.5.2.1 Anoxic Limestone Drains The technology of anoxic limestone drains (ALDs) is still very much in its infancy. Turner & McCoy ( 1 990) were the first to describe the use of buried limestone beds to treat AMD. They function by promoting the contact of AMD with limestone under anoxic conditions and are designed to increase the water' s bicarbonate alkalinity and pH to circumneutral levels (6 to 7). ALDs currently being used to treat AMD in the mining industry are typically about 60 m long and 1 m2 in cross-section. They often contain about 70-80 tonnes of high grade limestone in the form of 5 to 1 0 cm diameter cobbles (Brodie & Hedin, 1 994) . The limestone cobbles are usually contained in geofabric or high density plastic lining that is buried within compacted clay soils or substrates to prevent the ingress of oxygen. ALDs are typically designed to retain water for about 1 5 hours, a time usually sufficient to allow the water to attain an alkalinity level of approximately 300 mglL. Current design guidance precludes the use of successive ALDs to generate alkalinity much in excess of this value because of concerns with dissolved oxygen (Kepler & McCleary, 1 994). 97 After exiting the ALD the water is usually feel into an aerobic pond which promotes the precipitation of metals via hydrolysis as ferrous iron is oxidised. The water' s induced bicarbonate alkalinity is designed to neutralise the acid produced by the hydrolysis of Fe3+ (Hedin & Watzlaf, 1 994; Hedin et al., 1 994) . One of the main limitations of ALDs is that they are poorly suited to mine drainage that contains appreciable concentrations of dissolved Fe3+, AI, or oxygen (>2 mgIL). The presence of these constituents in water entering an ALD is likely to result in the formation of FeOOH, Fe(OH)3 and/or Al(OH)3 which tend to rapidly block the drains (Hedin et al., 1 994; Robb & Robinson, 1 995). Failure to identify high concentrations of dissolved Al levels in input waters has seen ALDs constructed in Pennsylvania fail within the first 4 months of use due to the formation of Al(OH)3 (Brodie & Hedin, 1 994). Similar problems may arise with the formation of CaS04 if sulphate levels in the AMD exceed about 4000 mg/L. The maintenance of anoxic conditions in the drains is crucial in that it inhibits the oxidation of Fe2+ and therefore limits the likelihood of ferric hydroxide formation (Hedin et al., 1 994). Since the function of ALDs is based on the dissolution and thus removal of limestone they have a finite working life. Whilst many ALDs are currently constructed with sufficient limestone to theoretically function for 20 to 30 years, plugging by metal hydroxides is likely to result in hydrologic failure of the ALD well before the limestone is exhausted (Hedin et al., 1 994). 3.5.2.2 Aerobic & Anaerobic Wetlands As alluded to above, ALDs are often used in conjunction with aerobic wetlands which are also useful for facilitating the precipitation of metals (particularly Fe and AI) from circurnneutral mine drainage. Aerobic wetlands are designed to encourage the oxidation of Fe2+ to Fe3+ which is highly insoluble at pH levels above about 3 (Hedin et al., 1 994). They are consequently typically shallow « 1 m) in construction and are designed to aerate the water and dissolved constituents. 98 In contrast to aerobic treatment ponds, anaerobic wetlands are constructed to promote the activity of anaerobic bacteria which precipitate metal sulphides and generate bicarbonate alkalinity (Eger et ai. , 1 994; Hedin et ai. , 1 994; Hellier et ai. , 1 994) The wetlands are typically lined with up to 0.5 m of organic-rich material such as spent mushroom compost, manure, composted sewage sludge and sawdust. Organic substrates are often stabilised with vegetation such as cattails (Typha latifolia) which also contribute organic matter and enhance anaerobic conditions (Hedin et al. , 1 994). The organIc material in anaerobic wetlands is designed to accommodate sulphate reducing bacteria such as Desulfovibrio desulfuricans which can reduce sulphate in mine water to produce hydrogen sulphide (H2S) and bicarbonate gas (HC03-) as per the following equation: D.desulfuricans The resulting H2S reacts with heavy metals in mine drainage causing them to precipitate as sulphides as exemplified by the following equation: Zn2+ + H2S � ZnS./, + H+ Although acidity i s produced during the precipitation of sulphides the alkalinity produced by the reduction of sulphate to H2S buffers against pH change to help maintain circumneutral to alkaline conditions (Robb & Robinson, 1 995). The pH buffering of this system is important because sulphate reducing bacteria will not flourish below pH 5.0. Limestone is usually added to organic substrates to introduce an additional source of bicarbonate and to further ensure that circumneutral pH levels are maintained in the wetland (Hedin et al., 1 994) Where the pH of the water entering the wetland system is >4, anaerobic wetlands are often preceded by a single aerobic pond to precipitate Fe. This encourages the subsequent precipitation of Mn and other heavy metals in the anaerobic wetland. Manganese can be difficult and/or costly to remove in both passive and active treatment systems because unlike Fe2+ no natural mechanism exists to rapidly oxidise Mn2+ under acidic conditions. In anaerobic wetlands Mn is usually precipitated as Mn02 but this is 99 unlikely to occur unless other metals are present in minor amounts and pH levels are maintained above 6.0 (Robb & Robinson, 1 995). Anaerobic wetlands provide a viable means of treating drainage waters containing dissolved oxygen, Fe, Al and/or high acidity concentrations (>300 mg/L) which can not be treated using ALDs (Hedin et ai., 1 994) . Their biggest limitation, however, is that they require a relatively large surface area of land for operation (at least six times the area required by aerobic wetlands constructed to treat alkaline waters) (Hedin et ai., 1 994). The surface area requirements of anaerobic wetland systems often exceed 1 ha and are governed by the volume and metal loading of the mine drainage. Much of the current research associated with passive treatment systems is focusing on the development of 'hybrid ' ALD-compost wetlands or SAPs (Successive Alkalinity Producing Systems) (Kepler & McCleary, 1 994) . The theory behind these systems is that organic substrates can be used to strip Fe3+ and dissolved oxygen from the contaminated water to facilitate the flow of water into underlying anoxic limestone drains (Hedin et ai., 1 994; Kepler & McCleary, 1 994). If these systems are successfully developed it may be possible in the future to cost-efficiently treat highly acid mine drainages by cycling them between anoxic alkalinity-generating environments and aerobic metal-removing environments (Hedin et al. , 1 994). SAPs also have the added benefit of requiring substantially less surface area than anaerobic wetlands and are capable of generating much higher levels of alkalinity (Kepler & McCleary, 1 994). Useful reVIews of the use of wetland areas and paSSIve treatments systems in the treatment of AMD are provided by Sengupta ( 1 993) and Hedin et al. ( 1 994). 3.6 Preventing Acid Mine Drainage From the preceding literature review it is evident that iron-oxidising bacteria, oxygen and water all play a vital role in the oxidation of sulphide minerals and ultimately in the production of AMD. Whilst there is a wide variety of techniques employed to prevent mine wastes from generating AMD, the methods are similar in that most try to eliminate either bacteria, oxygen and/or water from the oxidation system. 1 00 Methods used to control or limit the ingress of oxygen into sulphide-bearing materials include; deep burial, subaqueous deposition, organic matter application, vegetative covers, compacted clay and synthetic barriers. Barrier methods, pH modifiers (including alkaline recharge zone techniques) and bactericides are commonly used to inhibit the activity of iron-oxidising bacteria. The ingress of water into sulphidic wastes is often minimised via the use of diversion drainage systems, impermeable clay and synthetic barriers and/or vegetative covers. The re-mining of sUlphide deposits and/or the selective removal and treatment of acid generating sulphide minerals are also methods that are being researched and used at some sites (Raicevic, 1 979; Caruccio et ai., 1 990; Humber & Eng, 1 995). The sulphide content of tailings may be reduced by means of bulk sulphide flotation prior to placement (Sengupta, 1 993). Considerable research is also being focused on the use of hyperaccumulating plants for phytoremediation of metal contaminated soils and mine wastes (Brooks, 1 987; Baker et ai., 1 988, 1 995; Chaney et ai., 1 995 ; Nicks & Chambers, 1 995; Reeves et ai., 1 995). The primary factor currently limiting the application of phytoremediation is the need to find rapid growing, high biomass crops that can hyperaccumulate problematic metals such as Cu, Pb and Zn. Research currently being undertaken at Massey University is investigating the combined use of chelating agents and non-hyperaccumulating plant species to phytoremediate metal contaminated soils (Brooks, pers. comm.) . Two novel approaches to prevent sulphide oxidation are micro-encapsulation9 of sulphides with FeP04 (Backes et ai. , 1 987; Pulford, 1 99 1 ; Huang & Evangelou, 1 992; Evangelou, 1 994; Georgopoulou et ai., 1 995) and cathodic protection of sulphide­ bearing deposits (Shelp et ai. , 1 995). Although both technologies are in the developmental stage, laboratory and small scale fields studies indicate that they show potential for inhibiting sulphide oxidation and arresting the formation of AMD. 9 Micro-encapsulation techniques exploit the ability of some compounds such as FeP04 to form protective films on the surface of sulphide grains. Their formation effectively inhibits sulphide oxidation by preventing the attachment of iron-oxidising bacteria and by reducing oxygen contact. 1 0 1 3.6. 1 Subaqueous & Subterranean Deposition Clearly one of the most successful and cost effective means of preventing the oxidation of sulphide-bearing wastes is to selectively bury such materials and preferably encapsulate them in low permeability material. This method is commonly used to isolate reactive (acid generating) material and often simply involves identification and selective handling of reactive material during mining operations (Geidel & Caruccio, 1 984). The handling difficulties involved with tailings may preclude their disposal by deep burial although redeposition into underground workings and/or open pits has been practiced at some sites (Harries & Ritchie, 1 988). The blending of acid-generating and acid-consuming rock units is also used at some sites as an alternative to adding neutral ising agents to the acid-generating waste during operations and waste placement (Sengupta, 1 993 ; Skousen & Larew, 1 994). Subaqueous tailings deposition is a method used to minimise oxygen diffusion into the acid generating tailings. It is being increasingly used in North America and involves the disposal of tailings into existing bodies of water (Pedersen et ai. , 1 994) or alternatively flooding subaerial tailings deposits upon abandonment (Dave & Vivyurka, 1 994; St­ Arnaud, 1 994). Marine disposal of pyritic wastes has been used at mining sites in Norway (Sengupta, 1 993) and is likely to be used at Lihir gold mine in Papua New Guinea (PNG) once ore processing operations commence (Harris, 1 997). Although river systems are not commonly used as depositories for pyritic materials in most current mining operations, river disposal is used in some Third World countries as a cheap alternative to terrestrial dumping. Mining operations at the Bougainville copper mine in PNG, for example, involved the disposal of an estimated 5 1 0 million tons of tailings into the Kawerong/Jaba river-system between 1 972 and 1 985 . The extent of dumping was on such a large scale that a 7 km2 marine delta formed 30 km from the mine site at the mouth of the river-system (Archer et ai. , 1 988). Whilst river disposal of tailings continues to be used at the Porgera (gold) and Ok Tedi (copper) mines in PNG, little is known about its efficacy at controlling sulphide oxidation (Harris, 1 997). Water, with its low oxygen diffusion coefficient (2 x 1 0-9 m2/s) and inability to dissolve large quantities of oxygen (8.6 g/m3 @ 25 °C) is, however, considered to be one of the 1 02 most effective oxygen-limiting covers for sulphidic wastes (Dave & Vivyurka, 1 994). Even partial saturation of tailings is likely to substantially reduce sulphide oxidation as the rate of oxygen diffusion through water is approximately 1 0,000 times less than through air (James & Eng, c. 1 992). The rate of oxygen transport through water is thus considered to be sufficiently low to be of "no concern in terms of acid generation" (Sengupta, 1 993). This type of disposal is considered to be suitable at locations where the topography "lends itself to stable containment against hydrostatic pressure" (Ritcey, 1 989) and obviously where evapotranspiration is less than rainfall and/or inflow of water into the impoundment. The advantages associated with subaqueous disposal include :- (i) limits oxygen and hence controls acid generation, (ii) the water body acts a sedimentation basin, (iii) outlet drainage flows can be controlled, (iv) eliminates surface erosion by wind and water action, (v) temperature reduction can occur and inhibit bacterial activity and (vi) facilitates bacterial reduction of nitrates and sulphates (Garga et al., 1 983; Dave & Vivyurka, 1 994). The disadvantages of water covers include the need to ensure the water is adequately contained and protected against flood and the potential for infiltrating water to destabilise materials (via increased interstitial pore pressures) and/or to act as a conduit for contaminants (Garga et al., 1983) . 3.6.2 Bactericides Singer & Stumm' s ( 1 970) suggestion that acid mine drainage may best be controlled by inhibiting the bio-catalytic oxidation of Fe2+ to Fe3+ is perhaps the reason for much of the research into the development of bactericides. Studies investigating the use of bactericides in land reclamation indicate that they are useful agents for reducing sulphide oxidation and acid mine drainage (Kleinmann et al. , 1 980 : Watzlaf, 1 986, 1 987, 1 988 ; Parisi et al., 1 994; Splittorf & Rastogi, 1 995). Many types of bactericides are used in the mining industry to inhibit sulphide oxidation including, food preservatives (benzoate and sorbate compounds), phosphate compounds and anionic surfactants such as sodium lauryl SUlphate and dodecylbenzene sulphonate (Sengupta, 1 993 ; Parisi et a!., 1 994) . 1 03 The primary purpose of most bactericides is to create conditions that inhibit the activity of the bacterium Thiobacillus ferrooxidans which is responsible for the catalytic oxidation of Fe2+ to Fe3+. As discussed in section 3 .3 , the formation of Fe3+ is considered to be the rate limiting step involved in the oxidation of pyrite. Many bactericides inhibit the biocatalysis of pyrite by inducing physico-chemical conditions that prevent micro-organisms from attaching to the mineral surfaces and/or by detrimentally altering the permeability of the cells' cytoplasmic membrane (Kleinmann et al., 1 980; Ritcey, 1 989; Pulford, 1 99 1 ) . Since bactericides do not usually effect the abiotic oxidation of sulphide minerals, they do not eliminate AMD and are thus generally used in conjunction with other AMD control techniques (Sengupta, 1 993; Splittorf & Rastogi, 1 995). One common drawback with most bactericides is that they degrade rapidly with time and are potentially depleted by runoff and leaching. To overcome this problem bactericidal compounds are often impregnated into rubber pellets or encapsulated in plastic pellets formulated to slowly degrade with time (Kleinmann et al. , 1980; Sengupta, 1 993 ; Splittorf & Rastogi, 1 995). The time-release bactericide ProMac (developed by BF Goodrich) is one most widely used commercial bactericides in the mining industry (Sengupta, 1 993). Overseas research outlined by Sengupta ( 1993) indicates that Pro Mac can improve water quality by significantly reducing metal concentrations, acidity and sulphate levels and encourage the establishment of plants and heterotrophic bacteria populations. The high cost of applying bactericides (often in the order of several thousand dollars per hectare) and the recognised need to use these substances in conjunction with other forms of sulphide oxidation control techniques are considered to be two of the primary drawbacks associated with the use of bactericides (Sengupta, 1 993). Additionally, research outlined by Watzlaf ( 1 988) indicates that bactericides may be ineffective when applied to partially oxidised materials. This finding indicates that bactericides may be ineffective at significantly inhibiting the formation of AMD at the Tui mine site. 1 04 3.6.3 pH Modification The surface application and/or incorporation of acid neutralising compounds is the most commonly used method for treating pyritic mining wastes!o. In principal base addition is designed to inhibit sulphide oxidation by precipitating Fe3+ and by creating an environment that is not conducive to the growth of iron-oxidising bacteria which oxidise Fe2+ to Fe3+. As discussed in section 3 .3 , the rate of sulphide oxidation (and hence AMD formation) is believed to be dramatically reduced in the absence of soluble Fe3+. Research outlined by Ziemkiewicz & Skousen ( 1 995), however, indicates that the addition of alkaline materials may not significantly inhibit sulphide oxidation. Instead they suggest that sulphide oxidation may continue largely unhindered and that AMD is mitigated by the precipitation of metals (primarily as hydroxides) under alkaline conditions. The most commonly used base additives are calcite (CaC03), burned lime (CaO), hydrated lime Ca(OH)2) and sodium hydroxide (NaOH) (Sengupta, 1 993). These and other bases may be added in 'pure' chemical form or as constituents in a variety of industrials wastes that are increasingly being used to amend acid-generating mine wastes. Trials investigating the use of alkaline waste products such as lime kiln dust (Rich & Hutchison, 1 994; Evans & Rose, 1 995 ; Ziemkiewicz & Skousen, 1 995), coal combustion bypro ducts (Dick et al., 1 994; Ziernkiewicz & Skousen, 1 995) and Leblanc waste! ! (Gemmell, 1 98 1 ) indicate they are useful pH amendments and may be an extremely cost effective means of controlling AMD where a ready source is available. An added benefit of using wastes rich in CaO (eg. lime kiln dust) is that they increase the shear strength of saturated or semi-saturated tailings within a few weeks of application (Ritcey, 1 989) . 10 See section 3 .7.3 for discussion pertaining to the use of l imestone in the reclamation of pyritic mine wastes. I I The Leblanc process (now obsolete) was used to produce Na2C03 for glass manufacture. The process produced highly alkaline (pH 1 2.7) waste containing large quantities of calcium hydroxide and calcium carbonate (Gemmell, 1 98 1 ). 105 The success of base additives at controlling AMD depends primarily on:- (i) the path of water through the system, (ii) the nature (eg. duration) of contact of acid generating rock and pore waters with the neutral ising agent, (iii) the proportion of excess alkalinity and (iv) the type and purity of the neutral ising agent (Gemmell, 1 98 1 ; Costigan et al., 1 984; Sengupta, 1 993). AMD is usually best controlled when highly reactive alkaline compounds are thoroughly mixed with the acid source to facilitate ' continual' dissolution of the additive in response to acid generation (Sengupta, 1 993). If AMD control is the primary objective of pH amendment, excess neutralising agents should be incorporated to compensate for that lost from the system due to consumption by infiltrating water and by that effectively ' lost' due to armouring by oxidation products (Sengupta, 1 993). Two potential disadvantages associated with the blanket application of neutralising agents in terms of their ability to inhibit the oxidation of sulphide minerals are firstly, that the ameliorant (neutralising agent) is not necessarily concentrated in major water recharge zones and, secondly, that the major flow path of water will not necessarily intercept a large potion of the neutral ising agent unless the amended surface is highly permeable. The use of alkaline recharge zones are designed to overcome these problems by intercepting infiltrating water and converting it into highly alkaline solutions which can potentially neutralise acidity and displace iron-oxidising bacteria (Carnccio et al., 1 984; Carnccio et al., 1 988 ; Nawrot et al., 1 994). Trials which have effectively altered the geochemistry of upslope groundwater recharge zones with alkaline materials including hydrated lime (Ca(OH)2) and lime (CaC03) have shown "extremely encouraging results" by reducing seepage acidity by 23% and acidity in recharge zone groundwater by up to 90% (Nawrot, 1 994). 3.6.4 Dry Cover Barriers The capping of sulphidic tailings and wastes with geological or synthetic barriers aims at preventing AMD by firstly restricting the entry of oxygen and/or secondly by preventing the migration of AMD by controlling water (conduit) entry. S ince the Thiobacilli bacteria involved in sulphide oxidation are micro-aerophilic and can survive at reduced oxygen tensions, partial elimination of oxygen is considered to poorly control AMD (Singer & Stumm, 1 970). Thus, barriers are believed to be more effective at 1 06 controlling AMD by reducing in the infiltration of water and ultimately preventing the migration of sulphide oxidation products to the surrounding environment. Barriers vary considerably in their structure and range from simple geological caps constructed of compacted ' soil' to synthetic membrane liners and high technology composite covers. 3.6.4. 1 Synthetic Barriers Synthetic barrierslliners are fabricated from a variety of materials including polyvinyl chloride (PVC), high density polyethylene (HDPE), butyl rubber, asphalt emulsions and cementitious materials (Ritcey, 1 989; Sengupta, 1 993 ; Jones & Wong, 1 994). They have the potential to provide covers with an extremely low conductivity of oxygen and water. The high cost of installation and their susceptibility to mechanical rupture, however, are major l imitations governing their use. To prolong their effective life synthetic liners are often used in conjunction with geological covers designed to host plants and protect the underlying synthetic liners from ultraviolet light and/or the direct effects of trafficking. Synthetic mineral polymers (geopolymers) are compounds which when added to mineralised tailings achieve strength through molecular bonding (Davidovits & Davidovics, 1 987). Under alkaline conditions these alumino silicate compounds are converted to very reactive silico-aluminates which polycondense to form three dimensional stone-like materials (Ritcey, 1 989). Whilst this technology is still in its developmental stage, geopolymers may in the future be used to create low permeability, high strength covers for the long term chemical and physical stabilisation of tailings. Ahmed ( 1 994, 1 995) reports on similar studies which are developing techniques to enhance the formation of hardpans which naturally occur in some pyrrhotite-rich tailings. These synthetic hardpans are seen as a potential means of inhibiting water and oxygen ingress into sulphide bearing materials. 3.6.4.2 Geological Barriers Geological barriers are usually constructed from non-sulphidic geological materials such as overburden, soils and clay substrates (Pulford, 1 99 1 ; Sengupta, 1 993). Single layer barriers constructed of compacted clay or silt have been frequently used to cap and 1 07 isolate sulphide bearing materials from infiltrating water and oxygen (Pulford, 1 99 1 ; Sengupta, 1 993) . Problems associated with the use of such covers include that:- (i) they are generally only effective at limiting oxygen ingress if they are maintained in a saturated or near saturated state, (ii) they do not prevent capillary rise of contaminated water from underlying materials and (iii) they are prone to erosion (Sengupta, 1 993). Compacted geological barriers also tend to be poor growing media for plants and where vegetation does establish penetrating roots can rapidly breech the barrier and facilitate the ingress of air and water to the underlying waste material. To eliminate problems associated with single-layered geological barriers, composite covers have been developed to cap and isolate acid generating wastes (Harries & Ritchie, 1 988 ; Sengupta, 1 993; Bell et al., 1 994; Aachib et al., 1 994). Composite covers are typically constructed of 3 or 4 layers of geological materials of variable thickness which each have a specific function. Collectively these layers may be a metre or more thick and are often designed to simultaneously host vegetation, inhibit water and air infiltration, prevent capillary rise of contaminated water and to prevent desiccation cracking (Sengupta, 1 993). Wind and water erOSIOn of composite covers is usually prevented by establishing vegetation on the surface of the cover. Thus, a surface horizon with chemical and physical properties conducive to plant growth must be incorporated. Alternatively where rIM"' -' vegetation is not used, a coarse gravel layer is usually ron the surface of the cover to reduce the effects of erosion. Another main design feature of many composite covers is the incorporation of a capillary break layer that is designed to inhibit the capillary rise of AMD and to prevent the accumulation of metal compounds in the growing medium or at the surface of the cover. This is usually achieved via the use of a layer of coarse textured material such as a screened gravel with no fines. Whilst laboratory studies indicate that a 50 mm layer of gravel may be sufficient to prevent the upward movement of solutes, a 300 mm layer is considered to be more practical minimal depth (Johnson et al., 1 994). Capillary action is also used advantageously within composite covers to create moisture retention zones which potentially store water for vegetation, limit desiccation of underlying layers and reduce oxygen diffusion (Sengupta, 1 993) . Moisture retention 1 08 zones can be created by 'sandwiching' a fined grained material between two layers of well sorted coarse gravel . In this situation the middle layer will tend to maintain a comparatively high water content (Aachib et al. , 1 988). Infiltration of water and oxygen may be further reduced by the emplacement of a compacted fine-grained geological barrier and/or a synthetic liner immediately above the waste material. The lower the permeability of this layer the more effective the cover will be at reducing infiltration. Figure 3 .2 highlights the theoretical difference in water infiltration between bare tailings and tailings capped with a composite cover consisting of an infiltration barrier, a capillary break/lateral drain and vegetated soil . Mathematical modelling indicates that bare tailings will exhibit high runoff ( 1 9%) 12, modest evapotranspiration losses (63 .5%) and substantial net infiltration/seepage ( 1 7.5%). Capping the tailings with a vegetated composite cover will in theory reduce total seepage by about 62% to 6.6% (Bi Tech Publication, 1 989, cited in Sengupta, 1 993). Whilst composite covers are effective at reducing seepage it is important to note that unless they are used in combination with synthetic membranes, infiltration into underlying material will continue and thus AMD is unlikely to be totally arrested (Bi Tech Publication, 1 989, cited in Sengupta, 1 993) . The Rum Jungle mine site l 3 in Australia is considered to be one of the best documented examples of the use of composite cover technology (Sengupta, 1 993). Geochemical monitoring of the pyritic waste dumps at the site indicates that the composite covers have been effective at reducing the ingress of oxygen and particularly water into the dumps. The percentage of incident rainfall infiltrating the dumps was found to fall from about 50% to 60% prior to rehabilitation to just 3% after the emplacement of the covers (Harries & Ritchie, 1 988). The composite covers also appear to have reduced Cu, Mn and Zn loading of AMD emanating from the dumps by more than 60% (Sengupta, 1 993). 12 Bracketed values represent percentage of annual precipitation. 13 See section 3 .2. 1 . 1 for further discussion of the remedial work carried out at the Rum Jungle mine site. ------------ - 1 09 3.6.4.3 Organic Barriers The surface application of organic-rich materials onto acid generating mine wastes is another method that has shown promising results in reducing oxygen ingress (Pierce et al. , 1 994; Tremblay, 1 994) . Organic barriers are nonnally constituted of organic wastes such as composted municipal sewage sludge, manure or ligneous wastes including paper pulp and bark. In theory organic barriers reduce sulphide oxidation and AMD by:- (i) acting as a sink for atmospheric and dissolved oxygen which is consumed by heterotrophic microorganisms during organic matter decomposition, (ii) providing a physical barrier to oxygen if high moisture content is maintained in the cover, (iii) producing soluble organic compounds that leach into underlying material and inhibit the growth and metabolism of iron-oxidising bacteria and (iv) by encouraging the activity of sulphate reducing bacteria which precipitate sulphide minerals and lead to the removal of Fe3+ from solution (Pierce et al., 1 994). In tenns of cost one of the primary advantages of using organic wastes to reclaim mine sites is that the amendment is usually free and haulage and application costs may be offset by savings made with waste disposal. (a) Bare Tailings Precipitation 1 00% Evaporation 63.5% t Seepage 1 7 .5% (b) Capped Tailings Precipitation 1 00% Evaporation 70.9% � t Seepage 6.6% Runoff 1 9 % • 1 1 0 k = 6 X 1 0-5 [cmlS] Runoff 3 . 1% • k = 6 X 1 0-5 [cmlS] Figure 3.2 Infiltration rate, runoff and seepage for (a) bare tailings and (b) capped tailings. Material depths:- Soil layer = 1 .0 m, Coarse layer = 0.3 m, Barrier layer = 1 .0 m. (adapted from Bi Tech Publication, 1 989 cited in Sengupta, 1 993). 1 1 1 3.7 Revegetative Reclamation Since the goal of most land reclamation projects is to establish a permanently stable landscape that is aesthetically and environmentally compatible with surrounding undisturbed land, vegetation is often used to reclaim tailings and waste rock dumps (Sengupta, 1 993). Ideally vegetative reclamation should produce a self-perpetuating plant cover either directly, or by promoting the colonisation of plants which will form a self-regenerating community (Darmer, 1 973). It is widely recognised, however, that wastes from metalliferous mines, particularly acidic tailings, are very difficult to revegetate (Pulford, 1 99 1 ; Johnson et ai., 1 994) . There are several factors that may inhibit plants from naturally colonising tailings. These include:- 1 . Low pH and associated problems with H+ and metal toxicities. 2. Major plant nutrients, in particular nitrogen and phosphorus, are generally deficient in tailings. In situ cycling of nutrients is usually negligible due to the characteristically low organic matter content of tailings. 3 . The small uniform grainsize of tailings and lack of macroporosity results in poor soil aeration and drainage. 4. Wind blown tailings particles tend to sand blast, uproot and bury young plants. 5 . The thixotropic properties of tailings can make 'handling' and the use of heavy vehicles for amendment application difficult. 6. The high albedo of tailings may impose physiological stress on growing plants (Easton-Groves et ai., 1 982; Palmer, 1 99 1 ) To gain the full aesthetic and practical benefits of a vegetative cover, the cover should be continuous, dense and permanent. Research outlined by Sengupta ( 1 993), investigating the natural recolonisation of three abandoned lead and zinc tailings deposits in Wisconsin, USA, found that aggregation of vegetation was extremely pronounced at the sites and that numerous bare areas existed interspersed with well­ vegetated zones. This type of patchy vegetative cover is commonly observed on colliery spoil and metal contaminated sites where acidification and/or metal toxicity restricts the colonisation of plants or causes plant 'die-back' (Costigan et ai., 1 98 1 ; Winterhalder, 1 995b). 1 12 Some of the earliest research into the establishment of vegetative covers over fine­ grained mine wastes began in the South African Witwatersrand in the early 1 930s, where attempts were made to suppress dust from nearly 70,000 ha of gold tailings and sand dumps (Easton-Groves et al. , 1 982) . Although drought and acidity resulted in the early revegetation methods largely failing, over the last 30 years revegetation of gold mine residues has proved to be one of the more successful means of erosion control and pollution abatement in South Africa (Easton-Groves et al., 1 982) . The ameliorative approach is one frequently used to establish vegetation on mine wastes and refers to achieving 'optimum' plant growth conditions by improving the chemical and physical properties of the medium through the use of soil amendments which act to dilute and/or avoid toxicity (Johnson et al., 1 994) . The remainder of this chapter discusses the potential benefits and problems associated with vegetative covers and reviews the use of lime and organic waste amendments and the metal-tolerant plant species for reclamation purposes. 3.7. 1 Potential Beneficial Effects of Revegetation Besides being useful for stabilising mine wastes against wind and water eroSIOn, assisting in the immobilisation of potentially toxic cations and improving the aesthetic appearance of the site, vegetative covers may also serve to reduce the rate of biochemical sulphide oxidation (Ritcey, 1 989). The potential benefits associated with the establishment of vegetative covers on metalliferous mine wastes are discussed below. 3.7. 1 .1 Stabilises Against Erosion Land devoid of vegetation is much more vulnerable to erosion resulting from rainfall and wind (Verma & Thames, 1 978). Thus one of the primary objectives of reclaiming mine-wastes is to stabilise them against wind and water erosion. Vegetative stabilisation is often the preferred method for the reclamation of both active and abandoned mine sites due to its potentially more permanent nature, its aesthetic value and because it usually allows for a wider range of end uses (Norland et ai., 1 993). 1 1 3 In the United States, dust is considered to be a major environmental problem that affects land adjacent to inactive tailings ponds (Nielson & Peterson, 1 978). Wind erosion of fine-grained tailings « 1 00 microns) may also exacerbate AMD by continually exposing previously unweathered material and thereby accelerating the rate of sulphide oxidation (Johnson et al. , 1 977). Plant organs provide effective protection from the physical effects of wind and water erosion both above and below-ground (Norland et al., 1 993) . Below ground level, plants act to bind particles by firstly, physically holding the particles in place and secondly by encouraging the activity of soil organisms which promote soil aggregate formation. Physically, plant roots, root hairs and fungal hyphae bind particles together to form macro-aggregates. The growth and expansion of plant roots in the soil pores creates pressure on the proximal soil particles, physically holding them in place (McLaren & Cameron, 1 990). ' Soil' stability is promoted biochemically by vegetation through the effects of organic matter accumulation in the soil medium. Organic matter promotes the activity of some 200 genera of bacteria and 1 70 genera of fungi that exist in the soil microbial community (Land Resources Network Ltd. , 1 993). Organic matter breakdown by saprophytic bacteria results in the production of mucilage or 'glue-like' compounds in the plant root rhizosphere. These organic compounds effectively glue soil particles together to form larger aggregates. Polysaccharides exuded by plant roots bind particles together in much the same way (Molloy, 1 988). Another mechanism by which soil stability is promoted is through the flocculating characteristics of polysaccharides and other long chain polymers such as humic and fulvic acids. These organic polymers are derived from the degradation of organic matter and carry net negative charge. This charge, combined with their flexible structure, enables polymers to effectively link soil colloids and assist soil aggregate formation (McLaren & Cameron, 1990). 1 14 Above ground, vegetation aids in the prevention of water erosion by reducing the impact of raindrops and by reducing the volume and velocity of surface run-off. Rill erosion (one of the most common forms of erosion affecting mine tailings) is readily eliminated by the introduction of an appropriate vegetative cover. The severity of wind erosion can also be abated by vegetation which reduces wind velocities at the ' soil ' surface and also captures fugitive dust. In wind prone areas measures to revegetate the tailings often require steps being taken to prevent 'sand blasting' and/or burial of emerging seedlings. In South Africa this problem has been effectively dealt with by the use of dry reed windbreaks (Easton-Groves et al., 1982). 3.7.1 .2 Increases CEC and Specific Adsorption Humus derived from the breakdown of accumulated organic matter is responsible for about 20% to 70% of the CEC of soils (Land Resources _Network Ltd., 1 993). The ability of humic substances to act as cation exchangers arises from the presence of functional groups such as ionisable carboxyl and phenolic hydroxyl carried on the organic polymers (McLaren & Cameron, 1 990) . These predominantly negatively charged functional groups act as exchange sites not only for cationic plant nutrients such as Ca2+, K + and NH/, but also to some extent for potentially toxic non-essential metallic elements including AI3+, Cd2+ and Pb2+ (Land Resources Network Ltd. , 1 993 ; Ritcey, 1 989). Although required by plants in small amounts, micronutrients such as Cu2+, Mn2+, Cu2+ and Zn2+ become toxic to plants if high concentrations are present in soil solution. These micronutrients, along with the non-essential metallic cations are bound to soils by specific adsorption and by simple cation exchange. Specific adsorption, however, is considered to dominate soil adsorption of polyvalent micronutrients and heavy metals and refers to the formation of stable complexes with specific functional groups at the surface of inorganic and organic soil colloids (McLaren & Cameron, 1 990). Cation exchange and specific adsorption are thus extremely important mechanisms in that they serve to simultaneously reduce the bioavailability and mobility of the toxic metal cations and increase .a material ' s ability to retain plant nutrients which may otherwise be lost via leaching. 1 1 5 3.7.1 .3 Reduces Oxygen Ingress One of the primary variables that governs the rate at which oxidation occurs in sulphidic mine wastes is the availability of oxygen. In the case of fine grained wastes such as tailings, diffusion as opposed to convection is considered to be the primary mechanism of oxygen transport (James & Eng, c. 1 992) . The depth to which oxygen diffuses into a material is dependent chiefly on its texture and on the oxygen consuming characteristics of the material . Thus, a considerable reduction in the contaminant load of drainage water emanating from sulphide-bearing wastes can, in theory, be achieved by minimising the rate of oxygen diffusion and/or by ensuring biological oxygen demand (BOD) is maximised within surface horizons. The BOD of surface materials can be increased through the gradual accumulation of organic matter (via the establishment of a vegetative cover) or alternatively by the application of organic wastes (Pulford, 1 99 1 ). Research indicates that as vegetation becomes established, soil organisms, decaying organic matter and plant roots deplete oxygen in the plant root zone (Ritcey, 1 989; Pulford, 1 99 1 ) . As discussed in section 3 .6 .4 .3 , biological activity associated with the breakdown of organic matter thus effectively acts as a sink for atmospheric oxygen and dissolved oxygen in infiltrating water. The accumulation of organic matter in surface 'soil ' horizons is therefore generally encouraged as a means of restricting the amount of oxygen available for the oxidation of sulphide minerals within underlying strata. 3.7.1.4 Decreases Leaching As previously discussed, water plays an integral role in the oxidation of sulphide minerals and in the production of AMD. The ingress of water into tailings may be substantially reduced by vegetation which intercepts and returns rainfall to the atmosphere via evapotranspiration. Johnson et al., ( 1 994) suggest that in temperate climates, vegetation will intercept and return about 50% of the rainfall to the atmosphere. This is a very approximate figure only as the amount of incident rainfall returned to the atmosphere via evapotranspiration is highly dependent upon numerous site specific variables including the type of vegetation, climatic conditions and the nature of the rainfall. 1 1 6 3.7. 1 .5 Increases Plant Diversity Improved physicochemical 1 4 and biological properties afforded by organIc matter accumulation associated with the establishment of vegetation on mine wastes can accelerate the development of a viable nutrient cycle and lead to a more diverse self­ sustaining vegetative cover (Norland et al., 1 993; Winterhalder, 1 995a, 1 995b). Improving plant growth conditions will eventually enable less tolerant plant species to colonise once hostile growing media and in most cases will ultimately facilitate the growth of vegetation consistent with that hosted by surrounding land. Research conducted near Sudbury, Canada, investigating the reclamation of extensive areas of acidified metal-contaminated land, found that the addition of dolomite and grass/legume seeds at rates of 1 0 Mg/ha and 45 kg/ha respectively, promoted colonisation of the area by native trees and herbaceous plants (Winterhalder, 1 99 1 ) . It was found that over time the relative abundance of sown grasses on seeded areas decreased, while sown legumes and native plant species increased. 3.7.1 .6 Improves Site Aesthetics Perhaps the greatest attribute of vegetative covers is their ability to improve the aesthetic appearance of denuded mine sites. Even simple grass covers can significantly reduce the obtrusive impact of tailings and waste rock dumps, however, the planting of shrubs and trees may be more useful for concealing the topography of engineered structures (Scheltus, 1 990). Many reclamation schemes aim at establishing vegetation that is consistent with the surrounding landscape. This aids in the creation of a cover that merges with the adjacent vegetation and minimises any negative visual impact of the site. 1 4 Refer to sections 3 .7 .4 . 1 & 3 .7 .4.2 for specific discussion. 1 1 7 It should be noted, however, that revegetative reclamation does not always alm at ' camouflaging' mined land (Boffa, 1 990). In some instances reclamation is used as an opportunity to change the vegetative cover. Land revegetated with native bush, for example, can provide habitats for native flora and fauna and recreational areas for people living in deforested landscapes. Alternatively land mined within scrub may be reclaimed with improved pasture species to create highly productive pastures for grazing purposes. 3.7.2 Potential Negative Effects of Revegetation Although the establishment of a vegetative cover on mine wastes is generally considered to be beneficial at least in terms of its aesthetic value, negative effects may potentially arise. Two such effects that can potentially exacerbate environmental degradation are the reductive dissolution of weathered compounds and increased sulphide oxidation arising from improved aeration and water infiltration. 3.7.2.1 Reductive Dissolution As previously discussed, one of the primary benefits of establishing vegetation on mining wastes is that it introduces organic matter into the system. Some research suggests, however, that the build-up of humified organic matter in surface tailings either indirectly through the gradual decay of vegetation or directly via the application of sewage sludge or composted municipal waste, for example, may exacerbate acid mine drainage (Land Resources Network Ltd., 1 993 ; Ribet et al., 1 995). Treatments that induce reducing conditions may result in the release of potentially toxic metals (previously adsorbed or co-precipitated with ferric oxy-hydroxides) into surface and/or ground water (Ribet et al., 1 995). Research conducted at the Nickel Rim Mine located near Sudbury in Canada, found that more than 80 % of Fe, Ni, Cu and Co in the oxidised zone of tailings at the site were available for release by reductive dissolution (Ribet et al., 1 995). The mobility of metals released by reductive dissolution may be increased by specific adsorption of metallic cations by small, soluble organic polymers such as fulvic acid produced from decaying organic matter. 1 1 8 Increased metal loading of drainage water is not the only potential problem associated with reductive dissolution of metal compounds. Under oxidising conditions Fe2+ released into the water may oxidise and undergo hydrolysis to liberate acid. Thus, acidification of drainage waters can potentially occur. 3.7.2.2 Increases Oxygen & Water Ingress As previously discussed 15 vegetative covers can reduce the ingress of oxygen and water into materials via evapotranspiration and increased BOD respectively. The ability of planted surfaces to reduce the ingress of these substances may, however, be partially negated by improving ' soil' structure and macroporosity associated with accumulating organic matter and pedogenesis. Improved macroporosity and surface permeability may increase the ingress of oxygen and water into the surface horizons and potentially deeper into the profile to facilitate sulphide oxidation and the formation of AMD. The interaction between increased permeability and evapotranspirationIBOD is clearly antagonistic in terms of AMD generation and will be dependent on the interplay of parameters such as the physical nature of the material, vegetation type, climate and nature of incident rainfall. 3.7.3 Lime Application Neutralising agents such as limestone are the most commonly used amendments for aiding the establishment of vegetation upon sulphidic mine wastes (Pulford, 1 99 1 ; Dollhopf, 1 992). The rationale for the addition of limestone to ameliorate acidic conditions is as follows:- (i) the plant nutrient Ca is being added to the soil along with Mg if dolomitic limestone is used, (ii) mineral acidity is neutralised as well as acidity due to Ae+, (iii) P availability may be increased, (iv) Fe oxidising bacteria are displaced due to unsuitable pH conditions and (v) exchangeable AI, Fe and Mn are greatly reduced, (Jackson, 1 958 ; Smith & Sobek, 1 978; Johnson et ai., 1 994). Maclean & Dekker ( 1 976) found that for metalliferous tailings, providing sufficient limestone is applied to maintain the pH at circurnneutral levels, heavy metal toxicity is 15 See sections 3 .7 . 1 .3 & 3 .7 . 1 .4 1 1 9 not generally a problem. This finding is in accordance with the recognised relationship between pH and metal solubility in soils (Mortvedt et ai., 1 972; Alloway, 1 995). 3.7.3.1 Acid Neutralisation Acid is neutralised by limestone VIa one of two mechanisms. Under strong acid conditions limestone (CaC03) dissolves relatively rapidly to neutralise H+ in solution (equation 1 ), whereas under weak acid or even neutral conditions the presence of carbon dioxide (C02) favours the dissolution of calcium CaC03 by forming calcium bicarbonate (Ca(HC03)2) which in turn neutralises solution H+ (equations 2 & 3). ( 1 ) CaC03 + CO2 + H20 B Ca(HC03)2 (2) Ca(HC03)2 + 2 H+ B Ca2+ + 2 CO2 + 2 H20 (3) Although the liming of acid generating materials often enables plants to initially establish, plant die-back may occur if the liming agent is either consumed or incapable of reacting at a rate sufficiently rapid to neutralise acid released from sulphide oxidation. The second scenario may arise in situations where the grainsize of the limestone is too coarse (Gemmell, 1 98 1 ; Costigan et ai., 1 984), or alternatively when the limestone particles become armoured by weathering compounds (Costigan et ai., 1 984; Evangelou, 1 994). Plant die-back within a year or two of sowing, however, most commonly arises as a result of underestimating the liming requirement of pyritic wastes and applying insufficient limestone to maintain the pH at levels conducive to plant survival (Johnson et ai. , 1 977). In situations where very high liming rates (> 1 00 Mg/ha) are required to neutralise acid and to prevent reacidification of growing media (via sulphide oxidation), potential problems can arise that can negatively impact plant growth. These problems include enzyme inactivation, induced Cu, B, Mg, Mn and P deficiencies and accelerated N loss through leaching (Doubleday 1 972; MacLean & Dekker, 1 976; Gemmell, 1 977, Pulford, 1 99 1 ) . These problems are generally worst in materials that are poorly buffered and may in severe cases lead to total plant die-back. 120 Phosphorus is perhaps the most problematic of the macronutrients in tenns of its availability in mining wastes. Not only is P generally deficient in such materials but its availability is highly dependent on pH. At low pH « 4), P is largely imrnobilised by anion adsorption by colloidal Fe and Al materials, whereas at high pH (>7) the fonnation of precipitates of low solubility phosphates becomes an important mechanism governing its availability. In heavily limed materials (or in calcareous soils) the fonnation of highly insoluble tricalcium phosphate (Ca3(P04)2) often occurs which can fix P and render it largely unavailable for plant uptake. Research outlined by Mengel & Kirkby ( 1 987) indicates that phosphate uptake by some species is 1 0 times greater at pH 4 than at pH 8.7. This difference is attributable to the fixation of P in the highly alkaline conditions and the faster uptake of H2P04- under the acidic conditions which favour the fonnation of primary orthophosphate. Another widely recognised limitation of using limestone to neutralise acidity in growing media is that it is an ineffective means of ameliorating ' subsoil' acidity, due to the slow rate at which lime derived alkalinity leaches down the profile (Costigan et al., 1 98 1 ; Dollhopf, 1 992). Thus, amendment incorporation is considered to be vital, particularly in highly acidic growing media. Research indicates that whilst standard agricultural discs, sub-soilers and moldboard ploughs can incorporate lime efficiently to depths of about 25 cm, tractor mounted rototillers probably provide the most efficient means of incorporating lime into mining wastes (Dollhopf, 1 992). Whilst the shear strength of waste rock materials is usually adequate to support the weight of wheeled vehicles for lime spreading and incorporation, low ground pressure machines (ie. tracked vehicles) may be required on poorly consolidated tailings or saturated materials. 3.7.4 Organic Waste Application Soil organic matter has long been considered the "elixir of plant life" (Allison, 1 973) and is well recognised as an important factor governing the versatility and productivity of agricultural soils under pasture and more specifically land disturbed by mining activity (Halderson & Zenz, 1 978; Sopper, 1 992). Numerous studies conducted over the 1 2 1 past 60 years have shown that amending soils with organic waste materials increases organic matter content and improves soil structure and long term fertility (Joost et ai., 1 987). The effect of a variety of organic wastes including crop residues, farmyard manures, municipal composts and sewage sludges, lignite, peat, wood-wastes and papermill bypro ducts have been studied (Jozefaciuk et ai., 1 990; Pulford, 1 99 1 ). Specific examples pertaining to the application of such materials are outlined in an extensive review that summarises the use of organic materials as soil amendments in reclamation (Land Resources Network Ltd., 1 993). Sludge l 6 is one of the most commonly used organic amendments in land reclamation schemes around the world and literature pertaining to its use is readily found (Dinauer et aI. , 1 977; Gemmell, 1 977; Johnson et ai., 1 977; Halderson & Zenz, 1 978 ; Seaker & Sopper, 1 983, 1 984, 1 988a, 1 988b; Crites, 1 984; Atkinson et ai., 1 991 ; Forstner, 1 99 1 ; Moffat et ai., 1 991 ; Sopper, 1 992; Land Resources Network Ltd., 1993 ; Pichtel et ai., 1 994). Although most of this research is centred on the use of sewage sludge in colliery spoil reclamation, much of the information contained in the literature is relevant to the reclamation of other mining wastes including base-metal tailings. The disposal of sludge generated from sewage treatment plants is a serious problem facing numerous cities around the world (Sopper, 1 992). Within the last few decades the problem has been exacerbated by growing populations and environmental lobbying. This has forced governing bodies to cease the dumping of sludge into river systems, lakes and oceans. In the USA it is estimated that more than 7.7 million tons of dry municipal sludge are produced each year by 1 5,300 treatment plants scattered throughout the country (Sopper, 1 992). New Zealand produces about 52,000 tonnes of dry sewage per year (McLaren & Cameron, 1 996). Because the sheer volume of sludge produced by sewage treatment plants makes the disposal of sludge by other methods (eg. incineration, land-fills) costly, land based disposal of sludge onto agricultural and degraded land is encouraged as it is considered to be both cost-effective and potentially beneficial to soils (Forstner, 1 99 1 ) . 16 For the purposes of this review 'sludge' is used as a generic tenn to describe any aerobically or anaerobically treated solid waste derived from municipal sewage treatment plants. The material is assumed to contain negligible quantities of pathogenic organisms. 1 22 Sludge application to organically impoverished soils and mining wastes is considered to be an effective means of recycling organic matter and essential plant nutrients removed from soils as a result of cropping and grazing practices (Smith, 1 994). For these reasons some researchers suggest that sludge should be regarded as a resource rather than a waste product (Werner et al., 1 99 1 ; Sopper, 1 992; Lottermoser & Morteani, 1 993). The effects of sludge addition on the physical, chemical and biological characteristics of mine spoils are discussed below. 3.7.4.1 Beneficial Chemical Effects A variety of ' soil' chemical parameters are known to potentially benefit from the addition of sludge. These include plant nutrient levels, cation exchange capacity (CEC), electrical conductivity, pH buffering capacity and soil reaction (Dinauer et al., 1 977). (i) Nutrient Addition The ability of sludges to supply large quantities of organically bound nutrients to ' soil' media is one of the most widely recognised benefits associated with its application. It is estimated that in the vicinity of 90% of the total N and S and between 1 0% and 50% of the total P exist in organic forms in soil surface horizons (Land Resources Network Ltd., 1 993). Mineralisation facilitated by micro-organisms and soil fauna is responsible for transforming organically bound nutrients into compounds that are more readily available to plants (Land Resources Network Ltd. , 1 993). The enrichment of nutrients through sludge addition is particularly evident when used to amend mining wastes which commonly exhibit multiple nutrient deficiencies (Gemmell, 1 977; Johnson et al. , 1 994). Deficiencies of N and P in such wastes are often encountered and are potentially remedied by the addition of sludge which usually contains appreciable quantities of these macronutrients (Gilmour et al. , 1 977; Olsen & Barber, 1 977; Sopper, 1 992). In addition to N and P, sludges often contain variable quantities of Ca, K, Mg, S and a variety of other trace elements that may aid in the correction of nutrient deficiencies (Olsen & Barber, 1 977; Sopper, 1 992). There are numerous reviews and studies that specifically discuss the nutritional benefits of sewage sludge addition to both soils and mining wastes (Chaney & Giordano, 1 977; Gilmour et al., 1 977; Olsen & Barber, 1 977; 123 Sopper et ai., 1 982; Topper & Sabey, 1 986; Seaker & Sopper, 1 988aJ1 988b; 10zefaciuk et ai., 1 990; Atkinson et ai., 1 99 1 ; Moffat et ai., 1 99 1 ; Sopper, 1 992; Pichtel et ai. , 1 994). (ii) Increased CEC As discussed in section 3 .7. 1 .2, one of the key attributes of soil organic matter is its ability to exchange and electrostatically bind cations from the soil solution (Land Resources Network Ltd. , 1 993) . A material ' s CEC is primarily controlled by its colloidal organic matter and clay mineral content and is a very important parameter in that it governs the availability of nutrient cations and essential and non-essential 'trace' elements (Keeney & Wildung, 1 977; Sopper, 1 992). Since organic amendments, such as sludge, usually have high CEC their addition generally increases a substrates ability to retain cationic plant nutrients including NH/, Ca2+, K+, Mg2+, Mn2+, Fe2+, Cu2+ and Zn2+ (Land Resources Network Ltd . , 1 993). The humus component of organic matter is largely responsible for CEC due to its comparatively high surface area per unit mass and hence more chemically reactive functional groups. Since the charge exhibited by humus is largely generated by the presence of carboxylic, phenolic and hydroxyl functional groups, the CEC of such materials tends to increase with pH (Land Resources Network Ltd. , 1 993). In acid soils Ca2+, K+, Mg2+, Ae+ and Al[OHf+ generally dominate the exchange complex, whereas in calcareous soils Ca2+ and Mg2+ occupy most of the exchange sites (Keeney & Wildung, 1 977). The benefits of adding sludge in terms of CEC is therefore largely dependent on the pH of the amended substrate and the quantities of colloidal organic matter present in the sludge. Non-composted or 'fresh' sludge with a low humic colloid content is, therefore, less likely to ameliorate a substrate's CEC in the short term. Humification of the organic matter is, however, likely to improve CEC with time. (iii) Moderated pH The increase in CEC afforded by organic matter addition, not only improves the retention of nutrients and non-essential metal cations, but also serves to buffer pH change via the release and absorption of hydrogen (W) and aluminium (Ae+) ions from the exchange sites. The pH buffering capacity of a soil is largely governed by CEC and 1 24 the degree of saturation of the exchange complex with Ae+ and W (often termed reserve acidity; Keeney & Wildung, 1 977). Thus, soils with higher CEC tend to be more resistant to changes in pH than soils with lower CEC. Although pH buffering is known to reduce extreme acidity in mining wastes, the pH of such wastes is also ameliorated due to the liming action of sludge materials. Sludges with near neutral to alkaline pH can raise the pH of acidic mine wastes substantially by simple dilution. In one notable study, Sopper & Seaker ( 1 990) found that the addition of sludge at a rate of 1 84 Mg/ha increased the pH of colliery spoil from 3 . 8 to 6.2 within four months of application. Twelve years later, the pH of the substrate was still found to be comparatively high (5 .4). Several studies outlined by Sopper ( 1 992) indicate that sludge may be more effective at maintaining pH levels than limestone despite it having a substantially lower calcium carbonate equivalency. Joost et at. , ( 1 98 1 ), for example, reported that the application of sludge at rates of between 450 and 900 Mg/ha maintained pH for a longer period of time than limestone applied at a rate of 1 80 Mg/ha. Sludge has also been used to ameliorate the pH of highly alkaline substrates (Topper & Sabey, 1 986). pH buffering involving the liberation of acidic cations from the exchange sites is believed to be largely responsible for alkaline 'neutralisation' in such situations. Sopper ( 1 992) outlines research, however, which indicates that increased infiltration resulting from the addition of sludge may facilitate leaching of alkaline compounds from surface horizons. (iv) Heavy Metal Attenuation Sludge addition generally serves to lower the bioavailability of non-essential metals via mechanisms including cation exchange, specific sorption, chelation, precipitation and complexation with phosphates and hydrous oxides (Chaney & Giordano, 1 977; Corey et at. , 1 98 1 ; Hinkle, 1 982; Seaker & Sopper, 1 98311 984; Sopper, 1 992). Chaney ( 1 973) regarded organic matter chelation along with CEC as the most important factors controlling metal availability. Precipitation and other forms of specific adsorption processes involving covalent bonding to certain functional groups on clay colloids is also believed to dominate in some soils (Keeney & Wildung, 1 977). 125 Whereas single charged cations are held mainly by simple cation exchange, multi­ charged cations such as Cu2+, Mn2+ and Zn2+ tend to form complex chelation linkages with organic matter (Stevenson, 1982). Keeney & Wildung ( 1 977) describe chelation as a form of complexation that occurs between metal ions and organic ligands and results in the "the formation of a heterocyclic ring which includes the metal ion". Chelates tend to form more stable complexes than analogous complexes in which the metal is coordinated in a non-cyclic fashions with water, other ligands or both (Keeney & Wildung, 1 977). Regardless of its preCIse nature, organo-metal complexation is known to play a significant role in decreasing free-metal activity and often serves to palliate phytotoxicity in heavy metal contaminated substrates (Chaney & Giordano, 1 977; Forstner, 1 99 1 ; Alloway, 1 995). Elliot et al. ( 1 986) found that in terms of metal retention the addition of organic matter to growing media was generally most effective under acidic conditions where secondary metal sulphates, carbonates, oxides and hydroxides are unstable. Keeney & Wildung ( 1 977) provide a useful review of the various soils processes and mechanisms that are involved in the fixation and exchange of metals and plant nutrients. 3.7.4.2 Beneficial Physical Effects The potential physical benefits of adding sludge to mining wastes such as the Tui mine tailings are numerous and well documented. Soil physical parameters potentially improved by the incorporation of sludge include; bulk density, water holding capacity, particle aggregation (stability), soil porosity and aeration, hydraulic and thermal conductivity and surface water infiltration (Sopper, 1 992; Land Resources Network Ltd. , 1 993). The combined effects of incorporating sludge is often manifested in a substrate that is more resistant to wind and water erosion and has improved structure, aeration and water holding capacity. Clearly such improvements in the physical nature of a substrate would benefit seed germination and the subsequent establishment of plants. 126 Sopper ( 1 992) and Land Resources Network Ltd. ( 1 993) outlined several studies in which the benefits of sludge addition were shown to greatly improve the physical condition of various sludge amended substrates. Because greatest yield improvements are often observed when organic matter levels in sandy or clayey substrates are increased (Land Resources Network Ltd., 1 993), (i) Bulk Density The degree to which sludge addition reduces the bulk density of a given substrate is proportional to the amount of sludge added and is dependent on the relative bulk densities of the materials (Metzger & Yaron, 1 987). Because most types of sludge have comparatively much lower bulk densities than mining wastes, the addition of sludge to such materials invariably reduces the bulk density via simple dilution. The introduction of organic matter may, however, also indirectly reduce the bulk density by facilitating the establishment of soil organisms that serve to aggregate soil particles and to aerate and improve the structural integrity of the substrate (Land Resources Network Ltd., 1 993) . (ii) Aggregation & Porosity As discussed in section 3 .7. 1 . 1 several factors associated with the addition of organic matter assist soil aggregate formation. Joost et al. ( 1 987) reported that applications of high rates of sewage sludge (225-900 Mglha) reduced the bulk density of acid colliery spoil and increased the number of water-stable aggregates and the proportion of macropores present in the amended spoil. Land Resources Network Ltd. ( 1 993) cites several studies in which similar results were found. Although some research findings suggest that sludge improves the overall porosity of soils, there are conflicting findings regarding the ability of sludges to improve macroporosity (Pagliai et al., 1 98 1 ; Wei et al., 1 985) . Obviously the effects of sludge on macroporosity would depend heavily upon the initial condition of the ' soil' and upon the biological and physical nature of the sludge and the rates applied. 127 (iii) Hydraulic Conductivity In addition to its ability to ameliorate bulk density and porosity, several studies indicate sludge application can substantially improve water infiltration and saturated hydraulic conductivity (Sopper, 1 992). Some research indicates that the presence of organic matter at the ' soil ' surface aids infiltration by preventing crusting (Pagliai et ai., 1 983). Clearly any increase in macroporosity is likely to result in increased infiltration and higher hydraulic conductivity. Improved infiltration arising from the application of sludge has been shown to reduce run-off, erosion and sedimentation from mine spoils (Sutton & Vimmerstedt, 1 974) . Based on the preceding discussion (section 3 .7.4.2) it is clear that the physical parameters effected by sludge addition are highly inter-related and that sludge addition is likely to improve several soil parameters simultaneously. 3.7.4.3 Biological Effects The impact and role that soil micro-organisms play in recycling, fixing and liberating nutrients is perhaps the most poorly understood aspects of land reclamation by practitioners. Many practitioners fail to fully appreciate the complex and interactive role that soil microbial populations play in breaking down organic matter and regulating the chemical and physical fertility of a soil. A 'typical ' soil contains a vast number and variety of micro-organisms including bacteria, fungi, algae actinomycetes and protozoa (Ritcey, 1 989, Land Resources Network Ltd., 1 993). Thus if the ultimate long term objective of a reclamation program is to create a self-sustaining soil ecosystem that is capable of hosting a variety of plants and fauna, an integral part of achieving such a goal will be to ensure that the soil contains a functional microbial population (Sopper, 1 990). Although the composition of microbial popUlations is highly dependent on the micro­ environmental conditions prevailing in a soil medium, the majority of organisms in substrates containing organic matter are heterotrophic saprophytes which gain their energy source from carbon. In substrates that contain negligible organic matter, heterotrophic organisms are largely absent. Instead chemo-autotrophic organisms which gain their energy from inorganic chemical compounds are likely to be predominant. The 1 28 integral role that chemo-autotrophs such as Thiobacillus ferrooxidans play In the oxidation of sulphide minerals is discussed in section 3 . 3 . 1 . Sludge addition to sulphide bearing wastes is likely to adversely affect the stability of chemo-autotrophic bacterial populations due to increased pH and the presence of organic compounds that favour the growth of heterotrophic bacteria (Brown et al. , 1 990). The addition of heterotrophic bacteria capable of breaking down organic matter to release stored macronutrients (chiefly N & P) into the soil solution is also likely to ameliorate growing conditions in amended substrates (Seaker & Sopper, 1 988a). Many of the microbiological processes that govern the availability of individual nutrients, however, are highly complex and involve a multitude of micro-organisms. In the nitrogen cycle for example, the processes of fixation, nitrification, denitrification, mineralisation and immobilisation are all governed by soil microbes (Ritcey, 1 989; Sopper, 1 992). All of the above processes are totally dependent on the presence of at least one specific bacterial population and will not take place unless the bacteria are present and the environmental conditions are suitable (eg. Rhizobial fixation of nitrogen) . Since some bacteria are more sensitive to adverse growing conditions than others, specific nutrient deficiencies may arise as nutrient cycling processes that are catalysed or facilitated by bacteria are inhibited. At high concentrations metals such as Cd, Cr, Cu, Pb and Zn may interfere with microbial function and have been shown to inhibit soil bacterial and fungal activity (Babich & Stotsky, 1 977). Although sludge application to agricultural soils has been shown to decrease microbial biomass due to metal toxicity (Brookes & McGrath, 1984), the application of sludge to mine wastes is generally beneficial in that it increases the diversity and activity of soil micro-organisms. Most of the studies that have investigated the impact of organic wastes on the microbiological populations of mining wastes have indicated that sludge addition invariably enhanced microbial activity (Fresquez & Lindemann, 1 982; Visser et ai., 1 983 ; Stroo & Jenks, 1 985 ; Fresquez & Aldon, 1 986; Seaker & Sopper, 1 988a; Smith, 1 99 1 ; Hetrick et al., 1 994; Shetty et ai. , 1 994). 129 3.7.4.4 Effects of Sludge Addition on Plant Yield Sopper ( 1 992) reviewed several studies in which the combined effects of sludge addition on soils and mining wastes were manifested by improved plant growth yields. In general, the studies demonstrated that good plant cover can be established on many types of disturbed land using municipal sludge. Differences in substrate composition and climate were, however, found to significantly influence the performance of plants species. For example, whilst tall fescue (Festuca arundinacea) was found to perform well in a study investigating the addition of sludge to highly acidic strip-mine spoil (Sopper & Kerr, 1 982), it completely failed when used on soils contaminated by zinc (Franks et al. , 1 982). Research evaluating the use of composted sewage sludge/straw on three contrasting derelict sites in the UK, found that there was a linear relationship between plant yield and composted sewage sludge/straw applied (Atkinson et al. , 1 99 1 ) . The increased yield was attributed to an improvement in the nutrient status and physical properties of the spoils. Another study looking at the use of sewage sludge in the amendment of coal spoils (Moffat et al. , 1 99 1 ) found that the addition of sludge increased tree and ground cover growth. Dramatic increases in foliar N and P were also noted. 3.7.4.5 Problems associated with Sludge Application Detrimental agronomic and environmental problems can arise with the use of organic amendments. Problems associated with the inappropriate use of organic amendments such as sewage sludge include:- ( 1 ) Induced nutrient deficiencies (N, P and S) can arise in materials containing extremely high C levels. (2) Surface run-off of applied waste and/or the leaching of soluble salts (eg. nitrate) from organic wastes can pollute water ways and potentially result in eutrophication. The latter problem of nitrate leaching is usually most severe in situations where the sludge is applied in liquid form. (3) The addition of organic materials to poorly drained substrates can lead to nitrogen loss via biological denitrification. The generation of gaseous compounds such as nitrous oxide via denitrification not only reduces the nutrient content of the amendment but can contribute to ozone depletion and global warming. (4) Some wastes, particularly sludges sourced from industrial areas, contain elevated concentrations of heavy metals and persistent organic toxins. These contaminants may accumulate in soils and be 1 30 introduced into the human food chain by crop or stock uptake, or by the contamination of potable water. (5) In metalliferous wastes organic matter treatments can cause reductive dissolution of heavy metals previously adsorbed or co-precipitated with the iron oxy-hydroxides. This process can ultimately result in the release of potentially toxic metals compounds into surface and ground water (Sopper, 1 992; Land Resources Network Ltd. , 1 993; Ribet et ai. , 1 995). (i) Heavy Metal Contamination The most widely publicised and studied of the potential hazards associated with the use of sludge amendments is that of metal contamination. Sludges often contain appreciable concentrations of metals including Cd, Cr, Cu, Mo, Mn, Ni, Pb and Zn (Forstner, 1 99 1 ). Cu, Mo, Ni, Zn and particularly Cd tend to be the most problematic of these metals in terms of their tendency to inhibit microbial activity and to threaten livestock and human food chains where the soils are used for cropping or grazing. The interaction of sludge on the dietary uptake of metals such as Cd and Zn have been the focus of numerous studies (Allaway, 1 977; Dean & Suess, 1 985 ; Alloway, 1 995; Sopper, 1 992; Smith, 1 994). A working group organised by the World Health Organisation identified Cd as one of the most potentially hazardous contaminants in sludges due to its ability to accumulate in certain crops grown on sludge amended soils (Sopper, 1 992). Fortunately the translocation of most heavy metals into the human food chain is governed to a large degree by phytotoxicity which often results in crop failure in heavily contaminated plant growth media. There is a serious risk, however, of metals entering the food chain via ingestion of sludge amended soils along with forage (US EPA, 1 983). It is necessary to note that specific adsorption associated with sludge addition does not always result in the net immobilisation of metals within the root zone. As mentioned in section 3 .7.2. 1 , the chelation of metals with smaller organic molecules, such as fulvic acid, can result in the formation of small highly soluble complexes (Land Resources Network Ltd., 1 993; Ribet et ai. , 1 995). The mobility of small metal-chelates renders them susceptible to plant uptake and/or leaching from the 'regolith ' . Where excessive 1 3 1 concentrations of metal chelates are present in root zone, phytotoxicity is likely to ensue. Most of the US guidelines relating to the use of sludge are for agricultural soils and are based on trace metal loading and/or vegetation nutrient requirements (Sopper, 1 992). Although no federal guidelines have been issued specifically governing the use of sludge for mine reclamation purposes, some states have developed stringent guidelines concerning the use of sewage sludge in such circumstances (Sopper, 1 992). The Pennsylvanian Department of Environmental Resources, for example, limits sludge application rates based on crop nitrogen requirements and on the trace metal content of the sludge. Metal loading rates for reclaimed land in Pennsylvania are presented in Table 3 .2 . In Pennsylvania and other states, additional site and soil parameters such as topography, pH and CEC are employed to further reduce the environmental risks associated with the use of sludge. The New Zealand Department of Health (DOH) have issued similar guidelines based on soil pH and metal loading (DOH, 1 992). The DOH guidelines include the following controls : - (i) the soil to which the sludge is being applied must have a pH of 5 or greater, (ii) the maximum concentrations of heavy metals in the surface 200mm of arable soils must not exceed those values presented in Table 3 .2 and (iii) the maximum annual rates at which heavy metals contained in sludge can be applied to land must not exceed the bracketed values presented in Table 3 .2 (McLaren & Cameron, 1 996). Furthermore the DOH guidelines require that pH and heavy metal loadings are monitored on a 3 yearly and 5 yearly basis respectively. Clearly the ethical dilemma of using metal-contaminated sludge as a ' soil ' amendment is greatly reduced when the substrate itself contains comparable or higher concentrations of the 'contaminant' . 1 32 Table 3.2 Maximum recommended cumulative loading rates and maximum permitted total concentrations of heavy metals in sludge amended soils. Bracketed data represents the maximum recommended annual loading rate of heavy metal. * = Pennsylvania Department of Environmental Resources ( Sopper, 1 992). � = New Zealand Department of Health Guidelines (DOH, 1 992). Constituent Maximum Cumulative Loading Rate Maximum Soil [kg/ha] Concentration [mg/kgl Pennsylvania* New Zealand

2 mm) of the dam face cover material indicated that it contains comparable levels of Cu, Pb and Zn to the tailings (Table 4. 1 ). This finding indicates that if the material originated from Tui ore as discussed in section 4. 1 , then despite its coarse grain size it has undergone extensive oxidation and leaching since its extraction and deposition. Table 4.1 Selected properties of materials from the Tui mine site based on average values from duplicate analysis. See Appendix 4(C) for calculation of estimated sulphide percentages. ND = metal concentrations were not determined. Site Depth Total S S04- Sulphide S Cu Pb Zn [mm] [mg/kg] [mglkg] [%] Lmglkg] [mglkg] [mglkgl S I 0- 100 934 56 0.09 75 2498 1 12 1 00-400 2 1 30 245 0 . 1 9 147 4229 246 400+ 33977 1 740 3 .22 20 1 5 1 2336 1 83 1 S2 0-50 3483 36 0 .35 269 1 1 76 1 1 92 50- 1 50 1 782 94 0 . 1 7 166 1 2503 1 93 1 50-400 3 3 1 7 1 483 3 .27 644 1 0550 2880 400+ 404 1 6 1 1 85 3 .92 32 19 9502 2634 S3 0- 1 00 2 1 83 49 0.2 1 146 8852 430 1 00-400 32469 670 3 . 1 8 45 1 1 2372 1 3 1 9 400+ 4 1 457 1 200 4 .03 1 733 3227 2699 Bulk Sample 0- 1 00 1 624 20 0 . 1 6 ND ND ND Dam Face Fine « 2mm) 0- 1 00 426 1 1 4 1 0 .37 ND ND ND Coarse (> 2mm) 0- 100 3 865 l 3 1 0.4 1 732 70 1 0 1 668 0 1 00 E 200 E., ..c 0- 300 Q) 0 Q) r..;::; 0 .... 400 0.. 500 600 o 1 00 E 200 E., ..c � fr 300 o Q) r..;::; o 0: 400 0 I ] o • b (a) 6M Hel Extract. eu [mglkg] 1 000 2000 3000 • Site 1 • Site 2 hi Site 3 (b) 6M Hel Extract. Zn [mglkg] 1 000 2000 3000 4000 4000 500 I 600 �------------------------------� Figure 4.2 (a) 'Total' copper and (b) zinc versus depth at each of the 3 preliminary samplings sites 1 48 E S ..c -0-Q) a Q) t;:: 0 '- 0... 0 0 1 00 200 300 400 500 l I 600 I 2.0 6 M He] Extract.Pb [mglkg] 5000 1 0000 \ � 3 .0 8.. pH (Water) 4.0 5 .0 \ 6.0 1 5000 7.0 o _Ir-�----�--��--��--��� I n �:: 1 /1 300 1 "" 400 l \ � 500 J I • • A Site 1 Site 2 Site 3 I I I I I I L-..-____ 1 I 600 -'I _____________ ------l Figure 4.3 (a) 'Total' lead and (b) pH versus depth at each of the 3 preliminary sampling sites. 149 1 5 0 4.1 .2.2 Total Sulphur & Sulphate Concentrations Very high concentrations of total S and phosphate extractable S04 = were also detected in some samples. In a trend similar to that exhibited by Cu and Zn, both total S and S04 = sulphur were found to increase substantially with depth. Total S and S04= concentrations exceeding 33,000 and 1 1 00 mg/kg respectively were detected in each of the 400+ mm samples. The concomitant increase of Cu, Zn and total S concentrations with depth indicated that these metals were present predominantly in the form of sulphides (Table 4 . 1 ). _ Analysis of a bulk sample collected from the surface 1 00 mm of the tailings adj acent to site S3 indicated that the material contained similar S04 = and total S concentrations to the surface sample at site S3 (Table 4. 1 ). The concentrations of both total S and S04 = sulphur were consistent with the concentrations of Cu, Pb and Zn in the tailings and indicated that the bulk sample contained relatively low concentrations of sulphide minerals (Table 4. 1 ) . Sulphate and total S concentrations extracted from the cover material were comparable with levels found in the tailings and confirmed that both grain size fractions of the material contained relatively low concentrations of sulphide minerals (Table 4. 1 ). 4.1 .2.3 Plasma Emission Spectrometry ICP-AES analyses were conducted on two samples obtained from preliminary sampling site S2 (Figure 4. 1 ). The analyses were carried out by the New Zealand Pastoral Agriculture Research Institute on an Applied Research Laboratories 34000 simultaneous emission spectrometer. The total concentrations of 22 elements in the tailings were determined at two depths (0-50 mm and 400+ mm) to assess elemental variability with depth (Table 4.2). With the exception of calcium (Ca), sodium (Na) and strontium (Sr), the concentrations of the measured elements were all found to increase with depth. Although elemental concentrations were generally far higher at lower depth, certain elements including As (arsenic), B (boron), Mo (molybdenum), Cu, Pb and Zn were present in the surface tailings at levels considered to adversely affect plant metabolism (Table 4.2). Of these 1 5 1 elements, As, B and Pb levels were particularly high and found to fall well outside the 'normal range' for soils (Table 4.2). In contrast, total Al (aluminium) and Fe (iron) concentrations in the surface tailings were relatively low. Below 400 mm, however, the concentrations of Al and Fe increased to levels comparable with mean soil levels. Other elements that were found to exhibit substantial concentration increases with depth were As, Cd (cadmium), Cr (chromium), Cu, Se (selenium) and Zn. These elements were all present at concentrations well above the critical levels which delineate the potential onset of phytotoxicity (Table 4.2). Table 4.2 Elemental concentrations of the Tui Mine tailings at preliminary sampling site S2 as determined by ICP-AES [mglkg] . Bracketed data represent Cu, Pb ruId Zn levels determined by F AAS and S concentrations determined by sodium bicarbonate fusion. Sources : * = Bowen ( 1 979); t = Kabata-Pendias & Pendias ( 1 984) "" = Forstner ( 1 995). Element Sampling Depth Sampling Median Normal Range in Onset of (0-50 m m) Depth Level in Soils* Soils* P hytotoxic (400+ mm) Levelst Al 2 1 3 3 4 49223 T 0.45 - 1 0 % As 252 1 524 6 . 0 0 . 1 - 4 0 1 5 - 5 0 B 2 1 3 0 2396 1 0.0 0.9- 1 000· 2 5 - 1 00 Ca 1 64 87 Cd < 1 . 8 26.5 0 . 3 5 0 . 0 1 - 2 . 0 3 - 8 Co 3 . 8 1 9.2 8 . 0 0.5 - 6 5 25 - 5 0 Cr 1 3 . 7 1 44 70 5 - 1 500 75 - 1 00 Cu 149 (269) 4554 (32 1 9) 3 0 2 - 250 60- 1 2 5 Fe 1 0 224 3 7 692 40000 K 2466 3666 Mg 3 6 1 439 Mn 26.0 38.0 1 000 2 - 1 0 000 1 50 0 - 3 000 Mo 1 6 .9 45.0 1 .2 0 . 1 - 40 2 - 1 0 Na 1 5 3 6 1 404 Ni <3 . 5 1 3 .6 50 2 - 750 1 00 P 1 80 7 1 9 Pb 2082D 1 76 1 ) 2988 (9502) 3 5 2 - 3 00 1 00 - 400 S 3927 (34 8 3) 42 1 86 (404 1 6) Se <3 5 . 5 3 8 . 0 0. 1 - 5 5 - 1 0 Sn <3 . 8 5 . 1 1 - 200 50 Sr 1 4 . 8 7.4 2-600T Zn 204 ( 1 92) 3 8 1 9 (2534) 90 1 - 900 70 - 400 Predictably ICP-AES sulphur concentrations in the tailings were high and comparable to the levels obtained using the sodium bicarbonate fusion method of Steinbergs et al. ( 1 962). Cu and Zn concentrations determined by ICP-AES were similar to those determined by F AAS (Flame Atomic Absorption Spectrometry), whereas in the case of 1 52 Pb the fonner technique resulted in substantially lower values than the latter technique (Table 4.2). Subsequent F AAS and ICP-AES analysis confinned that the Pb concentrations were very high in the surface tailings and that ICP-AES detennined Pb concentrations in the preliminary samples were erroneously low. The reason for the disparity between the preliminary Pb results remains unclear but may have resulted from the fonnation of highly insoluble PbS04 during the ICP-AES extraction and dilution procedures. 4.1 .2.4 pH (water) The Tui tailings were found to be extremely acidic, particularly in the 50 to 400 mm sampling zone where pH values were typically between 2.5 and 3.0 and tended to be slightly lower than those found in the surface samples (Table 4.3) . The pH of the tailings was found to increase below 400 mm indicating reduced hydrogen ion (W) activity. Similar pH profiles were obtained from each ofthe 3 sampling sites (Figure 4 .3b). Both grain size fractions of the dam face material were found to exhibit similar pH values in the vicinity of pH 4 and were thus slightly less acidic than the majority of the tailings samples (Table 4 .3) . 4.1 .2.5 Acid Generating Potential Upon treatment with 30% hydrogen peroxide (H202) the pH of all samples was found to fall to below pH 2.5 indicating the presence of appreciable quantities of acid generating sulphides (Table 4.3). The pH data obtained from preliminary characterisation of the tailings was further used to estimate the relative reactivity of the Tui Mine tailings in tenns of their Acid Generating Potential (AGP). Based on the known liberation of hydrogen ions (H+) during sulphide and H202 reaction it was theorised that samples containing high concentrations of acid generating sulphides would liberate greater amounts of H+ than samples containing relatively low concentrations. Using the pH of the samples after reaction with both water and 30% H202, the change in hydrogen ion concentration (�H+) was calculated as follows: Where !!1 [W] = [W]H202 - [H+]H20 [W] = 1 0 (-pH) 1 53 The estimated concentrations of W [mollL] liberated during the oxidation reaction of each sample are presented in Table 4.3 . Very approximate estimates of the Acid Generating Potential (AGP) of the tailings are also presented in Table 4.3 and were made by converting the m+ values for each sample into calcium carbonate equivalents in a similar fashion to the example presented in Appendix 6(B). The values presented in Table 4.3 were considered to be very crude because they were based on the assumption that changes to the sample to reactant ratio will affect the solution pH in a linear fashion. The relative magnitude of the values were, however, expected to be useful in identifying general AGP trends amongst the samples. Estimates of the tailings AGP based on H+ activity indicated that the AGP of the tailings was variable and apparently unrelated to depth (Table 4.3) . This finding was contrary to what was expected as the trends demonstrated by eu, Zn and total S were postulated to reflect an increase in the concentrations of acid generating sulphide minerals with depth. Higher concentrations of W release and AGP values were therefore expected in the samples obtained from deeper within the profile. To assess the validity of the AGP values obtained from H202 oxidation data, the AGP of the tailings was calculated stoichiometrically using the estimated sulphide S values as per the method outlined in Appendix 4(C). Although as discussed in Chapter 3 (section 3 .4. 1 . 1 ) estimates of AGP using sUlphur data are widely employed in the coal mining industry, it was postulated that such a method would overestimate the AGP of the Tui tailings due to the presumed occurrence of relatively high concentrations of the non-acid generating sulphide minerals in the tailings3 . The AGP (2) values presented in Table 4.3 were based on the assumption that all sulphide S in the tailings was present as pyrite 3 Based on literature review outlined in Chapter 2 (section 2.6) galena (PbS) and sphalerite (ZnS) were predicted to be the primary non-ferrous sulphides present in the tailings. 1 54 Table 4.3 pH data and Acid Generating Potential (AGP) estimates [Mg CaC03/ 1 000 Mg material] of the Tui dam materials based on ( 1 ) hydrogen ion liberation and (2) sulphur analysis. Site Depth pH pH � [W] AGP AGP [mm] (H2O) (H202) [mollLl (1) (2) S l 0- 1 00 3 .45 l .62 2.36 x 1 0-2 30 2 .7 1 00- 400 3 . l 6 1 .84 1 .38 x 1 0-2 1 8 5 .9 400+ 5 .74 2.06 8 .71 x 1 0-3 1 1 1 0 1 S2 0-50 3 .55 l .59 2.54 x 1 0-2 32 1 0 .8 50 - 1 50 3 .0 1 1 .97 9.74 x 1 0-3 1 3 5 . 3 1 50 - 400 2.78 1 . 8 1 1 .3 8 x 1 0-2 1 9 1 02 400+ 3 .43 1 . 89 1 .25 x 1 0-2 1 6 1 23 S3 0- 1 00 3 . 1 4 2.36 3 .64 x 1 0-3 5 6.7 1 00 - 400 2.52 1 .77 1 .40 x 1 0-2 20 99 400+ 4.06 2.04 9.03 x 1 0-3 1 1 1 26 Bulk Sample 0- 1 00 3 .43 1 .98 l .0 1 x 1 0-2 1 3 5 .0 Dam Face Fine « 2mm) 0-1 00 3 .89 2.48 3 . l 8 x 10�3 4 1 2.9 Coarse (> 2mm) 0- 1 00 4.07 2.27 5 .29 x 1 0-3 7 1 1 .7 The AGP of the tailings, as predicted from sulphur/sulphate analysis, was found to increase substantially with depth at each of the 3 sites (Table 4.3) . The AGP values of the 400+ mm samples were extremely high; at least 1 0 times higher than the surface samples at corresponding sites. Interpretation of the results presented in Table 4.3 indicate that for some samples liming rates in excess of 1 00 Mg of CaC03 would have been required to neutralise the acid theoretically produced by complete oxidation of 1000 Mg of tailings. Regression analysis of values generated by the two AGP methods (ie. H+ generation versus sulphur) indicated that there was no significant interaction between the two parameters. The major reason for the disparity between the AGP estimates is believed to have resulted from the inability of the added volume of 30% H202 to fully oxidise sulphide minerals contained by some samples. Subsequent research was to show that substantially larger volumes of H202 were required to fully oxidise subsurface samples rich in sulphides (Chapter 4.3). Although this problem could have been readily 1 55 overcome by reducing the sample to H202 ratio no attempt was made to modify the procedure in the present study. Another limitation identified with the assessment of AGP based on H202 oxidation was the variability of the pH values following reaction with the oxidant. The pH values obtained for the surface tailings at site S 1 for example, were found to vary from 1 .62 to 2 . 1 1 after H202 addition and reaction. When these values were used to estimate the AGP of the tailings, calcium carbonate equivalent values of 29.8 and 9.6 Mg CaC0311 000 Mg of tailings were obtained respectively. The large disparity between duplicate samples highlights the crudity of this method to predict AGP values. Clearly, pH data generated by simple H202 methods such as that outlined section 4. 1 . 1 .2 must be used cautiously. It is evident that screening procedures should refrain from using pH measurements to rank the likely AGP of sulphide-bearing materials once identified. It is postulated that the AGP estimates based on SUlphur analysis better predicted the potential of the tailings to generate acidity. This hypothesis is substantiated by observed increases in 6M HCl extractable Cu and Zn concentrations with depth and by field observations which confirmed the presence of visible sulphide minerals within tailings originating from beneath 300-400 mm . depth. 4.1.3 Conclusions on Preliminary Characterisation Preliminary characterisation of the Tui Mine tailings indicated that the tailings are a hostile plant growth medium due primarily to their extremely acidic pH and elevated heavy metal concentrations. Chemical analyses conducted on the various samples highlighted the extremely variable chemical nature of the tailings, particularly with depth. Samples obtained from corresponding depths at each site, however, were found to contain comparable concentrations of Cu, Zn, total S and S04 = indicating that certain chemical parameters varied with depth in a predictable manner. Whilst total S, S04 =, Cu and Zn concentrations were lowest in the surface samples and increased substantially with depth, Pb concentrations in the surface samples were comparatively very high and well above the range normally found in soils. 1 56 Although the phyto-availability of Cu, Pb and Zn could not be estimated from the values obtained using the 6M HCl acid extraction procedure, it was postulated that the extremely acidic conditions prevailing in the upper tailings profile would render most labile metal forms readily available to plants. Strongly acidic conditions were found to persist throughout the profile but were particularly severe in the central region of the 3 profiles where pH values as low as 2.52 were encountered. Preliminary pH data indicates that this parameter is also depth dependent. Hydrogen peroxide oxidation and elemental analysis indicated that the tailings contain appreciable quantities of acid generating sulphide minerals within 0.5 metres of the tailings surface. Correlation of these variables, however, was poor. Determination of the Cu, Zn and sulphide S concentrations indicated that the use of this H202 method to estimate the AGP of the tailings is inadvisable and that it should not be used to rank the relative reactivity of sulphide-bearing samples. Stoichiometric AGP estimates based on sulphur analysis were found to correlate well with Cu and Zn concentrations within the profile and are believed to far more accurately reflect the 'true' AGP of the Tui tailings. Despite the disparity between the AGP results usmg the two different estimation approaches it is clear the tailings contained substantial amounts of reactive sulphide minerals and that more detailed chemical analysis was required to elucidate the tailings' AGP and the distribution and availability of heavy metals within the tailings dam. The lack of Fe concentration data with depth prevented better interpretation of AGP results. Preliminary characterisation of the material covering the western and southern slopes of the dam (Plate 3 .4) indicate that this material contains relatively low concentrations of sulphides minerals and has limited ability to produce acid. Based on this finding a decision was made to focus research on characterisation of the tailings as they were envisaged as being far more problematic to revegetate. 4.2 Physical Characterisation 4.2.0 Introduction 1 57 To facilitate further chemical characterisation of the Tui tailings, a series of twelve sampling sites (grid sites 1 to 1 2) was set up on the dam in a grid arrangement (Figure 4 . 1 ). At each of the 1 2 sampling sites a piezometer was installed. Although the primary purpose of the piezometers was to act as location markers for the sampling sites, they yielded valuable information relating to the position of the water-table within the Tui tailings impoundment. The water-table data collected from the 1 2 piezometers was compared with that obtained from four previously installed piezometers and is discussed below in combination with other physical characterisation work that was carried out during the course of this study. 4.2.1 Piezometer Installation As a part of the monitoring program that has been conducted at the Tui Mine site, four piezometers have been periodically monitored by Works Consultancy of Hamilton over the past two decades. The piezometers are located at western end of the tailings embankment and it is believed they were installed during the dam's construction to monitor de-watering. Their installation, however, may have been prompted by a rainstorm in 1 97 1 which seriously eroded the tailings. Subsequent storms during 1 974 and 1 976 also caused localised slumping of the dam and raised concerns regarding the stability of the dam (WRCM, 1 99 1 ) . Other factors that may have contributed to erosion of the tailings during these storm events include; the presence of artesian springs beneath the dam, the fine grain size of the tailings and the steepness of the surrounding terrain and tailings dam batters. It is the author's opinion that these factors, combined with the site's high rainfall, render the tailings long-term stability highly questionable despite the extensive stabilisation work that was carried out by the Catchment Board in 1 976 (WRCM, 1 99 1 ) . Water level data recorded from the 4 piezometers, which are now mostly dysfunctional, is held by the Matamata-Piako District Council. It indicates that the phreatic water-level in the tailings has generally rested more than 5 metres below the tailings' surface at the . south-western comer of the dam. The water-table was found to rise towards the eastern end of the dam and intercept the surface of the tailings in the vicinity of site 1 1 (Figure 1 5 8 4. 1 ). To obtain water-table data from elsewhere within the dam, 12 additional piezometers were installed in a grid arrangement that encompassed most of the dam's surface. The piezometers were constructed from 1 .2 m lengths of 0 . 1 5 m internal diameter PVC drainage pipe. The bottom metre of the PVC lengths were drilled with a series of 1 0 rnm diameter holes to enable ground water to flow freely into the tubes from the surrounding tailings. The perforated end of the piezometer tubes were inserted into pre-dug holes of approximately 1 .0 m depth4 at each of the twelve sampling sites (Figure 4. 1 ) . After backfilling, removable plastic caps were fastened to the tops of the tubes and the piezometers were left to equilibrate for 24 hours prior to initial water level readings being taken. 4.2. 1 .1 Piezometer Results The results obtained from the 1 2 piezometers were consistent with those obtained from the pre-existing piezometers and indicates that the water-table lies deeper than a metre below the tailings surface over much of the dam (Table 4.4). The 8 piezometers located at the western end of the dam were mostly dry at each inspection and are postulated to have remained largely devoid of water throughout the monitoring period. Although the exact depth of the water-table could not be determined due to the shallow installation depth of the piezometers, the absence of water within the top metre of the tailings profile indicates that a rising water-table is unlikely to adversely affect plant growth in the central and western regions of the tailings dam. In contrast to these sites, the water-table at sites 1 , 2, 5 , 8, and 1 1 was usually found to lie within a half a metre of the tailings surface (Figure 4. 1 ) . On several occasions both during and after the monitoring period the water-table was observed either at or within a few centimetres of the tailings surface at these locations. The slight lowering of water levels in the piezometer tubes at sites 1 , 2, 5, 8 and 1 1 during the summer months indicates that some seasonal variation may occur (Table 4.4). 4 This depth was considered adequate for the purposes of this study as it provided an insight into water­ table effects within the potential plant rooting zone (ie. within the surface 1 .0 m of the profile). 1 5 9 The piezometric data obtained during the monitoring period suggests that regardless of any seasonal variation, the water-table at and around these sites is likely to remain sufficiently close to the surface of the tailings to cause problems to establishing plants throughout the year. Permanent or extended periods of root zone saturation is a problem that any tailings reclamation scheme will have to address at the Tui mine site, particularly if it involves direct planting of the tailings. Table 4.4 Piezometric data for the Tui tailings dam. Readings are in mm and refer to the distance of the water-table below ground level at the site. 'Dry' indicates that the water-tables lay below 1 000 mm . Piezometers 1 , 1 1 & 12 were not installed until the 3 0- 1 1 -92 . Date 51 52 53 54 55 56 57 58 59 510 51 1 512 1 0- 1 1 -92 - 20 Dry Dry 1 80 Dry Dry 2 5 0 Dry Dry - - 1 3 - 1 1 -92 - 1 4 0 Dry Dry 1 00 Dry Dry 1 5 0 Dry Dry - - 3 0- 1 1 -92 2 5 0 0 400 Dry 0 Dry Dry 3 0 Dry Dry 1 00 Dry 1 5-02-93 4 5 0 3 5 0 Dry Dry 460 Dry Dry 280 Dry Dry 3 50 Dry 07-03-93 400 300 Dry Dry 400 Dry Dry 1 5 0 Dry Dry 3 00 Dry 1 8-06-93 2 3 0 1 00 Dry Dry 1 00 Dry Dry 4 0 Dry Dry 1 40 Dry 03 - 1 2-93 Dry 340 Dry Dry 600 Dry Dry 800 Dry Dry 530 Dry 26-08-94 70 40 Dry Dry 3 0 Dry Dry 5 0 Dry Dry 40 Dry Traditional drainage methods that use perforated pipes and/or gravel as a conduit for drainage water in subsoils are unlikely to lead to successful, long-term drainage of the Tui tailings. The potential inadequacy of such systems was highlighted by the piezometers installed at sites 2, 5 , 8 and 1 1 . These piezometers were found to continually refill with sediment upon cleaning due to the ingress of finely-ground tailings through the water access holes. Based on these observations it is therefore highly probable that sub-surface drains would quickly clog with tailings of low saturated hydraulic conductivity and cease to effectively drain the tailings. The function of sub-surface drainage is likely to be further hindered by the distinctively layered nature of the Tui Mine tailings (Chapter 4.2.2). Due to observed textural variation between the layers it is probable that the coefficient of hydraulic conductivity in the horizontal direction is far greater than that in the vertical direction in many parts of the dam. In such circumstances exclusively vertical downward movement under 1 60 leaching can not be assumed (Easton-Groves et al., 1 982) and the efficacy of sub­ surface drains is thus likely to be further diminished. 4.2.2 General Description The distinctively layered nature of the Tui tailings is immediately obvious upon excavation of the tailings (Plate 4 . 1 ). The layers are readily distinguished by discrete changes in colour and/or texture and generally vary in thickness from a few mm to several cm. Although the colour of the individual layers vary considerably, two distinctive colour zones are evident in cross-sectional view (Plate 4. 1 ) . Chemical characterisation of the tailings will show that these two colour zones broadly relate to the oxidative state of the tailings and indicate a dramatic change in the AGP of the Tui tailings. Plate 4. 1 Tui mine tailings exposed in cross-section at grid sampling site 7 . This photograph highlights the layered nature of the tailings. Smith & Sobek ( 1 978) outlined the use of Munsell chroma values of powdered samples to evaluate the extent of oxidation in colliery spoils. They found a chroma value of 3 or more generally indicates that Fe oxidation has taken place and that there is probably an absence of pyrite. Chroma values of 2 or less were used to identify spoils in which Fe is absent or in a reduced state. 1 6 1 The laminae and layers o f Tui tailings in the top 200 typically exhibit high chroma colours indicative of aerobic conditions (Figure 4.4). The tailings in this zone are characteristically pale yellow (2.5 Y 7/3)5 or very pale brown ( 1 0 YR 7/4). An abrupt colour change is generally observed at about 200 mm depth due to a presumed change in redox potential and the onset of predominantly anaerobic conditions deeper within the tailings' profile. The tailings below this depth are largely unoxidised and are characteristically grey (N 5/ ) or dark grey (N 4/) in colour. The boundary between the two colour zones almost invariably occurs within 200 to 300 mm of the tailing's surface over most of the dam. The boundary is characterised by interlayered horizons of low and high chroma tailings that generally exhibit well­ defined, non-gradational boundaries. The colour of the layers within this zone (200 to 300 mm) appears to be at least partially governed by grain size, with layers comprised finer-grained particles tending to exhibit grey colouring and coarser-grained layers, pale yellow and brown colours characteristic of the upper horizon. Although most prevalent at the boundary of the two colour zones, isolated layers of grey (unoxidised) tailings were also found in the tailing' s upper profile and at some locations exposed at the dam's surface. Much of the area bounded by sites 8 , 9, 1 0 and 1 1 (Figure 4. 1 ) for example, is covered by a 1 0 to 30 mm layer of fine-grained , dark-grey tailings. This layer was first noted during initial characterisation of the tailings in March 1 993 and was still present in October of 1 996. The fact that this layer of unoxidised tailings overlies about 300 mm of pale brown (oxidised) tailings and grey (unoxidised) tailings at depth indicates that some layers are relatively very resistant to oxidation and that air and moisture must be infusing laterally into the underlying tailings. Shattered crusts of Fe-rich tailings were also found on the tailing' S surface at several locations. These crusts are typically only a few to several millimetres in thickness and are generally dark brown in colour (strong brown 7.5 YR 4/6). XRD analysis of the crusts indicate they comprise primarily quartz, with minor kaolin, goethite and possibly some haematite. 5 Munsell Soil Colour Charts (USA) O m 0.2 m 0.4 m 0.6 m 0.8 m l .O m 162 Oxidised Zone - Tailings in the upper 0.2 to 0 .3 m of the profile are typically pale yellow (2.5 Y 7/3) or very pale brown ( 1 0 YR 7/4) in colour, however, pink (5 YR 7/3), white (2 .5 Y 811 ) and pinkish white (7.5 YR 8/2) are common colour variations exhibited by some layers. Isolated horizons of grey unoxidised tailings are often present in this zone. The surface tailings characteristically contain relatively low concentrations of sulphide minerals and most heavy metals, however, pH levels are generally extremely acidic. Unoxidised Zone - Below about 0.2 m depth the tailings are largely unoxidised and are characterised by grey (N 5/) and dark grey (N 4/) colours. Layers in this zone often exhibit a bluish hue, particularly where the texture of the tailings is very fine. Tailings in this zone are characterised by relatively high concentrations of pyrite and associated sulphide minerals that are often visible to the naked eye. Heavy metal concentrations are accordingly high and pH conditions moderately acidic. Figure 4.4 Cross-sectional diagram of the Tui mine tailings profile at grid sampling site 7 - - - ------------� 1 63 The majority of the Fe in the crusts is probably in the form of ferrihydrite, an amorphous Fe/AI compound not detected readily by XRD. It is postulated that these crusts formed as a result of the rapid oxidation of sulphide-rich tailings exposed at the tailings surface as the overlying, unconsolidated tailings particles were eroded by wind and water movement. Several forms of hardpans are known to form in pyritic tailings including the reddish brown variety present at the Tui mine and a grey coloured variety that forms under reducing conditions from ferrous iron (Ahmed, 1 995). Surface pans (such as those occurring at the Tui Mine) that form from the oxidation of pyrite-rich tailings are considered unstable as their electrochemical reactivity and structure render them susceptible to hydrolysis and dissolution in water (Ahmed, 1 995). 4.2.2.1 Tailings Grain size Despite obvious textural variation between individual layers, the grain size of the Tui tailings was relatively uniform, with the majority of particles falling in the range of fine sand (0.25- 0. 1 mm) to silt (0 .05-0.002 mm). The only site to show any marked variation in grain size was grid sampling site 1 (Figure 4. 1 ) which was found to contain about 25 % angular pebbles (>4 mm) by weight in the surface 200 mm of the profile. The presence of this coarse material is attributable to the erosion and down-slope movement of crushed rock from the site of the old Tui ore processing plant (Plate 4.2). A full-dispersion gram size determination was conducted on the six 1 00-mm incremental tailings samples6 to investigate the variation of grain size with depth (Table 4.5). The grain size fractionation method used was based on the wet pipette method described by Thomas ( 1 973). As further discussed in Chapter 6 (section 6 .4.2 . 1 ), the surface tailings (0- 1 00 mm) at site 7 were found to contain a lower proportion of silt than most of the subsurface samples (Table 4.5). The relatively low concentrations of silt in the surface sample is believed to reflect post-depositional removal of the finer­ grained tailings by wind and water erosion. Evidence of wind resorting is visible at the western end of the tailings dam adjacent to sites 4, 7 and 1 1 (Figure 4. 1 ) where welI- 6 The incremental samples were obtained from between 0 and 600 mm depth at grid sampling site 7 as discussed in section 4.3 .L2 1 64 sorted fine sand-sized tailings particles are found deposited on the inside edge of the wall of the tailings dam. These deposits are commonly greater than 300 mm thick and are protected from the prevailing westerly wind by the front wall of the dam. Table 4.5 Particle size determination of site 7 incremental samples. Analyses are presented as a percent of oven-dry sample. Bracketed data = grain size range [mm] based on USDA classification. Depth Clay Silt Very Fine Fine Sand Medium Coarse Very Coarse [mm] Sand Sand Sand Sand « 0 .002) (0 .002-0 .05) (0.05-0 . l ) (0 . 1 -0 .25) (0.25-0.50) (0 .50- l .0) ( 1 .0-2.0) 0- 1 00 4.6 2 1 .4 32.2 34.6 6 .9 0 .2 0.0 1 00-200 8 .8 26.4 24. 1 27.5 12 .6 0.6 0.0 200-300 7.4 53 .3 27.6 1 0.0 1 .6 0 . 1 0 .0 300-400 8 .5 60.2 23 . 1 7.4 0 . 5 0 . 3 0 . 0 400-500 4 .5 3 8.2 3 1 . 1 2 1 . 1 1 .4 1 .7 2 . 1 500-600 0.4 27.0 3 9.4 27.0 3 . 0 1 .5 1 .6 The oxidation and subsequent dissolution of fine-grained sulphide minerals in the surface tailinf9may have also contributed to the inferred depletion of silt-sized particles from the surface tailings. The high surface area of silt-sized particles renders them more susceptible to biochemical oxidation than coarser particles (Caruccio, 1 975). Thus, any silt-sized sulphide minerals in the upper profile may have been preferentially oxidised and removed from the tailings by lateral or vertical drainage of the oxidation products in solution. With the exception of the sample from the 500-600 rum depth, the clay fraction « 0.002 mm) of the tailings was found to be relatively uniform with depth, averaging about 6% (Table 4 .5).XRD analysis of the tailings samples indicates that the clay in the samples is primarily kaolinite (Chapter 5) originating from the Tui ore (Courtney et ai., 1 987). 1 65 Plate 4.2 Debris and degradation at the Tui mine site. The foundations of the Tui ore mill and processing site are visible in the top right hand comer of this photograph. Debris from this site including wood and angular pebbles were encountered at grid sampling site 1 (Figure 4. 1 ) . 4.2.2.2 General Discussion The two distinctive colour zones observed in the Tui tailings are postulated to broadly equate with the oxidation and reduction zones described by Boorman & Watson ( 1 976) . During their investigation of abandoned base metal tailings they also identified a third zone, the hardpan zone. Ritcey ( 1 989) cites several additional studies where similar mineralogical zones have been identified in materials ranging from uranium tailings to colliery spoil. These studies and the recent study by Lin ( 1 997) indicates that tailings deposits often exhibit progressive change in mineralogical composition with depth to form distinctive zones. These zones develop with time and Ritcey ( 1 989) likens the processes that form them to those that occur during pedogenesis and the weathering of mineral veins. 1 66 The oxidation zone (or oxygenated zone as it is referred to in the study of Garga et ai. , 1 983) i s situated at the top of the profile where the tailings are exposed to atmospheric oxygen and moisture (Figure 4.4). Boorman & Watson ( 1 976) found that base-metal tailings in this zone were characterised by depleted concentrations of Cu and Zn minerals and the presence of secondary anglesite (PbS04). They also identified that with the exception of Pb, the concentrations of metallic cations and sulphate present in the pore waters increased with depth. Because the exchange of oxygen and the populations of aerobic bacteria are generally depth dependent the control of sulphide oxidation is considered by many as a surface problem (Sorensen et al., 1 980). The thickness of the oxidation zone is therefore governed to a large extent by the depth to which oxygen can ingress into the tailings. The low chroma colour of the Tui tailings indicates that diffusion of oxygen into the Tui tailings is largely confined to the upper 200 to 300 rum of the profile. The shallow oxidised zone is believed to reflect the Tui tailings' fine grain size which inhibits oxygen diffusion into the tailings. The hardpan zone described by Boorman & Watson ( 1 976) is often referred to as the transitional zone or enrichment zone and lies directly beneath the oxidation zone (Ritcey, 1 989). Hardpans or transitional zones generally develop as a result of the precipitation of relatively insoluble compounds (eg. silica and metal oxides) from leachates percolating down the profile (Ritcey, 1 989). The geochemistry of the transitional zone is therefore strongly influenced by the biogeochemical interactions and products of the overlying oxidised zone (Garga et at., 1 983) . The deposition of alteration minerals in the transition zone may eventually fully impede drainage and root penetration (Lin, 1 997). As the various reactions that take place during sulphide oxidation are quite dependent on the presence of oxygen and water the formation of a hardpan can potentially reduce the oxidation rate of tailings beneath the pan (Ahmed, 1 995; Lin 1 997). The reduction zone of Boorman & Watson ( 1 976) is sometimes referred to as the redox or hypogene zone (Garga et at. , 1 983; Ritcey, 1 989). This zone is analogous to the C horizon of a soil in that it represents the unweathered parent material of the tailings 1 67 profile (Figure 4.4) . The mineral constituents of tailings within the reduction zone remain essentially unaltered. At the Tui mine site this zone is believed to prevail relatively near the tailings surface (generally within 300 mm of the surface) and is characterised by the onset of continuous and relatively uniform dark grey tailings. 4.2.3 Penetration Resistance To investigate the presence of a sub-surface hardpan that could potentially inhibit plant growth by impeding root penetration and drainage, penetration resistance (PR) data was collected from sites 1 to 9 . Although hardpans occur naturally in pyritic (FeS2) tailings they are more commonly associated with deposits rich in pyrrhotite (FeS) (Ahmed, 1 995). The development and/or existence of a sub-surface hardpan in the transitional zone is dependent on a number of factors including; the mineralogical composition of the tailings deposit, the climate, the genre and abundance of bacteria, the disposal technique (eg. subaerial versus aerial tailings discharge) and the addition of liming agents (Ritcey, 1 989). The occurrence and height of water-tables are also usually of critical importance to the formation of hardpans as they influence the precipitation and redissolution of reaction compounds within metalliferous tailings (Lin, 1 997). Although no distinct hardpans were detected during the excavation of sampling pits at grid sites 7 and 9, layers of more indurated tailings were encountered. The use of a cone penetrometer facilitated rapid quantification of the tailings PR at these sites and at other sites where high water-tables were known to persist throughout much of the year (section 4.2. 1 . 1 ). 4.2.3.1 Penetration Resistance Methodology Penetration resistance (PR) at the nine sites was measured with a Rimik CPI O Cone penetrometer to a depth of 450 mm. An in-built data-logger recorded penetration data at 1 5 mm intervals down the profile. Between 1 2 and 1 8 passes with the penetrometer were made within 2 metre radius of the piezometers at each site. The PR data obtained from the 9 sites was used to calculate an average PR value for each depth. 1 68 Tailings samples were taken from 3 zones (0- 1 50 mm, 1 50-300 mm, 300-450 mm) to calculate the gravimetric water content of the tailings at each of the sampling sites. This enabled the effect of tailings' moisture content on PR to be assessed. 4.2.3.2 Penetration Resistance Results The average penetration resistance (PR) of the tailings varied substantially, ranging from 605 kPa to 1 545 kPa at sites 5 and 1 respectively (Table 4.6). The PR of the tailings also varied significantly with depth. Average resistance values at site 3, for example, ranged from approximately 450 kPa to nearly 3050 kPa (Figures 4.5b) . Although initially there appeared to be little similarity between the sites' penetration resistance profiles, closer examination of the plots indicated that most of the sites could be separated into two groups. Sites 1 ,2 ,5 , and 8 all exhibited maximum resistance to penetration between 1 50 and 200 mm depth (Figure 4.5a), whereas the maximum penetration resistance at sites 3 , 4, 7, and 9 was encountered between 350 to 400 mm depth (Figures 4.5b). Site 6 was the only site to exhibit no distinct profile maxima within either of these depth zones (Figure 4.6). The fact that those sites exhibiting maximum PR deeper within the profile (ie. between 350 and 400 mm depth) were all found to lie in the drier central and western areas of the dam (ie. water-tables > 1 .0 m deep) indicated that soil moisture content was possibly affecting penetration resistance. To further investigate the influence of the tailings' moisture content on cone penetration, penetration resistance data were averaged over 0- 1 50 mm, 1 50-300 mm and 300-450 mm depths at each site and regressed against the gravimetric water content ((j) of the tailings assessed over the corresponding depths. Because ' soil ' shear strength generally decreases with increasing ' soil ' moisture content (McLaren & Cameron, 1 990), the tailings' penetration resistance was predicted to negatively correlate with the gravimetric water content of the tailings Six of the nine sites were found to show some correlation (ie. r2>0.5) between water content and penetration resistances. Due to the low number of regression pairs, however, a significant relationship (P<0.05) was only identified at one of the nine sites investigated (Table 4.6). Three of the sites (3 , 4, 9) were found to exhibit a positive o o 1 00 � 200 300 - 400 - 500 - 600 � o o 1 00 � 200 - 300 400 500 600 -' Penetration Resistance [kPa] 1 000 2000 • Site 1 Site 2 A Site 5 --� Site 8 Penetration Resistance [kPa] 1 000 2000 + Site 3 ---e- S ite 4 Site 7 ----.-- S ite 9 3000 3000 Figure 4.5 Tailings penetration resistance versus depth at specified grid sampling sites. 1 69 (a) (b) E .s ..c o o 1 00 � 0.. (!) 300 -i o (!) t;:: o 0::: 400 � i 500 -l 600 Penetration Resistance [kPa] 1 000 2000 3000 Site 6 Figure 4.6 Tailings penetration resistance versus depth at grid sampling site 6. 1 70 1 7 1 relationship between PR and water content (ie. the PR of the tailings was found to increase with increasing water content). This finding contradicts predicted results in that high PR at these sites did not correspond with drier zones within the tailings' profiles. Other factors such as bulk density, grain size and cementation were obviously collectively outweighing the effects of moisture content at these 3 sites which lay in the unsaturated central and western regions of the dam. Table 4.6 Penetration Resistance CPR) data from Sites 1 to 9 . CO = gravimetric water content; S = significant (P<0.05); PS = possibly significant (PO. l ) . Sampling PR PR PR r 2 Site Average Range Correlation (significance) (kPa] [kPa] with CO 1 1 545 230 - 2628 - 0 .05 (NS) 2 78 1 3 3 1 - l 386 Negative 0.984 (PS) 3 1 4 1 0 45 1 - 3 044 Positive 0.983 (PS) 4 1 365 400 - 2093 Positive 0 .968 (NS) 5 605 335 - 992 Negative 0.999 (S) 6 947 498 - 1 222 - 0.0 1 (NS) 7 l 34 1 424 - 2202 - 0 .44 (NS) 8 1 1 95 447 - 2323 Negative 0.64 (NS) 9 1 487 241 - 3 1 05 Positive 0.95 (PS) Sites 2, 5 and 8, which demonstrated a negative correlation between PR and gravimetric water content, were situated at the eastern end of the dam where piezometric data indicates that the water-table remains close to the tailings' surface throughout most of the year. The PR trends observed at these sites are postulated to simply reflect the relatively low water content of the tailings in the upper profile (Figure 4.5a). The results of PR characterisation indicated that although no distinct subsurface hardpans have fonned within the tailings, incipient cementation of the tailings may be occurring within specific horizons at some sites. At most sites, however, the average PR resistance of the tailings was below the widely recognised critical value of 2000 kPa 7 indicating that plants roots should readily explore this medium should they establish. 7 PR values in excess of 2000 kPa are considered to restrict root penetration (Unger & Kaspar, 1 994) 1 72 4.2.4 Infiltration Capacity To assess the ability of the surface tailings to absorb rainfall, the infiltration capacity of the tailings was estimated in the field using ring infiltrometers . The infiltration capacity of the tailings was assessed at sites 7 and 9, where six ring infiltrometers were placed randomly within a 2 metre radius of each piezometer. Infiltration rates were measured over the first 5 minute period and then again after water had been continually ponded above the tailings for 3 0 minutes. The infiltration rates at both sites were found to have decreased markedly after this period and were believed to be at or very close to equilibrium. The average infiltration capacity of the surface tailings at sites 7 and 9 were substantially different. Initial infiltration rates measured over the first 5 minutes were found to vary from 49 mmJhr at site 7 to as much as 1 272 mmJhr at site 9 . Although rates were found to vary at the individual sites, the surface tailings at site 7 generally had far greater capacity to absorb ponded water than the tailings in the vicinity of site 9 (Table 4 .7). After the equilibration period the average infiltration rate at site 7 decreased dramatically, however, it was still relatively high (269 mmlhr) compared to site 9 where an average infiltration rate of less than 20 mmIhr was obtained (Table 4.7) . Although intra-site variability was also significant the final infiltration rates at the two sites were again clearly different (Table 4.7). Table 4.7 Average initial and [mal surface infiltration rates at grid sampling sites 7 and 9 based. The measurements were conducted in the field using ring infiltrometers. The standard deviation values presented are based on six replicate measurements made at each site. Infiltration Site 7 Site 7 Site 9 Site 9 Initial Rate Final Rate Initial Rate Final Rate [mmIhr] 773 ± 3 14 269 ± 50 35 ± 9 1 9 ± 8 Site 7 - The surface tailings at this site are predominantly fine-sand sized (0.25 and 0 . 1 mm) and exhibit relatively high infiltration rates (269 mmlhr). I'-- � � --' J ! � "-.,. \0. Site 9 - The occurrence of a layer unoxidised tailings dominated by fine silt-sized particles substantially reduces the infiltration capacity of the tailings in the vicinity of site 9. Figure 4.7 Schematic d iagram explaining infiltration rate variability at grid sampling sites 7 and 9. 1 73 1 74 The substantial difference between the sites' infiltration capacities was explained upon removal of the infiltrometer rings from the tailings. It was found that the base of the infiltrometer rings at site 9 had penetrated either into or through a layer of extremely fine-grained tailings of low permeability (Figure 4.7) . Although no more than a few cm in thickness, this layer was effectively controlling the infiltration rate of water into the tailings. Because no such layer was present in the surface tailings (0- 1 00 mm) at site 7, the infiltration capacity of the tailings at this site was largely dictated by the rate at which the water could permeate through the comparatively coarse particles that generally dominate the surface 1 00-mm of the profile at most sites. These results agam demonstrate the variable nature of the tailings. As previously discussed (Chapter 4.2.2) certain areas of the dam (including site 9) have layers of finely ground tailings at or within a few centimetres of the dam's surface. The infiltration capacity of water at these sites is relatively low as evidenced by visible rilling and the formation of dendritic drainage patterns on the tailings following moderate to heavy rainfall events. Clearly the presence of low permeability layers near the dam' s surface largely dictates the 'effective' rate of water infiltration at a given site. 4.2.5 Bulk Density To assess the variability of the tailings bulk density with depth, 30 cores were taken from the field at grid sites 7 and 9 from 3 depths (Table 4 .8) . Samples were collected using aluminium cores of approximately 88 cm3 volume. The pre-weighed cores were driven vertically into the tailings at the 3 sampling depths with a wooden block and hammer. The samples were carefully removed by excavating around the core to prevent disintegration of the sample. After the cores had been trimmed and cleaned, the moist samples were weighed and dried in an oven at 1 05 °C oven for 48 hours . The samples were then reweighed to enable the dry bulk density of each sample to be calculated. Results indicate that there is a slight increase in the bulk density of the tailings with depth at both sites (Table 4.8) . In addition to greater compaction, this increase is believed to reflect the increasing abundance of higher density sulphide minerals with depth as discussed in section 4.4. Although there was some inter-site variation, the bulk density of Tui tailings at corresponding depths was comparable at sites 7 and 9 . The 1 75 average bulk density of tailings at both sites were approximately 1 .3 1 Mg/m3 . This value although moderately high is comparable to many mineral soils (especially within subsoils horizons) and is therefore not considered likely to adversely effect plant root exploration. Table 4.8 Average bulk density values [Mglm3] of the Tui tailings at various depths. Values and sample standard deviations are based on 5 replicate cores. DeQthJmmJ Site 7 Site 9 0 - 50 1 . 30 ± 0.04 1 .27 ± 0 .02 200 - 250 1 . 32 ±O. O l 1 .32 ± 0 .02 400 - 450 1 . 33 ± 0.03 1 . 3 5 ± 0.05 Site Average 1 . 32 ± 0.03 1 . 3 1 ± 0 . 05 4.2.6 Conclusions on Physical Characterisation Physically, it appears that the Tui Mine tailings pose no major limitations to plant growth. Characterisation conducted at the site indicates that the surface tailings have moderate bulk density and generally low to moderate resistance to penetration. Although layers of fine, uniformly textured tailings are believed to restrict oxygen diffusion into the tailings at some localities, the surface tailings (0-200 mm) generally appear to be sufficiently aerated to sustain plant growth. Two factors that may be inhibiting plant establishment, however, are the shallow water-table at the eastern end of the dam and the unconsolidated nature of the tailings. The depth to the water-table was to found vary greatly about the dam. Piezometer results indicate that over most of the dam the water-table lies at least a metre below the tailings surface throughout much of the year. At sites 2, 5, 8 and 1 1 , however, the water-table is believed to persist within 50 cm of the tailings' surface during most months (Figure 4 . 1 ) . The close proximity of the water-table to the tailings surface at these sites is of concern as it may lead to prolonged saturation of the plant root zone with potentially acidic and heavy-metal contaminated ground water. Problems arising from evaporative concentration of soluble salts during dry summer spells are envisaged to occur within areas proximal to sites 2, 5, 8 and 1 1 where evaporated pore-water may be quickly replenished via capillary action from shallow groundwater. To aid plant establishment in this region of the tailings dam, steps may have to be taken to either inhibit capillary rise or to substantially reduce the water-table height via open-channel drainage. The 1 76 thixotropic, fine-grained nature of the tailings may prohibit the use of 'traditional' subsurface drainage systems. The unconsolidated nature of the tailings is likely to hinder attempts to revegetate the site by sand-blasting, upheaving andJor burying establishing seedlings. Methods to reduce wind velocity and hence the movement of abrasive particles across the surface of the tailings may be needed to improve seedling establishment. Water erosion will likewise pose a threat to establishing vegetation at the site. The Tui tailings were generally found to exhibit penetration resistances of less than 2000 kPa within the potential root zone of the tailings (0-200 mrn). This finding indicates that the roots of establishing vegetation should be able to adequately explore the root zone over most of the dam's surface. Although there is no clear evidence suggesting the presence of a distinct hardpan in the tailings profile, most sites exhibited indurated zones which generally occurred at one of two depths ( 1 50-200 mm or 350-400 mm). Because at several sites these zones coincided with those of maximum gravimetric water content the results indicate that some incipient cementation of the tailings may be occurring at specific depths . 4.3 Chemical Characterisation 4.3.0 Introduction 1 77 Based on the findings of the physical and preliminary chemical characterisation of the Tui Mine tailings (sections 4 . 1 and 4.2) it is evident that certain chemical properties are primarily responsible for inhibiting plant growth. No major physical limitations were identified, although as discussed in the previous section, the tailings do not provide an optimum growing medium in some areas of the dam due to poor aeration and drainage. One of the primary findings of preliminary characterisation of the Tui mine tailings was that the tailings tended to vary considerably in their chemical composition. To elucidate the spatial variability of the tailings, further chemical characterisation was carried out at each of the 1 2 piezometer/grid sites. The variability of selected tailings' parameters with depth was also investigated at two of these sites. 4.3.1 Sampling LocationslProcedures 4.3 .1 .1 Grid Sampling at Sites 1 to 12 Composite tailings samples were collected from each of the 1 2 grid sampling sites at 0- 1 00 mm and 1 00-200 mm depths for laboratory analyses. Characterisation concentrated mainly on the surface 200 mm of the tailings as it is this region of the profile that is generally most important in terms of seed germination and plant root growth. As discussed in section 4 . 1 and 4.2, the surface tailings were generally found to exhibit moderate to high chroma colours and relatively low 'total ' concentrations of Cu, Fe and Zn, indicating that the tailings within this zone are aerobic and have probably undergone significant oxidation since their deposition. The composite samples (each weighing about 1 .0 kg) were made up of about 1 0 sub­ samples collected within a 2 metre radius of each piezometer with a soil corer. The samples from each site were air-dried, passed through a 2 mm sieve and refrigerated at 4°C prior to thorough mixing and analysis . 4.3.1 .2 Incremental Sampling (i) 50-mm Incremental Samples 1 78 To study the variability of pH, phosphate extractable sulphate and total SUlphur with depth, open pits approximately 1 metre deep were excavated at two randomly chosen sites (site 7 and 9). Sixteen 50-mm incremental samples were obtained from each site for analysis. pH analyses were conducted on an additional 6 samples at site 7 to a depth of 1 200 mm. At the time of incremental sampling four 'bulk' samples were also collected from two depths (0- 1 50 mm and 1 50-300 mm) at each site. These samples each weighed approximately 1 kg and were collected and analysed to assess variability between the bulk and incremental samples. (ii) lOO-mm Incremental Samples On a subsequent visit to the Tui mine site an additional six 1 00-mm incremental samples were collected at site 7 to a depth of 600 mm for the lime incubation trial reported in Chapter 6. To help interpret the results of the incubation trial the samples were characterised to assess the variability of the tailings' Acid Generating Potential (AGP), chemical, physical and mineralogical properties with depth. The excavation pit from which these samples were obtained is shown in Plate 4 . 1 . 4.3.2 Methodology The pH of the tailings were determined in distilled water as per the method outlined in section 4. 1 . 1 . 1 . The total S and phosphate-extractable sulphate (S04=) content of the tailings at sites 7 and 9 were again determined by the methods outlined in Appendices 4(A) and 4(B). A literature review (Chapter 3 .4.2) was undertaken to identify suitable extractants to characterise the tailings and to assess their AGP. The metal concentrations associated with the sulphidic and non-sulphidic fractions of the tailings were estimated from aliquots using Flame Atomic Absorption Spectrometry (F AAS) and are expressed in terms of the original air-dry mass. 1 79 Total and 'plant available' concentrations of selected metals were similarly assessed in the various tailings samples. With the exception of As (arsenic), metal concentrations in the extractants were determined using a GBC 904AA atomic absorption spectrophotometer. In the case of As, concentrations of this element were determined using a hydride generator. Arsine produced by the rapid addition of ca 5 mL of sodium borohydride solution8 to between 0 . 1 and 0.5 mL of sample solution was directly introduced to a GBC 909AA atomic absorption spectrophotometer via a heated silica tube (Robinson, 1 994) . The peak height of the absorption curve was used to measure As concentrations in the samples (Godden & Thomerson, 1 980). The tube was heated to 950 °C using a GBC EHG3000 heating system to provide good atomisation of arsine. With the exception of AI, As and Mn9 an oxidising air-acetylene flame was used to determine the various metals (Table 4.9). Recommended lamp currents, absorption wavelengths and a slit width of 0 .5Jlm were used during the analyses (Athanasopoulos, undated) . Where a wide range of metal concentrations were encountered more than one wavelength was often used to avoid excessive dilution of the samples. The methodology of extraction procedures hence forth discussed in this section are fully reported in Appendices 4(D), 4(E), 4(F) and 4(G). Due to their very fine-grained nature (predominantly <250 Jlm), metal analysis of the tailings was generally carried out on the tailings after 2mm sieving unless otherwise stated. 8 The sodium borohydride solution contained 2% NaBH4 and 1 % KOH. 9 AI, Cd and Mn concentrations are discussed in Chapter 7. Table 4.9 Flame type and lamp current settings generally used for atomic absorption spectrometry in the present study. Metal Flame Type Flame Lamp Current r rnA 1 Al Nitrous oxide-acetylene reducing 1 0.0 As Flameless - 3 .0 Cd Air-acetylene oxidising 3 .0 Cu Air-acetylene oxidising 3 .0 Fe Air-acetylene oxidising 7.0 Pb Air-acetylene oxidising 5 .0 Mn Air-acetylene stoichiometric 5 .0 Zn Air-acetylene oxidising 5 .0 4.3.3 Extractant Selection 1 80 One of the primary objectives considered during the selection of extractants that could be used to estimate the AGP of the Tui tailings was to identifY extractants that would provide equally useful information relating to the 'plant available' and 'total ' metal content of the tailings. In addition, the speed, simplicity and versatility of the extractants and/or extraction procedures was also considered. The TAGP method outlined in section 4.4 required quantification of two forms of Cu and Fe in the tailings; acid generating (sulphidic) and non-acid generating (non­ sulphidic) Cu and Fe. To estimate the ratios of non-sulphidic to sulphidic Cu and Fe, two extractants of different severity were thus required. A literature review undertaken to determine the extractants which were likely to fulfil the afore-mentioned requirements revealed that although a great deal of research has been undertaken investigating micronutrient extractants for soils, comparatively little research has been conducted looking at extractants suitable for estimating the plant-availability of 'trace elements' in metal-rich mining wastes (Chapter 3 , section 3 .4.2). Upon completing review of literature, however, it was decided to use and/or further evaluate the following three acid extractants; concentrated HFIHN03, 25% (v/v) HN03 and O. 1M HCl (Appendices 4(D), 4(E) and 4(F), respectively). The 25% HN03 and HF extractants were compared to investigate the ability of 25% HN03 to assess the 'total ' concentrations of Cu, Fe, Pb and Zn in the tailings. The two 1 8 1 extractants were predicted to release similar heavy metal concentrations because most heavy metals in the tailings were believed to be present in the form of sulphide minerals and thus not bound within silicate minerals (Chapter 3 .4.2). To elucidate the ability of hydrogen peroxide (H202) to oxidise and liberate sulphide­ bound metals several samples were also reacted with this oxidant as per the method outlined in Appendix 4(0). After filtration the resulting supernatants were analysed for Cu, Fe, Pb and Zn. 4.3.3.1 Extractant Comparison The materials upon which the various extractants were assessed included, surface (0- 1 00 mm) and subsurface (500 - 600 mm) Tui mine tailings obtained from grid sampling site 7, two sulphide-bearing rock standards and a sample of pulverised pyrite. Three additional samples were prepared by combining the surface and subsurface tailings at 3 different ratios (7:3, 1 : 1 , 1 :3) to obtain samples containing approximately 20,000, 30,000 and 40,000 mglkg of Fe respectively. The inclusion of these samples was designed to ensure that the data was normally distributed and to provide materials containing a range of metal concentrations comparable to those likely to be found within the upper metre of the tailings profile. 4.3.3.2 Validation of Analytical Results The efficacy of the HF extraction and flame atomic absorption spectrometry (F AAS) procedures were validated by running CANMET10 (MP- 1 a) and CAAS l I (Sulphide Ore- of 1 ) ro�k standards with the tailings samples. Due to the limited availability 'these standards, a 2 kg sample of Tui ore was ground to <1 50 !-Lm and analysed by both ICP­ AES and F AAS techniques for subsequent use as a standard. The certified standards proved to be useful reference material for the standardisation of the Tui Ore material and provided a means of optimising extraction procedures during the initial phases of the research program. 10 Reference data pertaining to this Zinc-Tin-Copper-Lead Ore standard can be obtained from the Canada Centre for Mineral and Energy Technology, 555 Booth Street, Ottawa, Canada. 1 1 Analytical data for the eAAS standard is reported in Sine et al. ( 1 969). 1 82 Analysis of HFIHN03 extracts showed levels similar to the standard values for Cu and Pb in the two standards (Table 4 . 1 0) . Measured Zn concentrations in the CAAS standard were also fOW1d to agree well with the expected value, however, the procedure was found to overestimate the Zn content of the CANMET standard by about 1 2 %. This error is believed to have arisen during dilution of the supernatant which involved a 1 000-fold serial dilution to bring Zn concentrations into the analytical range of the spectrometer. Table 4.1 0 Comparison of average FAAS values with CANMET Ore standard MP- I a and CAAS Sulphide Ore- l rock standard. VNR = values not reported Metal MP-la CAAS Standard HN03 HF Standard HF Cu 1 4400 1 1 778 l 3345 82 1 5 862 1 Fe VNR 56835 5 8292 VNR 279068 Pb 43300 43577 43607 249 233 Zn 1 90200 1 92637 2 1 3439 294 304 Although total Fe values were not reported for either of the standards, HF extraction results indicated that the CANMET and CAAS standards contained about 60,000 and 280,000 mg/kg of Fe respectively and that the 25% HN03 solution extracted about 97 % of the total Fe in the CANMET standard. 4.3.3.3 Assessment of 25% Nitric Acid Extractant The Cu, Pb and Zn content of the MP- l a ore standard was also determined using the HN03 extraction procedure (Appendix 4(E» . This extractant was fOW1d to underestimate Cu levels in the standard, however, Pb and Zn concentrations compared well with recommended values (Table 4 . 1 0). When the concentrations of 25% HN03 extractable Cu, Fe, Pb and Zn were regressed against the concentrations extracted by the HF method, a very highly significant (P 1 . 5 litres) of H202 to fully oxidise the samples. Due to the clear shortcomings of the method only four samples were subjected to hydrogen peroxide oxidation (Table 4 . 1 2) . Where the measured values lie within the bounds of the regression graphs they have been plotted (Figures 4 .8 and 4.9). A linear curve was fitted to the Fe data and indicates that less than 1 0% of the total Fe in the samples was rendered soluble using this method (Figure 4.8b). The concentrations of Cu and Zn liberated into solution were generally found to be much greater than for Fe .....; u (Ij .b 2500 2000 r3 1 500 .... o o 60000 50000 40000 30000 :-g -< 20000 u .;::: ....- Z � If) N 1 0000 o o � I (a) Y = 0 .8798 * X + 22.20 (n=6) r 2 = 0 .99 1 (P Cu. 'Total ' Fe concentrations were thus substantially higher than Cu and Zn concentrations and ranged from about 3 ,000 mg/kg in the uppermost sample to nearly 50,000 mg/kg in the 500-600 mm sample (Figure 4. 1 Oa). Average Cu and Zn concentrations were both found to peak between 400 and 500 mm depth where levels exceeded 2 1 00 and 5400 mg /kg respectively. Like Fe, the abundance of Cu and Zn were lowest in the surface 1 00 rnm of the profile where respective concentrations of approximately 75 and 350 mglkg were obtained (Figures 4. l 1 a). Similar to preliminary results reported in section 4 . 1 .2, the distribution of Pb down the profile was found to contrast with that of Cu, Fe and Zn. Lead tended to be most abundant in the upper half of the profile and was the least variable of the major metals constituting the tailings (Figure 4. 1 Oa). Concentrations of Pb were found to peak at over 1 1 ,000 mg/kg between 200 and 300 mm depth. Although least abundant in the deepest sample (500-600 mm) Pb concentrations were still very high, exceeding 4000 mglkg (Figure 4. 1 Oa). 1 88 4.3.4.2 O.lM HCl Extractable The concentrations of O . IM HCI extractable eu, Fe, Pb and Zn in the 1 00-nun incremental samples were generally much less than the 'total ' concentrations of these metals (Figures 4. 1 Ob and 4. 1 1 b). The amount of Fe, Pb and Zn extracted by O. IM HCI was in most cases less than 1 0% of the total concentrations of these metals in the tailings (Table 4. l 3) . In the case of Cu, however, the proportion of metal extracted by O. I M HCI was generally far greater and increased markedly below 3 00 nun depth. The large proportion of Cu extracted by the relatively weak acid (O. IM Hel) indicates that beneath 300 nun depth more than half the eu in the tailings is in a form that is potentially available to plants, presumably as relatively soluble Cu sulphates l2. The majority of Fe, Pb and Zn in the tailings, however, appears to be non-labile and is likely to be present in the form of largely insoluble sulphates, metal oxides and/or sulphides (Table 4 . 1 3) . Whilst total As levels were not assessed in the RN03 extractants, O. lM Hel extractable As was determined and found to be present in similar concentrations to Fe (Figure 4. l 0b). Table 4.13 Percentage oftotal metal extracted by O. lMHCl in site 7 incremental samples and grid samples. NA = not analysed Depth [mm] Cu % Fe % Pb % Zn % Site 7 0 - 1 00 32 6 1 4 9 1 00 - 200 23 5 8 1 5 200 - 300 1 0 1 5 3 300 - 400 75 3 4 1 3 400 - 500 52 6 4 4 500 - 600 69 3 6 5 Grid Samples 0- 1 00 12 . 8 2 .8 NA NA (range) (4.3 -28.9) ( l .45 - 5 .2) 1 00-200 1 9. 1 6. 1 NA NA (range) (7.2 - 3 1 .7) ( 1 .7 - 35 . 1 ) The distribution of O. IM extractable Cu, Fe Pb an d Zn within the profile were similar to that of the total concentrations (Figures 4. 1 0 & 4 . 1 1 ) . In contrast to Cu, Fe and Zn 12 The absence of organic matter and specifically humus which is responsible for strongly complexing Cu in soils (Alloway, 1 995) undoubtedly increased the availability of this metal in the tailings. o o 1 00 .-J 200 300 � 400 500 600 � o o 400 500 (a) 25% Nitric Acid Extract. Fe & Pb [rng/kg] • i • I • 20000 40000 Lead Iron r (b) O. lM Hel Extract. Fe & Pb [rng/kg] 500 1 000 1 500 • • ' Lead • 1 89 60000 2000 Figure 4.10 (a) 25% Nitric acid extractable Fe and Pb and (b) O . lM Hel extractable As, Fe and Pb versus depth. o 300 � 400 500 � 600 -' 0 0 � 1 00 I I i 200 I : (a) 25% Nitric Acid Extract. Cu & Zn [mglkg] 2000 4000 6000 .� Copper .. � Zinc (b) O. lM HCr Extract. Cu & Zn [mglkg] 500 1 000 1 500 3 00 '� . . . .. � 400 500 , 600 -' .A.. Zinc i' Copper 1 90 Figure 4. 1 1 (a) 25% Nitric acid extractable Cu and Zn and (b) O. IM HCl extractable Cu and Zn versus depth. 1 9 1 concentrations, O . IM HCI extractable Pb concentrations were greatest in the top of the profile where XRD analysis indicated that the majority of Pb was present as highly insoluble angle site (PbS04) (Chapter 5). Despite the high concentrations of PbS04 in the surface tailings, O . IM HCI was found to extract relatively little of the 'total ' Pb in the tailings, indicating that this extractant only partially solubilises angle site (Table 4. 1 1 ). This result is consistent with findings from the H202 oxidation extraction procedure (Chapter 4 .3 . 3 .4) and Eh-pH data which indicates PbS04 is stable at pH levels induced by O . lMHCl (ie. pH � 1 .0) (Brookins, 1 988) . Regression analyses of the Cu fractions identified a highly significant (P 0. 1 -3 .08 P < 0 .01 Cu (total) 0.06 0 .20 3 .45 P < 0.05 -2.24 P < 0.05 Fe (total) 0.01 0. 1 3 1 2 .26 P < 0.00 1 - 1 . 1 9 P > 0 . 1 Pb (total) 0.09 0 . 1 1 1 . 1 8 P > 0. 1 -0. 1 6 P > 0. 1 Zn (total) 0.05 0. 1 57 2 .85 P < 0.05 -2.40 P < 0.05 13 If the variances of two data sets are significantly different one can hypothesis that the two data sets were drawn from different parent populations. 1 96 To assess whether or not the difference in mean metal concentrations between the 0- 1 00 and 1 00-200 mm samples were significantly different Student's t-tests were conducted. Since the t-test is sensitive to variance, the F -test results were used to identify data sets of equal or unequal variance (Brooks, 1 983) . With the exception of Pb the t-tests were conducted on the logarithms of the elemental ' total ' metal concentrations as most of the data was strongly log-normally distributed. The t-test results indicate that there is no significant difference (P > 0. 1 ) between the mean total Fe concentrations in the surface (O- l OOmm) and subsurface ( l 00-200 mm) samples. S imilarly the difference between the mean ' total' Pb level in the surface samples was not significantly different to the mean level in the subsurface samples. Mean concentrations of eu and Zn, however, were found to be significantly different (P < O .OS) between the two sampling depths (Table 4 . 1 S) . The lack of significant difference between the mean Fe concentrations of each depth, is believed to reflect the highly variable Fe content of the tailings, particularly at the 1 00-200 mm sampling depth (Table 4 . 1 S) . Highly significant (PO. l ) . A significant relationship (P0. 1 Fe (Hel) 0 .02 P > 0. 1 Fe (total) 0 .04 P > 0 . 1 Pb (total) 0 . 33 P < 0. 1 Zn (total) 0 .26 P < 0. 1 4.3.5.2 O.lM Hel Extractable Predictably 0. 1M Hel extractable metal concentrations were generally much lower than 2S % HN03 extractable levels (Table 4 . 1 4) . This indicates that the majority of eu and Fe is not in a form readily available to plants. The average concentrations of O . IM Hel extractable Fe in both the 0- 1 00 mm and 1 00-200 mm sampling zone were greater than those of copper. As a percentage of the total Fe in the tailings, however, the concentrations of Fe extracted by O. I M Hel was relatively low in the surface tailings (Table 4. 1 3). Figure 4. 14b indicates that there is no significant relationship between the total and O. I M HCI extractable Fe concentrations in the 1 00-200 mm tailings and only a possibly significant (PO. l ) 5 .0 Log 25% Nitric Acid Extract. Fe [mglkg] Figure 4. 1 4 (a) Log O. IM Hel eu versus Log 'Total' eu and (b) Log O . lM Hel Fe versus Log 'Total' Fe Seperate correlations are shown for individual depths (0- 1 00 mm and 1 00-200 mm). :::l U o o 400 (a) o 250 500 750 100-200 mm -j -- Y = 0 .3 1 8 1 * X - 1 6.95 (n= 1 2) r 2 = 0.99 (PO . 1 ) 'Upper x-axis 0-100 mm Y == 0.0268 * X + 4.544 (n== 12) r 2 = 0.25 (P 600 � � 800 Figure 4.23 Total extractable sulphur versus depth at (a) grid sampling site 7 and (b) grid sampling site 9 . 2 1 1 (a) Sulphate [mg/kg] 0 500 1 000 1 500 2000 0 r � 200 � 1 9 1 7 :;::; I e �---- I ..--- S • ..c: ,/ a. 400 Q) Q Q) t;:::: -' � 0 .... 0... , " 600 �� • • • 800 --II- ---e- (b) Sulphate [mg/kg] 0 500 1 000 1 500 2000 0 , " I �'� S 860 ,,� 200 -" " "'. 3067 e � s ---/�/ � ..c: ...... ." 0.. 400 Q) - �� Q � • t;:::: 0 • ..... 0... / 600 -' /-/ • • • 800 Figure 4.24 Phosphate extractable sulphate versus depth at (a) grid sampling site 7 and (b) grid sampling site 9. Site 7 Site 9 -- - - -- - ------- - - - - - 2 1 2 The tailings at these sites were fOlU1d to exhibit similar pH profiles (Figure 4 .25) that were comparable to those obtained from the 3 preliminary sampling sites (Figure 4.3b). pH values at sites 7 and 9 were fOlU1d to vary from about pH 4 in the uppermost samples to less than 2.5 between 1 50 and 250 nun depth, to approximately 6 below 600 nun depth. Profile pH minima of 2 . 1 4 and 2.45 were obtained for sites 7 and 9 respectively at approximately 200 nun depth. Below this depth the pH of the tailings was found to gradually increase to reach about pH 6 at 600 mm depth. Only a very slight rise in pH was noted further down the profile (Figure 4.25) . 4.3.6.4 pH/Sulphate Correlation When the incremental pH values of each site were plotted against S04 = concentrations, it became evident that the zones of low pH coincided with S04= maxima within the profiles (Figures 4.26). The respective pH minima and S04= maxima were found to occur between 1 50 nun and 250 nun depth at both sites which indicates that the majority of sulphide oxidation may be taking place within this depth zone at both sites. Although regression analysis indicated a possible relationship existed between pH and sot at site 7, no significant relationship between the two parameters was found to exist at site 9. A highly significant (PO. l ) 6 .0 8.0 Y == - 0 .08 1 5 * X + 3 .06 (n== 1 3) r 2 = 0.5 3 1 (PO. I ) .A .A .A ----- .A ��f � f 4.0 6 .0 8.0 pH 2 1 5 Figure 4.27 Log Phosphate extractable sulphate versus pH at (a) grid sampling site 7 and (b) grid sampling site 9 Correlation was conducted both with and without the uppermost 3 surface samples. 2 1 6 (Tan, 1 993). The relative mobility of H+ and S04=, thus explains the relationship between these two parameters and depth. Previous studies investigating the reclamation of pyritic mmmg wastes have also identified l inear relationships between pH and sulphate concentrations (Pulford & Duncan, 1 978 ; Arora et aZ. , 1 979). Pulford & Duncan ( 1 978) used this relationship in an attempt to predict the acid generating potential of coal-mine wastes originating from Central Scotland. Their research indicated that samples from the same site, or samples from different sites of similar age tended to have uniform pH and sulphate levels, and that the method was potentially useful for assessing the relative reactivity of pyrite­ bearing mine wastes. 4.3.6.5 Spatial pH Variation In addition to assessing the variation of pH with depth at sampling sites 7 and 9, duplicate pH determinations were conducted on the surface (0- 1 00 mm) and subsurface ( 1 00-200 mm) samples at each of the 1 2 grid sites. All of the grid samples exhibited acidic pH which ranged from 2.77 to 3 .85 (Table 4. 1 7) . At all bar one of the 1 2 sampling sites, the pH of the surface tailings was greater than that of the subsurface samples. The only site not to demonstrate this trend (site 2) exhibited a pH of 3 .38 in both the surface and subsurface tailings (Figures 4.22). Table 4 .17 Average pH values and range for grid sampling sites 1 to 12 . pH values of the grid samples and bulk samples (bracketed data) are also included for sites 7 and 9. Depth [mm] Average pH Range Site 7 Site 9 0- 1 00 3 .35 ± 0.2 1 2.95 - 3 . 85 2.96 3 .25 (0- 1 50) (2.5 0) (2.44) 1 00-200 3 . 1 7 ± 0.25 2 .76 - 3.73 2.77 3 . 1 6 ( 1 50-300) (2.68) (2. 1 1 ) An F-test conducted on the two groups of data indicated that there was no significance difference between the variance of the surface and subsurface tailings (Table 4. 1 6) . A significant difference (P < 0.05) was noted, however, between the means of the samples 2 1 7 (Table 4. 1 8). This indicates that at the time of sampling the temporal variation i n pH was greater than the spatial variability of the tailings. Table 4.18 Pearson' s correlation coefficient (r), Student's t-test and F-test results based on pH data for 0- 1 00 rom and 1 00-200 mm depths (grid sampling sites 1 to 1 2). Variance F value t value r2 0- 1 00 1 00-200 mm mm 0.05 0 .06 1 .3 3 1 . 88 0.66 P >0. 1 P< 0 .05 P < 0 .0 1 Regression analysis of the surface and subsurface samples indicates that a highly significant positive relationship between surface and subsurface pH exists (Figure 4.47). No such correlation was identified between the sites' pH and 0 . 1M HCl or 25% HN03 acid extractable Cu and Fe in the tailings at either depth. This finding is consistent with previous studies that have identified other factors such as particle grain size, bacterial populations and oxygen accessibility as being the primary factors controlling sulphide oxidation and hence W activity within sulphidic mining wastes (Chapter 3, section 3 .3 .3) . A lack of relationship between pH and the concentrations of O. lM HCl Cu and .th e Fe in the tailings is explained byY relative immobility of H+ compared to Cu2+ and F e2+ which would result in preferential leaching of these metallic cations from the tailings. Although the bulk samples were found to exhibit comparable pH values to the corresponding incremental samples, the grid samples obtained from sites 7 and 9 were notably less acidic than the bulk samples obtained from the corresponding sites (Table 4. 1 7) . The lower pH values obtained from the grid samples is possibly just a result of chemical variability within the tailings dam, however, other factors such as climate induced leaching prior to grid sampling may have reduced sample acidities . 2 1 8 4.3.7 Conclusions o n Chemical Characterisation The 25% (v/v) RN03 procedure of Smith & Bradshaw was compared against an HF based extractant and found to adequately estimate the heavy metal contents of the Tui tailings. The procedure was found to dissolve between about 90 and 1 00% of the total metals within the tailings (Figures 4 . 8 & 4.9). The HN03 extractant was particularly effective at liberating Fe, indicating that a negligible proportion of the Fe is bound within silicate minerals (section 4.3 . 3 .3) . In contrast to the 25% RN03 procedure, the quantity of metal extracted by the trial H202 procedure was found to vary substantially between samples. Following H202 oxidation, comparatively very low levels of Fe and Pb were detected in solution due to the formation of highly insoluble oxidation products (4.3 .3 .4). Characterisation of incremental samples at site 7 identified that 25% RN03 ('total ') Fe levels increased from relatively low concentrations of about 3000 mglkg in the surface tailings (0- 1 00 mm) to about 50,000 mglkg at 500-600 mm depth (Figure 4 . l 0a) . Copper and Zn levels were much lower than Fe concentrations in the surface tailings but like Fe increased substantially down the profile to reach levels of approximately 2 1 00 and 5400 mglkg respectively at 400-500 mm depth (Figure 4. 1 1 a). The distribution of Pb in the tailings was found to contrast those of Cu, Fe and Zn in that it was found in slightly higher total concentrations near the surface of the profile. Pb concentrations were extremely elevated in the tailings at site 7, ranging from about 1 1 ,000 mglkg to 4000 mglkg (Figure 4 . 1 Oa). With the exception of Cu, O. lMHCl was generally found to extract less than 1 0% of the 'total ' metals in the tailings indicating that metals are predominantly in the form of sulphide minerals, oxides or highly insoluble sulphates. Below 300 mm depth >50% of the 'total' Cu was extracted by the dilute acid extractant indicating the presences of high concentrations of labile Cu. Like 'total' metal concentrations, O . IM HCI extractable metal levels were generally found to increase with depth at site 7 (Figures l Ob & 1 1 b). Interpretation of grid sampling site data indicates that the positive correlation between 'total' and 0 . 1M HCI extractable Cu, Fe and Zn concentrations with depth reflects translocation of soluble sulphide weathering products down the profile from where sulphide oxidation (depletion) is predominantly occurring (0-300 mm). The contrasting 2 1 9 trend observed for 0 . 1M HCI extractable Pb concentrations i s believed to reflect the immobility of PbS04 and the resulting higher concentrations of this sparingly soluble mineral near the surface of the tailings (Figure l Oa). Like Cu, Fe and Zn concentrations, total S and sot levels were similarly found to increase with depth (Figures 4.23 & 4.24) . Total S levels were generally considerably higher than those of S04 � indicating that S is present predominantly as sulphides. Observed Fe and total S concentration trends at site 7 (characterised by a sharp increase in these elements below 200 mm depth) indicate that FeS2 is by the far the most prevalent sulphide mineral in the Tui tailings. A positive and significant correlation between pH and S04 = was identified at site 7 and 9. These parameters were used to identify what was postulated to be a zone of maximum sulphide oxidation between 200-300 mm depth where, very low pH values were found to coincide with maximum sulphate levels (Figures 4.26) . Characterisation of tailings samples obtained from 0- 1 00 mm and 1 00-200 mm depths at the 1 2 grid sampling sites indicated that metal variability amongst the sites was high (Figures 4. 1 8 to 4.2 1 ) . Average Cu, Fe, Pb and Zn concentrations in the surface samples were approximately 70, 5200, 1 1 900, 260 mglkg respectively. Much greater concentrations were, however, identified at some sites. Total Pb concentrations, for example, were found to be in excess of 27,000 mglkg at site 4 (Figure 4.2 1 ) . Without exception average 'total' metal levels were higher in the subsurface tailings than the surface samples and tended to be much more variable. Average Cu, Fe, Pb and Zn concentrations in the 1 00-200 mm zone were approximately 1 40 , 6800, 1 2400 and 500 mgikg. With the exception of Fe and Pb, these concentrations were significantly higher than those found in the surface tailings. O. lM HCI extractable Cu and Fe levels were similarly present at higher concentrations in the subsurface tailings. The average concentrations of ' labile' Cu and Fe in the surface tailings reached about 40 and 350 mglkg respectively in the surface tailings and about 3 1 0 and 770 in the subsurface tailings (Figures 4. 1 6 & 4. 1 7) . It is postulated that the differences (trends) observed between ' total' and O . lM HCI extractable metal 220 concentrations in the surface and subsurface grid samples reflects that the surface tailings at the site have undergone more extensive oxidation than subsurface samples. Presumably the better aeration and warmer conditions prevailing in the surface samples have promoted biochemical oxidation of sulphide minerals within the surface tailings. With the exception of Pb, metals are postulated to have been extensively leached from the surface tailings as indicated by the generally significantly lower concentrations in which they are found in the surface 1 00 mm of the profile. 4.4 Acid Generation Potential 4.4.0 Introduction The importance of assessing the net Acid Generating Potential (AGP) of sulphidic mine wastes is widely recognised, as this variable dictates the materials ' plant growth characteristics and strongly influences mine site drainage quality (Chapter 3 , section 3 .4) . Prior to this study very little was known about the extent and depth to which the Tui tailings had undergone oxidation and their potential to generate further acidity via oxidation. Considering that acid generation data is often used to estimate liming requirements of sulphidic material, knowledge about how the AGP of a material varies with depth has important implications for the incorporation of liming agents. As discussed in Chapter 3 (section 3 .4 . 1 ) there are several approaches to estimating the AGP of a material. Due to the perceived shortcomings of these traditional methods at estimating the AGP of polysulphidic materials, an alternative method was used to assess the AGP of the Tui tailings. The TAGp14 method is based largely on the acid-base accounting method described by Smith & Sobek ( 1 978) and utilises the stoichiometry of pyrite and chalcopyrite oxidation outlined by Bruynesteyn & Hackl ( 1 982). Although it was specifically designed to assess polysulphidic materials, containing two or more acid generating sulphide minerals, it is considered equally useful for determining the AGP of pyritic (monosulphidic) materials such as coal spoils. This section describes the devised method and compares the results with those obtained from a distinctly different method of AGP assessment based on hydrogen peroxide 14 For the purposes of this study the devised AGP prediction method based on the tailings Cu and Fe content will be referred to as the T AGP (Theoretical Acid Generating Potential) method 22 1 oxidation (Miller et al. , 1 994). The AGPHPOl5 method involved the use of hydrogen peroxide (H202) to rapidly oxidise remnant sulphides in the tailings. Following oxidation the AGP of the tailings was then estimated by back titrating the solution with a standardised base to calculate the amount of acid liberated during the reaction, per given quantity of material (Appendix 4(1) . 4.4.1 TAGP Method: Theory & Methodology The T AGP prediction method is based on the principles used to assess the AGP of pyritic spoil in ABAl6 (Chapter 3, section 3 .4. 1 . 1 ) . The TAGP method is similar to ABA methods in that it uses stoichiometry to estimate the AGP of a given material. The TAGP method differs from traditional ABA methods in that:- ( 1 ) it enables the acid produced from non-pyritic sulphide minerals to be determined and (2) the material ' s AGP is estimated from sulphidic metal concentrations as opposed to sulphur concentrations. The choice to base AGP estimates on metal concentrations rather than SUlphur concentrations was made due to the occurrence of what were believed to be appreciable concentrations of galena (PbS) and sphalerite (ZnS) in the Tui Mine tailings. Although these sulphide minerals undergo oxidisation, the reaction is not believed to liberate significant amounts of acid (Bruynesteyn & Hackl, 1 982; Kwong, 1 995). Literature review and preliminary characterisation of the Tui mine tailings (Chapter 4.2) indicate that pyrite (FeS2) and chalcopyrite (CuFeS2) are the two primary sources of acid in the Tui mine tailings. Thus, the determination of the sulphidic iron (Fe) and copper (Cu) content of the tailings enabled the relative concentrations of acid generating sulphide minerals to be determined. Sulphidic sulphur associated with the non-acidifYing sulphides (galena and sphalerite) was effectively eliminated from the TAGP calculations. The oxidative dissolution reactions of pyrite and chalcopyrite were used to stoichiometrically estimate an upper and lower TAGP value for the tailings (Bruynesteyn & Hackl, 1 982). 15 The term AGPHPO will, for the purposes of this study, be used to denote the Hydrogen Peroxide Oxidation Method of Miller et ai., 1 984. 222 The upper limit (a worst case scenario) assumes that ferric iron (Fe3+) completely hydrolyses to form ferrihydrite, whereas the lower limit assumes that all of the Fe is precipitated as hydronium-jarosite (Bruynesteyn & Hackl, 1 982). The latter reaction liberates a comparatively low rate of acid (Table 4. 1 9) . If complete oxidation of the sulphide minerals occurs, the amount of acid produced in reality is likely to fall somewhere between the upper and lower limits (Chapter 3 , section 3 .3 .2). Lime amendment rates, however, are often based on the upper limit as this provides the maximum potential oxidation (Bruynesteyn & Hackl, 1 982). The sulphidic Fe and Cu content of the tailings was estimated using two extractants; O. 1M HCl and 25% (v/v) HN03 . As discussed in chapter 4.2, one of the primary objectives borne in mind during the selection of these extractants was that they would not only satisfactorily estimate the AGP of the tailings but also provide useful information pertaining to the material' s plant growth characteristics. The speed, safety and simplicity of the extractants and/or extraction procedures were also considered. The Cu and Fe extracted by the O . lM HCl was assumed to be non-sulphidic and therefore non-acid generating. The 'total' Cu and Fe content of the tailings was estimated using 25% HN03. By assuming the majority of Fe and Cu in the tailing was present as pyrite, chalcopyrite and/or as occluded or soluble sulphate salts, it was postulated that the sulphidic Fe and Cu content of the tailings could be satisfactorily predicted as follows; Sulphidic Fe & Cu = Fe & Cu (25% HN03) - Fe & Cu (O. 1MHCI) [Equation 4. l ] If it is assumed that eu associates with the Fe and sulphur (S) in a ratio of 1 : 1 :2 to form chalcopyrite (CuFeS2), and that Fe associates with S in a ratio of 1 :2 to form pyrite (FeS2), the results obtained from equation ( 1 ) above can be used to stoichiometrically calculate the amount of p)Tite and chalcopyrite in the tailings. The T AGP of the tailings was then estimated using the upper and lower stoichiometric values in Table 4. 1 9. [6 ABA = Acid Base Accounting (see chapter 3, section 3.4. 1 . 1 for further discussion). Table 4 . 1 9 Acid potentially released from the oxidation of 1 .0 mole of chalcopyrite and pyrite (Bruynesteyn & Hackl, 19:::2). Sulphide Lower Limit Upper Limit [moles H2SO4) [moles H2SO4] Pyrite 1 .3 3 2 . 0 Chalcopyrite 0.33 1 .0 223 To calculate the TAGP of the Tui mine tailings in this fashion several assumptions have to be made:- ( 1 ) It is assumed that the entire sulphidic Cu content of the tailings is present as 'pure' chalcopyrite and that based on empirical formula the Fe component of chalcopyrite is associated with Cu in a simple atomic weight ratio of 0 .879: 1 .000 (ie . for every 1 g ofCu, 0 .879 g of Fe is complexed in the chalcopyrite molecule). Therefore the pyritic Fe content of the tailings can be calculated as follows, Pyritic Fe = Fe[mglkg] - (Cu[mg/kg] * 0 .879). (2) The HN03 acid digestion does not render significant quantities of Fe and Cu associated with the silicate lattice layers soluble. ie. any Fe and Cu that is released from this phase is likely to be negligible. (3) The 0 . 1M HCl extracts predominantly Cu and Fe associated with sulphate. The method for determining the sulphidic Fe and Cu content of the Tui Mine tailings is summarised below:- ( 1 ) The tailings samples were air-dried and sieved to < 2mm. (2) The non-sulphidic Cu and Fe content of the tailings was extracted using O. lMHCl as per the method outlined in Appendix 4(F). (3) Total eu and Fe concentrations were extracted with 25% VN nitric acid as per the method outlined in Appendix 4(E). (4) The sulphidic Cu and Fe content of the tailings was then estimated using Equation 4. 1 above. Appendix 4(H) includes working examples of the devised method for samples containing 1 000 mglkg Cu, 1 000 mg/kg Fe, and a mixed sample containing 1 000 mglkg Cu and 1 0,000 mg/kg Fe as sulphides. The examples demonstrate how the sulphidic Fe and Cu concentrations are used to determine the acid generation potential of the tailings. 224 4.4.2 Acid Generating Potential Results The TAGP of the tailings was assessed at each of the twelve sampling sites to enable the liming requirements of the surface 200 mm of the tailings profile to be( Analysis was conducted on two samples from each site obtained from 0 - 1 00 mm and 1 00-200 mm depths (section 4.3 . 1 . 1 ) . TAGP results based on Cu and Fe metal analysis data presented in Figures 4. 1 6-4. 1 9, indicate that the TAGP of the surface tailings (0- 1 00 rum) is less variable between sites and generally lower than that of tailings obtained from 1 00 to 200 mm depth (Table 4.20). Upper limit TAGP values (based on the complete hydrolysis of Fe3+), for example, were found to range from 1 2.2 to 30 .7 Mg CaC03/ 1000 Mg of tailings in the surface tailings, where as the range in the 1 00 to 200 mm depth zone was far greater, varying from 3 .6 to 79.6 Mg CaC0311 000 Mg of tailings. Not all of the sites, however, were found to exhibit greater TAGP at depth. The surface tailings at sites 2, 6, 1 0 and 1 2 were found to exhibit higher T AGP values than the sub­ surface tailings at the corresponding sites . This is postulated to reflect the occurrence of isolated layers of highly sulphidic tailings at or near the tailings surface as discussed in Chapter 4.2.2. Table 4.20 AGP data [Mg CaC0311 000 Mg of tailings] based on the assessment of sites 1 - 12 . Presented data is based on 48 ana1ys(�s conducted at sampling sites 1 to 1 2 . Depth Average AGP Average AGP Range Range [mm] Lower Limit Upper Limit Lower Upper Limit Limit 0- 1 00 1 2 .4 ± 3 .2 1 8 .7 :± 4 .8 8. 1 - 2004 1 2 .2 - 30 .7 1 00-200 2 1 .0 ± 1 5 .0 3 1 . 8 :±: 22.7 2 .3 - 52.5 3 .6 - 79.6 Spatial representation of the AGP values indicates that there are no obvious trends over the surface of the dam (Figure 4.28) . This suggests that the prolonged saturation of the tailings at sites 1 , 2, 8 , and 1 1 has had little influence on the oxidation of acid generating sulphide minerals at these sites. 225 (a) (b) Figure 4.28 Tailings AGP [Mg CaC03 / 1000 Mg of tailings] in (a) 0- 100 nun zone and (b) 1 00-200 mm zone Assumes complete hydrolysis of ferric iron to ferrihydrite. 226 Table 4.20 indicates that complete oxidation of 1 000 Mg of the surface tailings (0- 1 00 mm) at the Tui mine site would l iberate an an10unt of acid that would require between 12.4 and 1 8 .7 Mg of CaC03 to neutralise it. Assuming a bulk density of 1 .3 Mg/m3 and a 1 00 mm liming depth, these rates are equivalent to and Mg CaC03 per hectare respectively. Similarly, neutralisation of the acidity potentially produced by the sub-surface tailings ( 1 00-200 mm) would require the addition of lime at a rate of between and Mg of CaC03 per hectare. If the 'upper' and ' lower' limit values for each sampling depth are averaged and combined, an TAGP value of 20. 1 MgCaC03!1 000 Mg tailings is attained for the surface 200 mm of the tailings. Based on this value and neutralisation of the tailings to 200 mm depth, it is postulated that a lime application rate in the vicinity of 50 Mg/ha would prevent tailings re-acidification. Although very high by agricultural standards, liming rates such as these are commonly exceeded in mine reclamation schemes abroad (DoUhopf, 1 992). The relative contribution of pyrite and chalcopyrite to the Tui tailings' AGP was estimated from the concentrations of Fe and eu respectively in the extractions. Pyrite was found to be by far the major contributor to the tailings' AGP, generally accounting for over 99% of the acidity potentially produced at each site (Table 4.2 1 ). Although as discussed in Chapter 3 (section 3 .3 .2) chalcopyrite releases comparatively lower rates of acid upon oxidation than pyrite, the primary reason for the much greater influence of pyrite on the T AGP of the tailings is due simply to its presence in far higher concentrations. Much of the Cu in the tailings was readily extracted with O . lMHCl and was thus deemed to be non-sulphidic and hence non-acid generating (Figure 4. 1 1 ) . Whilst the pyrite content of the tailings varied from 0.2 1% to as much as 4 .7 1% in the upper 200mm of the profile, estimated chalcopyrite concentrations were generally far lower, ranging from less than 0 . 0 1% to about 0 .2%. 227 Table 4.21 Average chalcopyrite and pyrite concentrations in the surface 200 rnm of the Tui Mine tailings. t Refers to the percentage of the tailings TAGP attributable to pyrite oxidation. Presented data is based on 48 analyses conducted at sampling sites 1 to 12. Chalcopyrite PYrite Pyritet Sulphide Ratio (%] r%l Contribution rCu/Fe*IOOl Average 0.04 ± 0.04 L50 ± 1 .04 99. 1 1 ± 1 . 14 2.08 ± 2.77 Minimum 0.01 0.2 1 93 . 8 1 0.54 Maximum 0.20 4.7 1 99.76 1 5 .02 Although the TAGP attributable to chalcopyrite was negligible at most grid sampling sites, results from site 2 suggest that about 6% of the potential acidity is ' stored' as chalcopyrite in the 1 00-200 mm sample. Samples from this site account for the high maximum sulphide ratio and large standard deviation for this parameter which was otherwise relatively constant from site to site (Table 4.2 1 ). 4.4.2. 1 AGP of IOO-mm Incremental Samples To elucidate the results presented in sections 4. 1 .2 .5 which indicated that the AGP of the Tui tailings increased with depth, the TAGP of the six- l OO mm incremental samples obtained from site 7 was assessed. TAGP results based on Cu and Fe metal analyses presented in Figures 4 . 1 0 and 4 . 1 1 confirm that the TAGP of the tailings tends to increase with depth and indicate that an almost ten-fold increase occurs below 200 mm (Figure 4 .29). Estimates of the TAGP of the surface tailings' (0-200mm) indicate that between 7.4 and 1 1 .2 Mg CaC0311 000 Mg of tailings would be required to neutralise the acid potentially produced from the oxidation of pyrite and chalcopyrite within this zone. The composite sample (0- 1 00 rum) obtained at grid sampling site 7 was found to exhibit comparable but slightly higher TAGP values ( 1 5 .8 and 1 8 .2 Mg CaC03/1 000 Mg of tailings). Due to the comparatively low concentrations of acid generating sulphide minerals in the surface tailings, the difference between the upper and lower T AGP limits appears relatively small (Figure 4 .29). As the concentrations of pyrite and chalcopyrite increase with depth, however, the difference becomes more apparent. At 500-600 mm depth, for e .§. ..c:: .... c.. Q.I � Q.I iE 0 - � o 1 00 200 300 400 o x a. ; � 500 - 600 228 Calculated Lime Requirements [Mg CaC03/1000 Mg tailings] 50 1 00 1 \ !l� I • EE sB • � x 1 50 200 Liming Requirements 1 pH-6 H PO-1 50 HPO-2000 -)(- Sulphide-S -e- TAGP-Low -II- T AGP- High Figure 4.29 Liming requirements of the Tui tailings versus depth at site 7 as determined by: 48 hour pH buffering incubation (J pH-6), hydrogen peroxide oxidation at I SO microns (HP-l SO) and 2000 microns (HP-2000), stoichiometry based on sulphur analysis (Sulphide-S) and Fe/eu analysis (TAGP). Upper and lower limits for the latter method are specified by the suffixes 'High' and 'Low' respectively. 229 example the lower and upper AGP estimates differ by more than 58 Mg CaC03/1 000 Mg of tailings. This difference is equivalent to about 75 Mg ofCaC03 per hectare. Analysis of the incremental san1ples indicates that between 98.5 and 99.5% of the potential acidity of the tailings is associated with pyrite. Chalcopyrite, in contrast is thought to contribute less than 2% of the ' stored' acidity of the tailings. Pyrite was found to constitute approximately 0.59% of the tailings in the upper 1 00 mm of the profile and about 1 0 . 1 3 % at 500-600 mm depth. At corresponding depths, chalcopyrite constituted just 0 .02% and 0 .38% of the tailings (Table 4.22). Table 4.22 Sulphide content and metal ratios of tailings from Site 7 estimated from Fe and eu extractions. Depth Chalcopyrite Pyrite Sulphide [mm] [%] [%] Ratio rCuiFe*1001 0- 1 00 0.02 0 .59 2 .5 1 00-200 0.03 0 .73 3 .4 200-300 0 . 1 9 6.76 2 .9 300-400 0. 1 1 6 .8 1 1 .5 400-500 0.30 6 .39 4 .8 500-600 0.38 1 0. 1 3 3 .7 Although pyritic Cu and Fe concentrations were found to vary substantially with depth, the relative proportions of these minerals remained fairly constant at each of the sampling depths (Table 4 .22). The sulphide ratios obtained from this site are similar to those obtained from the composite samples (Table 4.2 1 ), however, they contrast with results in Chapter 5 which suggest that sulphidic Cu has been preferentially leached from the surface tailings (0-200 mm). To demonstrate the extent to which traditional soil liming requirement methods can severely underestimate the liming requirements of sulphidic material, included in Figure 4 .29 are calcium carbonate equivalents for each of the incremental samples based on the pH buffering capacity of the tailings (Chapter 6.2) . Whilst the amount of lime required to raise the pH of the tailings to 6.0 is seen to increase with depth, as the pH buffering capacity of the tailings improves, the increase is negligible in comparison to substantial 230 increase in the tailings TAGP (Figure 4 .29). The tendency for pH buffering methods to grossly underestimate the liming requirement of sulphide bearing materials is further discussed in Chapter 6 . 4.4.2.2 Comparison of AGP Results For comparative purposes, the AGP of the six incremental samples were also assessed using two alternative methods; ( 1 ) sulphide sulphur analysis (Chapter 4 . 1 ) and (2) hydrogen peroxide oxidation (Miller et ai. , 1 994), The AGP values stoichiometrically estimated from the sulphur data (as per the method outlined in Appendix 4(C) were found to correlate well with the upper limit values of the TAGP method based on Fe and Cu content (Figure 4.29) . Strong correlation between these two methods reflects the fact that pyrite is by far the most abundant sulphide mineral in the tailings and indicates that the AGP of the Tui tailings may have been adequately predicted by determining the sulphidic sUlphur content of the tailings. The occurrence of non-acidifYing sulphide minerals , galena and sphalerite in the tailings does not appear to have resulted in this method overestimating the AGP of the tailings as was predicted (Chapter 4 . 1 ). The AGPHPO method of Miller et al. ( 1 994) involved the addition of 1 5% hydrogen peroxide (H202) to rapidly oxidise the sulphide minerals in the tailings (Appendix 4(1)). The AGP of the tailings was then estimated by back titrating the solution with a standardised base to calculate the amount of acid l iberated during the reaction per given quantity of tailings (see Appendix 4(J) for similar methodology). Initially, the AGPHPo method was found to produce comparatively low AGP values in all but the two uppermost samples (0- 1 00 mm and 1 00-200 mm). More comparable values were attained for the sub-surface samples by increasing the H202 to tailings ratio . This was achieved by decreasing the tailings sample mass to 1 .Og as suggested by Miller et ai. ( 1 994). A sample mass of2.5 g was retained for the 0- 1 00 mm and 1 00-200mm samples as the initial results indicated that the oxidising capacity of the H202 was sufficient to fully oxidise these samples which contained comparatively low concentrations of pyrite and chalcopyrite. 23 1 After modification of the H202 : tailings ratio, the results of the AGPHPO method were also found to correlate well with those obtained from the T AGP method and confirmed that the tailings' AGP markedly increases below 200 rom depth (Figure 4.29). AGPHPO values were generally found to lie between the upper and lower limits of the TAGP method. A notable difference was evident, however, in the 500-600 rom sample. At this depth the AGPHPo values were found to be much lower than those obtained from the TAGP method (Figure 4 .29). Whilst a number of factors are known to influence the analytical results obtained from AGP procedures based on HzOz oxidation (Finkelman & Giffin, 1 986), it was considered beyond the scope of this study to isolate the exact cause of the disparity between the two methods. A noticeable increase in the sulphidic sulphur content of the tailings (Figure 4.29), however, indicates that the AGPHPO method is likely to have underestimated the AGP of the 500-600 rom sample due presumably to the exhaustion of added H202 resulting in incomplete oxidation of pyrite and chalcopyrite. The grain size of the samples was also found to affect the AGP of the tailings when assessed using the AGPHPO method. Grinding the tailings to < 1 50 !-lm was found to appreciably increase the AGP of 4 of the 6 incremental samples (Figure 4.29). No significant difference, however, was noted in the sulphide-depleted surface samples (0- l OO and 1 00-200 mm samples). These results are consistent with findings of previous studies which have identified surface area (grain size) as one of the dominant factors effecting pyrite reactivity with hydrogen peroxide (O'Shay et ai., 1 990; Finkelman & Giffin, 1 986). In contrast to the AGPHPO method, the grain size of the tailings was not found to influence the AGP of the tailings when estimated using the method based on metal content (TAGP method). This result is readily explained by the comparatively severe nature of25% HN03 and its ability to more effectively oxidise pyrite and chalcopyrite. 4.4.3 Neutralisation Potential The Neutralisation Potential (NP) of a material is defined as the ability of a material to neutralise strong acid (Sengupta, 1 993). The inherent neutralisation potential of a material is related primarily to the amounts of alkaline carbonates, exchangeable bases, 232 and weatherable silicates that it contains. The most effective minerals for neutralising acid are the carbonate minerals such as calcite (CaC03), magnesite (MgC03), and dolomite (CaMg(C03)2) (Lapakko, 1 994). Calcite, however, is usually by far the greatest contributor to the NP of mine wastes due to its common occurrence, relatively high rate of reaction and high neutralising capacity (Dent, 1 986) . The most commonly used methods to assess the NP of mine wastes are static procedures which involve reacting a sample with a known amount of standardised acid, heating the solution to assure complete dissolution of the bases and back-titrating with a standard base (Sobek et at. , 1 987). Although several variations exist (Smith et ai., 1 974; Sobek et at., 1 978 ; Ritcey, 1 989) most of these procedures are based largely on the method of Jackson ( 1 958) which was initially designed to assess the neutralising potential of agricultural limestone (Costigan et ai., 1 98 1 ). The method involves heating the sample in an acid solution to ensure complete dissolution of the bases which allows the maximum neutralisation potential to be measured including that contributed by silicates and dolomite (Sobek et al., 1 987). Calculation of the proportion of added acid that has been neutralised by the sample's base components allows for the ready conversion to CaC03 equivalents. The NP of a material is then combined with its AGP to estimate its NAGP (Net Acid Generating Potential). 4.4.3.1 Neutralising Potential Assessment Due to the very low pH of the tailings, the high quartz component and absence of significant quantities of Ca (Table 4.2), the neutralising potential of the various Tui tailings samples was expected to be extremely low. Although small quantities of the clay mineral kaolinite were detected in the tailings by XRD analyses (Chapter 5), this mineral has low ion-exchange capacity and therefore a low ability to buffer pH change. An initial assessment of the neutral ising potential of the tailings was conducted by reacting l .Og of tailings with 50 mL of standardised 0 . 1 0M HCl at 8 0 °C for approximately 1,12 an hour. The solution was allowed to cool and the excess acid was then titrated to the end point of phenolphthalein (PH = 8 .3 ) using a standardised 0 . 1 0M NaOH solution. pH 8 .3 was used to ensure that all the Fe2+ produced from neutralisation by siderite (if present) would be oxidised to Fe3+ (Costigan et at., 1 98 1 ). 233 Problems arose with this method due to the formation of ferric and ferrous hydroxides which masked the end-point colour change. At low pH values (PH ::; 2.3) ferric ion (Fe3+) is stable and remains in solution coordinated with 6 water molecules as the following complex; It was found as the pH of the solution was raised much above pH 2.5, by the addition NaOH, colloidal gels began to precipitate out of solution. Ferric iron is renowned for its tendency to hydrolyse and/or form complexes and even at pH values of about 2.5 - 3 . 0 the extent of hydrolysis is substantial (Cotton & Wilkinson, 1 972). The binuclear species shown in Figure 4.30 is thought to form at these relatively low pH values, however, at higher pH more highly condensed species are · formed resulting in the precipitation of colloidal gels and ultimately hydrous ferrous oxides (Fe(OH)3) . Figure 4.30 Binuclear Fe complex (source Cotton & Wilkinson, 1 972) As the Fe3+ complex is hydrolysed, hydrogen ions are released into solution which act to buffer pH change as they react with the NaOH titrate. The precipitation of Fe3+ and the subsequent liberation of H+ is an integral reaction in the overall process of acid mine drainage (Sengupta, 1 993) . The colour change of phenolphthalein from colourless to pink proved impossible to detect due to the large amount of iron hydroxides present in the solution at pH values in the range of the 234 titration endpoint (8 .3) . Thus, the formation of the iron hydroxide was not only found to hinder the visual detection of the end-point but also tended to buffer pH change by liberating H+ into solution. Whilst the overall effect of this buffering reaction may be negligible if Fe3+ or Fe2+ concentrations are low, the effect of the reaction may result in an underestimation of the samples neutralising potential if high concentrations of these ions are present in solution. To avoid the problem of the endpoint of the titration being obscured by colloidal precipitates the method of Sobek et al. ( 1978) was alternatively used to assess the NP of the Tui Mine tailings. An endpoint of 7.0 was used and attained using a pH meter as opposed to an indicator (Appendix 4(1)). The lower endpoint pH and use of pH meter allowed the endpoint of the reaction to be accurately assessed and was postulated to decrease the effects of the buffering action of the iron hydroxide complexes. Both Hel and water titrations were performed in triplicate, and the volume results were averaged and converted to molar units. 4.4.3.2 Neutralising Potential Results The NP of two incremental samples from site 7 were assessed (0- 1 00 and 500-600 nun) using the method outlined in Appendix 4(J). Both samples gave slightly negative neutralising potential results indicating that the NP of the tailings samples was negligible and that the samples were liberating small amounts of acid during the reaction. Table 4.23 Neutralising Potential of sampling site 7 tailings. The results are the averages of 4 replicate analyses conducted on each sample. Sample Moles of HCI Moles of NaOH Moles of NaOH N eutralising Depth initially consumed by consumed by Potential (mm] added HCI sample blank (Mg CaC03/ 1 000 Mg) 0-1 00 4.97 x 1 0-3 5 . 1 35 x 1 0-3 7 .0 X 1 0-) -2.38 500-600 4 .97 x 1 0-3 5 .244 x 1 0-3 1 .68 X 1 0-4 -2.65 235 The negative NP results are postulated to have occurred as a result of H+ liberation due to the hydrolysis of Fe3+ and Fe2+ complexes as previously discussed. The titration of the tailings samples involved raising the pH of the solution to 7 .0 . This value exceeds the initial pH of the tailings and is sufficiently high to cause mass precipitation of colloidal gels . Whilst clearly erroneous results were obtained using the NP titration method of Sobek et al. (1 978), the results confirmed that the Tui tailings samples have negligible ability to neutralise acid produced from the oxidation of sulphide minerals. The inability of the tailings to neutralise significant amounts of liberated acid stems from the fact that they contain negligible quantities of alkaline carbonates, organic matter, exchangeable bases and high CEC clays. Because the Tui mine tailings have negligible NP the net acid generating potential of the Tui tailings can be adequately estimated from the materials' ability to produce acid. 4.4.4 Conclusions on AGP Characterisation Analyses of incremental tailings samples at site 7 indicates that the AGP of the tailing dramatically increases below 200mm depth (Figure 4.29). Whilst predicted TAGP values were not considered to be particularly high within the surface 200 mm of the profile at this site, very high TAGP values (> 1 00 Mg CaC031 l OOO Mg of tailings) were encountered in the subsurface tailings. Predicted AGP results obtained using the hydrogen peroxide method of Miller et al., ( 1 994) and sulphur analysis were found to agree very well with stoichiometrically estimated values obtained from Fe and Cu data (T AGP values) and confirm that the Tui tailings have considerable potential to generate acidity below 200 mm depth. In accordance with site 7 data, T AGP values in the grid samples (sites 1 to 1 2) were generally found to increase with depth (Figure 4.28). By averaging the upper and lower T AGP values over the 200 mm depth an average T AGP value of about 20 Mg CaC03/ 1 000 Mg of tailings was obtained. Using a tailings bulk density value of 1 .3 Mg/m3 and a liming depth of 200 mm this equates to a liming rate of about 50 Mg per hectare. Incorporation of this rate of lime into the surface 200 mm of the tailings should 236 in theory be sufficient to prevent long term reacidification of the tailings by neutralising acid generated by pyrite and chalcopyrite oxidation. Based on Fe stoichiometry it was estimated that the pyrite content of the surface tailings (0-200 mm) ranges from about 0.2% to 4.7%. Chalcopyrite concentrations in the tailing were estimated to be comparably much lower (0.0 1 % to 2%). These estimates indicate that pyrite is by far the primary source of potential acidity in the Tui tailings. The very high proportion of pyrite in the tailings explains the close approximation of upper limit TAGP values using predicted sulphide S data (Figure 4.29). Estimation of the Tui tailings Neutralising Potential (NP) based on methods commonly used to estimate the NP of colliery spoil proved problematic due to the precipitation of Fe2+ and Fe3+ hydroxide complexes and concomitant liberation of H+ during Fe hydrolysis. The results, however, combined with XRD analysis (Chapter 5) and ICP­ AES analysis confirmed that the NP of the Tui tailings is negligible and need not be incorporated in net AGP calculations. The TAGP method based upon Cu and Fe stoichiometry proved to be a useful and rapid method to characterise the Tui tailings. Not only did the method provide a useful indication of the materials ability to produce acidity but also yielded important information about the quantities and potential availability of Cu and Fe in the tailings. In this respect the TAGP method is clearly superior to those methods based on sulphur analysis or hydrogen peroxide oxidation. 4.5 Chapter 4 Summary & Conclusions • Characterisation of the Tui mine tailings indicates that they do not exhibit any obvious physical traits likely to severely inhibit the establishment of vegetation. The surface tailings (0-200 mm) with their moderate bulk density, low to moderate resistance to penetration and predominantly fine sand to silt-sized texture are considered to be an adequate rooting medium. The upward migration of contaminated pore water (via capillary rise) and the presence of a very high water table at the eastern end of the dam are, however, potential problems that may hinder the establishment of plants should attempts be made to revegetate the tailings without 237 the emplacement of capping materials. The tailings' lack of cohesion may similarly hinder attempts to revegetate the site. Surface erosion may damage, upheave or even bury emergent seedlings during severe wind or rain storms. • In terms of their chemical characteristics the Tui tailings are considered to be an extremely hostile and chemically variable potential plant growth medium. ICP-AES and F AAS analysis indicates that they contain elevated concentrations of several potentially phytotoxic elements. At several sampling sites, for example, Pb was found to constitute more than 2% of the tailings by weight. Whilst acidic conditions and very high Pb concentrations prevail in the surface tailings at most sites, they contain comparably low concentrations of most other heavy metal elements. Likewise the AGP of the surface tailings as determined by static tests was relatively low. • Characterisation of subsurface tailings (>200 mm depth) in contrast, indicated that they contain very high concentrations of metals primarily as sulphides and have a very high capacity to generate acidity via sulphide oxidation. The large increase in AGP below 200 mm indicates that the tailings below this depth may be largely unoxidised and capable generating large volumes of AMD in the environmental conditions prevailing at the site unless remedial steps are taken to reclaim the site. The disturbance and exposure of unoxidised tailings at the site must be avoided to minimise the likelihood of exacerbating sulphide oxidation at the site and AMD. • Based on the results of chemical characterisation it is postulated that the surface tailings (0-200 mm) have undergone extension oxidation and leaching which has substantially decreased the concentrations of most metals. • In terms of the potential of the tailings to host vegetation the results clearly indicate that the surface tailings are most suited to plant growth providing extreme acidity and nutrients imbalances are addressed and Pb remains in a form that is highly insoluble and thus largely unavailable to plants. • The surface tailings are considered to be relatively inert in terms of their AGP, although it is estimated that a liming rate of approximately 50 Mg per hectare would be required to prevent the long term reacidification of the tailings resulting from sulphide oxidation. 238 Chapter 5 Mineralogical Analysis & Scanning Electron Microscopy 5.0 Introduction This chapter further characterises the Tui mine tailings in terms of their mineralogy using X-ray Diffraction Analysis (XRD) and Energy Dispersive X-ray Spectrometry (EDS). In addition, Scanning Electron Microscopy (SEM) is used to investigate the variation in mineral surface morphology with horizon depth to establish if chemical trends observed in Chapter 4 have been manifested physically. 5.1 X-ray Diffraction Analysis The technique of XRD mineral identification is based on principle that each crystalline substance has a characteristic arrangement of atoms which diffract X-rays in a unique and hence characteristic pattern (Whitton & Churchman, 1987). Although XRD analyses has been used to investigate the mineralogy of ore materials, andesitic rocks and stream sediments associated with the Tui Mine (Cochrane, 1969; Wodzicki & Weissberg, 1 970; Tay, 1 980), there appears to have been very little analysis carried out on the tailings material per se. Whilst Pang ( 1 992) mentions the identification of quartz, pyrite, some sphalerite and kaolinite and a "little" feldspar in a bulk sample of the Tui mine tailings using XRD, the author stops short of discussing the relative proportions of the minerals and the source of the data. Cochrane (1 969) used XRD to determine the purity of sphalerite and the presence of "well­ crystallised kaolinite" in the Tui ore. Wodzicki & Weissberg ( 1 970) similarly used XRD to identify, quartz, chlorite, illite and inter-layered illite-montmorillonite in a study investigating the alteration and origin of sulphide mineralisation at the Tui Mine. Although Tay ( 1 980) did not conduct XRD analyses on the Tui tailings per se, analysis of sediments derived from Tui mine adits and the Tui and Tunakohoia Streams was undertaken. In Tay's study, quartz galena, sphalerite and cerussite were identified in No. 7 adit sediments. The XRD traces of Tui Stream sediments (taken from below the confluence of the tailings dam 239 stream), however, did not indicate the presence of sulphide minerals despite the high concentrations of Cu, Cd, Fe, Pb and Zn in the sediments. Iron was therefore postulated to be predominantly in the form of ferric oxy-hydroxides which are amorphous to X-rays and have a tendency to ' scavenge' and adsorb heavy metals such as Cu, Cd, Pb and Zn. Heavy liquid (tetrabromoethane) separation of sediments from the Tunakohoia Stream conducted by Tay ( 1 980) facilitated the identification of sphalerite and pyrite in the sediments. An attempt by Tay to quantify pyrite concentrations by relative peak intensity failed, however, due to complications relating to the overlapping peaks of sphalerite and pyrite at 1 .63 , 1 .9 1 and 3 . 1 2 A o . Tay ( 1 980) concluded that based on XRD analysis of the sediments, peak intensity values did not satisfactorily predict pyrite content. 5.1 . 1 Aims of X-ray Diffraction Study Analytical data reported in Chapter 4 indicates that Cu, Fe and Zn are present in relatively low concentrations in the top 200 mm of the tailings profile. The distribution of Pb, in contrast, was found to be relatively uniform with depth. It was postulated that the oxidation of sulphide minerals and the subsequent dissolution and leaching of weathering products into the aquatic environment has lead to the specific depletion of some metals (Table 5 . 1 ) . Sulphur analyses and pH data from the profile samples support this theory. Thus the primary aims of the XRD analyses reported in this chapter was to :- ( 1 ) Identify the different mineral phases i n the tailings within the surface 600 mm of the profile. (2) Confirm observed trends from previous analytical work that indicates that concentrations of sulphides increase with depth. (3) To ascertain if and why Pb appears to be resistant to leaching from the surface tailings (0-200 mm). 5.1 .2 Preliminary X-ray Diffraction Method The samples used for this study correspond to those described in Chapter 4 (section 4.3 . 1 .2) and were obtained from an open pit excavated near the centre of the tailings at the western end of the dam (grid sampling site 7). Six, 1 00 mm incremental samples were collected at 240 this site to a depth of 600 nun . The samples were air-dried, ground to < 1 50 �m with an agate mortar and pestle and prepared for XRD analysis by packing holders with non­ oriented powder samples. A sample ore from the Tui Mine was similarly prepared. Analysis of the samples was undertaken using a Phillip' s PW 1 729 X-ray Generator and a PW 1 840 Diffractometer, operating at 40 kV and 40 rnA. A cobalt anode was used and the samples were run from 2° to 75° (28). Sietronics XRD trace processing software (Version 1 .0) was used to analyse the traces and to identify the mineral phases present in the tailings samples. 5.1 .3 Preliminary X-ray Diffraction Results Results of the XRD analyses indicated that the tailings were predominantly quartz, with minor kaolinite (Figure 5 . l a) . Samples below 200 nun were shown to contain small quantities of pyrite. This trend is consistent with F AAS 1 data which show an almost tenfold increase in total Fe content below the 200 mm depth (Table 5 . 1) . The XRD trace indicated that the 0- 1 00 mm sample contained no detectable pyrite ( Figure 5 . 1 a), whereas the 500- 600 nun sample was shown to contain a small but significant amount of pyrite as indicated by the presence of an additional 7 peaks associated with this mineral (Figure 5 . 1 b). The Tui ore sample analysed was found to consist primarily of quartz, galena, sphalerite and chalcopyrite. The presence of kaolinite or pyrite in the ore were not detected. It is postulated that the presence of pyrite was masked by sphalerite which, as previously mentioned, has similar major d-spacings to pyrite. F AAS data indicates that the sample contained over 4 times as much Zn as Fe (Table 5 . 1 ). The presence of relatively high proportions of sphalerite in the Tui ore is consistent with the findings of previous mineralogical studies reported in Chapter 2 (section 2.6). I Flame Atomic Absorption Spectrometry 241 5.2 Sodium Poly tungstate Separation Inspection of the XRD traces indicated that the various tailings samples were dominated by quartz which effectively diluted the heavy mineral fraction (HMF) of the tailings. To mitigate this problem density separation was used to isolate the HMF of the tailings from the light mineral fraction (LMF). This was achieved using a solution of sodium polytungstate (SPT) with a bulk density (pb) of approximately 2.8 g/cm3 (Callahan, 1 987). Thus for the purposes of this study the LMF is defined as the fraction of the tailings made up of particles with a pb of less than 2 .8 glcm3• Conversely, the HMF is defined as the fraction of tailings composed of particles with a pb of greater than 2.8 glcm3 . Table 5.1 Total elemental concentrations of selected metals and minerals detected by XRD in unseparated Tui mine tailings. The minerals are listed in alphabetical order and abbreviated as follows: ch = chalcopyrite (CuFeSz); gn = galena (PbS); ka = kaolinite (AhSi20S(OH)4); py = pyrite (FeSz); qtz = quartz (SiOz); sl = sphalerite (ZnS). Sample Depth Minerals Fe [ppm] Cu [ppm] Pb [ppm] Zn [ppm] [mm] Detected 1 0- 1 00 ka, qtz 2967 75 7538 347 2 1 00-200 ka, qtz 3660 1 1 3 1 0568 486 ,., 200-300 ka, py, qtz 32443 746 1 1 590 1 885 .) 4 300-400 ka, py, qtz 32892 1 465 5857 1 893 5 400-500 ka, py, qtz 32475 2 1 85 5060 5462 6 500-600 ka, py, qtz 49877 1 96 1 4058 4388 Tui ore NA ch, gn, qtz, s1 43279 1 0 1 69 1 86486 1 85266 5.2.1 Sodium Poly tungstate Separation Method Half gram (0.500g) tailings samples were placed in centrifuge tubes with narrow stems. Approximately 5 mL of SPT was added to each tube and the solutions were stirred gently to disperse the grains and break up aggregates. The tubes were then centrifuged for 5 minutes at approximately 1 500 rpm. The LMF of the samples were decanted from the test-tubes into funnels containing filter paper (MFS, No. 1 ). This was achieved by creating a seal at the narrow section of the stem containing the HMF, with a glass rod. The tubes were rinsed to ensure that any of the tailings LMF adhering to the walls of the tubes was removed. The samples were rinsed thoroughly with deionised water to remove the SPT and then placed in Sample: A:\jm1 28/06/94 C 1200,-------------------------------------------------------------, o u n t s 800 Sample: A:\jm6 28/06/94 Q Q Q K C 1 2001,- ------------------------------------------------------------- o u n t s 800� 400 2 Q 12 22 Q Q p p 32 42 Degrees 2-Theta Q Q K Q 62 72 242 Figure 5.1 (a) XRD trace of 0- 1 00 nun tailings sample and (b) XRD trace of 500-600 nun tailings sample containing detectable pyrite (P) as well as kaolinite (K) and quartz (Q). 243 an oven at 1 05 °C and left to dry over night on filter papers (Whitton & Churchman, 1 987). Following the removal of the LMF, the HMF was rinsed from the bottom of the centrifuge tubes with distilled water and similarly collected and dried. Both fractions of each sample were then weighed to 4 decimal places to enable the relative proportions of each fraction to be calculated (Table 5 .2). 5.2.2 X-ray Diffraction Method Due to the greater number of samples a Phillip' s 4-kVa PW 1 7 1 0 instrument with a PW 1 050/80 goniometer operating at 40 kV and 60 rnA was used in conjunction with a 42 sample changer system (a PW 1 775 unit). A broad-focus cobalt anode tube was used and the samples were scanned from 2° to 90° 28. The tailings and ore samples were run as acetone slurry mounts which were prepared by mixing ground tailings with a few drops of acetone on a glass slide before evenly distributing the slurry over the whole slide and allowing it to air-dry (Whitton & Churchman, 1 987) . 5.2.3 Sodium Poly tungstate Separation Results Density separation of the tailings enabled a semi-quantitative estimate of the relative proportions of heavy to light minerals in the mine tailings to be made and provided useful insight into the probable distribution of sulphide minerals within the profile. Results from the SPT density separation of the tailings indicate that the HMF generally increases with depth. The HMF was found to increase from approximately 1 .5% in the surface sample (0- 1 00mm) to over 1 6% in the bottom sample (500-600 mm) (Figure 5 .2). Figure 5.2 indicates that there is a substantial increase in the HMF of the tailings below 200 mm depth. XRD analysis of the LMF also indicates a distinct change in the composition of the tailings below 200 mm where pyrite and/or sphalerite is detected (Table 5 .2). The increase in heavy minerals with depth is consistent with F AAS analyses which shows a corresponding increase in the concentrations of Cu, Fe and Zn with depth (Table 5 . 1 ) and 244 correlates well with AGP values presented in Figure 4.29. The non-depletion ofPb from the upper profile is readily explained by the detection of anglesite (PbS04) in the HMF of the tailings2 . Table 5.2 Heavy mineral fraction (HMF) and XRD identified minerals. Minerals are listed alphabetically and abbreviated as follows: an-anglesite (PbS04); ca-calcite (CaC03); ch-chalcopyrite (CuFeSz); gn-galena (PbS); ka-kaolinite (AhSizOs(OH)4); qtz- quartz (Si02); slfpy- sphalerite (ZnS) and pyrite (FeSz). Sample Depth HMF LMF HMF (mm] [%] Minerals Detected Minerals Detected 1 0- 1 00 1 .45 ka. qtz an, gn, ka, qtz, sIlpy 2 1 00-200 1 .55 ka, qtz an, gn, ka. qtz, sIlpy 3 200-300 7.23 ka, qtz, sIlpy an, ch, gn, ka, qtz, sIlpy 4 300-400 8 . 1 1 ka, qtz, sIlpy an, ch, ka, qtz, sIlpy, 5 400-500 9.85 ka. qtz, sIlpy an, ch, gn, ka, qtz, sl/py 6 500-600 1 6.25 ka. qtz, slJpy an, ch, gn, ka. qtz, slJpy Tui ore NA 77.7 gn, ka, qtz, slJpy ca, ch, gn, ka, qtz, sl/py, The presence of pyrite and/or sphalerite in LMF indicates that SPT density separation was not 1 00% efficient at separating the quartz and kaolinite from the sulphide minerals. Pyrite and sphalerite have specific gravities of approximately 5 .0 and 4 . 1 g/cm3 respectively (Deer et ai., 1 992) and are therefore far denser than the SPT solution used. Contamination of the LMF with these minerals, in particular pyrite, was unexpected and explained only after examination of the sample grains under an optical microscope (transmitted light). Temporary mounts of the tailings were made for examination by sprinkling a small quantity of the material on an optical slide and adding two or three drops of clove oil and a cover slip (Whitton & Churchman, 1 987). Examination of the LMF grains at l O x magnification showed a significant portion of the quartz grains contained inclusions of opaque material, presumably pyrite and/or sphalerite. The sulphide inclusions were sufficiently small to prevent the overall density of the particle to rise above 2 . 8 g/cm3, thus providing a mechanism for the 'contamination' of the LMF with detectable concentrations of sulphide minerals. The intrinsic association of the quartz and pyrite also explains the presence of a 2 See further discussion in section 5 .2.4. 245 significant amount of quartz in the HMF of the tailings. Due to the comparatively high density of the heavy minerals, a relatively large amount of quartz was carried through into the HMF fraction of the tailings by the sodium polytungstate solution. The efficacy of this method at separating the mineral fractions of the tailings is therefore dependant on the sample containing discreet mineral grains and could theoretically be improved by finer grinding of the tailings prior to separation. Excessive grinding, however, is undesirable as it reduces the quality of the XRD patterns generated by mineral samples. o 1 00 200 ..c: - fr 300 --..; Q 0.> is o "" Q,. 400 500 ", 600 _I Heavy Mineral Concentrations [%] 5 10 I S 20 <> i Figure 5.2 Heavy mineral contents versus depth in incremental samples. 5.2.4 X-ray Diffraction Results XRD analyses of the heavy mineral fraction (HMF) of the Tui mine tailings identified an additional 4 minerals that were not detected in the non-separated tailings samples. These 246 included anglesite, chalcopyrite, galena and sphalerite. An increase in the relative abundance of pyrite and/or sphalerite with depth was indicated by a significant increase in the intensity of peaks corresponding to these minerals . No attempt, however, was made to quantify the relative proportions of the individual minerals as comprehensive F AAS analyses of the tailings provided an adequate indication of the approximate proportions of minerals in the various samples (Table 5 . 1 ). Interpretation of the XRD patterns indicates that anglesite (PbS04) is the most prevalent Pb mineral in the top 300 mm of the profile. Although anglesite has been identified in Tui ore samples in several studies (Courtney et al. , 1 990), its limited occurrence in the ore indicates that the origin of this mineral in the tailing is primarily from in situ weathering of galena. The identification of anglesite confirms that the surface tailings are actively oxidising. The relatively high concentrations of Pb in the upper profile of the Tui mine tailings compared to Cu, Fe and Zn is fully explained by the extremely low solubility and high stability of anglesite (Table 5 .3) . Under oxidising conditions anglesite (PbS04) is stable between pH values of about 0 .4 and 5 (Brookins, 1 988) . Comparable weathering products of sphalerite, pyrite and chalcopyrite are generally highly soluble under similar pH conditions (Table 5 .3) and are likely to have been leached from the upper profile by infiltrating rain and surface water. The relative depletion of Cu, Fe and Zn from the upper profile and immobility of Pb is therefore largely a function of the stability/solubility of the various weathering products of the remnant sulphide minerals within the tailings. Table 5.3 Solubility and density of selected sulphate minerals (Aylward & Findlay, 1 974). Compound Name Solubility (gil OOg] Specific gravity [g cm-3] CuS04. 5H2O Chalcanthite 22 2 .3 FeS04·7H20 Melanterite 29.5 1 .9 PbS04 Anglesite 0.0045 6.4 ZnS04.7H20 Goslarite 5 8 2 .0 247 The water quality studies of Tay ( 1 9 8 1 ) and Pang ( 1 992) confirm that substantial amounts of Cu, Fe, Zn and other elements are currently being leached into surface and ground water at the Tui mine site. The relative immobility of Pb is reflected by the fact that Fe and Cu concentrations in leachates emanating from the tailings dam tend to be at least two orders of magnitude greater than Pb concentrations (Table 2. 1 ). Density separation of the Tui ore using SPT facilitated the identification of kaolinite and pyrite in the LMF and HMF respectively. XRD analysis of the ore sample also indicated the presence of calcite in the HMF sample. Calcite is a common gangue mineral associated with hydrothermal sulphide deposits and has been identified in several previous mineralogical studies of the Tui ore (Courtney et aI., 1 990). The fact that calcite was not detected in the LMF of the ore material is surprising given that due to its relatively low specific gravity (2.7 g/cm3) it should have been separated into this fraction rather than the HMF. XRD analysis of the tailings does, however, clearly indicate that calcite was not a major constituent of the Tui ore sample. The absence of XRD-detectable calcite in the tailings was not surprising considering the extremely acidic conditions which prevail at most depths. Most calcite initially deposited as gangue derived from the processing of ore has undoubtedly reacted with hydrogen ions produced by the oxidation of pyrite and chalcopyrite in the tailings, resulting in its depletion from at least the upper 600 mm of the tailings profile. 5.3 Scanning Electron Microscopy & Energy Dispersive X-ray Spectrometry Scanning Electron Microscopy (SEM) has emerged as a powerful tool for examination of material surfaces. It is particularly useful for producing images of the surface textures of mineral grains and delicate structures such as microfossils or micro biota and thus has considerable application within the petroleum industry, for example, for the paleontological dating of marine sediments. 248 In SEM an electron beam is rastered across the surface of a specimen to build up a picture of the entire area. Electrons are used because they have a much shorter effective wavelength than visible light and therefore provide much higher resolving power (Lapidus, 1 990). The microscope operates by counting the secondary electrons emitted when the primary electron beam strikes and excites atoms at the specimen's surface which is generally coated with a thin film of gold or carbon. The secondary electrons are detected by a scintillation counter which produces a voltage signal proportional to the number of electrons emitted from the specimen surface. The quantity of electrons emitted is governed by the specimen' s surface morphology which results in signal strength variation from the surface profile of the specimen. The signals are then amplified and used to produce an image on a monitor. Many scanning electron microscopes have an associated analytical facility. The most common of these is Energy Dispersive X-ray Spectrometry (EDS) which is based on detecting the X -ray energy spectrum characteristics of the elements present in the area of electron beam impact. This technique combined with SEM makes it possible to identify elements in situ. Although the use of SEM has increased steadily since its commercial introduction in the 1960' s, the amount of reference material available to aid in SEM mineral identification is relatively small (Welton, 1 984). Texts such as that by Hearle et al., ( 1 972) and Wells ( 1 974) give excellent descriptions of SEM theory but information regarding the application of SEM to geological studies is often very limited in contrast to biological studies. Whilst Hearle et al. ( 1 972) provide an excellent review of the application of SEM to metallurgy, the literature is heavily biased towards industrial applications, such as the detection of metal fatigue, stress fractures and surface wear studies. Welton ( 1 984) published an atlas to provide SEM users in the petroleum industry with a guide to aid in the identification of minerals which are commonly found in clastic reservoir rocks. Because the atlas was almost exclusively designed for use within the petroleum industry it was of limited use in this present study. 249 Over the past decade the use of SEM and EDS in geological and pedological related studies has steadily increased as researchers have begun to incorporate these technologies into their repertoire of analytical techniques. For example, Goenadi & Tan ( 1 99 1 ), Robertson & Eggleton ( 1 99 1 ) and Watanabe et al. ( 1 992), have used SEM in studies to investigate the weathering of clay minerals such as muscovite, kaolinite and montmorillonite. In these studies SEM (and EDS in the first two studies) was used in conjunction with other analytical techniques to identify and characterise various primary and secondary clay minerals and to investigate weathering processes. SEM analyses were also used in a pedological study by Karathanasis ( 1 99 1 ) to investigate phosphate mineral weathering in Alfisols developed in limestone. In this study SEM was used to study apatite surface weathering and to investigate the formation of secondary Ca-Al and Al and/or Fe­ phosphates. Golden et al. ( 1 99 1 ) have similarly used SEM and EDS to investigate phosphate dissolution in soils. In a recent study by Inskeep et al. ( 1 993), the surface morphology and secondary coatings of naturally weathered oligoclase grains in soils, sediments and rocks were characterised using SEM and EDS . The aforementioned techniques were used to characterise etch pits and secondary coatings on a sequence of oligoclase grains at various stages of weathering. The results were used to evaluate the efficiency of laboratory studies in predicting weathering intensities in natural watersheds. In four notable studies investigating pyrite oxidation, SEM was used to elucidate the role of Thiobacillus ferrooxidans (Bennett & Tributsch ( 1 978); Keller & Murr ( 1 982); Rodriguez­ Leiva & Tributsch ( 1 988) and Mustin et al. ( 1 992). In the study of Rodriguez-Leiva & Tributsch ( 1 98 8) Tferrooxidans was cultivated on synthetic pyrite and SEM was used to examine the morphology of the pyrite-bacteria interface with time. SEM facilitated the identification of bacterial corrosion pits at various stages of development and helped to identify the presence of organic films in and around corrosion pits created by the bacteria. The organic compounds found between the bacterial outer membrane and the sulphide 250 interface are believed to be secreted by Tferrooxidans during sulphide oxidation to facilitate breakdown of the sulphide moiety. Rodriguez-Leiva & Tributsch ( 1 988) believed that previous SEM studies of Tferrooxidans on natural sulphide surfaces were largely unsuccessful due to the complex surface morphology of non-synthetic sulphide crystals. Research aided by SEM indicates that Tferrooxidans does not absorb randomly onto pyrite but instead preferentially attacks crystal defects, impurities and mineral inclusions on the crystal surface (Andrews, 1 988) . Based partially on SEM analyses, Rodriguez-Leiva & Tributsch ( 1 988) suggested that T jerrooxidans has some sort of "chemotactic" sense which allows them to locate a favourable site for attachment on the sulfide interface. Tferrooxidans is thought to be able to distinguish between surface areas that mechanically or chemically deviate from ideally crystalline sulphide substrates (Bennett & Tributsch, 1 978). Research outlined by Natarajan ( 1 990) also indicates that surface defects, crystallographic orientation and surface heterogeneity are the main parameters influencing bacterial corrosion patterns on sulphide minerals. Mustin et ai. (1 992) applied SEM in an investigation of bacterial oxidation of pure pyrite by Tferrooxidans. Pyrite samples were biologically leached over a 60 day period to evaluate elution front analysis of intragranular porosity. SEM microphotographs of individual pyrite grains indicated that after the 60 day period a great number of corrosion tunnels were created by the bacteria, resulting in a surface area increase of approximately 42%. Further analysis of the sulphide particles indicated that some 25 km of tunnels were formed as a result of pyrite dissolution by T. ferrooxidans per gram of pyrite. The mean diameter of tunnels was of the order of 2 )lm and the shapes were predominantly hexagonal or square, probably relating to the cubic lattice of pyrite (Keller & Murr, 1 9 82) . The corrosion tunnels showed preferred orientation, evidenced in microphotographs by the anisotropic parallel alignment of pores . SEM images of the burrow borders showed little sign of bacterial or chemical degradation indicating the reactive area is located at the base of the holes. 25 1 In a study investigating the oxidation of pyrrhotite (Fel -xS), SEM was used to obtain images of biologically oxidised mineral surfaces (Kwong et al. , 1 995). SEM revealed that the bio­ oxidised pyrrhotite surfaces are appreciably different to pyrite surfaces in that they are characterised by micro-fractures and circular deposits of what was believed to be bacterially generated elemental S . The recent studies o f Willet e t al. ( 1 992) and Lin ( 1 997) also provide useful examples of the application of SEM and/or EDS technology in studying the oxidation of sulphide minerals in acid sulphate soils and mine tailings, respectively. The study of Lin ( 1 997) is similar to the present study in that it used SEM and EDS to complement more traditional analytical and mineralogical techniques (including F AAS and XRD) to characterise base­ metal tailings. 5.3 . 1 Aims of SEM & EDS Study Although F AAS, chemical and XRD data indicate that the surface tailings (O-200mm) contain comparatively low concentrations of metal-bearing sulphide minerals, it provided little insight into the original distribution of sulphide minerals in the profile. The primary aim of the SEM and EDS research reported in this chapter was to : - ( 1 ) establish if the distribution of sulphide minerals in the profile was originally relatively uniform with depth and (2) to investigate if the depletion of metals in the surface tailings resulted from sulphide oxidation and the subsequent leaching of soluble metal compounds. In addition, it was envisaged that SEM and EDS analyses would help identify a relationship between the Tui tailings chemistry, depth and surface morphology, should one exist. 5.3.2 Sample Preparation As previously reported in section 5 . 1 . 3 , XRD analyses of the six l OO-mm incremental tailings samples indicated they were composed primarily of quartz and kaolinite. To isolate sulphide particles for SEM inspection and EDS analyses, sodium polytungstate (SPT) was again used to separate the maj ority of quartz and kaolinite from the heavy mineral fraction (sulphide component) of the tailings samples as per the method outlined in section 5 .2 . 1 252 (Callahan, 1 987). The separation method was this time conducted using unground tailings samples. Thus additional care was taken to ensure loosely bound aggregates were adequately broken apart to facilitate dispersion of the particles in solution. Samples of the heavy mineral fraction of the tailings were mounted on aluminium stubs and sputter-coated with gold. The use of highly conductive gold for sample coating minimises static electricity build up and thus minimises 'flaring' and improves image quality. In total, seven samples were mounted for SEM examination, including six tailings samples and a specimen of ground « 1 50 !-lm) Tui ore (Table 5 .4). Table 5.4 Tailings Samples Examined by SEM & EDS. Sampl Depth SEM EDS HM e [mm] [%] 1 0- 1 00 ./ ./ 1 .45 2 1 00-200 ./ x 1 .55 3 200-300 ./ x 7.23 4 300-400 ./ x 8 . 1 1 5 400-500 ./ x 9.85 6 500-600 ./ ./ 1 6.25 Tui ore NA ./ x 77.7 (v" = analysis conducted, x = analyses not conducted, NA = not applicabJe) For EDS analysis two samples were sputter-coated with carbon instead of gold. Although carbon-coated samples generally produce an inferior SEM image to gold-coated samples (attributed to carbon's poorer conductive properties) it is more suited to EDS analysis as it prevents the masking of the X-ray spectra back-scattered from the samples. Tailings from the 0- 1 00 mm and 500-600 mm depths were chosen for EDS analysis because, (i) XRD analyses indicated that there was a relatively large difference in sulphide content between the two samples, (ii) chemical characterisation of the tailings indicated that they were quite dissimilar in terms of their pH, sulphate and metal content and theoretical acid generating potential and (iii) field observations indicated that these samples were obtained from contrasting oxidising environments. 253 5.3.3 Scanning Electron Microscopy Method SEM examination and EDS analyses of the samples were made with a Cambridge Stereoscan 250 Mark 3 , with an attached Link E543 1 EDS detector and 8 /lm beryllium window. Fine-grained black and white micrographs of the mineral surfaces were taken using a 3 5 -mm Nikon camera focused on a high-definition monitor. Exposure times of 60 seconds were used to maximise the signal to noise ratio and to improve image quality. 5.3.4 Results & Discussion 5.3.4.1 Scanning Electron Microscopy SEM examination of HMF samples provides evidence that the change in the distribution of sulphide minerals with depth is attributable to differences in the extent of oxidation and subsequent dissolution of sulphide minerals. Sulphide particles in the upper 200 mm of the profile generally exhibited very irregular and craggy morphologies (Plate 5 . 1 ). In contrast many of the mineral grains between 200 and 600 mm depth exhibited angular well-defined edges which enabled their crystal habit to be readily recognised. EDS analysis conducted on several cubic particles yielded spectra that were characterised by two Fe peaks and a single S peak (Appendix 5 (A)). Comparison of these spectra with that of a standard iron pyrite (FeS2) spectrum confirmed that the cubes were indeed pyritic (Welton, 1 984). Cubic pyrite particles were observed only in the 200-600 mm zone and became more prevalent towards the bottom of the profile (Plate 5 . 1 ). Under moderate magnification (420x) pyrite cubes isolated in the 500-600 mm depth sample show few signs of 'weathering' (Plate 5 .2). At higher magnification, however, there is some evidence that dissolution of the mineral edges has occurred (Plate 5 .3 ) . The lack of cubic pyrite in the upper 200 mm of the tailings profile indicates that this form of pyrite is susceptible to oxidation and has been 'weathered' beyond recognition. The effects of surface weathering of a pyrite particle isolated in the 300-400 mm zone is evident in Plates 5 .4 & 5 .5 . Small hexagonal shaped etch holes are clearly visible in the middle and upper left hand comer of Plate 5 .5 The most probable mechanism for the 600 mm Depth .... 011(�--------------------------------------------... � Surface Plate 5.1 A photo-mosaic of six microphotographs correlating to each of the sampling depths from 0-600 mm. Particles within the top 200 mm of the profile are generally seen to exhibit a craggy and irregular morphology (C). The occurrence of cube-shaped pyrite minerals (P) and angular grains is seen to increase w ith depth (Magnification 1 80 x). 255 Plate 5.2 A SEM image showing three cuboidal pyrite grains (P) iso lated in the 500-600 mm zone. Their we l l defined cube shape indicates that they are resistant to the micro-environmental weathering conditions prevailing at this depth (Magnification 470 x). Plate 5.3 A magnified view of central pyrite cube in P late 5.2 confirms that the surface is largely unweathered and that i t is partia l ly coated with plates of kaol in ite (Magnification 4700 x). Plate 5.4. Surface etching of a pyrite grain (centre) indicates that l imited biochemical oxidation is occurring at 300-400 mm depth (Magnification 480 x). 256 Plate 5.5 A magnified view of the pyrite grain in P late 5 .4 indicates the presence of several small hexagonal etch holes (E). Like the smal ler holes, the larger non-symmetrical etch p it (N) is thought to have resulted from biological and/or chemical oxidation of the pyrite along a chemically active zone (Magnification 4800 x). 257 creation of these holes is that they are the result of bacterial and/or chemical oxidation of the pyrite. Similar hexagonal etch-pits were observed in SEM microphotographs from an experiment investigating the role of crystallographic orientation in pyrite oxidation (Keller & Murr, 1 982). Hexagonal and sub-hexagonal etch pits in pyrite were also observed with SEM in research conducted by Mustin et al. ( 1 992). In the latter study the etch pits were believed to have formed as a result of bacterial oxidation, with the hexagonal shapes and anisotropy of the corrosion pits being related to the cubic lattice of pyrite. Burrow-like weathering parallel to the mineral surface is also evident in Plate 5 . 5 . The irregular shape of the etch pit indicates that preferential weathering of the mineral may have occurred at a chemically active zone. Such zones are believed to result from crystal dislocations or fracture lines in the mineral which enhance the rate of sulphide oxidation by providing conduits for the diffusion of S atoms (Andrews, 1 988). Bacteria adsorbed at these sites are thought to be advantaged because S oxidation has a much higher yield of free energy than Fe oxidation (Andrews, 1 988). At pure pyrite sites the diffusivity of S atoms through pyrite crystals is several orders of magnitude lower. This is a possible explanation of the observed selectivity shown by Tferrooxidans for specific sites (Bennett & Tributsch, 1 978; Andrews, 1 988; Rodriguez-Leiva & Tributsch, 1988 ; Mustin et al. , 1 992). Wiggering ( 1 993) reported similar leaching channels in pyrite which he suggested resulted from the selective dissolution of individual zones within the crystals. An alternative explanation for the formation of the irregular etch-hole, however, is that it may have resulted from the release of surface fluid inclusions during pyrite dissolution. Although the surface tailings (0-200 mm) were devoid of cubic pyrite they were found to contain a greater proportion of distinctive 'aggregate grains' than the underlying tailings (Plates 5 .6 & 5 .8) . At high magnification (9200x) these grains are seen to be composed of individual sub-angular crystals less than 1 !lm in width (Plate 5 .7). EDS analysis of an aggregate grain isolated in the 0- 1 00 mm zone indicated that it comprised primarily AI, Cu, Fe, Pb, Si and S (Appendix 5 (B». Although the dominant form of the S present in the particle (ie. sulphide or sulphate) could not be interpreted from the EDS spectrum for the 258 mineral, it is highly unlikely that tiny sulphide crystallites would persist in the strongly oxidising conditions present in the surface tailings. These particles were therefore postulated to be clusters of metal sulphate crystallites precipitated from the weathering solution. XRD and chemical analysis of the 0- 100 mm samples indicates that angle site is likely to be a major constituent of the grain, however, this could not be confirmed from the EDS spectra due to the major EDS peak for Pb coinciding with that of S (Appendix 5 (B» . SEM examination of the HMF of the tailings indicated that all the samples again contained appreciable proportions of quartz after density separation (Plate 5 .9) . Although quartz' s characteristic conchoidal fracture and absence of cleavage made visual identification of particles relatively straightforward (0 'Donoghue, 1 976), the composition of these grains was confirmed using EDS. Whilst the quartz (Si02) spectra were dominated by a single silicon peak, minor AI, Fe and Sn were identified in several samples (Appendix 5 (C» . As discussed in section 5 .2.3 the inability of the heavy mineral separation method to completely segregate quartz from the HMF was explained by the presence of sulphide inclusions in the quartz grains. EDS analyses confirmed the presence of phyllosilicate clay minerals in the tailings. It appears that X-ray amorphous iron-alumino-silicate gels were also present as coatings on some grains in the surface tailings (0- 1 00 mm). EDS spectra associated with these grains exhibited large Fe peaks with relatively small S peaks, indicating that the Fe was not intimately associated with S in the form of a sulphide or sulphate compound but rather with AI, S i and presumably O2 (Appendix 5 (D) The detection of iron-alumino-silicate gels in the tailings is significant in that it is indicative of strong oxidising conditions which are conducive to sulphide oxidation. A potential source for the Al and Si identified in these spectra is the clay mineral kaolinite. Although kaolinite was readily detected in the samples by XRD analyses, well crystallised kaolinite was difficult to identify in the SEM microphotographs. Some of the mineral 259 Plate 5.6 An 'aggregate grain' comprised of sub-cuboidal crystal lites isolated in the 0- 1 00 111m depth zone. These distinctively shaped grains are more prevalent in the surface 200 m111 of the profile (Magnification 470 x). Plate 5.7 A magnified view of the sub-cuboidal crystal l ites constituting the aggregate grain i n Plate 5 .6 (Magnification 9200 x). 2(iO Plate 5.8 An 'aggregate grain' isolated in the 0- 1 00 mm depth zone exhibiting mami!lary growth habit (Magnification 460 x). Plate 5.9 A SEM image of a quartz particle ' contaminating' the heavy mineral fraction of the tailings. Quartz partic les such as these constitute > 80 - 90% of the tailings and are characterised by conchoidal fracturing and angular morphologies (Magnification 360 x). 26 1 Plate 5. 1 0 A h igh resolution image of p late-like kaolinite located on the surface of a m ineral grain in the 500-600 mm sample (Magnification 4700 x). Plate 5.1 1 A SEM microphotograph of the surface of a sulphide mineral isolated in the Tui ore sample (Magnification 9500 x). 262 surfaces were, however, coated with plate-like particles which based upon their size and morphology were probably kaolinite (Plate 5 .3 & 5 . 5) . Increased magnification of selected mineral grains revealed what appeared to be localised 'pockets' of poorly crystallised kaolinite (Plates 5 . 1 0). It is possible that the kaolinite flakes were difficult to identify because they had undergone partial dissolution as a result of edge attack under the acidic conditions prevailing in the tailings (Barnhisel et al. , 1 982). SEM images of the Tui ore were characterised by highly angular particles. Whilst the majority of the grains were indistinctive, some grains exhibited unique cleavage patterns (Plate 5 . 1 1 ) 5.3.5 Low Magnification Energy Dispersive X-ray Spectrometry EDS analyses confirmed the findings of XRD analysis which indicate that the tailings contain a variety of minerals . This finding was not unexpected given the multitude of minerals known to occur in the 'parent material ' (the Champion and Ruakaka ore bodies) from which the tailings were derived (Courtney et al., 1 990; Wodzicki & Weissberg, 1 970; Weissberg & Wodzicki, 1 970). A total of 12 elements were detected in the tailings by EDS including AI, Ba, Ca, Cu, Fe, K, Pb, S, Si, Sn, W and Zn. Whilst tungsten (W) may be an artefact of the sodium polytungstate used to separate the minerals fractions, wolframite {(Fe2+, Mn)W04} has been identified in the ore body and thus W may be a 'genuine' elemental constituent of the Tui tailings (Courtney et al. , 1 990). Low magnification EDS analyses were conducted on the two samples to establish if there was any marked variation in the elemental composition of the tailings . Three EDS analyses were conducted at each depth at 1 80 x magnification. Although minor differences were noted between the replicate EDS spectra, the differences were negligible compared to those between the two sampling depths. EDS analyses of the 0- 1 00 mm sample (Appendix 5(E» indicated that it contained no Cu or Zn but small quantities of Ba presumably in the form of 263 barite (BaS04) which is a common constituent of hydrothermal vein deposits (O'Donoghue, 1 976). In contrast, the 500-600 rom tailings sample contained Cu and Zn but no Ba (Table 5 .5 ; Appendix 5 (F)). Trace quantities of Ca were also identified in the surface sample. Table 5.5 EDS analysis of Tui tailings particles Depth AI Ba Ca Cu Fe K Pb S i S Sn W Zn [mm] 0- 1 00 ..r ..r ..r .Ie ..r ..r ..r ..r ..r .Ie .Ie .Ie 500-600 ..r .Ie .Ie ..r ..r .Ie ..r ..r ..r .Ie ..r ..r The absence of EDS detectable Cu and Zn in the upper sample (0- 1 00 rom) reflects the fact that the sulphide minerals associated with these metals are more susceptible to weathering than pyrite. Research has shown that galvanic interactions come into play in poly-sulphidic leaching systems (Natarajan, 1 990). These interactions lead to the preferential weathering of electrochemically active sulphide minerals due to the formation of galvanic cells (Natarajan, 1 990; Kwong, 1 995). Table 5 .6 lists six base-metal sulphides in a bioleaching medium, in order of decreasing stability based on their relative electrochemical activity. Table 5 . 6 shows that pyrite and chalcopyrite are relatively 'noble' sulphides and are thus more difficult to oxidise in an acid medium than say pyrrhotite or sphalerite, which are electrochemically less stable. Table 5.6 Galvanic series of some common sulphide minerals (adapted from Natarajan, 1 990). Pyrite Noble Chalcopyrite 1 Pentlandite Decreasing Galena Rest Potential Pyrrhotite Sphalerite Active The galvanic dissolution process is driven by the potential differences existing in a galvanic couple, while the rate of anodic oxidation is a function of the galvanic current flowing in the circuit (Natarajan, 1 990). The rate at which selective dissolution of sphalerite in the presence of cathodic pyrite for example, depends on several factors such as the rest 264 potential differences and conductivity of the minerals, the relative surface areas of the anodic and cathodic minerals, the distance between the sulphidic electrodes, the duration of mineral contact, the environmental parameters of the medium (eg. pH, Eh and oxygen availability) and the presence and activity of acidophilic bacteria such as T ferrooxidans (Natarajan, 1 990). The phenomenon of selective galvanic bioleaching is one possible explanation why Cu and Zn were not detected in the upper tailings sample by ED S and may partially explain the absence of XRD detectable galena in the 300-400 mm tailings zone. The greater concentration o f pyrite in the tailings and the cathodic protection afforded to this mineral by galena and sphalerite has undoubtedly contributed to the selective oxidation of the less noble sulphides in the upper profile. S phalerite is particularly electrochemically unstable due to its low (negative) rest potential and thus tends to be rapidly oxidised and ' lost' from polysulphidic materials (Natarajan, 1 990; Kwong, 1 995) Although the distribution of Cu, Fe, Pb and Zn within the tailings profile i s readily explained by the solubility of the oxidation products and/or the electrochemical reactivity of the sulphide minerals, the reason for the low concentrations of B a at depth is unknown. Like Pb, Ba is immobile over most of the Eh-pH field of water and tends to form highly insoluble sulphate (BaS04)' Although BaS04 dissociates to form Ba2+ below pH 1 .2 , the pH and Eh conditions present at 500-600 mm depth are not thought to be conducive to the formation of this cation (Brookins, 1 98 8) . In fact, Ba is more likely to be immobile in the conditions prevailing at the 500-600 mm depth than it is in the considerably more acidic conditions present in the surface tailings (0- 1 00 mm). It is thus postulated that the surface tailings were derived from ore containing relatively high concentrations of barite and that this mineral has remained in situ due to its chemical stability. Presumably Ba may have also undergone some isomorphous substitution with sulphate-bound Pb in the surface tailings . Oxidation of sulphide minerals in the top 200 mm of the profile is also likely to have resulted in the selective enrichment of barite due to the leaching of soluble metal compounds from the upper profile. The absence of XRD detectable concentrations of barite 265 or other Ba-bearing minerals, however, indicates that this element is not a major constituent of the tailings at either depth. Comparison of the EDS spectra for the respective surface (0- 1 00 mrn) and subsurface (500- 600 mm) samples indicates that Pb concentrations are slightly higher in the surface sample (Appendices 5(E) & 5(F)). This trend is consistent with finding ofXRD and FAAS analyses of the tailings. 5.4 Chapter 5 Summary & Conclusions • XRD analyses of the Tui mine tailings indicate that they are composed primarily of quartz, with subordinate kaolinite and pyrite. Sodium polytungstate (SPT) density was required to facilitate the identification of several additional minerals including; chalcopyrite, galena, sphalerite and anglesite. • Cross contamination of the heavy and light mineral fractions of the tailings with quartz and sulphide minerals respectively, continued to hamper mineral identification after SPT density separation. Contamination of the samples was explained by examination of the tailings particles under an optical microscope which showed that a significant portion of the quartz grains were found to contain inclusions of sulphide minerals . • Quantification of the light and heavy mineral fraction of the tailings after SPT density separation demonstrated a general trend of increasing heavy mineral concentration with depth. On a mass basis the tailings increased from about 1 .5% heavy minerals near the surface to over 1 6 % in the 500-600 mm depth zone. • The results gained from XRD analysis and sodium polytungstate separation of the Tui mine tailings were found to substantiate analytical data from the chemical characterisation of the tailings that indicates substantial oxidation of the tailings has occurred in the top 200 mm of the profile (Chapter 4). Supporting observations include; ( 1 ) the general increase in heavy mineral concentration with depth as determined by SPT density separation, (2) the lower abundance of sulphide minerals in the upper profile as evidenced by variations in XRD peak height intensity, (3) the absence of pyrite and/or 266 sphalerite in the LMF of the tailings derived from the upper 200 rom of the profile and (4) the greater concentrations of anglesite near the profile 's surface . .. The identification of the highly insoluble angle site in the tailings profile explains why Pb has remained largely in situ in the upper profile despite the strong leaching conditions. .. The major limitation of using XRD analyses to identify the mineral constituents of the Tui mine tailings was that the similarity of the XRD spectra for pyrite and sphalerite. Several coincidental d-spacings for these minerals l imited the use of XRD to quantify the relative concentrations of these minerals in the tailings based on peak intensity. .. As hypothesised, SEM indicated that sulphide minerals in the top 200 rom of the tailings profile are significantly more weathered than those below 200 rom. The increased degree of weathering in the top 200 rom of the profile is reflected by:- ( 1 ) the very craggy and extremely irregular morphology of the grains, (2) the absence of cubic and angular shaped sulphide minerals and (3) the presence of a greater proportion of what are postulated to be aggregated sulphate minerals . .. SEM results indicate that the original distribution of sulphide minerals in the tailings profile was relatively uniform and that depletion of Cu, Fe and Zn from the upper profile i s attributable to the oxidation of sulphides and the subsequent leaching of their weathering products. The selective dissolution of chalcopyrite and sphalerite due to galvanic interactions is postulated to have played a role in selectively depleting these minerals from the upper profile. .. The greater abundance of Pb and Ba in the upper profile is believed to result from a combination of:- ( 1 ) selective enrichment, as other more mobile metal compounds are oxidised and leached from the profile and/or (2) the possible deposition of Pb and Ba enriched tailings towards the end of mining operations. .. EDS analyses of the HMF of the tailings indicates that the particles tend to vary considerably in elemental composition, a finding believed to reflect the complex geochemistry of the Champion and Ruakaka ore bodies. .. Both SEM and EDS proved useful tools in assessing, qualitatively, the extent of weathering that has occurred in the tailings as a function of depth. There is undoubtedly 267 such scope for the use of these tools in characterisation and weathering studies of tailings and mine wastes. 268 Chapter 6 Lime Incubation Experiment 6.0 Introduction It is well recognised that traditional tests for assessing the liming requirements of soils which involve short term pH buffering experiments (Peech, 1 965; McLean, 1 973) often seriously underestimate the requirements of pyrite-bearing materials (Berg, 1 978; Caruccio et al. , 1 988) . Whilst the application of lime at rates sufficient to elevate pH to circumneutral levels may ameliorate acidity and decrease the bioavailability of heavy metals in sulphidic wastes in the short term, the effects are all too often temporary in nature. Reacidification of sulphidic growing media can occur extremely rapidly upon the exhaustion of added neutralising agents (Costigan et al. , 1 98 1 ; Dent, 1 986; Wiggering, 1993) . Vegetation die-back often quickly ensues as phytotoxic metals remobilise and become increasingly available for plant uptake. As discussed in Chapter 3 (section 3 .4. 1 ) ' static tests' have been specifically developed to assess the liming requirements and Acid Generating Potential (AGP) of sulphide­ bearing mine wastes. These methods are commonly criticised, however, in that they fail to yield important information relating to the reactivity of materials in terms of their ability to produce and neutralise acid with time. For instance, static tests mostly fail to consider the effects and interactions of variables such as pH, oxygen concentrations, bacteria, grain size (surface area) and temperature. These and other parameters are known to significantly influence the rate of sulphide oxidation under field conditions (Sorensen et al. , 1 980; Caruccio et al. , 1 988, Kwong, 1 995) . To elucidate the reactivity and AGP of mining wastes, kinetic tests are frequently conducted whereby samples are weathered under controlled conditions for periods of up to a year or more (Chapter 3 ; section 3 .4. 1 .3) . Several tailings and/or leachate parameters are generally monitored with time to gain an understanding of reaction kinetics. This chapter outlines the results of a simple kinetic experiment that was conducted on samples of the Tui mine tailings to assess the relevance of AGP prediction results obtained using static tests (Chapter 4; section 4.4). Furthermore, it was envisaged that the kinetic test outlined in this chapter would be useful in elucidating chemical and 269 mineralogical findings, reported in Chapters 4 and 5, that indicated that the surface tailings (0-200 mm) have undergone extensive oxidative-weathering and contain relatively low concentrations of acid generating sulphide minerals. 6.1 Aims & Hypotheses The primary aim of the lime incubation experiment reported in this chapter was to assess the relative 'reactivity' of six tailings samples collected at 1 00-mm incremental depths (0-600 mm) in terms of their ability to oxidise and release acidity and heavy metals. It was anticipated that the results of this incubation experiment would confirm that the surface tailings are comparatively ' inert' and may be readily amended with low rates of lime (CaC03) to prevent reacidification. In contrast the subsurface samples (400-600 mm) and particularly the 500-600 mm sample, were predicted to be most reactive under aerobic conditions and thus most likely to reacidify after the application of liming materials. An equally important objective of the experiment was to assess the efficacy of different rates of lime at preventing reacidification of the tailings and the release of selected heavy metals under conditions considered to be conducive to biochemical sulphide oxidation. The incubation experiment was therefore also aimed at quantifying the liming requirements of the tailings in terms of facilitating chemical conditions suitable for long term vegetative growth. 6.2 Experimental Design A completely randomised, factorially designed incubation experiment was used to investigate the interaction of various liming rates on the selected properties of tailings (Table 6. 1 ). Six liming rates were assessed, including a nil rate (control treatment) and five rates that were based upon the samples' individual pH buffering capacities (see section 6. 1 .3 . 1 ). Each treatment was replicated four times. 270 Table 6. 1 Design Treatments for Lime Incubation Experiment. Liming Rates Nil Yzj)H-6 I pH-6 2�H-6 4 pH-6 8 pH-6 Depths 6 6 6 6 6 6 Replicates 4 4 4 4 4 4 Total 24 24 24 24 24 24 (Total Number of Samples: 6 depths x 6 liming rates x 4 replicates == 144 ) 6.3 Methodology The Tui tailings material used for this incubation experiment corresponded to the six 1 00-mm incremental samples collected at grid sampling site 7 (Chapter 4; section 4.3 . 1 .2). Approximately 1 5 kg of tailings were obtained at 1 00-mm sampling depths for the incubation experiment. Prior to lime addition the samples were air-dried and passed through a 2 mm sieve to break up loosely aggregated particles. The bulk samples were then thoroughly mixed and weighed out (400g sub-samples) into a series of numbered polyethylene bags to which the lime amendments were added. 6.3.1 Lime Application Rates The 400 g tailings samples within the individual bags were treated with one of six rates of finely ground « 1 50!J.m) analytical grade calcium carbonate (lime). High quality, finely ground lime was used to ensure liming requirements were accurately met and that the effects of hydrous ferric oxide 'armouring' of limestone were minimised. The liming rates were based upon the pH buffering capacities of the samples, or more specifically, upon the amount of lime required to raise the pH of the samples to approximately 6.0 ( 1 pH-6 rate) . This liming rate was estimated for each of the six 1 00-incremental samples using a short-term pH buffering experiment. The buffering experiment (the methodology of which is outlined in Appendix 6(A)) was conducted over a 48 hour period and involved incubating 20g tailings samples with varying concentrations of dilute NaOH. The pH' s of the samples were plotted against NaOH concentrations to obtain sample-specific pH buffering curves from which the liming rates required to raise the pH of the samples to 6 ( 1 pH-6) were estimated. 271 None of the Tui tailings samples were anticipated to exhibit high pH buffering capacityl due to their negligible content of 2: 1 clay minerals and organic matter. Although XRD analysis detected the presence of appreciable quantities of kaolinite (A14Si40lO(OH)s) throughout the tailings profile, this 1 : 1 clay mineral develops very little surface charge via isomorphic substitution of Si4+ and thus exhibits low cation-exchange capacity ( 1 - 1 0 cmoleslkg) (Brownlow, 1 979; Pierzynski e t al. , 1 994). The results of the pH buffering experiment confirmed that the Tui tailings samples have little capacity to buffer against pH change, as indicated by the comparatively small volumes of dilute NaOH that were required to induce relatively large shifts in pH (Table 6.2). The pH buffering curves (Figure 6 . 1 a) showed that there were differences in the initial pH (Y intercepts) and the rate of change in pH with hydroxide addition (slope) between the samples. Figures 6. 1 b indicates that a relationship between pH buffering capacity and sampling depth exists. Tailings obtained at lower depths within the profile were found to exhibit a greater capacity to resist pH change than surface samples. The difference was most pronounced between the 0- 1 00 mm and the 400-500 mm samples which exhibited 1 pH-6 liming requirements of 1 . 1 7 and 7.00 Mg CaC031 1 000 Mg of tailings respectively (Figure 6 . 1 b). This trend is believed to reflect a combined increase in the relative concentrations of fine-grained materials (silt and very fine sand), the pH buffering mineral kaolinite and most importantly hydroxide-consuming sulphide minerals with depth. It should be noted that variations amongst the initial pH of the sanlples also influenced the magnitude of the various 1 pH-6 liming requirement values presented in Figure 6. 1 b . Tailings samples were also amended with l ime at rates of nil, half, one, two, four and eight times the 1 pH-6 liming rate (Table 6.2). It was postulated that in the presence of high concentrations of acid generating sulphide minerals (ie. at sub-200 mm depths) the 1 pH-6 liming rate would be insufficient to maintain the pH of the tailings at circumneutral levels for the duration of the experiment. Thus, the higher liming rates 1 The pH buffering capacity of a growing medium is an extremely important characteristic in that it has a s ignificant bearing on the availability of nutrients and phytotoxic metals to plants. A material' s capacity to buffer against pH change is largely governed by its clay and organic matter content, as these materials tend to have relatively high cation exchange capacities and thus have an inherent ability to release and absorb ' acidic' ions. 272 were included to try and identify an 'optimum' liming rate for the Tui tailings to maintain the pH for long term and to observe the effect of relatively high rates of lime on pH and availability of 'plant available' metals with time. With the exception of tailing obtained from the 0-200 mm depth zone, it was found that the 8 pH-6 liming rate (the highest used in the incubation experiment) was substantially lower than those theoretically required to prevent eventual reacidification of the tailings2 (Figure 6.2a). It was not, however, deemed necessary to treat subsurface (200-600 mm) tailings with rates of lime � TAGP rate, as the 8 pH-6 liming rate was predicted to be sufficiently high to prevent the various samples from reacidifying within the short to medium term (ie. for the 1 'li year period over which the experiment was run). Table 6.2 pH buffering capacity, T AGP (theoretically acid generating potential) upper and lower limits and various liming rates based on the 1 pH-6 buffering rate. Depth pH TAGP Nil � pH-6 1 pH-6 2 pH- 4 pH-6 [mm] 0- 1 00 1 00-200 200-300 300-400 400-500 500-600 Buffering t t t t 6t * Lower Upper 1 5 .6 6.6 1 0.0 0 0 .59 1 . 1 7 2.34 22.8 8.2 12 .4 0 0 .88 1 .75 3 .50 24.4 75.4 1 1 3 .9 0 1 .47 2.94 5 .88 26.3 75.8 1 1 4.3 0 1 .7 1 3 .41 6.82 55.2 7 1 .4 1 08.2 0 3 .50 7 .00 14.00 53.9 1 1 3 .0 1 7 1 .0 0 3 . 1 6 6.32 12 .64 -* T . . = = unIts rnrnol OH /kg/pH, umts Mg CaCOi1 000 Mg taIlmgs 6.3.2 Water Regime t 4.68 7.0 1 1 .76 1 3 .64 28 .00 25.28 8 pH-6 t 9.36 14.0 23 .52 27.28 56.00 50.56 After lime incorporation the moisture content of the admixes were adjusted to approximately 70% of the 0.05 bar water content value3 using distilled water. The handling nature and aeration of the 200-300 mm , 300-400 mm and 400-500 mm samples was further improved by reducing the water content to between 60-65% of the 0.05 bar value (Table 6.3) . Maintaining the samples in a moist but largely unsaturated condition�s designed to provide conditions conducive to the biochemical oxidation of sulphide minerals constituting the tailings. 2 As estimated by TAGP prediction outlined in Chapter 4 (section 4.4.2. 1 ) 3 Determined using Haine's apparatus. Table 6.3 Percentage water content (oven dry basis) of the amended tailings at 0.05 bar pressure. 70% (0.05 bar) values and the approximate water content used in the Lime Incubation Experiment are also provided. Depth Water content 70% 0.05 bar Approx. Water [mm1 at 0.05 bar content Content Used 0- 1 00 29.4 % 20.6 % 21 % 1 00-200 27.6 % 19 .3 % 20 % 200-300 28.5 % 20.0 % 1 8 % 300-400 30.6 % 2 1 .4 % 1 8 % 400-500 32.3 % 22.6 % 1 8 % 500-600 29.8 % 20.9 % 2 1 % 273 During the first 8 weeks of the experiment the water content of the samples were checked twice weekly and the admixes were aerated by trapping air within the polyethylene bags and vigorously shaking for 1 0 to 20 seconds. During the final stages of the experiment, the bags were watered and aerated once per week. Throughout the experiment an attempt was made to ensure that the 'neck' of the polyethylene bags remained open so as the samples were in contact with atmospheric oxygen. This practice was designed to enhance the aerobic oxidation of the tailings. Plate 6.1 Lime Incubation Experiment January 1 995. Polyethylene bags containing lime amended tailings 274 6.3.3 Incubation Environment During the lime incubation experiment the amended tailings samples were stored in a glasshouse maintained at an ambient temperature 25°C (Plate 6 . 1 ) . The tailings were, however, subject to diurnal temperature variations of between 1 0- 1 5°C in the glasshouse, particularly during the summer months. The polyethylene bags containing the amended tailings were arranged in randomised order and rotated weekly to minimise the positional effects within the glasshouse associated with temperature and lighting. In laboratory leaching experiments Geidel ( 1 979) demonstrated that the time interval between leaching events on pyrite does not affect the rate of pyrite oxidation. Pyrite is thus believed to convert to oxidation products at a constant rate independent of the concentration of oxidation products in the ' soil solution' . Burt & Caruccio ( 1 986) suggested that this behaviour is expected of an irreversible process limited by the rate of the oxidation mechanism4. Geidel' s ( 1 979) research indicates that the use of a non­ leaching environment in the present study may not have substantially effected the rate of sulphide oxidation. 6.3.4 Monitoring of the Treatments The lime incubation experiment was commenced in July 1 993 and terminated 76 weeks later (1 Yz years), in January 1 995 . During this time, periodic sampling was conducted to monitor pH and determine the concentrations of 'plant available' heavy metals with time. Sulphate (S04 levels were also assessed in the treated tailings 1 6 weeks after the commencement of the experiment (Table 6 .4). The moisture contents of the samples were adjusted periodically to accommodate diminishing sample masses resulting from sub-sampling. 6.3.4.1 pH The pH of the amended tailings in each of the 4 replicate bags was measured 13 times during the course of the experiment using a 1 :2.5 tailings to water ratio as per the method outlined in Chapter 4 (section 4. 1 . 1 . 1 ) . To reduce processing time prior to sampling, the water contents of the amended tailings were adjusted to their required levels (60 to 70 % field capacity) and left to equilibrate for 3 0 minutes. Weight adjusted 275 sub-samples (the equivalent of 1 0 .0 g of air-dried material) were then measured out into propylene beakers. Samples of standardised soil were included with each batch for comparative purposes. 6.3.4.2 Metals Samples from 2 of the 4 replicate bags were analysed for concentrations of 0. 1M Hel acid extractable eu, Fe, Pb and Zn on four occasions during the course of the experiment (Table 6.4). The extraction was carried out according to the method outlined in Appendix 4(F) and the resulting supernatant solutions were analysed using a GBe 904AA atomic absorption spectrophotometer. Recommended standard slit width, flame types, lamp currents and wavelengths were used (Table 4.9). 6.3.4.3 Sulphate The concentrations of O . IM Hel acid extractable sulphate determined colourimetrically using an auto-analyser (Johnson & Nishita, 1 952). Again, only 2 of the 4 treatment replicates were analysed. Table 6.4 Sampling times for Lime Incubation Experiment. Weeks & Sampling Dates Parameter 0 1 2 3 4 6 8 12 1 6 24 32 48 76 Measured 2 1 .7.93 28.7 . . 93 4.08.93 1 1 .8.93 1 8.8.93 1 .09 .93 1 5 .9.93 1 3 . 1 0.93 1 0. 1 1 .93 5 . 01 .94 2.03.94 22.6.94 5.0 1 .95 S04= ,/ Metals ,/ ,/ ,/ ,/ pH ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ ,/ • Metals analysed include O. lMHCI extractable Cu, Fe, Pb and Zn 4 See Chapter 3 (section 3 . 3 . 1 ) for discussion of sulphide oxidation and the rate detennining reactions. 276 1 1 Figure 6.1(a) Sample Depth [mm] 3 o 1 00 S 200 S 500 600 0.00 • o 0.50 1 .00 1 .50 mmol of hydroxide added/20g tailings Figure 6.1 (b) • 2 4 6 Mg CaC03 11000 Mg Material • 2.00 8 Figure 6.1 (a) pH buffering curves for tailings at various sampling depths . (b) Relationship between profile depth and the amount ofl ime required to attain a pH of 6 at each depth increment. This figure clearly demonstrates that the pH buffering capacity of the tailings tended to increase with depth. 0-1 00 • 1 00-200 200-300 • 300-400 400-500 .. 500-600 8 E -= .... Q. � � � is <:) 10.. � 0 1 00 200 300 400 500 600 o 600 277 Liming Requirements [Mg CaC03 /1000 Mg tailings] o 50 1 00 1 50 200 Oi.+ ; , \ \ i<>:.: <> --! I 1<) <> I <> 0.0 • • (a) Liming Requirements � 1 pH-6 Rate -+- 8 pH-6 Rate TAGP (low) TAGP (high) Acid Generation [Mg CaC03 /1000 Mg of tailings1 0.5 1 .0 1 .5 LSD(5%) - 2.0 (b) Liming Rates ---.A- Nil Rate 1 12 pH-6 Rate -II- 1 pH-6 Rate 2 pH-6 Rate 4 pH-6 Rate -e- 8 pH-6 Rate Figure 6.2 (a) Graph comparing Theoretical Acid Generating Potential (T AGP) and the rate of lime required to amend the pH of the tailings to 6.0 (I pH-6 Rate). The 8 pH-6 rate represents highest rate of lime used in this trial. (b) Acid generated by the incubation samples over 76 week period calculated from hydrogen ion liberation. 6.4 Results 6.4.1 Variation in pH and Metal Levels with Time 6.4. 1 . 1 pH 278 The pH versus time curves presented in Figures 6.3 and 6.4 clearly indicate that under non-leaching conditions H+ ions were liberated from all control samples (non-limed) during the course of the experiment. The greatest drop in pH was noted in the 300-400 rum sample where the pH decreased from approximately 3 .4 to less than 2 . 1 in 3 8 weeks (Figure 6 .4a). The decrease in pH amongst the other control samples ranged from 0.8 to 1 .2 . Whilst the application of lime at the Yz pH-6 rate initially increased the pH of each tailings sample to at least 5 .5 , the pH of all tailings samples (with the notable exception of the 0-1 00 rum) dropped to levels comparable with the control treatment within 1 6 to 24 weeks of lime application (Figures 6.3 & 6.4). The application of lime at the 1 pH-6 rate similarly failed to maintain the pH of the tailings at circurnneutral levels for the duration of the experiment. By week 48 the pH of all samples (again with the exception of the surface sample) had declined to levels approaching those of the control treatments (Figures 6 .3 & 6.4). Whilst the 2 pH -6 liming rate adequately maintained the pH of most of the treated tailings samples at levels greater than 6, the pH of tailings obtained from the 300-400 rum depth began to rapidly fall between weeks 24 and 32 of the experiment (Figure 6 .4a) . By week 76 of the experiment the average pH of this treatment had dropped almost 5 pH units to pH 2.03, a level lower than (but not significantly different from) that of the control treatment. In contrast to the preceding liming rates, the 4 pH-6 and 8 pH-6 lime treatments were found to adequately maintain the pH of all amended tailings at levels in the vicinity of 7 .5 for the duration of the experiment (Figures 6.3 & 6.4). At week 76 of the experiment there was no evidence to suggest that tailings treated with either of these rates of lime were beginning to reacidify. 279 An observation made during the course of the experiment that is not portrayed in Figures 6.3 & 6.4 was that while treatments were in a state of rapid pH decline, replicate samples often exhibited pH values that varied by more than 2 or 3 pH units. Without exception, however, the pH of the replicated samples were eventually found to reach an equilibrium pH very close to one another. To cite a specific example, at week 24 of the experiment, the 200-300 mm sample treated with lime at the 1 pH-6 rate was found to contain 2 replicate samples with pH values of 2 .63 and 6.47. By week 48 of the experiment, however, these samples exhibited pH values of 1 .84 and 1 .85 respectively. Like most samples beyond week 48 of the experiment the tailings generally exhibited little further pH change. The non-synchronous pH decline observed between some replicate samples may have resulted from the uneven establishment of bacterial populations responsible for the catalytic oxidation of pyrite. Graphs presented in Figures 6.3 & 6.4 show that where acidification of samples occurred, pH levels did not drop below about 2 regardless of the initial rate of pH decline or lime treatment. The buffering of sulphide bearing materials at similar pH levels has been reported in previous studies (Miller, 1 979; Wig gering, 1 993) and has been attributed to the dissolution of jarosite (KFe3(OH)6[S04]z), kaolinite and silica. Research indicates that even when sulphide-bearing materials are incubated in polyethylene bags under moist non-leaching conditions, rarely will pH levels fall below 1 . 5 (Dent, 1 986). In acid-sulphate soils, acid hydrolysis of alumino silicate clays is considered to be the primary pH buffering mechanism (Dent, 1 986). In colliery spoil, no strong buffering systems generally exist between pH 6.5 and 3 .5 , thus pH levels tend to drop very rapidly from circurnneutral levels to strongly acid conditions upon exhaustion of acid neutralising carbonate minerals (Wiggering, 1 993; Walder & Chavez, 1 995). The Tui tailings are similarly considered to be devoid buffering systems within this pH zone and to be buffered at low pH chiefly by kaolinite and possibly j arosite and silica dissolution. 5 :r: (a) 0·100 mm � 4 ' 3 ' �"'_---"*I<--____ *___ ------.6A :r: � :r: � 2 0 8 1 6 24 32 40 48 Weeks (b) 1 00·200 mm 8 i ��� � �r � 7 C? '� � � - � 6 'T • 5 G 4 t-- -=------ • 56 3 """, 8, � 2 . �� 1 0 8 1 6 24 32 40 48 56 Weeks (c) 200·300 mm 8 �� =� • 7 " -. ,j, i --� 6 '� ) " ". 5 4 i :'��"-- .. � 0 8 1 6 24 32 40 48 56 Weeks 64 72 • II 64 72 !. � II 64 72 280 80 80 80 Figure 6.3 pH versus time in (a) 0-1 00 mm sample, (b) 1 00-200 mm sample and (c) 200-300 mm sample. Liming rates: nil (.6.), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (D), 4 pH-6 (\7 ) and 8 pH-6 (e). 5 :x: Q. 4 2 1 2 1 8 2 • • • o 8 1 6 24 o 8 1 6 24 o 8 1 6 24 (c) 300-400 mm 32 40 48 Weeks (b) 400·500 mm 32 40 48 Weeks (c) 500- 600mm 32 40 48 Weeks 281 56 64 72 80 56 64 72 80 56 64 72 80 Figure 6.4 pH versus time in (a) 3 00-400 mm sample, (b) 400-500 mm sample and (c) 500-600 mm sample. Liming rates: nil (A.), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (0), 4 pH-6 (V ) and 8 pH-6 (e). 282 6.4.1 .2 Metal Concentrations In contrast to sample pH, concentrations of O . lM HCl extractable Cu, Fe and Zn were generally found to rise or remain relatively constant throughout the course of the experiment (Figures 6 . 5 to 6 . 1 0). The availability of these metals appears to have been related to both sampling depth and in some cases treatment. Metals were generally most readily available in samples below 200 mm depth and in tailings treated with the nil, Y2 pH-6 or 1 pH-6 liming rates (Figures 6.5 to 6 . 1 2) . (i) Cu In the case of Cu, the increase in the availability of this metal with time tended to be slightly higher in the control and 12 pH-6 lime treatments in samples obtained from the surface 300 mm of the profile (Figure 6.5). Tailings sampled from deeper in the profile, however, failed to demonstrate this same trend (Figure 6.6). One notable trend evident in Figures 6 .6 b,c is the significantly lower concentrations of O. lM HCl extractable Cu in the 8 pH-6 lime treated 400-600 mm zone tailings. The application of this relatively high rate of lime was found to reduce Cu levels from about 1 000 mg/kg (or more) to less than 1 00 mg/kg in these samples (Figures 6 .6 b,c). In contrast to the 8 pH-6 liming rate, the 2 and 4 pH-6 rates appeared to actually increase the amount of O . lM HCl extractable Cu at corresponding depths. It is postulated that the low availability of Cu in the 8 pH-6 treatments reflects the presence of excess, unreacted lime in the tailings which would have reduced the acidity of the O . IM HCI extractant and therefore its ability to solubilise and extract Cu. (ii) Fe Like Cu, extractable Fe levels generally remained constant or increased during the course of the experiment depending upon treatment and depth (Figures 6 .7 & 6.8). Whilst Fe levels in the uppermost sample (0- 1 00 mm) remained relatively constant throughout the experiment regardless of treatment (Figure 6 .7a), Fe levels in subsurface samples ( 1 00-600 mrn) that received either 12, nil or 1 pH-6 lime rates were substantially higher at week 48 of the experiment than at week 0 (Figures 6 .7 b,c & 6.8). The greatest ' liberation' of Fe occurred in the 200-300 mm sample where O . lM HCl extractable Fe concentrations in the liz pH-6 treatment increased by about 5000 mglkg 283 during the course of the experiment (Figure 6.7c). At depths where relatively high levels of O . 1MHCl extractable Fe were initially present the effects of liming on Fe availability were immediately evident (Figures 6.8 a,b,c). Like Cu, levels of extractable Fe clearly tended to be much lower in those samples treated with the 8 pH -6 liming rate. In the 400-600 mm sampling zone the availability of Fe in the 2 pH-6 and 4 pH-6 treatments was also substantially lower (Figures 6 .8 b,c). The 3 highest rates of lime were similarly appeared to prevent O . lM HCI extractable Fe levels from appreciably increasing in the 1 00-200 and 200-300 mm samples during the course of the experiment (Figures 6 .7 b,c). (iii) Zn In keeping with Cu and Fe, the levels of extractable Zn were low in the uppennost sample (0- 1 00 mm) and remained relatively constant throughout the experiment regardless of treatment (Figure 6 .9a). In most other samples, however, O . IM HCI extractable Zn levels increased with time and were greatest in samples treated with either of the 3 lowest rates of lime (nil, 12 pH-6, 1 pH-6). The difference between these rates is most readily observed in the sub-300 mm samples (Figures 6 . 1 0 a,b,c) where substantial increases in Zn are noted at week 48 of the experiment. (iv) Ph In contrast to O. IM HCl extractable Cu, Fe and Zn concentrations, the availability of Pb in most treatments and at most depths declined with time (Figures 6. 1 1 & 6. 1 2) . This trend is most readily observed in Figures 6 . 1 1 b in which Pb levels are seen to fall from about 1 000 mg/kg at week 0 to between 1 00 and 400 mg/kg by week 32 of the experiment. In most Pb plots it is apparent that where low quantities of lime were added to the tailings, available Pb levels were correspondingly low (Figures 6. 1 1 & 6. 1 2) . This trend is readily seen in Figures 6 . 1 1 b and 6 . l2a and once again contrasts with trends generally observed for Cu, Fe and Zn (Figures 6.5 to 6 . 1 0) . Whilst the addition of lime at the 8 pH-6 rate seemed to increase the availability of Pb (relatively to other treatments) at most depths, within the 400-500 and 500-600 mm samples the addition of I t!�1t this rate substantially reduced the extractability of Pb 50 40 Ci 30 .x: C, E ...... <3 20 1 0 0 0 50 40 Ci 30 .x: C, .§. <3 20 1 0 0 0 200 50 o o 8 1 6 8 16 8 (a) 0-100 mm 24 Weeks (b) 100-200 mm 24 Weeks (c) 200-300 mm 1 6 24 Weeks 284 32 40 48 32 40 48 32 40 48 Figure 6.5 O. lMHCl extractable Cu versus time (a) 0-1 00 mm sample, (b) 1 00-200 mm sample and (c) 200-300 mm sample. Liming rates: nil (A), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (D), 4 pH-6 ('\7 ) and 8 pH-6 (e). (a) 300-400 mm 2000 ! 500 0 0 8 1 6 24 32 40 48 Weeks (b) 400-500 mm 2500 2000 0; 1 500 � C, .§. (5 1000 500 0 0 8 1 6 24 32 40 48 Weeks (c) 500-600 mm 2000 1 500 SOD O�==�==��======��========� a 8 1 6 24 Weeks 32 40 48 285 Figure 6.6 O . lMHCl extractable Cu versus time (a) 300-400 mm sample, (b) 400-500 rum sample and (c) 500-600 rum sample. Liming rates: nil (.A), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (0), 4 pH-6 (\1) and 8 pH-6(e). (a) 0-100 mm 300 o ------------------------------------- o 1 500 1 250 1 000 ]i E 750 Q) u.. 500 ! 250 o o 6000 5000 4000 i Ol ..:x: 0, E 3000 I .... Q) u.. 2000 o o 8 8 8 1 6 1 6 1 6 24 Weeks (b) 1 00-200 mm 24 Weeks (b) 200-300 mm 24 Weeks 32 40 48 32 40 48 32 40 48 286 Figure 6.7 O. 1MHCI extractable Fe versus time (a) 0- 1 00 mm sample, (b) 1 00-200 rum sample and (c) 200-300 nun sample. Liming rates: nil CA), halfpH-6 CO), 1 pH-6 C-), 2 pH-6 (0), 4 pH-6 CV ) and 8 pH-6 (e). 4000 i 3000 o 2500 500 4000 3000 01 x ]:2000 Q) u.. o o 8 1 6 8 16 8 1 6 (a) 300-400 mm 24 Weeks (b) 400-500 mm 24 Weeks (e) 500-600 mm 24 Weeks 32 32 32 287 40 48 40 48 40 48 Figure 6.8 O . lMHCI extractable Fe versus time (a) 300-400 nun sample, (b) 400-500 nun sample and (c) 500-600 nun sample. Liming rates: nil (A.), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (D), 4 pH-6 (\7 ) and 8 pH-6(.). (a) 0-1 00 mm 60 I 40 01 .:.: � � ..... 'f c: N 20 o �----------------------------__________ _ o 8 1 6 8 0 i 20 : 24 Weeks (b) 100·200mm 32 40 48 o -------------------------------------------- o 8 1 6 150 o o 8 1 6 24 Weeks 32 (e) 200·300 mm 24 Weeks 32 40 48 40 48 288 Figure 6.9 O. lMHCI extractable Zn versuS time (a) 0- 1 00 rum sample, (b) 1 00-200 rum sample and (c) 200-300 rum sample. Liming rates: nil C£.), halfpH-6 (0), I pH-6 (-), 2 pH-6 (D), 4 pH-6 (\7 ) and 8 pH-6 (e). (a) 300-400 mm 1 500 1200 0 0 8 16 24 32 40 48 Weeks (b) 400-500 mm 1 500 1 200 � 900 I c, .§. c 600 N 300 0 0 8 1 6 24 32 40 48 Weeks (c) 500-600 mm 2000 1 500 500 o�====�--------------- ------------- o 8 1 6 24 Weeks 32 40 48 289 Figure 6.10 O. lMHCl extractable Zn versus time (a) 300-400 mm sample, (b) 400-500 mm sample and (c) 500-600 mm sample. Liming rates: nil (.6.), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (0), 4 pH-6 (\7 ) and 8 pH-6 (e). (a) 0-100 mm 1 500 300 o ''------------------------- o 8 1 6 1 500 1 200 o o 8 1 6 400 1 00 o o 8 1 6 24 Weeks 32 (b) 100-200 mm 24 Weeks (e) 200-300 mm 24 Weeks 32 32 40 48 40 48 40 48 290 Figure 6. 1 1 O . lMHCI extractable Pb versus time (a) 0-1 00 nun sample, (b) 1 00-200 nun sample and (c) 200-300 nun sample. Liming rates: nil (.A), halfpH-6 CO), 1 pH-6 C-), 2 pH-6 (D), 4 pH-6 (\7 ) and 8 pH-6 (e). 200 ! 1 50 o o 200 1 50 Ci .:.:: C, .§. 1 00 ..Q a. 50 8 1 6 (a) 300-400 mm 24 Weeks (b) 400·500 mm 32 40 48 o ����--- --- - -- -===�=========-__ o 8 16 400 i o 8 16 24 Weeks 32 (c) 500·600 mm 24 Weeks 32 40 48 40 48 29 1 Figure 6.12 O . lMHCl extractable Pb versus time (a) 300-400 rom sample, (b) 400-500 rom sample and (c) 500-600 rom sample. Liming rates: nil (.t..), halfpH-6 (0), 1 pH-6 (_), 2 pH-6 (0), 4 pH-6 (V ) and 8 pH-6 (e). 292 (Figures 6. 1 2 b,c). In this respect trends of Pb availability in the 400-600 mm zone were similar to those of Cu, Fe and Zn in 8 pH-6 treatments and may again have resulted from the extractant being effectively 'neutralised' due to the presence of excess lime and the precipitation of carbonates under highly alkaline conditions. 6.4.2 Variation in pH, Sulphate and Metal Levels with Depth and Liming Rate The variation in pH and extractable metal concentrations with depth are most readily observed in Figures 6 . 1 3 to 6. 1 6 which present parameter values versus depth on 4 separate occasions during the course of the experiment (weeks 0 , 1 6, 32 & 48). Although as discussed in the previous section, treatment pH and extractable metal concentrations were in some samples highly time dependent, the overall trends with depth were consistent at most sampling times. The following discussion on the effects of lime and depth on pH and metal concentrations is based primarily on data (plots) presented in Figures 6 . 1 3 & 6 . 1 6 (weeks ° and 48 of the experiment respectively) and upon ANOV A conducted on data presented in Figures 6 . 1 3 to 6 . 1 6 . 6.4.2.1 pH Figures 6. 1 3e & 6. l 6e clearly indicate that pH levels were lowest in the 200-300 mm control sample. Tailings above and below this depth were consistently less acidic that this sample throughout the experiment. Whilst the addition of lime to the tailings was initially found to significantly increase the pH of tailings at each depth to levels in excess of pH 5, by week 48 of the experiment the effects of lime application at the Yz pH-6 and 1 pH-6 rates were insignificant except in the surface sample (Figure 6. 1 6e). By week 76 of the experiment the pH of the 300-400 mm tailings sample treated with the 2 pH-6 lime rate had similarly fallen to levels comparable with that of the control treatment (Figure 6.4a). Incremental samples treated with the 4 and 8 pH-6 liming rates, however, exhibited pH values that were consistently significantly higher than those of the nil, Yz and 1 pH-6 treatments. Depth-wise pH variation was insignificant at the two highest liming rates. ANOVA indicated that there was a significant interaction (P 0 a.> '+= 400 0 '- 0.. 600 0 E ..s 200 .r:::. -0-a.> 0 a.> '+= 400 0 "- 0.. 600 0 E ..s 200 .r:::. -0-a.> 0 a.> '+= 400 0 '- 0.. 600 0 .� . I : II 0 (a) 600 �/ (c) Cu [mg/kg] 1 200 1 800 2400 LSD - - '� . " � '� ::.. , � /'/ ////./ / / ////� / / / . . / ./ C'li: vc::J Pb [mg/kg] 500 1 000 1 500 ��. �v � , . r l " " 'i i�' LSD - ' /, � \/! ff '� Ii i , r ,1 , \ . : l. ' i' \ '1ZJ � 2 (e) pH 4 6 8 ,.. � � � " _______ �---..----- I i ''fi�'r c '» I, ! LSD - � I i • �� � I , , '» � I , ,. " 1 i ' l " ! ' \ " .t. :::::. 0 E ..s 200 .r:::. 0.. a.> 0 a.> '+= 400 0 L- 0.. 600 0 E ..s 200 .r:::. -0-a.> 0 a.> � 400 0 '- 0.. 600 300 (b) Fe [mg/kg] 0 1 000 2000 3000 J�� I � " LSD - 1 \\ iIIS'J" V . /� K ' // --, // / r 4iJE 0 �;::;/ r,,:�:;-, '& " : " • 8 (d) Zn [mg/kg] 500 1 000 1 500 LSD -i �"' t "�'" ';r � /// \ i . ./ � j I , � I • "'II: i - Liming Rates -A- Nil Rate 1 /2 pH-6 Rate -II- 1 pH-6 Rate 2 pH-6 Rate 4 pH-6 Rate -e- 8 pH-6 Rate Figure 6. 15 Variation of pH and O. IM Hel extractable metal concentrations with depth and liming rate at week 32. LSD = DepthiRate Interaction (5%) 4000 2000 301 (a) Cu [mg/kg] (b) Fe [mg/kg] 0 1 000 2000 3000 0 1 500 3000 4500 6000 : 0 0 � E .s 200 .s::::. -Q. (J) 0 (J) 400 l;::: 0 '- 0... 600 0 E .s 200 .s::::. -Q. (J) 0 (J) t;::: 400 0 .... 0... 600 o E .s 200 .s::::. -Q. (J) o (J) i;:; 400 e 0... LSD - � I �",/ -j . ./ //'" C � �! . �/ d7 ;. (c) Pb [mg/kg] 0 500 1 000 �r P? , �� -j / 1/ LSD -1/ t� -�!!� / \, � ! \ ,� , 75 % gennination with seedlings exhibiting fully-expanded, green cotyledons were classed as having high gennination. The medium, low and nil classes represent 25 - 75% gennination, < 25% gennination and nil germination respectively. 7.3.2 Trial Results (i) Germination No obvious differences in germination time were noted amongst the five plant taxa. The majority of seeds germinated within ten days of sowing. With the exception of 3 Synonymous with Agrostis tenuis. 32 1 A. capillaris var. Goginan there was a high germination rate for all other taxa on the nil treatments (distilled water). This indicated that a significant proportion of the Goginan seed was non-viable. The low viability of the seed was confirmed by germination tests run subsequent to the trial. These tests indicated that only about 30% of the Goginan seed was viable compared with 93% and 90% for A. capillaris Parys Mountain and Egmont respectively and 85% and 78 % for the two varieties of F rubra (Merlin and Lobi). (ii) Lead & Zinc Tolerance With the exception of A. capillaris (Goginan), all five plant taxa were found exhibit high germination rates in solutions of up to 1 00 mgIL Pb and Zn nitrate. Despite expected tolerance to relatively high concentrations of Pb, Goginan exhibited only medium germination and vigour in the 0, 5 , 1 0, 20 and 50 mgIL Pb treatments. A visual assessment of the 1 00 mgIL Pb treatment indicated that less than 1 0% of the Goginan seed germinated. (iii) Copper Tolerance Of the three heavy metals investigated, eu was found to have the most detrimental effect on the seed germination, with three out of the five plant taxa exhibiting nil or low germination at eu concentrations above 1 0 mgIL (Appendix 7(A)). The only variety to germinate above 20 mg/L eu was Merlin, the metal-tolerant cultivar of F rubra. This taxa demonstrated high germination at 50 mgIL, however, no germination was evident at 1 00 mg/L. Of the two metal-tolerant varieties of A.capillaris, var. Parys Mountain again faired better that Goginan. Results suggest that the germination threshold for Goginan lies between 10 and 20 mg/L eu as indicated by low germination at 10 mgIL and no germination at 20 mgIL. Parys Mountain exhibited medium germination at 20 mgIL eu, however, no germination was evident on the 50 mgIL treatment. The non metal-tolerant varieties of F rubra and A. capillaris showed little tolerance to eu concentrations above 1 0 mg/L, with Lobi failing to germinate and Egmont demonstrating low germination only at 20 mg/L (Appendix 7(A)). -322 7.3.3 Discussion of Germination Results It is apparent that a greater range of solution concentrations were required to fully investigate the tolerance of the plant taxa to Pb and Zn. High rates of germination were noted for all seeds on the Pb and Zn treatments with the notable exception of A. capillaris var. Goginan. In the case of Cu, the 0- 1 00 mg/L range used in the experiment proved adequate to assess the tolerance of the seed to this metal, as Cu was found to inhibit seed germination at much lower concentrations than Pb and Zn. It is widely accepted that Cu tends to be more phytotoxic than Pb and Zn due to its tendency to displace other metal ions and particularly Fe from physiologically important centres (Neuman et ai. 1 987; Mengel & Kirkby, 1 98 7) . Based on experimental results the taxa seed have been placed in the following predicted order of increasing tolerance to heavy metals : Most Tolerant +- Least Tolerant Merlin < Parys Mountain < Egmont < Lobi < Goginan F. rubra var. Merlin and A. capillaris var. Parys Mountain were found to exhibit greater tolerance to heavy metals than their equivalent pastoral varieties (Egmont and Lobi). The poor performance of A.capillaris var. Goginan is believed to reflect the low viability of the seed. Plants germinated from low viability seeds often exhibit poor vigour and thus tend to fair relatively poorly in adverse growing conditions. Based on the results of this trial it was decided to exclude Agrostis capillaris var. Goginan from the plant growth trials due to its low seed viability and poor performance. A.capillaris var. Parys Mountain was obviously likely to show greater tolerance to the extreme growing conditions provided by the tailings. The characteristics of the various taxa used in the trials are outlined below (Table 7.4 ) . 323 Table 7.4 Description of Plant Taxa used in Glasshouse Trials. Plant Taxa Characteristics Agrostis capillaris A. capillaris is a low fertility perennial grass often found growing on acidic to neutral var. Egmont substrates in temperate regions. It is generally shade tolerant and moderately resistant (bent grass, browntop) to drought (Bennett et aI., 1 978). Most varieties do well under heavy traffic, however the Egmont variety is best suited to amenity areas that do not get severe wear (DSIR Grasslands, 1989). Agrostis capillaris A. capillaris var. Parys Mountain (PM) is classified as a pseudometallophyte because var. Parys Mountain popUlations can be found on both metalliferous and 'normal' sites (Antonovics et aZ., (bent grass, browntop) 1 97 1) . The PM variety was isolated from the Parys Copper Mine located in the Anglesea region of north em Wales. It is commercially available and has been used extensively in Great Britain to cover metalliferous wastes (Smith & Bradshaw, 1 979; Whiteley & Williams, 1 993). PM is recognised mainly for its Cu tolerance, although it does exhibit some tolerance to other metals including Pb and Zn. Metal-tolerant bent grass has been shown to successfully root in solutions containing up to 0.5 mgIL Cu and in tailings containing 1 % Pb and 1% Zn (Bradshaw & Chadwick, 1 980). Festuca rubra F rubra is a low fertility grass that is generally found on neutral or slightly calcareous var. Lobi substrates . It is resistant to both drought and shade, developing a deep fibrous root (Chewing or red fescue) system ideal for holding soils in place (Bennett et al., 1 978). Festuca rubra Although this taxa has been isolated growing in substrates of low pH (4.5), Frubra var. Merlin var. Merlin generally prefers alkaline growing conditions. This cultivar was i solated (Chewing or red fescue) on abandoned mine-sites in the UK and was specifically bred for revegetating Pb and Zn contaminated alkaline substrates (Bradshaw & Chadwick 1 980; Gemmell, 1 977; Bradshaw & McNeilly, 1 9812. Lotus corniculatus L. corniculatus is an acid tolerant, low ferti lity perennial forage legume used for var. pasture, hay and silage in the US and in grass-legume mixtures on disturbed land (Birdsfoot trefoil) areas, particularly acid strip mine soils. It is adaptable, growing on poorly drained, droughty and both acid and alkaline soils. Isolated plants have reportedly been found growing on acid strip soils with pH 3 .5 (Bennett et al., 1 978). L. comiculatus is useful for vegetative stabilisation as it often exhibits well developed taproot systems with numerous lateral branches. Rooting depths can exceed I m. The rationale for the use of metal-tolerant plants and the mechanisms and limitations of metal tolerance are reviewed in Chapter 3 (section 3 .7 .5) . 7.4 Sowing Rates & Dates In both pot trials the equivalent to 1 00 viable seeds were sown into each of the pots. The variation in seed viability4 meant that the number of seeds sown had to be varied for each species. The total number of seeds required for each species were counted and weighed in triplicate to attain an average weight for each taxa (Table 7.5) , The seeds for each treatment were then weighed-out and placed in 4 ml vials until the time of sowing. The seeds of the two Agrostis capillaris varieties were mixed with a small quantity of quartz sand (2 grams) prior to sowing to improve seed distribution. In addition to 4Seed viability tests were conducted independently by the Seed Technology Centre at Massey University. 324 regular watering, the surface of the pots were sprayed with distilled water at least once a day during the germination period to ensure the seeds were kept moist. Table 7.5 Seed viability and sowing rates for each of the plant taxa. Seeds per pot = 1 00/ proportion viable seed Plant Taxa Approximate Seed Seeds Mass of Sowing Number of viability required seeds rate seeds/gram [%] per pot [gJ [kglha] A. capillaris 1 2000 93 1 08 0,009 7 (Parys Mountain) A. capillaris 1 4000 90 1 1 1 0.008 6 (Egmont) F rubra 1 1 00 85 1 1 8 0. 1 1 1 84 (Merlin) F rubra 1 1 00 78 1 28 0. 1 1 8 89 (Lobi) L. corniculatus 833 53 1 89 0.227 1 7 1 After the 3 week incubation period which commenced on the 26-01 -95 the various treatments were transferred to the trial pots and the seeds were sown on the 1 6-02-95. The trials were terminated 8 weeks later, on the 1 3-04-95 when the final herbage cut was made. The unsown control pots were maintained for a further 1 3 weeks after harvesting and finally sampled on the 1 3 -7-95, 7.5 Monitoring of the Growth Medium During the course of the trial a number of tailings properties were measured to assess chemical changes in the growth medium with time. The properties measured included, pH, electrical conductivity (BC), exchangeable Ca, K, Mg and Na, cation exchange capacity (CEC), Olsen P, calcium phosphate extractable SO/ concentrations, O . 1MHCl and 0 .02M CaCh extractable metal (As, Cd, Cu, Fe, Pb, Mn, Zn) concentrations and 1M KCI and 0 .02M CaCh extractable Al levels (Table 7 .6) . Monitoring of the trial concentrated primarily on assessmg the tailings chemical conditions in treated but non-sown pots (control pots). Four such pots were required for Trial One and 8 for Trial Two. The control pots were stored and treated similarly to sown pots but because the control pots were destructively sampled, water addition to 325 these pots had to be adjusted after each sampling to ensure that they were maintained at the desired water content (section 7 .2 . 1 ). Table 7.6 Sowing, termination and sampling dates for Glasshouse Trials. * The fertility measurements include Olsen P, CEC, exchangeable Ca, K, Mg and Na. t Metals include O. lMHCI and O.02MCaCI2 extractable As, Cd, Cu, Fe, Pb, Mn and Zn levels and IMKCI and O.02M CaCh extractable Al concentrations. Parameters Week 0 Week 3 Week 4 Week 5 Week 7 Week 9 Week 1 1 Week 24 26-1 -95 16-2-95 23-2-95 2-3-95 16-3-95 30-3-95 1 3 -4-95 1 3-7-95 (Sowing) (Final Cut) F ertili:tY.* ./ ./ Metalsr ./ ./ ./ S042- ./ ./ ./ EC ./ ./ ./ IJH ./ ./ ./ ./ ./ ./ ./ ./ 7.5.1 pH The pH of the amended control pots were measured 7 times during the course of the two trials using a 1 :2 .5 tailings to water ratio as per the method outlined in Chapter 4 (section 4 . 1 ) . To reduce processing time of the samples, prior to sampling, the water content of the various treatments were adjusted to 70 or 80 % field capacity and left to equilibrate for 30 mins. Weight adjusted samples (the equivalent of 1 0 .0 g of air-dried material) of each treatment were then measured out into propylene beakers to ensure a constant tailings to water ratio amongst the treatments. Samples of standardised soil (Chapter 4 . 1 . 1 ) were included with each batch to ensure that comparisons could be made between the batches at different sampling periods. In addition to measuring pH levels in the control pots, at the termination of the growing period pH measurements were also made from material taken from each of the 1 60 sown pots constituting the two trials . This enabled a comparative assessment between the control and the sown pots to be made (Table 7.7). Comparison of the pH values in Table 7 .7 indicate that at the end of the growing period the pH values in the control pots were very close to values observed in the sown pots. This findings is significant in that it indicates growing conditions in the control pots were similar to those in the sown pots and justifies the use of extrapolated data from the control pots to examine the effect of tailings characteristics on plant growth. 326 To assess the long tenn pH stability of the various treatments, an additional set of pH measurements were made in the control pots approximately 3 months after the tennination of the trials. During this time maintenance of the control pots at 25°C and at 70 - 80 % field capacity moisture was continued. Graphs tracing the variation of pH with time for the various lime and sludge amendments are included and discussed separately in the two respective trials. 7.5.2 Electrical Conductivity The electrical conductivity (EC) of the lime and sludge amended control pots were assessed at week 3 (sowing), week 7 and week 1 1 (termination) of the trial . To corroborate EC results obtained from the unsown control pots, 1 8 sown pots from both trials were destructively sampled after harvesting (Table 7 .7) . Two samples , a surface (0- 1 0mm) and subsurface sample (>50 mm) were collected from each of the pots to investigate depth-wise EC variation within the amended tailings. The conductivity of the samples was assessed using a 1 :5 ratio of sample to water. The method involved weighing 5 g air-dry ground « 2mm) samples of the amended tailings into 50 ml centrifuge tube. Twenty-five ml of deionised water was then added and the samples were shaken for 30 minutes prior to gravity filtration through Whatman # 1 filter paper. The conductivity and temperature of the filtrates were measured with a Hanna Instruments HI-8633 portable conductivity meter. Because the meter had an in-built temperature compensation mechanism, standardisation of the individual EC values to 25°C was unnecessary. Prior to each use, however, it was necessary to calibrate the meter using a solution of standardised O.OIM KCl. The electrical conductivity of each sample was then calculated as per equation 7 . 1 EC == ECs x j EC = sample electrical conductivity ECs = solution electrical conductivity j = meter calibration coefficient [7 . 1 ] Based on a ranking system outlined in Blakemore et ai., ( 1 987), the amended materials were found to exhibit EC's that were either high (>800 J.lS/cm) or very high (>2000 327 )lS/cm) throughout the duration of the trial (Table 7.7). Comparison of the sludge's EC ( 1 73 )lS/cm) to that of the unamended Tui tailings ( 1 257 )lS/cm) indicates that the high electrical conductivity of the treatment solutions is largely attributable to the inherently high soluble salt content of the tailings . Table 7.7 The pH and electrical conductivity (EC) of the tailings treated with different amendments in the presence of plants (sown pots) and in the absence of plants (control pots). The average pH values of the 1 60 sown pots are also presented along with standard deviation values. Average pH calculations: n =0 24 and n = 8 for glasshouse Trials 1 and 2 respectively. Electrical Conductivity (IlS/cm] pH (8 weeks) Trial S ludge Lime Sown Pots Sown Pots Controls Sown Pots Controls [Mg/ha] [Mg/ha} (surface) (subsurface) Pots Pots 1 NA 0 3 522 2272 1 748 2 . 1 9 ± 0.02 2. 1 3 1 NA 8 .25 3487 23 1 9 1 993 2.27 ± 0.03 2. 1 6 1 NA 1 6. 5 1 069 1 290 1 1 1 2 6 .60 ± 0. 1 3 6.59 1 NA 1 1 2 1 749 1 806 1468 7 .59 ± 0.05 7.49 2 0 0 3363 2266 1 948 2 .22 ± 0.02 2.08 2 1 1 0 0 3 894 2266 25 1 3 2.22 ± 0.03 2. 1 3 2 220 0 794 1 074 1054 3 .78 ± 0.35 3 . 80 2 440 0 1 1 5 6 1 297 1 2 1 5 4.7 1 ± 0 .05 4.56 2 0 1 6 .5 1 292 1 1 8 1 1 267 6.56 ± 0. 10 6.46 2 1 1 0 1 0. 1 989 1 208 1 048 4.36 ± 0. 1 6 4.30 2 220 8 .5 1 3 82 1 1 85 1 270 4.73 ± 0. 14 4.37 2 440 5 . 1 1 566 1292 1 384 4.98 ± 0.06 4.87 Comparison of the surface and subsurface EC values in Table 7.7 indicates that those pots receiving low rates of lime and sludge generally exhibited higher EC at the surface of the pots. The higher EC' s encountered suggest that soluble salts were migrating upwards in the substrates due presumably to convective flow (capillary rise) resulting from the evaporative loss of water at the surface of the pots. Such effects are commonly noted in disturbed mining materials and are a frequently encountered problem in glasshouse experiments in which downward salt movement associated with convection (leaching) is minimal (Carlstrom et at. , 1 987). The minimal build-up of salts at the surface of the heavily limed or sludge treated tailings is believed to reflect the lower concentrations of electrolytes in solution. 328 The changes in EC values for various treatments during the growth periods are presented in Figure 7 .2 . In general there was an initial increase in EC values followed by a decrease towards the termination of the trial . The trends in EC values were similar to those of phosphate extractable sulphate contents of various treatments (Figure 7.3) . This clearly indicates that a considerable proportion of the salts responsible for conducting electricity in the tailings were sulphates. Flushing the pots midway during the experiment (as discussed in section 7.2. 1 ) clearly lowered the concentrations of sulphate in the tailings and simultaneously reduced EC in most treatments markedly. Comparison of the EC values of the control pots with the sown pots once again indicates that growing conditions in the control pots were similar to those in the sown pots and further justifies the extrapolation of metal concentration data from the control pots. 7.5.3 Metal Concentrations 25% HN03 and O. lMHCl acid extractable metal concentrations were determined by the methods outlined in Appendices 4(E) and 4(F). The metal extraction procedure using 0.02M CaCh was based on the method of Hoyt & Nyborg ( 1 972) and involved the addition of 20 m} of the extractant to 1 0. 0 g of tailings in a SOml polypropylene centrifuge tube. The solutions were shaken end-over-end for 1 hour at 25°C before being centrifuged at 9000 rpm for 5 minutes and filtered through Whatman # 42 filter paper. 1M KCl extractable Al concentrations were assessed using a method based on that of McLean ( 1 965) and outlined in Blakemore et al. ( 1 987). The extraction procedure involved weighing 5 .0 g of air-dried tailings into a 1 25-ml conical flask to which 25 ml of IMKCl was added. The contents of the flasks were gently mixed and then allowed to stand for 1 6 hours before being filtered through Whatman # 42 filter paper. Aliquots of the supernatants were diluted as required and concentrations determined using flame atomic absorption spectrometry. An Al hollow-tube cathode lamp set at 1 0 .0 rnA was used in conjunction with a reducing nitrous oxide-acetylene flame to 329 determine 1M KCl and O.02M CaCh Al concentrations. Typical flame type and lamp current settings for other metals are provided in Table 4.9. 7.5.4 Seed Germination The numbers of germinated seedlings were counted in each pot on 3 occasions during the first month of the experiment (1 , 2 and 4 weeks after sowing).The results of germination are individually reported in each trial. Throughout the duration of the trials, established seedlings were visually monitored for signs of interveinal chlorosis, discolouration and necrotic foliar tissue symptomatic of metal toxicities and/or nutrient deficiencies (Johnson et al. , 1 977). Figure 7.2 (a) 5000 .�-------�---=--------------, 4000 e � b 3000 ·s f 2000 :fZ:-�-=:::;�Z�::< � . 8 ,. .... _---, • / �0 1 000 .. " . . . . - . . . _ - - . . " - - . . . . . . _ _ . . . . . . , Termination i I e '-' 00 o o .�---------------------------------� o 2 3 4 5 6 7 8 9 1 0 1 1 1 2 Figure 7.2 (b) 5000 .�------------------�---------- 4000 5000 4000 Sowing Termination o 2 3 4 5 6 7 8 9 1 0 1 1 1 2 Figure 7.2 (c) t; 3 000 ·s C i I �� ' l o 2 3 4 5 6 7 8 9 1 0 1 1 12 Weeks Nil ---- 112 pH-6 1 pH-6 TAGP nil, nil 1 10, nil -..- 220, nil ----- 440, nil nil, I pH-6 -£- 1 1 0, 1 pH-6 -"9- 220, 1 pH-6 -fB- 440, 1 pH-6 330 Figure 7.2 The changes in electrical conductivity of various glasshouse trial treatments. (a) Trial One: limed treatments, (b) Trial Two: non-limed/sludge amended treatments and (c) Trial Two: limed ( 1 pH-6 rate)/sludge amended treatments. � � ..::: c:.. J; 8000 t 4000 � C "" r.l � � -E. 2000 if> o ..::: Q., Figure 7.3 (ll) Sowing Termination o . L_�_�._ .. �_�_�_� 8000 o � � -E.2000 '" ] � () o 2 3 4 / .L/�_c: Sowing o 2 3 4 Sowing 5 6 7 8 9 1 0 I I ) 2 Figure 7.3 (b) ., .. - . . ��""---��� �. -.-��� -=::.':=..-:-'-- • it n Termination 5 6 7 8 9 l O ) 1 1 2 Figure 7.3 (c) Termination () 2 3 4 5 6 7 8 9 ) 0 ) J J 2 Wed,-� -.­ Nil �.- J /2 pH-6 1 pH-6 -t)- T!\GP �. ni l , n i l �E}- 1 1 0, n i l � .. - 220, n i l .. � 440. n i l n i l . 1 pl l-6 1 1 0 , l pII-6 220, 1 pH-6 · EE 440. 1 pH-6 Figure 7.3 The changes i n sulphate levels within yarious glasshouse trial treatments. (a) Trial One: l imed lreailnents, (b) Trial Two: non-limed/sludge amended treatments and (c) Trial Two: l imed (I pH-6 rate)/sludge amended treatments � � l 332 7.6 Trial One - The Effects of Lime Application Rate and Plant Metal tolerance 7.6. 1 Aim This plant trial was specifically designed to investigate the impact of lime (CaC03) addition on the plant growth characteristics of the Tui mine tailings. Lime is an extremely useful soil amendment as it not only neutralises acidity but can also increase nutrient availability and hence plant uptake, reduce metal bioavailability and inhibit biooxidation of sulphide minerals (Chapter 3 ; section 3 .7.3). Four rates of lime were assessed in this trial in conjunction with 5 varieties of plants, including two exotic metal-tolerant grass cultivars. This trial provided an excellent opportunity to assess the merits of using metal-tolerant grass varieties in place of conventional cultivars. Additionally it was anticipated that the results would aid in identifying an optimal liming rate that would facilitate germination and sustained plant growth. 7.6.2 Trial Design To investigate the interaction of various liming rates, plant growth and metal tolerance, a completely randomised, factorially designed pot trial was conducted . Four liming rates and 5 plant taxa were assessed in the trial (Table 7.8) . The characteristics of the plant taxa used in this trial, which included one legume (Lotus corniculatus), two metal­ tolerant grass species (Agrostis capillaris var. Parys Mountain & Festuca rubra var. Merlin) and two non-metal-tolerant grass species (Agrostis capillaris var. Egmont & Festuca rubra var. Lobi), are discussed in Table 7.4. An additional treatment was included for L. corniculatus that consisted of inoculating the treated glasshouse trial tailings with rhizobium. A total of one hundred pots were used in this trial (Plate 7 . 1 ). This included four replicates of each treatment and an additional four amended but non-sown control pots that were prepared for destructive analysis during the course of the trial. 333 Table 7.8 Design Treatments for Trial One Liming Rate Nil Yz pH-6 Rate 1 pH-6 Rate TAGP Rate Mg/ha 0 8 .25 1 6.5 1 12 g/kg tailings 0 3 . 1 7 6 .33 43.2 Plant Taxa 1 , 2, 3 , 4, 5 1 , 2, 3 , 4, 5 1 , 2, 3 , 4, 5 1 , 2, 3 , 4, 5 Lotus Rhizobium + -, + -, + -, + -, Treatment Number of treatments 6 6 6 6 7.6.3 Methodology 7.6.3.1 Determination of Liming Rates The glasshouse trial tailings were amended with four rates of analytical grade CaC03 (Table 7.8). The low and intennediate liming rates were based upon the pH buffering capacity of the tailings, with the intermediate rate of 1 6.5 Mg CaC03/ha ( 1 pH-6 rate) representing the amount of lime required to raise the pH of the tailings to 6.0. The low rate (8.25 Mg CaC03/ha) is equivalent to 50 % of the 1 pH-6 rate. The TAGP rate was the highest of the liming rates used and is based on the amount of lime required to neutralise the acid potentially produced from the oxidation of remnant sulphide minerals within the tailings. In theory the TAGP rate ( 1 1 2 Mg CaC03/ha) is sufficiently high to prevent re-acidification of the tailings resulting from the total oxidation of acid generating sulphide minerals within the tailings . . The stated liming rates are based on lime incorporation to 200 mm depth and assume that the tailings have a uniform bulk density of 1 .3 Mg/m3 . 7.6.3.2 Lime Incorporation & Pot Preparation The equivalent of 1 kg of oven dry tailings was weighed and placed into a series of numbered plastic bags. The various rates of lime (Table 7.8) were then added to the air­ dried tailings and thoroughly mixed. After adjusting the water content of the various treatments to about 70% field capacity (section 7.2. 1 ) with distilled water, the 1 00 bags were incubated at 25°C for 3 weeks. J II •• � I 1/1,." '-'"1111!1' [l.It.' Ir. lh \!Jl\,h l(jQ( 334 Plate 7. 1 Trial One 4 weeks after sowing. Comparative growth of the various taxa on uncontaminated soil is seen in the 4 plastic trays in the upper left hand corner of the photograph. Upon completion of the 3 week incubation period the water content of the treatments was re-adjusted to 70% field capacity. The various treatments were then transferred to a series of numbered plastic pots of approximately 1300 cm3 which were lined with filter paper to prevent the loss of the amended growing media through the drainage holes. The combined weights of the plastic pots and treatments were then recorded to enable the watering requirements of the pots to be accurately met. 7.6.3.3 Sowing Approximately 100 viable seeds of each taxa were sown on the surface of the appropriate treatment pots as per the methods outlined in Section 7.4. 7.6.4 Hypotheses It was predicted that germination would fail on the untreated tailings due to the extreme acidity of the medium and its associated impact on the solubility of phytotoxic metals. The addition of lime to the tailings was predicted to improve germinating and growing conditions in the tailings. The tailings treated with the low and moderate (8.25 & 1 6 .5 335 Mglha) rates of lime, however, were expected to reacidifY due to the ongoing oxidation of remnant sulphide minerals. Plant die-back was therefore anticipated on these treatments during the course of the trial. The addition of lime at a rate equivalent to 1 12 Mglha was predicted to prevent reacidification of the tailings and to facilitate plant growth throughout the growing period. This very high rate of lime, however, was predicted to detrimentally affect the growth of the various plants due to alkaline induced nutrient deficiencies. The metal-tolerant plant varieties were predicted to out-yield the non-tolerant varieties on the low and moderate lime treatments due to their greater tolerance of heavy-metals. The rationale was that under acidic conditions the bioavailability of phytotoxic metals generally increases. 7.6.5 Results 7.6.5.1 Germination Since all plant species failed to germinate on the untreated tailings and the only taxa to germinate on the 12 pH-6 treatment was L. comiculatus, the following discussion of germination trends relates to the 1 pH-6 and TAGP treatments that received 1 6 .5 and 1 1 2 Mg of lime per hectare respectively (Table 7.8). The probable causes of complete germination failure on those pots that received ::; 8 .25 Mg limelha are discussed in section 7.6 .5 .2 . Over the four week period that seed germination was monitored, marked variability in germination trends amongst the species was evident (Figure 7.4). Very little variation, however, was noted between the metal-tolerant and non-metal-tolerant cultivars of the two grass species investigated (Frubra and A.capillaris) . Seed germination of both varieties of Frubra peaked at week 2, with average germination rates of 50 - 60 % being attained. Survival rates of F. rubra were high with less than 10% of the plants dying during the monitoring period (Figure 7.4). Similar germination trends were observed for L. com iculatus, although slightly lower germination rates of approximately 40% were attained. Lime application rates had little effect on the germination rates of L.comiculatus and Frubra, although a slight drop in germination was noted for 336 L. corniculatus at the highest liming rate. Slightly higher seedling mortality was also noted for L. corniculatus at the TAGP liming rate (Figure 7.4 b). At the 1 pH-6 liming rate germination rates of close to 1 00% were attained for both varieties of A . capillaris within a week of sowing. The survival rate of the seedlings, however, was low, with only about 50% of the Parys Mountain seedlings and 36% of the Egmont seedlings surviving through to week 4 of the trial (Figure 7.4a). A. capillaris germination rates of approximately 80% were attained in the TAGP treatments indicating a notable deterioration in germination conditions to those found at the lower liming rate (Figure 7.4b). 7.6.5.2 Yield Response (i) Nil Lime Treatment As predicted, seed germination on the nil lime treatment failed. The pH of the untreated tailings was found to remain extremely low and relatively constant throughout the duration of the trial, dropping only slightly over the 1 1 week trial period from a pre­ incubation level of 2.29 in the control pots to about 2 . 1 3 . S imilar pH values of 2 . 1 9 ± 0.02 were found in the trial pots which were analysed at the end of the 8 week growing period. As discussed in section 7.5 . 1 , the pH of the control pots was monitored for a further 3 months after the final herbage cut was made. Measurements of pH made at the end of this period indicate that the pH of the non-limed control pot rose slightly despite what were believed to be near optimal conditions for biochemical sulphide oxidation (Figure 7.5). The lack of a substantial decrease in the pH of the unamended tailings suggests that a buffering system exists in the Tui Mine tailings at low pH. This finding is consistent with other research that has found that below pH values of about 3 .5, pH buffering in acid sulphate soils, pyritic tailings and spoil occurs due to the acid hydrolysis of alumino silicate clays, jarosite and silica (Miller, 1 979; Dent, 1 986). All oxides, carbonates and silicates are generally alkaline in reaction (Mengel & Kirkby, 1 987). The hydrolysis of these minerals serves to buffer pH change and generally prevents the pH of sulphide-bearing materials falling below 1 .5 even under moist, non-leaching conditions (Dent, 1 986). 1 00 Cd i • I 1.l 80 I C!) I (tJ I - I l:: 50 )..tglml within a few 352 weeks of crop flooding (Mengel & Kirkby, 1 987). Clearly the Fe concentrations encountered at the nil and Y2 pH liming rates were extremely high and likely to have been directly toxic to most plant species. Concentrations of 1M KCI Al and O. IM HCI extractable As, Cu, Fe and Zn metal concentrations were also found to decrease markedly at the highest two liming rates, particularly at the TAGP rate (Figure 7 . 1 2) . O. IM HCI extractable Pb concentrations were reduced to 2 mg/kg or less at this rate, whereas lime application at the I pH-6 rate appeared to have little effect on the availability of O . IM HCl extractable Pb concentrations (Figure 7. 1 2). The neutralisation of O . lM HCI with CaC03 potentially explains the relatively low quantities of metals extracted in the tailings. heavily limed Plate 7.4 The Lotus corniculatus seen growing in this photograph (pots 53-55) are about 4 weeks old and are c learly perfonning poorly. Grown in tailings limed at a rate of 1 6.5 Mglha the plants showed signs of chlorosis and a generally yellow stunted appearance, symptoms consistent with N deficiency. 20 0; ..II:: c, E ....... E 1 0 :::I 'c 'E :::I < 0 1 00 ...... en ..II:: -en .§. 50 ... (I,) c. c. 0 () 0 4 0; ..II:: C, E ..... � 2 (J) c:: n:I en c:: n:I :E 0 -, 2 + � � + u 'c (I,) I!? « 0 0 28 56 2000 ,., -, -i ,� � � '-¥ , I I 0 28 56 I • I 353 6 , I I - - -. - E 4 -. - - - -. :::I 'E I "C n:I 2 () 0 0 28 56 300 • - - - - -* - - - -e • 200 ... 0 • 1 0 c:: "C ---0 1 000 � n:I .: .- (I,) ...J -.- - 1 00 0 0 28 56 0 28 56 0 28 56 500 l J Extractants - - -. • � u - . - 0 . 1 M HCI • c:: 250 N T 0.02M CaCI2 �/ � - + - 1 M KCI 0 28 56 0 28 56 Days Days Figure 7,9 Changes in metal concentrations as extracted by O. lMHCl Ce), O.02MCaCIz Co), and IMKCl C+) for 1 pH-6 lime treated tailings during the growing period. 2 0 ...... O"l � O"l E ...... E 1 0 ::l 'c 'E ::l < ...... O"l .:.c: C, 0 2 E ';:' 1 III c. C. o (,) c; .:.c: C, E ...... o 4 � 2 III c: III O"l c: III :E 0 354 0.5 l 4 - ! A (.) E • 'c � ::l / Q) 'E 2 I!? "C � « III ., (,) , / • 0 .0 0 0 28 56 0 28 56 0 28 56 1 , 4 • o o o 28 56 o 28 56 o 28 56 50 • Extractants . - - - -. - . - 0 • 0. 1 M HCI (.) • ,S: 25 -. ,� 0.02M CaCI2 N -c---t--'-../ + - 1 M KCI 0 28 56 0 28 56 Days Days Figure 7, 1 0 Changes in metal concentrations as extracted by O. l M HCl (e), O .02M CaCI2 (0), and I M KCl (+) for TAGP lime treated tailings during the growing period. 400 ..... OJ � OJ .§. E 200 :::J 'i: 'E :::J :;;: 0 1 50 '0) =a, 1 00 .§. ... Q) -] I r7 • 0 g; 50 -' o o '0) ..x: -OJ E ..... Q) II) Q) c:: o III o 6 , 4 j/// � 2 , c:: III :E *= 0 I 0 ./ ./ .. /!I., " ," " '" ·v "''''' ... '< '. i • 28 56 :::- - - - - 'i • • 28 56 J-" �, • :======-I ' 28 56 Days 1 0 t,) 'i: Q) 5 I!! « � /-. . ". D- - ... .. .. \ - .. - - -r 0 • ===r= Ii II I 0 28 56 5000 c:: g 2500 � o �i�' --�--�.�, -- ,---II' o 800 l t,) c:: 400 I N I .1 � , 1 0 • 0 28 56 �. __ L:::J --. " -''; i =I) 28 56 Days 1 0 E :::J 'E 5 '" t1I O 0 20 355 l ,., '\ J . \ /// , '\ /'/f I f/ \ i � � � II I : i 0 28 56 I 3 1 0 _L� ��-==4 � • o o 28 56 Liming Rates Nil Lime • 1 /2 pH-6 1 pH-6 • TAGP Figure 7,1 1 Concentrations of various metals extracted by 0 ,02M CaCI2 at different lime rates, c; 400 l ..:.:: 0, .s E 200 ,:: s:: 'E :::! � o I 300 c; ..:.:: 0, 200 E ...... .... Q) §: 1 00 o u o 28 56 • 1b_// ______ �-----� \j 1 5 (.) 1 0 '2: Q) f! « 5 o 8000 -\ s:: ,g 4000 356 1 5 - 28 56 o 28 56 300 200 1 00 o �.�--��.�, --�--�, o � •• ' ---,--,.,---�--• o �.��--�.��----. 6 c; ..:.:: 0, E 4 ...... o 28 56 o 28 Days 56 o 28 56 1 000 -; 500 o �,�====�==�=­ o o 28 Liming Rates Ni l Lime • 1 /2 pH-6 1 pH-6 • TAGP Figure 7.12 Concentrations of various metals extracted by O. l M HCI and 1M KCI (aluminium only) at different liming rates. 56 357 (iv) TAGP Treatment The pH of the control pot treated with 1 1 2 Mg/ha of lime (TAGP rate) was found to almost parallel that of the tailings treated with 1 6.5 Mg/ha of lime (Figure 7.5) . The pH of the tailings was found to increase to about 7 .5 by the end of the 3 week incubation period where it remained for the entire monitoring period. Comparison of the pH of the planted pots (7 .59 ± 0.05) with that of the equivalent control pot (7 .49) again indicated that there was good correlation between the control and trial pots and that the effect of plant growth on pH was minimal over the duration of the experiment. Analysis of variance indicates that addition of the T AGP rate of lime had a negative and significant (P < 0.00 1 ) impact on plant yield. Inspection of Figure 7 .8 shows that the average herbage yields of all five plant taxa decreased notably at the TAGP liming rate. The taxa demonstrating the greatest negative response to this rate of lime addition was the metal-tolerant variety of A. capillaris (Parys Mountain). Dry matter yields of this taxa decreased from nearly 0.7g per pot at the 1 pH-6 rate to about 0 .01 g per pot at the higher liming rate (Figure 7 .8) . Although the two effective liming rates had no significant effect (P=0.328) on the relative yields of A.capillaris and Frubra, a highly significant (P>O.OO I ) interaction was identified between lime treatment and plant metal tolerance. This interaction is readily observed in Figure 7 .8 where at the highest liming rate ( 1 12 Mg/ha) a marked decrease in the relative performance of the metal-tolerant plants in relation to the non­ metal-tolerant taxa is evident. This interaction is most marked for A. capillaris and suggests that the growth of this species is adversely affected by high alkalinity. In contrast to A. capillaris, no dramatic decrease in dry matter yields were noted for Frubra at the TAGP liming rate and the metal-tolerant variety of Frubra again significantly out-yielded the non-metal-tolerant variety by a ratio of more than 3 to 1 . The reduction in yield of Frubra vaL Merlin at the higher liming rate was, however, significant (P<0.05). Lime was also found to have a significant (P .s <{ -() co '--x W _N 0 co 0 0) ..;.c:; --0> .s Q.l u... -() co ..... x W N 0 co 0 C) ..;.c:; --0> .s c: N ....; () co .... -x w N 0 co 0 1 000 1 00 1 0 1 1 0000 1 000 1 00 1 0 1 . 0 0 . 1 0.0 1 1 000 1 00 1 0 1 � -- -:j - � ::J -- i Aluminium //� ' / � .: liS (A-// //� ·.o // , 0 / � .�.< . 1 , / 1 1 1 § � 1 � § � 1 -, .. 0· 1 ., - � ::J ...) - 1 1 0 1 00 1 000 1 M KCI Extract. AI Iron [j-o � . ' oo�. / / , , I i , l 1 00 B · � . 1 000 0. 1 M HCI Extractable Zn 1 000 � � ::l 1 00 � 0 j ....; � () 1 0 co � ..... x W 1 .0 N =0 0 co -0 0. 1 "2 ::J � - 0.01 0 . 1 1 00 ...J .0 D... -() 1 0 --=' co ..... -x W N 0 1 .0 co -0 0. 1 0 · 1 <> A 0 • 359 Copper � I A..' . �OO I � ... . . \ • • 1 A.. o · 0 1 0 1 00 0. 1 M HCI Extract. Cu Lead • • � 1 000 /� '�-_J]// I • �// / 1 1 0 1 00 1 000 0 . 1 M HCI Extract. Pb Liming Rates Nil Lime Half (8.25 Mg/ha) 1 pH ( 1 6.5 Mg/ha) TAGP ( 1 1 2 Mg/ha) Figure 7.13 Relationships between 1M KCI and O.02M CaCh extractable AI, and between O. IM HC] and O.02M CaCl2 extractable Cu, Fe, Pb and Zn (mglkg]. The 3 data points for each treatment originate from analysis of unsown control pots conducted at week 3 (sowing), week 7 and week 1 1 (termination) of the trial. 360 As previously discussed, the addition of the equivalent of 1 1 2 Mg CaC03/ha was found to increase the pH of the tailings to about 7 .5 where it remained relatively constant for the duration of the experiment. Potential problems affecting plant growth associated with high pH are well recognised and include; enzyme inactivation, induced B, Cu, Fe, Mg, Mn, Zn and P deficiencies, accelerated N loss through leaching and volatilisation and secondary growth problems particularly with legumes (Doubleday 1 972; McLean & Dekker, 1 976; Gemmell, 1 977; Mengel & Kirkby, 1 987; Pulford, 1 99 1 ). 7.6.5.3 Plant Nutrition To investigate whether the 'heavy' liming rates were adversely affecting plant nutrition, foliar analysis was conducted on selected herbage from the two effective lime treatments (Table 7 . 1 0) . The nutrient values presented in the Table 7 . 1 0 represent either average nutrient concentrations from individual analysis conducted on two replicate pots or average values from composite samples (ie. combined herbage from all 4 replicate pots). Low dry matter yields meant that composite samples were required for L.comiculatus at both liming rates and for A.capillaris var. Egmont. and F rubra var. Lobi at the 1 pH-6 and T AGP liming rates respectively. The quantities of dry matter produced by both A. capillaris cultivars at the T A GP liming rate were unfortunately too low for foliar analyses to be carried out. Included in Table 7. 1 0 are nutrient data for Ca, K, Mg, N and P and estimated total uptake values calculated as per equation 7.2. The nutrient status of the herbage was classified as satisfactory, low or very low by comparing the foliar nutrient content of the herbage grown on the amended tailings with ' critical' nutrient values for Trifolium repens and Latium perenne grown in soil (McLaren & Cameron, 1 996). Herbage was classified as having satisfactory nutrient status where herbage concentrations � critical value, low nutrient status where herbage concentrations < critical value and very low status where herbage concentrations < 50% critical value. It is recognised that the critical values reported by McLaren & Cameron ( 1 996) are based on field trials as opposed to glasshouse trials and that their use as anything but a rudimentary guide to indicate potential deficiencies in the herbage derived from the pot trials would be inappropriate . Uptake [mglkg] = mean pot yield [g] * nutrient [%] * 1 0,000 [7.2] 36 1 Table 7. 1 0 Nutrient content and estimated total nutrient uptake by herbage from selected treatments. sat. = satisfactory; v.low == very low; nd == values not determined due to insufficient dry matter product. Agrostis capillaris Festuca rubra Lotus Parys Mtn. Egmont Merlin Lobi Liming Rate pH-6 TAGP pH-6 TAGP pH-6 TAGP pH-6 TAGP pH-6 TAGP Phosphorus [%] 0 . 1 0 0.08 0.27 nd 0.27 nd 0.34 0.27 0 .25 0 . 1 2 Critical P [%] 0.3 0.3 0.3 - 0.3 - 0.3 OJ 0 .3 0 . 3 P Status v.low v.low low - low - sat. low low v.low Uptake [mglkg] 1 8 1 97 1 49 1 - 1 64 - 3 104 1259 787 168 Nitrogen [%] 3.7 1 6.67 4.53 nd 4.29 nd 4.39 4.54 4 .6 1 4.9 1 Critical N [%] 4.4 4.4 4.0 - 4.0 - 4.0 4.0 4 .0 4 .0 N Status low sat. sat. - sat. - sat. sat. sat. sat. Uptake [mglkg) 6680 800 1 24940 - 2577 - 39734 2 1 5 5 1 1 4738 6877 Calcium [%] 0.78 1 .65 0 .4 1 nd 0.6 1 nd' 0.39 0.64 0 .73 1 .07 Critical Ca [%) 0.3 0.3 0.2 - 0.2 - 0.2 0.2 0 .2 0 .2 Ca Status sat. sat. sat. - sat. - sat. sat. sat. sat. Uptake [mglkg) 1401 1 984 2250 - 368 - 3 548 3054 2323 1 502 Magnesium [%] 0.20 0.45 0 . 1 2 nd 0. 1 5 nd 0 . 1 1 0.2 1 0 . 1 6 0.25 Critical Mg [%] 0 . 1 3 0 . 1 3 0 . 1 5 - 0 . 1 5 - 0 . 1 5 0 . 1 5 0 . 1 5 0 . 1 5 Mg Status sat. sat. low - sat. - low sat. sat. sat. Uptake [mglkg) 355 544 660 - 89 - 1 032 10 17 525 354 Potassium [%) 1 .40 1 .06 4.07 nd 3 .28 nd 3 .06 2.28 2 .22 1 . 1 8 Critical K [%) 1 .7 1 .7 1 .7 - 1 .7 - 1 .7 1 .7 1 .7 1 . 7 K Status low low sat. - sat. - sat. sat. sat. low Uptake [mglkg] 25 12 1271 2239 1 - 1967 - 27693 1 0849 7 1 0 1 1 648 (i) Phosphorus The most severe nutrient deficiency observed in the herbage was that of P which was found to be deficient in 7 of the 8 treatments analysed. Analyses of the tailings sampled from the equivalent control pots confirmed that the total P and Olsen P content of the tailings were very low over the duration of the trial despite the addition of soluble P fertiliser (Table 7 .9) . The deficiency of P is considered by many researchers to be the main factor limiting plant colonisation of mining wastes closely fol lowed by N deficiencies (Smith & Bradshaw, 1 979; Bradshaw & Chadwick, 1 980; Barnishel et ai. , 1 982) . 362 Herbage of the metal-tolerant cultivar of Frubra (Merlin) grown on the lower of the two liming rates was the only foliage found to contain what were considered to be satisfactory levels of P (Table 7 . 1 0) . This observation is important in that it was this treatment that yielded the greatest levels of dry matter which strongly suggests that P nutrition was an important factor governing the growth of the plants on the various treatments. P deficiency was most marked in herbage originating from tailings amended with the TAGP rate of Eme (Table 7 . 10) . L. com iculatus and F.rubra var. Lobi were particularly deficient in P at the higher of the two liming rates, presumably due to the effective immobilisation of P in the tailings . In the case of L. com iculatus, very low P concentrations were observed in herbage obtained from both lime treatments. The inability of L. corniculatus to take up greater quantities of P may have at least in part reflected their poorly developed root systems which were clearly incapable of fully exploiting the substrate' s P reserve. As previously discussed (Chapter 3 ; section 3 .7.3 . 1 ) the uptake of P is highly pH dependent. Soil pH not only regulates the ratio of H2P04 -IHPO/- in the soil solution but also governs the adsorption of P and formation of phosphatic complexes. In heavily limed growth media or in calcareous soils, the formation of precipitates of low solubility phosphates, such as trica1cium phosphate (Ca3(P04)2), becomes an important mechanism governing the availability of this macronutrient. Most plants grown on the tailings showed some symptoms of P deficiency which is characterised by overall stunting of the leaf and root system, dark green mature foliage and reddening or purpling of the stems, petioles and leaves resulting from the accumulation of anthocyanins (Mengel & Kirkby, 1 987; Salisbury & Ross, 1 992). The strongest visual evidence of P deficiency was manifest in the dark green foliage of the grasses and discolouration of stems. In other studies the application of heavy rates of lime to pyritic mine wastes has been found to have variable impact on the P nutrition of pastoral plant species (Doubleday, 1 972; Gemmell, 1 98 1 ; Costigan et ai. , 1 982). One common finding of these studies was 363 that the performance of legumes was more adversely effected by the application of high rates of lime than the grass species investigated. In the study of Costigan et ai. ( 1 982), for example, the application of lime in excess of 5 Mg/ha was found to significantly reduce the growth of Trifolium repens by as much as 90%, where as in the case of Lotium perenne this amount of lime was the minimum required to attain satisfactory germination and growth. Costigan et al. ( 1 982) proposed that the application of high rates of P fertiliser and the substitution of Mg limestone (dolomite) for calcitic limestone could improve the growth response of L.perenne and reduce the inhibitory effects of high rates of calcitic limestone on T.repens. In contrast to Gemmell ( 1 98 1 ) and Costigan et ai. ( 1 982), Doubleday ( 1 972) found that that the growth of Lolium perenne was significantly reduced by the application of high rates of lime. Doubleday ( 1 972) reported that the growth responses of L.perenne to phosphate application on some spoils were reduced by over 80% after liming to pH 7 .8 . Even moderate rates of lime addition (sufficient to attain a pH of 6 .5 ) were found to significantly reduce dry matter yields. Doubleday ( 1 972) attributed this inhibition to lime-induced Fe deficiency and possibly the imrnobilisation of P and nutrient imbalances. This finding is consistent with those of the present study which indicate that the performance of A. capillaris and F. rubra as well as L. corniculatus was governed to a large degree by P availability. Although the herbage did not show obvious symptoms of Fe deficiency (interveinal chlorosis and in severe cases total bleaching of young foliage) analysis of the growing medium clearly indicates that availability of Fe in the tailings treated with the two highest rates of lime was very low (Figure 7. 1 3 ) . Because the herbage was not analysed for Fe, however, it is impossible to speculate whether the performance of the plants were adversely effected by Fe deficiencies. (ii) Nitrogen In contrast to phosphate levels, herbage N levels appeared to be mostly adequate at both the 1 pH-6 and TAGP liming rates (Table 7 . 1 0) . L. comiculatus grown on the lower of the two rates, however, was found to contain less than 3 .8% N, a level considered to indicate deficiency. This analytical finding is consistent with the chlorotic yellow appearance of the L. com iculatus seedlings grown on this treatment and strongly suggests that the low yields and poor performance of this species were at least in part 364 related to the inability of Lotus to take up sufficient N under the circumneutral pH conditions induced by applying 1 6. 5 Mg CaC03/ha. Conversely, the heavier rate of lime appeared to improve the N status of L. corniculatus (Table 7 . 1 0) . This findings is somewhat surprising in that the foliage of Lotus plants grown on this treatment also showed signs of chlorosis consistent with N deficiency. It is postulated that lime have induced molybdenum deficiencies mayf alternatively adversely affected L. comiculatus growth. Analysis of the unplanted control pots indicated that total N concentrations in the tailings were extremely low in all treatments at the end of the growing period despite the addition of appreciable quantities of ammonium nitrate (Table 7 .9). It is postulated that in the absence of plant uptake, N was being lost from the tailings as a result of leaching and/or ammonia volatilisation. Coarsely grained materials devoid of organic matter and high CEC clays, such as the Tui mine tailings, are susceptible to leaching of certain plant nutrients including nitrate (Johnson et al., 1 977; Bradshaw & Chadwick, 1 980). In those pots receiving the equivalent of 1 1 2 Mg CaC03 per hectare, gaseous losses of N may have further depleted N in the tailings. Previous research indicates that the volatilisation of ammonia (NH3) produced by the dissociation of ammonium carbonate (NH4)2C03 is enhanced at high pH and can lead to significant losses of N in calcareous soils or in heavily limed mine-wastes (Reeder & Sabey, 1 987). (iii) Calcium & Magnesium Predictably, the Ca status of the various taxa grown on tailings amended with both rates of lime were adequate (Table 7. 1 0). In the case of L. corniculatus the percentage of Ca in the herbage was found to more than double at the higher rate of lime and was present in concentrations far in excess of that considered necessary for optimum growth ( 1 .65% cf. 0.3%). The trends observed in the herbage are consistent with the concentrations of exchangeable Ca in the control pots (Table 7 .9). The presence of high concentrations of Ca in the tailings appears to have had little impact on the uptake of Mg. Like K+ and N� +, excessive concentrations of Ca2+ in soils can adversely affect the uptake the Mg2+ due to ion competition. This phenomenon is most commonly associated with calcareous soils (Mengel & Kirkby, 1 987), but has 365 been noted in several studies investigating the reclamation of heavily limed colliery spoils (Doubleday, 1 972; Gemmell, 1 98 1 ; Costigan et aI., 1 982). Previous research indicates that the substitution of dolomite for calcitic limestone helps prevent Mg deficiencies by reducing the competitive uptake of Ca2+ (Costigan et al. , 1 982). The relatively high levels of Mg found in the TAGP control pot indicates that Mg may have been immobilised at high pH by free carbonate (Table 7.9). It is highly unlikely that significant quantities of Mg were introduced through liming because analytical grade limestone (CaC03) was used in this experiment. (iv) Potassium Although the K status of some plants was low, the results presented in Table 7. 1 0 indicate that K levels were generally adequate or only marginally deficient i n most treatments. In the case of L. com iculatus, however, the herbage was found to be deficient at both liming rates. The non-tolerant Frubra cultivar also appeared to exhibit K deficiency at the highest liming rate (Table 7. 1 0) . Like Mg, K uptake and retention in plant cells is competitively affected by other ions. High concentrations of H+, Ca2+, Mg2+ and Na+ are known to adversely effect the uptake of this essential plant nutrient (Mengel & Kirkby, 1 987) . Foliar analysis of the herbage indicates that high concentrations of Ca may have induced K deficiency in the plants grown on the amended tailings, particularly those grown on the T AGP lime treated tailings. 7.6.6 Conclusions of Trial One • The failure of all plant taxa to germinate on the unamended Tui tailings, including two varieties of metal-tolerant grass, confirmed that the tailings are an extremely hostile growth medium even for plants adapted to adverse growing conditions. • The addition of lime at a rate equivalent to 8 .25 Mg/ha (Y:z pH-6) did not adequately ameliorate the tailings' adverse germination conditions. Although pH levels initially improved upon lime incorporation, this rate of lime was found to, in fact, increase the relative bioavailability of some metals as the tailings re-acidified. This finding has important implications for any future attempts to reclaim the Tui Mine site. It 366 indicates that the application of what, by agricultural standards would be considered to be a high rate of lime (8 .25 Mglha) to the tailings, may actually exacerbate growing conditions and acid mine drainage. • It is evident that low pH, high electrical conductivity and phytotoxic metal concentrations in the soil solution all contributed to germination failure on the nil and 1S pH -6 lime treatments. • The addition of lime at rates z 1 6. 5 Mg/ha facilitated germination and plant growth of all taxa by reducing the pH and bioavailability of metals in the tailings. Plant nutrition, however, was adversely effected by these relatively heavy rates of lime application, particularly at the TAGP liming rate ( 1 1 2 Mg/ha) as indicated by foliar analysis. • It appears that induced P deficiency was the primary reason for reduced plant growth yields on the TAGP lime treatment and that F. rubra (var. Merlin) was better adapted to absorbing this nutrient at circumneutral pH levels induced by heavy rates of lime application. • Both varieties of A. capillaris were found to be poorly suited to the moderately alkaline conditions induced by the TAGP liming rate ( 1 1 2 Mg/ha). Likewise L. corniculatus performed poorly in this treatment, as it did at the lower liming rate ( 1 6 . 5 Mg/ha). i mprove • Rhizobial inoculation of L. comiculatus failed to significantly�he performance of this species at either liming rate and visual inspection indicated no evidence of root nodulation. The poor performance of Lotus was particularly disappointing in that it is recognised that the establishment of a legume on the Tui tailings is fundamentally important to the formation of self-sustaining vegetative cover. • In the short term (ie. <24 weeks), the 1 pH-6 liming rate was identified as being the superior treatment in terms of its ability to maintain the pH of the tailings in a range conducive to plant growth. The ongoing potential of the tailings to reacidify, however, was recognised as a major limitation to the long-term survival of plants established on this treatment. Contrary to what was predicted, reacidification of this treatment did not occur during the growth period, nor in the ensuing 3 months that the control pots were monitored. • Whilst the growth of F. rubra var. Merlin was not considered to be optimal this taxa formed and maintained what was considered to be a ' satisfactory' vegetative cover on 367 tailings treated with 1 6.5 Mg CaC03/ha throughout the 2 month growing period. Foliar analysis of the Merlin herbage grown on this treatment indicated that, with the possible exception of Mg, Merlin plants were receiving adequate macronutrients via the application of growth solution (Figure 7. 1 0) • Merlin was without doubt the most successful of the plant taxa trialed in terms of biomass produced and vigour. Whilst Merlin significantly out-yielded the non-metal­ tolerant variety of Frubra (var. Lobi) at both effective liming rates, a significant difference in yield between the metal-tolerant and non-metal tolerant A. capillaris cultivars was only noted at the lowest effective liming rate ( 1 pH-6 rate) . Both cultivars of A. capillaris preformed very poorly upon the TAGP lime treatments (Figure 7 . 8) . • Whilst literature pertaining to the reclamation of pyritic mines wastes commonly advocates the use of very high rates of lime to facilitate the growth of vegetation and to lessen the likelihood of acid mine drainage, the results of this trial suggest that there is a trade-off between diminished plant response resulting from the application of high rates of lime and the need to ensure sufficient lime is added to prevent reacidification of the growing medium and the generation of AMD. Although the addition of lime at the TAGP rate is likely to have effectively prevented reacidification of the tailings, the adverse nutritional effects associated with this liming rate are unlikely to facilitate the establishment of an effective vegetative cover. 368 7.7 Trial Two - Organic Matter Amendment 7.7.1 Aim The primary aim of this trial was to investigate the effect of the addition of an organic matter amendment on the plant growth characteristics of the Tui mine tailings. Various rates of composted sewage sludge were evaluated, both with and without the addition of lime, to assess the impact of these treatments on the growth performance of two plant species, Lotus corniculatus and Festuca rubra var. Merlin. The liming rate was decreased with increasing rates of sludge addition to account for the neutralising effect of the sludge. 7.7.2 Trial Design The growth responses of L. comiculatus (Lotus) and Frubra (Merlin) were monitored over an 8 week period on tailings amended with four rates of composted sewage sludge and two rates of lime. A completely randomised block design trial comprised of 4 sludge rates, 2 liming rates and 2 plant species with four replicates was used (Table 7 . 1 1 ). An additional 8 amended but unsown control pots were included in the trial, thus a total of 72 pots were used in Trial Two (Plates 7.5 & 7 .6) . The control pots were destructively sampled and analysed during the course of the trial to monitor the growth conditions of the tailings(section 7 .5) . Table 7.1 1 Trial Two design and amendment rates. Liming rates: n i l = n i l lime treatment, 1 pH 6 = amount of lime required to raise the pH of the sludge amended tailings to approximately 6.0. Fr = Festuca rubra (var. Merlin), Lc = Lotus corniculatus Sludge Rate [Mglha] 0 1 1 0 220 440 glkg tailings 0 78 1 56 3 1 2 Liming Rates nil, 1 pH 6 nil , 1 pH 6 nil , 1 pH 6 nil, 1 pH 6 Plant Species Fr, Lc Fr, Lc Fr, Lc Fr, Lc Treatment replicates 4 4 4 4 The characteristics of the two plant taxa used in the glasshouse trial are outlined in Table 7.4. It was deemed important to include a legume such as Lotus in this trial because as discussed in Chapter 3 (section 3 . 7) legumes aid in the development of a 369 self-sustaining soil-plant ecosystem by reducing or eliminating the need for supplementary nitrogen application. 7.7.3 Trial Methodology 7.7.3.1 Sludge Addition The North Shore sludge7 used in this trial was incorporated into the tailings at 4 different rates; 0, 1 1 0, 220 and 440 Mg/ha. The highest rate (440 Mg/ha) was based on the amount of sludge required to elevate the organic carbon content of the tailings to approximately 4 % (Appendix 7(B») . This level is comparable to many of New Zealand's soils which typically contain between about 4 and 1 0% organic carbon (Blakemore et at. , 1 9 87) and is in excess of the minimal amount that Johnson et al. ( 1 994) suggest should be incorporated into mining wastes ( l . 8%) . As discussed in Chapter 3 (section 3 .7.4) sludge application rates in excess 400 Mg/ha are commonly used in the reclamation of mine wastes . In some overseas trials outlined by Sopper ( 1 992) application rates in excess of 900 Mg/ha have been evaluated. The sludge rates used in this trial were calculated based upon a 200 mm incorporation depth and used organic carbon analysis of the sludge (Table 7.2) conducted by Limpus­ O'Reilly ( 1 997). Prior to amendment addition, samples of the sludge were oven-dried at 1 05 °C for 24 hours to assess its water content. The sludge was found to contain about 46 % water by mass, therefore the highest sludge rate treatments (440 Mg/ha) required the addition of approximately 3 1 2 g of moist sludge to each kilogram of oven-dry tailings (Table 7 . 1 1 ) . To ensure thorough mixing of the plant growth media the various sludge treatments were weighed out individually and then added to numbered plastic bags containing one kilogram of oven-dry tailings and one of two rates of lime; either a nil rate, or a rate calculated to raise the pH of the sludge amended tailings to 6.0 (Table 7 . 1 2) . 7 S e e description of sludge in section 7. 1 .3 370 Plate 7.5 Trial Two: 4 weeks after sowing. Plate 7.6 Trial Two: 8 weeks after sowing immediately prior to the herbage harvest. 3 7 1 7.7.3.2 Liming Requirements & Lime Addition To ensure that the liming requirements of the amended tailings were accurately met and that the lime rapidly reacted with the tailings, finely ground « 1 50) analytical grade CaC03 was again used in this trial. Due to the liming effect of the sludge, the amount of lime required to achieve pH 6 ( 1 pH-6 rate) as determined by a short-term pH buffering experiment (section 7. 1 . l . 1 ), varied in relation to the rate of sludge added. Figure 7 . 1 b clearly indicates that as the sludge rate was increased less lime was required to raise the pH of the tailings to 6.0. The different liming requirements of the various sludge amended treatments are presented in Table 7 . 12 . As previously mentioned, the four rates of sludge outlined in Table 7 . 1 2 were evaluated both with and without lime addition, thus a total of 8 amendment treatments were assessed in Trial Two. Table 7.1 2 Lime and sludge application rates for Trial Two treatments. The area rate assumes amendment incorporation to 200 mm depth and a tailings' bulk density of 1 .3 Mglm3. SludgelLime Levels 1 pH -6 Liming Rate Sludge Application Rate nil 1 1 0 220 440 [Mglha] Lime [Mglha] 1 6 .5 1 0 . 1 8 .5 5 . 1 Lime [MglI OOO Mg tailings] 6 .33 3 . 88 3 .25 1 .97 7.7.3.3 Sowing Each of the trial pots was sown with a rate of seed designed to apply approximately 1 00 viable seeds of the appropriate taxa (section 7.4). To help optimise the growth of Lotus all pots sown with this species were treated with an appropriate commercial Rhizobial inoculum. 7.7.4 Hypotheses Dry matter yields for both plant species were expected to increase with increasing sludge application rates as the physical and chemical growing conditions in the tailings improved (Chapter 3 ; section 3 .7 .4). Sludge addition not only increases the organic matter content of the tailings but also acts as a liming agent and therefore reduces the 372 liming requirement of the tailings. The metal tolerant grass cultivar Merlin was predicted to substantially out-yield Lotus on the non-limed treatments due to Merlin's presumed lower Ca nutritional requirements and the supposed ability of this grass cultivar to tolerate Cu and to a lesser degree Pb and Zn contaminated substrates (Chapter 3; section 3 .7. 5). The addition of lime to the sludge treated tailings at rates sufficient to increase the pH of the amended tailings to 6.0 was predicted to substantially improve plant performance and to reduce dry matter yield differences between Lotus and Merlin The legume was predicted to perform better in the limed treatments due to the greater calcium concentrations and the immobilisation of potentially toxic heavy metals under the more neutral pH conditions. 7.7.5 Germination Results 7.7.5.1 Germination Trends on Non-Limed Treatments Predictably and in keeping with the results of Trial One, both Lotus and Merlin failed to germinate on the untreated tailings (ie. nil sludge and nil lime treatment) . The addition of 1 1 0 Mglha sludge only marginally improved germinating conditions in the tailings (Figure 7 . 14a) . The few seedlings that emerged on this treatment were found to die back within 4 weeks of sowing. The addition of sludge at rates of ;::: 220 Mglha, however, was found to substantially improve germination conditions and seedling establishment conditions in the short-term at least. Germination rates in the vicinity of 40% were attained for both Lotus and Merlin at the 220 Mg sludge rate, however, the rate of seedling mortality was high, particularly for Lotus. By week four of the trial all of the Lotus seedlings had died and little more than half of the Merlin seedlings remained (Figure 7 . 1 4a) . The surviving Merlin plants were generally unhealthy and relatively small and many of the tillers exhibited brown and purple hues. These symptoms are characteristic of phosphorus (P) deficiency and various metal toxicities. Regression (ie. die-back) of the Merlin plants continued to occur through the growing term. At the 440 Mg/ha sludge rate, Lotus exhibited similar germination trends to those observed at the 220 Mglha sludge rate and despite the high rate of sludge addition most of the seedlings died-back within two weeks of sowing (Figure 7. 1 4a). Regression of the 373 Lotus seedlings continued through the growing period to the extent that at harvesting only one of the four treatment pots yielded sufficient dry matter (O.O I g) to warrant measurement. In contrast, Merlin germinated and fared well on the 440 Mglha treatment. A germination rate of about 60% was attained and seedlings grew vigorously throughout the trial. The plants showed no obvious signs of regression or nutrient imbalances. 7.7.5.2 Germination Trends on Limed Treatments The overall effect of lime addition was to improve seed germination and seedling survival rates. The impact of liming was most noticeable on treatments that received nil, 1 1 0, or 220 Mglha rates of sludge (Figure 7. 1 4b). The addition of lime to the non-sludge treated tailings enabled Lotus and Merlin to germinate moderately well (Figure 7. 1 4b). Whilst relatively high average germination rates in the vicinity of 60% were attained for both species and seedling survival rates were high, dry matter yields from this treatment were poor (Figure 7 . 1 4b) . These results corroborate those of Trial One in which the nil sludge and I pH-6 lime treatment was effectively duplicated (Figures 7.4b & 7.5) . The germination and seedling survival rates of Merlin showed little variability amongst those treatments that received both lime and sludge. Germination rates for this taxa were in the vicinity of 50 to 60 % in each of the 3 sludge amended treatments and seedling mortality was low during the 4 week monitoring period (Figure 7 . 14b) . The germination results indicate that the variations in Merlin dry matter production between the limed treatments was primarily a function of plant size rather than seedling numbers. The average germination rate of Merlin on the 440 Mglha treatment, for example, was comparatively low but significantly out yielded the other treatments (Figures 7. 1 4b & 7 . l sb). In terms of germination, Lotus showed a greater degree of variability between the various limed treatments than Merlin (Figure 7 . 1 4b). Within one week of sowing the average germination rate attained from those pots treated vvith 1 1 0 Mglha sludge and the equivalent of 1 0 . 1 Mglha of lime was approximately 67 ± 9% (this compares with a germination rate of just 7 ± 3% on the equivalent non-limed treatment, Figure 7 . 1 4a) . Despite the addition of this relatively high rate of lime and sludge by week 4 of the trial 80 o 374 (b) 1 pH-6 Liming Rate • • 1 2 3 Number of Weeks after Sowing 1 ---...- Merlin 0 --- Lotus 0 ... Merlin 1 1 0 • Lotus 1 1 0 1 -6- Merlin 220 -e- Lotus 220 -6 Merlin 440 0 Lotus 440 [ Figure 7.1 4 Percentage germination at various periods after sowing in (a) non-limed, sludge amended tailings and (b) sludge amended tailing limed to pH 6.0. 4 375 all of the seedlings had died. Similar germination and seedling survival trends were exhibited by Lotus sown on the limed 220 and 440 Mg/ha sludge treatments. The initial germination rates for these two treatments were in the vicinity of 40%. The majority of the seedlings, however, died prior to the end the growing period (Figure 7 . l 4a & 7 . 14b). At the 440 Mg/ha sludge rate the addition lime at a rate equivalent to 5 . 1 Mg/ha had little effect on the germination of Merlin and Lotus (ie. germination and seedling survival trends were comparable to those at the 440 Mg/ha sludge treatments that received no l ime). Although Merlin seedlings established well on the limed 440 Mg/ha sludge treatment and grew vigorously throughout the experiment, the majority of the Lotus seedlings that germinated died-back during the germination monitoring period and in the ensuing weeks. 7.7.6 Dry Matter Yields 7.7.6.1 Non-Limed Treatments Without lime addition the performance of Lotus on the various sludge treatments was extremely poor (Figure 7 . l 5a) . No significant quantities of dry matter were produced from any of the non-limed pots sown with Lotus. The only pot to successfully yield any dry matter was one of four pots that had been amended with sludge at rate equivalent to 440 Mg/ha. The quantity of dry matter produced from this pot (O.O l g) originated from a few seedlings that were clearly unhealthy and regressing. In the case of Merlin, analysis of variance (ANOV A) indicated that significant yields of this taxa were only attained from those pots treated with 440 Mg/ha sludge (Figure 7. 1 5a) . Although small yields of Merlin were also produced from each of the four pots treated with 220 Mg/ha sludge, they were not significant. In the absence of lime these pots produced an average of just 0.032 ± 0 .005 grams/pot of dry matter. As previously discussed (section 7.7. 5 . 1 ) the seedlings established on this treatment were unhealthy and exhibited symptoms of P deficiency characterised by dark green foliage and numerous brown and purple tillers. The identification of high Olsen P levels in the treatment control pot (ie. in the absence of plants), however, casts doubt as to whether the plants were indeed P deficient or whether Olsen P is suitable for assessing P availability in sludges (Table 7. 1 3) . 3.0 (a) Non-Limed Treatments LSDl 376 I . . . . . . . . . . . . . . . j I I DB DB 0.0 r----t-----+---t---+-----"----=-=--+- 4.0 � 3.0 Q. Oro ...... 0.0 1 1 0 220 440 Sludge Application Rates [Mg/ha] (b) 1 pH-6 Liming Rate LSD2 DB Nil 1 1 0 220 440 Sludge Application Rates [Mg/ha] Merlin Lotus Figure 7.1 5 Dry matter yields of Festuca rubra (Merlin) and Lotus corniculatus in (a) non-limed tailings and (b) tailings limed initially to pH 6.0. DB = indicates that total .Qie J2.ack of seedlings occurred. LSD I = Least significant difference (speciesirate interaction) on non-limed treatments (5%); LSD2 = Least significant difference (speciesirate interaction) on 1 pH-6 treatments (5%). 3 7 7 Doubling the rate of sludge application from 220 Mglha to 440 Mg/ha increased dry matter yields of Merlin dramatically (Figure 7. 1 5a). At the higher sludge rate a 90-fold increase in yield was noted with each pot producing an average of 2.94 ± 0.39 grams of dry matter. ANOVA confirmed that the sludge had a very highly significant (P0.05). Merlin yields were, however, significantly higher on limed treatments that were amended with sludge at rates ;;:: 220 Mg/ha. Comparison of Figures 7 . 1 Sa and 7 . I Sb indicates that the addition of S . 1 Mg/ha of lime to the 440 Mglha sludge treated tailings did not substantially improve the performance of Merlin in terms of dry matter production. Lime treated pots yielded an average of just 0.3 grams additional dry matter per pot. Analysis of sown pots at the end of the growing period also indicated that there was little difference between the pH of limed and non­ limed 440 Mglha sludge treatments (Table 7 . l 4). The availability of metals (as determined by extraction with O. lM HCI, 1M KCl and 0.02M CaCb) were similarly, comparable between the two treatments (Figures 7.20 & 7.26). In comparison to Merlin the performance of Lotus on the limed treatments was poor. Although lime addition enabled Lotus to ' successfully' germinate and establish on 3 of the 4 treatments, dry matter yields of Lotus were significantly (P 1/1 Q,) c: � 20 .. c: II! :E 0 0 200 r: 'C g 500 -j II! Q,) ..J .. - - 1 00 - -e 0 a 28 56 0 28 56 0 28 600 • - Extractants 400 -, (J - . -r: 0. 1 M HC! - . N .� 200 O.02M CaCI2 e - + - 1 M KCI ----: 0 28 56 0 28 56 Days Days Figure 7.25 Changes in metal concentrations as extracted by O. 1MHCl (e), 0 .02MCaClz (0) and IMKCI (+) for l imed, 220 Mglha sludge treated tailings during the growth period 56 20 'Ci � 0) E ...... E 1 0 ::s 'c 'E ::s <{ 0 200 ...... 0) ..:.:: 0, E ';:' 1 00 Cll C. c. 0 () 0 40 ....... 0) ..:.:: 0, E ..... � 20 (II c «l 0) c III :E 0 396 � 6 1 0 - -+ (.) 4 .. � E .- 'c ::s � Cll -e 'E 5 / I!! • "0 e - - - - --e- - - - .... • « 2 «l \ () 0 0 0 28 56 0 28 56 0 28 56 1 000 400 -\ . � - � _e - - - _ __ -e -. - i - -e i • c \ "0 g 500 \ «l 200 l Cll ..J - e - - - - - . i 1 0 0 0 28 56 0 28 56 0 28 56 600 � e Extractants / e _ _ - - e 400 -' - . - O. 1 M HC! . - (.) c N 0 . 02 M CaCI2 200 � e - . - 1 M KCI 0 0 28 56 0 28 56 Days Days Figure 7.26 Changes in metal concentrations as extracted by O. lMHCl (e), O.02MCaCh (0) and IMKCI (.) for limed, 440 Mglha sludge treated tailings during the growth period 1 0 I 5 6 � 0, 4 E ...... ... Q) §: 2 o () 20 o i o o o 28 56 • 28 56 0.2 (.,) '2 � 0 . 1 i « o 3 --; 2 o 300 28 28 56 � 14P / / (/ 56 4 -, E :::5 'E 2 "'C ftl () o 1 5 1 0 5 o o o 397 28 56 28 56 Treatments Sludge I lime 28 Days 56 (.,) t: N 200 1 00 o o 28 Days 56 nil / 1 pH-6 ______ 1 1 0 ! 1 pH-6 220 / 1 pH-6 -II- 440 I 1 pH-6 Figure 7.27 Concentrations of various metals extracted by O .02MCaClo at various rates of sludge treatments which have been limed to pH 6.0 . 20 ....... Ol .:.:: en .§. E 1 0 :::l 'c 'e :s ;;a: 0 200 � en E -= 1 00 C1> 0. 0. o () o 40 o 398 6 � :'" � � �:>�� � . .. () 4 'c E :s 1 0 - � �// ,// C1> e « 2 -; ���� �> ./ ./ 1 "e 5 " III () 0 o o - -=-+ -u:; - - -". I 0 -+- o 28 56 0 28 56 o 28 56 1 500 300 200 1 00 o o 28 56 a 28 56 o 28 56 600 l • Treatments Sludge I lime 28 Days 56 () s::: N 400 200 o o 28 Days 56 - nil / 1 pH -6 ___ 1 1 0 / 1 p H -6 220 / 1 p H-6 -II- 440 / 1 p H -6 Figure 7,28 Concentrations of various metals extracted by O. lMHCI at various rates of sludge treatments which have been limed to pH 6.0. 399 (ii) Extractable Metals The addition of lime to the various sludge treatments was found to decrease the ready availability (0.02M CaCb extractable) of metals in the amended tailings, particularly in the nil and 1 1 0 Mg/ha treatments (Figures 7.23-7.26). Comparison of AI, Cu and Zn levels 4 weeks after sowing in the 1 1 0 Mg/ha treatment for example, indicates these metals were reduced from 53 1 , 240 and 1 353 mglkg in the non-limed pots to 4, 1 . 5 , 2 1 1 mg/kg respectively in the limed pots (Figures 7 . 1 8 & 7 .24). The decrease in metal availability was least marked in the 220 and 440 Mg/ha sludge treatments but still appreciable (Figures 7.25 & 7 .26). Comparison of Figures 7 .2 1 and 7 .27 clearly indicates that the addition of lime reduced metal concentrations and variability between availability of metals between the treatments, particularly in the cases of AI, Cu, Fe and Zn. In several instances the 0. 1M HCl extractable levels were also substantially reduced (Figure 7.28). Comparison of Figures 7.22 and 7.28 indicates that liming notably reduced the quantity of AI, Cd, Fe and Zn extracted in most treatments. (iii) Nutrients Soil chemical analysis of control pots again indicated that sludge addition substantially increased the tailing's nitrogen (N) and phosphorus (P) reserves (Table 7 . 1 5) . Total P levels were again increased from very low concentrations in the non-sludge amended tailings to very high levels in the 220 and 440 Mg/ha sludge treatments. At the 440 Mg/ha sludge rate total N levels in the amended tailings were approaching those typically found in soils (Table 7. 1 5) . Liming appeared to have little effect on the availability of Olsen P levels which were similar albeit slightly lower than levels in the non-limed treatments (Table 7. 1 3) . Levels of exchangeable calcium were predictably higher in all limed treatments. Exchangeable K and Mg levels were found to increase slightly with sludge addition. In contrast to Mg levels which decreased during the course of the trial, K levels were found to increase slightly during the growth period. Sodium levels were not notably effected by the combined application of sludge and lime (Table 7 . 1 5) . The CEC of the various limed-treatments was found to increase slightly with increasing sludge addition. 400 Table 7.1 5 Soil fertility parameters of limed control pots determined at sowing (Week 3) and at the end of the growing period (Week 1 1 ) . Medium values for New Zealand soils are based on the data of Blakemore et aI. ( 1 987). *Values are expressed as mg/kg. t Values expressed as meq/I OOg . nd =:0 not determined Sludge Application Levels fMg/kg] Soil Medium nil 1 1 0 220 440 Parameter Values Week 3 Week 1 1 Week 3 Week 1 1 Week 3 Week 1 1 Week 3 Week 1 1 pH 5.3-6 .5 6 .9 1 6.46 6 . 1 8 4.30 6.00 4 .37 5 .71 4 .87 Olsen P* 20-30 4.5 2 .0 3 1 27 52 58 70 86 Total p* 400-800 nd 1 02 nd 569 nd 1 1 74 nd 2302 Total N* 3000-6000 nd 1 3 .0 nd 467 nd 1 282 nd 2654 Sulphate * 1 5-50 1 770 1 9 1 0 1 485 1 809 1 470 2 1 1 1 1 500 2 1 1 1 Exch. Cat 5 .0- 1 0.0 1 2.9 1 2.0 12 .7 1 1 . 5 1 5 .8 1 3 .6 1 7.4 1 6 .5 Exch. Kt 0.5-0.8 0.02 0. 1 4 0 .08 0. 1 5 0 .04 0 . 1 7 0.09 0.20 Exch. Na! 0 .3-0.7 0 . 1 0. 1 0.2 0. 1 0 .2 0 .2 0.2 0.2 Exch. Mgt 1 .0-3 . 0 0. 1 7 0. 1 2 0.46 0.3 1 0 .58 0 .44 0.90 0.75 CECt 1 2-25 nd 1 3 1 7 1 6 2 1 20 24 23 7.7.7 Conclusions of Trial Two • A 'satisfactory' cover of Merlin was established on the non-limed 440 Mg/ha sludge treatment. Below this rate, however, Merlin either failed to germinate (nil and 1 1 0 Mg/ha sludge treatments) or seedlings died-off during the growing period. The addition of lime to the various sludge treatments facilitated the establishment of Merlin on all limed treatments. • Whilst moderate germination rates were attained for Lotus on all treatments (bar that which received no lime nor sludge), this species performed very poorly on both the limed and non-limed sludge treatments in terms of dry matter yield produced. In the non-limed treatments, Lotus in fact failed to produce any dry matter at all and although small yields of Lotus were attained from 3 of the 4 sludge treatments that were limed, the established plants were clearly unhealthy and showed symptoms of metal toxicity and/or nutrient imbalance. • As the sludge to lime ratio increased so too did Merlin dry matter yields. The positive effects of increasing this ratio is believed to reflect an improvement in the physico­ chemical properties of the amended tailings. Specifically it appears that sludge rates of 440 Mg/ha (both with or without the addition of lime) were capable of maintaining 401 the pH at a level more conducive to nutrient uptake while simultaneously ensuring metal concentrations were sufficiently low in the tailings solution to facilitate the growth of metal-tolerant Merlin. • The treatment of the tailings with 1 1 0 Mg/ha of sludge, either with or without the addition of lime ( 1 0 . 1 Mg/ha), failed to facilitate the establishment of Lotus. Whilst Merlin established on the limed treatment, the level of growth was not significantly greater than that treated with lime only ( 1 6.5 Mg/ha) . In the absence of lime, the addition of 1 1 0 Mg/ha sludge increased the already high availability of some metals in the tailings. • The significantly higher yields of Merlin obtained from the limed 220 and 440 Mglha rate indicate that growth conditions are superior in high sludge to lime ratio treatments . Results from this trial clearly indicate that whilst the application of lime at a rate equivalent to 16 .5 Mg/ha may facilitate the growth of plants, the resulting physico-chemical conditions are far from optimal in terms of potential production, particularly in the case of Merlin. • It is evident that a satisfactory cover of Merlin is attainable through the amendment of tailings with the equivalent to 440 Mg/ha of North Shore sewage sludge. The results of this trial indicate that this treatment may inhibit severe acidification of the tailings and facilitate plant growth for at least 24 weeks. 7.8 Chapter 7 Summary & Conclusions To briefly summarise the conclusions of the plant growth trials (see sections 7.66 and 7.77), it is evident that on tailings treatments sown with metal-tolerant varieties of F. rubra or A. capillaris, satisfactory cover was achieved by the addition of lime at a rate of 1 6 .5 Mg/ha ( 1 pH-6 rate) or composted sewage sludge at rates 2220 Mg/ha. On the lime amended tailings metal-tolerant plant taxa were found to substantially out yield their non-metal-tolerant counterparts. Whilst plants were at least temporarily established on lime amended tailings, dry matter yields were relatively much lower than on treatments receiving high rates of sludge. The addition of lime at a rate of 1 1 2 Mg/ha (TAGP rate) was found to detrimentally affect plant growth due to pH-induced nutrient deficiencies. In contrast, plant growth improved significantly with increasing sludge addition. 402 CHAPTER S Thesis Summary & Conclusions 8.0 Introduction The primary objective of this thesis was to determine the potential for revegetating the Tui mine tailings. To achieve this objective the relevant literature was reviewed, the tailings were physically and chemically characterised, small scale incubation and plant growth experiments were developed and conducted in the glasshouse and the overall findings discussed. The main conclusions of the literature review and the experimental work are summarised below. 8.1 Literature Review • The Tui Mine is situated on the flanks of the Kaimai Ranges and was last actively mined between 1 967 and 1 974. During this time an estimated 1 63 ,000 tonnes of base-metal ore was extracted and processed on site. The Tui ore deposits comprised two brecciated quartz reefs which were effectively cemented by an aggregate of sulphide minerals; predominantly sphalerite, galena chalcopyrite and pyrite. The ore also contained appreciable quantities of gold and silver. • The mining operation generated an estimated 1 00,000 m3 of sulphide-bearing tailings. These finely-ground tailings are contained in a rudimentary dam located at an altitude of 350 m in one of numerous steep stream valleys which dissect hydrothermally altered andesitic country rocks on Mt. Te Aroha. • Rainfall at the Tui mine site is high (c. 2000 mm1yr), well distributed but, at times, very intense. The Tui and Tunakohoia streams, which flank the Tui mine site, intercept catchment rainfall as well as highly contaminated drainage from the tailings dam and mine adits. The inflow of acid mine drainage (AMD), characterised by low pH and high concentrations of Cd, Cu, Fe, Mn and Zn, has rendered the Tui Stream largely devoid of aquatic life . • The vegetation type surrounding the mine site is predominantly dense, secondary native bush which contains a variety of hardwood species. The ground cover is dominated by mosses and ferns which testify to the moist conditions prevailing at the site. 403 • The Matamata-Piako District Council and the Department of Conservation, who currently own the Tui mine site, are responsible for mitigating environmental problems generated by the past mining operation. • Literature pertaining to mine site degradation and reclamation indicates that the certain problems encountered at the Tui mine site, including the generation of AMD and the scarcity of vegetation on disturbed areas, are common features of abandoned mine sites where sulphide-bearing mine wastes have been excavated and left exposed to the atmosphere. • The severe acidification of mining wastes and the concomitant release of AMD are problems most likely to arise from materials which have a high pyrite to carbonate mineral ratio and are situated at sites where moist, well aerated and warm conditions prevail . Such conditions will generally facilitate abiotic pyrite oxidation at circmnneutral pH and biologically catalysed oxidation under extremely acidic conditions « pH 3 .0). • Whilst a wide range of technologies exist to treat metal contaminated mine drainage (including AMD), the most promising systems in terms of cost and ongoing maintenance appear to be passive water treatment systems. These systems exploit natural processes, that occur in aerobic and anaerobic wetlands, to precipitate and complex heavy metals. Anoxic Limestone Drains (ALDs) may be used, in conjunction with wetlands, to increase the bicarbonate alkalinity and pH of drainage waters and to ultimately reduce acidity and metal loading. • Preventative rather than palliative measures are, however, widely recognised as being more effective means of combating sulphide oxidation and AMD generation. Low permeability capping materials combined with surface vegetation are common methods used to prevent or reduce sulphide oxidation and to improve the aesthetic value of degraded mine sites. • Previous research indicates that the addition of liming agents and/or organic wastes to sulphide generating materials may not only serve to ameliorate growing conditions but may also retard the rate of pyrite oxidation and thus inhibit the onset or continued generation of AMD. • In circumstances where lime is added to sulphide-bearing material intended for use as a ' surrogate soil' , the material ' s inherent Acid Generating Potential (AGP) must be evaluated to ensure that sufficient lime is incorporated to prevent rapid 404 reacidification of the material. Whilst munerous methods have been developed to assess the AGP of mining wastes, there is currently no universally accepted method. • Although literature pertaining to the reclamation of colliery wastes is abundant, comparatively little literature exists that specifically discusses the characterisation and reclamation of base-metal tailings. 8.2 Characterisation of the Tui Mine Tailings • Physical characterisation of the Tui tailings usmg laboratory and field based measurements indicated that whilst they are not an optimal plant growth medium, they pose no major physical limitations to plant growth. Poor aeration and susceptibility to erosion are, however, problems that could potentially affect establishing plants in certain areas of the dam. • In terms of their chemical properties, the tailings in their unamended state were identified to be an extremely hostile plant growth medium. The tailings' very low pH combined with their high base-metal content are the two main factors inhibiting plants from co Ionising the site. Growing conditions within the tailings are further exacerbated by their negligible organic matter content and low clay content. • Elemental analysis of the tailings indicated that both the total and labile (O. lM HCI extractable) metal contents of the tailings are spatially highly variable. Metal concentrations were also found to vary substantially with depth. • With the notable exception of Pb, the concentrations of most metals were found to dramatically increase below about 200 rom depth. Accordingly, total sulphur levels and the theoretical AGP of the tailings also increased with depth as sulphide mineral concentrations increased. • Whilst the surface tailings (0-200 rom) were found to contain relatively low concentrations of sulphide minerals, it i s estimated that a liming rate in the vicinity of 50 Mg/ha would be required to prevent long term reacidification of the surface tailings. • Chemical characterisation of the tailings indicated that, in terms of their AGP and metal content, the surface tailings (0-200rom) are far better suited for use as a growing medium than tailings obtained from deeper within the profile. Plants, however, are not envisaged to begin colonising the tailings until the adverse chemical growing conditions are ameliorated via some form of amendment( s) which would 405 have to simultaneously provide nutrients, mcrease pH and reduce metal bioavailability. • XRD analysis of the Tui mine tailings showed that they comprise primarily quartz with subordinate kaolinite and pyrite. Fractionation of the tailings using a solution of sodium polytungstate (density = 2.8 glcm3) facilitated the detection of small but appreciable quantities of sulphide minerals such as anglesite, chalcopyrite, galena and sphalerite in most samples. • The mass of the heavy mineral fraction of the Tui tailings was found to significantly increase with depth from about 1 .5% at 0-1 00 mm depth to more than 1 6% at 500- 600 mm depth. • The detection of highly insoluble PbS04 (anglesite) in the tailings samples provided a ready explanation for the presence of very high total Pb concentrations in surface 200 mm of the profile where other heavy metals were contrastingly found in relatively low concentrations. • XRD results substantiate the findings of the chemical characterisation that surface tailings (0-200 mm) have been extensively oxidised. Evidence includes; (i) the increasing proportion of heavy minerals with depth, (ii) the lower levels of XRD­ detectable sulphide minerals in the surface tailings and (iii) the presence of relatively high concentrations of secondary anglesite in the surface tailings. • SEM images of remnant sulphides isolated in the Tui tailings provided further evidence that the surface tailings (0-200 mm) have undergone extensive weathering. Sulphide minerals isolated from this zone were characterised by their craggy and highly irregular shape indicating that they have been extensively weathered via oxidation. In contrast, sulphide minerals obtained from deeper within the profile generally exhibited highly angular, and in some instances cubic morphology, characteristic of unaltered pyrite and other sulphide minerals • Based on SEM and EDS analysis it i s thus postulated that the original distribution of base-metals within the Tui tailings profile was relatively uniform with depth and that the depletion of Cu, Fe and Zn in the surface tailings (0-200 mm) has resulted from the oxidation of remnant sulphide minerals within this relatively well aerated zone and the subsequent dissolution and leaching of weathering products. • Analyses conducted using EDS support XRD findings which indicate that the Tui tailings comprise primarily quartz but in addition contain a variety of heavy metal- 406 bearing minerals. The processing of the Tui ore, however, has selectively enriched the tailings with pyrite. Relatively high concentrations of pyrite persist below 200 mm depth and form a significant source of potential acidity. .. The results of a lime incubation trial conducted over a 112 years period confirmed that the uppermost tailings sample assessed (0- 100 mm) had very low AGP and that the sample's pH was readily maintained at circumneutral levels via the addition of relatively low rates of lime. • Tailing samples obtained from below 100 mm depth, however, were found to reacidifY within a year when amended with lime at rates based on those required to raise the pH of the tailings to approximately 6.0 ( 1 pH-6 liming rate). Two or four times the 1 pH-6 liming rate were required to prevent reacidification of subsurface tailings over the trial period. • Contrary to estimated TAGP values, which indicated that tailings obtained from between 500-600 mm were likely to be the most reactive (in terms of acid generation and metal liberation), the incubation study suggested that samples obtained from between 200 and 400 mm depth were most reactive. The greater reactivity of these samples under aerobic incubations conditions (considered to be conducive to the biochemical oxidation of sulphides) is attributed to their appreciably finer grain-size and hence larger reactive surface area. .. Whilst high rates of lime were generally found to decrease the availability of labile heavy metals in the tailings, the lowest rate in contrast was found to increase the availability of eu, Pb and Zn in the tailings and hence exacerbate potential growing conditions. The availability of H+, S04 = and soluble Fe was generally greater in samples treated with the 12 pH-6 liming rate than in the control samples. This finding has important implications for reclaiming the Tui mine sites as it indicates that the application of rates of lime below a certain threshold level may in fact enhance sulphide oxidation rates and further degrade growing conditions. 8.3 Plant Growth Trials • The failure of all plant taxa (including metal-tolerant varieties) to germinate on fertilised but otherwise unamended T ui tailings confirmed that the tailings are an extremely hostile plant growth medium. Extreme acidity and the high availability of 407 phytotoxic metals (AI, Cu, Fe and Zn) were considered to be the primary factors inhibiting gennination. • The addition of a ' low' rate of lime (8.25 Mglha) to the tailings, however, was not found to substantially improve growth conditions. To the contrary, it was found to increase the relative bioavailability of some metals as the tailings re-acidified. This finding is consistent with observations made in the lime incubation trial. • The addition of lime at rates equivalent to 1 6.5 Mglha ( l pH-6 rate) or 1 1 2 Mglha (TAGP rate) facilitated gennination and plant growth of all taxa by reducing the pH and bioavailability of metals in the tailings. Substantially greater mean yields were, however, obtained on tailings amended with the lower of these two liming rates; a finding that is attributed to the adverse effects of the very high liming rate on plant nutrition and particularly on P uptake. • Metal tolerant Festuca rubra (var. Merlin) was clearly the most successful of the p lant taxa trialed in this study in tenns of the biomass produced and vigour. Merlin significantly out-yielded the non-metal-tolerant variety of Frubra (var. Lobi) at both effective liming rates. A significant difference in yield between the metal-tolerant and non-metal-tolerant Agrostis capillaris cultivars was noted only at the lowest effective liming rate ( 1 pH-6 rate) . At the higher liming rate both cultivars of A. capillaris perfonned very poorly. Rhizobial inoculation of Lotus corniculatus failed to significantly improve this species' poor perfonnance at both liming rates. • Attempts to establish plants on the lime amended (and fertilised) tailings clearly demonstrated that the liming requirements of the Tui tailings (in tenns of optimising plant grovvih) is a trade-off between diminished plant response resulting from the application of high rates of lime and the need to ensure that sufficient lime is added to prevent reacidification of the growing medium. • The amendment of the tailings with various rates of composted sewage sludge and lime did little to improve the growth perfonnance of L. comiculatus. This species failed to establish on all non-limed sludge treatments. In the case of Frubra var. Merlin, however, very satisfactory plant growth was attained on 440 Mglha sludge treatments both with or without the addition of lime. Satisfactory Merlin growth was also achieved on 220 Mglha sludge treatments where lime was added. • The satisfactory performance of Merlin on treatments amended with high rates of sludge is believed primarily to reflect the ability of sludge to :- (i) supply plant 408 nutrients, particularly N & P, (ii) increase and maintain pH at levels more conducive to the growth of plants, (iii) reduce the bioavailability of heavy metals and (iv) improve the physical characteristics of the rooting medium. • Results from the plant trials clearly indicate, that whilst the application of lime at a rate equivalent to 1 6.5 Mglha may facilitate the growth of plants, the resulting physico-chemical conditions are far from optimal in tenns of potential production, particularly in the case of Merlin. It is evident that much better Merlin establishment was attained through the amendment of tailings with high rates of composted sewage sludge. 8.4 Overall Conclusion Based on the information presented in this thesis, it is evident that the Tui mining operation was conducted in an extremely sensitive mining environment. Environmental degradation was exacerbated at the site by the failure of the mining company to implement what would now be considered to be very rudimentary measures to minimise environmental degradation. Characterisation of the tailings, that were effectively dumped at the processing site, indicate that they are chemically very poorly suited to plant growth due to their very low pH, high content of phytotoxic metals and their potential to generate further acidity through sulphide oxidation. Whilst very high levels of acid generating sulphide minerals were found to persist in the tailings below 200-300 mm depth, it was demonstrated that plants can be established on the highly weathered surface tailings (0-200 mm) via the application of sludge and/or lime under controlled ot conditions. The usefulness of amending the tailings with a sourcetorganic matter and utilising metal-tolerant plant species to revegetate the tailings was clearly demonstrated by the results of plant growth trials. It is envisaged that unless attempts are made to reclaim the site, sulphide oxidisation and concomitant acid generation and metal liberation will continue to contaminate ground and stream waters indefinitely. The establishment of vegetation on the tailings would, at the very least, serve to physically stabilise the tailings and improve the aesthetic value of the site. 8.5 The Scope for Future Work Whilst the research presented in this thesis adds to our ' reclamation knowledge base' by elucidating environmental problems at the Tui mine site and by providing insight into 409 the viability of revegetating the tailings, by its very nature it has created niche opportunities for future research. Perhaps the most obvious and important question generated by this study is . . . "Can vegetation be established on tailings under the environmental conditions prevailing at the Tut mine site?" It is envisaged that the next logical step leading on from this study would be to investigate this research question by establishing in situ field trials at the Tui mine site. Clearly, the very promising plant growth results obtained in the glasshouse using metal-tolerant Festuca rubra in conjunction with high rates of sludge should strongly influence future field trial designs. The incorporation of native plants into such field trials would have obvious merit considering the nature of the vegetation surrounding the tailings dam. The effects of various surface treatments in terms of their ability to inhibit subsurface sulphide oxidation rates (and ultimately AMD) require investigation. Whilst there is much speculation about the ability of organic matter-based covers to inhibit sulphide oxidation, hard evidence from in situ field trial work is scarce. It is envisaged that the use of in situ leaching columns and/or vegetated lysimeters may provide an opportunity to investigate the impact of vegetative covers on oxidation rates as well as the possible alteration of chemical and biological processes within the tailings profile. Other areas of research that require and/or warrant further investigation at the Tui mine site include:- • The determination of suitable fertiliser regimes for revegetative reclamation. • The apparent ability of low rates of lime (or sludge) to enhance sulphide oxidation. • The impact of capillary rise and the effects of a high water table in the field on plant establishment. • Suitable amendments/strategies for revegetating dam batters. • Assessment of indigenous metal-tolerant plant species for revegetation. • Identification of leguminous species tolerant to growing in amended tailings. • The effectiveness of a capping layer(s) in reducing capillary rise and isolating sulphidic materials from water and oxygen. • The potential use of passive treatment systems to treat water from the site clearly needs assessing as a part of the overall rehabilitation strategy. 4 10 References Aachib M., Aubertin M. & Chapius R.P. ( 1 994) Column tests investigation of milling waste properties used to build cover systems. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. Adams G.l.D. , Wodzicki A. & Weissberg B.G. ( 1 974) K-Ar Dating of Hydrothermal Alteration at Tui Mine, Te Aroha, New Zealand. New Zealand Journal of Science 1 7: 1 93 - 1 99. Ahmed S.M. ( 1 994) Surface Chemical Methods of Forming Hardpan in Pyrrhotite Tailings and Prevention of Acid Drainage. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Ahmed S.M. ( 1 995) Chemistry of Pyrrhotite Hardpan Formation. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 -June 1 , 1 995 . Sudbury, Ontario, Canada, vol 1. CANMET, Ottawa. Ahonen 1. & Tuovinen O.H. ( 199 1 ) Temperature Effects on Bacterial Leaching of Sulphide Minerals in Shake Flask Experiments. Applied and Environmental Microbiology 57 ( 1 ): 1 38-145. Allaway W.H. ( 1 977) Food Chain Aspects of the Use of Organic Residues. In: Dinauer R.C.M.E., Stelly M., Padrutt 1.M., Nauseef J.H., Elliot 1.F., Stevenson F.J., Frink c.R. , Hill R.R., Hortenstine C.C., Kilmer VJ., Monke E.l., Viets F.GJ. (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin (N). Allison F.E. ( 1 973) Soil Organic Matter and its Roll in Crop Production. Developments in Soil Science, vol 3 . Elsevier Scientific Publishing Company, Amsterdam. Alloway B.l. (ed) ( 1 995) Heavy Metals in Soils, 2nd edn. Blackie Academic & Professional, New York. Ammons 1.T. & Shelton P.A. ( 1 988) A Comparison of Results from Acid-Base Accounting Versus Potential Acidity Measured by the Peroxide Oxidation of Weathered and Unweathered Soils Containing Pyrite. In: ASSMRJUSDI (ed) Mine Drainage and Surface Mine Reclamation: Mine Water and Mine Waste. Conference Proceedings. Information Circular 9 1 83, vol 1 . ASSMRlUSDI, Pittsburgh. 41 1 Andrew C.S. & Hegarty M.P. (1 969) Comparative Responses to Manganese Excess of Eight Tropical and Four Temperate Pasture Legume Species. Australian Journal of Agricultural Research 20: 687-696. Andrews G.F. ( 1 988) The Selective Adsorption of Thiobacilli to Dislocation Sites on Pyrite Surfaces. Biotechnology and Bioengineering 3 1 : 378-3 8 1 . Anon. ( 1 974) Mine Waste Pollution of the Molongolo River. Final Report on Remedial Measures. Joint Government Technical Committee. 68 p. Australian Government Publishing Service, Canberra. Antonovics J., Bradshaw A.D. & Turner RG. ( 1 97 1 ) Heavy Metal Tolerance in Plants. Advances in Ecological Research 7: 1 -85 . Arand J.G. ( 1 986) Mass movement hazard at Te Aroha, North Island, New Zealand. MSc. Thesis, University ofWaikato, Hamilton, New Zealand. Archer LM., Marshman N.A. & Salomons W. ( 1 988) Development of a Revegetation Programme for Copper and Sulphide-Bearing Mine Wastes in the Humid Tropics. In: Salomons W., Forstner U. (eds) Environmental Management of Solid Waste - Dredged material and mine tailings. Springer-Verlag, New York. Arora H.S., Pugh C.E., Dixon J.B. & Hossner L.R ( 1 979) Pyrite and Acidity Fonnation. In: Hossner L.R (ed) Reclamation of Surface-Mined Lignite Spoil in Texas. RM- 1 0. Athanasopoulos N. (undated) Flame Methods Manual for Atomic Absorption. GBC Scientific Equipment Pty Ltd. 1 2 Monterey Road, Dandenong, Victoria, Australia. Part No: 0 1 -001 9-00. Atkinson S.L., Buckley G.P. & Lopez-Real J.M. ( 199 1 ) Evaluation of Composted Sewage Sludge/Straw for the Revegetation of Derelict Land. In: Davies M.C.R. (ed) Land Reclamation. An End to Dereliction? Elsevier Applied Science, London. Aylward G.H. & Findlay T.1.V. (1 974) SI Chemical Data, 2nd edn. John Wiley & Sons Australasia Pty. Ltd., New York. Babich H. & Stotsky G. (1 977) Sensitivity of Various Bacteria, including Actinomycetes and Fungi, to Cadmium and the Influence of pH on Sensitivity. Applied and Environmental Microbiology 33 : 68 1 -688. Backes c.A., Pulford LD. & Duncan H.J. ( 1 986) Studies on the Oxidation of Pyrite in Colliery Spoil. 1 . The Oxidation Pathway and Inhibition of the Ferrous-Ferric Oxidation. Reclamation and Revegetation Research 4: 279-291 . Baker A., Brooks RR & Reeves R. ( 1 988) Growing for Gold . . . and Copper. . . and Zinc. New Scientist 1 0: 44-48. 412 Baker A.J.M., McGrath S .P . , Sidoli C.M.D. & Reeves R.D. ( 1 995) The Potential for Heavy Metal Decontamination. Mining Environmental Management 3(3): 12- 14. Baker D.E. & Amacher M.C. ( 1 982) Nickel, Copper, Zinc and Cadmiwn. In: Page A.L., Miller RH., Keeney D.R (eds) Methods of Soil Analysis. Part 2 - Chemical and Microbological Properties, 2nd edn. American Society of Agronomy, Madison, Wisconsin. Baldi F. & Olson G.J. ( 1 987) Effects of Cinnabar on Pyrite Oxidation by Thiobacillus ferrooxidans and Cinnabar Mobilization by a Mercury-Resistant Strain. Applied and Environmental Microbiology 53(4): 772-776. Bamhisel R.I., Powell J.L., Akin G.W. & Ebelhar M.W. ( 1 982) Characteristics and Reclamation of Acid Sulphate Mine Spoils. In: Acid Sulphate Weathering SSSA 10. (N). Bartlett RJ. & Riego D.C. ( 1 972) Effect of Chelation on the Toxicity of Aluminium. Plant and Soil 37: 4 19-423. Baskiran, S. , Bolan, NS., Rahman, A & Tillman, RW (1 996) Effect of Exogenous Carbon on the Sorption and Movement of Atrazine and 2,4-D by Soils. Australian Journal of Soil Research 34: 609-622. Bastias J.G. ( 1 987) Retreatment of Radioactive Gold Bearing Tailings and Rehabilitation of Mill and Tailings Dump Sites at Rockhole and Moline, Northern Territory (a personal view). In: Australian Mining Industry Council Environmental Workshop Papers (Adelaide, September 2 1 -25, 1 987). Australian Mining Industry Council, Canberra. Bates T.E. ( 1 989) Te Aroha goldfield - Tui and Waiorongomai GoldiSilverlBase Metal Prospects. In: Kear D. (ed) Mineral deposits of New Zealand. Monograph 1 3 . Australian Institute of Mining and Metallurgy. Bell A.V., Riley M.D. & Yanful E.K. ( 1 994) Evaluation of a Composite Soil Cover to Control Acid Waste Rock Pile Drainage. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. Bell L.C., Ward S.C., Kabav E.D. & Jones C.J. ( 1 989) Mine Tailings Reclamation in Australia - An Overview. In: Walker D.G., Powter C.B., Pole M.W. (eds) Proceedings of the Conference - Reclamation, a Global Perspective. 23-3 1 August, 1 989, Calgary, Canada, 2nd edn. Alberta Land Conservation and Reclamation Council. Report RRTAC 89-2, Edmonton, Canada. Bell P.RF., Greenfield P.F., Kelly G.J. & Krol A.A. ( 1 982) Treatment of Acid Mine Drainage. In: Darwin Environmental Workshop, Discussion Papers (September 1 982). Australian Mining Industry Council, Darwin. 413 Bennett J .C . & Tributsch H. ( 1 978) Bacterial Leaching Patterns on Pyrite Crystal Surfaces. Journal of Bacteriology 1 34 ( 1 ): 3 1 0-3 1 7. Bennett J.W., Harries J.R. & Ritchie A.I.M. ( 1 987) Monitoring the Effectiveness of Rehabilitating the Waste Rock Dumps at Rum Jungle. In: Proceedings of "Mining and Environment - A Professional Approach" National Conference (Brisbane 20-2 1 July 1 987), Brisbane. Bennett O.L., Mathias E.L., Armiger W.H. & Jones IN.J. ( 1 978) Plant Materials and Their Requirements for Growth in Humid Regions. In: Schaller F.W., Sutton P. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison, Wisconsin. Berg W.A. ( 1 978) Limitations in the Use of Soil Tests on Drastically Disturbed Lands. In: Schaller F.W., Sutton P. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison. Bi Tech Publications ( 1 989) Draft Acid Rock Drainage Technical Guide. Volume l . Vancouver, Canada. Blaiklock K. ( 1981) The Environmental Impact of Mining. New Zealand Environment 29 Autumn: 6- 1 6. Blakemore L.c., Searle P.L. & Daly B.K. (1 987) Methods for Chemical Analysis of Soils. NZ Soil Bureau Scientific Report 80. Department of Scientific and Industrial Research. , Lower Hutt, New Zealand. Boffa F. ( 1 990) Creating New Landscapes for Disturbed Land. In: Gregg P.E.H., Stewart R.B., Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Bokich lC. ( 1 995) Golden Cross Mine - Gold Mining in a Sensitive New Zealand Environment. In: Mauk J.L., St-George J.D. (eds) Proceedings of the 1 995 PACRIM Congress. 1 9-22 November 1 995. The Australasian Institute of Mining and Metallurgy, Carlton, Australia. Bollard R.J. ( 1991) Tribunal Report on Prospecting Application by Tasman Gold Limited. Report to the Minister of Energy, Decision No AK 52/91 : 25 . Boorman R.S . & Watson D.M. ( 1 976) Chemical Processes in Abandoned Sulphide Tailings Dumps and Environmental Implication of New Brunswick. elM Bulletin 69: 86-96. Bowen H.J.M. ( 1 979) Environmental Chemistry of the Elements. Academic Press, London. 414 Boyle RW. ( 1 979) The Geochemistry of Gold and its Deposits (together with a chapter on geochemical prospecting for the element). Canadian Geological Survey, Ottawa. Bradshaw A.D. ( 1 952) Populations of Agrostis tenuis resistant to Lead and Zinc. Nature 1 69 : 1 098. Bradshaw A.D. & Chadwick M.J. ( 1 980) The Restoration of Land. The Ecology and Reclamation of Derelict and Degraded Land. Studies in Ecology Volume 6. Blackwell Scientific, Oxford. Bradshaw A.D. & McNeilly T. ( 1 98 1 ) Evolution and Pollution. Studies in Biology No. 1 30. Edward Arnold, London. Brady J.E. & Humiston G.E. ( 1 982) General Chemistry - Principles and Structure, 3rd edn. John Wiley & Sons, Inc. , New York. Brady K.B. , Perry E.F., Beam RL., Bisko D.C., Gardner M.D. & Taratino J.M. ( 1 994) Evaluation of Acid-Base Accounting to Predict the Quality of Drainage at Surface Coal Mines in Pennsylvania, USA. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Brathwaite RL. & Blattner P. ( 1 995) The Waihi Epithermal Gold-Silver-Quartz Vein System, New Zealand: a High Throughput Geothermal System of Late Miocene Age. In: Mauk J.L., St George J.D. (eds) Proceedings of the PACRIM Congress 1 995. The Australian Institute of Mining and Metallurgy, Bayswater. Brathwaite R.L. & Christie A.B. ( 1 996) Geology of the Waihi Area. Institute of Geological & Nuclear Sciences Limited, Lower Hutt, New Zealand. Geological Map 2 1 . Scale 1 : 50 000. Brathwaite RL., Christie A.B. & Skinner D.N.B. (1 987) The Hauraki Goldfield - Regional Setting, Mineralisation and Recent Exploration. In: Kear D. (ed) Mineral Deposits of New Zealand. Australasian Institute of Mining and Metallurgy Monograph 1 3 . Australasian Institute of Mining and Metallurgy, NA. Brodie G.A. & Hedin R.S. (1 994) Constructed Wetlands for Treating Acid Drainage Workshop. In: Workshop held at the International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage. Pittsburgh, Pennsylvania, April 25-29, 1 994. Brookes P.C. & McGrath S.P. ( 1 984) Effects of Metal Toxicity on the Size of the Soil Microbial Biomass. Journal o/Soil Science 35 : 341 -346. Brookins D.G. ( 1 988) Eh-pH Diagrams for Geochemistry. Springer-Verlag, Berlin. 4 1 5 Brooks R.R. ( 1 983) Chapter 19: Statistical Interpretation of data from biological prospecting methods. In: Brooks R.R. (ed) Biological Methods of Prospecting for Minerals. John Wiley & Sons, New York. Brooks R.R. ( 1 987) Serpentine and its Vegetation. A Multidisciplinary Approach, vol 1 . Dioscorides Press, Portland, Oregon. Brooks R.R. ( 1995) Plants that Scavenge Minerals. Commercial Horticulture, November: 49-50. Brown E.J., Rasley B.T., Dixon D.P., Hong S., Luong H.V. & Braddock IF. ( 1 990) Microbial Ecology of Thiobacillus ferrooxidans. Final Technical Report to U.S. Department of the Interior. Geological Survey, Washington, D.C. Report IWR - 1 1 3 . National Technical Information Service, Springfield, VA. Brownlow A.H. ( 1 979) Geochemistry. Prentice-Hall, Englewood Cliffs, New Jersey. Bruere A.N. ( 1981) Investigation into Animal Health on Farms Associated with Mining 0/ Waipipi Iron Sands (unpublished report). Massey University, Palmerston North, New Zealand. Bruynesteyn A. & Hackl R.P. ( 1 982) Evaluation of Acid Production Potential of Mining Waste Materials. Minerals and the Environment 4: 5-8. Burns M.S. ( 1 970) Determination of Pyritic Sulphur in Australian Coals. Fuel 49: 126- 133 . Burt R.A. & Caruccio F.T. ( 1 986) The Effect of Limestone Treatments on the Rate of Acid Generation from Pyritic Mine Gangue. Environmental Geochemistry and Health 8 (3): 7 1 -78. Callahan J. ( 1 987) A Non Toxic Heavy Liquid and Inexpensive Filters for Separation of Mineral Grains. Journal o/Sedimentary Petrology 57: 765-766. Carlstrom M.G., Amendola F.A., Shay D.A. & Dollhopf D.J. ( 1 987) Sodium. In: Williams R.D., Schuman G.E. (eds) Reclaiming mine soils and overburden in the Western United States: Analytical parameters and procedures. Soil Conservation Society of America, Iowa. Carter D.A. ( 1 98 1 ) Submissions to Commerce and Energy Select Committee. Hauraki Catchment Board and Regional Water Board. New Zealand Carter L. & Pharo C.H. ( 1 965) Tui Mine, Te Aroha. Tane 2: 55-59. Caruccio F.T. ( 1 973) Characterisation of Strip Mine Drainage by Pyrite Grainsize and Chemical Quality Existing Groundwater. In: Hutnik R.J., Davis G. (eds) Ecology and Reclamation of Devastated Land, vol 1 . Gordon & Breach, London. 4 16 Caruccio F.T. ( 1 975) Estimating the Acid Potential of Coal Mine Refuse. In: Chadwick M.J., Goodman G.T. (eds) The Ecology of Resource degradation and Renewal. Blackwell Scientific Publications. London. Caruccio F.T., Ferm J.C., Home J., Geidel G. & Baganz B. ( 1 977) Paleoenvironment of Coal and its Relation to Drainage Quality. Intergency Energy-Environment researach and Development Program. Rep., EPA-60017-77-067, Environmental Protection Agency, Cincinnati. Caruccio F.T., Geidel G. & Pelletier M. ( 198 1 ) Occurrence and Prediction of Acid Drainages. Journal of the Energy Division, American Society of Civil Engineers 1 07(EYI): 1 67- 1 68. Caruccio F.T., Geidel G. & Williams ( 1 984) Induced Alkaline Recharge Zones to Mitigate Acid Seeps. In: Graves D. (ed) Proceedings of National Symposium on Surface Mining Hydrology, Sedimentology and Reclamation. Lexington, University of Kentucky, December 2-7. OES Publications, Lexington, Kentucky. Caruccio F.T., Hossner L.R. & Geidel G. ( 1 988) Pyritic Materials: Acid Drainage, Soil Acidity and Liming. In: Hossner L.R (ed) Reclamation of Surface-Mined Lands, vol 1 . CRC Press, Florida. Caruccio F.T., Geidel G. & Bradham W.S. ( 1 990) Accelerated Pyrite Oxidation/Enhanced Alkalinity Couple to Reduce Acid Mine Drainage. US Bureau of Mines Report No. J0178026, August 1 990, Pittsburgh Research Center. Chadwick MJ. ( 1 973) Methods of Assessment of Acid Colliery Spoil as a Medium for Plant Growth. In: Hutnik RJ., Davis G. (eds) Ecology and Reclamation of Devasted Land, vol 1 . Gordon & Breach, New York Chamber of Mines of South Africa ( 198 1 ) Guidelines for the Rehabilitation of Land Disturbed by Surface Coal Mining in South Africa. Chaney R.L. & Giordano P.M. ( 1977) Microelements as Related to Plant Deficiencies and Toxicities. In: Dinauer RC.M.E., Stelly M., Padrutt lM., Nauseef J.H., Elliot L.F., Stevenson F.J., Frink C.R, Hill R.R, Hortenstine C.C., Kilmer V.J., Monke E.J., Viets F.GJ. (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison,Wisconsin (N). Chaney R., Brown S. , Li Y., Angle J.S., Homer F. & Green C. ( 1 995) Potential use of Metal Hyperaccumulators. Mining Environmental Management 3(3): 9- 1 1 . Christie A.B. & Robinson B.W. ( 1 992) Regional Sulphur Isotope Studies of Epithermal Au-Ag-Pb-Zn-Cu Deposits in the Hauraki Goldfield, South Auckland. New Zealand Journal of Geology and Geophysics 35 : 145- 1 50. 4 17 Christie T . & Brathwaite B . ( 1 997) Geology of the Gold Deposits. In: Louthean R . (ed) Minerals Resources of New Zealand, 1 997 edn. Louthean Publishers Pty Ltd. New Zealand Crown Minerals, Perth, W A. Clevenger T.E. ( 1 990) Use of Sequential Extraction to Evaluate the Heavy Metals in Mining Wastes. Water, Air and Soil Pollution 50: 24 1 -254. Cochrane R.H.A ( 1 969) Geology of the Tui Mine, Mt. Te Aroha. MSc.Thesis, University of Auckland, New Zealand. Cole J.W. & Lewis K.B. ( 198 1 ) Evolution of the Taupo-Hikurangi Subduction System. Tectonophysics 72: 1 -2 1 . Corey R.B. , Fuj i R. & Hendrickson L.L. ( 198 1 ) Bio-A vailability of Heavy Metals in Soil-Sludge Systems. In: Proceedings of the Fourth Annual Conference of Applied Research on Practical Industrial Waste, Madison, Wisconsin 28-38 September. University of Wisconsin. pp 449-465. Costigan P.A, Bradshaw AD. & Gemmell R.P. ( 1 98 1 ) The Reclamation of Acidic Colliery Spoil; I Acid Production Potential. Journal of Applied Ecology 1 8 : 865- 878. Costigan P.A, Bradshaw AD. & Gemmell R.P. ( 1 982) The Reclamation of Acidic Colliery Spoil; III Problems Associated with the Use of High Rates of Limestone. Journal of Applied Ecology 1 9 : 1 93-20 1 . Costigan P.A, Bradshaw A.D. & Gemmell R.P. ( 1 984) The Reclamation of Acidic Colliery Spoil; IV The Effects of Limestone Particle Size and Depth of Incorporation. Journal of Applied Ecology 2 1 : 377-385. Cotton F.A & Wilkinson G. (1 972) Advanced Inorganic Chemistry. Interscience Publishers, New York. Courtney S.F., King P. & Rodgers K.A. (1 990) A Check List of Minerals from the Tui Mine, Te Aroha, New Zealand. New Zealand Natural Sciences 1 7 : 95-98. Crampton A.T. ( 1993) The Macraes Gold Mine. New Zealand Mining 9: 1 3- 1 7 . Crampton AT. (1 996) Remedial Measures at Golden Cross Following Ground Movement. New Zealand Mining 19: 7-8. Cravotta C.A, Brady K.B.C., Smith M.W. & Beam R.L. ( 1 990) Effectiveness of the Addition of Alkaline Materials at Surface Coal Mines in Preventing or Abating Acid Mine Drainage: Part 1 Geochemical Considerations. In: Skousen J., Sencindiver J. , Samuel D. (eds) Proceedings of the Mining and Reclamation Conference and Exhibition, voi r. West Virginia University Publications Service, Morgantown, USA 4 1 8 Craze B . ( 1 979) Rehabilitation of Tailings Dumps at Captains Flat, New South Wales. In: Hore-Lacey 1. (ed) Mining Rehabilitation. Australian Mining Industry Council, Canberra. Craze B . ( 1 977) Restoration of Captains Flat Mining Area. Journal o/Soil Conservation Service o/New South Wales 33 : 98- 105. Creagh C. ( 1 992) Preventing Acid Spills from Farmland. ECOS - CSIRO Quarterly Science Magazine 70. Crites RW. ( 1 984) Land Use of Waste Water and Sludge. Environmental Science and Technology 1 8(5) : 1 40A- 147A. Dacey P.W. & Colbourne P. ( 1 979) An Assessment of Methods for the Determination of Iron Pyrites in Coal Mine spoil. Reclamation Review 2: 1 1 3- 1 2 1 . Darmer G . ( 1 973) Grasses and Herbs for Revegetating Phytotoxic Material. In: Hutnik RJ. (ed) Ecology and Reclamation of Devastated Land, vol 2. Gordon & Breach, London. Dave N.K. & Vivyurka AJ. ( 1 994) Water Cover on Acid Generating Uranium Tailings - Laboratory and Field Studies. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Davidovits J. & Davidovics M. ( 1 987) Geopolymer Poly (sialate)lPoly (sialate-siloxo) Mineral Matrices for Composite Materials. In: Matthews F.L., Buskell N.C.R, Hodgkinson J.M., Morton J. (eds) Proceedings of 6th International Conference on Composite Materials combined with the 2nd European Conference on Composite Materials. July 20-24, Imperial College, London. Elsevier, London. Davidson N. ( 1 997) The Alluvial Mining Picture in New Zealand. In: Louthean R. (ed) Minerals Resources of New Zealand, 1 997 edn. Louthean Publishers Pty Ltd. New Zealand Crown Minerals, Perth, W A. Davies B.E. ( 1 995) Lead. In: Alloway B.J. (ed) Heavy Metals in Soils, 2nd edn. Blackie Academic & Professional, London. Davies B.E. & Ballinger RC. ( 1 990) Heavy Metals in Soils in North Somerset, England, with Special Reference to Contamination from Base Metal Mining in the Mendips. Environmental Geochemistry and Health 1 2: 291 -300. Davis M., Ross C.W. & Langer L. ( 1 995) Mine Site Restoration to Native Forest in Westland. New Zealand Mining 17 : 1 1 - 1 7. Dean RB. & Suess M.J. ( 1 985) The Risk to Health of Chemicals in Sewage Sludge Applied to Land. Waste Management Research 3 : 25 1 -278. 4 1 9 Deer W.A., Howie R.A. & Zussman J . ( 1 992) An Introduction to the Rock-Forming Minerals, 2nd edn. Longman Scientific & Technical, Hong Kong. Dent D. ( 1 986) Acid Sulphate Soils- A Baseline for Research and Development. IRLI Publication 39, Netherlands. Dick W.A., Stehouwer R.C., Beeghly J.H., Bigham lM. & Lal R. ( 1 994) Dry Flue Gas Desulfurization Byproducts as Amendments for Reclamation of Acid Mine Spoil. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 3 . United States Department of the Interior, Pittsburgh. Dinauer R.C., Stelly M., Padrutt lM., Nauseef lH., Elliot L.F., Stevenson F.J., et al. (eds) ( 1 977) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin. DOH ( 1 992) Public Health Guidelinesfor the Safe Use of Sewage Effluent and Sewage Sludge on Land. New Zealand Department of Health (DOH), Wellington. Dollhopf DJ. ( 1 992) Deep Lime Incorporation Methods for Neutralisation of Acidic Mine Spoils, vol Report Number RRU 9201 . Reclamation Research Unit, Montana State University, USA, Bozeman, Montana (N). Doubleday G.P. ( 1 972) Development and Management of Soils on Pit Heaps. Landscape Reclamation 2: 25-35 . Downey 1.F. ( 1 935) Gold-Mines of the Hauraki District New Zealand. Government Printers, Wellington. DSIR Grasslands ( 1 989) DSIR Grasslands Range ofCultivars. Department of Scientific and Industrial Research, Wellington. Duck C.W. & Burton A.S. (1 985) The Practicality of Restoring the Silt Loam Soils of the Waimea Plains, Nelson, Following Extraction of Subsurface Gravel. In: Campbell I.B. (ed) Proceedings of the soil dynamics and land use seminar. Blenheim, May 1 985 . New Zealand Society of Soil Science, New Zealand Soil Conservators Association, Lower Hutt. Dudas MJ. & Pawluk S. (1 977) Heavy Metals in Cultivated Soils and in Cereal Crops in Alberta. Canadian Journal of Soil Science 57 : 329-339. Easton-Groves J., van der Westhuizen M., Bristow K.L. , van der Bank D.J. & Mills J.W. ( 1 982) Tracing the Leaching of Acidity from Surfaces of Pyritic Residue Dams During Revegetation. In: Darwin Environmental Workshop, Discussion Papers (September 1 982). Australian Mining Industry Council, Darwin. Edmeades D.C. , Smart C.E. & Wheeler D.M. ( 1983) Aluminium Toxicity in New Zealand Soils: Preliminary Results on the Development of Diagnostic Criteria. New Zealand Journal of Agricultural Research 26: 493-501 . 420 Eger P. & Lapakko K. ( 1 98 1 ) The Leaching and Reclamation of Low Grade Mineralised Stockpiles. In: Proceedings of symposium on Surface Mining, Hydrology, Sedimentology and Reclamation. University of Lexington, Kentucky, USA. Eger P., Wagner J.R., Kassa Z. & Melchert G.D. ( 1 994) Metal Removal in Wetland Treatment Systems. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Ehrlich H.L. ( 1 995) Microbial Leaching of Ores. In: Brooks R.R., Dwm C.E., Hall G.E.M. (eds) Biological Systems in Mineral Exploration and Processing. Ellis Horwood, New York. Elliot H.A., Liberati M.R. & Huang C.P. ( 1 986) Competitive Adsorption of Heavy Metals by Soils. Journal of Environmental Quality 1 5 : 2 14-2 19. Enright N.F. ( 1 984) Aluminium Toxicity Problems on Mine Waste. Journal of Soil Conservation July: 1 08-1 1 5 . Erickson P.M., Ladwig K.J. & Kleinmann R.L.P. ( 1 985) Acid Mine Drainage from Inactive Eastern Coal Operations. Environmental Geochemistry and Health 7(1) : 1 6. Eslick L.J. ( 1 995) Heavy Metal Uptake Patterns in Riparian Vegetation. MSc. Thesis, Boise State University. Evangelou V.P. ( 1 984) Controlling Iron and Manganese in Sediment Ponds. Reclamation New & Views, Vol. 2, No. 1 . University of Kentucky, College of Agriculture. Evangelou V.P. ( 1 994) Potential Microencapsulation of Pyrite by Artificial Inducement of FeP04 Coatings. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. Evans D.R. & Rose A.W. ( 1 995) Experiments on Alkaline Addition to Coal Mine Spoil. In: Hynes T.P., Blanchette M.e. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol I . CANMET, Ottawa, Canada. Evans P. & Skousen J. ( 1 995) Effect of Digestion Method on Neutralisation Potential of Overburden Samples Containing Siderite. In: Skousen J., Ziemkiewicz P.F. (eds) Acid Mine Drainage - Control & Treatment. West Virginia University and the National Land Reclamation Centre, Morgantown, West Virginia. EWM ( 1995) Environment Waikato Memorandum : Tui Mine. 28 June 1 995, File 60 1 1 04A, New Zealand. 42 1 Ferguson K.D. ( 1 986) Static and Kinetic Methods to Predict Acid Mine Drainage. In: Lawrence R.W., Branon R.M.R., Ebner H.G. (eds) Fundamental and Applied Biohydrometallurgy. Elsevier, New York. Finkelman R.B. ( 1 987) A Rapid Procedure for Determining Pyritic Sulfur in Coal. Journal oJCoal Quality April: 50-52. Finkelman R.B. & Giffin D.E. ( 1 986) Hydrogen Peroxide Oxidation: An Improved Method for Rapidly Assessing Acid Generating Potential of Sediments and Sedimentary rocks. Reclamation and Revegetation Research 5 : 521 -534. Fitzgerald R.E. ( 1 98 1 ) The Potential Role of Legwnes in the Rehabilitation of Levelled Gold Dredge Tailings, Taramakau River, New Zealand. In: Proceedings of the New Zealand Grasslands Association 42nd Conference, Grey Mouth 1980, vol 42. New Zealand Grasslands Association Publishers. Forstner U. ( 1 99 1 ) Soil Pollution Phenomena-Mobility of Heavy Metals in Contaminated Soil. In: Bolt G.H., De Boodt M.F., Hayes M.H.B., McBride M.B. (eds) Interactions at the Soil Colloid - Soil Solution Interface. Kluwer Academic Publishers, London. Forstner U. ( 1995) Land Contamination by Metals: Global Scope and Magnitude of Problem. In: Allen H.E., Huang C.P., Bailey G.W., Bowers AR. (eds) Metal Speciation and Contamination of Soil. Lewis Publishers, London. Foy e.D., Chaney R.L. & White M.e. ( 1 978) The Physiology of Metal Toxicity in Plants. Annual Reviews of Plant Physiology 29: 5 1 1 -566. Franks W.A, Persinger M., lob A & Inyangetor P. ( 1 982) Utilisation of Sewage Effluent and Sludge to Reclaim Soil Contaminated by Toxic Fumes from a Zinc Smelter. In: Sopper W.E., Seaker E.M., Bastian R.K. (eds) Land Reclamation and Biomass Production with Municipal Wastewater and Sludge. Pennsylvania University Press, University Park, P A Frazer W.l. & Keoghan 1.M. ( 1 990) Legwne Growth on Mine Tailings in the Naseby Forest. In: Gregg P.E.H., Stewart R.B. , Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Fresquez P.R. & Aldon E.F. (1 986) Microbial Reestablishment and the Diversity of Fungal Genera in Reclaimed Coal Mine Spoils and Soils. Reclamation and Revegetation Research 4: 245-258. Fresquez P.R. & Lindemann W.C. ( 1 982) Soil and Rhizosphere Microorganisms in Amended Coal Mine Spoils. Soil Science Society oj America Journal 46: 75 1 - 55. 422 Fricker A.G. ( 1 986) Mining in the New Zealand Environment. DSIR, Industrial Processing Division Report No.IPDIRII1 5 1 7. Wellington. Fuge R , Laidlaw LM.S. , Perkins W.T. & Rogers K.P. ( 199 1 ) The Influence of Acidic Mine and Spoil Drainage on Water Quality in the mid-Wales area. Environmental Geochemistry and Health 1 3 : 70-75. Fyson H. ( 1991) The Tui Comes Home to Roost. Terra Nova September: 4 1 -44. Galvin P. (ed) ( 1 906) The New Zealand Mining Handbook. Government Printers, Wellington. Garga V.K., Smith H.R & Scharer J.M. ( 1 983) Abandonment of Acid Generating Mine Tailings. In: Proceedings of Seventh Panamerican Conference on Soil Mechanics and Foundation Engineering, PAN AM'83. Canadian Geotechnical Society, Engineering Institute of Canada, Montreal. pp 6 16-626. Geidel G. (1 979) Alkaline and Acid Production Potentials of Overburden Material: the Rate of Release. Reclamation Review 2 : 1 0 1 - 1 07. Geidel G. & Caruccio F.T. ( 1 984) A Field Evaluation of the Selective Placement of Acidic Material within the Backfill of a Reclaimed Coal Mine. In: Graves D. (ed) Proceedings of National Symposium on Surface Mining Hydrology, Sedimentology and Reclamation. Lexington, University of Kentucky, December 2-7. OES Publications, Lexington, Kentucky. Gemmell R.P. ( 1 977) Colonisation o/Industrial Wasteland. Edward Arnold, London. Gemmell RP. ( 1981 ) The Reclamation of Acidic Colliery Spoil; II The Use of Lime Wastes. Journal of Applied Ecology 1 8 : 879-887. Georgopoulou Z.1., Fytas K., Soto H. & Evangelou B. ( 1 995) Pyrrhotite Coating to Prevent Oxidation. In: Hynes T.P., Blanchette M.C (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol 1 . CANMET, Ottawa, Canada. Gibbs H.S. ( 1 980) New Zealand Soils. An Introduction. Oxford University Press, Wellington. Gilmour C.M., Broadbent F.E. & Beck S.M. ( 1 977) Recycling of Carbon and Nitrogen through Land Disposal of Various Wastes. In: Dinauer RC.M.E., Stelly M., Padrutt J.M., Nauseef J.H., Elliot L.F., Stevenson F.J., Frink C.R, Hill RR, Hortenstine CC., Kilmer V.1., Monke E.1., Viets F.G.1. (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin. Godden RG. & Thomerson D.R ( 1980) Generation of Covalent Hydrides in Atomic Absorption Spectrometry. The Analyst 1 05 : 1 1 37- 1 1 56. 423 Goenadi D.H. & Tan K.H. ( 1 99 1 ) The Weathering of Paracrystalline Clays into Kaolinite in Andisols and Ultisols in Indonesian. Indonesian Journal o/Tropical Agriculture 2: 56-65. Gough L.P. , McNeal J.M. & Severson Re. ( 1 980) Predicting Native Plant Copper, Iron, Manganese, and Zinc Levels using DTP A and EDT A Soil Extractants, Northern Great Plains. Soil Science 44: 1 030- 1 036. Gow A.J. ( 1 987) Petrographic Studies of Ironsands and Associated Sediments. New Zealand Journal o/Geology and Geophysics 1 0 : 675-696. Gregg P.E.H. ( 1 983) Report to Waipipi Ironsands Ltd., on a Subtractive Soil Test of Restored Land (unpublished). Massey University, Palmerston North, New Zealand (N). Gregg P.E.H. & Stewart R.B. ( 199 1 ) Land Reclamation at Waihi - Principles. In: New Zealand Branch of the Australasian Institute of Mining and Metallurgy, Proceedings of 25th Annual Conference, 1 99 1 . Gregg P.E.H.,Stewart RB. & Stewart G. (1 990) Determination of a Suitable Plant Growth Medium on Gold and Silver Mining Waste, Martha Hill, Coromandel Peninsula -1. Agronomic and Soil Chemical Aspects. In: Gregg P.E.H., Stewart R.B. , Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Gregg P.E.H., Stewart RB. & Mason K.A. ( 1 995) Pastoral Reclamation of Tailings and Overburden Material from a New Zealand Open Cast Gold Mine. In: Hynes T.P., Blanchette M.e. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1995, vol 3 . CANMET, Ottawa, Canada. Gregg P.E.H., Stewart R.B., Denize P.R, Drury RG. & Keating R.D. (in press) The Practice of Tailings Revegetation in New Zealand. In: Bengsen S .A. (ed) Principles and practices of tailings revegetation. American Society for Surface Mining and Reclamation. Gregg R. ( 1 997) The Macraes Flat Mine 1 997 Update. New Zealand Mining 2 1 : 1 0- 1 5 . Gurung S.R., Stewart R.B., Loganathan P . & Gregg P.E.H. ( 1 996) Aluminium-Organic Matter-Fluoride Interactions During Soil Development in Oxidised Mine Waste. Soil Technology 9: 273-279. Halderson J.L. & Zenz D.R. ( 1 978) Use of Municipal Sewage Sludge in Reclamation of Soils. In: Schaller F.W., Sutton P. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy. Crop Science Society of America. Soil Science Society of America, Madison. Hansen P.V. ( 1 989) Mining in the Hauraki district. New Zealand Association o/Soil and Water Conservation Broadsheet Easter, Part II: 6-7. 424 Haq A.U., Bates T.E. & Soon YK. ( 1 980) Comparison of Extractants for Plant­ Available Zinc, Cadmium, Nickel, and Copper in Contaminated Soils. Soil Science 44: 772-777. Harries 1.R & Ritchie A.I.M. ( 1 987) The Effect of Rehabilitation on the Rate of Oxidation of Pyrite in a Mine Waste Rock Dump. Environmental Geochemistry and Health 9: 27-36. Harries 1.R & Ritchie A.I.M. ( 1 988) Rehabilitation Measures at the Rum Jungle Mine Site. In: Salomons W., Forstner U. (eds) Environmental Management of Solid Waste - Dredged Material and Mine Tailings. Springer-Verlag, New York. Harris C. ( 1 997) Plunder Islands. New Internationalist 29 1 : 1 4- 1 5 . Haynes RJ. & Swift RS. ( 1 985a) Effects of Liming on the Extractability of Fe, Mn, Zn and Cu from a Peat Medium and the Growth and Micronutrient Uptake of Highbush Blueberry Plants. Plant and Soil 84: 2 1 3-223 . Haynes RJ. & Swift RS. ( 1 985b) Effects of Soil Acidification on the Chemical Extractability of Fe, Mn, Zn and Cu and the Growth and Micronutrient Uptake of Highbush Blueberry Plants. Plant and Soil 84: 20 1 -2 12. Hearle 1.W.S. , Sparrow 1.T. & Cross P.M. ( 1 972) The Use of Scanning Electron Microscope. Pergamon Press, Oxford. Hedin RS. & Watzlaf G.R ( 1 994) The Effects of Anoxic Limestone Drains on Mine Water Chemistry. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Hedin RS., Narin R.W. & Kleinmann R.L.P. ( 1 994) Passive Treatment of Coal Mine Drainage. Bureau of Mines Information Circular, Ie 9389. United States Department of the Interior, Pittsburgh. Hellier W.W., Giovannitti E.F. & Slack P.T. ( 1 994) Best Professional Judgment Analysis for Constructed Wetlands as a Best Available Technology for the Treatment of Post-Mining Groundwater Seeps. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Helz G.R, Dai J.H., Kijak P.J . & Fendinger N.J. ( 1 987) Processes Controlling the Composition of Acid Sulfate Solution Evolved from Coal. Applied Geochemistry 2: 427-436. Henderson 1 . & Bartrum 1.A. ( 1 9 1 3) The Geology of the Aroha subdivision, Hauraki, Auckland. In: New Zealand Geological Survey Bulletin 1 6. , Wellington (Government Printers; N). 425 Hendy C.H. ( 1 98 1 ) The Tui Mine- After the Miners Have Left. New Zealand Environment 29: 1 7- 19. Hetrick B.A.D., Wilson G.W.T. & Figge D.A.H. ( 1 994) The Influence of Mycorrhizal Symbiosis and Fertiliser Amendments on Establishment of Vegetation in Heavy Metal Mine Spoil. Environmental Pollution 86: 1 7 1 - 1 79. Hewitt A.E. ( 1 993) New Zealand Soil Classification. Landcare Research Science Series 1 . Manaaki Whenua - Landcare Research Ltd., Dunedin, New Zealand. Hillel D. ( 1 992) Out of the Earth: Civilization and the Life of the Soil. University of California Press, Los Angeles. Hinkle K.R. ( 1 982) Use of Municipal Sludge in the Reclamation of Abandoned Pyrite Mines in Virginia. In: Sopper W.E., Seaker E.M., Bastian R.K. (eds) Land Reclamation and Biomass Production with Municipal Wastewater and Sludge. Pennsylvania State University Press, University Park, P A. Hochstein M.P. & Nixon LM. ( 1 979) Geophysical Study of the Hauraki Depression, North Island, New Zealand. New Zealand Journal of Geology and Geophysics 22 : 1 - 1 9. Hodder A.P.W. & Scadden P.G. ( 1 98 1 ) Waihi Goldfields - Te Aroha Mineralisation. In: Briggs R.M. (ed) Hamilton Conference - Field Trip Guides. Geological Society of New Zealand Inc., Hamilton. Home D.J., Stewart R.B. & Gregg P.E.H. ( 1 990) Detennination of a Suitable Plant Growth Medium on Gold and Silver Mining Waste, Martha Hill, Coromandel Peninsula - II Physical Properties. In: Gregg P.E.H., Stewart R.B. , Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Hoyt D.B. & Nyborg M. ( 1 972) Use of Dilute Calcium Chloride for the Extraction of Plant Available Aluminium and Manganese from Acid Soil. Canadian Journal of Soil Science 52: 1 63-167 . Huang X. & Evangelou V.P. ( 1 992) Abatement of Acid Mine Drainage by Encapsulation of Acid Producing Geological Materials. United States Bureau of Mines Contract J0309013 . 60 pages. Humber A.J. & Eng P. (1 995) Separation of Sulphide Minerals from Mill Tailings. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol I . CANMET, Ottawa, Canada. 426 Hume L.J., Of so ski N.J. & Reynolds J. ( 1988) Influence of pH, Exchangeable Aluminium and 0.02M CaCh-Extractable Aluminium on the Growth and Nitrogen-Fixing Activity of White Clover (Trifolium repens) in Some New Zealand Soils. Plant and Soil 1 1 1 : 1 1 1 - 1 1 9. Hunter G.D. & Whiteman P.C. ( 1 975) Revegetation of Mine Wastes at Mount Isa, Queensland. 2. Amendment of Nutrient Status and Physical Properties of Tailing for Plant Growth. Australian Journal of Experimental Agriculture and Animal Husbandry 1 5 : 803-8 1 1 . IALR (1 993) Mining vs. Nature. lALR Newsletter 3 : 6. Inskeep W.P., Clayton J.L. & Mogk D.W. ( 1 993) Naturally Weathered Plagioclase Grains from the Idaho Batholith: Observations U sing Scanning Electron Microscopy. Soil Science 57: 85 1 -860. Jackson M.L. ( 1 958) Soil Chemical Analysis. Prentice-Hall, London. James A.R. & Eng P.R. (c. l 992) A Comparison of the Relative Importance of different Mechanisms of Oxygen Entry into Acid Generating Mine Waste Material: 1 - 1 1 Source Unknown. John M.K. ( 1 974) Extractable and Plant-Available Zinc in Horizons of Several Fraser River Alluvial Soils. Canadian Journal o/Soil Science 54: 1 25 - 132. Johnson C.M. & Nishita H. ( 1 952) Microestimation of Sulfur in Plant Material, Soil and Irrigation Waters. Analytical Chemistry 24: 736-742. Johnson M.S., McNeilly T. & Putwain P.D. ( 1 977) Revegetation of Metalliferous Mine Spoil Contaminated by Lead and Zinc. Environmental Pollution 12 : 261 -277. Johnson, WR & Proctor, J. ( 1 977) Short Communication - Metal Concentrations in Plants and Soils from Two British Serpentine Sites. Plant and Soil 46: 275-278. Johnson M.S., Cooke J.A. & Stevenson J.K.W. ( 1 994) Revegetation of Metalliferous Wastes and Land After Metal Mining. In: Hester R.E., Harrison R.M. (eds) Issues in Environmental Science & Technology, vol 1 . Royal Society of Chemistry, London. Jones C.E. & Wong J.Y. ( 1 994) Shotcrete as a Cementitious Cover for Acid Generating Tailings. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Joost R.E., Jones J.H. & Olsen FJ. ( 1 98 1 ) Physical and Chemical Properties of Coal Refuse as Affected by Deep Incorporation of Sewage Sludge and/or Limestone. In: Proceedings of Symposium on Surface Mining Hydrology, Sedimentology and Reclamation. pp 307-3 12. University of Kentucky, Lexington. 427 Joost RE., Olsen FJ. & Jones lH. (1 987) Revegetation and Minespoil Development of Coal Refuse Amended with Sewage Sludge and Limestone. Journal of Environmental Quality 16 : 65-68. Jopony M. & Young S . ( 1 993) Assessment of Lead Availability in Soils Contaminated by Mine Spoil. Plant and Soil 1 5 1 : 273-278. Joshi D.C. , Dhir RP. & Gupta B.S. ( 1983) Influence of Soil Parameters on DTPA Extractable Micronutrients in Arid Soils. Plant and Soil 72 : 3 1 -38 . Jozefaciuk C., Jozefaciuk A. & Pochy1czuk B. ( 1 990) Influence of Different Fertilising Substances on Regeneration of Degraded Loessial Soil. Pamientnik-Pulawsld 95 : 53-60. Kabata-Pendias A. & Pendias H. (1 984) Trace Elements in Soils and Plants. CRC Press, Inc., Boca Raton, Florida. Kanehiro Y. & Sherman G.D. (1 967) Distribution of Total and 0. 1 Normal Hydrochloric Acid-Extractable Zinc in Hawaiian Soil Profiles. Soil Science Society of America Proceedings 3 1 : 394-399. Kapland D.1. & Estes G.O. ( 1 985) Organic Matter Relationship to Soil Nutrient Status and Aluminium Toxicity in Alfafa. Journal of Agronomy 77 : 735-738. Karathanasis A.D. ( 1 99 1 ) Phosphate Mineralogy and Equilibria in Two Kentucky Alfisols Derived from Ordovician Limestones. Soil Science 55 : 1 774- 1 782. Keating R.D. & Buckley P .K. (198 1 ) Land Restoration on the Taramakau River. In: Proceedings of Annual Conference of the New Zealand Branch of Australian Institute of Mining and Metallurgy, Thames, 1 98 1 . Keeney D.R. & Wildung R.E. ( 1 977) Chemical Properties of Soils. In: Dinauer R.C.M.E., Stelly M., Padrutt lM., Nauseef lH., Elliot L.F., Stevenson F.l, Frink C.R. , Hill RR, Hortenstine C.C., Kilmer V.l, Monke E.J., Viets F.G.l (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin (N). Kelepertsis A.E. & Bibou A. ( 199 1 ) Heavy Metal Contamination of Soils at Old Mining Sites on Thasos Island, Greece. Environmental Geochemistry and Health 1 3 : 23- 28. Keller G. ( 1993) Radiological Aspects of Former Mining Activities in the Saxon Erzgebirge, Germany. Environment International 19 : 449-454. Keller L. & Murr L.E. ( 1 982) Acid-Bacterial and Ferric Sulfate Leaching of Pyrite Single Crystals. Biotechnology and Bioengineering XXIV: 83-96. 428 Kepler D.A. & McCleary E.C. ( 1 994) Successive Alkalinity-Producing Systems (SAPS) for the Treatment of Acidic Mine Drainage. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Kleinmann R.L.P. ( 1 99 1 ) Acid Mine Drainage: An Overview. In: Hobbs B.F. (ed) Proceedings of Energy in the 90s. March 1 0- 1 3 Pittsburgh, Pennsylvania. American Society of Civil Engineers, New York. Kleinmann R.L., Crerar D.A. & Pacelli R.R. ( 1 980) Biogeochemistry of Acid Mine Drainage and a Method to Control Acid Fonnation. Mining Engineering 33 : 300-306. Knabe W. ( 1 964) Methods and Results of Strip-Mine Reclamation in Gennany. Ohio Journal o/Science 64 (2) : 75- 1 05. Kucha H., Martens A., Ottenburgs R. , De Vos W. & Viaene W. ( 1 996) Primary Minerals of Zn-Pb Mining and Metallurgical Dumps and their Environmental Behaviour at Plombieres, Belgium. Environmental Geology 27: 1 - 1 5 . Kwong E.C.M., Scharer lM., Bryerley J.J. & Nicholson R.Y. ( 1 995) Prediction and Control of Bacterial Acitivity in Acid Mine Drainage. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol I . CANMET, Ottawa, Canada. Kwong Y.T.J. ( 1 995) Influence of Galvanic Sulphide Oxidation on Mine Water Chemistry. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 -June 1 , 1 995. Sudbury, Ontario, Canada, vol II. CANMET, Ottawa. Lacey D.T. & Lawson F. ( 1 970) Kinetics of the Liquid-Phase Oxidation of Acid Ferrous Sulfate by the Bacterium Thiobacillus ferrooxidans. Biotechnology and Bioengineering 12 : 29-50. Lagerwerff lV. ( 197 1 ) Uptake of Cadmium, Lead and Zinc by Radish from Soil and Air. Soil Science 1 1 1 (2) : 129- 1 33 . Lagerwerff lV. & Specht A.W. ( 1 970) Contamination of Roadside Soil and Vegetation with Cadmium, Nickel, Lead and Zinc. Environmental Science and Technology 4: 583-586. Land Resources Network Ltd. ( 1 993) OrganiC Materials as Soil Amendments in Reclamation: A Review of Literature. Alberta Conservation and Reclamation Council Report No.RRTAC 93-4, Edmonton, Alberta. 429 Lapakko K.A. ( 1 994) Subaqueous Disposal of Mine Waste: Laboratory Investigation. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage. Bureau of Mines Special Publication SP 06A-94, vol 1 . United States Department of the Interior, Pittsburgh. Lapidus D.F. ( 1 990) Collins Dictionary of Geology. Collins, London and Glasgow. Lapwood P. ( 1 99 1 ) Land Reclamation at Waihi - Practice. In: New Zealand Branch of the Australasian Institute of Mining and Metallurgy, Proceedings of 25th Annual Conference, 199 1 . Lim C.H. & Jackson M.L. ( 1 982) Dissolution for Total Element Analysis. In: Page A.L., Miller R.H., Keeney D.R. (eds) Methods of Soil Analysis. Part 2 - Chemical and Microbiological Properties, 2nd edn. American Society of Agronomy, Madison, Wisconsin. Limpus-O'Reilly J. ( 1 997) An Incubation Study to Assess the Effect of Waste Sludge Additions on Some Chemical Characteristics of Mine Soils. MSc. Thesis, Massey University, Palmerston North, New Zealand. Lin Z. ( 1 997) Mobilization and Retention of Heavy Metals in Mill-Tailings from Garpenberg Sulphide Mines, Sweden The Science of the Total Environment 1 98 : 1 3 -2 1 . Lindsay W.L. & Norvell W.A. ( 1 969) A Micronutrient Soil Test for Zinc, Iron, Manganese and Copper. Agron. Abstr. Journal of the Soil Science Society of America. p.84. Lindsay W.L. & Norvell W.A. (1 978) Development of a DTPA Test for Zinc, Iron, Manganese, and Copper. Soil Science 42: 421 -428 . Livingston M.E. (ed) (1 987) Preliminary Studies on the Effects of Past Mining on the Aquatic Environment, Coromandel Peninsula. Water & Soil Miscellaneous Publication No. 1 04 . Water and Soil Directorate, Ministry of Works, Wellington, New Zealand. Lottermoser B.G. & Morteani G. ( 1 993) Sewage Sludges: Toxic Substances, Fertilizers, or Secondary Metal Resources? Episodes 16( 1&2): 329-333 . Lowson R.T., Evans J.V., Sarbutt J.V. & Sinclair G. ( 1 987) The Rum Jungle Tailings Dam - Chemical Profile of the Subsoil. Australian Nuclear Science and Technology Organisation ANSTO Report E663, Lucas Heights. Luo Y. & Rimmer D.L. ( 1995) Zinc-Copper Interactions Affecting Plant Growth on a Metal-Contaminated Soil. Environmental Pollution 88 : 79-83. 430 Ma Q.Y. & Lindsay W.L. ( 1 993) Measurements of Free Zinc2+ Activity in Uncontaminated and Contaminated Soils using Chelation. Soil Science Society of America Journal S7: 963-967. MacGrath T. ( 1 973) State Holds Key to Fate ofTe Aroha Mine. New Zealand Herald 21 November: section 3, page 3 (N). Mackay A.D., Caradus lR & Wewala S. ( 1 99 1 ) Aluminium Tolerance of Forage Species. In: Wright R.J. et al. (ed) Plant-Soil Interactions at Low PH. Kluwer Academic Publishers, Netherlands. MacLean A.J. & Dekker AJ. ( 1 976) Lime Requirement and Availability of Nutrients and Toxic Metals to Plants Grown in Acid Mine Spoils. Canadian Journal of Soil Science 56: 27-36. MacLean D.C., Hansen K.S. & Schneider RE. ( 1 992) Amelioration of Aluminium Toxicity in Wheat by Fluoride. New Phytology 1 2 1 : 8 1 -88 . Main lV. ( 1 979) Precious Metal Bearing Veins of the Maratoto-Wentworth Area, Hauraki Goldfield, New Zealand. New Zealand Journal of Geology and Geophysics 22 ( 1 ) : 4 1 -S 1 . Mare B. ( 1 997) Golden Cross Mine, Waihi - Mining Operations and Recent Events. New Zealand Mining 2 1 : 1 9-20. Mason K.A. ( 1 996) Rehabilitation of Unoxidised Pyritic Waste Rock and Tailings at Martha Hill Gold Mine, N.Z. Master of Horticultural Science Thesis, Massey University, Palmerston North, New Zealand. Mason K.A., Gregg P.E.H. & Stewart RB. ( 1 993) Soil Formation on Tailings, Waihi. In: The Changing World of Mining. The Australasian Institute of Mining and Metallurgy (AusIMM). New Zealand Branch Annual Conference, Wellington 24 to 27 August 1 993, Wellington. Mason K.A., Gregg P.E.H. & Stewart R.B. ( 1 99S) Land Reclamation Trials and Practices at Martha Hill Gold Mine, Waihi, New Zealand. In: Mauk J.L., St­ George J.D. (eds) Proceedings of the 1 995 PACRIM Congress. 1 9-22 November 1 995 . The Australasian Institute of Mining and Metallurgy, Carlton, Australia. McCalla T.M., Peterson J.R & Lue-Hing C. ( 1 977) Properties of Waste. In: Dinauer RC.M.E., Stelly M., Padrutt J.M., Nauseef J.H., Elliot L.F., Stevenson F.J., Frink C.R, Hill RR, Hortenstine C.c., Kilmer V.J., Monke EJ., Viets F.GJ. (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin. McLaren ' RG. & Cameron K.C. ( 1 996) Soil Science. Sustainable production and environmental protection, 2nd edn. Oxford University Press, Auckland. 43 1 McLauchlan G. (ed) ( 1 989) The Illustrated Encyclopedia of New Zealand. Windsor Watt, Auckland. McLean E.D. ( 1 973) Testing Soils for pH and Lime Requirement. In: Walsh L.M., Beaton l.D. (eds) Soil Testing and Plant Analysis. Soil Science Society of America, Madison, Wisconsin. McLean E.O. ( 1 965) Aluminium. In: Black C.A. et al. (ed) Methods of Soil Analysis. American Society of Agronomy, Wisconsin, USA. McOnie A. ( 1 987) Tui Mine prospect ML 's 32-184 and 32-185: Progress Report to 31 July 1987, Including Results of Diamond Drilling Programme. Gold Resources Ltd. Unpublished open-file mining company report, Institute of Geological and Nuclear Sciences Limited, MR 1 029. Wellington, New Zealand. McQueen D.l. ( 1 982) Land Reclamation - Physical Restoration of Soil after Mining. A selected bibliography. New Zealand Soil Bureau Bibliographic Report 29. Department of Scientific and Industrial Research, New Zealand. McQueen D.l. ( 1 983) Land Reclamation After Gravel Extraction on Ranzau Soils, Nelson, New Zealand. New Zealand Soil Bureau Scientific Report 58. McQueen D.l. & Ross C.W. ( 1 982) Effects of Stockpiling Topsoils Associated with Opencast Mining 1 . Physical Properties. New Zealand Journal of Science 25 : 295-302. Mehlich A. ( 1 984) Mehlich 3 Soil Test Extractant: A Modification of Mehlich 2 Extractant. Communications in Soil Science and Plant Analysis 1 5 : 1409- 14 16. Mengel K. & Kirkby E.A. ( 1 987) Principles of Plant Nutrition, 4th edn. International Potash Institute, Bern, Switzerland. Metcalfe L. & Godfrey W. ( 1 990) Land Disturbed by Alluvial Mining in Westland. In: Gregg P.E.H., Stewart R.B. , Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Metzger L. & Yaron B. ( 1 987) Influence of Sludge Organic Matter on Soil Physical Properties. Advances in Soil Science 7 : 1 4 1 - 1 63 . Mew G. & Ross C.W. ( 1 99 1 ) Beech Forests After Mining. Terra Nova 4: 53-53. Mew G. & Ross C.W. ( 1 992a) From Beech to Coal and Back. New Zealand Forestry August: 1 8- 1 9. Mew G. & Ross C.W. ( 1 992b) Minding the Miners. Terra Nova 1 5 : 43-45. Middleton K.R. ( 1 973) Nutrient Solution. Plant and Soil 38: 2 1 9. 432 Miller S.D. (1 979) Chemistry of a Pyritic Coal Mine Spoil. Ph.D. Dissertation, Yale University, New Haven. (N). Miller S.D. & Murray G.S. ( 1 988) Application of Acid-base Analysis to Wastes from Base-Metal and Precious Metal Mines. In: ASSMRlUSDI (ed) Mine Drainage and Surface Mine Reclamatiom: Mine Water and Mine Waste. Conference Proceedings. Information Circular 9 1 83 , vol I . ASSMRJUSDI, Pittsburgh. Miller S.D., Jeffery J.J. & Donohue T.A ( 1 994) Developments in Predicting and Management of Acid Forming Mine Wastes in Australia and Southeast Asia. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Miller S.K. ( 1 992) Mines with a Sting in their Tailings. New Scientist 1 35, No. 1 830: 5 . Mineral Resources ( 1 997) Mineral Resources of New Zealand. Louthean Publishers Pty Ltd. New Zealand Crown Minerals, Perth, W A. Ministry of Commerce ( 1 995) Draft Minerals Program for Metallic and Non-Metallic Minerals, Coal and Industrial Rocks and Building Stones - Discussion Document. August 1 995. Energy and Resources Division, Wellington. Ministry of Commerce ( 1 996) New Zealand Annual Mining Review 1 995. Published by the Resource Information Unit of the Ministry of Commerce. Moffat AJ. , Bending N.AD. & Roberts C.J . ( 1 99 1 ) The Use of Sewage Sludge as a Fertiliser in the Afforestation of Opencast Coal Spoils in South Wales. In: Davies M.C.R. (ed) Land Reclamation. An End to Dereliction? Elsevier Applied Science, London. Molloy L. (1 988) The Living Mantle. Soils in the New Zealand Landscape. Mallinson Rendel, Wellington. Moore P., Ritchie N. & Homer L. (1 996) Coromandel Gold - A Guide to the Historic Goldfields of Coromandel Peninsula. The Dunmore Press Ltd. , Palmerston North, New Zealand. Morrell W.J., Gregg P.E.H., Stewart R.H., Bolan N. & Home D. ( 1 995) Potential for Revegetating Base-Metal Tailings at the Tui Mine Site, Te Aroha, New Zealand. In: Maule J.L., St-George J.D. (eds) Proceedings of the 1 995 PAC RIM Congress . 1 9-22 November 1 995. The Australasian Institute of Mining and Metallurgy, Carlton, Australia. Morton H.D. & Harrison R.W. (1 990) Restoration of Alluvial Soils for Pasture Production on the West Coast, South Island. In: Gregg P.E.H., Stewart R.H. , Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. 433 Mortvedt J.J., Giordano P.M. & Lindsay W.L. (eds) ( 1 972) Micronutrients in Agriculture. Soil Science Society of America, Madison, Wisconsin. Murdock DJ., Fox J.R.W. & Bensley J.G. ( 1 995) Treatment of Acid Mine Drainage by the High Density Sludge Process. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol 2. CANMET, Ottawa, Canada. Mustin C. , Donato P.de, Berthelin J. & De Donato P. ( 1 992) Quantification of the Intragranular Porosity Formed in Bioleaching of Pyrite by Thiobacillus ferrooxidans. Biotechnology and Bioengineering 39: 1 12 1 - 1 127. Natarajan K.A. ( 1 990) Electrochemical Aspects of Bioleaching of Base-Metal Sulfides. In: Ehrlich H.L., Brierley C.L. (eds) Microbial Mineral Recovery. McGraw-Hill Publishing Company, New York. Nawrot J.R. , Conely P .S . & Sandusky J.E. ( 1 994) Concentrated Alkaline Recharge Pools for Acid Seep Abatement: Principles, Design, Construction and Performance. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Nelson J.L., Boawn L.e. & Viets F.G. ( 1959) A Method for Assessing Zinc Status of Soils Using Acid-Extractable Zinc and Titratable Alkalinity Values. Soil Science 88 : 275-283 . Neuman D.R., Schrack J .L . & Gough L.P . ( 1 987) Copper and Molybdenum. In: Williams R.D. , Schuman G.E. (eds) Reclaiming Mine Soils and Overburden in the Western United States: Analytical Parameters and Procedures. Soil Conservation Society of America, Iowa, USA. New Zealand Herald ( 1 98 1 ) Newspaper Article. February 20. Nicks L.J. & Chambers M.F. ( 1 995) Farming for Metals? Mining Environmental Management 3(3): 1 5- 1 8 . Nielson R.F. & Peterson H.B. ( 1 978) Vegetating Mine Tailings Ponds. In: Schaller F.W., Sutton P . , et al. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, USA. Nordstrom D.K. ( 1 982) Aqueous Pyrite Oxidation and the Consequent Formation of Secondary Iron Minerals. In Kral D. (ed) Acid Sulphate Weathering: SSSA Special Publication Number 1 0. SSSA, Wisconsin. 434 Norland M.R, Veith D.L. & Dewar S.W. (1 993) Standing Crop Biomass and Cover on Ammended Coarse Taconite Iron Ore Tailings. In: Zamora B.A., Connolly RE. (eds) The Challenge of Integrating Diverse Perspectives in Reclamation. Proceeding of the 1 0th Annual National Meeting of the American Society for Surface Mining and Reclamation ASSMRI 0-93 , vol 1 . ASSMR, Spokane. Norpac Mining Ltd. (undated )Tui Base Metal Mine & Mill. Publicity Pamphlet (c. 1 970). Norton D.A. ( 1991) Restoration of Indigenous Vegetation on Sites Disturbed by Alluvial Gold Mining in Westland. Resource Allocation Report 3 . Resource Information Section, Energy and Resources Division, Ministry of Commerce, Wellington, New Zealand. Norvell W.A. ( 1 984) Comparison of Chelating Agents as Extractants for Metals in Diverse Soil Materials. Soil Science 48 : 1 285- 1292. Nriagu J.O. (ed) (1 984) Changing Metal Cycles and Human Health. Springer-Verlag, Berlin. NZMS ( 1 983) Summaries of Climatological Observations to 1980. New Zealand Meteorological Service, Wellington. O'Donoghue M. (1 976) The Colour Dictionary of Gemstones & Minerals. Orbis Publishing Limited, London. Okereke A. & Stevens E. ( 1 99 1 ) Kinetics of Iron Oxidation by Thiobacillus ferrooxidans. Applied and Environmental Microbiology 57(4): 1 052- 1 056. Olsen S .R & Barber S.A. (1 977) Effect of Waste Application on Soil Phosphorous and Potassium. In: Dinauer RC.M.E., Stelly M., Padrutt J.M., Nauseef J.H., Elliot L.F., Stevenson FJ., Frink C.R, Hill RR., Hortenstine C.C., Kilmer VJ., MOnke EJ., Viets F.G.J. (eds) Soils for Management of Organic Wastes and Waste Waters. Soil Science Society of America, American Society of Agronomy, Crop Science Society of America, Madison, Wisconsin (N). Olsen S .R, Cole C.V., Watanabe F.S. & Dean L.A. ( 1 954) Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate. USDA Department Circular 939. Orsman B. ( 1 991) Evils of Past Kill Mining Plan. New Zealand Herald 1 6 July (N). O'Shay T. , Hossner L.R & Dixon J.B. (1 990) A Modified Hydrogen Peroxide Method for the Determination of Potential Acidity in Pyritic Overburden. Journal of Environmental Quality 1 9: 778-782. Pagliai M., Guidi S. , La Marca M., Giachetti G. & Lucamante G. ( 198 1 ) Effects of Sewage Sludges and Composts on Soil Porosity and Aggregation. Journal of Environmental Quality 1 0(4) : 556-56 1 . 435 Palmer J.P. ( 1 99 1 ) Techniques for Reclaiming Metalliferous Tailings. In: Davies M.C.R. (ed) Land Reclamation. An End to Dereliction? Elsevier Applied Science, London. Palmer K.A. ( 1 982) An Outline of Mining Law. Auckland University Law Facility, Seminar Series. Auckland, New Zealand. Pang L. (1 992) A Hydrogeological Study of Heavy Metal Contamination in the Groundwater of the Te Aroha Area. MSc. Thesis, University of Waikato, Hamilton. Pang L. (1 995) Contamination of Groundwater in the Te Aroha Area by Heavy Metals from an Abandoned Mine. New Zealand Journal of Hydrology 33( 1 ) : 1 7-34. Parisi D., Horneman J. & Rastogi V. ( 1 994) Use of Bactericides to Control Acid Mine Drainage from Surface Operations. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. Pedersen T.F., McNee J.l., Mueller B., Flather D.H. & Pelletier C.A. ( 1 994) Geochemistry of Submerged Tailings in Anderson Lake, Manitoba: Recent Results. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Peech M. ( 1 965) Hydrogen-Ion Activity. In: Black C.A. (ed) Methods of soil analysis. Part 2. Agronomy 9. American Society of Agronomy, Madison, Wisconsin. Penny S.F. ( 1 983) A Preliminary Investigation into the Use of Biological Communities as Indicators of Water Quality, with Particular Reference to the Effects of Mining on Coromandel Streams. Unpublished Report. Water and Soil Directorate. Ministry of Works and Development, Wellington. Perez A. & Calvo de Anta R. ( 1 992) Soil Pollution in Copper Sulphide Mining Areas in Galicia (NW Spain). Soil Technology 5 : 27 1 -28 1 . Pichtel 1.R., Dick W.A. & Sutton P. (1 994) Comparison of Amendments and Management Practices for Long-Term Reclamation of Abandoned Mine Lands. Journal of Environmental Quality 23 : 766-772. Pierce W.G., Belzile N., Wiseman M.E. & Winterhalder K. ( 1 994) Composted Organic Wastes as Anaerobic Reducing Covers for Long term Abandonment of Acid­ Generating Tailings. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. 436 Pierzynski G.M., Sims J.T. & Vance G.F. ( 1 994) Soils and Environmental Quality. Lewis Publishers, London. Pollock K.M. ( 1 986) Plant Materials Handbook/or Soil Conservation. Water and Soil Miscellaneous Publication No. 95, vol 3 . National Water and Soil Conservation Authority, Wellington, NZ. Pugh P.E., Hossner L.R & Dixon J.B. ( 1 984) Oxidation Rate of Iron Disulphides as Affected by Surface Area, Morphology, Oxygen Concentration and Autotrophic Bacteria. Soil Science 137(5) : 309. Pulford 1.D. ( 1 99 1 ) A Review of Methods to Control Acid Generation in Pyritic Coal Mine Waste. In: Davies M.C.R (ed) Land Reclamation. An End to Dereliction? Elsevier Applied Science, London. Pulford 1.D. & Duncan H.J. ( 1 978) A Qualitative Assessment of Potential Acidity in Coal Mine Waste. Reclamation Review 1 : 5 1 -54. Raea S. ( 1 997) Dam on Shaky Ground. New Zealand Herald Wednesday Magasine, April 9 :Gl . Raicevic D. ( 1 979) Decontamination of Elliot Lake Uranium tailings. CIM Bulletin 72(808): 1 09- 1 1 5 . Reeder J.D. & Sabey B . ( 1 987) Nitrogen. In: Williams R.D., Schuman G.E. (eds) Reclaiming Mine Soils and Overburden in the Western United States: Analytical Parameters and Procedures. Soil Conservation Society of America, Iowa. Reeves RD., Baker A.J.M. & Brooks RR ( 1 995) Abnormal Accumulation of Trace Metals by Plants. Mining Environmental Management 3(3): 4-8. Ribet 1. , Ptacek e.J., Blowes D.W. & Jambor J.L. ( 1 995) The Potential for Metal Release by Reductive Dissolution of Weather Mine Tailings. Jow:nal of Contaminant Hydrology 17 : 239-273 . Rich D.H. & Hutchison K.R. ( 1 994) Coal Refuse Disposal Using Engineering Design and Lime Chemistry. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Ritcey G.M. ( 1 989) Tailings Management- Problems and Solutions in the Mining Industry, 1 st edn, vol 6. Elsevier, Amsterdam. Riveros P .A. ( 1 995) Applications of Ion Exchangers to the Treatment of Acid Mine Drainage. In: Hynes T.P., Blanchette M.e. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol 2. CANMET, Ottawa, Canada. 437 RMA ( 1991) New Zealand Resource Management Act. Wellington, New Zealand. Robb G. & Robinson J. ( 1 995) Acid Mine Drainage Prediction and Remediation. Mining Environmental Management 3 (3) : 1 9-2 1 . Robertson A.L ( 1 985) The Poisoning of Roots Zea Mays by Nickel Ions and the Protection Afforded by Magnesium and Calcium. New Phytology 1 00: 1 73-1 89. Robertson LB. & Eggleton R.A. ( 1991 ) Weathering of Granitic Muscovite to Kaolinite and Halloysite and of Plagioclase-Derived Kaolinite to Halloysite. Clays and Clay Minerals 39: 1 1 3- 1 26. Robinson B.W. ( 1 974) The Origin of Mineralisation at the Tui Mine, Te Aroha, New Zealand in Light of Stable Isotope Studies. Economic Geology 69: 9 1 0-925 . Robinson B.W. & Christie A.B. ( 1 980) Epithermal Silver-Gold Mineralisation, Maratoto Mine, New Zealand: Stable Isotopes and Fluid Inclusions. In: Ridge J.W. (ed) Proceedings of the Fifth Quadrennial IAGOD Symposium, vol 1 . Nagele und Obermiller, Stuttgart. Rodriguez-Leiva M. & Tributsch H. ( 1 988) Morphology of Bacterial Leaching Patterns by Thiobacillusferrooxidans on Synthetic Pyrite. Archives of Microbiology 149: 401 -405 . Ross C.W. ( 1 988) West Coast Alluvial Gold Mining Rehabilitation Study. Report No.3 Guidelines for Land Rehabilitation. DSIR, Division of Land & Soil Sciences Report to the Ministry of the Environment. 80 p. Ross C.W. & Mew G. ( 1 990) Land Restoration after Mining on the West Coast, South Island. In: Gregg P .E.H., Stewart R.B., Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. Ross C .W. & Mew G. ( 1 99 1 ) Land Rehabilitation After Alluvial Gold Mining in Westland, New Zealand. In: Proceedings of Australian Mining Industry Council Enviromental Workshop, Volume I. Ross C.W. & Widdowson J.P. ( 1 985) Restoration Research After Open Cast Mining in Southland. New Zealand Soil News 33(5): 1 63- 1 69. Ross C.W., Davis M. & Langer L. ( 1 995) Rainforest Restoration After Gold and Coal Mining in Westland, New Zealand. In: Proceedings of 20th Annual Minerals Council of Australia Environmental Workshop, Darwin, Northern Territories, Australia. Ryan P. ( 1 987) Rum Jungle Mine Rehabilitation - Northern Territory. Journal of Soil Conservation Service of New South Wales 43 : 1 8-27. 438 Safaya N.M., McLean J.E. & Halvorson G.A. ( 1 987) Cadmium and Zinc. In: Williams RD., Schuman G.E. (eds) Reclaiming Mine Soils and Overburden in the Western United States : Analytical Parameters and Procedures. Soil Conservation Society of America, Iowa, USA. Salisbury F.B. & Ross C. ( 1 992) Plant Physiology, 4th edn. Wadsworth Publishing Company, Belmont, California. Salmon J.H.M. ( 1 963) A History of Gold-Mining in New Zealand. Government Printers, Wellington. Salmon 1.T. ( 1 980) The Native Trees of New Zealand, 2nd edn. Heinemann Reed, Auckland. Scharer J.M., Kwong E.C.M., Nicholson RV., Pettit C.M. & Chambers D.W. ( 1 995) Factors Affecting ARD Production: Kinetics of Sulphide Oxidation. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1995, vol 2. CANMET, Ottawa, Canada. Scheltus H.W. ( 1 990) Landscape Rehabilitation of Disturbed Land Within Protected Natural Areas. A Case Study of the Ohakune-Horopito Rail Deviation. In: Gregg P.E.H., Stewart RB., Currie L.D. (eds) Proceeding of a Workshop: Issues in the Restoration of Disturbed Land. Fertiliser & Lime Research Centre, Palmerston North. Schofield J.e. ( 1 967) : Sheet 3-Auckland. Department of Scientific and Industrial Research, Wellington. Geological Map of New Zealand. Scale 1 :250,000. Schultze L.E., Zamzow M.J. & Bremner P.R ( 1994) AMD Cleanup Using Natural Zeolites. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2. United States Department of the Interior, Pittsburgh. Seaker E.M. & Sopper W.E. ( 1 983) Reclamation of Deep Mine Refuse Banks with Municipal Sewage Sludge. Waste Management Research 1 : 309-322. Seaker E.M. & Sopper W.E. ( 1 984) Reclamation of Bituminous Strip Mine Spoil Banks with Municipal Sewage Sludge. Reclamation and Revegetation Research 3 : 87- 100. Seaker E.M. & Sopper W.E. ( 1 988a) Municipal Sludge for Minespoil Reclamation: I Effects on Microbial Populations and Activity. Journal of Environmental Quality 17 : 591 -597. Seaker E.M. & Sopper W.E. ( 1 988b) Municipal Sludge for Minespoil Reclamation: II Effects on Organic Matter. Journal of Environmental Quality 1 7 : 598-602. 439 Searle P.L. ( 1 979) Measurement of Adsorbed Sulphate in Soils - Effect of Varying Soil­ Extractant Ratios and Method of Measurement New Zealand Journal of Agricultural Research 22: 287-290. Sengupta M. ( 1993) Environmental Impacts of Mining: Monitoring, Restoration and Control. Lewis Publishers, London. Severson R.C. & Gough L.P. ( 1 983) Rehabilitation Materials from Surface-Coal Mines in Western USA. LChemical Characteristics of Spoil and Replaced Cover-Soil. Reclamation and Revegetation Research 2: 83- 1 02. Shaw S.c., Rorison LH. & Baker A.J.M. ( 1 984) Solubility of Heavy Metals in Lead Mine Spoil Extracts. Environmental Pollution Series B (8): 23-33 . Shelp G., Chesworth W., Spiers G . & Liu L. ( 1 995) Cathodic Protection of a Weathering Ore Body. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 - June 1 , 1 995, vol 3 . CANMET, Ottawa, Canada. Shetty K.G., Hetrick B.A.D., Figge D.A.H. & Schwab A.P. ( 1 994) Effects of Mycorrhizae and Other Soil Microbes on Revegetation of Heavy Metal Contaminated Mine Spoil. Environmental Pollution 86: 1 8 1 - 1 88. Shuk-Ching T. & McColl J.G. ( 1 990) Aluminium and Calcium Binding Affinities of Some Organic Ligands in Acidic Conditions. Journal of Environmental Quality 1 9 : 5 14-520. Simcock R.C . ( 1 993) Reclamation of Aggregate Mines in the Manawatu, Rangitikei and Horowhenua Districts, New Zealand. Ph.D. Dissertation, Massey University, Palmerston North, New Zealand. Simcock R. ( 1 995) Undervalued Aggregate Mines. New Zealand Mining 1 8 : 33 -38 . Simcock R. & Ross C.W. ( 1 995) Methods of Restoring New Zealand Native Forests after Mining. In: Mauk J.L., St-George J.D. (eds) Proceedings of the 1995 PACRIM. Congress. 1 9-22 November 1 995. The Australasian Institute of Mining and Metallurgy, Carlton, Australia. Sine N.M., Taylor W.O., Webber G.R. & Lewis C.L. ( 1 969) Third Report of Analytical Data for CAAS Sulphide Ore and Syenite Rock Standards. Geochimica et Cosmochimica Acta 33 : 1 2 1 - 1 3 1 . Singer P.C. & Stumm W. ( 1 970) Acidic Mine Drainage: The Rate Detennining Step. Science 1 67: 1 12 1 - 1 1 23 . Skinner D.N.B. ( 1 986) Neogene Volcanism in the Hauraki Volcanic Region. In: Smith LE.M. (ed) Late Cenozoic volcanism of New Zealand. Royal Society of New Zealand Bulletin, 23 . Royal Society of New Zealand, (N). 440 Skousen J. ( 1 995) Acid Mine Drainage. In: Skousen J., Ziemkiewicz P.F. (eds) Acid Mine Drainage - Control & Treatment. West Virginia University and the National Land Reclamation Centre, Morgantown, West Virginia. Skousen J. & Larew G. ( 1 994) Alkaline Overburden Addition to Acid-Producing Materials to Prevent Acid Mine Drainage. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Skousen J. & Ziemkiewicz P.F. (eds) (1 995) Acid Mine Drainage - Control & Treatment. West Virginia University and the National Mine Reclamation Center, Morgantown, West Virginia. Skousen J., Politan K., Hilton T. & Meek A. ( 1990) Acid Mine Drainage Treatment Systems: Chemicals and Costs. Green Lands Fall: 3 1 -37. Skousen J., Smith R. & Sencindiver J. ( 1 995) The Development of the Acid-Base Account. In: Skousen J., Ziemkiewicz P.F. (eds) Acid Mine Drainage - Control & Treatment. West Virginia University and the National Land Reclamation Centre, Morgantown, West Virginia. Skow J. ( 1 993) Mother Lode vs. Mother Nature. Time (US Edition) November, 22: 58- 60. Smith D.G. ( 1 986) Heavy Metals in the New Zealand Aquatic Environment: A Review. Water & Soil Miscellaneous Publication No. 1 00 . Water Quality Centre. Ministry of Works and Development, Wellington, New Zealand. Smith G.S. , Edmeades D.C. & Upsdell M. (1 983) Manganese Status of New Zealand Pastures. l .Toxicity in ryegrass, white clover and lucerne. New Zealand Journal of Agricultural Research 26: 2 1 5-22 1 . Smith M.J. ( 1 974) Acid Production in Mine Drainage Systems. In: Deju R.A. (ed) Extraction of Minerals and Energy. Ann Arbor Science, Ann Arbor, Michigan Smith M.R. & Sobek A.A. ( 1 978) Physical and Chemical Properties of Overburdens, Spoils, Wastes, and New Soils. In: Schaller F.W., Sutton P. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison, Wisconsin. Smith R.A.H. & Bradshaw A.D. ( 1 979) The Use of Metal Tolerant Plant Populations for the Reclamation of Metalliferous Wastes. Journal of Applied Ecology 1 6 : 595-6 12. Smith R.M., Grube W.E. & Arkle T. ( 1 974) Mine Spoil Potentials for Soil and Water Quality. USEPA Report 670- 12-74-070. USEPA, Cincinnati. 441 Smith S.R. ( 1 99 1 ) Effects of Sewage Sludge Application on Soil Microbial Processes and Soil Fertility. Advances in Soil Science 1 6: 1 9 1 -2 1 2. Smith S.R. ( 1 994) Effect of Soil pH on Availability to Crops of Metals in Sewage Sludge-Treated Soils. II. Cadmium Uptake by Crops and Implications for Human Dietary Intake. Environmental Pollution 86: 5 - 1 3 . Sobek AA., Schuller W.A, Freeman lR. & Smith R.M. ( 1 978) Field and Laboratory Methods Applicable to Overburdens and Mine Soils. USEP A Report 600/2-78- 054. USEP A, Cincinnati. Sobek AA, Rossner L.R., Sorensen D.L., Sullivan PJ. & Fransway D.F. ( 1 987) Acid­ Base Potential and Sulfur Forms. In: Williams R.D., Schuman G.E. (eds) Reclaiming Mine Soils and Overburden in the Western United States : Analytical Parameters and Procedures. Soil Conservation Society of America, Ankeny, Iowa. Sopper W.E. ( 1 992) Reclamation of Mine Land Using Municipal Sludge. In: Lal R., Stewart B.A (eds) Advances in Soil Science - Soil Restoration, vol 1 7. Springer-Verlag, New York. Sopper W.E. & Kerr S .N. ( 1 982) Mine Land Reclamation with Municipal Sludge - Pennsylvania's Demonstration Program. In: Sopper W.E., Seaker E.M., Bastian R.K. (eds) Land Reclamation and Biomass Production with Municipal Wastewater and Sludge. Pennsylvania University Press, University Park, PA. Sopper W.E. & Seaker E. ( 1990) Long-Term Effects of a Single Application of Municipal sludge on Abandoned Mine Land. In: Skousen J. (ed) Proceedings of the 1 990 Mining and Reclamation Conference and Exhibition, vol 2. West Virginia University, Morgantown, WV. Sorensen D.L., Kneib W.A, Porcella D.B. & Richardson B.Z. ( 1 980) Determining the Lime Requirement for the Blackbird Mine Spoil. Journal of Environmental Quality 9. Splittorf D. & Rastogi V. ( 1 995) Ten Year Results from Bactericide-Treated and Reclaimed Mine Land. In: Proceedings of 1 995 National Meeting of the American Society for Surface Mining and Reclamation. Gillete, Wyoming, June 5-8 . American Society for Surface Mining and Reclamation. St-Arnaud L. ( 1 994) Water Cover for the Decommissioning of Sulfidic Mine Tailings Impoundments. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 1 . United States Department of the Interior, Pittsburgh. Steinbergs A., Lismma 0., Freney J.R. & Barrow N.J. ( 1 962) Determination of Total Sulphur in Soils and Plant Material. Analytica Chimica Acta 27: 1 58- 1 64. 442 Stevenson F.J. ( 1982) Humus Chemistry: Genesis, Composition, Reactions. John Wiley & Sons, New York. Stroo H.F. & Jenks E.M. ( 1 985) Effect of Sewage Sludge on Microbial Activity in an Old, Abandoned Minesoil. Journal o/Environmental Quality 1 4: 301 -304. Stumm W. & Morgan 1.L. ( 1 98 1 ) Aquatic Chemistry. Wiley, New York. Sturey C.S., Freeman J.R., Keeney T.A. & Stem J.W. ( 1 982) Overburden Analysis by Acid-Base Accounting and Simulated Weathering Studies as a Means of Determining the Probable Hydrological Consequences of Mining and Reclamation. In: Graves D.H., De Yore R.W. (eds) Symposimn on Surface Mining, Hydrology, Sedimentology and Reclamation. University of Lexington, December 5 - 1 0. University of Lexington, Lexington, Kentucky, USA. Sudhakar c., Syamalabai L. & Veeranjaneyulu K. ( 1 992) Lead Tolerance of Certain Legmne Species Grown on Lead ore Tailings. Agriculture, Ecosystems and Environment 4 1 : 253-26 1 . Sullivan P.J. & Yelton J.L. ( 1 988) Iron Sulphide Oxidation and the Chemistry of Acid Generation. Environmental Geology and Water Science 1 1 (3) : 289-295. Sullivan P.J. & Sobek A.A. ( 1 982) Laboratory Weathering Studies of Coal Refuse. Minerals and the Environment 4: 9- 1 6. Sullivan P.J., Yelton J.L. & Reddy K.J. ( 1 988) Solubility Relationships of Almninimn and Iron Minerals Associated with Acid Mine Drainage. Environmental Geology and Water Science 1 1 (3): 282-287. Sutton P. & Vimmerstedt J.P. ( 1 974) Treat Stripmines Spoils with Sewage Sludge. Compost Science 1 5( 1 ) : 22-23 . Takijima Y. & Kasumi F. ( 1 973) Cadmium Contamination of Soils and Rice Plants Caused by Zinc Mining. I. Production of high cadmimn rice on the paddy fields in lower reaches of the mine station. Soil Science and Plant Nutrition 19 : 29-38. Tan K.H. (1 993) Principles of Soil ChemiStry, 2nd edn. Marcel Dekker, Inc. , New York. Tay K.A. ( 1 980) Geochemistry and Environmental Impact of the Discharge of Heavy Metals from the Tui Mine and its Tailings. MSc. Thesis, University of Waikato, Hamilton. Taylor N.H. & Pohlen I. ( 1 962) Classification of New Zealand Soils. In: Soils of New Zealand, Part 1 . Soil Bureau Bulletin 26( 1 ) with 1 : 1 000 000 scale soil map of New Zealand. Department of Scientific and Industrial Research, Wellington. Taylor P.C. & Walker B.V. ( 1 987) The Development of Mining in New Zealand. In: Proceedings of Pacific Rim Congress. pp 679-687. 443 Tester C.F. ( 1 990) Organic Amendment Effects on Physical and Chemical Properties of a Sandy Soil. Soil Science Society of America Journal 54: 827-83 1 . Thomas, RF ( 1973) Determination of Particle Size Distribution. In: New Zealand Bureau Report No. 1 0. DSIR, Wellington. Thurman D. & Hardwick K. ( 1988) How Plants Survive an Overdose of Metal . New Scientist 10: 44-45 . Tonkin & Taylor ( 1974) Hauraki Catchment Board Stability Study of Norpac Mine Tailings Dam. Unpublished report prepared for the Hauraki Catchment Board, Te Aroha, New Zealand. Topper K.F. & Sabey B.R. ( 1 986) Sewage Sludge as a Coal Mine Spoil Amendment for Revegetation in Colorado. Journal of Environmental Quality 1 5 : 44-49. Tremblay R.L. (1 994) Controlling Acid Mine Drainage using an Organic cover: The case of the East Sullivan Mine, Abitibi, Quebec. In: International Land Reclamation and Mine Drainage Conference and the Third International Conference on the Abatement of Acidic Drainage, vol 2 . United States Department of the Interior, Pittsburgh. Tucker G.B., Berg W.A. & Gentz D.H. (1 987) pH. In: Williams R.D., Schuman G.E. (eds) Reclaiming mine soils and overburden in the western United States. Soil Conservation Society of America, Iowa, USA. Tui Meeting (1 993) Tui Mine - Notes of a Meeting ( 1 8-5-93). Facilitated by Environment Waikato and Department of Conservation (DOC) Staff. Held at Te Aroha DOC Office. Tuovinen O.H. (1 990) Biological Fundamentals of Mineral Leaching Processes. In: Ehrlich H.L., Brierley C.L. (eds) Microbial Mineral Recovery. McGraw-Hill Publishing Company, New York. Turner D. & McCoy D. ( 1 990) Anoxic Alkaline Drain Treatment Systems, a Low Cost Acid Mine Drainage Treatment Alternative. In: Graves D.H., DeVore R.W. (eds) Proceedings of the 1 990 National Symposium on Mining. Lexington, Kentucky, May 14- 1 8 . Unger, P.W. & Kaspar, T.C. ( 1 994). Soil Compaction and Root Growth: A Review. Agronomy Journal 86: 759-766. Ure A.M. ( 199 1 ) Atomic Absorption and Flame Emission Spectrometry. In: Smith K.A. (ed) Soil Analysis: Modern Instrumental Techniques, 2nd edn. Dekker, New York. US EPA (1983, October) US Environmental Protection Agency process Design Manual for Land Application of Municipal Sludge. Office of Research and Development. Municipal Environmental Research Laboratory, Cincinatti. 444 Van Breeman N. ( 1973) Soil-Forming Processes in Acid Sulphate Soils. In: Dost H. (ed) Acid Sulphate Soils, vol 1 . International Institute for Land Reclamation and Improvement, Publication 1 8 . Van Loon lC. & Lichwa L. ( 1 973) A Study of the Atomic Absorption Detennination of Some Important Heavy Metals in Fertilisers and Domestic Sewage Sludges. Environmental Letter 3 : 1 -8. Verma T.R & Thames J.L. ( 1 978) Grading and Shaping for Erosion Control and Vegetative Establishment in Dry Regions. In: Schaller F.W., Sutton P. (eds) Reclamation of Drastically Disturbed Lands. American Society of Agronomy, Madison, Wisconsin. Viets, F.G. Jr. & Boawn, L.C. ( 1 965) Zinc In: Black et aI. , (eds.) Methods of Soil Analysis: Part 2. Agronomy 9 : 1 090- 1 1 0 1 . Visser S . , Griffiths C.L. & Parkinson D . ( 1 983) Effects of Mining on the Microbiology of a Prairie Site in Alberta, Canada. Canadian Journal of Soil Science 63: 1 77- 1 89. Waihi Gazette ( 1 995) Gold Finns Join in Tui Mine Clean-Up. Waihi Gazette, 25th July. Walder I.F. & Chavez W.X.J. ( 1 995) Mineralogical and Geochemical Behaviour of Mill Tailing Material Produced from Lead-Zinc Skarn Mineralisation, Hanover, Grant County, New Mexico, USA. Environmental Geology 26: 1 - 1 8 . Wallace C. ( 198 1 ) Who's who? New Zealand Environment 29. Ward N.E., Brooks RR & Reeves RD. ( 1 976) Copper, Cadmium and Zinc in Soils, Stream Sediments, Waters, and Natural Vegetation Around the Tui Mine, Te Aroha, New Zealand. New Zealand Journal of Science 19 : 8 1 -89. Ward N.!., Brooks R.R. & Roberts E. (1 977) Contamination of a Pasture by a New Zealand Base-Metal Mine. New Zealand Journal of Science 20: 41 3-4 19. Watanabe T., Sawada Y., Russel J .D. , McHardy W.J. & Wilson M.J. ( 1 992) The Conversion of Montmorillonite to Interstratified Halloysite-Smectite by Weathering in the Omi Acid Clay Deposit, Japan. Clay Minerals 27: 1 59- 1 73 . Watzlaf G.R ( 1 986) Control of Acid Drainage from Mine Wastes using Bacterial Inhibitors. In: Proceedings of the 1 986 National Meeting of the American Society for Surface Mining and Reclamation, Jackson, Mississippi, March 17- 20. American Society for Surface Mining and Reclamation. Watzlaf G.R ( 1 987) Control of Acidic Drainage from Waste Rock and Sulfide Tailings by Chemical Inhibition of Iron-Oxidising Bacteria. In: Proceedings of the Twelfth Annual Meeting of the Canadian Land Reclamation Association. Sudbury, Ontario, June 7- 1 1 , 1 987. Canadian Land Reclamation Association, Sudbury, Canada. 445 WatzlafG.R. ( 1 988) Chemical Inhibition of Iron-Oxidising Bacteria in Waste Rock and Sulfide Tailings and Effect on Water Quality. In: Mine Drainage and Surface Mine Reclamation. Volume I : Mine Water and Mine Waste. Conference Proceedings April 1 9-2 1 , 1 988. Bureau of Mines Information Circular 9 1 83 . United States Department of the Interior, Pittsburgh. Wear J.1. & Evans C.E. ( 1 968) Relationship of Zinc Uptake by Corn and Sorghum to Soil Zinc Measured by Three Extractants. Soil Science Society of America Proceedings 32: 543-546. Wei Q.F. , Lowery B. & Peterson A.E. ( 1 985) Effect of Sludge Application on Physical Properties of a Silty Clay Loam Soil. Journal of Environmental Quality 1 0 : 556- 56 1 . Weissberg B.G. & Wodzicki A. ( 1 970) Geochemistry of Hydrothermal Alteration and Origin of Sulphide Mineralisation at the Tui Mine, Te Aroha, New Zealand. New Zealand Journal of Science March: 36-60. Wells O.C. ( 1 974) Scanning Electron Microscopy. Magraw-Hill Company, New York. Welton lE. ( 1 984) SEM Petrology Atlas. The American Association of Petroleum Geologists, Tulsa, Oklahoma, USA. Werner W., Scherer H.W. & Reinartz F. ( 1 99 1 ) Experiences of the Usage of Sewage Sludge for Reclaiming Open Cast Mining Areas and Amelioration of Very Steep and Stony Vineyards. In: Hall J.E. (ed) Alternative uses for sewage sludge. Pergamon Press. Weston R.M. ( 1 995) Golden Cross Mine - Mining and Processing Operations, Waihi, New Zealand. In: Mauk J.L., St-George J.D. (eds) Proceedings of the 1 995 PACRlM Congress. 1 9-22 November 1 995. The Australasian Institute of Mining and Metallurgy, Carlton, Australia. Whiteley G.M. & Williams S . ( 1 993) Effects of Treatment of Metalliferous Mine Spoil with Lignite Derived Humic Substances on the Growth Responses of Metal Tolerant and Non Metal Tolerant Cultivars of Agrostis capillaris L. Soil Technology 6: 163- 1 7 1 . Whitton J.S. & Churchman G.J. ( 1 987) Standard Methods for Mineral Analysis of Soil Survey Samples for Characterisation and Classification in NZ Soil Bureau. NZ Soil Bureau Scientific Report 79. NZ Soil Bureau, Department of Scientific and Industrial Research, Lower Hutt. Widdowson J.P. & McQueen D. ( 1 990) Rehabilitation After Opencast Mining in Southland. In: Gregg P .E.H., Stewart R.B., Currie L.D. (eds) Proceedings of the Workshop 'Issues in the Restoration of Disturbed Land' Occasional Report No.4. Fertiliser and Lime Research Centre, Massey University, Palmerston North, New Zealand. 446 Widdowson J.P., Gibson EJ. & Healy W.B. ( 1 982) Effects of Stockpiling Topsoils Associated with Opencast Mining 1 . Chemical Properties and the Growth of Ryegrass and White clover. New Zealand Journal of Science 25: 287-294. Widdowson J.P., Ross C.W., McQueen D. & Watts H.M. ( 1 984) Plant Growth Potential of Tailings and Waste Materials From Martha Hill, Waihi. NZ Soil Bureau Report, DSIR, Lower Hutt, New Zealand. Wiggering H. ( 1993) Sulphide Oxidation - An Environmental Problem within Colliery Spoil Dwnps. Environmental Geology 22: 99- 105 . Willet LR., Crockford RH. & Milnes AR (1992) Transformation of Iron, Manganese and Alwniniwn During Oxidation of a Sulfidic Material from an Acid Sulfate Soil. Catena-Supplement 2 1 : 287-302. Williams G.J. (1 965) Economic Geology of New Zealand, vol 4. Australian Institute of Mining & Metallurgy. Melbourne. Williams K. (1 985) Mining in New Zealand. New Zealand Environment 46: 40-43 . Williamson A & Johnson M.S. ( 1 98 1 ) Reclamation of Metalliferous Mine Waste. In: Lepp N.B. (ed) Effect of heavy metal pollution on plants, vol 2. Applied Science, London. Wilson A.D. ( 1 980) Soils of Piako County, North Island, New Zealand. NZ Soil Survey Report 39. New Zealand Soil Bureau. DSIR, Wellington. Winterhalder K. ( 1 991) Minimal Soil Amelioration as a Trigger Factor in the Revegetaion of a Derelict Landscape near Sudbury, Canada. In: Davies M.C.R (ed) Land Reclamation. An End to Dereliction? Elsevier Applied Science, London. Winterhalder K. (1 995a) Approaches to Engendering Biodiversity in Tailings Revegetation. In: Hynes T.P. , Blanchette M.C. (eds) Proceedings of Sudbury '95 Mining and the Environment. May 28 - June 1, 1995 . Sudbury, Ontario, Canada, vol III. CANMET, Ottawa. Winterhalder K. ( 1 995b) Minimal Soil Amelioration and Native Sod Transplanting as a Means of Restoring Biodiversity on Smelter-Affected Lands. In: Hynes T.P., Blanchette M.C. (eds) Proceedings of Sudbury '95 - Mining and the Environment. May 28 -June 1 , 1995. Sudbury, Ontario, Canada, vol III. CANMET, Ottawa. Wodzicki A & Weissberg B.G. ( 1 970) Structural Control of Base Metal Mineralisation at the Tui Mine, Te Aroha, New Zealand. New Zealand Journal of Geology and Geophysics 13 : 6 1 0-630. 447 Wong M.H. ( 1 982) Metal Cotolerance to Copper, Lead and Zinc in Festuca rubra. Environmental Research 29: 42-47. World Health Organisation (1 984a) Guidelines For Drinking-Water Quality, vol l . World Health Organisation, Geneva. World Health Organisation ( 1 984b) Guidelines For Drinking-Water Quality, vol 2. World Health Organisation, Geneva. WRCM ( 199 1 ) Waikato Regional Council Memorandum : Tui Mine Tailings. 1 July 1 99 1 , File 60 1 1 04A. Wu L. & Lin S.L. ( 1 990) Copper Tolerance and Copper Uptake of Lotus purshians (Benth.) Clem & Clem. and its Symbiotic Rhizobium loti Derived from a Copper Mine Waste Population. New Phytology 1 1 6: 53 1 -539. Zajic J.E. ( 1 969) Microbial Biochemistry. Academic Press, New York, London. Ziemkiewicz P.F. & Skousen lG. ( 1 995) Prevention of Acid Mine Drainage by Alkaline Addition. In: Skousen J., Ziemkiewicz P. (eds) Acid Mine Drainage - Control & Treatment. West Virginia University and the National Mine Reclamation Center, Morgantown, West Virginia. Appendices Appendix 4(A) Total Sulphur Determination by Sodium Bicarbonate Fusion (Adapted from Steinbergs et al., 1 952) Reagents (i) Oxidising Agent: Mix NaHC03 and Ag20 at 1 2.5 : 1 ratio. (ii) 5MHCl (iii) NaHC03 Method (i) Grind samples to less than 1 50 /lm. (ii) Accurately weigh about 0.25 g of sample into a pyrex tube and add 0.5 g of oxidising agent. Thoroughly mix contents with vortex mixer. (iii) Cover resulting mixture in test tube with approximately 0.5 g ofNaHC03. (iv) Fuse in muffle furnace at 550°C for 3 hours. (v) After cooling very slowly add 1 0 mL of 5MHCl. (vi) Mix solution with a vortex mixer and heat solution at 90°C for 1 hour. (v) Filter with Whatman No. 41 and quantitatively dilute. Analysis 448 (i) Analyse total sulphur in digest using auto-analyser and methylene blue method of Johnson & Nishita ( 1 952) (Blakemore et al., 1 987). Appendix 4(B) Calcium Phosphate Extractable Sulphate (Adapted from Searle, 1 979) Reagents (i) 0 .0 IMCa(H2P04)2.H20 solution . Conditions (i) Solution to sample ratio 5 : 1 (ii) Temperature 20 ± 2°C (iii) Period of shaking � hour (iv) Shaking speed 80- 1 00 rpm Method (i) Weigh 5 .0 g of sample into a centrifuge tube. (ii) Add 25 rnL of phosphate solution to the sample. (iii) Shake the solution for � an hour in an end of end shaker. (iv) Centrifuge at 9000 rpm for 5 minutes. (v) Decant solution through Whatman No. 41 filter paper. and quantitatively dilute samples with phosphate solution. Analysis (i) Analyse phosphate extractable sulphur in the digest using auto-analyser and methylene blue method of Johnson & Nishita ( 1 952) (Blakemore et ai., 1 987). Appendix 4(C) Acid Generating Potential (AGP) Based on Sulphur Analysis (Adapted from Sobek et al., 1 987). Method (i) Determine total sulphur content of sample(Appendix 4. 1A). (ii) Determine phosphate extractable sulphate content of sample (Appendix 4. 1B) . (iii) Estimate % Sulphide S using Equation 1 . [Equation 1 ] Sulphide S [%] = Total S [mglkgl - sulphate S [mglkg] 10,000 449 Based on the stoichiometry of FeS2 oxidation a 1 000 Mg of a material containing O. l % S (all as pyrite) will oxidise to produce a quantity of H2S04 which requires 3 . 125 Mg of CaC03 to neutralise (Sobek et al., 1 987). Thus, the AGP of a sample can be calculated using Equation 2. [Equation 2] AGP [Mg CaC03/1 000 Mg of tailings] = Sulphide S [%] * 0.3 1 25 Appendix 4(D) HFIHN03 Total Metal Extraction (Massey University Soils Dept. Standard Method) Reagents (i) Concentrated Nitric Acid (HN03) (ii) Concentrated Hydrofluoric Acid (HF) (iii) 2M Hydrochloric Acid (Hel) Method Day 1 (i) Accurately weigh about O. lg to 0.2g of air-dry sample into a polypropylene beaker. (ii) Add 5 mL of concentrated HN03 and leave the beaker to stand in a fume hood for about 1 � to 2 hours. (iii) Add 5 mL of concentrated HF acid (40%) and leave the beaker to stand overnight in a fume hood. Day 2 (iv) Evaporate the sample to dryness on a water bath at 80°C (approx. 3 hours). (v) Add 5 mL of concentrated HF acid (40%) and again take to dryness. (vi) Dissolve the residue in 5 mL of concentrated HCl. (vii) Make to 1 00 mL volume in volumetric flask with 2MHCl (rinse beaker residue). (viii) Analyse. Appendix 4(E) Reagents (i) 25% v/v RN03 Method 25% (v/v) HN03 Pseudo-Total Metal Extraction (Adapted from Smith & Bradshaw, 1 979) (i) Grind sample to <1 50 )lm (optional). (ii) Take 0.5 g of air-dry sample into a 200 mL digestion tube. (iii) Add 25 mL of25% v/v nitric acid and mix thoroughly with vortex mixer. (iv) Note total mass, and then cover with small glass funnel. (v) Heat the samples for 1 hour at approximately 1 00°C. (vi) Transfer to centrifuge tubes and centrifuge at 9000 rpm for 5 minutes. (viii) Filter sample through Whatman No. 4 1 filter paper. (ix) Quantitatively dilute to 1 00 mL with 1 : 1 H20 / RN03. (x) Analyse. 450 Appendix 4(F) Reagents (i) O . lMHCl Method O.lMHCI Metal Extraction Procedure (Adapted from Viets & Bowen, 1 965) (i) Accurately weigh a 5.00g air-dry sample into 50-mL polypropylene centrifuge tubes. (ii) Add 25 mL of the extractant and shake the samples end-over-end for 1 6 hours. (iii) Remove and centrifuge samples at 9000 rpm for 5 minutes. (iv) Filter solutions Whatman No. 4 1 . (v) Dilute as necessary and analyse. Appendix 4(G) Reagents 30% H202 Metal Extraction Procedure (Trial Procedure) (i) 30% Analytical Grade Hydrogen Peroxide (H202) (ii) O . lMHCl Method (i) Accurately weigh 0. 1 to 0.5g air-dry sample into a large wide-mouth conical flask, recording mass of flask. (ii) Add 20 mL ofH202 and gently warm on a hot plate until reaction ceases. (iii) Add a further 20 mL of H202 allow reaction to terminate and boil until near dryness. (iv) Repeat the above step until no further reaction occurs. 45 1 (v) Take sample to near dryness and gravimetrically readjust extractant volume to 25 mL with O. lM HCl. (vi) Centrifuge sample at 9000 rpm for 5 minutes and filter through Whatman No. 4 1 paper. (v) Dilute as necessary with O. lMHCl and analyse. 452 Appendix 4(H) Theoretical Acid Generating Potential (AGP) - Working Examples This appendix includes working examples that demonstrate how the AGP of the Tui Mine tailings was estimated based on their sUlphidic Fe & Cu content. Three examples are included for samples containing l OOO mg/kg pyritic Fe, 1000 mg/kg sulphidic Cu and a mixed sample containing 1 000 mg/kg and 1 0 000 mg/kg of sulphidic Cu and Fe respectively. Assumptions made with respect to the use of this method are discussed in Chapter 4 (section 4 .4). AGP of Pyrite (FeS2) This method assumes that one mole of pyrite oxidises to form between 1 .33 and 2.0 moles of H2S04 (Bruynesteyn & Hackl, 1 982). Therefore for every mole of pyrite oxidised between 1 .33 and 1 .0 mole of CaC03 is required to neutralise the acid produced. General Data pb (tailings) Area (hectare) Depth of sample :. Volume of tailings\ Ha :. Mass oftailings\Ha Molecular Masses FeS2 EXAMPLE 1 (pyrite) = 1 .3 Mg\m3 = 10 000 m2 = 0. 1 0 m 0. 1 * 1 0 000 = 1 000 m3 _ 1 000 * 1 .3 _ 1 300 MglHa = 55 .847 + 2(32.06) = 120 g = 40.08 + 12.0 1 1 + 3(1 5.9994) = 1 00. 1 g If tailings sample is analysed and found to contain 1 000 mg/kg Fe as pyrite then the AGP of the sample can be estimated as follows: Fe FeS2 (%) 1 000 mg/kg = ( 1 000/55 .847)* 1 1 9.967(MW FeS2) = 2148 mg/kg (2148 mg/kg 11 000 000 mg/kg )* 1 0011 = 0.2 148 % Mass (FeS2/1 000 Mg) = = = Moles (FeS211 000 Mg) = = 1 000 Mg * 0.2 148 % 2. 148 Mg 2 148 000 g 2 1 48 000/1 1 9.967 (MW FeS2) 17.9 k moles FeS2 :. Moles of sulphuric acid potentially produced Lower limit Upper limit = 1 7905 * 1 .33 23 . 8 k moles of H2S04 = 1 7.9 * 2 .0 = 35.8 k moles of H2S04 453 . . Between 23 .8 and 35 .8 k moles ofCaC03 will be required to neutralise the acid produced from 1 000 Mg of tailings containing 0.2 148 % pyrite ( 1 000 mglkg pyritic Fe) Liming requirement in Mg CaC03\1000 Mg of material Mass ofCaC03 [Mg] = (moles ofCaC03* molecular weight ofCaC03) 1 1 000 000 Lower limit AGP 1 ° Lower limit AGP 2° (238 13 * 1 00. 1 )/ 1 000 000 2.3 83 Mg CaC03\1 000 Mg = 2.4 Mg CaC03\1 000 Mg of tailings = (358 10 * 1 00.0892)/ 1000 000 = 3 .584 Mg CaC03\1 000 Mg = 3.6 Mg CaC03\1 000 Mg of tailings EXAMPLE 2 (Chalcopyrite) It is assumed that one mole of chalcopyrite (CuFeS2) oxidises to form between 0.33 and 1 .0 mole of H2S04 (Bruynesteyn & Hackl, 1 982). Therefore for every mole of chalcopyrite oxidised between 0.33 and 1 .0 mole of CaC03 is required to neutralise the acid produced. Atomic Masses Cu Fe S CuFeS2 = = = 63 .546 g 55 .847 g 32.06 g 1 83.5 1 3 g The amount of S associated with the Cu in chalcopyrite; = (1 000/63 .546) * 64. 12 = 1 009 mglkg 454 The amoWlt of Fe associated with the Cu in chalcopyrite; = (1 000/63 .546) * 55 .847 = 879 mglkg If the tailings sample is analysed and found to contain 1 000 mglkg Cu all as chalcopyrite then the AGP of the sample can be estimated as follows: Cu = :. CuFeS2 = = CuFeS2 (%) = Mass = (CuFeS211 000 Mg) = = Moles = (CuFeS211 000 Mg) 1 000 mglkg ( 1 000/63 .546 ) * 1 83 .5 1 3 2888 mglkg (2888 mglkg 11 000 000 mglkg) * 1 0011 0.2888 % 1000 Mg * 0.2888 % 2.888 Mg 2 888 000 g 2 888 0001183 .5 1 3 (MW CuFeS2) 1 5 .7 k moles .' . Moles of sulphuric acid potentially produced from the oxidation of 1 5 .7 k moles of chalcopyrite; Lower limit 1 5 .7 * 0.33 = 5.2 k moles Upper limit 1 5 .7 * 1 .0 = 1 5 .7 k moles Between 5 .2 and 1 5 .7 k moles ofCaC03 will be required to neutralise the acid produced from 1000 Mg of tailings containing 0.2888% chalcopyrite ( 1 000 mglkg Cu) Liming requirement in Mg CaC03\1000 Mg of material Mass of CaC03 [Mg] Lower limit AGP 1 ° (moles of CaC03 * molecular weight of CaC03)/ 1000 000 (5246 * 1 00.0892)11 000 000 = 0.525 1 Mg CaC03\1 000 Mg = 0.5 Mg CaC03\1000 Mg Lower limit AGP 2° = ( 1 5737 * 1 00.0892)/1 000 000 = 1 .575 Mg CaC03\1 000 Mg = 1 .6 Mg CaC03\1 000 Mg EXAMPLE 3 (Mixed Sulphide Sample) 455 If tailings sample is analysed and found to contain 1 000 mglkg and 10 000 mglkg sulphidic Cu and Fe respectively, the AGP of the sample can be estimated as follows: Pyritic AGP If Cu = 1 000 mg/kg the amount of Fe associated with the Cu in chalcopyrite; Fe (1 000/63 .546) * 55 .847 879 mg/kg Therefore for every I g of Cu analysed in solution 0.879 g of Fe will be associated with the Cu in the case of chalcopyrite, that is assuming all the Cu is derived from chalcopyrite. The remaining Fe is assumed to be associated with pyrite. :. Fe (associated with F eS2) AGP 1 0 (FeS2) (lower limit) AGP 20 (FeS2) (upper limit) Chalco pyritic AGP AGP 10 (CuFeS2) (lower limit) AGP 20 (CuFeS2) (upper limit) Total AGP Lower limit = Fe (total) - {(CU(total) * Atomic weight ratio(Fe/Cu)} = 1 0 000 - (1 000 * 0.879) = 9 12 1 mg/kg = ( [Fe]/1 000) * 2.383 = (9 121/ 1 000) * 2.383 = 2 1 .7 Mg CaC03/ 1 000 Mg of material = ([Fe]/1 000) * 3 .584 = (912 1 /1 000) * 3 .584 = 32.7 Mg CaC03/ 1 000 Mg of material = ([Cu]/1 000) * 0.525 1 = ( 1 000/1000) * 0.525 1 = 0.53 Mg CaC03/ 1 000 Mg of material = ([Cu]/1 000) * 1 .575 = ( 1 000/1 000) * 1 .575 = 1 .6 Mg CaC03/ 1 000 Mg of material AGP 1 0 (FeS2) + AGP 1 0 (CuFeS2) 2 1 .7 + 0.53 22.2 Mg CaC03/1 000 Mg material Upper limit = AGP 2° (FeS2) + AGP 2° (CuFeS2) 3 1 .8 + 1 .6 = = 34.3 Mg CaC03/1000 Mg material 456 Stoichiometrically, the calcium carbonate equivalent of acid produced from a material containing, 1 0 000 mglkg of Fe, and 1 000 mglkg of Cu in the form of FeS2 and CuFeS2 is between 22 and 35 Mg of CaC03 per 1 000 Mg of material. Appendix 4(1) Reactants Hydrogen Peroxide Oxidation AGP Method (NAG Method of Miller et at., 1 994) (i) 1 5% Analytical Grade (unstabilised) H202 Method (i) Place 2 .5 ± 0. 1 g sample of pulverised * waste rock or tailings into a 500-mL conical glass beaker and add 250 mL of unstabilised 1 5% H202. Cover with watch-glasses and place in fume hood. (ii) Allow reaction to occur until either, (a) the sample boils then cools or (b) after at least 1 2 hours has elapsed. (iii) Heat sample on a hotplate at 1 00 °C until all effervescence ceases and the solution boils. (iv) After the solution has cooled measure the pH and the titratable acidity of the solution. (v) The titratable acidity of the solution is measured by titrating the solution to an endpoint of pH 7.0 using standardised 0. 1 00 N NaOH. If the solution pH is < 2.5 use 0.500 N NaOH. If the sample gives a net acid generating potential equivalent value of > 25 Mg H2SOJI000 Mg of material repeat the procedure using 1 .0 g samples. The method provided did not state the grain-size to which the samples should be ground. Finkelman & Giffin ( 1986) recommended grinding the sample fine enough to remove the pyrite grain size effect; a grainsize of < 325 mesh (44 J.lIn) was used. They state, however that the grain-size effect can yield useful information on the mean or effective pyrite grain-size, a useful parameter for spoil handling strategies as sample reactivity is strongly dependent on the size factor. However for the purposes of this study, the samples were ground to < 1 50 Ilm with a tungsten-carbide ring grinder. Unground samples were also run. Appendix 4(J) Method Neutralising Potential Method (Adapted from Sobek et ai., 1 978) (i) Add 50 mL of O. lM HCI to a 2.0 g tailings sample « 1 50 /lm). 457 (ii) Heat the solution at near boiling, stirring occasionally until CO2 evolution ceases. (iii) Make the solutions up to 1 25 mL with distilled water. (iv) Boil samples for 1 minute. (v) Allow samples to cool to room temperature. (vi) Titrate samples with standard O. IMNaOH solution to pH 7.0 using a pH meter as per the titration method outlined below. (vii) Repeat titration on duplicate sample reacting with distilled water. Titration Method (i) A few drops of methyl red indicator were placed in the cooled solutions prior to titration to indicate when the solution was approaching pH 7.0. (Methyl red changes colour from red to yellow in the pH range of 4.8 to 6.0) (ii) The samples were gently agitated during the titration by a flea and magnetic stirrer to mix the solution as the NaOH was added by burette. A pH meter was used to take the solution to the endpoint pH 7.0, after solution colour change occurred. Calculation of CaC03 equivalent from NaOH consumption The calcium carbonate equivalent of a material can then calculated stoichiometrically based on its consumption of HC!. Firstly the moles of 0. 1 0M HCI added to the sample are calculated using Equation 1 . [Equation 1 ] Moles Hel added = initial volume HCI added [mL] x concentration [M]/ 1 000 mL = A Secondly the number of moles of NaOH required to neutralise the acid remaining in solution after the reaction are calculated using Equation 2. [Equation 2] Moles NaOH required = volume NaOH added [mL] x concentration [M]/ 1 000 mL = B Since 1 mole of NaOH reacts with 1 mole of Hel, it is possible to calculate the number of moles ofHCI consumed by the reaction of the material with the acid by Equation 3. [Equation 3] Moles of HCI consumed by sample = A - B [Equation 4] CaC03 + 2 Hel => eaCh + C02 + H20 458 Based on Equation 4 it is seen that 1 .0 mole of CaC03 neutralises 2.0 moles of HCl. Therefore the number of moles of CaC03 required to neutralise the moles of acid consumed by the reaction is shown in Equation 5. [Equation 5] Moles of CaC03 = A - B / 2 Since one mole of CaC03 = 1 00 g, the amount of CaC03 [g CaC03/g material] required to neutralise the acidity produced by a given mass of material equals; [Equation 6] Calcium Carbonate equivalent = moles CaC03 x 1 �Og / sample mass [g] Conversion to Mg CaC03/1 000 Mg material is made by multiplying the answer by 1 000. The above calculations do not take into consideration the inherent acidity of a sample. Failure to take the materials free-acidity into account may result in the underestimation of the materials neutralising capacity, because a small amount of NaOH will be required to neutralise the free acidity of the material if present. The sample 's inherent acidity can be calculated by running a duplicate sample treated with distilled water rather than acid and then titrated in the same fashion (Equation 7). [Equation 7] Moles NaOH consumed by samples free-acidity = volume NaOH added [mL] x concentration [M] / 1 000 mL = c These equations can be combined to enable allow rapid calculation of the theoretical NP of a sample as follows; Neutralising potential = (A -B) + C x 1 X 1 05 / 2 x sample mass [g] [Mg CaC03 /1 000 Mg of material] Where A = moles of acid originally added to the sample, B = moles of NaOH required to neutralise excess acid and C = moles ofNaOH neutralised by sample acidity. ,-" -, .. .. � ::' ·""21 1 0 0 ::­ D E:ad 459 O s ! ,� __________________________ �i , '--�I '''''I lil'!1 " i: : :1 � Appendix 5(A) EDS spectrum for a pyrite cube such as those visible in Plate 5 .2. The 3 major peaks are associated with the S and Fe constituents of iron pyrite. Very little ' contamination' is evident. F � � .•. . , .. , ! U U S ; -:;: ; .;:- � _. 1 ... � 1 0 0 ::: li ..:: . .:. ,..oj ..... �� .... i .-, I " .,:: I ., -' i 460 Appendix 5(B) EDS spectrum for an 'aggregate grain' such as that seen in Plate 5.6 . The major peaks are associated with AI, Fe, Si, S and possibly Pb. Trace quantities of eu were also identified in the sample. " ':.:' -�; �n-' : . . '" j " . 1 0 ! 1 . .... : � , " -;. ; I - ! '.' - ; .i;' .:;. � , � ! i � • • _ .... :\., 1 0 0 ::: ., -�' I� ._ i _� . ' ,.:. �-J �-R ;: .· � : : .�'!I f-----vr-·----,·------------ I �ll I i L.I I I Iii II , , t A q til •• " .; � . .. ! ; "0. n ... ! ."i 1 • c h 3 1 2= i i 46 1 -. I l) S \ . ._-----j i Appendix 5(C) EDS spectrum for the quartz grain shown in Plate 9. The major peak is that of Si, however, minor quantities of AI, Fe and Sn were also detected in the sample. 1 0 0 :=; 1 ....,.., .. .I -= -1_' _' k" 1 \ 1 0 0 :=; D ea.d. F 462 � I I I i I i I ' i ._. L U :=; 1 1 .... - 1 1 .::. -:" :' I�1 I" � l l 1 1 � .. " , ." . • ... .. .:....:... . r·. 1I::: '.' a _, ...- l i FS= 1 K c h 3:;' 1 = 1 30 c t. ::: i r l r·1 I:. t'1 1 : t'il] 1 I -.�==========���===================== ' " . . . �-: ' ; . . , ' . . Appendix 5(D) This EDS spectrum identifies the presence of K in the tailings probably in the form of mica. The occurrence of a large Fe peak and relatively small S peak indicate that the Fe is probably in the form of oxides or alumino-silicate clays as opposed to sulphides. i .'�" I ',· ! 7""- .-. ' j- '- - i t' ' �-:-1 i : '1 '- or' I ' ._ 1 11 i L-. 6 ,. 220 pI , '\ 1 'I 1-' " : .J .,. il I ! I I i ! ,-. ! U S ! , 463 Appendix 5(E) Low magnification ( 1 80 x) EDS spectrum for the 0- 1 00 mm tailings sample, A significant amount of Ba was identified in 3 EDS taken in the surface horizon, Pb concentrations also tended to be higher relative to the 500-600 mm sample, I : , : ;:: ... ;:. -::: ;::. t ! . / - - - ." 1 0 0 ::- �. , � �ma. l n i 464 r� l:J ; . . -- .: • . - 8... O· -=_. d.- '.1· .:...t • . · • r -----.."'M.!---------- -- -....;: ! , FI' ' i �:l <0 , :,' 1' � :j c;: � I I ' , I Appendix 5(F) Low magnification ( 1 80 x) EDS spectrum for the 500-600 mm tailings sample. The 3 EDS samples showed little or no Ba and unlike the surface sample eu, W and Zn were detected in all 3 low magnification EDS analyses at this depth. Appendix 6(A) Liming Requirement Based on pH Buffering (Massey University Soils Dept. Standard Method) 1 . Duplicate 20g tailing samples were weighed into a set of six 120-mL plastic beakers. 2. The following volumes of water and O.2MNaOH were added. Beaker # Volume of Distilled H2O Volume of Distilled NaOH Added rmL per beaker] Added rmL per beakerl 1 1 0 0 2 8 2 3 6 4 4 4 6 5 2 8 6 0 1 0 465 4. The samples were stirred, covered with Parafilm, and left to incubate at 25°C for 48 hours. 5 . After the incubation period 40 mL of distilled water was added to each of the samples to obtain a 1 :2.5 tailings-to-solution ratio. 6. The solution was stirred and allowed to settle for approximately Yz an hour, and then the pH of the solutions was measured. 7. A graph of pH against mmol ofNaOH added was plotted. 8. From the individual graphs for each sample the theoretical quantity of lime required to increase the pH of the tailings to 6.0 was calculated. 9. Graphs were plotted using Grapher for Windows version 1 .20. A linear best fit curve was plotted to obtain the pH buffering curve data. 466 Appendix 6(B) Acid Generation Estimates Based on Hydrogen Ion Liberation This appendix includes a working example of a method used to compare the relative reactivity of incubated Tui tailings samples in terms of their ability to generate acidity. The method involved the use of pH values to estimate the quantity of hydrogen ion (W) liberated over the course of a 1 'li year lime incubation experiment. The estimated acidity produced by the tailings samples was converted into units of calcium carbonate equivalence [Mg CaC03/1 000 Mg tailings] . General Data • The example outlined below assumes that the hypothetical tailings sample had an initial pH (PHi) of 5 .0 and a final pH (PHf) of 2.0 at the end of the incubation period. • pH measurements were made using a 1 09 to 25 mL ratio of tailings to distilled water as per the method outlined in Chapter 4 (section 4. 1 . 1 . 1 ) • Molecular weight of calcium carbonate (CaC03) == 1 00. 1 g • The concentration of W = 10 iH [moleslL] Working pHi [W] i = = = = = 5 .0 5.0 X 10-5 moles/litrellOg of tailings 5.0 2.0 * 10-2 moles/litrell Og of tailings 2.0 * 10-2 - 5 .0 * 1 0-5 1 .995 * 1 0-2 moles/litrell Og of tailings :. Mole ofW per kg of tailings = 1 .995 * 1 0-2 * 25 mL * 1000 g 1 000 mL * 1 0 g 4.988 * 1 0-2 moleslkg tailings 49.88 moleslMg of tailings 49880 moles/l OOO Mg of tailings As discussed in Chapter 4 the dissociation of 1 .0 mole of CaC03 in water effectively results in the formation of products that can neutralise 2.0 moles of W. Therefore 1 00. l g of CaC03 will neutralise 2.0 moles of W. The acid produced by from the tailings in units of Mg CaC03/1000 Mg tailings is thus calculated as follows; 1 00. 1 g * 49880 moles 2 moles * 1 000* 1 000 2.5 Mg CaC03/1 000 Mg tailings 467 Appendix 7(A) Seed Germination Trial Results Metal A. capillaris A. capillaris A. capillaris F. rubra F. rubra (Egmont) (Goginan) _