Elsevier

Ecological Engineering

Volume 64, March 2014, Pages 66-76
Ecological Engineering

Amelioration of soil chemical constraints for revegetation on gold oxide processing residues

https://doi.org/10.1016/j.ecoleng.2013.12.039Get rights and content

Abstract

Gold oxide processing residues are highly alkaline, sodic and saline and initially require amelioration for plant establishment. The aim of this study was to determine the efficacy of soil covers and residue ameliorants in alleviating chemical constraints for revegetation. The effects of residue covers (10 cm topsoil plus 0, 15 or 30 cm gravelly sub-soil referred to as gravel), combined with gypsum (30 and 60 t ha−1), and compost (0 and 50 m3 ha−1) on chemical properties (pH, electrical conductivity (EC), exchangeable Na percentage (ESP), cation exchange capacity (CEC), total N, extractable N, P, K, S, cations and micronutrients) of gold oxide processing residue were studied in a large-scale field trial at Boddington, Western Australia in a Mediterranean climate. After the first season's rainfall, the EC1:5 decreased substantially (from 4.5–5.0 to about 1.0 dS m−1) across all treatments in the 0–10 cm residue layer (RES1), while topsoil and gravel materials which had low initial salinity (0.04–0.07 dS m−1) increased their EC1:5 to 0.40–0.45 dS m−1 in the 10 cm above the residue interface in the 2nd hot-dry summer season. pH of both RES1 and the 10–20 cm layer of residue (RES2) dropped from 9.0 to 8.0 over the trial period but topsoil pH remained at 5.2–5.5, regardless of treatments. Exchangeable Na percentage in RES1 and RES2 decreased, but increased from 17 to 70% in topsoil placed directly on the residue surface, compared with no increase in topsoil where there was a gravel layer separating topsoil from residue. Fertiliser added to topsoil raised extractable nutrients to levels adequate for plant growth. Nutrient elements did not differ with gypsum rate and compost amendment except for S, on account of the gypsum applied, and extractable P and Mn, which were elevated by compost. These findings indicate the efficacy of gravel plus fertilised topsoil cover on residue treated with surface broadcast gypsum to alleviate the adverse properties of alkaline, saline, sodic gold oxide processing residue as a medium for plant establishment and early growth.

Introduction

Ore processing residues are often inhospitable to plant growth, due to alkalinity or acidity, saline–sodic conditions, elevated contents of metals, low nutrient availability, low plant available water, negligible soil microbial activity, coarse or excessively fine texture and poor structure (Hossner and Shahandeh, 2002). Phytostabilisation or revegetation is often an effective means of remediating residues but field investigations are needed to understand the key chemical constraints of the residue (Mendez and Maier, 2008). Typical constraints in gold processing residues were reported by Hossner and Shahandeh (2002). However, the processing residue on most gold mine residue disposal areas (RDA) is typically sourced from crushed rock. By contrast, during the previous operating phase at the Boddington Gold Mine (now Newmont Boddington Gold), processing residues were generated from regolith ore comprising mainly non-swelling kaolinitic clays (Rayner et al., 1996).

Revegetation on gold oxide processing residue requires the reconstruction of a soil profile that will support both establishment and longer term survival of vegetation. Initially, processing residue consists of 60–70% saline liquor and 30–40% milled ore. Following consolidation and drying (to about 70% solids), the residue lacks many of the characteristics of soil, as it dries to form massive apedal blocks with low hydraulic conductivity, separated by relatively wide cracks that are open to approximately 2 m below the residue surface (Ho et al., 1999). It is also characterised by chemical properties that are not conducive to soil biological activity or plant growth as a result of contact with the saline process water, and treatment with sodium hydroxide (NaOH) and sodium cyanide (NaCN) during processing (Ho et al., 1999).

Salinity of the pore water of gold oxide processing residue at Boddington could be as much as 10,000 mg of total dissolved salts L−1 (TDS) (equivalent to electrical conductivity ECe3 8.7–17.0 dS m−1 or EC1:5 of 1.45–2.83 dS m−1) (Rayner et al., 1996). A higher EC1:5 in the residue surface (>3 dS m−1) is expected due to surface evaporation and subsequent capillary rise of pore water transporting salts to the surface. The concentrated salt layer may restrict seed germination and root growth if not removed or dissipated during rehabilitation.

Below the surface layer, the recorded EC1:5 is still likely to be at levels that will limit growth of most species (between 1.4 and 2.8 dS m−1: see Ho et al., 1999). For agricultural crops and pasture, the residue would be regarded as very saline to highly saline and only very salt tolerant crops would be likely to yield satisfactorily (Shaw, 1999). According to Moore (1998), an EC1:5 of 1.4–3.5 dS m−1 severely limits the growth of most plants.

Processing of ore uses NaOH to extract gold, resulting in a residue pH of approximately 9–9.5. Generally, a pH above 8.5 is considered to severely limit the growth of most plants (Purdie, 1998) due directly to high pH or to induced nutrient deficiencies by immobilising nutrients such as zinc (Zn), manganese (Mn) and iron (Fe). Gypsum can reduce soil pH by removing bicarbonates and carbonates from soil solution as insoluble calcium carbonate (Barrow, 1982). Earlier research with gold oxide processing residue indicated that gypsum at 30 t ha−1, when well incorporated into residue (to 30 cm depth), increased plant growth (Ho et al., 1999).

Sodium accumulates as the dominant cation in processing water from the brackish source water, and is increased by direct addition of NaOH and NaCN in the refining process. The exchangeable sodium percentage (ESP) of the residue has been previously recorded at 62% (Ho et al., 1999), while an ESP of 5% is sufficient to cause the dispersion of clay particles (Rengasamy and Churchman, 1999). Hence the high concentrations of exchangeable Na contribute to the lack of pedality in the residue and poor infiltration of water. Problems arising from these effects can include waterlogging, low plant-available water storage (Shaw et al., 1994), poor aeration and hence poor root development (Naidu and Rengasamy, 1993). Gypsum is commonly added to soils to reduce sodicity, as it is a source of Ca2+ ions which replace Na+ ions on soil exchange sites, reducing the ESP of the residue (Qadir et al., 2001, Walker and Bernal, 2008, Lakhdar et al., 2008).

Residues may have levels of nutrients that cause deficiencies or toxicities in plants. Nutrient disorders potentially affecting revegetation of residue include: low levels of phosphorus (P), although jarrah forest vegetation has adapted to these conditions (Handreck, 1997); zinc, manganese (Mn) and iron deficiency (Fe) due to high pH causing elements to transform to unavailable forms; and elevated copper (Cu) due to Cu mineralisation together with gold (Rayner et al., 1996) that may cause Cu toxicity reactions in susceptible plants (Jones et al., 2010).

The adverse properties of residue mean that a cover of topsoil, sub-soil or other benign substrate may improve plant establishment and growth on residue storage areas. However, topsoil and sub-soil may be in short supply at the sites of residue disposal, and in any case generally have been in long term storage stockpiles which is not optimal for plant growth (Carrick and Krüger, 2007). A risk from placement of topsoil or sub-soil over residue is that over time migration of salts from residue into the cover compromises its suitability for plant growth unless the cover forms an effective capillary break for saline pore water in residue. Hence there is a need to determine the optimum thickness of cover and its properties over time following placement. Considering the value and availability of topsoil or gravel covering, the reconstructed soil profile was predicted at Boddington to consist of a 10–40 cm layer of topsoil and/or sub-soil to act as a seed bed, rooting medium and capillary break, overlying residue that had been solar dried for 2–3 years after deposition and treated with 30–60 t of gypsum ha−1 incorporated to 30 cm depth (Ho et al., 1999).

Application of gypsum to the residue to increase both flocculation of clay particles and hydraulic conductivity should encourage leaching of salts from the residue surface enhancing the prospects for plant growth. The effectiveness of gypsum when incorporated into residue has been previously demonstrated by improved plant growth (Ho et al., 1999) but the efficacy of surface broadcast application has not. Broadcasting of compost on the residue surface was also investigated in this trial as an additional organic form of nutrient supply.

In this study, we investigated the effects of gravel thickness, two rates of gypsum application, and the addition of organic material on amelioration of adverse chemical properties of the underlying residue over time. The aim of the study was to determine the effect of soil cover thickness (comprising topsoil alone with or without underlying gravel) and broadcasting of gypsum or compost on the residue on the edaphic properties of the reconstructed soil profile, and its likely suitability for plant growth and ecosystem function. A companion paper examines the consequences of these properties for vegetation establishment, vigour and diversity (Ni et al., 2014).

Section snippets

Site location

The study was conducted at Newmont Boddington Gold (NBG) mine near Boddington, 125 km south-east of Perth, Western Australia (Rayner et al., 1996, McGrath et al., 2004). The climate of the southwest of Western Australia is typically Mediterranean, with cool wet winters (June–August) and hot dry summers (December–February). Mean annual rainfall at the nearby Marradong station (32.86° S 116.45° E) is 724 mm with 88% falling between April and October while mean maximum/minimum temperatures in January

Salinity

Prior to the application of treatments, RES1 was characterised by high salinity (mean EC1:5 of 4.5 dS m−1, ranging from 1.4 to 6.0 dS m−1). After gypsum and topsoil/gravel application in March 1999 and winter rainfall (Fig. 1), EC1:5 decreased substantially across all treatments in this surface layer of the residue to 0.90 dS m−1 (ranging from 0.37 dS m−1 under 30 cm of gravel to 1.6 dS m−1 under topsoil only-no gravel) in October 1999 (Fig. 3a).

In RES1, EC1:5 continued to decrease under 15 and 30 cm of

Discussion

The initial gold oxide processing residue had levels of salinity, pH, sodicity and nutrient availability potentially limiting to vegetation establishment and subsequent growth. The effectiveness of gypsum and compost broadcast to the surface of the residue, various depths of topsoil and gravel covers and fertilizers broadcast on topsoil in overcoming these constraints are discussed below.

Conclusions

Prior to application of residue cover and ameliorant treatments, the gold oxide processing residue surface was extremely saline (EC1:5 4.5–5 dS m−1), alkaline (pH 10) and sodic (ESP 92%). After gypsum application on the residue surface and the first seasonal rainfall, the EC1:5 and pH in residue decreased substantially regardless of cover thickness. Thirty tonnes of gypsum per ha contributed to a decrease in residue surface pH, salinity, and sodicity, but 60 t gypsum ha−1 had no additional effect.

Acknowledgements

Boddington Gold Mine (now Newmont Boddington Gold) and Alcoa of Australia Ltd provided logistic, in-kind and cash support for the conduct of the study reported here. The research of a number of undergraduate students who worked on the project under the supervision of the authors is acknowledged: K. Barnes (2000), A. Brion (2000), K. Briggs (2000).

References (41)

  • R.W. Bell et al.

    Rehabilitation of Residue Storage Areas at Boddington Gold Mine and Hedges Gold Mine – End of Project Report

    (1999)
  • Bureau of Meteorology

    Rainfall and Temperature. Comparison and Summary Statistics for Wandering 1999–2000

    (2001)
  • C.R. Chen et al.

    Behaviour and dynamics of diammonium phosphate in bauxite processing residue sand in Western Australia – I, NH3 volatilisation and residual nitrogen availability

    Environ. Sci. Pollut. Res.

    (2009)
  • R. Courtney et al.

    Growth of Trifolium pratense in red mud amended with process sand, gypsum and thermally dried sewage sludge

    Int. J. Min. Reclam. Environ.

    (2003)
  • C.J.P. Gourley

    Potassium

  • I. Hammond

    Nutrient Cycling in Revegetated Bauxite Pits at Quindanning

    School of Environmental Science

    (1996)
  • A. Hanay et al.

    Reclamation of saline–sodic soils with gypsum and MSW compost

    Compost Sci. Utiliz.

    (2004)
  • K.A. Handreck

    Phosphorus requirements of Australian native plants

    Aust. J. Soil Res.

    (1997)
  • K.A. Handreck et al.

    Growing Media for Ornamental Plants and Turf

    (2002)
  • G.E. Ho et al.

    Direct revegetation of salt-affected gold ore refining residue: technology evaluation

  • Cited by (0)

    A preliminary account of the results was presented in McGrath et al. (2003).

    1

    Present address: Eco Logical, West Perth, Australia.

    2

    Formerly.

    View full text