Adequacy of Sampling and Volume Estimation for Pre-mining Evaluation of Potentially Acid Forming Waste: Statistical and Geostatistical Methods

Danny Kentwell, Andrew Garvie and John Chapman
Thursday, July 12, 2012
First presented: 
Life-of-Mine Conference, July 2012
To assess the risk of acid and metalliferous mine drainage (AMD), most new mining projects require an initial assessment of the locations and volumes of potentially acid forming (PAF) waste that is likely to be moved and dumped in waste stockpiles. This requires sampling of the waste rock zones and submitting the samples for a series of tests. Key to the outcome of the waste characterisation process is obtaining sufficient samples that are representative of the various rock types, both in number and spatial distribution. Various regulatory bodies throughout Australia and the rest of the world are currently struggling with how to define minimum amounts of sampling to achieve realistic estimates without imposing excessive costs and time constraints on a project.

This paper addresses two aspects of the process:
1. What constitutes a ‘sufficient’ number of samples
2. How to use those samples in determining volumes of likely PAF material
Geostatistical methods of characterising ore bodies with large numbers of samples are routinely used in the mining industry by geologists and geostatisticians to estimate and classify mineral resources. In theory, the same methods can be used to estimate PAF waste locations and volumes. In practice, the number and spatial distribution of dedicated PAF waste samples at feasibility stage is typically so small that it is almost impossible to estimate the spatial distribution of PAF material with any degree of confidence. Consequently, the common approach is to characterise a few samples from a lithological unit and then assume that they represent the unit as a whole. This approach, however, can be fraught with risk.

This paper describes some of the findings of SRK’s evaluation of several coal projects with respect to PAF waste evaluation and highlights some common misconceptions.

Feature Author

Danny Kentwell

Danny Kentwell is a geostatistician with a background in geological modelling, mine planning and surveying. He has 25 years’ international experience with varied commodities including gold, copper, mineral sands, iron ore, nickel laterites, nickel sulphides and phosphate. Danny’s skills cover, 3D modelling, Resource estimation, open pit optimisation scheduling and design. His geostatistical expertise includes standard and recoverable resource estimation techniques such as uniform conditioning, indicator kriging and conditional simulation as well as multivariate estimation and simulation. As a geostatistician and engineer, he has an excellent understanding of the advantages and limitations of different resource estimation techniques, their resulting block grade, tonnage and value curves and their use in mine planning. Danny also has experience in applying geostatistical techniques to waste characterisation and determination of sampling adequacy from very small data sets.

Principal Consultant (Geostatistics)
MSc (Mathematics & Planning; Geostatistics), FAusIMM
SRK Melbourne
Andrew Garvie

Andrew Garvie has more than 24 years’ experience providing scientific and technical assessments in acid and metalliferous drainage (AMD) and heap leach oxidation.  More recently he has undertaken assessments of self-heating and the potential for spontaneous combustion of coal wastes and carbonaceous black shales associated with sulfide minerals.  Studies have included the quantification of physical processes that support the oxidation of mine wastes and heap leach piles by measurement and predictive modelling.  He has assessed strategies used at mines to control oxygen supply and water flux into dumps and heap leach piles using the same methods.  Andrew’s experience includes use of geostatistics to assess the adequacy of sampling, geochemical characterisation to examine the potential of mine wastes to produce AMD, and assessment of contributors to pit lake water quality, including wall rock oxidation and in-pit waste rock disposal.  He has applied his understanding of the above processes to the development of conceptual waste landform closure strategies for the control of AMD production and spontaneous combustion.

Principal Consultant (Geoenvironmental)
PhD (Physics), MAusIMM
SRK Sydney
John Chapman

John Chapman has more than 26 years of relevant experience and has developed multi-disciplinary skills relating to mining environmental impacts assessment and environmental controls.  He is a recognised expert in ARD assessment and prediction, mine waste characterisation and management, and mine closure with recent project experience in Australia, South East Asia, Canada, USA and Europe.  John has facilitated enterprise-wide risk assessment workshops, undertaken due diligence and environmental risk assessments for a wide range of mining projects.  Recent projects include detailed geochemical assessment of the tailings properties for the Olympic Dam Project in SA, design and development of the geochemical characterisation program for the Cannington Life Extension project in QLD as well as for the Yeelirrie Project in WA.

Principal Consultant (Geochemistry and Environment)
MSc (Chemical Engineering), P Eng (British Columbia), P.Eng (Yukon Territory)
SRK Brisbane
SRK Turkey