Simultaneous stochastic optimisation of mining complexes: integrating waste management and progressive reclamation with encapsulation

Managing environmental performance of waste dump facilities in mining complexes is an integral part of long-term production planning. Sustainable long-term production scheduling solutions are desired to mitigate risk and return the environment to a productive post-mining state. A simultaneous stochastic optimisation framework for long-term production scheduling in mining complexes is developed that integrates waste management and progressive reclamation. The waste dump placement schedule is jointly optimised with the extraction sequence, destination policy, and stockpiling decisions in a single stochastic mathematical programming framework. This includes the timing of progressive reclamation activities in parallel with production to enhance waste dump rehabilitation. Uncertainty related to the production of acid rock drainage is quantified by simulating geochemical properties of waste and managing the blending of uncertain waste properties within the optimisation framework. With respect to the framework for simultaneous stochastic optimisation, contextual bandits are explored to improve the metaheuristic solution approach and solve the corresponding large-scale optimisation model. The framework is tested in a multi-mine copper-gold mining complex leading to improved environmental performance. Risk of acid rock drainage is decreased by 52.5% in the waste dump facilities. Reclamation planning activities for meeting environmental requirements are scheduled prior to closure. The solution approach more effectively improves the objective function with contextual bandits leading to a 24% improvement in the study presented.

Acid mine drainage (AMD), wherein acidic water is generated from pyrite-containing waste rock, can be mitigated by encapsulating pyritic waste rock with cover materials to restrict the inflow of oxygen and water. However, acidic water inevitably forms during the construction of waste rock dumps before applying cover materials. Considering that the presence of waste rock containing carbonate minerals contributes to acid neutralization, a mixture of carbonate minerals and pyritic waste rock can be utilized to reduce AMD generation before the completion of the cover system as a temporary management strategy. This paper examines waste rock management using blending scenarios. Kinetic NAG and column leaching tests were employed to evaluate the blending ratio necessary to prevent acidic water generation. Geochemical analyses were conducted on rock and leachate samples, including pH and temperature measurements, XRD and XRF analyses, and Ion Chromatography. Consequently, the pH and temperature measurement results obtained during the kinetic NAG test are valuable for expressing the balance between acid generation and acid neutralization by the mixture material. Furthermore, the column leaching test demonstrated that the pH of the leachate remained neutral when the acid generation and acid neutralization reactions were well balanced. Blending waste rocks is an effective method for AMD reduction during the construction of waste rock dumps.

Recent developments in modelling and optimization approaches for the production of mineral and energy resources have resulted in new simultaneous stochastic optimization frameworks and related digital technologies. A mining complex is a type of value chain whereby raw materials (minerals) extracted from various mineral deposits are transformed into a set of sellable products, using the available processing streams. The supply of materials extracted from a group of mines represents a major source of uncertainty in mining operations and mineral value chains. The simultaneous stochastic optimization of mining complexes, presented herein, aims to address major limitations of past approaches by modelling and optimizing several interrelated aspects of the mineral value chain in a single model. This single optimization model integrates material extraction from a set of sources along with their uncertainty, the related risk management, blending, stockpiling, non-linear transformations that occur in the available processing streams, the utilization of processing streams, and, finally, the transportation of products to customers. Uncertainty in materials extracted from the related mineral deposits of a mining complex is represented by a group of stochastic simulations. This paper presents a two-stage stochastic mixed integer nonlinear programming formulation for modelling and optimizing a mining complex, along with a metaheuristic-based solver that facilitates the practical optimization of exceptionally large mathematical formulations. The distinct advantages of the approach presented herein are demonstrated through two case studies, where the stochastic framework is compared to past approaches that ignore uncertainty. Results demonstrate major improvements in both meeting forecasted production targets and net present value. Concepts and methods presented in this paper for the simultaneous stochastic optimization for mining complexes may be adopted and applied to the optimization of smart oil fields.

Waste and lithological rocks were subjected to aqueous leaching, acid base accounting (ABA), and net acid generation (NAG) tests, and detailed mineralogical investigations were conducted to predict acid mine drainage (AMD) formation at Turkey’s largest historical copper deposits. The field water chemistry from springs and seeps on the mine site were compared with the static and long term aqueous leaching test results. During the ABA, NAG and long term paste pH tests, ore rich and ore bearing wastes showed a paste pH <4, implying their acid generating nature. The relationship between net neutralization potential and acid producing potential revealed that waste rocks with a low sulfur content were generally low potential sources of AMD. Consistent with the static test results, aqueous leaching tests revealed that greater concentrations of Fe, Cu, Zn, Ni, Pb, Cd, Co and As were released from the wastes rich in pyrite. The test methods all generally agreed that the ore-rich wastes (O, P1C, P1D) were the main sources of AMD and metal contamination in the district.

Acid Rock Drainage (ARD) prediction is complicated by the number and complexity of influencing parameters. The objective of this study was to determine the effect of particle size and mineral liberation on the production of ARD. This information was used to classify waste rocks from an Australian mine site for their potential use as cover material. Samples were crushed and sieved into different particle size fractions, and were subjected to varying static and kinetic tests including paste-pH, Acid-Neutralization-Capacity tests, kinetic Net Acid Generation tests and Humidity Cell tests. The results showed a strong influence of particle size, which can be attributed to the widely different mineral liberation characteristics of acid generating and neutralizing minerals. Whereas standard Acid Base Accounting (ABA) resulted in an uncertain or Non Acid Forming classification of open-pit waste rocks, the application of modified ABA test procedures using different particle size fractions resulted in an uncertain or Potentially Acid Forming classification of the very same material. The study demonstrated that the standard ABA procedure may lead to erroneous classification with potentially costly environmental consequences by overlooking particle size specific mineral liberation effects. Testing different particle size classes proved to be a practicable method to investigate the ARD characteristics of waste rocks and to gain a clearer insight into the consequences of mineral liberation.

Acid mine drainage (AMD) is the environmental issue that generates the greatest public concern regarding the mining industry. Thus, characterization of mine waste rock according to acid generation potential is necessary for mining operations to ensure proper waste rock storage and to avoid future adverse environmental effects. Therefore, this study was conducted to estimate the potential of AMD generation in the largest operating gold mine in Thailand by using acid base accounting and net acid generation tests. Representative samples of six types of waste rock classified by mining geologists for mineral processing and waste dumping were collected for this study: volcanic clastic, porphyritic andesite, andesite, silicified tuff, silicified lapilli tuff, and sheared tuff. Under various conditions, experimental results indicate that only silicified lapilli tuff and shear tuff are potentially acid-forming materials. The results indicate that AMD generation may possibly occur a long time after mine closure due to the lag time of the dissolution of acid-neutralizing sources. Acidic generation from some waste rocks may occur in the future based on environmental conditions, particularly the oxidation of sulphide minerals by the combination of oxygen and water. Therefore, a proper design for waste rock dumping and storage is necessary to reduce the risk of AMD generation in future. It is advisable to install a surface management system to control the overland flow direction away from the waste dump area and tailing storage facility and to install a second water storage pond next to the main storage pond to store the spilled water during storms and the rainy season. A water quality monitoring plan that focuses on disturbed areas such as water storage ponds and mine pits should be put in place.

Waste rock and tailings of a mine generate acid mine drainage due to oxidization during long-term stacking, and this may cause serious harm to the environment. 1-year-old, 5-year-old, 10-year-old and 20-year-old waste rock of Dexing Copper Mine are chosen in the experiment for acid generation research. The static evaluation result shows that the 4 types of waste rock have the acid generating potential, and 3 types (except for 1-year-old waste rock) have generated acid. The 9-month dynamic evaluation result indicates: the pH values of waste rock of 4 different ages show remarkable negative correlation, the detected ions are mainly of fluorides, Cu 2+ , Zn 2+ , SO4 2- and Fe 3+ . 1-year-old waste rock doesn't generate acid during the entire experiment, the acid generation of 5-year-old and 10-year-old waste rock is stable, but that of 20-year-old waste rock is slow. Thus the acid generation process of waste rock can be inferred as: no acid generation - high-rate acid generation - stable acid generation - lower-rate acid generation.

Chemical precipitation within pyritic waste rock

Rare earth element recovery in hard-rock acid mine drainage and mine waste: A case study in Idaho Springs, Colorado

The Mount Morgan Mine, in central Queensland, was in operation for over a hundred years and generated 134 Mt of waste rock and tailings. The last ten years of activity at the site involved re-treatment of historic tailings. The Mount Morgan mine site is a significant source of Acid Rock Drainage (ARD) in the Dee River catchment. The State of Queensland, Department of Natural Resources and Mines (NR&M) manages the site, and in January 2000 NR&M developed the Rehabilitation Plan for the Mount Morgan mine site, Central Queensland (Unger, 2003). The objectives of the plan are to: improve water quality downstream; avoid managing ARD interception and treatment indefinitely; manage the site in accordance with its significant mining heritage; and develop and apply best-practice rehabilitation and management at this site. A seepage interception and pump-back system is currently in place, and the amount of ARD entering the groundwater system and ultimately reaching the Dee River is being quantified. A number of projects are in progress, or have been completed, towards achieving these objectives. One of these projects is the development of a cover system for waste rock material at the mine site. This project involved the construction of two large-scale cover system field trials in 2003 on acid-generating waste rock, as well as installation of field performance monitoring systems (site-specific meteorological conditions, net percolation, interflow, moisture content, matric suction, temperature, and surface runoff). In addition, a monitoring system was installed at a “natural” site in order to compare performance of an undisturbed area, in terms of storage and release of moisture, to the cover system field trials. The two cover systems are designed to limit the infiltration of meteoric water to the underlying waste rock as a means of controlling long-term acidic drainage from the waste rock dump. Test Plot 1 is referred to as a “water-shedding” cover system with a nominal 1.0 m thick layer consisting of a compacted clay layer overlain by a layer of non-compacted growth medium (weathered granodiorite), while Test Plot 2 which is referred to as a “moisture store-and-release” cover system consists of a nominal 2.0 m thick layer of growth medium. This paper will briefly describe the design and construction of the field trials, as well as the field performance monitoring system. Discussion and field performance monitoring data will be provided on the difference in performance between the two cover system field trials. Emphasis will be placed on the impact on performance of the two cover systems as a result of runoff volumes. In particular, it has been demonstrated that runoff volumes negated the intended performance of the respective water-shedding and moisture store-and-release cover design. A complete water balance for each field trial will be presented illustrating the difference in performance between the cover system field trials. Mine Closure 2006 ― Andy Fourie and Mark Tibbett (eds) © 2006 Australian Centre for Geomechanics, Perth, ISBN 0-9756756-6-4 Mine Closure 2006, Perth, Australia 427

Blending of potentially acid generating (PAG) waste rock with non-PAG waste rock to create a rock mixture which performs as non-PAG is a possible approach to permanent prevention of acid rock drainage (ARD) for PAG waste rock. In 2012, a field weathering study using 300 kg samples was implemented at Teck Coal's Quintette Project located in northeastern British Columbia, Canada to test the prevention of acid generation in the PAG waste rock by dissolved carbonate leached from overlying non-PAG waste rock and direct neutralization of acidic water from PAG waste rock by contact with non-PAG waste rock. After eight years of monitoring the experiments, the layered non-PAG on PAG barrels provided proof-of-concept that as the thickness of the PAG layer increases relative to the thickness of the non-PAG layers, acidic waters are more likely to be produced. The PAG on non-PAG layering has resulted in non-acidic water and no indications of metal leaching despite accelerated oxidation in the PAG layer shown by sulfate loadings. The study has demonstrated that the scale of heterogeneity of PAG and non-PAG materials is a critical consideration for providing certainty that rock blends designed to be non-PAG will perform as non-PAG in perpetuity. This is contrary to the standard paradigm in which an excess of acid-consuming minerals is often considered sufficient alone to ensure ARD is not produced.

ABSTRACTSimultaneous stochastic optimisation manages risk and capitalises on the unique interactions that occur in a mining complex, where materials are transferred between mines, processors, stockpiles, and waste facilities to achieve a marketable product. Typically, when optimising the production schedule, the primary focus is to deliver valuable products to the market. However, this tends to ignore the environmental and economic impact of simplifying waste management requirements, including the storage and disposal of waste material. Stricter regulations and engineering requirements are transforming past mining practices to develop more sustainable operations. These transformations increase the financial cost of waste management and identify the requirement to integrate waste management into the production schedule. Additionally, misrepresenting the material uncertainty and variability associated with the amount of waste produced can impact, both, the stakeholders and the profitability of a mining complex. In this case study, a simultaneous stochastic optimisation approach is applied in a gold mining complex that integrates waste management into the long-term production schedule. The resulting schedule leads to a 6% increase in the net present value when compared to a conventional approach, while minimising the likelihood of deviating from production targets and ensuring permit constraints are satisfied.

Since 1995, cover system designs and waste rock management plans have been developed for the waste rock dumps (WRDs) at a mine in the Pilbara region of Western Australia. A major component has been investigation of acid metalliferous drainage (AMD) from the WRDs and development of management strategies to limit impacts to receiving environments. The site’s management strategies have focused on cover system design as well as landform and watershed designs. WRD landform design, and associated cover system design, has evolved several times since 1995. This has occurred in response to evolution of the conceptual models that have been developed at the site with respect to the mechanisms, and the controls on those mechanisms, resulting in AMD. The information gathered over a period of fifteen years led to changes in the conceptual model. Prior to 1995, AMD was thought to be, in general, a ‘non‐issue’ in arid environments similar to the Pilbara region. However, in 1995, the first appearance of AMD at the site occurred in response to a major cyclone. The first conceptual model for cover system performance was developed in 1995 with the main aim of limiting net percolation (NP) to the underlying WRD. This design was a ‘simple’ cover system that used the ‘moisture store‐and‐release’ concept, relying on evaporation to cycle moisture back to the atmosphere and a hummocky surface to limit runoff and runoff‐induced erosion. At that time, no vegetation was included in the conceptual model as it was considered that bare surface evaporation would be sufficient. The conceptual model for cover system performance has evolved since 1996. The model still utilises the ‘moisture store‐and‐release’ concept; however, incorporating cover construction QA/QC, optimising transpiration through sustainable vegetation, and managing runoff through catchment design have been refined and incorporated into full‐scale WRDs. Since 1996, several research test piles, cover system field trials, and vegetation test plots have been constructed and monitored to measure their effectiveness in meeting AMD management objectives at the site. Fifteen years of data from the large cover system field trials constructed in 1996 and 2002 have resulted in key findings influencing long‐term closure plans. NP into WRDs has been limited to 5% of average annual rainfall over 13 years of monitoring for a 2 m thick run‐of‐mine (ROM) cover system. Where a 4 m thick cover was constructed, NP has been limited to substantially less than 5% over 13 years. It has also been found that segregation of the ROM cover material, as well as selective handling of potentially acid forming wastes, strongly influences AMD management. Short, medium, and long‐term strategies were developed in 1996 to manage AMD at the site, identify mechanisms (and their controls) for AMD, and develop waste management practices to mitigate AMD. Short‐term strategies that were introduced included containing runoff, identifying research goals, characterising the waste, and initiating management and rehabilitation plans. Research into strategic selective handling and placement of acid forming and non‐acid forming overburden has also been monitored and evaluated. Long‐term closure strategies are being implemented on site based on confirmation of the effectiveness of the field research and rehabilitation strategies.

Geochemical characterization of coal and waste rocks from a high sulfur bearing coalfield, India: Implication for acid and metal generation

Complex electrical measurements of waste rock during acid mine drainage generation and release: Kinetic column tests