Chapter 5b

DRAFT DOCUMENT

5.0 Prediction

5.1 Introduction

5.2 Objectives of Prediction Program

5.3 The Acid Rock Drainage Prediction Approach

5.3.1 Acid Rock Drainage/Metal Leaching Characterization

5.3.2 Description of Phases

5.3.3 Water Quality Prediction

5.4 Prediction Tools

5.4.1 Introduction

5.4.2 Geological and Lithological Investigations

5.4.3 Hydrogeological/Hydrological Investigations

5.4.4 Geochemical Testing Methods

5.4.5 Data Management

5.4.6 Quality Assurance/Quality Control

5.4.7 Screening and Evaluation Criteria

5.4.8 Reporting

5.5 Modeling of Acid Rock Drainage, Neutral Mine Drainage, and Saline Drainage for Characterization and Remediation

5.5.1 Introduction

5.5.2 Geochemical Modeling

5.5.3 Hydrological Modeling

5.5.4 Hydrogeological Modeling

5.5.5 Gas Transport Modeling

5.5.6 Statistical Evaluation

5.6 Conclusions

5.7 References

List of Tables

List of Figures

First Page: Sections 5.1, 5.2, and 5.3

This is the Second Page: Section 5.4 Prediction Tools

Third Page: Sections 5.5, 5.6, and 5.7, Lists of Tables and Figures

Top of this page

5.4 Prediction Tools

5.4.1 Introduction

This Section 5.3 describes the main methods of estimating the environmental water-quality consequences of mineral extraction and processing and how these tools could be used to aid in remediation planning and remedial action. These tools build on the approaches described in Chapter 4.

The primary prediction tools discussed in this chapter include the following:

• Geological and lithological investigations
• Hydrogeological investigations
• Geochemical testing methods:
  • Laboratory static and short-term methods
  • Laboratory kinetic methods
  • Field methods
• Modelling

Top of this page

5.4.2 Geological and Lithological Investigations

Mineral deposits are categorized according to their temperature of origin, their mineralogy, their lithology, and their structure. These categorizations are the basis for the development of geo-environmental models described in Chapter 2. A thorough understanding of the mineral deposit is critical to the characterization of mine wastes and geologic materials and the prediction of mine drainage quality. This information is typically available from the project geologist. Therefore, the characterization and prediction programs often begin with assembly of geological reports and interviews with the project geologists.

The elements likely to be of concern in water-quality assessments have a source in the rock and minerals that are exposed to weathering because of mining activities. Qualitative predictions on what those elements are can be gained from the rock type, its type and degree of alteration (e.g., hydrothermal, weathering, metasomatic), and the structural controls, including those that affect permeability and surface and groundwater flow. Examples of important geological characteristics that can affect the drainage quality, and hence the characterization program, include the following:

• The presence of a pyrite halo around the mineralized zone
• The role of alteration (e.g., potassic vs. propylitic vs. quartz-sericite-pyrite alteration in porphyry copper deposits) in the presence and distribution of sulphide and carbonate minerals
• Vein vs. disseminated deposit
• The presence and role of faults in displacing mineralized and nonmineralized zones and as conduits for water
• Depth of weathering (e.g., supergene vs. hypogene alteration)
• Sedimentary/stratigraphic sequence of coal deposits

These factors will ultimately determine the chemical composition of the mine drainage source material, which is an important step toward predicting the chemical composition of the mine drainage. An example of geological information that can be gathered by mine geologists during their exploration programs, and is relevant to ARD prediction, is presented as Table 5-2.

Table 5-2

Geologists Observations and Logging of Core for ARD Analysis

Important data relevant to the prediction of ARD can be gathered during the core logging process. Much of this information is already collected by or can he obtained by interviewing exploration geologists. The following summarizes work recommended:

Quantitative Data:

• Visual sulphide content (primarily pyrite) with an estimate of accuracy
• Visual carbonate content with an estimate of accuracy

Semi-Quantitative Data:

• Mineralogy, grain size, mode of occurrence of sulphides
• Mineralogy, grain size, mode of occurrence of carbonates
• "fin." reaction of carbonates (strong, weak, none - powdered and unpowdered)
• extent of oxidation if any of rocks
• presence of gypsum, barite, graphite or siderite
• RQD or other tests of rock competence
• Limit of oxidation and supergene zones
• Presence of water (depth to water table)
• Rock hardness/competence

Qualitative Data:

• Presence of secondary sulphate minerals and identification where possible
• Weathering or slaking potential (unusual observations such as rapid oxidation or weathering) in core as recovered or after storage
• Potential for breakage along fracture planes and for preferential exposure of sulphides and/or carbonates
• Presence of coating on sulphides and carbonates
• Potential problems in compositing samples for ARD testing (e.g., core loss, concentration of holes near deposit versus waste, differences in specific gravity, biasing by sulphide/carbonate stringers etc.)
• Observations at outcrops of deposit (sulphide/carbonate content, extent of weathering, staining, coatings etc.)
• Presence of acidic/stained streams or seeps draining the deposit

Quantitative data should be compiled for each drill interval and entered into a geologists log. Semi-quantitative information should be collected periodically through the core when significant changes are noted and could be entered into the "comments" section of log records. Qualitative information relates to unusual conditions that may be encountered while logging or storage of the .samples and could be described in a covering memo from the exploration geologist.

Geology staff should also advise environmental staff and ARD consultants of any samples submitted for whole rock, metal scans, mineralogical or petrographic analysis as data are often also relevant to ARD/ML prediction.

Top of this page

5.4.3 Hydrogeological/Hydrological Investigations

Contaminants in surface water and groundwater are driven by hydrologic and geochemical processes. The conceptual site model (as discussed in Chapter 4) of the hydrologic system includes recharge (precipitation, snowmelt, less evapotranspiration or infiltration), flow path, and discharge (springs, abstraction boreholes, seeps, portal flow, base flow to a river or stream). These water fluxes should be estimated (flux-reservoir diagram) and pump tests are usually needed to determine the geohydrological characteristics of aquifer material. Often a potentiometric surface for underground workings, waste piles, and open pit or other excavations needs to be estimated to determine the current or future potential conditions for water flow and changes in direction of that flow. Determining the groundwater table in fractured rock terrain with or without mine voids (i.e., an open pit or underground mine) can be challenging but very useful, even in a crude form.

Top of this page

5.4.4 Geochemical Testing Methods

5.4.4.1 Introduction to Geochemical Characterization Program

This Section 5.4.4 describes the geochemical testing methods and how the test results can be used for prediction of ARD risk (or potential) and mine water quality. Other possible outcomes of a geochemical testing program include identifying materials suitable for construction uses, for plant growing horizons, and options for management and mitigation.

This section represents a high-level overview of available test methods rather than a detailed account of individual procedures, and focuses on the interpretive and predictive value resulting from geochemical tests. Table 5-1 provides a summary description of various test methods used globally and brief discussions of advantages and limitations of the test methods.

Figure 5-5 (Maest and Kuipers, 2005) schematically presents the components of a typical geochemical characterization program aimed at developing water quality predictions and the general sequence in which these components should be conducted. This flowchart in Figure 5-5 provides more detail on the Phase 1 and Phase 2 testing programs illustrated in Figure 5-1.

Figure 5-5: Schematic Illustration of Geochemical Characterization Program
(modified from Maest and Kuipers, 2005)

Phase 1 consists of a screening-level program, while Phase 2 is more detailed. In some cases, a Phase 1 program may be sufficient for mine water and waste management, whereas in more complex settings, a Phase 2 program is generally required. When a Phase 2 program is required, the results from the Phase 1 program are used to identify samples for kinetic testing or additional static testing, such as those identified in Figure 5-1 and Figure 5-5.

Therefore, not all components of the geochemical testing program may be necessary depending on site-specific characteristics and prediction needs. Individual test methods are described in more detail in the Sections 5.3.4.2 through 5.3.4.8, and are summarized in Table 5-1. Not all test methods presented in the table are appropriate for evaluation of mine wastes, even though they occasionally are requested by regulatory authorities. Such methods include the Toxicity Characteristic Leaching Procedure (TCLP) and Waste Extraction Test (WET), as explained in more detail in Table 5-1.

The geochemical characterization program starts with bench-scale testing, which generally involves whole rock analysis to determine chemical composition. In addition, mineralogical examination, evaluation of acid generation potential, and evaluation of metal leachability are used to determne the ARD/ML Potential. Detection limits in tests must be low enough to measure contaminants at potential concern levels. Depending on the complexity of the geology and variation in ARD potential, the results from the acid generation testing might be combined to develop a 3-dimensional representation of the quantity and geochemical characteristics of ore and waste rock. The information from the whole rock analysis is used to identify categories of rock in support of development of a waste management plan, which aims to handle mining wastes in such as manner as to prevent or minimize environmental impacts (see Chapters 6 and 9).

The next important step in the geochemical characterization program is kinetic testing, which can take the form of laboratory testing, field testing, or both laboratory and field testing, supplemented by on-site water quality monitoring. All materials involved in the kinetic testing should undergo a comprehensive characterization before the test begins, including surface area, particle size distribution, mineralogy, chemical composition, acid neutralization potential, and acid generation potential. At the completion of kinetic testing, the interpretive value of the kinetic testing program is greatly enhanced by repeating the determination of mineralogy, chemical composition, and acid generation potential.

In combination with water, and sometimes oxygen flux calculations, the results from the geochemical characterization programs are used to generate predictions regarding short-term and long-term acid generation potential, leachate quality, and loadings from individual waste type units. These predictions can be extrapolated to full-size mine facilities by incorporating a site-specific water balance based on information on hydrology, hydrogeology and climate, and a block model. Use of scaling factors may be required to account for differences in mass, surface area, rock to water ratio and temperature between testing arrangements, and mine facilities. The resulting water quality estimates can be used as inputs to geochemical models to account for geochemical processes that may affect dissolved concentrations such as mineral precipitation and dilution, sorption, and interaction with atmospheric gases. Ultimately, the findings of the geochemical characterization program contribute to development of mine waste and water management plans.

Any water quality prediction program needs to be customized for a particular situation and problem. Depending on the mine phase, commodity, climate, or mine facility, all or a subset of geochemical characterization tests may be required for the prediction effort and, although not indicated in Figure 5-5, multiple iterations may be required. Also, the issue of water transport might outweigh the geochemical testing as in very arid or arctic conditions with limited or infrequent generation of mine discharges. In that case, the focus of the program might be on determining the site hydrology, the site hydrogeology, or the hydraulics of the mine facility of interest rather than the range of geochemical characteristics. However, the ARD/ML potential of material that will comprise the upper horizon needs to be determined because of its importance for reclamation.

In general, the earlier in the life of a mine, the greater the reliance on use of laboratory tests for water quality prediction. As the mine matures, use of direct field measurements of material geochemistry and from water quality monitoring becomes feasible and is advocated. Accordingly, the comprehensive characterization program presented in Figure 5-5 is most appropriate for proposed operations, while characterization at inactive or orphaned mines would instead focus on observations regarding existing site water and soil quality.

5.4.4.2 Summary of Testing Requirements

In summary, the evaluation of mine waste ARD/ML potential and prediction of resulting water quality requires an understanding of the following characteristics of the mining wastes and geologic materials:

• Physical characteristics
• Chemical characteristics
• Mineralogical characteristics
• Acid neutralization potential
• Acid generation potential
• Leaching potential

For ease of presentation in this GARD Guide, tests aimed at determining acid generation potential and leaching potential are categorized as follows:

• Laboratory static and short-term methods
• Laboratory kinetic methods
• Field methods

Sections 5.3.4.3 through 5.3.4.8 present a brief overview of the components of a comprehensive geochemical characterization program and their significance for mine water quality prediction. Useful references related to static and kinetic testing methods and their interpretation include AMIRA (2002), BCAMDTF (1989), Jambor (2003), Lapakko (2003), Maest and Kuipers (2005), Mills (1999), Morin and Hutt (1997), Price (1997), USEPA (2003), and White et al. (1999).

5.4.4.3 Physical Characteristics

The physical characteristic of most significance for water quality prediction is the particle size. Particle size distributions impact both mineral reaction rates and reaction duration by affecting the reactive surface area, the distances between potentially reactive particles, and the porosity and permeability of a solid. Porosity and permeability of a solid are particularly important with regard to movement and transport of air, water, and reaction products from weathering reactions.

The particle size distribution should be measured before any testing, both for laboratory and field-scale tests. To enable scale-up of test results, estimates of particle size distribution in mine facilities, such as waste rock repositories and heap leaches, are also required. These can be determined from direct measurement or calculated from the blasting plan. The “reactive” surface area of a material (i.e., that portion of the total surface that is actively available for chemical reaction) both in the laboratory and in the field may be significantly smaller than the surface area measured by standard techniques.

Permeability, specific gravity, and porosity should be determined in the laboratory for tailing material. The soil water characteristic curve (SWCC) and air entry value for oxygen diffusion might also be determined in the laboratory (see Chapter 6).

5.4.4.4 Chemical Characteristics

The primary purpose of determining chemical composition is the identification of constituents of interest. Determining chemical composition requires that a wide range of metals be analyzed. Fortunately, modern ICP-MS scans provide a large number of parameters at relatively low costs.

Identifying parameters that might be of concern is accomplished through comparison against average values for reference materials (e.g., crustal abundance, composition ranges for specific lithologies and soils). The soluble or leachable proportion of constituents of interest can be determined by combining the results from the chemical analysis with those from leach tests.

Other uses of chemical analyses include evaluation of sample representativeness and determination of all or part of the bulk mineralogy. Chemical analyses may also provide a surrogate for acid base accounting parameters (e.g., Ca for NP; total sulphur for AP). Table 5-3 is an example table of results from chemical analysis of various rock types, including a comparison against crustal values.  

Table 5-3: Example Chemistry Table

5.4.4.5 Mineralogical Characteristics

The mineralogical characteristics of a mine waste and geologic material are often the most important control on ARD/ML and mine water quality. It is therefore critical to evaluate the mineralogical characteristics of a mine waste.

Several minerals often contain the same element of water-quality concern but minerals have different degrees of solubility, reactivity, and weatherability. As discussed in Chapter 2, “reactive” mineralogy, mineral chemistry, mineral assemblage, texture, morphology, and grain-size effects will all affect the composition of the drainage coming from the source material.

Mineralogical investigations provide valuable data that assist in interpreting other laboratory tests. The purposes of a mineralogical assessment include the following (Thompson et al., 2005):

• Confirm presence of minerals contributing to static and kinetic laboratory test results.
• Identify sulphide minerals that may contribute acidity and metals.
• Determine presence of carbonates and silicates that may consume acidity versus those that may not (e.g., calcite vs. siderite).
• Identify potential galvanic effects that may impact acid production or metal leaching
• Assess relative distribution of acid producing and consuming minerals in fractures and veins that could result in waste rock fines of different composition form the whole rock.
• Identify evidence of previous weathering and coatings.

Types of mineralogical investigations include petrographic analysis, X-ray diffraction (XRD), scanning electron microscope (SEM) and microprobe. Optical petrographic analysis conducted on thin sections is included in most metallurgical and ARD assessments.

5.4.4.6 Laboratory Static Methods

5.4.4.6.1 Acid Generation Potential

Two basic types of test for determination of acid generation are available: those that measure acid generation potential through independent determination of acid generating and neutralizing content, and those that generate a single value that can be used to indicate the likelihood of acid generation or release of stored acidity. The first type of test is collectively referred to as ABA test, while the latter type of test includes the net acid generation (NAG) test and paste pH. On a global scale, use of ABA and paste pH predominates, with the exception of Southeast Asia and Australia, where the NAG test is also used in conjunction with ABA for the prediction of ARD potential.

Both tests are relatively inexpensive and can be applied to large numbers of samples. The results from both types of test can be used for identification of samples requiring additional testing (e.g., kinetic testing) to more definitively determine acid generation potential (AP). In addition, the tests may provide operational screening criteria for mine waste classification and management. However, some differences exist in the ability of the tests to predict acid generation potential. The choice of test may therefore depend on site-specific considerations related to mineralogy, material characteristics, information requirements, or regulatory expectations.

ABA methods were initially developed for the coal mining industry and later adapted for use in metal mining. Although all methods incorporate an independent determination of AP and NP, many different protocols are available and in use. Table 5-1 presents the most common methods and summarizes advantages and limitations associated with each type of test. Results from ABA methods need to be interpreted in context with mineralogical information.

In general, the determination of the AP as part of ABA testing is conducted through analysis of one or more sulphur species. The theoretical relationship between sulphur content and AP is as follows:

AP (kg CaCO3/tonne) = 31.25 x S (%).

Sulphur species identified generally include total sulphur and pyritic (or sulphide) sulphur. Other sulphur species frequently determined (either through direct analysis or calculated by difference) include sulphate sulphur, organic (or residual) sulphur, and sulphate associated with barite. The acid potential can be calculated from total sulphur content (the most conservative approach) or the acid potential can be based on the concentration of one or more sulphur species to provide a more refined estimate of the amount of reactive sulphur present. In the case of coal, it is important to discount the proportion of sulphur associated with organics when determining AP. Similarly, sulphur occurring in the form of sulphate minerals, such as gypsum and barite, should be discounted when information on sulphur speciation is available.

The measurement of AP is relatively simple and generally not prone to significant subjectivity. However, tests developed to measure NP are not as straightforward in their interpretation because of the widely variable solubilities and reaction rates of potentially neutralizing minerals (e.g., carbonate and silicates), the differences in aggressiveness of the various methods used to determine NP, and the different reaction conditions and titration endpoints prescribed for each test. Because the resulting value for the NP is highly sensitive to test protocol, it is important that any ABA program makes use of the methodology that is most appropriate for a given objective and application. It is also important that at least one single test method is used throughout the program to ensure that the results are internally consistent. Although perhaps imperfect, the advantage of using “standard” methods for determination of NP, such as the Sobek and modified Sobek methods (see Table 5-1 for description), allows for comparison against a vast body of references values from other sites. The values for AP and NP are combined mathematically to indicate whether a sample has a stoichiometric balance that favours net acidity or net alkalinity. Table 5-4 is an example of ABA results, including summary statistics. Figure 5-6 provides an example comparison of NP calculated from total carbon measurements vs. NP using the modified Sobek method, while Figure 5-7 compares total sulphur content against sulphide sulphur content. These are just two of the many graphs that can be used to interpret ABA results.

Table 5-4: Example ABA Table

Figure 5-6: Example Plot of NP from Total Carbon vs. NP from Modified Sobek

Figure 5-7: Example Plot of Total Sulphur vs. Sulphide Sulphur

The NAG procedure uses a strong oxidant (hydrogen peroxide) to rapidly oxidize sulphide minerals in a crushed sample of the entire rock. The NP of the sample then can be directly consumed by the acidity generated by rapidly oxidising sulphide minerals. Although an overall balance is obtained directly with regard to a net acid or net alkaline potential, the test offers no indication of the individual values of AP and NP. A temporal component can be added to the NAG test by conducting a multistage NAG test (“sequential NAG”) or monitoring of diagnostic parameters (temperature, pH, EC) during reaction with the oxidant (“kinetic NAG”). Shaw (2005) recommends that the NAG test results be calibrated with other laboratory tests and that sequential NAG should be used as a check for samples with high NP or net neutralization potential (NNP). Figure 5-8 shows an example of the comparison between ABA and NAG test results.

Figure 5-8: Example Plot of ABA vs. NAG Results

Paste pH is a simple, rapid, and inexpensive screening tool that indicates the presence of readily available NP (generally from carbonate) or stored acidity. The outcome of the test is governed by the surficial properties of the solid material being tested, and more particularly, the extent of soluble minerals, which may provide useful information regarding anticipated mine water quality. For example, acidic paste pH values in combination with elevated sulphate sulphur generally suggest the presence of acidic sulphate salts that could cause short-term or long-term water quality issues.

5.4.4.6.2 Short-Term Metal Leaching

Although protocols for static (or short-term) leach tests vary widely, all tests measure readily soluble constituents of mine wastes and geologic materials. The short-term nature of static leach tests provides a snapshot in time of a material’s environmental stability. Test results depend entirely on the present disposition of the sample (e.g., unoxidized vs. oxidized; oxidation products absent vs. oxidation products present). For reactive rocks (e.g., material that contains oxidizable sulphur), the transient processes that lead to changes in solution chemistry during water-rock interactions often develop over periods of time that are much greater than is stipulated in the testing protocols. Therefore, the results from short-term leach tests generally cannot be applied to develop reaction rates and predict long-term mine water quality, but should instead be used to get an initial indication of parameters of constituents of interest. In addition, metal loadings can be calculated from short-term leach tests, as illustrated in Figure 5-9, where loading rates (in milligrams per kilogram [mg/kg]) are compared against initial sulphate content.

Figure 5-9: Example Plot of Metal Loadings vs. Sulphate Content

It is important to select the method that most closely simulates the site-specific ambient environment and conditions (e.g., solution to solid ratio, nature of lixiviant, grain size, agitation). In addition, selection of a test method has to take into account the anticipated use of the leach test results (e.g., for prediction of seepage vs. runoff quality, incipient vs. terminal water quality).

Regulatory requirements and expectations may also govern selection of a particular methodology. Many jurisdictions have well-defined regulations for evaluation of metal mobility and potential impacts to water resources and in such cases use of a test with regulatory status may be compulsory. In instances where such a test is required but where the mandated protocol has no bearing on site-specific conditions (e.g., the prescribed use of acetic acid in the TCLP test), use of an additional, and more appropriate, alternative short-term leach test is recommended to allow for a more realistic estimate of future mine water quality. Similarly, modifications to standard leach test protocols should be considered to take into account site-specific considerations and improve the tests’ predictive ability.

5.4.4.7 Laboratory Kinetic Methods

Laboratory kinetic testing methods are used to validate static test methods, to estimate long-term weathering rates, and to estimate the potential for mine wastes and geologic materials to release discharges that may have impacts on the environment. Both acid generation and metal leaching can be evaluated through kinetic testing. To generate the required information within a reasonable time frame, the testing procedures are designed to accelerate the natural weathering process.

The results from kinetic testing are frequently used in combination with geochemical modeling to evaluate geochemical controls on leachate composition and conduct water quality prediction under a range of conditions. Similarly, kinetic testing results are often scaled up and used in combination with water balances for mine facilities to determine loadings and associated potential impacts to the receiving environment. Depending on the end use of the kinetic test results, results may be expressed in terms of leachate quality (mass released/unit leachate volume), mass-based loadings (mass released/total mass/unit time), or surface-area-based loadings (mass released/total surface area/unit time). For loading calculations, a water balance for the test cell and information on the mass and the surface area of the test charge is required. Geochemical reactions and reaction rates most commonly monitored throughout the testing include sulphide oxidation, depletion of neutralization potential, and mineral dissolution.

Kinetic testing procedures are complex, time-consuming, and require operator skill to generate consistent results. For any kinetic test conducted, the objectives and limitations of the method used should be acknowledged before starting the program so that it is clear what information will be delivered from the tests conducted. This will ensure accountability and value for efforts and costs expended.

There is no single test that produces all of the chemical information required to evaluate all mine wastes under all conditions of disposal. In all cases, a sample is subjected to periodic leaching and leachate is collected for analysis, but the various methods available may differ in the amount of sample used, the particle size of the sample, effluent sample volume, test duration, degree of oxygenation, or nature of the lixiviant. Therefore, it is important that the objectives of kinetic testing are clearly defined so that an appropriate test method is selected and adjusted to simulate site-specific conditions and the intended use of the data produced. By the same token, conducting standard humidity cell tests (e.g., using the ASTM protocol – see Table 5-1) is very useful to allow comparison with the significant amount of information on kinetic test results available in the literature. A second phase of kinetic testing may be implemented or field testing may be considered if it is decided that tests representing site-specific conditions are required.

Two “end-member” types of kinetic tests generally recognized are the humidity cell tests (HCT) and the column tests. HCTs represent a standardized test under fully oxygenated conditions with periodic flushing of reaction products. No standards are available for column tests, and column tests can simulate different degrees of saturation, including flooded and oxygen-deficient conditions. Column tests are typically larger scale than humidity cell tests. Figure 5-10 is a photo of a typical HCT setup.

Figure 5-10: Humidity Cells

Both of the HCT and column tests may be used to estimate longer term potential for acid generation and metal leaching. HCTs are primarily intended to generate information on weathering rates of primary minerals (e.g., sulphides) only. However, HCTs can also be used to assess dissolution rates of readily soluble primary and secondary minerals present at the onset of testing (e.g., gypsum, hydrothermal jarosites). In theory, column tests provide information on combined weathering rates of primary and secondary minerals. Column tests may also be more suited to evaluation of attenuation and environmental performance of mitigation measures such as covers and amended mine wastes. Transfer of oxygen, which is not limiting in HCTs but may be in columns, must be understood in column testing. For determination of lag times to acid generation, HCTs are the preferred approach. Figure 5-11 is an example of concentration trends over time and presentation of HCT results.

Figure 5-11: Example Plot of HCT Results

For both HCT and column tests, it is imperative that the test charges be characterized before kinetic testing begins and after kinetic testing has been completed. The information on the test charges may provide important constraints to assist in the interpretation of test results, and may also provide information that can be used for quality control purposes by comparing measured mass removal against calculated mass removal from the leachates.

The required duration of kinetic testing is an area of frequent controversy. Although a minimum length of 20 weeks is universally referenced, there is little technical basis for this recommendation. In actuality, the kinetic testing duration depends on the specific objective of the test. If the objective is to determine whether a sample will generate acid, kinetic tests should be conducted until acidity is produced or until depletion calculations can be used reliably to predict acid generation potential. Another common endpoint for the kinetic testing is when leachate parameters are relatively constant with time (e.g., over a 5-week period).

5.4.4.8 Field Methods

Field methods to determine acid generation and metal leaching potential range from rapid very small-scale tests to monitoring of full-size mine facilities for extended periods of time. In all cases, the advantage of the field methods is that on-site materials are used and an added benefit is that that most field tests allow for evaluation of weathering reactions under ambient conditions, including seasonal effects and discrete events such as intense storms or snowmelt. The larger the amount of material being tested, the greater the likelihood that a well-designed test is representative in terms of chemical and mineralogical composition and that the physical properties of a mine facility are being simulated. The larger amount of material will better represent particle size distribution, porosity, hydraulic conductivity, gas ingress, and transport. Disadvantages of field cells are related to the time required to generate reliable field reaction rates, the challenges with comprehensive geochemical characterization of the large test charges, and the inability to test a large number of different material types.

The simplest “field” test is the 5-minute field leaching test (FLT) recently developed by the USGS to simulate the chemical reactions that occur when geological materials are leached by water. The test is considered by the USGS a useful screening procedure that can be used as a surrogate for laboratory leach tests such as the Synthetic Precipitation Leaching Procedure (SPLP), (see Table 5-1).

Wall washing allows for evaluation of runoff quality from an isolated section of in situ rock face after application of a controlled amount of irrigation (Figure 5-12). This wall washing test is considered to represent a very useful order-of-magnitude estimate of contributions from exposed mine waste, in particular open pit walls or underground mine faces.  

Figure 5-12: Wall Washing

Pilot cells (Figure 5-13), test piles, test plots (Figure 5-14), or test pads are constructed for long-term monitoring of relatively large quantities of material. Large-scale field columns (field lysimeters), to be operated under natural precipitation conditions, can also be useful.

 

Figure 5-13: Test Cells for Waste Rock

Figure 5-14: Test Plot for Paste Tailings – Somincor Neves Corvo Mine, Portugal

Monitoring can be conducted under ambient field conditions, or under controlled conditions, using artificial irrigation. The large scale results in use of a more representative sample, and minimizes impacts from boundary effects, sample heterogeneity, and reduced grain size relative to laboratory tests. A comprehensive characterization of the test charge is required. In combination with a good understanding of the water balance for the test pad (achievable through meteorological monitoring or controlled application of infiltration, or both), reaction rates and loadings can be developed for extrapolation to full-scale mine facilities. Longer monitoring durations are generally required because of the reduced reactivity of field cell test charges relative to the finer-grained materials commonly included in laboratory tests. It may be advantageous to operate field tests during the complete life of mine to identify potential long-term releases.

On-site monitoring of historical and newly-constructed mine facilities (e.g., waste rock pile, tailings impoundment, pit wall, adits) can provide very useful information regarding weathering rates and discharge quality under ambient conditions. By definition, monitoring results of this nature are representative of the facility and existing conditions as a whole, but prediction of future conditions may be hindered by the sluggish rate of reaction relative to smaller scale tests. Also, a comprehensive understanding of chemical and physical material characteristics is not generally feasible, nor is a comprehensive understanding of the water balance, water movement and the role of atmospheric gases. This may limit the interpretive value of direct monitoring of mine facilities for the prediction of future water quality and potential impacts to receptors.

Top of this page

5.4.5 Data Management

Proper data management is critical to any geochemical characterization and mine water quality prediction effort, and setup and maintenance of a database is an integral component of such a program (Wolkersdorfer, 2008). The primary requirements for a useful and reliable database are that it should be in electronic format, it should be implemented from the very beginning of the study, and it should be maintained and augmented throughout all phases of a mining project.

A database should be managed from a central location, with routine backups. The data should be presented in a format that is readily accessible, and appropriate safeguards should be in place to maintain the integrity of the information stored in the database and prevent unauthorized use. Although most databases are designed to store numeric information, increasing use of geospatial data is incorporated by use of GIS. GIS provides a means for integrating and interpreting geochemical data within a geospatial context for land use, climate, topography, or ecosystem.

The primary function of a database for geochemical data is to act as a comprehensive data repository that can be used to check and maintain data integrity (see Section 5.3.6 on QA/QC), support data manipulation and data interpretation, support and guide water quality and other monitoring programs, enable evaluation of compliance with regulatory requirements, and allow for evaluation of historical trends and prediction of future conditions.

One type of database unique to mining is the so-called block model, which is a 3-dimensional computerized representation of the quantity and characteristics of the pit walls, ore, and waste rock. Historically, block models have been resource focused, and have included information on ore grade, lithology, alteration types, principal minerals, fracture density and orientation, and rock competency, all of which are aimed at optimizing resource recovery. To this end, data from exploration drill holes are subjected to a variety of geostatistical analysis methods, such as kriging to quantify the 3-dimensional distribution of ore throughout the mine. However, increasingly, the same block models and geostatistical techniques are also used for environmental purposes, such as development of waste rock management plans and mine water quality prediction. Results of geochemical characterization programs are incorporated in block models, including inputs such as sulphur and sulphide content, NP, paste pH, NAG pH, NCV, carbon, and carbonate content. The combination of resource and environmental parameters in block models allows for prediction of environmental behaviour of mined materials in time and space and identification of requirements for mitigation actions in time and space. Environmental block models should be developed when a 3-dimensional understanding of ARD potential is required, and should then be maintained and refined throughout the life of mine through the ongoing acquisition of additional data. Examples of use of block models are presented in and . Figure 5-15 shows the ARD potential of a highwall remaining exposed after pit lake formation. Figure 5-16 shows the ARD potential of pit walls at the cessation of mining. In both cases, a block model incorporating ABA parameters formed the basis for the evaluations.

Figure 5-15: Example of Block Model Use: ARD Potential of Pit Highwall Above Final Pit lake

 

Figure 5-16: Example of Block Model Use: ARD Potential of Pit Wall after Cessation of Mining

Top of this page

5.4.6 Quality Assurance/Quality Control

A rigorous QA/QC program is needed to ensure that geochemical data are reliable and defensible, and that such data can be used for their intended purpose, such as mine water quality prediction.

QC is defined as the application of good laboratory practices, good measurement practices, and standard procedures for sampling. QC is also defined as sample preparation and analysis with control points within the sample flow to prevent the reporting of erroneous results. The sampling should include specifications for chain of custody procedures and documentation, sample holding time verification, drying, comminution, storage and preservation, sample labelling, and use of proper sample containers. Physical and chemical tests conducted using approved methods and accredited laboratories should produce analytical results with sufficient accuracy and precision for their intended usages. Analytical methods and their repeatability, reproducibility, quantitation, and detection limits should meet anticipated requirements (e.g., for comparison against water quality standards). Replicate samples, standards, certified reference materials, and blanks should be routinely submitted to ensure and confirm that the analytical results are of acceptable quality. QA is the process of monitoring for adherence to quality control protocols. The DQO of a quality assurance project plan (QAPP) are as follows: accuracy, precision, bias, representativeness, completeness, and comparability. A QAPP will ensure that the proper procedures are established before initiating sample collection and analysis, and that procedures are maintained throughout all stages of a geochemical program. In addition, corrective actions are prescribed through a QAPP. A defensible QA/QC program will add costs to an ARD study, but it will also enhance the confidence of operators, regulatory agencies, and other reviewers in assessing the data. A QAPP will help balance the costs of implementing a quality-assured program against the potential liabilities associated with a poorly-designed and executed geochemical characterization program.

The data validation and assessment protocols for geochemical data generated in support of prediction of ARD and metal leaching potential are similar to those used in any type of study that relies on use of analytical results, and the data validation and assessment protocols include a variety of statistical analyses and graphical tools. Geochemical modeling can be useful (e.g., through calculation of the ion balance), while cross checking using results from different types of testing also may provide insight in data quality (e.g., calcium content vs. NP, sulphur content vs. mineralogical composition, measured vs. calculated TDS, NP titration vs. TIC).

Top of this page

5.4.7 Screening and Evaluation Criteria

Use of screening and evaluation criteria is generally required to assess whether results from geochemical characterization studies represent a potential impact or risk to a receiving environment at a mine site. These criteria can be based on professional and empirical experience, guidance documents, and regulations promulgated for the express purpose of protecting the environment.

Screening and evaluation criteria are commonly used at mine sites for water and mine waste management. Mine waste management involves identification of potentially acid generating (PAG) and NAG waste. PAG material is either acidic or predicted to become net acidic in the future. A material will become net acidic if the rate of acid neutralization is unable to keep pace with the rate of acid generation. This inability to maintain neutral conditions may be due to a decrease in the rate of acid neutralization or an increase in the rate of acid generation, or both. NAG material is predicted to generate near-neutral or alkaline drainage in the future. Materials will be net neutral or alkaline if the rate of acid neutralization keeps pace with the generation of acidity (Price, 2009).

Site-specific operational parameters and threshold values are established for waste classification (i.e., PAG vs. NAG) based on regulatory requirements, literature, and the geochemical test program. Examples of commonly used operational parameters for waste rock management include the sulphur content, paste pH, NNP, net potential ratio (NPR), NCV, NAG value, or NAG pH and metal content.

Professional and empirical experience may be an acceptable basis for establishing a screening or evaluation criterion. For example, if a quantitative relationship can be reliably established between ARD potential and sulphur content, a sulphur cutoff can be determined to segregate between PAG and non-PAG waste rock. Similarly, if a relationship between metal leachability and metal content is identified, a metal concentration cutoff can be established to discriminate between material that will or will not affect receiving water quality. Sometimes a combination of methods is needed to classify problematic material, such as paste pH and NPR.

Guidance documents are available that provide screening criteria for evaluating geochemical test results, in particular those tests related to prediction of ARD potential: ABA and NAG (Price, 1997; AMIRA, 2002). These criteria are generally related to specific values for NNP, NPR, NAG pH, and NCV, and can be used to classify mine wastes and geologic materials in terms of their ARD potential. Special care is required when dealing with mining wastes that exhibit both low sulphur contents and low NP because small changes in analytical results can dramatically affect the calculated NPR and the mine waste classification. Therefore, the screening process should generally consider use of multiple criteria and tests, such as those based on NNP and NAG.

Figure 5-17 is the Australian AMIRA (2002) decision tree for determining acid generation potential. Through use of a combination of results from ABA testing, NAG testing, and professional judgment, samples are categorized into a number of classes with a range of ARD potentials. Another example of guidelines widely used throughout North America is based on NPR values as follows (Price, 2009):

Figure 5-17: Decision Tree for the Determination of Acid Generation Potential (AMIRA, 2002)

Potential for ARD	Initial Screening Criteria	Interpretation
Likely	NPR <1	Likely acid generating, unless sulphide minerals are nonreactive
Possible (uncertain)	1<NPR<2	Possibly acid generating if NP is insufficiently reactive or is depleted at a rate faster than sulphides
Nonacid generating	NPR>2	Not potentially acid generating unless significant preferential exposure of sulphides along fractures planes, or extremely reactive sulphides in combination with insufficiently reactive NP.

In Europe, an NPR value of 3 is conservatively assumed to be the threshold between potential acid generating and nonacid generating mine waste. However, use of a lower ratio is acceptable if it can be proven, based on site-specific information, that such a value is sufficiently protective. As with all screening criteria, the burden in on the proponent to prove that these criteria are appropriate and defensible based on site-specific considerations.

Worldwide regulatory jurisdictions have adapted criteria for ARD potential, and some have been promulgated into law. When such criteria exist, their application is generally mandatory, unless use of appropriate and defensible site-specific criteria is allowed under the law. The selected criteria can vary and an understanding of applicable regulations is needed when evaluating results from ABA and NAG tests for the purpose of prediction of ARD potential and identification of mine waste management requirements. Examples of such regulated criteria include an NPR threshold of 3 for nonacid generating waste in New Mexico, an NPR threshold of 1.2 in Nevada, (i.e., 20% excess base), and a three-pronged approach in Quebec based on sulphide content, NNP, and NPR. In Quebec, acid generating material is characterized by sulphide content greater than 0.3%, and, in the absence of confirmatory kinetic testing results, an NNP less than 20 kg CaCO3/tonne or an NPR less than 3. Figure 5-18 is an example plot of ABA results in which a number of screening criteria have been included, delineating the boundaries between materials with a different potential for ARD.

Figure 5-18: Example Plot of ABA Results and ARD Criteria

Regulatory criteria also exist for interpretation of results from certain leach tests specifically designed for classification of waste materials and compliance with water quality standards, as indicated in Table 5-1 and (AMIRA, 2002). Examples of such tests include the TCLP, meteoric water mobility procedure (MWMP), and WET tests in the United States, the CEN-series tests in Europe, the Chinese GB tests, and the Brazilian Norma Brasileira Registrada (NBR) tests.

In general, kinetic test results need to be interpreted in the context of all available geochemical information. The following evaluation steps may be of assistance in the assessment of kinetic test results:

• Temporal trends of acidity, alkalinity, sulphate, and pH used to assess rates of acid production and consumption
• Ratio of acid production (using sulphate) vs. acid consumption (using calcium, magnesium, alkalinity) to assess relative rates
• Comparison between observed sulphate generation rate and literature values (Morin, 1997)
• Comparison between observed metal concentrations and water quality objectives (A direct comparison generally should only be used as a screening tool, and should take into account the differences in solid to liquid ratio between the test and the ambient environment.)
• Comparison between kinetic test results and findings from ABA, NAG, mineralogy, static leach testing, and field water quality
• Comparison between kinetic test results and water quality from analog sites (i.e., geo-environmental approach)
• Geochemical modeling to identify controls on leachate composition
• Development of relationships between sulphate concentrations and those of constituents of interest that can be extrapolated to field conditions through sulphide oxidation modeling or calibrated against field measurements of sulphide oxidation

In the absence of regulatory criteria, and frequently in addition to regulatory criteria, site-specific screening criteria should be developed. These criteria should be based on a thorough geochemical characterization of the material at hand. Results from ABA, NAG testing, mineralogical examination, leach testing, and kinetic testing are used to develop an internally consistent understanding of acid generating potential, culminating in identification of a small number of criteria (generally one or two) that can be used to reliably classify mining wastes and geologic materials according to their ARD potential. To be of value in an operational setting, these criteria need to be based on parameters that can be rapidly determined onsite with a high degree of confidence. Visual methods (e.g., rock type, pyrite content) and laboratory determination of total sulphur, NCV, and NAG pH are probably the most commonly used operational waste management tools.

Although development of screening criteria is commonly aimed at identifying the acid generation potential of a mine waste or geologic material, the process of evaluation of potential environmental impacts should not stop there. The material classified as nonacid generating should still be assessed for drainage quality. NMD and SD from nonacid generating material may continue to be a cause for concern even in the case of waste management strategies that include, for example, segregation of PAG from non-PAG waste rock or encapsulation of PAG rock by non-PAG rock.

Top of this page

5.4.8 Reporting

Reporting is an integral part of an ARD-related study. In addition to including tabulations of the analytical results, the reported information needs to be presented in a format that provides proper interpretation. This requires calculation of descriptive statistics and use of a variety of graphical representations developed for evaluation of results from ABA, NAG, and kinetic testing. Price (1997) or Wolkersdorfer (2008) provide a comprehensive overview of the most commonly used table templates, calculation sheets, and graphs.

These procedures must be documented and submitted as part of the report because the reviewer of an ARD study may not be familiar with all analytical and sampling procedures. Also important is a discussion of QA/QC aspects and their bearing on data reliability and defensibility.

At a minimum, the report needs to contain all predictions of environmental behaviour, including the approach and tools used (e.g., geochemical modeling code, statistical software), assumptions incorporated in the predictions, the prediction results, and a discussion of uncertainties and limitations associated with the predictions. Frequently, a report will also include recommendations for further activities related to data collection or evaluation, an interpretation of results in terms of potential environmental impacts, and an assessment of measures that can be used to prevent, minimize, or mitigate such potential effects.

Previous Page (Page 1) of Chapter 5

Last Page (Page 3) of Chapter 5

Top of this page