Environmental & Earth Sciences
All mine processes require water for the extraction and concentration of ore. No matter how good the liner/capture system or the environmental management, spillage and leakage will eventually occur. Groundwater monitoring is usually undertaken to detect leaks, although the nature of the leachate can make detection difficult. The recent trend towards using statistical methods leads to confusion and concentration on deleterious contaminants rather than facilitating the detection of leaks. This type of monitoring is known as reactive monitoring and too often the interaction chemistry is ignored.
It must be accepted that zero discharge is impossible and leakage will occur. The best case seepage for artificial liners is calculated as 2,500 L/d/ha and the average case is 25,000 L/d/ha. In terms of water loss, compared to evaporation from a heap-leach pad, this is inconsequential. However, in terms of the impact on groundwater it may be significant. With this in mind, monitoring should be undertaken to ensure that leakage is occurring at or below a rate at which the environment can absorb any deleterious constituents. Prior to the start of monitoring, a conceptual model should be created, based on the physics and chemistry of the soil and groundwater, the waste and the interaction between them.
In most countries, the potential impact of leachate from processing plants and refineries (on groundwater and surface water) has resulted in monitoring being made a statutory requirement. Tailings dams, evaporation ponds and heap-leach operations tend to be licensed on a site specific basis and the usual groundwater impacts are often accepted as minor or unimportant. However, some countries, such as New Zealand, require that there be no measurable impact on the ground and surface water resulting from mining activities. Other states, such as Victoria in Australia, require that there be no adverse impact on potential beneficial users. This is a more practical approach as it must be understood that all liners will fail, possibly many decades after closure.
Regulatory authorities should recognise that tailings dams and heap-leach pads will leak and so should emphasise the protection of aquifers from pollution, without ignoring potentially adverse impacts from groundwater in aquitards. This goal is the basis behind the proposed monitoring strategy. The objectives of monitoring are two-fold: firstly to ensure that the leachate migration occurs at the predicted rate; and secondly, to provide early warning of
adverse plumes so that previously planned correction measures can be put into place. Effective monitoring compares sampling results with threshold levels that require a planned response. This could be increased frequency of monitoring, through to remediation measures. To achieve effective monitoring, an understanding of background groundwater chemistry, leachate chemistry and local geology is required.
When planning an environmental groundwater investigation it is important to assess the permeability, gradient, geological medium and direction of groundwater flow, to help determine piezometer installation design and location. Naturally occurring preferred pathways can be discovered by defining lineaments from aerial photographs, geological structural mapping and geophysical techniques. Monitoring bores should be positioned with respect to preferred pathways and sub-surface conditions, with more being placed down-gradient than up-gradient.
In many mines, interface drainage (perched water table) is often ignored as groundwater is perceived to be many metres below the level of the interface. Bores should be placed into the interface between extremely weathered rock and competent rock, if there is a significant difference in permeability between the soil and the weathered rock (B/C horizon in soil), at the topographical low. Water will only collect in these bores during periods of prolonged rainfall, seasonal wet periods, or after a groundwater mound has developed. The effect of groundwater mounding is especially important in ground of varying permeability, as leachate may flow along strata that is not normally saturated.
The aim of the monitoring bore is to ensure representative sampling so, not only is placement important, but so is the construction.
- Idealised piezometer construction.
Ideally, to ensure that remediation or corrective measures have sufficient time to be implemented, placement should involve a double ring monitoring system. The first ring of bores is placed close enough to detect leachate rapidly and the second ring is placed to confirm the conceptual model and to allow sufficient time for correction measures to be put in place. Based on experience, the distance from the inner ring of monitoring bores to the second ring should be no less than that travelled by groundwater in two years.
As well as bore placement, construction of the piezometers (particularly slot placement) is crucial. If the slots in a bore are placed so that they span the geological strata, their ability to detect changes in water quality may be limited. For example, if the monitoring bores are constructed with slots to intercept both the clay and the upper silty portion of a sand aquifer, the water monitored represents a mix of both waters.
- time required for leachate to flow various distances
h = 10 m
|Liner and collection
h = 0.5 m
|Liner alone||–||309 days||7 years|
k = 10-4 m/sec
k = 10-9 m/sec
k = 10-6m/sec
k = 3.1 x 10-9m/sec
k = 1.0 x 10-6m/sec
k = 2 x 10-7m/sec
k = 5 x 10-5 m/sec
- Permeability from Fetter 1988, AGC, 1984, Porosity from Fetter.
- Assume conditions are static for the interval.
Frequency of monitoring
Frequency of monitoring normally depends on government or licence requirements. However, in terms of monitoring effectiveness, the most important criteria is the permeability of the rock or soil. Table 1 shows the time taken for a leachate to reach a piezometer, at a given distance from a tailings dam after breaching the liner. Two examples are shown: one a tailings dam with a 10 m head and the other an evaporation pond with 0.5 m head. Both have a base with a 0.8 m compacted clay liner. A variety of geologic media is represented. It is assumed that the leakage from the base is instantaneous and that there is no unsaturated flow.
When there is sand, highly fractured rock or permeable aquifers immediately beneath the facility, the leachate plume is rapidly transported once it penetrates the liner. Composite liner systems are required when the leachate is hazardous, and the inner ring of monitoring bores should be placed on the wall of the dam or holding facility and the outer ring at greater than 100 m from the inner ring.
Tailing dams, located in weathered shale, siltstone, and fractured clays should ideally have monitoring bores placed at approximately 10 m and 50 m from the edge of the wall, while a shallow evaporation pond in the same environment should have piezometers located at 5 m and between 15 m and 20 m from the edge of the facility.
Tailings dams that are on or above a thick deposit of low permeability clay or competent rock, with little primary porosity in the absence of preferred pathways, require piezometers within the dam wall. If the storage facility is located within an excavation or void, drilling closer than 5 m may well create preferential fractures from the storage facility. Bores beyond 5 m should be positioned in preferred groundwater pathways.
From Table 1 it can be seen that sampling times for different geological conditions need to vary significantly. Dams and ponds located above sand or gravel lenses, weathered sandstone and permeable aquifers essentially require semi-continuous monitoring equipment, particularly when over or near a potable aquifer. Continuous monitoring probes are limited to electrolytic conductivity (EC) which is only useful when the total dissolved solids in the groundwater are below 2,500 mg/L and the leachate itself contains significant inorganic compounds, but can nevertheless be a useful indicator of leachate. Probes require regular maintenance and calibration to remain reliable. Though pH probes would also be useful, we have found that they have yet to live up to manufacturers’ claims.
If semi-continuous monitoring is not possible, pH, redox potential (pc), EC, odour and colour should be measured in the field weekly or fortnightly and appropriate constituents ideally monthly or, at worst, quarterly.
In all other instances, except stiff clay and unweathered shale, the location of the inner bores should be placed to allow quarterly monitoring. Outer bores should only be added for background sampling or if there is a response in the inner bores. By using quarterly monitoring, an early breakthrough can be detected and an appropriate strategy put into place, depending on the quality prior to the groundwater reaching the outer ring, which should be inside the site boundary if possible. In a controlled seepage designed dam or pond, remediation should be unnecessary with monitoring being used to collaborate the tailings dam conceptual model and confirm successful attenuation.
For stiff clay and competent rock with little to no primary porosity, the inner bores and bores on preferred pathways should be monitored quarterly for pH, pc, EC, odour and colour and six monthly or even annually for appropriate constituents.
To understand the concept of controlled seepage, the generation of leachate and variation of leachate with time needs to be understood. The nature of the reactions in the tailings, the types of original rock and ore and additives from the mill partially decide the constituents of the leachate. Groundwater chemical behaviour is controlled by its constituents, which are determined by the initial source of the water (rain, sea, connate water, leachate) and the medium through which it travels (soil, rock, fill). Currently, the common practice is to monitor the groundwater for the toxic substances known to be in the storage or disposal facility. These tend to be the most reactive and are usually the last substances to appear in a breakthrough front, while other contaminants may not be detected. With this type of monitoring, the contamination must actually occur before it can be detected, usually resulting in an expensive cleanup programme.
Effect of medium on groundwater: except in the case of highly permeable material, groundwater usually has sufficient time to reach equilibrium with the minerals in the medium and to allow ionic exchange with colloids. Therefore, the groundwater will reflect chemical attributes associated with the medium through which it has travelled or is travelling.
There is often a correlation of rock type and water quality, with the exchangeable cations of the weathered rock having similar ratios to that in the groundwater. During groundwater monitoring at the Nabarlek uranium mine, this was found to be the case with the magnesium/calcium ratio and, to a lesser extent, the mean potassium concentration, as indicated in Figure 2. Shallow monitoring around the evaporation ponds at Nabarlek indicated that the normal leachate breakthrough, as characterised by elevated sulphate, Mg/Ca and manganese, was not occurring. In a small group of bores in one area, there was no rise in sulphate levels.
- Variation of Mg:Ca ratio in natural groundwater
Investigations of the soil found that this area contained elevated barium as barium carbonate. Ionic substitution was occurring, creating barium sulphate. This was confirmed by the rise in bicarbonate concentration in the groundwater. For groundwater monitoring, it is therefore important that the mineralogy of the soil/rock be known.
Baseline water chemistry: complete identification of baseline water chemistry is also important. Analysis of groundwater down-gradient from a metal smelter showed that the water was high in sulphuric acid and metals, thought to be due to the smelter. Several sites had similar water quality, but, when a complete water analysis was carried out, the area of greatest groundwater flow exhibited some important differences. The water in the gully down-gradient of the plant, was acidic with a different chemical analysis to the rest of the site. Fluoride concentrations in this area were unusually high and not apparent elsewhere on the site.
A fertiliser plant is located immediately up-gradient from the smelter and over the infilled gully where the acid groundwater was flowing. Fertiliser manufacture uses sulphuric acid and produces fluoride as a waste material as well as waste acid. In addition, the ‘background’ groundwater has unusually high sulphate concentrations for this area, thought to be due to the regional atmospheric pollution from many years of smelter operation.
This clearly demonstrates that the definition of baseline data is important in interpreting the presence of a contaminant plume.
Groundwater chemical processes: the key to monitoring is not so much the contaminants themselves, but the reaction processes they are involved in. These reaction processes in groundwater may be grouped together as the process of attenuation. If there are no physical and chemical mechanisms of attenuation, the leachate would arrive at any point at maximum concentration, that is a slug. However, various mechanisms result in a decrease in concentration of the ions in the leachate. The physical mechanisms are dilution and longitudinal and transverse dispersion.
Dilution is caused by a highly concentrated plume of leachate discharging into ground water or surface water containing significantly lower concentrations. This results in a reduction in the concentration of leachate in the resulting fluid. Longitudinal dispersion is caused by the interaction of the fluid with the soil or rock medium. The fluids pass through the pores in the medium which vary greatly in size. Some parts of the leachate move in a straight path while others have a more tortuous path, resulting in a sigmoidal increase in concentration with time. The same mechanism will spread the plume laterally and this is known as transverse dispersion which, like dilution dispersion, results in an initial reduction in leachate concentration but does not change the relative relationship of ions. If the source of leachate continues, the effect of dispersion is gradually eliminated and the concentration increases with time.
Ions dissolved within the leachate may have an affinity for the soil or the rock it travels through. Clays, in particular, attract ions such as heavy metals which exchange with ions, usually calcium and sodium, already on the clay. Some ions attenuate or slow through their spread in the soil, while other ions may not be affected at all. This is called chemical attenuation. Although most hydrology practitioners are aware of the processes involved in chemical attenuation they may be unaware of some of the implications. Briefly, the mechanisms involved in attenuation are: adsorption by ionic exchange, precipitation (both pH dependent and solubility controlled), co-precipitation, ionic exchange, and bacterial degradation and redox reactions.
These mechanisms, while reducing the concentrations of some ions, introduce new ions and leave others to migrate unimpeded, depending on the chemical conditions of the groundwater and the constituents of the soil. Precipitation and co-precipitation removes ions from solution and alters the relative relationship of the ions. Ionic exchange removes some ions and introduces more of the relatively common ions. Bacterial degradation may remove some ions but, because they respire carbon dioxide, they introduce carbonate species to the water, usually bicarbonate. Bacterial action normally affects the redox state of the water (loosely defined as the presence or absence of oxygen in water). Redox reactions have a major effect on the nature of ions in solution. These reactions fragment a leachate front but also provide early warning precursors to the front.
Characteristics of a front: as the chemical characteristics of the leachate are altered by reactions occurring in the aquifer and soil as it passes, reaction by-products and non-reactive ions should be monitored for. Normally the contaminants that are most deleterious to humans and animals are the most reactive in the soil. Heavy metals, cyanide, acidity, alkalinity and arsenic are the most common contaminants in the mining industry. These ions are also the ions most likely to be attenuated and the last to come through on a leachate plume.
Ratios, rather than absolute ionic concentrations, are used to demonstrate the passing of a front. In the initial stages of a plume, the concentration may not be above the background variation, and it is in this area that the use of ratios and an understanding of chemistry is most useful in detecting the existence and advancement of a plume.
The native cations in pore water increase dramatically through the exchange process during the passage of seepage water. Initially, the dominant native cations are displaced by ammonium and heavy metals from the seepage water, with small amounts of less dominant ions also being displaced. This process is dependent on the dynamics of ionic exchange and is also subject to the ionic activity of ions in solution as well as the hydrated ionic radius and valency of the ions.
An alteration in the concentration and ratio of native cations, detected by groundwater monitoring, is seen as a precursor to a contaminant front, as the native cations are displaced by cations in the plume. The most commonly used ratios for this purpose are Na:Ca, Ca:Mg, Ca:K, Na:Mg and Na:K.
Similarly, an alteration in the ratio of anions usually heralds a front. This is caused by either bacterial by-products (carbonate and bicarbonate); products of a chemical reaction (carbonate and bicarbonate); reduction by bacteria or chemical reaction (NH4 from NO3 and hydrogen sulphide produced from sulphate) or introduction, by the seepage plume, of highly mobile ions such as Cl. The most common ions monitored are Cl, SO4 and HCO3 and ratios used are Cl/SO4, SO4/ HCO3 and Cl/ HCO3.
When the leachate differs considerably from exchangeable or groundwater cations (native cations), which can occur in many of the mine wastes, the ML/N (mine leachate/native cation) ratio is used to monitor the groundwater:
ML = 100 x 3 most dominant, least reactive cations in leachate
N = 3 most dominant groundwater or soil/rock exchangeable cations
The requirement to monitor other ions depends on the contaminants, the background groundwater and the medium. However, ions such as nitrate, ammonium, fluoride, manganese and zinc should be routinely considered for monitoring. Heavy metals, cyanide, TPH, PAHS, aromatics, when present in the waste water or products stored, should be monitored after the indicator front has passed. The pH, pe (Eh) and conductivity should be monitoring in the field, at the time of sample collection, for every hole.
As discussed previously, attenuation results in differing velocities for different contaminants, such that at any given time a series of fronts may occur within the one contaminant plume.
The nature of the plume, both in concentration and how it is attenuated, depends on the initial concentration and local ground conditions. A heavy, clayey soil or rock with clay minerals and low permeability will attenuate and fragment the plume, resulting in widely separated distinct phases of arrival. A non-reactive porous medium, such as sand, will result in a poorly attenuated plume. A positive detection in a bore may not be due to the groundwater that is thought to be monitored. If a plume in an attenuating environment, such as a clayey weathered rock, arrives as a pulse, rather than in distinct phases, or the phases are only slightly separated, leakage down the hole from interface drainage or surface runoff may be the cause. Normally in such a bore, standing water levels show greater variation than in other bores.
Pro-active monitoring: by monitoring the reaction by-products, bacterial waste products and displaced native cations, as well as the more mobile contaminants, it is possible to detect a contaminant front as it eventuates or, occasionally, even before it arrives. This type of monitoring is based on the development of a simple groundwater model. To be pro-active responses to plume breakthroughs need to be planned. For example, a slightly faster response than predicted may only require an increased frequency of monitoring of the outer ring of bores to confirm that environmental impact is minimal. However, the rapid breakthrough of an adverse contaminant at a rate where it may have an adverse impact of the environment would require installation of a recovery system. Planning would need to be undertaken to ensure that the nature and type of recovery system is available in cases where such a breakthrough is likely to occur.
- Effect of above ground tailings dam in an arid environment. The monitoring bore (shown) was supposedly installed to ensure that what is pictured did not occur. Once surface precipitation of has occurred, the chemical reaction becomes irreversible and environmental damage is permanent.
- Tailings dam evaporation pond. From a distance the water looks pristine, however, crystal clear can simply mean that all solids are in soluble form due to low pH. This water dissolved a 200-litre drum in two weeks.
Monitoring programmes should be designed to recognise the phase of a leachate plume and to evaluate the impact on the ground and surface water. Although pro-active monitoring does this, it may initially cost more. However, long-term costs are considerably reduced because, once the background water chemistry, the medium chemistry and the contaminant chemistry are defined, key indicators can be selected for long-term monitoring. These are usually considerably cheaper to monitor than contaminants such as low-level detection of heavy metals.
In the last five years, a number of well published events at mines, such as the leakage from a tailings dam in South Australia and fuel storage leak in the Northern Territory, illustrate the outcome of inappropriate or non-existent groundwater monitoring. Not only was there adverse publicity, but the investigation and the remediation costs were high. What to monitor becomes a question of risk assessment and economics.
Groundwater monitoring should be undertaken to confirm that controlled seepage of leachate is behaving as predicted and that it is having little or no measurable effect on the natural ground and surface water. There should also be an action plan based on the conceptual model of the leachate plume. Groundwater monitoring, bore location and the frequency of monitoring depends on the nature of the surrounding soil or rock in which the facility is located. Location and frequency of monitoring should not be haphazard or uniformly applied, but should be tailored for local site conditions. A simplistic, local conceptual model should be derived to highlight the monitoring requirements.
When interpreting the monitoring data, focus should be centred on recognising the phase and nature of leachate breakthrough fronts and designing action responses accordingly. Constituents monitored for early warning should be pH, electrical conductivity and redox potential in the field as well as ionic balance and TDS analysed in the laboratory.
- Looking away from the dam wall, the bore (located 80 m from the wall) has a water level of 2 m, while, regionally, groundwater is at 40 m.
- Looking back at the wall, salt is precipitating along the base of the dam wall, leading to geotechnical/stability concerns. Options such as discharge trench or dewatering/extraction bores need to be implemented immediately, but should have been considered sooner.
Only after the early warning indicators have been detected should heavy metals, cyanide or organics be monitored. Ratios, particularly the ML/N cation ratio, should be used to highlight differences in the anions and cations, particularly when background water salinity is high.
© Copyright 1999 – 2004 Environmental & Earth Sciences International
All rights reserved. No part of this paper may be reproduced or copied in any form or by any means without the written permission of the publisher.