Context
Subaqueous disposal (SAD) or flooding of sulphide-rich tailings in constructed facilities is one of a number of mitigation methods used by mines. The primary mitigation mechanism resulting from subaqueous disposal is limitation of oxygen ingress into water filled pores, which greatly reduces sulphide oxidation, minimizes metal leaching and prevents the development of acidic drainage.
To date, most studies and monitoring activities related to closed SAD facilities have focused on the initial physical and geochemical performance of the facilities and the resulting chemistry of the surface water. Longer-term aspects, such as the progressive addition of natural organic matter by sedimentation on top of the tailings and the overall biological performance of such facilities, are not as well understood. In particular, a major gap in understanding relates to the biological colonization of such facilities, the health of biological communities that are established, and the influence of those communities on water and sediment geochemistry.
Literature reviews in this general area were commissioned by the Mine Environment Neutral Drainage (MEND) programme in 1993 and 2009. These reviews focused on possible means of evaluating the potential biological effects of subaqueous disposal of mine tailings, with an emphasis on the fauna and flora that had colonized the tailings impoundment. Building on these earlier reports the present study was designed (i) to provide updated information on potential diagenetic changes in submerged tailings and on biogeochemical interactions between submerged tailings and overlying aquatic communities, and (ii) based on the information obtained in part (i), to provide guidance regarding recommended tools and methodologies that could be used to predict and/or monitor the biological effects of submerged tailings.
Contaminants addressed in the present preview fall into two categories: (i) the data-rich cationic trace elements that were covered in the previous reviews (Cd, Cu, Pb, Ni and Zn) and (ii) trace elements forming oxyanions and neutral polyhydroxy species that were not dealt with in the previous reports (As, Mo, Sb, Se). Other elements, such as Al, Fe and Mn, have been considered in the context of the diagenetic changes that occur in sediments and tailings after their deposition in a subaqueous tailings facility.
Content
The present report includes an Introduction (Section 1), Literature Review (Sections 2 to 7) and a set of Recommendations (Section 8). The Literature Review first considers how waterborne and diet-borne metals interact with living organisms (Sections 2 and 3), and then integrates these two uptake vectors and considers metal bioavailability in sediments and in submerged tailings (Section 4). This latter section also includes examples of various biomonitoring approaches that can be used to assess metal bioavailability in a SAD facility. The Literature Review also addresses the relative sensitivity of different aquatic organisms to the metals of concern (Section 5) and offers an overview of the environmental chemistry of various oxyanions and their bioavailability (Section 6). The final part of the Literature Review deals with data generated from the assessment of SAD facilities in Canada and elsewhere, found in company and government reports and conference proceedings (Section 7). The Recommendations section draws upon all this information and provides guidance regarding recommended tools and methodologies that could be used to assess the biological effects of submerged tailings.
Results
Tools
In the earlier MEND reviews, various approaches or tools were presented, notably: (i) the Biotic Ligand Model (BLM) to evaluate the bioavailability of metals present in the overlying water or in the tailings pore water; (ii) the Acid Volatile Sulphide – Simultaneously Extracted Metals (AVS–SEM) model to evaluate the potential toxicity of the metals present in submerged, aged and anoxic tailings; (iii) sorptive equilibrium models to calculate the free cation concentrations in the oxic tailings pore waters.
- The BLM was already reasonably mature by 2008 and the approach has undergone considerable improvement since that time, notably in its ability to handle chronic toxicity (refinements to the metal speciation component of the model) and in its application to metal mixtures. The review in the MEND (2009) report is still an accurate and useful summary, particularly for the well-studied metals (Cd, Cu, Ni, Pb, Zn). To evaluate the bioavailability of these metals, in dissolved form (i.e., in the water that overlies submerged tailings and in the pore water residing in the tailings interstices), the BLM approach remains the recommended tool to use.
- The AVS–SEM model and the sorptive equilibrium model are both designed to yield estimates of the free metal cation concentration in sediment or tailings pore water, this being recognized as the best way to evaluate the bioavailability of a sediment-associated metal. The former applies to anoxic conditions and assumes that pore-water metal concentrations are controlled by precipitation-dissolution reactions with reactive amorphous sulphides (AVS), whereas the latter was calibrated for oxic conditions and assigns control of the pore-water metal concentrations to sorption reactions on such sorbants as Fe-, Mn-oxyhydroxides or sediment organic matter. In the 20+ years since the MEND (1993) report appeared, there has been some oscillation between the two approaches (oxic vs. anoxic control of pore-water metal concentrations), culminating in the recent emergence of a more nuanced appreciation that both sorption and precipitation reactions can intervene as the oxygenation of the surface sediment changes (e.g., seasonally or as a function of drought) and as the sediment ages.
- Linked to this general appreciation of the importance of pore-water metal concentrations as an estimate of the geochemical and biological availability of sediment-associated metals, considerable research has been devoted to understanding the post-depositional mobility of metals in sediments (reviewed in section 4.4). For several metals and oxyanions (e.g., Pb, Mo and As), there is evidence for their mobilization in the zone where reductive dissolution of iron oxyhydroxides occurs, and for their removal lower down in the sediment profile by precipitation as a solid sulphide. However, the overall influence of these diagenetic reactions on the metal concentration profiles in the sediment cores was negligible for most of the studied metals, and thus post-depositional mobility of the metal did not exert a significant effect on the observed solid-phase metal profiles in the sediment. Similar conclusions would also be expected for tailings, given the very high background concentration of metals in the solid phase.
Biomonitoring
In addition to the geochemical approaches described above, the earlier MEND reports also considered indigenous aquatic organisms, i.e. those living in the subaqueous tailings facility, and their use as biomonitor organisms; this information is updated in section 4.5 of the present report. In principle, properly chosen biomonitors could be used to determine if the metals present in the tailings are available to the indigenous plants and animals and to assess if the biomonitor organisms are metal-stressed or not. This can be done by sampling the indigenous fauna and flora at the site (‘passive’ biomonitoring) or by transplanting metal-naïve plants or animals into the water body and allowing them to interact with the tailings (‘active’ biomonitoring). In both approaches, there is an obvious need to have nearby reference sites that can be used for comparison purposes. The biomonitor species can be chosen to target either the tailings or the overlying water. An obvious advantage of the biomonitoring approach is that the bioaccumulated metal concentrations provide an estimate of the overall metal bioavailability in a given water body (integrating waterborne and diet-borne metals) – use of the organism itself obviates the need to estimate the ‘bioavailability’ of the metal in the water or in the ingested food. Measures of bioaccumulated metals have an additional potential advantage in that they can be used within the framework of a Tissue Residue Approach for toxicity assessment, where metal concentrations in whole organisms or specific organs are compared to threshold values that are known to trigger deleterious effects.
As a refinement to the Tissue Residue Approach, which normally deals with the total bioaccumulated metal concentration, it is also possible to consider the subcellular fate of the metal. In effect, the bioaccumulated metals may be detoxified, for example by sequestration in insoluble granules, by isolation in membrane-bound vesicles, or by complexation with cytosolic ligands. The form in which the bioaccumulated metal exists may also affect its fate if the plant or animal is consumed as food. Simple techniques that have the potential to differentiate between detoxified metals and biologically available metals are reviewed in section 4.5.2.
Parameters that have been used to detect metal exposure and potential effects in aquatic organisms include tissue metal concentrations (as described above), metallothionein concentrations and indicators of overall organism health, such as condition indices, growth or reproductive endpoints. Metal stress also may induce subtle effects at the subcellular level that do not necessarily translate into changes in these traditionally measured variables, but that may still have a detrimental metabolic cost. Several new ecotoxicological tools, introduced in the post-2008 period, are being used to probe for new biomarkers of metal-induced stress.
One of these tools is ‘transcriptomics’, the study of the ‘transcriptome’ of native organisms chronically exposed to metals in their environment. The transcriptome reflects the genes that are being expressed in the organism and thus will tend to vary as function of the ambient conditions and the organism’s physiological state. In this context, the transcriptome can be considered as a potential addition to the small number of metal-sensitive ‘biomarkers’ that existed prior to 2008. Using this approach, laboratory-reared animals can be exposed to water or sediment obtained from the field and changes in their transcriptome can be interpreted as a response to bioavailable contaminants present in the field sample. In a variant of this bioassay approach, laboratory-raised test organisms (e.g., fathead minnows or amphipods) can be caged in the field, at reference sites and at sites that are subject to contaminant inputs; differences in their transcriptomes can then be examined. This use of the transcriptome to identify metal-induced deleterious effects is a promising but not yet mature approach.
Metal sensitivity of different organisms
Among the organisms that attempt to colonize or could be wanted in a given SAD facility, which species are the most sensitive to metals and the most likely to be adversely affected? Since some organisms may show marked sensitivity to one metal but tolerance to others, the answer to this question will vary from one ore body to another (or from one mine operation to another), as a function of the specific metals (and their form) present in the tailings. In addition, these differences in sensitivity to various metals are not necessarily the same for all target species. To address this question, the ‘species sensitivity distributions’ or SSDs were examined for many of the metals of interest, both essential and non-essential (section 5). For each metal, the most and least sensitive species have been compiled in Table 5-1; small crustaceans (e.g., daphnids, amphipods) and unicellular algae appear frequently in the ‘most sensitive’ column, whereas the ‘least sensitive’ column includes a greater variety of organisms. The limitations of the SSD approach are also explored in section 5.3.
Oxyanions
Arsenic (As), molybdenum Mo), antimony (Sb) and selenium (Se) were chosen as representative oxyanions, none of which were considered in the previous MEND reviews (1993; 2009). These elements share a number of properties: absence of cationic forms; existence as oxyanions; existence of multiple oxidation states. As a result, their geochemical behaviour differs markedly from that of the other (cationic) elements considered in the present review.
Changes between oxidation states, i.e. ‘redox reactions’, are often slow. For example, redox couples such as As(III/V) and Sb(III/V) are frequently not at equilibrium and individual species can be found at concentrations that differ markedly from those predicted thermodynamically. However, many redox reactions can be microbiologically catalyzed, in which case the reaction rate will be determined by microbial growth and metabolic activity. For trace concentrations at pH 8, the sequence of redox reactions follows the approximate order O2 > NO3– > Mn(IV) > Se(VI) > Fe(III) >As(V) > SO4 ≈ Mo(VI).
Another commonality among this group of elements is the presence of methylated forms (for As, Sb and Se). These are not metal complexes, where the cation and the ligand are bound together by a coordinate bond, but rather true organometallic molecules where the non-metallic element ‘M’ and the methyl group are linked by covalent bonds (both as M-O-CH3 bonds and direct M-CH3 bonds). Because of the presence of a covalent bond, these methylated forms are much more stable than typical metal cation complexes. The elements also react with reduced sulphur in sediments (either to precipitate as sulphides or to adsorb onto authigenic sulphides), and in this case their behaviour mirrors that of the cationic metals considered in this review. This reaction with reduced sulphur tends to immobilize As, Sb and Mo. The biogeochemical behaviour of the oxyanions at the sediment-water or tailings-water interface has been summarized, in an element-by-element approach, in Sections 6.2 to 6.5.
In their interaction with living organisms, elements such as As, Mo, Sb and Se clearly do not conform to the Biotic Ligand Model (BLM), since they exist in natural waters as oxyanions and as neutral polyhydroxy species rather than as free metal cations. The mechanisms by which they are taken up by living organisms are less well understood than is the case for (essential) metal cations, but a number of known uptake routes are described in section 6, notably uptake via anion transporters present in the cell membrane (e.g., SeO42- via a sulphate transporter; AsO43- and SbO43- via a phosphate transporter) and by aquaporins (As(OH)30; Sb(OH)30).
Subaqueous disposal facilities
The final part of the Literature Review includes a detailed compilation of the anticipated differences between a SAD facility and a natural water body, followed by a synthesis of the data generated from the assessment of SAD facilities in Canada and elsewhere (Section 7). Based on the available data, subaqueous disposal of unoxidized tailings has met the primary objectives of limiting sulphide oxidation to within a few millimetres or centimetres of the sediment-water interface and minimizing metal/oxyanion release. Where tailings predicted to be acid-generating under aerial conditions were flooded prior to significant oxidation, post-closure metal/oxyanion levels in SAD facilities in British Columbia meet discharge criteria set by the Ministry of Environment without additional mitigation or drainage treatment. However, evidence supporting this conclusion is from studies of relatively short duration; long-term studies of the effects of biological colonization are lacking. Where underwater disposal of mine tailings has been less successful (i.e., where marked remobilization of metal cations has been observed), water quality degradation has resulted from tailings oxidation during processing, an earlier extended period of sub-aerial exposure prior to flooding, or periodic exposure to the atmosphere due to water level changes (reviewed in Section 7.3). The success in limiting metal/oxyanion release, coupled with an apparent lack of attention to aquatic reclamation, has, however, resulted in limited monitoring or study of tailings diagenesis and biological colonization in the post-closure period; there proved to be a dearth of biomonitoring data for SAD facilities.
Recommendations
The recommendations presented in Section 8 and Appendix E provide guidance regarding recommended tools and methodologies to predict and/or monitor the biological and ecological effects of submerged tailings. This section is not designed to be a detailed guide to field or laboratory work, but rather should be viewed as a general outline of how to approach the challenge of monitoring and assessing possible biological effects of the subaqueous disposal of mine tailings. A site-specific, phased approach is recommended, focusing on the properties and processes of interest. Given the paucity of data for most existing SAD facilities, an initial survey followed by some exploratory sampling and analysis will undoubtedly be necessary. In addition, field trials should be carried out with a number of the monitoring techniques that have been recommended. These techniques have been shown to be useful in metal-contaminated lakes, but they haven’t been field-tested on submerged tailings.
‘Data and knowledge gaps’ that were identified during the preparation of the present report, and during its subsequent review by the MEND steering committee, can also be found in Section 8. Chief among these was a general lack of information about the hydrobiology of SAD facilities, their inherent heterogeneity, possible impediments to becoming productive habitat, and their variability over time (diagenetic changes). To address this situation, it is recommended that a limited campaign of field measurements be carried out in a representative selection of SAD facilities, focusing on mine-made facilities where the tailings were not exposed sub-aerially before flooding or on the edge of the facility and where disposal of mine tailings ceased at least 5 years ago. Process chemicals (e.g., cyanide; xanthates) and thiosalts have been detected in active SAD facilities (i.e., in the period before mine closure), but little is known about the possible persistence of these substances once active disposal of mine tailings has ceased. Samples of the tailings pore water and from the overlying water column should be tested for the residual presence of such process chemicals.
