Monitoring moisture dynamics in multi-layer cover systems for mine tailings reclamation using autonomous and remote time-lapse electrical resistivity tomography
Abstract
The dynamics of moisture content in cover systems constructed on mining wastes were monitored at the pilot scale using 2D autonomous, remote, and noninvasive time-lapse electrical resistivity tomography combined with conventional point sensors. A methodology was proposed to process the daily hydrogeophysical datasets from 23 m long instrumented sections of covers with capillary barrier effects (CCBEs) designed to act as oxygen barriers, and covers with low saturated hydraulic conductivity layers (LSHCCs) designed to limit the water infiltration rate. Hydrogeophysical monitoring suggested that CCBEs were able to maintain high degrees of saturation in the moisture-retaining layer throughout the 1 year monitoring period, which would make it an efficient oxygen barrier. Larger spatio-temporal changes in moisture content were observed in LSHCCs and most of the low hydraulic conductivity layers remained below 85% saturation, which was attributed to the combined effect of low precipitation, rapid vegetation development, and water percolation through the cover. The methodology proposed in this pilot-scale ”proof-of-concept” study allowed the hydrogeological behavior of the cover systems to be monitored in the 23 m long instrumented sections using continuous geoelectrical data, which demonstrated that this innovative monitoring technique could be useful for geochemical and geotechnical monitoring programs in large-scale mining waste storage facilities.
Résumé
L’évolution de la teneur en eau dans les systèmes de recouvrements construits sur des rejets miniers a été surveillée à l’échelle pilote par la méthode de tomographie de résistivité électrique en 2D de manière autonome, à distance et non-invasive, combinée à des capteurs ponctuels conventionnels. Une méthodologie a été proposée pour traiter les données hydrogéophysiques quotidiennes provenant de sections instrumentées de 23 m de long de couvertures avec effets de barrière capillaire (CEBCs) conçues pour agir comme des barrières à l’oxygène, et de couvertures avec des couches à faible conductivité hydraulique saturée (LSHCCs) conçues pour limiter le taux d’infiltration de l’eau. La surveillance hydrogéophysique a suggéré que les CEBCs étaient capables de maintenir des degrés de saturation élevés dans la couche de rétention d’eau tout au long de la période de surveillance d’un an, ce qui en ferait une barrière à l’oxygène efficace. Des changements spatio-temporels plus importants de la teneur en eau ont été observés dans les LSHCCs et la plus grande partie des couches à faible conductivité hydraulique sont restées en dessous de 85% de saturation, ce qui a été attribué à l’effet combiné de faibles précipitations, d’un développement rapide de la végétation et de la percolation de l’eau à travers le recouvrement. La méthodologie proposée dans cette ”preuve de concept” à l’échelle pilote a permis de surveiller le comportement hydrogéologique des systèmes de recouvrements dans les sections instrumentées de 23 m de long à l’aide de données géoélectriques continues, ce qui a démontré que cette technique de surveillance innovante pourrait être utile pour les programmes de surveillance géochimique et géotechnique dans les aires d’entreposage de rejets miniers à grande échelle.
Introduction
Mining operations generate two types of wastes during the extraction and processing of the ore: waste rocks and tailings (Bussière 2007; Bussière and Guittonny 2021a). Waste rocks are generally coarse material with noneconomical mineralization that have been blasted and extracted from the ground to reach the ore deposit. Tailings refer to the rocks extracted from the ore deposit that have been crushed into fine particles and mixed with water and chemicals to extract the metals (Aubertin et al. 2002a, 2002b; Dimech et al. 2022). Large volumes of mining wastes are generated worldwide since the proportion of metals in the ore is generally well below 10% for most metals (down to a few grams per tons for silver, platinum, or gold), and these ore grades are expected to drop in the future due to global ore depletion and increasing demand (Mudd 2007; Rötzer and Schmidt 2018). In 2019, it was estimated that over 220 billion tons of tailings have been generated worldwide, and 50 billion tons were expected to be generated over the following 5 years (World Mine Tailings Failures). Tailings and waste rocks are generally stored in large-scale waste facilities that can extend across tens of square kilometers and measure hundreds of meters in height (Aubertin et al. 2016; Vriens et al. 2020; Kossoff et al. 2014), and have been described as ”the largest man-made structures on earth” (Bowker and Chambers 2015).
Mining wastes are associated with major environmental concerns for two main reasons: the geotechnical and geochemical instabilities of tailings storage facilities (TSFs) and waste rock piles (WRPs) (Bussière 2007; Aznar-Sánchez et al. 2018). On the one hand, the poor geotechnical properties of tailings make TSFs vulnerable to failure if the dams are not properly designed and/or exposed to extreme precipitation, earthquakes, or landslides (Azam and Li 2010); several catastrophic failures have been reported in recent years (Rotta et al. 2020). On the other hand, the sulfides generally present in mining wastes can oxidize when exposed to water and oxygen, which is commonly referred to as acid mine drainage (AMD) generation (Blowes et al. 2003; Plante et al. 2021a). If poorly controlled, AMD can have significant impacts on both surface water and groundwater, decreasing the pH below 7 and increasing the solubility of most metal species (Nordstrom et al. 2015; Rezaie and Anderson 2020).
In recent years, several reclamation techniques have been developed to manage the risk of AMD, both at the source and over the long term, by controlling water and/or oxygen fluxes from the atmosphere toward the tailings, which significantly reduces the oxidation reaction (Bussière and Guittonny 2021a). Among other approaches, the construction of covers with capillary barrier effects (CCBEs) at the surface of potentially AMD generating tailings is particularly promising to control the availability of oxygen to tailings in humid climates (Aubertin et al. 1995; Bussière et al. 2003). CCBEs are based on the capillary barrier effects that develop at the interface between fine and coarse materials under unsaturated conditions, which tend to reduce vertical water flow across this interface (Morel-Seytoux 1992). In most CCBEs, a layer made of fine materials, referred to as the moisture-retaining layer (MRL), is installed between two layers of coarser materials, referred to as the capillary break layers (Aubertin et al. 1995). The capillary break layers tend to drain rapidly because of their poor water retention capacity, whereas the MRL tends to remain nearly saturated, which in turn greatly reduces oxygen migration (Bussière et al. 2003; Aachib et al. 2004; Demers and Pabst 2021). Low saturated hydraulic conductivity covers (LSHCCs) are an alternative reclamation approach, also referred to as water infiltration barriers or impermeable barriers, which aim to limit water infiltration into the wastes (Maqsoud et al. 2021). The low saturated hydraulic conductivity layer of the LSHCC is generally made of fine-grained soils (clay or fine silt) and/or man-made materials (e.g., geomembrane or geosynthetic clay liner), with a suggested saturated hydraulic conductivity less than or equal to 10−7 cm/s (Aubertin et al. 2016; Maqsoud et al. 2021). As presented by Aubertin et al. (1995), additional layers made of coarser material are essential to optimize the performance of LSHCCs.
The performance of CCBEs and LSHCCs has been documented in the literature in recent years across different scales (e.g., numerical studies (Bussière et al. 2003; Aubertin et al. 2009), laboratory columns (Aachib et al. 1994; Kalonji-Kabambi et al. 2017; Larochelle et al. 2019), field pilot-scale cells (Bussière et al. 2007), and field large-scale cover systems (Bussière et al. 2003; Dagenais 2005). However, the performance of CCBEs and LSHCCs is dependent on the water budget at a specific mining site and on the cover design (e.g., hydrogeological properties of the material used, thickness of the layers, topography of the cover system) (Demers and Pabst 2021; Maqsoud et al. 2021). Moreover, the performance can be locally or globally affected by physical processes (e.g., extreme precipitation and droughts, freeze–thaw cycles, or preferential pathways) or biological processes (e.g., evapotranspiration from vegetation, root-water uptake, and root or burrowing animal intrusions), both in the short or long-term (MEND 2004; Rykaart et al. 2006; Bussière and Guittonny 2021b). As a result, it is generally recommended to construct one or more test cover systems at the pilot scale to assess the hydrogeological behavior under real meteorological conditions and monitor the performance in the short term (Bussière 2007; Bussière et al. 2021). More generally, reclamation cover systems installed on TSFs and WRPs should also be properly monitored over the long term to provide early detection of local or global decreases in cover system performance (MEND 2004).
The tools deployed to monitor the stability of mining wastes are mostly based on local measurements (e.g., point sensors measuring volumetric water content (VWC) or piezometers) or skin-deep surface observations (e.g., visual inspections, photogrammetry, or remote sensing) (Hui et al. 2018; Clarkson and Williams 2020). While surface observation cover large scales with relatively low spatial resolution, local measurements allow physical parameters to be monitored within a few centimeters around the sensors (Vereecken et al. 2008). As a result, many monitoring stations using dense networks of sensors might be needed to cover TSFs and WRPs, which could represent significant costs for the monitoring programs (Rykaart et al. 2006). Geophysical techniques provide a promising approach to bridge the gap between local measurements and surface observations since they allow imaging of key physical parameters in the subsurface across intermediate scales, which can range from centimetric to kilometric surveys (Binley et al. 2015; Parsekian et al. 2015). Time-lapse electrical resistivity tomography (TL-ERT) has emerged in recent years as one of the most promising geophysical techniques for subsurface monitoring (Chambers et al. 2022; Dimech et al. 2022; Dimech 2023). Indeed, TL-ERT is generally robust, cost-effective, and readily deployable for permanent, continuous, and remote monitoring for various types of applications as highlighted by the reviews from Falzone et al. (2019) and Dimech et al. (2022). Moreover, ERT has been used in the context of mining environment to characterize and monitor mining wastes (Martinez-Pagan et al. 2021; Dimech et al. 2022). However, the potential that TL-ERT represents as a large-scale complementary monitoring technique remains largely untapped for mining wastes (Dimech et al. 2022), despite recent developments of remote, permanent, and automated geoelectrical monitoring systems in many other fields of geosciences (Slater and Binley 2021; Chambers et al. 2022).
This study presents the results from a pilot-scale experimental investigation at an active mining site where permanent ERT arrays were deployed to continuously monitor moisture content in two multi-layer cover systems. This paper aims to: (i) present and validate the methodology followed to predict moisture content distribution using daily ERT measurements, (ii) assess the accuracy of ERT-predicted moisture content distribution using co-located point VWC sensors, and (iii) demonstrate how continuous remote geoelectrical monitoring can be used in conjunction with conventional techniques to assess the performance of mine tailings reclamation cover systems across large scales. To our knowledge, this survey is one of the first attempts to use permanent, automated, and remote TL-ERT for monitoring mining wastes at the pilot scale. As a result, this study provides a ”proof-of-concept” that geoelectrical monitoring is a robust technique that could be successfully combined with conventional techniques to spatially extend the monitoring of moisture dynamics in mining wastes for environmental monitoring.
Site description
Canadian Malartic Mine (CMM) is a world-class large-tonnage and low-grade gold deposit located in the Abitibi region, Quebec, Canada (48.11°N, 78.12°W) as shown in Fig. 1. It is expected that over 620 Mt of waste rocks and nearly 300 Mt of tailings will have been produced by the end of operations in 2029 (Canadian Malartic Mine 2020). These wastes will be stored in WRPs (3.7 km2) and in TSFs (6.2 km2), as shown in Fig. 1a with a satellite view of the mine (Canadian Malartic Mine 2020). The tailings have average sulfur and carbon contents of 1.2% and 0.6%, respectively. This corresponds to a neutralizing potential ratio below 2 (Plante et al. 2021a), which places the tailings in an uncertainty zone for their AMD generating potential (Plante et al. 2021b). The region is characterized by a cold and temperate continental climate with a typical annual precipitation of 900 mm (Larchevêque et al. 2014). The mean annual temperature is around 1 °C, with cold winters (mean of −17 °C in January) and hot summers (mean of 17 °C in July) (Canadian Malartic Mine 2015).
Fig. 1.
As part of its mine reclamation decision-making process, CMM conducted a failure mode and effects analysis exercise. As a result, it was decided that four large-scale field experimental cover systems would be built in order to address various identified risks, notably in terms of constructability and performance of the cover systems. Two CCBE-type and two LHSCC-type covers were constructed in 2019 and 2020 in the southwestern portion of the TSF (Fig. 1). Each cover was about 18 m wide and 280 m long, with an 80 m long flat section (referred to as ”plateau”) and a 200 m long inclined section with a 10% slope. At this location, the maximum elevation of the TSF was approximately 30 m above the natural ground. Four 23 m long sections of the field experimental multi-layer covers were instrumented with a dense network of point VWC sensors and geophysical electrodes as illustrated in Fig. 1. Two hydrogeophysical sections were located in one CCBE: in the plateau section (A1) and at the top of the slope (A2) while two sections were located in one LSHCC: at the top of the slope (B1) and at the bottom of the slope (B2). The locations of the four sections were selected in order to assess the potential of geoelectrical measurements to monitor moisture content dynamics in covers for a broad range of field conditions at the pilot scale. Both the multi-layer cover design and the slope were expected to affect the hydrogeological behavior of the field cover systems (Bussière et al. 2003). Moreover, the bottom part of the slope was revegetated in October 2020 and June 2021 using oat seeds as indicated in Fig. 1, which was also expected to play a role in the moisture content distribution.
Materials and methods
Physico-chemical, hydrogeological, and electrical characterizations
Detailed physico-chemical, hydrogeological, and electrical characterizations were performed on the samples collected on the field. In total, 44 samples of tailings were collected. 47 samples of overburden and eight samples of waste rocks were also collected. The samples were mostly collected from the 23 m long hydrogeophysical sections during the construction of the cover systems, according to a regular sampling grid. The grain size distribution (GSD) was determined using laser diffraction for the tailings and overburden samples, and a combination of sieve analysis and laser diffraction for the waste rocks (ASTM D-422). The specific gravity (Gs) was determined for all materials using a helium pycnometer (ASTM D-5550). The sulfur and carbon contents of the samples were determined using a LECO CS-2000 induction furnace (ASTM UOP-703), and were used to compute the acidity and neutralization potentials of the tailings and waste rocks materials (Plante et al. 2021b).
A systematic quality control strategy was carried out during the cover section construction to assess material compaction and gravimetric moisture content (GMC) using a nucleodensimeter (86 measurements for the tailings and 33 for the overburden). The GMC of the material used for the construction was also obtained by comparing the weight of wet samples collected on the field and after 48 h in an oven at 60 °C (ASTM D-2216). The saturated hydraulic conductivities (ksat) for the tailings and overburden materials were predicted using the modified Kozeny–Carman model (KCM) (Mbonimpa et al. 2002; Chapuis and Aubertin 2003) and determined in the laboratory using flexible wall permeameters from material samples. Lavoie-Deraspe (2019) also determined ksat of waste rocks with a similar grain size distribution using a rigid wall permeameter (ASTM D-5084). Standard Tempe cells were used to assess the water retention curves (WRC) for the tailings and overburden (ASTM D-6836) at a respective porosity of n = 0.40 and n = 0.45. Incremental steps of pressure were applied to the Tempe cells using air pressure controllers and high-pressure nitrogen bottles to drive changes in the VWC of the samples. Finally, the electrical properties of the tailings and overburden samples were characterized using a laboratory apparatus derived from classical Tempe cells and described in detail by Dimech (2023). The modified Tempe cells were used to measure VWC and bulk EC for different pressures following the methodology presented by Dimech et al. (2023), which allowed to determine petrophysical relationships for each material.
Design, instrumentation, and monitoring of experimental covers
Figure 2 presents the geometry of the multilayer covers and the location of point VWC sensors and electrodes. From bottom to top, the CCBEs (Design A) consisted of a 0.3 m thick bottom capillary break layer made of waste rocks (0–100 mm), a 1.0 m thick MRL of tailings acting as the oxygen barrier, a 0.5 m thick upper capillary break layer made of waste rocks (0–50 mm), and a layer of material that supported the growth of vegetation, made of 0.4 m of overburden overlain by 0.1 m of topsoil. Similarly, from bottom to top, the LSHCCs (Design B) consisted of a 0.3 m thick support layer made of waste rocks (0–100 mm), a 1.0 m thick low saturated hydraulic conductivity layer made of compacted overburden, a 1.0 m thick protective layer made of uncompacted overburden, and a 0.1 m thick surface layer made of topsoil. Each multi-layer cover sections were equipped with hydrogeological monitoring stations located at the center of the sections, connected to vertical profiles of six Teros 12 sensors monitoring unfrozen VWC, temperature, bulk EC (two-points method), and pore water EC (MeterGroup (2019)) every 30 min. Specific material calibration curves were obtained in the laboratory and the accuracy for calibrated VWC was ± 0.03 m3/m3 (Isabelle 2022). The Teros 12 sensors were connected to ZL6 autonomous dataloggers (MeterGroup) and data were transferred automatically to remote servers once a day. For each cover section, two longitudinal profiles of electrodes were also buried within the tailings and the overburden materials to protect the electrodes from coarse waste rocks, ensure good electrical grounding, and maximize the resolution within the tailings and compacted overburden layers, respectively. Each longitudinal profile contained 24 flat rectangular stainless-steel electrodes measuring 6 cm x 2.5 cm and separated from each other by 1 m. The hydrogeophysical datasets presented in this study extended over a 1-year period from July 2021 to August 2022 and a set of four measurements was recorded for each cover section daily. Finally, air temperature, precipitation, and relative humidity were also monitored using a meteorological station.
Fig. 2.
Construction of pilot-scale field experimental cover systems
The first step of the construction was to prepare a 1–3 m thick base layer made of coarse waste rocks (0–1000 mm) to support the cover systems using a mechanical shovel and a bulldozer. A 0.3 m thick layer of finer waste rocks (0–100 mm) was then placed to act as a transition layer between the coarse waste rocks and the overlying fine material. As illustrated by Fig. 3, the covers were constructed in several lifts from 15 to 50 cm in elevation using a mechanical shovel. Nucleodensimeter measurements were systematically carried out after the construction of each lift to ensure that proper compaction was achieved. Point VWC sensors were installed at the specified depths by excavating a small hole after each lift installation; the needles of the sensors were then inserted into the compacted material and surveyed. They were held in place by backfilling the holes with the material that was previously excavated to ensure representative compaction. The longitudinal profiles of electrodes were buried in the covers during the construction and all electrodes were surveyed. More information on the pilot-scale cells construction can be found in Dimech (2023). The 256 electrodes were connected to a PRIME system, an autonomous resistivity meter developed by the British Geological Survey (PRoactive Infrastructure Monitoring and Evaluation) (Holmes et al. 2020, 2022b; Chambers et al. 2022). The PRIME system was installed in a cabin connected to a power line and equipped with a router and an antenna to transfer ERT data to remote servers.
Fig. 3.
Time-lapse ERT data acquisition and processing
ERT measurement protocols
A Wenner-type measurement configuration was used in this study to provide a strong signal/noise ratio. The spacing between measuring electrodes ranged from 1 to 5 m and both bottom and top electrodes were used separately to carry out measurements. Each ERT datasets contained 150 measurements for each cover section, and took less than 10 min to be collected. Sensitivity analysis showed high sensitivity was achieved in the tailings layer for the CCBEs, and both in the compacted overburden and loose overburden for the LSHCCs. Lower sensitivity was observed in the waste rocks layers and little to no sensitivity was achieved in the overburden layer at the top of the CCBE. The ERT datasets presented in this study were recorded autonomously every day for the four cover sections from June 2021 to July 2022 and an additional set of reciprocal ERT measurements was recorded every day for data quality assessment (c.f. Fig. S1 presenting raw ERT datasets in Supplementary files).
ERT data error assessment and filtering
ERT datasets were recorded along with a daily set of reciprocal measurements, for which the current and potential electrodes were interchanged (Tso et al. 2017). As reported in Uhlemann et al. (2016), any difference between direct and reciprocal measurements can be used to assess ERT data quality. ERT data with reciprocal error greater than 10% were removed from the datasets (Tso et al. 2019). Moreover, Hampel filtering was applied to each time-series to identify and remove any temporal outliers (Hampel 1974). The filtered measurements were then replaced in their respective time-series using spatio-temporal inverse distance weighting interpolation to obtain ERT datasets with the same number of measurements for each snapshot (Uhlemann et al. 2016). This procedure yielded 397 full datasets, each containing 150 measurements, which corresponded to a total of ≃240 000 data points to be inverted for each cover section.
Time-lapse ERT data inversions
ERT datasets were inverted using pyGIMLi/BERT, a smoothness-constrained least-squares inversion open-access software (Rücker et al. 2017). Two-dimensional unstructured meshes were defined according to the known geometry of the covers (see Dimech et al. (2023)). The meshes contained 10 600 and 11 400 triangular cells for the CCBE and the LSHCC, respectively. Anisotropic spatial smoothing was applied within each layer to allow sharper variations of inverted bulk EC along the vertical direction than in the horizontal direction. No spatial smoothing constraint was applied at the interface between two successive layers to enable sharp contrasts between two different materials. The normalized χ2 defined in Johnson et al. (2012b) was used to assess the misfit error between modeled and measured data weighted by their respective measurement errors. Most of the χ2 values obtained by the inversions ranged between 0.6 and 1.4, which denoted that ERT datasets were appropriately fitted (Johnson et al. 2012b).
Temperature and pore water corrections
Both temperature and pore water EC can have significant effects on the subsurface bulk EC (Hayley et al. 2010; Dimech et al. 2023). This is especially true for long-term monitoring surveys in areas where significant seasonal changes in air temperature can be observed (Dimech et al. 2022). A similar observation can be made if significant changes in interstitial pore fluid geochemistry are typically monitored, such as in mining wastes. In such cases, the inverted bulk EC models need to be corrected to standard temperature and standard pore water EC to properly convert bulk EC into moisture content and avoid misinterpretation of geoelectrical monitoring results.
The inverted bulk EC distribution σ was corrected to a standard temperature Tstd of 25 °C using:
where σT std is the corrected bulk EC and δT is the fractional change in σ per degree celsius (Uhlemann et al. 2016). A value of δT = 0.02 °C−1 was used in this study, which means that bulk EC increases by a factor of 2% for a temperature increase of 1 °C (Hayley et al. 2010). The temperature measured by the Teros 12 sensors was used to determine the vertical profiles of temperature for each multi-layer cover section (c.f. Fig. S2 in Supplementary files presenting the spatio-temporal variations of temperature in the CCBE section on the plateau). This temperature model was assumed to be representative of the 23 m long instrumented cell and was used to correct the inverted bulk EC values using eq. 2. Since, the maximum amplitude of variation of temperature was close to 10 °C in the tailings layer, such a seasonal temperature variation would have affected bulk EC in the tailings up to a factor of 20% if not properly accounted for (e.g., Dimech et al. (2023)).
(1)
The inverted bulk EC was also corrected to a standard pore water EC σw std of 4.0 mS/cm using
where 
is the corrected bulk EC and σw is the pore fluid EC normalized at 25 °C, following the approach discussed by Hayley et al. (2010) for temperature correction. The pore water EC values were calculated from the moisture content, bulk EC, and temperature data recorded by the Teros 12 sensors using a semi–empirical relationship developed in Hilhorst (2000) (see Dimech et al. (2019) for details). The accuracy of pore water EC determined from this relationship was expected to be around ±20% in media with a moisture content greater than 0.10 m3/m3. The Teros 12 measurements were used to assess the spatio-temporal variations of pore water EC along a vertical profile at the center of each cover section (c.f. Fig. S2 in Supplementary files). The pore water EC model estimated at the center of the hydrogeophysical cover sections was assumed to be representative of the 23 m long cover section to be used to correct the inverted bulk EC.
(2)
is the corrected bulk EC and σw is the pore fluid EC normalized at 25 °C, following the approach discussed by Hayley et al. (2010) for temperature correction. The pore water EC values were calculated from the moisture content, bulk EC, and temperature data recorded by the Teros 12 sensors using a semi–empirical relationship developed in Hilhorst (2000) (see Dimech et al. (2019) for details). The accuracy of pore water EC determined from this relationship was expected to be around ±20% in media with a moisture content greater than 0.10 m3/m3. The Teros 12 measurements were used to assess the spatio-temporal variations of pore water EC along a vertical profile at the center of each cover section (c.f. Fig. S2 in Supplementary files). The pore water EC model estimated at the center of the hydrogeophysical cover sections was assumed to be representative of the 23 m long cover section to be used to correct the inverted bulk EC.Conversion to volumetric water content
The Archie model was used to convert the inverted bulk EC to VWC expressed in m3/m3 (Archie et al. 1942) as a previous study carried out with the same materials suggested that surface conduction could be neglected (Dimech et al. 2023). The Archie relationship is generally expressed as
where σw is the pore fluid EC, n is the porosity (-), and Sw is the saturation (VWC = n · Sw). The parameters mA and nA are unitless, commonly referred to as the cementation exponent and the saturation exponent, respectively (Glover 2016). Equation 3 can be rearranged to convert the corrected bulk EC (σcorr) at standard temperature and standard pore fluid EC (σw std) into VWC using
(3)
(4)
Results
Physico-chemical, hydrogeological, and electrical properties
Table 1 summarizes the results from the multi-component characterization approach carried out on material samples. The dry density and GSD obtained for the tailings and waste rocks materials were typical of these materials (e.g., Bussière (2007); Sylvain et al. (2019); Lavoie-Deraspe (2019), c.f., GSD results shown in Fig. S3 in Supplementary files). Notably, the 44 tailings samples shared similar properties with low standard deviations, which denoted the homogeneity of the MRL in the CCBE. By comparison, a greater variability was reported among the samples of overburden for the LSHCC. However, systematic sampling showed that the properties were randomly distributed in the LSHCC and no spatial correlation was evidenced (Dimech 2023). Systematic sampling and systematic nucleodensimeter measurements provided similar results in terms of GMC. GMC at installation was 0.21 ± 0.01 for the tailings layer and 0.27 ± 0.07 for the overburden layer. The in situ porosity calculated from nucleodensimeter data were n = 0.39 ± 0.02 and n = 0.46 ± 0.06 for the tailings and overburden layers, respectively. Notably, a slight difference in porosity was observed between the plateau and the inclined sections of the CCBE (n = 0.38 and n = 0.41, respectively).
Table 1.
The saturated hydraulic conductivity predicted from basic geotechnical properties were in good agreement with permeameter experiments (see Table 1). Similarly, WRC obtained from the Tempe cells were consistent with those determined from the modified electrical resistivity Tempe cells and allowed the determination of the unsaturated hydraulic properties such as saturated and residual VWC (θsat and θr), air-entry value (ψa), and suction at residual moisture content ψr. As expected, the waste rocks (0–100 mm) had a high saturated hydraulic conductivity (ksat around 10−1 cm/s) and low water retention capacity (ψa lower than 0.1 kPa). In this study, the waste rocks with grades 0–100 mm and 0–50 mm were supposed to exhibit similar hydrogeological behaviors. On the contrary, the fine materials exhibited lower saturated hydraulic conductivities (ksat around 10−5 and 10−6 cm/s for tailings and overburden, respectively) and higher water retention capacities (ψa around 20–30 kPa for tailings and around 60–100 kPa for overburden). Such strong contrasts between coarse and fine materials suggested that CCBEs would be likely to develop at the interface between the waste rocks and tailings for the CCBE and between the waste rocks and the compacted overburden materials for the LSHCCs (Bussière et al. 2003). The electrical resistivity Tempe cells allowed the electrical properties of the tailings and overburden materials to be determined. The parameters of Archie petrophysical models (c.f. eq. 3) were optimized to fit the experimental measurements of bulk EC and VWC in the modified Tempe cells following the methodology presented in Dimech et al. (2023). The tailings were compacted to a porosity of n = 0.40 in the Tempe cell to reproduce similar conditions as in the field CCBE (n = 0.45 for the overburden). The values mA = 1.2 and nA = 3.5 provided the best fit to the experimental data for the tailings at a reference temperature of 25 °C and a reference pore water EC of 4 mS/cm (mA = 0.5 and nA = 3.4 for the overburden).
Moisture content dynamics from the point VWC sensors
The daily precipitation and VWC dynamics measured by the point VWC sensors in the four experimental cover sections are shown in Fig. 4. As expected, VWC was much lower (typically less than 0.15 m3/m3) in the waste rocks layers (blue and purple sensors for CCBE and blue sensor for LSHCC) when compared to the fine materials. In particular, almost no temporal variations in VWC were observed in the bottom waste rocks layers, which suggested limited percolation rates from the overlying layers in the CCBE sections. VWC remained between 0.30 and 0.40 m3/m3 in the tailings layer of the CCBE section on the plateau and greater than 0.35 m3/m3 in the inclined area during the 1-year monitoring period. Each significant precipitation event (i.e., greater than 20 mm/day) was followed by a sharp increase in VWC at the top of the tailings layer in the plateau section, whereas VWC was mostly stable in the rest of the MRL and in the inclined area. VWC in the overburden layer of the CCBE remained near 0.30 m3/m3 throughout the year and VWC slightly increased after each precipitation event. A sharp decrease in VWC was observed during winter in the overburden layer of the plateau section. This was interpreted as a partial freezing of the pore water in this superficial layer, given that the Teros 12 sensors measured unfrozen water content, which was consistent with the measured temperature near 0 °C at the same location during winter. In the LSHCCs, VWC remained between 0.30 and 0.45 m3/m3 in the compacted overburden material at the top of the slope. In comparison, the compacted overburden material at the bottom of the slope was slightly drier in the hydrogeophysical section, with VWC ranging from 0.25 to 0.38 m3/m3. On the contrary, significant changes in VWC were observed in the uncompacted overburden material and at the top of the compacted overburden layers in the LSHCC, with variations of up to 0.15 m3/m3 between the wettest and driest periods. The lowest values of VWC in the uncompacted overburden layers of the LSHCCs were recorded at the end of summer 2021. VWC in the transition layer of waste rocks was higher near the bottom of the slope than near the top part of the slope. Notably, strong increases in VWC were recorded in the waste rock layers of the LSHCC following each significant precipitation event, and the temporal variations of VWC in the bottom waste rocks layers were stronger at the bottom of the slope of the LSHCC, both in terms of absolute values and amplitude of variation.
Fig. 4.
Time-lapse electrical resistivity tomography results
Spatial distribution of moisture content from inverted conductivity
Electrical resistivity tomography in the field multi-layer covers allowed the visualization of VWC dynamics across the 23 m long cover sections. Figure 5 presents the distribution of VWC obtained from inversion of ERT data for the four cover sections on 1 July 2021 (left panel) and on 28 August 2021 (right panel). As indicated by the dashed lines on Fig. 4, 1 July 2021 corresponded to a wet period, with cumulative rain of ≃ 55 mm during the previous week (blue dashed line) whereas 28 August 2021 corresponded to a dry period with cumulative rain of <1 mm during the previous two weeks and ≃40 mm during the previous month (red dashed line).
Fig. 5.
The VWC distribution predicted from geoelectrical monitoring was consistent with hydrogeological measurements for both dates shown on Fig. 5. VWC was high in the MRL composed of tailings for the CCBE (between 0.30 and 0.40 m3/m3) and of overburden for the LSHCC (between 0.25 and 0.45 m3/m3), while estimated VWC was less than 0.10 m3/m3 in the waste rocks. Higher VWC values were predicted from ERT inversions in July after the strong precipitation events, as opposed to August, which seemed to be more affected by the drought, particularly in the overburden material of the LSHCC. ERT inversions produced VWC distributions that were globally smooth in the longitudinal direction and showed stronger changes in the vertical direction (Fig. 5). This was consistent with the inversion anisotropic smoothing and with the increase of suction with elevation expected in such multi-layer covers (Bussière et al. 2007). Nonetheless, some horizontal variations in VWC were noted for the LSHCC, which might have been caused by the higher variability in hydrogeological properties for the overburden material evidenced by the systematic characterization. For instance, higher VWC values were predicted from ERT inversions at the left upslope part of the profile for the top slope LSHCC (as opposed to the right bottom slope part of the profile).
Spatio-temporal dynamics of moisture content
Figure 6 presents the VWC distribution on 25 July 2021, which was considered as a baseline for the calculation of relative variations of VWC during the last week of July (wet period in blue) and August (drier conditions indicated in red). For the CCBEs, only the top layers were affected by the precipitation with a VWC increase greater than 0.05 m3/m3 in the overburden and top waste rocks layers. The top of the tailings was also slightly affected by the precipitation with an increase in VWC around 0.05 m3/m3 in the plateau section while almost no variations were recorded in the tailings layer in the inclined area. The decrease in VWC during the dry period in the tailings was less than 0.05 m3/m3 for both CCBE sections. Greater spatio-temporal changes were observed in the LSHCCs. The largest variations in VWC were in the upper part of the uncompacted overburden layer, with a relative increase in VWC around 0.03 m3/m3 following precipitation events and a relative decrease in VWC greater than 0.05 m3/m3 during the dry period. By contrast, the 1 m thick layer of compacted overburden did not show significant variations after precipitation, particularly for the cell located at the top of the slope. Nonetheless, the decrease in VWC that was recorded at the end of August 2021 affected both the uncompacted and the compacted overburden layers. No horizontal trends in VWC spatio-temporal dynamics were evidenced from TL-ERT for any cover section although more lateral variations were observed for LSHCCs than for CCBEs.
Fig. 6.
Accuracy of ERT-predicted moisture content
The accuracy of the geoelectrical monitoring results was investigated in the four experimental cover sections by comparing the VWC measured by the Teros 12 sensors and the VWC predicted by ERT at the same locations. Following the methodology described in Dimech et al. (2023), the ERT-predicted VWC values were extracted from the 2D inverted conductivity models at the sensor location, knowing their volume of investigation (approximately 1 L). The left panel of Fig. 7 compares the temporal evolution of the measured VWC (dashed lines) and the ERT-predicted VWC (solid lines) at the location of point VWC sensors in the tailings layers for the CCBEs and in the overburden material for the LSHCCs. The right panel shows the comparison between measured VWC (horizontal axis) and ERT-predicted VWC (vertical axis). Overall, the ERT-predicted VWC matched the hydrogeological measurements well, both in absolute value and relative temporal variations. Indeed, most of the ERT-predicted VWC values were close to the black solid line that represents a perfect fit between predicted and measured VWC (right panel of Fig. 7). The difference between the two datasets was less than ± 0.05 m3/m3 and the root mean square errors ranged from 0.01 to 0.03 m3/m3.
Fig. 7.
Discussion
Hydrogeological behavior of the covers interpreted from hydrogeophysical monitoring data
Covers with capillary barrier effects (CCBEs)
The spatio-temporal dynamics of VWC in the CCBEs were consistent with the expected behavior for that type of multi-layer cover. Both hydrogeological and geophysical datasets were consistent although TL-ERT allowed the VWC measurements to be spatially extended across the 23 m long cover sections. The degree of saturation in the MRL layer made of fine tailings remained almost saturated most of the time during the 1-year monitoring period, whereas VWC was much lower in the coarser-grained waste rocks layers. This behavior suggested that CCBEs were present at the top and at the bottom of the tailings layers and efficiently reduced both water infiltration and evaporation from the MRL (e.g., Nicholson et al. (1989); Aubertin et al. (1995)). Overall, the CCBE sections investigated in this study were acting as efficient oxygen barriers since most of the MRL layer made of tailings remained near full saturation during the monitoring period of 400 days (Nicholson et al. 1989; Demers and Pabst 2021). The slight decrease of VWC observed during the summer period was limited to the top part of the MRL in the plateau section, which would likely not affect the performance of the oxygen barrier. The two sections of CCBEs investigated in this study showed similar behavior, which suggested that the slope did not greatly affect the performance of the CCBEs. Nonetheless, the MRL in the inclined section seemed to be slightly less affected by precipitation than the MRL in the plateau section, which could suggest that a fraction of the precipitation was laterally diverted within the top waste rocks layer toward the bottom of the slope (e.g., Bussière et al. (2003); Aubertin et al. (2009)). This type of reclamation covers using tailings and waste rocks is promising since no natural materials are needed for the construction (e.g., natural sand, silt, or clay from nearby borrow pits), which reduces the environmental impact of reclamation and provides logistical and economic benefits (Bussière et al. 2007; Larochelle et al. 2019).
Low saturated hydraulic conductivity covers (LSHCCs)
Hydro-geophysical results suggested greater spatio-temporal variations of VWC within the LSHCC sections compared to the CCBE sections. The lowest VWC values were observed in the overburden material at the bottom of the slope during summer and most of the compacted overburden layer remained with a degree of saturation below 80% throughout the year. This could be explained, at least in part, by the presence of vegetation in the lowest part of the slope (see Fig. 1), which might have modified the water budget by adding evapotranspiration for this area, as it has been documented for CCBEs (e.g., Bussière and Guittonny (2021b)). These observations highlighted the impact of the top waste rocks layer placed above the fine-material layer in a CCBE, which acts as an evaporation barrier (not present in the LSHCCs). In addition, the strong variations in moisture content monitored by the sensors in the waste rocks layer at the bottom of the cover suggested that a non negligible portion of the precipitation infiltrated toward the tailings. The larger variability in overburden physical properties (c.f. Table 1) did not seem to play a major role in the LSHCC hydrogeological behavior since no clear spatial trends or strong anomalies of VWC distribution were evidenced in the overburden material, as has been documented in previous work for heterogeneous materials (e.g., Beauvais et al. (2004); Bechtold et al. (2012)). These observations would be helpful in the future to assess the performance of the LSHCC sections and calibrate hydrogeological numerical models which would allow to assess the ability of the LSHCCs to reduce water infiltration toward the tailings under different hydrogeological conditions (e.g., Bussière et al. (2003); Pabst et al. (2017); Power et al. (2017)).
Advantages and limitations of TL-ERT for multi-layer cover system performance monitoring
Reliability and accuracy of TL-ERT monitoring of VWC
This study demonstrates the reliability of geoelectrical monitoring for assessing the hydrogeological behavior of multi-layer covers at the pilot scale under field conditions. The proposed methodology allowed the prediction of VWC from ERT, given that the temperature, pore water conductivity, and petrophysical relationships were properly accounted for. The ERT-predicted VWC was similar to the VWC measured with calibrated point VWC sensors, both in terms of absolute value and temporal variations, with root mean square errors between measured and predicted VWC ranging from 0.01 to 0.03 m3/m3. Although to our knowledge this study was the first of its kind for mining waste monitoring, similar accuracy values for ERT-predicted VWC have been reported for other types of applications (e.g., Garré et al. (2011); Beff et al. (2013); Dietrich et al. (2014); Fan et al. (2015)). Nonetheless, we expect a decrease in ERT-predicted accuracy at larger scales since a larger electrode spacing may be needed to cover larger sections of a TSF cover system. This study is a field-scale ”proof-of-concept” which demonstrates that TL-ERT could be applied to provide continuous and remote monitoring of the VWC in mine tailings and mine reclamation covers over the long term for environmental applications.
Spatio-temporal resolution and monitoring costs for TL-ERT
The spatial imaging capacity of ERT is also a strong advantage of this monitoring technique since wide areas can be monitored using noninvasive, continuous, and remote geoelectrical monitoring (up to several hundreds of meters) (Auken et al. 2014; Chambers et al. 2022; Holmes et al. 2022a). As discussed in Dimech et al. (2022), this perspective is promising for monitoring large structures such as mining waste storage facilities to spatially extend the performance assessment and provide complementary data to traditional local point VWC sensors. As a result, the integration of geoelectrical imaging techniques within traditional surveillance programs could allow large-scale monitoring at reduced costs (Bussière et al. 2021; Tresoldi et al. 2021). In particular for this pilot-scale study, the geophysical instrumentation costs were lower than 50 000 US$ for cables, electrodes, and data acquisition system (autonomous resistivity meter, batteries, antennas, and routers) (Dimech 2023). As discussed by Dimech (2023), this is relatively affordable considering (i) the cost of multi-layer cover systems construction (from 100 000 to >300 000 US$ per hectare, MERN (2019)), (ii) the cost of each conventional moisture content sensor and data acquisition systems (from 200 to >1 000 US$ per sensor), and (iii) the remediation or water treatment costs which would be associated with a poorly-performing multi-layer cover system (Bussière et al. 2021). In the future, ERT arrays could be deployed permanently in the storage facilities across profiles that could measure up to several hundreds of meters to yield autonomous and continuous monitoring of the moisture content (in combination with point VWC sensors for moisture content, suction, temperature, and pore water EC). In particular, the recent development of autonomous, off-grid, remote resistivity meters such as the PRIME system used in this study (Chambers et al. 2022; Holmes et al. 2022b) is likely to play a role in the development of geoelectrical monitoring applications since hundreds of electrodes can be used with a single resistivity meter, which in turn improves the spatial resolution and increases the area covered while ensuring high temporal resolution (Slater and Binley 2021; Chambers et al. 2022).
Influence of temperature, pore water conductivity, and uncertainty of inversion
Since ERT is based on low-frequency electrical measurements, the relationship between bulk electrical conductivity and VWC is also dependent on temperature and water electrical conductivity, which can vary simultaneously in the material and must be properly taken into account (Hayley et al. 2007; Ma et al. 2011; Dimech et al. 2019). Dimech et al. (2023) demonstrated that not taking into account these variations reduces the accuracy of moisture content predictions by a factor of 10. Such conclusions highlight the importance of using complementary point VWC sensors to monitor temperature and pore water conductivity at different locations, which will be needed to correct the ERT datasets (Dimech et al. 2022). However, high uncertainties could be associated with the extrapolation of local temperature and pore water EC data in the imaging domain, especially for pore water EC, which can vary sharply in mining wastes, both in space and time (e.g., Dimech et al. (2019)). Local variations of porosity in the material are also likely to occur in such cover systems (cf. Table 1), which means that the degree of saturation predicted from ERT-predicted VWC could be locally under or over-estimated. As discussed in Wagner and Uhlemann (2021), the integration of other geophysical methods could help reduce this ambiguity, as it has successfully been done for permafrost imaging using electrical and seismic co-located characterization (Hauck et al. 2011; Wagner et al. 2019; Mollaret et al. 2020). For instance, active seismic, ground penetrating radar or electromagnetic surveys could be carried out frequently where ERT arrays are installed (Wagner and Uhlemann 2021). Permanent passive seismic or fiber-optic arrays could also be installed alongside ERT arrays to carry out monitoring, although only a few examples are reported in the literature (Daley et al. 2013; Olivier et al. 2017; Pevzner et al. 2021; Wu et al. 2021).
In this study, most of the geoelectrical data processing was done independently from the hydrogeological measurements. As a result, the hydrogeological and physical constraints associated with this kind of multi-layered unsaturated media might not be fully represented by the numerical solution from the inversion, which is neither perfect nor unique (Samouëlian et al. 2005; Loke 2018) and depends on the spatio-temporal regularization constraints applied (Johnson et al. 2012a; Loke et al. 2022). For instance, some artifacts could be present in the inverted distribution of bulk EC, especially in areas of low sensitivity such as at the boundaries of the imaged volume, or due to the complex geometries of these media (Nimmer et al. 2008; Dimech et al. 2017, 2018; Hojat et al. 2020). Moreover, the inverted results might differ locally from the hydrogeological measurements because of the difference in volume of investigation. Indeed, the sensors only sample a small volume (typically less than 1 L), whereas ERT measurements provide information averaged across larger volumes (typically several m3) (Robinson et al. 2008). The latter could explain some differences between the measured and ERT-predicted VWC temporal evolution reported at some locations in this study (Fig. 7). In this regard, Dimech et al. (2022) identified several alternative data processing (e.g., coupled hydrogeo-physical inversion) and assimilation techniques (e.g., Ensemble Kalman Filters) that are promising to combine hydrogeological data alongside numerical modeling tools (e.g., hydrogeo-thermal, transport flow, and geophysical modeling) and geoelectrical datasets to provide more realistic distributions of VWC in the subsurface (Camporese et al. 2015; Kang et al. 2018, 2019; Tso et al. 2020; Isabelle 2022). Finally, some uncertainties remain associated with the petrophysical models in tailings, mainly because of the potential effects of mineralogy on solid matrix conduction and the importance of the electrical double layer on surface conduction for clay-like materials (Revil et al. 2012; Mollehuara-Canales et al. 2020).
Conclusion
This study presents the first pilot-scale application of autonomous and remote time-lapse geoelectrical monitoring of moisture dynamics within experimental engineered multi-layer cover systems on mine tailings. We presented the methodology followed to install the hydrogeophysical instruments, record daily ERT measurements, and obtain 2D distributions of bulk EC for which temperature and pore EC variations were taken into account. The petrophysical models calibrated in the laboratory were used to convert bulk EC into 2D distributions of VWC. This approach provided data that were combined with conventional moisture sensors to assess the hydrogeological behavior of the covers. In particular, the 2D distributions of moisture content were used to assess the ability of the cover systems to act as oxygen barriers to control the risk of AMD generation. This study provided a ”proof-of-concept” demonstration of how continuous geophysical monitoring can be used to help assess the performance of mine tailings reclamation covers across larger scales. Both hydrogeological and geophysical datasets were consistent, with a root mean square error between 0.01 and 0.03 m3/m3. Despite some limitations specifically identified for the application of TL-ERT in mining wastes, the high accuracy, high temporal resolution, and large-scale monitoring capacity of TL-ERT make it promising for monitoring the geochemical and geotechnical stability of mining waste storage facilities in the near future. Geoelectrical monitoring could be combined with conventional sensors to spatially extend the monitoring of these large structures to help control the risk of environmental contamination and provide early warning to prevent catastrophic failures.
Acknowledgements
The authors would like to thank the anonymous reviewers for their insightful comments and helpful suggestions which have helped to improve the structure and the content of the initial manuscript. We would like to thank Canadian Malartic Mine teams for continuous support and assistance during the construction and the monitoring of the experimental covers. We would also like to acknowledge the partners of the UQAT-Polytechnique Research Institute on Mines and Environment (RIME) for their financial support. Finally, the authors would like to thank the URSTM/UQAT staff for their assistance in the laboratory and in the field.
References
Aachib M., Aubertin M., Chapuis R.P. 1994. Column tests investigation of milling wastes properties used to build cover systems. In Proceedings of the International Land Reclamation and Mine Drainage Conference and 3rd International Conference on the Abatement of Acidic Drainage. Pittsburgh. pp. 128–137.
Aachib M., Mbonimpa M., Aubertin M. 2004. Measurement and prediction of the oxygen diffusion coefficient in unsaturated media, with applications to soil covers. Water Air Soil Pollut. 156(1): 163–193.
Archie G.E. 1942. The electrical resistivity log as an aid in determining some reservoir characteristics. Trans. AIME, 146(01): 54–62.
Aubertin M., Chapuis R., Aachib M., Bussière B., Ricard J., Tremblay L. 1995. Évaluation en laboratoire de barrières sèches construites à partir de résidus miniers. Ècole Polytechnique de Montréal.
Aubertin M., Bussière B., Bernier L. 2002a. Environnement et gestion des résidus miniers. Presses Internationales de Polytechnique, Corporation de l’École Polytechnique de Montréal, Montréal.
Aubertin M., Fala O., Bussière B., Martin V., Campos D., Gamache-Rochette A., et al. 2002b. Analyse des écoulements de l’eau en conditions non saturées dans les haldes à stériles. Symposium Rouyn-Noranda: L’Environnement et les Mines.
Aubertin M., Cifuentes E., Apithy S., Bussière B., Molson J., Chapuis R. 2009. Analyses of water diversion along inclined covers with capillary barrier effects. Can. Geotech. J. 46(10): 1146–1164.
Aubertin M., Bussière B., Pabst T., James M., Mbonimpa M. 2016. Review of the reclamation techniques for acid-generating mine wastes upon closure of disposal sites. In Geo-Chicago 2016. pp. 343–358.
Auken E., Doetsch J., Fiandaca G., Christiansen A.V., Gazoty A., Cahill A.G., Jakobsen R. 2014. Imaging subsurface migration of dissolved co2 in a shallow aquifer using 3-d time-lapse electrical resistivity tomography. J. Appl. Geophys. 101: 31–41.
Azam S., Li Q. 2010. Tailings dam failures: a review of the last one hundred years. Geotech. News, 28(4): 50–54.
Aznar-Sánchez J., García-Gómez J., Velasco-Muñoz J., Carretero-Gómez A. 2018. Mining waste and its sustainable management: advances in worldwide research. Minerals, 8(7): 284.
Beauvais A., Ritz M., Parisot J.C., Bantsimba C., Dukhan M. 2004. Combined ert and gpr methods for investigating two-stepped lateritic weathering systems. Geoderma, 119(1-2): 121–132.
Bechtold M., Vanderborght J., Weihermüller L., Herbst M., Günther T., Ippisch O., et al. 2012. Upward transport in a three-dimensional heterogeneous laboratory soil under evaporation conditions. Vadose Zone J. 11(2).
Beff L., Günther T., Vandoorne B., Couvreur V., Javaux M. 2013. Three-dimensional monitoring of soil water content in a maize field using electrical resistivity tomography. Hydrol. Earth Syst. Sci. 17(2): 595–609.
Binley A., Hubbard S.S., Huisman J.A., Revil A., Robinson D.A., Singha K., Slater L.D. 2015. The emergence of hydrogeophysics for improved understanding of subsurface processes over multiple scales. Water Resour. Res. 51(6): 3837–3866.
Blowes D., Ptacek C., Jambor J., Weisener C. 2003. The geochemistry of acid mine drainage. Environ. Geochem. 9: 149–204.
Bowker L.N., Chambers D.M. 2015. The risk, public liability, & economics of tailings storage facility failures. Earthwork Act. pp. 1–56.
Bussière B. 2007. Colloquium 2004: Hydrogeotechnical properties of hard rock tailings from metal mines and emerging geoenvironmental disposal approaches. Can. Geotech. J. 44(9): 1019–1052.
Bussière B., Guittonny M. 2021a. Hard rock mine reclamation: from prediction to management of acid mine drainage. CRC Press.
Bussière B., Guittonny M. 2021b. Long-term evolution of reclamation performance. In Hard rock mine reclamation: from prediction to management of acid mine drainage. CRC Press.
Bussière B., Aubertin M., Chapuis R.P. 2003. The behavior of inclined covers used as oxygen barriers. Can. Geotech. J. 40(3): 512–535.
Bussière B., Maqsoud A., Aubertin M., Martschuk J., McMullen J., Julien M. 2003. Results from the monitoring program at the LTA site: hydraulic behavior of the cover. Proceedings of the 105th Annual General Meeting of CIM–ICM, Montréal, QC, pp. 4–7.
Bussière B., Aubertin M., Mbonimpa M., Molson J.W., Chapuis R.P. 2007. Field experimental cells to evaluate the hydrogeological behaviour of oxygen barriers made of silty materials. Can. Geotech. J. 44(3): 245–265.
Bussière B., Pabst T., Boulanger-Martel V., Guittonny M., Plante B., Neculita C.M., et al. 2021. Monitoring the performance of mine site reclamation. Hard rock mine reclamation: from prediction to management of acid mine drainage.
Camporese M., Cassiani G., Deiana R., Salandin P., Binley A. 2015. Coupled and uncoupled hydrogeophysical inversions using ensemble k alman filter assimilation of ert-monitored tracer test data. Water Resour. Res. 51(5): 3277–3291.
Canadian Malartic Mine 2015 Canadian malartic reclamation plan - revision 2. p. 127. Available from https://gestim.mines.gouv.qc.ca/documents/009_PDR_Canadian_Malartic_MAJ2015_extension_2016-04-14_0000055515.pdf.
Canadian Malartic Mine 2020 Canadian malartic reclamation plan - revision 3. p. 3068.
Chambers J., Holmes J., Whiteley J., Boyd J., Meldrum P., Wilkinson P., et al. 2022. Long-term geoelectrical monitoring of landslides in natural and engineered slopes. Lead. Edge, 41(11): 768–776.
Chapuis R.P., Aubertin M. 2003. On the use of the Kozeny Carman equation to predict the hydraulic conductivity of soils. Can. Geotech. J. 40(3): 616–628.
Clarkson L., Williams D. 2020. Critical review of tailings dam monitoring best practice. Int. J. Min. Reclam. Environ. 34(2): 119–148.
Dagenais A.M. 2005. Techniques de contrôle du drainage minier acide basées sur les effets capillaires. Ph.D. thesis, École Polytechnique.
Daley T.M., Freifeld B.M., Ajo-Franklin J., Dou S., Pevzner R., Shulakova V., et al. 2013. Field testing of fiber-optic distributed acoustic sensing (das) for subsurface seismic monitoring. Lead. Edge, 32(6): 699–706.
Demers I., Pabst T. 2021. Covers with capillary barrier effects. In Hard rock mine reclamation: from prediction to management of acid mine drainage. p. 167.
Dietrich S., Weinzettel P.A., Varni M. 2014. Infiltration and drainage analysis in a heterogeneous soil by electrical resistivity tomography. Soil Sci. Soc. Am. J. 78(4): 1153–1167.
Dimech A. 2023. Application of electrical resistivity tomography for monitoring water content in mining reclamation cover systems [Application de la tomographie de résistivité électrique pour la surveillance de la teneur en eau volumique dans les ouvrages de restauration minière]. Ph.D. thesis.
Dimech A. 2024. Monitoring moisture dynamics in multi-layer cover systems for mine tailings reclamation using autonomous and remote time-lapse electrical resistivity tomography. Mendeley Data V1.
Dimech A., Chouteau M., Chou E., Aubertin M., Martin V., Bussière B., Plante B. 2017. Monitoring water infiltration in an experimental waste rock pile with time-lapse ert and multi-parameter data collection. Symposium on the Application of Geophysics to Engineering and Environmental Problems 2017. pp. 195–203.
Dimech A., Chouteau M., Bussière B., Martin V., Aubertin M., Plante B. 2018. 3D time-lapse geoelectrical monitoring of moisture content in an experimental waste rock pile: validation using hydrogeological data. FastTIMES Special Issue Mine Site and Mining Geophysics.
Dimech A., Chouteau M., Aubertin M., Bussière B., Martin V., Plante B. 2019. Three-dimensional time-lapse geoelectrical monitoring of water infiltration in an experimental mine waste rock pile. Vadose Zone Journal, 18(1).
Dimech A., Cheng L., Chouteau M., Chambers J., Uhlemann S., Wilkinson P., et al. 2022. A review on applications of time-lapse electrical resistivity tomography over the last 30 years : perspectives for mining waste monitoring. Surv. Geophys.
Dimech A., Isabelle A., Sylvain K., Liu C., Cheng L., Bussière B., et al. 2023. A multiscale accuracy assessment of moisture content predictions using time-lapse electrical resistivity tomography in mine tailings. Sci. Rep. 13(1): 20922.
Dubois-Roy F. 2021. Caractérisation hydrogéotechnique préliminaire de la cellule expérimentale de mort-terrain (2m) de la mine Canadian malartic. Research Institute of Mines and Environment.
Falzone S., Robinson J., Slater L. 2019. Characterization and monitoring of porous media with electrical imaging: a review. Transp. Porous Media, 130(1): 251–276.
Fan J., Scheuermann A., Guyot A., Baumgartl T., Lockington D.A. 2015. Quantifying spatiotemporal dynamics of root-zone soil water in a mixed forest on subtropical coastal sand dune using surface ert and spatial tdr. J. Hydrol. 523: 475–488.
Garré S., Javaux M., Vanderborght J., Pagès L., Vereecken H. 2011. Three-dimensional electrical resistivity tomography to monitor root zone water dynamics. Vadose Zone J. 10(1): 412–424.
Glover P.W. 2016. Archie’s law–a reappraisal. Solid Earth, 7(4): 1157–1169.
Hampel F.R. 1974. The influence curve and its role in robust estimation. J. Am. Statistical Assoc. 69(346): 383–393.
Hauck C., Böttcher M., Maurer H. 2011. A new model for estimating subsurface ice content based on combined electrical and seismic data sets. Cryosphere, 5(2): 453–468.
Hayley K., Bentley L.R., Gharibi M., Nightingale M. 2007. Low temperature dependence of electrical resistivity: Implications for near surface geophysical monitoring. Geophys. Res. Lett. 34(18).
Hayley K., Bentley L., Pidlisecky A. 2010. Compensating for temperature variations in time-lapse electrical resistivity difference imaging. Geophysics, 75(4): WA51–WA59.
Hilhorst M. 2000. A pore water conductivity sensor. Soil Sci. Soc. Am. J. 64(6): 1922–1925.
Hojat A., Arosio D., Ivanov V.I., Loke M.H., Longoni L., Papini M., et al. 2020. Quantifying seasonal 3D effects for a permanent electrical resistivity tomography monitoring system along the embankment of an irrigation canal. Near Surf. Geophys. 18: 427–443.
Holmes J., Chambers J., Meldrum P., Wilkinson P., Boyd J., Williamson P., et al. 2020. Four-dimensional electrical resistivity tomography for continuous, near-real-time monitoring of a landslide affecting transport infrastructure in british columbia, canada. Near Surf. Geophys.
Holmes J., Chambers J., Wilkinson P., Dashwood B., Gunn D., Cimpoiaşu M., et al. 2022a. 4d electrical resistivity tomography for assessing the influence of vegetation and subsurface moisture on railway cutting condition. Eng. Geol. 307: 106790.
Holmes J., Chambers J., Wilkinson P., Meldrum P., Cimpoiaşu M., Boyd J., et al. 2022b. Application of petrophysical relationships to electrical resistivity models for assessing the stability of a landslide in british columbia, canada. Eng. Geol. 301: 106613.
Hui S., Charlebois L., Sun C. 2018. Real-time monitoring for structural health, public safety, and risk management of mine tailings dams. Canadian J. Earth Sci. 55(3): 221–229.
Isabelle A. 2022. Hydrogeophysical data assimilation using Ensemble Kalman filters to predict moisture content in mining reclamation covers [Assimilation de données hydrogéophysiques par filtre de Kalman d’ensemble pour l’estimation de la teneur en eau en restauration minière]. Master’s thesis, École Polytechnique de Montréal.
Johnson T.C., Slater L.D., Ntarlagiannis D., Day-Lewis F.D., Elwaseif M. 2012a. Monitoring groundwater-surface water interaction using time-series and time-frequency analysis of transient three-dimensional electrical resistivity changes. Water Resour. Res. 48(7).
Johnson T.C., Versteeg R.J., Rockhold M., Slater L.D., Ntarlagiannis D., Greenwood W.J., Zachara J. 2012b. Characterization of a contaminated wellfield using 3d electrical resistivity tomography implemented with geostatistical, discontinuous boundary, and known conductivity constraints. Geophysics, 77(6): EN85–EN96.
Kalonji-Kabambi A., Bussière B., Demers I. 2017. Hydrogeological behaviour of covers with capillary barrier effects made of mining materials. Geotech. Geol. Eng. 35(3): 1199–1220.
Kang X., Shi X., Deng Y., Revil A., Xu H., Wu J. 2018. Coupled hydrogeophysical inversion of dnapl source zone architecture and permeability field in a 3d heterogeneous sandbox by assimilation time-lapse cross-borehole electrical resistivity data via ensemble kalman filtering. J. Hydrol. 567: 149–164.
Kang X., Shi X., Revil A., Cao Z., Li L., Lan T., Wu J. 2019. Coupled hydrogeophysical inversion to identify non-gaussian hydraulic conductivity field by jointly assimilating geochemical and time-lapse geophysical data. J. Hydrol. 578: 124092.
Kossoff D., Dubbin W., Alfredsson M., Edwards S., Macklin M., Hudson-Edwards K.A. 2014. Mine tailings dams: characteristics, failure, environmental impacts, and remediation. Appl. Geochem. 51: 229–245.
Larchevêque M., Desrochers A., Bussière B., Cimon D. 2014. Planting trees in soils above non-acid-generating wastes of a boreal gold mine. Ecoscience, 21(3-4): 217–231.
Larochelle C.G., Bussière B., Pabst T. 2019. Acid-generating waste rocks as capillary break layers in covers with capillary barrier effects for mine site reclamation. Water Air Soil Poll. 230(3): 1–16.
Lavoie-Deraspe J. 2019. Étude du comportement hydrogéologique de couvertures avec effets de barrière capillaire sur une halde à stériles de grande dimension. Master’s thesis, Polytechnique Montréal.
Loke M., Wilkinson P., Chambers J., Uhlemann S., Dijkstra T., Dahlin T. 2022. The use of asymmetric time constraints in 4-d ert inversion. J. Appl. Geophys. 197: 104536.
Loke M.H. 2018. Tutorial: 2D and 3D electrical imaging surveys.
Ma R., McBratney A., Whelan B., Minasny B., Short M. 2011. Comparing temperature correction models for soil electrical conductivity measurement. Precis. Agric. 12(1): 55–66.
Maqsoud A., Bussière B., Mbonimpa M. 2021. Low saturated hydraulic conductivity covers. In Hard rock mine reclamation: from prediction to management of acid mine drainage.
Martinez-Pagan P., Gómez-Ortiz D., Martín-Crespo T., Martín-Velázquez S., Martinez-Segura M. 2021. Electrical resistivity imaging applied to tailings ponds: an overview. Mine Water Environ. 1–13.
Mbonimpa M., Aubertin M., Chapuis R., Bussière B. 2002. Practical pedotransfer functions for estimating the saturated hydraulic conductivity. Geotech. Geol. Eng. 20(3): 235–259.
MEND. 2004. Project 2.21.4 - design, construction and performance monitoring of cover systems for waste rock and tailings. Canadian Mine Environment Neutral Drainage Program.
MERN. 2019. Plan de travail 2019-2020: Restauration des sites miniers abandonnés. Gouvernement du Qu.bec. Available from https://numerique.banq.qc.ca/patrimoine/details/52327/4032549?docref=jd_emZ7itVQKQY3u4wHx4w.
MeterGroup. 2019. Teros 12 datasheet.
Mollaret C., Wagner F.M., Hilbich C., Scapozza C., Hauck C. 2020. Petrophysical joint inversion applied to alpine permafrost field sites to image subsurface ice, water, air, and rock contents. Front. Earth Sci. 8.
Mollehuara-Canales R., Kozlovskaya E., Lunkka J.P., Guan H., Banks E., Moisio K. 2020. Geoelectric interpretation of petrophysical and hydrogeological parameters in reclaimed mine tailings areas. J. Appl. Geophys. 181: 104139.
Morel-Seytoux H. 1992. The capillary barrier effect at the interface of two soil layers with some contrast in properties. Hydrowar report 92.4. Hydrology Days Publications, 57: 94027–3926.
Mudd G.M. 2007. Global trends in gold mining: Towards quantifying environmental and resource sustainability. Resour. Policy, 32(1-2): 42–56.
Nicholson R.V., Gillham R.W., Cherry J.A., Reardon E.J. 1989. Reduction of acid generation in mine tailings through the use of moisture-retaining cover layers as oxygen barriers. Can. Geotech. J. 26(1): 1–8.
Nimmer R.E., Osiensky J.L., Binley A.M., Williams B.C. 2008. Three-dimensional effects causing artifacts in two-dimensional, cross-borehole, electrical imaging. J. Hydrol. 359(1-2): 59–70.
Nordstrom D.K., Blowes D.W., Ptacek C.J. 2015. Hydrogeochemistry and microbiology of mine drainage: an update. Appl. Geochem. 57: 3–16.
Olivier G., Brenguier F., de Wit T., Lynch R. 2017. Monitoring the stability of tailings dam walls with ambient seismic noise. Lead. Edge, 36(4): 350a1–350a6.
Pabst T., Aubertin M., Bussière B., Molson J. 2017. Experimental and numerical evaluation of single-layer covers placed on acid-generating tailings. Geotech. Geol. Eng. 35(4): 1421–1438.
Parsekian A., Singha K., Minsley B.J., Holbrook W.S., Slater L. 2015. Multiscale geophysical imaging of the critical zone. Rev. Geophys. 53(1): 1–26.
Pevzner R., Isaenkov R., Yavuz S., Yurikov A., Tertyshnikov K., Shashkin P., et al. 2021. Seismic monitoring of a small co2 injection using a multi-well das array: Operations and initial results of stage 3 of the co2crc otway project. Int. J. Greenhouse Gas Control, 110: 103437.
Plante B., Schudel G., Benzaazoua M. 2021a. Generation of acid mine drainage. Hard rock mine reclamation: from prediction to management of acid mine drainage.
Plante B., Schudel G., Benzaazoua M. 2021b. Prediction of acid mine drainage. In Hard rock mine reclamation: from prediction to management of acid mine drainage.
Power C., Ramasamy M., MacAskill D., Shea J., MacPhee J., Mayich D., et al. 2017. Five-year performance monitoring of a high-density polyethylene (hdpe) cover system at a reclaimed mine waste rock pile in the sydney coalfield (nova scotia, canada). Environ. Sci. Pollut. Res. Int. 24(34): 26744–26762.
Rücker C., Günther T., Wagner F.M. 2017. pyGIMLi: an open-source library for modelling and inversion in geophysics. Comput. Geosci. 109: 106–123.
Revil A., Karaoulis M., Johnson T., Kemna A. 2012. Some low-frequency electrical methods for subsurface characterization and monitoring in hydrogeology. Hydrogeol. J. 20(4): 617–658.
Rezaie B., Anderson A. 2020. Sustainable resolutions for environmental threat of the acid mine drainage. Sci. Total Environ. 137211.
Robinson D., Campbell C., Hopmans J., Hornbuckle B.K., Jones S.B., Knight R., et al. 2008. Soil moisture measurement for ecological and hydrological watershed-scale observatories: a review. Vadose Zone J. 7(1): 358–389.
Rotta L.H.S., Alcantara E., Park E., Negri R.G., Lin Y.N., Bernardo N., et al. 2020. The 2019 brumadinho tailings dam collapse: possible cause and impacts of the worst human and environmental disaster in brazil. Int. J. Appl. Earth Obs. Geoinf. 90: 102119.
Rötzer N., Schmidt M. 2018. Decreasing metal ore grades—is the fear of resource depletion justified? Resources, 7(4): 88.
Rykaart M., Hockley D., Michel Noel M.P. 2006. Findings of international review of soil cover design and construction practices for mine waste closure. In International Conference on Acid Rock Drainage (ICARD). St. Louis, MO.
Samouëlian A., Cousin I., Tabbagh A., Bruand A., Richard G. 2005. Electrical resistivity survey in soil science: a review. Soil Tillage Res. 83(2): 173–193.
Slater L., Binley A. 2021. Advancing hydrological process understanding from long-term resistivity monitoring systems. Wiley Water, 8(3): e1513.
Sylvain K., Pabst T., Dimech A. 2019. Waste rock valorization in monolayer covers with elevated water table. In Proceedings of GeoEnvironmental Engineering 2019 Concordia Montréal.
Tresoldi G., Hojat A., Zanzi L., et al. 2021. G. Re. Ta solution to monitor the integrity of tailings dams. In 7th International Conference on Tailings Management. pp. 1–8.
Tso C.H.M., Kuras O., Wilkinson P.B., Uhlemann S., Chambers J.E., Meldrum P.I., et al. 2017. Improved characterisation and modelling of measurement errors in electrical resistivity tomography (ert) surveys. J. Appl. Geophys. 146: 103–119.
Tso C.H.M., Kuras O., Binley A. 2019. On the field estimation of moisture content using electrical geophysics: the impact of petrophysical model uncertainty. Water Resour. Res. 55(8): 7196–7211.
Tso C.H.M., Johnson T.C., Song X., Chen X., Kuras O., Wilkinson P., et al. 2020. Integrated hydrogeophysical modelling and data assimilation for geoelectrical leak detection. J. Contam. Hydrol. 234: 103679.
Uhlemann S., Smith A., Chambers J., Dixon N., Dijkstra T., Haslam E., et al. 2016. Assessment of ground-based monitoring techniques applied to landslide investigations. Geomorphology, 253: 438–451.
Vereecken H., Huisman J., Bogena H., Vanderborght J., Vrugt J., Hopmans J. 2008. On the value of soil moisture measurements in vadose zone hydrology: a review. Water Resour. Res. 44(4).
Vriens B., Plante B., Seigneur N., Jamieson H. 2020. Mine waste rock: Insights for sustainable hydrogeochemical management. Minerals, 10(9): 728.
Wagner F., Mollaret C., Günther T., Kemna A., Hauck C. 2019. Quantitative imaging of water, ice and air in permafrost systems through petrophysical joint inversion of seismic refraction and electrical resistivity data. Geophys. J. Int. 219(3): 1866–1875.
Wagner F.M., Uhlemann S. 2021. An overview of multimethod imaging approaches in environmental geophysics. Adv. Geophys. 62: 1–72.
Wu R., Martin V., McKenzie J.M., Broda S., Bussière B., Selker J., Aubertin M. 2021. Fiber optic measurements of soil moisture in a waste rock pile. Groundwater, 59(4): 549–561.
Supplementary material
Supplementary Material 1 (PDF / 1.28 MB).
- Download
- 1.29 MB
Information & Authors
Information
Published In
Canadian Geotechnical Journal
Volume 61 • Number 11 • November 2024
Pages: 2505 - 2522
History
Received: 5 October 2023
Accepted: 20 February 2024
Accepted manuscript online: 8 April 2024
Version of record online: 23 August 2024
Copyright
© 2024 Dimech, Bussière, Cheng, Chouteau, Fabien-Ouellet, Chevé, Isabelle, and British Geological Survey © UKRI. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
Data Availability Statement
The datasets presented in this study are freely available at the Mendeley Data repository at doi:10.17632/bcnn88fw7z.1, an open-source online data repository hosted with CC BY 4.0 licence (Dimech 2024).
Key Words
Mots-clés
Authors
Author Contributions
Conceptualization: AD, BB, LC
Data curation: AD
Formal analysis: AD
Funding acquisition: AD, BB, LC, MC, GF
Investigation: AD
Methodology: AD, BB, LC
Project administration: BB, LC, MC
Resources: BB, LC, MC, GF
Supervision: BB, LC, MC
Software: AD
Validation: AD
Visualization: AD
Writing – original draft: AD, BB, LC
Writing – review & editing: AD, BB, LC, MC, GF, NC, AI, PW, PM, JC
Competing Interests
The authors declare no conflict of interest.
Funding Information
Université du Québec en Abitibi-Témiscamingue Foundation: Jean-Descarreaux merit scholarship
Natural Sciences and Engineering Research Council of Canada: Collaborative Research and Development
RIME industrial partners, Canadian Malartic Mine, IAMGOLD, RTFT and Newmont.
Fonds de recherche du Québec – Nature et technologies: Merit scholarship program for foreign students
Canadian Exploration Geophysical Society: Bourse des pionniers de la géophysique québéc.
Society of Exploration Geophysicists
Funding for this study was provided by the Collaborative Research and Development Grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the RIME industrial partners, CMM, IAMGOLD, RTFT, and Newmont. AD acknowledges the financial support from the ”Merit scholarship program for foreign students” (PBEEE) by the Fonds de recherche du Québec – Nature et Technologies (FRQNT), the support from the ”Bourse des pionniers de la géophysique québécoise” scholarship by the Canadian Exploration Geophysical Society (KEGS Foundation), the support from the Society of Exploration Geophysicists (SEG), and from the Jean-Descarreaux merit scholarship in mining environment (UQAT Foundation).
Metrics & Citations
Metrics
Other Metrics
Citations
Cite As
Adrien Dimech, Bruno Bussière, LiZhen Cheng, Michel Chouteau, Gabriel Fabien-Ouellet, Nathalie Chevé, Anne Isabelle, Paul Wilkinson, Philip Meldrum, and Jonathan Chambers. 2024. Monitoring moisture dynamics in multi-layer cover systems for mine tailings reclamation using autonomous and remote time-lapse electrical resistivity tomography. Canadian Geotechnical Journal.
61(11): 2505-2522. https://doi.org/10.1139/cgj-2023-0112
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
There are no citations for this item
View Options
View options
Login options
Check if you access through your login credentials or your institution to get full access on this article.