Clayton Larkins, Geological Survey of Finland GTK, P.O.Box 1237, FI-70211, Kuopio, FINLAND, email: clayton.larkins(at)gtk.fi
The Stekenjokk mine site in the Caledonian mountains of Lapland, Sweden is above the tree line at 800 m elevation and has a mean annual temperature of 0°Celsius. Operations at the mine lasted from 1976 to 1988, during which time approximately 8 million tons of metal sulphide ore was mined from Caledonian, strata bound volcanogenic host rock and concentrated on-site (Zachrisson 1971, Broman & Göransson 1994). The orerock included pyrite, chalcopyrite, sphalerite, galena, pyrrhotite, covellite and minor amounts of arsenopyrite (Holmström & Öhlander 1998). Ore processing included grinding in a two stage autogeneous mill and ore concentrate recovery using froth flotation (Eriksson et al. 2001). Mining was conducted primarily underground, as cut and fill, using the coarse fraction of tailings to fill abandoned mine workings. The fine tailings fraction, mostly silt and clay with grain size < 60 µm, was deposited as a slurry in a 110 ha tailings and clarification pond (Eriksson et al. 2001). Throughout the course of mine operations a total of 4.4 million tons of sulphide rich tailings were deposited in the tailings pond on site (Broman & Göransson 1994).
The pond tailings were assessed to be net acid producing, with high sulphide and relatively high metals concentrations (Eriksson et al. 2001). Tailings solids were composed of approximately 35% sulphides and 5 to 15% buffering material by weight (Holmström & Öhlander 1998). They also contained 0.65% zinc, 0.23% copper, 0.15% lead, and 0.14% arsenic by weight (Ljungberg et al. 1997).
During mine operations measures were taken to limit environmental impacts to the surrounding area, including water recycling and controlling the pH of tailings (Broman & Göransson 1994). From 1978 to 1987 the tailings pond discharge included average annual sulphate concentrations that ranged from 129 to 444 mg/l, zinc concentrations from 0.127 to 0.347 mg/l, and pH that varied from 7 to 8.4. Zinc loading in discharge was approximately 800 kg per year throughout the course of mine operations (Holmström & Öhlander 1998). Mine closure objectives were to prevent the site from becoming a major source of metals and acid emissions to the environment, to remove potentially hazardous surface facilities, and to integrate the site into the natural surroundings (Broman & Göransson 1994).
Mine decommissioning was completed in 1991 and included Sweden’s first application of a water cover in a waste disposal facility as a tailings reclamation measure (Bjelkevik 2005). Decomissioning also included reclamation of surface waste rock dumps, closure of a small open pit, and removal of surface structures. Permanent flooding of the tailings pond was assessed to be the safest, most effective and cost efficient reclamation strategy (Broman & Göransson 1994). Additional construction on the tailings pond was required to facilitate a permanent water cover for subaqueous disposal. Construction measures included raising and reinforcing the impoundment dams, adding an erosion resistant spillway, constructing water breaks across the pond surface, and capping tailings with coarse material in shallow areas of the pond. Construction materials for these measures were in-part sourced from the waste rock dumps on site (Broman & Göransson 1994).
The subaqueous disposal of tailings at Stekenjokk has been reported to exceed expectations for limiting ARD emissions (Lindvall 2005). At the time of decommissioning, approximately 5% of pond tailings were partially oxidized (Homlström & Öhlander 1998). Additionally, metals and sulphates in the tailings pore water resulted from significant quantities of CuSO4, SO2, and Ca(OH)2 that were added to concentrator process water during the final year of operation. Water discharged from the impoundment in 1988 had average annual concentrations of SO42-, Zn, Pb, and Cu of 293, 0.180, 0.015, and 0.25 mg/l, respectively (Holmström & Öhlander 1998). As part of the closure process, the pond surface water was largely drained to allow construction on the impoundment dykes. When the impoundment was reflooded, sulphate concentration in the free water column had decreased to approximately 60 mg/l (Eriksson et al. 2001). Zinc concentrations in the pond effluent have subsequently decreased 90% since the time of decommissioning (Lindvall 2005). Reintegration of the site into the local environment is considered successful, both on the basis of aesthetics and re-established arctic char populations in the pond (Lindvall 2005). Monitoring and maintenance at the site is ongoing (Bjelkevik 2005).
Mine closure objectives
The primary mine closure objectives, as described by Broman & Göransson (1994), were to
- prevent the site from becoming a major source of ARD emissions. This included the goal of limiting the combined discharge of Pb, Zn and Cu from the tailings and clarification pond to below 800 kg per year (Lindvall 2005).
- remove mine-related facilities that could be hazardous to humans. In addition to the tailings pond, the mine site included a small open pit mine, multiple rock dumps, and a dyke that held the mine’s raw water.
- in general, adapt the mine area to the state of its natural surroundings.
Evaluation of alternatives
The primary focus for remediation alternatives was to address the oxidation of sulphide minerals in the tailings and clarification pond. At the Stekenjokk site the following four remediation alternatives were evaluated: permanently flooding the tailings pond, capping the tailings with a dry cover, tailings depyritization, and tailings buffering. As described by Broman and Göransson (1994), flooding and use of a dry cover are means of preventing sulphide oxidation by limiting oxygen availability within the tailings, while depyritization and buffering mitigate the effects of oxidation by altering the tailings composition.
Both buffering and dypyritization would require implementation during tailings processing to alter the bulk composition of the tailings. Buffering could be accomplished by incorporating olivine, available if quarried from the site, into the milling process. Depyritization could be accomplished through pyrite flotation during ore processing. (Broman & Göransson 1994). Both of these methods were determined to be infeasible at the Stekenjokk site, because there was insufficient time between decommissioning planning and the closure of milling operations to accommodate the required composition adjustments.
The dry cover alternative was evaluated on the basis of using locally sourced moraine material for construction. However, due to limited construction material availability, the construction of a dry cover was determined to be prohibitively costly. Creating a water cover by permanently flooding the impoundment was determined to be the safest, most effective, and most efficient method for impoundment reclamation (Broman & Göransson 1994).
Closure technology and design
An evaluation of alternative methods to meet the mine closure objectives revealed that permanent subaqueous disposal achieved by flooding the tailings impoundment was the most effective and cost efficient reclamation alternative. The largest design considerations for implementation of this method included the ability to sustain a permanent water cover, even through extreme drought conditions, and to ensure the submerged tailings remained undisturbed to prevent resuspension (Broman & Göransson 1994).
Contractors were hired to evaluate site specific water balance conditions through detailed hydrologic and hydrogeological site studies. The final pond design drew on geochemical testing and modelling results to ensure that the water cover would facilitate acceptable discharge water quality to the receiving stream (ICMM 2006). The design ensured adequate water balance in the event of a 1,000 year drought, and that the pond would not develop a complete ice cover [Knutsson 1989 (as referred to in Bjelkevik 2005), Broman & Göransson 1994].
Decommissioning of the tailings pond required lowering the impoundment water table to enable construction. Construction measures included raising and stabilizing the impoundment dykes and constructing an erosion resistant spillway. Dyke walls were raised 3 meters and stabilized to sustain an intensity 6 Richter scale earthquake (Broman & Göransson 1994). Stabilization included bringing dyke slope angles to 1:3 using sulphide free waste rock and constructing breakwaters using sulphide bearing waste rock to inhibit erosion on the inner wall of dyke (Broman & Göransson 1994). To achieve the design water cover depth, 90,000 m3 of tailings were redistributed from shallow to deeper areas within the pond (Broman & Göransson 1994). Additional measures taken to ensure tailings in shallow areas would not undergo resuspension by wave action included construction of breakwaters across shallow areas to reduce wave action, and construction of a coarse rock cap over tailings in shallow water areas (Broman & Göransson 1994). In total, the coarse material cap was constructed over approximately 1/10 of the tailings pond bed surface area. After construction, the water level was raised again. The final water cover was from 0.6 to 9 m deep, with an average depth of approximately 2 m (Holmström & Öhlander 1998).
The capacity of the subaqueous disposal method is dependent on site specific variables. At the Stekenjokk site, to achieve sufficient water cover the impoundment capacity was increased by raising the dyke height, and redistributing tailings to lower tailings elevation (Broman & Göransson 1994). The limited availability of dyke construction material limited the height to which the dykes could be raised. Therefore the additional measure of redistributing tailings from shallow to deeper pond areas was used to meet the designed water cover depth (Broman & Göransson 1994).
Since the initial design was completed in 1991, multiple maintenance operations have been conducted to address issues recognized during monitoring. Maintenance has included reconstruction (deepening and widening) of the spillway to prevent ice clogging, construction of an emergency spillway, and construction of stabilizing berms and filters on the downstream dam wall (Bjelkevik 2005, EC 2009).
According to Bjelkevik (2005) the tailings pond dam at the Stekenjokk site should not be considered indefinitely stable due to the potential for internal erosion. Additionally, because of the age of the dam, its design was not in accordance with modern ICOLD recommendations (ICMM 2006). However, ongoing monitoring has expanded to include dam safety issues, and maintenance has been conducted in response to identified issues (Bjelkevik 2005).
Environmental cost aspects
Subaqueous disposal of mine waste at the Stekenjokk site has exceeded remediation objectives for limiting ARD production. For example, Zn concentrations in pond water have fallen 90% since decommissioning. The goal for zinc loading in impoundment discharge water was 800 kg per year, while according to Lindvall (2005), Zn loadings observed for the five consecutive years prior to reporting were between 50 to 100 kg per year. The Pond’s effective integration into the natural environment (Figure 1) is also indicated by the reestablishment of an Arctic char population within the pond. The char within the tailings pond have been found to have lower metals concentrations than those in the pond upstream of the mining operation (Lindvall 2005).
Figure 1. Stekenjokk tailings pond disposed under water after remediation. Photo © M.L. Räisänen, GTK.
Advantages and disadvantage
The advantages of this methodology were assessed during the testing of alternative methods (Broman & Göransson 1994). Subaqueous disposal is considered the most effective method of reducing sulphide oxidation in mine waste (Tremblay & Hogan 2001). At the Stekenjokk site subaqueous disposal also reduced the need for borrow material, provided a use for on-site waste rock that otherwise required disposal, and was the most cost efficient remediation alternative. Subaqueous disposal was determined to be 8 times more cost effective than construction of a dry cover, 3 times more cost effective than depyritization, and twice as cost effective as the buffering alternative.
The disadvantage of subaqueous disposal is that its application is strongly limited by site-specific variables. The site geologic and hydrologic conditions must allow the construction of an impoundment that is robust enough to be flooded indefinitely, and accommodate an adequate water balance to maintain a sufficient water cover. Ongoing site monitoring and occasional maintenance has been necessary at Stekenjokk to ensure the long term effectiveness of these design measures.
Monitoring / control needs
As part of the decommissioning plan, site monitoring was scheduled to be conducted for 5 years following closure. Monitoring has, however, been ongoing since decommissioning, and is expected to continue for decades into the future (Bjelkevik 2005). The original focus of monitoring was on effluent water quality and flow, water level fluctuations within the pond, resuspension of tailings within the pond, and breakwater stability (Eriksson et al. 2001). Eventually monitoring was expanded to include dam safety issues, the success of re-vegetation, and fish populations (ICMM 2006). Control measures as the result of monitoring have included reconstruction of the impoundment spillway to prevent ice clogging, construction of an emergency spillway, and the strategic emplacement of dam stabilization and filter systems to reduce erosion and seepage (Bjelkevik 2005).
Application of subaqueoues disposal of tailings successfully limited ARD related emissions from mine waste, and has facilitated tailings pond integration into the natural environment. The success of the tailings pond’s integration has been assessed on aesthetics and the establishment of a native fish population (ICMM 2006).
The evaluation of alternative remediation strategies identified permanent subaqueous disposal by flooding the tailings pond as the most suitable reclamation strategy at the Stekenjokk site (Broman & Göransson 1994). The suitability of subaqueous disposal of mine waste is generally dependant on hydrologic and geologic site characteristics, and the nature of the mine waste. A detail hydrogeologic site evaluation was conducted to determine the suitability of the site’s water balance for maintaining a permanent water cover (Broman & Göransson 1994). Geochemical testing and modelling confirmed that a water cover would prevent unacceptable levels of ARD emissions from the site (ICMM 2006). Holmström and Öhlander (1998) identified that the highest concentrations of dissolved metals in the tailings pond water after decommissioning were associated with previously oxidized waste, but predicted that the emissions would diminish with continued deposition of organics over the tailings surface.
The ongoing monitoring results from Stekenjokk provide insight to the application of subaqueous disposal in a constructed tailings pond. ICMM (2006) summarizes lessons learned from Stekenjokk based on 15 years of post closure monitoring. Their recommendations include (adapted from ICMM 2006):
- dam construction according to current ICOLD recommendations
- covering subaqueous tailings with a thin layer of non-reactive material to limit metals diffusion
- avoiding potentially reactive construction materials for dam construction, and
- routing the adjacent stream through the impoundment to accelerate stabilization of tailings and the pond’s integration into the surrounding environment.
Bjelkevik, A. 2005. Water Cover Closure Design for Tailings Dams. State of the Art Report. Luleå University of Technology, Department of Civil and Environmental Engineering, Division of Geotechnology. 2005:19. http://epubl.ltu.se/1402-1528/2005/19/LTU-FR-0519-SE.pdf
Broman, G. & Göransson, T. 1994. Decommissioning of Tailings and Waste Rock areas at Stekenjokk, Sweden. Proceedings of International Land reclamation and Mine Drainage Conference and the Third Conference of on the Abatement of Acid Drainage, April 24-29, 1994. Pittsburgh, USA. 2:32-40.
EC 2009. Reference Document on Best Available Techniques for Management of Tailings and Waste-rock in Mining Activities. European Commission. January 2009. http://ec.europa.eu/environment/waste/mining/bat.htm
Eriksson, N., Lindvall M., & Sandberg M. 2001. A Quantitative Evaluation of the effectiveness of the Water Cover at the Stekenjokk Tailings Pond in Northern Sweden: Eight Years of Follow-up. Proceedings to Securing of the future, International Conference on Mining and the Environment, Skellefteå, June 25-July 2, 2001.
Holmström, H. & Öhlander, B. 1998. Oxygen Penetration and Subsequent Reactions in Flooded Sulphideic Mine Tailings: a Study at Stekenjokk, northern Sweden. Applied Geochemistry. 14: 747-759.
ICMM 2006. The Closure of the Stekenjokk Mine in Northern Sweden and 15 Years of Post-Closure Follow-up. International Council on Mining and Metals. https://circabc.europa.eu/webdav/CircaBC/env/wg_non_energy/Library/natinal_guidelines/input_from_members/euromines/R_StekenjokkFinal.pdf
Knutsson, S. 1989. Tjälens inverkan på sandmagasin i Stekenjokk. Written for Boliden AB by Knutsson S. at Luleå University of Technology, Sweden (as referred to in Bjelkevik 2005).
Lindvall, M. 2005 Some Examples from Boliden’s Programme for Mine Sites Reclamation. 29th Annual Mine Reclamation Symposium, Abbotsford, BC. http://partner.boliden.com/www/bolidense.nsf/(LookupWebAttachment)/Library%20Lectures/$file/Bolidens_progr_for_Recl.pdf
Ljungberg, J., Lindvall, M., Holström, H., & Öldander, B. 1997. Geochemical field Study of Flooded Mine Tailings at Stekenjokk, Northern Sweden. Proceedings of Fourth International Conference on Acid Rock Drainage, May 31 – June &, 1997. Vancouver, B.C., Canada. 3: 1401-1417.
Tremblay, G.A. & Hogan, C.M. 2001. MEND Manual, Volume 4. Prevention and Control. MEND Report 5.4.2d.
Zachrisson, E. 1971. The Structural Setting of the Stekenjokk Ore Bodies, Central Swedish Caledonides. Economic Geology. 66: 642-652.