ARD control using waste rock layering - Samatosum mine site, British Columbia, Canada

Clayton Larkins, Geological Survey of Finland GTK, P.O.Box 1237, FI-70211, Kuopio, FINLAND, email: clayton.larkins(at)gtk.fi

Introduction

The Samatosum mine site is located approximately 80 kilometres north of Kamloops, British Columbia, where it produced silver, gold, lead, copper, and antimony concentrate from 1989 to 1992 (Morin & Hutt 1997). The Samatosum deposit is a strata bound sulphide rich quartz carbonate vein deposit within the metasedimentary and volcanogenic units of the Eagle Bay Assemblage (Bailey et al. 1999). Polymetallic sulphide orewas mined originally from an open pit, and to a lesser extent, from underground workings in the later stages of mine operations (Morin & Hutt 1997). The primary mineralogy of the ore was quartz, dolomite, and pyrite. Ore also included tetrahedrite (a copper-iron-zinc-silver-antimony sulphide), sphalerite, galena, chalcopyrite, and electrum (Morin & Hutt 1997). Over the course of operations 566,000 tons of ore was extracted, producing 23,864 tons of ore concentrate. Approximately 9.6 million tons of waste rock and tailings were produced on site.

Waste rock and tailings were derived from the following seven rock types, which were assessed as part of acid base accounting (ABA) studies conducted prior to mine closure: mafic pyroclastics, sericitic tuff, argillite, muddy tuff, quartz vein, quartzite, and chert (Denholm & Hallam 1991). The waste materials included high concentrations of pyrite, zinc, lead, and copper (Morin & Hutt 1997). Waste rock was deposited in a waste rock dump (WR dump) adjacent to the open pit mine, and 542,000 tons of tailings were deposited in a tailings impoundment northwest of the pit, adjacent to the process plant (Holmes et al. 2012). The tailings were flooded at mine closure to facilitate long term subaqueous disposal. The waste rock was disposed of by layering acid producing rock types within rock with high neutralization potential (NP) (Denholm & Hallam 1991).

The WR dump was designed based on ABA studies and waste volume estimates were made for each rock type. The design concept was to layer waste rock with high acid potential (AP) within acid consuming (AC) waste rock to utilize the net WR dump chemistry to buffer seepage and prevent the release of acid rock drainage (ARD) (Denholm & Hallam 1991). AP waste rock was deposited in two lifts, layered within AC waste rock (Holmes et al. 2012). The average NP in the WR dump was designed to be the equivalent of approximately 165 kg CaCO₃ per ton, and provide 3:1 NP to AP (Denholm & Hallam 1991, Holmes et al. 2012).

The WR dump was constructed on undisturbed till and organic soil with hydraulic conductivity ranging from 10^-5 to 10^-7 m/s. The WR dump design entailed seepage moving vertically through all layers and being discharged from springs at the low permeability base of the dump (Morin & Hutt 1997). The seepage was channelled to a retention pond before being discharged to the receiving Johnson Creek (Holmes et al. 2012).

Neutralized seepage from the waste rock exits the dump from two springs at the toe of the dump. Seepage is routed to a mine water sedimentation pond. After seepage water quality monitoring revealed the onset of ARD, a water treatment facility (WTF) was constructed in 1996. Water from the sedimentation pond is gravity fed to the WTF and discharged to a series of clarification ponds, from which it is discharged to the receiving creek.

The layered WR dump failed as a prevention measure for ARD generation at the Samatosum mine site. The water discharged from the WR dump became acidic during the freshet of 1995 (Holmes et al. 2012). At the onset of acid production from the waste dump control measures were taken, including the construction of the WTF in 1996. Trench networks capture and route site surface water for treatment (Holmes et al. 2012). Subsequent WTF upgrades, and continued monitoring has ensured that water discharged from the mine site has consistently met regulatory water quality standards (MEMPR 2008, Holmes et al. 2012). The tailings management, crisis management, and outreach work conducted at the Samatosum site has been recognized and awarded as exemplary (MEMPR 2008).

Mine closure objectives

The Samatosum mine site included a small open pit mine, a 32 hectare WR dump, a tailings impoundment containing 542,000 tons of tailings, multiple borrow pits and a processing plant site area (Morin & Hutt 1997). The mine closure objectives were to reclaim land surfaces and water ways to facilitate commercial logging, grazing, and wildlife habitat (Tashe 2012). Strategies for land and water reclamation, including ARD management, were developed prior to closure and integrated into the waste management design. Design measures for controlling ARD emissions from the site included flooding the tailings impoundment for permanent subaqueous tailings disposal, and disposing of waste rock using a layered dump design (Holmes et al. 2012). Additionally, native top soil was salvaged during mining for use as a growing medium as part of revegetation efforts (MEMPR 2008).

The mine site consists of an upper and lower basin that discharge independently into Johnson Creek. The upper basin contains the tailings impoundment, while the lower basin contains the open pit mine and WR dump (Figure 1). Water discharged from the flooded tailings impoundment of the upper basin has consistently met regulatory water quality standards, indicating that subaqueous tailings disposal is effectively preventing ARD generation (Holmes et al. 2012). Discharge from the WR dump showed the onset of ARD generation in the years following mine closure, and acidification during the freshet of 1995 (Morin & Hutt 1997). Additionally, the flooded open pit mine began generating ARD in 1992, and was amended with lime and NaOH between controlled discharges from 1993 to 1996 (Morin & Hutt 1997).

Figure 1. Samatosum Mine site and remedial features (adapted from Holmes et al. 2012, basemap copyright © ESRI 2015)

The development of ARD from the lower basin required adaptive management in response to monitoring results. In 1996 the mine was issued a pollution abatement order by the Canadian Ministry of the Environment to address the lower basin’s increasingly acidic discharge (MEMPR 2008). The subsequent control measures have included installation of the WTF, expanding the surge pond that collects untreated water, and developing management systems (MEMPR 2008).

Closure technologies

Plans for mine closure included salvaging native soil for site revegetation and ARD controls. Initial measures to control ARD included flooding the tailings pond and the layered WR dump design. Due to ARD generation from the lower basin (WR dump and open pit mine), the WTF was constructed after mine closure for ARD treatment.

Tailings pond

By flooding the tailings pond, a permanent water cover was induced as a means to prevent ARD generation. A properly designed water cover is considered the most effective method of ARD prevention (EC 2009). Due to the relatively low molecular oxygen concentration in water relative to air, and the low rate of oxygen diffusion in water, a properly designed water cover can prevent oxidation of sulphides in previously unoxidized mine waste (EC 2009).

Throughout the course of mine operation and a couple of years following mine closure, tailings pond water was retained and treated with NaOH. Monitoring results show that sulphate and metals concentrations in supernatant tailings pond water had stabilized below regulatory threshold levels by 1994 (Holmes et al. 2012). As the supernatant water approached acceptable water quality standards, natural stream water was rerouted through the tailings impoundment to maintain a permanent water cover and to facilitate pond water discharge to the receiving watershed. The permanent water cover was designed to be a minimum depth of 1 meter. At the time of flooding, pond water depths ranged from 1.5 to 8 meters, with the exception of localized areas at the edges of the pond that were slightly shallower than 1 meter (Holmes et al. 2012).

Layered WR dump

The concept for the layered WR dump design was to utilize the net NP of the waste rock to neutralize ARD generated from the AP fraction of the waste rock by strategically layering the different rock types (Denholm & Hallam 1991). Acid base accounting (ABA) and waste volume estimates of the site’s different rock types provided the basis for the layered WR dump design (Denholm & Hallam 1991). Six of the seven rock types were identified to be AP, while the mafic pyroclastic rock was determined to be non-AP, and have a high NP. Subsequently, 3.9 million tons of AP waste rock was deposited within 5.7 million tons of the AC mafic pyroclastic waste rock (Holmes et al. 2012). The WR dump was designed to have a 3:1 NP to AP ratio, as shown in table 1.

The WR dump was constructed on undisturbed till and organic soil that had a permeability ranging from 10^-5 to 10^-7. A 2 m thick base layer of AC mafic pyroclastic rock was overlain by four 6 m thick lifts of waste rock. A total of two lifts of AP rock were constructed in alternation with two lifts of AC rock. The upper lift of AC rock was capped with a 1 m thick till cover. The cover material provided a growth medium, but was not designed to inhibit water infiltration. The layers were not mixed, and the surface of each lift was smoothed prior to the construction of subsequent lifts (Denholm & Hallam 1991).

Table 1 Total AP and NP for WR dump design, adapted from Morin & Hutt (1997)

Water treatment

In response to increasing ARD generation from the open pit mine, and the development of ARD from the WR dump, the WTF was constructed for treatment of lower basin surface water discharge from these two sources. Water from the WR dump is channelled to a surge pond and then gravity fed to the WTF. The WTF was upgraded to a high density sludge (HDS) treatment plant in 1998 to increase treatment efficiency. The HDS plant neutralizes water using a series of two reactor tanks and a clarifier, from which water is decanted into a series of settling ponds prior to discharge to the receiving creek. Excess sludge is piped into sludge dewatering ponds (Martin et al. 2013). Every 4 to 6 years, sludge is excavated from the sludge pond for long term disposal in a sludge storage facility, where subaerial disposal will entail capping and revegetation (Martin et al. 2013).

Performance

Capacity

Capacity of the waste management technologies used at the Samatosum is discussed below in the context of NP within the WR dump for the prevention of ARD, and water management capacity related to water treatment.

The 3:1 NP to AP of the WR dump was designed to have more than sufficient neutralization capacity for the AP waste. A detailed review of the WR dump design using both laboratory and field data found that, although the neutralization capacity of the AC fraction was likely lower than intended by design, alkalinity was still being generated from the WR dump when the discharge became acid (Morin & Hutt 1997). However, the results of this study highlight important considerations regarding neutralization capacity calculations. Based on column experiments using material from the Samatosum site, NP consumption is predicted to proceed faster than acid is consumed, and up to 10 tons CaCO₃/1000 tons is not available for acid neutralization at this site (Morin & Hutt 1997). Despite the net geochemical capacity for neutralization at this site, seepage from the WR dump still became acidic. The development of ARD emissions was attributed to the physical characteristics of the WR dump and preferential flow (Morin & Hutt 1997).

In 2001 the capacity of the surge pond, which collects the untreated water of the lower basin, was increased to 8 times its previous size. The surge pond has sufficient capacity to store water for up to three months without discharge, which may be necessary if water treatment ceases during the coldest months of the year (Holmes et al. 2012). Periods of high water inputs to the surge pond generally occur at times of the year when the WTF can be operated to maintain sufficient freeboard within the pond (Holmes et al. 2012).

Maintenance needs

ARD generation was observed within the open pit mine the same year as mine closure, 1992 (Holmes et al. 2012). Subsequently, open pit mine water was treated with lime or NaOH from 1993 to 1996 prior to water release to the receiving creek (Holmes et al. 2012).

The development of ARD from the WR dump in combination with the progressively degrading water quality within the open pit mine resulted in the need for the installation of a WTF. The WTF was constructed in 1996 and upgraded to a HDS lime treatment plant in 1998 to increase treatment efficiency (Holmes et al. 2012). On average, 100 tons of lime is used annually in the HDS WTF (Holmes et al. 2012). Sludge discharged from the HDS WTF is excavated from sludge ponds every 4 to 6 years and deposited in a subaerial sludge storage facility (Martin et al. 2013). Additional maintenance has included treatment plant process automation, installation of remote monitoring systems, and expansion of the surge pond for collection of open pit and WR dump seepage waters (Holmes et al. 2012). Water treatment in the lower basin is anticipated to be required indefinitely (INMET 2004).

Reliability/Malfunction

This site illustrates both successful and failed ARD management strategies. In the case of the failed management approach, subsequent control measures have been recognized as exemplary (MEMPR 2008). Ongoing monitoring of the flooded tailings impoundment indicates that subaqueous disposal in a waste facility has successfully prevented tailings oxidation and subsequent ARD emissions at the Samatosum site (Holmes et al 2012). The gradual acidification of the layered WR dump after mine closure illustrates the malfunction of this ARD management strategy.

A detailed review of the WR dump failure is provided in MEND report 2.37.3 (Morin & Hutt 1997). Some important points of consideration taken from the conclusions of this report include:

• Neutralizing potential should be expected to be consumed faster than the consumption of acidity because not all alkalinity will be available for neutralization
• The WR dump began leaching ARD prior to the consumption of all alkalinity, indicating that preferential flow can lead to ARD leaching regardless of large scale NP and AP mass balance
• Careful physical design considerations, including the hydrogeologic characteristics of the waste, are necessary for a layering design to successfully control leachate pH
• Column experiments show that acid generation continues from AP layers layered as thin as 0.2 m with NP layers
• Layering may not be suitable for preventing metals leaching, depending on the solubility of secondary metal precipitates

This work concludes that the most likely cause for the development of acidic drainage from the Samatosum WR dump resulted from the development of preferential flow of acidic leachate within the dump (Morin & Hutt 1997). One assumption in the WR dump design was that water infiltration would occur vertically. However, the presence of coarse, low permeability waste rock, especially in the AP layers, likely facilitates preferential flow of leachate that enables acidic discharge, despite the net NP of the WR dump.

Laboratory column analyses showed that acid production from AP materials could proceed unattenuated when layered with NP material down to 0.2 m intervals. Therefore, while carefully designed and implemented layering may be used to control the pH of discharge, it may not prevent metals mobilization, depending on factors such as the site-specific solubility of secondary minerals formed within the waste rock (Morin & Hutt 1997).

The WTF installed in response to the onset of ARD emissions from the lower basin effectively removes metals by maintaining an optimal pH of 9.6 (Holmes et al. 2012). Since installation, the water treatment system has been upgraded to improve efficiency and reliability. In 1998 the plant was upgraded to a HDS system. In 2001 the surge pond was expanded to increase retention capacity of untreated water on site. Additionally, remote monitoring systems were installed and plant processes were automated to ensure discharge consistently meets water quality standards (Holmes et al. 2012). The Samatosum operation has received recognition and awards for its reclamation efforts, environmental management programs, tailings management, crisis management planning, and external outreach (MEMPR 2008).

Environmental cost aspects

Since the time of closure the Samatosum mine site has been largely revegetated with self-sustaining, native plant communities that provide diverse wildlife habitat (MEMPR 2008). Revegetation efforts include corridors constructed across the WR dump. However, there has not been revegetation around the surge pond that retains untreated water, nor around the WTF (Tashe 2012). Sludge from the HDS WTF is deposited in a sludge storage facility which will be capped and revegetated (Holmes et al. 2012). Monitoring of metals content in site vegetation started in 2006, and shows that most metals concentrations are within the tolerance range for beef cattle (Tashe 2012). The environmental costs of mining at Samatosum have been largely contained, reflecting the effectiveness of monitoring efforts and adaptive waste management strategies (Tashe 2012).

Advantages and disadvantages

The advantages of waste rock layering as an ARD prevention measure at the Samatosum site were precluded by its failure, which required subsequent treatment measures. The hypothetical advantages of layering waste rock include utilization of waste generated on-site and relatively easy construction (INAP 2009). As assessed in the MEND report 2.37.3, the failure of the WR dump to prevent ARD leaching likely resulted from the formation of preferential leachate flow (Morin & Hutt 1997). This assessment illustrates the primary disadvantage of layering as an ARD prevention measure. It is difficult to ensure adequate mixing of porewater for neutralization of acidic leachate and to prevent preferential flow (INAP 2009). Additionally, because layering does not attenuate acid generation, it may not be a suitable method to prevent the leaching of metals (Morin & Hutt 1997).

Monitoring / control needs

Monitoring at the site has included surface water monitoring from multiple points prior to discharge in both the upper and lower basins (Morin & Hutt 1997, Holmes et al. 2012). Groundwater in the upper basin and metal concentrations in site vegetation have also been monitored (Holmes et al. 2012, Tashe 2012). Monitoring results have guided management strategies and informed the need for control measures.

The water within the open pit mine of the lower basin first became acidic in 1992, after mine closure. The open pit pond water was treated with lime and NaOH to control pH. Treatments were conducted in batches, prior to releasing pond water to the receiving creek. The water quality became progressively worse between batch treatments from 1993 to 1996, until in-situ treatment was no longer sufficient for achieving water quality standards in discharge (Holmes et al. 2012).

Water discharged from the WR dump of the lower basin was monitored at several points after mine closure. Water monitoring was conducted at two seep springs from the toe of the WR dump, from a lined trench that bisects the WR dump, and from the surge pond that collects lower basin seepage (Morin & Hutt 1997). Seepage from the WR dump showed signs of ARD development following closure, and became acidic during the freshet of 1995 (Morin & Hutt 1997, Holmes et al. 2012).

To control the acidic discharge from both the waste rock pile and the open pit mine, a WTF was constructed in 1996. The WTF was upgraded to a HDS treatment plant in 1998 (Holmes et al. 2012). The HDS WTF is gravity fed from the surge pond and the open pit mine. Influent water is routed through 2 stage treatment before entering the clarifier. From the clarifier treated sludge is piped to one of two sludge ponds for dewatering, and supernatant treated water is decanted to a polishing pond. From the polishing pond, water is decanted to a final settling pond, from which it is discharged to the receiving creek (Martin et al. 2013). Further upgrades to the WTF conducted in 2001 included process automation and expansion of the surge pond (Holmes et al. 2012)

In the upper basin water is monitored from the tailings pond supernatant, a water quality pond (WQP) that captures seepage from the tailings pond, and two down-gradient groundwater wells (Holmes et al. 2012). Both surface water and groundwater monitoring indicate that oxidation has been curbed in tailings, and discharge has remained within regulatory thresholds since pond water amendment was stopped in 1994 (Holmes et al. 2012). Sulphate concentrations in the surface water discharged from the tailings pond reflect meteoric water and suggests that the water barrier is preventing ARD generation from tailings (Holmes et al. 2012). The WQP has elevated sulphate concentrations relative to the tailings pond, which have fluctuated between 10 and 270 mg/l since the beginning of operations (Holmes et al. 2012). The gradual stabilization of sulphate concentrations in the WQP indicates that sulphate oxidation is not ongoing, while the relatively elevated sulphate concentrations are interpreted to reflect the gradual flushing of tailings pore water impacted during mining operations (Holmes et al. 2012). Monitoring wells show sulphate concentrations generally between that of the WQP and meteoric water, which is interpreted as the mixing of tailings leachate with local groundwater recharge.

Conclusion

While the application of layered waste rock did not provide long term effective ARD prevention at the Samatosum mine site, subsequent monitoring and adaptive management practices have minimized impacts from mine emission to the surrounding environment. Controls on lower basin ARD generation include the construction of a WTF and advanced monitoring efforts (Holmes et al. 2012). Careful site monitoring, and subsequent upgrades to the WTF have ensured site discharge meets water quality standards (MEMPR 2008). Revegetation efforts have been largely successful (Holmes et al 2012).

The work presented in MEND report 2.37.3 provides insight on the suitability of layering waste rock as an ARD prevention strategy specific to the Samatosum site (Morin & Hutt 1997). MEND report 2.37.1 provides additional case studies of similar ARD reduction strategies from other sites (MEND 1998). While the geochemical basis for this approach has been recognized as valid, its practical application is limited by the difficulty of ensuring adequate pore water mixing (Morin & Hutt 1997, INAP 2009). Additionally, because acid generation within the AP portions of waste will continue without attenuation, this approach may not effectively immobilize metals, and therefore may not be appropriate when metals emissions in leachate are a concern (Morin & Hutt 1997).

References

Bailey, S., Paradis, S., Johnston, S., & Höy, T., 1999. Geologic Setting of the Devonian-Mississippian, Rea and Samatosum VMS Deposits of the Eagle Bay Assemblage, Adams Lake Area, South Central British Columbia. British Columbia Geological Survey. Geological Fieldwork, 1999.

Denholm, E. & Hallam, R. 1991. A Review of Acid Generation Research at the Samatosum Mine. Proceedings of the 15th Annual British Columbia Mine Reclamation Symposium in Kamploops, BC, 1991.

EC 2009. Reference Document on Best Available Techniques for Management of Tailings and Waste-rock in Mining Activities. European Commission. January, 2009.

ESRI 2015. Image available at: http://www.arcgis.com/home/webmap/viewer.html?useExisting=1

Holmes, A., Hamblin, B., Hogarth, J., Ford, C. & Anderson, T. 2012. Samatosum- A Review of Two Early Acid Drainage Prevention Approaches. Proceedings of 9th International Conference on Acid Rock Drainage (ICARD 2012). May, 2012. Ottawa, Canada.

INAP 2009. The GARD Guide. The Global Acid Rock Drainage Guide. The International Network for Acid Prevention (INAP). http://www.gardguide.com

INMET 2004 Report on Economic, Environmental, and Social Performance. INMET Mining Corporation

Martin, A., Loomer, D., Fawcett, S., Rollo, A., Gault, A., Jamieson, H., Simpson, S. & Al, T. 2013. Characterization and Prediciton of Trace Metal Bearing Phases in ARD Neutralization Sludges. MEND Report 3.44.1. May, 2013.

MEMPR 2008. Annual Report of the Chief Inspector of Mines 2008. British Columbia Ministry of Energy, Mines and Petroleum Resources. p 25-26.

MEND 1998. Blending and Layering Waste Rock to Delay, Mitigate or Prevent Acid Rock Drainage and Metal Leaching: A Case Study Review. MEND Project 2.37.1. April, 1998.

Morin, K.A. & Hutt, N.M. 1997.Control of Acidic Drainage in Layered Waste Rock: Laboratory Studies and Field Monitoring. MEND Project 2.37.3. September 1997.

Tashe, N. 2012. Evolution of Reclamation for Wildlife Habitat at Sulphide Mines. presented at Maine Metallic Minerals Conference, Nov 9, 2012.