Tailings and Mine Waste Conference

Building on soft tailings : east lake TMF expansion Kurylo, John Bohdan; Podaima, Trevor; Rykaart, Maritz Oct 31, 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


59368-Kurylo_J_et_al_Building_soft_tailings_TMW_2015.pdf [ 1.89MB ]
JSON: 59368-1.0320852.json
JSON-LD: 59368-1.0320852-ld.json
RDF/XML (Pretty): 59368-1.0320852-rdf.xml
RDF/JSON: 59368-1.0320852-rdf.json
Turtle: 59368-1.0320852-turtle.txt
N-Triples: 59368-1.0320852-rdf-ntriples.txt
Original Record: 59368-1.0320852-source.json
Full Text

Full Text

Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015    Building on soft tailings: east lake TMF expansion John B. Kurylo, Trevor Podaima, Maritz Rykaart SRK Consulting (Canada) Inc., Vancouver, BC, and Saskatoon, SK Canada ABSTRACT To accommodate an increased mine life at a northern Saskatchewan mine, an expansion to its existing tailings storage capacity was required. Innovative solutions were necessary due to operational and space limitations associated with the mine’s existing tailings management facilities (TMF). Specifically, staged construction of a waste rock containment dike directly over previously deposited unconsolidated tailings coupled with a complex seasonal tailings deposition plan for new tailings was required.  Despite efforts to characterize the tailings foundation conditions, significant uncertainty remained regarding the variability of its properties. The impact of this variability was examined through a sensitivity analysis; however, in an effort to minimize risk, a full-scale test section was constructed within the existing facility footprint and across the most sensitive section of the tailings foundation.   This paper presents the findings of the test section monitoring and overviews settlement and deformation back analysis completed to confirm and refine the dike design and overall tailings operational plan. It also provides a case study on how state-of-practice desktop analysis can be effectively coupled with field verification to support design. Keywords: back analysis, field verification, tailings foundation, deformation, dike expansion, case history 1. INTRODUCTION The Seabee Mine is located about 120 km northeast of La Ronge, Saskatchewan Canada (Fig. 1) and is operated by Claude Resources Inc. (CRI).  Seabee has been in operation since 1991 with an original project life of approximately five years. Additional reserves at Seabee Mine and at nearby Santoy 7, Santoy 8, Santoy Gap, and Porky West and Main deposits were discovered in the interim. As a result, the life of the mine for the Seabee Project is projected to continue past year 2020. Ore from all zones is being processed at the Seabee mill and tailings are deposited into the East Lake and Triangle Lake TMFs (Fig. 2). The current capacity of the combined TMFs is not sufficient to handle the increased mine life and; therefore, expansion of the facilities are required. The design of the East Lake TMF expansion was carried out by SRK Consulting (Canada) Inc.  The design consisted of constructing a waste rock containment dike directly on top of existing tailings in the East Lake TMF. To improve uncertainty associated with the strength characteristics of the tailings, and how that might affect the proposed dike, the initial phase of construction served as a test section allowing for rigorous data collection on actual performance of the foundation. This paper presents how this field monitoring/verification was coupled with Figure 1. Project location Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   state-of-practice desktop analysis to support ongoing design of the dike.   Figure 2. Site plan and general arrangement of existing tailings management facilities. 1.1 Historical use of East Lake East Lake was a natural lake converted to a TMF when Seabee Mine was first developed. Prior to initial tailings deposition, the lake was partially dewatered to provide the necessary capacity. Over the years as new reserves were found, a series of earthen and concrete containment walls were constructed along the TMFs low-lying edges to increase its capacity. The last series of raises brought the retaining structures to their current final crest elevation of 460 m, which includes 0.5 m of freeboard. Tailings were hydraulically deposited in East Lake TMF from one location via single pipe discharge. 2. EAST LAKE TMF EXPANSION DESIGN HISTORY Several design concepts to expand the existing East Lake TMF were evaluated, which consisted of utilizing and/or raising the existing retaining structures, constructing a series of internal waste rock containment cells, and various waste rock dike alignments for internal tailings storage. The design selected consisted of a continuous waste rock dike, approximately 900 m in length that would be constructed entirely within the footprint of the existing TMF.  Following preliminary design work, a field investigation was carried out that consisted of 16 cone penetration tests (CPTs) along the alignment of the proposed dike. The design was subsequently optimized using field data coupled with limit equilibrium and bearing capacity calculations. Specifically, areas of potentially weaker/softer unconsolidated tailings were identified and the alignment of the dike was adjusted to areas of higher strength tailings. Staged construction of the expansion dike was subsequently undertaken, with the first step being a starter dike (aka the test section) that would allow for collection of actual performance data to alleviate some of the uncertainty associated with the weak foundation properties. The test section data subsequently allowed for back analysis to determine suitable tailings properties for finalization of the expansion dike design to confirm short and long-term stability requirements. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   3. TEST SECTION CONSTRUCTION The test section was constructed over the weakest foundation zone as identified through the field investigation. Construction of the test section commenced early October 2012 and was largely completed by late April 2013. The target crest elevation of the test section was 460 m and the base width was the full final design width as illustrated in the design cross section of Figure 3.  Figure 3. Typical cross section through test section. 4. MONITORING DATA 4.1 Survey monuments A total 47 fixed survey monuments were placed near the upstream and downstream crests of the test section to assess deformation of the test section after construction. The location of these monuments, as illustrated in Figure 4, had to accommodate ongoing construction traffic over the test section.  Figure 4. Test section as-built and settlement monument locations. 4.2 Ground survey An initial survey of the as constructed dike test section is shown in Figure 3 and an isopach of the as-built waste rock fill thickness is illustrated in Figure 5.  Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   Figure 5. Isopach of test section waste rock thickness.  Following construction, monthly ground surveys of the test section were carried out using global navigation satellite system (GNSS) survey equipment coupled with a global position system (GPS).  These ground surveys had an accuracy of about ± 0.1 m and were used to monitor large-scale displacements of the dike.   4.3 Monitoring duration Primary data collection was stopped in August 2013, four months after completion of the test section. Ongoing review of deformation and primary settlement data showed an attenuation of movement towards steady state conditions by that time.  4.4 Monitoring results Vertical displacements ranged from 0.03 to 0.93 m (Fig. 6), while horizontal displacements ranged from around 0.00 to 0.38 m (Fig. 7). The bulk of the displacements occurred within the first two and a half months of monitoring up to the end of June 2013.  Displacements after this time were noticeably smaller and by August 2013 primary settlement appears to have stopped.  A histogram timeline plot of settlement is presented in Figure 8. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   Figure 6. Vertical displacements of test section survey monuments.  Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   Figure 7. Horizontal displacements of test section survey monuments.  Figure 8. Settlement histograms for test section survey monuments. Survey monuments M24, 25, 26, 32, 35, and 38 (Figures 6 through 8) were placed very close to the edges of the test section crest. Some slumping/relaxing of the waste rock was observed at these locations due to over steepened slopes. Therefore, some of the ongoing changes in these locations are attributed not only to primary settlement, but also from deformations as a result of the monuments moving as the slope relaxed.  Figure 9 presents an isopach of how the settlement data varied spatially across the test section at the end of the monitoring period.     Figure 9. Settlement isopach based on vertical settlements at test section survey monuments. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   5. INCORPORATION OF MONITORING DATA  For the final proposed full height expansion dike arrangement, the test section monitoring data was used to complete empirical settlement calculations using Terzaghi’s one-dimensional theory of consolidation (REFERENCE). The data was then incorporated into finite element deformation and shear strength reduction models to determine ‘safety factors’ for each of the dike construction stages. 5.1 One-dimensional consolidation analysis One-dimensional consolidation settlement calculations were carried out to facilitate an improved understanding of the test section data, and to help define critical two-dimensional cross section locations for completing the more rigorous finite element analysis.   Given the large variability in dike geometry and foundation conditions (physical extent and material variability), consolidation calculations were completed at several key locations across the dike section (Fig. 10), and about every 50 m along the dike centerline. Calculations were also completed for interim dike heights at elevations 460 and 463 m. These were done to assess the potential benefits of staged construction and to determine the maximum expected settlement during construction. For these calculations, the foundation material was assumed to consist entirely of variable thickness tailings with consolidation index values (Cc) from 0.3 to 0.4.  Figure 10. Simplified cross-section assumed for 1D consolidation.  5.2 Finite Element Analysis 5.2.1 Model set-up Deformation analyses were carried out using PLAXIS 2D Anniversary Edition two-dimensional finite element software (Plaxis 2014).  The analyses included pore water pressures induced by compression and consolidation of the foundation tailings, solved simultaneously with the soil and rock mass stress field (Plaxis 2014). The results were also coupled with the computation of a strength factor via strength reduction techniques. This was done to get a better understanding of expected failure mechanisms and identification of areas where larger displacements may occur.  For the tailings foundation the nonlinear Hardening Soil model with a Small Strain Stiffness (HS-Small) constitutive soil model was used. This constitutive model allows for improved evaluation of increases in tailings stiffness at small strains while obtaining more reliable displacement estimates in the finite element models, specifically for working load conditions (Plaxis 2014).  5.2.2 Model calibrations Dimensionless ratios were used to simplify and group interrelated variables (dimensionless groups). This was done to facilitate comparison of test section data and modelled finite element Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   results, and allowed for theoretical solutions to be more readily identified. To approximate deformations (ε), the ratio of settlement (S) divided by tailings thickness (d) was tabulated (i.e. ε = S/d ratio). These values were then plotted against the dimensionless ratio of load (h or waste rock dike height) divided by the tailings thickness (i.e. dimensionless ratio ε plotted against h/d). This plot revealed two distinct groups of data. One group (Group 1) appeared to be correlated with settlement of the tailings foundation and the second group (Group 2) had influences from tailings settlement as well as from slumping/relaxing of the waste rock dike slopes. Specifically, it was confirmed that the highest S/d and h/d ratios corresponded to areas with thinner tailings foundations, while the highest displacements were primarily due to relaxing of the outer slopes. Group 1 showed less horizontal displacements (primarily vertical settlement) and correlated well with the tailings foundation settlement. Group 2 data primarily consisted of monuments affected by slope relaxation. Since deformations for Group 2 could not be attributed to only foundation settlement, the deformation data was excluded in the back analysis. Two Plaxis calibration models were subsequently set up to allow for comparison to the monitoring data as follows: (1) Uniform 5 m thick waste rock dike with a tailings foundation that increased in thickness from 0 to 5 m; and (2) Uniform 5 m thick tailings foundation that had an overlying waste rock dike thickness that increased from 1 to 6 m.  Using the initial Plaxis calibration model results, the tailings material properties were adjusted until the results best fit the collected monitoring data. The primary values that were adjusted included void ratio, compression index, recompression index and internal friction angle. These values, along with engineering judgement and published benchmark values, were used to calculate some of the secondary material properties for the tailings (based on EPRI 1990, Plaxis 20014, Santos et al. 2001). These back analyzed tailings material properties were then adopted and used in the subsequent deformation models for the two critical design sections.  All monitoring data (including Group 2 data) were used as a check to review deformations predicted in the tailings and in the waste rock dike from the deformation model.  5.2.3 Modelled sections  Two critical dike sections were modelled (see Figure 11 and Figure 14). The first, Station 0+750 was chosen because it represents an area close to the zone of maximum observed displacements, and the second, Station 0+525 represents and area of maximum dike thickness overlying a thick tailings foundation (approximately 12 m thick). Sensitivity analyses were completed for these two sections varying the water table, tailings foundation stiffness, and tailings strength parameters. 6. BACK ANALYSIS RESULTS 6.1 One dimensional consolidation analysis For the 1D dimensional settlement calculations, the footprint of the expansion dike was divided into a series of sections, approximately 50 m apart. Along each section settlement was calculated at about nine points (Fig. 10). Figure 11 illustrates the predicted maximum settlement based on these calculations that ranged from 0.3 to 2.1 m. These settlement values assume instantaneous full construction of the expansion dike to its maximum crest elevation of 466 m.  Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015    Figure 11. Predicted settlements based on 1D consolidation calculations. 6.2 Finite element analysis Displacements due primarily to consolidation (i.e. vertical displacement of the dike crest), were less than 2.1 m for the worst case deformation model and were typically less than 1.6 m. The largest displacements typically occurred near the crest of the expansion dike and on the side that had the thicker tailings foundation. The analysis suggests that the majority of primary consolidation for a given raise occur over a period of approximately 7 to 12 months. Based on these calculations, supported by observations during construction of the test section, the critical time period for settlement is during the first four months following construction. Settlement after this period slows down in rate and magnitude. As the dike is raised, adding more load, further primary consolidation is triggered resulting in an increased total settlement over the life of the structure. Finite element modeling of horizontal sections across the expansion dike indicated differential settlements may range from 0.8 to 1.2 m.  These differential settlements lead to some strain development within the expansion dike; however, these strains do not appear to result in the development of a large failure plane in the waste rock itself.  Base case deformation results for the two critical sections are shown in Figure 12. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   Figure 12. Deformation sections - example finite element model results. Based on the settlement, deformation and stability analysis, Safety Factors for all modelled cases under static conditions, and without consideration of staged construction, were found to be above 1.4. With staged construction, the proposed method to construct the expansion dike, the finite element models yielded minimum Safety Factors of 1.6 or greater. These results were used to confirm that the expansion dike design, as proposed, will function appropriately. Figure 13 shows select results from the analyses completed for the critical dike section at Station 0+750.  Figure 13. Example finite element model result for modelled deformation and corresponding safety factors. 7. EAST LAKE EXPANSION DESIGN OVERVIEW The initial design work and the subsequent back analysis has confirmed that in almost all cases the failure mode for the East Lake TMF expansion dike would be complete foundation failure governed by the low strength tailings or failure of a portion of the dike as a result of the poor mechanical response of the underlying tailings foundation material. To final expansion dike design entails a 6 m high, 900 m long waste rock dike, constructed in 1.5 m high raises (see Figure 15 for typical cross section). Tailings deposition will done to ensure development of a uniform beach immediately upstream of the dike as illustrated in Figure 14.  This beach will reduce seepage rates through the dike and ensure dike stability by moving the pool as far from the dike as practical. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015    Figure 14. Overview of the East Lake TMF expansion design and proposed tailings deposition . Figure 15. Typical expansion dike design section. The analysis described in the paper confirmed that to ensure adequate performance of the design, the following considerations must be adhered to: 1. Staged Dike Construction: Staged construction allows for increases in the sheer strength properties in the foundation tailings over time, and assist dissipation of foundation pore pressures. 2. Tailings Deposition and Water Management Plan: The deformation analysis confirmed the importance of ensuring a proper beach immediately upstream of the expansion dike to move the pool as far as practical from the dike. 3. Dike Monitoring: As long as overall stability remain within acceptable limits (i.e. to ensure freeboard is maintained), deformation of the dike is not a concern. To track this, continued dike monitoring during each construction phase will be required. Proposed monitoring includes piezometers along the upstream slope (driven into the tailings foundation), deep settlement points at or near the tailings surface (expected to consist of flat metal plates affixed to vertical rods -the top of which would be surveyed), and fixed survey monuments. Proceedings Tailings and Mine Waste 2015 Vancouver, BC, October 26 to 28, 2015   8. CONCLUSION This paper demonstrates how state of practice desktop analysis can be effectively coupled with field monitoring and verification to improve the design and associated operational plans for a structure founded on soft tailings. Inherent uncertainty associated with the strength characteristics of the tailings foundation was addressed by designing the structure to allow the first construction stage to double as a test section to allow in-situ performance testing of the materials. Next, using the collected field data, multiple desktop analysis methods ranging from empirical to rigorous finite element models was used to back analyze the tailings foundation strength properties and as such allow for a well-understood assessment of the likely performance of the structure. For the East Lake TMF expansion it was established that the dike should be constructed in stages to allow for expected deformations and that a detailed tailings deposition, water management and monitoring plan should be completed to ensure successful construction and operation of the TMF.  9. ACKNOWLEDGEMENTS The authors are grateful to Claude Resources Inc. for permission to publish this material. The authors also wish to acknowledge Arcesio Lizcano for ongoing reviews and guidance, and Murray McGregor for site and design support provided throughout this project. Finally, the authors could not have presented this material were it not for the efforts of their other colleagues at SRK Consulting (Canada) Inc. and Seabee Mine site staff who have been involved with the fieldwork, design, construction, data collection, and administrative support for this project. 10. REFERENCES  [EPRI] Electric Power Research Institute, 1990. Manual on Estimating Soil Properties for Foundation Design. Product ID: EL-6800.  Plaxis bv. 2014. Plaxis 2D Anniversary Edition, Finite Element Software Package, Reference Manual, Material Models Manual and Scientific Manual. 2014. Delft, Netherlands.  Santos, J.A., Correia, A.G. 2001. Reference threshold shear strain of soil - its application to obtain a unique strain-dependent shear modulus curve for soil. In proceedings for the 15th International Conference on Soil Mechanics and Geotechnical Engineering. Istanbul, Turkey, volume 1, 267-270. SRK Consulting (Canada) Inc. (SRK) 2015. “Seabee Mine - East Lake TMF Expansion Geotechnical Design. Revision 3”. Project Number: 4CC005.015. Report submitted to Claude Resources Inc. and Saskatchewan Ministry of Environment, July 2015.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items