- Email: [email protected]
Underground geotechnical and geological investigations at Ekati Mine–Koala North: case study
Lithos 76 (2004) 347 – 357 www.elsevier.com/locate/lithos
Underground geotechnical and geological investigations at Ekati Mine–Koala North: case study Jaroslav Jakubec a,*, Larry Long b, Tom Nowicki c, Darren Dyck b a
SRK Consulting, Suite 800, 1066 West Hastings Street, Vancouver, B.C., Canada V6E 3X2 b BHP Billiton Diamonds Inc., Canada c Mineral Services Canada Inc., Canada Received 25 June 2003; accepted 3 December 2003 Available online 10 June 2004
Abstract Since 1998, BHP Billiton has mined diamonds at the Ekati Diamond Minek near Lac de Gras in the Northwest Territories of Canada. Current operations are based on mining multiple pipes by the open-pit method, but as some pits deepen, converting to underground mining is being considered. As a test of underground mining methods and to provide access to the lower elevations of the Panda and Koala pipes, the Koala North pipe is being developed for underground mining. Initially, the top 40 m of the pipe were mined as an open pit to provide grade information and a prepared surface for the transition to underground mining. Currently, Koala North is being developed as an open-benching, mechanized, trackless operation. Although the method was successfully used at several De Beers diamond operations in South Africa, it has never been tested in an Arctic environment. This case study describes basic geology, mining method layout and ongoing geological and geotechnical investigation. From the beginning of underground development, geotechnical daily routines have been fully integrated within the technical services department, which supports the operation. Geotechnical, geological and structural information obtained from underground mapping and core logging is compiled, processed, reviewed and analyzed on site by the geotechnical staff. Conclusions and recommendations are implemented as part of the operations in a timely manner. This ongoing ‘‘live’’ process enables the operators to make the most efficient use of resources both for ground support and excavations as well as to address safety issues, which are the top priority. D 2004 Published by Elsevier B.V. Keywords: Kimberlite geology; Diamond mining; Rock mass rating; Kimberlite pipe contact zone; Kimberlite weathering
1. Project background Since its opening in 1998, the Ekati Diamond Minek has produced more than 10 million carats. Nearly all of the diamond production has been from * Corresponding author. Tel.: +1-604-681-4196; fax: +1-604687-5532. E-mail address: [email protected] (J. Jakubec). 0024-4937/$ - see front matter D 2004 Published by Elsevier B.V. doi:10.1016/j.lithos.2004.03.052
open-pit mining of multiple pipes. However, as some pits deepen, it is planned to convert to underground mining. The Koala North pipe has been selected as a trial underground mine for the purposes of testing mining methods and to provide access to the lower elevations of the Panda and Koala pipes. The upper 40 m of the Koala North pipe was mined in late 2000 as a small open pit to provide grade information and a prepared surface for the transition to underground
348
J. Jakubec et al. / Lithos 76 (2004) 347–357
mining. The Koala North underground mine, North America’s first underground diamond mine, formally opened in November 2002.
2. General geology Koala North forms part of the Lac de Gras kimberlite field that occurs within the central Slave Structural Province in the Northwest Territories of Canada. The Lac de Gras kimberlites range in age from f 45 to 75 Ma and intrude Archean granitoids as well as metamorphosed supracrustals (primarily metagreywackes) of the Yellowknife Supergroup. Multiple Proterozoic diabase dike swarms with ages varying from 2.4 to 1.27 Ga intrude the Archean rocks. Kimberlites are the only Phanerozoic rocks known in the Lac de Gras region, although fossil-bearing mudstone xenoliths within the kimberlites indicate that
sediments must have formed a thin veneer over the older rocks at the time of kimberlite emplacement. A model Rb –Sr age of 53.1 F 0.3 Ma (Geospec, 2002) has been determined for the Koala North kimberlite. The kimberlites in the Lac de Gras province generally form steep-sided, diatreme-shaped pipes. They are relatively small features with surface areas ranging from approximately 0.1 to 20 ha in size. With few exceptions, they are composed almost exclusively of volcaniclastic kimberlite material. This includes fine- to medium-grained crater sediments, ash/mudrich to olivine-rich resedimented volcaniclastic kimberlite (RVK) and primary volcaniclastic or pyroclastic kimberlite (PVK). 2.1. Koala North pipe geology The Koala North kimberlite is a very small body (0.4 ha at surface) situated between the Panda and
Fig. 1. Oblique view looking south across the central development area at Ekati. The Panda open pit is in the foreground; the Koala North underground project is located beneath the existing test pit mid-frame; and the Koala pit and Ekati plant site are in the background. Photo taken in mid-2002.
J. Jakubec et al. / Lithos 76 (2004) 347–357
349
Fig. 2. An isometric southwest-looking view of the Koala North underground mine model. Note that as of end of 2003, the ore has been extracted from the first three levels.
Fig. 3. Brown and black mudstone xenoliths in the drill core. Note the slicken-sided surfaces in disintegrated brown mudstone (top photo).
350
J. Jakubec et al. / Lithos 76 (2004) 347–357
Koala pipes in the central portion of the Ekati property (Fig. 1). The pipe lies within a depression formerly occupied by the Koala Lake and was originally covered by 8 m of water and 15– 20 m of boulder- and gravel-dominated glacial till overburden. Koala North intrudes competent, syn-tectonic biotite granodiorite of the Koala Batholith. Modeling of the Koala North body, based on mapping of the kimberlite/granodiorite contact in the test pit and on interpretation of kimberlite drill intersections, suggests that the lower portion of the pipe has an overall steep-sided inverted cone morphology typical of most kimberlites in the Lac de Gras area (Fig. 2). In plan view, the kimberlite outline is roughly circular with no consistent elongation evident. Drilling data suggest that, below the 2370-m mining level (370 m a.m.s.l.), wall rock contacts typically dip inwards at a regular angle of approximately 86j. Above this elevation, however, the eastern margin of the pipe flattens out considerably. It is likely the irregular morphology of the upper portion of the pipe is controlled to a large extent by fault/joint zones that predate kimberlite emplacement. Drilling at Koala North has intersected kimberlite down to the 155-m elevation ( f 270 m below the top of the pipe) and indicates that it is comprised exclusively of volcaniclastic kimberlite material to this depth. The dominant infill lithology at Koala North is crudely bedded to massive, relatively mud-rich volcaniclastic kimberlite with varying, but generally moderate to high olivine contents. These assemblages appear to represent a ‘‘package’’ of resedimented volcaniclastic kimberlite (RVK). The RVK is gener-
ally dark gray and typically contains between 10% and 20% glassy pale green, partially serpentinized olivine grains. However, olivine contents vary locally with minor intervals of both mud-rich ( < 10% coarse olivine) and significantly more olivine-rich (30 – 60% coarse olivine) material evident in most drill intersections. Olivine grains range in size from < 0.3 to f 5 mm in diameter, but commonly exceed 1 mm. In addition to olivine, a minor amount of quartz (generally less than 0.5 mm in size) is evident. The olivine and quartz occur in a matrix consisting primarily of gray to black very fine grained material comprising serpentine group minerals and relatively abundant probable disaggregated mud (65 – 80% of the rock). The RVK is typically xenolith-poor with small (generally < 3 cm) angular to rounded fragments of biotite granodiorite and black to brown mudstone (Fig. 3) usually constituting less than 10% of the rock. Rare zones with increased xenolith content (up to f 20%) are present in places. Centimeter-scale bedding is moderately developed in places (see Fig. 4), but, in general, the RVK has a massive appearance. Dark brown to black fine-grained sedimentary material occurs as rare blocks/bands ranging in width from tens of centimeters up to 5 m. The morphology and lateral continuity of these intervals are not constrained by the current drilling. However, exposure of the kimberlite in underground workings indicates that mudstone occurs both as xenolithic blocks and in situ lenses/bands of fine-grained sediment. This material is indurated, laminated and fissile. It generally lacks olivine ( < 1%) and other obvious kimberlite components. While the RVK material shows variations in coarse-olivine and, to a lesser extent, xenolith content, it retains a very similar character throughout the
Fig. 4. Layers of fine and coarse clasts in resedimented volcaniclastic kimberlite. The core diameter is approximately 63 mm (HQ size core).
J. Jakubec et al. / Lithos 76 (2004) 347–357
drilled intersections examined. Contacts between olivine-rich and ash-rich sections of the core, as well as between xenolith-rich and xenolith-poor zones, are typically gradational and distinct geological boundaries or markers are not evident. For the purposes of geological modeling, all RVK material (as well as associated minor intersections of mudstone and siltstone) has been allocated to a single, crater phase that dominates the Koala North pipe. This material is believed to have formed by slumping and reworking of pyroclastic material and surface sediments into the kimberlite vent. In addition to the RVK described above, a small lens of coarse-grained olivine-rich rich primary volcaniclastic kimberlite occurs internally in the upper northern portion of the pipe. This rock type is similar in appearance to olivine-rich RVK in that it is dominated by macrocrystic olivine set in a dark finegrained matrix. It is distinguished by a more uniform olivine distribution and massive appearance, a lack of any bedding, an absence of quartz, a less friable texture and a greater degree of competence (i.e., apparent greater rock strength). Xenolithic components commonly make up between 10% and 20% of the kimberlite and include small mudstone clasts, typically well-rounded unaltered granodiorite cobbles and rare carbonized black wood fragments. The interclast matrix is dominated by fine-grained serpentine. This material is interpreted to represent a remnant of an earlier pyroclastic deposit (PKV) preserved on the margin of the pipe.
351
very clear that both RVK and PVK kimberlite are relatively competent rock masses. However, within the RVK, clay-rich intervals show low rock mass competency and also high weathering susceptibility. Geotechnical data were input into spreadsheets and various parameters were calculated. Data were also entered into a 3D computer modeling program and database (Gemcom) and used to construct a geotechnical domain model. 3.2. Geotechnical domain model Geotechnical data were superimposed onto the basic geological model. Based on the physical properties of the rock mass, eight geotechnical domains were identified. The rock mass parameters are illustrated in Table 1 and the schematic geotechnical domain model is shown in Fig. 5. 3.3. Geotechnical issues Two main geotechnical issues were identified during the investigation:
Stability of the kimberlite pipe contact zones. Competency and weathering susceptibility of the weak clay-rich kimberlite.
3.3.1. Kimberlite pipe contact zones The contact zones are variable both in the country rock granodiorite and in the kimberlite. Typically, as the development in the country rock approaches the kimberlite pipe contact, there is a zone of granodi-
3. Premine development geotechnical investigation Feasibility-level geological and geotechnical information was obtained from the surface core drilling program as no underground development was available. 3.1. Drill core geotechnical data Basic geotechnical information was obtained from the exploration drill holes. Laubscher’s (1990) rock mass rating (RMR) classification system was used to log the drill core. This classification system enables capturing of individual geotechnical parameters that are used for rock mass competency estimates. It was
Table 1 Koala North rock mass ratings (RMR) values for the individual geotechnical domains Geotechnical domain
RMR
Overburden Near-surface granodiorite Granodiorite External contact zone Internal contact zone Upper RVK Lower RVK Clay-rich RVK
– 35 – 55 60 – 75 40 – 55 18 – 30 20 – 40 45 – 65 15 – 30
RMR number represents the rock mass competency index from 0 (poorest) to 100 (most competent).
352
J. Jakubec et al. / Lithos 76 (2004) 347–357
Fig. 5. Schematic geotechnical domain model of Koala North kimberlite pipe. Note that both external (outside the kimberlite pipe) and internal (inside the kimberlite pipe) geotechnical contact zones of poorer quality rock mass were considered for the mine design. Rock mass above the dashed line was influenced by the surface weathering and glacier rebound.
orite approximately 1 to 5 m (or greater in some areas) in width with increased joint frequency (Fig. 6). It appears that the emplacement of the kimberlite has overprinted or enhanced the regional jointing. In some areas, there is also a narrow (approximately 1 to 2 m wide) transition zone at the pipe contact in which kimberlite stringers occupy joints within the granodiorite. Although the contact zones are not equally developed around the entire perimeter of the pipe, their stability will negatively impact on the development cost and amount of country rock dilution in the recovered ore. 3.3.2. Kimberlite weathering The weathering susceptibility in the mining sense is rock strength deterioration within the mine life. The degree of weathering is influenced mainly by the clay minerals (that are product of geological weathering) within the kimberlite rock types and by the moisture in the mining environment. Weathering of the kimberlite can adversely impact on several mining activities such as ground support, production blasting and trafficability. Weathering
can also promote the generation of mud in the muck pile and it is clearly an important issue that needs to be addressed in any kimberlite mining. To assess the impact of weathering on mining operations, a series of accelerated weathering tests was conducted on the drill core (Fig. 7). Tests revealed that the majority of the Koala North kimberlite has low weathering susceptibility from the mining point of view; that is, the rock mass will not deteriorate significantly during the mining cycle. It was noticed, however, that certain clay-rich kimberlite and mudstone will disintegrate after exposure to the elements. Such materials appear to form zones of limited dimensions (up to 5 m); it was estimated that they will form less than 10% of the materials being mined.
4. Mining method Koala North is being developed and operated as a highly mechanized operation utilizing a trackless diesel-powered fleet of equipment. A key function
J. Jakubec et al. / Lithos 76 (2004) 347–357
353
Fig. 6. Example of less competent contact zones being developed inside and outside the kimberlite pipe contact. Lightly colored rocks represent the country rock granodiorite.
Fig. 7. Accelerated weathering test of the clay-rich kimberlite reveals high weathering susceptibility of this rock type. Top photo illustrates fresh rock and bottom photo shows the same core after several wetting and drying cycles. The core diameter is approximately 42 mm (NQ size core).
354
J. Jakubec et al. / Lithos 76 (2004) 347–357
of the mine is to serve as a learning experience for underground kimberlite mining in the northern Arctic environment. The project will be dealing with a unique set of mining conditions in this harsh environment.
5.0-m wide 5.0-m high. Because subhorizontal joints are present in the country rocks, the development drift profile has a ‘‘flat back’’ profile to increase back stability. 4.3. Draw points
4.1. Ramp Access to the underground workings is from a ramp developed from the surface. The profile of the ramp is 5.5 5.5 m (Fig. 2). All of the drilling in the granodiorite is conducted using a brine solution due to the presence of permafrost and cold air temperatures. 4.2. Level access drifts Production access drives are being developed off the ramp at regular intervals to access the kimberlite pipe. These drives provide access for stope production, exploratory diamond drilling and installations of mining infrastructure, including sumps, refuge bays and electrical installations (Fig. 8). In country rock granitoids, the current designs call for the level access drifts to be developed at
Production draw points are being developed into and across the kimberlite pipe for slot access, stope drilling and production mucking. These cross-cuts are designed in an arched profile at 4.0-m wide 4.0-m high. Due to the geomechanical and weathering characteristics of the kimberlite, development and production blast-hole drilling must be completed dry. A two-stage dust filtering system has been developed to collect and filter return air from drill holes and remove dust in coarse and then fine fractions. 4.4. Production stopes The mining method to be used at Koala North is open benching. As noted above, the unique mining environment and climatic conditions at Ekati Diamond Minek will involve adaptation of the method.
Fig. 8. Schematic layout of development on a production level. Each grid cell is 50 50 m for scale.
J. Jakubec et al. / Lithos 76 (2004) 347–357
355
Fig. 9. Schematic vertical section with individual production level geometries.
The method is a top-down retreat mining system, similar to sublevel caving but without the caved waste behind the drawn ore. The schematic vertical section with basic dimensions is illustrated in Fig. 9; an aerial view of the 2385 level being extracted at the bottom of the pit is shown in Fig. 10.
5. Geotechnical investigation during mine development
5.1. Geological and structural mapping Basic geology data are collected on a daily basis as mining faces advance. Sketch maps of the development including major geological contacts and structures are constructed and a digital photograph of the face is taken. Geological contacts and major structures are plotted on plans and projected to the next mining level. 5.2. Geotechnical mapping
Underground excavations combined with the underground drilling program provide truly 3D geological and geotechnical data. Such information is input into a computerized database (Gemcom) and the initial 3D pipe geological model is adjusted accordingly. This information provides invaluable technical input to the Mine Planning and Engineering Department as well as to the development contractor. It facilitates adjustment and optimization of mine layout and ground support design.
The geological sketch map of the development acts as a foundation for the rock mass classification a s s e s s m e n t . L a u b s c h e r ’s u p d a t e d M R M R (Laubscher and Jakubec, 2001) classification system is used to describe the main geotechnical parameters, such as intact rock strength, joint frequency, number of main joint sets and joint conditions. Field sheets are compiled into rock mass classification maps and geotechnical domains are delineated. Information from the geotechnical mapping is
356
J. Jakubec et al. / Lithos 76 (2004) 347–357
Fig. 10. Aerial view of the first production level 2385 being extracted at the bottom of the pit.
used for optimizing the layout and for assessing support requirements. Such requirements are communicated on a daily basis with the development crew. Both geological and geotechnical mapping have to be completed on a daily basis because advancing ground support (including shotcrete) rapidly covers the entire face and prohibits any further rock mass assessment.
6. Conclusions The Koala North pipe has been selected as a trial underground mine for the purposes of testing mining methods and to provide access to the lower elevations of the Panda and Koala pipes. The geotechnical and mining investigations will serve as a learning experience for underground kimberlite mining in a northern Arctic environment. Although this mining method is successful when used on several De Beers diamond
operations in South Africa, it has not been tested in this setting. Ongoing geological and geotechnical investigation during the mine development is vital for a mining operation to succeed. At Koala North, both geotechnical and geological sections are fully integrated into a Technical Services Department. This high level of integration enables important information and suggestions related to the ground support and mine development to be forwarded to the mining contractors without delays. Effective communication between the miners and engineers is of paramount importance to ensure a safe and successful operation.
Acknowledgements The authors would like to thank BHP Billiton Diamonds for permission to publish this paper.
J. Jakubec et al. / Lithos 76 (2004) 347–357
References Geospec, 2002. Rubidium – strontium isotope analyses for BHP Billiton Diamonds Inc. Geospec Consultants Limited, internal report. Laubscher, D.H., 1990. A geomechanics classification system for
357
the rating of rock mass in mine design. J. S. Afr. Inst. Min. Metall. 90 (10), 257 – 273 (South Africa). Laubscher, D.H., Jakubec, J., 2001. The MRMR rock mass classification for jointed rock masses. Underground Mining Methods. Society For Mining Mettalurgy, and Exploration (SME), Littleton, CO, pp. 475 – 483.
Underground geotechnical and geological investigations at Ekati Mine–Koala North: case study Jaroslav Jakubec a,*, Larry Long b, Tom Nowicki c, Darren Dyck b a
SRK Consulting, Suite 800, 1066 West Hastings Street, Vancouver, B.C., Canada V6E 3X2 b BHP Billiton Diamonds Inc., Canada c Mineral Services Canada Inc., Canada Received 25 June 2003; accepted 3 December 2003 Available online 10 June 2004
Abstract Since 1998, BHP Billiton has mined diamonds at the Ekati Diamond Minek near Lac de Gras in the Northwest Territories of Canada. Current operations are based on mining multiple pipes by the open-pit method, but as some pits deepen, converting to underground mining is being considered. As a test of underground mining methods and to provide access to the lower elevations of the Panda and Koala pipes, the Koala North pipe is being developed for underground mining. Initially, the top 40 m of the pipe were mined as an open pit to provide grade information and a prepared surface for the transition to underground mining. Currently, Koala North is being developed as an open-benching, mechanized, trackless operation. Although the method was successfully used at several De Beers diamond operations in South Africa, it has never been tested in an Arctic environment. This case study describes basic geology, mining method layout and ongoing geological and geotechnical investigation. From the beginning of underground development, geotechnical daily routines have been fully integrated within the technical services department, which supports the operation. Geotechnical, geological and structural information obtained from underground mapping and core logging is compiled, processed, reviewed and analyzed on site by the geotechnical staff. Conclusions and recommendations are implemented as part of the operations in a timely manner. This ongoing ‘‘live’’ process enables the operators to make the most efficient use of resources both for ground support and excavations as well as to address safety issues, which are the top priority. D 2004 Published by Elsevier B.V. Keywords: Kimberlite geology; Diamond mining; Rock mass rating; Kimberlite pipe contact zone; Kimberlite weathering
1. Project background Since its opening in 1998, the Ekati Diamond Minek has produced more than 10 million carats. Nearly all of the diamond production has been from * Corresponding author. Tel.: +1-604-681-4196; fax: +1-604687-5532. E-mail address: [email protected] (J. Jakubec). 0024-4937/$ - see front matter D 2004 Published by Elsevier B.V. doi:10.1016/j.lithos.2004.03.052
open-pit mining of multiple pipes. However, as some pits deepen, it is planned to convert to underground mining. The Koala North pipe has been selected as a trial underground mine for the purposes of testing mining methods and to provide access to the lower elevations of the Panda and Koala pipes. The upper 40 m of the Koala North pipe was mined in late 2000 as a small open pit to provide grade information and a prepared surface for the transition to underground
348
J. Jakubec et al. / Lithos 76 (2004) 347–357
mining. The Koala North underground mine, North America’s first underground diamond mine, formally opened in November 2002.
2. General geology Koala North forms part of the Lac de Gras kimberlite field that occurs within the central Slave Structural Province in the Northwest Territories of Canada. The Lac de Gras kimberlites range in age from f 45 to 75 Ma and intrude Archean granitoids as well as metamorphosed supracrustals (primarily metagreywackes) of the Yellowknife Supergroup. Multiple Proterozoic diabase dike swarms with ages varying from 2.4 to 1.27 Ga intrude the Archean rocks. Kimberlites are the only Phanerozoic rocks known in the Lac de Gras region, although fossil-bearing mudstone xenoliths within the kimberlites indicate that
sediments must have formed a thin veneer over the older rocks at the time of kimberlite emplacement. A model Rb –Sr age of 53.1 F 0.3 Ma (Geospec, 2002) has been determined for the Koala North kimberlite. The kimberlites in the Lac de Gras province generally form steep-sided, diatreme-shaped pipes. They are relatively small features with surface areas ranging from approximately 0.1 to 20 ha in size. With few exceptions, they are composed almost exclusively of volcaniclastic kimberlite material. This includes fine- to medium-grained crater sediments, ash/mudrich to olivine-rich resedimented volcaniclastic kimberlite (RVK) and primary volcaniclastic or pyroclastic kimberlite (PVK). 2.1. Koala North pipe geology The Koala North kimberlite is a very small body (0.4 ha at surface) situated between the Panda and
Fig. 1. Oblique view looking south across the central development area at Ekati. The Panda open pit is in the foreground; the Koala North underground project is located beneath the existing test pit mid-frame; and the Koala pit and Ekati plant site are in the background. Photo taken in mid-2002.
J. Jakubec et al. / Lithos 76 (2004) 347–357
349
Fig. 2. An isometric southwest-looking view of the Koala North underground mine model. Note that as of end of 2003, the ore has been extracted from the first three levels.
Fig. 3. Brown and black mudstone xenoliths in the drill core. Note the slicken-sided surfaces in disintegrated brown mudstone (top photo).
350
J. Jakubec et al. / Lithos 76 (2004) 347–357
Koala pipes in the central portion of the Ekati property (Fig. 1). The pipe lies within a depression formerly occupied by the Koala Lake and was originally covered by 8 m of water and 15– 20 m of boulder- and gravel-dominated glacial till overburden. Koala North intrudes competent, syn-tectonic biotite granodiorite of the Koala Batholith. Modeling of the Koala North body, based on mapping of the kimberlite/granodiorite contact in the test pit and on interpretation of kimberlite drill intersections, suggests that the lower portion of the pipe has an overall steep-sided inverted cone morphology typical of most kimberlites in the Lac de Gras area (Fig. 2). In plan view, the kimberlite outline is roughly circular with no consistent elongation evident. Drilling data suggest that, below the 2370-m mining level (370 m a.m.s.l.), wall rock contacts typically dip inwards at a regular angle of approximately 86j. Above this elevation, however, the eastern margin of the pipe flattens out considerably. It is likely the irregular morphology of the upper portion of the pipe is controlled to a large extent by fault/joint zones that predate kimberlite emplacement. Drilling at Koala North has intersected kimberlite down to the 155-m elevation ( f 270 m below the top of the pipe) and indicates that it is comprised exclusively of volcaniclastic kimberlite material to this depth. The dominant infill lithology at Koala North is crudely bedded to massive, relatively mud-rich volcaniclastic kimberlite with varying, but generally moderate to high olivine contents. These assemblages appear to represent a ‘‘package’’ of resedimented volcaniclastic kimberlite (RVK). The RVK is gener-
ally dark gray and typically contains between 10% and 20% glassy pale green, partially serpentinized olivine grains. However, olivine contents vary locally with minor intervals of both mud-rich ( < 10% coarse olivine) and significantly more olivine-rich (30 – 60% coarse olivine) material evident in most drill intersections. Olivine grains range in size from < 0.3 to f 5 mm in diameter, but commonly exceed 1 mm. In addition to olivine, a minor amount of quartz (generally less than 0.5 mm in size) is evident. The olivine and quartz occur in a matrix consisting primarily of gray to black very fine grained material comprising serpentine group minerals and relatively abundant probable disaggregated mud (65 – 80% of the rock). The RVK is typically xenolith-poor with small (generally < 3 cm) angular to rounded fragments of biotite granodiorite and black to brown mudstone (Fig. 3) usually constituting less than 10% of the rock. Rare zones with increased xenolith content (up to f 20%) are present in places. Centimeter-scale bedding is moderately developed in places (see Fig. 4), but, in general, the RVK has a massive appearance. Dark brown to black fine-grained sedimentary material occurs as rare blocks/bands ranging in width from tens of centimeters up to 5 m. The morphology and lateral continuity of these intervals are not constrained by the current drilling. However, exposure of the kimberlite in underground workings indicates that mudstone occurs both as xenolithic blocks and in situ lenses/bands of fine-grained sediment. This material is indurated, laminated and fissile. It generally lacks olivine ( < 1%) and other obvious kimberlite components. While the RVK material shows variations in coarse-olivine and, to a lesser extent, xenolith content, it retains a very similar character throughout the
Fig. 4. Layers of fine and coarse clasts in resedimented volcaniclastic kimberlite. The core diameter is approximately 63 mm (HQ size core).
J. Jakubec et al. / Lithos 76 (2004) 347–357
drilled intersections examined. Contacts between olivine-rich and ash-rich sections of the core, as well as between xenolith-rich and xenolith-poor zones, are typically gradational and distinct geological boundaries or markers are not evident. For the purposes of geological modeling, all RVK material (as well as associated minor intersections of mudstone and siltstone) has been allocated to a single, crater phase that dominates the Koala North pipe. This material is believed to have formed by slumping and reworking of pyroclastic material and surface sediments into the kimberlite vent. In addition to the RVK described above, a small lens of coarse-grained olivine-rich rich primary volcaniclastic kimberlite occurs internally in the upper northern portion of the pipe. This rock type is similar in appearance to olivine-rich RVK in that it is dominated by macrocrystic olivine set in a dark finegrained matrix. It is distinguished by a more uniform olivine distribution and massive appearance, a lack of any bedding, an absence of quartz, a less friable texture and a greater degree of competence (i.e., apparent greater rock strength). Xenolithic components commonly make up between 10% and 20% of the kimberlite and include small mudstone clasts, typically well-rounded unaltered granodiorite cobbles and rare carbonized black wood fragments. The interclast matrix is dominated by fine-grained serpentine. This material is interpreted to represent a remnant of an earlier pyroclastic deposit (PKV) preserved on the margin of the pipe.
351
very clear that both RVK and PVK kimberlite are relatively competent rock masses. However, within the RVK, clay-rich intervals show low rock mass competency and also high weathering susceptibility. Geotechnical data were input into spreadsheets and various parameters were calculated. Data were also entered into a 3D computer modeling program and database (Gemcom) and used to construct a geotechnical domain model. 3.2. Geotechnical domain model Geotechnical data were superimposed onto the basic geological model. Based on the physical properties of the rock mass, eight geotechnical domains were identified. The rock mass parameters are illustrated in Table 1 and the schematic geotechnical domain model is shown in Fig. 5. 3.3. Geotechnical issues Two main geotechnical issues were identified during the investigation:
Stability of the kimberlite pipe contact zones. Competency and weathering susceptibility of the weak clay-rich kimberlite.
3.3.1. Kimberlite pipe contact zones The contact zones are variable both in the country rock granodiorite and in the kimberlite. Typically, as the development in the country rock approaches the kimberlite pipe contact, there is a zone of granodi-
3. Premine development geotechnical investigation Feasibility-level geological and geotechnical information was obtained from the surface core drilling program as no underground development was available. 3.1. Drill core geotechnical data Basic geotechnical information was obtained from the exploration drill holes. Laubscher’s (1990) rock mass rating (RMR) classification system was used to log the drill core. This classification system enables capturing of individual geotechnical parameters that are used for rock mass competency estimates. It was
Table 1 Koala North rock mass ratings (RMR) values for the individual geotechnical domains Geotechnical domain
RMR
Overburden Near-surface granodiorite Granodiorite External contact zone Internal contact zone Upper RVK Lower RVK Clay-rich RVK
– 35 – 55 60 – 75 40 – 55 18 – 30 20 – 40 45 – 65 15 – 30
RMR number represents the rock mass competency index from 0 (poorest) to 100 (most competent).
352
J. Jakubec et al. / Lithos 76 (2004) 347–357
Fig. 5. Schematic geotechnical domain model of Koala North kimberlite pipe. Note that both external (outside the kimberlite pipe) and internal (inside the kimberlite pipe) geotechnical contact zones of poorer quality rock mass were considered for the mine design. Rock mass above the dashed line was influenced by the surface weathering and glacier rebound.
orite approximately 1 to 5 m (or greater in some areas) in width with increased joint frequency (Fig. 6). It appears that the emplacement of the kimberlite has overprinted or enhanced the regional jointing. In some areas, there is also a narrow (approximately 1 to 2 m wide) transition zone at the pipe contact in which kimberlite stringers occupy joints within the granodiorite. Although the contact zones are not equally developed around the entire perimeter of the pipe, their stability will negatively impact on the development cost and amount of country rock dilution in the recovered ore. 3.3.2. Kimberlite weathering The weathering susceptibility in the mining sense is rock strength deterioration within the mine life. The degree of weathering is influenced mainly by the clay minerals (that are product of geological weathering) within the kimberlite rock types and by the moisture in the mining environment. Weathering of the kimberlite can adversely impact on several mining activities such as ground support, production blasting and trafficability. Weathering
can also promote the generation of mud in the muck pile and it is clearly an important issue that needs to be addressed in any kimberlite mining. To assess the impact of weathering on mining operations, a series of accelerated weathering tests was conducted on the drill core (Fig. 7). Tests revealed that the majority of the Koala North kimberlite has low weathering susceptibility from the mining point of view; that is, the rock mass will not deteriorate significantly during the mining cycle. It was noticed, however, that certain clay-rich kimberlite and mudstone will disintegrate after exposure to the elements. Such materials appear to form zones of limited dimensions (up to 5 m); it was estimated that they will form less than 10% of the materials being mined.
4. Mining method Koala North is being developed and operated as a highly mechanized operation utilizing a trackless diesel-powered fleet of equipment. A key function
J. Jakubec et al. / Lithos 76 (2004) 347–357
353
Fig. 6. Example of less competent contact zones being developed inside and outside the kimberlite pipe contact. Lightly colored rocks represent the country rock granodiorite.
Fig. 7. Accelerated weathering test of the clay-rich kimberlite reveals high weathering susceptibility of this rock type. Top photo illustrates fresh rock and bottom photo shows the same core after several wetting and drying cycles. The core diameter is approximately 42 mm (NQ size core).
354
J. Jakubec et al. / Lithos 76 (2004) 347–357
of the mine is to serve as a learning experience for underground kimberlite mining in the northern Arctic environment. The project will be dealing with a unique set of mining conditions in this harsh environment.
5.0-m wide 5.0-m high. Because subhorizontal joints are present in the country rocks, the development drift profile has a ‘‘flat back’’ profile to increase back stability. 4.3. Draw points
4.1. Ramp Access to the underground workings is from a ramp developed from the surface. The profile of the ramp is 5.5 5.5 m (Fig. 2). All of the drilling in the granodiorite is conducted using a brine solution due to the presence of permafrost and cold air temperatures. 4.2. Level access drifts Production access drives are being developed off the ramp at regular intervals to access the kimberlite pipe. These drives provide access for stope production, exploratory diamond drilling and installations of mining infrastructure, including sumps, refuge bays and electrical installations (Fig. 8). In country rock granitoids, the current designs call for the level access drifts to be developed at
Production draw points are being developed into and across the kimberlite pipe for slot access, stope drilling and production mucking. These cross-cuts are designed in an arched profile at 4.0-m wide 4.0-m high. Due to the geomechanical and weathering characteristics of the kimberlite, development and production blast-hole drilling must be completed dry. A two-stage dust filtering system has been developed to collect and filter return air from drill holes and remove dust in coarse and then fine fractions. 4.4. Production stopes The mining method to be used at Koala North is open benching. As noted above, the unique mining environment and climatic conditions at Ekati Diamond Minek will involve adaptation of the method.
Fig. 8. Schematic layout of development on a production level. Each grid cell is 50 50 m for scale.
J. Jakubec et al. / Lithos 76 (2004) 347–357
355
Fig. 9. Schematic vertical section with individual production level geometries.
The method is a top-down retreat mining system, similar to sublevel caving but without the caved waste behind the drawn ore. The schematic vertical section with basic dimensions is illustrated in Fig. 9; an aerial view of the 2385 level being extracted at the bottom of the pit is shown in Fig. 10.
5. Geotechnical investigation during mine development
5.1. Geological and structural mapping Basic geology data are collected on a daily basis as mining faces advance. Sketch maps of the development including major geological contacts and structures are constructed and a digital photograph of the face is taken. Geological contacts and major structures are plotted on plans and projected to the next mining level. 5.2. Geotechnical mapping
Underground excavations combined with the underground drilling program provide truly 3D geological and geotechnical data. Such information is input into a computerized database (Gemcom) and the initial 3D pipe geological model is adjusted accordingly. This information provides invaluable technical input to the Mine Planning and Engineering Department as well as to the development contractor. It facilitates adjustment and optimization of mine layout and ground support design.
The geological sketch map of the development acts as a foundation for the rock mass classification a s s e s s m e n t . L a u b s c h e r ’s u p d a t e d M R M R (Laubscher and Jakubec, 2001) classification system is used to describe the main geotechnical parameters, such as intact rock strength, joint frequency, number of main joint sets and joint conditions. Field sheets are compiled into rock mass classification maps and geotechnical domains are delineated. Information from the geotechnical mapping is
356
J. Jakubec et al. / Lithos 76 (2004) 347–357
Fig. 10. Aerial view of the first production level 2385 being extracted at the bottom of the pit.
used for optimizing the layout and for assessing support requirements. Such requirements are communicated on a daily basis with the development crew. Both geological and geotechnical mapping have to be completed on a daily basis because advancing ground support (including shotcrete) rapidly covers the entire face and prohibits any further rock mass assessment.
6. Conclusions The Koala North pipe has been selected as a trial underground mine for the purposes of testing mining methods and to provide access to the lower elevations of the Panda and Koala pipes. The geotechnical and mining investigations will serve as a learning experience for underground kimberlite mining in a northern Arctic environment. Although this mining method is successful when used on several De Beers diamond
operations in South Africa, it has not been tested in this setting. Ongoing geological and geotechnical investigation during the mine development is vital for a mining operation to succeed. At Koala North, both geotechnical and geological sections are fully integrated into a Technical Services Department. This high level of integration enables important information and suggestions related to the ground support and mine development to be forwarded to the mining contractors without delays. Effective communication between the miners and engineers is of paramount importance to ensure a safe and successful operation.
Acknowledgements The authors would like to thank BHP Billiton Diamonds for permission to publish this paper.
J. Jakubec et al. / Lithos 76 (2004) 347–357
References Geospec, 2002. Rubidium – strontium isotope analyses for BHP Billiton Diamonds Inc. Geospec Consultants Limited, internal report. Laubscher, D.H., 1990. A geomechanics classification system for
357
the rating of rock mass in mine design. J. S. Afr. Inst. Min. Metall. 90 (10), 257 – 273 (South Africa). Laubscher, D.H., Jakubec, J., 2001. The MRMR rock mass classification for jointed rock masses. Underground Mining Methods. Society For Mining Mettalurgy, and Exploration (SME), Littleton, CO, pp. 475 – 483.