The following DRAFT paper has been submitted by BGC Engineering for consideration to GeoManitoba, the 2012 Canadian Geotechnical Conference. First author is Gabe Hensold, who has granted permission to post this draft.
Potential Development at Dublin Gulch, Yukon: Part 1 – Engineering Geology
G. Hensold1, P. Quinn2, D. Welkner1, and W. Newcomen3
BGC Engineering Inc., 1Vancouver; 2Victoria, and 3Kamloops, British Columbia, Canada
Dublin Gulch, located in the heart of central Yukon, is the site of potential development of an open pit heap leach gold mine. The site’s geotechnical characteristics are affected by a complex geological history. The local metasedimentary and intrusive bedrock is variably folded, faulted, and contact metamorphosed. Dublin Gulch remained ice-free during the most recent continental glaciations; as a result, the rock mass has been left in place for at least 200,000 years and has developed a thick weathering profile. Surface sediments and bedrock are further modified by ongoing periglacial and colluvial processes, and excess ice is commonly encountered in widely-distributed, discontinuous permafrost. Valley bottom sediments have also been re-worked by a century of human activity. An empirical classification system, based on simple index parameters, was used to separate bedrock into three general engineering types at the project scale for design purposes.
Situé dans la zone centrale du Territoire du Yukon, Dublin Gulch est un projet de mine d’or a ciel-ouvert en cours d’étude et développement. Le faciès local est le résultat d’un passé géologique complexe en raison de quoi les formations metasédimentaires et intrusives locales sont typiquement affectées par des déformations de failles, de plis ou par contact métamorphique. Le site, hors des zones d’érosions glaciaires de ces 200,000 dernières années, présente un profil de météorisation de la roche en profondeur. Les couches sédimentaires et rocheuses y sont sujettes à la périglaciation et redistribution colluviale, impliquant la présence de glace excédentaire dans un pergélisol discontinu. Les dépositions alluvionnaires ont de plus été déplacées par un siècle d’exploitation. Un système de classification fondé sur des paramètres d’index simples, a permis de dissocier la masse rocheuse en trois grandes catégories de roche qui permettent un dimensionnement géotechnique plus systématique à l’échelle du projet.
Dublin Gulch, Yukon, is the site of a potential open pit heap leach gold mine development. The property is located at the confluence of Haggart Creek and Dublin Gulch, approximately 40 km north of Mayo, and 15 km northwest of Elsa, YT (Figure 1, inset). The site presents a number of interesting geotechnical engineering challenges related to its geologic history. The present paper, Part 1 of a set of two companion papers, describes the geologic setting, with a focus on creating a framework for understanding potential engineering challenges associated with the planned development. A second paper, Part 2, describes the associated engineering challenges in more detail and discusses some of the methods used to better understand and solve these challenges.
2 GEOLOGIC HISTORY
2.1 Bedrock geology
Dublin Gulch lies within a region that was deformed and contact metamorphosed in the late Jurassic to early Cretaceous period by north-directed folding and thrusting. The site sits on a hanging wall to the south of major thrust faults that accommodated north-south shortening (Murphy 1997). Intrusive plutons in the area, including a granodiorite stock immediately south of Dublin Gulch, were emplaced by subsequent magmatism associated with the deformation (Mortensen et al. 2000). Most of the modern rock structure at Dublin Gulch can be attributed to this event, and to a period of north-south extension shortly afterwards – around the time of gold mineralization at Dublin Gulch – that caused the development of steeply dipping, E- to NE-striking extension veins and NNW-striking strike-slip fault veins (Stephens et al. 2004). These structures occur pervasively across the project area and form important structural discontinuity sets for consideration in engineering designs.
The bedrock in the Dublin Gulch area is primarily clastic metasedimentary rock of the Hyland Group, consisting of intercalated and deformed quartzites and phyllites, and to a lesser degree, schists and carbonates. These rocks are intruded by a granodiorite intrusive stock belonging to the Tombstone Plutonic Suite (Murphy 1997) that outcrops above and east of Haggart Creek, south of Dublin Gulch. The stock is elongated 70E, covers an area 2 km wide by 5.5 km long (Tetra Tech 2012), and is dated at approximately 93 million years (Smit et al. 1995; Figure 1). The Eagle Gold mineral deposit is located at the narrowest extent of the pluton. A hornfelsed thermal aureole surrounds the intrusion, within which the Hyland Group rocks have been altered and contact metamorphosed. The aureole extends about 300 – 1000 m outward from the intrusion along the ground surface (Stephens 2004; Fig. 1). Figure 2 shows typical examples of rock encountered in outcrops around the project site.
2.1.1 Structural domains
Structural data were compiled from pre-2011 investigations by BGC (surface mapping and oriented-core borehole mapping), outcrop mapping during the 2011 field season, and mapping published by Stephens et al. (2004). This information was gathered to support slope stability analysis of cut-slopes at certain proposed mine facilities, and to divide the project area into general domains with similar rock type and structures (Figure 1). The intrusive rocks comprise one single domain (B), and the metasedimentary rocks were divided into four separate domains (A, C, D, and E). Domains A and C occupy the southern three quarters of the study area and are separated by a major west-plunging anticline that runs from northeast to southwest, with its estimated axis passing just north of the proposed open pit. In domain A on the south limb of the anticline, the average foliation dips shallowly to steeply southwest; in domain C on the north limb of the anticline, the foliation dips moderately northwest. Domain D covers an area around Tin Dome where the bedrock is phyllitic and intensely folded, resulting in a distribution of somewhat irregular foliation orientations. Domain E, covering the upper eastern side of Ann Gulch, is the southwest corner of an area stretching to the north and east of the study area in which the foliation dips mostly north.
2.2 Surficial geology and glacial history
Three general periods of glacial advance have been described in Yukon Territory. From oldest to youngest these are the pre-Reid, Reid, and McConnell glaciations. The pre-Reid glaciation consisted of multiple undifferentiated episodes, beginning at least 2.58 million years ago (Froese, 1997) and was the most extensive of the three glaciations in the area. During the Reid glaciation from approximately 300,000 to 200,000 years ago (Bond 1996), fingers of ice extended up Haggart Creek valley and into Dublin Gulch from major trunk valleys to the north and south where ice flowed in a west-southwest direction. (Bond 1999). These fingers were oriented transverse to the regional ice flow direction and terminated in the Haggart Creek valley. Their extent is well-mapped near Dublin Gulch, reaching approximately 1000-1050 m elevation along the hillsides (Figure 3). During the most recent (McConnell) glaciation, the Stewart Plateau remained ice-free (Bond 1999). As a result, the project site is somewhat unique in that the valley bottoms were last glaciated at least 200,000 years ago, with the uplands last glaciated even longer ago, i.e. approximately 800,000 to 1,000,000 years ago.
The surficial geology of the Dublin Gulch area has been mapped by Bond (1998) (Figure 4). Colluvium is found on sloping uplands, with thickness varying depending on slope aspect and elevation. Pleistocene and Holocene colluvial deposits are abundant in the project area and generally consist of diamicton, gravel, shattered bedrock, and lenses of sand and silt derived from bedrock and surficial materials by a variety of chemical and physical weathering processes. Transport of surface material occurs as solifluction or creep, sheetwash, and mass wasting; these processes are common on all slopes in the area.
Glacial till (moraine) is infrequently observed in the study area (Figure 4) because Haggart Creek Valley is aligned transverse to the regional Cordilleran ice flow (Figure 3). Where till does occur, it is generally either a silty or sandy clay matrix with varying proportions of larger clasts up to cobble size. The valley bottom in Dublin Gulch is dominated by alluvial/fluvial soils and placer mining tailings (anthropogenic soils in Figure 4). The north facing uplands are covered by a blanket of colluvium over bedrock, in contrast with the southern facing uplands, where bedrock is nearer to surface and covered by a thinner veneer of colluvium. The Haggart Creek Valley to the west of the project site is similarly filled with a mix of alluvial deposits and placer tailings. A till blanket has been mapped along the east side of Haggart Creek, south of its confluence with Dublin Gulch.
2.2.1 Effect of geologic history on engineering geology
The geomechanical behaviour of the bedrock at Dublin Gulch is strongly influenced by the site’s geologic history. The intrusive stock that includes the ore body contains relatively competent granodiorite, which is highly weathered near the surface and becomes stronger and more massive with depth. The hornfels aureole, or “halo,” surrounding the intrusion consists of relatively competent silicified metasedimentary rocks; however, a narrow contact zone of completely deteriorated altered granodiorite, often observed as coarse sand, occurs at the contact between the silicified metasediments and the granodiorite. An intrusive dike within the hornfelsed metasediments was observed to be similarly altered and deteriorated.
Outside the hornfels aureole bedrock consists of foliated metasedimentary rock, which grades from very weak phyllite to stronger quartzite. Often, these lithologies are interbedded at the outcrop scale in 0.1 m to 1 m thick beds. Rock mass quality gets progressively worse going northwards from the intrusive body, with weak to moderately strong intact material (R2-R3; Hoek and Brown 1997) and discontinuity spacing averaging 0.03 m to 0.2 m (closely to very closely spaced; ISRM 1978).
Due to the absence of recent glacial activity, a deep weathering horizon exists across the area which affects the character and behaviour of local bedrock and surficial materials. Rock mass quality in weathered rocks is typically low, ranging from very poor to fair, i.e. with Rock Mass Rating, or RMR (Bieniawski, 1976) values estimated as low as 10 to 20 and occasionally as high as 40 to 60, with rare observations of good or very good quality rock in the quartzite or granodiorite. The depth of the weathering horizon is variable but often deep, on the order of tens of metres; however, fresh/competent rock is observed in isolated outcrops or near the surface along certain sections of road cuts. These more competent zones likely represent the uppermost grade of rock quality, and their visibility stems from being surrounded by weaker, more erodible material. In many areas, the rock is sufficiently weathered, altered, and/or disintegrated as to exhibit soil-like mechanical behaviour. Figure 5 shows an example of crushed, weathered rock exposed in a test pit.
The tectonic and glacial history lead to several engineering challenges associated with bedrock conditions. These are discussed in more detail in the companion (Part 2) paper; however, there is a general absence of durable rock for use in high quality engineering applications, such as concrete aggregate, large block riprap, drain rock, structural fill or durable rock fill. The typically poor quality rock mass has important implications for the design of foundations, particularly those with heavy static or vibratory loads, and where foundations will need to be placed on sloping ground in rugged terrain. The pervasive foliation and associated joints and faults in the rock also present challenges for the design of engineered cut slopes. For example, the orientation of foliation follows natural slope surfaces in much of the area south of Dublin Gulch, forming potential planes of weakness of concern for engineered cut slopes.
2.3 MODERN PROCESSES
2.3.1 Periglacial activity
The Dublin Gulch project site is located within a zone of extensive discontinuous permafrost (Natural Resources Canada 1993). Permafrost occurs typically on north- and east-facing slopes at higher elevations, and within poorly drained valley bottoms. Frozen ground, when observed, is generally encountered immediately below the organic cover, although frozen organics are also encountered. The permafrost is relatively warm, with measured ground temperatures in permafrost typically ranging between 0C and -1C.
The project site has been the subject of geological and engineering investigation since the late 1970s. This work has included exploration drilling, as well as geotechnical drilling and test pitting. A total of 377 subsurface observations, indicating whether frozen ground is present or absent, are available from various engineering reports. The majority (n = 271) of these observations are from test pit investigations, and the remainder are from drill holes, including both diamond drilling (with or without triple tube core barrels), and auger drilling (solid stem, hollow stem and CRREL barrel) with standard penetration testing, where appropriate. Figure 4 shows the distribution of observations of frozen and unfrozen ground. Of the 377 observations, ground was frozen in 161 (43 % of observations) and not frozen in 216 (57 %).
Most drill holes and test pits were completed in summer months, typically between June and September. The peak thaw at the bottom of the active layer occurs in mid to late September, as observed in selected borehole thermistors. Therefore it can be expected that some of the observations represent seasonally frozen ground, rather than permafrost. The distribution of actual permafrost can therefore be expected to be somewhat less pervasive from what is illustrated. However, this distinction may not be particularly important for design and construction purposes, which will be influenced by the presence of warm, easily disturbed, and potentially ice-rich frozen ground late into the normal construction season.
Permafrost is almost completely absent in the valley bottoms where placer mining activities have resulted in reworking of sediments and removal of the insulating organic cover.
Construction challenges will be most significant where warm permafrost that is also ice-rich is disturbed and exposed to thawing during the summer, since ice-rich permafrost can suffer significant loss of strength and stiffness upon thawing. Of the 161 subsurface observations of frozen ground, 145, or roughly 90 %, were noted to contain excess ice. Where excess ice is present, it is usually confined to a relatively thin horizon within 2 m to 5 m of the ground surface; however, there are isolated observations of thicker ice-rich permafrost at select borehole locations. Figure 6 shows massive ice extruded from a CRREL barrel, encountered at a localized feature in a north-facing valley bottom, consisting of ice-rich colluvium with variable ice content to about 30 m depth. The local presence of ice-rich permafrost in selected areas of the site will require careful planning in the location and timing of mine earthworks activities, and due care in planning for management of ice-rich spoil materials.
2.3.2 Anthropogenic Influence
Placer miners discovered gold in Dublin Gulch around 1898, and tungsten placer concentrates in 1904. Since formal documentation of placer mining began in 1978, about 110,000 ounces of placer gold have been recovered (Scott Wilson RPA, 2010). The long history of placer mining activity in the area has transformed the valley bottom, leaving extensive deposits of re-worked materials, or placer tailings. These man-made deposits are highly variable in texture, density and thickness, and cover an area roughly 1.5 km long by 200 m wide, extending along much of the Dublin Gulch valley bottom. Typical placer tailings are shown in Figure 7, and the extent of placer tailings in the valley bottom is illustrated as “Anthropogenic soils” in Figure 4.
The placer tailings are distributed along the valley somewhat haphazardly, in numerous deposits of different evident origin. This includes a small number of piles of boulders and cobbles, extensive deposits of gravelly sand or sandy gravel, and isolated deposits of soft, wet sandy silt. These latter fine-grained materials appear to have developed in former settling ponds. Coarse clasts within the tailings are typically derived from more durable local rocks, such as granodiorite and quartzite. The finer matrix contains a mix of those materials, but is dominated by deteriorated schist and phyllite.
The placer tailings comprise both a valuable engineering resource and a construction challenge. Much of the sand and gravel is expected to be suitable for re-use in various engineering applications, once suitably processed by crushing, screening and/or potentially washing. However, these deposits, being of non-uniform density, and largely below the valley-bottom water table, may be susceptible to differential settlements, or possibly to liquefaction under heavy static loads or extreme earthquake loads.
3 EMPIRICAL ROCK TYPES
General engineering characterization of Dublin Gulch bedrock is needed for a variety of recommendations for the proposed mine infrastructure, such as building foundation grades, cut-slope angles, and estimation of effort required for excavation. The rock mass quality depends on lithology, weathering, alteration, and tectonic damage; however, these properties are highly variable, sometimes across short distances. Furthermore, available subsurface data coverage is limited, being particularly sparse in areas of low rock mass quality where drilling recovery is low, and is compiled from a variety of sources including several previous site investigations in addition to BGC’s own work. To address these challenges, a site-specific empirical classification system was devised to identify three broad geotechnical “rock types” from simple index parameters available in most subsurface data. These rock types divide site rocks based on expected mechanical behaviour, and are independent of lithology and structure. They are used to approximate “first-pass” estimates for general, qualitative engineering recommendations (Table 1). Detailed investigations were undertaken at key mine infrastructure sites for the purpose of developing more specific recommendations to account for local conditions, in addition to the general rock “type”. See the companion paper, Part 2 – Engineering challenges, for further detail.
Table 1. Empirical classification for Dublin Gulch rock types, and typical generic engineering recommendations
Note: 1. These are typical recommendations for simple situations, provided as illustrative examples only.
“Type 3” rock is usually the first rock-like material underlying the overburden soil materials (Figure 8); however, sharp contacts between overburden and Type 2 or Type 1 rock have been observed occasionally. Type 3 rock is defined as rock that is “highly weathered” or better (i.e. weathering grade 4 or better, ISRM 1978), and has intact strength grade greater than R0 (Hoek and Brown 1997); i.e. has UCS strength of > 1 MPa. In-situ rock material that does not meet these criteria is termed “highly to completely weathered bedrock” and classified as an overburden material (example shown in Figure 5). “Type 2” rock is defined as rock that meets the minimum requirements of Type 3 rock, and with Geological Strength Index (GSI, Hoek and Brown 1997) or Rock Mass Rating (RMR, Bieniawski, 1976) of 30 or greater, and core recovery during drilling of 50% or greater. Note that the % recovery criterion was used simply to allow confidence in interpretations from recovered core. Where GSI and/or RMR data are unavailable, an average Rock Quality Designation (RQD) index of 10% or greater serves as an alternate, equivalent criterion. “Type 1” rock is defined as rock meeting the minimum requirements of Type 2 rock, but also having GSI, RMR or average RQD exceeding 40. Figure 8 illustrates the typical differences in rock types, with highly to completely weathered rock overlying Type 3 rock, which overlies a zone that varies between Type 2 and Type 1 rock, depending on joint density.
Figure 8. Outcrop at Dublin Gulch showing transition to competent bedrock with depth. A) Dilated and partially transported bedrock (colluvium), B) Highly weathered and fractured, partly dilated bedrock (Type 3), C) Moderately weathered and fractured bedrock (Type 2 and Type 1). Spacing between white bars along base of outcrop surface is 1 m.
The engineering geologic conditions at Dublin Gulch, and their associated unique challenges for development, are the product of a complex geologic and modern history. The local rock mass quality is highly variable, being modified by tectonic faulting and deformation, alteration and contact metamorphism related to an intrusive pluton, and a deep weathering profile developed during a long interglacial period. The area is affected by ongoing periglacial processes, with nearly half of the site underlain by permafrost, most of which contains excess ice. Recent human activity extending over a hundred years into the past has resulted in extensive reworking of a very large volume of valley bottom sediments, leaving a wide footprint of anthropogenic soils. The variable but generally poor quality bedrock, extensive ice-rich permafrost and large volumes of unsorted fill present a number of interesting engineering challenges for proposed development of an open pit-heap leach gold mine. A site-specific, empirical classification system has been devised that separates rocks into three engineering types based on simple index parameters available in data from BGC studies and various historical studies. These rock types are used as a “first-pass” estimate to recommend foundation grades, design cuts and predict excavation difficulty. Efforts to understand the engineering behaviour of these complex materials for the design of specific mine facilities are described in greater detail in a companion paper (Part 2) that extends the present discussion of engineering geology.
The writers would like to acknowledge Victoria Gold Corporation for permission to use its data in development of this paper.
Bieniawski, Z.T. 1976. Rock mass classification in rock engineering. In Exploration for Rock Engineering, proc. of the symp., (ed. Z.T. Bieniawski) 1, 97-106. Cape Town: Balkema.
Bond, J.D., 1999. Glacial limits and ice flow patterns, Mayo area, central Yukon. Exploration and Geological Services Division, Yukon Region, Indian and Northern Affairs Canada, Geoscience Map 1999-13, 1:250,000.
Bond, J.D. 1998. Surficial Geology of Dublin Gulch, Central Yukon. Geoscience Map 1998-6. Indian and Northern Affairs Canada, Exploration and Geological Services Division, Yukon Region.
Bond, J.D., 1996. Quaternary history of McQuesten map area, central Yukon. In: Yukon Quaternary Geology, Volume 1, W.P. LeBarge (ed.), Exploration and Geological Services Division, Yukon Region, Indian and Northern Affairs Canada, p. 27-46.
Froese, D.G., 1997. Sedimentology and paleomagnetism of Plio-Pleistocene lower Klondike valley terraces, Yukon Territory. Unpublished M.Sc. thesis, Dept. of Geography, University of Calgary. 153 pp.
Hoek, E. and Brown, E.T., 1997. Practical estimates of rock mass strength. International Journal of Rock Mechanics and Mining Sciences, 34(8): 1165-1186.
International Society of Rock Mechanics (ISRM), 1978. Suggested methods for the quantitative description of discontinuities in rock masses. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 15(6): 319-368.
Mortensen, J.K., Hart, C.J.R., Murphy, D.C., Heffernan, S., 2000. Temporal evolution of Early and mid-Cretaceous magmatism in the Tintina Gold Belt. In: The Tintina Gold Belt: Concepts, Exploration and Discoveries. British Columbia and Yukon Chamber of Mines, Special Volume 2, pp. 49–57.
Murphy, D.C., 1997. Geology of the McQuesten River region, northern McQuesten and Mayo map areas, Yukon Territory (115P/14, 15, 16; 105M/13, 14). Exploration and Geological Services Division, Yukon, Indian and Northern Affairs Canada, Bulletin 6.
Natural Resources Canada. 1993. Canada – Permafrost [map], Fifth Edition, National Atlas of Canada. Available: http://atlas.nrcan.gc.ca/auth/english/maps/environment/land/permafrost [Accessed March 2012].
Scott Wilson RPA, 2010. Pre-feasibility study on the Eagle Gold project, Yukon, Canada. Unpublished report prepared for Victoria Gold Corp. 366 pp.
Smit, J. Sieb, M. and Swanson, C., 1995. Summary Information on the Dublin Gulch Project, Yukon Territory. In: Yukon Exploration and Geology, 1995, Exploration and Geological Services Division, Yukon Region, Indian and Northern Affairs Canada, p. 33-36.
Stephens, J.R., Mair, J.L., Oliver, N.H.S., Hart, C.J.R., and Baker, T., 2004. Structural and mechanical controls on intrusion-related deposits of the Tombstone Gold Belt, Yukon, Canada, with comparisons to other vein-hosted ore-deposit types. Journal of Structural Geology 26: 1025-1041.
Tetra Tech WEI Inc. (as Wardrop Engineering), 2012. Technical Report – Feasibility Study, Eagle Gold Project, Yukon. Unpublished report prepared for Victoria Gold Corp. 419 pp.