Pb and S isotopic composition of indicator minerals in glacial sediments from NW Alberta, Canada: implications for Zn-Pb base metal exploration Open Access

The sediment samples used in this study were collected throughout the Zama Lake and Bistcho Lake map sheets (Canadian National Topographic System (NTS) 84L and 84M, respectively). This area lies within the Fort Nelson Lowland region of NW Alberta (Bostock 1967), and is characterized a number of uplands (Buffalo Head Hills, Cameron Hills, Caribou Mountains, and Clear Hills) interspersed amongst broad lowland areas (Pettapiece 1986). The area is drained by the Peace, Hay and Petitot rivers, forming part of the southern Mackenzie River drainage basin which drains northward to the Beaufort Sea (Fig. 1). The flat nature of most of the region reflects the underlying horizontal to gently dipping, sedimentary bedrock.

The uppermost bedrock in NW Alberta is a Cretaceous succession of near-horizontal and poorly-indurated carbonaceous, laminated, non-bioturbated, marine black shales of the Fort St. John Group (Loon River and Shaftesbury formations) and Smoky Group, separated by deltaic to marine sandstones of the Dunvegan Formation (Green et al. 1970; Dufresne et al. 2001; Okulitch 2006). The transition from Lower to Upper Cretaceous strata occurs within the Shaftesbury Formation. A large prominent crustal-scale structure, the Great Slave Lake Shear Zone (GSLSZ), cuts across the study area (Eaton & Hope, 2003; Morrow et al. 2006). The GSLSZ is a northeasterly trending structure that extends more than 700 km from NE British Columbia to Great Slave Lake, Northwest Territories. The GSLSZ is recognized as a cratonic boundary between the Archean Slave microcraton and the Archean Churchhill Province (Hoffman 1987). It occurs as a broad band (up to 50 km wide in places) of intensely sheared rocks bounded by several major basement lateral strike-slip faults with a complex tectonic history dating back to the Early Proterozoic (Hoffman 1987). The GSLSZ system experienced reactivation during the Lauramide Orogeny (Morrow et al. 2006) and created relatively young pathways for hydrothermal fluids, opening up an exploration fairway along the length of the shear zone (Lonnee & Al Aasm 2000; Nelson et al. 2002; Hitchon 2006). The Steen River impact structure, dated at 95 Ma (Carrigy & Short 1968) occurs just to the east of the study area.

NW Alberta is mantled by an extensive cover of unconsolidated sediments deposited during the glacial and interglacial periods of the Quaternary. Sediment thicknesses are variable, ranging from 0–450 m (Pawlowicz et al. 2005, 2007). The cover of unconsolidated sediments completely obscures the bedrock topography which generally reflects Tertiary–early Quaternary drainage. During the Late Wisconsin glaciation (25.0–10.0 ¹⁴C ka BP), the Keewatin-sourced Laurentide Ice Sheet flowed west and SW across northern Alberta towards the Rocky Mountains (Dyke et al. 2002; Dyke 2004; Paulen et al. 2007; Bednarski 2008). Ice retreated from the study area between 12.0–11.0 ¹⁴C ka BP (Dyke 2004), during which time extensive glacial lakes developed over the lowland areas as a consequence of damming of the regional eastward drainage by the retreating glaciers. Extensive fine-grained glacial lake sediments overlying till in the lower portions of the Hay and Peace river drainage basins document the existence of these glacial lakes. Today, the region is poorly drained, secondary streams are not deeply incised, and organic deposits in the form of fens and bogs abound.

A total of 90 large till samples (c. 30 kg) were collected in C soil-horizon (>1 m depth) from hand-dug pits, natural bluffs, and man-made exposures. An additional sample was collected from till (2 m depth), down-ice and proximal to a subcropping prismatic Zn-Pb orebody at Pit N41, at the past producing Pine Point Zn-Pb mine for comparison purposes. Samples were sent to Overburden Drilling Management Limited for sample processing and recovery of indicator minerals. The heavy mineral fraction of each till sample was isolated in a two-step process using a shaking table and heavy liquids (specific gravity 3.2; McClenaghan 2011). Kimberlite indicator minerals, gold grains and other indicator minerals such as metamorphosed/magmatic massive sulphide indicator minerals were identified in the 0.25–2.0 mm non-ferromagnetic fraction of the till samples under binocular microscopes by staff mineralogists at the laboratory. To monitor the accuracy of the heavy mineral separation and mineral identification procedures, several blank till samples as well as till samples spiked with kimberlite indicator minerals were included in the batch (Plouffe et al. 2006a, 2008).

Selected sphalerite grains (n=15) from the heavy mineral concentrates of two till samples were analyzed by electron microprobe to determine their mineral chemistry (Plouffe et al. 2007). The grains were mounted on epoxy stubs, polished, and analysed at SGS Minerals Services. The analyses were conducted with a JEOL 733 Superprobe using an accelerating voltage of 30 kV, a cup electron beam of 30 nA, and a measuring time of 20 seconds. Triplicate analyses were used for quality assurance. Microprobe calibration was completed with mineral standards from CANMET and SPI Supplies including chalcopyrite (Cu Kα measured with the LiF crystal), arsenopyrite (As Lα measured with the TAP crystal), galena (Pb Lα measured with the PET crystal), synthetic AgBiSe2 Cabri-499 (Ag Lα measured with the PET crystal), sphalerite (Zn Lα measured with the TAP crystal), pyrrhotite Cabre-241 (Fe Kα measured with the LiF crystal), greenokite (Cd Lα measured with the PET crystal), hertzenbergite (Sn Lα measured with the PET crystal), synthetic GaAs (Ga La measured with the TAP crystal), pure indium metal (In Lα measured with the PET crystal), and synthetic TlBrI (Tl Lα measured with the LiF crystal).

Geochemical analyses were performed on the silt plus clay-sized fraction (<0.063 mm) of till samples that were separated by dry sieving at the Alberta Geological Survey laboratory. Duplicate and analytical standard samples were added to the batch, and then the material was sent to Acme Analytical Laboratories Limited for analyses. Analyses conducted on the <0.063 mm sized fraction include: (1) 15 g aliquots were submitted for inductively coupled plasma mass spectrometry (ICP-MS) analysis for a suite of 37 minor elements following an aqua regia digestion; and (2) 0.2 g aliquots were analyzed for major elements by ICP emission spectrometry (ICP-ES) minor element abundances by ICP-MS after a LiBO₂ fusion and dilute nitric acid digestion. Detailed field sampling methods, laboratory procedures for heavy mineral separation and identification, quality control and reproducibility are described in Plouffe et al. (2006a).

Isotopic analyses of Pb were conducted on galena (n=10) and sphalerite grains (n=2) from 11 till samples at the Department of Earth Sciences, Carleton University. Each grain was powdered and dissolved in 50 % hydrofluoric acid – 12N nitric acid, then attacked with 8N nitric acid and finally 6N hydrochloric acid. The residue is taken up in 1N hydrobromic acid for Pb separation. Pb is separated using 1N hydrobromic acid to elute other elements and 6N hydrochloric acid to elute Pb. The collected Pb solution is dried, redissolved in 1N hydrobromic acid, and the above procedure is repeated with a small volume resin bed. Samples are loaded onto single Re filaments with phosphoric acid and silica gel, and are analyzed in a TRITON mass spectrometer. All mass spectrometer runs are corrected for fractionation using NIST Standard Reference Material SRM981 Pb standard. The average ratios measured for SRM981 are ²⁰⁶Pb/²⁰⁴Pb = 16.892 + .010, ²⁰⁷Pb/²⁰⁴Pb = 15.431 + .013, and ²⁰⁸Pb/²⁰⁴Pb = 36.512 + .038 (2 s.d.), based on 20 runs between April 2005 and November 2006. The fractionation correction, based on the values of Todt et al. (1984), is +0.13%/amu. Isotopic results are reported in conventional per mil notation (‰).

Sulphur isotope analyses (n=11) were performed on sphalerite powders made from single grains selected from the 0.25–0.5 mm heavy mineral fraction of 10 till samples at the Department of Earth Sciences, University of Ottawa. Samples were weighed into tin capsules with tungstic oxide and then loaded into a Costech elemental analyser to be flash combusted at 1800 °C. Released gases are carried by helium through the elemental analyser to be cleaned and separated. SO₂ gas is carried into the DeltaPlus isotope ratio mass spectrometer for analysis. Analytical precision is +/−0.2 per mil. Isotopic results are normalized with international standards of argentite (IAEA-S1 and IAEA-S2) and sphalerite (IAEA-NBS-123) and are reported in conventional per mil notation (‰).

Reconstruction of glacial ice flow was determined from orientations of streamlined landforms identified on air photographs and satellite imagery, measurement of striations on boulder pavements and clast fabric analysis of tills surficial geology maps of the study area (Plouffe et al. 2004, 2006b; Paulen et al. 2005a, b, 2006a, b; Smith et al. 2005, 2007), and published data (Mathews 1980; Dyke 2004; Bednarski 2008). Bedrock striation data were also collected from the Pine Point mine site in order to document local potential dispersal trajectories from the region's only major Zn-Pb deposits.

Dark grey to black, angular, brittle grains of sphalerite, with rare grains of orange to honey sphalerite, were found in high concentrations (>100 grains) in nine of ninety till samples from the study area (Fig. 2). This zone of high concentration of sphalerite grains extends over an area of approximately 4000 km² and defines a SW-oriented sphalerite dispersal train. One till sample contained 1047 sphalerite grains (normalized to 30 kg sample weight; Plouffe et al. 2006a). The absence of sphalerite grains in the regional till samples collected within the study area and outside of the dispersal train indicates that the background grain concentration is zero. Grains exhibit pristine crystal shapes and fragile morphologies with angular to sub-angular edges (Fig. 3). One to four angular to sub-angular galena grains were recovered in eight of the till samples from the anomalous region. The galena grains are also angular to sub-angular.

Zinc content of the silt plus clay-sized fraction (<0.0063 mm) of till samples shows no correlation with those samples that contain abundant sphalerite in the sand-sized fraction. For example, the till sample with >1000 grains of sphalerite contained only 150 ppm Zn which is well within the Zn background concentration of tills from this region, defined by Plouffe et al. (2006a) at c. 170 ppm (95th percentile). On a regional scale, Zn concentrations in tills are slightly elevated (≥200 ppm) in a broad band oriented NE–SW (>98th percentile) extending subparallel to, and down-ice from, the GSLSZ (Fig. 4; Plouffe et al. 2006a).

A selection of dark black sphalerite grains were submitted for electron microprobe analyses to confirm their identification (Plouffe et al. 2007). A total of 180 analyses were performed for the core and periphery areas of 15 sphalerite grains from two till samples (Plouffe et al. 2007). The average composition of the sphalerite is 33.4 wt.% S, 65.4 wt.% Zn, 0.7 wt.% Fe and 0.43 wt.% Cd with trace amounts (0.3–0.1 wt.%) of Cu, Ag, Se, and In ( Table 1). Compared to the composition of sphalerite from the world class Pine Point Mississippi Valley-type Zn-Pb deposits located 330 km to the NE (Kyle 1981), sphalerite from this study contains, on average, lower levels of Pb and Fe coupled with higher Cd concentrations. Furthermore, colour of sphalerite from bedrock samples at the Pine Point deposits varies from tan, yellow, light red-brown, dark red-brown to dark brown (Fig. 5). In contrast, the majority of sphalerite grains recovered from the northwestern Alberta till samples are dark grey to black.

Lead isotope analyses of sphalerite and galena from till samples are listed in Table 2 and plotted on Figure 6. The analyses are plotted on the average ‘crustal evolution curve’ of Stacey & Kramers (1975) and the ‘shale curve’ of Godwin & Sinclair (1982), a growth curve unique to the ancient western margin of North America. The ‘shale curve’ was constructed using the Pb values from SEDEX deposits with well constrained ages within basinal sedimentary sequences of the Canadian Cordillera and ranging in age from middle Proterozoic to Mississippian. The curve models the isotopic evolution of the upper crustal Pb in the source regions for clastic sedimentary rocks within the North American miogeocline.

Of the ten galena grains recovered from the two till samples, nine plot as one population which exhibits very small variations in ²⁰⁶Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴Pb, and ²⁰⁸Pb/²⁰⁶Pb. These nine galena grains plot within the ‘Pine Point lead’ cluster that includes data from the Pine Point orebodies (Cumming et al. 1990), other Zn-Pb occurrences in the Pine Point district (Paradis et al. 2006), and subsurface sulphide occurrences indrill cores from carbonate sequences of the Western Canada Sedimentary Basin located along the GSLSZ. One galena grain is more enriched in ²⁰⁶Pb/²⁰⁴Pb and plots below the ‘shale curve’ (Fig. 6a) The only two sphalerite grains analyzed for Pb isotopes plot below the ‘shale curve’ away from the ‘crustal evolution curve’. They have higher ²⁰⁶Pb/²⁰⁴Pb and ²⁰⁷Pb/²⁰⁴Pb values than the galena grains (Fig. 6a) and on the ²⁰⁸Pb/²⁰⁶Pb versus ²⁰⁷Pb/²⁰⁴Pb plots, the two sphalerite grains plot within the ‘Cordilleran carbonate lead’ field (Fig. 6b). The Pb in the sphalerite may come from a different reservoir than the Pb in the galena or that the Pb comes from different mineralizing events. The ‘Pine Point Pb’ isotopic signature is quite distinctive. It shows remarkable homogeneity, with ²⁰⁶Pb/²⁰⁴Pb values in a narrow range of +18.167 to +18.189 ‰, indicating either a homogeneous source or a thorough mixing of Pb during extraction, transport, and precipitation (Cumming et al. 1990; Paradis et al. 2006). Modelling of U and Pb concentrations and ratios suggest a depleted lower crustal source consistent with Pb derivation from basement rocks (Nelson et al. 2002; Paradis et al. 2006).

Sphalerite from the 10 till samples displays a range of δ³⁴S values between −14.1 to −6.0‰ with a mean value of −9.0‰ (Table 3). These low values could be explained by the bacterial reduction of coeval seawater sulphate in the depositional site or could reflect remobilization of sulphides in the till during postglacial weathering. Sulphur in the latter case would have come from a source where the bacterial reduction of coeval seawater sulphate had occurred. Another possibility is that mineralization occurred under oxidizing conditions, which leads to fractionation of ³⁴S from ³²S. With the present data, no geological evidence permits us to discriminate between these processes. The low values are different than those reported for Mississippi Valley-type deposits in the northern and southern Cordillera, which are dominantly much heavier (Fig. 7). Sulphur isotope values from the Pine Point deposits range from +12.2 to +27.0‰ (Evans et al. 1968; Sasaki & Krouse 1969; Paradis et al. 2006). Alternatively, δ³⁴S values in sphalerite grains recovered from the dispersal train vary from −14.1 to −6.0‰ (Table 3; Fig. 7) and are clearly lower than at Pine Point. These negative values may reflect the incorporation of isotopically light H₂S (aq) derived from the oxidation of the sulphide-bearing solution. In this case, the primary source of sulphur could have been evaporitic sulphur much like for the Pine Point deposits.

At the onset of glaciation, lobes of ice advanced in a general southwestward to westward direction likely mimicking the trends of the small arrows in Figure 8. At glacial maximum, the Laurentide Ice Sheet flowed westward across the region towards the Rocky Mountains where it abutted the Cordilleran Ice Sheet and was deflected north and south along the mountain front (Bednarski 2008). The Zama Lake area lies almost 300 km east of the confluence of both ice sheets. At glacial maximum, ice-flow trajectories across the Pine Point Pb-Zn deposits were westward (255°) as indicated by striations observed at the abandoned Pine Point mine (Fig. 8). Any mineralized debris derived from the Pine Point deposits and entrained by ice was transported in that general direction, north of the Cameron Hills and well north of the Zama Lake sphalerite dispersal train (thick black dashed lines, Fig. 8). Extensive fluted and otherwise glacially-sculpted terrain (smaller arrows, Fig. 8) exhibit evidence of a number of crosscutting, and at times topographically confined ice flows (Paulen et al. 2007, 2008; Bednarski 2008). Many of these streamlined landforms formed during deglaciation when the ice sheet retreated as a series of streaming lobes most likely in a fashion similar to the onset of glaciation. Thus, it is possible that material from the Pine Point region was transported for short distances towards Zama Lake during either early or late stages of the last glaciation. Clast fabric analyses at three different levels in till, at a site near the head of the dispersal train where the sphalerite grain content is 676 grains/30 kg (Fig. 9; refer also to Fig. 2), indicate that glacial transport and deposition was towards the WSW and SW, along the same trajectories defined for the glacial maximum (see large arrows in Fig. 8).

Surficial mapping and ice flow studies in the region indicate that the sphalerite dispersal train is not likely derived from the Pine Point Zn-Pb deposits 330 km to the NE (Fig. 8). Aside from the regional ice-flow history, there are several factors that argue against the sphalerite anomalies being the product of long-distance glacial transport, comminution, and deposition of erratic material from the Pine Point area, and instead favour a proximal bedrock source (Paulen et al. 2007; Plouffe et al. 2008). First, the nine sample sites with high sphalerite grain counts (and eleven with lesser concentrations) are situated within a geographically restricted area north of Zama Lake and the grain concentration does not increase up-ice towards the Pine Point locality. Second, the silt plus clay-sized fraction of the tills does not contain elevated concentrations of Pb and Zn, suggesting that glacial comminution of sand-sized sphalerite and galena grains has been limited. Third, close examination of the mineral grains shows that some grains have a combination of primary crystal structure and sub-angular to angular or pristine morphologies which would not likely have survived extensive glacial erosion and transport. Galena (hardness = 2.5) and sphalerite (hardness = 3.5–4) are relatively soft minerals, and would not be expected to survive glacial comminution for great distances. Lastly, the sphalerite grains in till in NW Alberta have dissimilar optical and chemical properties compared to the sphalerite at the Pine Point deposits (Kyle 1981).

The Steen River impact structure (Winzer 1972) in early Late Cretaceous rock is up-ice of, and proximal to, the sphalerite dispersal train. The impact structure is thought to have potential to host base metal mineralization along fault systems within the outer and inner rims of the crater (Germundson & Fischer 1978; McCleary 1997). However, the impact structure is completely covered by more than 200 m of Loon River and Shaftesbury Cretaceous sediments (Molak et al. 2001). Thus, the Laurentide Ice Sheet would not have had erosional access to the impact structure and associated mineralized rocks.

Several Zn-Pb deposits, including the past producing Pine Point mine, and subsurface sulphide occurrences in Devonian carbonates in NW Alberta, occur along the NW-trending GSLSZ suggesting a direct link between mineralization emplacement and this regional structure. The elevated concentrations of sphalerite and galena in till are located immediately north of Zama Lake where an occurrence of sulphides was previously identified in the Chevron Lutose well at 1265.4–1304.9 m depth within subsurface Palaeozoic carbonate rocks (Dubord 1987; Turner & McPhee 1994; Nelson et al. 2002; Paradis et al. 2006). The reported concentrations of metal in this interval was 3.1% Zn and 0.05% Pb occurring ashoney-coloured disseminated sphalerite and no visible galena (Turner & McPhee 1994).

The distinctive Pb isotopic signature of the sulphides recovered from the drill-core (Nelson et al. 2002; Paradis et al. 2006) and from the till in this study support the contention that the GSLSZ and other subparallel NE-trending faults are major structures that most likely played a role in focusing hydrothermal fluid to depositional sites.

Geological and isotopic evidence points to a key role for the fault systems in the localization of base metal mineralization within the Pine Point district and other areas along the GSLSZ (Rhodes et al. 1984; Morrow et al. 2006). Recent research on Pb and Zn in northern Alberta formation waters concluded that exploration should focus on these shear zones and faults, up which geothermal fluids might have migrated (Hitchon 2006). Nelson et al. (2002) and Paradis et al. (2006) speculated that the ‘Pine Point lead’ was leached directly from rocks of the basement by fluids circulating in deep fault structures, such as the GSLSZ and/or other NE-trending faults. We hypothesize that such mineralizing fluids might have precipitated sphalerite with some galena in the Shaftesbury Formation black shale in a marine anoxic environment which might have existed in this region near the transition from the Lower–Upper Cretaceous (Dufresne et al. 2001). In their study on black shales, Arthur & Sageman (1994) have identified a number of oceanic anoxic events in the United States Western Interior Seaway, necessary for the preservation of sulphide mineralization, near the transition from the Lower to Upper Cretaceous time. Therefore, the environment might have been favourable at given times for mineralization, possibly a vein system given the grain size and morphologies recovered from the till samples, in the shale of the Shaftesbury Formation. The bedrock mineralization, yet to be discovered, was exposed to glacial erosion resulting in large amounts of sphalerite and galena in till of the Zama Lake region as here reported. Despite recent exploration activity (Hawkins 2008), the bedrock source(s) of the sphalerite and galena grains recovered from till in the study area remains unknown.

Isotopic fingerprinting of indicator minerals is a relatively new, powerful exploration tool to help determine provenance of indicator minerals grains. Other isotopic work, such as age dating, is a relatively new technique to assist in deciphering past glacial flow trajectories and to ultimately delineate the provenance of erratics and mineral grains. Additional research will be conducted using Zn isotopic analysis (δ⁶⁶Zn) which may also be a useful tracer for deciphering the hydrothermal system from which the sphalerite grains in till are derived (e.g. Kelley et al. 2009).

Lead and S isotopic composition of sphalerite and galena grains recovered from till samples in NW Alberta serve to demonstrate that the 4000 km² sphalerite dispersal train in that region is clearly not derived from the Pine Point Mississippi Valley-type Zn-Pb deposits located 330 km to the northeast, but rather, is interpreted to be the result of glacial erosion of undiscovered proximal Zn mineralization hosted in the sedimentary bedrock. Firstly, the Pb isotopic values of the galena recovered from the till in NW Alberta are identical to those of the Pine Point district and other occurrences along the GSLSZ providing a direct link between the undiscovered mineralization and this structural break. Furthermore, because the sphalerite dispersal train occurs in close proximity to the GSLSZ, we suggest that the bedrock source(s) for the sphalerite and galena grains in till is probably a local sulphide occurrence(s) situated along the GSLSZ, hosted in the Cretaceous bedrock. Secondly, the δ³⁴S values determined on sphalerite grains recovered from till in NW Alberta are lower than the levels measured in the mineralization at Pine Point, indicating that the sphalerite in till is not derived from that deposit.

In general, sulphide indicator minerals such as sphalerite and galena can be useful for exploring for base metal deposits in reconnaissance, regional and property-scale surveys in glaciated terrain (cf. Shilts 1984; Paulen 2009). Analyzing the isotopic composition of indicator minerals recovered from surficial sediments remains to be fully developed for a variety of isotopes from various minerals derived from a multitude of deposit types. Analyzing the isotopic composition of specific minerals could serve to fingerprint specific deposits and potentially provide indication of economic potential in the case of undiscovered mineralization. Such approaches could be used for indicator minerals recovered from glacial and non-glacial sediments (e.g. stream sediments) in glaciated and other terrains. In this study, using these indicator minerals may eventually lead to the discovery of one or more Zn-rich base metal deposits hosted within Cretaceous shale rocks in the Western Canada Sedimentary Basin.

The authors would like to acknowledge M.M. Fenton and J.G. Pawlowicz of the AGS for insight into the regional Quaternary geology of northern Alberta. We thank G.J. Prior and D. Goulet at the AGS for their contribution of sample preparation and rigorous quality control of our indicator mineral and geochemistry samples. B. Cousens at Carleton University conducted the Pb isotope analysis and P. Middlestead at the Ottawa University conducted the S isotope analysis. Discussions with P. Hannigan and K. Dewing (GSC Calgary) served to orient this study. Capable field assistance was provided by T. Ahkimnachie, L. Andriashek, H. Campbell, B. Griffiths, C. Kowalchuk, R. Metchooyeah, R. Peterson, J. Sciarra, M. Tarplee, M. Trommelen and J. Weiss. This study is part of R. Paulen's PhD thesis (University of Victoria), funded by the Alberta Geological Survey and the Geological Survey of Canada. A thorough review by I. McMartin (GSC Ottawa) greatly benefited an early version of this manuscript. Helpful manuscript reviews by W. Goodfellow, D. Kontak and B. McClenaghan greatly benefitted this paper. ESS/GSC Contribution No. 20090458.

The copyright has been updated to Open Access.