Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Origin of sulphides associated with the Spy Sill, Klu Property, Kluane Belt, Southwest Yukon Bell, Cameron C. 1999

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

Item Metadata

Download

Media
831-ubc_1999-0460.pdf [ 65.14MB ]
Metadata
JSON: 831-1.0052721.json
JSON-LD: 831-1.0052721-ld.json
RDF/XML (Pretty): 831-1.0052721-rdf.xml
RDF/JSON: 831-1.0052721-rdf.json
Turtle: 831-1.0052721-turtle.txt
N-Triples: 831-1.0052721-rdf-ntriples.txt
Original Record: 831-1.0052721-source.json
Full Text
831-1.0052721-fulltext.txt
Citation
831-1.0052721.ris

Full Text

ORIGIN OF SULPHIDES ASSOCIATED WITH THE SPY SILL, KLU PROPERTY, KLUANE BELT, SOUTHWEST YUKON By Cameron C. Bell B.Sc. (Honours), McMaster University, 1987 A Thesis Submitted In Partial Fulfillment Of The Requirements For The Degree Of Master of Science In The Faculty of Graduate Studies (Department of Earth and Ocean Sciences) (Geology) We accept this thesis as conforming to the required standard The University of British Columbia August 1999 © Cameron Charles Bell, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Earth and Ocean Sciences (Geology) The University of British Columbia 129-2219 Main Mall Vancouver, Canada V6T 1Z4 Date: 11 ABSTRACT Ni-Cu-PGE rich sulphide mineralization and associated ultramafic sills on the Klu property in Southwest Yukon were studied to determine their origin in the context of regional and worldwide examples of various Ni-Cu-PGE deposits. This sulphide mineralization is related to the Spy Sill, a 100 m thick ultramafic sill of Triassic age, which is believed to be co-magmatic with the Triassic Nikolai basalt. A number of reasonably well studied Ni-Cu-PGE deposits and showings (including the Wellgreen and Canalask deposits) also occur in the Kluane Belt. The host intrusions to these deposits and showings suffer variable degrees tectonism, serpentinization and overburden cover which can obscure the characteristics of the intrusions and associated Ni-Cu-PGE mineralization. The Spy Sill on the Klu property provides an opportunity to study one of the least tectonized and serpentinized intrusions in the Kluane Belt. In addition, an entire section through the Spy Sill is well exposed from top to base. Another ultramafic sill occurs on the Klu property, which appears to be depleted in Ni, Cu, Pt and Pd relative to the Spy Sill. The two intrusions have similar ranges of MgO (25 to 35%) in their peridotite phases, but the Lewis Intrusion has Ni values in the 300 to 800 ppm range, whereas the Spy Sill has Ni values in the 1,100 to 1,400 ppm range. Cu, Pt and Pd show similar levels of depletion. The depletion in whole rock metal contents in the Lewis Intrusion is mirrored by low Ni levels in olivine grains relative to their Fo # (Fo 78 to 80.7 with Ni from 0.05 to 0.11%). The geochemistry of the Nikolai basalt and Kluane-type intrusions on the Klu property indicate that these rocks have originated from a common parental melt. Strong co-genetic trends are displayed in plots of immobile, incompatible elements between samples from the Nikolai basalt and various phases of the mafic to ultramafic Kluane-type intrusions. A gabbro phase of the Spy Sill, marginal to the main peridotite part of the sill, displays geochemical evidence of contamination (including elevated Ce/Yb, Ba/Zr and Th). No evidence of Ni, Cu or Co depletion similar to that observed in the Noril'sk area of Russia is evident in any particular part of the Nikolai basalt. However, the Nikolai basalt does display significant overall depletion in Cu and Co relative to the basalt occurring in the western facies of the terrane, the Karmutsen basalt. The parental composition predicted by olivine compositions from the two intrusions, is close to close to that of the Nikolai basalt (averages of Mg# 55.3 and 57.9 for the two sections). Ni-Cu-PGE rich marginal gabbro contains higher MgO levels (average of 12.8%) than non-mineralized marginal gabbro. Within the suite of mineralized marginal gabbro, there is no correlation between MgO and Ni. The highest Ni and Co levels occur in massive sulphide lenses at the lower marginal gabbro-sediment contact. The highest Cu, Pt and Pd levels occur in massive sulphide lenses in I l l the footwall sediment. Ni, Cu and Co display an overall increase with increasing S contents in Ni-Cu-PGE rich samples whereas Pt and Pd values correlate poorly with S. Ni-PGE-Cu-Au values in 100% sulphide/chondrite for the Spy showings show patterns which are most similar to those from flood basalt-related deposits. Pd/lr versus Ni/Cu and Ni/Pd versus Cu/lr plots suggest PGE distribution has an overall control from olivine and chromite fractionation and a local control from sulphide fractionation. Ultramafic sills and related gabbro sills on the Klu property are emplaced within siltstone, shale and limestone of the Permian Hasen Creek Formation. This formation also contains Ni-Cu-PGE barren sulphide layers close to the stratigraphic level of the Spy Sill. Sulphur isotopic data from Ni-Cu-PGE mineralization and potential sulphur sources show that the sulphur isotopic signature of the showings is the result of mixing between mantle and synsedimentary sulphur. The various types of Ni-Cu-PGE mineralization on the Klu property have median S^S values between -1.5 and -7 700. Ni-Cu-PGE barren synsedimentary sulphide from the Hasen Creek Formation is the probable sulphur contaminant. Ni-Cu-PGE rich material appears to be related to an early magma pulse. Specific flows of Ni-Cu-PGE depleted magma are not recorded in the Nikolai basalt due to ancient (Triassic) erosion of such depleted flows or later tectonism. Also, there may be a lack of connectivity between the mineralized sills and magma that reached the surface. Metals were stripped from part of a lower magma chamber by sulphide equilibration; this magma is represented by the Lewis Intrusion. During sill emplacement, synsedimentary sulphur was assimilated from sulphidic layers in the Hasen Creek Formation. Most Ni-Cu-PGE mineralization is associated with marginal gabbro pulses. The magma represented by the Lewis Intrusion is either an earlier sill (before the Spy Sill) or the leading edge of a magma pulse that also formed the Spy Sill which contains magma with normal or elevated Ni-Cu-PGE levels. No economically significant quantities of Ni-Cu-PGE rich sulphide are known on the Klu property, but the nature of the mineralization suggests a process capable of producing Ni-Cu-PGE enriched sulphide took place. This is known to have occurred on a larger scale at the nearby Wellgreen deposit. iv Table of Contents Page Abstract ii Table of Contents iv List of Tables vi List of Figures vi List of Plates xi List of Appendices xii Acknowledgements xiii Section 1: Introduction 1.1: Purpose of Thesis 1 1.2: Scope of Thesis 1 Section 2: Geological Setting 2.1: Location, Access and Topography 3 2.2: Exploration History 3 2.3: Previous Studies 2.3.1 Regional Mapping 5 2.3.2 Metallogenic Studies 5 2.4: Regional Geology 5 2.5: Geology of the Klu property 2.5.1 Property Geology 9 2.5.2 Mineralization Associated with the Spy Sill 16 Section 3: Petrology and Geochemistry of the Spy Sill and Associated Intrusions 3.1: Introduction 18 3.2: Lithological Description 18 3.3: Petrography 3.3.1 Peridotite 21 3.3.2 Marginal Gabbro 21 3.3.3 Maple Creek Gabbro 25 3.4: Geochemistry (Major Element, Trace Element, REE, Ni-Cu-PGE, Sulphur Isotope) 3.4.1 Methods 25 3.4.2 Results 27 3.5: Olivine Geochemistry 3.5.1 Methods 58 3.5.2 Results 58 V Page Section 4: Petrology and Geochemistry of the Nikolai Basalt 4.1: Introduction (descriptions of sections sampled: 63 Burwash Creek, Halfbreed Creek and Klu property) 4.2: Lithological Description and Petrography 63 4.3: Geochemistry (Major Element, Trace Element, REE, 63 Ni-Cu) 4.4.1 Methods 63 4.4.2 Results 64 Section 5: Geochemical Comparison between Peridotite, Marginal Gabbro, Maple Creek Gabbro and Nikolai Basalt 5.1: Genetic Relationships and Contamination 87 5.2: Metal Depletion (Ni-Cu-Co) 93 Section 6: Petrology and Geochemistry of Ni-Cu-PGE Mineralization Associated with Spy Sill and Sulphide/Oxide Layers in Hasen Creek Formation Sediments 6.1: Introduction 97 6.2: Lithological Description 97 6.2: Petrography 99 6.3: Geochemistry 6.3.1 Methods 104 6.3.2 Results (Ni, Cu, Co, Zn, PGE, Se, 106 Te, As, S isotopes) Section 7: Conclusions 7.1: Introduction 129 7.2: Major Conclusions 7.2.1 Geochemical variation between the Spy Sill 129 and the Lewis Intrusion 7.2.2 Relationship between the Nikolai volcanics and 129 Triassic intrusions 7.2.3 Metal depletion and contamination in the 130 Nikolai basalt 7.2.4 Emplacement model for Triassic intrusions 130 7.2.5 Olivine geochemistry 131 7.2.6 Sulphide geochemistry 131 7.2.7 Sulphur isotopes 133 7.2.8 Emplacement and significance of Ni-Cu-PGE 133 mineralization References 135 vi List of Tables Page Table of Formations 12 List of Figures Page Figure 2-1: Klu property location map 4 Figure 2-2: Stratigraphic section 7 Figure 2-3: Kluane Belt geology and section location map 8 Figure 2-4: Klu property geology map 10 Figure 3-1: Spy Sill stratigraphic section 19 Figure 3-2: Lewis Intrusion stratigraphic section 20 Figure 3-3a: Plot of stratigraphic height versus Ni in the Spy Sill 28 Figure 3-3b: Plot of stratigraphic height versus Cu in the Spy Sill 28 Figure 3-3c: Plot of stratigraphic height versus MgO in the Spy Sill 28 Figure 3-3d: Plot of stratigraphic height versus Cr in the Spy Sill 28 Figure 3-3e: Plot of stratigraphic height versus Ni/MgO - Spy Sill 29 Figure 3-3f: Plot of stratigraphic height versus Mg# - Spy Sill 29 Figure 3-3g: Plot of stratigraphic height versus Ni/Cu - Spy Sill 29 Figure 3-3h: Plot of stratigraphic height versus Cu/Zr - Spy Sill 29 Figure 3-3i: Plot of stratigraphic height versus T i 0 2 - Spy Sill 31 Figure 3-3j: Plot of stratigraphic height versus Co - Spy Sill 31 Figure 3-3k: Plot of stratigraphic height versus Au - Spy Sill 32 Figure 3-3I: Plot of stratigraphic height versus Pd - Spy Sill 32 Figure 3-3m: Plot of stratigraphic height versus Pt - Spy Sill 32 Figure 3-3n: Plot of stratigraphic height versus Pt/Pd - Spy Sill 32 Figure 3-3o: Plot of stratigraphic height versus Ce/Yb - Spy Sill 33 Figure 3-3p: Plot of stratigraphic height versus S - Spy Sill 33 Figure 3-3q: Plot of stratigraphic height versus As - Spy Sill 34 Figure 3-3r: Plot of stratigraphic height versus 5 3 4S - Spy Sill 34 Figure 3-3s: Plot of stratigraphic height versus Se - Spy Sill 34 Figure 3-3t: Plot of stratigraphic height versus S/Se - Spy Sill 34 Figure 3-3u: Plot of stratigraphic height versus Ba - Spy Sill 35 Figure 3-3v: Plot of stratigraphic height versus Th - Spy Sill 35 Figure 3-3w: Plot of stratigraphic height versus U - Spy Sill 35 Figure 3-4a: Plot of stratigraphic height versus Ni - Lewis Intrusion 38 Figure 3-4b: Plot of stratigraphic height versus MgO - Lewis Intrusion 38 Figure 3-4c: Plot of stratigraphic height versus Mg# - Lewis Intrusion 38 vu Page Figure 3-4d: Plot of stratigraphic height versus Ni/MgO - Lewis Intrusion 38 Figure 3-4e: Plot of stratigraphic height versus T i 0 2 - Lewis Intrusion 39 Figure 3-4f: Plot of stratigraphic height versus Cr - Lewis Intrusion 39 Figure 3-4g: Plot of stratigraphic height versus Cu - Lewis Intrusion 39 Figure 3-4h: Plot of stratigraphic height versus Co - Lewis Intrusion 39 Figure 3-4i: Plot of stratigraphic height versus Au - Lewis Intrusion 40 Figure 3-4j: Plot of stratigraphic height versus Pd - Lewis Intrusion 40 Figure 3-4k: Plot of stratigraphic height versus Pt - Lewis Intrusion 40 Figure 3-41: Plot of stratigraphic height versus Pt/Pd - Lewis Intrusion 40 Figure 3-4m: Plot of stratigraphic height versus S - Lewis Intrusion 42 Figure 3-4n: Plot of stratigraphic height versus Cu/Zr - Lewis Intrusion 42 Figure 3-4o: Plot of stratigraphic height versus Ni/Cu - Lewis Intrusion 42 Figure 3-4p: Plot of stratigraphic height versus Ba - Lewis Intrusion 42 Figure 3-5a: Plot of T i 0 2 versus Zr - Spy and Lewis Intrusions 43 Figure 3-5b: Plot of T i 0 2 versus A l 2 0 3 - Spy and Lewis Intrusions 43 Figure 3-5c: Plot of P 2 0 5 versus A l 2 0 3 - Spy and Lewis Intrusions 44 Figure 3-5d: Plot of Zr versus AI203 - Spy and Lewis Intrusions 44 Figure 3-5e: Plot of Zr versus P205 - Spy and Lewis Intrusions 44 Figure 3-5f: Plot of Ce versus Yb - Spy and Lewis Intrusions 45 Figure 3-5g: Plot of Zr versus Yb - Spy and Lewis Intrusions 45 Figure 3-5h: Plot of Zr versus Y - Spy and Lewis Intrusions 45 Figure 3-5i: Plot of S versus Ni - Spy and Lewis Intrusions 47 Figure 3-5j: Plot of S versus Ni - Spy and Lewis Intrusions 47 Figure 3-5k: Plot of S versus Cu - Spy and Lewis Intrusions 48 Figure 3-51: Plot of S versus Cu - Spy and Lewis Intrusions 48 Figure 3-5m: Plot of S versus Pt - Spy and Lewis Intrusions 49 Figure 3-5n: Plot of S versus Pt - Spy and Lewis Intrusions 49 Figure 3-5o: Plot of S versus Pd - Spy and Lewis Intrusions 50 Figure 3-5p: Plot of S versus Pd - Spy and Lewis Intrusions 50 Figure 3-5g: Plot of Pt versus Ni - Spy and Lewis Intrusions 51 Figure 3-5r: Plot of Pt versus Cu - Spy and Lewis Intrusions 51 Figure 3-5s: Plot of Pd versus Cu - Spy and Lewis Intrusions 52 Figure 3-5t: Plot of Pd versus Ni - Spy and Lewis Intrusions 52 Figure 3-5u: Plot of Ni versus MgO - Spy and Lewis Intrusions 54 Figure 3-5v: Plot of Ni versus Mg# - Spy and Lewis Intrusions 54 Figure 3-5x: Plot of Ni versus MgO - Spy and Lewis Intrusions 55 Figure 3-5y: Plot of Ni versus Mg# - Spy and Lewis Intrusions 55 Figure 3-5z: Plot of S versus Se - Spy and Lewis Intrusions 56 Figure 3-5aa: Plot of 834S versus S/Se - Spy Sill 56 Figures 3-5abi-vi: REE Spidergrams for units from the Spy Sill 57 Figure 3-6a: Plot of average MgO-rim versus average MgO-core 59 from olivine microprobe data - Spy and Lewis Intrusions viii Page Figure 3-6b: Plot of average FeO-rim versus average FeO-core 59 from olivine microprobe data - Spy and Lewis Intrusions Figure 3-6c: Plot of average NiO-rim versus average NiO-core 60 from olivine microprobe data - Spy and Lewis Intrusions Figure 3-6d: Plot of Ni versus Fo# from olivine microprobe data - 60 Spy and Lewis Intrusions Figure 4-1 a: Plot of stratigraphic height versus Cr - Nikolai basalt 65 Figure 4-1 b: Plot of stratigraphic height versus Cu - Nikolai basalt 65 Figure 4-1 c: Plot of stratigraphic height versus MgO - Nikolai basalt 65 Figure 4-1 d: Plot of stratigraphic height versus Ni - Nikolai basalt 65 Figure 4-1 e: Plot of stratigraphic height versus Cu/Zr - Nikolai basalt 66 Figure 4-1 f: Plot of stratigraphic height versus Mg# - Nikolai basalt 66 Figure 4-1 g: Plot of stratigraphic height versus Ni/Cu - Nikolai basalt 66 Figure 4-1 h: Plot of stratigraphic height versus Ni/MgO - Nikolai basalt 66 Figure 4-1 i: Plot of stratigraphic height versus Ce/Yb - Nikolai basalt 67 Figure 4-1 j : Plot of stratigraphic height versus Co - Nikolai basalt 67 Figure 4-1 k: Plot of stratigraphic height versus T i02 - Nikolai basalt 67 Figure 4-11: Plot of stratigraphic height versus Ba - Nikolai basalt 69 Figure 4-1 m: Plot of stratigraphic height versus Ba/Zr - Nikolai basalt 69 Figure 4-1 n: Plot of stratigraphic height versus Th - Nikolai basalt 69 Figure 4-1 o: Plot of stratigraphic height versus Th/Ta - Nikolai basalt 69 Figure 4-2a: Plot of Zr versus P205 - Nikolai basalt 71 Figure 4-2b: Plot of Zr versus Y - Nikolai basalt 71 Figure 4-2c: Plot of Zr versus Yb - Nikolai basalt 71 Figure 4-2d: Plot of T i02 versus AI203 - Nikolai basalt 72 Figure 4-2e: Plot of T i02 versus Zr - Nikolai basalt 72 Figure 4-2f: Plot of Zr versus AI203 - Nikolai basalt 72 Figure 4-2g: Plot of Ce versus Yb - Nikolai basalt 73 Figure 4-2h: Plot of T i02 versus P205 - Nikolai basalt 73 Figure 4-2i: Plot of Cu versus MgO - Nikolai basalt 74 Figure 4-2j: Plot of Ni versus Mg# - Nikolai basalt 74 Figure 4-2k: Plot of Ni versus MgO - Nikolai basalt 74 Figure 4-2I: Plot of Cu versus Mg# - Nikolai basalt 75 Figure 4-2m: Plot of Co versus MgO - Nikolai basalt 75 Figure 4-2n: Plot of Co versus Mg# - Nikolai basalt 75 Figure 4-2o: Plot of Ni versus MgO - Nikolai and Karmutsen basalt 77 Figure 4-2p: Plot of Ni versus Mg# - Nikolai and Karmutsen basalt 77 Figure 4-2q: Plot of Ni versus LOI - Nikolai and Karmutsen basalt 77 Figure 4-2r: Plot of Cu versus MgO - Nikolai and Karmutsen basalt 78 Figure 4-2s: Plot of Cu versus Mg# - Nikolai and Karmutsen basalt 78 Figure 4-2t: Plot of Cu versus LOI - Nikolai and Karmutsen basalt 78 Figure 4-2u: Plot of Co versus LOI - Nikolai and Karmutsen basalt 79 Figure 4-2v: Plot of Co versus Mg# - Nikolai and Karmutsen basalt 79 ix Page Figure 4-2w: Plot of Co versus MgO - Nikolai and Karmutsen basalt 79 Figure 4-2x: Plot of Cu/Zr versus LOI - Nikolai and Karmutsen basalt 80 Figure 4-2y: Plot of Cu/Zr versus Mg# - Nikolai and Karmutsen basalt 80 Figure 4-2z: Plot of Ni/MgO versus LOI - Nikolai and Karmutsen basalt 80 Figure 4-3a: REE spidergram - Nikolai basalt, Halfbreed Creek section 82 Figure 4-3b: REE spidergram - Nikolai basalt, Burwash Creek section 82 Figure 4-4: Extended Element Profile - Nikolai basalt 83 Figure 4-5: A F M Plot - Nikolai basalt 84 Figure 4-6: Jensen Cation Plot - Nikolai basalt 84 Figure 5-1: Plot of T i02 versus Zr - Kluane intrusions and Nikolai basalt 88 Figure 5-2a: Plot of Ce/Yb versus MgO - Kluane intrusions and Nikolai 88 basalt Figure 5-2b: Plot of Th versus Ce/Yb - Kluane intrusions and Nikolai basalt 89 Figure 5-2c: Plot of Ba/Zr versus Ce/Yb - Kluane intrusions and Nikolai 89 basalt Figure 5-3: 0.5Fe+0.5Mg/Ti versus Si/Ti PER Plot - Kluane intrusions and 91 Nikolai basalt Figure 5-4: 0.5AI+0.5Mg+1.5Ca+2.75Na/Ti versus Si/Ti PER Plot - 92 Kluane intrusions and Nikolai basalt Figure 5-5a: Plot of Cu versus MgO - Kluane intrusions and Nikolai basalt 94 Figure 5-5b: Plot of Co versus MgO - Kluane intrusions and Nikolai basalt 94 Figure 5-5c: Plot of Ni versus MgO - Kluane intrusions and Nikolai basalt 94 Figure 5-5d: Plot of Co versus Mg# - Kluane intrusions and Nikolai basalt 95 Figure 5-5e: Plot of Cu versus Mg# - Kluane intrusions and Nikolai basalt 95 Figure 5-5f: Plot of Ni versus Mg# - Kluane intrusions and Nikolai basalt 95 Figure 5-5g: Plot of Cu /Zr versus Mg# - Kluane intrusions and Nikolai 96 basalt Figure 5-5h: Plot of Ni/MgO versus Mg# - Kluane intrusions and Nikolai 96 basalt Figure 6-1: Location of Ni-Cu-PGE showings and barren 98 sulphide/oxide layers Figure 6-4a: Plot of Ni versus S - Klu property 107 Figure 6-4b: Plot of Cu versus S - Klu property 107 Figure 6-4c: Plot of Ni versus MgO - Klu property 107 Figure 6-4d: Plot of Pt versus S - Klu property 108 Figure 6-4e: Plot of Pd versus S - Klu property 108 Figure 6-4f: Plot of Co versus S - Klu property 108 Figures 6-4g and 6-4h: Plots of metals in 100% sulphide/chondrite - 110 Archean and Proterozoic komatiitic deposits and deposits in large igneous complexes X Page Figure 6-41: Plot of metals in 100% sulphide/chondrite - flood basalt 111 related deposits Figure 6-4j: Plot of metals in 100% sulphide/chondrite - Klu property 111 Figure 6-4k: Plot of metals in 100% sulphide/chondrite - Kluane Belt 112 Figure 6-4I: Plot of metals in 100% sulphide/chondrite - Klu property: 112 sediment hosted massive sulphide Figure 6-4ml: Plot of metals in 100% sulphide/chondrite - Klu property: 113 sediment hosted fracture mineralization Figure 6-4n: Plot of metals in 100% sulphide/chondrite - Klu property: 113 gabbro hosted semi-massive sulphide Figure 6-4o: Plot of metals in 100% sulphide/chondrite - Klu property: 114 gabbro hosted disseminated sulphide Figure 6-4p: Plot of metals in 100% sulphide/chondrite - Klu property: 114 massive sulphide at the gabbro/siltstone contact Figure 6-4q: Plot of Pd/lr versus Ni/Cu - various Ni-Cu-PGE deposits 117 Figure 6-4r: Plot of Pd/lr versus Ni/Cu - Klu property 117 Figure 6-4s: Plot of Ni/Pd versus Cu/lr - various Ni-Cu-PGE deposits 118 Figure 6-4t: Plot of Ni/Pd versus Cu/lr - Klu property 118 Figure 6-4u: Box and Whisker Plot of 534S sorted by mineralization type - 120 Klu property Figure 6-4v: Box and Whisker Plot of S/Se sorted by mineralization type - 120 Klu property Figure 6-4w: Plot of S versus 834S - Klu property 122 Figure 6-4x: Plot of Se versus S - Klu property 122 Figure 6-4y: Plot of Se versus S - Klu property (decreased scale) 122 Figure 6-4z: Plot of Co versus 534S - Klu property 123 Figure 6-4aa Plot of Ni versus 534S - Klu property 123 Figure 6-4ab: Plot of Pt versus 534S - Klu property 123 Figure 6-4ac: Plot of Zn versus 534S - Klu property 124 Figure 6-4ad: Plot of Cu versus 534S - Klu property 124 Figure 6-4ae: Plot of As versus 534S - Klu property 125 Figure 6-4af: Plot of Sb versus 534S - Klu property 125 Figure 6-4ag: Plot of S/Se versus 534S - Klu property 125 Figure 6-4ah: S/Se - 634S mixing model (Lower Claimpost showing - 127 mantle) Figure 6-4ai: S/Se - 634S mixing model (Claimpost showing - mantle) 127 Figure 7-1: Schematic section for sill emplacement and Ni-Cu-PGE 132 mineralization xi List of Plates Page Plate 3-1: Photomicrograph of peridotite from the Spy Sill 22 Plate 3-2: Photomicrograph of peridotite from the Spy Sill 22 Plate 3-3: Photomicrograph of peridotite from the Spy Sill 23 Plate 3-4: Photomicrograph of peridotite from the Spy Sill 23 Plate 3-5: Photomicrograph of peridotite from the Spy Sill 24 Plate 3-6: Photomicrograph of marginal gabbro from the Spy Sill 24 Plate 3-7: Photomicrograph of marginal gabbro from the Spy Sill 26 Plate 3-8: Photomicrograph of Maple Creek gabbro adjacent to the Spy Sill 26 Plate 6-1: Photomicrograph of massive sulphide in footwall sediment 100 Plate 6-2: Photomicrograph of massive sulphide in footwall sediment 100 Plate 6-3: Photomicrograph of massive sulphide in footwall sediment 101 Plate 6-4: Photomicrograph of massive sulphide in footwall sediment 101 Plate 6-5: Photomicrograph of massive sulphide occurring along marginal 102 gabbro/siltstone contact Plate 6-6: Photomicrograph of massive sulphide occurring along marginal 102 gabbro/siltstone contact Plate 6-7: Photomicrograph of massive sulphide occurring along marginal 103 gabbro/siltstone contact Plate 6-8: Photomicrograph of net-textured sulphide in marginal gabbro 103 Plate 6-9: Photomicrograph of massive sulphide from the 105 Claimpost showing Plate 6-10: Photomicrograph of massive sulphide from the 105 Claimpost showing xii List of Appendices Page Appendix 1: Analytical Results from the Spy Sill and Lewis Intrusion 142 Appendix 2: Olivine Microprobe Data from the Spy Sill and Lewis 148 Intrusion peridotite samples Appendix 3: Analytical Results from the Nikolai basalt samples 153 Appendix 4: Analytical Results from Ni-Cu-PGE showings and barren 157 sulphide/oxide layers X l l l Acknowledgements The field work and geochemical analyses involved in this study were entirely supported by Inco Limited. Phil Rush and Richard Alcock (currently and formerly of Inco respectively) supported and encouraged the initiation of this thesis. Numerous other current and former Inco employees provided suggestions and assistance including: Walter Peredery, Peter Lightfoot, Catherine Farrow, Paul Golightly, Ed Pattison and Herb Mackowiak. Dr. Larry Hulbert of the Geological Survey of Canada provided valuable direction and assistance in the field and with interpretation of the thesis data. I would like to thank Dr. Al Sinclair, my thesis supervisor, for providing encouragement and direction. Dr. Sinclair was very patient with the length of time taken to complete this thesis, which stretched into his retirement. 1 Section 1: Introduction 1.1 Purpose of Thesis The Klu property was staked in the fall of 1994 after Ni-Cu-PGE rich sulphide showings were located during a regional exploration program in the Kluane Belt of Southwest Yukon. These showings are related to the Spy Sill, a 100 m thick ultramafic sill of Triassic age which is believed to be co-magmatic with the Triassic Nikolai basalt. A number of reasonably well studied Ni-Cu-PGE deposits and showings (including the Wellgreen and Canalask deposits) also occur in the Kluane Belt. The host intrusions to these deposits and showings suffer variable degrees tectonism, serpentinization and overburden cover which can obscure the characteristics of the intrusions and associated Ni-Cu-PGE mineralization. The Spy Sill on the Klu property provides an opportunity to study one of the least tectonized and serpentinized intrusions in the Kluane Belt. In addition, an entire section through the Spy Sill is well exposed from top to base. Another advantage of studying the Kluane-type sills on the Klu property is the presence of an ultramafic sill which appears to be depleted in Ni, Cu, Co, Pt and Pd relative to the Spy Sill. This sill, the Lewis Intrusion, was recognized as being particularly metal-poor during lithogeochemical sampling which was part of the 1995 exploration program on the property. The presence of the Ni-Cu-PGE barren Lewis Intrusion provided an excellent comparison to the Spy Sill. A further reason for studying the Klu property is the occurrence of Ni-Cu-PGE barren sulphide layers close to the stratigraphic level of the Spy Sill. The study of this material made it possible to examine the probability and effect of the assimilation of country rock sulphur by the Triassic intrusions. A significant component of this study is to compare the Nikolai basalt with the Kluane-type intrusions on the Klu property and to determine if evidence of the Ni-Cu-PGE mineralization exists within the volcanic pile. Two measured sections of Nikolai basalt outside the property were sampled for this purpose. The Ni-Cu-PGE rich sulphide mineralization associated with the Spy Sill is described and put in the context of Ni-Cu-PGE mineralization elsewhere in the Kluane Belt and of worldwide examples of various types of Ni deposits. 1.2 Scope of Thesis Much of the fieldwork involved in this study was carried as part of regional and property-scale exploration in the Kluane Belt and on the Klu property. The geology on the Klu property was mapped by the author and Scott Casselman (formerly of Inco Exploration) in the summer of 1995. The sampling of Ni-Cu-2 P G E mineralization, represented by a suite of 41 samples was also done in conjunction with the mapping. These samples were analyzed for an extensive suite of elements including: Ni, Cu, Co, Pt, Pd, Au, S, Se and 8 3 4S. Polished thin sections were examined from a number of these samples for petrographic characterization. Samples were collected from the measured section of the Spy Sill by the author and Larry Hulbert (of the Geological Survey of Canada) in the summer of 1995. Twenty five samples were collected from the Spy Sill and were analyzed for major elements, trace elements, REE, Pt, Pd, Au, As, S, Se and 534S. Seventeen samples were collected from a measured section of the Lewis Intrusion by the author in the summer of 1996. These samples were analyzed for major elements, trace elements, Ni, Cu, Co, Pt, Pd and Au. In addition to geochemical analysis, thin sections were examined from a number of samples from the these measured sections for petrographic characterization. Olivine compositions from a suite of peridotite samples from these intrusions were determined by electron microprobe analysis. Sampling of two measured sections of the Nikolai basalt was completed by the author and Walter Peredery (formerly of Inco Exploration) in the summer of 1993. Twenty six samples collected from the two sections were analyzed for an large suite of elements including: majors, XRF traces, Ni, Cu, Co, REE, Th and Ta. As a majorpart of this study, geological mapping, geochemistry and petrographic observations are used to characterize the Kluane-type intrusions on the Klu property which are associated with the Ni-Cu-PGE mineralization. The geochemistry of the Nikolai basalt is used to explore the genetic relationship between the basalt and the Kluane-type intrusions. Ni-Cu-PGE mineralization as well as barren synsedimentary sulphide is described in terms of its geochemistry, petrography, spatial distribution and host rock. The above information is evaluated and synthesized into a proposed model for the emplacement of the Ni-Cu-PGE mineralization and the associated Kluane-type intrusions. 3 Section 2: Geological Setting 2.1 Location, Access and Topography The Klu property is centred 8 km south of the village of Destruction Bay and 20 km southeast of the village of Burwash Landing, Yukon Territory. Whitehorse is approximately 200 km east of the property. The property parallels the Alaska Highway with its northeast boundary being 5 km southwest from the highway. Kluane National Park borders the property to the southeast and southwest. The property is completely within the Kluane Game Sanctuary (Figure 2-1). Primary access to the Klu property is by helicopter. A permanent helicopter base is in Haines Junction, 70 km to the southeast. Secondary gravel roads extend from the Alaska Highway along the southeast banks of Nines Creek and Bock's Brook; both roads end approximately two km short of the northeast boundary of the property. Topography on the property is extremely rugged. Elevations range from 1,050 to 2,460 m. Treeline is generally at about 1,260 m with treed areas only occurring in the Congdon, Nines, Bock's and Lewis Creek valleys. Several glaciers occur on the property above the 2,100 m level. 2.2 Exploration History The earliest recorded exploration on the present Klu property was carried out by John S. Vincent Ltd. for the Nickel Syndicate during 1972-73 (MacEy et al, 1973). This work consisted of geological mapping and rock sampling on the Spy 1-12 claims. The claims were located on the northeast facing slope above the southern branch of Nines Creek. John S. Vincent Ltd. mapped a series of gabbro to peridotite sills that intrude the Hasen Creek Formation. Values up to 1.47% Ni and 0.49% Cu were reported for sulphides occurring in "quartz xenoliths" at the base of the sill. Additional sulphide mineralization consisting of disseminated chalcopyrite, pyrrhotite and pyrite with values of 0 .5% Ni and 0.5% Cu was reported to occur in gabbro along strike from the previous showing. The above mineralization is in the same area as the Ni-Cu-PGE rich sulphide showings discovered by Inco personnel in 1994. Elsewhere on the Spy claims, sphalerite-galena mineralization was reported on the margins of a quartz vein. This showing returned values of 1.25% Zn and 0.25% Pb. Aurum Geological Consultants Inc. explored part of the property for Walhala Explorations Ltd. in 1987. Assessment was filed on the Tony 1-28 and Tony 29-60 claim blocks. This work consisted of geological mapping and 5 lithogeochemical sampling (Keyser, 1987). Also in 1987, the Duke 1-44 claims were staked over part of the current property by the Kluane Joint Venture (Chevron Minerals-All North Resources); work consisted of prospecting and soil geochemistry. 2.3 Previous Studies 2.3.1 Regional Mapping The first comprehensive regional mapping in the belt was done by Muller (1967) of the Geological Survey of Canada (GSC). His report contains descriptions of the various stratigraphic formations, structural geology, glaciation and economic geology in the belt; included with the report is a geology map of NTS 115G and 115F East Vz. A further refinement of Muller's work was completed by Read and Monger (1976) Dodds and Campbell (1992) compiled previous published regional mapping and combined this with some of their unpublished work in the belt to produce G S C Open File 2188 in 1992 (115G and 115F East 1/2). 2.3.2 Metallogenic Studies One of the earliest economically oriented studies is that of F.A. Campbell (1960) who compared the structural and petrological relationships of host intrusions and Ni-Cu-PGE mineralization in the White River and Quill Creek Complexes. S.W. Campbell (1981) studied the genesis of copper deposits and occurrences associated with late Paleozoic and Triassic strata in the belt. Her study presents field relationships, S isotopic data and petrological data. Miller (1991) studied the Wellgreen Deposit in detail and focused on PGE geochemical variations within the various zones of the deposit. The most detailed metallogenic study of the belt is that of Hulbert (1995). Hulbert's research documents ultramafic intrusions and associated Ni-Cu-PGE mineralization throughout the Kluane Ultramafic Belt (KUB). A large amount of original petrographic, geochemical and isotopic data is presented and discussed in the context of regional geology and tectonics. Data and conclusions from these previous studies will be compared with the results of this study in the appropriate sections which follow. 2.4 Regional Geology The Klu property is situated within Wrangellia, an accreted terrane extending 2,340 km from Alaska to southern B.C.. Certain geological elements are common throughout the terrane including an Upper Paleozoic island arc basement overlain by a thick Triassic flood basalt sequence. The eastern part of Wrangellia (in Southwest Yukon) is bounded to the northeast by the Denali Fault System and to the southwest by the Duke River Fault. Oldest 6 Wrangellian rocks in the belt are the Pennsylvanian to Lower Permian Skolai Group. The Station Creek Formation occurs at the base of the Skolai Group and consists of tuffs, pyritic black tuff, mafic volcanlcs and argillite. This is overlain by the Hasen Creek Formation which consists of tuffs, mafic volcanics, argillite and limestone. The Skolai Group is stratigraphically overlain by Pennsylvanian to Triassic mafic meta-volcanic rocks, Upper Triassic Nikolai basalt, and Upper Triassic McCarthy Formation of limestone and phyllite. Tertiary volcanics and sediments unconformably overlie the Triassic sequence. Quaternary surficial deposits locally cover the Paleozoic, Mesozoic and Cenozoic strata. A stratigraphic section showing the formations and their approximate thicknesses is shown in Figure 2-2. The Nikolai basalt is an oceanic flood basalt believed to be related to a mantle plume (Lassiter et al, 1995). The basalt attains a thickness of up to 1,000 m in southwestern Yukon. The base of the Nikolai basalt was deposited subaqueously whereas most of the formation was deposited under subaerial conditions. The distribution of Nikolai basalt exposures in Southwest Yukon is shown in Figure 2-3. There are two major suites of intrusive rocks in the belt: the oldest is the mafic to ultramafic Triassic suite which includes ultramafic sills, marginal gabbro (i.e. marginal to ultramafic sills) and the Maple Creek Gabbro. The Triassic suite is thought to be oogenetic with the Nikolai basalt. Mortensen and Hulbert (1991) obtained a U-Pb date from zircon of 231+/-1 Ma from a Maple Creek gabbro dyke. Cretaceous Kluane Range Intrusions are dioritic to granodioritic in composition and occur throughout northern Wrangellia. Minor Tertiary sills, dykes -and stocks of felsic to intermediate composition are also present The major Triassic ultramafic intrusions (Kluane-type) are sill-like bodies which intrude the Hasen Creek and Station Creek Formations. The dips of the sills range from vertical to steeply overturned to as shallow as 30 degrees. Maximum dimensions of the sills are estimated to be up to 18 km in length and 600 m in thickness. These intrusions occur throughout the extent of Wrangellia in Southwest Yukon; the distribution of major Kluane-type ultramafic intrusions is shown in Figure 2-3. This belt of rocks is informally known as the Kluane Ultramafic Belt (KUB). Peridotite is the dominant ultramafic phase with lesser dunite and pyroxenite. The peridotite ranges in composition from wehrlitic to Iherzolitic and contains varying amounts of olivine, clinopyroxene, orthopyroxene, plagioclase, phlogopite and oxides. The degree of serpentinization varies locally from minor to total. Many of the ultramafic sills have a marginal gabbro at their base which makes up approximately 4-8% of the thickness of each sill. Clinopyroxenite layers locally are present above the marginal gabbro layer. Some of the sills also have marginal 7 Stratigraphic Section Figure 2-2 CO CO 3 .P . Is .to. c ."3 c c c a. Nizina & Chitistone Lms O F V V V v rrx o o o 0 o Basalt pillows, breccia and flows, amygdaloidal basalt and minor tuffs Gabbro, olivine gabbro (Maple Creek Gabbro) Ultramafic-mafic cumulates Limestones Conglomerate; +/- pebbly greywacke Argillie, shale, siltstone, cherty, argillite Argillie, shale, siltstone, cherty, argillite I I I 1000 800 600 400 200 0 m From Hulbert(1994) 8 9 gabbro at their upper contacts. Field relationships suggest that these marginal gabbros represent an initial pulse of magma which was followed by progressively more ultramafic magma. Permo-Triassic rocks of the belt are faulted and tightly folded about steeply dipping axial planes and shallow northwest trending axes. Faulting includes bedding-plane slip faults and strike-slip faults which trend normal to the Denali Fault (a terrane-bounding transcurrent fault). Ni-Cu-PGE mineralization is associated with Kluane-type intrusions in numerous locations throughout the belt. The locations of Ni-Cu-PGE deposits and showings (including the Spy showing) in the KUB are shown in Figure 2-3. The most significant deposit in the KUB occurs at the Wellgreen Mine which operated during parts of 1972 and 1973. At the time production commenced, Wellgreen had "reserves" of 669,150 tonnes grading 2.04% Ni, 1.42% Cu, 0.073% Co, 1.3 g/t Pt and 0.9 g/t Pd (Came, 1992). This deposit consists of massive sulphide mineralization at the base of a 600 m thick peridotite sill with gabbroic margins. Exploration in the late 1980s outlined a low-grade disseminated sulphide deposit hosted in basal marginal gabbro and overlying peridotite. This deposit has "probable reserves" of 42,326,030 tonnes grading 0.36% Ni, 0.35% Cu, 0.5 g/t Pt and 0.3 g/t Pd (Carne, 1992). The other significant deposit in the KUB is the Canalask deposit. This deposit has a "geological reserve estimate" of 1,800,000 tonnes grading 1.1% Ni (Came, 1992) contained in a septum of country rock within a 500 m thick peridotite sill. Detailed descriptions of the Wellgreen and Canalask deposits are given by Campbell (1981), Hulbert (1995) and Carne (1992). 2.5 Klu property Geology and Mineralization Associated with the Spy Sill 2.5.1 Property Geology The geology of the property is dominated by several fault-bounded slices of folded Paleozoic and Mesozoic strata (Figure 2-4). These rocks are overlain by gently-dipping Tertiary sedimentary and volcanic rocks. The bounding faults trend southeasterly and are believed to dip steeply. The axial planes of the folds also trend from the southeast to the northwest and appear to dip steeply; fold axes are assumed to be nearly horizontal. Much of the folding is inferred; no large scale folds were observed on the property. The lack of outcrop in the valley bottoms of Congdon Creek, Nines Creek, Bock's Brook and Lewis Creek make some structural interpretations tenuous. 10 11 A Table of Formations present on the Klu property is shown in Table 2-1. Geological age, map symbol, unit name and a brief description of each unit are listed in the table. 12 TABLE OF FORMATIONS (Table 2-1) STRATIFIED ROCKS AGE SYMBOL UNIT DESCRIPTION Tertiary (Miocene-Pliocene) Nw Wrangell Lava (Undivided) Basalt to andesite flows, minor white to yellow felsic pyroclastics and flows Tertiary (Oligocene) Os Amphitheatre Formation Yellow-buff to gray-buff sandstone, pebbly sandstone, polymictic conglomerate U. Triassic-Cretaceous uTrKp Dark gray phyllite, minor greywacke and Conglomerate U. Triassic uTrM McCarthy Formation Argillaceous limestone and dark gray argillite U. Triassic uTrc Chitistone and Nizina Formations Massive light gray limestone, limestone breccia, and dark gray well bedded limestone U. Triassic uTre White to creamy-white gypsum and anhydrite U. Triassic uTrN Nikolai Greenstone Dark green and maroon amygdaloidal to massive basaltic and andesitic flows, locally interbedded with tuff, breccia, shale, limestone; pillow lava and conglomerate occur at base L. Permian Ps Hasen Creek Formation Thin bedded siliceous argillite, siltstone, shale, greywacke, conglomerate, local thin basalt flows L. Permian Pc Hasen Creek Formation Buff to gray bioclastic limestone Pennsylvanian Pv Station Creek Formation Andesitic to basaltic tuffs and flows INTRUSIVE ROCKS AGE SYMBOL UNIT DESCRIPTION Tertiary (Miocene) IMf Wrangell Plutonic Suite Buff to creamy-white granodiorite, diorite, gabbro dykes and sills, fine grained Tertiary (Miocene) IMdi Bock's Brook Stock Light buff-gray biotite diorite, medium grained Triassic Trb Maple Creek Gabbro (Kluane-Type) Gabbro and anorthositic gabbro sills, medium Grained Triassic Trub Kluane-Type Ultramafics Peridotite, feldspathic peridotite sills with minor pyroxenite and dunite, medium grained Triassic Trmg Kluane-Type Marginal Gabbro Gray medium grained to fine grained locally chilled gabbro, forms along margins of peridotite 13 The Triassic mafic and ultramafic magma intruded the Hasen Creek Formation (and, to a lesser extent, the Station Creek Formation) as sills. Virtually all exposures of peridotite in Hasen Creek strata appear to occur in the same position relative to the other rocks, which raises the possibility that there is only one main sill in fault-repeated sections of the stratigraphy. If so, then the original sill would have covered an area of 25 km by 10 km (250 km 2). The geology of the Klu property is summarized below with reference to the major ultramafic intrusions whose locations are shown on Figure 2-4. Spy Area The Spy Sill in the southeastern part of the property is a 75 to 100 m thick, 8 km long intrusion of dominantly unserpentinized feldspathic peridotite emplaced within Hasen Creek siltstone. Marginal gabbro up to 10 m thick is locally present at the top and base of the sill. The Spy Sill has an Ar-Ar date from intercumulus phlogopite of 232 + /-1 Ma (personal communication from Hulbert, 1995). Maple Creek gabbro sills occur stratigraphically above and below the Spy Sill as well as directly at its base. The most continuous Maple Creek gabbro sill occurs 230 m below the base of the peridotite, is up to 1 60 m thick, and is intermittently exposed over a 10 km strike. The Spy Sill is cut locally by Maple Creek gabbro sills up to 200 m thick. Maple Creek gabbro also forms lens-shaped bodies within the peridotite. Three stacked lenses of the peridotite (up to 600 m long by 30 m thick) occur below the main sill on the ridge between Congdon Creek and the southern branch of Nines Creek. Smaller lenses of peridotite occur below the main sill elsewhere. The lower contact of the main peridotite sill occurs 100 m above a distinctive chert/siltstone pebble conglomerate bed, while the upper contact is approximately 20 m below a buff coloured limestone bed with positive weathering relief. Aeromagnetic data (MacGowan, 1996) suggest that the Spy Sill connects to the Bock's Brook Intrusion. Down-section from the Spy Sill on the opposite side of the southern branch of Nines Creek, two apparently unconnected feldspathic peridotite intrusions up to 65 m thick occur within the Station Creek Formation. The Hasen Creek and Station Creek Formations have a constant southeasterly strike and dip at an average of 40° to the southwest. Nikolai basalt caps the ridge above the Spy Sill. The lower contact of the basalt is approximately 450 m above the top of the peridotite. The contact between the Hasen Creek Formation and the Nikolai basalt appears to be disconformable. 14 This sill is one of the targets of this study as is among the best preserved ultramafic sills in the KUB, both in terms of having preserved contacts and lack of serpentinization. Bock's Brook Area The Bock's Brook Intrusion, located in the central part of the property between the Spy Sill and the Lewis Intrusions, is the thickest on the property (500 m maximum, but may be exaggerated due to fault-repetition). At least one smaller sill occurs below the main sill. The peridotite is strongly serpentinized and appears to be fault-bounded along part of its northern (basal) contact. Marginal gabbro along the base of the intrusion may have been removed by this fault near surface. If the fault has a near-vertical orientation, marginal gabbro may be preserved down-dip. The top of the sill and part of the base are in contact with Hasen Creek limestone, siltstone and conglomerate. These beds have variable orientations, but generally strike easterly and dip approximately 50° to the south. Much of the intrusion underlies a boulder-choked cirque at the head of a tributary of Bock's Brook. Downstream from the main sill, along the tributary, are complicated fault slices of Hasen Creek Formation, Nikolai basalt, Upper Triassic shale, peridotite and Upper Triassic gypsum and limestone. Aeromagnetic data form the 1996 survey shows the Bock's Brook Intrusion to have double the strike length suggested by mapping (MacGowan, 1996). The southeastern part of the intrusion is covered by Hasen Creek Formation strata and overburden. The aeromagnetic data also suggests that the Spy Sill connects with the southeastern part of the Bock's Brook Intrusion. The Nikolai basalt to the north of the Bock's Brook Intrusion contains numerous small ( 2 x 3 x 1 m) rafts of gypsum. At the eastern end of the Bock's Brook Intrusion, the peridotite is in contact with a gabbro sill, which may be a continuation of the thick Maple Creek sill which occurs below the Spy Sill. Right-On Mountain Area The Right-On Mountain Intrusions occur at the southern extremity of the property in Hasen Creek shale, chert and limestone which strikes southeast-northwest and dips to the southwest at 60-90° (possibly overturned). There are two major peridotite sills, the largest of which is 200 m thick (due, in part, to fault repetition?). The large sill ends a short distance (approximately 1 50 m) into the property whereas the narrower peridotite sill (60 m thick) strikes into Kluane Park to the northwest where it is locally obscured by Tertiary cover. The southwestern contact of the narrower peridotite sill is locally part of the Duke River Fault which forms the boundary between 15 Wrangellia and Alexander Terrane. A Maple Creek gabbro sill up to 250 m thick occurs to the northeast of the peridotite sills. Duke - Halfbreed Area Portions of the Duke and Halfbreed Intrusions occur in the northwestern part of the property. Both intrusions trend northwest into the Native Land Claim Staking Withdrawal. These sills are emplaced within Hasen Creek Formation siltstone and conglomerate which are part of a synclinal structure cored by overlying Nikolai basalt. The Duke Intrusion is a mafic/ultramafic sill emplaced within Hasen Creek Formation which strikes southeast-northwest and dip at approximately 50° to the southwest. The intrusion strikes below Tertiary cover to the southeast. Its base consists of a partially to totally serpentinized, 200 m thick lower peridotite section. This is overlain by a 30 to 50 m thick screen of Hasen Creek siltstone and chert (or silicified siltstone?). Above these sediments is a 350 m thick section of gabbro, believed to be of the Maple Creek-type and contains multiple phases. Marginal gabbro is not exposed at the base of the peridotite in the Duke Intrusion. The portion of the Halfbreed Intrusion on the Klu property consists of two peridotite sills, each up to 125 m thick. They intrude siltstone, conglomerate, and limestone beds of the Hasen Creek Formation which strike southeasterly and dip to the southwest at 50 to 60°. Thin discontinuous marginal gabbro lenses occur at the base of these sills. Maple Creek gabbro sills occur both above and below the peridotite sills. The peridotite and gabbro sills here are poorly exposed due to extensive glacial moraine, beneath which the sills appear to terminate. Lewis Area The Lewis Intrusions occur in the north-central part of the property, 2 km southeast of the Halfbreed Intrusion. The Lewis Intrusions appear to consist of three relatively unserpentinized peridotite to pyroxenite sills of complex morphology. They intrude Hasen Creek siltstone, chert and argillite, which generally strike southwest-northeast and dip between 20 and 85° to the southeast. An irregular body of Maple Creek gabbro with local cumulate layering appears to intrude the western-most sill. A thin marginal gabbro zone occurs at the base of the two eastern-most sills. The thickest sill is approximately 300 m thick. The sills are overlain by Tertiary volcanics to the southeast. A circular stock of Tertiary diorite approximately 1.2 km in diameter lies south of the Lewis Intrusions. 16 The Lewis Intrusion was selected for inclusion in this study due to its relatively unserpentinized character and the absence of local tectonic effects. Several exposures of Upper Triassic gypsum and limestone beds occur directly above the Nikolai basalt between the Bock's Brook-Lewis trend and the northeast boundary of the property. The gypsum is locally interbedded with basalt, but is likely also tectonicly-repeated. 2.5.2 Mineralization Associated with the Spy Sill The mineralization described below along with Ni-Cu-PGE analytical results are intended as an overview. Analytical results and sample locations are documented in Bell (1996). A more detailed discussion of Ni-Cu-PGE geochemistry specifically relevant to this study is given in later sections Ni-Cu-PGE mineralization on the property is associated with the basal marginal gabbro phase of the Spy Sill. Sulphide mineralization at the Spy showing occurs in siltstone in the footwall of the sill, marginal gabbro and feldspathic peridotite. Massive chalcopyrite-pyrrhotite lenses in footwall siltstone are up to 2.0 x 0.25 m; these lenses grade up to 2.6% Ni, 10.45% Cu, 0.09% Co, 75.8 g/t Pt, 7.9 g/t Pd and 7.0 g/t Au. The host siltstone is weakly altered, but highly fractured with chalcopyrite-pyrrhotite mineralization occurring along the fractures. The basal marginal gabbro unit hosts massive sulphide lenses, net-textured sulphide and disseminated sulphide. Gabbro hosted massive pyrrhotite-chalcopyrite lenses grade up to 3.1% Ni, 2.8% Cu, 0.2% Co, 3.1 g/t Pt, 1.4 g/t Pd and 1.0 g/t Au. These lenses are up to 20 cm thick and occur at the gabbro-siltstone contact and within zones of gabbro-hosted semi-massive sulphide. Disseminated and net-textured pyrrhotite and chalcopyrite within marginal gabbro grade up to 1.2% Ni, 1.0% Cu , 0.009% Co, 290 ppb Pt, 180 ppb Pd and 88 ppb Au in grab samples. A 1.2-m chip sample of net-textured pyrrhotite-chalcopyrite mineralization in marginal gabbro returned values of 0.2% Ni, 1.4% Cu, 1.9 g/t Pt, 1.0 g/t Pd and 0.7 g/t Au. Ni-Cu-PGE mineralization at the base of the Spy Sill has been traced over a strike of 3.6 km. Gabbro/sulphide talus from the base of the sill, 900 m northwest of the Spy showing, returned values of 1.5% Ni, 0.4% Cu, 0.7 g/t Pt and 1.4 g/t Pd. A gabbro boulder with coarse blebs of pyrrhotite was found at the base of the Spy Sill, 2.7 km southeast of the Spy showing. This boulder returned values of 0.3% Ni, 0.4% Cu, 0.4 g/t Pt and 0.7 g/t Pd. Several pyrrhotite-magnetite layers occur between the Spy Sill and the base of the Nikolai basalt. One magnetite layer is up to 10 m thick, while pyrrhotite layers are up to 4 m thick. A 4-m thick pyrrhotite layer discovered in 1995 was named the Claimpost showing. This pyrrhotite body is hosted within silicified siltstone and is capped by gabbro and magnetite. Minor magnetite and 17 chalcopyrite occur with the pyrrhotite. Copper values from chip samples across the layer returned values from 0.1 to 0 .3%; Co values are anomalous (330-640 ppm). The maximum Ni value from these samples is 520 ppm. Pt and Pd are not anomalous. Common skarn minerals such as garnet, epidote and clinopyroxene are not present in sediments adjacent to the pyrrhotite-magnetite bodies. The sediments are locally highly silicified, probably due to the presence of numerous Maple Creek Gabbro sills. Most of the pyrrhotite-magnetite layers are not in contact with gabbro sills. The 10-m thick magnetite layer abruptly changes to a thinner pyrrhotite-rich zone. No bedding is visible in either pyrrhotite or magnetite rich layers. Thus, it is not clear whether these layers represent skarn-type mineralization or are a type of syngenetic iron formation. Exposure on the property is abundant at elevations above glacial sediments and valley filling talus fans, but the bases of the various peridotite intrusions are generally poorly exposed. The best-exposed peridotite intrusion on the property is the Spy Sill. Even here, the base of the sill is relatively poorly exposed as mineralized marginal gabbro and massive sulphide lens weather recessively relative to the peridotite. A thin layer of fine talus was observed covering gossanous material at several places along the base of the sill; this makes locating sulphide mineralization difficult. Other intrusions such as the Duke Intrusion have almost no exposure at their basal contacts. 18 Section 3: Petrology and Geochemistry of the Spy Sill and Associated Intrusions 3.1 Introduction Although a general description of Kluane-type intrusions on the property has been given, this study will focus on two intrusions, namely the Spy and Lewis Intrusions. These intrusions were selected because of their relative lack of alteration, good exposure, spatial separation and variation in associated sulphide mineralization. Section 3 deals with the petrology and geochemistry of the intrusions and not their associated sulphide mineralization. Samples were collected from measured sections through these intrusions. The location of the sections sampled is shown in Figure 2-4. Stratigraphic sections through the two measured sections including sample locations are shown in Figures 3-1 and 3-2. 3.2 Lithological Description Spy Intrusion Section: A marginal gabbro unit is preserved at the top of the intrusion where it is in contact with brown to dark grey shale and siltstone of the Hasen Creek Formation. The upper marginal gabbro is 14 m thick and locally contains up to 4 % pyrrhotite. Under this unit is a thin, 4 m thick unit of what appears to be Maple Creek Gabbro. This Maple Creek Gabbro unit is coarser grained and more chloritic than the marginal gabbro above. Below this unit is a 7 m thick, brown to buff coloured, silicified siltstone layer. Beneath this is a 4.5 m thick buff to white coloured limestone unit. Much of the limestone has been altered to calc-silicate minerals. Below this carbonate unit is 90 m of feldspathic peridotite which comprises the main part of the sill. A 10 m thick marginal gabbro unit underlies the peridotite. 15 -20 cm of olivine gabbro marks the transition between the peridotite and the marginal gabbro. The lower marginal gabbro is leucocratic, fine grained to chilled and contains between 2 - 3 % sulphide (pyrrhotite and chalcopyrite). Small (3.5 x 0.3 m) sulphide lenses (dominantly pyrrhotite with lesser chalcopyrite) locally occur within this unit. A thin (9 m) discontinuous feldspathic peridotite unit separates the lower marginal gabbro from a 30 m thick Maple Creek Gabbro sill. The Maple Creek Gabbro is medium grained, chloritic and contains local patches containing quartz-carbonate veinlets. In the siltstone immediately below the marginal gabbro are chalcopyrite-rich lenses of sulphide up to 2.0 x 0.25 cm. These lenses occur within 5 m of the gabbro - siltstone contact. The sill contacts appear to have an orientation which is generally conformable with that of the Hasen Creek sediments which have an average strike of 315° and an average dip of 35° to the southwest. Lewis Intrusion Section: A 380 m thickness of the Lewis intrusion is preserved where it underlies Tertiary volcanic cover rocks. The intrusion is 1995 Spy Sill Section - Klu Property Figure 3-1 Legend H m J Shale Chert Siltstone -20 m S = S = ~ ^ ^ r ^ ^ l Shale, Siltstone Marginal Gabbro u\ Maple Creek Gabbro Maple Creek Gabbro Siltstone meters 20 1996 Lewis Intrusion Section - Klu Property Figure 3-2 v v v v * v v v V v y v •500 m Rx225892 • • • • '89.1B ; 890B'. •889B • •400 m '.' . . • • ~ l\<:r~ 888V J f i l l . . ( V V v v v Tertiary Volcanics V v - . • Triassic Peridotite •887B'. • 300 m Triassic Marginal Gabbro ; 886B'. • ".•'.-Permian Hasen Creek Formation Siltstone • 885" : • 200 m • Sample Location (Rx225???) 884B•; ! 883 B; • ;882B ;" .' 881 • ' . ; 880 B • •100 m (1 878 II r 0 50 877,. -0 m meters -LU. — — " — — ——r_ 21 composed of two separate sills of dominantly pyroxenitic peridotite. Both of these sills have a gabbro unit at their base; these gabbro units appear to be of marginal-type. The marginal gabbro at the base of the lower sill is 25 m thick. This unit is overlain by 275 m of pyroxenitic peridotite. This peridotite has hackly weathering texture and macroscopically appears to contain a higher modal abundance of pyroxene than the Spy Sill. Above this unit is a 70 m thick section of Hasen Creek Formation siltstone. The sills have an orientation similar to bedding in the siltstone which trends at 358/36. The gabbro unit at the base of the upper sill is 20 m thick. This is overlain by 60 m of pyroxenitic peridotite after which an erosional surface occurs. The Tertiary volcanics occur above this surface. 3.3 Petrography 3.3.1 Peridotite The peridotite in the Spy section contains approximately 60% olivine in equant to slightly elongate anhedral grains up to 1 mm and as poikiolitic grains within clinopyroxene. Serpentinization to antigorite is generally limited to fractures. This unit contains approximately 35% clinopyroxene (probably augite or diopside) which form grains up to 2.5 mm. This unit generally contains less than 5% interstitial material giving it an adcumulate texture. Secondary magnetite forms along fractures in olivine whereas chromite (1-2%) occurs interstitially. Accessory phlogopite is present in this unit along with minor plagioclase (up to 5%). The peridotite contains up to 2% disseminated pyrrhotite. The modal percentage of the major mineral constituents in this unit defines it a wehrlite. The mineralogy and textures of the peridotite unit from the Spy Sill are illustrated in photomicrographs in Plates 3-1 through 3-5. The Lewis Intrusion differs from the Spy Sill in that it contains less olivine and a greater amount of interstitial material. The peridotite in the Lewis Intrusion has approximately 40% olivine in equant anhedral grains up to 1.5 mm. Locally olivine also occurs as poikilitic grains within clinopyroxene. Clinopyroxene oikocrysts up to 5 mm make up approximately 30% of the rock. Plagioclase locally forms up to 10% of this unit. Accessory phlogopite is also present. Serpentinization in the Lewis peridotite varies from occurring only along fractures to local partial alteration of entire olivine grains. The modal percentage of the major minerals in this unit make it a pyroxenitic wehrlite. The amount of interstitial material in this unit gives it a mesocumulate texture. 3.3.2 Marginal Gabbro The marginal gabbro unit from the Spy Sill is generally a fine-grained leucocratic phase with 30 to 55% plagioclase. Plagioclase laths are up to 0.5 mm and are 22 Plates 3-1 and 3-2: Photomicrographs of peridotite from the Spy Sill (samples Rx222114 and Rx222108). Olivine (ol) occurs in equant anhedral grains and as poikolitic grains in clinopyroxene (cpx). Clinopyrexene oikocrysts up to 5 mm are also present. Crossed polars. 2 2 * Plate 3-2 23 Plates 3-3 and 3-4: Photomicrographs of peridotite from the Spy Sill (sample Rx222110). Plain light and crossed polars. 13* Plate 3-4 24 Plate 3-5: Photomicrograph of peridotite from the Spy Sill (sample Rx222120). Crossed polars. Plate 3-6: Photomicrograph of marginal gabbro from the base of the Spy Sill (sample Rx222116). This unit is generally a fine-grained leucocratic phase with 30 to 5 5 % plagioclase (pig). Plagioclase laths are up to 0.5 mm and are typically saussuritized. Clinopyroxenite (cpx) grains (up to 0.7 mm) make up approximately 4 0 % of the rock and are dominantly altered to secondary amphibole (tremolite). Crossed polars. Plate 3-6 25 typically saussuritized. Clinopyroxenite grains (up to 0.7 mm) make up approximately 4 0 % of the rock and are dominantly altered to secondary amphibole (tremolite). Chlorite locally forms from clinopyroxene and groundmass alteration. Secondary quartz is also present. A subophitic texture is locally present with plagioclase laths partially intruding into clinopyroxenite grains. The marginal gabbro commonly contains up to 5% opaques. Most of the opaque grains are pyrrhotite which average about 0.4 mm; these grains are locally rimed by chalcopyrite. Chalcopyrite also occurs in discrete interstitial grains and as inclusions in pyrrhotite. Minor secondary magnetite with internal skeletal ilmenite is also present. Marginal gabbro with significant mineralization is described in Section 6. Photomicrographs of marginal gabbro are shown in Plates 3-6 and 3-7. Marginal gabbroic rocks from the Lewis Intrusion are petrographically similar except they contain a higher modal abundance of clinopyroxene and less plagioclase. This gabbro is a non-cumulus phase which was emplaced marginal to the peridotite. This unit locally is chilled near the country-rock contact. 3.3.3 Maple Creek Gabbro This unit is typically coarser grained than the marginal gabbro, has a darker (mesocratic) colour index and does not have chilled margins. Clinopyroxene grains in the Maple Creek gabbro are up to 1.2 mm. Alteration of clinopyroxene grains to tremolite and chlorite varies from weak to total on a thin section scale. Clinopyroxene and its alteration products account for 6 5 % of the rock. Plagioclase is invariably saussuritized. Accessory phlogophite occurs in this unit. The Maple Creek Gabbro also typically does not contain more than 1% pyrrhotite. A photomicrograph of Maple Creek gabbro is shown in Plate 3-8. 3.4 Geochemistry (Majors, Traces, REE, Ni-Cu-PGE, Sulphur Isotopes) 3.4.1 Methods Sample locations are shown relative to stratigraphic height and lithologies in Figures 3.1 and 3.2 for the Spy and Lewis sections. Approximately 3 kg of material was collected from each site (25 samples from the Spy Sill, 17 samples from the Lewis Intrusion) and an attempt was made to chip off or avoid weathering rinds. During fieldwork, duplicate samples were collected from both gabbro and peridotite in the Spy Sill Section and submitted for analysis. A pulp duplicate from one of the peridotite samples was submitted from the Lewis Intrusion Section. In addition, Inco and international standards were submitted for analysis. All samples were prepared at Chemex Labs in North Vancouver, B.C.. 26 Plate 3-7: Photomicrograph of upper marginal gabbro from the Spy Sill (sample Rx222103). This unit is generally a fine-grained leucocratic phase with 30 to 55% plagioclase (pig). Plagioclase laths are up to 0.5 mm and are typically saussuritized. Clinopyroxenite (cpx) grains (up to 0.7 mm) make up approximately 40% of the rock and are dominantly altered to secondary amphibole (tremolite). Crossed polars. Plate 3-8: Photomicrograph of Maple Creek gabbro adjacent to the Spy Sill (sample Rx222121). Clinopyroxene grains in the Maple Creek gabbro are up to 1.2 mm. Alteration of clinopyroxene grains to tremolite and chlorite varies from weak to total on a thin section scale. Clinopyroxene and its alteration products account for 65% of the rock. Plagioclase in this unit is invariably saussuritized. Crossed polars, Plate 3-8 27 Samples were crushed and then pulverized in a ring pulverizer to -150 mesh. Tungsten carbide pulverizing equipment was used for samples from the Spy Section. Samples from the Spy Section were analyzed for whole rock (Al 20 3, CaO, Cr 2 0 3 , Fe 20 3 , K 20, MgO, MnO, Na 20, P 2 0 5 , Si0 2 , T i0 2 and LOI) by Chemex Labs of North Vancouver, B.C.. Whole rock determinations were made by x-ray fluorescence (XRF). Trace element (Nb, Zr, Y, Sr, Rb and Ba) analyses by pressed pellet XRF analyses were done at the Inco Exploration Lab in Copper Cliff, Ontario. Inductively Coupled Plasma (ICP) analyses for metals and major elements (Ag, Al, Ba, Be, Bi, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, Sr, Ti, V, W and Zn) were carried out by Chemex Labs. Samples were digested with HF resulting in a total digestion for most elements. Sulphur analyses were done at Chemex Labs using a Leco induction furnace. Trace-level precious metals (Pd, Pt and Au) were analyzed by fire assay with an ICP-mass spectrometer (ICP-MS) finish at Actlabs in Ancaster, Ontario. Rare Earth Element (REE - La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb and Lu) plus Th and U analyses were completed by ICP-MS at XRAL Laboratories in Don Mills, Ontario. Se and As were analyzed by the ICP-hydride method at Acme Labs in Vancouver, B.C.. Digestion was by a 3-1-2 mixture of HCI, HN0 3 and H 20. Sulphur isotopic analyses were done on the mass spectrometer at the University of Calgary in Calgary, Alberta. Sulphur extraction was done using KIBA reagent. An analysis of error (precision and accuracy) is included with the analytical data in Appendix 1. 3.4.2 Results Various elements, oxides and ratios are plotted versus stratigraphic height for the Spy Sill are presented in Figures 3-3a to 3-3w and for the Lewis Intrusion in Figures 3-4a to 3-4p. Geochemical variations between elements are shown in Figures 3-5a to 3-5z. Different symbols are plotted for each rock type at stratigraphic heights coinciding with sample locations; each x value of these sample locations is set equal to 1. In this way, a crude stratigraphic column is shown for reference on each of the above graphs. Element abundances below detection limit are assigned values of one half detection limit for plotting purposes. Whole rock data (major element oxides) have been recalculated on a LOI free basis; these recalculated values are what is plotted and discussed in the text. 28 Figure 3-3a: Plot of Stratigraphic Height versus Ni in the Spy Sill. Ni values in peridotite range from 1105 to 1365 ppm and grossly vary with MgO. Legend abbreviations are as follows: TrPrd= Triassic peridotite, TrMG= Triassic marginal gabbro, TrMCG= Triassic Maple Creek gabbro, PHCSH= Permian Hasen Creek Formation shale, PHCLST= Permian Hasen Creek Formation limestone, PHCSLT= Permian Hasen Creek Formation siltstone. Figure 3-3b: Plot of Stratigraphic Height versus Cu in the Spy Sill. Cu values in peridotite range from 78 to 313 ppm and have a chaotic profile. Figure 3-3c: Plot of Stratigraphic Height versus MgO in the Spy Sill. MgO levels in peridotite range from 25 to 3 5 % and form three crude plateaus, which show an overall decrease towards the base of the sill. MgO levels in the marginal gabbro are approximately 3%, whereas the Maple Creek gabbro contains approximately 1 5 % . Figure 3-3d: Plot of Stratigraphic Height versus Cr in the Spy Sill. Cr values in the peridotite range from 1510 to 2610 ppm and display no obvious pattern. 28* Height (m) vs. Ni (ppm) • Spy SiH Height (m) vs. Cu (ppm) 220 140 100 60 20 -200 200 600 1000 1400 0 400 800 1200 1600 ° PHCSH ° TRMG » PHCSLST 4 PHCLST • TRPRO • TRMCG • NIPPM 220 140 £ 100 60 20 a t • • o PHCSH P TRMG » PHCSLST 4 PHCLST • TRPRD « TRMCG "-50 0 60 100 150 200 250 300 350 * CU_PPM Height (m) vs. MgO% In Spy Sill 220 r 180 140 I 100| 60 1 If .20 1 o PHCSH B TRMG « PHCSLST » PHCLST • TRPRD • TRMCG 5 10 15 20 25 30 36 40 • MOO_% 180 £ 100 O 60 Height (m) vs. Cr (ppm) •20 -200 400 1000 1600 2200 2800 ° PHCSH o TRMG o PHCSLST a PHCLST • TRPRD » TRMCG • CR_PPM 29 Figure 3-3e: Plot of Stratigraphic Height versus Ni/MgO in the Spy Sill. The Ni/MgO ratio shows a similar profile to that of Ni versus stratigraphic height reflecting the expected control of olivine on Ni contents. Figure 3-3f: Plot of Stratigraphic Height versus Mg# in the Spy Sill. Mg# show a virtually flat profile through the peridotite section of the sill. Figure 3-3g: Plot of Stratigraphic Height versus Ni/Cu in the Spy Sill. Ni/Cu ratios in the peridotite range from 4.4 to 15.2, whereas Ni/Cu ratios in marginal gabbro vary from 0.2 to 4.5. Figure 3-3h: Plot of Stratigraphic Height versus Cu/Zr in the Spy Sill. Cu/Zr ratios display a similar profile to Cu suggesting the abundance of Cu is controlled by fractionation of low-sulphur rocks of the sill. 21* Height (m) vs. NiMgO In Spy Sill Height (m) vs. Mg#-Spy Sill 220 r 180 140 53 tu x 60 ° PHCSH D TRMQ ° PHCSLST 4 PHCLST • TRPRO « TRMCG 1S 25 35 46 55 * NlwMGO 220 100 a so PHCSH T R M G PHCSLST PHCLST TRPRO TRMCG MG 220 r 180 140 Height (m) vs. Ni/Cu - Spy Sill 10 14 PHCSH TRMG PHCSLST PHCLST TRPRD TRMCG NI CU 220 140 Height (m) vs. Cu/Zr in Spy Sill • • • • o PHCSH o TRMG o PHCSLST » PHCLST • TRPRD » TRMCG • CU_ZR 30 Chemical Variations with Stratigraphic Height - Spy Sill MgO levels within the peridotite of the Spy Sill range between 25 and 3 5 % . MgO levels decrease towards the base of the sill in a series of three crude plateaus (see Figure 3-3c). The plot of Mg# versus height (Figure 3-3f) shows a virtually flat profile through the peridotite. MgO levels in the marginal gabbro are approximately 3%. MgO levels drop very quickly across the lower marginal gabbro-peridotite contact over a distance of less than 2 m. Samples from the Maple Creek gabbro have MgO levels of approximately 15% . Ni levels within the sill vary between 1105 and 1365 ppm (see Figure 3-3a). The Ni levels grossly vary with MgO; the Ni/MgO versus height plot (Figure 3-3e) displays a similar trend to the Ni versus height plot. Cu values in the peridotite (Figure 3-3b) vary between 78 and 313 ppm and have a somewhat chaotic profile. Ni/Cu values (Figure 3-3g) in the peridotite vary between 4.4 and 15.2, while Ni/Cu values in the marginal gabbro vary from 0.2 to 4.5. Cu/Zr values (as per Hawkesworth et al, 1995) were plotted versus height (Figure 3-3h) to monitor the variation of Cu with respect to fractionation. The profile is very similar to the Cu versus height profile suggesting that the Cu levels are controlled by fractionation in the low-sulphur rocks of the sill. The plot of Co versus height (Figure3-3j) has a very similar profile to that of Ni. Co values in the peridotite vary between 85 and 105 ppm; values in the marginal gabbro and Maple Creek gabbro vary between 17 to 67 ppm. T i 0 2 levels in the peridotite vary from 0.3 to 0.7 and gradually increase from the top of the peridotite to its base. Marginal gabbro and Maple Creek gabbro have T i 0 2 values between 0.9 and 1.5. Cr values in the peridotite (Figure 3-3d) vary from 1510 to 2610 ppm. Au, Pd, Pt and Pt/Pd are plotted versus height in Figures 3-3k to 3-3n respectively. Pt and Pd profiles in the main peridotite part of the sill do not display any obvious systematic pattern; the highest values are in the middle and lower parts of the unit. Pt and Pd vales vary from 8.5 to 44.7 and 20.7 to 81 ppb respectively within the peridotite; Pt/Pd values vary from 0.4 to 1.0. Au values within the peridotite range from 1 to 14 ppb. The narrow lower peridotite lens has a significantly higher Pt/Pd ratio (4.5) and a higher Au level (17 ppb). The marginal gabbro unit has Pt values from 0.7 to 4.9 ppb and Pd values vary from 1.8 to 7.5 ppb; Pt/Pd values range from 0.4 to 1.5. The Maple Creek gabbro is differentiated from the marginal gabbro by its higher Pt/Pd ratio (>2). Figure 3-3o is a plot of normalized Ce/Yb ratios versus height. Ce and Yb are normalized by chondrite values listed by Sun (1982). Ce/Yb N ratios in peridotite samples are all less than 3, as are ratios from the Maple Creek gabbro. Ce/Yb N ratios in the marginal gabbro are significantly higher being in the 3 to 4 range. Ce/Yb N ratios in the Permian (Hasen Creek Formation) in the same range as the gabbros. This suggests that the marginal gabbros have been contaminated by these Island Arc derived sediments. Figures 3-3u to 3-3w show profiles of Ba, Th and U versus height. These plots all show low levels of these elements in the 3 1 Figure 3-3i: Plot of Stratigraphic Height versus T i 0 2 in the Spy Sill. T i 0 2 levels in the peridotite vary from 0.3 to 0.7% and gradually increase from the top of the unit to the base. Marginal gabbro and Maple Creek gabbro have T i 0 2 between 0.9 and 1.5%. Figure 3-3j: Plot of Stratigraphic Height versus Co in the Spy Sill. Co values vary between 85 and 105 ppm in the peridotite.; values in the marginal gabbro and Maple Creek gabbro vary between 17 and 67 ppm. The Co versus height profile is very similar to the profile for Ni. 3k Height (m) vs. T i 0 2 (%) in Spy Sill 220 180 140 I E 100 O LU X 60 20 -20 • 0 8 • • ft • • • • • • • • • • • • ri • • • a • • B • 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 o PHCSH a TRMG o PHCSLST A PHCLST • TRPRD * TRMCG • T I02 % Height (m) vs. Co (ppm) in Spy Sill 220 180 140 100 O LU I 60 20 -20 o • B o • 4 • • -20 20 40 60 80 100 120 o PHCSH a TRMG o PHCSLST A PHCLST • TRPRD * TRMCG • CO PPM 32 Figure 3-3k: Plot of Stratigraphic Height versus Au in the Spy Sill. Au values within the peridotite range from 1 to 14 ppb. The small peridotite lens below the main part of the sill has a higher Au abundance (17 ppb). Figure 3-3I: Plot of Stratigraphic Height versus Pd in the Spy Sill. Pd values in the peridotite vary from 20.7 to 81 ppb and show no systematic pattern. Pd values in the marginal gabbro vary from 1.8 to 7.5 ppb. Figure 3-3m: Plot of Stratigraphic Height versus Pt in the Spy Sill. Pt values in the peridotite vary from 8 to 44.7 ppb and show no systematic pattern. Pd values in the marginal gabbro vary from 0.7 to 4.9 ppb. Figure 3-3n: Plot of Stratigraphic Height versus PtVPd in the Spy Sill. Pt/Pd ratios in the peridotite vary from 0.4 to 1.0. The peridotite lens below the main sill has a Pt/Pd ratio of 4.5, which is significantly higher than the in the main part of the sill. Marginal gabbro has Pt/Pd ratios ranging from 0.4 to 1.5. The Maple Creek gabbro is differentiated from the marginal gabbro by higher Pt/Pd ratios (>2). 3Z<K Height (ra) vs. Au (ppb) in Spy Sill 220 r 140 • * • • m • o • o PHCSH a TRMG « PHCSLST A PHCLST • TRPRD • « TRMCG 6 10 14 18 22 * AUPPB 220 180 fe 100 a -20 Height (m) vs. Pd (ppb) -10 10 30 50 70 90 o PHCSH D TRMG o PHCSLST * PHCLST • TRPRD » TRMCG • PO PPB Height (m) vs. Pt (ppb) in Spy Silt 180 60 20 o PHCSH o TRMG o PHCSLST » PHCLST • TRPRD » TRMCG -5 5 15 25 35 45 55 * PT-PPB Height (m) vs. Pl/Pd in Spy Sill o PHCSH 0 T R M G « PHCSLST 1 PHCLST • TRPRD • TRMCG • PT PD Figure 3-3o: Plot of Stratigraphic Height versus Ce/Yb (normalized) in the Spy Sill. Values are normalized by chondritic values of Sun (1982). Ce/YbN ratios in peridotite and Maple Creek gabbro are less than 3, whereas ratios in marginal gabbro are in the 3 to 4 range suggesting the marginal gabbro has experienced contamination from Island Arc derived sediments. Figure 3-3p: Plot of Stratigraphic Height versus S in the Spy Sill. Peridotite and Maple Creek gabbro samples have S contents less than 0.1%, whereas most marginal gabbro samples have S contents greater than 0.2%. 33* Height (m) vs. Ce/Yb (Normalized) - Spy Sill 1-0 1.5 2.0 2.5 3.0 3.5 4.0 o PHCSH n TRMG o PHCSLST A PHCLST • TRPRD » TRMCG 4 5 • CE_YB_N 220 j • Height (m) vs. S (%) - Spy Sill 0 180 B • 140 100 60 • • 0 HEIGHT.M : • ; • • • • • i • • • • • • • • • • • • • 1 • o D PHCSH TRMG 20 O PHCSLST  A PHCLST • O • TRPRD -20 TRMCG -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 * S - % 34 Figure 3-3q: Plot of Stratigraphic Height versus As in the Spy Sill. Arsenic levels in peridotite and Maple Creek gabbro are less than 3.5 ppm, whereas marginal gabbro samples have As in the 2 to 9 ppm range. Figure 3-3r: Plot of Stratigraphic Height versus 8 3 4S in the Spy Sill. 5 3 4S values in peridotite cluster at -5 7 0 0 and range from -7 to +3 7 0 0. The upper marginal gabbro has 8 3 4S values in the -15 to -10 7 o o range, whereas the lower marginal gabbro has a wide range of 5 3 4S values (-5 to +25 700). Samples of Maple Creek gabbro have positive 8 3 4S values in the +5 to +207 o o range. Figure 3-3s: Plot of Stratigraphic Height versus Se in the Spy Sill. Se levels in the Spy Sill section are highest in the upper marginal gabbro unit with Se up to 2.4 ppm. Se levels in the peridotite range between 0.2 and 0.9 ppm. The Hasen Creek Formation sediments have low Se levels with 2 of 3 samples having values below the detection limit. Figure 3-3t: Plot of Stratigraphic Height versus S/Se in the Spy Sill. S/Se ratios in the peridotite samples are less than 4,500. The lower marginal gabbro unit has anomalously high S/Se with ratios up to 14,000. 3Vq Height (m) vs. As (ppm) in Spy Sill Height (m) vs. Del 34S (per mil) 220 o PHCSH Q TRMG » PHCSLST A PHCLST • TRPRD • TRMCG • AS_PPM 180 140 100 60 •20 -16 -10 -5 0 5 10 15 20 25 PHCSH TRMG PHCSLST PHCLST TRPRD TRMCG DEL 34S Height (m) vs. Se (ppm) in Spy Sill Height (m) vs. S/Se in Spy Sill ° PHCSH 0 TRMG o PHCSLST A PHCLST • TRPRD • TRMCG -0.2 02 0.6 1.0 1.4 1.8 2.2 26 * SE.PPM 2000 2000 6000 10000 14000 0 4000 8000 12000 16000 * S - S E 35 Figures 3-3u, 3-3v and 3-3w: Plots of Stratigraphic Height versus Ba, Th and U in the Spy Sill. The peridotite and Maple Creek gabbro have low abundances of these elements, whereas the marginal gabbro and Hasen Creek Formation sediments have significantly higher levels. This may be related to the marginal gabbro being contaminated by the sediments. 3S<K 180 140 J 100 60 -20 Height (m) vs. Ba (ppm) in Spy Sill o PHCSH a TRMG o PHCSLST * PHCLST • TRPRD « TRMCG -200 200 600 1000 1400 1800 * BA.PPM 140 5E' 100 a Height (m) vs. Th (ppm) in Spy SHI -0.5 0.5 1.5 2.5 3.5 4 5 5.5 o PHCSH Q TRMG o PHCSLST A PHCLST • TRPRD « TRMCG • TH PPM 220 140 60 20 Height (m) vs. U (ppm) in Spy Sill t* -20 1 -0.2 0.2 0.6 1.0 1.4 1.8 2.2 2.8 3.0 PHCSH T R M G | PHCSLST PHCLST TRPRD TRMCG U PPM 36 peridotite and Maple Creek Gabbro while the marginal gabbro has significantly elevated levels. The elevated Ba, Th and U may also be due to contamination from Hasen Creek Formation sediments. Sulphur is plotted versus height in Figure 3-3p. All samples from the peridotite, sediments and Maple Creek Gabbro have S contents less than 0 .1% . 6 of the 7 marginal gabbro samples have S contents greater than 0.2%. S contents in the marginal gabbro are up to 1.2%. Figures 3-3q to 3-3t show plots of As, 5 3 4S, Se and S/Se versus height. Arsenic levels in the peridotite and Maple Creek gabbro are less than 3.5 ppm, while the marginal gabbro samples have As in the 2 to 9 ppm range. 8 3 4S values in the peridotite cluster about the -5 7 0 0 level and range between -7 and +37^. The upper marginal gabbro has 5 3 4S values in the -10 to -15 7 0 0 range while the lower marginal gabbro unit has a wide range of 8 3 4S values. The lower marginal gabbro has 5 3 4S values in the -5 to +25 7 0 0. The upper samples from the lower marginal gabbro have 8 3 4S values in the similar to most of the peridotites at about -5 7 0 0 and the samples from closer to the base of the lower marginal gabbro have S 3 4S values in the 0 to +15 7^ range. The Maple Creek gabbro has positive 8 3 4S values in the +5 to +207 o o range. Typical 8 3 4S values in magmatic rocks range from -5 to +5 7 0 O and average approximately 07 o o ; sedimentary sulphates have 8 3 4S values ranging from +10 to +30 7 0 0 depending on geological age and syn-sedimentary sulphides typically range from -30 to +5 7 0 0 (Faure, 1986). The strongly positive 8 3 4S values in the Hasen Creek sediment samples, the Maple Creek gabbro and one of the marginal gabbro samples may be related to low amounts of sedimentary sulphate in Hasen Creek Formation sediments which are assimilated into gabbroic and ultramafic magma. It must be noted that this contamination scenario is probably only applicable when magma with very low levels of magmatic sulphur mixes on a local scale with isotopically heavy synsedimentary sulphur; the S levels from the strongly isotopically positive gabbro samples (83 4S > + 5 7 J are all less than 0.025%. Se levels in the Spy Sill section are highest in the upper marginal gabbro unit with Se up to 2.4 ppm. The Hasen Creek sediments generally contain low Se levels with 2 of the 3 samples having Se values less than the detection limit. Se levels in the main peridotite section of the sill range from 0.2 to 0.9 ppm. S/Se ratios in the peridotite unit are less than 4,500 and are typical of Kluane Belt ultramafics (Hulbert, 1995) and mafic-ultramafic rocks with magmatic sulphide in general (Loftus-Hills and Solomon, 1967 and Thompson, 1982). The lower marginal gabbro unit has anomalously high S/Se ratios up to 14,000. These high ratios are probably caused by the marginal gabbroic magma encountering and assimilating synsedimentary sulphide at some point during its ascent or lateral intrusive path. 37 Geochemical Variations with Stratigraphic Height - Lewis Intrusion MgO levels in the peridotite of the Lewis Intrusion (Figure 3-5b) vary between 25 and 3 5 % ; MgO levels in the marginal gabbro vary between 6 and 16% . MgO levels show an overall decline within the peridotite from top to base in both of the individual sills. A similar trend is reflected in the Figure 3-4c which shows the variation in Mg# with respect to height. Mg# within the intrusion varies between 54 and 82 with the peridotites varying between 77 and 82. The plot of Ni versus height (Figure 3-4a) shows an overall trend of Ni decreasing from top to base in each of the sills. Ni levels with respect to height change in a very similar pattern to MgO, suggesting Ni levels are controlled by a systematic decrease in forsterite content of olivine or of modal olivine from top to base. This relationship is further illustrated in the plot of Ni/MgO versus height (Figure 3-4d). T i 0 2 values in the peridotite vary between 0.25 and 0.6% (Figure 3-4e). The T i 0 2 values show an overall increase towards the base of the sills. Marginal gabbros have T i 0 2 values up to 1.0%. Cr is plotted versus height in Figure 3-4f. This plot shows Cr values varying from 111 ppm in the marginal gabbro to 2690 ppm in the peridotite. There are no obvious variations in Cr levels with position in the sill. Cu values versus height are shown in Figure 3-4g. In peridotite, Cu varies between 49 and 440 ppm while gabbro values vary from 31 to 117 ppm. Cu values have a somewhat erratic profile, but the lower Cu values generally occur at the base of the two individual sills and near the top of the lower sill. The variation of Co values with height for the Lewis Intrusion is shown in Figure 3-4h. Co values in the peridotite vary between 87 and 129 ppm while values in the gabbro range from 38 to 58 ppm. Co varies almost exactly as Ni does with stratigraphic height; Co shows gradually decreasing values towards the base of both sills. Au, Pd, Pt and Pt/Pd are plotted versus height in Figures 3-4i to 3-4I. In the peridotite phase, the lower values of each of the precious metals occur at the base of both sills and near the top of the lower sill. Pt, Pd and Au values in the peridotite of the Lewis Intrusion vary between 4.1 and 63 ppb, 2 and 18.8 ppb, and 2 and 31 ppb respectively. Pt, Pd and Au levels in the gabbro are generally low except for one of the gabbro samples from the base of the lower sill which has a Pd value of 20.6 ppb. These precious metals vary with stratigraphic height in a similar way to Cu, especially Pt and Pd. Pt/Pd ratios in gabbro and peridotite samples vary from 0.4 to 4.4. Lower ratios occur near the base of both sills with low ratios also occurring near the top of the lower sill. A plot of S versus height is presented in Figure 3-4m. This plot shows S having a similar profile with height to Cu, Pt, Pd and Au; suggesting S levels have a significant control on copper and precious metal contents. S levels vary from detection limit (0.01%) to 0.12%. 38 Figure 3-4a: Plot of Stratigraphic Height versus Ni in the Lewis Intrusion. Ni values in peridotite range between 300 and 800 ppm and show an overall decreasing trend from top to base in the two sills. Figure 3-4b: Plot of Stratigraphic Height versus MgO in the Lewis Intrusion. MgO levels in the peridotite vary between 25 and 3 5 % , whereas MgO levels in the marginal gabbro vary between 6 and 16% . MgO levels vary systematically with Ni, suggesting Ni abundance is controlled by the modal amount of olivine or changes in the forsterite content of olivine. Figure 3-4c: Plot of Stratigraphic Height versus Mg# in the Lewis Intrusion. The Mg# profile shows a similar pattern that of MgO. The Mg number of the peridotite varies between 77 and 82 and the range of the entire intrusion is from 54 to 82. Figure 3-4d: Plot of Stratigraphic Height versus Ni/MgO in the Lewis Intrusion. The Ni/MgO plot further illustrates the close correlation between Ni and MgO. 33* Height (m) vs. Ni (ppm) - Lewis Intrusion 550 450 150 50 -50 -100 100 TRMG TRPRD NI PPM Height (m) vs. MgO (%) - Lewis Intrusion 350 5 fe' 250 2 150 50 -50 o TRMG D TRPRD -5 0 5 10 15 20 25 30 35 40 0 MSO_% 550 I 450 350 fe 250 | © a i 150 50 -50 1 Height (m) vs. Mg* - Lewfs Intrusion -10 10 30 70 Height (m) vs. Ni/MgO - Lewis Intrusion 550 450 £ 250 <2 150 TRMG TRPRD 6 10 14 18 22 26 TRMG TRPRD NI MGO 39 Figure 3-4e: Plot of Stratigraphic Height versus T i 0 2 in the Lewis Intrusion. T i 0 2 levels in peridotite vary between 0.25 and 0.6% and shown an overall increase towards the base of the sills. Marginal gabbro has T i 0 2 values up to 1.0%. Figure 3-4f: Plot of Stratigraphic Height versus Cr in the Lewis Intrusion. Cr values vary from 111 ppm in the marginal gabbro to 2690 ppm in peridotite. There are no obvious variations in Cr with position in the sills. Figure 3-4g: Plot of Stratigraphic Height versus Cu in the Lewis Intrusion. Cu values vary between 49 and 440 ppm in peridotite and between 49 and 440 in gabbro. The Cu profile in somewhat erratic, but the lower values occur at the base of the two sills and near the top of the lower sill. Figure 3-4h: Plot of Stratigraphic Height versus Co in the Lewis Intrusion. Co values in peridotite vary between 87 and 129 ppm whereas values in gabbro range from 38 to 58 ppm. Co shows a trend of decreasing values towards the base of both sills. Co varies with stratigraphic height almost exactly the same as Ni. 31* Height <m) vs. TiCH (%) - Lewis Intrusion 550 Height (m) vs. Cr (ppm) - Lewis Intrusion 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 9 1.0 1.1 TRMG TRPRD Tf02_% -200 200 600 1000140018002200 26003000 o TRMG ° TRPRD o CR PPM Height vs. Cu (ppm) -Lewis Intrusion 550 Height (m) vs. Co (ppm) - Lewis Intrusion 150 250 350 450 550 TRMG TRPRD CU PPM o TRMG o TRPRD 80 100 120 140 0 CO_PPM 40 Figure 3-4i: Plot of Stratigraphic Height versus Au in the Lewis Intrusion. Au values in peridotite vary between 2 and 31 ppb. Figure 3-4j: Plot of Stratigraphic Height versus Pd in the Lewis Intrusion. Pd values in peridotite vary between 2 and 18.8 ppb and decrease towards the base of both sills. One of the gabbro samples from the base of the lower sill had a Pd value of 20.6 ppb; the other gabbro samples have significantly lower Pd values than the peridotite. Figure 3-4k: Plot of Stratigraphic Height versus Pt in the Lewis Intrusion. Pt values in peridotite vary between 4.1 and 63 ppb and decrease towards the base of both sills in a similar trend to that of Cu and Pd. Figure 3-4I: Plot of Stratigraphic Height versus Pt/Pd in the Lewis Intrusion. Pt/Pd ratios in peridotite and gabbro vary from 0.4 to 4.4. The lower ratios occur at the base of both sills with low ratios also occurring near the top of the lower sill. Height (m) vs. Au (ppb) - Lewis Intrusion o o • o o o TRMG o TRPRD -6 0 5 10 15 20 25 30 36 0 A U - P P 8 Heigtrt (m) vs, Pd (ppb) • Lewfe Intrusion 550 350 fe 250 -50 10 14 18 22 o TRMG o TRPRD o PD PPB Height (m) vs. Pt (ppb) - Lewis Intrusion 10 20 30 40 50 60 70 o TRMG a TRPRD o PT PPB Height (m) vs. Pt/Pd - Lewis Intrusion 550 -50 TRMG TRPRD PT PD 41 Cu/Zr ratios are plotted versus height in Figure 3-4n. This measure of the variation in Cu with respect to fractionation varies with height in a similar pattern to Cu. This suggests that fractionation and Cu levels change in a parallel manner with respect to position in the sill. This does not prove that Cu abundances are controlled by fractionation. Cu levels are probably controlled more by S contents due to the chalcophile nature of Cu. Ni/Cu ratios in the Lewis Intrusion vary between 0.4 and 7.3. This large variation occurs within the peridotite as well as between the gabbro and peridotite. The variation in Ni/Cu versus height in the sill does not follow changing MgO levels showing that the ratio between Ni and Cu is not caused solely by changes in silicate mineralogy. The Ni/Cu ratio pattern changes with height in the opposite way to S. This trend demonstrates that at low S levels, silicate mineralogy (olivine abundance) has a large effect on Ni levels whereas S has a large effect on Cu levels. Ba is plotted versus height in Figure 3-4p. Ba values vary from 70 to 860 ppm in the Lewis Intrusion. The highest Ba value occurs in the gabbro at the base of the upper sill. This high Ba level may be due to contamination from Ba-rich Island Arc basement rocks. Ba does not show any systematic variation with height in the peridotite section of the sill. General Geochemical Variations Various high field strength, incompatible elements and oxides are plotted against each other (Ti0 2 vs. Zr, T i 0 2 vs. A l 2 0 3 i P 2 0 5 vs. A l 2 0 3 , Zr vs. A l 2 0 3 , Zr vs. P 2 0 5 , Ce vs. Yb, Zr vs. Yb, Zr vs. Y) in Figures 3-5a to 3-5h. Samples from both sections are plotted and the various rock types are represented by different symbols. Samples of Hasen Creek Formation sediment from the Spy Section which are derived from Island Arc volcanics are plotted for reference. Co-magmatic (co-genetic) rocks should plot along a single trend which passes through the origin of the graphs where both x and y axis elements are immobile (conserved). All of the above plots show a general co-magmatic trend; most points plot within 2 standard deviations of a trend formed by the magmatic samples, which passes through the origin. The standard deviation of repeated analyses of the Inco Bas standard at various labs is listed with the analytical data in Appendix 1. This is a measure of precision through many analytical batches at the same lab; this precision measurement is relevant here as the data analyzed in this study are from more than one analytical batch. This measurement of error is not total as it does not include sampling error (precision). Duplicate field samples were collected from the Spy Sill to determine the combined sampling and analytical error (precision); these results are listed in Appendix 1. 42 Figure 3-4m: Plot of Stratigraphic Height versus S in the Lewis Intrusion. S levels vary from detection limit to 0.12%. Cu, Pt and Pd have similar profiles to S, suggesting S levels have significant control on copper and PGE contents. Figure 3-4n: Plot of Stratigraphic Height versus Cu/Zr in the Lewis Intrusion. The Cu/Zr profile is similar to the profile for Cu suggesting that the fractionation of silicate minerals is a possible control on Cu abundance. The control of Cu abundance is probably more closely related to S contents due to the chalcophile properties of Cu. Figure 3-4o: Plot of Stratigraphic Height versus Ni/Cu in the Lewis Intrusion. Ni/Cu ratios vary between 0.4 and 7.3. Large variability in Ni/Cu ratios occurs within both the peridotite and gabbro units. The Ni/Cu ratios do not follow changes in MgO suggesting that at low sulphur levels, silicate mineralogy (modal olivine) has a large effect on Ni abundance whereas S has a large effect on Cu abundance. Figure 3-4p: Plot of Stratigraphic Height versus Ba in the Lewis Intrusion. Ba levels vary from 70 to 860 ppm. The highest Ba value occurs at the base of the upper sill. The elevated Ba in some samples may be the result of contamination of the sill from Ba-rich Island Arc country rocks. Height (m) vs. S (ppm) - Lewis Intrusion Height (m) vs. CuZr - Lewis Intrusion o TRMG -200 200 600 1000 1400 ° T R P R D 0 400 800 1200 1600 " S_PPM Height (m) vs. Ni/Cu - Lewis Intrusion Height (m) vs. Ba (ppm) - Lewis Intrusion TRMG TRPRD NI CU -100 100 300 500 700 900 1100 TRMG TRPRD BA PPM 43 Figures 3-5a and 3-5b: Plots of Ti0 2 versus Zr and Ti0 2 versus Al203- Spy and Lewis Intrusions. These plots show a co-magmatic trend for the Kluane-type intrusive units. The marginal gabbro samples show the greatest deviation from the co-magmatic trend. This is probably due to contamination by Hasen Creek Formation sediments which also do not plot along the co-magmatic trend. H3<K Ti02 (%) vs. Zr (ppm).- Spy and Lewis Intrusions 1.6 1.4 1.2 1. ' o. 0. 0. 0. 0. • to o 6 • • A i — 20 40 60 80 100 120 140 160 180 ZR PPM o ROCK_UNIT=TrPRD'" o. ROCK_UNIT='TrMG' o ROCK_UNIT=TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST Ti02 (%) vs. AI203 (%) - Spy and Lewis Intrusions 1.6 1.4 1.2 1.0 • 0.8 0.6 0.4 0.2 0.0 • • • i o o 0 • • Q | 0 11 • : • A • 0 0 0 12 16 20 o ROCK_UNIT=TrPRD' : n ROCK_UNIT='TrMG' o ROCK_U N IT-TrMCG' A ROCK_UNIT='PHCSHV '.• ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST A L 2 0 3 _ % 44 Figures 3-5c, 3-5d and 3-5e: Plots of P 2 0 5 versus A l 2 0 3 , Zr versus A l 2 0 3 and Zr versus P 2 0 5 - Spy and Lewis Intrusions. These plots show a co-magmatic trend for the Kluane-type intrusive units. The marginal gabbro samples show the greatest deviation from the co-magmatic trend. This is probably due to contamination by Hasen Creek Formation sediments which also do not plot along the co-magmatic trend. O 0.24 0.20 0.16 0.12 ' I I 0.08 0.04 0.00 P205 (%) vs. AI203 (%) - Spy and Lewis Intrusions C D D n a o O O . . . o o o ; rp <D o rr> o cao o o • 8 AL203 % 12 16 20 o ROCK UNIT= ='TrPRD' • ROCK UNIT= =TrMG' 0 ROCK UNIT= ='TrMCG' A ROCK UNIT= ='PHCSH' ROCK UNIT= ='PHCLST' a ROCK UNIT= ='PHCSLST' 180 160; 140 120 100 80 60 "40 20 0 Zr (ppm) vs. AI203 (%) - Spy and Lewis Intrusions iff. fa • 8 12 AL203 % 16 20 o ROCK UNIT= ='TrPRD' • ; ROCK UNIT= ='TrMG' O : ROCK UNIT= =TrMCG' A ROCK UNIT= ='PHCSH' • ROCK UNIT= ='PHCLST • ROCK UNIT= ='PHCSLST' 180 160 140 120 C L . 100 C L 80 r ct-N 60 40 20 0 0.00 Zr (ppm) vs. P205 (%)•- Spy and Lewis Intrusions p o O 0.04 0.08 0.12 ..•5 B r ? • 0.16 0.20 0.24 o ROCK UNIT= •TrPRD' • : ROCK UNIT= ='TrMG' o ROCK UNIT= ='TrMCG' A ROCK UNIT= ='PHCSH' e ROCK UNIT= :'PHCLST' • ROCK UNIT= ='PHCSLST' P205_% 45 Figures 3-5f, 3-5g and 3-5h: Plots of Ce versus Yb, Zr versus Yb and Zr versus Y - Spy and Lewis Intrusions. These plots show a co-magmatic trend for the Kluane-type intrusive units. Ce (ppm) vs. Yb (ppm) - Spy Sill 1.4 2.0 YB PPM 2.6 3.2 o ROCK_UNIT='TrPRD' •: • ROCK_UNIT=TrMG' o ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST' ; • ROCK UNIT='PHCSLST' Zr (ppm) vs. Yb (ppm) - Spy and Lewis Intrusions 1.4 2.0 YB PPM ' 2.6 3.2 o ROCKJJNIT-TrPRD' • R0CK_UNIT=TrMG' o R0CK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' .• ROCK_UNIT='PHCLST' • ROCK UNIT='PHCSLSTV 180 Zr (ppm) vs. Y (ppm) - Spy andLewis Intrusions 10 14 18 22 26 30 o ROCKJJNIT-TrPRD' a ROCK_UNIT='TrMG' o ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST'. Y_PPM 46 In the above plots (Figures 3-5a to 3-5h), marginal gabbro samples show the greatest deviation from the co-genetic trend. This is probably due to their increased interaction with country rocks compared to the peridotite phase of the sill. Some of the marginal gabbro samples tend to deviate from the co-genetic trend towards data points representing samples of Hasen Creek sedimentary rocks. These plots suggest that part of the marginal gabbro phase has been contaminated by Hasen Creek sedimentary material. Sulphur is plotted versus Ni, Cu, Pt and Pd in Figures 3-5i to 3-5p; separate figures are shown for each element with symbols representing rock type and section. Ni values within the peridotite samples, especially from the Spy section, show a large range with relatively no change in S content. Several marginal gabbro samples from the Spy Sill show no increase in Ni content despite large increases in S content. These marginal gabbro samples also show a negative relationship between S and Cu, Pt and Pd. Cu levels within the peridotite phases of both sections display a very weak increase in Cu with increasing S. Pt and Pd levels show a large variation over an absent to small variation in S content. Overall, marginal gabbro samples from the Spy Section show a markedly different trend in their Ni, Cu, Pt and Pd levels with respect to S than gabbros from the Lewis section and peridotites from both sections. Pt is plotted versus Ni and Cu in Figures 3-5q and 3-5r. Ni values do not appear to vary with Pt in either gabbro or peridotite samples at the low concentrations considered here. The peridotite samples display a relatively large variation in Pt values with little or no change in Ni concentrations. The peridotite samples form two different groups on the plot related to the low Ni levels in the Lewis Intrusion and relatively high Ni levels in the Spy Sill. Pt shows approximately the same range in concentration in both intrusions. Cu levels vary more closely with Pt; Cu values tend to increase with increasing Pt values. Pd is plotted versus Cu and Ni in Figures 3-5s and 3-5t. The Pd versus Cu plot shows two distinct trends, one with a very low slope (m < 0.05), and the other with a significantly higher slope (m > 0.25). The population with the low slope is made up of samples from the Lewis Intrusion while the higher slope population is comprised of Spy Sill samples. These trends may record Pd depletion relative to Cu in the Lewis Sill. Maier et al (1998) outline the use of Cu/Pd ratios as an exploration tool; they note that magmas with Cu/Pd ratios less than 6,500 should contain Pd-rich sulphide. Magmas with Cu/Pd ratios less than 6,500 have lost Pd due to sulphide segregation. The usefulness of this ratio is related to the fact that the partition coefficient (D s u i p h i d e / S i| i c a t e ) of Pd is much greater than Cu. Due to this, a segregating sulphide melt will have a low Cu/Pd ratio and the residual silicate magma will have a higher Cu/Pd ratio. This ratio indicates the presence of Pd-rich sulphide related to the Spy Sill and the depleted nature of the Lewis Intrusion. Cu/Pd ratios in peridotite in the Spy Sill have a range from 2,103 to 5,452 and an average of 4,173. Cu/Pd ratios in peridotite from the 47 Figures 3-5i and 3-5j: Plots of S versus Ni - Spy and Lewis Intrusions. Peridotite samples from the Spy Sill show a large range in Ni without a significant variation in S content. Several marginal gabbro samples from the Spy Sill show no increase in Ni despite large increases in S content. H7* S (ppm) vs. Ni (ppm) - Spy and Lewis Intrusions 14000 12000 10000 8000 2 Q- 6000 CL co' 4000 2000 0 O-0 0 o 0 o n n o o o O O QQO -2000 L — — ' — ' ° Intrusion-Spy" -200 0 200 400 600 800 1000 1200 1400 1600 D lntrusion='Lewis' NI PPM S (ppm) vs. Ni (ppm) - Spy and Lewis Intrusions 2 CL CL co1 14000 12000 10000 8000 6000 4000 2000 0 -2000 • • D • a o o o 0 o o o o ROCK_UNIT=TrPRD' a ROCK UNIT='TrMG' o ROCK UNIT=TrMCG' A ROCK_UNIT='PHCSH' ROCK UNIT='PHCLST -200 0 200 400 600 800 1000 1200 1400 1600 8 1 ROCKJJNIT='PHCSLST NI PPM 48 Figures 3-5k and 3-51: Plots of S versus Cu - Spy and Lewis Intrusions. Peridotite samples from both sections display a weak increase in Cu with increasing S. Some marginal gabbro samples from the Spy Sill display a negative relationship with S. S (ppm) vs. Cu (ppm) - Spy and Lewis Intrusions 14000 150 250 CU PPM 350 :: ° Intrusion-Spy' 450 550 D lntrusion='Lewis' S (ppm) vs. Cu (ppm) - Spy and Lewis Intrusions 14000 12000 10000 8000 6: 6000 LV •• I W 4000 2000 0 -2000 B d c B £&<P Q ° 0 coo o o ° o -50 50 150 250 350 450 550 o ROC K_U N IT-TrP R D' • ROCK_UNIT='TrMG' ; o ROCK_U N IT='TrMCG' : A ROCK_UNIT='PHCSHr; • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST CU_PPM 49 Figures 3-5m and 3-5n: Plots of S versus Pt - Spy and Lewis Intrusions. Pt values from peridotite samples from both intrusions show a large variation over an absent to small variation in S content. Some marginal gabbro samples from the Spy Sill display a negative relationship between Pt and S. S (ppm) vs. Pt (ppb) - Spy and Lewis Intrusions 14000 12000 10000 8000 EL 6000 o. . to 4000 2000 0 -2000 -10 o o O : (5b o • o o fi cPQ O • o lntrusion='Spy' 10 20 30 40 50 60 70 P Intrusion-Lewis' PT PPB S (ppm) vs. Pt (ppb) - Spy and Lewis Intrusions 2 0 . 0 . I w 14000 12000 10000 8000 6000 4000 2000 0 -2000 -10 1 3 [ 1 ! • • i • i j o °*P o° 1 g o o o 10 20 30 40 50 60 70 o ROCK_UNIT='TrPRD' • ROCK_UNIT='TrMG' o ROCK_UNIT=TrMCG' ' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST' • / R O C K UNIT='PHCSLST PT_PPB 50 Figures 3-5o and 3-5p: Plots of S versus Pd - Spy and Lewis Intrusions. Pd values from peridotite samples from both intrusions show a large variation over an absent to small variation in S content. As with Cu and Pt, some marginal gabbro samples from the Spy Sill display a negative relationship between Pd and S. So d S (ppm) vs. Pd (ppb) - Spy and Lewis Intrusions 14000 12000 10000 8000 o- 6000 Q_ co 4000 2000 -2000 ^10 u S o B..&. D D o oo 0 O O .10: 30 50 PD PPB 70 :•: ° lntrusion='Spy' : 90 D Intrusion-Lewis' 14000 12000 10000 8000 Q- 6000 o. i CO 4000 2000 -2000 S (ppm).vs. Pd (ppb) - Spy and Lewis Intrusions -10 6° o &<t> ° Q O CD o o OOo 10 30 50 PD PPB 70 90 o ROCK_UNIT='TrPRD' o ROCK_UNIT='TrMG' .,' o ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLSr • ROCK UNIT='PHCSLSr 51 Figure 3-5q: Plot of Pt versus Ni - Spy and Lewis Intrusions. Ni does correlate positively with Pt at the concentrations shown in this diagram. Peridotite samples display a large range in Pt with little or no change in Ni concentrations. Peridotite samples fall into two groups which have similar ranges of Pt; one group consists of samples with low Ni from the Lewis Intrusion and the other group with relatively enriched Ni is from the Spy Sill. Figure 3-5r: Plot of Pt versus Cu - Spy and Lewis Intrusions. Pt has a positive correlation with Cu. This trend of increasing Pt with increasing Cu is distinct from the relationship between Pt and Ni. Pt (ppb) vs: Ni (ppm) - Spy and Lewis Intrusions 200 400 600 800 1000 1200 1400 1600 NI PPM o ROCK_UNIT='TrPRD' • ROC K_U N IT='TrMG' o ROCK_U NIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST Pt (ppb) vs. Cu (ppm) - Spyiand Lewis Intrusions 70 60 50 40 & 30 20 10 0 -10 o ! O O o o Dn CP tP o o m -50 50 150 ; 250 350 450 CU PPM 550 o ROCK_UNIT='TrPRD': :f • ROCK_UNIT='TrMG' o ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST B ROCK UNIT='PHCSLSr: 5 2 Figure 3-5s: Plot of Pd versus Cu - Spy and Lewis Intrusions. This plot shows two distinct trends in peridotite samples, one with a slope (m<0.05) which consists of samples from the Lewis Intrusion, and one with a significantly higher slope (m>0.25) which consists of samples from the Spy Sill. These trends probably reflect a record of Pd depletion relative to Cu in the Lewis Intrusion. Cu/Pd ratios in the marginal gabbro are erratic and the fields of samples from the two intrusions overlap. Figure 3-5t: Plot of Pd versus Ni - Spy and Lewis Intrusions. This plot does not show a strong relationship between Pd and Ni. A group of samples from the Spy Sill forma separate field defined by Ni values in excess of 1000 ppm and a large range in Pd concentrations. A separate field is formed by samples from the Lewis Intrusion, which contain Ni values below 1000 ppm and Pd values less than 30 ppb. Lewis Intrusion range from 7,000 to 84,286 and average 32,175. The range of Cu/Pd ratios in the marginal gabbro from the two intrusions overlaps and is somewhat erratic. 5~2* Pd (ppb) vs. Cu (ppm) - Spy and Lewis Intrusions co o. o. I Q Q. 90 70 50 30 10 -10 o o o : O ; ° • 0 : 6 \ o . oL ° ° I O O O o O : C @§L CP o o ; O I ° -50 50 150 250 350 450 CU PPM 550 O : ROCKJJNIT-TrPRD' • ROCKJJNIT-TrMG' o ROCK_UNIT='TrMCG' : ' A ROCKJJNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST': Pd (ppb) vs. Ni (ppm) - Spy and Lewis Intrusions co o. a. Q' CL 90 70 50 30 10 -10 C D ° -200 0 200 400 600 800 1000-1200 1400 1600 o ROCK_UNIT='TrPRD' a ROCKJJNIT-TrMG' o ROCKJJNIT-TrMCG' A : ROCKJJNIT-PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST NI PPM 53 The Pd versus Ni plot shows no strong relationship between the two elements. A group of peridotite samples from the Spy Sill forms a separate field with Ni values greater than 1,000 ppm which have a large range in Pd concentrations. Peridotite samples from the Lewis Intrusion form a field with relatively lower Ni (<1,000 ppm) and Pd (<30 ppb) concentrations. Ni vs. MgO and Mg# plots for the Spy and Lewis intrusions are shown in Figures 3-5u to 3-5y. Figures 3-5u and 3-5v show symbols representing the various rock types while Figures 3-5x and 3-5y have symbols representing the intrusion the sample was collected from. These plots show that within the same range of MgO levels (25 to 35%), peridotites from the Spy Sill have much higher Ni levels (800 to 1400 ppm range) than peridotites from the Lewis Intrusion (350 to 800 ppm range). Marginal gabbro from both intrusions have similar Ni levels at a given MgO concentration. The Maple Creek gabbro samples from the Spy Sill plot very close to marginal gabbro samples from the Lewis Intrusion. A similar trend is evident in the Ni versus Mg# plots. The peridotite in the Lewis intrusion has a slightly lower average Mg# than the Spy Sill peridotite (Mg# 79 versus Mg# 82). This depletion in Ni noted in the Lewis Intrusion peridotites suggests that this magma has equilibrated with sulphide. A S versus Se plot is presented in Figure 3-5z for samples from the Spy Sill. A reasonably well-developed trend with a low slope (m=~ 0.225) is formed by the peridotite samples. The trend for marginal gabbro samples is less well defined, but has a higher slope indicating a higher S/Se ratio. A plot of 5 3 4S versus S/Se is shown for samples from the Spy Sill section in Figure 3-5aa. The majority of peridotite samples plot in the field defined by magmatic sulphides (S/Se<4,500 and 8 3 4S from -5 to +5 700). All the marginal gabbro samples plot outside this field; the low levels of sulphide in these samples shows contamination from non-magmatic sulphide. The sulphide in these marginal gabbro samples does not define any coherent mixing trend. The points representing Maple Creek Gabbro and Hasen Creek Formation shale and limestone samples in Figure 3-5aa have S values near the detection limit and thus these points are of questionable relevancy. REE spidergrams (samples normalized by chondrite values of Sun, 1982) of samples from the Spy Section are presented in Figures 3-5ab/' to 3-5abw. Overall, REE abundances appear to be controlled by the amount of plagioclase. Peridotite samples have the lowest abundances while the marginal gabbro has the highest. There are differences in the profiles of the various units which probably are related to their origin and contamination history. The Maple Creek 54 Figures 3-5u and 3-5v: Plots of Ni versus MgO and Ni versus Mg# - Spy and Lewis Intrusions. Over a similar range in MgO (25 to 35%) peridotites from the Spy Sill have much higher Ni (800 to 1400 ppm) than peridotites from the Lewis Intrusion (350 to 800 ppm). Marginal gabbro from both intrusions have similar Ni levels at various MgO concentrations. Maple Creek gabbro samples from the Spy Sill plot adjacent to samples from marginal gabbro from the Lewis Intrusion. The Ni versus Mg# plot shows similar relationships with the Lewis Intrusion peridotite having slightly lower average Mg# than the Spy Sill peridotite (Mg# 79 versus Mg# 82). The relative Ni depletion in the Lewis Intrusion suggests that this magma has equilibrated with sulphide. Ni (ppm) vs. MgO (%) - Spy and Lewis Intrusions .,, 1600 1400 1200 1000 5 800 O L • 0. -' 600 400 200 0 -200 o i o 0 8 o : p p i i i o j S ; O^J ; che A ! _ ! • • S C D L I !•• 10 15 20 25 30 35 40 MGO % o ROCK_UNIT='TrPRD' n ROCK_UNIT='TrMG' «.',• ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST 2 CL 0. 1600 1400 1200 1000 800 600 400 200 . 0 ,200 Ni (ppm)vs. Mg#- Spy and Lewis Intrusions • •DO. Q 40 50 i O ! o \ 8 p 60 70 MG 80 90 o ROCK_U N IT-TrPRD' • ROC K_U N IT='TrMG' o ROC K_U N IT-TrMCG' A ROCK_UNIT='PHCSH': • : ROCK_UNIT='PHCLST' • ROCK UNIT='PHCSLST 55 Figures 3-5x and 3-5y: Plots of Ni versus MgO and Ni versus Mg# - Spy and Lewis Intrusions. Plots of Ni versus MgO and Ni versus Mg# - Spy and Lewis Intrusions. Over a similar range in MgO (25 to 35%) peridotites from the Spy Sill have much higher Ni (800 to 1400 ppm) than peridotites from the Lewis Intrusion (350 to 800 ppm). Marginal gabbro from both intrusions has similar Ni levels at various MgO concentrations. Maple Creek gabbro samples from the Spy Sill plot adjacent to samples from marginal gabbro from the Lewis Intrusion. The Ni versus Mg# plot shows similar relationships with the Lewis Intrusion peridotite having slightly lower average Mg# than the Spy Sill peridotite (Mg# 79 versus Mg# 82). The relative Ni depletion in the Lewis Intrusion suggests that this magma has equilibrated with sulphide. Ni (ppm) vs. MgO (%) - Spy and Lewis Intrusions 1600 1400 1200 1000 5 800 CL Q L -' 600 400 200 0 -200 o o o _ 0 8 o o % o o [ 3 a ° y D • o 6 -5 o Intrusion-Spy' 10 15 20 25 30 35 40 ° lntrusion='Lewis' MGO % Ni (ppm) vs. Mg# - Spy and Lewis Intrusions 1600 1400 1200 1000 5 800 o_ o. -' 600 400 200 0 -200 40 50 60 o o o % 8 c q • J oh • • "oo o o Q c 70 80 o lntrusion='Spy' gg D Intrusion-Lewis' MG 56 Figure 3-5z: Plot of S versus Se - Spy Sill. Peridotite samples form a trend with a low slope (m~0.225). Marginal gabbro samples form a less well-defined trend with a higher slope indicating a higher S/Se ratio. Figure 3-5aa: Plot of 534S versus S/Se - Spy Sill. Most peridotite samples plot in the field for magmatic (mantle) derived sulphur - S/Se<4500 and MS between -5 and +5 7^ . Marginal gabbro samples plot outside this field suggesting contamination from country-rock sulphur. 1.4 S (%) vs. Se (ppm) - Spy Sill 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 bo o o -0.2 0.2 0.6 1.0 1.4 SE PPM 1.8 2.2 2.6 Q ROCKJJNIT-TrPRD' • ROC KJJ N IT='TrMG' o ROCKJJNIT='TrMCG" A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST Del 34S vs. S/Se - Spy Sill 15 10 -20 -2000 oo D D \ • 2000 6000 10000 14000 4000 8000 12000 16000 o ROCK_UNIT='TrPRD' • ROCK_UNIT='TrMG' : o ROCK_UNIT='TrMCG' A ROCK_UNIT='PHCSH' • ROCK_UNIT='PHCLST • ROCK UNIT='PHCSLST S S E 57 Figures 3-5ab/-w: REE Spidergrams for units from the Spy Sill: Upper Marginal gabbro, Maple Creek gabbro, Lower Marginal gabbro, Peridotite, Hasen Creek Formation siltstone. REE concentrations are normalized by values from Sun (1982). Overall REE abundances appear to be controlled by the modal amount of plagioclase. Peridotite samples have the lowest abundances while marginal gabbro has the highest. Maple Creek gabbro and peridotite samples have relatively flat profiles with LREE having a similar slope to HREE similar to MORB and OIB. Marginal gabbro samples have a steeper slope in LREE, typical of Island Arc magmas whereas the HREE have a slope similar to those from peridotite and Maple Creek gabbro samples. The REE profile for from siltstone at the base of the sill is similar to the marginal gabbro profile suggesting contamination of the magma by Hasen Creek Formation sediment. S7« REE Spldergrtm - Upper Marginal GetXfO REE Spidergram Maple Crae* Gabbro REE Spidergram S JtWooe at base of Spy Sii REE Spidergram Lower Margtnej Gabbro 58 gabbro and peridotite samples have relatively flat profiles with light REE (LREE) part of the spidergram having a similar slope to the heavy REE (HREE) part. This type of pattern is similar to those of MORB and Oceanic Island Basalts (Wood et al, 1979). The marginal gabbro samples have steeper slopes in the LREE part of the profile (as reflected in the Ce/Yb ratio) than the peridotite and Maple Creek Gabbro samples. A steep slope in the LREE part of the profile is typical of Island Arc volcanics (Wood et al, 1979). As discussed in the previous section, the marginal gabbro samples have Ce/Yb N ratios in the 3 to 4 range, whereas peridotite samples have lower ratios. As with the peridotite samples, the profile in the HREE part of the spidergrams is relatively flat for the marginal gabbro samples. The sample of siltstone from the footwall of the Spy Sill has a very similar REE profile to the marginal gabbro samples. Since the siltstone is derived from Island Arc volcanics, this suggests that the marginal gabbros have picked up some of the geochemical signature of the country rocks through contamination or that the marginal gabbros have an Island Arc affinity. The former theory seems most plausible as the peridotites in the core of the sill have different REE profiles than the marginal gabbros. 3.5 Olivine Geochemistry 3.5.1 Methods Electron microprobe analyses of olivine grains from peridotites of the Spy and Lewis intrusions were done at the University of British Columbia on a Cameca SX-50 probe. Operating parameters were 15 kV and 20 nA; a count time of 100 seconds was used. Each grain was probed with two points in the core and two points near the rim. 3.5.2 Results Olivine microprobe analyses are listed in Appendix 2. Only mineralogically reasonable results are listed. The occasional bad analysis is believed to be the result of inclusions, alteration or grain edge effects. Average core and rim values were calculated for MgO, FeO and NiO for each grain. Average rim versus core values for MgO, FeO and NiO were plotted in Figures 3-6a to 3-6c. Coefficients of variation (CV) were calculated using the average core and rim compositions; these values are listed in Appendix 2. Overall, there does not appear to be a significant difference between the core and rim compositions as shown in Figures 3-6a to 3-6c. The largest variation between rim and core compositions occurs with NiO. The largest CV for MgO is 3.352% while the largest CV for FeO is 4.301%; the largest CV for NiO is 61.481%. 59 Figure 3-6a: Average MgO-Rim versus Average MgO-Core from olivine microprobe data - Spy and Lewis Intrusions. There does not appear to be a significant difference between core and rim compositions. Figure 3-6b: Plot of Average FeO-Rim versus Average FeO-Core from olivine microprobe data - Spy and Lewis Intrusions. There does not appear to be a significant difference between core and rim compositions. Spy and Lewis Intrusions - Olivine E a. o > < 45.000 44.500 44.000 43.500 43.000 42.500 42.000 41.500 41.000 40.500 J. 40.500 41.000 41.500 42.000 42.500 43.000 Avg. MgO Core 43.500 44.000 44.500 45.000 Spy and Lewis Intrusions - Olivine 25.000 20.000 E 15.000 K O a la. 310.000 5.000 0.000 0.000 5.000 10.000 15.000 Avg. FeO Core 20.000 25.000 60 Figure 3-6c: Plot of Average NiO-Rim versus Average NiO-Core from olivine microprobe data - Spy and Lewis Intrusions. There does not appear to be a significant difference between core and rim compositions. Figure 3-6d: Plot of Ni versus Fo# from olivine microprobe data - Spy and Lewis Intrusions. Average (of core and rim) Fo# and Ni for each grain are shown on diagram. Grains from the Spy Sill have Ni values between 0.15 and 0.21% and Fo# between Fo79.8 and Fo83.2. Grains from the Lewis Intrusion have Ni values between 0.05 and 0.11% and Fo# between Fo78 and Fo08.7. All grains from the Lewis intrusion have compositions that fall below the reference line of Simkin and Smith (1970) which represents the lower limit of olivine compositions from typical layered intrusions. Spy and Lewis Intrusions 0.450 0.400 0.350 0.300 0.250 0.200 ^ 0.150 0.100 0.050 0.000 0.000 0.050 0.100 0.150 Avg. NIO Core 0.200 0.250 0.300 Spy and Lewis Intrusions - Olivine 0.25 0.15 0.1 0.05 78 L o w e r limit o f o l i v ine f ie ld in typica l l a ye red in t rus ions (S imk in a n d S m i t h , 1970) . 79 80 81 Fo# 82 83 84 «.Ni% Spy • Ni% Lewis 61 The average Ni and forsterite number (Fo#) of each grains (including both core and rim values) are plotted in Figure 3-6d. Olivine grains from the Spy and Lewis Intrusions plot in distinct fields. Grains from the Spy Sill have Ni values between 0.15 and 0.21% whereas grains from the Lewis Intrusion vary between 0.05 and 0.11%. Forsterite content of the olivines in the Spy Sill range from Fo79 .8 to Fo83.2. In the Lewis intrusion, olivine compositions range from Fo78 to Fo80.7. The lower limit of the field of olivine compositions from typical layered intrusions from Simkin and Smith (1970) is shown for reference in Figure 3-6d. All grains from the Lewis Intrusion fall below this line as well as some grains from the Spy Sill. This suggests that the "depleted" olivine grains have lost Ni at some point in their emplacement history and that equilibration with sulphide has taken place. The samples containing the lowest Ni contents in the Spy Sill are from the middle of the sill. Olivine data from Hulbert (1995) for other intrusions in the Kluane Belt show that the Quill Creek Complex at Wellgreen has olivine compositions (in terms of Fo# and Ni) which overlap those from both the Spy and Lewis sills. Limited olivine data from Miller (1991) for the Quill Creek Complex partially overlap the field from the Spy Sill. Hulbert found that the White River Complex at Canalask had the highest Fo contents (82.5 to 88.5) and the highest Ni abundances (1000 to 4000 ppm) in the belt; limited olivine data from Campbell (1981) show similar Fo levels. The distribution coefficient (KD) for Fe and Mg between olivine and basaltic melt was established by Roeder and Emslie (1970). This relationship is relatively independent of temperature. The distribution coefficient is described in molar terms by the equation below. K D = (Mg/Fe2+)me„ x (Fe 27Mg) o l i v i n e = 0.30 +/- 0.03 In order the determine the most primitive possible composition of melt that was parental to the sills, olivine compositions from grains with the maximum Fo# were used from each of the Spy (Fo83.2) and Lewis (F08O.68) sills to calculate molar MgO/FeO ratios (Spy=1.4891, Lewis=1.2431). These ratios correspond to maximum parental melt Mg numbers of 55.42 for the Lewis Sill and 59.82 for the Spy Sill. The marginal gabbro from the Spy Sill has Mg#s ranging from 46.53 to 69.37. The sample with the Mg# of 69.37 was from a thin transitional phase between the peridotite and the marginal gabbro along the base of the sill. All other marginal gabbro samples from the Spy Sill have Mg#s which fall into a range from 46.53 to 51.99. Thus, the composition of the marginal gabbro from the Spy Sill is not consistent with the parental melt. This may be due to contamination of the marginal gabbro with Island Arc derived sediments of the Hasen Creek Formation. The marginal gabbro from the Lewis Intrusion has Mg#s ranging from 54.10 to 74.29. The sample with the Mg# of 54.10 is from the base of the lower sill and may be more reflective of the parental melt than the 62 marginal gabbro from the base of the upper sill with Mg# of 74.29. The Mg# of the lower marginal gabbro (54.10) closely matches the parental melt composition predicted by the olivine composition (55.42). The composition of the Maple Creek Gabbro at the base of the Spy Sill is significantly more magnesian (Mg# 70.06-72.7) than the parental melt composition predicted by the olivine compositions (Mg# 59.82). The Maple Creek Gabbro here may not be representative of typical Maple Creek Gabbro. Hulbert (1995) states the average Mg# of the Maple Creek Gabbro from his study was 62.1. Exploration data from other Maple Creek gabbros on the Klu property also have an average composition which is less magnesian (Bell, 1996). The parental melt composition predicted from olivine composition will be compared to the Nikolai basalt composition in a following section. 63 Section 4: Petrology and Geochemistry of the Nikolai Basalt 4.1 Introduction Two basalt sections were sampled near the Klu property along Burwash Creek and Halfbreed Creek in order to study relatively complete sections. No complete sections in areas which were safe to sample are present on the Klu property. Locations of the Burwash Creek and Halfbreed Creek sections are shown in Figure 2-3. The stratigraphic heights of samples from the two sections are listed in Appendix 3, which contains the analytical results. The Burwash section has a stratigraphic thickness of approximately 1,000 m whereas the Halfbreed section is approximately 500 m thick. 4.2 Lithological Description and Petrography The Nikolai basalt has a stratigraphic thickness up to 1,000 m in southwest Yukon. The age of the basalt is constrained to Middle or Upper Triassic by fossils in overlying and underlying formations (Read and Monger, 1976). The lower 100 m of the Nikolai basalt locally consists of a volcanic breccia and pillow lava (Read and Monger, 1976). The subaqueous base of the Nikolai basalt is overlain by volcanic material deposited in subaerial conditions. The bulk of the formation contains poorly defined flows which have amygdaloidal to vesicular tops and bases. Flows range in thickness from 2 to 15 m. The amygdules consist of chlorite, epidote, calcite, prehnite and quartz. Amygdules/vesicules range from 1 mm to 2 cm and form up to 25% of the rock. The interior of flows are massive to porphyritic; phenocrysts are dominantly plagioclase and locally clinopyroxene, and can form up to 30% of the rock. Campbell (1981) noted rare serpentine pseudomorphs after olivine; no serpentine pseudomorhs were noted in the samples from the Burwash and Halfbreed sections. The groundmass consists of plagioclase, clinopyroxene, opaques and glassy material. The basalt has been regionally metamorphosed to prehnite-pumpellyite facies resulting in chlorite, epidote, prehnite and calcite mineralogy. Outcrops are typically rubbly weathering and maroon coloured due to the presence of hematite. 4.3 Geochemistry (Majors, Traces, REE, Ni-Cu-Co) 4.3.1 Methods Approximately 3 kg of material was collected from each of 26 sample sites and an attempt was made to chip off or avoid weathering rinds. An Inco standard was submitted for analysis along with the basalt samples. All samples were 64 prepared at Chemex Labs in North Vancouver, B.C.. Samples were crushed and then pulverized in a ring pulverizer to -150 mesh. Samples from the Spy Section were analyzed for whole rock (Al 2 0 3 , CaO, C r 2 0 3 , F e 2 0 3 , K 2 0 , MgO, MnO, Na 2 0, P 2 0 5 , S i0 2 , T i 0 2 and LOI) by Chemex Labs of North Vancouver, B.C.. Whole rock determinations were made by ICP. Ba was analyzed by ICP after a lithium metaborate fusion. Trace element (Nb, Zr, Y, Sr and Rb) analyses by pressed pellet XRF analyses were done at the Inco Exploration Lab in Copper Cliff, Ontario. ICP analyses for metals and certain trace elements (Nb, Rb, Sr, Y, Zr, Ag, Co, Cu, Fe, Mn, Mo, Ni, Pb and Zn) were carried out by Chemex Labs. Samples were digested with Aqua Regia resulting in a partial digestion. Ni was also analyzed by Atomic Absorption (AA) after a total digestion by a perchloric-nitric-HF digestion. Rare Earth Element (REE -La, Ce, Nd, Sm, Eu, Tb, Yb and Lu) plus Th and Ta analyses were completed by Instrumental Neutron Activation Analysis (INAA) at Actlabs in Ancaster, Ontario. U was analyzed at Actlabs by DNC. An analysis of error (precision and accuracy) is included with the analytical data in Appendix 3. 4.3.2 Results Various elements, oxides and ratios are plotted versus stratigraphic height for the basalt sections are presented in Figures 4-1 a to 4-1 o. General geochemical variations between elements are shown in Figures 4-2a to 4-2z. REE spidergrams are presented in Figure 4-3a and 4-3b. An extended element profile is shown in Figure 4-4. These diagrams are shown to study cogenetic trends, contamination and metal depletion. Ternary diagrams analyzing magma-types are shown in Figures 4-5 and 4-6. Geochemical Variations with Stratigraphic Height Stratigraphic height versus Cr, Cu, MgO, Ni, Cu/Zr, Mg#, Ni/Cu, Ni/MgO, Ce/Yb, Co and T i 0 2 is plotted in Figures 4-1 a to 4-1 k. Samples from the Halfbreed Creek and Burwash Creek sections are shown on each plot. Cr values vary from 130 to 330 ppm and generally decrease towards the top with the exception of the sample at the base of the Burwash section. This Ca-rich sample is highly altered/contaminated and occurs directly above the basal contact and will not generally be considered in the interpretation of these diagrams. MgO levels in most samples varies between 6 and 1 0 % and show no obvious systematic variation with stratigraphic height. Mg# displays a similar pattern versus height with Burwash section having an average of 57.9 and the Halfbreed section having an average of 55.3. These average Mg numbers match closely with the Mg number of the melt predicted by the compositions of olivine in peridotite from the Spy and Lewis intrusions (59.82 and 55.42). The average Mg numbers of 65 Figure 4-1 a: Plot of Stratigraphic Height versus Cr - Nikolai Basalt. Cr values vary from 130 to 330 ppm and generally decrease towards the top with the exception of the sample at the base of the Burwash section. Figure 4-1 b: Plot of Stratigraphic Height versus Cu - Nikolai Basalt. Cu values in the Halfbreed section vary from 27 to 343 ppm with average of 108 ppm. The higher Cu values occurring near the base of the section. Cu values in the Burwash section vary from 20 to 474 ppm with an average of 81 ppm. No systematic variation in Cu with stratigraphic height is apparent in the Burwash section. Figure 4-1 c: Plot of Stratigraphic Height versus MgO - Nikolai Basalt. MgO levels in most samples varies between 6 and 1 0 % and show no obvious systematic variation with stratigraphic height. Figure 4-1d: Plot of Stratigraphic Height versus Ni - Nikolai Basalt. Ni values do not vary systematically with stratigraphic height in either section. The Ni levels vary over a relatively narrow range with the average value in the Burwash section being 74 ppm and the average value in the Halfbreed section being 82 ppm. Height (m) vs. Cr (ppm) - Nikolai Basalt 1200 1000 x 2 600 400 200 •200 o location*'Burwash-20 80 140 200 260 320 380 ° locatiorp'Haltbree' Height (m) vs. Cu (ppm) - Nikolai Basalt 1200 r 1000 800 600 X a Ui X 200 0 location='Burwash' -50 50 150 250 350 450 550 ° loeation='•Hallb^ee• CU Height (m) vs. MgO% - Nikolai Basalt 1200 r — 1000 800 600 x a I 2 3 4 5 6 7 MGO 0 locatk>n='Burwash' 8 9 10 11 D tocatlon-'Halfbree' Height (m) vs. Total Ni (ppm). Nikolai Basalt 1200 | 800 200 -200 ° tocaliorp'BurwasrT -20 0 20 40 60 80 100120140 0 tocatiorv'Halfbree' NLTOTAL 66 Figure 4-1 e: Plot of Stratigraphic Height versus Cu/Zr - Nikolai Basalt. The Cu/Zr profile is considerably smoother than the Cu versus height profile suggesting some of the variation in the Cu versus height profile is due to fractionation. Figure 4-1f: Plot of Stratigraphic Height versus Mg# - Nikolai Basalt. Mg# displays a similar pattern versus height with Burwash section having an average of 57.9 and the Halfbreed section having an average of 55.3. These average Mg numbers match closely with the Mg number of the melt predicted by the compositions of olivine in peridotite from the Spy and Lewis intrusions (59.82 and 55.42). Figure 4-1g: Plot of Stratigraphic Height versus Ni/Cu - Nikolai Basalt.. Ni/Cu ratios show a steadily increasing trend from the base of both sections to approximately 200 m in height. Above that height, the profiles are somewhat chaotic. The Ni/Cu ratios do not vary systematically with MgO levels. Figure 4-1 h: Plot of Stratigraphic Height versus Ni/MgO - Nikolai Basalt. Ni/MgO ratios versus height vary closely with Ni suggesting that Ni levels may be related to accessory olivine in the basalt. Height (m) vs. Cu/Zr - Nikolai Basalt 1200 1000 I a 600 400 200 o o 0 o 0 o o • o a a 0 a • a • • oo • • Height (m) vs. Mg# - Nikolai Basalt 1200 r -2 0 2 4 6 CU ZR o locatton='Burwash' S 10 12 D locationa'Halfbree' 800 600 o tocatton='BurwasrT 32 38 44 50 56 62 68 ° localiari«>tallbfe»; MG Height (m) vs. Ni'Cu - Nikolai Basalt 1200 1000 t— x o 800 600 400 200 -200 -0.5 0£ 15 2S 3.5 ° looation«'Burwash' 0.0 1.0 2.0 3.0 4.0 0 loeatiorv'Halrbree' NI CU Height (m) vs. Ni/MgO - Nikolai Basalt 1200 600 400 200 -200 ro • O D o loca1k>n*'BurwaslV .2 2 6 10 14 18 22 ° looation='Halfbree' NI MGO 67 Figures 4-1 i and 4-1 k: Plots of Stratigraphic Height versus Ce/Yb and T i 0 2 -Nikolai Basalt. Ce/Yb ratios and T i 0 2 display similar trends with stratigraphic height. The entire Halfbreed section and the lower 300 m of the Burwash section have Ce/Yb ratios and T i 0 2 in the 6.5 to 9.5 and 1.6 to 2.5% range respectively. Above this stratigraphic level, the Ce/Yb ratios and T i 0 2 decrease towards the top of the Burwash section. Figure 4-1j: Plot of Stratigraphic Height versus Co - Nikolai Basalt. Co does not vary systematically with height. The two sections have very similar average Co values (Halfbreed= 38 ppm, Burwash= 39 ppm). .7A Height (m) vs. Ce/Yb - Nikolai Basalt 1200 1000 x 400 200 0 • • o • o locatlon-'Bumash' 2 3 4 5 6 7 8 9 10 a locattorp'Haltbree' CE YB Height (m) vs. Co (ppm) - Nikolai Basalt 1200 800 600 t-x O m r 200 0 3 0 o o 0 0 o • • 0 a o • • • o • • • DO • • o 16 20 24 28 32 36 40 44 CO ° tocatforF'Bufwash' o location='HaHbfee' Height (m) vs. Ti02% - Nikolai Basalt 1200 r 1000 400 200 -200 o location-'Burwash' 0.2 0.6 1.0 1.4 1.8 2.2 2.6 D locatk>n»'Har(bree' TI02 6 8 the basalts also occur roughly in the middle of the Mg# range in the marginal gabbro from Klu (46.52 to 69.37). Cu values in the Halfbreed section vary from 27 to 343 ppm with average of 108 ppm. The higher Cu values occurring near the base of the section. Cu values in the Burwash section vary from 20 to 474 ppm with an average of 81 ppm. No systematic variation in Cu with stratigraphic height is apparent in the Burwash section. Ni values do not vary systematically with stratigraphic height in either section. The Ni levels vary over a relatively narrow range with the average value in the Burwash section being 74 ppm and the average value in the Halfbreed section being 82 ppm. Ni/MgO ratios versus height vary closely with Ni suggesting that Ni levels may be related to accessory olivine in the basalt. Cu/Zr ratios were plotted to normalize Cu values by a measure of fractionation. The Cu/Zr profile is considerably smoother than the Cu versus height profile suggesting some of the variation in the Cu versus height profile is due to fractionation. Ni/Cu ratios show a steadily increasing trend from the base of both sections to approximately 200 m in height. Above that height, the profiles are somewhat chaotic. The Ni/Cu ratios do not vary systematically with MgO levels. Co also does not vary systematically with height. The two sections have very similar average Co values (Halfbreed= 38 ppm, Burwash= 39 ppm). Ce/Yb ratios and Ti02 display similar trends with stratigraphic height. The entire Halfbreed section and the lower 300 m of the Burwash section have Ce/Yb ratios and Ti02 in the 6.5 to 9.5 and 1.6 to 2.5% range respectively. Above this stratigraphic level, the Ce/Yb ratios and Ti02 decrease towards the top of the Burwash section. This trend reflects the more primitive nature of the basalt in the top 700 m of the Burwash section relative to the Halfbreed section and the lower part of the Burwash section. Ba, Ba/Zr, Th and Th/Ta are plotted versus height in Figures 4-11 to 4-1 o. These diagrams are intended to assess the degree of contamination in the basalt. The base of both sections have low Ba levels. The level of Ba increases erratically up-section in both sections; the maximum Ba level in the Burwash section is 270 ppm which occurs at a stratigraphic height of 662 m. Ba is a possible contaminant from sediments within the Hasen Creek Formation. When Ba is normalized by Zr to account for fractionation, the resulting profile is smoother with Ba/Zr ratios varying from near 0 at the base of the section to approximately 2 at the top of the Burwash Section. The highest Ba/Zr ratios, possibly representing the greatest contamination, occur in the upper part of the Burwash section at 585 and 662 m (4.7 and 7.3 Ba/Zr respectively). Th values do not increase systematically with stratigraphic height. Trace levels of Th generally decrease with increasing height in the first 400 m of both sections. In order to determine if Th levels are related to fractionation, a Th/Ta ratio was plotted versus height. This diagram shows consistently low ratios (below 1.5) for the entire Halfbreed section and the lower 500 m of the Burwash section. The upper 500 m of the Burwash section has Th/Ta ratios between 1.5 and 105. Thus, the 69 Figure 4-11: Plot of Stratigraphic Height versus Ba - Nikolai Basalt. The base of both sections have low Ba levels. The level of Ba increases erratically up-section in both sections; the maximum Ba level in the Burwash section is 270 ppm which occurs at a stratigraphic height of 662 m. Figure 4-1 m: Plot of Stratigraphic Height versus Ba/Zr - Nikolai Basalt. Ba/Zr ratios vary from near 0 at the base of the section to approximately 2 at the top of the Burwash Section. The highest Ba/Zr ratios, possibly representing the greatest contamination, occur in the upper part of the Burwash section at 585 and 662 m (4.7 and 7.3 Ba/Zr respectively). Figure 4-1 n: Plot of Stratigraphic Height versus Th - Nikolai Basalt. Th values do not increase systematically with stratigraphic height. Trace levels of Th generally decrease with increasing height in the first 400 m of both sections. Figure 4-1 o: Plot of Stratigraphic Height versus Th/Ta - Nikolai Basalt. This diagram shows consistently low ratios (below 1.5) for the entire Halfbreed section and the lower 500 m of the Burwash section. The upper 500 m of the Burwash section has Th/Ta ratios between 1.5 and 105. Thus, the upper part of the Burwash section displays some evidence of contamination in terms of trace element ratios. a Height (m) vs. Ba (ppm) - Nikolai Basalt 1200 i i — -20 20 60 100140180220260300 BA location='Burwash' location='Halfrjree' Height (m) vs. Ba/Zr - Nikolai Basalt 1200 1000 600 600 200 e o 0 0 o o o o • • • a DO • • a a D 0 -1 0 1 2 3 4 5 6 7 8 BA ZR location=Burwash' focation='Haffbreo" Height (m) vs. Th (ppm) - Nikolai Basalt 1200 r ° location='BurwasrT 0.0 0.2 0.4 0.6 0.8 1 0 D tocation»'HaHbree' TH Height (m) vs. Th/Ta - Nikolai Basalt 1200 r 1000 -200 8888 88SE3B ° tocation-'Burwash' location='Halfbree' TH TA 7 0 upper part of the Burwash section displays some evidence of contamination in terms of trace element ratios. This part of the sections does not display any marked metal depletion compared to the lower part of the section or the Halfbreed section. Therefore, it is difficult to link such contamination with the presence of sulphide showings in intrusions related to the basalts. General Geochemical Variations Immobile trace and major elements were plotted against each other to determine if the Nikolai basalt samples within and between the sections sampled are oogenetic and to test for the presence of a conserved element. Zr vs. P 20 5, Zr vs. Y, Zr vs. Yb, Ti0 2 vs. Al 20 3, Ti0 2 vs. Zr, Zr vs. Al 20 3, Ce vs. Yb and Ti0 2 vs. P 20 5 are shown in Figures 4-2a to 4-2g. Clearly, many of the major and traces are mobile under the metamorphic and hydrothermal conditions to which the Nikolai basalt has been subject to. However, the Ti0 2 vs. Zr and Zr vs. Y plots suggest that the basalt samples are oogenetic and that certain elements are conserved. Ni, Cu and Co are plotted against MgO and Mg# to evaluate the behavior of metals with how magnesium-rich or mafic the basalts are. These diagrams (shown in Figures 4-2i to 4-2n) can be used to detect metal depletion as Ni, Cu and Co trend to concentrate in mafic minerals (olivine and pyroxene) in the absence of sulphide. Ni, Cu and Co should show a trend of increasing concentration with increasing MgO and Mg# unless the magma has come in to contact with significant amounts of sulphide. Basalts with depleted metal abundances may indicate that the magma equilibrated with sulphide and that the metals may be concentrated in nearby Ni-Cu-Co-PGE deposits. The tendency for certain metals to accumulate in sulphide melts relative to silicate melt can be described by a metal's partition coefficient or Dsulphlide/Sil icaite (Naldrett, 1989). The Dsuiphide/siiicate values for Ni and Cu are 200-500 and 200-1400 respectively for basaltic magmas (Francis, 1994). In comparison, PGE are even more chalcophile with D S U | P n i d e / Sj| i c a t e values in the order of 100,000 for basaltic magmas (Maieretal, 1998). Ni, Cu and Co levels in the Nikolai basalt do not increase with increasing MgO or Mg#. Also, there do not appear to be significant differences in metal concentrations between the two sections sampled. The majority of samples from both sections have Ni values between 60 and 120 ppm over a range in Mg# from 50 to 63 (6 to 10% MgO). Similar trends are displayed by Co and Cu relative to MgO and Mg#. Metal depletion has been documented in certain basalt flows in the Noril'sk area of Siberia, Russia which are oogenetic with gabbroic intrusions which host world-class Ni-Cu-PGE deposits (Naldrett et al 1992, Lightfoot et al, 1994 and Wooden et al, 1993). At Noril'sk, a contaminated and depleted flow is proposed to be 7 1 Figure 4-2a: Plot of Zr versus P205 - Nikolai Basalt. Based on these elements, a oogenetic trend is not observed in samples from the Burwash section. Figure 4-2b: Plot of Zr versus Y - Nikolai Basalt. These elements have a strong positive correlation, but do not display a strict oogenetic trend. Figure 4-2c: Plot of Zr versus Yb - Nikolai Basalt. These elements do not display a oogenetic trend. 7/« Zr (ppm) vs. P205 - Nikolai Basalt 160 140 120 100 80 60 40 20 0.02 • o n • o o o ° • ^ a • <& a a o o 0 0 8 0 0.04 0.06 0.08 0.10 0.12 P205 0.14 0.16 0.18 0.20 ° location-Burwash' a location-Halfbree' 160 140 120 100 a, N 80 60 40 20 14 Zr (ppm) vs. Y (ppm) - Nikolai Basalt 8 p Q ° ; • o o o o o o ( o 3 ; 18 22 26 30 34 38 42 location-Burwash' location-Halfbree' Zr (ppm) vs. Yb (ppm) - Nikolai Basalt or location-Burwash' location-Halfbree' 72 Figure 4-2d: Plot of Ti0 2 versus A l 2 0 3 - Nikolai Basalt. These elements do not display a oogenetic trend. Figure 4-2e: Plot of Ti0 2 versus Zr - Nikolai Basalt. These elements display a reasonably well defined cogenetic trend showing that the two separate basalt sections are co-magmatic. Figure 4-2d: Plot of Zr versus A l 2 0 3 -display a cogenetic trend. Nikolai Basalt. These elements do not 7Z+ Ti02% vs. AI203% - Nikolai Basalt 2.6 2.2 1.8 S 1.4 1.0 0.6 0.2 12 13 • • • n • a • 6? _ o o 0 o ° 0 o 0 14 15 16 AL203 17 18 19 20 location-Burwash' location-Halfbree' 2.6 2.2 1.8 S 14 t-1.0 0.6 0.2 20 40 Ti02% vs. Zr (ppm) - Nikolai Basalt • 0 o° °a a J o o ° ° o o 0 o ^ 0 60 80 100 120 ZR 140 160 location-Burwash' location-Halfbree' Zr (ppm) vs. AI203% - Nikolai Basalt 160 140 120 100 N 80 60 40 20 12 13 O Qj D O • o o o • ^ B o o 0 • o o o 14 15 16 AL203 17 18 19 20 o location-Burwash' o location='Halfbree' 73 Figure 4-2g: Plot of Ce versus Yb - Nikolai Basalt. These elements show positive correlation, but do not display a cogenetic trend. Figure 4-2h: Plot of Ti02 versus P 2 0 5 display a cogenetic trend. - Nikolai Basalt. These elements do not 73c 22 18 14 LU o 10 Ce (ppm) vs. Yb (ppm) - Nikolai Basalt • o • a • • o o a o o • o • o 1.0 1.2 1.4 1.6 1.8 YB 2.0 o location='Burwash' 2 2 24 2 6 D location-Halfbree' Ti02% vs. P205% - Nikolai Basalt 2.6 2.2 1.8 8 1-4 1.0 o 0.6 0.2 • D o • Gh 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 o location-Burwash' a location-Halfbree' P2Q5 74 Figures 4-2i, 4-2j and 4-2k: Plots of Cu versus MgO, Ni versus Mg# and Ni versus MgO - Nikolai Basalt. Plots of Cu versus Mg#, Co versus MgO and Co versus Mg# - Nikolai Basalt. Ni, Cu and Co levels in the Nikolai basalt do not increase with increasing MgO or Mg#. No significant differences in metal concentrations between the two sections sampled are apparent. The majority of samples from both sections have Ni values between 60 and 120 ppm over a range in Mg# from 50 to 63 (6 to 10% MgO). Similar trends are displayed by Co and Cu relative to MgO and Mg#. 550 450 350 D 250 O 150 50 -50 32 Cu (ppm) vs. MgO% - Nikolai Basalt o • D 0 \ V? \a ° fl | °cPo 0 0 0 • o o <tf 38 44 50 56 62 68 o location-Burwash' • location-Halfbree' MG 140 120 100 80 < 60 H , I 40 20 0 -20 32 38 Ni (ppm) vs. Mg# - Nikolai Basalt ;°. ...*g • • n o o n o 44 50 56 MG n o o • o 62 68 o location='Burwash' • location-Halfbree' Ni (ppm) vs. MgO% - Nikolai Basalt 140 120 100 80 < o 60 K I 40 Z 20 0 -20 • * o rf>° „ ~ ° o 0 P o o • oo • 0 0 o location-Burwash' • • • ni. in. . >• ..* MGO 75 Figures 4-21, 4-2m and 4-2n: Plots of Cu versus Mg#, Co versus MgO and Co versus Mg# - Nikolai Basalt. Ni, Cu and Co levels in the Nikolai basalt do not increase with increasing MgO or Mg#. No significant differences in metal concentrations between the two sections sampled are apparent. The majority of samples from both sections have Ni values between 60 and 120 ppm over a range in Mg# from 50 to 63 (6 to 10% MgO). Similar trends are displayed by Co and Cu relative to MgO and Mg#. 7f* 550 450 350 D 250 o 150 50 -50 32 38 Cu (ppm) vs. Mg# - Nikolai Basalt O o D o C? o 0 • ° a o aPo • o CJD 44 50 MG 56 62 68 o location-Burwash' a location='Halfbree' Co (ppm) vs. MgO - Nikolai Basalt 44 40 36 32 O " 28 24 20 16 • q. o . s • D • o° 0 o D • C o ° a o • o 6 7 MGO 8 9 10 11 o location-Burwash' n location- Haltbree' Co (ppm) vs. Mg#- Nikolai Basalt o location-Burwash' n location^Halfbree1 76 related to the Ni-Cu-PGE mineralization. This particular flow, the Lower Nadezhdinsky, has an average Ni value of 23 ppm Ni and an average Cu value of 33 ppm with an average MgO value of 6.07% (Naldrett et al, 1992). These metal concentrations are in contrast to flows which are believed to have normal metal abundances (the Mokulaevsky flow) which has an average of 103 ppm Ni and 132 ppm Cu with an average of 6.83% MgO. In comparison with the undepleted Siberian volcanics and given its range in MgO levels, the bulk of the Nikolai basalts would appear to be weakly to moderately depleted in Ni and moderately to strongly depleted in Cu. Rather than directly compare possible metal depletion in the Nikolai basalt with the Siberian volcanics, it seems more appropriate to compare metal abundances in the Nikolai basalt with those of the Karmutsen basalt in the western part of Wrangellia. Figures 4-2o to 4-2w show plots of Ni, Cu and Co versus MgO, Mg# and LOI for samples from the Burwash and Halfbreed sections in the Yukon and a suite of samples from the Karmutsen basalt in the Buttle Lake area of Vancouver Island. The Buttle Lake data are unpublished Inco Ltd. data from stratigraphically controlled sampling carried out in 1993. Ni, Cu and Co values are from total digestion-ICP analysis and MgO values were done by XRF. The Inco basalt standard analyzed with both the Buttle Lake data and the Kluane data had a relative error of 2.7%, 5.1% and 0 % for Cu, Co and Ni respectively. These plots show the Nikolai basalt to be more magnesian with the Nikolai basalt having an MgO range of 6 to 10% , while the bulk of the Karmutsen basalt has an MgO range of 5 to 8%. Despite the Nikolai basalt being more magnesian, the two formations have approximately the same range in Ni values. Cu and Co are significantly lower in the Nikolai basalt compared with the Karmutsen, despite the Nikolai basalt having an overall more magnesian composition. Cu values average 166 ppm in the Karmutsen basalt versus 81 ppm in the Burwash section and 108 ppm in the Halfbreed section. The Cu/Zr versus Mg# plot (Figure 4-2y) shows Cu/Zr ratios in the Nikolai basalt to be consistently lower than Cu/Zr ratios in the Karmutsen despite the Nikolai basalt having higher Mg numbers. Ni, Cu and Co were plotted versus LOI to explore the possibility that Cu and Co were removed during metamorphism or hydrothermal alteration. The Nikolai basalt samples consistently have higher LOI values than the Karmutsen samples. Thus, it is possible that Cu and Co have been removed by metamorphism and alteration, but the fact that Ni has not been similarly effected and that the Cu and Co values are consistently (and not erratically) depleted suggests otherwise. Also, there is very little correlation between Ni, Cu and Co values and LOI as would be expected if metals were removed progressively by greater degrees of hydrothermal alteration. Plots of Cu/Zr and Ni/MgO versus LOI (Figures 4-2x and 4-2z) also show no correlation between these ratios and LOI. Thus, while the depletion at Noril'sk occurs in a particular unit within the basaltic volcanic pile, the Nikolai basalt is depleted in metals relative to another part of 77 Figures 4-2o, 4-2p and 4-2q: Plots of Ni versus MgO, Mg# and LOI - Nikolai and Karmutsen Basalt. These plots show the Nikolai basalt to be more magnesian with the Nikolai basalt having an MgO range of 6 to 10% , while the bulk of the Karmutsen basalt has an MgO range of 5 to 8%. Despite the Nikolai basalt being more magnesian, the two formations have approximately the same range in Ni values. No correlation between Ni and LOI is apparent. 77a 140 120 100 80 < p o 60 K 1 40 Z 20 0 -20 Ni (ppm) vs. MgO% - Nikolai and Karmursen Basalt oo o o °>° § oo o • o 6 7 MGO 10 11 o location='Burwash' a location-Halfbree' « location='Buttle' 140 120 100 _ l 80 < t-o 60 K 1 40 z 20 0 -20 32 Ni (ppm) vs. Mg# - Nikolai and Karmutsen Basalt 38 o o o • t 0<^o%D O a % © o n o 44 50 56 62 o location-Burwash' • location-Halfbree' e location-Buttle' MG Ni (ppm) vs. LOI% - Nikolai and Karmutsen Basalts 140 120 100 80 < o 60 K I 40 z 20 0 -20 —, • —i 1 • • r - — • • — ' • ' • ! — • • • ' • ' ' • 1 ' 1 — • • o o o o o * o \ :- Q-O 0 • j ° 0 "D ; r?o ! 1 or? i ° 0 ° o o 3 1 2 3 4 5 6 7 8 < o location='Burwash' a location-Halfbree' o location='Buttle' LOI 78 Figure 4-2r, 4-2s and 4-2t: Plots of Cu versus MgO, Mg# and LOI - Nikolai and Karmutsen Basalt. Cu is significantly lower in the Nikolai basalt compared with the Karmutsen, despite the Nikolai basalt having an overall more magnesian composition. Cu values average 166 ppm in the Karmutsen basalt versus 81 ppm in the Burwash section and 108 ppm in the Halfbreed section. Cu is plotted versus LOI to explore the possibility that Cu was removed during metamorphism or hydrothermal alteration. The Nikolai basalt samples consistently have higher LOI values than the Karmutsen samples, but there is no correlation between Cu and LOI. A correlation should exist between Cu and LOI if Cu has been removed by alteration. 76c Cu (ppm) vs. MgO - Nikolai and Karmutsen Basalt location='Burwash' location-Halfbree' location-Buttle' MGO Cu (ppm) vs. Mg# - Nikolai and Karmutsen Basalt 550 o location-Burwash' a location-Halfbree' gg o location-Buttle' MG Cu (ppm) vs. LOI% - Nikolai and Karmutsen Basalts 550 450 350 Z) 250 O 150 50 -50 0 o o; • • o o \ OD<> ° 0 * P «s | °a o o ° o 3 4 5 6 LOI o location='Burwash' • location-Halfbree' o location='Buttle' 79 Figure 4-2u, 4-2v and 4-2w: Plots of Co versus LOI, Mg# and MgO - Nikolai and Karmutsen Basalt. Co is significantly lower in the Nikolai basalt compared with the Karmutsen, despite the Nikolai basalt having an overall more magnesian composition. Co is plotted versus LOI to explore the possibility that Co was removed (along with Cu) during metamorphism or hydrothermal alteration. The Nikolai basalt samples consistently have higher LOI values than the Karmutsen samples, but there is no correlation between Cu and LOI. A correlation should exist between Co and LOI if Co has been removed by alteration. 7U Co (ppm) vs. LOI% - Nikolai and Karmutsen Basalts 44 40 j 36 32 O O 28 24 20 ie o o o o o o o o o o o o o o o o o o o o o 4 5 LOI o location-Burwash' a location-Halfbree' o location-Buttle' 44 i 40 36 32 O O 28 24 20 16 32 Co (ppm) vs. Mg#- Nikolai and Karmutsen Basalt 38 o o o o o O 00 o c o o o o o o o • o • o 0 % o a** 0 % a o a o o o 44 50 MG 56 62 68 o location-Burwash' o location-Halfbree' o location='Buttle' Co (ppm) vs. MgO% - Nikolai and Karmutsen Basalt 44 40 36 32 O O 28 24 20 16 — . — , — , — , — . — ^ — , — . — . — . — • o o o O 0 o o o o o o o o i O O 0 I <>o 0 0 o o • 0 o ^ S 0 o o ° ° 0 • O o 2 2 3 4 5 6 7 8 9 10 11 o location-Burwash' • location-Halfbree' o location-Buttle' MGO 80 Figure 4-2x: Plot of Cu/Zr versus LOI - Nikolai and Karmutsen Basalt. No correlation between Cu/Zr and LOI is apparent. Figure 4-2y: Plot of Cu/Zr versus Mg# - Nikolai and Karmutsen Basalt. Cu/Zr ratios in the Nikolai basalt are consistently lower than Cu/Zr ratios in the Karmutsen despite the Nikolai basalt having higher Mg numbers. Figure 4-2z: Plot of Ni/MgO versus LOI - Nikolai and Karmutsen Basalt. No correlation between Ni/MgO and LOI is apparent. 80* Cu/Zr vs. LOI% Nikolai and Karmutsen Basalt o location-Burwash' n location-Halfbree' o location-Buttle' 12 10 8 6 4 2 0 32 Cu/Zr vs. Mg# - Nikolai and Karmutsen Basalts o • o • : • o < > * "^aSh no°o o o ° • 38 44 50 MG 56 62 68 o location-Burwash' o location='Halfbree' o location-Buttle' Ni/MgO vs. LOI% - Nikolai and Karmutsen Basalts O o • o a o • o 0 s j % o 8 rf3 • o D o o o location-Burwash' Q location-Halfbree' 0 1 2 3 4 5 6 7 8 9 ° location='Buttle' 81 the terrane. As suggested by Hulbert (1995), the Karmutsen is probably a more distal, offshore accumulation (as evidenced by the abundant pillows) relative to the Nikolai. REE spidergrams (samples normalized by chondrite values of Sun, 1982) are shown for the Burwash and Halfbreed sections in Figures 4-3a and 4-3b. Both sections have REE spidergrams with relatively flat profiles similar to MORB basalt (Sun and McDonough, 1989). Samples from the base of the Burwash section and all of the Halfbreed section have slightly steeper slopes in the LREE part of the profiles. This is possibly reflecting the more evolved composition of the initial magma relative to more primitive upper part of the formation and also contamination of the initial magma as it passed through the underlying Island Arc volcanics. An Extended Element Profile (samples normalized by primitive mantle values of Taylor and McLennan, 1985) for the average basalt from both sections is shown in Figure 4-4. The profiles in this plot do not match profiles of N (normal) or E (enriched) -MORB rocks. The plots from both sections have positive anomalies in Rb and K which are distinctly different from both N and E-MORB patterns and are probably the result of contamination. Ba is not particularly anomalous in either section. The Burwash profile has a positive Sr anomaly. A similar Sr anomaly noted by Hulbert (1995) in his Extended Element Profiles for samples of Nikolai basalt, Maple Creek Gabbro and marginal gabbro chilled phases was attributed to plagioclase accumulation. Normalized elements in the Nb-La-Ce part of profiles appear to be depleted relative to E-MORB. Low levels of Nb are common in rocks which are the result of partial melting (Island Arc volcanics); this signature in the Nikolai basalt is again probably the result of contamination. Figure 4-5 is an AFM plot of basalts from both sections. The calc-alkaline/tholeiitic boundary from Irvine and Baragar (1971) is shown for reference. The Nikolai basalt samples from this study evenly straddle the boundary. This is also believed to be due to contamination from the underlying Island Arc volcanic units. An AFM plot is shown in Barker et al (1989) for a suite of Karmutsen basalt samples from Vancouver Island and the Queen Charlotte Islands. These rocks all plot in the tholeiitic field, suggesting the Nikolai basalt has been subject to a greater degree of contamination than the coeval Karmusten. A similar trend is shown in Nikolai basalt and Kluane Belt gabbro data presented by Hulbert (1995). Whole rock compositions of Nikolai basalt from both sections are plotted on a Jensen Cation Plot (Jensen, 1967) in Figure 4-6. The samples plot in a field which straddles the High Iron Tholeiitte (HFT) and High Magnesium Tholeiite (HMT) sections of the plot. 82 Figures 4-3a and 4-3b: REE Spidergrams - Nikolai Basalt, Halfbreed Creek and Burwash Creek Sections. REE values are normalized by chondrite values from Sun (1982). Both sections have REE spidergrams with relatively flat profiles similar to MORB basalt. Samples from the base of the Burwash section and all of the Halfbreed section have slightly steeper slopes in the LREE part of the profiles. 62<K R E E Spidergram - Nikolai Basalt, Halfbreed Ck. Section 20 or o 5 OT o z UJ o 5 or o z or a a z 2 or O z 2 5 or o z UJ 5 or o z Q e> s or o z CQ 2 or o z 2 or O z O x. 2 or O z or ui or o z 2 2 or o z f 2 or O z 3 Rx224055 Rx224056 R X 2 2 4 0 5 7 Rx224058 Rx224059 R X 2 2 4 0 6 0 Rx224061 Rx224062 Rx224063 Rx224064 Rx224065 Rx224066 R E E Spidergram - Nikolai Basalt, Burwash Ck. Section 2 2 2 2 2 2 2 or or or or or or or o o o o O o c z z z z z z z < UJ or Q 2 Q o Z w UJ 0 2 2 2 2 or or or or o o o o z z z z m >- o or i - o . I UJ or o z 2 r-2 or o z m >-2 or o z ID - o - RX224069 — D — Rx224070 —o— Rx224071 Rx224072 Rx224073 m- Rx224074 -*- RX224075 - - A - RX224076 — j . Rx224077 RX224078 - o — RX224079 - c - Rx224080 Rx224081 83 Figure 4-4: Extended Element Profile - Nikolai Basalt. Elements are normalized by primitive mantle values from Taylor and McLennan (1985). The profiles in this plot do not match profiles of N (normal) or E (enriched) -MORB rocks. The plots from both sections have distinct positive anomalies in Rb and K which are distinctly different from both N and E-MORB patterns and are probably the result of contamination. Ba is not particularly anomalous in either section. The Burwash profile has a distinct positive Sr anomaly. Normalized elements in the Nb-La-Ce part of profiles appear to be depleted relative to E-MORB. 84 Figure 4-5: AFM Plot - Nikolai Basalt. The calc-akaline /tholeiitic boundary from Irvine and Baragar (1971) is shown for reference. Nikolai basalt samples from this study evenly straddle the boundary. This is also believed to be due to contamination from the underlying Island Arc volcanics. Figure 4-6: Jensen Cation Plot - Nikolai Basalt. Whole rock compositions from Nikolai basalts from both sections are plotted on a Jensen Cation Plot (Jensen, 1967) in Figure 4-6. The samples plot in a field which straddles the High Iron Tholeiitte (HFT) and High Magnesium Tholeiite (HMT) sections of the plot. 85 Tectonic and Magmatic Setting of Nikolai Basalt Since the late 1970s, numerous authors have attempted to clarify the tectonic and magmatic setting of the Nikolai and Karmutsen basalts. While not the focus of this study, a discussion of the tectonic and magmatic setting of these basalts is relevant to the origin of Ni-Cu-PGE mineralization in the Kluane Belt. Jones et al (1977) contributed a paper which provided a stratigraphic framework which linked various sections throughout Wrangellia; a series of sections are presented which are up to thousands of kilometres apart with almost identical stratigraphic elements. This paper convincingly documents Wrangellia as an Allocthonous Terrane. Barker et al (1989) studied basalt sections and geochemical aspects of Karmutsen basalt sections on Vancouver Island and the Queen Charlotte Islands. Barker et al espouse the rifting of the Paleozoic island arc sequence. Based on trace element discrimination diagrams, they believe an initial Island Arc Tholeiite (IAT) magma was initially generated and extruded. E-MORB or OIB magma from a deep source then mixed with the IAT magma and fractionated to produce the Karmutsen basalt. Samson et al (1990) propose that evolved continental crust can not be part of this terrane on based Nd and Sr isotopic evidence. They believe that the Wrangellia terrane was derived in an intra-oceanic environment and then accreted. Richards et al (1991) describe a mantle plume origin for the Nikolai and Karmutsen basalts involving a diapir rising beneath an island arc which resulted in the partial melting of mantle, upper mantle and island arc material. This resulted in the extrusion of very large volumes of tholeiitic lava which fractionated from a picritic melt. Lassiter at al (1995) believe that the Nikolai and Karmutsen formations are oceanic flood basalts derived from a plume source. This conclusion is based on Nd, Sr and Pb isotopic evidence. Lassiter et al document contamination in the base of the Karmutsen and Nikolai basalts based on isotopic and trace element geochemistry. Hulbert (1995) reviews previous papers on the tectonic and magmatic setting of the Nikolai and Karmutsen basalts and provides a model to account for the origin of the basalts. He discounts the occurrence of rifting due to: the lack of crustal extension and graben formation, the occurrence of widespread sills rather than dykes and dyke swarms and the shallow nature of the volcanism (subaerial and shallow marine). Hulbert proposes a model similar to that of Richards et al (1991) with the basalts and related intrusions being the results of a mantle plume. This study suggests that contamination is not limited to the lower part of the Nikolai basalt as documented by Lassiter et al (1995). Ce/Yb ratios are highest in the lower part of the section, whereas Ba/Zr and Th/Ta are highest in the upper part of the section. The high Ba/Zr and Th/Ta levels in the upper part of the volcanic pile may be due to contamination of the basalt by Hasen Creek 8 6 Formation sedimentary rocks that initial more evolved magma did not encounter due to varying conduit morphologies and paths. Ce/Yb ratios probably record both the degree of contamination and how evolved a magma is. The high Ce/Yb in the base of the basalt pile (in the sections sampled as part of this study) was likely caused by mixing of mantle derived and Island Arc magma. Widespread structural evidence of rifting was not noted on the Klu property or in regional mapping efforts discussed in Section 2. A mantle plume seems to be a probable source of the tholeiitic magma. The above geochemical data have shown that the Nikolai basalt is, on average, more magnesian than the Karmutsen, which suggests (as noted by Hulbert, 1995) that the Nikolai basalt may be closer to the axial jet of a mantle plume. The tectonic setting in Wrangellia is clearly different from that in the Noril'sk camp where Ni-Cu-PGE deposits are associated with continental flood basalts. It is not obvious that the tectonic setting of a given area is related to metal depletion. ____ 87 Section 5: Geochemical Comparison between Peridotite, Marginal Gabbro, Maple Creek Gabbro and Nikolai Basalt 5.1 Genetic Relationships and Contamination Previous sections have demonstrated that the Kluane-type intrusions (peridotite, marginal gabbro and Maple Creek gabbro) from the Spy and Lewis intrusions on the Klu property are co-genetic and that the entire basalt sections in Halfbreed and Burwash creeks are also co-genetic. This section will explore the genetic relationship between the Nikolai basalt and Kluane-type intrusions. Figure 5-1 shows T i 0 2 vs. Zr for the Kluane-type intrusions and the Nikolai basalt. The various units clearly show a co-genetic trend with most of the samples plotting within 2 standard deviation error (see appendices) of a line passing through the origin. The obvious exception to the co-genetic trend are certain of the marginal gabbro samples. Various plots (Ce/Yb vs. MgO, Th vs. Ce/Yb, Ba/Zr vs. Ce/Yb) were constructed to compare the degree of contamination in the Kluane-type intrusions and Nikolai basalts; these are shown in Figures 5-2a to 5-2c. Ce/Yb, Ba/Zr and Th were shown in previous sections to monitor contamination. Ce/Yb was plotted versus MgO to test if the Ce/Yb levels were related to compositional differences rather than contamination. The marginal gabbro samples have MgO values within the range of the Nikolai basalt, but the Ce/Yb levels are the highest of any rock type. This suggests that the Ce/Yb levels are related to contamination. Ba/Zr was plotted versus Ce/Yb to further test if the high Ce/Yb values are in fact related to contamination. The Ba/Zr ratio was used so as to normalize Ba levels by an index of fractionation. Ba was analyzed by different methods in the basalt and intrusive samples. The Ba levels in the basalt analyses appear to be biased low (see appendix) relative to the standard, but when normalized, this bias has only a relatively minor effect and the overall integrity of the diagram is not adversely effected. The Ba/Zr vs. Ce/Yb plot shows that a group of marginal gabbros have high values for both ratios. The Th vs. Ce/Yb plot shows marginal gabbro having the highest Ce/Yb and Th values. The elevated Ba/Zr, Ce/Yb and Th are proposed to be a geochemical signature imparted by contamination of the tholeiitic Nikolai magma by volcanics and sediments from the island arc rocks of the Hasen Creek and Station Creek Formations. These plots show a trend of contamination increasing from basalt to peridotite and Maple Creek gabbro to marginal gabbro. This is consistent with the marginal gabbro being the earliest phase with the greatest degree of interaction (contamination) with country rock. Subsequent magma in the interior of sills (peridotite) would not be as exposed to the contaminant effects of the country rocks. The volcanic rocks appear to have been extruded from well 8 8 Figure 5-1: Plot of T i 0 2 versus Zr - Kluane Intrusions and Nikolai Basalt. The various units clearly show a co-genetic trend with most of the samples plotting within 2 standard deviation error (see appendices) of a best fit line passing through the origin. The obvious exception to the co-genetic trend are certain of the marginal gabbro samples. Figure 5-2a: Plot of Ce/Yb versus MgO - Kluane Intrusions and Nikolai Basalt. Ce/Yb was plotted versus MgO to test if the Ce/Yb levels were related to compositional differences rather than contamination. The marginal gabbro samples have MgO values within the range of the Nikolai basalt, but the Ce/Yb levels are the highest of any rock type. This suggests that the Ce/Yb levels are related to contamination. 680, Ti02% vs. Zr (ppm) - Kluane Intrusions and Nikolai Basalt 20 40 60 80 100 120 140 ZR PPM 160 o ROCK UNIT=TrMG' a ROCK UNIT-TrMCG* o ROCK UNIT-TrPRD' A ROCK UNIT='Basalf Ce/Yb vs. MgO% - Kluane Intrusions and Nikolai Basalt o ROCKJJNIT-TrMG' a ROCK_UNIT='TrMCG* o ROCKJJN IT=TrPRD' A ROCK UNIT-Basalt' MGO_% 89 Figure 5-2b: Plot of Th versus Ce/Yb - Kluane Intrusions and Nikolai Basalt. This plot shows marginal gabbro having the highest Ce/Yb and Th values. The elevated Ce/Yb and Th are proposed to be a geochemical signature imparted by contamination of the tholeiitic Nikolai magma by volcanics and sediments from the island arc rocks of the Hasen Creek and Station Creek Formations. Figure 5-2c: Plot of Ba/Zr versus Ce/Yb - Kluane Intrusions and Nikolai Basalt. This plot shows that a group of marginal gabbros have high values for both ratios. The elevated Ba/Zr and Ce/Yb are proposed to be a geochemical signature imparted by contamination of the tholeiitic Nikolai magma by volcanics and sediments from the island arc rocks of the Hasen Creek and Station Creek Formations. Wo Th (ppm) vs. Ce/Yb - Kluane Intrusions and Nikolai Basalt 5.5 4.5 o oo o 3.5 I 2.5 1.5 B o 0.5 -0.5 A a-g 0 A •? 0o A A oAO 0 < S > < > 8 10 CE YB 12 14 16 18 o ROCK_UNIT=TrMG' a ROCK_UNIT=Tr1v1CG• o ROCK UNIT='TrPRD' A ROCKUNIT-Basalt' Ba/Zr vs. Ce/Yb - Kluane Intrusions and Nikolai Basalt 16 14 12 10 * 8 3 4 2 0 -2 o o , A *o o°*<> A o A Cp o ROCK_UNIT='TrMG' a ROCK_UNIT=TrMCG' o ROCK UNIT=TrPRD' 10 12 14 16 18 A ROCK_UNIT='Basalf CE YB 90 established conduits through which large volumes of magmas would pass, thus minimizing the interaction of the magma with country rock. Pearce Element Ratio (PER) plots as described by Russell et al (1990) and Russell and Stanley (1990) were used to explore the mineralogical controls on the compositions of the Klaune-type intrusions and the Nikolai basalt. Ti was chosen as the denominator element in the PER diagrams as the binary plots of immobile elements in the preceding sections suggest that it is conserved. Error is represented by ellipses which are propagated as per the method of Stanley (1990); the plots were generated using MERLIN software (Stanley and Wong, 1996). One standard deviation error ellipses are shown-based on the repeated analysis of the Inco Basalt Standard for the basalt samples and an average error from field replicates of the intrusive samples. „v v •. Figure 5-3 is a 0.5Fe+0.5Mg (FM) PER plot of the Kluane-type intrusions and Nikolai basalt. On this diagram, samples whose error ellipses fall along a line with a slope of 1.0 are related entirely by olivine fractionation. The olivine, plagioclase and clinopyroxene mineralogical vectors are shown on the diagram for reference. Peridotite samples on this plot do not fall along such a line, instead they fall along a general trend with a slope of 0.96 showing that the composition of the peridotite almost entirely controlled by olivine fractionation, but with contributions from other mineralogical sources. It should be noted that the error involved in the slope calculation makes it difficult to distinguish between slopes of 0.96 and 1.0. Some groups of peridotite samples appear to be effected by plagioclase and pyroxene crystallization to a minor extent. All basalt samples and some of the gabbro samples fall about a line with a slope of approximately 0.15, indicating their composition is controlled by plagioclase and clinopyroxene fractionation. A group of 4 marginal gabbro samples from the Spy Sill plot just below the basalt trend. Figure 5-4 is a 0.5AI+0.5Fe+0.5Mg+1.5Ca+2.75Na PER diagram. Samples whose compositions are completely controlled by olivine, plagioclase and clinopyroxene fractionation will fall along a line with a slope of 1.0. In this plot, almost all the basalt, gabbro and peridotite samples plot within error of a line with a slope of 0.94. Again, it should be noted that the error involved in the slope calculation makes it difficult to distinguish between slopes of 0.94 and 1.0. The exceptions to this are the basalt samples from the upper part of the Burwash section and some of the marginal gabbro samples from the Spy Sill. Thus, almost the entire compositional variation of the Kluane-type intrusions and the Nikolai basalts are accounted for by a combination of olivine, plagioclase and clinopyroxene fractionation. While the samples from the upper part of the Burwash section could be interpreted to fall along a parallel trend to the remainder of the data, the marginal gabbro samples clearly do not. The separation of the Spy Sill marginal gabbro from the rest of the data fits with the contamination scenario described in previous sections. 91 Figure 5-3: 0.5Fe+0.5Mg/Ti versus Si/Ti PER Plot - Kluane Intrusions and Nikolai Basalt. On this diagram, samples whose error ellipses fall along a line with a slope of 1.0 are related entirely by olivine fractionation. Peridotite samples plot along a general trend with a slope of 0.96 showing that the composition of the peridotite almost entirely controlled by olivine fractionation, but with contributions from other mineralogical sources. All basalt samples and some of the gabbro samples fall about a line with a slope of approximately 0.15, indicating their composition is controlled by plagioclase and clinopyroxene fractionation. A group of 4 marginal gabbro samples from the Spy Sill plot just below the basalt trend. 92 Figure 5-4: 0.5AI+0.5Mg+1.5Ca+2.75Na/Ti versus Si/Ti PER Plot - Kluane Intrusions and Nikolai Basalt. In this plot, almost all the basalt, gabbro and peridotite samples plot within error of a line with a slope of 0.94. The exceptions to this are the basalt samples from the upper part of the Burwash section and some of the marginal gabbro samples from the Spy Sill. Thus, almost the entire compositional variation of the Kluane-type intrusions and the Nikolai basalts are accounted for by a combination of olivine, plagioclase and clinopyroxene fractionation. 91+ 0 40 80 120 160 200 240 Si/Ti 93 5.2: Metal Depletion (Ni-Cu-Co) Ni, Cu and Co are plotted versus Mg# (Figures 5-5a to 5-5f) for the Kluane-type intrusions and the Nikolai basalt. As documented in Section 3, the Lewis Intrusion samples appear to be depleted in Ni and Cu relative to peridotite from the Spy Sill. Ni and Co show relatively restricted ranges in abundance in the basalt and gabbro samples compared with peridotite samples. The Ni versus Mg# plot shows a logarithmic relationship between Ni and Mg#. Cu values in this plot appear to be noisier relative to Ni and Co, but overall have a more restricted range in the gabbro and basalt samples compared with peridotite samples. Cu/Zr and Ni/MgO are plotted versus Mg# (Figures 5-5g and 5-5h) to further explore possible metal depletion. The Cu/Zr ratio, which normalizes Cu to an index of fractionation, shows a much smoother trend versus Mg# than Cu. The Cu/Zr range for a given Mg# level is restricted for basalt and gabbro, but very large for peridotite. A similar trend is displayed in the Ni/MgO versus Mg# plot. The Ni/MgO ratio serves to normalize the Ni by a measure of olivine content and how forsteritic the olivine is. A probable explanation for the behaviour of these metals relative to Mg# is that some of the peridotite magma encountered sulphide at some location in the crust and equilibrated with it. The metal-rich peridotite must have had very little or no interaction with sulphide as it ascended through the crust. 94 Figures 5-5a, 5-5b and 5-5c: Plots of Cu, Co and Ni versus MgO - Kluane Intrusions and Nikolai Basalt. Lewis Intrusion samples appear to be depleted in Ni and Cu relative to peridotite from the Spy Sill. Ni and Co show relatively restricted ranges in abundance in the basalt and gabbro samples compared with peridotite samples. Cu values in this plot appear to be noisier relative to Ni and Co, but overall have a more restricted range in the gabbro and basalt samples compared with peridotite samples. 9V« 550 450 350 £ 250 150 50 -50 Cu (ppm) vs. MgO% - Kluane Intrusions and Nikolai Basalt 10 A 0 * A o • % * <> - o. A A | • * <6 o o 0 j A4,A j 9 o ^ A ^ ; a o o 0 0 o oo o> o o o o 15 20 MGO % 25 30 35 40 o ROCKJJ N IT=TrMG' a ROCK_UNIT='TrMCG' o ROCK_UNIT='TrPRD' A ROCK UNIT='Basalf Co (ppm) vs. MgO% - Kluane Intrusions and Nikolai Basalt 15 20 MGO % o ROCK_UNIT=TrMG' a ROCK_UNIT=TrMCG' o ROCK UNIT='TrPRD' A ROCK_UNIT='Basalt' Ni (ppm) vs. MgO - Kluane Intrusions and Nikolai Basalt o ROCK_UNIT=TrMG' o ROCK_UNIT='TrMCG' o ROCK_UNIT='TrPRD' A ROCK UNIT='Basalf MGO_% 95 Figures 5-5d, 5-5e and 5-5f: Plots of Co, Cu and Ni versus Mg# - Kluane Intrusions and Nikolai Basalt. Lewis Intrusion samples appear to be depleted in Ni, Cu and Co relative to peridotite from the Spy Sill. Ni and Co show relatively restricted ranges in abundance in the basalt and gabbro samples compared with peridotite samples. The Ni versus Mg# plot shows a logarithmic relationship between Ni and Mg#. Cu values in this plot appear to be noisier relative to Ni and Co, but overall have a more restricted range in the gabbro and basalt samples compared with peridotite samples. 95« 140 120 100 5 80 Q-Q-o1 60 o 40 20 0 30 Co (ppm) vs. Mg#- Kluane Intrusions and Nikolai Basalt 40 A.A4, o • • o 50 60 WIG 70 80 90 o ROCK_UNIT=TrMG' a ROCK_UNIT='TrMCG' o ROCK_UNIT=TrPRD' A ROCK UNIT='Basalf 550 450 350 a 250 ° 150 50 -50 30 Cu (ppm) vs. Mg# - Kluane Intrusions and Nikolai Basalt 40 A 8 A So A o ^ ° * A A A A A • O 0 o o o <><> A A AA 0 ° 50 60 70 80 90 o ROCK UNIT-TrMG' o ROCK_UNIT=TrMCG' o ROCK_UNIT='TrPRD' A ROCK UNIT^Basalf MG 1600 1400 1200 1000 800 600 400 200 0 -200 Ni (ppm)vs. Mg#- Kluane Intrusions and Nikolai Basalt $ o o j 40 50 60 70 80 o ROCK_UNIT=TrMG' • ROCK UNIT-TrMCG' o ROCK_UNIT='TrPRD' A ROCK UNIT='Basatf MG 96 Figures 5-5g and 5-5h: Plots of Cu/Zr and Ni/MgO versus Mg# - Kluane Intrusions and Nikolai Basalt. Cu/Zr shows a much smoother trend versus Mg# than Cu. The Cu/Zr range for a given Mg# level is restricted for basalt and gabbro, but very large for peridotite. A similar trend is displayed in the Ni/MgO versus Mg# plot. Cu/Zr vs. Mg# - Kluane Intrusions and Nikolai Basalt 0 O A O 0 o o o 0 0 0 * O A <*> o o 0 0 , o o !5 o o A * A o o o 30 40 50 60 MG 70 80 90 o ROCK UNIT-TrMG' o ROCK_UNIT='TrMCG' o ROCK_UNIT=TrPRD' A ROCK_UNIT=,Basalf Ni/MgO vs. Mg# - Kluane Intrusions and Nikolai Basalt 30 40 Q O * * A A 8o o o o • o 50 60 MG 70 80 o ROCK_UNIT=TrMG' a ROCK_UNIT='TrMCG' o ROCK_UNIT=,TrPRD' 9 0 A ROCK_UNIT='Basalt' 97 Section 6: Petrology and Geochemistry of Ni-Cu-PGE Mineralization Associated with the Spy Sill and Sulphide/Oxide Layers in Hasen Creek Formation Sediments 6.1 Introduction This section will focus on the petrography and geochemistry of sulphide and oxide mineralization adjacent to and within the Spy Sill. A general description of the Ni-Cu-PGE showings and barren sulphide/oxide layers adjacent to the Spy Sill is given in Section 2. The location of Ni-Cu-PGE showings, barren sulphide/oxide layers and samples discussed in this section are shown in Figure 6-1. 6.2 Lithological Description Ni-Cu-PGE rich sulphides occur in three principal locations/lithologies in and adjacent to the Spy Sill. These are: 1) within the basal marginal gabbro unit, 2) at the contact between the basal marginal gabbro and the footwall siltstone, and 3) within the footwall siltstone. Sulphides within the marginal gabbro occur in massive, semi-massive, net-textured and disseminated form. Sulphide occurring directly at the contact occurs in massive and semi-massive form. The footwall siltstone unit hosts massive sulphide lenses and semi-massive sulphide. Ni-Cu-PGE mineralization in the footwall is only observed within 10 m of the marginal gabbro/siltstone contact. The samples included in this study are from the Spy showings which include sulphide occurrences over a 1 km strike along the base of the Spy Sill in the area of the Spy lithogeochemical section. Two Ni-PGE barren pyrrhotite/magnetite bodies are located in an area 1.2 km northwest of the Spy Section. The lower 10-m thick magnetite layer is centred approximately 50 m up-section from the upper contact of the Spy Sill while the 4-m thick pyrrhotite body at the Claim Post Showing occurs approximately 140 m up-section. The magnetite layer contains internal blocks of carbonate (limestone) up to 1 m in diameter. This can be traced for approximately 150 m along strike. At either end it narrows (to 1 m) and becomes pyrrhotite-rich. This magnetite layer is hosted within silicified siltstone. The contacts above and below the massive pyrrhotite body at the Claim Post Showing are poorly exposed, but the footwall appears to consist of silicified and fractured siltstone; the hangingwall consists of gabbro. The thickness of this gabbro is believed to be less than 10 m. A magnetite layer approximately 30 cm thick occurs between the top of the pyrrhotite body and the overlying gabbro. The pyrrhotite body is talus covered along strike, but the strike extent is believed to be limited (probably less than 30 m). No layering is observed in the pyrrhotite body. Chalcopyrite is the only other sulphide observed in the pyrrhotite body; it occurs mainly along fractures and makes up less than 1% of the sulphide. 98 3 99 An additional Ni-PGE barren sulphide occurrence is located 800 m northwest of the Spy Section. A 1.5-m wide x 3-m long pyrrhotite lens occurs at the base of a limestone unit 25 m up-section from the upper contact of the Spy Sill. This lens contains local seams of magnetite up to 4 cm thick. The limestone unit and pyrrhotite lens overlie a siltstone unit. Several samples of gypsum were collected on the property to for whole rock geochemistry and sulphur isotope analysis. The gypsum samples are from Upper Triassic sulphate beds which directly overlie the Nikolai basalt. These samples were collected to determine if the gypsum beds could possibly be contributing sulphur to the Kluane-type magmas. Because of the structural complications on the property, it is not certain whether all gypsum beds were actually Upper Triassic. Where not complicated by faulting, the gypsum beds directly overlie the Nikolai basalt. 6.3 Petrography Polished thin sections were prepared for samples from most of the sulphide mineralized environments described above. Thin sections of samples Rx225655 and Rx222658 from massive sulphide lenses in the footwall sediments were examined. The pyrrhotite in Rx225655 (Plate 6-1) is highly strained with pervasive development of strain lamellae. Unstrained grains in this sample up to 800 fj.m are believed to be violarite which has developed after pentlandite (see Plate 6-2). Chalcopyrite in some of the lenses is believed to be a late phase as evidenced by chalcopyrite veinlets cutting pyrrhotite (see Plate 6-3). Sulphide in the chalcopyrite-rich lens from which Rx222658 was collected show 120° grain boundaries indicating equilibrium crystallization conditions (see Plate 6-4). Massive sulphide occurring along the marginal gabbro/siltstone contact shows pentlandite occurring as up to 30 LIITI flames within pyrrhotite (see Plate 6-5) and along the edges of pyrrhotite grains and gangue veinlets (see Plate 6-6). Chalcopyrite appears to be late in this environment as well as it is observed replacing and rimming pyrrhotite (see Plate 6-7). Net-textured sulphide in marginal gabbro consists of pyrrhotite and rare grains of violarite after pentlandite (see Plate 6-8). The sulphide in many of these samples envelopes or partially envelopes grains of serpentinized olivine and amphibole after clinopyroxene. Microprobe work by Jago (1995) resulted in the identification of various platinum group minerals (PGM) in the footwall-hosted sulphide lenses which occur as very fine grained (up to 50 jj.m) inclusions within pyrrhotite and chalcopyrite. PGM identified include: solid solution phases of Pd-Bi-Te-Sb-rich PGM (polarite-sudburyite-kolulskite), geversite (PtSb2), sperrylite (PtAs2), an unnamed palladian 1 0 0 Plate 6-1: Photomicrograph of massive sulphide lens in footwall sediment (sample Rx225655). Pyrrhotite in this sample is highly strained with pervasive development of strain lamellae. Plate 6-2: Photomicrograph of massive sulphide lens in footwall sediment (sample Rx225655). Unstrained grains in this sample up to 800 Lim are believed to be violarite (vl) which has developed after pentlandite (pn). /OOO, Plate 6-2 101 Plate 6-3: Photomicrograph of massive sulphide lens in footwall sediment (sample Rx225655). Chalcopyrite (cp) in some of the lenses is believed to be a late phase as evidenced by chalcopyrite veinlets cutting pyrrhotite (po). Plate 6-4: Photomicrograph of massive sulphide lens in footwall sediment (sample Rx225658). Sulphide in this chalcopyrite (cp)-rich lens show 120° grain boundaries indicating equilibrium crystallization conditions. / 0 / « Plate 6-4 102 Plate 6-5: Photomicrograph of massive sulphide occurring along the marginal gabbro/siltstone contact (sample Rx222668). Pentlandite (pn) occurs as up to 30 |o.m flames within pyrrhotite (po). Plate 6-6: Photomicrograph of massive sulphide occurring along the marginal gabbro/siltstone contact (sample Rx222668). Pentlandite (pn) is shown along the edges of pyrrhotite (po) grains and gangue veinlets. 103 Plate 6-7: Photomicrograph of massive sulphide occurring along the marginal gabbro/siltstone contact (sample Rx222668). Chalcopyrite (cp) is observed replacing and rimming pyrrhotite (po). Plate 6-8: Photomicrograph of net-textured sulphide in marginal gabbro (sample Rx222671). Pyrrhotite (po) and rare grains of violarite (vl) after pentlandite envelope and rim silicate grains. Plate 6-8 104 arsenide (PdAs3) and an unnamed palladian antimonide (Pd2Sb). Jago (1995) states that the Sb-rich nature of many of the PGM is due to the assimilation of graphitic and sulphidic sediments by the host magma. Massive sulphide from the Claimpost showing contains distinctly different textures than sulphide at the base of the Spy Sill. Pyrrhotite is unstrained and medium grained (>500 Lim); anhedral magnetite grains up to 200 p,m are common (see Plate 6-9). The magnetite grains dominantly occur along fractures with minor amounts of gangue (chlorite + serpentine?). Occasional chalcopyrite-rich sections show that chalcopyrite is a late phase which cross-cuts magnetite (see Plate 6-10). 6.4 Geochemistry 6.4.1 Methods Approximately 3 kg of material was collected from each of 41 sample sites and an attempt was made to chip off or avoid oxidized material. All samples were prepared at Chemex Labs in North Vancouver, B.C.. Samples were crushed and then pulverized in a ring pulverizer to -150 mesh. Sulphide mineralized samples with a significant silicate component were analyzed for major elements (Al203, CaO, Cr 20 3, Fe 20 3, K20, MgO, MnO, Na20, P205, Si02, Ti0 2 and LOI) by ICP-Whole Rock by Chemex Labs of North Vancouver, B.C.. Trace element (Zr, Y, Sr, Rb and Ba) analyses by pressed pellet XRF analyses were done at the Inco Exploration Lab in Copper Cliff, Ontario. Inductively Coupled Plasma (ICP) analyses for metals and major elements (Ag, Al, Ba, Ca, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, Sc, Tl, V, and Zn) were carried out by Chemex Labs. Samples were digested with Aqua Regia resulting in a partial digestion for most elements. Initial Pt, Pd and Au analyses were performed at Chemex Labs by fire assay with Pb collection and an ICP finish. Inco internal standards were included to monitor analytical quality. The QC results and an analysis of error are outlined with the data in Appendix 4. Sulphur analyses were done for samples with significant Pt and Pd values (from fire assay) at Chemex Labs using a Leco induction furnace. Samples with significant Pt and Pd values were also analyzed for Rh, Pd, Au, Pt, Ir, Os and Ru at the University of Toronto using a Ni-S collection and INAA determination. Inco Bulk Concentrate and Inco Tails analytical standards were included with these samples to monitor analytical quality. The QC results are outlined with the data in Appendix 4. Most samples (Ni-PGE barren, Ni-Cu-PGE rich and gypsum samples) were analyzed for As, Sb and Se by INAA at Actlabs and for sulphur isotope analyses at the University of Calgary. The sulphur isotope analyses were done on the 105 Plate 6-9: Photomicrograph of massive sulphide from the Claimpost showing. This sulphide has distinctly different textures than sulphide at the base of the Spy Sill. Pyrrhotite (po) is unstrained and medium grained (>500 jj.m); anhedral magnetite (mt) grains up to 200 urn are common. The magnetite grains dominantly occur along fractures with minor amounts of gangue (chlorite + serpentine?). Plate 6-10: Photomicrograph of massive sulphide from the Claimpost showing. Chalcopyrite-rich sections of the sulphide body show that chalcopyrite (cp) is a late phase which cross-cuts magnetite (mt). Plate 6-10 106 mass spectrometer at the University of Calgary after the direct conversion of sulphide and sulphate to S 0 2 as outlined by Ueda and Krouse (1986). Pulp duplicates were analyzed for analytical QC for the As, Sb, Se analyses. The QC results are outlined in Appendix 4. 6.4.2 Results General Geochemical Variations Plots of various metals versus sulphur and Ni versus MgO are shown in Figures 6-4a to 6-4f for sulphide mineralized samples from both Ni-Cu-PGE rich and Ni-Cu-PGE barren mineralization types. Marginal gabbro which hosts Ni-Cu-PGE mineralization is considerably more mafic than the fine grained leucogabbro that was studied in Section 3. The MgO range of sulphide mineralized gabbros is 5.99 to 17.94%; the average MgO is 12.77%. As is evident in Figure 6-4c, there is no correlation between MgO and Ni in the suite of mineralized samples collected. The Mg-rich nature of the mineralized marginal gabbro may be closer to the composition of the parental melt predicted by the olivine chemistry compared with the more leucocratic marginal gabbro described in Section 3. The Ni-Cu-PGE depleted nature of sulphides at the Claimpost and Lower Claimpost showings is obvious in the plots of metals versus sulphur. Samples from these showings have low Ni, Cu, Pt and Pd values and do not display any trend of increasing Ni, Cu, Pt or Pd with increasing sulphur. Samples from the Claimpost showing do show a trend of increasing Co with S level. The Ni versus S plot (Figure 6-4a) shows an overall trend of increasing Ni with S, but several types of mineralization have one or more samples where the Ni values drop with increasing S. The highest Ni values occur in massive sulphide at the gabbro-sediment contact and in sulphide lenses in the footwall sediment with the maximum Ni value (3.07%) occurring in sulphide along the gabbro-sediment contact. The Co versus S plot (Figure 6-4f) has a very similar distribution of samples. The Cu versus S plot (Figure 6-4c) shows a poor trend of increasing Cu with S content. The highest Cu values (>5%) occur in sulphide lenses in the footwall sediment, in massive lenses at the marginal gabbro-sediment contact and along fractures in the footwall sediments. The plots of Pt and Pd (by fire assay) versus S (Figures 6-4d and 6-4e) do not show a trend of increasing Pt and Pd with S. A similar range of Pt and Pd values exists over a large range in S content (1-25%). Samples with extreme Pt and Pd values greater than 10,000 ppb are not all shown. Most of these samples are from Cu-rich sulphide lenses in the footwall sediments. The lack of correlation between PGE and S is unusual considering the highly chalcophile nature of PGE. This may be due to part to the contamination the Ni-Cu-PGE enriched sulphide by variable amounts of synsedimentary sulphide. 107 Figure 6-4a: Plot of Ni versus S - Klu property. This plot shows an overall trend of increasing Ni with S, but several types of mineralization have one or more samples where the Ni values drop with increasing S. The highest Ni values occur in massive sulphide at the gabbro-sediment contact and in sulphide lenses in the footwall sediment with the maximum Ni value (3.07%) occurring in sulphide along the gabbro-sediment contact. Legend: CTMS= massive sulphide at marginal gabbro/sediment contact, CTSMS= semi-massive sulphide at marginal gabbro/sediment contact, GBSMS= marginal gabbro hosted semi-massive sulphide, SEDFS= sulphide along fractures in footwall sediment, SEDMS= massive sulphide lenses in footwall sediment, SEDSMS= semimassive sulphide in footwall sediment, BSCP= barren sulphide at Claimpost showing, BSLCP= barren sulphide at Lower Claimpost showing, GYPSUM= Upper Triassic gypsum. Figure 6-4b: Plot of Cu versus S - Klu property. This plot shows a poor trend of increasing Cu with S content. The highest Cu values (>5%) occur in sulphide lenses in the footwall sediment, in massive lenses at the marginal gabbro-sediment contact and along fractures in the footwall sediments. Figure 6-4c: Plot of Ni versus MgO. The MgO range of sulphide mineralized gabbros is 5.99 to 17.94%; the average MgO is 12.77%. As is evident in this plot, there is no correlation between MgO and Ni in the suite of mineralized samples collected. Ni (ppm) vs. S% - Klu Property o MIN_TYPE-CTMS' a MIN_TYPE='CTSMS' o MIN_TYPE-GBSMS' A MIN_TYPE='GBDS' • MIN TYPE-SEDFS' » MIN_TYPE='SEDMS' • MIN TYPE='SEDSMS' A MIN_TYPE='BSCP + MIN_TYPE=,BSLCP" * MIN TYPE='GYPSUM' 3e5 2.6e5 2.2e5 1.8e5 1.4e5 1e5 60000 20000 -20000 -2 Cu (ppm) vs. S% -= Klu Property A A A o O n A £*0 + O O A A ° + 10 16 22 28 34 MIN_TYPE-MIN_TYPE: MIN_TYP& MINJYP& MINJYP& MINJYP& MIN_TYPE; MINJTYPE MINTYPE MIN TYPE CTMS' •CTSMS' •GBSMS" •GBDS' 'SEDFS' :'SEDMS' :'SEDSMS' ^BSCF ='BSLCP' ='GYPSUM' S % 35000 30000 25000 20000 '- 15000 : 10000 5000 0 -5000 -2 Ni (ppm) vs. MgO% Plot - Klu Property 6 10 MGO % 14 18 22 o MIN_TYPE='CTMS' a MINJTYPE^CTSMS' o MIN_TYPE='GBSMS' A MIN_TYPE-GBDS' • MINJTYPE-SEDFS' • MINJTYPE-SEDMS' • MIN_TYPE='SEDSMS" * MIN_TYPE='BSCP' + MIN_TYPE=-BSLCP' * MIN TYPE='GYPSUM' 108 Figures 6-4d and 6-4e: Plots of Pt and Pd versus S - Klu property. These plots do not show a trend of increasing Pt and Pd with S. A similar range of Pt and Pd values exists over a large range in S content (1-25%). Samples with extreme Pt and Pd values greater than 10,000 ppb are not all shown. Most of these samples are from Cu-rich sulphide lenses in the footwall sediments. The lack of correlation between PGE and S is unusual considering the highly chalcophile nature of PGE . Figure 6-4f: Plot of Co versus S - Klu property. This plot has very similar distribution of samples to the Ni versus S plot. Samples from the Claimpost showing do show a trend of increasing Co with S level. Pt (ppb) vs. S% - Klu Property 0 MIN TYPE^CTMS" a MIN TYPE='CTSMS' o MIN TYPE-GBSMS' A MIN TYPE-GBDS' • MIN TYPE-SEDFS' MIN TYPE-SEDMS' • MIN TYPE='SEDSMS' A MIN TYPE='8SCP' + MIN TYPE='BSI_CP\ * MIN" TYPE='GYPSUM' Pd (ppb) vs. S% - Klu Property 10000 1000 if CD 0. a o' 0. 0 MIN TYPE='CTMS' D MIN TYPE='CTSMS' O MIN TYPE-GBSMS' A MIN TYPE-GBDS' • MIN TYPE-SEDFS' • MIN TYPE-SEDMS' MIN TYPE='SEDSMS' A MIN TYPE='BSCP' + MIN TYPE='BSLCP' X MIN TYPE='GYPSUM' Co (ppm) vs. S% - Klu Property 2200 1800 1400 ft 1000 o' ° 600 200 -200 o o o A O o + A * A " A -2 10 16 S % 22 28 34 o MIN TYPE="CTMS' • MIN TYPE-CTSMS' o MIN TYPE-GBSMS' A MIN TYPE-GBDS' • MIN TYPE-SEDFS' MIN TYPE-SEDMS* • MIN TYPE^SEDSMS' A MIN TYPE='BSCP" + MIN" TYPE=*BSLCP' MIN* TYPE='GYPSUM' 109 P G E Distribution The abundance's of the various elements in the PGE suite vary systemically in the different types of Ni-Cu-PGE deposits. This is also true of Au, whose distribution in Ni-Cu-PGE deposits is also examined here. The distribution of P G E and Au in ultramafic and mafic rocks is to a large degree controlled fractionation and segregation of sulphides, olivine and chromite (Barnes et al, 1985). The distribution of Ni, Cu, PGE and Au in a particular magma is governed to a large extent by the partition coefficients (D) of the various metals between sulphide and silicate phases and the magma to sulphide ratio (R). A detailed discussion of these values and their significance is given in Naldrett (1981). P G E and Au have significantly higher partition coefficients than Ni and Cu. In the absence of sulphides, olivine and chromite cumulates become enriched in Os, Ir and Ru. Partial melting can also effect the distribution of PGEs. If sulphides are retained in the mantle during partial melting, then deposits which later form from this magma will be depleted in PGE and Au relative to Ni and Cu. Chondrite Normalized Metal Profiles Normalization of PGE, Ni, Cu and Au values in 100% sulphide by chondrite or mantle values is a common method of genetically classifying sulphide mineralization associated with mafic and ultramafic magmas (Barnes et al, 1988 and Naldrett, 1981 and 1995). The approach used in this study is to plot PGE and Au in order of decreasing melting point with Ni and Cu on either end of the plot. The PGE values from each sample are recalculated to 100% using the method of Naldrett (1995b); the 100% sulphide values are then normalized by chondrite values listed in Naldrett (1981 and 1995). The recalculation of values to 100% sulphide allows comparison of sulphide from disseminated and massive occurrences. Samples from each type of mineralization on the Klu property are recalculated to 100% sulphide and normalized by chondrite values; these values are plotted in Figures 6-4k to 6-4p. Average values form the various types of sulphide mineralization on the Klu property are shown in Figure 6-4j. Values from various Ni-Cu-PGE deposits (Archean and Proterozoic komatiites, large ingneous complexes and flood basalt related deposits) and deposits in the Kluane Belt are plotted in Figures 6-4g to 6-4i for comparison (data for deposits shown is from Naldrett, 1995 and Hulbert, 1995). The 100% sulphide/C1 chondrite profiles for the deposit types listed above will be discussed briefly to put the results from the Klu property in context. "Reef-type PGE-rich layers such as Merensky and J-M which occur in large layered complexes have near flat profiles from Ni to Ir and then rise steeply to Pt and Pd. The JM reef is notably more Pd-rich than the Merensky. The UG-2 layer has higher Os, Ir and Ru due to its chromite-rich nature. This deposit type is characterized by high R factors (10,000-100,000); this results in Pt and Pd concentrations which are higher than most types of nickel deposits with only 110 Figures 6-4g and 6-4h: Plots of metals in 1 0 0 % sulphide/chondrite - Archean and Proterozoic komatiitic deposits and deposits in large igneous complexes. Archean and Proterozoic komatiites contain elevated Ni (approx. 10X chondrite) compared with most of the PGE suite (approx. 1X chondrite). These patterns are typical of low R deposits. The Pipe Deposit in the Thompson Belt has the lowest Ni-Cu-PGE concentrations in 1 0 0 % sulphide of any of the komatiite related deposits. "Reef-type PGE-rich layers such as Merensky and J-M which occur in large layered complexes have near flat profiles from Ni to Ir and then rise steeply to Pt and Pd. The JM reef is notably more Pd-rich than the Merensky. The UG-2 layer has higher Os, Ir and Ru due to its chromite-rich nature. This deposit type is characterized by high R factors (10,000-100,000); this results in Pt and Pd concentrations which are higher than most types of nickel deposits with only slightly higher or normal Ni concentrations. //Oo Archean and Proterozoic Komatiitic Deposits -100% Sulphide/Chondrite 10000.00 f- - ; ; ; • : • ! — - i : r-1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS CN RU CN PT CN AU CN : - e - . Lunnon • D : Juan ~ o - Nepean • Katiniq Donaldson pjpe Deposits in Large Igneous Complexes and Archean Komatiites -100% Sulphide/Chon. 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS_CN RU CN PT CN AU CN - e - Merensky •Q- UG-2 - o - ' JMReef Langmuirl Trojan I l l Figure 6-4i: Plot of metals in 100% sulphide/chondrite - flood basalt related deposits. Pd is up to 100X chondrite, while Ir is mainly 1X chondrite or less. Differences between the various deposits are likely related to fractionation. Insizwa appears to be the least fractionated of the group. Legend: lnsizwaM= Main Insizwa deposit, lnsizwaCu= Insizwa Cu-rich zone, TalnakhlAvg= Talnakh Average, TalnakhPic= Talnakh picritic gabbro hosted ore, TalnakhTax= Talnakh taxitic gabbro hosted ore, NorilskTax= Norilsk taxitic gabbro hosted ore. Figure 6-4j: Plot of metals in 100% sulphide/chondrite - Klu property. Samples from the Klu property have normalized metal patterns similar to those of flood basalt related deposits such as Noril'sk and Insizwa. While the various types of sulphide mineralization on the Klu property have 100% sulphide/chondrite normalized profiles which are similar to flood basalt related deposits, they show large variations in Pt, Pd, Au and Cu abundances. Flood Basalt Related Depoists (Naldrett, 1995) - 100% Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 i y . • / \ \a i 1 /.k-.-z M ...-•g/ « -^ra •"" •Vt\ *^~~~<> s /.••••m' NI_CN IR_CN RH_CN PD_CN CU_CN OS_CN RU CN PT CN AU CN GreatLakes o InsizwaM o InsizwaCu - * - TalnakhAvg • • • TalnakhPic •;»•-:: TalnakhTax -•— NonlskTax Klu - All Sulphide Types -100% Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS_CN RU CN PT CN AU CN Avg. CTMS D Avg. GBSDS - o - Avg. GBSMS -ft- Avg. SEDFS Avg. SEDMS 112 Figure 6-4k: Plot of metals in 100% sulphide/chondrite - Kluane Belt. A large variation exists between the Main Zone at Canalask and the remainder of the data. All P G E concentrations at Canalask are extremely low; particularly the Os, Ir and Ru. The profiles of the remainder of the data are generally similar to flood basalt related deposits. The Wellgreen East Zone Massive Sulphide Ore has a relatively flat profile with elevated Cu indicating it is among the least fractionated. The most fractionated mineralization types are the peridotite, dunite and gabbro hosted types. These profiles display elevated Pt, Pd, Au and Cu. The data for the Klu property samples also has a similar pattern to mineralization at Wellgreen, suggesting the mineralization at both Wellgreen and Klu is related to flood basalt magmatism. Legend: CanaSMS= Canalask deposit semi-massive sulphide, CanaPRD= Canalask deposit peridotite hosted mineralization, WelIDUN= Wellgreen deposit dunite hosted mineralization, WellGB= Wellgreen deposit gabbro hosted mineralization, WellMS= Wellgreen deposit massive sulphide, WellPRD= Wellgreen deposit peridotite hosted mineralization, WellSMS= Wellgreen deposit semi-massive sulphide. Figure 6-4I: Plot of metals in 100% sulphide/chondrite - Klu property: sediment hosted massive sulphide. This mineraliztion type has Os, Ir and Ru concentrations well below chondritic and strongly elevated Pt, Pd, Au and Cu abundances. Kluane Belt (from Hulbert, 1995) - 100% Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS_CN RU CN PT CN AU CN -e- CanaSMS ;Q CanaPRD ' o WellDUN WellGB ' . WellMS •••»-'• WellPRD -»- WellSMS Klu - Sediment Hosted Massive Sulphide -100% Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS_CN RU_CN PT_CN AU_CN -e- Rx225655 Q Rx225720 oRx222658 113 Figure 6-4m: Plot of metals in 100% sulphide/chondrite - Klu property: sediment hosted fracture mineralization. This mineraliztion type has Os, Ir and Ru concentrations well below chondritic and strongly elevated Pt, Pd, Au and Cu abundances. Figure 6-4n: Plot of metals in 100% sulphide/chondrite - Klu property: gabbro hosted semi-massive sulphide. Semi-massive sulphide mineralization at the sediment gabbro contact and massive sulphide lenses in the gabbro have the flattest profiles are can be considered the least fractionated. 113* Klu - Sediment Hosted Fracture Su lphide - 1 0 0 % Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 N I _CN I R _ C N R H _ C N P D _ C N C U _ C N O S _ C N R U C N PT C N A U C N - o - Rx225654 Q- Rx225657 o Rx225719 Klu - Gabbro Hosted Semi-Mass i ve Su lphide - 1 0 0 % Sulphide/Chondrite 10000.00 114 Figure 6-4o: Plot of metals in 100% sulphide/chondrite - Klu property: gabbro hosted disseminated sulphide. This mineraliztion type has Os, Ir and Ru concentrations well below chondritic and strongly elevated Pt, Pd, Au and Cu abundances. Figure 6-4p: Plot of metals in 100% sulphide/chondrite - Klu property: massive sulphide at the gabbro/sediment contact. Massive sulphide mineralization at the gabbro/sediment contact and massive sulphide lenses in the gabbro have the flattest profiles are can be considered the least fractionated. Klu - Gabbro Hosted Disseminated Sulphide -100% Sulphide/Chondrite 10000.00 | ; ' ; • : : : , : , 1000.00 100.00 10.00 1.00 0.10 0.01 NI_CN IR_CN RH_CN PD_CN CU_CN OS CN RU CN PT CN AU CN -o- RX222609 •a Rx222671 o Rx222673 -A-- Rx223182 Klu - Contact Massive Sulphide -100% Sulphide/Chondrite 10000.00 1000.00 100.00 10.00 1.00 0.10 0.01 P • if : / . / - / : i : * / r / //' .•• .•• / ,7 ; i ' . / •/ •••cf ' ! \ Ali \ /  // I — • I 0 \ yr. '•. \ I •• V ; • , / •••• • - • — \ v i O —a'.'' NI_CN IR_CN RH_CN PD_CN CU_CN 0S_CN RU CN PT CN AU CN: -e- Rx222607 a- Rx222604 o Rx222668 -ft- Rx222669 Rx222608 115 slightly higher or normal Ni concentrations. In contrast, Archean and Proterozoic komatiites contain elevated Ni (approx. 10X chondrite) compared with most of the PGE suite (approx. 1X chondrite). These patterns are typical of low R deposits. Variations among komatiites are mainly due to post-magmatic alteration. The Pipe Deposit in the Thompson Belt has the lowest Ni-Cu-PGE concentrations in 100% sulphide of any of the komatiite related deposits. This is probably due to significant contamination by synsedimentary sulphide. Profiles for flood basalt related reposits are relatively fractionated (Figure 1e). Pd is up to 100X chondrite, while Ir is mainly 1X chondrite or less. Differences between the various deposits are likely related to fractionation. Insizwa appears to be the least fractionated of the group. This may be related to the early saturation of sulphides which has the effect of preserving the metal pattern (Barnes et al, 1988). The Kluane data (Hulbert, 1995) shows a large variation between the Main Zone at Canalask and the remainder of the data. All PGE concentrations at Canalask are extremely low; particularly the Os, Ir and Ru. This confirms the theory that Canalask is probably a low-temperature replacement deposit. The profiles of the remainder of the data are generally similar to flood basalt related deposits. The Wellgreen East Zone Massive Sulphide Ore has a relatively flat profile with elevated Cu indicating it is among the least fractionated. The most fractionated mineralization types are the peridotite, dunite and gabbro hosted types. These profiles display elevated Pt, Pd, Au and Cu. Samples from the Klu property have normalized metal patterns similar to those of flood basalt related deposits such as Noril'sk and Insizwa. The data for the Klu property samples also has a similar pattern to mineralization at Wellgreen, suggesting the mineralization at both Wellgreen and Klu is related to flood basalt magmatism. While the various types of sulphide mineralization on the Klu property have 100% sulphide/chondrite normalized profiles which are similar to flood basalt related deposits, they show large variations in Pt, Pd, Au and Cu abundances. It must be noted that the Klu profiles are from individual samples whereas the data from profiles shown for comparison approximate deposit averages. The most fractionated mineralization types on the Klu property are sediment hosted massive and fracture filling sulphides and disseminated sulphides in gabbro. These mineralization types have Os, Ir and Ru concentrations well below chondritic and strongly elevated Pt, Pd, Au and Cu abundances. Semi-massive sulphide mineralization at the gabbro/sediment contact and massive sulphide lenses in the gabbro have the flattest profiles and can be considered the least fractionated. 116 Metal Ratio Diagrams Another method of examining the distribution of metals in Ni-Cu-PGE deposits is the use of metal ratio diagrams. The ratios used here (Pd/lr vs. Ni/Cu and Ni/Pd vs. Cu/lr) are from Barnes et al (1988). This approach is much easier for visually comparing data. The major premise for these diagrams is that Ir (and Os) is less soluble than Pt, Pd in silicate magmas (Barnes et al, 1985). This likely causes PGE fractionation during partial melting or crystal fractionation. A Pd/lr vs. Ni/Cu diagram is shown for various Ni-Cu-PGE deposits (Archean and Proterozoic komatiites, large igneous complexes and flood basalt related deposits) and deposits in the Kluane Belt in Figures 6-4q (data is from Naldrett, 1995 and Hulbert, 1995). A Pd/lr vs. Ni/Cu diagram for samples from the Klu property is shown in Figure 6-4r. Vectors are shown for trends resulting from partial melting, chromite removal, olivine addition, chromite addition and olivine removal as per Barnes et al (1988). In these plots, unfractionated mineralization is represented by low Pd/lr ratios. The Archean komatiite deposits plot in a distinct field on both diagrams which is governed by the olivine-rich character of the magma. This mineralization type has the highest Ni/Cu ratios and low Pd/lr. Most samples from the Klu property plot in the same part of the diagram as flood basalt related deposits such as Noril'sk and Insizwa. Certain mineralization types from Wellgreen also plot in this part of the diagram. Sulphide mineralization associated with dunite at Wellgreen and peridotite at Canalask plot in the same part of the diagram as some of the komatiite hosted deposits (Donaldson and Katiniq). The data from the Klu property cover a large field in the Pd/lr vs. Ni/Cu diagram with samples of sediment hosted sulphide lenses and disseminated sulphide in marginal gabbro showing high degrees of fractionation possibly due to olivine and chromite removal. A Ni/Pd vs. Cu/lr diagram is shown for various Ni-Cu-PGE deposits (Archean and Proterozoic komatiites, large igneous complexes and flood basalt related deposits) and deposits in the Kluane Belt in Figure 6-4s (data is from Naldrett, 1995 and Hulbert, 1995). A Ni/Pd versus Cu/lr diagram for samples from the Klu property is shown in Figure 6-4t. Vectors are shown for trends resulting from sulphide removal, chromite removal, olivine addition, chromite addition and olivine removal as per Barnes et al (1988). Large layered complexes form the most discrete field on the Ni/Pd vs. Cu/lr plot. These deposits have both low Ni/Pd and Cu/lr ratios. The UG-2 deposit has the lowest Cu/lr ratio reflecting its chromite-rich nature. Most of the differences in the flood basalt field in Figure 6-4s are due to increasing sulphide fractionation which causes higher Cu/lr ratios. In the Ni/Pd vs. Cu/lr diagram, samples from Klu property showings cover a large field which extends from flood basalt related deposits to Archean komatiite hosted deposits. A group of samples from the Klu 117 F i g u r e s 6-4q a n d 6-4r: Plots of Pd/lr versus Ni/Cu for various Ni-Cu-PGE deposits and the Klu property. Vectors are shown for trends resulting from partial melting, chromite removal, olivine addition, chromite addition and olivine removal as per Barnes et al (1988). In these plots, unfractionated mineralization is represented by low Pd/lr ratios. The Archean komatiite deposits plot in a distinct field on both diagrams which is governed by the olivine-rich character of the magma. This mineralization type has the highest Ni/Cu ratios and low Pd/lr. Most samples from the Klu property plot in the same part of the diagram as flood basalt related deposits such as Noril'sk and Insizwa. Certain mineralization types from Wellgreen also plot in this part of the diagram. Sulphide mineralization associated with dunite at Wellgreen and peridotite at Canalask plot in the same part of the diagram as some of the komatiite hosted deposits (Donaldson and Katiniq). The data from the Klu property cover a large field in the Pd/lr vs. Ni/Cu diagram with samples of sediment hosted sulphide lenses and disseminated sulphide in marginal gabbro showing high degrees of fractionation possibly due to olivine and chromite removal. 1 0e5 Pd/lr vs. Ni/Cu Plot of Various Ni-Cu-PGE Deposits JM Reef Great Lakes q, Talnakh Tax 0 ° Talnakh Pic Norilsk Tax Canalask SMS Insizwa Cu Talnakh Avg Wc ellgreen Dun Donaldson 0 0 Katiniq Wellgreen Prd ° Merensky L m n o n Wellgreen GB n Wellgreen SMS UG-2 Insizwa M 0 Wellgreen MS Juan _ Canalask Prd Trojan Langmuir 1 o o ) o Nepean o Pipe 0.1 1.0 10.0 NI CU 100 0 1000.0 Klu Property Ni-Cu-PGE Mineralization cc o Cu 1.0e5 10000.0 1000.0 100.0 10.0 1.0 0.1 partial melting trend ^ r . olivine removal • o chromite removal olivine addition chromite addition .1 1.0 10.0 NI CU 100.0 o MIN TYPE='CTMS' a MIN_TYPE='CTSMS' o MIN TYPE='GBSMS' A MIN_TYPE='GBDS' • Ml N_TYPE='SEDFS' 1000 0 m MIN_TYPE='SEDMS' 1 1 8 Figures 6-4s and 6-4t: Plots of Ni/Pd versus Cu/lr for various Ni-Cu-PGE deposits and the Klu property. Vectors are shown for trends resulting from sulphide removal, chromite removal, olivine addition, chromite addition and olivine removal as per Barnes et al (1988). Large layered complexes form the most discrete field on the Ni/Pd vs. Cu/lr plot. These deposits have both low Ni/Pd and Cu/lr ratios. The UG-2 deposit has the lowest Cu/lr ratio reflecting its chromite-rich nature. Most of the differences in the flood basalt field in this plot are due to increasing sulphide fractionation which causes higher Cu/lr ratios. Samples from Klu property showings cover a large field which extends from flood basalt related deposits to Archean komatiite hosted deposits. A group of samples from the Klu property of several types of mineralization show a similar trend to the sulphide fractionation trend in the world-wide flood basalt related deposits field. ne* 1e6 i 1eS 10000 Ni/Pd vs. Cu/lr of Various Ni-Cu-PGE Deposits Q a Pipe o Nepean Juan o o o Lunnon Langmulrl o o Trojan Wellgreen MS Q Wellgreen SMS ° Canalask Prd Talnakh Avg o Katlnlq 1000 Merensky Insizwa M Insizwa Cu o 9 Wellgreen Dun ° ° Q Wellgreen PRD Wellgreen GB Donaldson Talnakh Pic 0 o ° G r e a t U k e s Norilsk Tax „ TalnahkTax UG-2 100 Klu Property - Ni-Cu-PGE Mineralization 1 e 6 1e5 £ 10000 1 0 0 0 100 olivine addition X-chromite addition ^ sulphide removal chromite removal 1 0 0 0 10000 o MIN_TYPE='CTMS' • MIN_TYPE='CTSMS' o MIN_TYPE-GBSMS' A MIN_TYPE='GBDS' • MIN_TYPE='SEDFS, * MIN TYPE='SEDMS' CUJR 119 property of several types of mineralization show a similar trend to the sulphide fractionation trend in the world-wide flood basalt related deposits field. Sulphur Isotopic and Selenium Geochemistry Nickel sulphide mineralization in many districts (such as: Kambalda, Noril'sk, Thompson, Sudbury) appears to be related the country rock-magma interactions. This interaction often involves assimilation of sulphur-rich country rocks. Sulphur isotopic techniques have been used by numerous workers to determine if Ni-sulphide ores have sulphur isotopic signatures similar to: footwall country rocks, magmatic sulphur or a mixture of country rock and magmatic sulphur. In some districts such as Thompson (Eckstrand et al, 1989), there is no significant variation between the S isotopic composition of Ni sulphides and country rock sulphide. Here, another geochemical indicator is required to determine if country rock sulphur has been incorporated into the Ni-sulphides. In other districts such as Noril'sk, where sulphur-rich country rock is a plausible contaminant, other geochemical indicators are needed to confirm the sulphur source. S/Se ratios can be used for this purpose. S/Se ratios are particularly useful in Archean and Proterozoic rocks due the limited range of sulphur isotopic compositions due to the lack of biogenic fractionation of sulphur. Se is a strongly chalcophile element which has been shown to be preferentially enriched in magmatic sulphides relative to sedimentary sulphides (Loftus-Hills and Solomon, 1967). Se is less soluble and less mobile than sulphur and is usually depleted in clastic sedimentary rocks. Several studies report S/Se ratios of country rock-sulphide from districts with Ni-sulphide mineralization. Thompson (1982) reported S/Se values from the Hildreths Formation in Maine to range from 36,833 to 48,500. Hulbert (1995) gives range of 10,000 to 20,000 as being typical for Permian sediments in the Kluane Belt of the Yukon. Black shales tend to have lower S/Se ratios than most clastic rocks; Hulbert et al (1992) reports that Devonian black shales from the Selwyn Basin in Yukon range from <1000 to 6000. No selenium data were found in the literature for sulphates; Kluane Belt (U. Triassic) sulphates have S/Se ratios >700,000 (1995 Inco Limited data). There is no firm magmatic range for S/Se ratios, but most magmatic sulphides have ratios between 0 and 3,500 (Hulbert, 1995). 8 3 4S values in magmatic sulphides range from -5 to +5 7 0 0 and average approximately 0 °lQO (Faure, 1986). Typical mantle values are from -1 to +1 °l00. Sedimentary sulphates have 8 3 4S values ranging from +10 to +307 o o depending on geological age, while sedimentary sulphides typically range from -30 to +5 7^ (Faure, 1986). A 5 3 4S box and whisker plot for the various sulphide mineralization types as well as sulphate beds on the Klu property is shown in Figure 6-4u. Sulphides from 120 Figure 6-4u: Box and Whisker Plot of S^S sorted by mineralization type. Sulphides from the Ni-Cu-PGE showings have median values between approximately -1.5 and -7 700; disseminated sulphide in the marginal gabbro has the values closest to 0 700. Sulphides from the Upper and Lower Claimpost showings have median values of approximately -27 and -13 7 0 0 respectively. Based on their S/Se and 834S values, these sulphides can thus be regarded as potential sources of country rock sulphur. The high positive 834S values of the gypsum (median approx. +15700) precludes it from being a potential contaminant (as well, the gypsum is believed to be slightly younger than the sills associated with the Ni-Cu-PGE mineralization). Figure 6-4v: Box and Whisker Plot of S/Se sorted by mineralization type. The gypsum samples have extremely high estimated S/Se ratios (>700,000). All types of Ni-Cu-PGE mineralization on the property (massive sulphide and semi-massive sulphide from the gabbro-sediment contact, gabbro hosted disseminated and semi—massive sulphide and sediment hosted semi-massive and massive sulphide) except one sample of semi-massive sulphide from the contact have S/Se ratios close to assumed mantle values (<3,500). Disseminated sulphide in the marginal gabbro has the lowest S/Se ratios. Ni-Cu-PGE barren sulphide which is believed to be synsedimentary in origin (Claimpost and Lower Claimpost sulphide occurrences) has median S/Se ratios of approximately 6,000 and 7,000 respectively which are generally above the range of S/Se levels in the Ni-Cu-PGE bearing sulphides. Del 34S - Mineralization Type Box and Whisker Plot 25 | • • • ; , ; , ; , 1 15 s CTSMS GBDS BG SEDFS BSLCP SEDSMS CTMS BGDS GBSMS BSCP SEDMS GYPSUM Non-Outlier Max Non-Outlier Min 75% 25% Median Extremes MIN TYPE S/Se - Mineralization Type Box and Whisker Plot @ I Non-Outlier Max Non-Outlier Min CZ3 75% 25% • Median CTSMS GBDS BG SEDFS BSLCP SEDSMS CTMS BGDS GBSMS BSCP SEDMS GYPSUM ° Outliers MIN TYPE 121 the Ni-Cu-PGE showings have median values between approximately -1.5 and -7 7 0 0; disseminated sulphide in the marginal gabbro has the values closest to 0 7 0 0. Sulphides from the Upper and Lower Claimpost showings have median values of approximately -27 and -13 7 0 0 respectively. Based on their S/Se and 8 3 4S values, these sulphides can thus be regarded as potential sources of country rock sulphur. The high positive 8 3 4S values of the gypsum (median approx. +15 700) precludes it from being a potential contaminant (as well, the gypsum is believed to be slightly younger than the sills associated with the Ni-Cu-PGE mineralization). Figure 6-4v is a S/Se box and whisker plot for the various sulphide and sulphate types on the property. The gypsum samples have extremely high estimated S/Se ratios (>700,000). The Se was below the detection limit of 0.5 ppm, so one half of the detection limit was used in the ratio. All types of Ni-Cu-PGE mineralization on the property (massive sulphide and semi-massive sulphide 7 0 0 from the gabbro-sediment contact, gabbro hosted disseminated and semi— massive sulphide and sediment hosted semi-massive and massive sulphide) except one sample of semi-massive sulphide from the contact have S/Se ratios close to assumed mantle values (<3,500). Disseminated sulphide in the marginal gabbro has the lowest S/Se ratios. Ni-Cu-PGE barren sulphide which is believed to be synsedimentary in origin (Claimpost and Lower Claimpost sulphide occurrences) has median S/Se ratios of approximately 6,000 and 7,000 respectively which are generally above the range of S/Se levels in the Ni-Cu-P G E bearing sulphides. Sulphur is plotted versus 5 3 4S and Se in Figures 6-4w and 6-4x. The S vs. 8 3 4S plot shows the 8 3 4S values in the Ni-Cu-PGE enriched sulphides remaining relatively constant over a large range in sulphur contents. This plot also demonstrates the large range of 8 3 4S over similar S concentrations between the samples of gypsum, the Ni-Cu-PGE rich and Ni-Cu-PGE barren sulphides. The Se 8 3 4S vs. S plot (an expanded version of the plot at lower Se concentrations is shown in Figure 6-4y) shows an overall trend of increasing Se with increasing S concentrations. A subset of samples of Ni-Cu-PGE barren sulphide from the Claimpost and Lower Claimpost occurrences does not follow the overall trend, but instead has a flat slope. That is, the Se levels in these samples increase with increasing S. Co, Ni, Pt, Zn, Cu, As, Sb and S/Se are plotted versus 8 3 4S in Figures 6-4z to 6-4ag. The Co, Ni, Pt and Cu values show distinct maximums at 8 3 4S values of approximately -37^,. This suggests that mixing of synsedimentary and magmatic sulphide is important, but after a certain level of addition of synsedimentary sulphur, this process serves to dilute metal grades. The S/Se versus 8 3 4S plot suggests that the sulphides from the Ni-Cu-PGE showings fall along a crude binary mixing line between synsedimentary sulphide occurrences 122 Figures 6-4w, 6-4x and 6-4y: Plots of S versus Se and 5 3 4S - Klu property. The S vs. 8 3 4S plot shows the 8 3 4S values in the Ni-Cu-PGE enriched sulphides remaining relatively constant over a large range in sulphur contents. This plot also demonstrates the large range of 8 3 4S over similar S concentrations between the samples of gypsum, the Ni-Cu-PGE rich and Ni-Cu-PGE barren sulphides. The Se 8 3 4S vs. S plot shows an overall trend of increasing Se with increasing S concentrations. A subset of samples of Ni-Cu-PGE barren sulphide from the Claimpost and Lower Claimpost occurrences does not follow the overall trend, but instead has a flat slope. Se levels in these samples increase with increasing S. /22a S% vs. Del 34S (per mil) - Klu Property 34 28 22 •# 16 CO 10 4 -35 • <* o O m * * A* + o ° 5 A v A A A A -25 -15 -5 DEL 34S 15 25 0 MIN TYPE='CTMS' • MIN TYPE='CTSMS' o MIN* TYPE-GBSMS' A MIN* TYPE-GBDS' • MIN TYPE-SEDFS' a MIN TYPE-SEDMS' * MIN TYPE='SEDSMS' A MIN TYPE-BSCP' + MIN* TYPE='BSLCP' MIN TYPE='GYPSUM' Se (ppm) vs. S% - Klu Property • J * * + O O 0 A 4 W « p * A A + M -2 10 16 S % 22 28 34 o MIN TYPE= 'CTMS' • MIN* TYPE= 'CTSMS' o MIN* *TYPE= 'GBSMS' A MIN TYPE= 'GBDS' • MIN* TYPE= 'SEDFS' 8 MIN TYPE= 'SEDMS'  MIN TYPE= 'SEDSMS' A MIN TYPE= 'BSCP' + MIN* *TYPE= 'BSLCP' * MIN* TYPE= 'GYPSUM' Se (ppm) vs S% - Klu Property -2 0 *a • A o A A o O A A + • O 0 0 + A A m 10 16 S % 22 28 34 o MIN TYPE='CTMS' a MIN TYPE='CTSMS' « MIN* TYPE-GBSMS' A MIN TYPE-GBDS' . • MIN TYPE-SEDFS' MIN TYPE-SEDMS' • MIN TYPE='SEDSMS' A MIN TYPE='BSCP" MIN TYPE='BSLCP' MIN" TYPE='GYPSUM' 123 Figures 6-4z, 6-4aa and 6-4ab: Plots of Co, Ni and Pt versus 834S - Klu property. Co, Ni and Cu values show distinct maximums at 834S values of approximately -3700. This suggests that mixing of synsedimentary and magmatic sulphide is important, but after a certain level of addition of synsedimentary sulphur, this process serves to dilute metal grades. /23* Co (ppm) vs. Del 34S - Klu Property 2200 1800 1400 a 1000 o 1 ° 600 200 -200 o .A o o 0 * O A + . " * a A -30 -24 -18 -12 DEL 34S o a o A • . • A + * MIN_TYPE; MIN TYPE: MIN_TYPE: MIN_TYP& MIN_TYPE: MIN_TYPE= MIN TYPE: MIN~TYPE= MIN_TYPE= MIN TYPE: ='CTMS' ='CTSMS' ='GBSMS' ='GBDS' ='SEDFS' ='SEDMS' '^SEDSMS' ;'BSCP' 'BSLCP' :'GYPSUM' Ni (ppm) vs. Del 34S - Klu Property -30 0 Ai A © o o • D O O o -24 -18 -12 DEL 34S o MIN TYPE='CTMS' a MIN~TYPE='CTSMS' » MINJTYPE-GBSMS' A MIN TYPE-GBDS' • MINJYPE='SEDFS' « MIN_TYPE='SEDMS' • MINjrYPE='SEDSMS' A MINJYPE^ BSCP' + MIN_TYPE='BSLCP' * MIN TYPE='GYPSUM' Pt (ppb) vs. Del 34S - Klu Property O ^ A " AA A + • + <» o A 4 -30 -24 -18 -12 o MIN_TYPE='CTMS' • MIN_TYPE='CTSMS' o MIN_TYPE-GBSMS' A MI N_TYPE-GBDS' • MIN_TYPE-SEDFS' « MIN_TYPE-SEDMS' • MIN_TYPE='SEDSMS' A MIN TYPE='BSCP' + MIN_TYPE='BSLCP' * MIN_TYPE='GYPSUM' DEL_34S 124 Figure 6-4ac: Plot of Zn versus 8 3 4S - Klu property. This plot supports the addition of synsedimentary sulphide, but the data plots in a dispersed pattern and can not be used to discriminate between mixing with either the Upper Claimpost or Lower Claimpost occurrences. Figure 6-4ad: Plot of Cu versus 8 3 4S - Klu property. Cu values show a distinct maximum at 8 3 4S values of approximately -3 7 0 0 . This suggests that mixing of synsedimentary and magmatic sulphide is important, but after a certain level of addition of synsedimentary sulphur, this process serves to dilute metal grades. /2V« Zn (ppm) vs. Del 34S - Klu Property 550 450 350 t 250 z1 N 150 50 -50 0 OSS A + A + 0 0 o A A ° ^ A o A O * * * -35 -25 -15 -5 5 DEL 34S 15 25 o a o A MIN TYPE= MIN_TYPE= MIN_TYPE= MIN_TYPE= MIN_TYPE= MIN TYPE= MIN TYPE= MIN_TYPE= MIN TYPE1 MIN TYPE= 'CTMS' •CTSMS' 'GBSMS' !'GBDS' 'SEDFS' ''SEDMS' •SEDSMS' 'BSCPV 'BSLCP' 'GYPSUM' Cu (ppm) vs. Del 34S - Klu Property 3e5 2.6e5 2.2e5 1.8e5 2 8: 1.4e5 a 1e5 60000 20000 -20000 • 0 M A +£ 3 + <*> 0 § o A -30 -24 -18 -12 DEL 34S MIN_TYPE= MIN_TYPE= MIN TYPE' MIN TYPE= MIN_TYPE: MIN_TYPE= MIN_TYPE= MIN_TYPE= MIN_TYPE: MIN TYPE= 'CTMS' 'CTSMS' 'GBSMS' :'GBDS' :'SEDFS' 'SEDMS' 'SEDSMS' ;'BSCP' 'BSLCP' •'GYPSUM' 125 Figures 6-4ae, 6-4af and 6-4ag: Plots of As, Sb and S/Se versus 534S - Klu property. The S/Se versus 834S plot suggests that the sulphides from the Ni-Cu-PGE showings fall along a crude binary mixing line between synsedimentary sulphide occurrences in the Hasen Creek Formation and mantle-type sulphide. The Upper Claimpost showing would seem to be the better candidate for the contaminant due to its low As and Sb concentrations and the low As and Sb levels of most of the Ni-Cu-PGE rich sulphides. IZCc, As (ppm) vs. Del 34S - Klu Property MIN_JYPE='CTMS' MIN TYPE='CTSMS' M1N_TYPE~ GBSMS' MIN_TYPE='GBDS' MIN_TYPE='SEDFS' MIN_TYPE='SEDMS' MIN_TYPE='SEDSMS' MIN_TYPE='BSCP' MIN_TYPE='BSLCP' MIN TYPE='GYPSUM' Sb (ppm) vs. Del 34S - Klu Property 55 45 35 8: 25 cd' w 15 5 • c 1 i i A + + CO o 0 O ^ A / S A l A -30 -24 -18 -12 DEL 34S o MIN TYPE='CTMS-• MIN TYPE='CTSMS' o MIN TYPE='GBSMS' A MIN TYPE-GBDS' • MIN TYPE-SEDFS' a MIN" TYPE-SEDMS' • MIN TYPE='SEDSMS' A MIN TYPE='BSCP' + MIN TYPE='BSLCP' MIN TYPE='GYPSUM' S/Se vs. Del 34S - Klu Property 18000 14000 10000 6000 2000 2000 + A* A «• 0 a A if! A A -30 -24 -18 -12 DEL 34S 0 MIN TYPE='CTMS' D MIN" TYPE='CTSMS' o MIN" TYPE-GBSMS' A MIN TYPE-GBDS' • MIN TYPE-SEDFS' tt :' MIN TYPE-SEDMS' » MIN* TYPE='SEDSMS' A MIN TYPE='BSCP' + MIN* TYPE='BSLCP' X MIN TYPE='GYPSUM' 126 in the Hasen Creek Formation and mantle-type sulphide. The Upper Claimpost showing seems to be the better candidate for the contaminant due to its low As and Sb concentrations and the low As and Sb levels of most of the Ni-Cu-PGE rich sulphides. The Zn versus 8 3 4S plot supports the addition of synsedimentary sulphide, but the data plots in a dispersed pattern and can not be used to discriminate between mixing with either the Upper Claimpost or Lower Claimpost occurrences. A subset of the data from semi-massive and massive sulphide from the gabbro-sediment contact and massive sulphide in the footwall do contain elevated As and Sb. This raises the possibility of a multi-stage non-binary sulphur mixing scenario. For example, the sulphides within the gabbro may have assimilated sulphur from a different source at an earlier time than the sulphide at the contact and in the footwall. Two possible binary sulphur mixing scenarios are modeled in Figures 6-4ah and 6-4ai to illustrate the proportions of synsedimentary sulphide addition using S/Se versus 8 3 4S plots. The mixing scenarios shown model the mixing of 100% sulphide with 3 8 % weight percent sulphide in each end member. Figure 6-4ah shows the mixing curve for adding sulphide from the Lower Claimpost occurrence (-11.45 8 3 4S and 6,891 S/Se) to magmatic sulphide (0 8 3 4S and 1,450 S/Se). The S/Se level of magmatic (mantle) sulphide was picked to be 1,450 based on the low end of S/Se in the marginal gabbro. Figure 6-4ai shows the mixing of magmatic sulphide with sulphide from the Claimpost showing (-26.78 8 3 4S 7 0 0 and 5,447 S/Se). Of the two scenarios, the mixing line in the Lower Claimpost-mantle diagram fits the Klu data the best. In this diagram, the Klu samples represent the addition of between 20 and 8 0 % Lower Claimpost-type sulphur. The 8 3 4S value of -3 7 0 0 which is the value at which maximum metal values occur, would represent between 25 and 3 0 % mixing of Lower Claimpost-type sulphide with magmatic sulphide. On the Claimpost-mantle mixing curve, the Klu samples represent between 5 and 4 0 % addition of Claimpost-type sulphide to mantle sulphide. The 8 3 4S value of -3 7 0 0 represents a contribution of approximately 1 0 % Claimpost-type sulphide. Germane to this discussion are the S/Se and 8 3 4S values from peridotite in the core of the Spy Sill. The 8 3 4S values in the peridotite have average approximately -57 0 0 and have S/Se values between 1,500 and 4,000. The variation between these values and that observed in disseminated marginal gabbro (83 4S of -1 7 0 0 and S/Se of 1,487) suggests that contamination of the low-level sulphide in the peridotite took place in a different location with a different contaminant. Campbell (1981) studied the sulphur isotopic signature of the Wellgreen deposit. The average 8 3 4S values from 22 samples of massive and disseminated sulphide 127 Figure 6 -4ah : S/Se - 8 3 4S Mixing Model (Lower Claimpost - mantle). This model shows the mixing curve for adding sulphide from the Lower Claimpost occurrence (-11.45 7 ^ 8 3 4S and 6,891 S/Se) to magmatic sulphide (O7 0 0 5 3 4S and 1,450 S/Se). This mixing curve shows the Klu samples representing the addition of between 20 and 8 0 % Lower Claimpost-type sulphur. The 5 3 4S value of -3 7 0 0 which is the value at which maximum metal values occur, would represent between 25 and 3 0 % mixing of Lower Claimpost-type sulphide with magmatic sulphide. Figure 6 -4a i : S/Se - 8 3 4S Mixing Model (Claimpost - mantle). This model shows the mixing of magmatic sulphide (0 8 3 4S and 1,450 S/Se) with sulphide from the Claimpost showing (-26.78 7 0 0 8 3 4S and 5,447 S/Se). This mixing curve shows the Klu samples representing between 5 and 4 0 % addition of Claimpost-type sulphide to mantle sulphide. The 8 3 4S value of - 3 7 0 0 represents a contribution of approximately 1 0 % Claimpost-type sulphide. S/Se - Del34S Mixing Curve (Lower Claimpost - Mantle) -25 -20 -15 DEL 34S 10 MIN_TYPE= MIN_TYPE= MIN_TYPE= MIN_TYPE= MIN TYPE= MIN TYPE= MIN TYPE= MIN_TYPE= MIN_TYPE= MIN TYPE= MIN TYPE1 •CTMS 'CTSMS' 'GBSMS' 'GBDS' 'SEDFS' 'SEDMS' 'SEDSMS' 'BSCP' 'BSLCP' 'GYPSUM' 'LCP-Mix' S/Se - Del34S Mixing Curve (Claim Post - Mantle) 7000 6500 6000 5500 5000 4500 4000 co 3500 co' 3000 2500 2000 1500 1000 500 A -35 -30 -25 -20 -15 DEL34S -10 -5 o MIN TYPE='CTMS' • MIN TYPE='CTSMS' « MlN_TYPE='GBSMSV A MIN_TYPE='GBDS' • MIN_TYPE='SEDFS' • MIN_TYPE='SEDMS' • MIN TYPE='SEDSMS' A MIN TYPE='BSCP' + MIN~TYPE='BSLCP' * MIN TYPE='GYPSUM' o MIN TYPE='CP-Mix' 128 was found to be -4.5 7oo. This value is in good agreement with sulphur isotopic data from Miller (1991) who reported S^S values from 30 samples of disseminated sulphide from Wellgreen which average -4.9 7^,. Campbell (1981) analysed two samples of sulphidic Hasen Creek Formation from the "transition zone" below the Wellgreen ultramafic intrusion. These samples had an average S^S value of -19 7oo- Mass balance calculations by Campbell (1981) showed that 3 2 % of the sulphur in the Wellgreen deposit could be derived from synsedimentary sulphide from the Hasen Creek Formation. Hulbert (1995) outlined several binary mixing scenarios for Ni-Cu-PGE deposits and occurrences in the Kluane Belt utilizing 5 3 4S versus S/Se diagrams. In these scenarios, Hulbert assumed the parental magma had a b^S value of 0 0loo and S/Se of 3,500. For the Wellgreen deposit, Hulbert assumed the synsedimentary contaminant sulphide had a S^S value of -19 7oo and, a S/Se ratio of 20,000. Based on the present S^S and S/Se ratios in gabbro hosted disseminated sulphide, gabbro hosted massive sulphide and ultramafic hosted sulphide; the mixing models show that 3.5 to 10% , 2.5 to 8.3% and 1.5 to 9 % synsedimentary sulphur contamination has occurred in the respective mineralization types. Using a b^S value of -29 7* , and a S/Se ratio of 20,000 for the contaminant sulphide at Canalask, Hulbert determined the massive sulphide and semi-massive sulphide mineralization in the "offset" deposit has undergone 5 to 1 1 % and 16 to 2 5 % contamination respectively. Hulbert speculates that the lower grade of the Canalask mineralization in 100% sulphide compared to that in Wellgreen is related to an excess contribution of country rock sulphide at Canalask. Hulbert's mixing models support this statement. The suggested proportion of country rock sulphur contamination at Klu is higher than that proposed by Hulbert at Wellgreen due to the different S^S and S/Se ratios used for both the mantle and country rock end-members. A comparison of the results from this study with data from other studies of Ni-Cu-PGE mineralization in the Kluane Belt shows that mantle-country rock mixing scenarios for sulphur can not be summarized on a belt-scale; local country rock and parental magma characteristics are key in developing realistic mixing models. It is also apparent that a simple binary mixing model for sulphur contamination does not reflect the multi-stage emplacement of the intrusions or the effect of several isotopically different sources of sulphur which may progressively influence the final isotopic and geochemical composition of the magma. 129 Section 7: Conclusions 7.1 Introduction The conclusions offered below attempt to describe the characteristics of the Ni-Cu-PGE mineralization on the Klu property and to determine its relationship with the Kluane-type intrusions, the country rock to the intrusions and the regional geology. To accomplish this, the geochemistry of a barren intrusion and a Ni-Cu-PGE mineralized intrusion are contrasted and the geochemistry of the Kluane-type intrusions is compared with that of the Nikolai basalt. The characteristics of the Ni-Cu-PGE mineralization in terms of metal ratios, metal abundances in 100% sulphide, metal co-relations with sulphur and MgO and sulphur isotopic composition are also summarized. Proposed models for the emplacement of the mafic to ultramafic Kluane-type intrusions, the related Nikolai basalt and the Ni-Cu-PGE mineralization are outlined. 7.2 Major Conclusions 7.2.1 Geochemical Variation between the Spy and Lewis Intrusions Ni and Pd are strongly depleted throughout the Lewis Intrusion relative to the Spy Sill; Cu and Pt are depleted in parts of the Lewis Intrusion. This depletion is not recorded (or was missed during sampling) in the preserved Nikolai basalt stratigraphy. Stratigraphic sampling of intrusions to test for metal depletion may be a more effective exploration technique than sampling partially eroded coeval basalt sequences. Highly depleted intrusions are devoid of associated mineralization whereas non-depletion implies potential for mineralization. The depletion recorded in the Lewis Intrusions appears to be independent of sulphur contents. Most samples from both intrusions have S levels between 0 and 2,000 ppm. The intrusions have similar ranges of MgO and Mg#. The Lewis Intrusion has consistently higher Cu/Pd ratios relative to the Spy Sill suggesting it is a residual material from a magma which has had its Pd preferentially stripped away relative to Cu by equilibration with sulphide. 7.2.2 Relationship between the Nikolai Volcanics and Triassic Intrusions The Nikolai basalt, peridotite, Maple Creek Gabbro and some of the marginal gabbro samples from this study show a cogenetic trend on plots of immobile, incompatible elements. A strong cogenetic trend is shown in the T i 0 2 versus Zr plot. This relationship suggests that the Nikolai basalt and Triassic mafic to ultramafic intrusions orginated from a common parental melt. The marginal 130 gabbro samples show a greater degree of contamination compared with the Nikolai basalt based on Ce/Yb, Ba/Zr and Th values. The data shows a trend of contamination increasing from basalt to peridotite and Maple Creek gabbro to marginal gabbro. This trend is probably related to the amount of interaction between the magma and country rock for each of the above units. PER diagrams show peridotite compositions controlled by olivine fractionation whereas the basalt and gabbro sample compositions are controlled by plagioclase and clinopyroxene fractionation. Certain of the marginal gabbro samples plot off the fractionation trends in the PER diagrams; this is probably the effect of contamination. Ni, Cu and Co have restricted ranges in the basalt and gabbro relative to the peridotite. 7.2.3 Metal Depletion and Contamination in Nikolai Basalt In terms of metal depletion (as indicated by Ni, Cu, Co vs. Mg and Mg# relationships), there appears to be no significant difference between the Burwash and Halfbreed basalt sections. The basalt sections sampled are not depleted in Ni and Cu to the degree that part of the stratigraphy in the Norils'sk district is. The eastern basalt facies of Wrangellia (the Nikolai basalt) is strongly depleted in Cu and Co compared with basalt from the western facies of Wrangellia (the Karmutsen basalt). Ni appears to be moderately depleted in the Nikolai basalt relative to the Karmutsen. The Nikolai basalt is overall more Mg-rich than the Karmutsen (6-10% and 5-8% respectively). LOI values for both basalt units suggest that the Nikolai has undergone a greater degree of hydrothermal alteration and metamorphism than the Karmutsen, which may account for the low Cu and Co values. Alternatively, the lack of correlation between metal concentration and LOI suggests that the low metal values are not related to alteration and metamorphism. The upper part of the Burwash section displays the greatest degree of contamination based on Th/Ta and Ba/Zr ratios. There is no co-incident metal depletion accompanying the contamination. 7.2.4 Emplacement Model for Triassic Intrusions A large volume of basaltic magma, possibly related to a mantle plume passed through Island Arc related Upper Paleozoic volcanic and sedimentary rocks in various pulses. An early gabbroic pulse lead to extensive sill development. Initial gabbroic magma passed through conduits and spread laterally in the form of sills. This material is preserved in the marginal gabbro phase. The basal part of the Nikolai basalt was likely fed by this intrusive phase. Continued flow 131 through these sills and conduits led to cumulate olivine-rich ultramafic material occupying the bulk of the sills in the Hasen Creek Formation. Later Maple Creek Gabbro sills and conduits fed the remainder of the basalt sequence along conduits largely established by earlier magma pulses. This material locally formed an extensive series of sills. The Lewis Intrusion probably represents an initial pulse of Ni-Cu-Co-PGE depleted magma from a lower chamber which was stripped of metals by equilibration with sulphide. A schematic section illustrating the above features is shown in Figure 7-1. 7.2.5 Olivine Geochemistry Olivine grains from both the Spy Sill and Lewis Intrusion have core and rim compositions which do not differ significantly. Olivine grains from the Lewis Intrusion are strongly depleted in Ni relative to their Fo#. Olivines from the Spy Sill have compositions from Fo 79.8 to 83.2 with Ni from 0.15 to 0 .21%. Olivines from the Lewis Intrusion have compositions from Fo 78 to 80.7 with Ni from 0.05 to 0 .11%. The olivines from both intrusions have overlapping Fo numbers, but the Lewis Intrusion olivines have much lower Ni suggesting that the magma represented by the Lewis Intrusion equilibrated with sulphide. Grains with maximum Fo# were used to calculate the molar MgO/FeO ratio and Mg# (Lewis: Fo55.42 and Spy: Fo59.82) of the parent melt. The Mg number of the marginal gabbro is less magnesian than predicted by the olivine composition; this appears to be due to contamination by country rock sediments. The Mg number of the Nikolai basalt (averages of 55.3 and 57.9 for the Burwash and Halfbreed sections respectively) are close to the parental melt composition predicted by the olivine compositions. 7.2.6 Sulphide Geochemistry Ni-Cu-PGE rich marginal gabbro contains higher MgO levels (average of 12.8%) than non-mineralized marginal gabbro. Within the suite of mineralized marginal gabbro, there is no correlation between MgO and Ni. The highest Ni and Co levels occur in massive sulphide lenses at the lower marginal gabbro-sediment contact. The highest Cu, Pt and Pd levels occur in massive sulphide lenses in the footwall sediment. Ni, Cu and Co display an overall increase with increasing S contents in Ni-Cu-PGE rich samples whereas Pt and Pd values correlate poorly with S. Ni-PGE-Cu-Au values in 100% sulphide/chondrite for the Spy showings show patterns which are most similar to those from flood basalt-related deposits. Schematic Section for Sill Emplacement and Ni-Cu-PGE Mineralization Figure 7-1 Chitistone - Nizina Limestone Nikola Basalt Hasen Creek Formation Station Creek Formation j-i-LJ- Synsedimentary Sulphide & Ni-Cu-PGE rich Massive Sulphide 133 Metals in 100% sulphide/chondrite patterns are highly variable between different mineralization types. The sulphide lenses and fracture fillings in the footwall sediment show elevated 100% sulphide/chondrite Cu.PGE and Au relative to the other mineralization types; this is attributed to sulphide fractionation. Semi-massive and massive sulphide in marginal gabbro have the flattest profiles suggesting they are the least fractionated. Pd/lr versus Ni/Cu and Ni/Pd versus Cu/lr plots suggest PGE distribution has an overall control from olivine and chromite fractionation and a local control from sulphide fractionation. 7.2.7 Sulphur Isotopes Sulphur isotopic data from Ni-Cu-PGE mineralization and potential sulphur sources show that the sulphur isotopic signature of the showings is the result of mixing between mantle and synsedimentary sulphur. Ni-Cu-PGE barren synsedimentary sulphide from the Hasen Formation is the probable sulphur contaminant. Plots of 634S versus As and Sb suggest that the Upper Claimpost showing type sulphide is the more likely contaminant while the 834S versus S/Se plot shows the Lower Claimpost showing type sulphide is the probable contaminant. A binary mixing model between Lower Claimpost type sulphide and mantle sulphide suggests an addition of between 20 and 80% country rock sulphur. If the Upper Claimpost showing type sulphide mixed with mantle sulphide, the country rock S contribution would be 5 to 40%. In nature, a simple binary mixing of mantle sulphide and sulphide from a particular country rock layer is not considered likely, but the contribution of some amount of country rock sulphur appears to be important in the formation of the Ni-Cu-PGE mineralization. Plots of 834S versus Ni, Cu, Co, Pt and Pd show that maximum metal values occur at 834S values of near -3 7^ . Thus, an excess contribution of country rock sulphur appears to have the effect of lowering metal grades in Ni-Cu-PGE mineralization. 7.2.8 Emplacement and Significance of Ni-Cu-PGE Mineralization Ni-Cu-PGE rich material appears to be related to an early magma pulse. Specific flows of Ni-Cu-PGE depleted magma are not recorded in the Nikolai basalt due to ancient (Triassic) erosion of such depleted flows or later tectonism. Also, there may be a lack of connectivity between the mineralized sills and magma that reached the surface. Metals were stripped from part of a lower magma chamber by sulphide equilibration; this magma is represented by the Lewis Intrusion. During sill 134 emplacement, synsedimentary sulphur was assimilated from sulphidic layers in the Hasen Creek Formation. Most Ni-Cu-PGE mineralization is associated with marginal gabbro pulses. The magma represented by the Lewis Intrusion is either an earlier sill (before the Spy Sill) or the leading edge of a magma pulse that also formed the Spy Sill which contains magma with normal or elevated Ni-Cu-PGE levels. The salient parts of the model presented here are shown in the schematic section in Figure 7-1. No economically significant quantities of Ni-Cu-PGE rich sulphide are known on the Klu property, but the nature of the mineralization suggests a process capable of producing Ni-Cu-PGE enriched sulphide took place. This is known to have occurred on a larger scale at the nearby Wellgreen deposit. The lack of serpentinization and tectonism over much of the Klu property provide an ideal location to study the Kluane-type Ni-Cu-PGE showings and associated mafic to ultramafic intrusions which occur here. 135 References Barnes, S.J., Naldrett, A.J. and Gorton, MP. (1985). The origin of the fractionation of platinum-group elements in terrestrial magmas. Chemical Geology, v. 53, pp. 303-323. Barnes, S.J., Boyd, R., Korneliussen, A., Nilsson, LP., Often. M., Pederson, R.B. and Robins, B. (1988). The Use of Mantle Normalization and Metal Ratios in Discriminating between the Effects of Partial Melting, Crystal Fractionation and Sulphide Segregation on Platinum Group Elements, Gold, Nickel and Copper: Examples from Norway in Geo-Platinum 87, edited by Pritchard, H.M., Potes, P.J. and Bowles, J.F.W., Elseiver Publishing, London, pp. 113-143. Barker, F., Southerland Brown, A., Budhan, J.R and Plafker, G. (1989). Back-arc with frontal-arc component origin of Triassic Karmutsen basalt, British Columbia, Canada. Chemical Geology, v. 53, pp. 303-323. Bell, C.C. (1996). 1995 Geological and geochemical surveys on the Klu property. Yukon Assessment Report (115G/2). Campbell, S.W. (1981). Geology and Genesis of Copper Deposits and Associated Host Rocks in and Near the Quill Creek Area, Southwestern Yukon. Unpublished PhD Thesis, University of British Columbia, Vancouver, B.C., 215p. Came, R.C. (1992). Kluane Range Ni-Cu-PGE Properties - Southwest Yukon Territory, Summary Report for All-North Resources. Distler, V.V., (1994). Platinum Mineralization in the Noril'sk Deposits in: Proceedings of the Sudbury Noril'sk Symposium, Ontario Geological Survey Special Volume 5., eds. Lightfoot, P.C. and Naldrett, A.J., pp. 243-260. Dodds, C.J. and Campbell, R.B., 1982, Geology - Southwestern Kluane Lake, Yukon Territory, Map, GSC Open File 829. Dodds, C.J. and Campbell, R.B., (1992). Overview, legend and mineral deposit tabulations for Geological Survey of Canada Open Files 2188, 2189, 2190 and 2191. Eckstrand, O.R. and Cogolu, E., (1986). Se/S Evidence Relating to the Genesis of Sulphides in the Crystal Lake Gabbro, Thunder Bay, Ontario, Program with Abstracts - Geological Association of Canada; Mineralogical Association of Canada; Canadian Geophysical Union Joint Annual Meeting, p.66. Eckstrand, O.R., and Hulbert, L.J., (1987). Selenium and the Source of Sulphur in Magmatic Nickel and Platinum Deposits, Program with Abstracts - Geological 136 Association of Canada; Mineralogical Association of Canada Joint Annual Meeting, p.40. Eckstrand, O.R., Grinenko, L.N., Krouse, H.R., Paktunc, A.D., Schwann, P.L., and Scoates, R.F.J., (1989). Preliminary data on sulphur isotopes and Se/S ratios, and the sources of magmatic sulphides from the Fox River Sill, Molson Dykes, and Thompson nickel deposits, northern Manitoba; in Current Research, Part C, GSC, Paper 89-1C, pp. 235-242. Faure, G., (1986). Principles of Isotope Geology, 2nd Edition, Chap. 28, John Wiley and Sons, N.Y.. Fedorenko, V.A. (1981). Petrochemical Series of Extrusive Rocks of the Noril'sk Region. Soviet Geology and Geophysics, v. 22(6). pp. 66-74. Geological Survey of Canada Aeromagnetic Series Map 4299G (1966). Geological Survey of Canada Open File Map 381 (1976). Geological Survey of Canada Open File Map 2188 (1992). Gorbachev, N.S., and Grinenko, L.N. (1973). Origin of the October sulfide ore deposits, Noril'sk region, in the light of sulfide and sulfate sulfur isotope compositions. Geochemistry, v. 10, pp. 843-851. Green, A.H., and Naldrett, A.J. (1981). The Langmuir Volcanic Peridotite-Associated Nickel Deposits: Canadian Equivalents of the Western Australian Occurrences. Economic Geology, v. 76, pp.1503-1523. Grinenko, L.N. (1984). Hydrogen Sulphide-Containing Gas Deposits as a Source of Sulphur for Sulphurization of Magma in Ore-Bearing Intrusives of the Noril'sk Area. v. 27, pp. 270-292. Grinenko, L.N. (1985). The Sulphur of the Nickeliferous and Barren Gabbro-Dolerite Intrusions of the Northwest Siberian Platform. International Geology Review, pp. 695-708. Groves, D.I., Barrett, F.M., and McQueen, K.G. (1979). The Relative Roles of Magmatic Segregation, Volcanic Exhalation and Regional Metamorphism in the Generation of Volcanic-Associated Nickel Ores in Western Australia. Canadian Mineralogist, v. 17, pp.319-336. Hawesworth, C.J., Lightfoot, P.C., Fedorenko, V.A., Blake, S., Naldrett, A.J., Doherty, W. and Gorbachev, N.W. (1995). Magma differentiation and mineralization in continental flood basalts. Lithos, v. 34, pp.61-88. 1 3 7 Hulbert, L.J., Gregoire, D.C., Paktunc. D., and Carne, R.C. (1992). Sedimentary Nickel, Zinc, and Platinum-group-element Mineralization in Devonian Black Shales at the Nick Property, Yukon, Canada: A New Deposit Type. Exploration and Mining Geology, v. 1, pp. 39-62. Hulbert, L.J. (1995). Geology and Metallogeny of the Kluane mafic-ultramafic belt, Yukon Territory, Canada; eastern Wrangellia - a new Ni-Cu-PGE metallogenic terrane. Geological Survey of Canada Open File 3057. Irvine, T.N. and Baragar, R.A. (1971). A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciencs, v. 8, pp.523-548. Jago, B.C. (1995). Yukon/Kluane Area: Mineralogy of platinum group minerals in PGE-rich samples from the Spy Complex, Southern Yukon Territory. Inco Exploration and Technical Services Inc. Internal Memo from B.C.J, to R.A.A., January 23, 1995. Jensen, L.S, (1976). A New Cation Plot for Classifying Subalkalic Volcanic Rocks. Ontario Division of Mines, MP66, 22p. Jones, D.L., Silberling, N.J. and Hillhouse, J. (1977). Wrangellia - A displaced terrane in northwestern North America. Canadian Journal of earth Sciences, v. 14, pp. 2565-2577. Keyser, H.J. (1987). Yukon Assessment Report 92528 (115G/2). Kovalenker, V.A., Gladyshev, G.D., and Nosik, L.P. (1974). Isotopic Composition of Sulphide Sulphur from Deposits of the Talnakh Ore Node in Relation to their Selenium Content. International Geology Review, v. 17, pp. 725-734. Lassiter, J .C, DePaolo, D.J. and Mahoney, J.J. (1995). Geochemistry of the Wrangellia Flood Basalt Province: Implications for the role of continental and oceanic lithiosphere in flood basalt genesis. Journal of Petrology, v. 36, pp.983-1009. Lesher, CM. (1989). Komatiite-Associated Nickel Sulphide Deposits; in: Ore Deposits Associated with Magmas, eds. Whitney, J.A. and Naldrett, A.J., v. 4, pp. 45-101. Lightfoot, P.C. (1996). The Giant Nickel deposits of Sudbury and Noril'sk. 1996 Prospectors and Developers Association of Canada, Nickel in a Nutshell Workshop Abstract. 138 Lightfoot, P.C , Naldrett, A.J., Gorbachev, N.S., Fedorenko, V.A., Hawkesworth, C.J., Hergt, J. and Doherty, W. (1994). Chemostratigraphy of Siberian Trap Lavas, Noril'sk District: Implications for the Source of Flood Basalt Magmas and their Associated Ni-Cu Mineralization; in Proceedings of the Sudbury - Noril'sk Symposium, eds. Lightfoot, P.C. and Naldrett, A.J., Ontario Geological Survey Special Volume 5. pp. 283-312. Loftus-Hills, G., and Solomon, M. (1967). Cobalt, Nickel and Selenium in Sulphides as Indicators of Ore Genesis. Mineralium Deposita, v. 2, pp. 228-242. MacEy, M., McLoughlin and Vincent, J.S. (1973). Yukon Assessment Report 60154 (115G/2). MacGowan, P. (1996). 1996 Geophysical Survey on the Klu property, Yukon Assessment Report (115G/2). Maier, W.D., Barnes, S.-J. and de Waal, S.A. (1998). Exploration for magmatic Ni-Cu-PGE deposits: a review of recent advances in the use of geochemical tools, and their application to some South African ores. South African Journal of Geology, v. 101(3). McQueen, K.G. (1981). Volcanic-Associated Nickel Deposits from around the Wdgiemooltha Dome, Western Australia. Economic Geology, v. 76, pp. 1417-1443. Miller, S.M. (1991). Geology, Petrology and Geochemistry of the Host Rocks and Associated Ni-Cu-PGE Disseminated Mineralization of the Wellgreen Deposit, Kluane Ranges, Yukon Territory. Unpublished MSc Thesis, McMaster University, Hamilton, Ontario, 249p. Mortensen, J.K. and Hulbert, J.K. (1991). A U-Pb zircon age for a Maple Creek gabbro sill, Tatamagouche Creek area, southwest Yukon Territory; in Radiogenic age and isotopic studies: Report 5, Geological Survey of Canada, Paper 91-2, pp. 175-179. Muller, J.E. (1967). Kluane Lake map-area, Yukon Territory (115G, 115F east half); Geological Survey of Canada, Memior 340. Naldrett, A.J. (1981). Nickel Sulphide Deposits: Classification, Composition and Genesis in Economic Geology 75th Anniversary Volume, edited by Skinner, B.J., pp. 628-685. 139 Naldrett, A.J. (1984). Mineralogy and Composition of the Sudbury Ores; in The Geology and Ore Deposits if the Sudbury Structure, Ontario Geological Survey Special Volume 1, eds. Pye, E.G., Naldrett, A.J., and Giblin, P.E., pp. 309-325. Naldrett, A.J. (1989). Ores Asociated with Food Basalts; in Ore Deposits Associated with Magmas, eds. Whitney, J.A. and Naldrett, A.J., v. 4. pp.103-118. Naldrett, A.J., Lightfoot, P.C, Fedorenko, V.A., Doherty, W. and Gorbachev, N.S. (1992). Geology and Geochemistry of Intrusions and Flood Basalts of the Noril'sk region, USSR, with Implications for the Origin of the Ni-Cu Ores. Economic Geology, v. 87, No.4, pp. 975-1004. Naldrett, A.J. (1995). A Short Course on Magmatic Sulphide Deposits - Theory, Reality, Exploration, Alaska Miners Association 1995 Convention. Naldrett, A.J., Lightfoot, P.C, Fedorenko, V.A., Doherty, W. and Gorbachev, N.S. (1992). Geology and Geochemistry of Intrusions and Flood Basalts of the Noril'sk region, USSR, with Implications for the Origin of the Ni-Cu Ores. Economic Geology, v. 87, No.4, pp. 975-1004. Nicholls, J. and Russell, J.K. (1991). Major element discrimination of magma-batches in lavas from Kilauea volcano, Hawaii, 1954-1971 eruptions. Canadian Mineralogist, v. 29, pp. 981-993. Paktunc, A.D., and Eckstrand, O.R. (1986). Nickel-Copper Deposits of the St. Stephen Intrusion; Sulphur Isotope and Se/S Relations. Program with Abstracts -Geological Association of Canada; Mineralogical Association of Canada; Canadian Geophysical Union Joint Annual Meeting, p. 111. Pearce, J.A. and Norry, M.J. (1979). Petrologic Implications of Ti, Zr, Y and Nb Variations in Volcanic Rocks. Contributions to Mineralogy and Petrology, v. 69, p33-47. Read, P.B. and Monger, J.W.H. (1976). Pre-Cenozoic volcanic assemblages of the Kluane and Alsek Ranges, southwestern Yukon Territory. Geological Survey of Canada Open File Report 381, 96p. Richards, M.A., Jones, D.L., Duncan, R.A. and Depaolo, D.J. (1991). A mantle plume initiation model for the Wrangellia Flood Basalt and other oceanic plateaus. Science, v. 254, pp. 263-267. Roeder, P.L. and Emslie, R.F. (1970). Olivine-Liquid Equilibrium. Contributions to Mineralogy and Petrology, v. 29, p275-289. 140 Russell, J.K. and Stanley, CR. (1990). Origins of the 1954-1960 Lavas, Kilauea Volcano, Hawaii: Major Element Constraints on Shallow Reservoir Magmatic Processes. Journal of Geophysical Research, v. 95, No.B4, p5021-5047. Russell, J.K., Nicholls, J, Stanley, C.R., and Pearce, T.H. (1990). Pearce Element Ratios: A Paradigm for Testing Hypotheses. Eos, v. 71, p234-6, 246-247. Samson, S.D., Patchett, P.J., Gehrels, G.E. and Anderson, R.G. (1990). Nd and Sr isotopic characterization of the Wrangellia Terrane and implications for crustal growth of the Canadian Cordillera. Journal of Geology, v. 98, pp. 749-762. Simkin, T. and Smith, J.V. (1970). Minor element distribution in olivine. Journal of Geology, v. 78, pp304-325. Stanley, CR. and Russell, J.K. (1989). Petrologic hypothesis testing with.Pearce element ratio diagrams: derivation of diagram axes. Contibutions to mineralogy and petrology, v. 103, pp. 78-89. Stanley, CR. (1990). Error Propagation on Pearce Element Ratio Diagrams, Modified from Chapter 9 of GAC Short Course Notes Theory and Application of Pearce Element Ratios to Geochemical Data Analysis'. GAC Annual Meeting, Vancouver, B.C., May 1990, 7p. Stanley, CR. and Wong, P.K.C. (1996). MERLIN: Molar Element Ratio Diagram MS-Windows Software for the Petrologic Evaluation of Lithogeochemical Data. Mineral Deposits Research Unit - Dept. of Geological Sciences, University of British Columbia. Sun, S.S. (1982). Chemical Composition and origin of the Earth's primitive mantle. Geochimica et Cosmochimica Acta, v. 46, pp. 179-192. Sun, S.S. and McDonough, W.F. (1989). Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes; in Magmatism in the ocean basins, eds. Saunders, A.D. and Norry, M.J., Geological Society Special Publication No. 42, pp. 3113-345. Taylor, S.R. and McLennan, S.M. (1985). The continental crust: its composition and evolution. Blackwell Scientific Publications, Oxford, 312p. Thompson, J.F.H., (1982). The Intrusion and Crystallization of mafic magmas, central Maine and genesis of their associated sulphides; unpublished Ph.D. thesis, University of Toronto. Thompson, J.F.H., and Naldrett, A.J., (1984). Sulphide-Silicate Reactions as a Guide to Ni-Cu-Co Mineralization in Central Maine, U.S.A. in Sulphide Deposits in 141 Mafic and Ultramafic Rocks, Eds. Buchanan, D.L., and Jones, M.J., The Institute of Mining and Metallurgy, pp. 103-113. Ueda, A. and Krouse, H.R. (1986). Direct conversion of sulphide and sulphate minerals to S0 2 for isotope analyses. Geochemical Journal, v. 20, pp. 209-212. Wood, D.A., Joron, J.L. and Treuil, M. (1979). A reappraisal of the use of trace elements to classify and discriminate between magma series erupted in different tectonic settings, Earth and Planetary Science Letters, v. 45, pp. 326-336. Wooden, J.L., Czamanske, G.K., Bouse, R.M., Likhachev, A.P., Kunilov, V.E. and Lyul'ko, V. (1992). Pb Isotope Data Indicate a Complex, Mantle Origin for the Noril'sk-Talnakh Ores, Siberia. Economic Geology, v. 87, pp.1153-1165. Wooden, J.L., Czamanske, G.K., Fedorenko, V.A., Arndt, NT., Chauvel, C , Bouse, R.M., King, B.W., Knight, R.J. and Siems, D.F. (1993). Isotopic and Trace-Element Constraints on Mantle and Crustal Contributions to Siberian Continental Flood Basalts, Noril'sk Area, Siberia. Geochimica et Cosmochimica Acta, v. 57, pp.3677-3704. Hz Appendix 1: Analytical Results from the Spy Sill and Lewis Intrusion Appendix 2: Olivine Microprobe Data from the Spy Sill and Lewis Intrusion peridotite samples |RX2261< IRX2261I |RX22S1( |RX2261( |RX2261( |RX2261( IRX2261I |RX2261( |RX2261( |RX2261( |RX2261( |RX2261( |RX2261( |RX2261I |RX2261( IRX2261I IRX2261I IRX2261I |RX2261( IRX226K |RX2261( |RX2261< |RX2261( |RX2261( |RX2261( |RX2261( |RX2261I IRX226K |RX2261( RX2231! RX2231! |RX2231! IRX2231! |RX2231! IRX2231! |RX2231( |RX2231i |RX2231( IRX2231! |RX2231( IRX2231! |RX2231! |RX2231! IRX2231! IRX2231! |RX2231( |RX2231! |RX2231! |RX2231! IRX2231! IRX2231! |RX2231! IRX2231! |RXNuml b> bi O 1 o b i cn CO bi b i bi c i b i U 109-1_2 | ! Cb107-4_4| I u I IO 1 I I M 1 u IO IO 8 4. 8 4> 8 dn IO 106-6J I l4 ± OJ ro 106-4_4 j 106-4_3 ] 106-4 2 ] 106-4 1 ] 106-3_4 j 106-3_2 I 106-3J I 197-2 3 197-1 3 197-1_2 | 195-2_1 | ! 195-1_3 ] ! ro 195-1_1 ] 194-2 4 | 194-2_3 | !! ! 194-1_3 | 194-1_2 | 194-1J | ! 1 OJ i 10 I 193-1_4 | 193-1_3 | ! 10 193-1_1 | 1 I' i 1 i I' I' 1 1 I' s i 1 1 i 1-1 1 I %l !9 I-i 1 1 i t Si W iS 1' 1 ! i I' 1 8^  i ! 1 1 1-1 I I 1- I' I s I' I | 0.003| 1 | 0.004 j 1 | 0.016 | 0.003] I 6000 I I 0.003] | 0.003 | 0.009] | 0.003) I 0.003J i i I! 0.003 0.003 1 0.003I ! l 1 0.003] | 0.015| I 0.013! 0.003 | 0.003| \ 0.0291 I i | 43.820] | 44.310| 1 I fc * 8 1 ft fc | 44.723] 1 CO OJ | 44.086| fc * | 44.047] | 44.013| I 43.327J f^c I 44.232I 42.128 k. ft i ! 1 1 1 1 1 1 II IS i | 41.484] 1 1 | 40.805] I p i I I | 0.005| 1 | 0.014 | 0.031 1! I i l l 1 i | 0.031 I I 1 i 8 I £ I 1 "S ! 8 a -. I I Ii ! 1 I 8 1 a • 8 8 I I! I w. I! ! 1 s 3 s 8 1 .8 s i 1 I! si 1 I I I s 1 i s a 1 I 1 1 1 i1 1 1 a i 1 1p s I 0.171 p ii = p p p 8 p P <= I 0.201 S f c p 8 =• p < w = p 15 p < §i 3 p H ! p 8 1 P 5 | 0.109 3! p P | I P p i I p t p s 1 !| 0.011] 1% i I I !! i i 1 i £ | 0.060| I 0.0021 5 p p 1 1 | 0.0521 | 0.043J 0.024 1 0.0021 l i I 1 i | 0.016| i I ! I I I p 1 1 1 1 1 P 1 1 1 | 0.2281 | 0.314 1 | 0.239 I 1 i 1 i i 3 O | 0.221 i l l i i 11 l l ? 1 1 1 1 1 1! i i 1 1 1 1 I 1 1 1 1 I -* -± 1 1 -\ 8 = 1 I I 1 CO CO s CO cn Ol 1 CD S CO s CO 3 CO 3 8 i i CO a 6.252 Si 7i CO 3S CO 3 CO CO • CO a CO 3 CO 8 Ol ( n co 38 B.911 » CD < i l l » 03 00 c 3i D CO IB CO s 1 i CO * CO CO § 1 1 p fi OO 1 1 1 1 1 1 1 i 1 1 1 1 1 1 | 0.216| 1 1i 1 i p i ! I 0.239 0.038 i i 1 0.094 = p 38 p s I i p 1 p s P a \ 0.062 = P ; i 1 0.1011 p 8 p 8 P 8 I 1 1 1 GO 1 1 I 1 1 1 1 s i 5 I 0.2621 1 i < i I 1 1 1 i | 0.4291 1 1 | 0.246] 1 0.245I 5 l P OJ cn P s | 0.126| | 0.095] | 0.082| I 1 37.4401 1 1 ro i 1 6.8861 • 1 1 17.214| | 32.184| | 4.607 | 5.374] |C.V.(%) 44.4341 e 2 § 1 | 44.239] | 43.801 j 0 1 8 1 | 41.287| I 41.297| I 40.728] | 40.805| |AvgCore_MgO 43.389| 44.115| fc g fc *• i fc 8 | 44.333 S . ! s ft S | 40.801 | 41.208 | 40.690 | 40.740 I 1 1.684| 1 o CO cn M p S p 3 | 0.853 1 ) p a 1 | 0.152 | 0.065 | 0.112 P < 16.400| 1 1 s 1 P 3 | 16.578| | 16.337| D 18.804 ! c ) 1 18.050| I | 19.421| | 20.3811 | 20.185| 1 8 iSo |RX2261 |RX2261 [RX2261 |RX2261 IRX2261 |RX2261 |RX2261 IRX2261 |RX2261 |RX2261 |RX2261 IRX2261 |RX2261 |RX2261 |RX2261 IRX2261 IRX2261 |RX2261 IRX2261 |RX2261 |RX2261 |RX2261 IRX2261 |RX2261 |RX2261 |RX2261 |RX2261 IRX2261 |RX2261 IRX2261 |RX2261 IRX2261 |RX2261 |RX2261 IRX2261 |RX2261 IRX2261 IRX2261 IRX2261 |RX2261 |RX2261 |RX2261 IRX2261 RX2261 [RX2261 |RX2261 |RX2261 |RX2261 |RX2261 |RX2261 |RX2261 [RX2261 |RX2261 |RX2261 [RX2261 | RXNum tn oi CO LJ u CO CO OJ LJ LJ CO LJ LJ CO LJ LJ CO CO - -- --- - - o o o o o o <o o o o o o o o o o-$ s s 5 s S S J ! J S i « to 2 J J i V ! i r w s fa t ! I i i 1 1 I I ! ! I 1 t 1 1 1' 1 1 1 S i t 1 I i- 8 t 1 8 S I 1- 1 1 i t 1 1 1 1- 8 1 t 3 1 I' 3 s ! t i i 1 § i- i I i- 1 1 I p o g CO O § p | 0.014 1 i Ol i p 8 | 0.010| I | 0.004] 1 1 fe § fe 1 fe 3 s 3 1 1 i 1 i | 43.051 i fe ro to fe fe | 43.047| 1 fe O) CO | 44.166| | 44.447| fi | 44.039| fi I I I 1 fi fi S fi | 44.218] fi § I 1 fi 1 I I fi I fi KJ 1 Ol 1 I 1 1 —i | 0.022 i 1 0.024 1 I| 0.030] I 0.023 -| 0.005| 1 1 1 I 1 | 0.012 I I 1 i 1 1 S ft I s 8 s 1 I I I I £ I I 1 I I I p a I 1 1 1 .8 & I 8 ft to I I 1 Ii 1 i 1 s1 i 1 p i i 1 i 1 1 1 1 1 1 11 1 p » p s P i p P 3 P s p i p 8 P : p 2 p £ p i p 8 p a P s p s p 8 p 3 I i II i I \ 0.002 j I i i 1 i i i i 1 i I I 1 1 11 p 1 i 1 1 1 1 p 1 i i 1 1 1 1 1 i 1 1 s 1 i p 8 p a 1 I 1 I I 1 p 3 1 I § LJ 1 1 1 1 1 1 i i $ a s cn O bi 01 CO i 2 1 00 CO s s i i s § s i i i i H i i a 1 « 1 CD 1 Ol & 1 Ol g § 1 1 2 o> Ol 8 CD CD I Ol Ol g CD 3 CD s i l 8 1 oi i CD 1 CD i 1 1 1 i 1 1 1 1 1 1 | 0.262J g CO 1 1 1 1 1 I p i p 3 1 1 p El 1 p 8 1 1 i 1 1 1 S 1 1 i 1 1 i 1 1 1 1 0.253 1 p 1 1 1 1 1 1 |AvgCore_NiO 1 1 1 1 1 0.254 | 0.193! | 0.213 1 I 0.258 | 0.274 I I § p s i 1 1 0.070i | 6.216 | 8.166 5 1 9.167 1 1.492 | 3.183 |C.V.(%) 1 1 i 1 I | 42.93 | | 44.215| | 44.1241 44.393 fi | 44.137] I 44.156 | 44.165| I i j ft H 1 1 1 | 44.449| | 44.252| | 44.330: I | 44.230 | 44.152 1 I 1 I 1 P 8 | 0.373J | 0.205| P 8 p P i I 0.118: | 0.020 1 (%)'A- 0| I 3 18.032| | 18.126 | 17.529 I 16.050 | 16.400 a cn « | 16.562 | 16.434 I 16.907 « 3 |AvgCore_FeO / S 7 |RX226' IRX226-|RX226-|RX226-IRX226' |RX226' |RX226-|RX226-[RX226-[RX226' |RX226-|RX226' |RX226-IRX226' IRX226" |RX226-IRX226' |RX226-|RX226' |RX226' |RX226 |RX226 |RX226 |RX226 |RX226 |RX226' |RX226-|RX226-|RX226 |RX226' IRX226-|RX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RX223 [RX223 |RX223 |RX223 |RX223 |RX223 [RX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RX223 IRX223 |RX223 |RX223 |RX223 |RX223 |RX223 |RXNun o o o o p o o o o o o o o o o 8 8 8 8 o 8 8 8 8 8 8 8 8 8 8 8 CD CO CD to CD CD to CD to CO CD CO CD CD CD CD CD CO CO CO CD CD CD CO CO CO CD CO | to 109-1_1 I cb107-4_3| cb107-4_2i | 1 1 1 t 1 r CO 1 ro IO |Cd t o -* 8 i> 106-6_3 106-6_2 106-6_1 106-5_4 106-5_2 106-5J 106-4_4 106-4_3 106-4 2 106-4_1 106-3_4 106-3 3 M 106-3_1 rb 4k 197-2_3 197-2_1 ! CO 197-1 2 1 195-2_1 1 Jk 195-1 3 ! to g —» 1 4k 1 .1 ro 194-2 1 194-1 4 1 CO 194-1 2 194-1 1 193-2_1 193-1 4 g CO g ro g ! 1 8 8 1 1 8 1 1 8 1 8 I I 8 1 f i 1 1 i I I 1 1 1-s I core | i i 1 1 I rim | s I rim | s 8 i 8 1 i t 8 |core | |rim | i 8 8 I rim j | rim | 8 1 |rim _ _ J i 8 1 Irim I i-I a CO 00 | 16.281 2 1 p 3 cr p is s OJ 01 | 20.116| I 20.535! I i i 1 I 1 P i 1 P s 1 1 1 1 1 1 |C.V.(%) 2 3 1 a 3 1 1 I g I I 1 1 I N 5 1 I s I i P 8 CO 1 1 CO 1 CO 1 1 I I I s I I p .8 2 I I 1 3 6I s 3 8 i^ 3 1 3 8 3 1 I 3 3 8 * 1 P 1 p ! p 1 p 3 1 p 1 p g o p 3 p 5 p g P a P 3 p a P 1 p 1 P s 8 Isozo 1 P p P 1 I 1 1 P 8 p i P 8 1 p Ueo'o I p g cn p 1 I1 0.0611 | 0.0681 | 0.0811 i c ! c % ISZ IRX226115 | IRX226115 | IRX226115 | IRX226115 | IRX226115 | |RX226115 | IRX226115 | IRX226115 | |RX226115 | IRX226113 | |RX226113 | IRX226113 | |RX226113 | IRX226113 | |RX226113 | IRX226113 | IRX226113 | |RX226113 | IRX226113 j |RX226113 j IRX226113 j IRX226113 ] |RX226113 ] IRX226113 ] IRX226113 | |RX226111 j IRX226111 | |RX226111 ] |RX226111 ] IRX226111 |RX226111 ] IRX226111 IRX226111 ] |RX226111 | IRX226111 |RX226111 IRX226111 | IRX226111 ] IRX226111 ] IRX226111 IRX226111 | [RX226109 IRX226109 ] IRX226109 |RX226109 |RX226109 IRX226109 IRX226109 ] |RX226109 | |RX226109 |RX226109 |RX226109 | IRX226109 IRX226109 |RX226109 | RXNumber 115-4_4 | 115-4 1 | |115-2_4 | 1115-2 3 | [115-2 2 | |115-2_1 | |115-1 4 | |115-1 3 | |115-1_2 | CJ k |113-4.3 I |113-4 2 | |113-3 4 j 1113-3_3 | 1113-3 2 | |113-3_1 ] 1113-2 4 ] |113-2 3 ] |113-2_2 ] |113-2 1 ] 1113-1 _4 ] |113-1_3 | |113-1 2 | OJ k k w k io k OJ OJ OJ IO OJ lo J> lb OJ lb M lb OJ IO |109-4_4 |109-4 3 | |109-4_2 | |109-4_1 ] |109-3_4 |109-3_3 |109-3_2 |109-3_1 ] |109-2_4 | |109-2_3 ] |109-2_2 ] 1109-2_1 ] |109-1_4 |109-1_3 |Grain-Pt core j rim j | core | |core' | | rim UJU | |core | |core UJU | | core | core | rim | rim | core 8 3 UJU| | rim | core | core | rim UJU | | core | core UJU | | rim | core I core UJU | | rim | core | core |rim 3. 3 | core I core UJU] I' | core | core | rim f | core Icore UIU i |core |core |rim I nm |core |core | rim 1' |core |core Core/Rim | 15.859| | 18.145| | 18.749| | 18.158| | 17.787| | 17.853J | 17.697| | 16.093] | 16.731J | 16.2681 | 16.353| | 16.469| | 16.898| | 16.613| | 17.159] > < EQ 7> •3 -n n O 0.0001 | 1.786| | 0.9581 | 0.0471 | 0.9681 | 1.074| | 0.6741 | 0.1911 1 1-415] I 0.977| | 0.995| | 0.396] [ 1.967] | 1.242] | 1.908| |C.V.(%) 80.922| 83.413| 80.145| 80.0931 80.6781 \ 80.704| | 79.913| | 79.583J | 79.894J | 80.4621 | 80.547] | 80.283] | 80.613] [ 80.852] | 81.049| | 81.109] | 81.190| | 80.701 | 80.820] | 81.044| | 81.074| | 81.266] | 81.458| \ 81.272] | 81.092] | 83.0581 | 83.104| | 83.047] | 83.187| \ 82.601 | 82.891 | 82.608| | 82.392| | 82.902] | 82.599] | 82.633! | 83.222. | 83.263 | 83.004! I 82.777: | 83.052 | 82.454 | 82.764 | 82.327: | 83.113 [ 82.844 I 82.610 | 82.426 | 82.273 | 82.161 | 82.481 I 82.544 f 82.596 | 82.731 | 82.352 -n o % 0.213| | 0.178| ! 0.183| | 0.187| | 0.174| | 0.170| | 0.199| | 0.194| | 0.183| | 0.2071 | 0.2041 I 0.220| | 0.196| I 0.213| | 0.207| | 0.205| | 0.207| | 0.192] | 0.224| | 0.212| 1 0.2021 | 0.190| | 0.208| | 0.217] | 0.1811 | 0.168| | 0.164| | 0.144| | 0.159| | 0.136| | 0.163| | 0.165| | 0.169] | 0.166| | 0.159| | 0.153| | 0.162| | 0.166J | 0.172| | 0.154| | 0.163| | 0.185| | 0.205| | 0.2291 | 0.215| | 0.213| | 0.2011 | 0.2071 | 0.198| | 0.218| | 0.194| | 0.218| | 0.213| | 0.209| | 0.2021 z % as Appendix 3: Analytical Results from the Nikolai basalt sampl / f i -l l . Appendix 4: Analytical Results from Ni-Cu-PGE showings and barren sulphide/oxide layers 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0052721/manifest

Comment

Related Items