Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Wrangellia flood basalts in Alaska, Yukon, and British Columbia : exploring the growth and magmatic history.. Greene, Andrew R. 2008-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2008_fall_greene_andrew_r.pdf [ 14.27MB ]
Metadata
JSON: 1.0052552.json
JSON-LD: 1.0052552+ld.json
RDF/XML (Pretty): 1.0052552.xml
RDF/JSON: 1.0052552+rdf.json
Turtle: 1.0052552+rdf-turtle.txt
N-Triples: 1.0052552+rdf-ntriples.txt
Original Record: 1.0052552 +original-record.json
Full Text
1.0052552.txt
Citation
1.0052552.ris

Full Text

WRANGELLIA FLOOD BASALTS IN ALASKA, YUKON, ANDBRITISH COLUMBIA: EXPLORING THE GROWTH AND MAGMATIC HISTORYOF A LATE TRIASSIC OCEANIC PLATEAUByANDREW R. GREENEA THESIS SUBMITTED iN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Geological Sciences)UNIVERSITY OF BRITISH COLUMBIA(Vancouver)August 2008©Andrew R. Greene, 2008ABSTRACTThe Wrangellia flood basalts are parts of an oceanic plateau that formed in theeastern Panthalassic Ocean (ca. 230-225 Ma). The volcanic stratigraphy presently extends>2300 km in British Columbia, Yukon, and Alaska. The field relationships, age, andgeochemistry have been examined to provide constraints on the construction of oceanicplateaus, duration of volcanism, source of magmas, and the conditions of melting andmagmatic evolution for the volcanic stratigraphy.Wrangellia basalts on Vancouver Island (Karmutsen Formation) form anemergent sequence consisting of basal sills, submarine flows(>3km), pillow breccia andhyaloclastite (<1 1cm), and subaerial flows (>1.5 km). Karmutsen stratigraphy overliesDevonian to Permian volcanic arc (—‘380-355 Ma) and sedimentary sequences and isoverlain by Late Triassic limestone. The Karmutsen basalts are predominantlyhomogeneous tholeiitic basalt (6-8 wt% MgO); however, the submarine part of thestratigraphy, on northern Vancouver Island, contains picritic pillow basalts (9-20 wt%MgO). Both lava groups have overlapping initial and ENd, indicating a common, oceanisland basalt (OIB)-type Pacific mantle source similar to the source of basalts from theOntong Java and Caribbean Plateaus. The major-element chemistry of picrites indicatesextensive melting (23 -27%) of anomalously hot mantle (‘—1500°C), which is consistentwith an origin from a mantle plume head.Wrangellia basalts extend —-‘450 km across southern Alaska (Wrangell Mountainsand Alaska Range) and through southwest Yukon where <3.5 km of mostly subaerialflows (Nikolai Formation) are bounded by Pennsylvanian to Permian volcanic (312-280Ma) and sedimentary strata, and Late Triassic limestone. The vast majority of the Nikolaibasalts are LREE-enriched high-Ti basalt (1.6-2.4 wt% Ti02)with uniform plume-typePacific mantle isotopic compositions. However, the lowest ‘-.400 m of stratigraphy in theAlaska Range, and lower stratigraphy in Yukon, is light rare earth element (LREE)depleted low-Ti basalt (0.4-1.2 wt% Ti02)with pronounced negative-HFSE anomaliesand highElfvalues that are decoupled from Nd and displaced well above the OIB mantlearray. The low-Ti basalts indicate subduction-modified mantle was involved in theformation of basalts exposed in Alaska and Yukon, possibly from mechanical andthermal erosion of the base of the lithosphere from an impinging mantle plume head.11TABLE OF CONTENTSfiTable of ContentsfflList of Tables viiList of Figures ixAcknowledgementsDedicationxviCo-authorship statementxviiCHAPTER 1Introduction 1Introduction and motivation for this study 2The Wrangellia oceanic plateau 7Methodology and rationale for this study 10Previous research 11The importance of LIPs and mantle plumes 12An overview of the four manuscripts in this dissertation and additional references...15Contributions to this project 18References 19CHAPTER 2Wrangeffia Flood Basalts on Vancouver Island: Significance ofPicritic and Tholeiitic Lavas for the Melting History and MagmaticEvolution of a Major Oceanic Plateau 26Introduction 27Geologic setting 28Wrangellia on Vancouver Island 28Age of the Karmutsen Formation 31Volcanic stratigraphy and petrography 31Sample preparation and analytical methods 37Whole-rock chemistry 44Major- and trace-element compositions 44Sr-Nd-Hf-Pb isotopic compositions 56Alteration 62Olivine accumulation in picritic lavas 63Discussion 63Melting conditions and major-element composition of primary magmas 65Source of Karmutsen lavas 68REE modeling: Dynamic melting and source mineralogy 71Magmatic evolution of Karmutsen tholeiitic basalts 75Conclusions 78Acknowledgements 79References 79111CHAPTER 3Wrangeffia Flood Basalts in Alaska: A Record of Plume-LithosphereInteraction in a Late Triassic Accreted Oceanic Plateau 86Introduction 87Geologic setting 89Wrangellia in Alaska 89Wrangell Mountains 89Eastern Alaska Range 91Age of the Nikolai Formation 93Volcanic stratigraphy and petrography 93Whole-rock chemistry 104Major- and trace-element compositions 104Sr-Nd-Hf-Pb isotopic compositions 115Alteration 115Flood basalt chemostratigraphy 123Discussion 125Source ofNikolai basalts 125Lithospheric involvement in derivation of the low-titanium basalts 128Nature of underlying Paleozoic arc lithosphere 128Trace-element and isotopic source constraints of the low titanium basalts... 129Origin of decoupled Hf and Nd isotopic compositions oflow-titanium basalts 134Melting conditions and estimated major-element composition ofprimary low-Ti magma 138Conclusions 142Acknowledgements 143References 143CHAPTER 4Geochemistry of Flood Basalts from the Yukon (Canada) Segment of theAccreted Wrangeffia Oceanic Plateau 151Introduction 152Geologic setting and age constraints 153Field relations and petrography 156Whole-rock chemistry 161Major- and trace-element compositions 161Sr-Nd-Hf-Pb isotopic compositions 171Discussion 171Effects of alteration and comparison to Nikolai basalts in Alaska 171Relationship between chemistry and stratigraphic position 179Source characteristics ofNikolai basalts in Yukon 180Melting of arc mantle in formation of the low-titanium basalts 182Conclusion 188Acknowledgements 189References 189ivCHAPTER 5The Age and Volcanic Stratigraphy of the Accreted Wrangeffia OceanicPlateau in Alaska, Yukon and British Columbia 195Introduction 196Wrangellia flood basalts: The volcanic stratigraphy of anaccreted oceanic plateau 198Geographic distribution and aerial extent of the Wrangellia flood basalts 200Geologic history of Wrangellia 200Stratigraphy of Wrangellia 203Wrangellia of southern Alaska 203Talkeetna Mountains and eastern Alaska Range 205Wrangell Mountains 209Wrangellia in southwest Yukon 215Wrangellia in southeast Alaska 218Wrangellia in the Queen Charlotte Islands (Haida Gwaii) 221Wrangellia on Vancouver Island 222Central and southern Vancouver Island 222Northern Vancouver Island 228Geochronology of Wrangellia 233Previous geochronology for Wrangellia flood basaltsand related plutonic rocks 233Information about samples analyzed by40Ar/39Arin this study 2354O,39geochronological results 236Summary of isotopic age determinations for Wrangellia flood basalts 243Paleontological studies 243Discussion 245Overview of geology and age of Northern and SouthernWrangellia 245Eruption environment for Wrangellia flood basalts 248Northern Wrangellia 248Southern Wrangellia 249The accumulation and subsidence of the Wrangellia flood basalts 251Conclusion 254Acknowledgements 255References 255CHAPTER 6ConclusionsConclusions and Directions for Future Research 267References 271vAppendices.Appendix A. Geologic map of the Mount Arrowsmith area 273Appendix B. XRF whole-rock analyses of a subset of Karmutsen basalts,Vancouver Island, B.C 274Appendix C. PCIGR trace-element analyses of Karmutsen basalts,Vancouver Island, B.C 276Appendix D. Sample preparation and analytical methods for Alaska samples 279Appendix E. Sample preparation and analytical methods for Yukon samples 284Appendix F. Major element (wt% oxide) and trace element (ppm) abundancesin whole rock samples of Late Paleozoic Station Creek Formation,Yukon 288Appendix G. Previous research on Wrangellia 290Appendix H. 40Ar/39Aranalytical methods 306Appendix I. Analytical results of reference material from Actlabs whole-rockanalyses for Vancouver Island and Yukon 308Appendix J. Description of supplementary electronic files on CD-ROM 310Supplementary electronic files on CD-ROMSupplementary data files (SD)SD 1- Endnote database for Wrangellia (.enl file)SD 2- Reference list for Wrangellia (.doc file)SD 3- Geochemistry for40Ar/39Arsamples (.xls file)SD 4-40Ar/39Aranalytical data (.xls file)SD 5- Wrangellia ages and biostratigraphy (.xls file)Supplementary Google Earth ifies (SGE)SGE 1- Mapped Wrangellia flood basalts (.kniz file)SGE 2- Major faults in Alaska and Yukon (.kml file)SGE 3- Major faults in southwest B.C. (.kml file)SGE 4- Alaska sample locations (.kml file)SGE 5- Yukon sample locations (.kml file)SGE 6- Vancouver Island sample locations (.kml file)SGE 7- Alaska Range photograph locations (.kmz file)SGE 8- Wrangell Mountains photograph locations (.kmz file)SGE 9- Yukon photograph locations (.kmz file)SGE 10- Vancouver Island photograph locations (.kmz file)Supplementary photo ifies (SP)SP 1- Alaska Range photographs (.pdf file)SP 2- Wrangell Mountains photographs (.pdf file)SP 3- Ed MacKevett Jr. Wrangell Mountains photographs (.pdf file)SP 4- Yukon photographs (.pdf file)SP 5- Vancouver Island photos (.pdf file)Greene_2008_PhD_dissertation_UBC (.pdf file of complete dissertation)viLIST OF TABLESCHAPTER 2Table 2.1 Summary of petrographic characteristics and phenocrystproportions of Karmutsen basalts on Vancouver Island, B.C 39Table 2.2 Major element (wt% oxide) and trace element (ppm)abundances in whole rock samples of Karmutsen basalts,Vancouver Island, B.C 46Table 2.3 Sr and Nd isotopic compositions of Karmutsen basalts,Vancouver Island, B.C 59Table 2.4 Hf isotopic compositions of Karmutsen basalts, VancouverIsland, B.C 60Table 2.5 Pb isotopic compositions of Karmutsen basalts, VancouverIsland, B.C 61Table 2.6 Estimated primary magma compositions for Karmutsen basaltsand other oceanic plateaus/islands 67CHAPTER 3Table 3.1 Summary of petrographic characteristics and phenocrystproportions ofNikolai basalts in Alaska 102Table 3.2 Major element (wt% oxide) and trace element (ppm)abundances in whole rock samples ofNikolai basalts, Alaska 106Table 3.3 Sr and Nd isotopic compositions ofNikolai basalts, Alaska 118Table 3.4 Hf isotopic compositions ofNikolai basalts, Alaska 119Table 3.5 Pb isotopic compositions ofNikolai basalts, Alaska 120Table 3.6 Estimated primary magma compositions for Nikolai basalts andother oceanic plateaus/islands 141CHAPTER 4Table 4.1 Summary of petrographic characteristics and phenocrystproportions ofNikolai basalts in Yukon 160Table 4.2 Major element (wt% oxide) and trace element (ppm)abundances in whole rock samples ofNikolai basalts, Yukon 163viiTable 4.3 Sr and Nd isotopic geochemistry ofNikolai basalts, Yukon 174Table 4.4 Hf isotopic compositions ofNikolai basalts, Yukon 175Table 4.5 Pb isotopic compositions ofNikolai basalts, Yukon 176CHAPTER 5Table 5.1 Areal extent and volumetric estimates for the Wrangellia flood basalts 201Table 5.2 Compilation of previous geochronology of Wrangellia floodbasalts and associated plutonic rocks 234Table 5.3 40Ar/39Ardating results for 13 samples of Wrangellia floodbasalts and 6 samples from the Wrangellia Terrane 241Table 5.4 Comparison of geology and ages ofNorthern and Southern Wrangellia 246vi”LIST OF FIGURESCHAPTER 1Figure 1.1 Distribution of Phanerozoic LIPs on Earth 3Figure 1.2 Schematic diagrams of mantle plumes in the Earth’s interior 4Figure 1.3 “Pin-pricking” the giant Ontong Java Plateau 6Figure 1.4 Map showing distribution of Wrangellia flood basalts inAlaska, Yukon, and British Columbia 8Figure 1.5 Estimated distribution of the continental landmasses in theMiddle to Late Triassic (226 Ma) 9Figure 1.6 Age difference between peak eruption ages of LIPs from thelast 260 Myr versus the age of a stratigraphic boundary 14Figure 1.7 Photographs showing the different modes of transportationused for field work in remote areas of Alaska, Yukon, and BCas part of this project 17CHAPTER 2Figure 2.1 Simplified map of Vancouver Island showing the distributionof the Karmutsen Formation 30Figure 2.2 Geologic map and stratigraphy of the Schoen Lake ProvincialPark and Karmutsen Range areas 33Figure 2.3 Photographs of picritic and tholeiitic pillow basalts from theKarmutsen Range area (Alice and Nimpkish Lake area),northern Vancouver Island 34Figure 2.4 Photographs showing field relations from the Schoen LakeProvincial Park area, northern Vancouver Island 35Figure 2.5 Photomicrographs of picritic pillow basalts, KarmutsenFormation, northern Vancouver Island 38Figure 2.6 Whole-rock major-element, Ni, and LOl variation diagrams forthe Karmutsen Formation 45Figure 2.7 Whole-rock REE and trace-element concentrations for theKarmutsen Formation 53ixFigure 2.8 Whole-rock trace-element concentrations and ratios for theKarmutsen Formation 55Figure 2.9 Whole-rock Sr, Nd, and Hf isotopic compositions for theKarmutsen Formation 57Figure 2.10 Pb isotopic compositions of leached whole-rock samples byMC-ICP-MS for the Karmutsen Formation 58Figure 2.11 Relationship between abundance of olivine phenocrysts andwhole-rock MgO contents for Keogh Lake picrites 64Figure 2.12 Estimated primary magma compositions for three KeoghLake picrites using the forward and inverse modelingtechnique of Herzberg et a!. (2007) 66Figure 2.13 Comparison of age-corrected (230 Ma) Sr-Nd-Hf-Pb isotopiccompositions for Karmutsen flood basalts on VancouverIsland to age-corrected OIB and MORB 69Figure 2.14 Evolution of8Hfwith time for picritic and tholeiitic lavas forthe Karmutsen Formation 72Figure 2.15 Trace-element modeling results for incongruent dynamicmantle melting for picritic and tholeiitic lavas from theKarmutsen Formation 74Figure 2.16 Forward fractional crystallization modeling results for majorelements from MELTS (Ghiorso & Sack, 1995) compared topicritic and tholeiitic lavas from the Karmutsen Formation 77CHAPTER 3Figure 3.1 Simplified map of south-central Alaska showing thedistribution of the Nikolai Formation 90Figure 3.2 Geologic map and stratigraphy of the Wrangell Mountains, Alaska92Figure 3.3 Photographs of the base of the Nikolai Formation in theWrangell Mountains, Alaska 94Figure 3.4 Photograph of —1000 m of continuous flood basalt stratigraphyat the top of the Nikolai Formation along Glacier Creek in theWrangell Mountains, Alaska 95Figure 3.5 Photographs showing the top of the Nikolai Formation in theWrangell Mountains, Alaska 97xFigure 3.6 Simplified geologic map and stratigraphy of the AmphitheaterMountains, Alaska 98Figure 3.7 Photographs of the base of the Nikolai Formation in theAmphitheater Mountains, east-central Alaska Range (TangleLakes, West), Alaska 99Figure 3.8 Geologic map and stratigraphy of the Clearwater Mountains, Alaska 101Figure 3.9 Representative photomicrographs ofNikolai basalts, Alaska 103Figure 3.10 Whole-rock major-element and Ni variation diagrams vs.MgO for the Nikolai Formation in Alaska with data for theNikolai Formation in Yukon 105Figure 3.11 Whole-rock REE and trace-element concentrations for theNikolai Formation in Alaska 113Figure 3.12 Whole-rock trace-element concentration variations and ratiosfor the Nikolai Formation in Alaska, with data for the NikolaiFormation in Yukon 114Figure 3.13 Whole-rock Sr, Nd, and Hf isotopic compositions for theNikolai Formation in Alaska 116Figure 3.14 Pb isotopic compositions of leached whole-rock samples byMC-ICP-MS for the Nikolai Formation in Alaska 117Figure 3.15 Loss-on-ignition versus MgO and isotopic ratios for theNikolai Formation in Alaska 122Figure 3.16 Chemostratigraphy of the Nikolai Formation in three areas ofAlaska (Clearwater, Amphitheater, and Wrangell Mountains) 124Figure 3.17 Comparison of age-corrected (230 Ma) Sr-Nd-Hf isotopiccompositions for Nikolai basalts in Alaska to age-correctedOIB and MORE 126Figure 3.18 Comparison of Pb isotopic compositions of the NikolaiFormation in Alaska to OIB and MORE 127Figure 3.19 Trace-element ratios and isotopic compositions of the NikolaiFormation in Alaska 130Figure 3.20 Th-Nb and Ti-Yb proxies of the Nikolai Formation in Alaskawith data compilation and modeling results 132xiFigure 3.21 Global Hf-Nd isotope systematics with age-corrected data ofthe Nikolai Formation in Alaska 136Figure 3.22 Comparison of initial Hf, Nd, and Sr isotopic compositions ofthe Nikolai Formation in Alaska to Pacific arcs (TongaKermadec, Mariana, Vanuatu, New Britain) and Pacific MORB 137Figure 3.23 Estimated primary magma compositions for two picrites fromthe Nikolai Formation using the forward and inversemodeling technique of Herzberg et al. (2007) 140ChAPTER 4Figure 4.1 Simplified map of southwest Yukon showing the distributionof the Nikolai Formation 154Figure 4.2 Geologic map and stratigraphy of the northern part of theKluane Ranges, Yukon 155Figure 4.3 Photographs of the Nikolai Formation in the Kluane Ranges, Yukon 158Figure 4.4 Photomicrographs of representative Nikolai basalts insouthwest Yukon 159Figure 4.5 Whole-rock major-element variation diagrams for the NikolaiFormation in Yukon with data for the Nikolai Formation in Alaska 162Figure 4.6 Whole-rock REE and trace-element concentrations for theNikolai Formation in Kluane Ranges, Yukon 170Figure 4.7 Whole-rock trace-element concentrations for the NikolaiFormation in Yukon, with data for the Nikolai Formation in Alaska 172Figure 4.8 Whole-rock Sr, Nd, and Hf isotopic compositions for theNikolai Formation in Yukon, with fields for the NikolaiFormation in Alaska 173Figure 4.9 Comparison of trace-element compositions of the Nikolaibasalts in Yukon to averages for basalts from Alaska 178Figure 4.10 Comparison of age-corrected (230 Ma) Sr-Nd-Hf isotopiccompositions of the Nikolai Formation in Yukon and Alaskato age-corrected OIB and MORB 181Figure 4.11 Th-Nb and Ti-Yb systematics for the Nikolai Formation inYukon with data compilation and modeling results 183xiiFigure 4.12 Trace-element ratios of high- and low-titanium basalts of theNikolai Formation in Yukon with Paleozoic arc samples andfields for Nikolai basalts in Alaska 184Figure 4.13 Schematic diagrams of two stages of melting ofNikolaibasalts in Yukon and Alaska that produced the low- andhigh-titanium basalts in Yukon and Alaska 185Figure 4.14 Trace-element abundances of low-titanium basalts fromYukon compared to arc mantle compositions of the Early toMiddle Jurassic Talkeetna arc and incongruent dynamicmelting modeling results 187CHAPTER 5Figure 5.1 Map showing the distribution of Phanerozoic large igneous provinces 197Figure 5.2 Simplified map showing the distribution of Wrangellia floodbasalts in Alaska, Yukon, and British Columbia 199Figure 5.3 Simplified map of eastern south-central Alaska showing thedistribution of Wrangellia flood basalts and stratigraphic columns 204Figure 5.4 Geology and magnetic map of the Amphitheater Mountains 207Figure 5.5 Photographs ofbase ofNikolai Formation in Tangle Lakesarea of the Amphitheater Mountains 208Figure 5.6 Photographs of the base ofNikolai Formation north of SkolaiCreek in Wrangell-St. Elias National Park 210Figure 5.7 Photographs of flood basalts in the Glacier Creek area inWrangell-St. Elias National Park and map of the southern partof the Wrangell Mountains 211Figure 5.8 Photographs ofthe top of the Nikolai Formation aroundHidden Lake Creek in Wrangell-St. Elias National Park 212Figure 5.9 Photographs and map of the Nikolai Formation in southwest Yukon 216Figure 5.10 Simplified map of southeast Alaska showing the distribution of Triassicbasalts that may be correlative with Wrangellia flood basalts 219Figure 5.11 Simplified map of Vancouver Island showing the distributionof the Karmutsen Formation 223xliiFigure 5.12 Field notes, photographs, and geologic map for the SchoenLake area, Vancouver Island 226Figure 5.13 Generalized geology and photographs of Buttle Lake area,Vancouver Island 227Figure 5.14 Generalized geology of the Karmutsen Formation on northernVancouver Island in the Port Alice-Robson Bight area 229Figure 5.15 Photographs of intra-Karmutsen sedimentary lens near the topof the Karmutsen southwest of Nimpkish Lake, northernVancouver Island 231Figure 5.16 Photographs from near Holberg Inlet, northern Vancouver Island 232Figure 5.17 40ArI39Arage spectra for 6 analyses of plagioclase separatesfrom Vancouver Island 237Figure 5.18 40Ar/39Ar age spectra of 1 biotite and 2 plagioclase separatesfrom Yukon 238Figure 5.19 Ar/Ar age spectra for 6 analyses of plagioclase separatesfrom Alaska 239Figure 5.20 40Ar/39Ar age spectra for 6 analyses of homblende separatesfrom Alaska 240Figure 5.21 40Ar/Ar and UiPb ages of Wrangellia flood basalts andplutomc rocks 244Figure 5.22 Summary of geochronology and biostratigraphy for Wrangellia 247xivACKNOWLEDGEMENTSThis project has involved help from numerous people who I am extremely grateful to.James Scoates and Dominique Weis proposed and guided this study. I am grateful for theopportunity they gave me and for their enthusiastic guidance and support.I also have greatly appreciated help from Graham Nixon of the BC Geological Surveyand Steve Israel of the Yukon Geological Survey. Advice was also offered from JeanineSchmidt (USGS Anchorage), David Brew (USGS Menlo Park), Nick Massey (BCGS),Mikkel Schau, and Don Carlisle.Kelly and Natalie Bay of Wrangell Mountain Air in McCarthy, Alaska providedextensive knowledge of the Wrangell Mountains. Jeff Trop (Bucknell University) andDanny Rosenkrans (Wrangell-St. Elias National Park) offered assistance withbackcountry advice about Wrangell-St. Elias National Park.Great thanks go to Bruno Kieffer, Frederico Henriques, and James Scoates for help withfield work and making time in the field lots of fun.Katrin Breitsprecher was helpful with reviewing and discussions. Bruno Kieffer, JaneBarling, and Bert Mueller provided training for analytical work.I would also like to thank Mark Jellinek and Jim Mortensen for being a part of my thesiscommittee and posing thought-provoking questions that inspired me to explore differentareas.xvDEDICATIONThis dissertation is dedicated to my Mom and Dad, Sue and Art Greene.It is special to share this achievement with them.xviCO-AUTHORSifiP STATEMENTThe four manuscripts in this dissertation are all co-authored by colleagues. Mysupervisors James Scoates and Dominique Weis are co-authors on each of themanuscripts and they significantly contributed to each manuscript. Their contributionsincluded ideas, advice, comments, and reviewing, as well as financial support. Thecontributions from other co-authors on each manuscript are described below.CHAPTER 2Wrangellia flood basalts on Vancouver Island: Significance of picritic and tholeilticlavas for the melting history and magmatic evolution of a major oceanic plateauAuthors: Andrew R. Greene, James S. Scoates, Dominique Weis, Graham T. Nixon,Bruno KiefferBruno Kieffer analyzed a suite of 24 samples from Vancouver Island for isotopic andtrace-element compositions.Graham Nixon helped with field work and advice on field relationships andgeochemistry.CHAPTER 3Wrangeffia flood basalts in Alaska: A record of plume-lithosphere interaction in aLate Triassic accreted oceanic plateauAuthors: Andrew R. Greene, James S. Scoates, and Dominique WeisJames Scoates and Dominique Weis contributed to many aspects of this manuscript.CHAPTER 4Geochemistry of flood basalts from the Yukon (Canada) segment of the accretedWrangeffia oceanic plateauAuthors: Andrew R. Greene, James S. Scoates, Dominique Weis, Steve IsraelSteve Israel assisted with field work, and contributed funding for field work andanalytical data. Steve provided major- and trace-element whole-rock analyses of twenty-six samples ofNikolai basalt and 8 Station Creek samples.CHAPTER 5The age and volcanic stratigraphy of the accreted Wrangeffia oceanic plateau inAlaska, Yukon and British ColumbiaAuthors: Andrew R. Greene, James S. Scoates, Dominique Weis, Erik C. Katvala, SteveIsrael, and Graham T. NixonErik Katvala assisted with field work on Vancouver Island, provided fossildeterminations, and helped with revisions on the manuscript.Steve Israel contributed a single40Ar/Ar age date and provided revisions on themanuscript.Graham Nixon offered advice and revisions on the manuscript.xviiIuornpornIHLJAVIDINTRODUCTION AND MOTIVATION FOR THIS STUDYThe largest melting events on Earth have led to the formation of transient largeigneous provinces (LIPs). LIPs are features that can extend over millions of squarekilometers of the Earth’s surface and consist of flood basalt sequences up to 6 km thick.These voluminous outpourings of lava occur over short intervals of geologic time (i.e.several million years) and have been linked to major changes in the global climate andbiosphere in Earth history.This project is an extensive field and geochemical study of one of the majortransient LIPs on Earth, the Triassic Wrangellia oceanic plateau. Transient and persistentLIPs cover large areas of the oceans and continents (Fig. 1.1). Approximately 13 majortransient LIPs have formed in the last 260 Myr (Courtillot & Renne, 2003), including theSiberian (ca. 251 Ma) and Deccan traps (ca. 65 Ma); volcanism usually takes place in lessthan a few million years. Examples of persistent LIPs, or hotspots, are Ninetyeast Ridgeand the Hawaiian-Emperor chain, where volcanism occurs over tens of millions of years.Large igneous province is an umbrella term that includes continental flood basalts(CFBs), volcanic rifled margins, oceanic plateaus, and aseismic ridges (Saunders, 2005).Common attributes of these provinces are that they form from an unusually highmagmatic flux, they are mostly basaltic in composition, and they are not directlyassociated to seafloor spreading processes. The origin of most LIPs is best explained bymantle plumes. Mantle plumes are buoyant upwellings of hot mantle that may risethrough the surrounding mantle as large spherical heads fed by narrower conduits(Griffiths & Campbell, 1990), or a variety of plume shapes, possibly without heads(Farnetani & Samuel, 2005). Transient LIPs are the initial eruptive products that formfrom melting of a new mantle plume head at the base of the lithosphere, and persistentLIPs form from melting within the narrower plume tail (Campbell, 2005) (Fig. 1.2).Continental flood basalts are closely associated with continental breakup. Theformation of the Paraná-Etendeka (ca. 133 Ma) preceded the opening of the SouthAtlantic Ocean and formation of the North Atlantic Volcanic Province (ca. 62-60 Ma)preceded the opening of the North Atlantic Ocean. The Ethiopian traps (ca. 30 Ma) are anexample of the eruption of flood basalts prior to rifling of Arabia and Africa. CFBs are atleast 1 km thick and greater than 100,000 km2 and are emplaced as hundreds (or over a2(Ca. 230 Ma)(16±1 Ma)(201±1 Ma) (56±1 Ma; 61±2 Ma)wrangelliaColumbia River •, North AtlanticFigure 1.1 Distribution of Phanerozic LIPs on Earth. Transient LIPs are indicated in yellow (continental) and orange(oceanic) with peak eruption ages. Persistent LIPs are mostly in red. Map modified from base map by A. Goodlife andF. Martinez in Mahoney and Coffin (1997). The peak eruption ages of each LIP are from the compiled referenceswithin Courtillot and Renne (2003) and Taylor (2006).(30±1 Ma) (65.5±0.5 Ma) (259±3 Ma) (250±1 Ma)Ethiopia Deccan Emeishan Siberiaa Ontong Java-H(122±1 Ma)Caribbean Paraná-Etendeka(123±1.5 Ma)(89±1 Ma) (133±1 Ma)3Figure 1.2 Schematic diagrams of mantle plumes in the Earth’s interior. (a) A view of the deep-mantle complexitiesbeneath the central Pacific Ocean showing the Hawaiian hotspot, a D” high-velocity reflector, and heterogeneity at thebase of the lower mantle where mantle plumes may originate. Mantle plume head is hypothetical. Drawing modifiedfrom Gamero (2004) with parts of a plume model by Farnetani and Samuel (2005). (b) A drawing of a model of plume-lithosphere interaction beneath oceanic lithosphere that produces basalts erupted in an oceanic plateau. Dashed linemarks the base of the mechanical boundary layer. Arrows indicate flattening of the plume head. Model developedprimarily after work of Saunders et al. (1992). Parts of the diagrams were adapted from a drawing by J. Holden in Fodor(1987) and a plume model by Fametani and Samuel (2005)._._r.loxvertical exaggeration4thousand) of inflated compound pahoehoe flow fields that form a tabular flowstratigraphy (Self et a!., 1997). CFBs have been well-studied due to their accessibility assubaerial land-based features.Oceanic plateaus are not nearly as well-studied as CFBs because they are mostlysubmerged in the oceans. Several extant oceanic plateaus in the ocean basins have beenbroken up by seafloor spreading since their formation (Fig. 1.1). The Ontong Java,Manihiki and Hikurangi plateaus may have formed together at ca. 120 Ma and togethercover 1% Earth’s surface (Taylor, 2006). Ontong Java (2x106k2 in area; 30-35 kmthick) formed from the largest magmatic event recorded on Earth and covers an area ofthe western Pacific Ocean comparable to the size of western Europe, or approximatelyone-third the size of the conterminous United States (Fitton eta!., 2004). The Kerguelen(2x 106km2 in area; 20 km thick) and Broken Ridge Plateaus formed together, beginningca. 118 Ma, and have since been separated by seafloor spreading along the SoutheastIndian Ridge (Weis & Frey, 2002; Fig. 1.1). An important feature of the basalts eruptedin some oceanic plateaus is that they are unaffected by continental lithosphere and arethus better suited for understanding the composition of the mantle source of LIPs thanCFBs.The basalt stratigraphy of oceanic plateaus remains largely unsampled because ofthe difficulty in sampling more than the uppermost few hundred meters of flows fromdrilling ships (Fig. 1.3). For example, the Ontong Java Plateau, which rises to depths of1700 m below sea level, has been the focus of several Deep Sea Drilling Project (DSDP;Site 289) and Ocean Drilling Program (ODP; Leg 192, Sites 803 and 807) legs to samplethe basaltic basement of the plateau (Fig. 1.3; Mahoney et a!., 2001). These efforts havebeen described as “pin-pricking the elephant” (Tejada eta!., 2004); drilling of OntongJava has only penetrated a maximum 338 m of approximately 6 km or more of the basaltstratigraphy (Mahoney et a!., 2001). Drilling of the Kerguelen Plateau has onlypenetrated a maximum 233 m of the basalt stratigraphy (Frey et a!., 2000). Ocean drillingof extant plateaus submerged in the ocean is an extremely difficult and expensive tool tostudy oceanic plateaus.The most efficient and comprehensive way to study oceanic plateaus is byobserving and sampling exposures of oceanic plateaus where they have been accreted at5Figure 1.3 “Pin-pricking” the giant Ontong Java Plateau. (a) Representation of drilled stratigraphic sections from 8DSDP and ODP drilling sites (locations shown in panel b). Diagram adapted from Mahoney et al. (2001). The mostvolcanic rock sampled from drilling is 338 m of volcaniclastic rock at Site 1184 and 217 m of pillowed and massivebasalt at Site 1185. (b) Predicted bathymetry (after Smith & Sandwell, 1997) of the Ontong Java Plateau andsurrounding region showing the location of drill sites from ODP Leg 192 (stars) and previous drill sites (white andblack circles). Map adapted from Mahoney et a!. (2001). The broken black line (labeled W-E) indicates a transectwhere multichannel seismic reflection (MCS) investigations acquired data for the plateau by RJV Hakuho MaruKH98-1 Leg 2 in 1998 (shown in panel c). (c) Composite east-west MCS transects of the Ontong Java Plateau withreflecting horizons (location shown in panel b). Figure modified from Inoue et a!. (2008). Sediments and the top ofthe igneous basement are indicated. Vertical red lines indicate depth of drilling at 3 labelled sites on the transect. Redcircle indicates location of ‘eye structure’ (see Inoue et a!., 2008). Vertical exaggeration = 1 00:1. This figure servesto illustrate the difficulty of sampling and studying extant oceanic plateaus in the ocean basins. The black rectangle inpanel c indicates the extent of stratigraphy that is exposed and accessible in obducted oceanic plateaus on land, suchas the Wrangellia oceanic plateau.Distance (km)6continental margins or accreted onto island arcs, as in the case of the Ontong Java Plateauand the Solomon island arc. However, Phanerozoic examples of accreted oceanicplateaus are rare. Accreted sections of flood basalts around the Caribbean Sea and inCentral and South America are several kilometers thick in areas (e.g. Kerr et a!., 1998;Révillon et al., 1999; Kerr, 2003). Several small sections of the Ontong Java Plateau areexposed on land in the Solomon Islands (Tejada et a!., 1996; Babbs, 1997; Petterson eta!., 1999; Tejada et a!., 2002). These are the two major oceanic plateaus where obducteciparts of the plateau have been studied.THE WRANGELLL4 OCEANIC PLATEAUWrangellia flood basalts in the Pacific Northwest ofNorth America are part ofone of the best exposed accreted oceanic plateaus on Earth and can provide informationabout oceanic plateaus that is rarely accessible elsewhere on Earth. Wrangellia floodbasalts formed as part of a transient LIP in the Middle to Late Triassic, with accretion towestem North America occurring either in the Late Jurassic or Early Cretaceous.Although their original areal distribution was likely considerably larger, currentexposures of the flood basalts extend over 2300 km in British Columbia (BC), Yukon,and Alaska (Fig. 1.4). Parts of the entire volcanic stratigraphy, from the base to the top,are accessible for close examination on land. From a geochemical perspective, theWrangellia flood basalts have remained relatively poorly studied and have been the focusof only one study in the last twenty years using modem analytical geochemistry (Lassitereta!., 1995).In this study, the Wrangellia oceanic plateau has been intensively studied toestablish the stratigraphical and geochemical architecture of an oceanic plateau and thenature of the episodic melting events that lead to the formation of oceanic plateaus. TheWrangellia flood basalts erupted mostly within the Late Ladinian and Camian stages ofthe Triassic (ca. 230 Ma; e.g. Carlisle & Suzuki, 1974; Parrish & McNicoll, 1992), as thecontinents were gathered into a great landmass referred to as Pangaea (Fig. 1.5). Theyerupted atop different-aged Paleozoic arc volcanic and marine sedimentary sequences andare overlain by Late Triassic limestone. Paleontological and paleomagnetic studiesindicate that the Wrangellia flood basalts probably erupted in the eastem Panthalassic7Figure 1.4 Map showing distribution of Wrangellia flood basalts in Alaska, Yukon, and British Columbia. Mapderived from Wilson et a!. (1998), Israel (2004), Massey et a?. (2005a, b), Wilson et a!. (2005), and Brew (2007,written comm.). Outlines for the Peninsular and Alexander Terranes are shown in orange and blue, respectively. Aslabeled, the Wrangellia Terrane is referred to as Northern and Southern Wrangellia in this study. Major faults inAlaska and parts of Yukon and BC are shown with purple lines. The Wrangellia, Alexander, Peninsular terranes,which are part of the Wrangellia Composite Terrane, share similar elements or have a linked geologic history [asdefined by Plafker et aL (1989), Nokleberg et a?. (1994), Plafker and Berg (1994), and Plaficer et a?. (1994)].Geochronology from a single pluton in Alaska is proposed to link the Proterozoic to Triassic Alexander Terrane toWrangellia by late Pennsylvanian time (Gardner et a?., 1988). The Wrangellia and Peninsular Terranes may havebeen in close proximity by the Late Triassic or Early Jurassic (Rioux eta!., 2007).8Figure 1.5 Estimated distribution of the continental landmasses in the Middle to Late Triassic (226 Ma). Mapmodified from Ogg (2004), based on the reconstruction of Scotese (2004). Global Stratotype Sections andPoints (GSSP) are indicated for stages of the Triassic. Blue arrow indicates estimated location of theWrangellia oceanic plateau.9Ocean in equatorial latitudes (Jones et al., 1977; Katvala & Henderson, 2002). Theeruption of Wrangellia flood basalts broadly coincides with major biotic andenvironmental changes worldwide that occurred at the end of Carnian time (Furin et al.,2006). A global-scale environmental crisis, referred to as the Reingrabener turnover,occurred in the Carnian (e.g. Furin et a!., 2006) This event is preserved by the collapse ofrimmed carbonate platforms, a global shift in sedimentological and geochemical proxies,and a strong radiation of several groups, including scleractinian reef builders, calcareousnannoplankton, and dinosaurs (Furin et al., 2006 and references therein). The accretion ofthe Wrangellia oceanic plateau to western North America was a major tectonic event andrepresents a significant addition of oceanic mantle-derived material to the NorthAmerican crust (Condie, 2001).The volcanic stratigraphy of the Wrangellia plateau is defined as the KarmutsenFormation on Vancouver and Queen Charlotte Islands (Haida Gwaii), and as the NikolaiFormation in southwest Yukon and south-central Alaska (Fig. 1.4). On Vancouver Island,the volcanic stratigraphy is a tripartite succession of submarine, volcaniclastic, andsubaerial flows approximately 6 km thick. In Alaska and Yukon, the volcanicstratigraphy is predominantly massive subaerial flows with a small proportion ofsubmarine flows along the base. Smaller elements in southeast Alaska may be correlativewith the Wrangellia flood basalts.Rapid eruption of the Wrangellia flood basalts is supported by the absence ofintervening sediments between the flows, except in the uppermost part of the stratigraphyas volcanism waned. Paleomagnetic measurements of the Wrangellia flood basalts havenot revealed any magnetic reversals, which also testifies to the short duration of eruptions(Hillhouse, 1977; Yole & Irving, 1980; Hilihouse & Coe, 1994).METHODOLOGY AND RATIONALE FOR THIS STUDYThis study represents the first comprehensive study of the Wrangellia oceanicplateau. The integration of geochemical, volcanological, stratigraphic, andgeochronological results provide constraints on the formation of the Wrangellia oceanicplateau. The volcanological and stratigraphic studies yield insights about theemplacement of flows, eruptive environment, original tectonic setting, and construction10of the plateau. The geochronological and paleontological studies allow for estimation ofthe age and duration of volcanism. The petrologic and geochemical studies provideinformation about the composition of the source, conditions of melting, and the magmaticevolution of lavas that formed the Wrangellia plateau. The interpretation of thesedifferent results will help to develop our understanding of the physical and chemicalprocesses that occur in mantle plumes, as they decompress and impinge on the base ofoceanic lithosphere and produce the basaltic magmas which erupt though oceaniclithosphere and form oceanic plateaus. The contributions in the four major chapters inthis dissertation (described below) address some of the enduring questions about theorigin of oceanic plateaus.PREVIOUS RESEARCHIn the 1 970s, Jones and co-workers (1977) defined the fault-bound blocks of crustthat contain diagnostic Triassic flood basalts in BC, Yukon, and Alaska as Wrangellia,named after the type section in the Wrangell Mountains of Alaska. Early paleomagneticand paleontological studies of Wrangellia indicated long-distance displacement of thebasalts from equatorial latitudes (Hillhouse, 1977) and similar Daonella bivalves werefound in sediments directly beneath the flood basalts on Vancouver Island and in theWrangell Mountains (Jones et aL, 1977). A back-arc setting was initially proposed for theformation of Karmutsen basalts on Vancouver and Queen Charlotte Islands based onmajor- and trace-element geochemistry of 12 samples (Barker et al., 1989). Richards andco-workers (1991) proposed a plume initiation model for the Wrangellia flood basaltsbased on evidence of rapid uplift prior to volcanism, lack of evidence of rifting associatedwith volcanism (few dikes and abundant sills), and the short duration and high eruptionrate of volcanism. A geochemical study of 36 samples of Wrangellia flood basalts, 29samples from Buttle Lake on Vancouver Island and 9 samples from the WrangellMountains in Alaska, was undertaken by Lassiter and co-workers (1995) as part of theonly modem geochemical and isotopic study of Wrangellia flood basalts until theinitiation of this project. Lassiter and co-workers (1995) suggested mixing of a plumetype source with low Nb/Th arc-type mantle could reproduce variations in the Wrangelliaflood basalts.11TUE IMPORTANCE OF LIPS AND MANTLE PLUMESSince the recognition of oceanic plateaus in the early 1 970s (Edgar et al., 1971;Donnelly, 1973; Kroenke, 1974), there have been significant advances in ourunderstanding of oceanic plateaus. A wide range of scientific approaches have been usedto study oceanic plateaus; however, we only partially understand the physical andchemical processes that lead to the formation of oceanic plateaus.Oceanic plateaus have formed for at least the last several hundred million yearsand likely well back into Earth history. Archean and Proterozoic greenstone belts havebeen interpreted to be remnants of oceanic plateaus (e.g. Kerr, 2003) and komatiites (>18wt % MgO, with spinifex-textured olivines) may be the high-temperature meltingproducts of ancient mantle plumes (e.g. Arndt & Nesbet, 1982) and some Phanerozoicplume heads (e.g. Storey et aL, 1991). Phanerozoic oceanic plateaus may form frommantle plumes that originate at the core-mantle boundary (Campbell, 2005). The largetemperature and density contrasts that exist between the outer core and mantle areexpected to produce an unstable boundary layer above the core that episodically leads tothe formation of mantle plumes (Jellinek & Manga, 2004). Oceanic plateaus are thus thesurface manifestations of a major mode of heat loss from the interior of the Earth. Thegeochemical study of flood basalts is one important way by which we gain informationabout mantle plumes rising from their origin to the eruption of their melting products atthe Earth’s surface.Although mantle plumes are considered the strongest hypothesis for the origin ofLIPs, there are a number of alternative interpretations: meteorite impacts (Rogers, 1982),enhanced convection at the edge of continental cratons (King & Anderson, 1998),delamination of lithosphere causing rapid mantle upwelling (Elkins-Tanton, 2005), andmelting of fertile eclogitic mantle (Korenaga, 2005). One example of evidence thatmantle plumes are responsible for the formation of most LIPs is the high melt productionrates that form LIPs, which make them difficult to explain by an origin related to platetectonic processes (Jellinek & Manga, 2004). Perhaps the best evidence for the highsource temperatures required to explain the melting that produces LIPs is the presence ofpicrites (e.g. Saunders, 2005), with MgO between 12 and 18 wt % and 1-2 wt % Na2O +K20 (Le Bas, 2000). In this study, high-Mg picritic lavas have been discovered for the12first time within the Wrangellia oceanic plateau on northern Vancouver Island and inAlaska and modeling of their compositions indicate high source temperatures and highdegrees of melting.The eruption ofbasalts in LIPs has the potential to significantly affect thecomposition of the atmosphere and oceans by releasing large amounts of gas (primarilyCO2 and SO2)and aerosols that may trigger heating via runaway greenhouse effect orcooling via the spread of stratospheric sulphate aerosols that backscatter and absorbe thesun’s radiation (e.g. Rampino & Stothers, 1988; Wignall, 2001). Although CO2 emissionsin LIPs are small compared to the amount of CO2 present in the atmosphere and the land-ocean-atmosphere flux of C02,the gradual buildup of CO2 from LIP eruptions may tipthe balance enough to initiate release of other greenhouse gases, such as methanehydrates on the seafloor (Saunders, 2005). The release of SO2 in LIP eruptions isincomparable to inputs at any other time during the Phanerozoic (Self et a!., 2006). Theeruption of a single <2400 km3 flow field is estimated to release as much as 6,500 Tg a1for a 10-year duration, which is enormous compared to the background amount of S inthe atmosphere (<1 Tg) (Self et a!., 2006). In the oceans, a series of variables that areconsequences of large releases of CO2 lead to warmer polar waters, decreased solubilityof CO2and 02, and increased biogenic productivity in surface waters (e.g. Kerr, 2005).These effects can potentially lead to a shutdown of the ocean circulation system andanoxic conditions. It is likely that the eruption of Wrangellia flood basalts would havesignificantly affected the chemical composition of the ocean and atmosphere.The close correlation between major mass extinctions in Earth history and theformation of most transient LIPs suggests a causal link between LIP eruptions andenvironmental change (Rampino & Stothers, 1988; Fig. 1.6). The three most recent largediversity depletion events in the Phanerozoic (Permian-Triassic, Triassic-Jurassic,Cretaceous-Tertiary) coincide with the timing of peak volcanic activity in the formationof a continental flood basalt province (Siberia, Central Atlantic Magmatic Province(CAMP), and Deccan, respectively) (Courtillot & Renne, 2003). The eruption of theSiberian traps may have played a key role in the End Permian extinction event because ofthe high proportion ofpyroclastic eruptions and intrusion of magmas into carbon- andmethane-bearing strata (Wignall, 2001). Recent studies of the CAMP and the Deccan13Age Difference (Myr)10 100. . . • . ILarge igneous province stratigraphic boundaryEthiopian Traps (30 Ma) 0i2 event50.End Early PalaeoceneEnd PalaeoceneNorth Atlantic Province (56 Ma)Deccan Traps (65.5 Ma)Caribbean (89 Ma)End CenEnd CretaceousMadagascar (88 Ma)omanian100Early AptianOntong Java (122 MaLIP Age Kerguelen/Rajmahal (118 MaFd_1dEarlyAptian(Ma)Paraná-Etendeka (133 Ma)————— EndValanginian150Karoo-Ferrar (183 Ma)— i— End Pliensbachian200Central Atlantic Magmatic Province (201 Ma)—•—End TriassicWrangellia (230 Ma)? —1 i—?End Camian250Siberian Traps (250 Ma) End PermianEmeishan Traps (259 Ma)— — End GuadalupianLIP pre-dates boundary LIP post-dates boundary300stratigraphic boundaryFigure 1.6 Age difference between peak eruption ages of LIPs from the last 260 Myr versus the age of astratigraphic boundary associated with major sudden environmental changes. Diagram modified from Kelley(2007). Ages are mostly from the compilation of Courtillot and Renne (2003). The possible connection betweenWrangellia and a climatic and biotic crisis recognized in the Carnian is hypothetical and needs to be exploredfurther (Furin eta!., 2006).14Traps support a close temporal relationship between the End Triassic and End Cretaceousextinctions, respectively.A recent study by Courtillot and Olsen (2007) suggests there may be a connectionbetween magnetic superchrons, where the Earth’s magnetic field does not reverse polarityfor abnormally long periods of time (30-50 Myr), and the timing of mantle plumes. The-‘50 Myr long Kaiman Long Reverse Superchron (310 to 260 Ma) ended 10 Myr beforeeruption of the Siberian Traps and the 35 Myr long Cretaceous Long NormalSuperchron (118 to 83 Ma) ended -l 8 Myr before eruption of the Deccan Traps(Courtillot & Olsen, 2007). Courtillot and Olsen (2007) suggest a mechanism whereby amajor change in heat flow at the core-mantle boundary may lead to the end of asuperchron and cause initiation of “killer” mantle plumes.Along with aerially-extensive flood basalt sequences, radiating dike swarms areimpressive aspects of transient LIPs which extend >2800 km and are a major mode oflateral transport of LIP magmas in the crust. Radiating dike swarms are diagnostic ofmajor magmatic events linked to mantle plumes and associated continental break-up andserve as indicators of plume centers and magma transport (Ernst et al., 2001). Dikeswarms have been identified in every Mesozoic and Cenozoic CFB on Earth and areexpected to form in oceanic plateaus (Ernst & Buchan, 2003). Where the flood basalts arelargely eroded, such as the CAMP, dike swarms can be used to reconstruct plate motionsand identify plume events through Earth’s history (Ernst & Buchan, 1997).There are many important and fascinating aspects of LIPs. The study of LIPsholds great relevance in a variety of large-scale Earth processes and this study providesconstraints which help to further our understanding of these geological phenomena.AN OVERVIEW OF THE FOUR CHAPTERS IN THIS DISSERTATION ANDADDITIONAL REFERENCESThe chapters in this dissertation were written in manuscript form for submissionto major international geological journals. Three of the first four manuscripts are detailedstudies of the Wrangellia flood basalts from different regions of western North America(Vancouver Island, Alaska, and Yukon). These three manuscripts parallel one another inthat similar analytical techniques are used, but each study has distinct results and15interpretations. Each study involved extensive field work using different modes oftransportation in remote areas of western North America (Fig. 1.7). Additionalcontributions from this project that are not included in the dissertation are fourgovernment geological survey papers (Greene et a!., 2005b, 2005c, 20060; Nixon eta!.,2008b) and eight abstracts to international geological conferences (Greene et a!., 2005a,2006a, 2006b, 2007a, 2007b, 2007c, 2008a, 2008b). Contributions to other conferenceproceedings include studies of platinum-group elements (Scoates et a!., 2006, 2007,2008) and stratigraphic successions of Wrangellia flood basalts (Nixon et a!., 2008a).The first manuscript in this dissertation is a comprehensive study of theKarmutsen Formation on Vancouver Island, which has major implications for theevolution of this large oceanic plateau. Karmutsen basalts on Vancouver Island compriseone of the thickest and most complete examples of the volcanic stratigraphy of anemergent oceanic plateau on Earth. My co-authors and I examine the field relationships,stratigraphy, petrography, and geochemical and isotopic compositions of the KarmutsenFormation to assess the nature of the mantle source and to evaluate the melting historyand subsequent magmatic evolution ofbasalts involved in the construction of this majoroceanic plateau.The second manuscript is a study of the Nikolai Formation in Alaska. This studyintegrates field relationships in widespread areas of Alaska and geochemistry ofcontinuous sections of volcanic stratigraphy to understand the physical and chemicalprocesses that occur as a mantle plume head impinges on the base of oceanic arclithosphere. This information provides insights into the temporal and spatial variation ofthe magmas in a major oceanic plateau that result from plume-lithosphere interaction.The third manuscript is a field, petrographic, and geochemical study ofNikolaibasalts in Yukon, where volcanic stratigraphy and bounding sedimentary sequences aresimilar to Alaska. Nikolai basalts in Yukon are more altered than their counterparts inAlaska and a comparative geochemical study of Nikolai basalts from Yukon and Alaskaallows for analysis of the effects of alteration and regional differences in basalts eruptedin the northern part of Wrangellia. This study also examines the geochemistry of theunderlying Paleozoic arc sequences and their role in the generation of the Nikolai basalts.16Figure 1.7 Photographs showing the different modes of transportation used for field work in remote areas ofAlaska, Yukon, and BC as part of this project. (a) Canoe access to the Amphitheater Mountains in the AlaskaRange, Alaska. (b) Helicopter access in the Kluane Ranges, southwest Yukon. (c) Transportation by Super Cub inWrangell-St. Elias National Park, Alaska. (d) Four-wheel drive vehicle access on logging roads on northernVancouver Island. Photograph by Graham Nixon.17The fourth manuscript is a stratigraphic and geochronological study of Wrangelliaflood basalts from throughout BC, Yukon, and Alaska. This synthesis integratesobservations from field work and dozens of government geological survey reports andmaps to provide detailed descriptions of the volcanic stratigraphy and the pre- and post-volcanic rock stratigraphy of Wrangellia. This overview is presented in the form ofstratigraphic columns and descriptions, compiled geologic maps, photographic databases,interactive Google Earth files, and a review and compilation of previous research onWrangellia. The maps, photographs, and archiving of information are presenteddifferently than in the previous chapters and offer users interactive electronic tools tovisualize and explore information about the Wrangellia oceanic plateau. This work, incombination with40Ar/39Argeochronology of flood basalts throughout Wrangellia,provides an over-arching contribution that brings together past and present research onWrangellia in a new light.CONTRIBUTIONS TO THIS PROJECTThis project benefited from the assistance of many people. Assistance with fieldwork in Alaska, Yukon, and BC was provided by Bruno Kieffer, Frederico Henriques,and James Scoates. Advice for field work was provided by Graham Nixon (BritishColumbia Geological Survey), Nick Massey (British Columbia Geological Survey), DonCarlisle, Steve Israel (Yukon Geological Survey), Jeanine Schmidt (United StatesGeological Survey-Anchorage), David Brew (United States Geological Survey-MenloPark), Danny Rosenkrans (Wrangell-St. Elias National Park), and Jeff Trop (BucknellUniversity). Isotopic and trace-element analyses for a suite of 24 samples on VancouverIsland were made by Bruno Kieffer and Jane Barling. Bruno Kieffer, Jane Barling, andBert Mueller also provided training for analytical work for Yukon and Alaska samples atthe Pacific Centre for Isotopic and Geochemical Research (PCIGR). The analyses for40Ar/39Argeochronology were carried out by Tom Ulfrich from samples that werecrushed, separated, picked, and leached by me.This project was made possible from funding provided by the BC & YukonChamber of Mines (Association of Mineral Exploration BC) from the 2004 Rocks toRiches Program, the BC Geological Survey, the Yukon Geological Survey, and by18NSERC Discovery Grants to James Scoates and Dominique Weis. The author was alsosupported by a University Graduate Fellowship at UBC. Major- and trace-elementanalyses for this project were performed by Activation Laboratories (ActLabs) inOntario, Canada and at the Ronald B. Gilmore X-Ray Fluorescence Laboratory (XRF) atthe University of Massachussetts. Claude Herzberg (Rutgers University), Julian Pearce(Cardiff University), and Haibo Zou (University of California Los Angeles) offeredassistance with geochemical modeling. My supervisors, James Scoates and DominiqueWeis, offered invaluable assistance and guidance with all aspects of this project.REFERENCESArndt, N. T. & Nisbet, E. G. (1982). Geochemistry of Munro township basalt. In: Arndt,N. T. & Nisbet, E. G. (eds.) Komatiites. Allen and Unwin: Winchester, MA,pp.19-27.Babbs, T. L. (1997). Geochemical and petrological investigations of the deeper portionsof the Ontong Java Plateau: Malaita, Solomon Islands. Unpublished Ph.D.dissertation, Leicester University, U.K., 254pp.Barker, F., Brown, A. S., Budahn, J. R. & Plafker, G. (1989). Back-arc with frontal-arccomponent origin of Triassic Karmutsen basalt, British Columbia, Canada.Chemical Geology 75, 8 1-102.Brew, D. A. C. (2007, written comm.). Unpublished map showing the distribution of theLate Triassic Wrangellia, Hyci Group, and Perserverance group rocks insoutheastern Alaska, scale 1:600,000 (Based on Brew, D. A. (Compiler),Unpublished bedrock geologic map of southeastern Alaska.)Campbell, I. H. (2005). Large igneous provinces and the mantle plume hypothesis.Elements 1, 265-269.Carlisle, D. & Suzuki, T. (1974). Emergent basalt and submergent carbonate-clasticsequences including the Upper Triassic Dilleri and Welleri zones on VancouverIsland. Canadian Journal ofEarth Sciences 11, 254-279.Condie, K. C. (2001). Mantle Plumes and Their Record in Earth History. CambridgeUniversity Press: Cambridge, 306p.Courtillot, V. & Olsen, P. E. (2007). Mantle plumes link magnetic superchrons toPhanerozoic mass depletion events. Earth and Planetary Science Letters 260,495-504.Courtillot, V. E. & Renne, P. R. (2003). On the ages of flood basalt events. ComptesRendus Geoscience 335, 113-140.Donnelly, T. W. (1973). Late Cretaceous basalts from the Caribbean: a possible floodbasalt province of vast size. EOS 54, 1004.Edgar, N. T., Ewing, J. I. & Hennion, J. (1971). Seismic refraction and reflection in theCaribbean Sea. American Association ofPetroleum Geologists Bulletin 55, 833-870.19Elkins-Tanton, L. (2005). Continental magmatism caused by lithospheric delamination.In: Foulger, G. R., Natland, J. H., Presnall, D. C. & Anderson, D. L. (ecis.) Plates,Plumes and Paradigms. Geological Society of America Special Paper 388,pp.449-462.Ernst, R. E. & Buchan, K. L. (1997). Giant radiating dyke swarms: Their use inidentifring pre-Mesozoic large igneous provinces and mantle plumes. In:Mahoney, J. 3. & Coffin, M. F. (eds.) Large Igneous Provinces: Continental,Oceanic, and Planetary Flood Volcanism. American Geophysical UnionGeophysical Monograph 100,pp.297-333.Ernst, R. E. & Buchan, K. L. (2003). Recognizing mantle plumes in the geologic record.Annual Review ofEarth and Planetay Sciences 31, 469-523.Ernst, R. E., Grosfils, E. B. & Mège, D. (2001). Giant dike swarms: Earth, Venus andMars. Annual Review ofEarth and Planetary Science 29,489- 534.Farnetani, C. G. & Samuel, H. (2005). Beyond the thermal plume paradigm. GeophysicalResearch Letters 32(L073 11), doi: 10.1 029/2005GL022360.Fitton, J. G., Mahoney, J. J., Wallace, P. J. & Saunders, A. D. (eds.) (2004). Origin andEvolution ofthe Ontong Java Plateau. Geological Society of London SpecialPublication 229. Y74pp.Fodor, R. V. (1987). Low- and high-Ti02flood basalts of southern Brazil: origin frompicritic parentage and a common mantle source. Earth and Planetary ScienceLetters 84, 423-43 0.Frey, F. A., et al. (2000). Origin and evolution of a submarine large igneous province: theKerguelen Plateau and Broken Ridge, southern Indian Ocean. Earth andPlanetary Science Letters 176, 73-89.Furin, S., Preto, N., Rigo, M., Roghi, G., Gianolla, P., Crowley, J. L. & Bowring, S. A.(2006). High-precision U-Pb zircon age from the Triassic of Italy: Implicationsfor the Triassic time scale and the Camian origin of calcareous nannoplankton anddinosaurs. Geology 34(12), 1009-1012, 10.11 30/g22967a. 1.Gardner, M. C., Bergman, S. C., Cushing, G. W., MacKevett, E. M., Jr. , Plafker, G.,Campbell, R. B., Dodds, C. J., McClelland, W. C. & Mueller, P. A. (1988).Pennsylvanian pluton stitching of Wrangellia and the Alexander terrane, WrangellMountains, Alaska. Geology 16, 967-971.Garnero, E. 3. (2004). A new paradigm for Earth’s core-mantle boundary. Science 304,834-836.Greene, A., Scoates, J. S., Weis, D. & Kieffer, B. (2006a). Wrangellia Flood Basalts:Exploring the architecture and composition of an accreted oceanic plateau.Geological Society ofAmerica Cordilleran Section Anchorage, Alaska. 38: 5,p.25.Greene, A. R., Scoates, J. S., Weis, D. & Kieffer, B. (2006b). Wrangellia Flood Basalts:Exploring the architecture and composition of an accreted oceanic plateau.Geological Association ofCanada 2006 Annual Meeting, Montreal, Quebec.Greene, A. R., Scoates, J. S., Nixon, G. T. & Weis, D. (2006c). Picritic lavas and basalsills in the Karmutsen flood basalt province, Wrangellia, northern VancouverIsland. In: Grant, B. (ed.) Geological Fieldwork 2005. British Columbia Ministryof Energy, Mines and Petroleum Resources Paper 2006-1,pp.39-52.20Greene, A. R., Scoates, J. S. & Weis, D. (2005a). Wrangellia Terrane in B.C. and Yukon:Where are its Ni-Cu-PGE deposits? Cordilleran Mineral Roundup Expo,Vancouver, BC.Greene, A. R., Scoates, J. S. & Weis, D. (2005b). Wrangellia Terrane on VancouverIsland: Distribution of flood basalts with implications for potential Ni-Cu-PGEmineralization. In: Grant, B. (ed.) Geological Fieldwork 2004. British ColumbiaMinistry of Energy, Mines and Petroleum Resources Paper 2005-1,pp.209-220.Greene, A. R., Scoates, J. S. & Weis, D. (2008a). Wrangellia flood basalts in Alaska: Arecord of plume-lithosphere interaction in a Late Triassic accreted oceanicplateau. Goldschmidt 2008 Conference, Vancouver, BC.Greene, A. R., Scoates, J. S., Weis, D. & Israel, S. (2005c). Flood basalts of theWrangellia Terrane, southwest Yukon: Implications for the formation of oceanicplateaus, continental crust and Ni-Cu-PGE mineralization. In: Emond, D. S.,Lewis, L. L. & Bradshaw, G. D. (eds.) Yukon Exploration and Geology 2004.Yukon Geological Survey, pp. 109-120.Greene, A. R., Scoates, J. S., Weis, D., Israel, S., Nixon, G. T. & Kieffer, B. (2007a).Geochemistry and geochronology of the Wrangellia Flood Basalts in BritishColumbia, Yukon, and Alaska. Cordilleran Mineral Roundup Expo, Vancouver,BC.Greene, A. R., Scoates, J. S., Weis, D. & Nixon, G. T. (200Th). Significance of picriticand tholeiitic lavas within Wrangellia Flood Basalts on Vancouver Island for themelting history and magmatic evolution of a major oceanic plateau. AGU 2007Fall Meeting San Francisco, CA. Abstract V33A-1 153Greene, A. R., Scoates, J. S., Weis, D. & Nixon, G. T. (2008b). Picritic and tholeiiticlavas within Wrangellia Flood Basalts on Vancouver Island for the meltinghistory and magmatic evolution of a major oceanic plateau. Cordilleran TectonicsWorkshop, Vancouver, BC.Greene, A. R., Weis, D. & Scoates, J. S. (2007c). Wrangellia Flood Basalts: Exploringthe architecture and composition of an accreted oceanic plateau. GeologicalAssociation ofCanada, NUNA 2007 Conference: The Pulse ofthe Earth andPlanetary Evolution, Sudbury, Ontario.Griffiths, R. W. & Campbell, I. H. (1990). Stirring and structure in mantlestartingplumes. Earth and Planetary Science Letters 99, 66-78.Hilihouse, J. W. (1977). Paleomagnetism of the Triassic Nikolai Greenstone, McCarthyQuadrangle, Alaska. Canadian Journal ofEarth Sciences 14, 2578-3592.Hillhouse, J. W. & Coe, R. S. (1994). Paleomagnetic data from Alaska. In: Plaficer, G. &Berg, H. C. (eds.) The Geology ofAlaska. Geological Society of AmericaGeology of North America G-l,pp.797-8 12.Inoue, H., Coffin, M. F., Nakamura, Y., Mochizuki, K. & Kroenke, L. W. (2008).Intrabasement reflections of the Ontong Java Plateau: Implications for plateauconstruction. Geochemistry Geophysics Geosystems 9(Q0401 4),doi:10. 1029/2007GC001780.Israel, S. (2004). Geology of Southwestern Yukon (1:250 000 scale). Yukon GeologicalSurvey Open File 2004-16.21Jellinek, A. M. & Manga, M. (2004). Links between long-lived hotspots, mantle plumes,D”, and plate tectonics. Reviews ofGeophysics 42(RG3002),doi: 10.1 029/2003RG000 144.Jones, D. L., Silberling, N. J. & Hillhouse, J. (1977). Wrangellia; a displaced terrane innorthwestern North America. Canadian Journal ofEarth Sciences 14(11), 2565-2577.Katvala, E. C. & Henderson, C. M. (2002). Conodont sequence biostratigraphy andpaleogeography of the Pennsylvanian-Permian Mount Mark and Fourth LakeFormations, southern Vancouver Island. In: Hills, L. V., Henderson, C. M. &Bamber, E. W. (eds.) Carbonferous and Permian ofthe World. Canadian Societyof Petroleum Geologists Memoir 19,pp.46 1-478.Kelley, S. (2007). The geochronology of large igneous provinces, terrestrial impactcraters, and their relationship to mass extinctions on Earth. Journal oftheGeological Society ofLondon 164, 923-93 6.Kerr, A. C. (2003). Oceanic Plateaus. In: Rudnick, R. (ed.) The Crust. Elsevier Science:Oxford Treatise on Geochemistry Vol. 3 Holland, H. D. & Turekian, K. K. (eds.),pp.537-565.Kerr, A. C. (2005). Oceanic LIPs: The kiss of death. Elements 1(5), 289-292.Kerr, A. C., Tarney, J., Nivia, A., Marriner, G. F. & Saunders, A. D. (1998). The internalstructure of oceanic plateaus: inferences from obducted Cretaceous terranes inwestern Columbia and the Caribbean. Tectonophysics 292, 173-188.King, S. D. & Anderson, D. L. (1998). Edge-driven convection. Earth and PlanetaryScience Letters 160, 289-296.Korenaga, J. (2005). Why did not the Ontong Java Plateau form subaerially? Earth andPlanetary Science Letters 234, 385- 399.Kroenke, L. W. (1974). Origin of continents through development and coalescence ofoceanic flood basalt plateaus. Eos 55, 443.Lassiter, J. C., DePaolo, D. J. & Mahoney, J. J. (1995). Geochemistry of the Wrangelliaflood basalt province: Implications for the role of continental and oceaniclithosphere in flood basalt genesis. Journal ofPetrology 36(4), 983-1009.Le Bas, M. J. (2000). [UGS reclassification of the high-Mg and picritic volcanic rocks.Journal ofPetrology 41(10), 1467-1470.Mahoney, J. J. & Coffin, M. F. (eds.) (1997). Large Igneous Provinces: Continental,Oceanic, and Planetary Flood Volcanism. AGU Geophysical Monograph 100.438 pp.Mahoney, J. J., Fitton, J. G., Wallace, P. J. & al., e. (2001). Proceedings of the OceanDrilling Program, Initial Reports Volume 192. Available from World Wide Web:http://odp.pangaea.de/publications/1 92_IR/VOLUME/CHAPTERS/IR192_0 1 .PDF.Massey, N. W. D., Macintyre, D. G., Desjardins, P. J. & Cooney, R. T. (2005a). DigitalGeology Map of British Columbia: Tile NM9 Mid Coast, B.C. B.C. Ministry ofEnergy and Mines Geofile 2005-2.Massey, N. W. D., Maclntyre, D. G., Desjardins, P. J. & Cooney, R. T. (2005b). DigitalGeology Map of British Columbia: Tile NM1O Southwest B.C. B.C. Ministry ofEnergy and Mines Geofile 2005-3.22Nixon, G. T., Larocque, J., Greene, A. R. & Scoates, J. S. (2008a). Stratigraphic settingand metallogenic significance of high-Mg lavas in the Karmutsen flood basalts,northern Vancouver Island, British Columbia. GA C-MA C 2008, Quebec City,Quebec.Nixon, G. T., Laroque, J., Pals, A., Styan, J., Greene, A. R. & Scoates, J. S. (2008b).High-Mg lavas in the Karmutsen flood basalts, northern Vancouver Island (NTS092L): Stratigraphic setting and metallogenic significance. In: Grant, B. (ed.)Geological Fieldwork 2007. B.C. Ministry of Energy, Mines and PetroleumResources Paper 2008-1,pp.175-190.Nokleberg, W. J., Plafker, G. & Wilson, F. H. (1994). Geology of south-central Alaska.In: Plafker, G. & Berg, H. C. (eds.) The Geology ofNorth America. GeologicalSociety of America: Boulder, CO The Geology of Alaska G-1,pp.311-366.Ogg, J. G. (2004). The Triassic Period. In: Gradstein, F. M., Ogg, J. G. & Smith, A. G.(eds.) A Geologic Time Scale 2004. Cambridge University Press: Cambridge,pp.271-306.Parrish, R. R. & McNicoll, V. J. (1992). U-Pb age determinations from the southernVancouver Island area, British Columbia. Geological Survey ofCanada.Radiogenic Age and Isotopic Studies: Report 5 Paper 9 1-2, 79-86p.Petterson, M. G., Babbs, T., Neal, C. R., Mahoney, J. J., Saunders, A. D., Duncan, R. A.,Tolia, D., Magu, R., Qopoto, C., Mahoa, H. & Natogga, D. (1999). Geological-tectonic framework of Solomon Islands, SW Pacific: crustal accretion and growthwithin an intra-oceanic setting. Tectonophysics 301, 35-60.Plafker, G. & Berg, H. C. (1994). Overview of the geology and tectonic evolution ofAlaska. In: G., P. & H.C., B. (eds.) The Geology ofNorth America. GeologicalSociety of America: Boulder, CO The Geology of Alaska G-l,pp.989-1021.Plafker, G., Moore, J. C. & Winider, G. R. (1994). Geology of the southern Alaskamargin. In: Plaficer, G. & Berg, H. C. (eds.) The Geology ofNorth America.Geological Society of America: Boulder, CO The Geology of Alaska G-l,pp.389-449.Plafker, G., Nokleberg, W. J. & Lull, J. S. (1989). Bedrock geology and tectonicevolution of the Wrangellia, Peninsular, and Chugach terranes along the TransAlaskan Crustal Transect in the northern Chugach Mountains and southernCopper River basin, Alaska. Journal of Geophysical Research 94, 4,255-4,295.Rampino, M. R. & Stothers, R. B. (1988). Flood basalt volcanism during the past 250million years. Science 241, 663-668.Révillon, S., Arndt, N. T., Hallot, E., Kerr, A. C. & Tarney, J. (1999). Petrogenesis ofpicrites from the Caribbean Plateau and the North Atlantic magmatic province.Lithos49, 1-21.Richards, M. A., Jones, D. L., Duncan, R. A. & DePaolo, D. 3. (1991). A mantle plumeinitiation model for the Wrangellia flood basalt and other oceanic plateaus.Science 254, 263-267.Rioux, M., Hacker, B., Mattinson, J., Kelemen, P., Blusztajn, 3. & Gehrels, G. (2007).Magmatic development of an intra-oceanic arc: High-precision U-Pb zircon andwhole-rock isotopic analyses from the accreted Talkeetna arc, south-centralAlaska. Geological Society ofAmerica Bulletin 119(9-10), 1168-1184,10.1 130/b25964.l.23Rogers, G. (1982). Oceanic plateaus as meteorite impact signatures. Nature 299, 341-342.Saunders, A. D. (2005). Large igneous provinces: Origin and environmentalconsequences. Elements 1, 259-263.Saunders, A. D., Storey, M., Kent, R. W. & Norry, M. 3. (1992). Consequencesof plume-lithosphere interactions. hi: Storey, B. C., Alabaster, T. & Pankhurst,R. J. (eds.)Magmatism and the Causes ofContinental Breakup. Geological Society ofLondon Special Publication 68: London,pp.4 1-60.Scoates, J. S., Greene, A. R. & Weis, D. (2006). Platinum group element geochemistry ofthe Karmutsen flood basalt province, Wrangellia terrane, Vancouver Island.Geological Society ofAmerica Cordilleran Section, Anchorage, Alaska. 38: 5,p.26.Scoates, J. S., Greene, A. R. & Weis, D. (2008). Platinum group element geochemistry oflarge igneous provinces. Goldschmidt 2008 Conference, Vancouver, BC.Scoates, J. S., Greene, A. R., Weis, D., Israel, S. & Nixon, G. T. (2007). PGEgeochemistry and suiphide saturation state ofthe Triassic Wrangellia basalts,Vancouver Island and Yukon. Cordilleran Mineral Roundup Expo, Vancouver,BCScotese, C. R. (2004). PALEOMAP project. http://www.scotese.com/research.htm.Self, S., T., T. & L., K. (1997). Emplacement of continental flood basalt lava flows. In:Mahoney, J. J. & Coffm, M. F. (eds.) Large Igneous Provinces: Continental,Oceanic, and Planetaiy Flood Volcanism. American Geophysical Union:Washington. Geophysical Monograph 100,pp.38 1-410.Self, S., Widdowson, M., Thordarson, T. & Jay, A. E. (2006). Volatile fluxes duringflood basalt eruptions and potential effects on the global environment: A Deccanperspective. Earth and Planetary Science Letters 248, 518-532.Smith, W. H. F. & Sandwell, D. T. (1997). Global Sea Floor Topography from SatelliteAltimetry and Ship Depth Soundings. Science 277, 1956-1962.Storey, M., Mahoney, J. J., Kroenke, L. W. & Saunders, A. D. (1991). Are oceanicplateaus sites of komatiite fonnation? Geology 19, 376-379.Taylor, B. (2006). The single largest oceanic plateau: Ontong Java-Manihilci-Hikurangi.Earth and Planetary Science Letters 241(3-4), 372-380.Tejada, M. L. G., Mahoney, 3. 3., Castillo, P. R., Ingle, S. P., Sheth, H. C. & Weis, D.(2004). Pin-pricking the elephant: evidence on the origin of the Ontong JavaPlateau from Pb-Sr-Hf-Nd isotopic characteristics of ODP Leg 192 basalts. In:Fitton, J. G., Mahoney, 3. J., Wallace, P. 3. & Saunders, A. D. (eds.) Origin andEvolution ofthe Ontong Java Plateau. Geological Society of London, SpecialPublication 229,pp.133-150.Tejada, M. L. G., Mahoney, 3. 3., Duncan, R. A. & Hawkins, M. P. (1996). Age andgeochemistry of basement and alkalic rocks of Malaita and Santa Isabel, SolomonIslands, Southern Margin of Ontong Java Plateau. Journal ofPetrology 37(2),361-394.Tejada, M. L. G., Mahoney, J. J., Neal, C. R., Duncan, R. A. & Petterson, M. G. (2002).Basement geochemistry and geochronology of Central Malaita, Solomon Islands,with implications for the origin and evolution of the Ontong Java Plateau. JournalofPefrology 43(3), 449-484.24Weis, D. & Frey, F. A. (2002). Submarine basalts of the northern Kerguelen plateau:interaction between the Kerguelen plume and the southeast Indian Ridge revealedat ODP Site 1140. Journal ofPetrology 43(7), 1287-1309.Wignall, P. B. (2001). Large igneous provinces and mass extinctions. Earth-ScienceReviews 53, 1-33.Wilson, F. H., Dover, J. D., Bradley, D. C., Weber, F. R., Bundtzen, T. K. & Haeussler,P. J. (1998). Geologic map of Central (Interior) Alaska. U S. Geological SurveyOpen-File Report 98-133-A. http://wrgis.wr.usgs.gov/open-file/of98- 1 33-al.Wilson, F. H., Labay, K. A., Shew, N. B., Preller, C. C., Mohadjer, S. & Richter, D. H.(2005). Digital Data for the Geology of Wrangell-Saint Elias National Park andPreserve, Alaska U S. Geological Survey Open-File Report 2005-1342,http://pubs.usgs.gov/of’2005/1 342/.Yole, R. W. & Irving, E. (1980). Displacement of Vancouver Island, paleomagneticevidence from the Karmutsen Formation. Canadian Journal ofEarth Sciences 17,1210-1228.25CHAPTER 2Wrangellia Flood Basalts on Vancouver Island:Significance of Picritic and Tholeiitic Lavas forthe Melting History and Magmatic Evolution ofa Major Oceanic Plateau‘A version of this chapter has been submitted for publication.26INTRODUCTIONThe largest magmatic events on Earth have led to the formation of oceanicplateaus. Oceanic plateaus are transient large igneous provinces (LIPs) that cover up totwo million square kilometers of the ocean floor and form crustal emplacements 20-40km thick with submarine or subaerial flood basalt sequences six or more kilometers thick(Coffin & Eldholm, 1994). LIPs typically form in one or more pulses lasting less than afew million years. Three key aspects of LIPs are that they form from unusually high meltproduction rates, they are predominantly basaltic in composition, and their formation isoften not directly attributable to seafloor spreading processes (e.g. Saunders, 2005). Thehigh melt production rates are best explained by high mantle source temperatures ofrapidly upwelling mantle (i.e. mantle plumes) and direct evidence of high mantletemperatures is eruption of high-MgO, near-primary lavas (Kerr & Mahoney, 2007).Studies of oceanic plateaus may provide important information about their constructionand growth history, the temperature and depth of melting of the mantle source, thevolume of magma and melt production rates, and the composition of the mantle source.Combined, this information serves to constrain aspects of the mantle plume hypothesisfrom the origin of components in the source to emplacement of voluminous lavasequences. From a geochemical perspective, study of oceanic plateaus is importantbecause the lavas erupted in oceanic plateaus are generally unaffected by continentalcontamination.Presently, one of the great challenges in studying oceanic plateaus in the oceanbasins is the difficulty in sampling more than the uppermost few hundred meters of flows(e.g. Kerguelen, Ontong Java, etc.). Accreted sections of oceanic plateaus (e.g.Caribbean, Ontong Java) provide an opportunity to closely examine their volcanicstratigraphy. Wrangellia flood basalts in the Pacific Northwest ofNorth America areparts of an immense LIP that erupted in a marine setting and accreted to western NorthAmerica in the Late Jurassic or Early Cretaceous (Richards et al., 1991). Although theiroriginal areal distribution was likely considerably larger, current exposures of the floodbasalts extend in a thin belt over 2300 km in British Columbia (BC), Yukon, and Alaskaand retain a large part of their original stratigraphic thickness (—‘6 km on VancouverIsland; —‘3.5 km in Alaska). The present extent of the Wrangellia oceanic plateau27remnants is primarily a result of transform fault motions that occurred along westernNorth America after accretion, rather than an indication of the original size of the plateau.The flood basalts erupted in the Middle to Late Triassic (—23 1-225 Ma; Richards et a!.,1991) atop different-aged Paleozoic arc volcanic and marine sedimentary sequences andare overlain by Late Triassic limestone. On Vancouver Island, a tripartite succession offlood basalts includes submarine, volcaniclastic, and subaerial flows formed as part of anenormous emergent oceanic plateau. Richards et a!. (1991) proposed a plume initiationmodel for the Wrangellia flood basalts based on evidence of rapid uplift prior tovolcanism, lack of evidence of rifting associated with volcanism (i.e. few dikes andabundant sills), and the short duration and high eruption rate of volcanism. Wrangelliaflood basalts are perhaps the most extensive aecreted remnants of an oceanic plateau inthe world where parts of the entire volcanic stratigraphy are exposed, but they have beenthe focus of only one study in the last 20 years using multiple types of isotopic andgeochemical data (Lassiter et al., 1995).In this study, we examine the field relationships, stratigraphy, petrography, majorand trace elements, and Sr-Nd-Hf-Pb isotopic compositions of Wrangellia flood basaltsfrom different areas of Vancouver Island to assess the nature of the mantle source and toevaluate the melting history and subsequent magmatic evolution of basalts involved inthe construction of this major oceanic plateau. The geochemistry of picritic and tholeiiticbasalts that form the volcanic stratigraphy of this oceanic plateau offers a view of themelting history of plume-derived magmas that does not involve continental lithosphereand where source heterogeneity does not have a major role. This study is one part of alarge research project on the nature, origin and evolution of the Triassic Wrangellia floodbasalts in British Columbia, Yukon and Alaska.GEOLOGIC SETTINGWrangeffia on Vancouver IslandWrangellia flood basalts form the core of the Wrangellia terrane, or Wrangellia,one of the largest outboard terranes accreted to western North America (Jones et a!.,1977). Middle to Late Triassic flood basalts extend in a discontinuous belt fromVancouver and Queen Charlotte Islands (Karmutsen Formation), through southeast28Alaska and southwest Yukon, and into the Wrangell Mountains, Alaska Range, andTalkeetna Mountains in east and central Alaska (Nikolai Formation) (Fig. 2.1).Wrangellia covers approximately 80% of Vancouver Island, which is 460 km longby 130 km wide (Fig. 2.1). Wrangellia is the uppermost sheet of a stack of southwestvergent thrust sheets that form the crust of Vancouver Island and has a cumulativethickness of>10 km (Monger & Journeay, 1994). Pre-Karmutsen units of Wrangellia onVancouver Island are Devonian arc sequences of the Sicker Group and Mississippian toEarly Permian siliciclastic and carbonate rocks of the Buttle Lake Group (Muller, 1980;Sutherland-Brown eta!., 1986). The Paleozoic formations have varied estimatedthicknesses (—P2-5 km) (Muller, 1980; Massey, 1 995a) and are only exposed on centraland southern Vancouver Island (Fig. 2.1). The Karmutsen Formation is overlain by theshallow-water Quatsino Limestone (30-750 m) and deeper-water Parson Bay Formation(-600 m), which is intercalated with and overlain by Bonanza arc volcanics (169-202Ma; Nixon et a!., 2006b). Bonanza arc plutonic rocks (167-197 Ma) also intrude theKarmutsen Formation and are some of the youngest units of Wrangellia that formed priorto accretion with North America (Nixon et a!., 2006b). Following accretion, Wrangelliaunits were intruded by the predominantly Cretaceous Coast Plutonic Complex (Wheeler&McFeely, 1991).The Karmutsen Formation (19,142 km2,based on mapped areas from digitalgeologic maps) is composed ofbasal sediment-sill complexes, a lower member ofpillowed and unpillowed submarine flows, a middle member of mostly pillow brecciaand hyaloclastite, and an upper member of predominantly massive subaerial flows(Carlisle & Suzuki, 1974). The pillow basalts directly overlie thick sediment-sillcomplexes composed of mafic sills intruding Middle Triassic pelagic sediments and LatePaleozoic formations. The boundary between pillow breccialhyaloclastite and massivelava flows represents the transition from a submarine to a subaerial eruptive environment.The uppermost flows of the Karmutsen are intercalated and overlain by shallow-waterlimestone and local occurrences of submarine flows occur within the upper subaerialmember. The Karmutsen Formation and Wrangellia on Vancouver Island were the focusof mapping efforts and stratigraphic descriptions by Carlisle (Carlisle, 1963; Carlisle,1972; Carlisle & Suzuki, 1974) and Muller (Muller et a!., 1974; Muller, 1977). Recent29Figure 2.1 Simplified map of Vancouver Island showing the distribution of the Karmutsen Formation(gray) and underlying Paleozoic formations (black; after Massey eta?., 2005a, 2005b). The main areasof field study are indicated with boxes or circles with capital letters (see legend). Ocean is light gray.The inset shows the extent of the Wrangellia flood basalts (gray) in British Columbia, Yukon, andAlaska.-5ON12W WMiddle—Late TriassicKarmutsen FormationPaleozoic—Middle TriassicSicker and Buttle Lake GroupsKarmutsen Range (Alice—Nimpkish Lake>Schoen Lake Provincial Park areaMount Arrowsmith areaL) Buttle LakeHolberg Inlet30descriptions of the Karmutsen Formation on northern Vancouver Island have been madeduring regional mapping studies (1:50,000 scale) on northern Vancouver Island by Nixonet al. (Nixon eta!., 2006b; Nixon & Orr, 2007; Nixon eta!., 2008).Age of the Karmutsen FormationThe age and duration of Karmutsen volcanism are constrained by fossils in theunderlying and overlying sedimentary units and by three U-Pb isotopic agedeterminations on intrusive rocks that are likely related to the Karmusten volcanics.Daone!la-bearing shale 100-200 m below the base of the pillow basalts on SchoenMountain and Halobia-rich shale interlayered with flows in the upper part of theKarmutsen indicate eruption of the flood basalts in the Upper Ladinian (Middle Triassic)to Upper Carnian (Late Triassic; Carlisle and Suzuki, 1974). The only published U-Pbage is based on a single concordant analysis of a multi-grain baddeleyite fraction from agabbro on southern Vancouver Island that yielded a206Pb/238Uage of 227.3± 2.6 Ma(Parrish & McNicoll, 1992). Two unpublished 206Pb/238Ubaddeleyite ages, also from agabbro on southern Vancouver Island, are 226.8 ± 0.5 Ma (5 fractions) and 228.4 ± 2.5 (2fractions; Sluggett, 2003).VOLCANIC STRATIGRAPHY AND PETROGRAPIIYField studies undertaken on Vancouver Island in 2004-2006 explored the volcanicstratigraphy of the Karmutsen flood basalts in three main areas: the Karmutsen Range(between Alice and Nimpkish lakes), the area around Schoen Lake Provincial Park, andaround Mount Arrowsmith (Greene eta!., 2005; Greene eta!., 2006), and also aroundHolberg Inlet, on northernmost Vancouver Island, and Buttle Lake (Fig. 2.1). Thecharacter and thickness of the flood basalt sequences vary locally, although the tripartitesuccession of the Karmutsen Formation appears to be present throughout VancouverIsland. The stratigraphic thicknesses for the pillow, pillow breccia, and massive flowmembers are estimated at —‘2600 m, 1100 m, and 2900 m, respectively, in the type areaaround Buttle Lake (Surdam, 1967); on northern Vancouver Island estimated thicknessesare >3000 m, 400-1500 m, and >1500 m, respectively (Nixon et al., 2008); on Mount31Arrowsmith and nearby areas on southern Vancouver Island estimates are 1100 m, 950m, and 1200 m, respectively (Yorath et a!., 1999; Fig. 2.1).Picritic pillow lavas occur west of the Karmutsen Range on northern VancouverIsland, in a roughly triangular-shaped area (30 km across) bounded by Keogh, Maynardand Sara lakes (Figs 2.2 and 2.3; Greene et a!., 2006; Nixon et al., 2008). Excellentexposures ofpicritic pillow lavas occur in roaclcuts along the north shore of Keogh Lake,the type locality (Greene eta!., 2006). The Keogh Lake picrites mostly form pillowedflow units (<15 m thick), with pillows and tubes of varied dimensions (typically <1 mwide), and unpillowed flows. Numerous thermal contraction features in the pillows arefilled with quartz-carbonate, such as drain-back ledges, tortoise-shell jointing, andinterpillow voids containing spalled rims (Fig. 2.3; Greene eta!., 2006). The picriticpillow basalts are not readily distinguishable in the field from basalt, except by theirdensity and non-magnetic character and minor interpillow quartz-carbonate. Recentfieldwork and mapping indicates that the picrites occur mostly near the transitionbetween pillow lava emplacement and hyaloclastite deposition (Nixon et a!., 2008).In and around Schoen Lake Provincial Park, there is a well-preserved sediment-sill complex at the base of the Karmutsen Formation (Figs 2.2 and 2.4). Middle Triassicmarine sedimentary rocks overlie Pennsylvanian to Permian limestone and siliceoussedimentary rocks, and both successions are intruded by mafic sills related to theoverlying flood basalts (Carlisle, 1972). Carlisle (1972) estimated the sediment-sillcomplex to be approximately 600-900 m thick with a total thickness of 150-200 m ofpreintrusive sedimentary rocks. The Triassic sedimentary rocks range from thinly-beddedsiliceous and calcareous shale to banded chert and finely-laminated, Daonella-bearingshale (Carlisle, 1972). The basal sediment-sill complex is immediately overlain by thicksuccessions of submarine flows (Fig. 2.4). Basal sills and pre-Karmutsen sediments arealso exposed around Buttle Lake.Exposures with large vertical relief around Mount Arrowsmith (Appendix A) andButtle Lake preserve thick successions of pillow basalt, breccias, and subaerial flows andthere are rarely sediments between flows or in interpillow voids. Unpillowed submarineflows interspersed with pillowed flows are locally recognizable by irregular, hacklycolumnar jointing. The massive subaerial flows form monotonous sequences marked32BONANZA VOLCANICSwaterlain pyroclastics, subaerial basalt,rhyolite flows, tuff, and intrusive suites(Island Plutonic Suite)HARBLEDOWN FORMATIONcalcareous siltstone, feldspathic wackePARSON BAY FORMATION1Jwell—bedded siliceous limestone and]wacke, with minor volcanicsQUATSINO FORMATIONmassiveto well—bedded micriticand locally bioclastic limestoneintra—Karmutsen limestone lensesKARMUTSEN FORMATIONsubaerial flows with minorpillow basalt and hyaloclastite(—3000 m)-massive lavaipillow lavapicritejabzT....._., Late Triassic-Middle Jurassic e to Late TriassicAlert Bay Volcanics ]Bonanza GroupL.._.uatsino FormationUpper CretaceousParson Bay FormationKarmutsen FormationNanaimo Group Early to Middle Jurassic/TJUndivided Parson Bayequivalentsand Bonanza sedimentsl5tsfld Plutonic SuiteFigure 2.2 Geologic map and stratigraphy of the Schoen Lake Provincial Park and Karmutsen Range areas(locations shown in Figure 2.1, northern Vancouver Island). (a) Stratigraphic column depicts units exposed inthe Schoen Lake area, derived from Carlisle (1972) and fieldwork. (b) Generalized geology for the SchoenLake area with sample locations. Map derived from Massey et a!. (2005a). The exposures in the Schoen Lakearea are the lower volcanic stratigraphy and base of the Kannutsen Formation. (c) Stratigraphic column forgeology in the Alice-Nimpkish Lake area, derived from Nixon & Off (2007). (d) Geologic map frommapping of Nixon & Orr (2006a) and Nixon et a!. (2008). Sample sites and lithologies are denoted in thelegend. The Keogh Lake picrites are exposed near Keogh, Sara, and Maynard lakes and areas to the east ofMaynard Lake (Nixon eta!., 2008).pillowed and unpillowed flows, brecciaand hyaloclastite(—2500 m)silicified shale, chert and limestonewith Daonella beds intruded by mafic sillsBUTTLE LAKE GROUPmassive bioclastic limestone (Mount MarkFm.), variety of chert, thinly——bedded shale, and limestone (Fourth Lake(a)Fm.)Early-MIddle JurassicIsland Plutonic SuiteLate Triassic- Early JurassicParson Bay FormationMIddle-Lower TriasslcQuatsino LimestoneKarmutsen FormationShale-chert-limestone(b)LEMARE LAKE VOLcANIcSMississlpplan-Permlan[]Buttle Lake Group(sedimentary rocks)massive lava/o sill I gabbro/fault•pillow lavao shale or chertriver— park boundaryIsubaerialbasalt andrhyolite flows, breccia,and tuffminor pillow lava, hyaloclastite,debris—flow and epiclasticdeposits— plag—megacrystic flows-interbedded volcaniclastic andsedimentary rocks‘UaIIt-JPARSON BAY FORMATIONwell—bedded shale, limestone,wacke, with minor volcanicsc breccia, tuff, reworkedcs, minor pillows andINOF Nlimestone lensesfKARMUTSEN FORMATIONsubaerial flows with minorpillow basalt and hyaloclastiteIpillow breccia and hyaloclastite—Keogh Lake picrite(mostly pillow lavas)pillowed and unpillowed flows(c)3334- 2.3 P1 - - of picritic and tholeiitic pillow basalts from the Karmutsen Range area (Alice andNimpkish Lake area), northern Vancouver Island. (a) Stack of dense, closely packed, asymmetric picriticpillows with radial vesicle infillings (circled sledgehammer 8O cm long for scale). (b) Picritic pillow basaltsin cross-section near Maynard Lake. Note vesicular upper margin to pillows. (c) Unpillowed submarine flowwith coarse, hackly columns (marked with arrows) draped over high-MgO pillow basalt. (d) Spalled rimsfrom pillows filling voids between tholeiitic pillow basalt (coin for scale). (e) Cross-section of a large picriticpillow lobe with infilling of quartz-carbonate in cooling-contraction cracks (sledgehammer for scale).(f) High-MgO pillow breccia stratigraphically between submarine and subaerial flows (lens cap for scale --7cm diameter).Figure 2.4 Photographs showing field relations from the Schoen Lake Provincial Park area, northernVancouver Island. (a) Sediment-sill complex at the base of the Karmutsen Formation on the north side ofMount Adam (see Fig. 2.2 for location). (b) Interbedded mafic sills and deformed, finely-banded chert andshale with calcareous horizons, from location between Mt. Adam and Mt. Schoen.35Arrowsmith and nearby areas on southern Vancouver Island estimates are 1100 m, 950m, and 1200 m, respectively (Yorath et al., 1999; Fig. 2.1).Picritic pillow lavas occur west of the Karmutsen Range on northern VancouverIsland, in a roughly triangular-shaped area (--‘30 km across) bounded by Keogh, Maynardand Sara lakes (Figs 2.2 and 2.3; Greene et a!., 2006; Nixon et al., 2008). Excellentexposures ofpicritic pillow lavas occur in roadcuts along the north shore of Keogh Lake,the type locality (Greene eta!., 2006). The Keogh Lake picrites mostly form pillowedflow units (<15 m thick), with pillows and tubes of varied dimensions (typically <1 mwide), and unpillowed flows. Numerous thermal contraction features in the pillows arefilled with quartz-carbonate, such as drain-back ledges, tortoise-shell jointing, andinterpillow voids containing spalled rims (Fig. 2.3; Greene et a?., 2006). The picriticpillow basalts are not readily distinguishable in the field from basalt, except by theirdensity and non-magnetic character and minor interpillow quartz-carbonate. Recentfieldwork and mapping indicates that the picrites occur mostly near the transitionbetween pillow lava emplacement and hyaloclastite deposition (Nixon et aL, 2008).In and around Schoen Lake Provincial Park, there is a well-preserved sediment-sill complex at the base of the Karmutsen Formation (Figs 2.2 and 2.4). Middle Triassicmarine sedimentary rocks overlie Pennsylvanian to Permian limestone and siliceoussedimentary rocks, and both successions are intruded by mafic sills related to theoverlying flood basalts (Carlisle, 1972). Carlisle (1972) estimated the sediment-sillcomplex to be approximately 600-900 m thick with a total thickness of 150-200 m ofpreintrusive sedimentary rocks. The Triassic sedimentary rocks range from thinly-beddedsiliceous and calcareous shale to banded chert and finely-laminated, Daonel!a-bearingshale (Carlisle, 1972). The basal sediment-sill complex is immediately overlain by thicksuccessions of submarine flows (Fig. 2.4). Basal sills and pre-Karmutsen sediments arealso exposed around Buttle Lake.Exposures with large vertical relief around Mount Arrowsmith (Appendix A) andButtle Lake preserve thick successions of pillow basalt, breccias, and subaerial flows andthere are rarely sediments between flows or in interpillow voids. Unpillowed submarineflows interspersed with pillowed flows are locally recognizable by irregular, hacklycolumnar jointing. The massive subaerial flows form monotonous sequences marked36mainly by amygdaloidal horizons and brecciated flow tops are rarely observed. A singlelocality with well-preserved pahoehoe flow features is exposed north of Holberg Inlet(Nixon & Orr, 2007; Nixon et al., 2008). There is no evidence of significant detritalmaterial from a continental source in sediments associated with the flood basaltsanywhere on Vancouver Island. The uppermost Karmutsen flows are interbedded withthin(>4m) lenses of limestone, and rarely siliciclastic sedimentary rocks (Nixon et a!.,2006b). Plagioclase-phyric (>0.8 mm) trachytic-textured flows are also commonly foundnear the top of the Karmutsen Formation (Nixon et a!., 2006b).A total of 129 samples were collected from the Karmutsen Formation onVancouver Island and 63 samples were selected for geochemical work based on therelative degree of alteration and geographic distribution of the samples. Fifty-six of thesesamples have been divided into four groups based on petrography (Table 2.1) andgeochemistry, including tholeiitic basalt, picrite, high-MgO basalt, and coarse-grainedmafic rocks. The tholeiitic basalts are dominantly glomeroporphyritic with an intersertalto intergranular groundmass (Table 2.1). Plagioclase forms most of the phenocrysts andglomerocrysts and the groundmass is fme-grained plagioclase microlites, clinopyroxenegranules, small grains of Fe-Ti oxide, devitrified glass, and secondary minerals; there isno fresh glass. The picritic and high-MgO basaltic lavas exhibit plagioclase andclinopyroxene with spherulitic morphologies with abundant euhedral olivine phenocrysts(Fig. 2.5; Table 2.1). Only olivine is strongly altered and is completelypseudomorphically replaced by talc/tremolite. Plagioclase and clinopyroxene phenocrystsare not present in the high-MgO samples. The coarse-grained mafic rocks arecharacterized petrographically by subophitic texture with average grain size typically>1mm, and are generally non-glomeroporphyritic. The coarse-grained mafic rocks aremostly from the interiors of massive flows, although some may be from sills.SAMPLE PREPARATION AND ANALYTICAL METHODSOnly the freshest rocks in the field were sampled and only the least alteredsamples were selected for chemical and isotopic analysis based on thorough petrographicinspection. Sixty-three of the 129 collected samples (including 7 non-Karmutsen37Figure 2.5 Photomicrographs of picritic pillow basalts, Karmutsen Formation, northern Vancouver Island.Scale bars are 1 mm long. (a) Picrite with abundant euhedral olivine pseudomorphs from Keogh Lake typelocality (Fig. 2.2) in cross-polarized transmitted light (sample 4722A4; 19-19.8 wt % MgO, 4 analyses).Samples contain dense clusters (24-42 vol %) of olivine pseudomorphs (<2 mm) in a groundmass of curvedand branching sheaves of acicular plagioclase and intergrown with clinopyroxene and altered glass. Inmany cases, plagioclase nucleated on the edges of the olivine phenocrysts. (b) Sheaves of intergrownplagioclase and clinopyroxene in aphyric picrite pillow lava from west of Maynard Lake (Fig. 2.2) in crosspolarized transmitted light (sample 4723A2; 10.8 wt % MgO).38Table 2.1 Summary of oetroaraohic characteristics and ohenocrist orooortions of Karmutsen basalts on Vancouver Island. B.C.SampieAreab FIOWC GroupdTexture voi%011PIag Cpx Ox Alteration9 Note”4718A1 MA PIL THOL intersertal 20 5 3 few plag glcr <2 mm, cpx <1 mm471 8A2 MA PIL THOL intersertal, glomero 15 1 plag glcr <2 mm, very fresh471 8A5 MA PLO THOL intersertal, glomero 10 2 mottled, few plag glcr <4 mm471 8A6 MA BRE THOL porphyritic 20 2 plag 5-6 mm, sericite alteration4718A7 MA PIL THOL glomero 10 3 plag glcr<3 mm4719A2 MA PIL THOL porphyritic, intersertal 5 3 plag glcr <2 mm471 9A3 MA PIL THOL glomero 10 2 plag glcr <3 mm4720A2 SL BRE THOL glomero 5 3 plag glcr <2 mm4720A3 SL FLO THOL glomero 3 3 plag glcr <3 mm4720A4 SL FLO THOL glomero 5 1 plag glcr <5 mm, slightly cg, very fresh4720A5 SL PLO THOL senate 20 3 plag glcr <2 mm, plag needles and laths4720A8 SL FLO THOL glomero 20 2 plag glcr <2 mm, plag needles and laths4720A9 SL FLO THOL glomero 10 2 plag glcr<1.5 mm4721A1 SL PIL THOL glomero 15 3 plagglcr<1.5mm, plag needles aligned4721A2 SL FLO THOL glomero 10 1 plag glcr<3 mm, very fresh4721A3 SL FLO THOL glomero 20 3 plag glcr <2 mm, plag needles4721A4 SL FLO THOL glomero, ophimottled 20 5 1 plag glcr <1.5 mm, cpx <1 mm (oik), very fresh4721A5 SL FLO THOL glomero, ophlmottled 5 15 3 plag glcr <1.5 mm, cpx <1 mm (01k)4722A2 KR FLO THOL intengranular, intersertal 1 1 few plag phenos <2 mm4723A10 KR PLO THOL intengranular 5 3 10 3 few plag glcr <1 mm, ox 0.5-1 mm4724A5 SL PLO THOL porphyritic 25 3 plag 7-8mm5614A10 KR FLO THOL intergranular, intersertal 2 5 3 ox <0.5mm5614A11 KR FLO THOL intergranular, intersertal 5 7 3 ox 0.5-1 mm5614A13 KR FLO THOL lntergranular, porphynitlc 5 5 3 ox 0.5-1 mm561 5A1 KR PIL THOL aphyric, Intersertal 1 2 f.g., no phenos5615A8 KR PIL THOL intengranular 3 1 f.g., abundant small ox, very fresh5615A10 KR PIL THOL intengranular 3 3 2 plag glcr <4 mm5616A2 KR FLO THOL intergranular, intersertal 5 7 3 ox 0.2-0.5mm5618A1 Cl FLO THOL intergranular 3 2 10 3 ox 0.5-2 mm, c.g., plag laths >2 mm93G171 KR PIL PlC subophitic 23 01<2.3 mm, cpx<1 .5 mm, partially enc plag4722A4 KR PIL PlC cumulus, intergranular 35 10 3 01<1.5 mm, cpx<2 mm enc plag4723A3 KR PIL PlC spherulitic 31 3 01<1 mm, swtl plag <1.5mm4723A4 KR PIL PlC intergranular, intersertal 0 3 swtl plag <1 mm, no ci phenos4723A13 KR PIL PlC spherulitic 24 3 swtl plag <1 mm5614A1 KR PIL PlC spherulitic 24 1 01<1.5 mm, swtl plag <1 mm5615A7 KR PIL PlC cumulus, intergranular 42 10 1 01<1.5 mm, cpx <1.5mm enc plag5615A12 KR PIL PlC spherulltic 13 1 01 <2 mm, swtl plag <1 mm5616A1 KR PIL PlC iritengranular, intersertal 25 1 01 <2mm4723A2 KR PIL HI-MG spherulitic 0 3 swtl plag <2 mm, no ol phenos5614A3 KR PIL HI-MG spherulitic, intrafasciculate 12 1 01<1.5 mm, swtl plag <2mm561 4A5 KR BRE HI-MG porphyritic, ophimottled 13 2 ol <2 mm, cpx <2 mm enc plag561 6A7 KR PIL HI-MG intersertal 2 2 ol <2 mm, plag needles <1 mm4722A5 KR FLO OUTLIER intersertal 3 mottled, very f.g., v. small p1 needles5615A11 KR PIL OUTLIER intersertal 3 mottled, very f.g., v. small p1 needles4720A6 SL4720A7 SL4720A10 SL4724A3 SL5614A14 KR5614A15 KR5615A5 KR561 5A6 KR5616A3 KR5617A1 SL5617A5 SL5617A4 SL SIL MIN intersertal 5 3 to.PLO CGFLO CGSIL CGFLO CGSIL CGSIL CGsubophiticsubophiticsubophltic, Intrafasciculateophimottled, subophiticintergranularsubophiticGAB CGGAB CGGAB CGSIL CGSIL CG15 1 plag glcr <3 mm, cpx <3 mm25 1 plag chad <1.5 mm, cpx oik <2mm, very fresh5 3 plag chad<2 mm, cpx oik <3 mm, ox <1 mm, c.g.5 30 5 3 plag glcr 4-5 mm, cpx olk <2mm, ox <1 mm10 5 3 plag<lmm,cpx<2mm25 3 1 plag<1 mm, CPX 01k <3 mm, ox <0.2mmintergranular, plag-phynicintergranular, plag-phynicsubophiticintergranular, lntergrowthsintergranular, intersertal30 5 3 ox <0.5 mm, plag laths <4 mm15 10 3 plag<3mm,oxo.5-1.5mm25 5 1 plag laths <1 mm, cpx 01k <2mm, ox 0.5-1 mm30 3 plag <3 mm, cpx <5 mm, cpx-plag intergrowths1 1 plag <1 mm, f.g., mottledasamplenumber: last digit year, month, day, initial, sample station (except93G171).bMA,Mount Arrowsmith; SL, Schoen Lake; KR, Karmutsen Range. °PIL,pillow; BRE, breccia; PLO, flow; SIL,sill; GAB,gabbro.dTHOL,tholeiitic basalt; PlC, picrite; HI-MG, high-MgO basalt; CG,coarse-grained; MIN, mineralized sill.°glomero, glomeroporphyrltic.OIivine modes for picrites based on area calculations from scans (see Olivine accumulation in picritic lavas section and Figure2.13).9Visual alteration index based primarily on degree of plagioclase alteration and presence of secondary minerals (1, least altered; 3, most altered).Plagioclase phenocrysts commonly altered to albite, pumpellyite, and chlorite; divine is altered to talc, tremolite, and clinochiore (determined using the Rietveldmethod of X-ray powder diffraction); clinopyroxene is unaltered; Fe-Ti oxide commonly replaced by sphene.’glcr, glomerocrysts; f.g., fine-grained; c.g., coarsegrained, oik, oikocryst; chad, chadacryst; enc, enclosing, swtl, swallow-tail. Mineral abbreviations: 01, olivine pseudomorphs; plag, plagioclase; cpx,clinopyroxene; ox, oxides (includes ilmenite + titanomagnetite).39samples) were crushed (400 g) into small pieces <2 mm in diameter in a Rocklabshydraulic piston crusher between WC-plates to minimize contamination. The coarse-crush was mixed and 100 g was powdered in a planetary mill using agate jars and ballscleaned with quartz sand between samples.ActLabs Analytical MethodsThe major- and trace-element compositions of the whole rock powders weredetermined at Activation Laboratories Ltd. (Actlabs) in Ancaster, Ontario. Analyticaltechniques and detection limits are also available from Actlabs(http://www.actlabs.com/methsub_code4ere.htm). The particular analytical method foreach of the elements analyzed is indicated in Table 2.2. For the major elements, a 0.2 gsample was mixed with a mixture of lithium metaborate/lithium tetraborate and fused in agraphite crucible. The molten mixture was poured into a 5% HNO3solution and shakenuntil dissolved (30 minutes). The samples were analyzed for major oxides and selectedtrace elements on a combination simultaneous/sequential Thermo Jarrell-Ash Enviro IIinductively coupled plasma optical emission spectrometer (ICP-OES). Internalcalibration was achieved using a variety of international reference materials (e.g. W-2,BIR- 1, DNC- 1) and independent control samples. Additional trace elements wereanalyzed by both the INAA (instrumental neutron activation analysis) and ICP-MS(inductively couple plasma mass spectrometry) methods. For the 1NAA analyses, 1.5-2.5g of sample was weighed into small polyethylene vials and irradiated with controlinternational reference material CANIVIET WMS- 1 and NiCr flux wires at a thermalneutron flux of 7 x1012n cm2sin the McMaster Nuclear Reactor. Following a 7-daywaiting period, the samples were measured on an Ortec high-purity Ge detector linked toa Canberra Series 95 multichannel analyzer. Activities for each element were decay- andweight-corrected and compared to a detector calibration developed from multipleinternational certified reference materials. For the ICP-MS analyses, 0.25 g of samplewas digested in HF, followed by a mixture of HNO3and HC1O4,heated and taken todryness. The samples were brought back into solution with HC1. Samples were analyzedusing a Perkin Elmer Optima 3000 ICP. In-lab standards or certified reference materials40(e.g. W-2, BIR- 1, DNC- 1) were used for quality control. A total of 15 blind duplicateswere analyzed to assess reproducibility and the results were within analytical error.University ofMassachussetts XRF Analytical MethodsFourteen sample duplicate powders (high-MgO basalts and picrites) were alsoanalyzed at the Ronald B. Gilmore X-Ray Fluorescence (XRF) Laboratory at theUniversity of Massachussetts. Major elements were measured on a fused La-bearinglithium borate glass disc using a Siemens MRS-400 spectrometer with a Rh X-ray tubeoperating at 2700 W. Trace element concentrations (Rb, Sr, Ba, Ce, Nb, Zr, Y, Pb, Zn,Ga, Ni, Cr, V) were measured on a separate powder pellet using a Philips PW2400sequential spectrometer using a Rh X-ray tube. Loss-on-ignition (LOl) and ferrous ironmeasurements were made as described by Rhodes and younger (2004). Precision andaccuracy estimates for the data are described by Rhodes (1996) and Rhodes andyounger (2004). Results for each sample are the average of two separate analyses(shown in Appendix B).PCIGR Analytical MethodsA subset of 19 samples was selected for high-precision trace-element analysis andSr, Nd, Pb, and Hf isotopic analysis at the Pacific Centre for Isotopic andGeochemicalResearch (PCIGR) at the University of British Columbia (UBC). Samples were selectedfrom the 63 samples analyzed for whole-rock chemistry at ActLabs, based on major- andtrace-element chemistry, alteration (low LOT and petrographic alteration index), andsample location. Samples were prepared for trace-element analysis at thePCIGR by thetechnique described by Pretorius et al. (2006) on unleached rock powders.Samplepowders (—100 mg) were weighed in 7 mL screw-top Savillex®beakers and dissolved in1 mL --‘l4N HNO3and 5 mL 48% HF on a hotplate for 48 hours at130°C with periodicultrasonication. Samples were dried and redissolved in 6 mL 6N HC1 on ahotplate for 24hours and then dried and redissolved in 1 mL concentrated HNO3for 24hours beforefinal drying. Trace element abundances were measured with a ThermoFinniganElement2 High Resolution-Inductively Coupled Plasma-Mass Spectrometer (HR-ICPMS) following the procedures described by Pretorius et al. (2006)within 24 hours of41redissolution. High field strength elements (HFSE) and large ion lithophile elements(LILE) were measured in medium resolution mode at 2000x dilution using a PFA teflonspray chamber washed with Aqua Regia for 3 minutes between samples. Rare earthelements (REE) were measured in high resolution mode, and U and Pb in low resolutionmode, at 2000x dilution using a glass spray chamber washed with 2% HNO3betweensamples. Total procedural blanks and reference materials (BCR-2, BHVO-2) wereanalyzed with the batch of samples. Indium was used as an internal standard in allsamples and standard solutions. Background and standard solutions were analyzed afterevery 5 samples to detect memory effects and mass drift. Results for PCIGR trace-element analyses are shown in Appendix C.Sample digestion for purification of Sr, Nd , Hf, and Pb for column chemistrybegan by weighing each sample powder (400-500 mg) prior to leaching. All sampleswere leached with 6N HC1 and placed in an ultrasonic bath for 15 minutes. Samples wererinsed two times with 18 mega 2-cm H20 between each leaching step (15 total) until thesupernatant was clear (following the technique of Mahoney, 1987). Samples were thendried on a hotplate for 24 hours and weighed again. Sample solutions were then preparedby dissolving -400-250 mg of the leached powder dissolved in 1 mL l4N HNO3and 10mL 48% HF on a hotplate for 48 hours at 130°C with periodic ultrasonication. Sampleswere dried and redissolved in 6 mL 6N HC1 on a hotplate for 24 hours and then dried. Pbwas separated using anion exhange columns and the discard was used for Sr, REE, andHf separation. Nd was separated from the other REE and Hf required two additonalpurification steps. Detailed procedures for column chemistry for separating Sr, Nd, andPb at the PCIGR are described in Weis et al. (2006) and Hf purification is described inWeis et a!. (2007). Sr and Nd isotope ratios were measured on a Thermo Finnigan TritonThermal Ionization Mass Spectrometer (TIMS) in static mode with relay matrix rotationon a single Ta and double Re-Ta filament, respectively. Four to 5 filaments per barrel of21 were occupied by standards (NBS 987 for Sr and LaJolla for Nd) for each barrelwhere samples were run. Sample Sr and Nd isotopic compositions were corrected formass fractionation using 86Sr/885r= 0.1194 and‘46NdJ1Nd= 0.7219. Each sample wasthen normalized using the barrel average of the reference material relative to the values of‘43Nd/’44Nd= 0.511858 and87Sr/86Sr=0.710248 (Weis et al., 2006). During the course of42the Vancouver Island analyses, the La Jolla Nd standard gave an average value of0.511856 ± 6 (n=7) and NBS987 standard gave an average of 0.710240 ± 8 (n=1 1) (2aerror is reported as times106).‘47SmJ’Nd ratio errors are approximately —1.5%, or—0.006. United States Geological Survey (USGS) reference material BHVO-2 wasprocessed with the samples and yielded Sr and Nd isotopic ratios of 0.703460 ± 7 and0.512978 ± 6, respectively. These are in agreement with the published values of 0.703479± 20 and 0.5 12984 ± 11, respectively (Weis et a!., 2006).Pb and Hf isotopic compositions were analyzed by static multi-collection on a NuPlasma (Nu Instruments) Multiple Collector-Inductively Coupled Plasma-MassSpectrometer (MC-ICP-MS). The detailed analytical procedure for Pb isotopic analyseson the Nu Plasma at the PCIGR is described in Weis et al. (2006). The configuration forPb analyses allows for collection of Pb, Tl, and Hg together. Ti and Hg are used tomonitor instrumental mass discrimination and isobaric overlap, repectively. All samplesolutions were analyzed with approximately the same Pb/Tl ratio (-.4) as the referencematerial NIST SRM 981. To accomplish this, a small aliquot of each sample solutionfrom the Pb columns was analyzed on the Element2 to determine the precise amount ofPb available for analysis on the Nu Plasma. The SRM 981 standard was run after everytwo samples on the Nu Plasma. During the time samples were run, analyses of the SRM981 Pb reference material gave values of206Pb/204Pb=16.9403 ± 22, 207Pb/204Pb=15.4958± 23, and208Pb/204Pb=36.7131 ± 64 (n=6 1; 2a error is reported as times 1 0); these valuesare in excellent agreement with reported TIMS triple-spike values of Galer & Abouchami(1998). Fractionation-corrected Pb isotopic ratios were further corrected by the sample-standard bracketing method or the in-In correction method described by White et a!.(2000) and Blichert-Toft eta!. (2003). Leached powders of USGS reference materialBHVO-2 yielded Pb isotopic ratios of206Pb/204Pb= 18.6454 ± 8, 207Pb/204Pb= 15.4910 ±5, and 208Pb/204Pb= 38.2225 ± 14 and BCR-2 yielded206Pb/204Pb= 18.8046 ± 6,207Pb/204Pb 15.6251 ± 8, and208Pb/204Pb= 38.8349 ± 6 (2a error is reported as times10). These values are in agreement with leached residues of BITVO-2 and BCR-2 fromWeis et a!., (2006).Hf isotopic compositions were analyzed following the procedures detailed inWeis eta!. (2007). The configuration for Hf analyses monitored Lu mass 175 and Yb43mass 172 to allow for interference correction to masses 174 and 176. Hf isotopic ratioswere normalized internally for mass fractionation to a‘79Hf/’77Hfratio of 0.7325 using anexponential correction. Standards were run after every two samples and sample resultswere normalized to the ratio of the in-run daily average and a‘76Hf/’77Hfratio for JMC475 of 0.282160. During the course of analyses, the Hf standard JMC-475 gave anaverage value 0.282 167 ± 9 (n=l30). USGS reference materials BCR-2 and BHVO-2were processed with the samples and yielded Hf isotopic ratios of 0.282867 ± 5 and0.283100± 5, respectively. Published values for BCR-2 and BHVO-2 are 0.28287 1 ± 7and 0.283104 ± 8, respectively (Weis et a!., 2007).WHOLE-ROCK CHEMISTRYMajor- and trace-element compositionsThe most abundant type of lava in the Karmutsen Formation is tholeiitic basaltwith a restricted range of major- and trace-element compositions. The tholeiitic basaltshave similar compositions to the coarse-grained mafic rocks, and both groups are distinctfrom the picrites and high-MgO basalts (Fig. 2.6). The tholeiitic basalts have lower MgO(5.7-7.7 wt % MgO) and higher Ti02 (1.4-2.2 wt % Ti02)than the picrites (13.0-19.8 wt% MgO, 0.5-0.7 wt % Ti02)and high-MgO basalts (9.1-11.6 wt % MgO, 0.5-0.8 wt %Ti02)(Fig. 2.6; Table 2.2). Almost all data plot within the tholeiitic field in a total alkalisversus silica plot, although there has been substantial K-loss in most samples from theKarmutsen Formation (Fig. 2.7), and the tholeiitic basalts generally have higher Si02,Na20+K0,CaO, and FeOT (total iron expressed as FeO) than the picrites. The tholeiiticbasalts also have noticeably lower LOl (mean 1.72±1.3 wt %) than the picrites (mean5.39 ± 0.8 wt %) and high-MgO basalts (mean 3.84 ± 1.5 wt %; Fig. 2.6) reflecting thepresence of abundant altered olivine phenocrysts in the latter groups. Ni concentrationsare significantly higher for the picrites (339-755 ppm) and high-MgO basalts (122-551ppm) than the tholeiitic basalts (58-125 ppm, except for one anomalous sample) andcoarse-grained rocks (70-213 ppm) (Fig. 2.6; Table 2.2). An anomalous tholeiiticpillowed flow (3 samples; outlier in Tables 2.1 and 2.2), from near the picrite typelocality at Keogh Lake, and a mineralized sill (disseminated sulfide), from the basal440006(C):00I I I I I 1455 10 15 20MgO (wt%)0x .‘cc••. +.()....I....I....IFigure 2.6 Whole-rock major-element, Ni, and LOT variation diagrams for the Karmutsen Formation. Newsamples from this study (Table 2.2) are shown by symbols with black outlines (see legend). Previously-published analyses are shown in small gray symbols without black outlines. The boundary of the alkalineand tholeiitic fields is that of MacDonald and Katsura (1964). Total iron expressed as FeO, LOT is loss-on-ignition, and oxides are plotted on an anhydrous, normalized basis. References for the 322 compiledanalyses for the Karmutsen Formation are G. Nixon (unpublished data), Barker et a!. (1989), Kuniyoshi(1972), Lassiter eta!. (1995), Massey (1995a; 1995b), Muller eta!. (1974); Surdam (1967), and Yorath et a!.(1999). Note that the compiled data set has not been filtered; many of the samples with high Si02 (>52 wt%) are probably not Karmutsen flood basalts, but younger Bonanza basalts and andesites. Three pillowedflow samples (x) and a mineralized sill (+) are distinguished separately because of their anomalouschemistry. Coarse-grained samples are indicated separately and generally have subophitic texture. Threetholeiitic basalt samples (4724A5, 4718A5, 4718A6) with >16 wt % A1203 contain 10-25 modal % of largeplagioclase phenocrysts (7-8 mm) or glomerocrysts(>4mm).454.03.5 -1102+3.02.52.01.51.0PicnteO High-MgO basalt0 Coarse-grained9 Tholelitic basalt- ,.• X Pillowed flow+ Mineralized sill.,* Karmutsen Form..:.00• (compiled data)+.111.111.1.. I....40 45II50Si02(wt%)II55Ni (ppm)—(b)IIliii600 -0.50.060151413121110400 -11111111111 I I IIIIIIIIIII- 0-CaO6*-(wt%) +•. •.—0A-•.1:LO (wtoo00Ocr.020001615141312111098765 10 15 20MgO(wt%)10MgO (wt%)3— I I •i—I—I II I I2FeO(T)• *(wt%):fE(f) -11111520 255 10 15MgO (wtLAA1203(w••l.. e25 0 ••*6*.. €0AA0O*OAAA.•*I — .l. — — — I —20161514131211-I.... I.... I0 5 10 15 20 250MgO (wt%)(g)5 10 15MgO (wt%)Table 2.2 Maior element (wt% oxide) and trace element (ppm) abundances in whole rock samples of Karmutsen basalts, Vancouver Island, B.C.Sample 4718A1 4718A2(1) 4718A2(2) 4718A5 4718A6 4718A7 4719A2 4719A3 4720A2 4720A3Group THOL THOL THOL THOL TI-IOL THOL THOL THOL THOL ThOLArea MA MA MA MA MA MA MA MA SL SLFlow Pillow Pillow Pillow Flow Brenda Pillow Pillow Pillow Brenda FlowUTM EW 5455062 5455150 5455150 5455459 5455518 5455280 5454625 5454625 5567712 5567305UTMNS 384113 384260 384260 383424 383237 382261 381761 381761 708686 707890Unnormalized Major Element Oxides (Wefght %):Si02 47.93 48.65 49.07 48.41 48.69 48.44 49.68 49.08 49.24 48.47Ti02 1.646 1.857 1.835 1.402 1.359 1.487 1.705 1.676 1.577 1.71620314.41 14.30 14.15 15.89 15.92 14.74 14.93 14.06 13.54 13.52Fe203*12.29 12.07 12.48 9.90 10.74 12.13 9.75 12.21 11.30 11.61MnO 0.171 0.183 0.181 0.155 0.157 0.167 0.185 0.178 0.212 0.185MgO 6.25 6.21 6.16 6.81 6.74 6.89 6.98 6.55 6.09 7.12CaO 10.22 11.85 11.70 11.31 10.76 11.56 10.96 10.94 14.00 11.94Na20 3.22 2.28 2.25 2.51 2.89 2.39 2.23 2.24 2.12 2.181(20 0.24 0.18 0.14 0.35 0.340.24 0.27 0.31 0.10 0.29P205 0.15 0.14 0.15 0.10 0.12 0.12 0.08 0.13 0.12 0.12LOl 2.72 1.38 1.22 2.17 2.28 1.64 1.97 1.63 1.15 1.95Total 99.24 99.09 99.33 99.01 99.98 99.81 98.75 99.00 99.45 99.11Trace Elements (ppm):La 7.88 9.03 8.72 6.50 6.58 6.10 7.17 7.02 720 7.88Ce 19.0 22.2 21.6 15.5 16.2 15.4 18.0 17.7 17.3 19.0Pr 2.71 3.10 3.02 2.15 2.20 2.23 2.62 2.52 2.42 2.72Nd 13.0 15.2 14.5 10.7 11.2 11.4 13.2 12.5 12.2 13.5Sm 3.57 4.11 4.05 3.03 3.13 3.42 3.67 3.66 3.43 3.85Eu 1.40 1.57 1.48 1.18 1.20 1.25 1.47 1.39 1.33 1.47Gd 4.39 4.77 4.87 3.59 3.64 3.90 4.32 4.24 4.12 4.48Tb 0.76 0.82 0.82 0.63 0.64 0.68 0.76 0.74 0.71 0.78Dy 4.55 5.02 4.79 3.80 3.87 4.12 4.68 4.41 4.284.78I-to 0.90 0.99 0.96 0.77 0.78 0.84 0.94 0.89 0.87 0.95Er 2.57 2.79 2.72 2.20 2.27 2.34 2.66 2.45 2.48 2.69Tm 0.36 0.40 0.39 0.32 0.32 0.33 0.37 0.35 0.34 0.38Yb 2.15 2.39 2.32 1.90 1.96 2.00 2.23 2.11 2.05 2.33Lu 0.31 0.33 0.32 0.27 0.27 0.28 0.30 0.30 0.29 0.33Sc 40.7 41.3 40.4 38.9 38.3 40.2 42.8 42.8 41.146.2V 328 351 353 285 279 317 341 341 325344Cr 93 146 125 290 297 252 130 131 151 168Co 48.7 49.9 48.9 46.1 45.4 51.4 53.1 53.4 50.556.2Ni 87 90 88 125 122 115 94 103 83 93Cu 177 201 198 167 158 159 180 168 167184Zn 83 91 91 77 73 82 8687 79 88Ga 19 19 20 17 19 18 16 18 19 17Ge 1.6 1.5 1.4 1.1 1.5 1.5 0.7 1.5 1.9 1.1Rb 4 2 2 6 6 4 5 5 6Sr 418 231 228 282 332 263 281 286 192 214Y 25 28 28 22 22 24 2626 25 27Zr 86 97 94 72 71 77 86 83 83 90Nb 8.4 10.1 10.0 7.0 6.8 6.5 7.4 7.3 7.9 8.6Cs 0.7 0.3 0.2 0.5 0.5 0.2 0.2 0.2 02Ba 173 50 49 71 98 62 143 133 28 81Hf 2.6 2.8 2.8 2.0 2.2 2.1 2.6 2.5 2.4 2.6Ta 0.59 0.68 0.65 0.45 0.46 0.41 0.53 0.49 0.51 0.60Pb 4 7 7 10 4 6 7 11 7 9Th 0.65 0.74 0.72 0.52 0.53 0.54 0.61 0.61 0.57 0.64U 0.20 0.23 0.24 0.17 0.16 0.17 0.22 0.190.18 0.20Abbreviations for group are: THOL, tholeiilic baaalt; PlC, picrite; HI-MG, high MgO basalt; CG, coarse-grained (sill or gabbro); MIN SIL,mineralized sill; OUTLIER, anomalous pillowed flow in plots. Abbreviations for area are: MA, Mount Arrowsmith; SL, Schoen Lake; KR, KarmutsenRange; 01, Quadra Island. Sample locations are given using the Universal Transverse Mercator (UTM) coordinate system (NAD83; zones 9 and10). Analyses were perfoiTned at Activation Laboratoiy (AcilLabs). Fe 203 is total ron expressed as Fe203.LOl is loss-on-ignition. AU majorelements, &, V, andY were measured by CF quadrupole OES on solutions of fused samples; Cu, Ni, Pb, and Zn were measured bytotal dilutionICP; Cs, Ga, Ge, Hf, Nb, Rb, Ta, Th, U, Zr, and REE were measured by magnetic-sector ICP onsolutions of fused samples; Co Cr, and Sc weremeasured by INAA. Blanks are below detection lwnit. Ni concentrations for these high-MgO samples by XRF. See Appendices B and C forcomplete XRF data and PCIGR trace element data, respectively. Major elements for sample 93G1 7 are normalized, anhydrous. Samples fromQuadra Island are not shown in figures.46Sample 4720A4 4720A5 4720A5 4720A6 4720A7(1) 4720A7(2) 4720A8 4720A9 4720A10 4721A1Group ThOL THOL THOL CG CG CG ThOL TI-IOL CG ThOLArea SL SL SL SL SL SL SL SL SL SLFlow Flow Flow Breccia Flow Flow Flow Flow Flow SIU PUlowUTM EW 5566984 5563304 5563304 5566161 5566422 5566422 5566800 5564002 5560585 5563843UTM NS 707626 705978 705978 704411 703056 703056 700781 703739 702230 704932Unnrxmalized Major Element Oxides (Weight %):5102 49.16 50.25 48.88 48.90 49.60 49.53 48.02 48.0647.38 48.17riO2 1.776 1.736 1.781 1.799 1.811 1.809 1.768 1.767 0.829 1.890A1203 14.00 13.43 13.84 14.21 14.34 14.36 14.34 14.11 14.42 14.96Fe203*13.30 13.03 13.46 11.78 13.37 13.39 13.18 13.73 10.58 11.40MnO 0.194 0.175 0.179 0.195 0.199 0.199 0.195 0.185 0.161 0.180MgO 6.39 5.93 6.12 6.40 5.93 5.94 6.73 6.35 7.69 6.39CaO 11.85 11.05 11.43 12.08 11.69 11.68 11.37 11.61 11.66 11.75Na20 1.86 2.12 2.16 1.82 1.97 1.97 1.97 2.08 2.77 1.931<20 0.14 0.12 0.140.31 0.15 0.13 0.11 0.11 0.12 0.10P205 0.15 0.15 0.16 0.12 0.16 0.15 0.15 0.16 0.06 0.13LOl 0.93 1.44 1.42 1.38 0.75 0.75 1.91 1.37 4.07 2.11Total 99.75 99.42 99.56 99.00 99.97 99.90 99.73 99.52 99.75 99.01Trace Elements (ppm):La 8.44 7.81 8.07 7.75 7.84 7.74 6.86 7.06 2.24 7.31Ce 20.5 18.8 19.5 19.0 19.0 19.3 17.5 17.3 5.6 18.9Pr 2.88 2.63 2.73 2.71 2.67 2.74 2.54 2.46 0.85 2.78Nd 14.3 13.1 13.5 13.9 13.3 13.7 12.6 12.34.8 13.5Sm 4.07 3.68 3.87 4.03 3.86 3.89 3.72 3.51 1.76 4.06Eu 1.49 1.43 1.45 1.43 1.42 1.43 1.37 1.36 0.75 1.47Gd 4.70 4.60 4.74 4.62 4.73 4.66 4.404.19 2.58 4.72Th 0.81 0.79 0.83 0.81 0.81 0.82 0.77 0.75 0.50 0.81Dy 4.90 4.77 4.92 5.01 4.95 4.97 4.86 4.62 3.384.99Ho 0.99 0.96 0.98 1.01 1.01 1.01 0.98 0.93 0.76 1.03Er 2.81 2.80 2.79 2.89 2.92 2.89 2.78 2.65 2.312.96Tm 0.39 0.40 0.40 0.41 0.42 0.40 0.40 0.37 0.34 0.41Yb 2.31 2.47 2.46 2.51 2.49 2.44 2.51 2.27 2.16 2.50Lu 0.33 0.35 0.35 0.36 0.34 0.35 0.35 0.330.31 0.36Sc 42.9 41.0 44.5 43.4 44.2 42.0 44.0 49.944.4V 349 354 362 363 366 367 362 362 288378Cr 122 133 156 152 154 155 166 311 158Co 53.7 50.8 54.8 52.5 52.6 54.5 55.5 49.554.8Ni 92 99 105 107 97 85 116 112 121106Cu 202 174 185 188 212 210 218187 134 208Zn 92 90 93 96 93 94 109 90 6486Ga 20 19 19 18 19 19 19 18 14 18Ge 1.6 1.6 1.6 1.2 1.6 1.7 1.81.4 1.4 1.3Rb 2 1 1 10 2 2 1 2 2Sr 179 183 188 155 173 174 174 196130 174Y 27 28 28 27 29 28 28 28 2130Zr 97 93 92 94 94 93 95 88 38 97Nb 9.1 8.6 8.2 8.7 8.6 8.7 8.37.8 1.7 8.5Cs 0.2 0.1 0.1 0.9 0.3 0.3 0.2 0.1 0.3 0.6Ba 34 37 38 39 45 43 33 35 6836Hf 2.7 2.7 2.6 2.7 2.7 2.8 2.7 2.6 1.2 2.9Ta 0.60 0.57 0.56 0.57 0.56 0.55 0.55 0.53 0.09 0.58Pb 7 8 8 7 14 6 88 14Th 0.68 0.63 0.61 0.64 0.61 0.61 0.62 0.61 0.240.68U 0.20 0.21 0.19 0.21 0.19 0.190.19 0.18 0.11 02247Sample 4721A2 4721A3 4721A4 4721A54722A2(1)* 4722A2(2)*472(3)*4722(4)* 4722A4(1)* 4722A4(2)*Group THOL THOL THOL THOL THOL THOL THOL THOL PlC PlCArea SL SL SL SL KR KR KR KR KR KRFlow Flow Flow Flow Flow Flow Flow Flow Flow Pillow PillowUTM EW 5563936 5564229 5564285 5564343 5590769 5590769 5590769 5590769 5595528 5595528IJTM NS 704941 704928 704896 705008 634318 634318 634318 634318 629490 629490Unnormalized Major Element Oxides (Weight %):Si02 49.56 48.17 49.49 49.37 48.33 48.67 47.23 48.16 43.85 43.841102 1.770 1.768 1.791 1.806 1.799 1.816 1.783 1.782 0.425 0.425Al203 13.92 14.16 13.98 13.98 13.59 13.74 13.77 13.62 11.56 11.74Fe203*13.07 13.24 13.13 13.53 13.75 13.05 14.13 13.76 10.11 9.65MnO 0.174 0.183 0.176 0.201 0.185 0.187 0.187 0.187 0.161 0.158MgO 6.06 6.61 6.46 5.98 6.97 7.02 7.09 7.09 17.74 17.51GaO 12.15 11.30 11.76 11.59 11.42 11.42 11.43 11.48 9.43 9.36Na20 2.01 2.03 1.92 1.92 2.03 2.05 2.02 2.08 0.53 0.53K20 0.09 0.15 0.12 0.15 0.14 0.13 0.06 0.14 0.10 0.13P205 0.16 0.14 0.14 0.15 0.15 0.14 0.14 0.12 0.04 0.00LOl 0.76 1.78 0.87 1.04 1.48 1.34 1.38 1.37 5.45 5.33Total 99.73 99.53 99.85 99.71 99.84 99.56 99.22 99.79 99.39 98.66Trace Elements (ppm):La 7.86 7.42 7.21 7.62 6.95 7.34 7.65 6.95 1.06 0.96Ce 19.2 18.3 17.9 18.8 17.8 18.5 19.3 17.8 2.6 2.5Pr 2.70 2.66 2.57 2.73 2.53 2.64 2.94 2.67 0.41 0.38Nd 13.5 13.3 13.0 13.9 12.9 12.7 13.6 13.3 2.6 2.3Sm 3.84 3.84 3.81 3.91 3.72 3.66 4.05 3.77 0.92 0.84Eu 1.46 1.40 1.38 1.47 1.41 1.42 1.55 1.42 0.38 0.37Gd 4.47 4.59 4.28 4.73 4.42 4.62 4.76 4.59 1.40 1.42Tb 0.79 0.79 0.77 0.83 0.77 0.77 0.83 0.80 0.30 0.30Dy 4.75 4.88 4.77 5.12 4.61 4.60 4.92 4.77 2.16 2.17Ho 0.96 0.98 0.97 1.04 0.94 0.94 0.92 0.95 0.49 0.50Er 2.72 2.82 2.80 2.99 2.65 2.70 2.64 2.69 1.53 1.54Tm 0.38 0.39 0.40 0.42 0.38 0.38 0.389 0.37 0.24 0.23Yb 2.35 2.42 2.40 2.55 2.32 2.27 2.41 2.28 1.58 1.52Lu 0.32 0.35 0.34 0.35 0.32 0.34 0.349 0.34 0.23 0.23Sc 42.1 42.3 42.2 39.7 39.2 29.2 41.6 41.5 38.3 40.1V 351 364 364 367 365 365 362 362 201 189Cr 118 164 151 122 143 98 162 165 1710 1830Co 51.6 54.9 53.3 46.5 50.0 37.7 55.7 53.5 80.3 84.6Ni 91 115 108 90 107 105 93 93 755 755Cu 195 206 211 210 189 189 178 174 92 83Zn 90 87 92 92 104 103 98 94 77 55Ga 19 19 19 19 19 19 20 17 10 9Ge 1.7 1.3 1.7 1.6 1.4 1.5 1.5 1.2 1.0 0.7Rb I 1 1 1 2 2 2 2 5 4Sr 187 162 160 171 190 193 178 190 100 97Y 26 28 29 29 27 28 26 27 16 13Zr 97 95 95 96 96 87 94 102 16 19Nb 9.0 8.5 8.6 8.5 8.3 8.4 8.5 8.5 0.7 0.9Cs 0.2 0.4 0.3 0.2 0.2 0.2 0.4 0.2 2.8 2.7Ba 39 34 28 34 34 34 28 33 19 18Hf 2.8 2.8 2.7 2.9 2.8 2.6 2.8 2.8 0.5 0.6Ta 0.60 0.57 0.55 0.59 0.55 0.54 0.6 0.57 0.09 0.03Pb 8 9 9 5 11 12 91 7 4Th 0.69 0.62 0.62 0.67 0.66 0.64 0.59 0.68 0.10 0.11U 0.21 0.20 0.19 0.20 0.19 0.20 0.23 0.15 0.05 0.0348Sample4722A4(3)* 4722A5(1)*4722A5(2r4723A2* 4723A3*--4723A104723A13(1)*4723A13(2)*4724A3Group PlC OUTLIER OUTLIER HI-MG PlC PlC THOL PlC PlC CGArea KR KR KR KR KR KR KR KR KR SLFlow Pillow Flow Flow Pillow Pillow Pillow Flow Pillow Pillow FlowUTMEW 5595528 5595029 5595029 5588266 5588274 5586081 5578863 5599233 5599233 5581870UTMNS 629490 627605 627605 626698 626641 626835 630940 616507 616507 704472Unnormalized Major Element Oxides (Weight %):Si02 42.94 48.95 48.45 46.73 44.41 44.39 50.03 44.62 44.71 49.081102 0.42 2.295 2.333 0.611 0.539 0.663 2.083 0.443 0.442 1.457A1203 11.26 13.61 12.51 15.24 12.75 14.93 12.31 13.71 13.75 14.77Fe203 10.82 12.56 15.24 10.26 10.33 10.11 15.21 10.38 10.37 11.60MnO 0.161 0.188 0.191 0.158 0.148 0.139 0.223 0.138 0.138 0.128MgO 18.28 6.18 5.99 10.27 15.42 13.02 5.66 14.47 14.48 8.73GaO 8.98 9.33 8.99 9.93 8.73 9.73 9.19 9.37 9.36 11.02Na20 0.54 3.26 3.30 2.26 0.78 1.56 2.94 0.86 0.86 1.99K20 0.02 0.30 0.30 0.38 0.07 0.07 0.70 0.06 0.07 0.150.03 0.13 0.19 0.06 0.05 0.06 0.18 0.05 0.05 0.12LOl 5.7 2.00 1.95 3.70 5.61 4.91 1.55 5.59 5.59 0.98Total 99.16 98.80 99.45 99.60 98.83 99.58 100.07 99.68 99.81 100.02Trace Elements (ppm):La 1.08 8.77 9.46 1.94 1.80 1.78 8.93 1.40 1.36 5.57Ce 2.6 23.0 25.2 5.0 4.5 4.5 21.5 3.4 3.4 14.3Pr 0.43 3.37 3.68 0.74 0.67 0.71 3.10 0.52 0.51 2.09Nd 2.5 17.2 17.9 4.2 3.8 4.1 15.5 3.0 2.9 10.4Sm 0.87 5.20 5.21 1.41 1.27 1.48 4.20 1.00 1.01 2.98Eu 0.405 1.74 1.79 0.52 0.51 0.63 1.62 0.46 0.44 1.21Gd 1.45 6.39 6.63 2.09 1.86 2.16 5.21 1.71 1.69 3.85Th 0.3 1.11 1.20 0.42 0.37 0.42 0.90 0.36 0.35 0.65Dy 2.11 6.59 7.08 2.91 2.60 2.79 5.40 2.54 2.50 4.01Ho 0.47 1.34 1.39 0.67 0.59 0.60 1.11 0.59 0.56 0.81Er 1.54 3.85 3.94 2.10 1.86 1.79 3.23 1.89 1.79 2.27Tm 0.242 0.55 0.57 0.32 0.29 0.26 0.45 0.29 0.28 0.33Yb 1.62 3.31 3.48 2.06 1.85 1.67 2.71 1.89 1.84 1.98Lu 0.244 0.45 0.51 0.31 0.28 0.25 0.39 0.29 0.28 0.28Sc 36.4 38.7 33.1 47.2 41.0 38.1 38.3 43.5 35.1V 194 481 495 261 218 235 496 219 222 287Cr 1750 79.7 59.0 358 1570 725 34.2 1370 352Co 80 45.6 40.7 48.8 72.9 60.4 47.1 72.2 52.4Ni 755 59 59 163 656 339 61 583 583 209Cu 83 116 114 111 110 106 306 142 98 208Zn 55 106 103 61 60 63 107 106 62 83Ga 11 17 20 13 12 13 20 12 12 19Ge 1.2 0.7 1.3 1.2 1.1 1.1 1.6 1.2 1.3 1.6Rb 6 6 6 10 2 2 12 2 1 2Sr 93 225 229 271 64 132 299 73 73 167Y 15 39 39 20 17 18 32 16 17 23Zr 16 127 126 33 29 36 110 24 22 73Nb 0.7 10.0 10.6 1.5 1.3 1.1 9.7 0.8 0.7 6.5Cs 5.8 0.3 0.4 6.5 0.9 0.8 0.4 0.7 0.7 0.3Ba 27 87 88 84 15 20 152 13 13 55Hf 0.6 3.7 3.7 1.0 0.9 1.0 3.1 0.7 0.7 2.1Ta 0.67 0.70 0.08 0.06 0.05 0.65 0.04 0.04 0.42Pb 22 6 7 4 5 8 7Th 0.09 1.08 1.08 0.23 0.20 0.10 0.77 0.15 0.14 0.43[I 007 0R oq 010 009 005 027 005 007 01549Sample 4724A5 5614AV 5614A3’ 5614A5 5I4A1Q 5614A11 5614A13 5614A14 5614A15 5615A1Group THOL PlC HI-MG HI-MG THOL THOL THOL CG CG THOLArea SL KR KR KR KR KR KR KR KR KRFlow Flow Pillow Pillow Breccia Flow Flow Flow Sill Flow PillowUTMEW 5580653 5599183 5599183 5599192 5595261 5593546 5588018 5588246 5589935 5599424UTMNS 704736 616472 616472 614756 615253 615098 618867 618183 615917 620187Unnormalized Major Element Oxides (Weight %):Sb2 48.08 46.24 48.43 47.1 46.69 48.06 47.78 46.17 46.2 47.611102 1.729 0.471 0.477 0.512 1.761 1.741 1.86 1.773 1.605 1.8732O3 16.39 14.57 14.9 14,52 13.32 13.35 12.91 13.8 14.77 13.55Fe203*11.13 8.57 7.59 9.12 12.72 12.96 14.23 13.53 10.89 14.34MnO 0.196 0.145 0.134 0.135 0.168 0.187 0.211 0.136 0.123 0.188MgO 5.54 12.11 10.66 10,92 7.09 6.38 6.47 8.25 9.37 7.42CaO 12.38 10.81 10.68 10.78 10.62 10.51 10.24 1041 10.22 9.86Na20 2.01 1.3 1.51 2.1 3.49 3.23 2.91 2.01 1.55 2.87K20 0.10 0.01 0.06 0.01 0.01 0.44 0.44 0.21 0.08 0.21p205 0.15 0.04 0.04 0.05 0.14 0.13 0.15 0.14 0.13 0.15LOI 1.29 4.91 4.13 4.67 3.53 2.39 2.02 2.87 4.07 1.92Total 99.01 99.14 98.6 99.89 99.43 99.37 99.21 99.3 99.03 100Trace Elements (ppm):La 7.98 1.79 1.71 0.91 8.04 7.23 8.51 7.31 6.88 8.17Ce 19.4 4.3 4.0 2.5 19.7 18.0 20.9 18.7 17.6 20.2Pr 2.72 0.65 0.59 0.43 2.96 2.68 3.08 2.81 2.68 3,01Nd 13.3 3.4 3.2 2.7 13.7 12.3 14.7 13.2 12.5 14.3Sm 3.69 1.11 1.09 1,02 3.88 3.6 4.2 3.89 3.43 4,12Eu 1.42 0.501 0.483 0.526 1.44 1,39 1.59 1.47 1.34 1.58Gd 4.58 1.74 1.71 1.6 4.47 4.24 4.79 4.61 4.03 4.88Th 0.75 0.37 0.37 0.34 0.79 0.77 0.87 0.81 0.7 0.85Dy 4.53 2.67 2.66 2.35 4.59 4.57 5.12 4.76 4.11 5.01Ho 0.89 0.6 0.59 0.52 0.85 0.89 0.98 0.9 0.76 0.94Er 2.54 1.87 1.87 1,67 2.47 2.53 2.83 2.55 2.18 2.7Tm 0.36 0.297 0.292 0.27 0.37 0.36 0.413 0.375 0.324 0.404Yb 2.26 2.01 1,99 1.8 2.27 2.26 2.54 2.35 2 2.51Lu 0.32 0.308 0.312 0.267 0.325 0.329 0.375 0.336 0.29 0.365Sc 34.4 45 48.8 43.7 43.8 43.8 424 39.3 32.3 44.5V 299 222 217 206 354 362 372 332 271 380Cr 207 1370 1420 797 195 76 71.9 328 400 172Co 43.1 68.7 76.1 68.6 48.9 53.6 51.5 58.9 56.8 57.4NI 98 564 551 315 93 67 65 147 213 86Cu 198 104 111 96 134 181 165 83 111 182Zn 76 53 51 48 80 93 102 86 72 84Ga 21 11 12 13 20 17 20 20 19 19Ge 1.5 0.9 0.8 0.9 1.4 0.9 1.6 1.3 1.1 1.3Rb 9 8 3 5Sr 272 120 130 141 86 381 274 283 183 287Y 25 18 17 16 26 27 29 28 23 29Zr 92 22 24 19 90 86 98 91 86 98Nb 8.6 0.7 0.8 0.5 8.1 8.0 9.0 8.2 7.7 8.6Cs 0.1 0.4 0.4 0.8 0.4 0.1 0.1 0.2 0.9Ba 32 18 20 9 11 80 85 62 32 39Hf 2.7 0.7 0.7 0.6 2.7 2.5 2.9 2.8 2.6 3.0Ta 0.60 0.6 0.5 0.7 0.5 0.5 0.6Pb 26 28 29 97 97 98 94 82 97Th 0.69 0.15 0.17 0.08 0.75 0.62 0.66 0.54 0.52 0.64ii fl9 nil nii nn n n4 n’i n92 (12750Sample 5615A5 5615A6 5615A7(1)5615A7(2)*5815A8 5615A10 5615A11 5615A15616A1*5616A2Group CG CG PlC PlC ThOL ThOL OUTLIER PIG PlC ThOLArea KR KR KR KR KR KR KR KR KR KRFlow Flow Sill Pillow Pillow Pillow Pillow Pillow Pillow Pillow FlowUTM EW 5601095 5601095 5595569 5595569 5595513 5595376 5595029 5586126 5598448 5585731UTM NS 624103 624103 629573 629573 629434 629069 627605 626824 616507 623077Unnomialized Major Element Oxides (Weight %):Si02 47.76 46.53 47.16 45.73 48.31 47.29 49.08 45.35 43.7 46.72riO2 2.125 2.119 0.466 0.442 1.745 1.807 2.304 0.643 0.439 1.939AJ2O 14.41 14.06 11.84 11.48 13.45 13.7 11.86 14.07 13.66 13.21Fe203*12.5 13.01 11.54 11.22 13.48 14.23 15.2 10.4 10.47 14.11MnO 0.191 0.177 0.172 0.166 0.196 0.195 0.215 0.142 0.158 0.235MgO 5.71 5.7 18.59 18.19 6.78 6.77 5.73 12.56 15.55 6.29CaO 8.81 10 9.45 9.14 11.77 10.65 9.84 9 8.71 11.19Na20 4.16 3.64 0.42 0.41 1.8 2.58 3.25 2.11 0.98 2.49K2O 0.2 0.09 0.01 0.16 0.03 0.21 0.04 0.02 0.04 0.01P205 0.18 0.19 0.03 0.05 0.14 0.14 0.22 0.06 0.04 0.16LOI 3.17 345 2.86 1.74 1.57 2.07 4.68 6.05 3.38Total 99.2 98.96 99.67 99.84 99.43 99.14 99.81 99.02 99.8 99.66Trace Elements (ppm):La 11.6 11.7 1.09 1.02 7.32 7.34 6.41 1.73 1.52 8.87Ce 28.4 26.9 2.7 2.6 18.4 18.5 17.3 4.7 3.6 21.0Pr 4.01 3.95 0.44 0.41 2.69 2.84 2.89 0.77 0.55 3.02Nd 17.7 17.7 2.5 2.4 12.9 13.2 15.64.4 3.0 14.4Sm 4.74 4.76 0.88 0.85 3.78 3.92 5.061.47 1.02 4.16Eu 1.78 1.73 0.381 0.361 1.47 1.48 1.88 0.607 0.454 1.59Gd 5.49 5.32 1.42 1.35 4.4 4.65 6.192.13 1.57 4.81Th 0.93 0.93 0.3 0.3 0.79 0.81 1.13 0.420.34 0.86Dy 5.49 5.31 2.16 2.12 4.65 4.79 6.65 2.78 2.44 5.07Ho 1.02 0.98 0.49 0.46 0.89 0.91 124 0.59 0.550.92Er 2.91 2.86 1.55 1.46 2.58 2.58 3.67 1,78 1.732.72Tm 0.441 0.43 0.253 0.236 0.379 0.38 0.5470.267 0268 0.413Yb 2.75 2.73 1.71 1.58 2.34 2.42 3.44 1.75 1.77 2.62Lu 0.392 0.391 0247 0.245 0.339 0.348 0.493 0.2690.277 0.378Sc 35.7 38.5 39.2 39.5 40.4 39.8 43.8 38.3 37.541.6V 381 376 224 222 353 363 520 215 204401Cr 127 164 1910 1850 173 166 107906 3000 76Co 39.4 43.7 87.1 89.3 53 52,4 53.4 67.267.9 50.8Ni 73 85 729 680 96 94 56 368 559 58Cu 105 148 86 80 175 185 208 83 77210Zn 86 102 57 54 88 86 94 50 53 85Ga 21 21 10 10 20 20 18 1412 22Ge 1.4 1.4 1.1 0.9 1.4 1.3 0.71.1 1.1 1.6Rb 2 6 6 2 9 2Sr 128 170 124 120 194 279 143189 112 67Y 32 31 15 15 28 28 3916 16 30Zr 116 121 16 15 91 96 124 35 22100Nb 12.7 13.0 0.9 0.7 8.0 8.5 9.91.4 0.9 9.2Cs 0.2 4.4 4.5 0.6 1.3 0.1 0.6 0.8Ba 36 34 27 24 39 61 2015 20 8Hf 3.4 3.5 0.6 0.5 2.7 2.8 3.8 1.1 0.72.9Ta 0.9 0.9 0.6 0.6 0.70.6Pb 110 118 24 22 91 92 11334 22 96Th 0.92 0.93 0.1 0.08 0.59 0.611.01 0.09 0.14 0.63U 0.34 0.33 0.07 0.07 - 0,22 0.24 0.41 - 0.07 0.10.2551Sample 5616A3 5616A7 5617A1 5617A45617A5(1)* 5617A5(2)*5618A1 5618A3 5618A4 93G171Group CG HI-MG CG MINSIL CG CG THOL THOL THOL PlCArea KR KR SL SL SL SL 01 QI QI KRFlow Flow Pillow Sill Sill Sill Sill Flow Pillow Breccia PillowUTM EW 5584647 5589833 5560375 5557712 5557712 5557712 5557892 5552258 5552258 5599395UTMNS 623236 626879 702240 700905 700905 700905 338923 341690 341690 616613Unnormalized Major Element Oxides (Weight %):Si02 46.77 46.75 49.1 48.06 49.31 47.4 48.61 47.9 40.29 47.931102 1.814 0.732 0.901 3.505 1.713 1.71 1.749 1.362 2.159 0.461Al203 14.23 15.92 13.72 12.79 13.39 13.33 12.82 13.79 16.27 14.02Fe203*12.85 10.88 11.34 16.78 12.39 12.71 13.81 11.62 18.1 9.9MnO 0.185 0.153 0.16 0.164 0.174 0.174 0.211 0.158 0.28 0.19MgO 7.28 8.7 6.99 5.25 7.5 7.46 6.96 6.57 9.7 15.8CeO 10.92 11.69 10.96 648 12.01 11.93 10.54 12.08 6.78 10.42Na20 1.86 2.12 3.46 2.31 1.83 1.8 2.76 3.13 0.58 1.02K20 0.16 0.07 0.03 0.69 0.22 0.17 0.5 0.12 0.52 0.11P205 0.13 0.04 0.08 0.35 0.13 0.14 0.14 0.1 0.18 0.08LOl 2.92 2.86 3.23 3.22 1.28 1.38 1.8 3.17 4.77Total 99.14 99.91 99.97 99.6 99.95 98.21 99.9 99.99 99.63 100.00Trace Elements (ppm):La 7.14 1.51 2.84 22.3 8.13 7.94 7.72 6.57 10.7 1.4Ce 17.9 3.8 6.9 51.5 19.7 19.2 19.1 15.4 24.3 3.6Pr 2.72 0.59 1.1 7.19 2.89 2.85 2.85 2.22 3.48 0.56Nd 12.9 3.5 6.1 32.7 13.3 13.3 13.5 10.5 15.9 2.9Sm 3.82 1.36 2.14 8.86 3.83 3.77 3.98 2.98 4.53 1.23Eu 1.47 0.638 0.888 3.33 1.45 1.43 1.47 1.21 2.04 0.47Gd 4.49 2.05 3.05 9.84 4.36 4.32 4.56 3.56 5.64 2.10Tb 0.79 0.43 0.63 1.64 0.75 0.75 0.79 0.63 0.99 0.32Dy 4.66 2.9 4.19 9.58 4.35 4.25 4.71 3.84 5.95 2.38Ho 0.9 0.61 0.86 1.78 0.83 0.81 0.88 0.75 1.15 0.64Er 2.6 1.91 2.65 5.06 2.37 2.31 2.56 2.16 3.36 2.11Tm 0.373 0.305 0.415 0.73 0.341 0.342 0.375 0.331 0.495 0.297Yb 2.35 2.02 2.66 4.56 2.13 2.14 2.35 2.12 3.11 1.97Lu 0.34 0.302 0.401 0.637 0.304 0.298 0.342 0296 0.455 0.282Sc 40 50.7 54.5 44.2 40.8 37.7 42.9 41.8 54.2V 349 270 327 517 342 338 367 313 499 140Cr 317 346 210 64.4 274 255 114 194 159 606Co 52.6 56.7 46.8 38.3 51.8 47.6 53 46 62Ni 110 122 70 39 98 97 79 67 68 491Cu 164 111 160 232 161 160 25 125 109 91Zn 87 51 69 160 77 76 89 92 75 57Ga 20 15 16 27 19 19 19 18 24Ge 1.3 1 1.3 2 1.1 0.7 1.5 1.5 1.3Rb 3 21 7 7 8 3 10Sr 237 181 58 197 255 254 227 170 220Y 25 19 27 52 24 23 27 21 38 13Zr 91 26 45 229 89 86 89 67 112 24Nb 8.1 1.1 2.1 19.7 8.9 8.5 8.1 6.6 9.8 0.9Ca 1.2 0.3 0.2 1.2 0.2 0.2 0.4 0.6Ba 39 23 12 771 38 37 103 17 109 20Hf 2.7 0.8 1.4 6.4 2.6 2.5 2.7 2 3.3 0.8Ta 0.6 0.1 1.4 0.6 0.6 0.6 0.5 0.7 0.1Pb 93 37 48 151 85 83 83 66 65Th 0.52 0.14 0.28 2.09 0.61 0.57 0.57 0.45 0.71 0.15U 0.23 0.09 0.16 0.88 0.23 0.25 0.22 0.22 0.3452(rooiuopspux)qjudniIUo!IApiojo&iu.iiuaijjippu‘(q>jpu;ossojAjjids)‘-jj’-jijuouoi.ijosoojjp‘suoidpuiiioiipQJ1u-33fluoquo3uSipijo•(c661)mispunouoaowoijStLJUAUO!z!1UUUOUliv•SUiUpjTJUTUWjotpwosjdmUsiojstuUd3uwp-ooI3pzquuou-uwA!w!.Idai(j)pu‘(p)‘(q)•pusjJA1LOOUEAuosaropjgWUW,jotpwogsjdwUsio;sudpzquuou-upuotpai()pu‘(a)‘to)uouuousnuujiojsuoirnIouoo1um-jqdmoouipOpu3J3pOJ-OuL1a.In,Irn13AOH1(53qlpD03‘1IHJZWSISPHd3)1ELqN)flqjq5)flWL13OH1(53qIp9553W5WdPN143),••II•••••••••,•,,•,•I•••.••••••••••1(1)()II!SP2Z!TeJ3Ui+MOp3MOfldXieseq!w3Ior1e3)OJiJewp3u1e16-3s13o)0eseqO 6 w-1 6 ’H03!J’IdV0L30LqWSMoJJUflOfiJP!WSMOIJVUflOjAJtill11111111111111,0L‘I1111111—00Lfl1cVJ3AoHXaqIpDn31;H,zwsJspNJda)e1eLqN)1flqLeaqus)flCIJWI13OH1(53(Ii.p9fl3W5WdPN143)3I•p•ipi•••IIIIII II(D)If,33__________________________________liii01(•t001•ee,uq)Sliii111111____________________________________________OOLfl53),13AoHXoqIp9flJ1.JHJZU.JSiSPNd3)l1qN)IflqjqSDfl(CI),Wj13OH1(53(Ijp9fl3WWdPN143)s,•,••,••,1111111liii111110•p•p•i••••••••(oo’3psirSSSIeS)e6eJ3Aeoowpesidea•001usnw111111111I__________________________________________ _________________________________________________________________________________IIIIIIIIIIIII001sediment-sill complex at Schoen Lake, are distinguished by higher Ti02 andFeOT thanthe main group of tholeiitic basalts. The tholeiitic basalts are similar in major-elementcomposition to previously published results for samples from the Karmutsen Formation;however, the Keogh Lake picrites, which encompass the picritic and high-MgO basaltpillow lavas, extend to —20 wt % MgO (Fig. 2.6, and references therein).The tholeiitic basalts from the Karmutsen Range form a tight range of parallellight rare earth element (LREE)-enriched patterns (La/YbcN=2.1-2.4; mean 2.2 ± 0.2),whereas the picrites and high-MgO basalts from the Karmutsen Range form a range ofparallel LREE-depleted patterns (LaJYbcN=0.3-0.7; mean 0.6 ± 0.2) with lower REEabundances than the tholeiitic basalts (Fig. 2.7). Tholeiitic basalts from the Schoen Lakeand Mount Arrowsmith areas have similar REE patterns to samples from the KarmutsenRange (Schoen Lake- LaJYbCN=2.l-2.6; mean 2.3 ± 0.3; Mount ArrowsmithLaJYbCN=1.9-2.5; mean 2.2± 0.4), and the coarse-grained mafic rocks have similar REEpatterns to the tholeiitic basalts (LaJYbCN=2.l-2.9; mean 2.5 ± 0.8). Two LREE-depletedcoarse-grained mafic rocks from the Schoen Lake area have similar REE patterns(LaJYbCN=0.7) to the picrites from the KeoghLake area, indicating that this distinctivesuite of rocks is exposed as far south as Schoen Lake (Fig. 2.7). The anomalous pillowedflow from the Karmutsen Range has lower La/YbCN (1.3-1.8) than the main group oftholeiitic basalts from the Karmutsen Range (Fig. 2.7). The mineralized sill from theSchoen Lake area is distinctly LREE-enriched (La/YbCN=3.3) with the highest REEabundances (La=94.0) of all Karmutsen samples, which may reflect contamination byadjacent sediment.The primitive mantle-normalized trace-element patterns (Fig. 2.7), and trace-element variations and ratios (Fig. 2.8), highlight the differences in trace-elementconcentrations between the four main groups of Karmutsen flood basalts on VancouverIsland. The tholeiitic basalts from all three areas have relatively smooth, parallel traceelement patterns; some samples have small positive Sr anomalies, relative to Nd and Sm,and many samples have prominent negative K anomalies relative to U and Nb, reflectingK-loss during alteration (Fig. 2.7). The large ion lithophile elements (LILE) in thetholeiitic basalts (mainly Rb and K, not Cs) are depleted relative to HFSE and LREE, andLILE segments, especially for the Schoen Lake area (Fig. 2.7d), are remarkably parallel.5415• • . . ,1.6Nb(ppm) Nb/La1.2-8o10 . -0.8- oo ‘5 Picrite•High-MgO basalt04 -0 Coarse-grained• Q0 Tholeiltic basalt(a)X Pillowed flow(b)0I I I I I I I I I I000 50 100 150 0 5 10 15 20 25Zr (ppm) MgO (wt%)1.25....i....i.,..i....i,.,.1.2iiiiiiiTh (ppm) Th (ppm) >X::0.250.2 (d)(C) 561 5A1 2_________________0.00I I • I • I• 0.0I I I I I I I I I0 1 2 3 4 5 0.0 0.1 0.2 0.3 0.4 0.5Hf(ppm) U(ppm)Figure 2.8 Whole-rock trace-element concentrations and ratios for the Karmutsen Formation (except panel b isversus MgO). (a) Nb vs. Zr. (b) Nb/La vs. MgO. (c) Th vs. Hf. (d) Th vs. U. There is a clear distinction betweenthe tholeiitic basalts and picrites in Nb and Zr, both in concentration and the slope of each trend.55The picrites and high-MgO basalts form a tight band of parallel trace-element patternsand are depleted in HFSE and LREE with positive Sr anomalies and relatively low Rb.The coarse-grained mafic rocks from the Karmutsen Range have identical trace-elementpatterns to the tholeiitic basalts. Samples from the Schoen Lake area have similar trace-element patterns to the Karmutsen Range, except for the two LREE-depleted coarsegrained mafic rocks and the mineralized sill (Fig. 2.7d).Sr-Nd-Hf-Pb isotopic compositionsThe nineteen samples of the Karmutsen Formation from the four main groups,selected on the basis of major and trace-element variation, stratigraphic position, andlocation to cover the range of these parameters, have indistinguishable age-corrected Hfand Nd isotope ratios and distinct ranges of Sr isotopic compositions (Fig. 2.9). Thetholeiitic basalts, picrites, high-MgO basalt, and coarse-grained mafic rocks have initial8Hf—+8.7 to +12.6 and5Nd=+6.6 to +8.8 corrected for in situ radioactive decay since 230Ma (Fig. 2.9; Tables 2.3 and 2.4). All the Karmutsen samples form a well-defined lineararray in a Lu-Hf isochron diagram corresponding to an age of 241 ± 15 Ma (Fig. 2.9).This is within error of the accepted age of the Karmutsen Formation and indicates that theHf isotope systematies have behaved as a closed system since ca. 230 Ma. Theanomalous pillowed flow has similar initial Epj (+9.8) and slightly lower initial ENd (+6.2)than the other Karmutsen samples. The picrites and high-MgO basalt have higherinitial87Sr/86Sr(0.70398-0.705 18) than the tholeiitic basalts (initial 87Sr/86Sr=0.70306-0.70381)and the coarse-grained mafic rocks (initial 87Sr/86Sr=0.70265-0.70428) overlap theranges of the picrites and tholeiitic basalts (Fig. 2.9; Table 2.3).The measured Pb isotopic compositions of the tholeiitic basalts are moreradiogenic than those of the picrites and the most magnesian Keogh Lake picrites havethe least radiogenic Pb isotopic compositions. The range of initial Pb isotope ratios forthe picrites is 206Pb/204Pb= 18.142-18.580, 207Pb/204Pb= 15.547-15.562, and 208Pb/204Pb= 37.873-38.257 and the range for the tholeiitic basalts is 206Pb/204Pb= 18.782-19.098,207Pb/204Pb= 15.570-15.584, and 208Pb/204Pb = 37.312-38.587 (Fig. 2.10; Table 2.5). Thecoarse-grained mafic rocks overlap the range of initial Pb isotopic compositions for thetholeiitic basalts with 206Pb/204Pb= 18.652-19.155, 207Pb/204Pb= 15.568-15.588, and560.5132 - - 0.28340.28330.51310.28320.51300.28310.5129 -0.28300.5128 ‘ 0.28290.15 0.17 0.19 0.21 0.23 0.25 0.00 0.02 0.04 0.06 0.08 0.101475m/144Nd 176Lu/177Hf10‘ • i •14(230 Ma)Nd1280 -0 6106X48(c)i0.702 0.703 0.704 0.705 0.7064• 60.702 0.703 0.704 0.705 0.706 5 6 7 89 1087Sr/ 86Sr(230 Ma)NdMa)Figure 2.9 Whole-rock Sr, Nd, and Hf isotopic compositions for the Karmutsen Formation. (a)‘43NW1Ndvs.‘47Sm/’Nd. (b) 176Hf/177Hfvs. 176LuJ’77Hf. The slope of the best-fit line for all samples corresponds to an ageof 241 ± 15 Ma. (c) Initial ENd VS. 87Sr/86Sr. Age correction to 230 Ma. (d) LOTvs. 87Sr/86Sr. (e) Initial CHf VS.6NdAverage 2a error bars are shown in a corner of each panel. Complete chemical duplicates,shown in Tables2.3 and 2.4 (samples 4720A7 and 4722A4), are circled in each plot.11•1J•143Nd/1NdaA PicriteHigh-MgO basaltX0Coarse-grained• tholeütic basaltXPillowed flow(a)i L_III I176Hf/177Hf- Age=241 ± l5Ma-(0=+9.90±0—I I I I iLOl (wt 96)tAA 4723A2dD(d) ê“Sri Sr‘ I ‘ I I I ‘E(230Ma)0Hf0•x-(e)I I i I i I I I i5715.6815.6015.6615.58 -15.6415.56 -15.6215.54 -15.6015.52 -15,584723A415.56 153018.6 19.0 194 19.8 202 20.6 21.0 17.8 18.0 18.2 18.4 18.6 18.8 19.0 19.2 19.4206pb/204Pb 206Pb/204Pb(230 Ma)39.6• ‘ i ‘ i • i ‘ i39.01111111111111114723A2 - (230 Ma)39.2 - 38.6 —_____________X638.8- 538.2384237.84723 -(c)c18619.0 19A 19.8 20.2 20.6 21.0(e)38.0‘ ‘ I I I I I I I I I ililililili liii18.6 19.0 194 19.8 20.2 20.6 21.0 17.8 18.0 18.2 18.4 18.6 18.819.0 19.2 19.4206pb/204pb(230 Ma)Figure 2.10 Pb isotopic compositions of leached whole-rock samples measured by MC-TCP-MS for theKarmutsen Formation. Error bars are smaller than symbols. (a) Measured 207Pb/204Pb vs. 206Pb/2°4Pb. (b) Initial207Pb/204Pb vs. 206PbP°4Pb. Age correction to 230 Ma. (c) Measured 208Pb/204Pb vs. 206PbP°Pb. (d) LOT vs.206Pb/204Pb. (e) Initial 208Pb/204Pb vs. 206Pb/204Pb. Complete chemical duplicates, shown in Table 2.5 (samples4720A7 and 4722A4), are circled in each plot. The dashed lines in panels b and e show the differences in age-corrections for two picrites (4723A4, 4722A4) using the measured U, Th, and Pb concentrations for eachsample (black triangles) and the age-corrections when concentrations are used from the two picrites (4723A3,4723A13) that appear to be least affected by alteration.58I I I I I I I I I I I4723A2(a)I I I I I I I I I I I• I • I • I•I•1•I•I•4723A2• 207pb/204pb0(230 Ma)•-— 4723A13—— _‘><•s4723A34722A4‘Picrite<>High-MgO basalt0Coarse-grainedTholeiitic basalt(b)Pillowed flowLOl (wt %)A4723A2000(d) • 20Pb’204Pb00Table2.3SrandNdisotopicgeochemistiyofKarmutsenbasalts,VancouverIsland,B.C.4722A5OUTLIERKR4.461910.70387770.06780.703666.1021.00.51292265.50.17560.512666.24723A2HI-MGKR8.232690.70546670.08850.705181.995.790.51305988.20.20810.512757.9Sample 4718A24718A74719A24719A3 4720A4 4721A2 4721A4GroupsAreabRbSr87SrISr2am87Rb/Sr815r/°6Sr1SmNd143NW1Nd2am6Nd147Sm/1Nd143Nd/1Nd(ppm)(ppm)230Ma(ppm)(ppm)230MaTHOL THOL THOL THOL THOL THOL THOIMA1.28MA3.78MA3.27MA3.49SL2.54SL1.58SL1.652020.70324762830.70370282440.70381192290.70375762150.70306682090.70307671720.70306270.0183 0.0387 0.0387 0.0441 0.0341 0.0220 0.0277CNd(t)0.703194.6016.80.51300460.703584.7916.60.51299860.703684.4816.10.51297260.703614.2214.80.51297260.702955.7721.00.51298470.703004.0514.60.51296570.702975.4819.60.51293367.10.16567.00.17426.50.16886.50.17186.70.16626.40.16755.80.16910.51275 0.51274 0.51272 0.51271 0.51273 0.51271 0.512688.1 7.7 7.3 7.2 7.6 7.2 6.64720A6CG5L6.111230.70312190.14380.702654.7516.80.51297676.60.17110.512727.34720A7CGSL2.181570.70307860.04010.702954.9617.30.51301267.30.17350.512758.04720A7(dup)CGSL1.791400.70306790.03690.702954.5616.00.51299466.90.17220.512737.74720A10CGSL0.11600.70429470.00550.704282.055.70.51312079.40.21910.512798.74724A3CGSL1.441350.70342180.03100.703323.8413.20.51296166.30.17570.512706.95616A3CGKR3.002370.70384970.03660.703733.8212.90.51298366.70.17900.512717.25617A1CGSL2.00580.70424170.09980.703912.146.060.51309779.00.21350.512788.54722A4PlCKR3.16770.70522080.11840.704831.072.700.51311569.30.23840.512768.14722A4(dup)PlCKR3.16770.70521470.11840.704830.972.480.51312269.40.23610.512778.34723A3PlCKR1.65690.70420990.06960.703982.095.910.51309879.00.21420.512788.54723A4PlCKR1.431100.70435280.03750.704232.145.900.513060108.20.21920.512737.64723A13PlCKR0.67630.70413280.03120.704031.333.540.51311879.40.22680.512788.5°THOL,tholenticbasalt,CG,coarse-grainedrnaficrock,PlC,picrite,HI-MG,high-MgObasalt;OUTLIER,anomalouspillowedflowinplots.bAbbreviationsforareaare:KR,KarmutsenRange,SL,SchoenLake,MA,MountArrowsmith.(dup)indicatescompletechemistlyduplicate.AllisotopicandelementalanalysescarriedoutatthePCIGRthecompletetraceelementanalysesareshowninAppendixC.Table 2.4 Hfisotopic compositions ofKarmutsen basalts, Vancouver Island, B.C.SampleGroupa AreabLu Hf 177Hff’76Hf2m 45 176Lu/1Hf 177Hf/176Hf45(t)(ppm) (ppm) 230 Ma4718A2 THOL MA 0.38 2.53 0.283007 4 8.3 0.0211 0.28291 10.24718A7 THOL MA 0.46 2.41 0.283001 7 8.1 0.0271 0.28288 9.14719A2 THOL MA 0.40 2.16 0.283079 5 10.9 0.0262 0.28296 12.04719A3 THOL MA 0.37 2.12 0.283024 9 8.9 0.0246 0.28291 10.24720A4 THOL SL 0.53 3.04 0.283002 7 8.1 0.0247 0.28289 9.54721A2 THOL SL 0.36 2.99 0.283004 10 8.2 0.0173 0.28293 10.74721A4 THOL SL 0.53 2.79 0.283006 6 8.3 0.0268 0.28289 9.34722A5 OUTLIER KR 0.56 3.15 0.283015 6 8.6 0.0254 0.28290 9.84720A6 CG SL 0.45 2.38 0.283012 5 8.5 0.0269 0.28289 9.54720A7 CG SL 0.46 2.23 0.283012 6 8.5 0.0291 0.28288 9.14720A7(dup) CG SL 0.43 2.35 0.283012 4 8.5 0.0260 0.28290 9.64720A10 CG SL 0.36 1.00 0.283179 7 14.4 0.0515 0.28295 11.54724A3 CG SL 0.34 1.41 0.283035 8 9.3 0.0339 0.28288 9.25616A3 CG KR 0.34 2.70 0.283024 7 8.9 0.0179 0.28294 11.35617A1 CG SL 0.40 140 0.283161 4 13.8 0.0407 0.28298 12.64722A4 PlC KR 0.29 0.48 0283321 10 194 0.0857 0.28294 11.24722A4(dup) PlC KR 0.29 0.45 0.283350 11 20.4 0.0915 0.28294 11.34723A3 PIG KR 0.45 0.98 0.283197 6 15.0 0.0649 0.28291 10.14723A4 PIG KR 0.35 0.90 0.283121 26 12.3 0.0561 0.28287 8.74723A13 PlC KR 0.35 0.63 0.283250 7 16.9 0.0791 0.28290 9.74723A2 HI-MG KR 0.43 1.06 0.283186 7 14.7 0.0579 0.28293 10.8aTHOL, tholeiitic basalt, CG, coarse-grained mafic rock, PlC, picrite, HI-MG, high-MgO basalt; OUTLIER, anomalous pillowedflow in plots.bAbbreviations for area are: KR, Karmutsen Range, SL, Schoen Lake, MA, Mount Arrowsmith. (dup) indicatescomplete chemistry duplicate. All isotopic and elemental analyses carried out at the PCIGR; the complete trace element analysesare shown in Appendix C.60Table2.5PbisotopiccompositionsofKarmutsenbasalts,VancouverIsland,B.C.SampleGroupskea’UThPb2°Pbfl°4Pb2m2°7Pb/204Pb2m2wPb/2o4Pb2m2U/2°Pb5U/2MPb232Thl204Pb206Pb/204Pb207Pb/204Pb208Pbl204Pb(ppm)(ppm)(ppm)230Ma230Ma230Ma4718A2THOLMA0.180.560.7119.53690.002115.61050.001739.06840.003516.00.11653.218.95715.58138.4554718A7THOLMA0.170.520.5419.51330.001615.60700.001539.06970.004220.10.14665.718.78215.57038.3124719A2THOLMA0.180.470.6819.57800.001515.61300.001239.03590.003217.60.12746.418.94015.58138.5014719A3THOL.MA0.140.450.5219.66140.001715.60660.00153920270.004218.00.13158.819.00715.57338.5244720A4THOLSL0.210.720.6519.81690.001215.62180.001039.39970.002621.00.15375.719.05315.58338.5274721A2ThOLSL0.250.720.9119.49720.001915.60840.001539.06610.003917.60.12753.118.85915.57638.4544721A4ThOLSL0.170.610.5419.85520.000915.62290.000839.47090.002020.90.15176.719.09815.58438.5874722A5OUTLIERKR0.330.820.8119.62720.001615.61640.001239.23220.003426.30.19168.018.67315.56838.4484720A6GGSL0.130.450.5619.62090.001715.61240.001339.23070.003915.40.11254.419.06215.58438.6034720A7CGSL0.130.480.6119.66400.000915.61380.000639.25470.001914.20.10353.519.14915.58838.6384720A7(dup)CGSL0.140.440.6119.69510.000615.61570.000639.29750.002214.90.10848.819.15415.58838.7354720M0CGSL0.050.110.5118.87450.001815.57880.001638.31760.00426.10.04414.318.65215.56838.1534724A3CGSL0.090.300.6019.25200.001415.58800.001338.90720.00329.70.07033.618.90015.57038.5195616A3CGKR19.48450.000715.59860.000639.06790.00155617A1CGSL19.12020.000715.59100.000638.44010.00164722A4PIGKR0.050.070.1718.95140.001715.58520.001738.40630.004120.80.15128.818.19715.54738.0744722A4(dup)PlCKR0.050.030.1618.95390.001315.58900.001238.41860.002922.40.16214.118.14215.54838.2574723A3PIGKR0.060.190.3218.98730.001215.58200.001138.44340.002711.40.08338.818.57315.56137.9964723A4PlCKR0.050.0919.17330.001315.58320.00123849840.003335.90.26155.517.86815.51737.8584723A13PlCKR0.030.110.1818.91310.001115.57840.001138.35180.00309.20.06741.518.58015.56137.8734723A2HI-MGKR0.060.190.0920.73210.003715.66560.003039.22950.007341.00.298134.719.24215.59037.676°THOL,tholeliticbasalt,CG,coarse-grainedmaficrock,PlC,picrite,HI-MG,high-MgObasalt;OUTLIER,anomalouspillowedflowinplots.bAbbreviationsforareaare:KR,KarmutsenRange,SL,SchoenLake,MA,MountArrowsmlth.(dup)indicatescompletechemistryduplicate.AllisotopicandelementalanalysescarriedoutatthePCIGR;thecompletetraceelementanalysesshowninAppendixC.208Pb/204Pb= 38.153-38.735. One high-MgO basalt has the highest initial 206Pb/204Pb(19.242) and 207Pb/204Pb(15.590), and the lowest initial 208Pb/204Pb(37.676). The Pbisotopic ratios for Karmutsen samples define broadly linear relationships in Pb isotopeplots. The age-corrected Pb isotopic compositions of several picrites have been affectedby U and Pb (andlor Th) mobility during alteration (Fig. 2.10).ALTERATIONThe Karmutsen basalts have retained most of their original igneous structures andtextures; however, secondary alteration and low-grade metamorphism generated zeoliticand prehnite-pumpellyite-bearing mineral assemblages (Table 2.1) (Cho et a?., 1987),which have primarily affected the distribution of the LILE and the Sr isotopicsystematics. The Keogh Lake picrites show the strongest effects of alteration and havehigher LOT (up to 5.5 wt %), variable LILE, and higher measured Sr, Nd, and Hf andlower measured Pb isotope ratios than tholeiitic basalts. As a result, Sr and Pb isotopesshow a relationship with LOl (Figs 2.9 and 2.10), whereas Nd and Hf isotopiccompositions do not correlate with LOI (not shown). The relatively high initial Srisotopic compositions for high-MgO lavas (up to 0.7052) likely resulted from an increasein8SriSrthrough addition of seawater Sr (e.g. Hauffet a?., 2003). High-MgO lavasfrom the Caribbean plateau have similarly high 87Sr/86Srcompared to basalts (Révillon eta?., 2002). The correction for in situ decay on initial Pb isotopic ratios has been affectedby mobilization of U and Pb (andlor Th) in whole rocks since their formation. A thoroughacid leaching during sample preparation was used that has been shown to effectivelyremove alteration phases (Weis et a?., 2006; Nobre Silva et a?., submitted). The HFSEabundances for tholeiitic and picritic basalts exhibit clear linear correlations in binarydiagrams (Fig. 2.8), whereas plots of Lll.E versus HFSE are highly scattered (not shown)due to the mobility of some of the alkali and alkaline earth LILE (Cs, Rb, Ba, K) and Srfor most samples during alteration.The degree of alteration in the Karmutsen samples does not appear to be related toeruption environment or depth of burial. Pillow rims are compositionally different frompillow cores (Surdam, 1967; Kuniyoshi, 1972), and aquagene tuffs are chemicallydifferent from isolated pillows within pillow breccia, but submarine and subaerial basalts62exhibit a similar degree of alteration. There is no clear correlation between the submarineand subaerial basalts and some commonly used chemical alteration indices [e.g. Ba/Rbvs.K20/P05(Huang & Frey, 2005)]. There is also no definitive correlation between thepetrographic alteration index (Table 2.1; 1-least altered, 3-intensely altered) and chemicalalteration indices, although 10 of 13 tholeiitic basalts with the highest K and LILEabundances have a petrographic alteration index of 3.OLIVINE ACCUMULATION IN PICRITIC LAVASGeochemical trends and petrographic characteristics indicate that accumulation ofolivine played an important role in the formation of the Keogh Lake picrites on northernVancouver Island. The Keogh Lake picrites show a strong linear correlation in plots ofAl203,Ti02,Sc, Yb, and Ni versus MgO and many of the picrites have abundant clustersof olivine pseudomorphs (Table 2.1; Fig. 2.11). There is a strong linear correlationbetween modal percent olivine and whole-rock magnesium content for the Keogh Lakepicrites (Fig. 2.11). Magnesium contents range between 10 and 19 wt % MgO and theproportion of olivine phenocrysts in these samples varies from 0 to 42 vol % (Fig. 2.11).With a few exceptions, the size range and median size of the olivine phenocrysts in mostsamples is comparable. The clear correlation between proportion of olivine phenocrystsand whole-rock magnesium contents indicates that accumulation of olivine was directlyresponsible for the high MgO contents (>10 wt %) for most of the picritic lavas.DISCUSSIONThe Wrangellia oceanic plateau on Vancouver Island was constructed in a three-layered structure of mostly tholeiitic basalt with a restricted range of composition. Theplateau formed rapidly during a single phase (ca. 230 Ma) and hiatuses between eruptionswere not long-lasting so there was very little accumulation of sediments. The pillowedand unpillowed flows that built up the submarine volcanic edifice from the deep seafloorresulted primarily from different effusive rates and local topography. Some of theunpillowed flows represent master tubes for the delivery of lava to distal parts ofsubmarine flow fields. Abundant picritic pillow lavas erupted during the middle and latterstages of submarine growth (Nixon et aL, 2008), in areas of the plateau exposed on63Figure 2.11 Relationship between abundance of olivine phenocrysts and whole-rock MgO contents forKeogh Lake picrites. (a) Tracings of olivine grains (gray) in six high-MgO samples made from high resolution(2400 dpi) scans of petrographic thin-sections. Scale bars are 10 mm in length and sample numbers areindicated in upper left. (b) Modal % olivine vs. whole-rock MgO. Black line is the best-fit line for ten highMgO samples. The areal proportion of olivine was calculated from raster images of olivine grains usingImageJ® image analysis software, which provides an acceptable estimation of the modal abundance (e.g.Chayes, 1954).MgO (wt %)2064northern Vancouver Island. The overlying volcaniclastic units formed as a consequenceof eruption in shallow water. A broad subaerial platform was constructed of well-layeredsheet flows and, as volcanism waned, local interfiow carbonate deposits developed, alongwith plagioclase megacrystic flows and local volcaniclastic deposits. The geochemicaland stratigraphic relationships observed in the Karmutsen Formation on VancouverIsland provide constraints on the construction of oceanic plateaus, the source of magmasfor a plume head impinging on oceanic lithosphere, and the conditions of melting andsubsequent magmatic evolution of basaltic magmas involved in the formation of anoceanic plateau. Studies of the formation of this oceanic plateau in the following sectionsexamine the (1) temperature and extent of melting of picritic lavas; (2) composition of themantle source; (3) the depth of melting and residual mineralogy in the source region; and(4) low-pressure evolution of the Karmutsen basalts.Melting conditions and major-element composition of primary magmasThe primary melts to most flood basalt provinces are believed to be picritic (e.g.Cox, 1980) and they have been found in many continental and oceanic flood basaltprovinces worldwide [e.g. Siberia; Karoo; Parana-Etendeka; Caribbean; Deccan(Saunders, 2005)). Near-primary picritic lavas with low total alkali abundances requirehigh-degree partial melting of the mantle source (e.g. Herzberg & O’Hara, 2002). TheKeogh Lake picrites from northern Vancouver Island are the best candidates for least-modified partial melts of the mantle plume source, despite having accumulated olivinephenocrysts, and can be used to estimate conditions of melting and the composition ofprimary magmas for the Karmutsen Formation. Herzberg et al. (2007) have provided athorough description of a technique for inverse and forward modeling for estimatingmantle temperature, melt fraction, primary magma composition, and source residuecomposition using major elements.The estimated melting conditions and primary magma compositions for thepicritic lavas of the Karmutsen Formation, and a brief summary of the modelingmethod,are presented in Figure 2.12 and Table 2.6. If Karmutsen picritic lavas were derived fromaccumulated fractional melting of fertile peridotite, the primary magmas would havecontained 15-17 wt % MgO and -40 wt % CaO (Fig. 2.12; Table 2.6), formed from 23-6514121086414121030MgO (wt%)8Karmutsen flood basaltsA Picrite6D High-MgO basalte Tholeiitic basalt— OIMne addition model4+ Olivine addition (5% increment)$ Pnmary magmaFigure 2.12 Estimated primary magma compositions for three Keogh Lake picrites (samples4723A4,4723A13, 5616A7) using the forward and inverse modeling technique of Herzberg et a!. (2007).Karmutsencompositions and modeling results are overlain on diagrams provided by C. Herzberg. (a) Whole-rockFeO vs.MgO for Karmutsen samples from this study. Total iron estimated to be FeO is 0.90. Gray linesshow olivinecompositions that would precipitate from liquid of a given MgO-FeO composition. Black lineswith crossesshow results from olivine addition (inverse model) using PRIMELT1 (Herzberg et al., 2007). (b)Si02 vs. MgOwith Kanntusen lavas and model results. (c) FeO vs. MgO with Karmutsen lava compositionsand results offorward model for accumulated fractional melting of fertile peridotite. (d) CaO vs. MgO withKarmtusen lavacompositions and model results. To briefly summarize the technique [see Herzberg et a!. (2007) for completedescription], potential parental magma compositions for the high-MgO lava series were selected (highest MgOand appropriate CaO) and, using PRIMELTI software, olivine was incrementally added tothe selectedcompositions to show an array of potential primary magma compositions (inverse model). Then, usingPRIMELT 1, the results from the inverse model were compared to a range ofaccumulated fractional melts forfertile peridotite, derived from parameterization of the experimental results ofWalter (1998) (forward model;Herzberg & O’Hara, 2002). A melt fraction was sought that was unique to both theinverse and forward models(Herzberg et al., 2007). A unique solution was found when there was a common meltfraction for both modelsin FeO-MgO and CaO-MgO-A1203-Si0 (CMAS) projection space. This modeling assumesolivine was theonly phase crystallizing and ignores chromite precipitation, and possible augite fractionationin the mantle(Herzberg & O’Hara, 2002). Results are best for a residue of spinel lherzolite (not pyroxenite). Thepresence ofaccumulated olivine in samples of Keogh Lake picrites used as starting compositions does notsignificantlyaffect the results because we are modeling addition of olivine. The tholeiitic basalts cannot beused formodeling because they are all plag+ cpx + ol saturated.Si02(wt%)Melting model— Dashed fines = initial melting pressure —Solidus7KR-4G03(b)Garnet Peridosite40Liquid compositionsFertile peridotite sourceAccumulated Fractional Melting modelThick black lines = initial melting pressureGray lines = final melting pressure66Table 2.6 Estimatedprimary magma compositionsfor Karmutsen basalts and other oceanic plateaus/islandsSample 4723A4 4723A13 561 6A7 93G171 AverageOJPaMaunaKeabGorgona(Weight %):Si02 47.0 47.6 46.9 47.8 47.348.0 46.3 46.1Ti02 0.68 0.46 0.60 0.46 0.550.62 1.93 0.56A1203 15.3 14.3 13.0 13.9 14.112.3 9.6 11.7Cr203 0.10 0.20 0.04 0.09 0.110.07 0.26 0.16Fe203 1.03 1.08 0.89 1.09 1.020.90 1.08 1.18FeO 8.7 9.0 9.5 8.9 9.09.2 10.3 10.1MnO 0.15 0.15 0.16 0.19 0.160.17 0.18 0.18MgO 15.3 16.3 17.4 16.1 16.316.8 18.3 18.8CaO 9.9 9.8 9.6 10.3 9.910.3 10.1 10.0Na20 1.59 0.90 1.74 1.01 1.311.36 1.67 1.04K20 0.07 0.06 0.05 0.11 0.070.08 0.41 0.03NO 0.05 0.08 0.08 0.07 0.070.10 0.08 0.11Eruption T(°C) 1354 1375 1397 1369 13741382 1415 1422Potential T(°C) 1467 1491 1517 1486 14901500 1606Fo content (olMne) 91.2 91.2 91.6 91.2 91.390.5 91.4 90.6Melt fraction 0.23 0.27 0.26 0.27 0.260.27 0.28%ol addition 4.2 2.5 24.3 0.8 7.918Ontong Java primary magma composition for accumulated fraction melting (AFM) from Herzberg (2004).Mauna Kea primary magma composition is average of 4 samples in Table I of Herzberg (2006).Gorgona primary magma composition for AFM for 1 F, fertile source, in Table 4 of Herzberg and OHara (2002).Total iron estimated to be FeO is 0.90.6727% partial melting, and would have crystallized olivine with a Fo content of --91 (Fig.2.12; Table 2.6). Calculated mantle temperatures for the Karmutsen picrites (—‘1 490°C)indicate melting from anomalously hot mantle (100-200°C) compared to ambient mantlethat produces mid-ocean ridge basalt (MORB; ‘--1280-1400°C; Herzberg et al., 2007).Several picrites are near-primary melts and required very little addition of olivine (e.g.sample 93G1 71, with measured 15.8 wt % MgO). These temperature and meltingestimates of the Karmutsen picrites in this study are consistent with the decompression ofhot mantle peridotite in an actively convecting plume head (i.e. plume initiation model).The estimated primary magma compositions and melting conditions for someother LIPs (e.g. the Ontong Java Plateau) are similar to estimates for the KarmutsenFormation (Table 2.6). Herzberg (2004) found that primary magma compositions forOntong Java basalts, assuming a peridotite source, would contain —17 wt % MgO and—l0 wt % CaO (Table 2.6), and these magmas would have formed by 27% melting withmantle potential temperatures of —1500°C, and first melting occurring at —-3.6 GPa. Fitton& Godard (2004) estimated 30% partial melting of the Ontong Java lavas based on the Zrcontents of primary magmas calculated by incremental addition of equilibrium olivine toanalyses of Kroenke-type basalt. If a considerable amount of eclogite was involved in theformation of the Ontong Java Plateau, the excess temperatures estimated by Herzberg eta!. (2007) may not be required (Korenaga, 2005). Estimated primary magmas for MaunaKea and Gorgona have higher MgO (18-19 wt % MgO) and comparable CaO (10 wt %CaO) (Table 2.6; Herzberg (2006)).Source of Karmutsen lavasKarmutsen samples have isotopic compositions belonging to the field of oceanisland basalts (OIB) and provide a sampling of the Pacific mantle Ca. 230 Ma (Fig. 2.13).The Karmutsen tholeiitic basalts have initial ENd and 87Sr/86Srthat fall within the range ofbasalts from the Caribbean Plateau (e.g. Kerr et a!., 1996; Kerr et al., 1997; Hauff et a!.,2000; Kerr, 2003), a range of Hawaiian isotope data (e.g. Frey eta!., 2005), and liewithin and just above the range for the Ontong Java Plateau (Tejada et a!., 2004; Fig2.13). Karmutsen tholeiitic basalts with the highest initial8Ndand lowest initial 87Sr/865rlie just below the field for northern East Pacific Rise (EPR) MORB (e.g. Niu eta!., 1999;680.70487SrI 86Sr(initial)206pb/204pbFigure 2.13 Comparison of age-corrected (230 Ma) Sr-Nd-Hf-Pb isotopic compositions for Karmutsen flood basaltson Vancouver Island to age-corrected OIB and MORB. (a) Initial8Ndvs. 87Sr/86Sr. (b) Initial8Hfvs.8NdBoth fieldswith dashed lines are Indian MORE. (c) Measured and initial 207Pb/204Pb vs. 206Pb/204Pb. (d) Measured and initial208Pb/204Pbvs. 206PbP°4Pb. Most of the compiled data was extracted from the GEOROC database (http://georoc.mpchmainz.gwdg.de/georocl). Data for Ontong Java from Mahoney et a?. (1993), Babbs (1997), and Tejada et a?. (2004);Indian MORE from Salters (1996), Kempton et a?. (2002), and Janney et a?. (2005); Pacific MORE from Mahoney eta?. (1992, 1994), Nowell et a?. (1998), Salters and White (1998), and Chauvel and Blichert-Toft (2001); ExplorerRidge data from Cousens and Weis (pers. conmi., 2007); OTB array line from Vervoort et a?. (1999). EPR is EastPacific Rise. Several high-MgO samples affected by secondary alteration were not plotted for clarity in Pb isotopeplots. Dashed lines indicate Bulk Silicate Earth (B SE). An extended reference list is available upon request._t •S14121086420I.I • •ft-’4•.1.-20.702II I liii16Indian__.EHfinitiah• Indian MORB6 .F • • East Pacific Rise• —Caribbean Plateau4 £ •Ontong JavaHawaii2 I• Karmutsen tholeiitic basalt0iOlB array • Karmutsen high-MgO lavai\2(b) II I I I I I I I I I I IENd(t1ao)•0.703 0.705 0.706 -4 -2 0 2 4 6 8 10 12 1415.6015.5515.5015.45I1F[ I I I III I I IIF F FI1 I 1 I I••I0•I..• I • East Pacific RiseJuan de Fuca/Gorda -• Explorer Ridge• Caribbean Plateau• Ontong Java• Hawaiio Karmutsen Formation (measured) —•Karmutsen Formation (initial)•Karmutsen Formation(different age-correction for 3 picrites)IIIIIIIIIIIII II I I I I I I I I17.5 18.0 18.5 19.0 19.5 20.0 17.5 18.0 18.5 19.0 19.5 20.069Regelous et a!., 1999). Initial Hf and Nd isotopic compositions place the Karmutsenbasalts on the edge of the field of OIB (age-corrected), with slightly higher initial8Ndandsimilar initial8Hfto Ontong Java, and slightly lower initial £Hf and similar initial ENd tothe range of compositions for Hawaii and the Caribbean Plateau (Fig. 2.13). Karmutsensamples are slightly offset to the low side of the OIB array (Vervoort et a!., 1999),slightly overlap the low8Hfend of a field for Indian MORB (Kempton et a!., 2002;Janney et aL, 2005), and samples with the highest initial8Hfand8Ndlie just at the edge ofthe presently defined field for Pacific MORB (Mahoney et aL, 1992, 1994; Nowell et al.,1998; Chauvel & Blichert-Toft, 2001). Karmutsen Pb isotope ratios overlap a broad rangefor the Caribbean Plateau and the linear trends for Karmutsen samples intersect fields forEPR MORB, Hawaii, and Ontong Java in 208Pb-206Pb space, but do not intersect thesefields in207Pb-206Pbspace (Fig. 2.13). The more radiogenic207Pb/204Pb for a given206Pb/204Pbof Karmutsen basalts indicate they are isotopically enriched in comparison toEPR MORB, Hawaii, and Ontong Java.The Hf-Nd isotopic geochemistry of volcanic rocks of the Karmutsen Formationon Vancouver Island indicates an isotopically rather uniform mantle source. The Hf andNd isotopic compositions of Karmutsen basalts indicate long-term depletion of the morehighly incompatible elements in the mantle source and are distinct from MORE; they areless depleted than the source of MORE and there is no clear evidence of involvement ofMORE-type mantle. The limited variation of Hf and Nd isotopic compositions is slightlygreater than in the isotopically very uniform mantle source of Ontong Java. Small-scaleheterogeneities in the mantle source of Karmutsen lavas may have been diluted by thehigh degrees of partial melting, similar to Ontong Java (Tejada et al., 2004). Lassiter eta!. (1995) suggested that mixing of a plume-type source with8Nd+6 to +7 with arcmaterial with low Nb/Th could reproduce variations in the Karmutsen basalts, but theabsence of low Nb/La ratios in most of the basalts restricts the amount of arc lithosphericinvolvement (based on major and trace elements and Sr, Nd, and Pb isotopiccompositions for a suite of 29 samples from Buttle Lake in Strathcona Provincial Park oncentral Vancouver Island; Fig. 2.1). The isotopic composition and trace-element ratios ofKarmutsen basalts in this study do not indicate significant involvement of Paleozoic arc70lithosphere on Vancouver Island. There are no clear HFSE-depletions in Karmutsentholeiitic basalts.The Sr-Nd-Hf-Pb isotope systematics for Karmutsen basalts distinguish thesource of Karmutsen basalts from MORB and some OIB. The Hf and Nd isotopesystematics provide the firmest constraints on the character of the source for Karmutsenlavas. The LU/Hf ratios of the picritic lavas (mean 0.08) are considerably higher than intholeiitic basalts (mean 0.02), but the small range of initial £Hf for the two lava suitessuggests that these differences in trace-element compositions were not long-lived, and donot correspond to intrinsic differences in the mantle plume source. Evolution of Hfisotopes with time shows that if the picritic and tholeiitic lavas originally had a similarrange of8Hfit would take —l55 m.y. for Hf isotope ratios to evolve so there would be nooverlap in8Hf(Fig. 2.14); this is similar to a concept emphasized by Arndt et a!. (1997)using Nd isotopes of lavas from Gorgona Island. Therefore, it is unlikely that the picriteshad high LU/Hf ratios long before ascent of the plume and the different Lu/Hf ratioslikely developed during the melting process within the plume during ascent. The similarinitial Hf and Nd isotopic ratios of the picrites and tholeiitic basalts preclude the limitedisotopic range being simply the result of mixing of magmas. The slightly different Pbisotope ratios of the picrites and tholeiitic basalts are at least partly a result of alteration-related effects. As observed in the Caribbean Plateau, the radiogenic Pb in Karmutsenbasalts is systematically different than in MORB, Ontong Java, and OIB from Hawaii,but Hf and Nd isotope compositions indicate a homogeneous, OIB-type enriched mantlesource.REE modeling: Dynamic melting and source mineralogyThe combination of isotopic and trace-element geochemistry of picritic andtholeiitic Karmutsen lavas indicates that differences in trace elements may haveoriginated from different melting histories within a predominantly homogeneous,depleted (but not MORE) mantle source. Dynamic melting models simulateprogressivedecompression melting where some of the melt fraction is retained by the residue andonly when the degree ofpartial melting is greater than the critical massporosity, and thesource becomes permeable, is excess melt extractedfrom the residue (Zou, 1998). For the7125PicriteHigh-MgO basalt20— • Tholelitic basalt11 - --15- —155 m.y.—+Ij.-’/ II -Hf10hypotheticalI actual230 Ma0• • ‘ I a I I I I500 400 300 200 100 0Age (Ma)Figure 2.14 Evolution of EHf with time for picritic andtholeiitic lavas for the Karmutsen Formation. Thedifferent Lu/Hf ratios of the Karmutsen lavas likelyoriginated during melting processes within the plume.For example, if the picritic and tholeiitic lavas possessedthe Lu/Hf ratios that they did at 230 Ma more than 155m.y. before their formation, they would have evolved tohave a different range of Hf isotopic compositions by230 Ma (shown by the gray and white vertical boxes fortholeiitic and picritic lavas, respectively). This is shownby hypothetical compositions at 400 Ma (using EHf(23OMa) for each sample) and evolution trends using176Lu!’77Hffor each sample. Picritic lavas have high Lu!Hf ratios so they accumulate radiogenic Hf within arelatively short geologic timespan. Decay constant of1.87 x 101yr’ used from Scherer et a!. (2001). Errorbars are smaller than symbols.72Karmutsen lavas, evolution of trace element concentrations was simulated using theincongruent dynamic melting model developed by Zou & Reid (2001). Melting of themantle at pressures greater than -4 GPa involves incongruent melting (e.g. Longhi, 2002)with olivine being produced during reaction of cpx+ opx + sp for spinel peridotites.Modeling was separated into two intervals of melting, for a garnet therzolite (gt lherz)and spine! lherzolite (sp lherz) source, using coefficients for incongruent meltingreactions based on experiments on lherzolite melting (e.g. Kinzler & Grove, 1992;Longhi, 2002). The model solutions are non-unique and a range in degree of melting,partition coefficients, melt reaction coefficients, and proportion of mixed sourcecomponents is possible to achieve acceptable solutions.The LREE-depleted high-MgO lavas require melting of a depleted spinellherzolite source (Fig. 2.15). The modeling results indicate a high degree of melting (22-25%), similar to results from major-element modeling using PRIMELT1 (discussedabove), from a LREE-depleted source, and melting of garnet was probably not involvedin formation of the high-MgO lavas. The enriched tholeiitic basalts involved melting ofboth garnet and spinel lherzolite and represent aggregate melts produced from continuousmelting throughout the depth of the melting column (Fig. 2.15), which were subsequentlyhomogenized and fractionated in magma chambers at low pressure. The modeling resultsindicate that these aggregate melts would have involved lesser proportions of enriched,small-degree melts (1-6% melting) generated at depth and greater amounts of depleted,high-degree melts (12-25%) generated at lower pressure (ratio of gt lherz to sp lherz meltwould have been approximately ito 3, or 1 to 4; see Fig. 2.15). The degree of meltingwas high (23-27%), similar to the degree of partial melting of primary melts that eruptedas the Keogh Lake picrites, from a source that was likely depleted in LREE.Although the modeling results indicate that the depleted high-MgO Karmutsenlavas were formed from high-degree partial melting within the spinel therzolite stabilityfield (—0.9-2.5 GPa), the source was not necessarily more depleted in incompatible traceelements than the source of the tholeiitic lavas. It is possible that early high-pressure,low-degree melting left a residue within parts of the plume (possibly the hotter, interiorportion) that was depleted in trace elements (e.g. Elliott et al., 1991); furtherdecompression and high-degree melting of such depleted regions could then have73100La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu0.0 03 1.0La/SmCNFigure 2.15 Trace-element modeling results for incongruent dynamic mantle melting for picritic and tholeiitic lavas from theKarmutsen Formation. Three steps of modeling for tholeiitic lavas are shown (order of modeling is shown going upwardsfrom panel e to c to a, following the arrows): (e) Melting of garnet lherzolite, (c) Mixing of garnet and spinel therzolite melts,and (a) Mixing of enriched tholeiitic and depleted picritic melts. (b) Melting of spinel therzolite for high-MgO lavas. Shadedfield is the range of REE patterns for Karmutsen tholeiitic basalts in panels a, c, and e and for picrites in panel b. Patterns withsymbols in panels a, b, c, and e are modeling results (patterns are 5% melting increments in panels a and c and 1% incrementsin panel e. Abbreviations are: PM, primitive mantle; DM, depleted MORB mantle; gt therz, garnet therzolite; sp lherz, spinellherzolite. The ratios of percent melting for garnet and spinel therzolite are indicated in panel c. A range of proportion ofsource components (0.7DM:0.3PM to 0.9DM:0.1PM) can reproduce the variation in high-MgO lavas. (d) Dy!YbCN vs. La!SmcN for results of spinel Iherzolite melting modeling and Karmutsen samples. Tickmarks are 5%increments and proportionsof mantle source components are labelled next to curves. (f) Dy/YbCN vs. La/SmCN for results of spinel lherzolite+ garnettherzolite melting modeling and Karmutsen samples. Melting modeling uses the formulation of Zou & Reid (2001), anexample calculation is shown in their Appendix. PM from McDonough & Sun (1995) and DM from Salters & Stracke (2004).Melt reaction coefficients of spinel lherzolite from Kinzler & Grove (1992) and garnet lherzolite from Walter (1998).Partition coefficients from Salters & Stracke (2004) and Shaw (2000) were kept constant. Source mineralogy for spinellherzolite (0.l8cpx:0.27opx:0.52o1:0.O3sp) from Kinzler (1997) and for garnet therzolite (0.34cpx:0.O8opx:0.53o1:0.OSgt)from Salters & Stracke (2004). A source composition with mantle components up to 0.3 PM and 0.7 depleted mantle canreproduce REE patterns similar to those of the high-MgO lavas, with best fits for a melt fraction of 0.22-0.25. Concentrationsof tholeiitic basalts cannot easily be reproduced from an entirely primitive or depleted source. A range of source compositions(0.7DM:0.3PM to 0.3DM:0.7PM) and garnet and spinel lherzolite melt mixtures generate similar results (x gtlherz + 3xsplherz to x gtlherz + 4x splherz; where x is percent melting in the interval of modeling; the ratio of the respective meltproportions in the aggregate melt was kept constant).hoiclavas0.8 enriched melt + 0.2 depleted high-MgO meltSpinel lherzolite+ Spirl lherzolite meltgarnet lherzolite melt (from panel B)(from panel C)(a) .. A.C.210UEHigh-MgO lavasMelting of spinel lherzolite 20% melting30% meltingurce(O.7DM+O3PMC0-CU0.EC,C,C0L)C,0.EC,0C0-CUC,0.EC,U,La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu101001010010I • I • I •Melting of spinel lherzolitemelting0.7DM +0.3PMmelting30% melting(d)La/Srn0 0.5 1.0 1.5 2.(% melting ratio,La Ce Pr Nd Pn-fSm Eu Gd Tb Dy[lo Er Tm Yb Lu1.5‘ ‘ ‘ ‘‘ o’Ie,it’ic lva141312xgtlherz+4xspiherz -—DyIYb1.1Spinel lherzolite +garnet lherzolite melt1.0.e(O.7DM÷O.3P0.9(C)L Ce Pr Nd P4Sm Eu Gd Tb Dy • ‘ •osHo Er TmYb Lu 0.1.5Tholeiitic lavas1 1 I 1 ‘x gnt lherz ÷ 3x sp lherz- - 1% melting 1.4. 25%20%10%____ 13•!eltin- Spinellherzolite+25.....J_12 .garnetlherzolitemelts %meltinp ratioDy/Yb1.1xgntlherz÷4xsplherz1.0 •A Picrite0 High-MgO basalt0.7DM+O.3PM)0.9 - 0 Coarse-g rained(e) •• I I I I I I • •.• Tholeiitic basaltA. •0.81.5 2.074generated the depleted high-MgO Karmutsen lavas. However, the modeling resultsindicate that this earlier melting was not necessary for their formation, depending on theoriginal trace-element composition of the source. Shallow, high-degree melting wouldpreferentially sample depleted regions, whereas deeper, low-degree melting may notsample more refractory, depleted regions (e.g. Caribbean; Escuder-Viruete et al., 2007).Decompression melting within the mantle plume initiated within the garnet stability fieldat high mantle potential temperature (T>l450°C) and proceeded beneath oceanic arclithosphere within the spinel field where more extensive degrees of melting could occur.Magmatic evolution of Karmutsen tholelitic basaltsPrimary magmas in LIPs leave extensive coarse-grained residues within or belowthe crust and these mafic and ultramafic plutonic sequences represent a significantproportion of the original magma that partially crystallized at depth (e.g. Farnetani et al.,1996). For example, seismic and petrological studies of Ontong Java (e.g. Farnetani etal., 1996), combined with the use of MELTS (Ghiorso & Sack, 1995), indicate that thevolcanic sequence may have been only -40% of the total magma reaching the Moho.Neal et al. (1997) estimated between 7 and 14 km of cumulate thickness correspondingwith 11 to 22 km of flood basalt, depending on the presence ofpre-existing oceanic crustand the total crustal thickness (25-3 5 km), and that 3 0-45% fractional crystallization tookplace for Ontong Java magmas. Interpretations of seismic velocity measurements suggestthe presence of pyroxene and gabbroic cumulates --9-16 km thick beneath the OntongJava Plateau (e.g. Hussong et al., 1979).Wrangellia is characterized by high crustal velocities in seismic refraction lines,which is indicative of mafic plutonic rocks extending to depth beneath Vancouver Island(Clowes et al., 1995). Wrangellia crust is —25-30 km thick beneath Vancouver Island andis underlain by a strongly reflective zone of high velocity and density that has beeninterpreted as a major shear zone where lower Wrangellia lithosphere was detached(Clowes et a!., 1995). The vast majority of Karmutsen flows are evolved tholeiitic lavas(e.g. low MgO, high FeOIMgO) indicating an important role for partial crystallization ofmelts at low pressure, and a significant portion of the crust beneath Vancouver Island hasseismic properties that are consistent with crystalline residues from partially crystallized75Karmutsen magmas. To test the proportion of the primary magma that fractionated withinthe crust, MELTS (Ohiorso & Sack, 1995) was used to simulate fractional crystallizationusing several estimated primary magma compositions from modeling results usingP1UMELT1. The major-element composition of primary magmas could not be estimatedfor the tholeiitic basalts due to extensive plag+ cpx + ol fractionation, and thus theMELTS modeling of the picrites is used as a proxy for the evolution of major elements inthe volcanic sequence. A pressure of 1 kbar was used with variable water contents(aithydrous and 0.2 wt % H20), calculated at the quartz-fayalite-magnetite (QFM)oxygen buffer. Experimental results from Ontong Java indicate that most crystallizationoccurs in magma chambers shallower than 6 km deep(<2 kbar; Sano & Yamashita,2004); however, some crystallization may take place at greater depths (3-5 kbar;Farnetani et aL, 1996).A selection of the MELTS results are shown in Figure 2.16. Olivine and spinelcrystallize at high temperature until —P15-20 wt % of the liquid mass has fractionated, andthe residual magma contains 9-10 wt % MgO (Fig. 2.16). At1235-1225°C, plagioclasebegins crystallizing and between 1235 and 1190°C olivine ceases crystallizingandclinopyroxene saturates (Fig. 2.16). As expected, the addition of clinopyroxene andplagioclase to the crystallization sequence causes substantial changes in the compositionof the residual liquid; FeO and Ti02 increase and A1203and CaOdecrease, while thedecrease in MgO lessens considerably (Fig. 2.16). Tn general, thecompositions of thepredicted residual liquids follow trends for compositions of Karmutsenlavas, but misssome of the low MgO, high CaO basalts (some of which are due to plag accumulation).Also, trends in Al203and CaO of Karmutsen basalts are not clearly decreasing withdecreasing MgO. Increasing the crystallization pressure slightly and water content of themelts does not systematically improve the match to the observed data.The MELTS results indicate that a significant proportion ofcrystallization takesplace before plagioclase (‘-15-20% crystallization) and clinopyroxene(-35-45%crystallization) join the crystallization sequence. These results suggest >45%crystallization of primary magmas occurred (estimated from residual liquid % in Fig.2.16) and, if the original flood basalt stratigraphy on Vancouver Island was—6 1cm, atleast an equivalent amount of complementary plutonic rocks should havecrystallized.76LLELVtEL9LLLSt6t9L86OLLLELVtEL9LLLssduizqjs&IjosiutpjqjinrndwSApmbijTP!so1uaoj().OZHouq(ETVZLI7jduius)urnjJjiuowo.sssdPz!TImsjosuoTijodoidTuI!N()coiipuuuoruuonJj%LsiOH%iotpA.iojiuod-puspucpuuotuoT.go,’isiusisnoip(qusiojuod-puS11.OZH%O!1I!IP5UOJP1(11I!‘II)1°1N1OPIt1O1‘OH%opuOHOU1{IAp5511SuAiqjjjoainsssidvsjnsaz‘W’!piIL10E6pu£VEZLssidwusiOjswuwi(iuiudpsuwsssqjijutuos‘EIvlLtsjdwsoj(/f)ivissqz.Is-{JOsnbimps&IiJspouIsiwoi,jssussuujstio;uwui&Iuuiudpswmtssjouonisodwoosqusn‘(sjaipi)uoiiisodmos&iiissuoiojuMo1ssmsijnsaisi‘i(iupiosuornsoduioouuussnuujopaidwoosprnbJJrnlpsa1josuornsodwoousuisjs-iofwsi(p)puu‘(o)‘(q)‘(s)uouuojussnuustpwoSA1siiiisJopuo!.Is!dopsidwoo(c661‘i°s‘oslon{9)S1iNWOSU5UI5S1018WJOjSflUSSJU!j5pOwRO13uZqSAJO1’°!3°pJAUO{91Z‘flLI(%M)06W(%IM)OWOE8L9LVLLOL89V0O8L9tVtLOL89V0(p)VVVV9L86OLLLEL(%IM)Qe3,III(3)VVV•!1!!IOq.1.@seq0 6 W-L1 6 !H(%M)S 0 Z 1 y1!J)!dVIIIIIIIII(%M)0 6 WOZ8L9tVLtOL89VZ0(%IM)0 6 W0St9LVLLOL891’0(q)0•00 6 Wc.00•(%)p!nb!IienpsiOLOOSO0009OLc•(8)0906OOL0ZH%1MO(J)/06110HOU’0ZLed6uIIsAjpue•• PiflbIIIeflP!S9°c.-i0OLL0.E0056(%)5 01j.SEIIO•VThe Karmutsen basalts and their plutonic residues thus represent a significant addition ofcrust (perhaps >12 km thickness), previously thickened by Paleozoic arc activity.CONCLUSIONThe Karmutsen Formation covers —20,000 km2 of Vancouver Island, BritishColumbia, and was constructed as a large oceanic plateau during a single phase over ageologically short interval (ca. 230 Ma). The tripartite volcanic stratigraphy onVancouver Island is upwards of 6 km of submarine flows, volcaniclastic deposits, andmassive sheet flows; these volcanological differences are primarily related to the eruptionenvironment (deep-water, shallow-water, subaerial). The rapid growth of the plateauprevented intervening sediments from accumulating, except in the uppermost stratigraphywhere isolated limestone lenses commonly associated with pillowed and volcaniclasticbasalts preserve a record of the subsidence history of the plateau.The Wrangellia plateau on Vancouver Island was constructed dominantly oftholeiitic basalt with a restricted major- and trace-element, and isotopic, composition.Picritic pillow basalts erupted in the middle to latter portion of the submarine phase.Modeling results of Karmutsen picrites indicate melting of anomalously hot mantle(—4500°C) and extensive degrees of partial melting (23-27%) and are consistent with aplume initiation model. The lavas that built the volcanic edifice were derived from anisotopically relatively uniform depleted mantle source distinct from the source of MORE.There are compositional similarities between the source of Karmutsen basalts onVancouver Island and the sources of ocean islands (e.g. Hawaii) and plateaus (e.g.Ontong Java) in the Pacific Ocean. The Karmutsen basalts erupted atop Paleozoic arcvolcanic sequences, but there is no clear evidence from our work on Vancouver Island ofsignificant involvement of arc or continental material in formation of the basalts. Picritesformed from melting of spinel lherzolite with a depleted mantle isotopic compositionsimilar to the source of the tholeiitic basalt. The tholeiitic basalts underwent extensivelow-pressure fractionation(<2-3 kbar) and seismic work indicates some of theWrangellia crust beneath Vancouver Island may correspond to the plutonic residues ofthe Karmutsen Formation.78ACKNOWLEDGEMENTSWe would like to thank Nick Arndt for helping us get this project started and NickMassey for insights into Vancouver Island geology. We would also like to thank ClaudeHerzberg for his kind help with modeling techniques. We are grateful to Mikkel Schaufor his insight and enthusiasm during fieldwork. Jane Barling assisted with analyses byMC-ICP-MS. Funding was generously provided by the Rocks to Riches Programadministered by the BC & Yukon Chamber of Mines in 2004, by the BC GeologicalSurvey in 2005, and by NSERC Discovery Grants to J. Scoates and D. Weis. A. Greenewas supported by a University Graduate Fellowship at UBC.REFERENCESBabbs, T. L. (1997). Geochemical and petrological investigations of the deeper portionsof the Ontong Java Plateau: Malaita, Solomon Islands. Unpublished Ph.D.dissertation, Leicester University, U.K.Barker, F., Brown, A. S., Budahn, J. R. & Plafker, G. (1989). Back-arc with frontal-arccomponent origin of Triassic Karmutsen basalt, British Columbia, Canada.Chemical Geology 75, 81-102.Blichert-Toft, J., Weis, D., Maerschallc, C., Agranier, A. & Albarède, F. (2003).Hawaiian hot spot dynamics as inferred from the Hf and Pb isotope evolution ofMauna Kea volcano. Geochemisfry Geophysics Geosystems 4(2), 1-27,doi: 10.1 029/2002GC000340.Carlisle, D. (1963). Pillow breccias and their aquagene tuffs, Quadra Island, BritishColumbia. Journal ofGeology 71, 48-71.Carlisle, D. (1972). Late Paleozoic to Mid-Triassic sedimentary-volcanic sequence onNortheastern Vancouver Island. Geological Society of Canada Paper. Report ofActvities 72-1, Part B, 24-30p.Carlisle, D. & Suzuki, T. (1974). Emergent basalt and submergent carbonate-elasticsequences including the Upper Triassic Dilleri and Welleri zones on VancouverIsland. Canadian Journal ofEarth Sciences 11, 254-279.Chauvel, C. & Blichert-Toft, J. (2001). A hafluium isotope and trace element perspectiveon melting of the depleted mantle. Earth and Planetaiy Science Letters 190, 137-151.Chayes, F. (1954). The theory of thin-section analysis. Journal ofGeology 62, 92-101.Cho, M., Liou, J. G. & Maruyama, S. (1987). Transition from the zeolite to prehnitepumpellyite facies in the Karmutsen metabasites, Vancouver Island, BritishColumbia. Journal ofPetrology 27(2), 467-494.Clowes, R. M., Zelt, C. A., Amor, J. R. & Ellis, R. M. (1995). Lithospheric structure inthe southern Canadian Cordillera from a network of seismic refraction lines.Canadian Journal ofEarth Sciences 32(10), 1485-15 13.Coffin, M. F. & Eldholm, 0. (1994). Large igneous provinces: Crustal structure,dimensions, and external consequences. Reviews ofGeophysics 32(1), 1-36.79Escuder-Viruete, J., Pérez-Estain, A., Contreras, F., Joubert, M., Weis, D., Ullrich, T. D.& Spadea, P. (2007). Plume mantle source heterogeneity through time: Insightsfrom the Duarte Complex, Hispaniola, northeastern Caribbean. Journal ofGeophysical Research 112(B04203), doi: 10.1 029/2006JB004323.Farnetani, C. G., Richards, M. A. & Ghiorso, M. S. (1996). Petrological models ofmagma evolution and deep crustal structure beneath hotspots and flood basaltprovinces. Earth and Planetary Science Letters 143, 8 1-94.Fitton, J. G. & Godard, M. (2004). Origin and evolution of magmas on the Ontong JavaPlateau. In: Fitton, J. G., Mahoney, 3. J., Wallace, P. 3. & Saunders, A. D. (eds.)Origin and Evolution ofthe Ontong Java Plateau. Special Publication 229Geological Society,pp.15 1-178.Frey, F. A., Huang, S., Blichert-Toft, J., Regelous, M. & Boyet, M. (2005). Origin ofdepleted components in basalt related to the Hawaiian hotspot: Evidence fromisotopic and incompatible element ratios. Geochemistry Geophysics Geosystems6(1), 23, doi: 10.1 029/2004GC000757.Galer, S. 3. G. & Abouchami, W. (1998). Practical application of lead triple spiking forcorrection of instrumental mass discrimination. Mineralogical Magazine 62A,491-492.Ghiorso, M. S. & Sack, R. 0. (1995). Chemical mass transfer in magmatic processes, IV.A revised and internally consistent thermodynamic model for the interpolationand extrapolation of liquid-solid equilibria in magmatic systems at elevatedtemperatures and pressures. Contributions to Mineralogy and Petrology 119, 197-212.Greene, A. R., Scoates, 3. S., Nixon, G. T. & Weis, D. (2006). Picritic lavas and basalsills in the Karmutsen flood basalt province, Wrangellia, northern VancouverIsland. In: Grant, B. (ed.) Geological Fieldwork 2005. Paper 2006-1 BritishColumbia Ministry of Energy, Mines and Petroleum Resources,pp.39-52.Greene, A. R., Scoates, J. S. & Weis, D. (2005). Wrangellia Terrane on VancouverIsland: Distribution of flood basalts with implications for potential Ni-Cu-PGEmineralization. In: Grant, B. (ed.) Geological Fieldwork 2004. Paper 2005-1British Columbia Ministry of Energy, Mines and Petroleum Resources,pp.209-220.Hauff, F., Hoernle, K. & Schmidt, A. (2003). Sr-Nd-Pb composition of Mesozoic Pacificoceanic crust (Site 1149 and 801, ODP Leg 185): Implications for alteration ofocean crust and the input into the Izu-Bonin-Mariana subduction system.Geochemistry Geophysics Geosystems 4(8), doi: 10.1 029/2002GC00042 1.Hauff, F., Hoernie, K., Tilton, G., Graham, D. W. & Kerr, A. C. (2000). Large volumerecycling of oceanic lithosphere over short time scales: geochemical constraintsfrom the Caribbean Large Igneous Province. Earth and Planetary Science Letters174, 247-263.Herzberg, C. (2004). Partial melting below the Ontong Java Plateau. In: Fitton, J. G.,Mahoney, 3. 3., Wallace, P. J. & Saunders, A. D. (eds.) Origin and Evolution ofthe Ontong Java Plateau. Special Publication 229 Geological Society, pp. 179-183.Herzberg, C. (2006). Petrology and thermal structure of the Hawaiian plume from MaunaKea volcano. Nature 444(7 119), 605.80Herzberg, C., Asimow, P. D., Arndt, N., Niu, Y., Lesher, C. M., Fitton, J. G., Cheadle,M. J. & Saunders, A. D. (2007). Temperatures in ambient mantle and plumes:Constraints from basalts, picrites, and komatiites. Geochemistry GeophysicsGeosystems 8(Q02006), doi: 10.1 029/2006GC00 1390.Herzberg, C. & O’Hara, M. J. (2002). Plume-associated ultramafic magmas ofPhanerozoic age. Journal ofPetrology 43(10), 1857-1883.Huang, S. & Frey, F. A. (2005). Trace element abundances of Mauna Kea basalt fromphase 2 of the Hawaii Scientific Drilling Project: Petrogenetic implications ofcorrelations with major element content and isotopic ratios. GeochemistryGeophysics Geosystems 4(6), 43, doi: 10.1 029/2002GC000322.Hussong, D. M., Wipperman, L. K. & Kroenke, L. W. (1979). The crustal structure of theOntong Java and Manihiki Oceanic Plateaus. Journal of Geophysical Research84(Bl 1), 6,003-6,0 10.Janney, P. E., Le Roex, A. P. & Carlson, R. W. (2005). Hafluium isotope and traceelement constraints on the nature of mantle heterogeneity beneath the CentralSouthwest Indian Ridge (l3°E to 47°E). Journal ofPetrology 46(12), 2427-2464,10.1 093/petrology/egiO6O.Jones, D. L., Silberling, N. J. & Hillhouse, J. (1977). Wrangellia; a displaced terrane innorthwestern North America. Canadian Journal ofEarth Sciences 14(11), 2565-2577.Kempton, P. D., Pearce, 3. A., Barry, T. L., Fitton, J. G., Langmuir, C. H. & Christie, D.M. (2002). Sr-Nd-Pb-Hf isotope results from ODP Leg 187: Evidence for mantledynamics of the Australian-Antarctic Discordance and origin of the Indian MORBsource. Geochemistry Geophysics Geosystems 3(12), 1074,doi: 10. 1029/2002GC000320.Kerr, A. C. (2003). Oceanic Plateaus. In: Rudnick, R. (ed.) The Crust. Treatise onGeochemistry Vol. 3. Holland, H. D. & Turekian, K. K. (eds.) Elsevier Science:Oxford,pp.537-565.Kerr, A. C. & Mahoney, J. 3. (2007). Oceanic plateaus: Problematic plumes, potentialparadigms. Chemical Geology 241, 332-353.Kerr, A. C., Marriner, G. F., Tarney, J., Nivia, A., Saunders, A. D., Thiriwall, M. F. &Sinton, A. C. (1997). Cretaceous basaltic terranes in Western Columbia:Elemental, chronological and Sr-Nd isotopic constraints on petrogenesis. JournalofPetrology 38(6), 677-702.Kerr, A. C., Tarney, J., Marriner, G. F., Klaver, G. T., Saunders, A. D. & Thiriwall, M. F.(1996). The geochemistry and petrogenesis of the Late-Cretaceous picrites andbasalts of Curacao, Netherlands Antilles; a remnant of an oceanic plateau.Contributions to Mineralogy and Petrology 124(1), 29-43.Kerr, A. C., Tarney, 3., Nivia, A., Marriner, G. F. & Saunders, A. D. (1998). The internalstructure of oceanic plateaus: inferences from obducted Cretaceous terranes inwestern Columbia and the Caribbean. Tectonophysics 292, 173-188.Kinzler, R. 3. (1997). Melting of mantle peridotite at pressures approaching the spinel togarnet transition: Application to mid-ocean ridge basalt petrogenesis. Journal ofGeophysical Research 102(B1), 853-874.Kinzler, R. J. & Grove, T. L. (1992). Primary magmas of mid-ocean ridge basalts 1.Experiments and methods. Journal ofGeophysical Research 97(B5), 6,885-6,906.81Korenaga, J. (2005). Why did not the Ontong Java Plateau form subaerially? Earth andPlanetary Science Letters 234, 385- 399.Kuniyoshi, S. (1972). Petrology of the Karmutsen (Volcanic) Group, northeasternVancouver Island, British Columbia. Unpublished Ph.D. dissertation, Universityof California, Los Angeles. 242pp.Lassiter, J. C., DePaolo, D. 3. & Mahoney, J. J. (1995). Geochemistry of the Wrangelliaflood basalt province: Implications for the role of continental and oceaniclithosphere in flood basalt genesis. Journal ofPetrology 36(4), 983-1009.Longhi, 3. (2002). Some phase equilibrium systematics of lherzolite melting: I.Geochemistry Geophysics Geosystems 3(3), 10.1029/2001 GC000204.MacDonald, G. A. & Katsura, T. (1964). Chemical composition of Hawaiian lavas.Journal ofPetrology 5, 82-133.Mahoney, J. J. (1987). An isotopic survey of Pacific oceanic plateaus: implications fortheir nature and origin. In: Keating, B. H., Fryer, P., Batiza, R. & Boehlert, G. W.(eds.) Seamounts, Islands, and Atolls. American Geophysical Union: Washington,D.C. Geophysical Monograph 43, pp. 207-220.Mahoney, 3., LeRoex, A. P., Peng, Z., Fisher, R. L. & Natland, 3. H. (1992).Southwestern Limits of Indian Ocean Ridge Mantle and the Origin of Low206Pb/204PbMid-Ocean Ridge Basalt: Isotope Systematics of the CentralSouthwest Indian Ridge (1 7°-50°E). Journal ofGeophysical Research 97(B 13),19771-19790.Mahoney, 3., Storey, M., Duncan, R., Spencer, K. & Pringle, M. (1993). Geochemistryand age of the Ontong Java Plateau. In: Pringle, M., Sager, W. & Sliter, W. (eds.)AGU Monograph on the MesozoicPacUIc:Geology, Tectonics, and Volcanism.77 AGU: Washington, pp. 233-261.Mahoney, J. J., Sinton, 3. M., Kurz, M. D., MacDougall, 3. D., Spencer, K. J. &Lugmair,G. W. (1994). Isotope and trace element characteristics of a super-fast spreadingridge: East Pacific rise, 1 3-23°S. Earth and Planetary Science Letters 121, 173-193.Massey, N. W. D. (l995a). Geology and mineral resources of the Alberni-NanaimoLakes sheet, Vancouver Island 92F/1W, 92F/2E, and part of 92F/7E. BritishColumbia Ministry ofEnergy, Mines and Petroleum Resources. Paper 1992-2,132 p.Massey, N. W. D. (1995b). Geology and mineral resources of the Cowichan Lake sheet,Vancouver Island 92C/16. British Columbia Ministry ofEnergy, Mines andPetroleum Resources. Paper 1992-3, 112p.Massey, N. W. D., MacJ.ntyre, D. G., Desjardins, P. J. & Cooney, R. T. (2005a).DigitalGeology Map of British Columbia: Tile NM9 Mid Coast, B.C. B.C. Ministry ofEnergy and Mines Geofile 2005-2.Massey, N. W. D., Maclntyre, D. G., Desjardins, P. 3. & Cooney, R. T.(2005b). DigitalGeology Map of British Columbia: Tile NM1O Southwest B.C. B.C. Ministry ofEnergy and Mines Geofile 2005-3.McDonough, W. F. & Sun, 5. (1995). The composition of the Earth. ChemicalGeology120, 223-253.Monger, 3. W. H. & Journeay, J. M. (1994). Basement geology and tectonic evolution ofthe Vancouver region. In: Monger, J. W. H. (ed.) Geology andGeological82Hazards ofthe Vancouver Region, Southwestern British Columbia. GeologicalSurvey of Canada Bulletin, 481,PP.3-25.Muller, J. E. (1977). Geology of Vancouver Island. Geological Survey ofCanada OpenFile Map 463.Muller, J. E. (1980). The Paleozoic Sicker Group of Vancouver Island, British Columbia.Geological Survey ofCanada. Paper 79-30, 22p.Muller, J. E., Northcote, K. E. & Carlisle, D. (1974). Geology and mineral deposits ofAlert Bay - Cape Scott map area, Vancouver Island, British Columbia. GeologicalSurvey ofCanada, Paper 74-8, 77.Neal, C. R., Mahoney, J. J., Kroenke, L. W., Duncan, R. A. & Petterson, M. G. (1997).The Ontong-Java Plateau. In: Mahoney, J. J. & Coffin, M. F. (eds.) Large IgneousProvinces: Continental, Oceanic, and Planetary Flood Volcanism. GeophysicalMonograph 100 AGU: Washington, pp. 183-2 16.Niu, Y., Collerson, K. D., Batiza, R., Wendt, 3. I. & Regelous, M. (1999). Origin ofenriched-typed mid-ocean ridge basalt at ridges far from mantle plumes: The EastPacific Rise at 1 1°20’N. Journal of Geophysical Research 104, 7067-7088.Nixon, G. T., Hammack, J. L., Koyanagi, V. M., Payie, G. J., Haggart, J. W., Orchard, M.J., Tozer, T., Friedman, R. M., D.A., A., Palf’, J. & Cordey, F. (2006a). Geologyof the Quatsino — Port McNeill area, northern Vancouver Island. BC Ministry ofEnergy, Mines and Petroleum Resources. Geoscience Map 2006-2, scale 1:50000.Nixon, G. T., Kelman, M. C., Stevenson, D., Stokes, L. A. & Johnston, K. A. (2006b).Preliminary geology of the Nimpkish map area (NTS 092L/07), northernVancouver Island, British Columbia. In: Grant, B. & Newell, J. M. (eds.)Geological Fieldwork 2005. Paper 2006-1 British Columbia, Ministry of Energy,Mines and Petroleum Resources,pp.135-152.Nixon, G. T., Laroque, J., Pals, A., Styan, 3., Greene, A. R. & Scoates, J. S. (2008). High-Mg lavas in the Karmutsen flood basalts, northern Vancouver Island (NTS 092L):Stratigraphic setting and metallogenic significance. In: Grant, B. (ed.) GeologicalFieldwork 2007. Paper 2008-1 British Columbia, Ministry of Energy, Mines andPetroleum Resources,PP.175-190.Nixon, G. T. & Off, A. J. (2007). Recent revisions to the Early Mesozoic stratigraphy ofNorthern Vancouver Island (NTS 1021; 092L) and metallogenic implications,British Columbia. In: Grant, B. (ed.) Geological Fieldwork 2006. Paper 2007-1British Columbia, Ministry of Energy, Mines and Petroleum Resources,pp.163-177.Nobre Silva, I. G., Weis, D., Barling, J. & Scoates, 3. S. (submitted). Leachingsystematics for the determination of high-precision Pb isotope compositions ofocean island basalts. Geochemistry, Geophysics, Geosystems 2007GC001 891.Nowell, G. M., Kempton, P. D., Noble, S. R., Fitton, 3. G., Saunders, A., Mahoney, 3. 3.& Taylor, R. N. (1998). High precision Hf isotope measurements of MORE andOIB by thermal ionisation mass spectrometry: insights into the depleted mantle.Chemical Geology 149, 211-233.Parrish, R. R. & McNicoll, V. 3. (1992). U-Pb age determinations from the southernVancouver Island area, British Columbia. Geological Survey ofCanada.Radiogenic Age and Isotopic Studies: Report 5 Paper 9 1-2, 79-86p.83Pretorius, W., Weis, D., Williams, G., Hanano, D., Kieffer, B. & Scoates, J. S. (2006).Complete trace elemental characterization of granitoid (USGSG-2,GSP-2)reference materials by high resolution inductively coupled plasma-massspectrometry. Geostandards and Geoanalytical Research 30(1), 39-54.Regelous, M., Niua, Y., Wendt, J. I., Batizab, R., Greiga, A. & Collerson, K. D. (1999).Variations in the geochemistry of magmatism on the East Pacific Rise at 1 0°30’Nsince 800 ka. Earth and Planetary Science Letters 168, 45-63.Révillon, S., Chauvel, C., Arndt, N. T., Pik, R., Martineau, F., Fourcade, S. & Marty, B.(2002). Heterogeneity of the Caribbean plateau mantle source: Sr, 0 and Heisotopic compositions of olivine and clinopyroxene from Gorgona Island. Earthand Planetary Science Letters 205(1-2), 91.Rhodes, J. M. (1996). Geochemical stratigraphy of lava flows samples by the HawaiiScientific Drilling Project. Journal ofGeophysical Research 101(B5), 11,729-11,746.Rhodes, J. M. & Vollinger, M. J. (2004). Composition of basaltic lavas sampled byphase-2 of the Hawaii Scientific Drilling Project: Geochemical stratigraphy andmagma types. Geochemistry Geophysics Geosystems 5(3),doi: 10.1 029/2002GC000434.Richards, M. A., Jones, D. L., Duncan, R. A. & DePaolo, D. J. (1991). A mantle plumeinitiation model for the Wrangellia flood basalt and other oceanic plateaus.Science 254, 263-267.Salters, V. J. & White, W. (1998). Hf isotope constraints on mantle evolution. ChemicalGeology 145, 447-460.Salters, V. J. M. (1996). The generation of mid-ocean ridge basalts from the Hf and Ndisotope perspective. Earth and Planetary Science Letters 141, 109-123.Salters, V. J. M. & Stracke, A. (2004). Composition of the depleted mantle.Geochemistry Geophysics Geosystems 5(Q05B07), doi: 10.1 029/2003GC000597.Sano, T. & Yamashita, S. (2004). Experimental petrology of basement lavas from OceanDrilling Program Leg 192: implications for differentiation processes in OntongJava Plateau magmas. In: Fitton, J. G., Mahoney, J. J., Wallace, P. J. & Saunders,A. D. (eds.) Origin and Evolution ofthe Ontong Java Plateau. SpecialPublications 229 Geological Society,pp.185-218.Saunders, A. D. (2005). Large igneous provinces: Origin and environmentalconsequences. Elements 1, 259-263.Scherer, E., Muiilcer, C. & Mezger, K. (2001). Calibration of the lutetium-hafluium clock.Science 293(5530), 683-687, 10.1126/science. 1061372.Sluggett, C. L. (2003). Uranium-lead age and geochemical constraints on Paleozoic andEarly Mesozoic magmatism in Wrangellia Terrane, Saltspring Island, BritishColumbia. Unpublished B.Sc. thesis, University of British Columbia. 84pp.Surdam, R. C. (1967). Low-grade metamorphism of the Karmutsen Group. UnpublishedPh.D. dissertation, University of California, Los Angeles. 288pp.Sutherland-Brown, A., Yorath, C. J., Anderson, R. G. & Dom, K. (1986). Geologicalmaps of southern Vancouver Island, LITHOPROBE I 92C/10, 11, 14, 16, 92F/l,2, 7, 8. Geological Survey ofCanada. Open File 1272.Tejada, M. L. G., Mahoney, J. J., Castillo, P. R., Ingle, S. P., Sheth, H. C. & Weis, D.(2004). Pin-pricking the elephant: evidence on the origin of the Ontong Java84Plateau from Pb-Sr-Hf-Nd isotopic characteristics of ODP Leg 192 basalts. In:Fitton, J. G., Mahoney, J. J., Wallace, P. J. & Saunders, A. D. (eds.) Origin andEvolution ofthe Ontong Java Plateau. Geological Society of London, SpecialPublication 229,pp.133-150.Vervoort, J. D., Patchett, P., Blichert-Toft, 3. & Albarède, F. (1999). Relationshipsbetween Lu-Hf and Sm-Nd isotopic systems in the global sedimentary system.Earth and Planetary Science Letters 168, 79-99.Walter, M. 3. (1998). Melting of garnet peridotite and the origin of komatiite and depletedlithosphere. Journal ofPetrology 39(1), 29-60.Weis, D., Kieffer, B., Maerschalk, C., Barling, J., de Jong, 3., Williams, G. A., Hanano,D., Mattielli, N., Scoates, 3. 5., Goolaerts, A., Friedman, R. A. & Mahoney, J. B.(2006). High-precision isotopic characterization of USGS reference materials byTIMS and MC-ICP-MS. Geochemistry Geophysics Geosystems 7(Q08006),doi: 10.1 029/2006GC00 1283.Weis, D., Kieffer, B., Hanano, D., Silva, I. N., Barling, J., Pretorius, W., Maerschalk, C.& Mattielli, N. (2007). Hf isotope compositions of U.S. Geological Surveyreference materials. Geochemistry Geophysics Geosystems 8(Q06006),doi: 10.1 029/2006GC00 1473.Weis, D., Kieffer, B., Maerschalk, C., Barling, J., de Jong, J., Williams, G. A., Hanano,D., Mattielli, N., Scoates, 3. 5., Goolaerts, A., Friedman, R. A. & Mahoney, J. B.(2006). High-precision isotopic characterization ofUSGS reference materials byTIMS and MC-ICP-MS. Geochemistry Geophysics Geosystems 7(Q08006),doi: 10.1 029/2006GC00 1283.Wheeler, 3. 0. & McFeely, P. (1991). Tectonic assemblage map of the CanadianCordillera and adjacent part of the United States of America. Geological SurveyofCanada Map 1712A.White, W. M., Albarêde, F. & Télouk, P. (2000). High-precision analysis of Pb isotoperatios by multi-collector ICP-MS. Chemical Geology 167, 257-270.Yorath, C. J., Sutherland Brown, A. & Massey, N. W. D. (1999). LITHOPROBE,southern Vancouver Island, British Columbia. Geological Survey ofCanada.LITHOPROBE, southern Vancouver Island, British Columbia Bulletin 498, 145p.Zou, H. (1998). Trace element fractionation during modal and non-model dynamicmelting and open-system melting: A mathematical treatment. Geochimica etCosmochimicaActa 62(11), 1937-1945.Zou, H. & Reid, M. R. (2001). Quantitative modeling of trace element fractionationduring incongruent dynamic melting. Geochimica et Cosmochimica Acta 65(1),153-162.85CHAPTER 3Wrangellia Flood Basalts in Alaska:A Record of Plume-Lithosphere Interaction in aLate Triassic Accreted Oceanic Plateau‘A version of this chapter has been submitted for publication.86INTRODUCTIONOceanic plateaus and continental flood basalts (CFBs) are produced from thelargest melting events recorded on Earth. Oceanic plateaus and CFBs are transient largeigneous provinces (LIPs) that form from unusually high magmatic fluxes over severalmillion years or less (Saunders, 2005). The presence of transient LIPs in a variety oftectonic settings attests to large thermal anomalies that are not directly attributable toseafloor spreading processes. A longstanding controversy in many transient LIPsworldwide is the role of the mantle lithosphere in generation of the basaltic magmas.CFBs have compositions that indicate involvement of subcontinental lithospheric mantleand continental crust (e.g. Peate & Hawkesworth, 1996). Compositional evidence ofplume-lithosphere interaction in oceanic plateaus, however, remains elusive becauseoceanic plateaus are less accessible. Oceanic plateaus are enormous volcanic edifices (2-4km high) that are isolated from continents and form upon mid-ocean ridges, extinct arcs,detached or submerged continental fragments, or intraplate settings (Coffin et a!., 2006).The basalts that form oceanic plateaus have a better chance of avoiding interaction withcontinental lithosphere than basalts erupted along the margins or in the interiors ofcontinents (Kerr & Mahoney, 2007).A significant issue in the geochemistry of flood basalt provinces has been theorigin of high- and low-titanium basalts within the flood basalt stratigraphy (e.g. Arndt eta?., 1993). Numerous flood basalt provinces have been found to possess two or moredistinguishable groups of basalts based on titanium contents [e.g. Siberia (Wooden et a?.,1993); Emeishan (Xu eta?., 2001); Karoo (Cox eta?., 1967); Ferrar (Hornig, 1993);Paraná-Etendeka (Peate, 1997); Deccan (Melluso et a?., 1995); Ethiopia (Pik et a?.,1998); Columbia River Basalts (looper & Hawkesworth, 1993)]. In several of theseprovinces, the high- and low-titanium basalts are geographically distributed, and inseveral provinces these different lava types have a distinct stratigraphic distribution.However, all of these LIPs formed upon continental crust and are thought to haveinvolved interaction with metasomatized lithospheric mantle or continental crust duringparts of their eruptive history.The Wrangellia flood basalts in Alaska erupted in the eastern Panthalassic Oceanover <5 Myr in the Middle to Late Triassic, with accretion to western North America87occurring in the Late Jurassic or Early Cretaceous (Jones et aL, 1977). The Wrangelliaflood basalts form thick successions of flood basalts bounded by marine sediments thatextend over widespread areas of Alaska, Yukon, and British Columbia (>2300 km inlength). In south-central Alaska, parts of the complete flood basalt stratigraphy overlieLate Paleozoic oceanic arc crust and marine sediments and are overlain by Late Triassiclimestone. Most of the 3.5-4 km of flood basalts in Alaska erupted subaerially, however,in areas of Alaska there are pillowed and volcaniclastic basalts in the lower part of thevolcanic stratigraphy.The Wrangellia flood basalts in Alaska provide an exceptional opportunity toexamine the volcanic stratigraphy of an accreted oceanic plateau. While Ocean DrillingProgram (ODP) and Deep Sea Drilling Program (DSDP) legs have drilled extant oceanicplateaus in the ocean basins, the vast majority of the stratigraphic sequence of oceanicplateaus remains generally unsampled and undescribed. This study examines volcanicstratigraphy, petrography, and geochemistry of flood basalts in the Wrangell Mountainsand Alaska Range in south-central Alaska to determine the source and origin of high- andlow-titanium basalts in the accreted Wrangellia oceanic plateau and the role of the preexisting Paleozoic oceanic arc lithosphere.A mantle plume origin was proposed for Wrangellia flood basalts by Richards etaL (1991) based on the large volume of flood basalts erupted in a short duration, theabsence of evidence of rifting, and evidence of uplift prior to eruption of the floodbasalts. The only previous modem analytical study of the Wrangellia flood basalts inAlaska involved major- and trace-element chemistry, and Sr, Nd, and Pb isotopicanalyses, of 9 basalts from the Wrangell Mountains (Lassiter et al., 1995). This presentstudy is part of a larger research project on the origin and evolution of the TriassicWrangellia flood basalts in British Columbia, Yukon, and Alaska (Greene et a!., 2008,submitted-b). The generation of compositionally distinct basalts in part of the Wrangelliaoceanic plateau has implications for the interaction of mantle plumes and oceanic mantlemodified by subduction.88GEOLOGIC SETTINGWrangeffia in AlaskaThe Wrangellia Terrane, or Wrangellia, was defined by Jones et a!. (1977) as a setof fault-bound crustal blocks with similar stratigraphy along the margin of western NorthAmerica (Fig. 3.1). The Wrangellia flood basalts have been mapped as the NikolaiFormation in Alaska and Yukon (Northern Wrangellia) and the Karmutsen Formation onVancouver and Queen Charlotte Islands (Southern Wrangellia). Three key aspects ofWrangellia led Jones and co-workers (1977) to suggest these crustal blocks formed aspart of a once-contiguous terrane: thick sections of tholeiitic flood basalts directly overlieshale with Middle to Late Ladinian Daonella; similar-aged Late Triassic limestonesoverlie the flood basalts; and paleomagnetic evidence indicated that eruption of the floodbasalts occurred at low latitude. Wrangellia may have joined with parts of the AlexanderTerrane, primarily in southeast Alaska, as early as the Late Pennsylvanian (Gardner et aL,1988) and may have been in close proximity to the Peninsular Terrane of southern Alaskaby the Late Triassic (Rioux et al., 2007).Wrangellia extends 450 km in an arcuate belt in the Wrangell Mountains, AlaskaRange, and Tailceetna Mountains in southern Alaska (Fig. 3.1). The northwest margin ofWrangellia is one of the most prominent geophysical features in south-central Alaskaand is exposed along the Talkeetna Suture Zone (Glen eta!., 2007). The suture betweenWrangellia and transitional crust to the northwest is well-defined geophysically by aseries of narrow gravity and magnetic highs along the Talkeetna Suture Zone, betweendense, strongly magnetic Wrangellia crust and less dense, weakly magnetic crust beneathflysch basins to the northwest (Glen et a!., 2007). This area lies directly in the axis of themajor orocline of southern Alaska, where structures curve from northwest- to northeast-trending (e.g. Plafker eta!., 1994; Fig. 3.1). The Wrangellia terrane is bounded by theDenali Fault to the northeast and extends more than 300 km to the southeast in Yukonwhere Wrangellia stratigraphy is very similar to Alaska (Fig. 3.1).Wrangell MountainsWrangellia stratigraphy is well-exposed in a northwest-trending belt extendingl00 km along the southern flank of the Wrangell Mountains in Wrangell-St. Elias89Figure 3.1 Simplified map of south-central Alaska showing the distribution of the Nikolai Formation (black),derived from the GIS-based digital map compilations of Wilson et a!. (1998, 2005) and Schmidt (pers. comm,2006). The three main areas that were studied are outlined with boxes and indicated in the legend. The insetshows the extent of the Wrangellia flood basalts (green) in Alaska, Yukon, and British Columbia. The red linesare faults.90National Park (Fig. 3.2). The Nikolai Formation disconformably overlies the SkolaiGroup, which comprises Pennsylvanian to Early Permian volcanic arc sequences andmarine sediments of the Station Creek and Hasen Creek Formations, respectively (Smithand MacKevett, 1970; Fig. 3.2). In most areas, Nikolai basalts unconfomably overlieparts of the Hasen Creek Formation called the Golden Horn Limestone Lentil (<250 mthick), which forms prominent yellow- and red-stained cliffs of bioclastic Early Permianlimestone. Isolated lenses of Daonella-bearing Middle Triassic argillite (<100 m thick)lie between the top of the Golden Horn Limestone and the base of the Nikolai Formation(Smith and MacKevett, 1970). The Skolai Group is intruded by mafic and ultramaficintrusive bodies related to the Nikolai basalts. The Nikolai basalts coverl057 km2(4.2% of all Wrangellia flood basalts from Vancouver Island to central Alaska) within theMcCarthy and Nabesna Quadrangles in Wrangell-St. Elias National Park and areapproximately 3.5-4 km in total thickness. A cumulative thickness of over 3.5 km ofmarine sedimentary rocks, ranging in age from Late Triassic to Late Jurassic, overlies theNikolai Formation in the Wrangell Mountains (MacKevett et al., 1997). Volcanicsuccessions of the Miocene to Holocene Wrangell volcanic field unconformablyoverlieJurassic and Cretaceous sedimentary sequences (MacKevett, 1978; Richter et al., 1990).Eastern Alaska RangeThe Nikolai Formation in the eastern Alaska Range and small areas of theTailceetna Mountains covers 666 km2 (2.6% of all Wrangellia flood basalts) mostly in theMount Hayes and Healy Quadrangles, and is 3.5-4 km thick in the Amphitheater andClearwater Mountains (Fig. 3.1). Volcanic and marine sedimentary sequences similar tothe Late Paleozoic successions in the Wrangell Mountains underlie the Nikolaibasalts inthe Alaska Range (Nokleberg eta?., 1994). In the Amphitheater Mountains, amajorfeeder system for the Nikolai Formation stratigraphically underlies flood basaltstratigraphy within a well-preserved synform that contains exposures of mafic andultramafic intrusive units. These are the most significant occurrence of plutonicrocksassociated with flood basalts within Wrangellia. The upper part of thevolcanicstratigraphy contains interbedded volcanic and sedimentary horizons (argillite and91Figure 3.2 Geologic map and stratigraphy of the Wrangell Mountains, Alaska (location shown in Figure 1).(a) Stratigraphic column depicts Late Paleozoic to Jurassic units on the south side of the Wrangell Mountains,derived from Smith and MacKevett (1970) and MacKevett (1978). (b) Simplified map showing thedistribution of the Nikolai Formation (green) in the Wrangell Mountains, derived from the GIS-based digitalmap compilation of Wilson et al. (2005). The four areas of field study are outlined with labelled boxes. Thered lines are faults.92limestone) that give way to fine-grained sedimentary strata, which are poorly exposed inthe Amphitheater Mountains.Age of the Nikolai FormationBiostratigraphy and geochronology provide bounds on the age and duration ofemplacement of the Nikolai basalts. Fossil assemblages in finely laminated shaleimmediately beneath the Nikolai basalts in the Wrangell Mountains indicate a Middle toLate Ladinian age (McRoberts, pers. comm. 2007; Jones etal., 1977) and fossils inlimestone disconformably overlying the Nikolai Formation are Late Carnian to EarlyNorian (Armstrong & MacKevett, 1977; Plaficer et al., 1989). Five 40Ar/39Arplateau agesfor hornblende and biotite from intrusive rocks in the Amphitheater Mountains in theAlaska Range interpreted to be comagmatic with Nikolai basalts indicate formation ofthese rocks at 23 1-225 Ma (Bittenbender et al., 2007; Schmidt & Rogers, 2007). ThreeNikolai basalt samples from the Wrangell Mountains yielded40Ar/39Arstep-heating agesof 228.3 ± 5.2, 232.8 ± 11.5, and 232.4 ± 11.9 Ma (Lassiter, 1995).VOLCANIC STRATIGRAPHY AND PETROGRAPHYField studies in Alaska focussed in three general areas where parts of the entireflood basalt stratigraphy are well-exposed: the southern flank of the Wrangell Mountains,and the Amphitheater and Clearwater Mountains in the southern part of the easternAlaska Range (Fig. 3.1).In the Wrangell Mountains, field studies focussed in 4 areas: Skolai Creek,Glacier Creek, Hidden Lake Creek, and Nugget Creek (Fig. 3.2). The base of the NikolaiFormation is well-exposed near Skolai Creek, the type section for the underlyingPaleozoic Skolai Group. At Skolai Creek, the base of the Nikolai Formation is basaltflow-conglomerate, pillow breccia, and minor pillow basalt. (Figs 3.2 and 3.3). Pebblesand cobbles comprise >30% of the basal flow-conglomerate and all the clasts appear tobe derived from the underlying Skolai Group (Fig. 3.3). Middle to upper portions of theflood basalt stratigraphy are well-exposed above Glacier Creek, where massive maroon-and green-colored flows form monotonous sequences with amygdaloidal-rich horizonsand no discernible erosional surfaces or sediments between flows (Fig. 3.4). The top of93Figure 3.3 Photographs of the base of the Nikolai Formation in the Wrangell Mountains, Alaska. (a)Westward-dipping Paleozoic arc volcanic rocks of the Station Creek Formation overlain by Early Permianshale and limestone (Phc-Hasen Creek Formation; Pgh-Golden Horn Limestone Lentil), isolated lenses ofMiddle Triassic ‘Daonella-beds’ (TRd), basalt flow-conglomerate with local pillows, and massive subaerialflows on the north side of Skolai Creek. Photograph by Ed MacKevett, Jr. (b) Close-up photograph of area bin photo a. (c) Close-up photograph of area c in photo b showing basal flow-conglomerate with clasts ofrounded cobbles of white limestone (<20 cm) derived from Golden Horn Limestone Lentil and red basalt(<40 cm) from Station Creek Formation. Pen (14 cm) in middle ofphoto for scale.S94Figure 3.4 Photograph of —4000 m of continuous flood basalt stratigraphy at the top of the NikolaiFormation along Glacier Creek in the Wrangell Mountains, Alaska. The yellow line marks the contactbetween Nikolai basalts and the overlying Chitistone Limestone.95the flood basalts are best exposed around Hidden Lake Creek where a sharp contactbetween Nikolai basalts and overlying Chitistone Limestone is mostly a smooth surfacewith minimal evidence of weathering (Fig. 3.5; Armstrong & MacKevett, 1982). Severaloccurrences of a thin zone (<1 m) of highly-oxidized subangular cobble-sized clasts ofNikolai basalt and minor thinly-bedded siltstone occur along the contact (Fig. 3.5).In the Amphitheater Mountains, fieldwork concentrated in five areas: Glacier GapLake, Landmark Gap Lake, Tangle Lakes (West), Sugarloaf Mountain, and Rainy Creek(Fig. 3.6). The Amphitheater Mountains are formed of exceptionally well-preserved floodbasalt sequences flanked by associated underlying mafic and ultramafic plutonic rocks.South of the Eureka Creek fault is a broad synform with approximately 3.5 km of basalticflows with basal sill complexes exposed along the outer margins (Fig. 3.6). The lower—‘500 m of volcanic stratigraphy are submarine flows and subaerial flows comprise mostof the remainder of the volcanic stratigraphy. Within the synform, several north-south-trending, U-shaped glacial valleys (e.g. Lower Tangle Lakes) provide excellent cross-sectional exposures of sediment-sill complexes and the base of the flood basaltstratigraphy (Figs 3.6 and 3.7). The lowest part of over 1000 m of continuous volcanicstratigraphy consists of unfossiliferous shale and siliceous argillite (<4 m thick)interbedded with massive mafic sills (2-30 mthick), in turn overlain by pillow basalt(Fig. 3.7). The basal pillowed flow is 13 m thick and pillows are typically <1 m indiameter with sediment between pillows along the base of the flow. Sills interbeddedwith thinly bedded basaltic sandstone and minor hyaloclastite also occur slightly higherin the stratigraphy, within the submarine stratigraphy (Figs 3.6 and 3.7). Middle andupper sections of the Nikolai Formation are massive amygdaloidal flows (mostly<15 mthick) with no discernible erosional surfaces or sediments between flows.A small segment of Wrangellia consisting of a heterogeneous assemblage ofmafic and ultramafic plutonic and volcanic rocks forms a wedge between the BroxsonGulch Thrust and the Eureka Creek fault in the northern part of the AmphitheaterMountains (Fig. 3.6). A complex steeply-dipping sequence of picritic tuff andvolcaniclastic rocks, mafic and ultramafic intrusives and dikes, and limestone occurswithin several ridges near Rainy Creek. These units have distinct lithologic characterfrom the volcanic stratigraphy of the Nikolai Formation south of the Eureka Creek Fault96:Figure 3.5 Photographs showing the top of the Nikolai Formation in the Wrangell Mountains, Alaska. (a)Chitistone Limestone overlying the Nikolai Formation above Hidden Creek in the Wrangell Mountains.Faulting has offset the contact. (b) Close-up photograph of uppermost Nikolai flow and base of theChitistone Limestone from where photo a was taken. (c) Cobbles<10 cm long along the contact betweenthe Chitistone Limestone and Nikolai Formation. The oxidized cobbles are subangular, closely packed,aligned along their long axis, and are glomeroporphyritic basalt identical to the uppermost flows of theNikolai Formation. Sledgehammer handle (4 cm wide) for scale.97Amphitheater MountainsFigure 3.6 Simplified geologic map and stratigraphy of the Amphitheater Mountains, Alaska (locationshown in Figure 3.1). (a) Stratigraphic column with sample lithologies and estimated vesicularity for floodbasalts from the lower part of the volcanic stratigraphy, derived from three traverses marked by red lines in b.Vesicularity estimated visually from thin-sections. (b) Generalized geology of the Nikolai Formation andrelated plutonic rocks in the Amphitheater Mountains. Five main field areas are outlined with numberedboxes (see legend). Map derived Nokleberg et a!. (1992) and digital compilation of Wilson et aL (1998). (c)Schematic cross-section of Amphitheater Mountains from A toA?in panel b, adapted from Nokleberg et aL(1985).98neater Mountains, Alaska RangeFigure 3.7 Photographs of the base of the Nikolai Formation in the Amphitheater Mountains, east-centralAlaska Range (Tangle Lakes, West), Alaska. (a) Basal sill and sediments beneath submarine flows. Lettersdenote locations of other photographs. (b) Fissile shale (—4 m thick) and a mafic sill from the lowermostexposure of shale. (c) Pillow basalt with shale between pillows lying directly above shale similar tophotograph b. (d) Pillow basalt (pillow tubes are <1 m diameter in cross-section) in the lowermost flow (13m thick) in the Tangle section. Photograph c is from the base of this flow. Sledgehammer (80 cm long) forscale. (e) Sequence of at least 4 massive sills (<8 m thick) interbedded with fine-grained tuff (<2 m thick).Tuff layers interbedded with sills contain plagioclase crystals (<0.5 mm), curvilinear shards, and local areasof volcanic breccia containing basaltic clasts with abundant small acicular plagioclase (<0.5 mm).99within the broad synform. A small suite of eight samples, including several highly alteredolivine-bearing picritic tuffs, were collected for comparison to Nikolai basalts within thesynform and are referred to as Rainy Creek picrites.In the Clearwater Mountains, a small mountain range 40 km west of theAmphitheater Mountains, the lowest level of exposure is pelagic sedimentary sequencesinterbedded with mafic sills, very similar to units in the Amphitheater Mountains (Fig.3.8). The lowest flows of the Nikolai Formation are pillow basalt directly overlying thinbeds of shale and argillite (<3 m thick) with sediment commonly filling interpillow voidsin the lowermost pillows. Picritic pillow lavas have been found in the submarinestratigraphy in the Clearwater Mountains. Similar to the Amphitheater Mountains, thelower <400 m of the volcanic stratigraphy is submarine flows and the remainder of thestratigraphy is primarily subaerial flows. Upper parts of the volcanic stratigraphy containsubaerial flows (or sills) with columnar jointing, minor occurrences of tuff and volcanicbreccia, and limestone and argillite lenses interbedded with flows are overlain by finegrained sediments with diagnostic index fossils (bivalve Halobia and ammonoidTropite; Smith, 1981).A total of 111 samples of the Nikolai Formation and several Paleozoic, LateMesozoic, and Cenozoic volcanic and sedimentary rocks were collected for petrographyand geochemical analysis. Fifty-three of these samples were selected for geochemistrybased on the visual degree of alteration (Table 3.1; 1-least altered, 3-intensely altered)and are grouped into high- and low-titanium basalts, sills, and picrites based ongeochemistry.The high- and low-titanium basalts have similar petrographic textures with simplemineralogy and aphyric or glomeroporphyritic texture (Fig. 3.9; Table 3.1). Half of the26 high-titanium basalts are glomeroporphyritic and about half of the low-titaniumbasalts are aphyric (Table 3.1). Phenocrysts and glomerocrysts in the Nikolai Formationare almost exclusively plagioclase and the primary minerals in the groundmass areplagioclase, clinopyroxene, and Fe-Ti oxide, and olivine is rarely present (Fig. 3.9; Table3.1). High-titanium basalts have a higher proportion of Fe-Ti oxide than low-titaniumbasalts (Table 3.1). Several of the uppermost flows in the Wrangell Mountains areplagioclase-rich (>50% plagioclase laths —1mm long with —5:1 aspect ratio) with100Figure 3.8 Geologic map and stratigraphy of the Clearwater Mountains, Alaska. (a) Stratigraphic columnwith sample lithologies and estimated vesicularity for flood basalts from the lower part of the stratigraphy,derived from three traverses and maps of Smith (1973) and Silberling et a!. (1981). Vesicularity estimatedvisually from thin-sections. (b) Generalized geology of the Nikolai Formation and related plutonic rocks inthe Clearwater Mountains, adapted from Silberling et a!. (1981). Sample locations are shown by red dots.Location of map shown in Figure 3.1. (c) Schematic cross-section of Clearwater Mountains from A to A’ inpanel b, adapted from Silberling eta!. (1981).101Table 3.1 Summaiyofpefrographic characteristics and phenoc,yst proportions of Nikolai basalts in AlaskaSample’AreabFlow’GroupdTexture’5710A2 WM FLO high-Ti glomero5801A2 CL PLO high-Ti glomero5806A3 GG PLO high-Ti aphyric, recrystallized5708A2 WM PLO high-Ti trachylic, glomero5801A9 CL PIL high-Ti aphyric, intersertal5725A2 TA SIL?? high-Ti sill glomero5716A2 WM PLO high-Ti intergranular, aphyric5806A5 GG PLO high-Ti aphyric, intersertal5716A3 WM PLO high-Ti plag-phyric, intersertal5707A3 WM PLO high-Ti intergranular, intersertal5710A3 WM PLO high-Ti glomero, intergranular5719A6 WM PLO high-Ti glomero, intergranular5719A5 WM PLO high-Ti glomero, intergranular5801A5 CL PLO high-Ti glomero, intergranular5806A6 GG PLO high-TI aphyric, intersertal5712A2 WM PLO high-Ti glomero5726A1 TA PIL high-Ti aphyric5726A6 TA PIL high-Ti glomero, intersertal5725A4 TA SIL?? high-Ti sill aphylic, ophimottled5726A3 TA PLO high-Ti intersertal5810AtO TA PIL high-Ti aphyiic, Intersertal57’15A1 WM PLO high-Ti trachylic, glomero5714A1 WM PLO high-Ti trachylic, glomero5716A1 WM PLO high-Ti plag-phyric5726A2 TA PLO high-Ti aphyric, intersertal5714A3 WM PLO high-Ti intergranular, glomero5808A3 RC TUP RCPIC tuffaceous5808A8 RC SIL RC tuffaceous5808A2 RC TUP RCPIC tuffaceous5808A1 RC TUP RC tutfaceous5808A6 RC DIK RC interaranular20 3 plag glcr <3 mm20 1 abundant ox10 3 abundant ox, few relict phenos3 15 2 abundant ox, plag glcr <2 mm3 aphyric15 10 2 plag glcr <2mm, cpx <0.5mm10 2 plag laths <1.5 mm, plag-rich, secondary mm15 3 abundantox,aphyric30 3 abundant plag phenos and glcr <4 mm3 10 3 fewplag<2mm10 15 3 plag glcr<3 mm, plag laths <1 mm, ox-rich10 15 2 abundant ox, plag glcr <3 mm, plag-rich10 15 1 abundant ox, plag glcr <3 mm, plag-rich15 20 2 abundant ox, plag glcr <2 mm3 plag <0.5 mm, cpx <0.5 mm5 15 2 plag glcr <3 mm, plag laths <0.5 mm2 vesicles 3%, very f.g., I plag glcr <3 mm20 2 plag glcr <2 mm, plag needles <0.5 mm7 2 plag <0.5 mm, cpx <0.5 mm1 10 3 plag phenos <2 mm5 1 plag <0.5 mm, cpx <0.5 mm20 10 1 plag glcr <3 mm, aligned p1 <0.5 mm10 10 2 plag glcr <3mm, aligned plag laths <0.5mm5 5 3 plag <6 mm, very c.g., plag laths <1 mm3 2 vesicles <0.5 mm, plag needles <0.5mm10 3 3 c.g., abundant sec mi plag glcr <3 mmvnl ôi Plan Cnx Os Alferatlnna Notch5715A5 WM PLO low-Ti relict glomero5801A8 CL PLO low-Ti intergranular, aphyric5802A5 CL PIL low-Ti intergranular, ophimotfied5731A5 CL PIL low-Ti intergranular, ophimottied5802A6 CL SIL low-Ti intergranular5810A4 TA SIL low-Ti sill glomero, intersertal5802A1 CL PIL low-Ti interaertal5731A6 CL PIL low-Ti intergranular, intersertal5810A6 TA PIL low-Ti glomero, intersertal5727A3 TA SIL low-Ti sill subophitic5810A1 TA PIL low-Ti aphyric, intersertal5727A5 TA PIL low-Ti aphyric, intersertal5727A7 TA PIL low-Ti aphyric, intersertal5810A2 TA SIL low-Ti sill aphyiic, intersertal5727A2 TA SIL low-Ti sill intergranular5731A3 CL PIL low-Ti intergranular, intersertal5802A3 CL PIL low-Ti aphyric, intersertal5727A6 TA SIL low-Ti sill subophilic, ophimottied5802A2 CL PIL low-Ti variolitic5731A4 CL PIL low-Ti subophitic, ophimotfied5811A1 TA GAB low-Ti sill intergranular571 SAl WM BRE basal trachytic5802A4 CL PIL CWPIC spherulitic35 35 2215 1 125 12 2310 15 3 321 22 21 25 32220 1 231 315 22235 33325230altered, abundant sec mm, plag glcr <3 mmblocky ox <0.5 mmaphyricaphyricplag laths <1 mm, cpx <1.5mmplag glcr <3 mmplag needles <0.5 mm, cpx <1.5mmplag <0.5 mm10% vesicles, plag glcr <3 mmplag <I mm, cpx <I mm, ox <0.5mm15% vesicles, tiny plag needles <0.3 mmplag needles <0.5 mmvesicular, plag needles <0.5 mm, cpx<0.5 mmf.g., plag needles <0.5mmplag laths <1 mm, cpx <1 mmfew ol<1.5 mmfew oI<1 mmcpx <2 mm enc plag <1 mmhighly altered, glomero, aligned plagcpx <1 mm enc tiny plag needlescpx <2 mm filling Intersticescalcite and qtz <3 mm, plag needles <1 mm01 <2 mm, swfl plag <1 mmblocky ox <0.5 mmelonoate hbl (<1.5 mmt with atz cement‘Sample number: last digit year, month, day, initial, sample station (except 93G171). ‘WM, Wrangell Mountains; TA, Tangle Lake; GG, Glacier Gap Lake; CL,Clearwater Mountains; RC, Rainy Creek; °PIL, pillow; BRE, breccia; FLO, flow; SIL,sill; GAB, gabbro; TUP, tuft; DIK, dike.dbasalbasal flow-conglomerate;RC, Rainy Creek; RCPIC, Rainy Creek picrite; CWPIC, Cleaiwater picrite. °glomero, glomeroporphyritic.Modal proportions were visually estimated.alteration index based primarily on degree of plagioclase alteration and presence of secondary minerals (1, least altered; 3, most altered). Plagioclase iscommonly replaced by epidote, chlorite, sericite, clmnozoisite, and clay minerals and zoning is obscured in many samples due to albitization; clinopyroxene ismosty unaltered compared to plagiodase; Pe-Ti oxides are altered to sphene and leucoxene minerals; amygdules (<20 volume %) are prevalent throughoutmost subaerial flows and are usually filled with quartz, calcite, epidote, prehnite and pumpellyite.hglcr,glomerocrysts; f.g., fine-grained; c.g., coarse-grained, oik, swtl, swallow-tail. Mineral abbreviations: 01, olivine pseudomorphs;plag, plagioclase; cpx, clmnopyroxene; ox, oxides (includes ilmenite + titanomagnetite). High- and low-titanium basalte are listed from highest (top)to lowest (bottom) TiO2 contents.102Figure 3.9 Representative photomicrographs of Nikolai basalts, Alaska. (a) Aphyric pillow basalt from theTangle Lake (West) in the Amphitheater Mountains in cross-polarized transmitted light (sample 5810A10).(b) Hyaloclastite from near the top of the submarine section at Tangle Lake (West) in plane-polarizedtransmitted light (sample 5810A9, no chemistry). (c) Hyaloclastite interbedded with sills from Tangle Lake(West) in plane-polarized transmitted light (sample 581 OA3b, no chemistry). (d) Glomeroporphyritic pillowbasalt with abundant amygdules (—10 vol %) from Tangle Lake (West) in cross-polarized transmitted light(sample 5810A6). (e) Variolitic picritic pillow basalt with olivine pseudomorphs (<2 mm) and acicularplagioclase with swallow-tail terminations from the Clearwater Mountains, in plane-polarized transmittedlight (sample 5802A4). (f) Rainy Creek picritic tuff with pseudomorphed olivine phenocyrsts, undeformedrecrystallized angular and cuspate shards, and lithic particles in a fine-grained matrix, in plane-polarizedtransmitted light (sample 5808A2). Some clasts appear to be clusters of grains, some of which contain thingreen rims.103trachytic-like texture and abundant oxides. Picrite from the Clearwater Mountainspreserve spherulitic textures with bow-tie and fan-shaped bundles of acicular plagioclase,typically with swallow-tail terminations. Olivine (<2 mm) pseudomorphed by secondaryminerals comprises —20 vol % of the picrite and plagioclase commonly radiates from theedges of olivine pseudomorphs and is intergrown with clinopyroxene. The samplepreparation and analytical methods for whole-rock chemistry, major elements, traceelements, Sr, Nd, Hf, and Pb isotopes are described in Appendix D.WHOLE-ROCK CHEMISTRYMajor- and trace-element compositionsThe most noteworthy feature of the major-element chemistry of the NikolaiFormation is two clearly distinguishable groups of high- and low-titanium basalt (Fig.3.10). The low-titanium basalts range from 0.4 to 1.2 wt % Ti02 and the high-titaniumbasalts range from 1.6 to 2.4 wt % Ti02 (Fig. 3.10; Table 3.2). The high-titanium basaltshave a limited range in MgO (5.7-7.9 wt % MgO, except for one plagioclase-rich flowwith 4.8 wt % MgO) and Si02 (49.2-52.1 wt %), whereas the low-titanium basalts extendto higher MgO and have a significantly larger range in MgO (6.4-12.0 wt %) and Si02(46.7-52.2 wt %). Almost all of the Nikolai basalts in Alaska fall within the tholeiiticfield in a total alkalis versus silica plot, with low-titanium basalts generally having lowertotal alkalis than the high-titanium basalts. The low-titanium basalts exhibit broadlydecreasing trends of Ti02,FeOT, and Na20with increasing MgO (Fig. 3.10) and havehigher loss on ignition (LOl; mean LOI=2.7 ± 1.3 wt %) than the high-titanium basalts(mean LOI=1.9 ± 1.5 wt %; Table 3.2). Low-titanium basalts with >8 wt% MgO havehigher concentrations ofNi than the high-titanium basalts and the three picrite sampleshave noticeably higher Ni concentrations (525-620 ppm) than all basalts (Fig. 3.10). Boththe high- and low-titanium basalts have a large range in CaO, which appears to beindependent of MgO variation. A single Clearwater picrite (13.6 wt % MgO) and twoRainy Creek picrites (15.5-16.2 wt % MgO) have higher MgO with similar Ti02,FeOT,and alkali contents to the low-titanium basalts, however, the Rainy Creek picrites havenotably lower A1203(Fig. 3.10). The basal flow-conglomerate from the WrangellMountains has distinct major-element chemistry compared to the other Nikolai basalts.1040(f)5102 (%)DoQDoVoLcW70(i600500.00000•_-8dj:l.0 2 4 6 8 10 124 6 8 10 12MgO(wt%) 121110987014 16 18 0 2 4— lillIllIll 111111111 —.— .• x•high-titanium basalt7 Na20- T102• high-titanium sill• +. 2.5D low-titanium basalt6K20 alkalic - 0 low-titanium sill(wt%)02.0o 0A ClearwaterpicriteRainy Creek picrite5 cPX basalfiow0000 Yukon high-Ti basalt4.D tholeiitic1.50o Yukon low-Ti basalt30 •00i.o2-1-0a)‘(b)0III I I I 1111111111 III0.040 45 50 55 60 0 2I I I I I I I I I I I I I I4 6 8 10 12 14 16 18MgO (wt%)I ICaO• (wt%)• (C)II20191817161514A1203 x• (wt%)x‘III’ 11111 III00CCo 000• •o°(e?I I i I I I 4001514131211109876506543202 4 6 8300MgO (wt%)I I • i2000.Ni(ppm)AoC;(d) •J1006 8 10 12 14 16 18MgO (wt%)Na20 X(wt%)0,. 1•1111•14 16FeO fr)(wt%)18 000DC000I I I - I -- I -CCI I I I6 8 10 12 14 16 18MgO (wt%) MgO (wt%)Figure 3.10 Whole-rock major-element and Ni variation diagrams vs. MgO for the Nikolai Formation in Alaskawith data for the Nikolai Formation in Yukon (see chapter 4). The boundary of the alkaline and tholeiitic fields isthat of MacDonald and Katsura (1964). Total iron expressed as FeO, LOT is loss-on-ignition, and oxides areplotted on an anhydrous, normalized basis. Note the clear distinction between the high- and low-titanium basaltsin panel b and the difference in alkali contents between the Nikolai basalts in Alaska and Yukon in panel a.105Table 3.2 Major element (wt% oxide) and trace element (ppm) abundances in whole rock samples of Nikolai basalts, Alaska.SAMPLE 5707A3 5708A2 5710A2 5710A3 5712A2 5714A1 5714A3 5715A1 (1) 5715A1 (2) 5715A5Group HI-TI HI-TI HI-TI HI-TI HI-TI HI-TI HI-TI HI-TI HI-TI LOW-TIArea WM WM WM WM WM WM WM WM WM WMFlow FLO PLO FLO FLO PLO FLO FLO FLO PLO FLOUTM EW 6816521 6813830 6838676 6836850 6838178 6826371 6826390 6827794 6827794 6828050UTM NS 430777 428595 356035 356266 353987 384783 384806 383399 383399 383641Unnormalized Major Element Oxides (Weight %):Si02 49.74 48.40 48.81 50.02 49.90 49.31 50.21 49.61 49.41 50.25TiO2 1.96 2.36 2.4 1.95 1.89 1.66 1.6 1.69 1.68 1.14A12O3 13.37 13.49 15.21 14.45 13.76 15.58 16.01 15.56 15.75 16.69Fe2O3*14.05 16.14 12.94 13.39 13.76 13.21 10.99 13 12.97 10.39MnO 0.27 0.27 0.23 0.18 0.24 0.17 0.16 0.2 0.2 0.18MgO 6.69 5.89 6.85 6.56 6.74 6.6 6.64 6.63 6.62 7.94CaO 10.61 10.09 9.16 8.56 11.35 10.8 9.97 11.09 11.04 9.48Na2O 2.64 2.47 3.15 3.56 1.87 2.08 3.77 1.84 1.82 3.361<20 0.29 0.61 0.58 1.03 0.22 0.42 0.28 0.2 0.2 0.670.16 0.23 0.21 0.18 0.16 0.15 0.14 0.15 0.14 0.09LOI 1.68 1.41 2.37 2.47 1.28 2.39 3.12 0.84 1.84 3.57Total 99.78 99.95 99.54 99.88 99.89 99.98 99.77 99.97 99.83 100.19Trace elements (ppm):La 10.32 8.08 6.56 6.56 3.74Ce 24.25 21.79 17.07 17.07 10.69Pr 3.66 2.90 2.35 2.35 1.49Nd 18.83 14.73 12.20 12.20 7.68Sm 5.22 4.19 3.51 3.51 2.32Eu 1.73 1.43 1.25 1.25 0.90Gd 5.57 4.17 3.80 3.80 2.55Tb 1.02 0.77 0.72 0.72 0.50Dy 6.84 5.04 4.77 4.77 3.34Ho 1.35 0.99 0.98 0.98 0.69Er 4.02 2.85 2.86 2.86 1.98Tm 0.51 0.33 0.36 0.36 0.25Yb 3.58 2.37 2.59 2.59 1.77Lu 0.53 0.36 0.40 0.40 0.28Sc 44.06 42.52 41.17 41.17 38.50V 362 411 349 337 346 329 276 354 352 249Cr 165 62 167 119 133 156 282 165 167 460Co 47 45 44 44 39Ni 75 57 89 69 73 82 81 85 85 116Cu 19 1067 126 126 64Zn 121 176 126 108 113 109 88 110 111 68Ga 19 20 21 19 19 19 17 20 19 15Rb 3.4 9.7 6.1 15 2.2 5.2 2.6 1 1 8.5Sr 237 162 263 481 184 334 394 184 184 437Y 28.1 36.1 31.6 27.8 25.8 26.2 22.6 26 26.2 18.3Zr 113 148 145 115 110 99 93 98 98 60Nb 10.1 12.5 12.2 9.7 9.9 7.7 7.3 7.7 7.7 4.4Cs 0.08 0.14 0.28 0.28 0.83Ba 79 108 147 246 70 63 56 55 56 129Hf 4.10 3.15 2.78 2.78 1.79Ta 0.74 0.67 0.48 0.48 0.30Pb 1.59 0.81 0.52 0.52 0.44Th 0.98 0.72 0.64 0.64 0.34U 0.31 0.20 0.19 0.19 0.09La(XRF) 6 7 8 7 6 6 4 6 6 2Ce(XRF) 21 27 24 22 21 16 13 17 16 8Abbreviations for group are: HI-TI, hIgh-titanium; LOW-TI, low-titanium; RC, Rainy Creek; RCPIC, Rainy Creek picnte; CWPIC, Clearwater picrite.Abbreviations for flow are: PLO, massive flow; PIL, pillow basalt; FLO-BRE, flow-conglomerate-pillow breccia; SIL, sill; TUF, tuff. Abbreviations forarea are: WM, Wrangell Mountains; TANGLE, Tangle Lakes; GLAC, Glacier Gap Lake; CLEAR, Clearwater Mountains; RAINY, Rainy Creek.Sample locations are given using the Universal Transverse Mercator (UTM) coordinate system (NAD83; zones 6 and 7). XRF analyses wereperformed at University of Massachusetts Ronald B. Gilmore XRF Laboratory.Fe2O3*is total iron expressed as Fe2O3.LOI is loss-on-ignition.Elements by XRF: Sc, V, Cr, Ni, Zn, Ga, Rb, Sr, Y, Zr, Nb, Ba. Elements by ICP-MS: REE, Co, Cu, Cs, Hf, Ta, Pb, Th, U.106SAMPLE 5716A1 5716A2 5716A3 5719A1 5719A5 (1) 5719A5 (2)Group HI-TI HI-TI HI-TI LOW-TI HI-TI HI-TIArea WM WM WM WM WM WMFlow FLO FLO FLO FLO-BRE FLO FLOUTM EW 6824771 6824809 6825071 6840607 6840632 6840632UTM NS 384884 384800 384853 430357 430204 430204Unnormalized Major Element Oxides (Weight %):SIO2 50.89 50.52 50.76 55.72 49.54 49.62TiO2 1.66 2.12 1.96 0.67 1.92 1.93A1203 15.82 13.56 15.57 18.59 14.57 14.55FC2O3*12.03 14.22 13.37 7.84 12.84 12.86MnO 0.18 0.19 0.18 0.16 0.21 0.21MgO 6.13 7.83 4.8 3.37 6.52 6.55CaO 7.94 7.01 8.1 6.07 11.47 11.5Na20 4.26 4.52 4.71 5.14 2.01 2.02K20 0.58 0.08 0.23 2.18 0.29 0.30.15 0.18 0.19 0.38 0.17 0.17LOI 2.96 3.57 3.54 5.14 1.05 1.09Total 99.64 100.23 99.87 100.12 99.54 99.71Trace elements (ppm):La 18.10 9.13 9.13Ce 36.95 21.43 21.43Pr 4.56 3.05 3.05Nd 20.65 16.26 16.26Sm 4.26 4.45 4.45Eu 1.31 1.48 1.48Gd 3.96 4.58 4.58Tb 0.52 0.79 0.79Dy 3.40 5.31 5.31Ho 0.66 0.97 0.97Er 2.12 2.94 2.94Tm 0.26 0.35 0.35Yb 1.94 2.46 2.46Lu 0.31 0.36 0.36Sc 11.02 41.37 41.37V 289 393 319 170 366 367Cr 157 133 70 0 162 160Co 16 42 42Ni 66 61 47 1 77 76Cu 25 167 167Zn 94 97 82 83 104 104Ga 15 17 21 17 19 19Rb 5.8 0.6 2 25.4 5.8 5.9Sr 321 56 97 158 200 200V 26.4 30.8 28.5 19.5 26.3 26.1Zr 98 120 118 81 113 113Nb 7.6 9.7 10.2 6.3 10.1 10Cs 0.06 0.09 0.09Ba 115 14 47 1277 94 91Hf 2.30 3.23 3.23Ta 0.36 0.63 0.63Pb 2.96 0.85 0.85Th 4.18 0.87 0.87U 1.91 0.25 0.25La(XRF) 4 4 7 17 7 8Ce(XRF) 16 18 18 34 23 235719A6 5725A2 5725A4 5726A1HI-TI HI-TI HI-TI HI-TIWM TANGLE TANGLE TANGLEFLO SIL SIL PIL6840632 7001750 7002024 7002334430204 556168 555867 55324549.76 50.53 50.43 49.011.92 2.19 1.82 1.8814.45 14.98 14.31 13.9612.79 11.73 12.36 13.130.21 0.2 0.2 0.236.65 5.99 6.66 7.0711.45 11.74 12.13 11.742.12 1.99 1.78 2.260.24 0.46 042 0.250.17 0.21 0.16 0.162.66 1.26 1.03 1.8899.76 100.02 100.27 99.698.57 7.5221.35 19.803.00 2.7415.16 14.354.30 4.071.43 1.084.32 4.150.77 0.744.89 4.920.94 0.912.69 2.590.32 0.302.17 2.030.30 0.2738.05 41.02340 358 325 359164 140 127 21742 4278 68 80 90160 138104 96 101 11019 20 20 174.1 16.2 16.5 2.1275 213 207 23925.6 29.6 24.3 24.3112 140 107 1109.9 12.1 9.2 9.40.53 0.0785 245 83 1913.08 2.660.57 0.520.66 0.950.83 0.860.25 0.238 8 5 521 23 19 18107SAMPLE 5726A2 5726A3 5726A6 5727A2 5727A3 5727A5 5727A6 (1) 5727A6 (2) 5727A7 5731A3Group HI-TI HI-TI I-Il-TI LOW-TI LOW-TI LOW-TI LOW-TI LOW-TI LOW-TI LOW-TIArea TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE CLEARFlow FLO FLO PIL SIL SIL PIL SIL SIL PIL PILUTM EW 7002466 7002475 7002270 6999661 6999663 6999658 6999646 6999646 6999646 6993085UTM NS 553936 553826 553495 550468 550377 550351 550324 550324 550270 483499Unnarmalized Major Element Oxides (Weight %):S1O2 51.41 49.90 50.11 50.70 50.97 47.43 48.89 48.89 46.13 50.94T1O2 1.63 1.81 1.85 0.54 0.58 0.57 0.48 0.48 0.56 0.53A12O3 13.83 14.73 14.74 14.49 15.85 15.81 15.06 15.06 17.97 15.51FC2O3*12.16 12.34 11.71 11.21 10.79 11.04 10.51 10.51 11.84 9.75MnO 0.19 0.21 0.19 0.19 0.19 0.19 0.18 0.18 0.17 0.17MgO 6.96 6.72 6.75 9.41 7.75 10.35 11.51 11.51 9.09 10.52CaO 11.25 11.96 12.7 12.3 11.96 13.12 12.43 12.43 12.78 10.68Na2O 2.2 1.95 1.94 1.09 1.76 1.06 0.79 0.79 1.19 1.22K2O 0.11 0.09 0.13 0.06 0.1 0.09 0.05 0.05 0.1 0.22P2O5 0.14 0.16 0.16 0.04 0.05 0.07 0.06 0.06 0.07 0.1LOI 1.6 1.35 1.29 2.18 2.3 2.87 3 3 2.75 3.15Total 99.88 99.87 100.28 100.03 100.00 99.73 99.96 99.96 99.90 99.64Trace elements (ppm):La 1.13 1.13 1.25 1.27 2.56Ce 3.19 3.22 3.56 3.58 5.66Pr 0.49 0.51 0.54 0.56 0.80Nd 2.92 2.92 3.05 3.09 3.86Sm 1.22 1.23 1.11 1.13 1.30Eu 0.56 0.49 0.37 0.36 0.67Gd 1.47 1.47 1.33 1.371.63Tb 0.36 0.34 0.34 0.32 0.38Dy 2.62 2.60 2.35 2.37 2.80Ho 0.56 0.60 0.55 0.53 0.64Er 1.67 1.70 1.63 1.58 1.93Tm 0.22 0.22 0.21 0.21 0.25Yb 1.61 1.61 1.53 1.53 1.91Lu 0.27 0.26 0.250.24 0.33Sc 44.90 43.02 45.67 49.00 50.33V 324 332 333 242 254 257 233233 268 217Cr 118 148 136 555 269 455 521 521199 605Co 51 44 5454 57Ni 79 84 85 178 106 157 214 214 163 191Cu 96 116 98 96131Zn 102 100 84 81 77 84 74 74 92 68Ga 18 20 19 14 14 1312 12 15 12Rb 0.7 0.4 3.4 1.7 1.8 1.2 0.6 0.61.5 4.6Sr 195 178 198 76 117 146 123 123155 205Y 22.2 24.6 24.7 15.1 15.5 15.814 14 17.2 17.2Zr 94 106 109 23 24 17 15 1520 38Nb 8 92 9.4 0.9 0.9 1 0.8 0.8 0.8 2.2Cs 0.36 0.28 0.120.12 0.23Ba 45 47 48 98 76 130 46 4654 125Hf 0.80 0.87 0.57 0.57 0.70Ta 0.04 0.05 0.04 0.04 0.05Pb 0.55 0.50 0.20 0.181.21Th 0.17 0.17 0.07 0.07 0.29U 0.07 0.07 0.03 0.030.10La(XRF) 6 7 6 0 1 1 1 1 24Ce(XRF) 17 18 19 2 2 2 3 33 11108Trace elements (ppm):LaCePrNdSmEuGdTbDyHoErTmYbLuScVCrCoNiCuZnGaRbSrVZrNbCsBaHfTaPbThULa(XRF)Ce(XRF)49.751.9115.1713.130.215.7611.841.690.280.171.2599.918.6521.653.0715.514.251.464.400.805.221.012.930.352.500.3638.99358584156182111205.918726.211310.30.09703.170.610.720.840.2472050.82 49.97 49.181.09 2.33 0.7115.06 13.95 16.2712.51 12.28 10.820.22 0.19 0.197.16 6.69 9.4110.44 11.77 101.99 1.85 2.820.27 0.38 0.630.15 0.22 0.172.13 1.5 3.6499.71 99.63 100.203.006.921.004.041.310.451.570.342.530.571.760.231.750.3950.28285 330 265 24769 129 360 364441278968112.130315.3231.30.30118 73 186 440.780.080.210.560.206 5 6 313 21 13 6242346151701316.926014.9291.7198485801A5HI-TICLEARFLO69927504803285801A8 5801A9 5802A1LOW-TI HI-TI LOW-TICLEAR CLEAR CLEARFLO PIL PIL6992452 6992443 6993172480516 480528 4838885802A2LOW-TICLEARPIL699314048386749.940.4815.3310.440.198.8512.671.880.110.063.4399.955802A3LOW-TICLEARPIL699310448384747.760.5216.7610.420.189.8512.311.20.590.062.799.65SAMPLE 5731A4 5731A5 5731A6 5801A2Group LOW-TI LOW-TI LOW-TI HI-TIArea CLEAR CLEAR CLEAR CLEARFlow PIL PIL PIL FLOUTMEW 6993131 6993230 6993320 6992862UTM NS 483495 483531 483566 480105Unnormalized Major Element Oxides (Weight %):Sb2 48.61 48.12 49.13 50.00TiO2 0.48 0.92 0.64 2.37A12O3 15.89 16.1 15.9 14.02Fe203*10.32 11.5 9.69 13.68MnO 0.19 0.19 0.16 0.21MgO 9.63 9.67 8.37 6.08CaO 12 11.71 13.54 11.77Na2O 1.87 1.35 2.13 1.741<20 0.63 0.28 0.29 0.130.06 0.1 0.16 0.22LOI 3.1 3.34 3.77 1.26Total 99.68 99.94 100.01 100.2212.7028.044.1421.105.551.795.490.936.211.133.250.372.660.3837.45232 256 237 386373 578 412 21143141 232 137 8020767 72 62 11512 15 13 2015.9 6.5 3.3 1.7392 272 484 22513.8 20.4 16.2 29.426 45 35 1461.4 1.6 2.8 13.10.12173 108 190 574.030.761.011.240.363 2 6 97 9 14 3156 81 142118 95 7316 19 124.6 5 16.3148 190 48526.7 24.8 18.159 109 403.9 9.4 3.1109SAMPLE 5802A4 (1) 5802A4 (2) 5802A5 5802A6 5806A3 5806A5 5806A6 5808A1 5808A2 5808A3Group LOW-TI CWPIC LOW-TI LOW-TI HI-TI HI-TI HI-TI RC RCPIC RCPICArea CLEAR CLEAR CLEAR CLEAR GLAC GLAC GLAC RAINY RAINY RAINYFlow PIL PIL PIL SIL? FLO FLO FLO TUF TUF TLJFUTMEW 6993057 6993057 6993004 6992950 7000695 7000232 6999795 7021484 7021213 7021298UTM NS 483801 483801 483762 483733 538926 538787 538622 556534 556314 556392Unnormalized Major Element Oxides (Weight %):Si02 48.26 48.28 46.68 51.77 49.67 49.86 50.25 49.93 46.44 46.89Ti02 0.5 0.5 0.93 0.9 2.37 1.97 1.89 1.01 1.06 1.18A1203 15.81 15.83 16.19 14.82 13.98 14.58 13.96 14.09 10.72 10.72Fe203*10.73 10.75 12.21 11 13.53 12.98 12.79 13.4 10.72 10.72MnO 0.19 0.19 0.2 0.19 0.21 0.21 0.21 0.22 0.15 0.18MgO 13.49 13.47 1021 7.95 6.4 6.6 6.62 7.31 15.42 14.78CaO 9.2 9.19 12.16 11.72 10.36 10.83 12.24 11.9 9.59 9.42Na20 1.48 1.51 1.23 147 2.14 2.25 1.59 1.61 1.34 0.881<20 0.05 0.05 0.12 0.3 0.68 0.57 0.29 0.31 0.35 1.12P205 0.09 0.09 0.1 0.09 0.22 0.17 0.17 0.1 0.09 0.11LOl 4.51 4.53 3.32 1.56 1.82 1.62 1.67 1.98 3.2 2.4Total 99.80 99.86 100.03 100.21 99.56 100.02 100.01 99.88 99.86 100.01Trace elements (ppm):La 5.18 5.18 3.58 5.99 6.52Ce 12.40 12.40 9.28 15.32 18.08Pr 148 148 1.32 1.96 2.33Nd 6.61 6.61 6.68 9.85 11.57Sm 1.69 1.69 2.19 2.69 3.01Eu 0.51 0.51 0.84 0.94 1.09Gd 1.84 1.84 2.53 2.77 3.02Tb 0.40 0.40 0.54 0.51 0.54Dy 2.62 2.62 3.76 3.25 3.23Ho 0.56 0.56 0.79 0.66 0.63Er 1.76 1.76 2.30 1.79 1.79Tm 0.23 0.23 0.29 0.21 0.22Yb 1.69 1.69 2.12 1.48 1.45Lu 0.27 0.27 0.33 0.22 0.25Sc 40.80 40.80 45.91 2729 29.39V 202 203 257 254 369 339 327 313 285.54 297.75Cr 1596 1577 633 448 213 141 123 138 1327 1250Co 60 60 58 67 63Ni 539 538 244 128 89 87 78 76 620 525Cu 78 78 103 123 79Zn 74 74 82 85 119 112 101 111 60 63Ga 13 13 16 16 21 20 19 16 13 13Rb 0.4 0.5 3 8.8 13.9 11.8 4.9 5.8 6.11 2643Sr 183 184 252 188 215 208 205 185 126.86 92.53Y 15.4 15.2 20.7 18.5 29.1 25.4 25.4 28.5 16.01 15.81Zr 35 35 46 49 144 116 112 41 58.73 65.10Nb 2 2.1 1.5 3 12.9 10.4 9.6 5.3 6.00 8.11Cs 0.50 0.50 0.30 0.43 1.10Ba 39 37 40 92 182 118 108 140 72 191Hf 1.05 1.05 1.51 1.71 1.88Ta 0.11 0.11 0.08 0.35 048Pb 0.13 0.13 0.50 0.76 0.65Th 0.89 0.89 0.34 0.70 1.06U 0.32 0.32 0.11 0.26 0.38La(XRF) 4 5 2 2 9 7 6 3 6 7Ce(XRF) 7 11 4 4 27 21 19 7 15 18110SAMPLE 5808A6 5808A8 5810A1 (1) 581 OAI (2) 5810A2 5810A4 58106(1) 5810A6(2) 5810A10 5811A1Group RC RC LOW-TI LOW-TI LOW-TI LOW-TI LOW-TI LOW-TI HI-TI LOW-TIArea RAINY RAINY TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE TANGLE TANGLEFlow DIK SIL PIL PIL SIL SIL PIL PIL PIL SILUTMEW 7021445 7021631 6999686 6999686 6999760 6999733 6999748 6999748 7000019 6999515UTMNS 556560 557160 550380 550380 550367 550310 550268 550268 550092 550374Unnormalized Major Element Oxides (Weight %):SiO2 49.27 50.04 47.02 46.82 48.39 48.45 48.75 48.69 50.27 49.18Ti02 0.93 1.14 0.57 0.57 0.54 0.73 0.64 0.64 1.74 0.35A1203 14.71 13.39 15.84 15.8 16.2 17.2 16.2 16.1 14.61 15.84Fe203*10.76 15.35 10.94 10.96 11.32 11.81 12.01 12.03 12.04 9.09MnO 0.2 0.25 0.18 0.18 0.2 0.2 0.24 0.24 0.19 0.17MgO 8.88 6.57 11.23 11.16 9.96 7.32 6.34 6.26 6.71 11.93CaO 11.31 11.27 13.14 13.09 12.21 12.5 13.95 13.98 12.81 12.38Na2O 1.73 1.52 1.19 1.04 1.22 1.25 1.86 1.77 1.3 0.88K2O 1.57 0.06 0.06 0.07 0.07 0.06 0.15 0.15 0.14 0.070.27 0.11 0.07 0.07 0.06 0.12 0.08 0.08 0.15 0.05LOI 2.16 1.96 2.68 2.65 2.62 2.19 1.73 1.72 1.49 3.27Total 99.63 99.70 100.24 99.76 100.17 99.64 100.22 99.94 99.96 99.94Trace elements (ppm):La 1.16 1.16 1.94 3.45 2.30 2.30 7.81Ce 3.40 3.40 4.53 8.87 5.88 5.88 20.71Pr 0.57 0.57 0.61 1.20 0.83 0.83 2.83Nd 3.31 3.31 3.21 6.12 4.52 4.52 13.98Sm 1.26 1.26 1.16 1.96 1.55 1.55 3.95Eu 0.43 0.43 0.46 0.73 0.51 0.51 1.45Gd 1.51 1.51 1.46 2.15 1.79 1.79 3.94Tb 0.36 0.36 0.35 0.45 0.42 0.42 0.76Dy 2.59 2.59 2.53 3.28 2.94 2.94 4.63Ho 0.57 0.57 0.56 0.72 0.64 0.64 0.95Er 1.74 1.74 1.78 2.13 1.95 1.95 2.50Tm 0.24 0.24 0.23 0.28 0.26 0.26 0.29Yb 1.70 1.70 1.75 1.99 1.88 1.88 2.05Lu 0.27 0.27 0.27 0.32 0.33 0.33 0.32Sc 50.28 50.28 47.06 41.96 43.29 43.29 35.80V 289 363 261 260 253 271 261 262 308 182Cr 629 58 493 491 182 81 131 129 133 380Co 53 53 50 45 43 43 41Ni 137 56 208 207 163 100 91 92 88 274Cu 98 98 110 116 85 85 151Zn 102 127 74 73 78 88 98 98 100 64Ga 18 17 13 13 13 15 14 14 19 10Rb 18.3 0.5 0.9 0.6 0.8 0.4 2 1.9 2.6 0.9Sr 801 58 151 150 141 198 170 170 231 114Y 16.7 32.6 15.5 15.5 15.5 18.5 17.2 17.2 23 11.8Zr 65 43 18 17 19 28 26 26 102 19Nb 4 4.8 1.1 1 1 1.4 1.3 1.3 8.8 1.3Cs 0.17 0.17 0.12 0.12 0.18 0.18 0.09Ba 674 39 62 62 62 33 75 78 59 102Hf 0.66 0.66 0.66 0.96 0.88 0.88 2.84Ta 0.04 0.04 0.05 0.06 0.10 0.10 0.59Pb 0.23 0.23 0.44 0.26 0.63 0.63 0.76Th 0.09 0.09 0.27 0.30 0.28 0.28 0.78U 0.04 0.04 0.10 0.09 0.12 0.12 0.24La(XRF) 8 3 1 0 2 3 3 0 6 2Ce(XRF) 20 7 3 1 3 7 2 4 19 3111The Nikolai Formation in Alaska has similar major- and trace-element chemistry toNikolai basalts from Yukon, which also have high- and low-titanium basalts (Fig. 3.10).Nikolai basalts in Alaska are notably lower in total alkalis than basalts from Yukon dueto alkali metasomatism during alteration of the Yukon basalts (see chapter 4).The low-titanium basalts are characterized by a range of flat and slightly light rareearth element (LREE)-depleted chondrite-normalized REE patterns (mean La/SmcN=0.8± 0.5, except one LREE-enriched sample) with flat, parallel heavy REE (HREE)segments (mean Dy/YbCN=l.0 ± 0.2; Fig. 3.11). The high-titanium basalts form a tightrange of parallel LREE-enriched patterns (mean LaJYbCN=2.3 ± 0.9) with higher REEabundances (mean YbCN= 16.1 ± 6.4) than the low-titanium basalts (mean Ybc=1 1.0 ±2.2; Fig. 3.11). Several low-titanium basalts from the Amphitheater Mountains havepositive Eu anomalies (Fig. 3.11). Several high-titanium basalts from the WrangellMountains have patterns with flatter HREE segments. The basal flow-conglomerate fromthe Wrangell Mountains has a distinct LREE-enriched pattern (LaJSmcN= 2.7) with a flatHREE segment (DyJYbCN=1.2). The Rainy Creek picrites are LREE-enriched(LaJYbcN=2.8-3.l) and the REE pattern of the Clearwater picrite is LREE-enriched(LaJSmCN= 1.9)with a flat HREE segment (Dy/YbCN=l.0).The low-titanium basalts have mostly parallel, primitive mantle-normalized trace-element patterns with pronounced HFSE depletions, especially for Nb, Ta, and Zr,relative to LILE and REE (Fig. 3.11), and HFSE (Nb and Zr) form linear trends (Fig.3.12). The large negative Zr anomalies in the low-titanium basalts are not accompaniedby comparably low Hf (Fig. 3.1 id). The LILE segments of trace-element patterns forlow-titanium basalts are parallel and each of the patterns has a pronounced positive Sranomaly, relative to Nd and Sm. The high-titanium basalts form a tight range of parallel,concave-downward trace-element patterns with negative Pb anomalies, relative to Ce andPr, and small negative K anomalies, relative to U and Nb (Fig. 3.11). The LILE in thehigh-titanium basalts are slightly depleted relative to the HFSE and LREE. TheClearwater picrite is depleted in HFSE with a positive Sr anomaly, similar to the low-titanium basalts, whereas Rainy Creek picrites have gently negative-sloping patterns withnegative Sr anomalies (Fig. 3.11).1124?C0104?aCevstitanium basalts90000 Wrangell Mountainsa Amphitheater MountainsA clearwater Mountains(a)1004?C:0.4?aa4?vsHigh-titanium basalts000Cs Rb8a ThU K NbTa La CePb Pr NdSrSmZr H 0 EuGdThDyHoY ErYbLuLa Ce Pr Nd PmSm Eu 0dm Dy Ho Er Tm Yb Lu4?0C010aa4?vs0e basal flowLow-titanium basaltso0/00(c)4?CCea0.4?aECevs10015La Ce Pr Nd Pm Sm Eu Gd Th Dy Ho & Tm Yb Lu1004?C0U4?aaCeviCsRb8aThU KNbTuLaCePbPrNdSrSmZrFfTiEuGdThDyHoY &YbLua?Ca0.4?aaCevs10La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu CsRbBaThU K NbTaL.aCePbPrNdSrSmZr HfT1 EuGdThDyHoY ErYbLuFigure 3.11 Whole-rock REE and other incompatible-element concentrations for the Nikolai Formation in Alaska.(a), (c), and (e) are chondrite-normalized REE patterns for high-titanium basalts, low-titanium basalts, andpicrites,respectively, from field areas in southern Alaska. (b), (d), and (f) are primitive mantle-normalized trace-elementpatterns for high-titanium basalts, low-titanium basalts, and picrites, respectively. All normalization values are fromMcDonough and Sun (1995). Depleted MORB average from Salters and Stracke (2004). Note the clear distinctionbetween LREE-enriched high-titanium basalt and mostly LREE-depleted low-titanium basalts. Trace-elementpatterns for the basal flow-conglomerate and one Clearwater sample are not shown in panel d for clarity.113IIT I I-I- I I I I I I I —I--—3.0150 ‘ 0 2 4 6 8 10 12 14 16 18Figure 3.12 Whole-rock trace-element concentration variations and ratios for the Nikolai Formation in Alaska(except panel b is versus MgO), with data for the Nikolai Formation in Yukon. (a) Nb vs. Zr. (b) Nb/La vs. MgO.(c) Th vs. Hf. (d) Sr vs. Y/Nb. Note the clear distinction between the high- and low-titanium basalt in each of theplots.j12Nb (ppm)1086 CjD4. CCo20502.52.01.51.00,5•high-titanium basaltNb/La• high-titanium sill•D low-titanium basalt0 low-titanium sillA Clearwaterpicrite0 RainyCreekpicrite••Ij X basal flow0 Yukon high-Ti basaltQC Yukon low-Ti basalt0-0-- —0x• (b)I I I I I I I I I I I I I I I I02.0100Zr (ppm)1.5MgO(wt%)1.00.5I,IQ I I ISr (ppm)400•CI300C•CC2004.QD0CCo$0•CI I I .0.001001 2 3 4 5Hf (ppm)(d)00 5 10Y/Nb15 20114Sr-Nd-Hf-Pb isotopic compositionsThe high- and low-titanium Wrangellia basalts in Alaska have distinct Hf and Srisotope ratios (Fig. 3.13). The low-titanium basalts have higher initialEHf(+13.7 to + 18.4)and 875r/86Sr(0.70422-0.70554) than the high-titanium basalts (initial8H=+9.7 to +10.7,initial87Sr/86Sr= 0.70302-0.70376), except for sample 571 5A5, which has anomalouschemistry (Fig. 3.13; Tables 3.3 and 3.4). The low-titanium basalts have a narrower rangein initial8Nd(+4.6 to +5.4) than the high-titanium basalts (initial8Nd=+6.0 to +8.1) and awider range in initial8Hf,except for three low-titanium samples with higher initial8Nd(+7.0 to +7.6). The basal flow-conglomerate in the Wrangell Mountains has similar initialSr, Nd, and Hf isotopic compositions to the high-titanium basalts. The Clearwater picritelies between the high- and low-titanium basalts in initial8Hf,and has the lowest initial6Ndand initial 875r/865rat the upper end of the range of all samples. The two Rainy Creekpicrites have similar Sr, Nd, and Hf isotope ratios to high-titanium basalts with slightlylower intial8Nd(Fig. 3.13).The high- and low-titanium basalts have indistinguishable age-corrected Pbisotopic compositions, although the high-titanium basalts show a narrower range (Fig.3.14). The range of initial Pb isotopic compositions for low-titanium basalts is206Pb/204Pb 18.421-19.418, 207Pb/204Pb = 15.568-15.609, and 208Pb/204Pb= 37.962-38.481 and the high-titanium basalts have 206Pb/204Pb= 18.504-18.888, 207Pb/204Pb=15.556-15.587, and 208Pb/204Pb= 38.008-38.451 (Fig. 3.14; Table 3.5). The basal flow-conglomerate has a slightly lower initial Pb isotopic composition than the high- and low-titanium basalts. The Clearwater picrite has a lower initial Pb isotope ratio than thebasalts and the Rainy Creek picrites have noticeably higher initial Pb isotope ratios thanthe basalts (Fig. 3.14).ALTERATIONThe Nikolai basalts generally preserve primary mineralogical, textural, andvolcanological features and have retained most of their primary magmatic composition.Secondary minerals have replaced variable, but generally small, proportions of primaryminerals in the Nikolai Formation and the basalts contain zeolite to prehnite-pumpellyitefacies alteration minerals (prehnite + pumpellyite + epiclote+chlorite + quartz ±1150.2835 9, ,‘ I I IIII571 5A50,2834 “6Hf/177Hf5727A3 5727A28727A25727A30.2833aENd230 Ma)5802A55802A570.2832 605132I I I I I I727A3143Nd/1Nd 5727A20.28315715A5051315802A2\5715A5C0.283005130,.5802A5C0.5129x0.2829+(a)05128(c)0.2828____________________________A I i • a I i0.00 0.02 0.04051270.703 0.704 0.705 0.706-4-176Lu/177Hf(b) • 87SrI 8tSr(230 Ma)0512C200.1 0.15 0.2 0.25 03_________________________________________________III4‘47Sm/’NdI I5802A5-18D18Ma) 5727A3—0E(230 Ma)5727A2\ E’5727A2Hf16 - 165802A25810A414- . 14\12- 12’,5715A5 ,5715A510 10• hi9h-titaniumbasalt• high-titanium sillQ low-titanium basalt0low-titanium sill8 - 8 A ClearwaterpicriteRainy Creek picrite(d)‘ (e) X basal flow6I I I I I I I I2 4 6 8 10 0.702 0.703 0.704 0.705 0.706ENd23°Ma)87SrI 86Sr(230 Ma)Figure 3.13 Whole-rock Sr, Nd, and Hf isotopic compositions for the Nikolai Formation in Alaska. (a)‘76Hf’177Hfvs.‘76LuJ’77Hf. (b) 143Nd/1Ndvs. 147Sm/1Nd. (c) Initial8Ndvs.t7SrIt6Sr. Age correction to 230 Ma. (d) Initial CHfvs.8Nd•(e) Initial vs. 87SrJ86Sr. Average 2a error bars are shown in a corner of each panel.11615.85‘ i ‘ i ‘ i i ‘ i‘ 15.68207Pb/204Pb 207Pb/204Pb(230 Ma)15.8015.64 -15.755802A2.58O2A215.6015.70•high-titanium basalt• high-titanium sill15.5615.65 D low-titanium basaltQlow-titanium sillX15.60.15.52X basalfiow(a) (b)15.55• I I I I I I15.48I I I I I18 19 20 21 22 23 24 17 18 19 20 21206Pb/204Pb 206Pb/204Pb(230 Ma)43.0 40.0i‘ iA 208nkI2O4nL..Z.1395nj! ni(230 Ma)42.0’39.041.038.5ö5802A20 —5802A240.038.0x37-539.0X37.0(c) (d)38.0‘ I 1 I36.518 19 20 21 22 23 24 17 18 19 20 21206Pb/204Pb(230 Ma)Figure 3.14 Pb isotopic compositions of leached whole-rock samples by MC-ICP-MS for the Nikolai Formationin Alaska. Error bars are smaller than symbols. (a) Measured 207PbP°4Pb vs. 206Pb/204Pb. (b) Initial 207Pb/204Pbvs.206Pb/204Pb. Age-correction to 230 Ma. (c) Measured 208Pb/204Pb vs. 206Pb/204Pb. (d) Initial 208PbP°4Pb vs.206pb/204Pb117Table3.3SrandNdisotopicgeochemistryofwholerocksamplesofNikolaibasalis.AlaskaSampleGroupsAreat)RbSr87SrlSr2m87Rb/86Sra7SrlwSrtSmNdNdlNd2amENd1475m/1”Nd143Nd/’tNdCNd(t)(nnmt(onm230Ma(nnm(onm230Ma5719A1BasalWM25.41580.70513870.46520.703624.2620.650.51289265.00.12480.512705725A4Sill(Hi-Ti)TA16.52070.70377580.23060.703024.3015.160.51291675.40.17150.512666.2GO5802A4PicriteCL041830.70553090.00630.705511.696.610.51274662.10.512515708A2Hi-TiWM9.71620.70377670.17330.703215.2218.830.51301157.30.16770.512768.15801A2Hi-TiCL1.72250.70336580.02190.703295.5521.100.51293475.80.15920.512696.95719A5HI-TIWM5.92000.70334480.08540.703064.4516.260.51298866.80.16530.512747.75801.8.5Hi-TiCL5.91870.70338980.09130.703094.2515.510.51293765.80.16590.512696.75712A2Hi-TiWM2.21840.70317980.03460.703074.1914.730.51299767.00.17200.512747.75726A1Hi-liTA2.12390.70384470.02540.703764.0714.350.51290875.30.17130.512656.05810A10Hi-TiTA2.62310.70344890.03260.703343.9513.980.51291865.50.17070.512666.25715A1Hi-TiWM1.01840.70361970.01570.703573.5112.200.51301777.40.17380.512768.15715A5Low-TiWM8.54370.70360280.05630.703422.327.680.51303057.60.18290.512758.15802A5Low-TiCL3.02520.70494260.03450.704832.196.680.51300067.10.19800.512707.05810A6Low-TiTA2.01700.70470470.03400.704591.554.520.51292465.60.20700.512615.35810A1Low-TiTA0.91510.70472170.01720.704661.263.310.51296356.30.23030.512625.45727A7Low-TiTA1.51550.70486590.02800.704771.303.860.51290675.20.20380.512605.05802A2Low-TiCL2.13030.70560480.02010.705541.314.040.51287274.60.19590.512584.6 7.15727.8.2Sill(Low-Ti)TA1.7760.70453280.06470.704321.222.920.51309869.00.25310.512727.35727A3Sill(Low-TI)TA1.81170.70487280.94450.704731.232.920.51311589.30.25530.512737.65727A6Sill(Low-Ti)TA0.61230.704857100.01410.704811.113.050.51293075.70.22030.512605.05727A6(dup)Sill(Low-Ti)TA0.61230.70485180.01410.704801.133.090.512928105.70.22020.512605.05810A2Sill(Low-Ti)TA0.81410.70484670.01640.704791.163.210.51291765.40.21870.512594.85810A4Sill(Low-Ti)TA0.41980.70423580.00580.704221.966.120.51288664.80.19370.512594.90.15443.35808A2PicriteRC6.11270.704124100.13950.703672.699.850.51288064.70.16490.512635.75808.8.3PlcriteRC26.4930.70630790.82650.703603.0111.570.51281163.40.15760.512574.5Hi-Ti,high-titaniumbasalt;Low-Ti,low-titaniumbasalt;Basal,basalfIow-congIomerate.Abbreviationsforareaare:WM,WrangellMountains;TA,TangleLake;GG,GlacierGapLake;CL,ClearwaterMountains;RC,RainyCreek.(dup)indicatescompletechemistryduplicate.Alltrace-elementandisotopicanalyseswerecarriedoutatthePCIGR.TheanalyticalmethodsaredescribedinAppendixD.Table 3.4 Hfisotopic compositions ofwhole rock samples ofNikolai basalts, Alaska571 9A1 Basal WM 0.31 2.30 0.28301 6 8.3 0.0188 0.28292 9.95802A4 Picrite CL 0.27 1.05 0.28315 6 13.2 0.0365 0.28298 12.0Hi-Ti, high-titanium basalt; Low-TI, low-titanium basalt; Basal, basal flow-conglomerate. Abbreviations for area are: WM, WrangellMountains; TA, Tangle Lake; GG, Glacier Gap Lake; CL, Clearwater Mountains; RC, Rainy Creek. All trace-element and isotopicanalyses were carried out at the PCIGR. The analytical methods are described in Appendix D.Sample GroupsAreabLu Hf 177HfP76Hf2m 8ijt 176Lu/177Hf 177HfP76Hf(ppm) (ppm) 230 Ma5708A2 Hi-Ti WM 0.53 4.10 0.28302 5 8.6 0.0184 0.28293 10.35801A2 Hi-Ti CL 0.38 4.03 0.28299 6 7.6 0.0134 0.28293 10.05719A5 Hi-Ti WM 0.36 3.23 0.28301 6 8.4 0.01 59 0.28294 10.55801A5 Hi-Ti CL 0.36 3.17 0.28302 7 8.7 0.0161 0.28295 10.75712A2 HI-Ti WM 0.36 3.15 0.28300 6 8.2 0.0162 0.28293 10.25726A1 Hi-Ti TA 0.27 2.66 0.28300 6 8.0 0.0146 0.28293 10.35810A10 Hi-Ti TA 0.32 2.84 0.28299 6 7.6 0.0160 0.28292 9.75715A1 Hi-Ti WM 0.40 2.78 0.28301 5 8.5 0.0202 0.28292 9.95715A5 Low-Ti WM 0.28 1.79 0.28304 6 9.6 0.0223 0.28294 10.65802A5 Low-Ti CL 0.33 1.51 0.28330 14 18.6 0.0309 0.28316 18.45810A6 Low-TI TA 0.33 0.88 0.28330 6 18.7 0.0537 0.28306 14.95810A1 Low-TI TA 0.27 0.66 0.28340 5 22.1 0.0581 0.28314 17.55727A7 Low-Ti TA 0.33 0.70 0.28339 7 21.9 0.0670 0.28309 15.95802A2 Low-Ti CL 0.39 0.78 0.28336 7 20.7 0.0717 0.28304 14.05725A4 Sill (HI-TI) TA 0.30 3.08 0.28299 5 7.7 0.0141 0.28293 10.15727A2 Sill (Low-Ti) TA 0.27 0.80 0.28333 7 19.7 0.0489 0.28311 16.55727A3 Sill (Low-Ti) TA 0.26 0.87 0.28333 9 19.8 0.0425 0.28314 17.75727A6 Sill (Low-Ti) TA 0.24 0.57 0.28335 8 20.6 0.0602 028309 15.75810A2 Sill (Low-Ti) TA 0.27 0.66 0.28337 6 21.3 0.0581 0.28311 16.7581 0A4 Sill (Low-Ti) TA 0.32 0.96 0.28325 8 16.7 0.0481 0.28303 13.75808A2 Picrite RC 0.22 1.71 0.28302 7 8.7 0.0184 0.28294 10.35808A3 Picrite RC 0.25 1.88 0.28298 6 7.2 0.0189 0.28289 8.8119Table3.5PbisotopiccompositionsofwholerocksamplesofNikolaibasalts,AlaskaSampleGroupsArea”UThPb2o6Pbl2mPb2m207Pbl204Pb2m208Pb/204Pb2°m2U/2MPb2U/2MPb232Thl204Pb206Pb/204Pb,207Pb/204Pb,208Pb/204Pb,(ppm)(ppm)(ppm)230Ma230Ma230Ma5708A2Hi-TiWM0.310.981.5919.25570.000615.60400.000538.76130.001412.390.0940.9518.80615.58138.2895801A2HI-TICL0.361.241.0119.44040.000815.61000.000739.02170.001923.110.1782.1718.60115.56738.0745719A5Hi-TiWM0.250.870.8519.31980.000715.60230.000638.91830.001318.640.1467.8618.64315.56838.1365801A5HI-TICL0.240.840.7219.60200.000815.61970.000739.19550.001921.580.1678.2918.81815.58038.2935712A2Hi-TiWM0.200.720.8119.38340.001115.59740.000939.00930.002416.000.1259.6118.80215.56838.3225726A1HI-TITA0.230.860.9519.45940.000815.61580.000739.15060.001815.730.1160.6818.88815.58738.4515810A10Hi-TiTA0.240.780.7619.45510.000715.61410.000539.05770.001420.200.1568.7418.72115.57738.2655715A1HI-TIWM0.190.640.5219.38150.000915.60020.000838.95570.002124.150.1882.2218.50415.55638.0085715A5Low-TIWM0.090.340.4419.25200.000815.59080.000738.80960.002713.880.1051.3318.74815.56538.2185802A5Low-TiCL0.110.340.5019.17080.001215.60190.001138.68610.002414.800.1145.7918.63315.57538.1585810A6Low-TiTA0.120.280.6319.01780.000715.59800.000638.54280.001711.840.0929.8718.58815.57638.1985810A1Low-TiTA0.040.090.2319.00250.000815.59270.000738.46970.001811.640.0824.4818.58015.57138.1885727A7Low-TiTA0.100.291.2119.11900.000715.60990.000538.66670.00145.540.0416.0718.91815.60038.4815802A2Low-TiCL0.200.560.2121.69090.000915.72480.000940.51190.002562.570.45185.2119.41815.60938.3765719A1BasalWM1.914.182.9619.88550.000715.61600.000638.77670.002041.910.3094.9318.36315.53937.6825725A4Sill(Hi-Ti)TA0.250.830.6619.60920.001015.62580.000939.20930.001824.330.1884.2218.72515.58138.2385727A2Sill(Low-Ti)TA0.070.170.5518.86250.000815.60210.000738.47110.00198.050.0620.1118.57015.58738.2395727A3Sill(Low-Ti)TA0.070.170.5018.90040.000915.59980.000838.46410.00248.680.0622.0418.58515.58438.2105727A6Sill(Low-Ti)TA0.030.070.2018.90760.000715.59670.000638.49600.00178.370.0622.3618.60315.58138.2385727A6(dup)Sill(Low-Ti)TA0.030.070.1818.90150.000515.59780.000538.48830.00149.400.0725.9918.56015.58138.1895810A2Sill(Low-Ti)TA0.100.270.4419.19680.000815.60890.000738.77760.001614.210.1040.9718.68115.58338.3055810A4Sill(Low-Ti)TA0.090.300.2619.26380.000815.61030.000738.85010.001823.200.1777.0418.42115.56837.9625802A4PicriteCL0.320.890.1323.73520.001815.82550.001342.37160.0043168.731.22491.9517.60615.51436.6995808A2PlcrlteRC0.260.700.7621.20950.001415.70830.001240.55400.003023.590.1764.9620.35315.66539.8055808A3PicriteRC0.381.060.6521.70020.000915.73790.000840.54680.002439.320.29114.8020.27215.66539.223Hi-Ti,high-titaniumbasalt;Low-Ti,low-titaniumbasalt;Basal,basalflow-conglomerate”Abbreviationsforareaare:WM,WrangellMountains;TA,TangleLake;GG,GlacierGapLake;CL,ClearwaterMountains;RC,RainyCreek.(dup)indicatescompletechemistryduplicate.Alltrace-elementandisotopicanalyseswerecarriedoutatthePCIGR.TheanalyticalmethodsaredescribedinAppendixD.t’J Claumonite), primarily making up the amygdules (Stout, 1976; Smith, 1981; MacKevett eta!., 1997). Many of the Nikolai basalts in the synform in the Amphitheater Mountains areexceptionally unaltered compared to flows in the Wrangell and Clearwater Mountains.Vesicles and interpillow voids in the Amphitheater Mountains commonly remain unfilledand secondary minerals are less common.Seventeen of the 21 low-titanium basalts have LOl greater than 2.5 wt % andgreater than 8 wt % MgO, whereas only four high-titanium basalts have greater than 2.5wt % LOT and all have less than 8 wt % MgO (Fig. 3.15). Three of the four high-titaniumbasalts that lie within the alkalic field are plagioclase-rich, highly amygdaloidal, and werecollected near a mineralized area at the top of the Nikolai Formation. Tight linear arraysare apparent on plots of HFSE and REE (not shown) indicating negligible affect ofelement mobility. Only a limited group of samples (5808A3, 5802A4, 5802A2, 5725A4,5726Al) have LTLE concentrations outside the narrow range of most high- and low-titanium basalts (Fig. 3.11) and there is no correlation between LOT and L]LE. All of thelow-titanium basalts have positive Sr anomalies that are complemented with smallpositive Eu anomalies in most samples, and none of the high-titanium basalts have Sranomalies (Fig. 3.11), which indicates Sr concentrations probably represent primaryvalues. U and Th show a linear relationship, whereas Pb and Th do not show a clearrelationship (not shown), indicating some secondary mobility of Pb, especially in thelow-titanium basalts.Almost all initial Nd and Hf, and to a lesser extent Sr and Pb isotopiccompositions represent close to magmatic compositions. Several of the more alteredsamples were not selected for isotopic analyses and leaching effectively removed most ofthe secondary alteration products (Weis et a!., 2006; Nobre Silva et a!., in revision). Asingle exception is the Clearwater picrite (5802A4), which was significantly affected byPb loss and has less radiogenic age-corrected Pb isotopic ratios (Figs 3.11 and 3.14). Thecorrelation of LOT and875r/865r,206Pb/204Pb, and238U/204Pb in the low- and high-titaniumbasalts is a primary feature that is not apparent with age correction (Fig. 3.15). The rathersmall range of initial Pb and distinct initial Hf and Sr isotopic compositions for high- andlow-titanium basalts clearly reflect the isotopic compositions of the sources of Wrangelliaflood basalts in Alaska.1215. i i • . i . I I I432054320—18.5 19.••....0.704 0.705 0.706Figure 3.15 Loss-on-ignition versus MgO and isotopic ratios for the Nikolai Formation in Alaska. (a) LOT vs.MgO. (b) LOl vs. measuredt7Sr/86Sr. (c) LOT vs. measured2°6Pb/204Pb. (d) LOT vs. measured 238U/206Pb. The twoinsets show the expanded x-axis for 206PbP°4Pb and 238U/206Pb. Note the generally higher LOT and MgO for thelow-titanium basalts. The differences in measured t7Sr/t6Sr and 206Pb/204Pb within the suites of high- and low-titanium basalts are mostly not apparent after age-correction.0•aLOl (wt %)(a)••.00I I I I I I I I0 3 6 9MgO (wtLOl (wt%)4 5715A503 00002••..1•(b)I I I I I I I I I - I•high-titanium basalt• high-titanium sill0 low-titanium basalt0 low-titaniumsillA Clearwater picrite0 Rainy Creek picrite012 15 18 0.70387Sr/ 86Sr5802A2I I ILOl (wt %)02001800• I I20 22(c)••19.5206pb/204pb20 5 10 15 20 25238u/204Pb122FLOOD BASALT CHEMOSTRATIGRAPIIYSamples of the Nikolai Formation were collected going up or downsection throughthe volcanic stratigraphy to provide an estimate of the relative stratigraphic position ofeach sample and to determine the relationship between stratigraphic position andchemical composition. Figure 3.16 shows sample numbers, lithologies, relativestratigraphic height, and Ti02 and MgO contents for Nikolai basalts from the three mainareas of Alaska where fieldwork was undertaken (Fig. 3.1). Each stratigraphic column isa combination of multiple traverses (separated by dashed lines in Figure 3.16). We arecoafident in the relative position of each section of stratigraphy because of the continuousexposure and minimal disruption by faults in these areas. Since the trace-element andisotopic variation of the basalts generally correspond with variation in Ti02 (Figs 3.10-3.13), only Ti02 and MgO are shown in Figure 3.16.In the Clearwater and Amphitheater Mountains, there is a clear relationshipbetween stratigraphic position and chemical composition of the flood basalts (Fig. 3.16).The low-titanium basalts form the lowermost several hundred meters of flows (10-15% ofstratigraphy) and the high-titanium basalts form the majority of the flows (—85-90% ofstratigraphy) above the lowest several hundred meters. More of the low-titanium basaltsand sills were sampled, partly because lower sections of volcanic stratigraphy were moreeasily accessible and partly because there are more interesting relationships with preNikolai sediments and mafic sills and submarine units preserved lower in thestratigraphy. The transition from low- to high-titanium basalts does not appear tocoincide with the transition from submarine to subaerial flows, but almost all of the low-titanium basalts that were sampled are submarine flows.In the Wrangell Mountains, there do not appear to be any low-titanium basalts,except for two anomalous samples (Fig. 3.16). A single sample of the basal flowconglomerate has a low titanium content (0.67 wt %), similar initial to the hightitanium basalts, and anomalous LaJYbCN (6.4), Ba (1277 ppm), and Th (4.18 ppm; Figs3.10 and 3.13; Tables 3.2 and 3.4). Field observations and several other geochemicalcharacteristics indicate the chemistry of this basal flow-conglomerate is the result ofconsiderable assimilation (-30 vol %) of material derived from underlying Paleozoic123Figure 3.16 Chemostratigraphy of the Nikolai Formation in three areas of Alaska (Clearwater, Amphitheater,and Wrangell Mountains). Each column shows lithology, sample numbers, relative stratigraphic height, andTi02and MgO contents (in wt %). Dashed lines in each column separate individual traverses.124•5801’295801A5Cleaiwater Mountains Amphitheater Mountains5714M5715A11794A35716A1•5801A855715A5Wrangeii Mountains5716A2571603I I55501A5157310615802A115802A215802031880204158025.51550206*570*5.2•5731A5I?IIIlIIll15707A3I5710*2!571053•5731A41573103.sed*sIif5712*2I II IMgOI II,IjO 15wt%1102MgOI I III I III I II1 2 3 5 10 1506% 06%87190*8719*5•5759A11102MgO1 2 33 8 906% 06%sequences. The next sample going upsection, 2O m above the uppermost exposure ofbasal flow conglomerate in the Wrangell Mountains, does not have visible assimilatedmaterial and is high-titanium basalt with unexceptional chemistry. A single sample with alow titanium content (1.14 wt %Ti02;571 5A5) was collected from near the top of thestratigraphy in the Wrangell Mountains, but this sample has similar isotopic compositionto the high-titanium basalts, atypical petrographic texture, and is at the upper range ofTi02of low-titanium basalts.DISCUSSIONSource of Nikolai basaltsThe Nikolai Formation in Alaska has two main lava types with distinct isotopiccompositions. The high-titanium basalts in Alaska have depleted Hf and Nd isotopiccompositions that are not as depleted as most Pacific and Indian mid-ocean ridge basalts(MORB) and are displaced just below the ocean island basalt (OIB) mantle array (Fig.3.17). The high-titanium basalts have similar initial Sr and Nd isotopic compositions toOntong Java Plateau, Hawaii, and Caribbean Plateau basalts and similar intial toOntong Java, with slightly lower initial8Hfthan most Hawaii and Caribbean basalts (Fig.3.17). In contrast, the low-titanium basalts are displaced well above the 0113 mantle arrayin a EHf(t)-ENd(t) correlation diagram and have a similar range of initial8Wto PacificMORB, with initial8Nd2 to 5 epsilon units lower than Pacific MORB. The Hf isotopiccompositions of the low-titanium basalts are 2 to 6 epsilon units higher than for mostsamples from Ontong Java, with slightly lower initial8Nd.Sr isotopic compositions forlow-titanium basalts extend to significantly higher initial 87Sr/86Srthan Ontong Java andHawaii. Three low-titanium basalts with particularly high initial ENd lie within a field forIndian MORB in EHf(t)-ENd(t) space. Two Rainy Creek picrite samples lie close to theEHf(t)-ENd(t) OIBmantle array with lower initial6Ndthan the high-titanium basalts (Fig.3.17). The high- and low-titanium basalts have uniform initial Pb isotopic compositionsthat overlap a field for Caribbean basalts and have more radiogenic initial 207Pb/204Pbthan Ontong Java, Hawaii, and a field for the East Pacific Rise (EPR; Fig. 3.18). Pbisotopic compositions form a linear trend in 208Pb-206Pbspace which intersects the fieldof Pacific MORB compositions and is slightly offset towards lower 208Pb/204Pbfrom1250.702 0.703 0.70487SrI Sr(1)ENdFigure 3.17 Comparison of age-corrected (230 Ma) Sr-Nd-Hf isotopic compositions for Nikolai basalts inAlaska to age-corrected OIB and MORB. (a) Initial vs. 87Sr/86Sr. (b) Initial vs.NdBoth fields withdashed lines are Indian MORB. All of the references for the compiled data are too numerous to cite here.Most of the compiled data was extracted from the GEOROC database (http://georoc.mpch-mainz.gwdg.de/georoc/). Data for Ontong Java from Mahoney et al. (1993), Babbs (1997), and Tejada et a!. (2004); IndianMORE from Salters (1996), Kempton et a!. (2002), and Janney et a!. (2005); Pacific MORE fromMahoney et a!. (1992, 1994), Nowell et a!. (1998), Salters and White, and Chauvel and Blichert-Toft(2001); Explorer Ridge data from Cousens and Weis (pers. comm., 2007); O]B array line from Vervoort(1999). EPR is East Pacific Rise. Dashed lines indicate Bulk Silicate Earth (BSE). An extended referencelist is available upon request.tI I• I IJ1 I I I—II..II.••.I.I• SI.x ‘.i• I..• East PadficRe• • Ontongiava • .1• Hawaii•I•Nikolai high-titanium basaIt•e•• •Nikolai low-titanium basalt%.I•Nikolai picrite.1XNikoiaibasalflow(WrangeilMountains) — — —I—(a)• • . • I • • • I • . • • I . •1412108(I) 6Nd420-2222018161412EHf1:6420-2-40.705 0.706-4 -2 0 4 6 8 10 12 1412615.65I I I II II I I II II II I I I I p I I I I15.6015.5515.5015.4515.4017.5 19.5 20.020.0Figure 3.18 Comparison of Pb isotopic compositions of the Nikolai Formation in Alaska to OIB andMORB. (a) Measured and initial 207Pb/204Pb vs. 206PbP°4Pb. (b) Measured and initial 208Pb/204Pb vs.206Pb/204Pb. See figure caption for Figure 3.17 for references for OIB and MORB data.:..60• East Pacific RiseJuan de Fuca/Gorda• Explorer RidgeCaribbean Plateau• Ontong JavaHawaii•Nikolai Formation (initial)oNikolai Formation (measured) -ii ii I I Ii I I ii II II Ii I I•p‘a -•(a)18.0 18.5 19.039.539.038.538.037.517.5127Ontong Java and Hawaii. The Pb in Nikolai basalts is more enriched in radiogenic Pbthan MORB, Ontong Java, and OIB from Hawaii, but similar to basalts of the CaribbeanPlateau.The initial Hf and Nd isotopic compositions of high-titanium basalts indicate auniform OIB-type Pacific mantle source derived from a long-term depleted source,distinct from the source of MORB. In contrast, the low-titanium basalts have initial Hfisotopic compositions that are clearly distinct from OIB and initial Nd isotopiccompositions that are distinct from the Pacific MORE source. The displacement of thelow-titanium basalts well above the OIB array indicates involvement of a depletedcomponent (mantle or crust), distinct from depleted MORE mantle, early in the formationof Nikolai basalts in Alaska. The origin of the distinct isotopic and geochemical signatureof the low-titanium basalts is the focus of subsequent discussion sections.Lithospheric involvement in derivation of the low-titanium basaltsThe stratigraphic relationship of the two contrasting lava types in the NikolaiFormation preserve a record of a shift in composition of the source of Nikolai basalts inAlaska and provide a rare opportunity to evaluate the role of oceanic arc lithosphere inthe formation of an oceanic plateau. Thus far, the low-titanium basalts have primarilybeen recognized in the lowermost part of the stratigraphy in the western part ofWrangellia in Alaska, where there is a substantial section of submarine flows (5OO m).The low-titanium basalts possess distinct negative-HFSE anomalies in normalized trace-element patterns and have high initial8Hfand high initial 87Sr/86Srcompared to the high-titanium basalts. This compositional and stratigraphic evidence suggests that theunderlying Paleozoic arc lithosphere may have played a significant role in the generationof early-erupted low-titanium basalts in the Wrangellia oceanic plateau of Alaska.Nature ofunderlying Paleozoic arc lithosphereThe Paleozoic arc (320-285 Ma) and marine sedimentary sequences (EarlyPermian to Middle Triassic) exposed underlying the Nikolai basalts in Alaska are >2.5km thick in areas. Recent geophysical studies in southern Alaska by Saltus et al. (2007)indicate Wrangellia crust is at least 50 km thick between the Denali and Border Ranges128Faults (Fig. 3.1). The arc crust that the Wrangellia oceanic plateau was built upon mayhave been 20-30 km thick and this would have included a substantial sub-arc mantlelithosphere that was metasomatized during arc activity.In the Alaska Range, the Nikolai basalts are underlain, in decreasing order ofdepth, by the Paleozoic Tetelna Volcanics, the Slana Spur and Eagle Creek Formations.Tetelna Volcanics (<1000 m) are anclesitic and dacitic flows, tuffs interbedded withvolcaniclastic rocks, and debris-flow deposits; the Slana Spur Formation (P4400 m) ismarine voicaniclastics, with lesser limestone and sandstone; and the Eagle CreekFormation (900 m) is Permian argillite and limestone (Nokieberg et a!., 1985).Numerous comagmatic intermediate to felsic piutonic rocks intrude Teteina Volcanicsand the Sian Spur Formation (Nokleberg eta?., 1994). In the Wrangell Mountains, thePaleozoic sequences include the Station Creek Formation (—P1200 m of mostly basalticand andesitic flows and 800 m of volcaniclastic sequences) and the sedimentary HasenCreek Formation (500 m of chert, black shale, sandstone, bioclastic limestone, andconglomerate) (Smith & MacKevett, 1970; Fig. 3.2).Trace-element and isotopic source constraints ofthe low-titanium basaltsThe trace element and isotopic compositions of the early-erupted low-titaniumbasalts are not typical of OIB and indicate involvement of a HFSE-depleted componentthat was different than the plume-type source of the high-titanium basalts. The arclithosphere is a key suspect for derivation of the low-titanium basalts because: 1) thegeochemical and isotopic signature of the low-titanium basalts is similar to rocks formedin subduction settings (e.g. Kelemen et a?., 2003); 2) arc crust is exposed beneath theNikolai basaits in Alaska; and 3) the low-titanium basalts only form -4 0-15% of thelowest part of the volcanic stratigraphy.Figure 3.19 highlights differences in trace elements and isotopic compositionsbetween the high- and low-titanium basalts and indicates a strong arc lithospheric, orsubduction-modified mantle, component in the low-titanium basalts. The high-titaniumbasaits form a concentrated cluster of points in each of the plots and show a remarkablysmall degree of variation, whereas the low-titanium basalts have a noticeablywider rangeof variation, which mostly does not overlap the range for the high-titaniumbasaits (Fig.129‘b.o 0.2 0.4 0.6 0.8Nb/LaPM ciciI1Jci(d)4 8 12Y/N b10 20 30 40Sr/NdFigure 3.19 Trace-element ratios and isotopic compositions of the Nikolai Formation in Alaska. (a) 176Lu/’77Hfvs.Nb/Th. (b) Initial5Ndvs. NbJTh. (c) Initial s vs. Nd/Zr. (d) Zr/Nb vs. YINb. (e) Initial vs. Nb/La. (f) Initial EHf VS.Sr/Nd. Primitive mantle (PM) from McDonough and Sun (1995), depleted mantle (DM) trace-element and isotopiccomposition estimates from Salters and Stracke (2004). Talkeetna arc lower crust compositions from Greene et a!.(2006) and Tailceetna arc lavas from Clift et aL (2005). Dashed circle in each panel outlines samples 5727A2,5727A3, and 5802A5.I III I I I0.08176Lu/177HfbaIflow(Amph.Mtns)Az.0.00• E(I)NdTalkeetna arclower crust8Talkeetna arclavas:::Nb/Th10I III I I I I I I I I I I I Ibasal flow (Amph. Mtns)Nb/Th10 15EHf:(c)PM:• •raiketnac •ç ) ••Z /NblavasciciciAxI I IIIIIIITalkeetna arcilower crust Ici6 ‘—‘3224168F18161412108a12018161412108)8 0.11 0.14 0.17 0.20 0.23 0.Nd/Zrcici•high-titanium basalt• high-titanium sillci low-titanium basalt*low-titanium sillA Clearwater picriteG Rainy Creek picriteX basal flow (Wrangell Mtns)I I I I I I I I I16 20 2 4I II I I IIIII I II:EHf(i) c:--. cici•A•()I I I I I I I 1I I Ir/Nd114• I I18EHf1614121o(f) PM61.4 0I I I I I I9ci• I I I I• I I I ITalkeetna arclower crust—5802A2,ci.ciI I I I1.0 1.2 50 60 70 801303.19). The low-titanium basalts have low Nb/Th and Nb/La relative to primitive mantle,which is characteristic of subduction-related rocks (Pearce, 1982). Except for threeanalyses from the basal flow in the Amphitheater Mountains, which has pelagic sedimentbetween pillow tubes derived from the directly underlying strata and has similar Nb/Th tothe high-titanium basalts, the low-titanium basalts have similar Nb/Th and Nb/La toaccreted arc crust from the Early Jurassic Talkeetna arc exposed in southern Alaska ‘-‘50km south of the Amphitheater Mountains (Fig. 3.19; Greene eta!., 2006). Low Nb/Th inarc magmas is commonly attributed to inheritance from subducted sediments (e.g.Kelemen eta!., 2003). Low Nb/La may be related to a process whereby migration of REEinto magma takes place, but mobilization of Nb is inhibited, such as by reaction betweenmagmas and metasomatized peridotite (e.g. Kelemen et al., 1990, 1993). The low-titanium basalts have high Sr/Nd and Nd/Zr relative to primitive mantle and the high-titanium basalts (Fig. 3.19). Elevated Sr relative to REE may indicate addition of Sr to arclithosphere through aqueous fluids, since Sr is more soluble than REE at high pressure(e.g. Johnson & Plank, 1999) or addition from Sr-enriched cumulates from gabbroiclower crust (e.g. Kelemen et a!., 2003). The trace-element and isotopic compositions ofthe low-titanium basalts have an arc-type geochemical signature.Figure 3.20 utilizes proxies described by Pearce (2008) for identifyinglithospheric input (Th-Nb) and assessing depth of melting (Ti-Yb). For the Th-Nb proxy,all the high-titanium basalts lie within a diagonal MORB-OIB array parallel to a meltingvector, whereas most of the low-titanium basalts are displaced above the array, oblique tothe melting vector. The low-titanium basalts follow a trend for lavas that have asubduction component, or have interacted with continental crust, and they are consistentwith a small amount of assimilation (F<0.9; F is melt fraction) combined with fractionalcrystallization (AFC), shown by the modeling curve of Pearce (2008; Fig. 3.20). A NbTh depleted source is indicated for the low-titanium basalts, which also have similar ThNb to Talkeetna arc lavas and lower crust. For the Ti-Yb proxy, high Ti/Yb ratios forhigh-titanium basalts indicate residual garnet from melting at high pressure, within theOIB melting array, whereas low-titanium basalts lie along a complementary mantle meltdepletion trend indicating shallow melting, similar to compositions of Talkeetna arc lavas(Pearce, 2008; Fig. 3.20).131Figure 3.20 Th-Nb and Ti-Yb proxies of the Nikolai Formation in Alaska with data compilation and modeling resultsfrom Pearce (2008). See Pearce (2008) for full reference list used to create MORB and OIB arrays, and OIB averagein panel b. Primitive mantle (PM) estimate in panel a from McDonough & Sun (1995). (a) Th/Yb vs. Nb/Yb. MORBOIB array and assimilation-fractional crystallization (AFC) model from Pearce (2008). (b) Ti02/Yb vs. Nb/Yb.Talkeetna arc lower crust from Greene et a!. (2006) and Tailceetna arc lavas from Clift et a!. (2005). Mariana arc datafrom Pearce et a!. (2007) and Woodhead et a!. (2001). The low-titanium basalts indicate a depleted source andinteraction with a subduction component combined with fractional crystallization, whereas the high-titanium basaltslie within an OIB array in panel b, parallel to a melting vector that indicates higher pressure melting. See Pearce(2008) for parameters of polybaric melting and assimilation-fractional crystallization (AFC) modeling. Blue line inpanel A represents an AFC model following the modeling of DePaolo (1981). Red line in panel b illustrates apolybaric melting trend (with changing composition of pooled melt extracted from the mantle that undergoesdecompression from the solidus to the pressure marked) for high and lower mantle potential temperatures whichcorrespond to representative conditions for the generation of present-day MORE and OIB (Pearce, 2008).Nb/Yb Nb/Yb132Lassiter et a!. (1995) suggested a minor role for the arc lithosphere in formationof the Wrangellia flood basalts based on a suite of nine samples from the WrangellMountains in Alaska. They inferred that assimilation of low ENd, low Nb/Th arc materialmay have affected the composition of the Wrangellia basalts, but that mixing of MORBmantle with low5Ndarc material did not reproduce the trends in the Wrangellia basalts.Rather, Lassiter et al. (1995) suggested mixing of a plume-type source, with8Nd+6 to +7,with arc material with low Nb/Th could reproduce variations in the Wrangellia floodbasalts. They noted that the absence of low Nb/La ratios in flood basalts from theirdataset suggests a restricted amount of lithospheric involvement. The lower FeO contentfor most of the low-titanium basalts also may reflect melting generated from refractoryarc lithophere (Lassiter & DePaolo, 1997). The widespread sampling of Nikolai basalts inAlaska in this study supports the interpretation of Lassiter et a!. (1995), that arclithosphere was involved in formation of the Nikolai basalts in Alaska.The low-titanium basalts may have developed an arc-type signature by substantialmelting of subduction-modified mantle, interaction of plume-derived melts with melts ormaterial derived from the arc lithospheric, and/o