UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Eocene volcanic response to the tectonic evolution of the Canadian Cordillera Bordet, Esther-Jeanne 2014

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

Item Metadata

Download

Media
24-ubc_2014_spring_bordet_esther.pdf [ 75.09MB ]
Metadata
JSON: 24-1.0166874.json
JSON-LD: 24-1.0166874-ld.json
RDF/XML (Pretty): 24-1.0166874-rdf.xml
RDF/JSON: 24-1.0166874-rdf.json
Turtle: 24-1.0166874-turtle.txt
N-Triples: 24-1.0166874-rdf-ntriples.txt
Original Record: 24-1.0166874-source.json
Full Text
24-1.0166874-fulltext.txt
Citation
24-1.0166874.ris

Full Text

   EOCENE VOLCANIC RESPONSE TO THE TECTONIC EVOLUTION OF THE CANADIAN CORDILLERA  by  Esther-Jeanne Bordet B.Sc., Université Pierre et Marie Curie, 2004 M.Sc., Université du Québec, Institut National de la Recherche Scientifique, 2007   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in The Faculty of Graduate and Postdoctoral Studies (Geological Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  March 2014  © Esther-Jeanne Bordet, 2014ii  Abstract The Ootsa Lake Group (OLG) represents a voluminous episode of Eocene volcanism across the Interior plateau of British Columbia (BC), in the Canadian Cordillera. Remarkable aspects of the rocks (lithology, texture, volume, extent, and geochemistry) suggest that even though they formed along an active continental margin, the tectonic setting was different from a classic arc. The OLG was defined by field mapping, U-Pb and 40Ar/39Ar geochronology, major and trace elements geochemistry, and three-dimensional modelling of the thickness and structure. A new tectonic model for the evolution of the Canadian Cordillera in the Paleogene is proposed using this comprehensive dataset. The OLG stratigraphy comprises a thick sequence of rhyolite and dacite lava, locally capped by andesite. Onset, duration and termination of volcanism are equivalent across the Interior plateau, and are constrained between 54.7 and 46.6 Ma by new U-Pb and 40Ar/39Ar geochronology.  OLG lavas yield a “volcanic arc” signature (diagnostic high-K calc-alkaline trend, and trace and rare earth elements patterns), suggesting supra-subduction zone contributions from a hydrated mantle wedge. However, a similar signature may be inherited from partial melting of crustal reservoirs composed of older accreted volcanic arc crust. This is supported by Sr isotope data indicating variable crustal contributions to melts across BC. OLG intermediate rocks were likely derived from mantle melting, but dominantly silicic compositions support partial melting of the crust as a dominant magma producing mechanism. Eocene volcanic rocks cover at least 65,000 km2 of BC, but their original extent may have been almost continuous from southwestern Yukon to Idaho. Coeval volcanism and extensional deformation contributed to the accumulation and preservation of volcanic products in extensional basins, up to 4000 m thick in some locations. With such dimensions, the OLG may have attained the status of a Silicic Large Igneous Province prior to erosion.  The cause of OLG volcanism and coeval extension is attributed to the sudden ingress of hot sublithospheric mantle within a previously metasomatized mantle region, following cessation of subduction and a slab break. The resulting thermal anomaly progressed across a “slab gap” beneath BC, leading to mantle and lower crustal melting, crustal anatexis and magmatism.  iii  Preface  All the work presented in this dissertation is the result from my own work carried out at The University of British Columbia, which also included field mapping in central British Columbia. Craig J.R. Hart, supervisor, and Mitchell G. Mihalynuk, co-supervisor, provided advice, supervision and editorial co-authorship. Jim K. Mortensen and Kelly J. Russell, committee members, contributed to comments and suggestions during committee meetings, and in personal discussions related to their fields of expertise (physical volcanology, geochronology). They also provided editorial feedback on initial drafts of the published papers, or directly contributed to some of them (see below).  The research technical objectives were identified following discussions between Craig J.R. Hart and myself during 2009. These objectives were designed to accommodate my research interests and technical skills, and as a response to a Request for Proposal issued by Geoscience BC in July 2009 to stimulate exploration activity and attract oil and gas investment in the Nechako Basin, British Columbia. A preliminary project report entitled“Preliminary lithological and structural framework of Eocene volcanic rocks in the Nechako Region, central British Columbia” was published in 2011. This report was entirely prepared by me, with advice and editorial supervision from Craig J.R. Hart, except for sections 5.3.1 and 5.3.2, which were prepared by Dianne Mitchinson. Components and datasets from this report are used to develop argumentations and interpretations in this dissertation, but none of the sections are literally reproduced. References to this report in the dissertation appear as Bordet et al. (2011b). Chapters 1, 4 and 7 of this dissertation are traditional thesis chapters. They represent my original work, and figures are original creations, or modified from accordingly cited sources. These chapters provide articulations and supplementary information between manuscript-based chapters 2, 3, 5 and 6. The repetition of information between manuscript-based chapters was inevitable, but the core ideas developed in each chapter are unique and do not overlap. Minor changes were applied to the originally submitted manuscripts to comply with UBC thesis requirements. Chapter 2 comprises a manuscript entitled “Stratigraphy of a voluminous silicic-intermediate volcanic sequence, Eocene Ootsa Lake Group, south-central British Columbia”, by Bordet, E., Hart, C.J.R. iv  and Mihalynuk, M.G. I am the lead author on this paper, and conducted field work, collected samples, and carried out all analytical studies. In addition, I designed the bulk of the text, figures, and tables. Co-authors provided technical advice and suggestions both in the field and during manuscript preparation. In addition, the manuscript benefited from improvements as suggested by committee members and an external reviewer. The Online Supplementary Material is entirely reproduced in Appendices B, C and E.  Chapter 3 reproduces integrally the manuscript “Chronostratigraphy of Eocene volcanism, central British Columbia”, by Bordet, E., Mihalynuk, M.G., Hart, C.J.R., Mortensen, J.K., Friedman, R.M., and Gabites, J. This manuscript was published in January 2014. I am the lead author on this paper, and am responsible for sample collection, design of the study and composition of the paper. Members of the supervisory committee are co-authors, and provided feedback, advice and editorial supervision throughout the preparation of the manuscript. Co-authors J.K. Mortensen, R.M. Friedman and J. Gabites were responsible for the geochronological analyses at the Pacific Centre for Isotopic and Geochemical Research, and I partly contributed to the sample processing and analyses under their supervision. They also assisted me with data interpretation, and provided feedback, advice and editorial suggestions during the preparation of the manuscript. Kelly J. Russell edited a final version of the manuscript before its submission, and his suggested changes were integrated to the manuscript. Complete data tables and analytical plots are entirely reproduced in thesis Appendix F.  Chapter 5 is a peer-reviewed technical paper entitled “Three-dimensional thickness model for the Eocene volcanic sequence, Chilcotin and Nechako plateaus, central British Columbia (NTS 92O, 92P, 93A, 93B, 93C, 93G, 93F, 93E, 93K, 93L)”, by Bordet, E., Mihalynuk, M.G., Hart, C.J.R. and Sanchez, M. It is reproduced entirely in its final published version, as of January 2014. I am the lead author on this paper, and am responsible for design of the model, analysis of the results, and composition of the paper. M.G. Mihalynuk originally proposed the methodology used in this chapter. Matias Sanchez is a research associate at the Mineral Deposit Research Unit, who guided structural interpretations throughout the research. All co-authors provided feedback, advice and editorial suggestions throughout the preparation of the manuscript.  Chapter 6, entitled “Eocene trigger to Eocene silicic-intermediate magmatism in the Canadian Cordillera”, is intended for submission in a peer-reviewed international journal. I am the lead author, and Craig. J.R. Hart and Mitchell G. Mihalynuk are co-authors. As the first author I v  designed the paper outline and conducted all the research, data compilation and data interpretation, with advice from my supervisory committee.  In addition, the following contribution was derived from work presented in this dissertation: “Bordet E. and Hart, C.J.R. 2011. Characterization and structural framework of Eocene volcanic sequences in the Nechako region, central British Columbia (NTS 092N, O, 093B, C, G). In Geoscience BC Summary of Activities 2010, Geoscience BC, Report 2011-1, p. 239–254.”   vi  Table of Contents Abstract ............................................................................................................ ii Preface ........................................................................................................... iii Table of Contents ........................................................................................... vi List of Tables ................................................................................................xiii List of Figures .............................................................................................. xvi List of Abbreviations ................................................................................ xxxvi Glossary .................................................................................................... xxxix Acknowledgements ....................................................................................... xli Dedication ...................................................................................................xliii Chapter 1 Introduction .................................................................................... 1 1.1. Paleogene tectonics, magmatism and deformation in the North American Cordillera ......................................................................................................................... 1 1.1.1. Tectonic models ...................................................................................................................... 1 1.1.2. Magmatism along the North American active continental margin .................................. 5 1.1.3. Structural elements ................................................................................................................ 12 1.2. Research questions ................................................................................................. 14 1.2.1. Eocene volcanic rocks as records of the Paleogene tectonics of western North America ............................................................................................................................................. 14 1.2.2. Ambiguous magmatic sources and evolution ................................................................... 15 1.2.3. Controls on the style of extensive, silicic volcanism ........................................................ 15 1.2.4. Eocene volcanic chronostratigraphy of central BC ......................................................... 16 1.3. Research objectives ................................................................................................ 17 vii  1.3.1. Chronostratigraphy of the OLG ......................................................................................... 17 1.3.2. Geochemical and physical properties of Eocene OLG lavas ......................................... 18 1.3.3. Thickness and regional structural framework of the Eocene volcanic sequence ........ 18 1.3.4. Paleogene tectonic and magmatic evolution of central BC ............................................ 18 1.4. Scientific and economic implications .................................................................... 19 Chapter 2 Stratigraphy of a voluminous silicic-intermediate volcanic sequence, Eocene Ootsa Lake Group, south-central British Columbia ..... 23 2.1. Introduction ............................................................................................................ 23 2.2. Definition of the OLG ............................................................................................ 26 2.3. Tectonic and geological setting for OLG volcanism ............................................. 27 2.4. Field and analytical methods ................................................................................. 28 2.5. OLG map units ....................................................................................................... 33 2.5.1. Tibbles rhyolite ...................................................................................................................... 34 2.5.2. Clisbako rhyolite .................................................................................................................... 41 2.5.3. Chezacut dacite sequence .................................................................................................... 44 2.5.4. Baezaeko dacite ..................................................................................................................... 46 2.5.5. Mount Sheringham dacite .................................................................................................... 49 2.5.6. Nechako andesite .................................................................................................................. 50 2.5.7. Summary of the OLG stratigraphy .................................................................................... 53 2.6. Major element geochemistry .................................................................................. 53 2.7. Thickness of the Eocene sequence ........................................................................ 54 2.8. Discussion .............................................................................................................. 57 2.8.1. Extensive areal distribution vs high viscosity of silicic lavas .......................................... 57 2.8.2. Thickness variations and structural controls ..................................................................... 59 2.8.3. Fissure-related silicic volcanism .......................................................................................... 59 2.9. Conclusion .............................................................................................................. 61 Chapter 3 Chronostratigraphy of Eocene volcanism, central British Columbia ....................................................................................................... 62 3.1. Introduction ............................................................................................................ 62 3.2. Tectonic setting ...................................................................................................... 67 viii  3.3. Regional geology .................................................................................................... 68 3.3.1. Basement and cover strata ................................................................................................... 68 3.3.2. Ootsa Lake and Endako groups ......................................................................................... 69 3.4. Eocene stratigraphy and geochronology ............................................................... 70 3.4.1. Lithologies and sampling framework ................................................................................. 70 3.4.2. Summary of analytical methods .......................................................................................... 72 3.4.3. Relevance of preexisting Eocene ages ............................................................................... 76 3.5. Results .................................................................................................................... 78 3.5.1. 40Ar/39Ar data ......................................................................................................................... 78 3.5.2. U-Pb data ............................................................................................................................... 78 3.5.3. Age of OLG lithostratigraphy ............................................................................................. 80 3.6. Discussion .............................................................................................................. 93 3.6.1. Age of OLG volcanism ........................................................................................................ 93 3.6.2. Thickness and eruption volume .......................................................................................... 94 3.6.3. Lateral compositional variations and temporal trends .................................................... 95 3.6.4. Implications for the Eocene tectonic setting .................................................................... 95 3.7. Conclusions ............................................................................................................ 97 Chapter 4 Geochemical and physical properties of Eocene Ootsa Lake Group lavas .................................................................................................... 99 4.1. Introduction ............................................................................................................ 99 4.2. Methods .................................................................................................................. 99 4.2.1. Sample selection .................................................................................................................... 99 4.2.2. Analytical procedures ......................................................................................................... 100 4.2.3. Data processing ................................................................................................................... 102 4.3. Geochemical signature and classification of OLG volcanic rocks ....................... 103 4.3.1. Rock classification ............................................................................................................... 103 4.3.2. Major elements compositional trends .............................................................................. 104 4.3.3. Trace, refractory and REE trends .................................................................................... 108 4.4. Physical properties of OLG magmas using geochemical compositions .............. 113 4.4.1. Eruption temperatures ....................................................................................................... 113 ix  4.4.2. Prediction of viscosities ..................................................................................................... 121 4.5. Discussion ............................................................................................................. 125 4.5.1. Regional geochemical patterns .......................................................................................... 125 4.5.2. Literature review of radiogenic isotopes geochemistry in Eocene rocks .................... 128 4.5.3. Magma source(s) and evolution ........................................................................................ 133 4.5.4. Tectonic setting ................................................................................................................... 138 4.6. Conclusion ............................................................................................................. 143 Chapter 5 Three-dimensional thickness model for the Eocene volcanic sequence ....................................................................................................... 146 5.1. Introduction ........................................................................................................... 146 5.2. Regional geology ................................................................................................... 150 5.3. Eocene structures .................................................................................................. 151 5.4. Methods and model constraints ............................................................................ 153 5.4.1. Thickness model ................................................................................................................. 153 5.4.2. Structural interpretations ................................................................................................... 153 5.5. Results ................................................................................................................... 154 5.6. Discussion ............................................................................................................. 157 5.6.1. Limitations of the model.................................................................................................... 157 5.6.2. Distribution and volume of the Eocene volcanic sequence ......................................... 158 5.6.3. Structural controls on regional variations in thickness .................................................. 159 5.6.4. Implications for exploration .............................................................................................. 160 Chapter 6 Eocene trigger to extensive silicic-intermediate magmatism in the Canadian Cordillera ............................................................................... 161 6.1. Introduction ........................................................................................................... 161 6.2. Is the OLG part of a Silicic Large Igneous Province? .......................................... 164 6.3. Tectonic constraints on the onset of OLG magmatism ....................................... 169 6.3.1. Constraints on tectonic plate and subduction geometry ............................................... 169 6.3.2. Slab discontinuities as a trigger to extensive volcanism ................................................ 173 6.4. Source(s) and evolution of OLG magmas ............................................................. 179 x  6.5. Mutual temporal and spatial controls between extension and volcanism, and effects on eruption style ............................................................................................... 183 6.6. Conclusion ............................................................................................................. 187 Chapter 7 Conclusion ................................................................................... 189 7.1. Summary of research findings ............................................................................... 189 7.2. Scientific significance and contribution of the research ....................................... 191 7.2.1. Geological contributions .................................................................................................... 191 7.2.2. Contribution to geophysical surveys interpretations and geophysical inversion ....... 193 7.2.3. Implications for mineral and hydrocarbon resources exploration............................... 193 7.3. Future research directions ..................................................................................... 194 7.3.1. Radiogenic isotope analyses............................................................................................... 194 7.3.2. Comprehensive studies of epithermal Au-Ag mineral occurences .............................. 194 7.3.3. Integrated regional mapping and resource exploration programs ............................... 195 7.4. Concluding remark ................................................................................................ 195 Bibliography ................................................................................................. 197 Appendix A  Field database ...................................................................... 221 Appendix B  Description and field photographs of Ootsa Lake Group map units  ............................................................................................ 243 Appendix C Petrographic microphotographs of selected Ootsa Lake Group map units ......................................................................................... 274 Appendix D  Field maps and stratigraphic columns ............................... 282 Appendix E Physical properties .............................................................. 302 Magnetic susceptibility ............................................................................................... 302 Density ........................................................................................................................ 303 Appendix F Geochronology .................................................................... 322 xi  Analytical methods ...................................................................................................... 322 Data tables and plots................................................................................................... 326 Appendix G Supplementary data and discussion on well logs stratigraphy and geochronology ...................................................................................... 369 Background and problems .......................................................................................... 369 Wells stratigraphy........................................................................................................ 370 Wells geochronology ................................................................................................... 374 Discussion ................................................................................................................... 382 Appendix H Scanning Electron Microscopy-Cathodoluminescence imagery (SEM-CL) ...................................................................................... 384 Method ........................................................................................................................ 384 Appendix I Geochemistry ...................................................................... 392 Data tables .................................................................................................................. 392 Analysis of data quality (QAQC) ................................................................................ 400 Appendix J Structural interpretations .................................................... 408 Structural interpretation workflow .............................................................................. 408 Appendix K Contributions to mineral exploration and Cordilleran tectonics conferences ................................................................................... 416 Epigenetic metallogeny of the Nechako Region, central British Columbia .............. 416 Characterization and structural framework of Eocene volcanic sequences                            in the Nechako Region, central British Columbia ..................................................... 422 Characterization and structural framework of Eocene volcanic sequences in the Nechako Region of central British Columbia ............................................................ 423 Epithermal-style Au-Ag mineralization in Cretaceous to Eocene felsic volcanic complexes, central British Columbia, western Canada .............................................. 425 Eocene volcanic and structural framework of the Nechako Region, central BC ...... 432 Eocene volcanic and structural framework of central British Columbia: insights for the tectonic evolution of the Canadian Cordillera ............................................................ 434 xii  Structural and stratigraphic framework for the Eocene volcanic rocks in central British Columbia ..................................................................................................................... 436 Eocene tectonic and magmatic evolution of central British Columbia ..................... 438 Stratigraphic framework for Late Cretaceous and Eocene Au-Ag mineralization                in central British Columbia ......................................................................................... 442 Tectonic trigger to Eocene volcanism in the Canadian Cordillera ............................ 444   xiii  List of Tables Table 1.1. Previous work and summary of existing data in central BC ............................................ 21 Table 2.1. Compiled thicknesses for the main Eocene Groups in central and southern BC ........ 24 Table 2.2. Synthesis of the lithological and analytical characteristics of the main OLG map units. References in table are: (1) Anderson and Snyder 1998; (2) Andrew 1988; (3) Armstrong 1949; (4) Barnes and Anderson 1999a,b; (5) Bordet et al. 2011b; (6) This study; (7) Dostal et al. 2001; (8) Faulkner 1986; (9) Grainger et al. 2001; (10) Haskin et al. 1998; (11) Metcalfe et al. 1997; (12) Mihalynuk et al. 2008c; (13) Mihalynuk et al. 2009b; (14) Rouse and Mathews 1988, 1989; (15) Tipper 1959. ...... 32 Table 3.1. Compositional and textural groups and lithologies forming the Eocene stratigraphy in central BC, as well as their age range and distribution. ............................................. 73 Table 3.2. Geochronological samples description and corresponding OLG lithology. Correlation between original sample numbers and simplified codes is indicated. Simplified sample numbers are formed of a two-letters prefix representing the mapping area, followed by a code indicating the dating method (ar = 40Ar/39Ar; z = U-Pb; dz = U-Pb on detrital zircon), and a sequential number (1 to 27). .............. 74 Table 3.3. Compiled 40Ar/39Ar geochronological ages and summarized interpretations. .............. 81 Table 3.4. Compiled U-Pb geochronological ages and summarized interpretations. ..................... 82 Table 4.1. Compilation of Sri for Eocene localities in BC and other volcanic provinces worldwide. Fields in white, bold letters correspond to BC localities, field in grey are from volcanic regions selected for comparison. References: (1) Brooks et al. 1976; (2) Rollinson 1993, and references therein; (3) Dickinson 1975; (4) Preto et al. 1979; (5) Armstrong 1988; (6) Petö 1974, and references therein; (7) Ewing 1981a; (8) Dostal et al. 2001; (9) Morris et al. 2000; (10) Lanphere et al. 1980; (11) Morris and Creaser 2003; (12) Dostal et al. 2003; (13) Ghosh 1995; (14) Pankhurst et al. 1998. ........................................................................................................ 132 xiv  Table 5.1. Datasets and sources used for building the thickness model of the Eocene volcanic sequence, Chilcotin and Nechako plateaus. ................................................................. 148 Table 5.2. Surface-area calculations for the Eocene volcanic sequence, for British Columbia and the thickness model area. Calculations are based on summation of polygon areas using Manifold™. ................................................................................................... 159 Table 6.1. Geological and tectonic criteria of SLIPs, with specific examples from the SMO, compared to compiled data from Eocene volcanic rocks in central BC (see references in text). SMO data compiled after Cameron et al. (1980), Cameron and Hanson (1982), Coffin and Eldholm (1993, 1994), Albrecht and Goldstein (2000), Bryan et al. (2002), Ferrari et al. (2002), Aguirre-Diaz and Labarthe-Hernandez (2003), Mason et al. (2004), Annen et al. (2006), Bryan (2007), Bryan et al. (2008), Bryan and Ernst (2008), Pankhurst et al. (2011), Bryan and Ferrari (2013).................................................................................................................................. 164 Table A 1. Field stations numbers and coordinates.......................................................................... 222 Table A 2. List of samples and analytical work. Geochronology samples indicate the method (U-Pb or 40Ar/39Ar), followed by the mineral dated (Zr = zircon, Bt = Biotite) ... 237 Table A 3. Summary of field traverses and corresponding number of outcrops, geophysical surveys and wells .............................................................................................................. 242 Table E 1. Calculated average magnetic susceptibilities per outcrop (Unit: 10-3 SI) .................... 307 Table E 2. Density laboratory measurements (unit = g·cc-1) .......................................................... 317 Table F 1. 40Ar/39Ar data tables. Abbreviations are: r.i., linear regression coefficient; f39Ar, % of 39Ar; NPT, Normal Pressure and Temperature ...................................................... 327 Table F 2. U-Pb LA-ICP-MS analytical data reported with 1σ error. ............................................ 355 Table F 3. U-Pb LA-ICP-MS detrital zircon analytical data reported with 1σ error. .................. 360 xv  Table F 4. U-Pb TIMS data reported with 2σ error. Abbreviations are: % err., percentage error; corr. coef., correlation coefficient ....................................................................... 365 Table F 5. KL standard analytical results (reported with 1σ error) ................................................ 367 Table G 1. Reported U-Pb zircon ages for wells B-22-K and B-16-J (adapted from Riddell 2010). Sample descriptions available in Riddell (2010). .............................................. 375 Table G 2. Selected intervals and results of independent geochronological analyses of B-22-K and B-16-J rock chips. ..................................................................................................... 380 Table G 3. U-Pb-Th isotopic data for sample B22K-3000 dated by ID-TIMS ........................... 381 Table I 1. Sample list for first (left) and second batch (right). Samples with a “X” suffixe are duplicates. MBX-1, P1 and WP-1 are MDRU standards. .......................................... 393 Table I 2. Analytical results for major elements normalized to anhydrous compositions. Notes: (1) Mapping areas located on figures of Chapter 2; (2) Method 4A-4B ...... 394 Table I 3. Analytical results for trace elements and REE. Notes: (1) Mapping areas located on figures of Chapter 2; (2) Methods: Trace elements (4A-4B); Rare earth and refractory elements (ICP-MS); CO2 and S (total) determined by Leco instrumentation; 1DX: aqua regia digestion, ICP-MS; (3) Units: ppm = part per million. ............................................................................................................................... 396 Table I 4. Compiled analytical results for MDRU standards MBX-1 (a), P-1 (b) and WP-1 (c) for selected elements (SiO2, K2O, Nb). ......................................................................... 401 Table I 5. Analytical results for randomly selected samples and their duplicates for selected elements (SiO2, K2O, Nb). .............................................................................................. 406 Table K 1. Compilation table of the significant Jurassic to Eocene epigenetic mineral deposits in the Nechako region and their characteristics (BC Geological Survey 2013). Deposits are sorted from the youngest to the oldest. ................................................. 420   xvi  List of Figures Figure 1.1. Pacific and North American plate geometries, major plate boundaries and approximate plate motion directions during the Late Cretaceous and Eocene. a) Rifting of the Pacific and Farallon plates in the Late Cretaceous formed the Kula plate. The subduction angle is estimated at 5° (Thorkelson and Taylor 1989); b) At 56 Ma, oblique subduction of the Kula and Resurrection plates was directed to the north-northeast. Subduction of the Farallon plate was relatively orthogonal with respect to the North American margin. The subduction angle is estimated to have been steeper than 5° (Thorkelson and Taylor 1989), and plate velocities are estimated between 100-116 mm/yr (Haeussler et al. 2003); c) The Kula Plate became fused to the Pacific Plate at 40 Ma (Madsen et al. 2006), and the Pacific plate began to move in an approximately transform motion with respect to the subduction trench (Madsen et al. 2006). ............................................................................. 4 Figure 1.2. Compilation map showing the distribution of Eocene volcanic and plutonic rocks and main structural elements in the North American Cordillera from Alaska to the northwestern USA. Eocene rocks are in places merged with Late Cretaceous or Paleogene rocks depending on the amount of detail available in compiled digital geological datasets. Metamorphic core complexes and volcanic fields referred to in text are indicated. Regional faults are D: Denali Fault; F: Fraser fault; I: Iditarod-Nixon Fork fault; K: Ketchikan fault; QC: Queen Charlotte fault; RM: Rocky Mountain Trench; T: Tintina fault; Y: Yalakom fault. Geology and structures are compiled from Huntting et al. (1961), Souther (1970), Bond et al. (1978), Beikman (1980), Walker and MacLeod (1991), Struik (1993), Green et al. (1994), Raines and Johnston (1995), Gordey and Makepeace (1999), Haeussler et al. (2003), and Massey et al. (2005). Accreted terrane assemblages are from Nelson and Colpron (2007). Age and origin of the Ruby Range Batholith and associated Paleocene volcanic complexes in southwestern Yukon after Israel et al. (2011)....................................................................................................................................... 8 xvii  Figure 1.3. Eocene volcanic groups and age correlations in central and southern British Columbia. Letters indicate references: a) Diakow and Koyanagi 1988; b) Rouse and Mathews 1988; c) Dostal et al. 2001, 2005; d) Grainger et al. 2001; e) Grainger et al. 2001; Diakow et al. 1997; f) Grainger and Anderson 1999; g) This study; h) Metcalfe et al. 1997; i) Dostal et al. 2003; Church 1973, 1985; j) Ewing 1981b; Read 2000. ............................................................................................................................ 10 Figure 2.1. Extent of OLG volcanic rocks in the Nechako and Chilcotin plateaus of central BC. Extents are mostly inferred from field mapping because of poor exposure, and possibly include the surface area covered by Miocene Chilcotin Group basalt. Geological boundaries, terrane outlines and regional structures adapted from Massey et al. (2005). Eocene areas referred to in the text are indicated by numbers: 1) Mount Greer/Kenney Dam; 2) Fawnie and Nechako ranges; 3) Tibbles; 4) Nazko; 5) Baezaeko; 6) Clisbako; 7) B-22-K exploration well; 8) Chezacut; 9) southern Chilcotin Plateau; 10) Blackdome Mine. Eocene epithermal prospects and deposits are EQ: Equity Silver; AL: Allin; UL: Uduk Lake; LN: Loon; TR: Trout; WO: Wolf; CL: Clisbako; BA: Baez; BD: Blackdome. Inset map: general extent of Eocene volcanic rocks in BC and distribution relative to accreted terranes (CPC: Coast Plutonic Complex; ST: Stikine terrane; CC: Cache Creek terrane; QN: Quesnel terrane; NP: Nechako Plateau; CP: Chilcotin Plateau; SBC: southern BC). ...................................................... 26 Figure 2.2. Regional stratigraphy of the Nechako and Chilcotin plateaus. Compiled after Riddell and Ferri (2008), Riddell (2011), Gabrielse and Yorath (1991). ...................... 28 Figure 2.3. Synthetic geological map for part of the Chilcotin Plateau highlighting the spatial distribution of OLG map units based on field mapping and observations. Mapping areas referred to in text are indicated by numbers: 3) Tibbles; 4) Nazko; 5) Baezaeko; 6) Clisbako; 7) B-22-K exploration well; and 8) Chezacut. Extent of undivided OLG is after Massey et al. (2005). Unit boundaries are inferred using total magnetic field data (Geological Survey of Canada 1994) when field relationships are insufficient. Eocene unit boundaries in the Chezacut area are after Mihalynuk et al. (2008b), some boundaries in the Clisbako area are after xviii  Metcalfe et al. (1997). Geological boundaries other than Eocene are from various sources: CGB after Dohanney et al. (2010a); Mesozoic after Massey et al. (2005) and Mihalynuk et al. (2008b) for the Chezacut area. Geochronological sample locations and ages are extracted from Chapter 3. Structural measurements are not reported, but major magnetic lineaments interpreted from aeromagnetic data (Geological Survey of Canada 1994; Collin Barrett, pers. comm. 2012) are traced. . 31 Figure 2.4. Whole rock major element classification diagrams for OLG volcanic rocks from the Chilcotin Plateau. Individual samples are classified by lithology, but broad map unit fields are indicated. Regional trends for Eocene volcanic rocks in central and southern BC (compiled after Ewing 1981a, Drobe 1991, Hickson 1992, Metcalfe et al. 1997, Anderson et al. 1998b, Dostal et al. 2001, Mihalynuk et al. 2009b), CGB (Bevier 1983; Anderson et al. 2001) and Chesltatta Lake suite basalt (Anderson et al. 2001) are reported for reference. a) Total Alkali Silica diagram, after Le Maitre et al. (1989); b) AFM diagram, after Irvine and Baragar (1971)..................................................................................................................................... 34 Figure 2.5. Field and microscopic photographs of Eocene OLG volcanic rocks of the Chilcotin and Nechako plateaus. (a) Exposure of biotite-quartz-feldspar-phyric rhyolite of the Tibbles rhyolite unit, Tibbles area; (b) Microscopic view of the biotite-quartz-feldspar-phyric rhyolite of the Tibbles rhyolite unit, Tibbles area, field of view 4.2 mm; (c) Brecciated, silicified, hematite stained rhyolite with vuggy quartz-veins at the Baez epithermal prospect, Clisbako rhyolite unit; (d) White, crumbly, clay weathered monomictic block and ash flow deposit with blocks of biotite-quartz-phyric rhyolite, Clisbako rhyolite unit, Clisbako area; (e) Microscopic view of the block and ash-tuff of the Clisbako rhyolite unit, displaying feldspar, quartz, biotite crystals and vesiculated glassy pyroclasts, field of view 4 mm; (f) East-dipping beds of volcanic conglomerate and tuffaceous sandstone of the Clisbako rhyolite unit, containing large angular blocks of brown-purple glassy plagioclase-phyric lava and biotite-quartz-phyric rhyolite; (g) Finely platy-jointed acicular hornblende dacite, Chezacut dacite unit; (h) Dacite ash tuff part of the Chezacut dacite unit,  B-22-K area. Crystals include plagioclase (3-4%), biotite (1%) and smoky quartz (1%); (i) Aphanitic dacite part xix  of the Baezaeko dacite unit; (j) Outcrop and microscropic view (field of view 10 mm) of flow-associated monomictic quench or hydrothermal breccia with a red-pink finely grained matrix and angular blocks of aphanitic dacite, Baezaeko dacite unit, Baezaeko area; (k) Steeply-dipping, columnar-jointed exposure of vitreous black dacite part of the Mount Sheringham dacite, Baezaeko area; (l) Perlitic fractures in vitreous black dacite, Mount Sheringham dacite, field of view is 4.2 mm; (m) White weathered brecciated facies of the Mount Sheringham dacite in the Baezaeko area; (n) Accretionnary lapilli-tuff part of the Mount Sheringham dacite, Baezaeko area; (o) Spectacular cliff of flow-banded andesite part of the Nechako andesite unit at Kenney Dam, and macroscopic view of a plagioclase-phyric andesite at Mount Greer; (p) Variegated monomictic pyroclastic breccia, Nechako andesite unit, Tibbles area. ................................................................................ 39 Figure 2.6. Composite stratigraphic column for the Tibbles area (3; Figure 2.3). Includes legend for Figures 2.7 to 2.11. Ages extracted from this study (see Chapter 3). ....... 40 Figure 2.7. Composite stratigraphic column for the Clisbako-B-22-K areas (6-7; Fig 3). Ages from Metcalfe et al. (1997), and this study. ...................................................................... 42 Figure 2.8. Composite stratigraphic column for the Chezacut area (8; Figure 2.3). Geology modified after Mihalynuk et al. (2008c, 2009b), ages extracted from this study......... 45 Figure 2.9. Composite stratigraphic column for the Nazko-Baezaeko areas (4-5; Figure 2.3). Ages from Metcalfe et al. (1997), and this study. ............................................................ 48 Figure 2.10. Composite stratigraphic column for the Kenney Dam/Mount Greer areas in the Nechako Plateau (1; Figure 2.1).  Compiled after Wetherup (1997), Haskin et al. (1998), Grainger et al. (2001). K/Ar and playnology ages from Rouse and Mathews (1988, 1989), other ages from Grainger et al. (2001). .................................... 52 Figure 2.11. Stratigraphic correlations between Eocene sections (numbered according to Figure 2.1) across the Nechako and Chilcotin plateaus of central BC, highlighting the lateral distribution of Eocene map units and lithologies. Vertical scale is approximate, horizontal scale is estimated at 350 km but distance between xx  individual sections is not representative. Individual synthetic logs are constructed from field observations and literature. Thickness constraints are from references indicated in the text, and include oil and gas wells or mineral exploration drilling constraints where available. All contacts are inferred due to poor exposure, either within the Eocene section, or between the Eocene sequence and Mesozoic rocks. Dashed lines indicate the temporal boundaries of the Eocene volcanic event, as defined in this study (see Chapter 3). Geochronological samples are compiled from this study and Breitsprecher and Mortensen (2004). ............................................ 57 Figure 3.1. Eocene volcanic and plutonic rocks in British Columbia (inset map) and distribution of the Ootsa Lake, Endako and other Eocene groups in central and southern BC (after Massey et al. 2005). Eocene rocks are coextensive with the Intermontane Belt (after Massey et al. 2005), and bounded by regional faults (Struik et al. 1993) and metamorphic core complexes (geological boundaries after Massey et al. 2005). The Miocene Chilcotin and Cheslatta groups basalts potentially overlie additional Eocene volcanic rocks. In legend, CPC = Coast Plutonic Complex. ............................................................................................................... 65 Figure 3.2. Distribution of 80-30 Ma isotopic cooling and crystallization ages determinations along the Canadian Cordillera of British Columbia (49-60⁰N) and from west to east across the Coast, Intermontane and Omineca Belts (compiled from Breitsprecher and Mortensen 2004, and data from this study). White symbols correspond to K/Ar or whole rock methods. ................................................................ 66 Figure 3.3. Temporal and spatial stratigraphic relationships in central BC on an isochronological framework. Mesozoic and Neogene stratigraphy is compiled from multiple sources indicated in text. Eocene stratigraphy is based on field mapping and regional correlations. Isotopic ages are selected and compiled from BC CordAge database (Breitsprecher and Mortensen 2004). ....................................... 69 Figure 3.4. Distribution of U-Pb and 40Ar/39Ar ages within the study area and their associated lithology. a) Simplified geological map with location of geochronological samples. Symbol colour represents assigned lithology. Existing 40Ar/39Ar and K-Ar xxi  isotopic Eocene ages are indicated by a simple black dot (after Breitsprecher and Mortensen 2004). b) Summary of U-Pb and 40Ar/39Ar dates from this study plotted against latitude for the six mapping areas. ......................................................... 78 Figure 3.5. SEM-CL imagery for representative Eocene zircons and examples of Jurassic xenocrysts dated by LA-ICP-MS. The red line indicates the laser track. .................... 79 Figure 3.6. Preferred ages for Eocene volcanic rock samples and lithological groups mapped in the Chilanko Forks, Chezacut and B-22-K areas, including 40Ar/39Ar step gas release plots, inverse isochron plots, and U-Pb concordia diagrams. Plateau steps are grey, rejected steps are white. Box heights are 2σ. Data-point error ellipses on the isochron plots are 2 σ. Data-point error ellipses on concordia diagrams are 2σ. .......................................................................................................................................... 89 Figure 3.7. Preferred ages for Eocene volcanic rock samples and lithological groups of the Clisbako, Baezaeko and Nazko areas, including U-Pb concordia diagrams and probability density plot for detrital zircons from sample BAdz-27. Data-point error ellipses on concordia diagrams are 2σ. Samples dated using the TIMS method are indicated. .......................................................................................................... 91 Figure 3.8. Preferred ages for Eocene volcanic rock samples and lithological groups of the Tibbles area, including 40Ar/39Ar step gas release plots, inverse isochron plots, and U-Pb concordia diagrams. Plateau steps are grey, rejected steps are white. Box heights are 2σ. Data-point error ellipses on the isochron plots are 2 σ. Data-point error ellipses on concordia diagrams are 2σ. ......................................................... 92 Figure 4.1. Distribution and average composition of whole rock geochemistry samples in the Chilcotin Plateau. Numbers correspond to mapping areas: 3 = Tibbles; 4 = Nazko; 5 = Baezaeko; 6 = Clisbako; 7 = B-22-K; 8 = Chezacut. Sample number prefixes (EB10- or EB11-) have been removed for clarity. Abbreviations: Bt=biotite; Hb=hornblende; Plag=plagioclase; Px=pyroxene; Qz=quartz. ........... 102 Figure 4.2. Bivariate diagram showing the concentration of Na2O with respect to LOI %. Samples with the highest LOI are also strongly depleted in Na2O, which is xxii  interpreted as alteration corresponding to the introduction of K-feldspar, and subsequent depletion in Na. ........................................................................................... 103 Figure 4.3. Whole rock composition of Eocene volcanic lithologies and map units. Compositional field for Eocene volcanic rocks in central and southern BC compiled after references indicated in Figure 4.15. Compositional fields for CGB and Cheslatta Lake basalt from Bevier (1983) and Anderson et al. (2001). a) Total Alkali Silica (TAS) classification diagram (Le Maitre et al. 1989), with alkaline and subalkaline field subdivisions (Irvine and Baragar 1971); b) K2O vs SiO2 diagram (alkaline series; Peccerillo and Taylor 1976); c) AFM diagram (Irvine and Baragar 1971). .................................................................................................................................. 105 Figure 4.4. Geochemical Harker diagrams for Eocene volcanic lithologies. Major elements (%) relative to SiO2 content: a) Al2O3; b) TiO2; c) Fe2O3; d) MgO; e) CaO; f) Na2O. .................................................................................................................................. 106 Figure 4.5. Diagram of MgO wt % versus Fe2O3 wt % for rocks of the Eocene OLG. CGB samples are plotted for reference. Kamloops Group field after Ewing (1981a); Chon Aike Silicic Large Igneous Province of Patagonia after Pankhusrt et al. (1998); typical trends for primitime intraplate arc (Cascades: Jakes and Gill 1970), tholetiitic arcs (Iceland: Wood 1978; Japan: Jakes and Gill 1970) and calc-alkaline arc (Santorini: Nicholls 1971) are shown. ..................................................................... 107 Figure 4.6. Minor element abundances in Eocene volcanic rocks. a) Primitive mantle-normalized incompatible trace elements abundances, plotted from left to right in order of increasing compatibility (after McDonough et al. 1992); b) Chondrite-normalized rare-earth-element abundance (after McDonough and Sun 1995); c) Comparison of incompatible trace element patterns of primitive arc basalts, typical arc andesite, and bulk continental crust in a primitive mantle norm (Davidson et al. 2005, and references therein). Abbreviations: LILE: Large Ion Lithophile Elements; HFSE: High Field Strength Elements; LREE: Light REE; HREE: Heavy REE. ........................................................................................................ 109 xxiii  Figure 4.7. Variation of Sr vs Ba for OLG rocks, and comparison with southern BC rocks of the White Lake Basin (WLB; Park Rill andesite, Maroon Fm.; Church 1973), Kamloops Group (Ewing 1981a), and Garibaldi Lake (Green 1977), and silicic rocks of the San Juan Mountains, Colorado (SJM; Summer Coon volcano; Zielinski and Lipman 1976). Sr and Ba are two incompatible, LIL elements. Sr is excluded from most common minerals except plagioclase. Ba similarly excluded except in alkali feldspar. ................................................................................................... 111 Figure 4.8. Variation of SiO2 content relative to Nb and Ti for samples of the Tibbles area versus all other areas of the Chilcotin Plateau. ............................................................ 111 Figure 4.9. Variation of Zr content relative to selected REE for samples of the Tibbles area versus all other areas of the Chilcotin Plateau. ............................................................ 112 Figure 4.10. Predicted eruption temperatures for OLG and CGB rocks from the study area as a function of MgO concentrations, calculated using the geothermometer of Sugawara (2000). ............................................................................................................... 115 Figure 4.11. Zircon saturation temperatures for a range of lithologies calculated after Watson and Harrison (1983). a) TZrM=1.3 versus SiO2 wt % shows the range of temperatures obtained for the different geochemical groups at M=1.3. Tzr are valid when they plot above average solidus for silicate melts; b) Solubility of zircon (ppm) as a function of temperature, TZrM=1.3, and melt composition M. M values between 1.3-1.9 are reported for reference (leucogranite to tonalite after Miller et al. 2003); c) Zr (ppm) versus calculated melt composition M. The shaded area represents the range of saturation solubility experiments of Watson and Harrison (1983). Altered samples (small dots) and mafic and intermediate samples (Nechako andesite, CGB) are clearly outside of this range, and their corresponding zircon saturation temperatures are likely geologically meaningless; d) TZrMcalc versus SiO2 wt % shows the range of temperatures obtained for the different geochemical groups using M calculated from whole rock compositions. The range of TZr is slightly wider than when calculated with a fixed M value; and e) Solubility of zircon (ppm) as a function of temperature, TZrMcalc, and calculated xxiv  melt composition M. Range of M values between 1.3-1.9 are reported for reference (typical leucogranite to tonalite; after Miller et al. 2003). The dispersion of M values from the reference value of 1.3 towards higher values shows the influence of variable cation contents on TZr. Andesitic and basaltic samples are clearly off the general trend. ........................................................................................... 120 Figure 4.12. Comparison between TZr (horizontal axis) and TMgO (vertical axis) for OLG rocks obtained from the MgO and Zr saturation geothermometers respectively. The grey shaded area represents the most significant temperature ranges obtained for average OLG silicic compositions. ................................................................................ 120 Figure 4.13. Viscosity (log η) variations as a function of H2O (wt %) and temperature (°C) for lithologies of the OLG, a) Dry lavas; b) Hydrated silicic and intermediate lavas (0.5 wt % H2O) and dry basalt. Viscosities are calculated as a function of estimated eruption temperatures (full lines) or extrapolated to the whole range of temperatures (dashed lines). For silicic lavas, temperature estimates between 750-850 ͦC are calculated from zircon saturation temperatures. Range of eruption temperatures are constrained at 950-1100 ͦC for andesite, and 1000-1100 ͦC for basalt (after Cas and Wright 1988). The range of viscosities as a function of temperature in some selected naturally occurring lavas are compiled from the literature (see also Chapter 2). ........................................................................................ 125 Figure 4.14. Eocene volcanic groups in central BC and areas assessed for geochemistry. Geological boundaries including Eocene groups are after Massey et al. (2005). Inset map: NP=Nechako Plateau; CP=Chilcotin Plateau; SBC= southern BC; CPC= Coast Plutonic Complex; ST=Stikine terrane; CC=Cache Creek terrane; QN=Quesnel terrane. ...................................................................................................... 126 Figure 4.15. TAS diagrams (after Le Maitre et al. 1989) displaying the trend of SiO2 versus alkalis for Eocene volcanic rocks in central BC (a, b) and southern BC (c). Compositional fields for Neogene basalt are indicated for reference. Geochemical data from this study are reported on all diagrams for comparison. Names of compositional field for TAS diagram appear on Figure 4.3. ................... 128 xxv  Figure 4.16. Variation of Nb vs Zr content supporting a signature of continental margin rhyolite for OLG rocks, as opposed to island arc or continental interior signatures. .......................................................................................................................... 139 Figure 4.17. Th/Yb vs Ta/Yb and Th/Ta vs Yb discrimination diagrams for arc-related and within-plate silicic to intermediate volcanic rocks (method of Gorton and Shandl 2000) ................................................................................................................................... 140 Figure 4.18. Tectonic discrimination diagrams for granitoid rocks after Pearce et al. (1984). OLG rocks of the Chilcotin Plateau lie in the same field than the majority of intermediate and silicic volcanic rocks of the Interior Plateau (Compiled in Metcalfe et al. 1997) in the volcanic arc or syn-collision granite fields. Field for Chon Aike SLIP of Patagonia after Pankhurst et al. (1998). ...................................... 141 Figure 5.1. Distribution of Eocene volcanic and intrusive rocks in the Chilcotin and Nechako plateaus of central British Columbia with respect to Cordilleran accreted terranes and major faults. The thickness model covers the Chilcotin and Nechako plateaus, as well as the Williams Lake area east of the Fraser fault. Geological boundaries, faults and terranes are after Massey et al. (2005). Map projection UTM Zone 10, NAD 83. ................................................................................................ 147 Figure 5.2. Datasets used to constrain the thickness model: a) simplified geology of the study area, showing Paleogene volcanic rocks superimposed on a digital elevation model (Centre for Topographic Information 1997); b) extent of the detailed interpretation. Eocene volcanic rock distribution after Massey et al. (2005), and updated from recent mapping by Dohaney et al. (2010a,b) for the CGB, Mihalynuk et al. (2008b, 2009a) and this study. Eocene mineral occurrences (compiled from BC Geological Survey 2013): AL, Allin; AP, April; BA, Baez; BB, Big Bar; BD, Blackdome; CL, Clisbako; ES, Equity Silver; FG, Free Gold; GA, Gaul; GR, Grubstake; LO, Loon; PM, Poison Mountain; SI, Silver Island; SQ, Silver Queen; TR, Trout; UL, Uduk Lake; WO, Wolf. Other: BW, Blackwater deposit. Map projection UTM Zone 10, NAD 83. ..................................................... 150 xxvi  Figure 5.3. Thickness model of the Eocene volcanic sequence for the Chilcotin and Nechako plateaus, and interpreted structural lineaments in a selected area of the Chilcotin Plateau (inset map; after Bordet et al. 2013a). Map projection UTM Zone 10, NAD 83. ............................................................................................................................ 152 Figure 5.4. Interpreted structural lineaments in the focused model area draped over a) Eocene thickness model, b) reduced to the pole (RTP) magnetic anomaly (Geological Survey of Canada 1994), and c) gravity anomaly (Ferri and Riddell 2006). Interpretations are developed from integration of magnetic and topographic lineaments (from the RTP and DEM), seismic reflectors (after Hayward and Calvert 2011; Appendix J) and tomographic models (Smithyman 2013; Smithyman et al. 2014). Two Eocene epithermal prospects are within the transfer zone located on Figure 5.3. Map projection UTM zone 10, NAD 83. .................... 157 Figure 6.1. Tomography (left column; Sigloch and Mihalynuk 2013) and schematic tectonic interpretation (right column) showing the geometrical evolution and mutual interactions between oceanic basins, arcs and Cordilleran terranes along the ancestral North America (ANA) margin from the Late Cretaceous to the Eocene, after the model of Sigloch and Mihalynuk 2013. Tectonic interpretations are schematic and approximate, and do not constrain absolutely plate convergence rates, distances and relative motions. The tomography imagery represents time-calibrated depth slices through a fixed mantle reference frame, and displays a three-dimensional high velocity isosurface interpreted as relicts of subducted oceanic lithosphere (color bands change every 200 km). Slab walls are sinking at an average rate of 60 mm·yr-1 near the trench, possibly decelerating to 10 mm·yr-1 by 440 km (Sigloch and Mihalynuk 2013). 1) at 80 Ma, ANA was moving westward relative to a fixed mantle reference frame, away from the opening Atlantic ocean. Oceanic crust at the leading edge of ANA had been consumed by westward subduction for >100 Myr, forming the Angayucham arc. Orientation of the subduction zone was oblique relative to the ANA margin, leading to diachronous closure from south to north. At the same time, northward motion of the Intermontane and Insular superterranes was occuring to the east of the east-dipping South Farallon subduction zone. First phases of collision between xxvii  the Angayucham arc and ANA circa 75 Ma triggered Carmacks Group high MgO volcanism (e.g, Johnston et al. 1996) between 72-69 Ma; 2) between 75-65 Ma, ANA moved towards the north-northwest, almost parallel with the Angayucham arc, then rapidly moved westward again. By 65 Ma, the consumption of the Angayucham oceanic crust at the leading edge of ANA was almost complete (it was by 60 Ma), and the Angayucham arc collided and was accreted along the margin; 3) at ~55 Ma, the Angayucham oceanic lithosphere was forced to rupture and was overridden by ANA and its hot, sublithospheric mantle. A flux of hot asthenosphere ingressed into a zone of relatively cool, metasomatized sublithospheric mantle through the slab break. In effect, melting of the metasomatized mantle was induced over a broad region corresponding to the extent of the slab gap. The CPC magmatic flare up between 55-50 Ma, and coeval initiation of OLG volcanism in central BC, were related to this event. In addition, evidence for coupling between ANA and the north-northeast subducting Farallon plate included the development of orogen-scale dextral strike-slip faults and extensional basins (ca. 57-34 Ma; Struik 1993; Umhoefer and Schiarizza 1996; Lowe et al. 2001), as well as the exhumation of metamorphic core complexes (ca. 63-42 Ma; e.g., Friedman and Armstrong 1988; Parrish et al. 1988; Vanderhaeghe et al. 2003); 4) Eocene volcanism across central BC continued until sublithospheric asthenosphere reached thermal equilibrium at ~45 Ma. .............................................................................................................................. 173 Figure 6.2. Schematic tectonic diagrams (left), thermal sections (center) and tomographic models (right) showing the possible interaction of a slab gap beneath central BC with westward moving Ancestral North America (ANA) during the Paleogene. Tomographic models (after Sigloch and Mihalynuk 2013) represent time slices at 55, 50 and 45 Ma, corresponding approximately to tectonic diagrams a, b and c. Position of cross-section A-B on tomographic models is approximate. Tectonic diagrams and thermal sections (after Duretz et al. 2011) represent the following evolution: a) immediately prior to cessation of subduction of the Angayucham slab at 55 Ma, a nearly vertical slab wall separated hot sublithospheric mantle coupled to the base of ANA from a metasomatized mantle wedge, which formed xxviii  as a result of subduction of the Angayucham slab, and was responsible in part for Coast Plutonic Complex (CPC) arc magmatism; b) immediately after cessation of subduction and the forced Angayucham slab break, ANA overrode the gap and hot asthenosphere ingressed through the gap, juxtaposing hot sublithospheric mantle to colder, metasomatized mantle wedge, and inducing CPC magmatic flare-up; and c) as ANA and accreted terranes progressed over the slab gap and the slab continued sinking, hot sublithospheric mantle was emplaced across the mantle wedge, and a voluminous volcanic province (OLG) was established across the extent of the gap. ANA = Ancestral North America; CPC = Coast Plutonic Complex; OLG = Ootsa Lake Group; INS = Insular superterrane; IMS = Intermontane superterrane. More details on lithospheric-scale processes are in Figure 6.3. .......................................................................................................................... 177 Figure 6.3. Schematic cross-section block diagram of the slab gap under BC at 50 Ma (after the model of Sigloch and Mihalynuk 2013). CPC = Coast Plutonic Complex; OLG = Ootsa Lake Group; INS = Insular superterrane; IMS = Intermontane superterrane. Red line on inset map shows the approximate location of the cross-section. More details on crustal-scale processes are in Figure 6.4. ............................ 178 Figure 6.4. Schematic representation of the magmatic evolution of OLG lavas (compiled from Annen et al. 2006; Calvert and Talinga 2014): 1) hot sublithospheric mantle ingress through a slab gap allowed melting of metasomatized mantle wedge, and the abrupt generation of a thermal anomaly at the base of the lithosphere; 2) mantle-derived mafic magmas were injected at the base of the crust; 3) in the lower crust, basaltic intrusions triggered partial melting, providing the source of silicic melts (the “hot zone” of Annen et al. 2006). Another alternative for production of silicic melts is the crystallization of basalt, with the first residual melt having the most silicic composition, which then evolves with time to dacite and andesite. However, huge magma volumes required by this process are problematic. Assimilation of crustal rocks imprinted distinctive trace element and isotopic signatures to the melts. In addition, the concentration of H2O in melts was limited by the content of hydrated minerals in the metasomatized asthenosphere and crustal sources, with implications for the physical properties xxix  of the generated melts; 4) magmas were segregated from the hot zone and ascended rapidly via dikes, then were stored in shallow crustal reservoirs; 5) ascent of evolved, silicic melts though upper crustal faults was facilitated by transcrustal trancurrent and extensional faulting. The products of OLG volcanism were preserved in extensional grabens. At this stage the crust was thick, hot, and dynamically unstable. Postorogenic gravitational collapse of this thick crust was accommodated by ductile spreading in the middle-lower crust, which played a major role in metamorphic core complex exhumation (Liu 2001). 183 Figure 6.5. Synthesis of tectonic events and timing relationships between magmatic activity, deformation and mineralization events in central British Columbia, from the Late Cretaceous to the Holocene. v = volcanic rocks; P = plutonic rocks. Compilation sources: (1) Evenchick et al. 2007; (2) Haeussler et al. 2003; (3) Madsen et al. 2006; (4) Thorkelson and Taylor 1989; (5) Souther 1986; (6) Bevier 1989; (7) Anderson et al. 2001; (8) Bevier 1983; (9) Mathews 1989; (10) Rouse and Mathews 1988; (11) This study; (12) Petö 1974; (13) Diakow 2006; (14) Leitch et al. 1992; (15) Dostal et al. 2005; (16) Coleman and Parrish 1991; (17) Struik 1993; (18) Umhoefer and Schiarizza 1996; (19) Parrish et al. 1988; (20) Spear 2004; (21) Vanderhaeghe et al. 2003; (22) Friedman and Armstrong 1988; (23) Friedman 1988; (24) Wetherup and Stuik 1996; (25) Wetherup 1997; (26) Anderson and Snyder 1998; (27) Andrew 1988; (28) Petersen 2013; (29) McClenaghan 2013; (30) Faulkner 1986; (31) Sigloch and Mihalynuk; (32) Ewing 1980; (33) Brown et al. (2012). ............................................................................................................................ 186 Figure B 1. Field photographs of the Tibbles Rhyolite unit ............................................................ 245 Figure B 2. Field photographs of the Clisbako Rhyolite unit ......................................................... 248 Figure B 3. Field photographs of the Chezacut Dacite unit ............................................................ 252 Figure B 4. Field photographs of the Baezaeko Dacite unit ........................................................... 256 Figure B 5. Field photographs of the Mount Sheringham Dacite unit .......................................... 260 Figure B 6. Field photographs of the Nechako Andesite unit ........................................................ 265 xxx  Figure B 7. Field photographs of the Cheslatta Falls basalt unit .................................................... 269 Figure B 8. Field photographs of Neogene and Quaternary volcanic units ................................. 271 Figure B 9. Simplified table to differentiate Mezosoic, Eocene, Neogene and Quaternary lithologies and textures in the Nechako and Chilcotin Plateaus of central British Columbia............................................................................................................................ 273 Figure C 1. Microphotographs of petrographic thin sections from the Tibbles rhyolite............ 275 Figure C 2. Microphotographs of petrographic thin sections from the Clisbako rhyolite ......... 276 Figure C 3. Microphotographs of petrographic thin sections from the Baezaeko dacite ........... 277 Figure C 4. Microphotographs of petrographic thin sections from the Mount Sheringham dacite .................................................................................................................................. 279 Figure C 5. Microphotographs of petrographic thin sections from the Nechako andesite ........ 280 Figure C 6. Microphotographs of petrographic thin sections from the Chilcotin Group basalt 281 Figure C 7. Microphotographs of petrographic thin sections from the Cheslatta Lake volcanic suite ..................................................................................................................................... 281 Figure D 1. General extent of the mapping area and location of detailed maps and stratigraphic columns presented in subsequent figures .............................................. 283 Figure D 2. Draft geological map and outcrops location east of the Clisbako prospect, along the 4200 forestry service road (FSR, red line) .............................................................. 284 Figure D 3. Lithological relationships in the Clisbako prospect area, along the 4200 FSR ........ 285 Figure D 4. Draft geological map and outcrops location in the Baezaeko area, along the Little Mountain FSR (red line) .................................................................................................. 286 Figure D 5. Stratigraphic relationships between lithologies along the Little Mountain FSR. Isotopic ages are indicated by a star. ............................................................................. 287 xxxi  Figure D 6. Draft geological map and outcrops location in the Baezaeko area, in the vicinity of the Michelle-Baezaeko roadside pit (Baezaeko FSR is the red line) .................... 288 Figure D 7. Stratigraphic relationships between lithologies in the vicinity of the Michelle-Baezaeko roadside pit. ..................................................................................................... 289 Figure D 8. Draft geological map and outcrops location in the Baezaeko area, in the vicinity of Crater Lake (Baezaeko FSR is the red line) ............................................................. 290 Figure D 9. Stratigraphic relationships between lithologies in the vicinity of Crater Lake. Isotopic ages are indicated by a star. ............................................................................. 291 Figure D 10. Draft geological map and outcrops location along the Nazko River valley .......... 292 Figure D 11. Stratigraphic relationships between lithologies along the Nazko River Valley. Isotopic ages are indicated by a star. ............................................................................. 293 Figure D 12. Draft geological map and outcrops location along the Nazko River valley, across from the Indian Head promontory .................................................................... 294 Figure D 13. Stratigraphic relationships between lithologies in the vicinity of the Indian Head Promontory, Nazko River Valley ................................................................................... 295 Figure D 14. Draft geological map and outcrops location along the Nazko River valley, along the Clisbako Mouth FSR ................................................................................................. 296 Figure D 15. Stratigraphic relationships between lithologies along the Clisbako Mouth FSR, Nazko River Valley. Isotopic ages indicated by a yellow star. ................................... 297 Figure D 16. Draft geological map and outcrops location in the northern part of the Tibbles area ...................................................................................................................................... 298 Figure D 17. Draft geological map and outcrops location in the southern part of the Tibbles area ...................................................................................................................................... 299 Figure D 18. Stratigraphic relationships between lithologies in the Tibbles area......................... 300 xxxii  Figure D 19. Simplified stratigraphic column for the Blackdome Mine area (after Faulkner 1986 and Rennie 2005) .................................................................................................... 301 Figure E 1. Cumulative frequency plots for density and magnetics susceptibility measurements ................................................................................................................... 305 Figure E 2. Synthetic percentile box-plots illustrating the physical property signature of Ootsa Lake Group lithologies in the Chilcotin and Nechako plateaus. a) Compiled measurements of average magnetic susceptibility values per outcrop; b) Compiled measurements of average density values per sample. Abbreviations are: Bt = biotite, Hb = hornblende, Plag = plagioclase, Px = pyroxene, Qz = quartz. ......... 306 Figure F 1. 40Ar/39Ar step gas release plots and inverse isochron plots for Eocene volcanic rock samples. Plateau steps are grey, rejected steps are white. Box heights are 2σ. Data-point error ellipses on the isochron plots are 2 σ. ............................................. 347 Figure F 2. U-Pb concordia diagrams and 2-points weighted mean diagram for Eocene volcanic rock samples dated by LA-ICP-MS, and U-Pb concordia diagram and probability density plot for detrital zircons from sample BAdz-27. Data-point error ellipses on concordia diagrams are 2σ. Black bars on weighted mean diagrams are selected fractions, white bars are not included in the calculation. conf., confidence; data-pt errs, data-point errors; rej., rejected; Wtd, weighted. .... 362 Figure F 3. U-Pb concordia diagrams and 2-points weighted mean diagram for Eocene volcanic rock samples dated by ID-TIMS. Data-point error ellipses on concordia diagrams are 2σ. ................................................................................................................ 366 Figure F 4. Concordia and weighted mean plots for LA-ICP-MS "KL" dates (represented with 2σ error) .................................................................................................................... 368 Figure G 1. Simplified well log for B-22-K. Adapted from original log and descriptions by Cosgrove (1982) and compiled descriptions from rock chips observations (this study). ................................................................................................................................. 372 xxxiii  Figure G 2. Simplified well log for B-16-J. Adapted from original log and descriptions by Cosgrove (1981) and compiled descriptions from rock chips observations (this study). Note different scale to Figure G 1. ................................................................... 373 Figure G 3. Comparative lithological well logs for B-22-K displaying palynology age data (Hunt 1992), isotopic ages (Riddell 2010), and intervals tentatively dated by LA-ICP-MS in this study. ....................................................................................................... 376 Figure G 4. Photographs of rock chip samples from selected intervals in wells B-22-K and B-16-J. a) Pale red-pink rhyolite lava and breccia with lesser fragments of dark grey aphanitic lava, green microcrystalline lava, and aphanitic black vitreous lava; b) Pale grey-beige-white volcanic tuff; c) Pink-pale grey very fine grained aphanitic lava or claystone; d) White to yellow to pink-altered rhyolite lava and breccia, lesser fragments of dark green lava; e) Rhyolitic tuff and lava, white-yellow to brown to pale-green clay altered biotite-plagioclase-phyric rhyolite. Few fragments of dark green-grey lava and volcanic sandstone; and f) Rhyolitic tuff and lava, white-yellow to brown to pale-green clay altered biotite-plagioclase-phyric rhyolite. .................................................................................................................. 378 Figure G 5. U-Pb Concordia diagrams sample B22K-3000 dated by ID-TIMS. Data-point error ellipses on Concordia diagrams are 2σ. ............................................................... 379 Figure H 1. SEM-CL imagery for zircons dated by the LA-ICP-MS method. The red line indicates the laser trace. Abbreviations in red represent ages of individual zircon grains: Eo  = Eocene; Jur = Jurassic; Kt = Cretaceous. ............................................ 391 Figure I 1. Graphs of analytical results for MDRU standard samples MBX-1 (a), P-1 (b) and WP-1 (c) for selected elements (SiO2, K2O, Nb). Compiled MDRU average and standard deviation obtained from analytical results obtained from various research projects at MDRU. ........................................................................................... 403 Figure I 2. Graphs of analyses of selected samples and their duplicate values, representing precision of the analytical method ................................................................................. 407 xxxiv  Figure J 1. Example of a structural interpretation workflow showing the parallel interpretation of different datasets, and their final integration to develop a comprehensive structural model. In this study, only the interpretation of first derivative (1VD) magnetic lineaments was not integrated to the final interpretation. .................................................................................................................... 409 Figure J 2. Structural interpretation along Geoscience BC vibroseis line 5, along the Nazko River Valley. ...................................................................................................................... 411 Figure J 3. Structural interpretation of a southwest-northeast seismic line from Canadian Hunter, intersecting GBC line 5 ..................................................................................... 412 Figure J 4. Structural interpretation of a northwest-trending seismic line from Canadian Hunter, in the vicinity of well B-22-K........................................................................... 413 Figure J 5. Structural interpretation of a series of southwest-northeast seismic lines from Canadian Hunter, in the vicinity of well B-16-J ........................................................... 414 Figure J 6. Structural “point map” showing symbols and vergence of structures interpreted along seismic lines, projected over the RTP aeromagnetic map and primary magnetic lineaments. Color coding of RTP background is detailed in Figure 5.4, and varies between 1725 nT (warm red colors) and -1250 nT (cold blue and purple). ............................................................................................................................... 415 Figure K 1. Significant mineral deposits in the Nechako region, central British Columbia, scale 1:1 500 000. Inset map: location of the Nechako region in British Columbia.419 Figure K 2. Mapped distribution of Cretaceous to Eocene volcanism in central British Columbia, and position relative to the accreted terrane and regional fault systems (adapted from Massey et al. 2005). Late Cretaceous and Eocene Au-deposits: B: Blackdome Mine; BD: Blackwater-Davidson; C: Capoose; Cl: Clisbako; N: Newton; W: Wolf. ............................................................................................................ 426 xxxv  Figure K 3. Distribution of Eocene magmatic rocks with respect to regional faults and metamorphic core complexes (after Massey et al. 2005; Struik et al. 1993). Eocene isotopic ages are selected from Breitsprecher and Mortensen (2004). ..................... 440 Figure K 4. Late Cretaceous to the Late Paleogene crystallization and cooling ages for the Insular-Coast, Intermontane and Omineca-Foreland belts (Compiled after Breitsprecher and Mortensen (1994), and recent data from this study) ................... 441    xxxvi  List of Abbreviations Chemical elements  Ag Silver Ar Argon Au Gold H2O Water Nd Neodymium Pb Lead Sr Strontium U Uranium Zr Zirconium ZrSiO4 Crystalline zircon HFSE High Field Strength Elements LILE Large Ion Lithophile Elements REE Rare earth elements  Geography  BC  British Columbia DEM Digital Elevation Model NAD North American Datum USA  United States of America xxxvii  UTM Universal Transverse Mercator WA Washington State  Geological groups  CGB Chilcotin Group Basalt CPC Coast Plutonic Complex OLG Ootsa Lake Group SMO Sierra Madre Occidental (Mexico)  Methods  ICP-ES Inductively coupled plasma - emission spectrometry ICP-MS Inductively coupled plasma - mass spectrometry ID-TIMS Isotope dilution – thermal ionization mass spectrometry LA-ICP-MS Laser ablation – inductively coupled plasma mass spectrometry MT  Magnetotelluric RTP  Reduced to the Pole (magnatic anomaly) SEM-CL Scanning Electron Microscopy – Cathodoluminescence  Minerals  Bt Biotite Cpx Clinopyroxene xxxviii  Hb Hornblende Kpar K-feldspar Ol Olivine Opx Orthopyroxene Plag Plagioclase Qz Quartz S glass shard Zr Zircon  Others  2-D two-dimensional 3-D three-dimensional FSR Forestry Service Road TAS Total Alkali versus Silica diagram   xxxix  Glossary Ignimbrite: represents a welded or unwelded, pumiceous, ash-rich deposit resulting from pyroclastic density current(s) and flows. The largest ignimbrites described are those associated with large calderas in the western USA or the Sierra Madre Occidental, Mexico, and are made for the most part of densely welded tuff. Rhyolite, dacite and andesite are the most common compositions (definition compiled from Cas and Wright 1988; Freundt et al. 2000). Intermediate volcanic rock: a rock with SiO2 content between 52 and 63 wt %, represented by andesitic compositions. Lava: a coherent volcanic rock formed as a result of an effusive outpouring of molten magma from a volcanic vent.  Lava dome: a proximal, relatively static feature formed following the emplacement of a mass of molten or partially solidified lava. Lava flow: a process associated with the passive flowing of a mass of molten or partially solidified lava, controlled by topography proximal to the vent. Lithospheric delamination: detachment of thickened lithospheric mantle from overlying crust during continental collision (Bird 1979), leading to lithosphere peeling off and sinking into the asthenosphere. Mafic volcanic rock: a rock with SiO2 content between 45 and 52 wt %, represented by basaltic compositions. Petrogenesis: the origin of rocks and processes leading to their formation. In this thesis, this term specifically relates to magma generation and evolution to explain the nature and composition of igneous rocks. Seismic tomography: a technique based on the use and interpretation of seismic velocities, to derive information about the internal structure of the Earth. Siliceous Large Igneous Province (SLIP): the silicic end-member of Large Igneous Provinces, which represent “magmatic provinces with areal extents >0.1 Mkm2, igneous xl  volumes >0.1 Mkm3, and maximum life spans of 50 m.y. that have intraplate tectonic settings and/or geochemical affinities and are characterized by igneous pulse(s) of short duration (1–5 m.y.), during which a large proportion (>75%) of the total igneous volume was emplaced” (Bryan and Ernst 2008). Silicic volcanic rock: a rock with SiO2 content above 63 wt %, represented by dacitic and rhyolitic compositions. Slab: refers to the lower plate portion of a subducting system, overridden by the upperplate. Therefore, when oceanic crust and lithosphere enter a subduction trench, the portion of this crust and lithosphere that penetrates the asthenosphere is referred to as a slab. Slab break-off: occurs when a slab becomes detached from the subducting plate and sinks into the asthenosphere. Slab window: slab-free region beneath the convergent margin of an overriding plate as a consequence of an oceanic spreading center interacting with a subduction zone (Haeussler et al. 2003). Slab gap: in this thesis, refers to a slab-free area in the mantle, defined by normal to anomalously slow seismic velocities. In contrast, the slab gap is bounded by anomalously fast seismic velocities (dVP/VP = +0.35 %) interpreted as relict subducted slabs (Sigloch 2011).  Tectonic setting: the environment defined by the geometry, velocity and relative motion between tectonic plates in a specific region, at a given time. Volcanic rock: a rock formed during a volcanic eruption (from a stratovolcano, vent, fissure, or caldera…), or as a result of the interaction of eruptive products with the surrounding environment. Volcanic rocks include primary coherent lavas and breccias formed during eruption, and for which all components are igneous. Volcaniclastic rocks comprise volcanic rock fragments originating from the syn- or post-eruption reworking of primary volcanic deposits. The formation of volcaniclstic rocks may be triggered by the eruption itself (e.g., lahars), or may form as a result of sedimentary processes (e.g., volcanic sandstones and conglomerates), as a result of interactions between the existing landscape and a dynamic volcanic environment.  xli  Acknowledgements I would like to thank Craig Hart, for providing me with the opportunity to study with the Mineral Deposit Research Unit, and for helping me design the project that would suit the best my research interests and skills. I am also thankful for the support and guidance of my supervisory committee, Mitch Mihalynuk, Jim Mortensen and Kelly Russell. I have learned a lot from all of you. This research would not have been possible without the support from Geoscience BC, who supplied the primary source of funding to the project, as well as two student grants. Equally important was the role of Golder Associates Ltd., particularly Kevin Palmer and Darren Kennard, for supporting me with a three-years Industrial Postgraduate Scholarship from the Natural Sciences and Engineering Research Council of Canada. Part of the analytical and field expenses was covered by two research grants from the Society of Economic Geologists Canada Foundation. I would like also to thank Endeavour Silver Corporation for granting me a graduate scholarship through the UBC Department of Earth, Ocean and Atmospheric Sciences. Many of my colleagues provided feedback for improving abstracts, presentations or technical reports, in particular Thomas Bissig and Diane Mitchinson. I am very grateful to Elizabeth Stock for helping me out with improving the final draft of this dissertation. Matias Sanchez contributions to the structural interpretations were extremely helpful, as well as follow up discussions with Andrew Calvert from Simon Fraser University, and co-authorship collaboration and discussions with Brendan Smithyman from UBC. Graham Andrews provided some useful advice on petrographic observations. Mati Raudsepp, Elisabetta Pani and Jenny Lai are thanked for their assistance for acquiring SEM-CL images of the zircons. Finally, I would like to thank Arne Toma and the rest of the MDRU staff for their support. Janet Riddell introduced me to the regional geology of the Chilcotin Plateau. Thank you for sharing your knowledge during my first field season, and for your time and help throughout this project. Access to several exploration projects and core was possible thanks to the collaboration of Mark Petersen and the Newgold geologists, Marc Rebagliati at HDI, and Wayne Murton and Eric Archie from Sona Resources. Thank you also to Robert Haley, from the Forestry Ministry, who supplied the key for this remote gate on the Redstone Chezacut FSR. I xlii  am extremely grateful for the help and company of Julia Smith, Lena Stelkia, and Erin Looby during fieldwork. Hard work during these four years was only bearable thanks to the laughts, mutual support, social and outdoor activities shared with office mates and graduate students from the EOAS Departement, MDRU, and the SEG Student Chapter. To this last one, I am particularly grateful for the unforgettable life and geological experiences during the Western Australia and Philippine international fieldtrips. And to my musician friends of the Family Band, thank you for the quality time playing fun tunes and drinking good wine. In addition, many parts of this dissertation were written under the inspiring melodies and beats of Eiffel, Bashung, Nirvana, Underworld, The Cat Empire… I would also like to thank my parents for their unconditional support throughout my life and studies, and for visiting me from so far away for the past ten years. Also, this PhD represents the outcome of 30 years of personal growth and education across continents. Martine Pernodet from Lycée Romain Rolland, Marc Fournier and the Lithosphere Research Laboratory from Université Pierre et Marie Curie, and several Appalachian geologists especially contributed to my motivation to lead a geologist carrier. Field stories from Pierre Bordet, my great uncle, were also a unique source of inspiration, and have helped me putting into perspective some days harder than others in the field. I am looking forward to going on his tracks in the Himalaya. Finally, Jaime, thank you for being here, and for your patience particularly during this past year. Many exciting times, travels around the world and discoveries are ahead of us.       xliii  Dedication   To all those, parents, brother, friends, professors,  who have contributed to who and where I am now.         1  Chapter 1   Introduction  In the North American Cordillera, spatially extensive, dominantly silicic to intermediate Paleogene-aged volcanic sequences are preserved in a 3000 km-long, 150-1000 km-wide belt that extends from eastern Alaska to Idaho, and even further south to include thick ignimbrite deposits of northern Mexico. Paleogene volcanism followed a complex Mesozoic history of terrane accretion along the North American margin, and was coeval with orogen-scale extensional deformation and dextral translation. This widespread volcanic event is the record of significant changes in relative plate tectonics motions and geometries.  Volcanic sequences of the Ootsa Lake Group (OLG) in south-central British Columbia (BC) are of Eocene age, and were investigated to: 1) provide geological constraints to the tectonic evolution of the Canadian Cordillera during the Paleogene; 2) emphasize the spatial and temporal relationship between volcanism and deformation; and 3) document the timing, duration, composition and style of volcanism to support geological evolution models of the Cordillera. This research also constitutes an improved foundation of knowledge for central BC which will contribute to the exploration of mineral and hydrocarbon resources, particularly Eocene epithermal Au-Ag deposits. First, a review of Paleogene (~66-23 Ma) tectonics and magmatism along and across the North American Cordillera outlines the broad tectonic and geological context for the specific study of Eocene (~56-34 Ma) volcanism in central BC.   1.1. Paleogene tectonics, magmatism and deformation in the North American Cordillera  1.1.1. Tectonic models The geology of the North American Cordillera reflects major changes of plate configuration and motion between the Late Cretaceous and Paleogene, but does not constrain absolutely the geometry, angles and rates of subduction. Plate tectonic models suggest different configurations of the Farallon-Kula (± Resurrection), and Pacific plates system with respect to 2  the North American margin (see references below). These models, and their particularly their influence on magmatism, are briefly summarized below. 1.1.1.a.  Eocene volcanic arc setting Traditional tectonic models suggest that from at least the Middle Jurassic to Paleogene times, the Cordilleran margin resembled the Andes (van der Heyden 1992), with nearly continuous orthogonal to oblique eastward subduction of the Farallon plate, and of the Kula plate from about 85 Ma, under the continental margin of ancestral North America (Figure 1.1a; Monger et al. 1972; Engebretson et al. 1985; Monger and Price 2002; Gabrielse and Yorath 1991). As a result, the rock record displays evidence of compression, terrane accretion, crust thickening and uplift, and subduction-related arc magmatism.  In contrast to the Mesozoic, the magmatic and structural records from the Late Cretaceous-Paleogene resulted from a significant change of relative plate motion and subduction geometry. The Kula plate began to move in a northerly direction relative to ancestral North America at about 59 Ma (Thorkelson and Taylor 1989). Oblique subduction of the Kula and Farallon plates under the North American margin resulted in a new tectonic regime characterized by crustal extension and dextral shearing (Figure 1.1b; Engebretson et al. 1985; Gabrielse and Yorath 1991). These changes in plate motion facilitated the subsequent capture of the Kula plate by the Pacific plate at 43 Ma (Engebretson et al. 1985) and demise of the Kula plate around 40 Ma (Figure 1.1c; Thorkelson and Taylor 1989; Madsen et al. 2006). An alternative model involves the amalgamation of the Pacific, Kula and northern Farallon plates around 56 Ma, resulting in cessation of the subduction and the development of a large transform Pacific-North America boundary (Ewing 1980). In this setting, widespread Eocene calc-alkaline volcanism in central and southern BC is inferred by many authors to have resulted from the subduction of the Kula and Farralon plates under the North American margin (e.g., Dickinson 1976; Ewing 1980; Armstrong and Ward 1993).  1.1.1.b.   “Slab window” inducing models The complex geometry and relative motions of plates subducting under the North American margin may induce the formation of slab windows, and locally influence volcanic arc 3  magmatism (Dickinson and Snyder 1979; Thorkelson and Taylor 1989; Breitsprecher et al. 2003; Madsen et al. 2006). A slab window “is a slab-free region beneath the convergent margin of an overriding plate and is a consequence of a spreading centre interacting with a subduction zone” (Haeussler et al. 2003).  The position and evolution of the Kula-Farallon slab window from Late Cretaceous to Middle Eocene time was defined using geochemical trace element contouring (Breitsprecher et al. 2003). In this study, regional geochemical variations along the Eocene volcanic arc are correlated with the position of the slab window at different times. At 50 Ma, arc volcanism originated above the subducting Kula plate, whereas volcanism located directly above the slab window had a highly alkaline affinity, possibly related to “asthenospheric upwelling, crustal extension, and partial fusion of Precambrian lithospheric mantle”. In contrast, the occurrence of adakites in southern BC is attributed to “partial melting of hot trailing edges of Kula plate”. Coeval near-trench magmatism in southern Alaska and the Cascadia margin around 56 Ma is attributed to the proximal interaction of the subducting ridges that separate the newly formed Resurrection plate from the Kula and Farallon plates (Figure 1.1b; Haeussler et al. 2003; Madsen et al. 2006), and provides evidence for the formation of two slab windows beneath the North American margin during the Eocene (Haeussler et al. 2003). Such slab windows geometries also account for the existence and discontinuous distribution of forearc magmatic sequences of Eocene age from Alaska to Oregon (Madsen 2004).  4   Figure 1.1. Pacific and North American plate geometries, major plate boundaries and approximate plate motion directions during the Late Cretaceous and Eocene. a) Rifting of the Pacific and Farallon plates in the Late Cretaceous formed the Kula plate. The subduction angle is estimated at 5° (Thorkelson and Taylor 1989); b) At 56 Ma, oblique subduction of the Kula and Resurrection plates was directed to the north-northeast. Subduction of the Farallon plate was relatively orthogonal with respect to the North American margin. The subduction angle is estimated to have been steeper than 5° (Thorkelson and Taylor 1989), and plate velocities are estimated between 100-116 mm/yr (Haeussler et al. 2003); c) The Kula Plate became fused to the Pacific Plate at 40 Ma (Madsen et al. 2006), and the Pacific plate began to move in an approximately transform motion with respect to the subduction trench (Madsen et al. 2006). 1.1.1.c.  Plate reconstruction constrained by seismic tomography imagery Alternate models suggest that oceanic crust at the leading (western) edge of the North American craton (Johnston 2001, 2008; Hildebrand 2009), or attached to an already loosely accreted Intermontane Belt (Sigloch and Mihalynuk 2013), subducted westward (rather than eastward) beneath the Jura-Cretaceous Angayucham intraoceanic arc (Sigloch and Mihalynuk 2013). These models are supported by data from the new, high-density USArray seismic network, which revealed high resolution images of the mantle structure beneath North America (Sigloch et al. 2008; Sigloch 2011). Mantle fast domains, interpreted as accumulation of fossil subducted slabs beneath North America, provide some constraints on subduction geometry 5  (Sigloch and Mihalynuk 2013). These slab relicts constitute massive physical discontinuities that isolate sublithospheric mantle domains from each other.  In this scenario, final collision of the Angayucham arc along an ancestral Alaska-BC margin between 59 and 55 Ma may have led to a slab break (Sigloch 2011; Sigloch and Mihalynuk 2013). Geometry and evolution patterns of this slab break differ from those described in the classic slab window models, but the effects, particularly on magmatism, may be similar. This model also supports weak coupling of the new Eocene margin with the Pacific plate in BC, or with subducting relicts of the Farallon plate in northwestern continental USA, resulting in crustal extension, dextral shear and Cascade arc magmatism (Engebretson et al. 1985; Gabrielse and Yorath 1991).  1.1.2. Magmatism along the North American active continental margin Fundamental variations in the style of magmatism existed along the North American Cordillera in the Paleogene. The following section is a review of selected areas from the Yukon to Mexico that best illustrate these differences, providing a broad context for the study of the Eocene OLG of central BC. Comprehensive background of the plutonic and volcanic evolution of the North American Cordillera is derived from Souther (1970), Petö (1974), Ewing (1980), Armstrong and Ward (1993), Gehrels et al. (2009), and Girardi et al. (2012).   1.1.2.a.  Northern BC and Yukon In northern BC and southwestern Yukon, most Eocene igneous rocks are intrusive, and located along the axial portions of the batholiths that form the Coast Plutonic Complex (CPC) (Figure 1.2; Woodworth et al. 1991; Gehrels et al. 2009). These magmatic suites, dominated by tonalite, granodiorite and granites (Resnick 2003), were emplaced from 100 to 50 Ma, after mid-Cretaceous juxtaposition of outboard and inboad terranes, and following the formation of a Late Jurassic-Early Cretaceous magmatic arc across the western and eastern portions of the CPC (Gehrels et al. 2009). High-flux Paleogene magmatism in the CPC resulted from rapid uplift of the axial portions of the batholiths, in an extensional setting (Gehrels et al. 2009); its cessation recorded the demise of plate convergence along this portion of the Cordilleran margin (Gehrels et al. 2009).  6  Eocene volcanic rocks northwestern BC and southwestern Yukon only form a narrow belt (70 km wide) along the Alaska-BC border, broadening toward the south (Ewing 1980). Part of this volcanic belt is composed of the Skukum Group of southwestern Yukon (Figure 1.2; Wheeler 1961; Morris and Creaser 2003), a volcanic and plutonic suite characterized by intermediate to silicic pyroclastic rocks and lavas deposited in a half-graben, and intruded by medium-grained granitic rocks (Mišković and Francis 2006). The Skukum Group comprises the Bennett Lake volcanic caldera complex (Lambert 1974), composed of a sequence of pyroclastic flows, andesite, dacite and rhyolitic lava flows, and boulder conglomerates. The Sloko Group of northwestern BC (Figure 1.2; Resnick 2003) consists of a 1500 m thick sequence of intermediate and lesser silicic volcanic and pyroclastic flows, also deposited in a half-graben. The volcanic sequence is intruded by dominantly hornblende and biotite bearing tonalite, granodiorite and quartz-monzonite (Resnick 2003). The Sloko Group is considered as a temporal and magmatic equivalent of the Skukum Group (Souther 1991).     7   Figure 1.2   8  Figure 1.2. Compilation map showing the distribution of Eocene volcanic and plutonic rocks and main structural elements in the North American Cordillera from Alaska to the northwestern USA. Eocene rocks are in places merged with Late Cretaceous or Paleogene rocks depending on the amount of detail available in compiled digital geological datasets. Metamorphic core complexes and volcanic fields referred to in text are indicated. Regional faults are D: Denali Fault; F: Fraser fault; I: Iditarod-Nixon Fork fault; K: Ketchikan fault; QC: Queen Charlotte fault; RM: Rocky Mountain Trench; T: Tintina fault; Y: Yalakom fault. Geology and structures are compiled from Huntting et al. (1961), Souther (1970), Bond et al. (1978), Beikman (1980), Walker and MacLeod (1991), Struik (1993), Green et al. (1994), Raines and Johnston (1995), Gordey and Makepeace (1999), Haeussler et al. (2003), and Massey et al. (2005). Accreted terrane assemblages are from Nelson and Colpron (2007). Age and origin of the Ruby Range Batholith and associated Paleocene volcanic complexes in southwestern Yukon after Israel et al. (2011). 1.1.2.b.  Central and southern BC Large Paleogene plutons in central and southern BC are restricted to the axial portion of the CPC, consistent with observations made in northern BC (Figure 1.2; Ewing 1980; Gehrels et al. 2009; Girardi et al. 2012). East of the CPC, widespread, Early and Middle Eocene, calc-alkaline and alkaline volcanism formed locally thick, discontinuous subaerial calc-alkaline volcanic sequences that unconformably overlie deformed Mesozoic rocks. Occurrences of Eocene volcanic rocks in central BC traditionally include silicic-intermediate successions of the OLG (Figure 1.2; Duffell 1959) and intermediate-mafic successions of the Endako Group (Armstrong 1949).  Detailed studies on volcanic rock successions correlated with the OLG (e.g., Drobe 1991; Metcalfe and Hickson 1995; Metcalfe et al. 1997; Diakow and Webster 1997; Grainger and Anderson 1999; Grainger et al. 2001; Dostal et al. 2001, 2005) emphasize the regional variability of this group, in terms of lithologies, stratigraphy, volcanic environment and ages (Figure 1.3). In particular, mafic-intermediate volcanic sequences of the Buck Creek basin (Figure 1.2; Dostal et al. 2001, 2005) are related to the development of an extensional pull-apart basin, and may represent a transition from an arc compressional to extensional intra-arc rifting environment (Dostal et al. 2005). In the Chilcotin Plateau, geochronology data coupled with seismic and gravity data interpretations (Riddell 2010; Hayward and Calvert 2011) suggest that great thicknesses of OLG rocks were deposited in northwest-trending pull-apart basins formed before and during the Middle to Late Eocene. Only a few occurrences of Eocene welded porphyrictic dacitic ash flows with ignimbritic textures were reported in the southern Chilcotin Plateau (Mihalynuk and Harker 2007; Mihalynuk et al. 2009a), and may indicate the presence of calderas. 9  Also, Au-Ag epithermal mineralization at the Wolf and Clisbako prospects is hosted in volcaniclastic, pyroclastic rocks and rhyolitic dome, lavas and breccias, and may be related to caldera collapse (Andrew 1988; Metcalfe et al. 1997). Volcanic rocks of Mount Clisbako (Metcalfe and Hickson 1995; Metcalfe et al. 1997) have been attributed also to a caldera environment. In southern BC, grabens and half-grabens developed between dextral northwest-trending transcurrent faults during northwest-directed extension (Souther 1970; Ewing 1980; Mathews 1991), leading to the deposition and preservation of the Kamloops (Figure 1.2; Ewing 1981b), Penticton (Church 1973, 1985) and Princeton (Thorkelson 1989; Souther 1991) groups (also see Chapter 2). The thickness of Eocene sedimentary and volcanic rocks ranges from 1.5 to 3 km in these grabens (Mathews 1991). Vertical displacements along graben-bounding faults are estimated up to 1800 m (Souther 1970). The Fig Lake Graben comprises a 3 km thick sequence of Eocene intermediate lavas and pyroclastic deposits, as well as polymictic conglomerate and sandstone in fault contact with Mesozoic rocks (Thorkelson 1989). Block rotation indicates syn-extensional deposition of the conglomeratic rocks, and at least 4.5 km of vertical displacement along normal faults (Thorkelson 1989). The Fig Lake Graben is inferred to have developed as a pull-apart basin in response to sinistral strike-slip displacement along a northeast-trending fault (Thorkelson 1989).  Several key studies focus on the petrogenesis of Eocene volcanic rocks in central and southern BC, including the OLG (Drobe 1991), the Penticton Group (Ickert et al. 2009), and a regional geochemical compilation study (Breitsprecher et al. 2003). These models are discussed in Chapter 4.   10   Figure 1.3. Eocene volcanic groups and age correlations in central and southern British Columbia. Letters indicate references: a) Diakow and Koyanagi 1988; b) Rouse and Mathews 1988; c) Dostal et al. 2001, 2005; d) Grainger et al. 2001; e) Grainger et al. 2001; Diakow et al. 1997; f) Grainger and Anderson 1999; g) This study; h) Metcalfe et al. 1997; i) Dostal et al. 2003; Church 1973, 1985; j) Ewing 1981b; Read 2000. 1.1.2.c.  Igneous complexes and volcanic fields of the northwestern USA The dominant type of magmatic activity in the northwestern USA was related to the formation of intrusive systems and wide calderas formed during explosive volcanic events in the Eocene (Lipman et al. 1972; Snyder et al. 1976; Ewing 1980). These large caldera systems contrast with the geological record of southern and central BC, where field evidence to formerly identify caldera structures is lacking.  The Colville Igneous Complex, WA (Figure 1.2), consists of Eocene intrusive and extrusive rocks formed and preserved in north-south trending grabens (Morris et al. 2000). A sequence of andesite and dacite flow units up to 1500 m thick is characterized by pyroclastic textures in the earlier members (Morris et al. 2000). A younger sequence made of basalt to 11  rhyolite flow and lava dome interbedded with sedimentary units is interpreted to have formed from graben-parallel fissures, and exceeds 2500 m in thickness (Morris et al. 2000). Calc-alkaline, subaerial andesitic lavas and volcaniclastic rocks of the Eocene Clarno Formation, Oregon, form a sequence up to 2 km thick (Rogers and Novitsky-Evans 1977; White and Robinson 1992). The rocks erupted from a number of small volcanic centers, and were deposited and preserved in sedimentary basins along a continental margin (White and Robinson 1992).  The volcanic setting is interpreted as diffuse volcanic arc undergoing extension (White and Robinson 1992). Volcanism is related to oblique subduction during the Eocene and eruption through a thin (20-30 km) crust (Rogers and Novitsky-Evans 1977). The Challis volcanic field of central Idaho (Figure 1.2) resulted from Eocene (51-40 Ma) effusive mafic to intermediate volcanism, followed by voluminous intermediate to silicic explosive volcanism associated with subsidence of large cauldron-complexes, calderas and northeast-trending volcano-tectonic depressions (Moye et al. 1988). The caldera structures are bounded by curvilinear faults and filled with ignimbrite units that record the emplacement of thick intracaldera pyroclastic flows and thin outflow units (Moye et al. 1988).  The Absaroka-Gallatin volcanic province of Wyoming and Montana (Figure 1.2) consists principally of Eocene to early Oligocene andesitic flows and breccias up to 1800 m thick over about 1300 km2 (Chadwick 1970). Igneous activity was dominantly associated with fissure eruption along pre-existing crustal weaknesses of Laramide (80-55 Ma; English and Johnston 2004) or older age, but also took place along a chain of volcanic vents (Chadwick 1970).   1.1.2.d.  Voluminous ignimbritic volcanism in the Sierra Madre Occidental, Mexico Silicic volcanism in the Sierra Madre Occidental (SMO) of northern Mexico took place during the Oligocene-Miocene (e.g., Ferrari et al. 1999; Ferrari et al. 2002; Aguirre-Diaz and Labarthe-Hernandez 2003; Bryan and Ferrari 2013), and marked the beginning of extension following the Laramide orogeny (Ferrari et al. 2002). This volcanic event generated a spectacular volume of rhyolitic ignimbrites, with an aerial surface of 400,000 km2 and an average thickness of 1 km; as such, the SMO is referred to as a Silicic Large Igneous Province (e.g., Bryan 2007; 12  Sheth 2007; Bryan and Ernst 2008; Pankhurst et al. 2011). Physical and chemical characteristics of this voluminous ignimbrite flare-up were controlled by the tectonic and crustal setting beneath Mexico in the Oligocene-Miocene, in particular by lithospheric processes inducing melting of the continental crust (see Chapter 6). Voluminous ignimbritic volcanism of the SMO was preceded by widespread Eocene silicic effusive volcanism (Aguirre-Diaz and McDowell 1991). Eruption of voluminous and widespread silicic ash flow tuffs and intermediate lavas were directly related to active subduction along the western margin of Mexico (Aguirre-Diaz and McDowell 1991, and references therein). 1.1.3. Structural elements 1.1.3.a.  Faults The onset of transtensional tectonism and deformation in the Canadian Cordillera in the Eocene led to the development of major, orogen-scale dextral strike-slip fault system including the Rocky Mountain Trench, the Tintina, Pinchi, Fraser, Yalakom and Denali-Queen Charlotte fault systems (Figure 1.2; Ewing 1980; Souther 1970; Struik 1993). Several tens to hundreds of kilometers of displacement are inferred along some of these faults (Eisbacher 1977; Ewing 1980). These faults represent the dextral strike-slip structures that contributed to between 1000-2000 km of northward translation of the Insular superterrane from the Late Cretaceous to the Eocene (Umhoefer and Schiarizza 1996; Hollister and Andronicos 1997). However, there are discrepancies between displacement amounts from geological and paleomagnetic evidence (Vandall and Palmer 1989; Enkin et al. 2006), with implications for the timing of docking of the Insular superterrane to cratonic North America (van der Heyden 1992) and, more generally, for tectonic evolution models of the Canadian Cordillera (Johnston 2001, 2008).  The present study focuses on major and secondary normal and strike-slip structures formed throughout central and southern BC during extensional and dextral-translation from the Late Cretaceous to the Oligocene (Struik 1993).  There are three major fault systems represented in central and southern BC, and briefly described below:  1) Early Eocene northwest-trending faults (Struik 1993; Umhoefer and Schiarizza 1996), with dextral strike-slip displacement up to 115 km estimated along the Yalakom Fault 13  (Umhoefer and Schiarizza 1996). These faults are inferred to have accompanied the development of Paleogene pull-apart basins (Struik 1993);  2) Eocene northeast-trending faults showing evidence of dip-slip extensional motion (Struik 1993) and locally associated with the older northwest-trending faults. These northeast-trending faults show normal or strike-slip displacement and are commonly intruded by Eocene mafic dikes (Lowe et al. 2001); and  3) Late Eocene to Early Oligocene north-trending dextral faults developed in association with northwest-directed extensional deformation. They form an en échelon fault system and include the Pinchi and Fraser regional faults (Struik 1993). Their last movements were coeval with the northwest-directed extension that exposed the Vanderhoof Metamorphic Complex (Struik 1993; Wetherup and Stuik 1996; Wetherup 1997). 1.1.3.b.  Metamorphic core complexes Eocene metamorphic core complexes represent large areas of high-grade metamorphic rocks that form a discontinuous sinuous belt along the eastern part of the North American Cordillera from southern Canada to northwestern Mexico (Figure 1.2; Davis and Coney 1979; Ewing 1980; Armstrong 1982; Coney and Harms 1984). Core complexes caused rapid uplift of hot metamorphic rocks and associated intrusive bodies, in association with tectonic and erosional denudation of an overthickened crustal welt of 50 to 60 km thick in an extensional regime (Davis and Coney 1979; Ewing 1980; Coney and Harms 1984). As a result they display a superimposition of ductile and brittle extensional fabrics (Armstrong 1982). Core-complex extension was also associated with sharp pulses of magmatism during the Eocene in the northern Cordillera and Oligocene-Miocene time in the southern Cordillera (Davis and Coney 1979; Coney and Harms 1984). Metamorphic core complexes in BC (Figure 1.2; Ewing 1980) include the Skeena (48 Ma), Wolverine (45 Ma), Shuswap (49 Ma), Vanderhoof (55-45 Ma; Wetherup and Stuik 1996; Wetherup 1997; Anderson and Snyder 1998) and Tatla Lake complexes (44-47 Ma; Friedman and Armstrong 1988; Friedman 1988). Exhumation of the Vanderhoof and Tatla Lake complexes was spatially and temporally associated with movements along the Fraser and Yalakom faults (Figure 1.2). 14  1.2. Research questions Paleogene volcanism in the Canadian Cordillera, and specifically Eocene volcanism in British Columbia, was coeval and spatially continuous with magmatic activity that formed broad volcanic fields in the northwestern USA and Mexico. Overall, this voluminous magmatic event marks the onset of a new tectonic, magmatic, and structural regime that affected broad parts of the North American Cordillera, complicating the existing geology, modifying the landscape, and promoting mineralization events.  1.2.1. Eocene volcanic rocks as records of the Paleogene tectonics of western North America  Eocene volcanic rocks constitute key elements in the evolution of the North American Cordillera, because they are the youngest exposed lithologies in many regions providing the most recent geological record of deformation and associated volcanism, metamorphism and mineralization. These rocks arised from magmatism interpreted as related to drastic changes in the organization and relative motions of tectonic plates subducting beneath, or interacting with, the North American margin. However, there is no consensus about the setting of volcanism and how it developed with respect to Cordilleran tectonics, and which were the tectonic mechanisms inducing volcanism. Therefore, specific research questions include: • Clarification of the geometry and number of tectonic plates involved in the subduction and responsible for Eocene magmatic activity in the North American Cordillera; • Implications of various tectonic models, such as those inducing slab windows (Ewing 1980; Thorkelson and Taylor 1989; Breitsprecher et al. 2003; Haeussler et al. 2003; Madsen et al. 2006), or recent tomography-constrained model (Sigloch and Mihalynuk 2013), on the inherited lithospheric structure beneath BC and the effects on induced Eocene volcanism; • Geological constraints on the onset of, and interaction between, volcanism, plutonism, coeval extension and strike-slip faulting in Eocene time as consequences of the tectonic evolution (Souther 1970; Ewing 1980). 15  1.2.2. Ambiguous magmatic sources and evolution Eocene volcanic rocks in BC have dominantly silicic-intermediate composition, with a consistent calc-alkaline signature (e.g., Ewing 1981a; Thorkelson 1989; Metcalfe et al. 1997; Dostal et al. 2003; Mihalynuk et al. 2009b; Bordet et al. 2011b). In most cases, trace elements indicate a hydrated mantle source and support crustal contributions in a subduction-related volcanic arc setting (e.g., Dostal et al. 2001). However, some characteristics of Eocene volcanism present crucial differences with classical volcanic arc settings. For instance, calc-alkaline rocks in arcs usually form elongate belts, with systematic cross-arc variation of magma chemistry (Jakeš and White 1972; Miyashiro 1974) and age (Gill 1981). In contrast, Eocene volcanic rocks in central BC are characterized by a wide areal distribution significantly inboard from the active paleo-margin. Also, the development of extensional basins in central BC reflects a gradual transition from an arc-compressional environment to an extensional intra-arc rift setting (Thorkelson 1989; Dostal et al. 2005). In addition, anomalously high alkaline compositions (Dostal et al. 2003) or adakites occur in few localities of southern BC (Thorkelson and Breitsprecher 2005; Ickert et al. 2009). Finally, the timing and duration of the volcanic event appears equivalent across parts of the province. These elements suggest that processes other than simple subduction took place, such as unique tectonic and/or magmatic mechanisms. The mappable characteristics, compositions and geochemical signatures of Eocene volcanic rocks must reflect: • The magmatic source(s) (mantle or crust) and their relative importance; • Fractionation and differentiation patterns of Eocene magmas. 1.2.3. Controls on the style of extensive, silicic volcanism In south-central BC, Eocene volcanic rocks cover an area up to 60,000 km2 (33,400 km2 of volcanic rocks and 27,300 km2 of intrusive rocks; Massey et al. 2005) between 49° and 55° N, with thicknesses up to 4 km locally (Riddell 2010, 2011). Thick sequences of Eocene volcanic rocks in fault contact with older rocks and the presence of synextensional sedimentary deposits suggest that the spatial extent, emplacement and deposition of OLG rocks could be coeval and partly controlled by these active fault systems and structural depressions (Ewing 1980; Thorkelson 1989; Gabrielse and Yorath 1991; Struik 1993; Dostal et al. 2005; Hayward and Calvert 2011). Alternatively, these thick sequences of volcanic deposits could suggest 16  emplacement within calderas (Andrew 1988; Metcalfe et al. 1997), although ignimbrite deposits are not representative of the OLG. Correct interpretation of structural lineaments together with constraints on the thickness, and spatial and temporal extent of Eocene volcanic rocks, should permit more complete interpretation of:  • A volcanic setting that explains the compositions, textures and large areal extent of the OLG; • Possible effects of coeval extension on the volcanic style (i.e., formation of effusive lavas versus ignimbrites), or a genetic link between deformation and volcanism; • The spatial and temporal relationship between volcanism, strike-slip faulting and the exhumation of metamorphic core complexes, and structural controls on the deposition and preservation of Eocene volcanic rocks. 1.2.4. Eocene volcanic chronostratigraphy of central BC The timing, frequency and total duration of Eocene magmatic activity is unclear, but generally constrained between 53-47 Ma (Ewing 1980; Grainger et al. 2001). Eocene volcanic rocks in central BC are traditionally divided into the dominantly silicic OLG, and the younger, dominantly mafic Endako Group on the basis of lithology and geochemistry (Anderson and Snyder 1998; Haskin et al. 1998; Grainger and Anderson 1999; Barnes and Anderson 1999b). However, evidence to support these distinctions is weak and inconsistent. The contact between Ootsa Lake and overlying Endako groups is locally unconformable (Struik et al. 1997), conformable, unweathered and structurally concordant (Haskin et al. 1998). In addition, the two groups display similar isotopic age ranges (Figure 1.3): from 56 Ma to 47 for the OLG (Grainger and Anderson 1999; Grainger et al. 2001) and from 51 to 37 Ma for the Endako Group (Haskin et al. 1998; Grainger et al. 2001). Similar age ranges of volcanic groups that are coextensive to a same area may suggest that they were formed during a same event, and only stratigraphic evidence could support their segregation in separate groups. However, the currently established stratigraphy for central BC significantly lacks:  17  • The identification of reference volcano-stratigraphic horizons to aid regional mapping and the development of a geological evolution model for this part of the Cordillera;  • Integration of existing regional fieldwork observations to support distinctions between groups on the basis of map units, textures, compositions and their spatial and temporal organization; • A geochronological framework for all Eocene volcanism in BC constrained by highly precise U-Pb or 40Ar/39Ar isotopic dating.  1.3. Research objectives  The dissertation is organized so that the research objectives are addressed from the rocks and field observations to the regional scale models and their application to broad tectonic concepts.  1.3.1. Chronostratigraphy of the OLG Primary objectives of the research aimed at: 1) improving the volcanostratigraphic framework for Eocene OLG rocks in the Chilcotin Plateau and inter-regional correlations; and 2) constrain the onset, duration and termination of Eocene volcanism.  Map units of the OLG are defined in Chapter 2, using a combination of field observations and surface mapping (Appendices A, B, D), petrography (Appendix C), physical properties (Appendix E), geochronology (Appendices F, G, H) and geochemistry (Appendix I). Integration of these datasets also supported preliminary interpretation of the volcanic setting. Detailed U-Pb and 40Ar/39Ar geochronological data are presented in Chapter 3 and Appendix F to support the stratigraphic framework presented in Chapter 2. Because of an extensive Quaternary volcanic and glacial cover throughout central BC and the lack of obvious widespread marker units, geochronology provided the backbone for regional lithostratigraphic correlations. Therefore, U-Pb and 40Ar/39Ar data tightly constrained the timing of the deposition of the different units and supported investigations of the tectonic processes at the origin of Eocene magmatism.  Supplementary information and discussion related to the stratigraphy and 18  age of rocks intersected by hydrocarbon exploration wells is provided in Appendix G, to address contradictory interpretations of the geological record in these wells.   1.3.2. Geochemical and physical properties of Eocene OLG lavas Important research objectives included the elucidation of: 1) the tectonic setting and nature of the OLG magma source region; and 2) petrogenetic processes operating during magma storage, emplacement and cooling; and 3) control of magma compositions on the physical properties and mode of eruption of OLG lavas, that may explain their stratigraphy, textures and widespread extent.  Detailed geochemical analyses for a selected suite of Eocene volcanic rock samples are presented in Chapter 4 and Appendix I. Geochemical major element compositions were primarily used to characterize OLG map units described in Chapter 2. Major, trace and rare earth element data supported preliminary interpretations of the magmas sources and their evolution. In addition, compositional data were utilized to predict chemical and physical properties of OLG lavas, such as eruption temperatures and viscosities, and to provide insights into the eruption style and volcanic setting.  1.3.3. Thickness and regional structural framework of the Eocene volcanic sequence A three-dimensional (3-D) thickness model for the OLG in the Chilcotin and Nechako plateaus was built to clarify structural controls on the distribution, extent and thickness of Eocene volcanic rocks in central BC, and is presented in Chapter 5. This model was generated from geological constraints, including field maps (Appendix D), cross-sections and stratigraphic columns (Chapter 2), and interpreted geophysical-survey data. The thickness model was improved by structural interpretations that were based on processed aeromagnetic and gravity anomaly maps, and enhanced by integrating structural constraints from selected seismic reflection and tomography surveys (Appendix J).  1.3.4. Paleogene tectonic and magmatic evolution of central BC The ultimate objective of this research was to identify a tectonic trigger that would explain Eocene magmatism in central BC, but also support extensive correlative volcanism from southwestern Yukon to southern Idaho. Such a tectonic trigger should explain characteristics 19  such as rapid melt generation, short temporal duration (cessation of magmatism within ~10 Mys), and OLG compositions and volcanic stratigraphy. This argumentation is developed in Chapter 6. Existing and new tectonic models were reviewed, especially those involving the paleogeography and kinematics of the Pacific, Kula and Farallon plates and their relationship with magmatism. In addition, interpretations of magmatic sources and the role played by crustal assimilation were discussed based on compositions and geochemical signatures of OLG rocks. Data from analogous volcanic environments, particularly Silicic Large Igneous Provinces (SLIPs) were synthetized to support the argumentation. Comparisons with areas undergoing coeval volcanism and extensional deformation aided the association between physical characteristics of the OLG and the eruptive style.  1.4. Scientific and economic implications This research builds upon work conducted by many authors on Paleogene volcanic rocks in the North American Cordillera, and particularly in the Canadian Cordillera. Original studies by Armstrong (1949), Duffel (1959), Tipper (1959, 1963, 1969) established a regional framework for Eocene volcanic rocks in central BC. Together with more recent literature (references in section 1.1.2.b.), these studies comprised the backbone of the present research, upon which new stratigraphic (Chapter 2), geochronological (Chapter 3), geochemical (Chapter 4) and structural (Chapter 5) data and interpretations were developed. New geological datasets include a new outcrop database (Appendix A), field and microscopic photographs (Appendices B, C), physical property data (density and magnetic susceptilility; Appendix E), geochronology data (Appendix F, G, H) and geochemical data (Appendix I). New geological maps of selected well-exposed areas (Appendix D) illustrate the lateral variations of lithologies and textures and significantly increase the detail of mapping and geological knowledge in parts of the Chilcotin Plateau. A new physical property dataset will help constraining the interpretations of seismic profiles, magnetotelluric (MT) profiles, aeromagnetic and gravity surveys, as it provides a direct link between the geology and geophysical models.  The “slab gap” model presented in Chapter 6 provides a tectonic explanation to the onset of OLG volcanism in central BC, and correlative volcanism in southern BC, southwestern Yukon and the northwestern USA. Seismic tomographic images of internal mantle structures 20  and their advanced tectonic interpretations, as proposed by Sigloch and Mihalynuk (2013), are utilized. The model of Sigloch and Mihalynuk (2013) is controversial because it questions several currently accepted tectonic models for the building and evolution of the Cordilleran orogen. However, the objective of this discussion is to investigate slab geometries that are different from those previously proposed (e.g., Ewing 1980; Thorkelson and Taylor 1989; Breitsprecher et al. 2003; Haeussler et al. 2003; Madsen et al. 2006), in an effort to minimize uncertainties related to the process of subduction and consumption of oceanic lithosphere, leaving no record of past tectonic plates geometry. The geometry of the slab gap and resulting magmatic processes differ from traditional models of Eocene arc magmatism (e.g., Dickinson 1976; Ewing 1980; Armstrong and Ward 1993), or slab windows (Breitsprecher et al. 2003; Haeussler et al. 2003; Madsen 2004; Madsen et al. 2006). The interpretation of OLG volcanism with this innovative tectonic model provides a different perspective into the understanding of an important epoch of Cordilleran geology, and could be applied to the recognition of orogenic and metallogenic processes globally. New datasets and interpretations will help designing more effective strategies for regional mineral and hydrocarbon exploration, which have so far been inhibited by the limited rock exposure. In particular, the new structural and thickness framework (Chapter 5) will help identification of pre-Eocene basement structure and reactivated faults during the Eocene, as well as prospective stratigraphic levels, constraining the location of existing and potential epithermal Au-Ag mineralization. Contribution to oil and gas exploration is indirect but significant, as the modelled extent and thickness of the Eocene sequence provide direct constraints on the depth of the Mesozoic basement that sourced and potentially hosts hydrocarbon and geothermal resources. For further information, research in the form of conference abstracts, with applications to natural resources explorations in BC, is reported in Appendix K. This research benefits from a considerable amount of previous work, including past projects by the Geological Survey of Canada (GSC), the British Columbia Geological Survey (BCGS) and Geoscience BC. Available datasets are summarized in Table 1.1.   21  Table 1.1. Previous work and summary of existing data in central BC   Dataset Reference● BCGS geological map Massey et al. (2005)● GSC Nechako NATMAP project Struik et al.  (2007)● QUEST surveys Geoscience BC (2007 to 2010)● Geology of the Nechako Basin Riddell (2006) ● Whitesail Lake (NTS 093E) Diakow and Mihalynuk (1987)● Nechako River and Fort Fraser (NTS 093F,K)Duffell (1959); Tipper (1963); Rouse and Matthews (1989); Diakow and Levson (1997); Diakow et al. (1997); Struik et al. (1997); Wetherup (1997); Anderson and Snyder (1998); Haskin et al. (1998); Barnes and Anderson (1999a,b); Grainger and Anderson (1999); Anderson et al. (2000); Dostal et al. (1998, 2001, 2005); Diakow (2006)● Quesnel (NTS 093B) Tipper (1959); Rouse and Matthews (1988); Metcalfe and Hickson (1995); Metcalfe et al. (1997)● Anahim Lake (NTS 093C) Tipper (1969); Mihalynuk et al.  (2008b, 2009a)● Tasako Lake and Bonaparte Lake (092O & P)Dohanney et al. (2010a,b);  Faulkner (1986); Hickson and Higman (1993); Mihalynuk and Harker (2007); Mahonney et al. (2013)Aerial aeromagnetic data and processed maps High-resolution regional aeromagnetic survey  in the Interior Plateau of British Columbia (1993-1994)Geological Survey of Canada (1994); Teskey et al. (1997); Geological Survey of Canada (2011)● High resolution Bouguer anomaly surveyDumont (2008a,b)● Hand-gridded Bouguer gravity map for the Nechako basinFerri and Riddell (2006)● CordAge database Breitsprecher and Mortensen (2004)● Recent isotopic dating of Eocene rocks Mihalynuk et al.  (2008a)● Dating of rock chips from oil and gas wells Riddell (2010, 2011) ● Whole rock and trace elements geochemistry for Eocene volcanic rocks Ewing (1981a), Green (1990), Drobe (1991), Hickson (1992), Metcalfe et al.  (1997), Anderson et al.  (1998), Breitsprecher et al. (2003), Dostal et al.  (2001, 2003), Ickert et al. (2009), Mihalynuk et al. (2009b), Bordet et al. (2011b)● Whole rock geochemistry and trace elements for the Chilcotin Group basalts and Cheslatta Lake suite basaltBevier (1983), Anderson et al.  (2001)Physical properties measurements BC Rock Property database (over 12000 measurements)Andrews et al. (2011a); Enkin et al. (2008)Data TypeGeological Maps and SurveysRegional geological compilations Bedrock and surficial geology maps - 1:50,000 or 1:250,000 scaleGeophysical SurveysGravity survey data and processed mapsAnalytical DatasetsIsotopic age dates for British Columbia Geochemical analyses 22  Table 1.1 (continued) Dataset Reference● 1100 km of seismic profiles (non public data)Canadian Hunter (1931-1986)● 5000 km of gravity surveys (non public data)● Drilling of 12 wells (logs and reports)Ministry of Energy and Mines website● Data compilation and reports Ferri and Riddell (2006), Riddell (2006), Riddell et al. (2007), Riddell and Ferri (2008)● 330 km of seismic reflection data Calvert et al . (2009), Hayward and Calvert (2009, 2011)● Magnetotelluric surveys Spratt and Craven (2009, 2010, 2011), Spratt et al.  (2012)● Tomography Smithyman and Clowes (2011, 2012, 2013), Smithyman (2013), Smithyman et al. (2014), Talinga and Calvert (2012)● List of mineral deposits, prospects and showings (MINFILE Mineral Inventory)BC Geological Survey (2013)● Mineral projects assessment reports (ARIS Database)BC Ministry of Energy and Mines (2013)● Regional metallogeny publicationsFriedman et al.  (2001), Mihalynuk (2007), Mihalynuk et al.  (2008c, 2009b)● Technical reports (NI 43-101) for Blackwater, Capoose, Blackdome, Clisbako, NewtonFaulkner (1986), Andrew (1988), Chapman and Kushner (2009), Awmack et al.  (2010), Simpson (2011), Looby et al. (2013), McClenaghan (2013), Petersen (2013)Mineral Exploration DataData TypeExploration for Natural ResourcesHydrocarbon exploration data23  Chapter 2  Stratigraphy of a voluminous silicic-intermediate volcanic sequence, Eocene Ootsa Lake Group, south-central British Columbia 2.1. Introduction The influence of Eocene Ootsa Lake Group (OLG) magmatism in central and southern British Columbia (BC) on the tectonic and metallogenic evolution of the North American Cordillera is poorly understood. As a result, the magmatic processes that led to the deposition of this broad volcanic province are also largely undefined. The OLG is characterized by: 1) predominantly silicic and lesser intermediate volcanic and volcaniclastic rock sequences; 2) geochemical variations along the Eocene volcanic belt, with arc signatures in central BC (Dostal et al. 2001, 2005) to intraplate signatures in southern BC (Breitsprecher et al. 2003); 3) volcanic activity constrained to a relatively short time span (~55-45 Ma; Grainger et al. 2001; this study); and 4) the extrusion of variably thick (Table 2.1), voluminous deposits, covering an extensive area of > 65,000 km2 (Figure 2.1). These characteristics do not support the formation of the OLG as a result of typical arc volcanism, as suggested in previous studies (e.g., Dickinson 1976; Ewing 1980; Armstrong and Ward 1993). Stratigraphic information exist for specific areas such as the Buck Creek basin in central BC (Figure 2.1; references in Table 2.1), and grabens of southern BC containing the Kamloops, Penticton and Princeton groups (references in Table 2.1). Based on these well-documented, widely-spaced localities, the Eocene volcanic sequence is spatially extensive, with measured thicknesses locally ≥2000 m (Table 2.1). Volcanic stratigraphy is essential for the correct identification of eruption and depositional modes, and structures controlling basin formation and preservation of this extensive volcanic sequence. However, regional scale integration of Eocene volcanic stratigraphy in BC has not been attempted so far; this study addresses this shortcoming. 24  Volcanic sequences of the OLG are widespread across the Chilcotin and Nechako plateaus of central BC (Figure 2.1; Massey et al. 2005), but are extensively covered by Neogene basalt and Quaternary glacial deposits. Six main map units of the OLG were defined based on discontinuous exposure, and are largely composite, comprising many petrographically, lithologically, texturally and geochemically distinct units. Map units thus defined are supported by geochronological constraints (see Chapter 3), physical property measurements made in-situ or in the laboratory (Bordet et al. 2011b; Appendix E), and correlations with Eocene localities elsewhere in BC. Using this stratigraphy and hydrocarbon exploration wells data, the thickness of individual units was estimated, and a thickness estimate for the OLG sequence across the Chilcotin Plateau was provided, with direct implications for regional resource exploration and targeting. However, the wide variability of structures across the region, with flat lying to significantly tilted sections (Appendix B), combined with the discontinuity of outcrop exposure typical of this region, make it difficult to assess the continuity of structural patterns, and their possible connection with regional faulting or folding events. Therefore the structural variations were not taken into account when assessing true vs apparent thicknesses of the units.   Table 2.1. Compiled thicknesses for the main Eocene Groups in central and southern BC    Orientation Group Name Locality Reported Thickness Range ReferencesEndako Group Nechako Plateau  ~ 600 m Diakow et al. 1997Buck Creek Complex Nechako Plateau > 1000 m Dostal et al. 2001, 2005Ootsa Lake Group Nechako Plateau  ~ 450 m Diakow et al. 1997Ootsa Lake Group Chilcotin PlateauPossibly up to 3700 mAverage 300-1500 m Armstrong 1949; Riddell 2010; Mihalynuk et al. 2009b; this study"Clisbako Volcanics" Chilcotin Plateau > 200 m Metcalfe et al. 1997Ootsa Lake GroupChilcotin Plateau - Blackdome ~ 400 mFaulkner 1986; Hickson et al. 1991Kamloops Group Southern BC  ~ 2500 mChurch 1973, 1985; Dostal et al. 2003Penticton Group Southern BC 100-2000 mChurch 1973; Ewing 1981b; Read 2000Princeton Group Southern BC 300-3400 mEwing 1981b; Thorkelson 1988; Souther 1991; Read 2000SouthNorth25   Figure 2.1  26  Figure 2.1. Extent of OLG volcanic rocks in the Nechako and Chilcotin plateaus of central BC. Extents are mostly inferred from field mapping because of poor exposure, and possibly include the surface area covered by Miocene Chilcotin Group basalt. Geological boundaries, terrane outlines and regional structures adapted from Massey et al. (2005). Eocene areas referred to in the text are indicated by numbers: 1) Mount Greer/Kenney Dam; 2) Fawnie and Nechako ranges; 3) Tibbles; 4) Nazko; 5) Baezaeko; 6) Clisbako; 7) B-22-K exploration well; 8) Chezacut; 9) southern Chilcotin Plateau; 10) Blackdome Mine. Eocene epithermal prospects and deposits are EQ: Equity Silver; AL: Allin; UL: Uduk Lake; LN: Loon; TR: Trout; WO: Wolf; CL: Clisbako; BA: Baez; BD: Blackdome. Inset map: general extent of Eocene volcanic rocks in BC and distribution relative to accreted terranes (CPC: Coast Plutonic Complex; ST: Stikine terrane; CC: Cache Creek terrane; QN: Quesnel terrane; NP: Nechako Plateau; CP: Chilcotin Plateau; SBC: southern BC).  2.2. Definition of the OLG  The OLG was originally defined from rock exposures southwest of Vanderhoof (Duffell 1959; Figure 2.1). It was later recognized as covering a vast portion of central BC, in the high-standing (~800-1600 m) region with subdued topography referred to as the Nechako and Chilcotin plateaus (Figure 2.1; Holland 1976).  The OLG comprises rhyolite and dacite lavas, silicic volcaniclastic rocks, minor basalt, andesite, and sedimentary rocks in variable proportions (Duffell 1959; Tipper 1959, 1969; Struik et al. 1997; Anderson and Snyder 1998; Haskin et al. 1998; Barnes and Anderson 1999a,b; Grainger and Anderson 1999). The OLG also hosts around ten epithermal Au-Ag prospects (Figures 2.1, 2.2; Faulkner 1986; Lane and Schroeter 1995, 1997; BC Geological Survey 2013). Eocene sequences dominated by mafic lavas were previously assigned to the Endako Group, originally defined near Vanderhoof (Figure 2.1) by Armstrong (1949), and later applied regionally to mafic to intermediate volcanic sections of broadly Eocene age (Figure 1.3; Anderson and Snyder 1998; Barnes and Anderson 1999a,b; Haskin et al. 1998; Grainger et al. 2001; Struik et al. 1997). However, investigations in the Chilcotin Plateau show that outside of the region in which the Endako Group was defined, the mafic to intermediate volcanic sections are commonly interleaved with silicic OLG strata (this study). Thefore, Endako rocks constitute a lithological subdivision of the Ootsa Lake Group rather than a disctinct group. In this paper, the OLG refers to Eocene volcanic rocks of central BC; the Endako Group is restricted to the region in which it was originally defined.   27  2.3. Tectonic and geological setting for OLG volcanism Eocene volcanic assemblages were deposited across tectonic terranes that were previously accreted to the ancestral North American margin during the Mesozoic and Paleogene; these terranes include Stikinia (island arc), Cache Creek (subduction-related accretionary complex) and Quesnellia (island arc) (Figure 2.1; Coney et al. 1980; Monger et al. 1982). The origin of the Eocene magmas was previously interpreted to be related to changes in the organization and relative motions between the Pacific, Kula and Farallon tectonic plates at the western margin of North America around 56 Ma (Ewing 1980; Engebretson et al. 1985). Inland, the Eocene tectonic regime was characterized by crustal extension and dextral faulting (Gabrielse and Yorath 1991; Struik 1993). Eocene OLG rocks are distributed between two regional dextral fault systems: the Yalakom Fault to the west and the Fraser Fault to the east (Figure 2.1). Jurassic strata of Stikinia include the Lower to Middle Jurassic volcanic arc successions of the Hazelton Group, formed of intermediate volcanic lavas and breccias and marine sedimentary rocks (Figure 2.2; Tipper and Richard 1976; Diakow 2006; Riddell 2011).  In the Nechako Plateau, these strata are overlain by Middle Jurassic to Lower Cretaceous deep to shallow water sedimentary successions of the Bowser Lake Group (Figure 2.2; Riddell 2011; Diakow and Levson 1997), and intruded by the Late Jurassic Laidman Pluton (Figure 2.2; Poznikoff et al. 2000), the Late Cretaceous Bulkley Intrusive Suite (85-74 Ma; Friedman et al. 2001; Souther 1991; Figure 2.2) and the Capoose Pluton (73-69 Ma; Friedman et al. 2001; Figure 2.2). A number of prospects for Au and base metals mineralization (e.g. Blackwater, Capoose; Figure 2.2) are associated with the young intrusions and comagmatic volcanic strata of the Kasalka Group (Figure 2.2, 71-68 Ma; Diakow 2006).  Thick beds of Cretaceous chert-pebble conglomerate and sandstone correlated with the Aptian to Albian Skeena Group (Figure 2.2; Riddell 2011) are exposed for ~150 km along the Nazko River (Figure 2.1) and are well-studied because of their potential to form hydrocarbon reservoirs (Ferri and Riddell 2006; Riddell 2011). Locally, they overlie Late Jurassic volcanic rocks of unknown affinity (Nazko Valley volcanic rocks, Figure 2.2; Riddell 2011). Various groups of Neogene to Holocene volcanic rocks overlie the OLG (Figure 2.2), including Chilcotin Group basalt (CGB, 28–1 Ma; Bevier 1983; Mathews 1989; Andrews and 28  Russell 2008), Cheslatta Lake Volcanic Suite basalt (21–11 Ma; Anderson et al. 2001), and alkaline rocks of the Anahim volcanic belt (14.5– 0.007 Ma; Bevier 1989).    Figure 2.2. Regional stratigraphy of the Nechako and Chilcotin plateaus. Compiled after Riddell and Ferri (2008), Riddell (2011), Gabrielse and Yorath (1991).  2.4. Field and analytical methods Field studies included geological mapping, measurements and description of stratigraphic sections. Fieldwork covered parts of the Quesnel (NTS 093B) and Anahim Lake (NTS 093C) map sheets in the Chilcotin Plateau (areas 3 to 10; Figures 2.1, 2.3) and the 29  Nechako Plateau (1 and 2; Figure 2.1). Access, location and descriptions of field traverses are detailed in Bordet et al. (2011b) and reproduced in Appendix A (Table A3). Gentle topography with extensive glacial till and forest cover limit rock exposures, but bedrock is exposed along ridges and small buttes, along streams, rivers and glacial outwash channels and along numerous logging roads. Regional mapping from road and foot traverse was conducted at 1:100,000 scale principally around the community of Nazko (Figures 2.1, 2.3). Detailed mapping at approximately 1:1,000 to 1:5,000 scales was conducted where key stratigraphic relationships are exposed (Appendix D). In addition, hydrocarbon exploration well logs B-22-K and B-16-J (Figures 2.1, 2.3) were evaluated and rock cuttings were examined for correlation with the surface geology and to constrain the stratigraphic thicknesses (Appendix G).  Major lineaments interpreted from aeromagnetic maps (Geological Survey of Canada 1994; Collin Barrett, pers. comm. 2012) offer some guidance for continuing geological contacts beneath glacial cover and for interpreting regional structures (Figure 2.3). Physical rock property measurements provide useful criteria for characterizing units. Direct magnetic susceptibility measurements (n ≈ 4650) were collected systematically in the field. Bulk densities of a suite of 219 samples were measured using the wet-dry density method. Analytical procedures and results are detailed in Bordet et al. (2011b) and Appendix E, and the range of magnetic susceptibility and density values for individual units are summarized in Table 2.2.  Whole rock geochemical analyses on a suite of 58 samples and petrographic observations illustrate the range of compositions of Eocene volcanic rocks and spatial compositional variations. Analytical methods are detailed in Bordet et al. (2011b) and in Chapter 4. Compositional ranges for individual units are summarized in Table 2.2. Detailed results are presented in Appendix I. U-Pb and 40Ar/39Ar geochronologic analysis of 33 igneous samples was conducted at the Pacific Centre for Isotopic and Geochemical Research (PCIGR) at The University of British Columbia to constrain the onset, duration and termination of Eocene volcanism. Complete analytical procedures for the U-Pb and 40Ar/39Ar methods, age results and interpretations are presented in Chapter 3 and Appendices F, G, H. Age ranges for each unit are summarized in Table 2.2. 30   Figure 2.3 31  Figure 2.3. Synthetic geological map for part of the Chilcotin Plateau highlighting the spatial distribution of OLG map units based on field mapping and observations. Mapping areas referred to in text are indicated by numbers: 3) Tibbles; 4) Nazko; 5) Baezaeko; 6) Clisbako; 7) B-22-K exploration well; and 8) Chezacut. Extent of undivided OLG is after Massey et al. (2005). Unit boundaries are inferred using total magnetic field data (Geological Survey of Canada 1994) when field relationships are insufficient. Eocene unit boundaries in the Chezacut area are after Mihalynuk et al. (2008b), some boundaries in the Clisbako area are after Metcalfe et al. (1997). Geological boundaries other than Eocene are from various sources: CGB after Dohanney et al. (2010a); Mesozoic after Massey et al. (2005) and Mihalynuk et al. (2008b) for the Chezacut area. Geochronological sample locations and ages are extracted from Chapter 3. Structural measurements are not reported, but major magnetic lineaments interpreted from aeromagnetic data (Geological Survey of Canada 1994; Collin Barrett, pers. comm. 2012) are traced.    32  Table 2.2. Synthesis of the lithological and analytical characteristics of the main OLG map units. References in table are: (1) Anderson and Snyder 1998; (2) Andrew 1988; (3) Armstrong 1949; (4) Barnes and Anderson 1999a,b; (5) Bordet et al. 2011b; (6) This study; (7) Dostal et al. 2001; (8) Faulkner 1986; (9) Grainger et al. 2001; (10) Haskin et al. 1998; (11) Metcalfe et al. 1997; (12) Mihalynuk et al. 2008c; (13) Mihalynuk et al. 2009b; (14) Rouse and Mathews 1988, 1989; (15) Tipper 1959. Range Median valueRhyolite lava, rhyolite porphyry, felsic tuff, volcaniclastic rocks (2,6,9,10)53-48 Ma (9) n/a n/a 0.4 - 4 < 1Biotite-quartz-plagioclase proximal rhyolitic ash tuff (9,10) 53-52 Ma (9) n/a n/a 3 - 20 10 1.7 - 2.1Basalt (Swans Lake Unit (7)) ~ 50 Ma (7) Basalt (7) 47-54% Andesite-basalt (Goosly Lake Fm. (7)) ~ 52 Ma (7) Basalt-Andesite (7) 49-60%Quartz-Biotite phyric rhyolite (6) 5.a-b 0.1 - 2 < 1Biotite-feldspar phyric rhyolite (6) 3 - 6 5Biotite-quartz phyric felsic block and ash flow breccia (6,11) 5.d-f ~ 49 Ma (6) Rhyolite (11) ~ 71%Biotite-feldspar ± quartz phyric rhyolite (6) 5.c ~ 49 Ma (6)250-800 mAcicular hornblende porphyritic dacite and platy-weathering dacite lavas (12,13)5.g 54-49 Ma (6) Dacite to rhyolite (11,12) 69-73%Chezacut-Chilanko Forks,B22K 8 - 15 11 2.4 - 2.6Biotite-Kfeldspar porphyritic stock (12) Andesite-dacite (12) 60-65% Chilanko ForksHornblende-Quartz-Biotite porphyry dykes (12) Chilanko Forks 100 mBiotite-phyric dacitic ash flow tuff with ignimbritic textures (12) 5.h ~ 48 Ma (6) Rhyolite (12) 71-75%Chezacut-Chilanko Forks,B22K 3 - 20 10 1.7 - 2.1Maroon and grey banded rhyolite lavas and domes (12) Chezacut-Chilanko Forks 4 - 14 10Conglomerate with wood fragments (13) Chezacut-Chilanko ForksAmygdaloidal pyroxene basalt  (12) Chezacut-Chilanko ForksAphanitic to plagioclase phyric dacite-rhyodacite (5,6) 5.i 51-47 Ma (6) Dacite to rhyodacite (5) 65-72% Chilcotin Plateau 4 - 15 9 2.4 - 2.5Hyloclastic breccias and autobreccias (5,6) 5.j Chilcotin PlateauPyroxene dacite (5,6) Nazko, Baezaeko 8-15 11 2.4 - 2.6Mount Sheringham pyroxene-dacite (13) Dacite (5,12) 63-66% Mt. Sheringham (B22K) 8 - 15 11 2.4 - 2.6Vitreous black dacite lava: perlitic lavas, flow breccias and dikes, black vitric ash flow tuffs (6, 12)5.k-m Dacite to rhyodacite (5-12) 64-70% Chilcotin Plateau 5 - 10 72.5 (coherent)2.3 - 2.4 (breccia)Rhyolitic ash tuff and lapilli tuff (6) 5.n ~ 50 Ma (6) n/a n/a Nazko, Baezaeko 3 - 20 10 1.7 - 2.1Andesite-basalt (Endako Gp. (1,4, 9,10,15))5.o ~ 49-43 Ma (9) Basaltic-andesite to andesite 52-62%Kenney Dam, Mount Greer (Nechako Plateau)7 - 25 18Andesite-basalt (6)~ 51-48 Ma (8)Eocene to Oligocene (15)Basaltic trachyandesite to trachyandesite (5)55-62%Tibbles, Nazko, BaezaekoBlackdome Mountain 7 - 25 18 2.4 - 2.6Variegated pyroclastic breccia 5.p Tibbles 8 - 28 15Cheslatta Falls basalt50 m Basalt (14) ~ 37 Ma (14) n/a n/aSouthwest of Fort Fraser (Nechako Plateau)2 - 20 8.5Intraformational or basal conglomerates and sandstones (3,6,12) Chilcotin and Nechako plateaus 2.3 - 2.4Epiclastic alluvial fan deposit (5,6) ~ 40 Ma (6) Baezaeko FigureMag. Susc. (10-3 SI)% SiO2 rangeAge range Geochemical classificationTibblesBuck Creek basinNechako PlateauClisbako 3 - 20 10 1.7 - 2.1Density range (g/cc)Mapping area70-77% 2.4 - 2.52.5 - 9 5Tibbles rhyolite Rhyolite (6)49-47 Ma (6)53-51 Ma (6)≥ 200 m55-52 Ma (6)300 mOotsa Lake Group undivided Chezacut dacite sequenceReworked Distal Volcaniclastic RocksLithologyNechako andesite Approximate Thickness Map unit≥ 100 mBaezaeko daciteMount Sheringham dacite Clisbako rhyolite ~ 400 m100-300 m33  2.5. OLG map units  OLG map units are composite rock assemblages comprising many petrographically, lithologically, texturally and geochemically distinct units, defined based on discontinuous exposure, that share similar characteristics of age, spatial extent and stratigraphic relationships. Individual units include coherent rocks and their associated brecciated facies, primary proximal volcaniclastic rocks and reworked distal volcaniclastic rocks (Table 2.2), described following the volcanic rock nomenclature of McPhie et al. (1993). Unit root names reflect the dominant geochemical composition of the unit (Figure 2.4), proceeded by a geographic qualifier referring to the area where the unit is the most widespread or has been defined. All the unit names are new and informal. In this study, rocks of the OLG are systematically subdivided across the combined Nechako and Chilcotin plateaus. The Chezacut dacite sequence and Clisbako rhyolite dominate the southwestern corner of the region (6, 7, 8; Figure 2.3), whereas the Tibbles rhyolite and the Nechako andesite are unique to the northeast (3; Figure 2.3). The Baezaeko dacite and Mount Sheringham dacite occupy the center of the mapped area (4, 5; Figure 2.3), but also occur in the southwest. Generally poor exposure across the Chilcotin and Nechako plateaus necessitate the development of composite sections (Figures 2.7 to 2.11) constructed from fragmentary sections that illustrate contacts and stratigraphic relationships (Appendix D). More complete descriptions of key units as well as those of lesser importance are presented, together with additional images from outcrop and petrography, in Appendix B and C. Units are described here in stratigraphic order. The Tibbles and Clisbako rhyolite and Chezacut dacite sequence constitute one of the lowest stratigraphic levels, commonly overlying a basal conglomeratic unit. The Baezaeko dacite unit is regionally at a higher stratigraphic level, although it represents the base of the Eocene sequence in the Baezaeko area. The Mount Sheringham dacite caps the Eocene stratigraphy in most areas of the Chilcotin Plateau. The Nechako andesite occupies the highest Eocene stratigraphic level on the Nechako Plateau.  34   Figure 2.4. Whole rock major element classification diagrams for OLG volcanic rocks from the Chilcotin Plateau. Individual samples are classified by lithology, but broad map unit fields are indicated. Regional trends for Eocene volcanic rocks in central and southern BC (compiled after Ewing 1981a, Drobe 1991, Hickson 1992, Metcalfe et al. 1997, Anderson et al. 1998b, Dostal et al. 2001, Mihalynuk et al. 2009b), CGB (Bevier 1983; Anderson et al. 2001) and Chesltatta Lake suite basalt (Anderson et al. 2001) are reported for reference. a) Total Alkali Silica diagram, after Le Maitre et al. (1989); b) AFM diagram, after Irvine and Baragar (1971). 2.5.1. Tibbles rhyolite The Tibbles rhyolite unit covers the western half of the Tibbles area (3; Figure 2.3). Compilation maps (Massey et al. 2005) indicate that  it overlies Carboniferous to Triassic rocks of the Cache Creek terrane (Figure 2.3; Monger et al. 1982), but the presence of zircon 35  xenocrysts derived from Jurassic rocks suggests that the rhyolite directly overlies Jurassic basement, from which zircons were incorporated into Eocene silicic magmas (see Chapter 3). Calculated on the basis of topographic intersections, the thickness of this unit is estimated at ≥ 200 m. A contact with the overlying Nechako andesite unit is marked by a baked zone at one exposure (Stratigraphic column on Figure 2.6).  The Tibbles rhyolite unit is dominantly flow-banded biotite-feldspar-phyric rhyolite. A quartz-phyric subunit has more limited extents, occurring mainly within the south-central part of the Tibbles area (Figure 2.5.a,b). Maroon and grey banded rhyolite, plagioclase-phyric dacite and rhyodacite are locally interfingered with rhyolite lava (3; Figure 2.3, Stratigraphic column on Figure 2.6). Orange-weathered, pale-yellow to white rhyolite contains euhedral and fresh biotite (1-2%, up to 5 mm), feldspar (1-3%) and magnetite (up to 7%).  Flow-top brecciation is observed locally, and internal flow-banding is common, and locally contains flattened vesicles. Smoky quartz crystals comprise up to 3% of the quartz-phyric rhyolite variety, which also contains between 1-2% biotite crystals and no magnetite. Silicification of the groundmass is obvious in hand sample and is confirmed by petrographic observations (Figure 2.5.b). Tibbles rhyolite is dated between 53-51 Ma (Table 2.2; see Chapter 3). SiO2 content is between 70 and 77% (Figure 2.4). Average magnetic susceptibility is of 5·10-3 S.I. (Table 2.2). Flow-banding plane directions are relatively consistently oriented E-W over the entire area, with most dips between 20° and 60° towards the north or south. The Tibbles rhyolite is interpreted regionally as a stratigraphic equivalent to the Clisbako unit and other undivided OLG rhyolite lavas (Table 2.2). 36   Figure 2.5 37   Figure 2.5 (continued) 38   Figure 2.5 (continued)   39  Figure 2.5. Field and microscopic photographs of Eocene OLG volcanic rocks of the Chilcotin and Nechako plateaus. (a) Exposure of biotite-quartz-feldspar-phyric rhyolite of the Tibbles rhyolite unit, Tibbles area; (b) Microscopic view of the biotite-quartz-feldspar-phyric rhyolite of the Tibbles rhyolite unit, Tibbles area, field of view 4.2 mm; (c) Brecciated, silicified, hematite stained rhyolite with vuggy quartz-veins at the Baez epithermal prospect, Clisbako rhyolite unit; (d) White, crumbly, clay weathered monomictic block and ash flow deposit with blocks of biotite-quartz-phyric rhyolite, Clisbako rhyolite unit, Clisbako area; (e) Microscopic view of the block and ash-tuff of the Clisbako rhyolite unit, displaying feldspar, quartz, biotite crystals and vesiculated glassy pyroclasts, field of view 4 mm; (f) East-dipping beds of volcanic conglomerate and tuffaceous sandstone of the Clisbako rhyolite unit, containing large angular blocks of brown-purple glassy plagioclase-phyric lava and biotite-quartz-phyric rhyolite; (g) Finely platy-jointed acicular hornblende dacite, Chezacut dacite unit; (h) Dacite ash tuff part of the Chezacut dacite unit,  B-22-K area. Crystals include plagioclase (3-4%), biotite (1%) and smoky quartz (1%); (i) Aphanitic dacite part of the Baezaeko dacite unit; (j) Outcrop and microscropic view (field of view 10 mm) of flow-associated monomictic quench or hydrothermal breccia with a red-pink finely grained matrix and angular blocks of aphanitic dacite, Baezaeko dacite unit, Baezaeko area; (k) Steeply-dipping, columnar-jointed exposure of vitreous black dacite part of the Mount Sheringham dacite, Baezaeko area; (l) Perlitic fractures in vitreous black dacite, Mount Sheringham dacite, field of view is 4.2 mm; (m) White weathered brecciated facies of the Mount Sheringham dacite in the Baezaeko area; (n) Accretionnary lapilli-tuff part of the Mount Sheringham dacite, Baezaeko area; (o) Spectacular cliff of flow-banded andesite part of the Nechako andesite unit at Kenney Dam, and macroscopic view of a plagioclase-phyric andesite at Mount Greer; (p) Variegated monomictic pyroclastic breccia, Nechako andesite unit, Tibbles area.   40   Figure 2.6. Composite stratigraphic column for the Tibbles area (3; Figure 2.3). Includes legend for Figures 2.7 to 2.11. Ages extracted from this study (see Chapter 3). 41  2.5.2. Clisbako rhyolite The Clisbako rhyolite unit occurs between the Clisbako and Baezaeko areas (6, 7; Figure 2.3), and covers an area of ~30×20 km2 as mapped and inferred from aeromagnetic data (Geological Survey of Canada 1994; Dumont 2008a,b). Minimum thickness of the unit is ~400 m, but a composite thickness of 2000-3000 m is attained according to B-22-K well log information (see following discussion and Appendix G), likely resulting from successive deposition of several eruptive units. The unit is inferred to cover Mesozoic layered rocks and intrusions although the contact has not been observed directly. Overlying units include dacitic lavas and breccias of the Baezaeko dacite and the Mount Sheringham dacite. Regionally, the Clisbako rhyolite unit is spatially distinct but coeval with the Tibbles rhyolite and Chezacut dacite sequence, and also temporally equivalent to the OLG silicic volcanic and volcaniclastic sequence mapped in the Nechako Plateau (1; Figure 2.1 and Table 2.2; Grainger et al. 2001; Haskin et al. 1998; Wetherup 1997). The Clisbako rhyolite is predominantly thick rhyolite lava and breccia (Figure 2.5.c; Stratigraphic column on Figure 2.7) interbedded with rhyodacite and dacite lavas, as shown by drill intersections at the Baez and Clisbako prospects (Brown and Strand 2008; Chapman and Kushner 2009). A coeval fragmental package of silicic pyroclastic breccia and volcaniclastic sandstone and conglomerate is mapped to the eastern extent of the unit (Figure 2.5. d-f; Stratigraphic column on Figure 2.7). Dacite ash tuffs are present in subordinate amounts.            42   Figure 2.7. Composite stratigraphic column for the Clisbako-B-22-K areas (6-7; Fig 3). Ages from Metcalfe et al. (1997), and this study. 43  2.5.2.a.  Rhyolite lava at Baez and Clisbako Typical rhyolite lava exposed around and at the Baez and Clisbako prospects (Figure 2.3) is yellow-orange weathered, white to-pale yellow flow-banded and brecciated, containing biotite (up to 10%, 2-3 mm) and plagioclase phenocrysts (up to 3%). Intervals of porphyritic rhyolite are distinguished by the abundance of purple to smoky quartz eyes (typically 2-4%) and milky white to transparent feldspar phenocrysts (4-5%; up to 1 cm) and biotite (4-7%). Flow-banding have dips between 30° and 60° with no consistent azimuth.  Rhyolite from the Baez property was dated at ~49 Ma (see Chapter 3). SiO2 contents in excess of 73 weight % (Figure 2.4) may reflect local silicification associated with hydrothermal alteration. In the extreme case, it is invaded by subvertical quartz veins within mineralized zones at both the Clisbako and Baez low-sulfidation epithermal prospects, where clay alteration may also be intense (Figure 2.5.c). Magnetic susceptibility measurements allow the distinction between quartz-phyric rhyolite lava with an average susceptibility <1·10-3 S.I., and biotite-phyric lava with a higher range of values from 3·10-3 to 7·10-3 S.I. (Table 2.2).  2.5.2.b.  Silicic pyroclastic breccia The coherent silicic sequence is laterally equivalent to a ≥ 30 m thick volcaniclastic sequence exposed mostly to the east of the Clisbako prospect (Figure 2.3) and previously reported by Metcalfe et al. (1997) as part of a “biotite-bearing assemblage” dated at ~49 Ma (Metcalfe et al. 1997; this study). Clay weathered, white-grey, crumbly, monomictic matrix-supported poorly-sorted volcanic breccia displays angular blocks of biotite-(5-6%) and quartz-(3-4%) phyric rhyolite in a tuffaceous matrix of identical composition (Figure 2.5.d, e). Sparse blocks of aphanitic black dacite and ochre lava comprise less than 1% of the rock volume. This sequence overlies beds of tuffaceous sandstone to coarse and chaotic volcanic conglomerate with large blocks of grey-red lava, brown-purple glassy plagioclase-phyric lava, and biotite-quartz-phyric rhyolite (Figure 2.5.f). The beds dip 30-40⁰ to the east. The top of the sequence is marked by a gradual redenning level, which could result from baking by overlying flows of the Mount Sheringham dacite (Stratigraphic column on Figure 2.7).  A circular magnetic anomaly associated with the extent of the Clisbako rhyolite unit suggests a large silicic dome. Rhyolitic tuffs are interpreted as part of a proximal block and ash 44  tuff or debris flow deposit overlying and interbedded with volcaniclastic intervals, and deposited coevally and at the edge of the silicic dome. The Tibbles and Clisbako rhyolite units share lithological and stratigraphic similarities. However, the Tibbles rhyolite is not known to be associated with hydrothermal alteration and mineralization, and trace element data suggest a different fractionation pattern, perhaps controlled by eruption through and deposition on different basement, Cache Creek versus Stikine terrane assemblage (Massey et al. 2005; see Chapter 4).  2.5.3. Chezacut dacite sequence The Chezacut dacite sequence here refers to the ~1500 m thick Eocene lithostratigraphic package described in the Chezacut and Chilanko Forks map areas (Figure 2.1) by Mihalynuk et al. (2008b,c, 2009a,b). A basal polymictic conglomerate constitutes locally the base of the sequence (Figure 2.3; Stratigraphic column on Figure 2.8; Mihalynuk et al. 2008b,c, 2009a,b), and was deposited on Triassic-Jurassic volcanic and intrusive complexes (Mihalynuk et al. 2008b,c). Acicular hornblende porphyric dacite lavas constitute the dominant exposed rock type of the Chezacut dacite sequence (Figure 2.5.g), and are cut by porphyritic intrusive rocks (Table 2.2; Stratigraphic column on Figure 2.8). Dacite lavas are regionally interfingered with dacitic ash-tuff (Figure 2.5.h) and overlain by sequences of maroon and grey rhyolite (Figure 2.3, Stratigraphic column on Figure 2.8). Other minor components of the Chezacut dacite sequence are listed in Table 2.2. Acicular hornblende porphyric dacite lava is characterized by exposures of black, with lesser pink to tan-grey or rusty-weathered lava, sometimes showing brown-grey flow laminations. Its thickness is estimated between 250 and 800 m (Mihalynuk et al. 2008b). Hornblende forms less than 1 to 3% of the rock; sanidine is present up to 2% to 7% and commonly weathers brown (Mihalynuk et al. 2008c). Brown-yellow quartz crystals comprise 1% to up to 5% of the rock (Mihalynuk et al. 2008c). Hornblende alignment with well-developed flow-banding layers is characteristic of this unit (Mihalynuk et al. 2008c); flow-banding planes commonly split up to form angular splintery pieces of rock (Figure 2.5.g).  Ages for the Chezacut dacite sequence are between 54-48 Ma (see Chapter 3). Older porphyritic intrusive rocks (~55-52 Ma; see Chapter 3) locally intersect the Eocene sequence. The geochemical signature of the Chezacut dacite ranges from dacite to rhyodacite, with 69 < 45  SiO2 < 73% (Figure 2.4). The average magnetic susceptibility of the hornblende porphyric lava is 11·10-3 S.I., slightly higher than other dacitic lithologies (Table 2.2).   Figure 2.8. Composite stratigraphic column for the Chezacut area (8; Figure 2.3). Geology modified after Mihalynuk et al. (2008c, 2009b), ages extracted from this study. 46  2.5.4. Baezaeko dacite The Baezaeko dacite is widespread across the Chilcotin Plateau, mostly in the Baezaeko and Nazko areas (4, 5; Figure 2.3). It is at least 100 m thick from exposed cliff sections in the Baezaeko area. Alternate, presumably interbedded sequences of dacite and rhyodacite form north-trending map patterns inferred from aeromagnetic surveys interpretations (Geological Survey of Canada 1994). The Baezaeko dacite represents the base of the Eocene sequence in these two areas: east-dipping flow-banded dacite unconformably overlies deformed south-east dipping Albian conglomerates in the Nazko area and in the east of the Baezaeko area; to the west, it overlies Jurassic rocks of the Hazelton Group (Figure 2.3). The unit is inferred to occur at a lower stratigraphic level than the Eocene Mount Sheringham dacite unit, but field relationships are commonly unclear and the two units may be locally interfingered.  The Baezaeko dacite comprises aphanitic to plagioclase-phyric dacite and rhyodacite (Figure 2.5.i,j), and pyroxene-phyric dacite in subequal amounts (Stratigraphic column on Figure 2.9). Subordinate amounts of biotite-quartz-phyric rhyolite and andesite lava occur, and the base of the unit is marked by a volcaniclastic interval in the Nazko area (Stratigraphic column on Figure 2.9).  2.5.4.a.  Aphanitic to plagioclase-phyric dacite  Flow-banded aphanitic to plagioclase-phyric dacite-rhyodacite lavas are present mostly east of the Baezaeko area and in the Nazko area (Figure 2.3). They form orange-dark brown weathered, flow-banded to platy-jointed lava that locally is highly silicified. Fresh rock is light to dark grey-purple to black glassy, massive to vesicular, aphanitic to microcrystalline dacite (Figure 2.5.i). Amygdules are commonly coated with yellow-orange crystals. Spectacular exposures in the Baezaeko area (5; Figure 2.3) display monomictic, matrix supported, silicified dacitic hyaloclastic breccia (Figure 2.5.j), spatially associated with coherent flow-banded aphanitic dacite lavas. Under the microscope, a microcrystalline groundmass contains dominantly plagioclase, as well as hornblende, pyroxene and opaques minerals.  2.5.4.b.  Pyroxene dacite Flow-banded to brecciated pyroxene (± hornblende) dacite is widely exposed across the Nazko, Baezaeko areas and also further south near the B-22-K exploration well (4, 5, 7; Figure 47  2.3). From field relationships in the Nazko valley, the pyroxene dacite is inferred to overlie a dacitic-rhyolitic sequence, and is exposed immediately below the basal contact of the Mount Sheringham dacite (Stratigraphic column on Figure 2.9). Coherent flow-banded to massive dark grey pyroxene-phyric dacitic lava interbedded with thick layers of proximal, block-supported monomictic scoriaceous breccia forms high relief, red-brown oxidized lava and breccia domes along the Nazko River valley (4; Figure 2.3). Columnar-jointed plagioclase-phyric andesite lava, biotite-quartz-phyric rhyolite, and layers of white lithic tuff, are interbedded with the lava breccia and flow lobes (Stratigraphic column on Figure 2.9). In the Baezaeko area (5; Figure 2.3), platy-jointed finely flow-banded, dark grey-green plagioclase-pyroxene-phyric dacite exposures with oxidized exposed surface are ascribed to the same unit. The flow-banded dacite commonly yields euhedral to subhedral plagioclase, pyroxene and feldspars in a cryptocrystalline chlorite-altered matrix. Phenocrysts are usually less than 1%, but plagioclase crystals can be up to 5% and black prismatic pyroxene crystals are up to 2% in an aphanitic groundmass. They are highly fractured and display quartz-chalcedony filling along fracture planes. An age of 51 Ma was obtained for lithologies of the Baezaeko dacite unit (Table 2.2; see Chapter 3), and an age of ~47 Ma was previously reported (Metcalfe et al. 1997). The geochemical signatures range from dacite with less than 65% SiO2, to rhyodacite with as much as 73% SiO2 (Figure 2.4). This range of compositions and trend is followed by most dacitic units of the Chilcotin Plateau, but is distinct from the trachytic trend followed by samples of the Tibbles area. The average magnetic susceptibility is 9·10-3 S.I. (Table 2.2; Figure 2.4). Physical property signatures are relatively homogeneous for the various coherent dacite that form the Baezaeko dacite unit (Appendix E). Aphanitic dacite, plagioclase-phyric dacite and hornblende-pyroxene-phyric dacite range from 6 to 15·10-3 S.I. (Table 2.2). Lithologies of the Baezaeko dacite are interpreted as coalesced silicic lava flow lobes or domes, formed in one or many silicic volcanic centers. 48   Figure 2.9. Composite stratigraphic column for the Nazko-Baezaeko areas (4-5; Figure 2.3). Ages from Metcalfe et al. (1997), and this study. 49  2.5.5. Mount Sheringham dacite The Mount Sheringham dacite unit includes two lithological assemblages originally defined by Mihalynuk et al. (2008c, 2009b) in the Chezacut map area: the vitreous black dacite and the “Mount Sheringham pyroxene-dacite” (Table 2.2; Stratigraphic column on Figure 2.8). The Mount Sheringham pyroxene-dacite is a subunit of the vitreous black dacite that sits at the highest stratigraphic level, and is part of the “Pyroxene-bearing assemblage” of Metcalfe et al. (1997). Similar lithologies cover a vast area in the Chilcotin Plateau, from the west of the Nazko valley to the B-22-K area (4, 5, 7, 8; Figure 2.3; Bordet et al. 2011b), and are referred to collectively as the Mount Sheringham dacite unit.  Intersections of field exposures with topography indicate a minimum thickness of 300 m. North-trending elongated map patterns are inferred from aeromagnetic surveys (Geological Survey of Canada 1994; Teskey et al. 1997), and outline possible interlayering with other dacitic units. However, the Mount Sheringham dacite caps the Eocene stratigraphy in all areas, as previously reported by Mihalynuk et al. (2008c, 2009b). Coherent, columnar-jointed, dark grey vitreous black dacite is the most common and distinctive lithology, and is exposed as dacitic domes, and possibly subvolcanic intrusions, but occur mainly as tabular lavas. Interflow pyroclastic facies include white-weathered basal breccia (Figure 2.5.k-n; Stratigraphic columns on Figures 2.8, 2.9, 2.10). Outcrops of vertically to horizontally columnar jointed black dacite lava typically cap the highest topographic points, forming ridges and buttes (Figure 2.5.k). This lithology has a yellow-tan-weathered color, whereas the fresh rock displays a characteristic vitreous black groundmass, with common perlitic fractures (Figure 2.5.l). Lavas can be aphanitic to plagioclase-phyric, locally with quartz or pyroxene phenocrysts. Quartz may be difficult to identify in hand sample, appearing brown, amber or olive-green because of glass inclusions within their outer growth zone (Mihalynuk et al. 2009b). Plagioclase form euhedral phenocrysts laths up to 10-15%, but most commonly ranges between 2-4%. Biotite or hornblende is present up to 2%. Titanite is a minor accessory phase.  Fragmental facies associated with the vitreous dacite lava are very distinctive: blocks of black vesicular dacite up to 50 cm across are supported by a crumbly highly weathered pale yellow to orange pelagonitic matrix (Figure 2.5.m). At the Baezaeko area, layers of lapilli and tuff 50  (Figure 2.5.n) are covered by a chaotic fragmental unit with meter-size blocks of vitreous black dacite lava (Stratigraphic column on Figure 2.9). U-Pb and 40Ar/39Ar geochronological age determinations from the Mount Sheringham dacite range between 49 and 47 Ma (see Chapter 3; Table 2.2). This is the youngest OLG lithology at most places across the Chilcotin Plateau (Stratigraphic columns on Figures 2.8, 2.9, 2.10). The lower contact is defined by an interval of volcanic sandstone, chalky lapilli tuff and conglomerate underlying columnar-jointed vitreous black lava in the Nazko area, dated at ~50 Ma, constraining the maximum age of eruption of the iverlying Mount Sheringham dacite (see Chapter 3; Table 2.2). SiO2 content is 64-71 weight %. Average magnetic susceptibility is ~5·10-3 S.I. (Table 2.2).  2.5.6. Nechako andesite The Nechako basaltic-andesite unit corresponds to mafic-intermediate rock sections of the Nechako Plateau (1, Figures 2.1, 2.5.o; Kenney Dam and Mount Greer), where they were referred to as Endako Group in former studies (Anderson and Snyder 1998; Barnes and Anderson 1999a,b; Grainger and Anderson 1999; Haskin et al. 1998). Because of their lithostratigraphic similarities, andesite sequences of the Tibbles area in the Chilcotin Plateau (3; Figure 2.3) are ascribed to the same map unit, with the Nechako andesite unit inferred to be temporally equivalent to the youngest Mount Sheringham dacite unit (see Chapter 3). In the Tibbles area, scattered andesite exposures are aligned along a ~40 km long NNE-trending ridge (Figure 2.3) outlined by aeromagnetic and topography patterns, and have an apparent thickness of 300 m. At Kenney Dam and Mount Greer (Figure 2.1), sections of 95 m to over 100 m in stratigraphic thickness were measured (Figure 2.5.o; Haskin et al. 1998). The Nechako andesite unconformably overlies the Tibbles rhyolite unit (Stratigraphic column on Figure 2.6) and Eocene Ootsa Lake rhyolite lavas and tuffs of the Nechako Plateau, such as at Mount Greer (Figure 2.1; Stratigraphic column on Figure 2.10; Grainger et al. 2001). In the Nechako Plateau, it is overlain by the Eocene Cheslatta Falls basalt unit (Stratigraphic column on Figure 2.10; Table 2.2). The Nechako unit is basaltic to andesitic in composition. In addition to lavas, characteristic breccias occur in the Tibbles area (Figure 2.5.p; Stratigraphic column on Figure 2.6).  Orange-weathered, massive to columnar-jointed basalt-andesite lavas are typical. Grey to tan-weathered or dark grey-purple, flow-banded aphanitic to microcrystalline andesite units 51  contain plagioclase crystals (1-2%, locally up to 10-20%), and pyroxene or hornblende crystals (4-7%) in a trachytic microcrystalline plagioclase to glassy groundmass (Figure 2.5.o). Orange weathered autobrecciated flow-tops may be scoriaceous. Outcrops are locally, highly fractured and faulted, and quartz-chalcedony is commonly observed along fracture plans and within vesicles. Flow-elongated vesicles are commonly coated with calcite, based on microscopic observations. Vertical breccia pipes are locally developed. Possible pillows with typical quenched rind occur at the base of the coherent basalt sequence at Kenney Dam (1, Figure 2.1; Haskin et al. 1998). If this interpretation is correct, at least part of the succession was deposited under water.  Age of the andesite package was estimated as Eocene to Oligocene by Tipper (1959), as it overlies rhyolitic to dacitic sequences of Middle Eocene age. Reported K/Ar whole rock ages for the Nechako andesite lava obtained from oil exploration wells near Mount Greer and Kenney Dam are between 49 to 43 Ma (Rouse and Mathews 1989). Andesite lavas in the Tibbles area are trachytic, with SiO2 content of 55 to 61 weight % (Figure 2.4). Locally magnetic (~30·10-3 S.I.) linear exposures of trachyandesite are interpreted as feeder dikes (Stratigraphic column on Figure 2.6), but the andesite lavas typically range between 10·10-3 to 25·10-3 S.I. (Table 2.2).  The stratigraphic position and age of the Nechako andesite unit makes it the youngest Eocene volcanic package in central BC, and indicates a magmatic evolution from silicic towards more mafic compositions.  52   Figure 2.10. Composite stratigraphic column for the Kenney Dam/Mount Greer areas in the Nechako Plateau (1; Figure 2.1).  Compiled after Wetherup (1997), Haskin et al. (1998), Grainger et al. (2001). K/Ar and playnology ages from Rouse and Mathews (1988, 1989), other ages from Grainger et al. (2001).    53  2.5.7. Summary of the OLG stratigraphy In the Chilcotin Plateau, the OLG is composed of six main composite map units. At the lowest stratigraphic levels, the Chezacut dacite sequence, Clisbako rhyolite and Tibbles rhyolite units and intrusive equivalents locally overlie (or cut) a basal conglomeratic sequence. These units are isotopically dated between 55 and 48 Ma using 40Ar/39Ar and U-Pb techniques (Table 2.2; see Chapter 3). The Baezaeko dacite unit is at a higher stratigraphic position, and overall displays younger ages of 51-47 Ma, but locally interfingers with the basal units. The Mount Sheringham dacite caps the OLG sequence across the Chilcotin Plateau, and returns ages ranging from 49 to 47 Ma (Table 2.2; see Chapter 3). Finally, spatially restricted basalt and andesite lavas previously ascribed to the Endako Group in the Nechako Plateau sit atop Tibbles rhyolite, and are interpreted to be stratigraphically coeval with the Mount Sheringham dacite. Unfortunately, only cooling ages (mostly whole rock analyses) are available for the Nechako andesite (~49-43 Ma; Table 2.2; Armstrong 1949; Rouse and Mathews 1989; Haskin et al. 1998) so age equivalence (or not) has yet to be firmly established.   2.6. Major element geochemistry Whole rock geochemical analyses illustrate the geochemical signature and trend of OLG lithologies, and support first order interpretations of their origin, and regional correlations. Chapter 4 and Appendix I present major and minor elements analyses in detail, and address the tectonic and petrogenetic implications of their geochemical variability. Eocene volcanic rocks from the Chilcotin Plateau are high-K calc-alkaline (Peccerillo and Taylor 1976) basaltic-trachyandesite to rhyolite with SiO2 contents between 55 to 77 % (Table 2.2; Figure 2.4). Most samples are dacite and rhyodacite (Figure 2.4). Geochemical discrimination of individual dacitic units, or between the Tibbles and Clisbako rhyolite units, is challenging based solely on major element geochemistry. Compositional trends in this study are consistent with data obtained from other areas of the Chilcotin Plateau (Figure 2.4a; Metcalfe et al. 1997, Mihalynuk et al. 2009b). In the Nechako Plateau and southern BC, compositions are locally, significantly more mafic (Figure 2.4a). For example, in the Buck Creek basin (Figure 2.1) compositions are predominantly basaltic and 54  andesitic-basaltic (Table 2.2; Dostal et al. 2001). Of all samples analyzed, only those from the Tibbles area follow a distinct, more alkaline trend (see Chapter 4). Overall, OLG temporal compositional trends in the Chilcotin Plateau are silicic (rhyolite) to intermediate (andesite), which may suggest an evolution opposite to what is expected with normally fractionating magma chambers, or the existence of distinct, non-cogenetic magmatic sources.   2.7. Thickness of the Eocene sequence In order to provide a thickness estimate for the OLG sequence in the Chilcotin Plateau, a stratigraphic section was generated (Figure 2.11) that integrates thickness data from regional and detailed mapping as well as hydrocarbon wells (Figure 2.1). Composite thickness of the OLG on the Chilcotin Plateau is estimated by adding up individual unit thicknesses as determined from surface exposures, and is conservatively evaluated at 1400 m, but could be as much as 2300 m (Table 2.2). Where well-exposed and constrained by drilling such as at the Blackdome epithermal gold deposit near the southern edge of the Chilcotin Plateau (Figure 2.1), the thickness of the Eocene sequence is 400 m (10; Figure 2.11; Faulkner 1986; Hickson et al. 1991). Thicknesses are >3000 m within interpreted grabens of southern BC (Table 2.1, and references therein). In the Nechako Plateau, the maximum composite thickness of the Eocene sequence is ~1050 m in the vicinity of the Wolf and Trout epithermal prospects (2; Figure 2.1 and Figure 2.11; Diakow et al. 1997). Hydrocarbon wells B-22-K and B-16-J (Figure 2.3), respectively, intersected 3778 m and 2700 m of rock. Seven U-Pb zircon ages between 50 and 60 Ma (Riddell 2010) indicate Eocene ages down to at least 3778 m in B-22-K (4 Eocene ages), and to about 2000 m in B-16-J (3 Eocene ages) where a contact with underlying Jurassic basement is established (Riddell 2011). However, palynology data constrain the depth of the Eocene sequence in B-22-K at a maximum of 720 m (Hunt 1992). From reexamination of the rock chip samples (Appendix G), both the diversity of Eocene rock types and textures mapped in the surface were represented in the cuttings and confident correlation with surface units could be extended down to ~1000 m depth in B-22-K. However, rock chips from greater depths have less clear correlatives in Eocene, nor within the Cretaceous or Jurassic units at surface. Attempts to resolve the geochronological 55  ambiguity were unsuccessful, so the stratigraphy as established by Riddell (2011) is retained, recognizing that ambiguities add large uncertainties may exist in assessing the stratigraphy in the vicinity of the wells.  56  Figure 2.11 57  Figure 2.11. Stratigraphic correlations between Eocene sections (numbered according to Figure 2.1) across the Nechako and Chilcotin plateaus of central BC, highlighting the lateral distribution of Eocene map units and lithologies. Vertical scale is approximate, horizontal scale is estimated at 350 km but distance between individual sections is not representative. Individual synthetic logs are constructed from field observations and literature. Thickness constraints are from references indicated in the text, and include oil and gas wells or mineral exploration drilling constraints where available. All contacts are inferred due to poor exposure, either within the Eocene section, or between the Eocene sequence and Mesozoic rocks. Dashed lines indicate the temporal boundaries of the Eocene volcanic event, as defined in this study (see Chapter 3). Geochronological samples are compiled from this study and Breitsprecher and Mortensen (2004).  2.8. Discussion 2.8.1. Extensive areal distribution vs high viscosity of silicic lavas  Areal extent of the OLG inferred from mapping is 6,000 km2 in the Chilcotin Plateau (Figure 2.3) and ~65,000 km2 for south-central BC (Figure 2.1). Based upon the stratigraphic columns (Figure 2.11), the average thickness is ~2 km. However, these measures are for exposed relics of an originally more extensive blanket that has been eroded away in many places, and represent minimum area and thickness estimates. In the Chilcotin Plateau, the OLG sequence is predominantly silicic lavas, and minor pyroclastic tuff, breccia and volcaniclastic deposits. Lava successions likely formed from coalesced, extensive subaerial silicic lavas and/or domes. Because of their high viscosities (Richet and Bottinga 1995), individual silicic lavas and domes are typically thick (~100 m), are less than a few kilometers in extent, and represent small volumes (<1 km3) (McPhie et al. 1993); OLG lavas greatly exceed these dimensions.  The formation of regionally extensive coherent rhyodacite or dacite lavas depends on physical properties of the magma, including supply rate and volume, and viscosity which is primarily controlled by composition, volatile content and temperature (Giordano et al. 2008; see Chapter 4). Several end-members exist: 1) silicic lavas with average eruption temperatures and viscosities, but which are particularly thick and remain hot, such as those with a voluminous magma supply; 2) low viscosity flows as a consequence of high eruption temperatures; and 3) low viscosity flows arising from high retained volatile contents, such as those erupted in deep subaqueous environments, where ambient pressure prevents exsolution of the gas phase (e.g., Scutter et al. 1998; Dinel et al. 2008). The last two end-members allow for greater mobility of 58  rhyolite lava by low viscosity. However, most OGL volcanic facies are subaerial, with no record of deep water sedimentation, so the third end-member is not considered further. 2.8.1.a.  Normal eruption temperature, thick flow Low viscosities are not required to generate extensive silicic lavas provided that the flows are thick enough (100-300 m) so that flow interiors retain sufficient heat, and that the available magma supply is large (Manley 1992). This is the case of the calc-alkaline, high-silica, highly porphytic Badlands rhyolite lava of Miocene age in southwest Idaho, USA, which covers 15 km3 and advanced up to 9 km from a fissure-style vent, and for which viscosity considerations are presented (Manley 1996). The viscosity of this lava was calculated at ~9.5-10 log ɳ Pa.s at ~830 °C, with a low magmatic water content (< 2.75 weight % H2O) having prevented a large explosive eruption (Manley 1996). Similarly, the highly porphyritic, calc-alkaline high-K Chao dacitic lava, Chile, is 14 km long with a thickness of 400 m at the flow-lobe front and viscosity estimated in the range of other known silicic lava extrusions (~9 log ɳ Pa.s at 840 °C; DeSilva et al. 1994). Eruption of this large volume of highly crystalline, dacite may have been mechanically forced by the intrusion of hot andesitic magma in the magma chamber (DeSilva et al. 1994).  2.8.1.b.  High eruption temperature Rhyolitic lavas of the Gawler Range Volcanic system, south Australia, are preserved over 12,000 km2, and result from eruption of large-scale lobate flows >100 km in length from a central cluster of feeder vents (Pankurst et al. 2001). Viscosities are low (3.4-5 log ɳ Pa.s), closer to those usually exhibited by basalt than typical silicic melts, as a result of elevated temperatures (950 to 1100 °C), high halogen concentration and low-moderate water content (Pankurst et al. 2001). The dimension and attributes of this volcanic province support interpretation of a well-mixed magma system/chamber of batholitic proportions. 2.8.1.c.  Estimates of eruption temperature and viscosity for OLG lavas OLG eruption temperatures for silicic lavas were calculated as between 765-840°C using zircon saturation temperatures as a proxy (see Chapter 4; Wastson and Harrison 1983; Barrie 1995; Hanchar and Watson 2003). Apparent viscosities were predicted using the model of 59  Giordano et al. (2008), and returned a range of 9 to 12 log η (Pa.s) for dry silicic lavas, at temperatures ranging between 750 and 850 °C; with the simulated addition of 0.5 weight % H2O, viscosities of the silicic lavas drop between 7 and 10 log η (Pa.s) (see details in Chapter 4). Higher volatile contents are uncertain; they are inconsistent with lack of extensive vesiculation within the dacite-rhyolite sections, although degassing could have occurred exclusively at the vent location. Compared with well-documented rhyolitic centers, both eruption temperatures and volatile content of OLG lavas were modest according to the model calculations, so they correspond to the first end-member of lavas yielding normal eruption temperature and forming thick flows.  2.8.2. Thickness variations and structural controls OLG rocks display significant lateral thickness variations, with individual sections reaching > 3000 m. Thickness variations in silicic volcanic terranes result from: 1) filling of paleotopographic depressions; 2) syn-volcanic extension producing grabens; 3) depositional variations such as proximal accumulations in calderas; or 4) amount of syn- and post-volcanic erosion and variations in preservation.  A constructed section (Figure 2.11) shows that the thickest OLG sequences are preserved in grabens, whereas Mesozoic horsts are established from a series of field exposures and drilling (Riddell 2006, 2011).  Syn-volcanism extensional deformation documented by Struik (1993) included Early Eocene northwest-trending dextral strike-slip faults that formed pull-apart basins bounded by northeast-trending extensional faults, and was followed by Late Eocene to Early Oligocene dextral strike-slip faults. Deposition of Eocene volcanic and volcaniclastic rocks in north-trending half grabens has also been reported by numerous workers in southern BC (Church 1973, 1985; Ewing 1981a,b; Thorkelson 1989; Souther 1991) and in the Buck Creek basin (Figure 2.1; Dostal et al. 2001). Modelled thicknesses and structural interpretations conducted in Chapter 5 suggest a similar scenario for the OLG. 2.8.3. Fissure-related silicic volcanism Unlike other large silicic centers such as the Challis volcanic field of central Idaho (Moye et al. 1988), the San Juan volcanic field of Colorado (Steven and Lipman 1976), the Chon Aike 60  province of Patagonia (Pankhurst et al. 1998) or the Sierra Madre Occidental of western Mexico (e.g., Aguirre-Diaz and McDowell 1991; Ferrari et al. 1999; Ferrari et al. 2002), the OLG lacks caldera complexes and voluminous rhyolite ignimbrite. Only few occurrences of welded porphyritic dacitic ignimbrites are reported in the southern Chilcotin Plateau (Mihalynuk and Harker 2007; Mihalynuk et al. 2009a). Instead, the northern part of the Chilcotin Plateau is covered by thick, extensive packages of flow-banded dacitic lavas such as the Mount Sheringham dacite. A number of reported cases of extensive silicic lavas globally involve fissure-related eruptions, as opposed to caldera-related eruptions. For instance, extension fissures served as conduits for the eruption of pyroclastic ignimbrite dikes, lava dikes and domes in the Sierra Madre Occidental (Aguirre-Díaz and Labarthe-Hernández 2003). The Eucarro Rhyolite in the Mesoproterozoic Gawler Range, South Australia forms a voluminous (>675 km3), thick (300 m) lava field, interpreted to have been emplaced as a single unit, as a result of effusive eruption in several source vents located along a fissure system (Allen and McPhie 2002). Also, voluminous (600 km3), high-silica rhyolite lavas infill the preexisting Yellowstone caldera, were effusively erupted along fissure-style vents aligned with regional faults (Girard and Stix 2010). Proposed eruption mechanisms invoke repeated, sustained heating pulses beneath the preexisting caldera, causing gradual melting and formation of a crystal-poor silicic magma that eventually erupted (Girard and Stix 2010). Also, the Badlands rhyolite lava of southwest Idaho, USA, was erupted from a fissure-style vent (Manley 1996). Extensional and transtensional Miocene continental arc basins of the southwestern USA correspond to elevated areas of intra-arc extensive volcanism and extensional deformation, and contain trachydacitic lavas erupted from fault fissures up to 15 km from their source conduits (Busby 2012). Comparisons with volcanic analogues worldwide suggest that OLG silicic lavas probably erupted from multiple subaerial fissure-related vents, represented by the network of regional NE-trending extensional faults and NW- to N-trending strike-slip faults (Struik 1993). However, field evidence, such as dikes that would support a fissure origin for OLG rocks, is lacking and limited by poor exposure across the region. A genetic relationship is suggested between the degree of extension and eruption volumes and styles. Extrapolating an average thickness of ~2 km for OLG lava across south-central BC yields an eruptive volume of ~1.3 x105 km3, attaining the status of a Silicic Large Igneous Province (SLIP); SLIPS are typically formed on continental 61  margins, and cover areas > 100,000 km2, with dacite and rhyolite ignimbrites > 1 km thick (Bryan et al. 2002; Bryan 2007; Bryan and Ernst 2008).  2.9. Conclusion  The OLG in the Chilcotin and Nechako plateaus of central BC is an extensive, voluminous, short-lived high-K calc-alkaline dominantly silicic volcanic province that formed between ~55-47 Ma (see Chapter 3), and is coeval and coextensive (locally interbedded) with more intermediate to mafic volcanic packages previously ascribed to the Endako Group. This work constitutes a first attempt at subdividing and correlating OLG strata regionally, but remains challenged by extremely limited field exposure. Variably thick (few hundred to several thousand meters) sequences of rhyolite and dacite lava are locally capped by andesitic lavas. Pyroclastic and reworked volcaniclastic deposits are mapped throughout the sequence, but they are volumetrically minor. This new view of the stratigraphy does not support the existence of large caldera complexes formed by explosive ignimbritic volcanism as observed in most silicic centers, particularly in silicic volcanic fields of the northwestern and southwestern USA and Mexico; the former are contiguous with the OLG. It is suggested that the eruption and unusually thick and extensive OLG dacitic and rhyolitic lavas were controlled by: 1) temperature and volatile content affecting magma viscosity –although evidence can be seen of neither having played a leading role, and the cause is still in question; 2) eruptions along fissure-style vents rather than calderas; and 3) genetic relationship between extensional deformation coeval with Eocene volcanic activity, and preservation of thousands of meters of volcanic rocks in extensional basins.  Recognition of a regional Eocene stratigraphy has major implications for the mineral exploration. In particular, low-sulfidation epithermal Au-Ag deposits occur at similar stratigraphic levels in the Clisbako rhyolite and in the OLG silicic sequence in the Nechako Plateau (Figure 2.11). Therefore, the mapped and inferred extents of these units provide many more opportunities for mineral exploration focused on Au-Ag epithermal mineralization.   62  Chapter 3   Chronostratigraphy of Eocene volcanism, central British Columbia  3.1. Introduction  Evidence of widespread Eocene magmatism in the North American Cordillera is preserved in a 3000 km long, 150-1000 km-wide belt, extending from eastern Alaska to Idaho. In south-central British Columbia (BC), Eocene magmatism was predominantly calc-alkaline and led to the deposition of mainly silicic and intermediate volcanic and volcaniclastic rocks over an area >65,000 km2 (Figure 3.1). Despite the imposing dimensions of this volcanic belt, correlation and integration of Eocene volcanism with Cordilleran tectonic events, magma genesis processes and metallogenesis have been hampered by inexact or nonexistent regional chronostratigraphy. Eocene regional stratigraphy is complex and lacks obvious widespread marker units, as a result of heterogeneous primary volcanic processes and modification by syn- to post-Eocene deformation and erosion. Coeval and presumably correlative Eocene volcanic groups include the Ootsa Lake (Duffell 1959) and Endako (Armstrong 1949) groups described in the Chilcotin and Nechako plateaus of central BC (Figure 3.1), and the Kamloops (Ewing 1981a,b), Penticton (Church 1973, 1985) and Princeton (Thorkelson 1989; Souther 1991) groups of southern BC. Variable local and regional nomenclatures impede the establishment of regional chronostratigraphic correlations, and suggest lithological and stratigraphic variations resulting from distinct, unrelated volcanic episodes. Existing isotopic age determinations from the Ootsa Lake Group (OLG) and Endako Group have significant overlap with a range of 56-47 Ma for the OLG (Grainger and Anderson 1999; Grainger et al. 2001) and 51-37 Ma for the Endako Group (Haskin et al. 1998; Grainger et al. 2001). These age ranges are biased by many reproducible isotopic age determinations from silicic OLG lithologies and few age determinations from the mafic Endako Group, from which crys