UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Glaciovolcanism at the Mount Cayley volcanic field, Garibaldi volcanic belt, Southwestern British Columbia Kelman, Melanie Catherine 2005

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

Item Metadata

Download

Media
831-ubc_2005-10516X.pdf [ 36.97MB ]
Metadata
JSON: 831-1.0052392.json
JSON-LD: 831-1.0052392-ld.json
RDF/XML (Pretty): 831-1.0052392-rdf.xml
RDF/JSON: 831-1.0052392-rdf.json
Turtle: 831-1.0052392-turtle.txt
N-Triples: 831-1.0052392-rdf-ntriples.txt
Original Record: 831-1.0052392-source.json
Full Text
831-1.0052392-fulltext.txt
Citation
831-1.0052392.ris

Full Text

GLACIOVOLCANISM AT THE MOUNT CAYLEY VOLCANIC FIELD, GARIBALDI VOLCANIC BELT, SOUTHWESTERN BRITISH COLUMBIA by MELANIE CATHERINE KELMAN B.Sc, The University of Saskatchewan, 1994 M.Sc, Oregon State University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Geological Sciences) THE UNIVERSITY OF BRITISH COLUMBIA July 2005 ©Melanie Catherine Kelman, 2005 Abstract This thesis investigates glaciovolcanism in the Mount Cayley volcanic field (MCVF) of southwestern British Columbia's Garibaldi Volcanic Belt (GVB). The MCVF is dominated by intermediate magma compositions, has been intermittently glaciated, and has extreme topography. No study to date has focused on intermediate composition glaciovolcanic deposits in a similar setting. The core of this thesis is a 1:20,000 volcanological map, which was used in conjunction with field observations, geochemical data, and petrographic examinations to investigate the volcanological history of the MCVF, to investigate the eruptive processes that produce specific glaciovolcanic landforms, and to make predictions about paleo-ice thicknesses and distributions. The MCVF (and the entire GVB) has three dominant intermediate composition glaciovolcanic landform types: subglacial domes, flow-dominated tuyas, and impoundment features. Subglacial domes are irregularly-shaped piles of lava flows representing subglacial eruptions that commenced beneath 100-650 m of ice and did not breach the surface; in many cases, the subglacial domes grew to within 150 m of the ice surface. Flow-dominated tuyas are steep-sided, flat-topped stacks of lava flows representing subglacial eruptions that ultimately breached the ice surface. Subglacial domes and flow-dominated tuyas represent the same eruptive process, the primary difference being whether or not the ice was breached. Impoundment features are subaerial lava flows with steep flanks or termini representing channeling or ponding of lava by ice. Al l three landforms have intense fine-scale jointing that indicates cooling surfaces inconsistent with apparent paleotopography. Additionally, all lack features recording eruption into water (pillows and hyaloclastite), indicating efficient and continuous meltwater drainage during eruptions. This is different from glaciovolcanic deposits in most other settings. It also makes these deposits distinct from basaltic glaciovolcanic ii deposits in the GVB. A combination of ice geometry, topography, and magma composition is responsible; thin ice promotes the formation of drainage pathways around subglacial vents, while thin ice coupled with steep bedrock topography creates hydraulic gradients away from vents. Magma composition probably also plays a role, due to its effect on quantities of ice melted and the pressure in subglacial vent cavities and drainage conduits. iii Table of contents Abstract ii Table of contents iv List of figures xviii List of tables xv Acknowledgements vii Parti'. 1. Introduction 1 2. Garibaldi Volcanic Belt 2.1 INTRODUCTION 8 2.2 GLACIATION IN THE GARIBALDI VOLCANIC BELT 14 2.3 INTRODUCTION TO THE MOUNT CAYLEY VOLCANIC FIELD 21 3. Mount Cayley Volcanic Field 3.1 INTRODUCTION 25 3.2 GEOLOGY AND STRATIGRAPHY OF INDIVIDUAL CENTRES 25 3.2.1 Centres erupted prior to the Late Wisconsinan (Fraser) Glaciation (>30,000 BP) 25 3.2.2 Centres erupted during the Late Wisconsinan (Fraser) Glaciation (-30,000-10,000 BP) 45 3.3 GEOCHEMISTRY 117 iv 3.3.1 Introduction 117 3.3.2 Major elements 117 3.3.3 Trace and rare-earth elements 122 3.4 AGE 128 Part II: 4. Glaciovolcanic Features and Processes in the Mount Cayley Volcanic Field 4.1 INTRODUCTION 131 4.2 PRODUCTS OF WELL-DRAINED INTERMEDIATE COMPOSITION GLACIOVOLC ANISM 132 4.2.1 Introduction .' 132 4.2.2 Subglacial domes 135 4.2.3 Flow-dominated tuyas. 140 4.2.4 Impoundment features 143 4.3 DRAINAGE AT MCVF SUBGLACIAL VOLCANIC VENTS 146 4.3.1 Introduction 146 4.3.2 Drainage pathways 146 4.3.3 Hydraulic potential 15 4.4 ABILITY OF ERUPTIONS TO BREACH THE ICE SURFACE 15 4.4.1 Introduction 1-4.4.2 Quantities of ice melted 16 4.5 ICE THICKNESS OVER SUBGLACIAL DOMES 1' 4.5.1 Introduction 17 v 4.5.2 Ice thickness based on assumptions about ice cavity size 170 4.5.3 Ice thickness based on field observations and ice quantities melted 171 4.6 MAGMA-RELATED INFLUENCES 173 4.6.1 Introduction 173 4.6.2 Total erupted volume 173 4.6.3 Effusion rate and style, viscosity, and volatile content 174 4.7 MODELS FOR INTERMEDIATE COMPOSITION GLACIOVOLCANISM 178 4.7.1 Introduction 178 4.7.2 Subglacial to emergent eruptions 178 4.7.3 Impoundment eruptions 185 5. Conclusions and Future W o r k 5.1 CONCLUSIONS 191 5.2 FUTURE WORK 196 References 198 Appendix 1: Compi la t ion of volcanic events shown in Figure 4 232 Appendix 2: Appendix 2a. Sample location list. 235 Appendix 2b. Summary of petrography. 238 Appendix 3: M a j o r and trace element chemistry 243 Appendix 4: Ana ly t ica l methods 248 Appendix 5: Calculations pertinent to ice thickness over subglacial domes 250 Appendix 5a. Ice cavity wall meltback rates. 250 vi Appendix 5b. Ice cavity wall deformation rates 253 Appendix 5c. Equilibrium cavity sizes 255 Plate 1: Geological M a p of the Mount Cayley volcanic field, southwestern British Columbia, Canada vii List of figures Figure 1.1 Tectonic setting of southwestern British Columbia's Garibaldi Volcanic Belt ( G V B ) 2 Figure 1.2 Quaternary volcanic deposits in the G V B 3 Figure 2.1 Chemical composition of lavas from the Mount Cayley volcanic field ( M C V F ) and the rest of the G V B . (a) Total alkalies versus silica, after LeBas et al. (1986). (b) Tholeiitic-calc-alkaline classification of Irvine and Baragar (1971) 11 Figure 2.2 Global climate change over the last 2.6 M a , and the record of volcanism in the G V B . (a) The oxygen isotope record (8 1 8 0) , a proxy for global ice volume, (b) The last 100,000 years' record of mean eustatic sea level changes, a proxy for global ice volume 15 Figure 2.3 Style of advance and retreat of continental glaciers during the Fraser Glaciation 18 Figure 2.4 Changing ice configurations during the Fraser Glaciation (Clague and James 2002). (a) Expansion of ice. (b) Retreat of ice 19 Figure 2.5 The Quaternary Mount Cayley volcanic field ( M C V F ) 22 Figure 3.1 Mount Cayley Quaternary volcanic deposits (Souther 1980) 31 Figure 3.2 Photographs of the Mount Cayley volcanic complex, (a) V i e w from a ridge west of Mount Fee. (b) V i e w from a point on the ridge between the Turbid and Shovelnose Creek valleys 33 Figure 3.3 Mount Fee Quaternary volcanic units 36 Figure 3.4 Photographs of Mount Fee. (a) The southern side of the east-west limb, (b) Subvertical planar joints and steep irregular erosional forms on the south side of Q v p d l . (c) Detail of subvertical planar joints of Q v p d l . (d) The main spine of Mount Fee (Qvpd2) 39 Figure 3.5 The development of Mount Fee 44 Figure 3.6 The Slag H i l l main mass (Qvpaul) and the Slag H i l l tuya (LQvpa7) 47 Figure 3.7 Two cross-sections of Slag H i l l perpendicular to the long axis of the main mass, viewed looking south, and photographs, (a) The high elevation ice-flanked ridge of Qvpaul 3. (b) Qvpaul 1, at the base of the gully between the east and west lobes, (c) The margin of the west lobe, Qvpauh 48 Figure 3.8 Pali Dome East and Pali Dome West 56 Figure 3.9 Cross-section through Pali Dome East, (a) Finely jointed upper surface at the northeast end of the glassy cliffs, (b) Glassy cliffs formed by impoundment of L Q v p a u U against valley-filling ice. (c) Coarsely-jointed lower portion of L Q v p a u U 59 Figure 3.10 Three-dimensional views of Pali Dome East, (a) A view from east of the glacier, (b) A view from the south 62 Figure 3.11 Cauldron Dome, and the inferred subglacial distribution of upper Cauldron Dome 65 Figure 3.12 Cross-section through Cauldron Dome, along the long axis of the sequence, (a) The lower sequence, LQvpau3. (b) The upper sequence, LQvpau2 66 Figure 3.13 The eight lava masses comprising Ember Ridge 72 Figure 3.14 Cross-section through Ember Ridge Southwest, LQvpa3, without vertical ix exaggeration 74 Figure 3.15 Photographs of Ember Ridge, (a) Ember Ridge Northwest, (b) Tiny columnar joints (<5 cm diameter) at the summit of Ember Ridge North, (c) Ember Ridge North, (d) Ember Ridge Southeast, (e) Pseudopillow joints at Ember Ridge West, (f) Intercalated lava flows and breccia at Ember Ridge Southeast 75 Figure 3.16 Cross-sections of five of the mapped Ember Ridge masses 78 Figure 3.17 Little Ring Mountain 86 Figure 3.18 Cross-section through Little Ring Mountain, (a) The southeast side of Little Ring Mountain, (b) Columnar joints on the upper east side of Little Ring Mountain 87 Figure 3.19 Pali Dome East and Pali Dome West 90 Figure 3.20 Three-dimensional views of Pali Dome West, (a) A view from east of the glacier, (b) A view from the south 92 Figure 3.21 Cross-section through Pali Dome West, (a) Ice-impounded lava flow of LQvpau62. (b) Pseudopillow jointing at the north end of Pali Dome West, within LQvpau62. (b) Subaerial flow LQvpau63 at the north end of Pali Dome West 93 Figure 3.22 Ring Mountain (LQvpau7) and the small flow to the northwest (LQvpa8) 98 Figure 3.23 Cross-section through Ring Mountain (LQvpau7) and Ring Mountain Northwest (LQvpa8). (a) Ring Mountain, viewed from below and to the southwest, (b) Ring Mountain Northwest, (c) Ring Mountain Northwest 99 Figure 3.24 Tricouni Southeast 104 x Figure 3.25 Two cross-sections through Tricouni Southeast (LQvpadu). (a) Lava knob of LQvpadiL). (b) Columnar joints in LQvpadu2. (c) Fanning columnar joints near Freeman Lake, in LQvpadu6 106 Figure 3.26 Tricouni Southwest 113 Figure 3.27 Cross-sections through Tricouni Southwest (LQvpba). (a) Columnar joints along the bottom of the gully, (b) The entire wall o f the gully, viewed from the gully floor 114 Figure 3.28 Three-dimensional view of Tricouni Southwest (LQvpba) 116 Figure 3.29 Total alkalies versus silica rock classification (after LeBas et al. 1986) for M C V F lava samples 118 Figure 3.30 A F M diagram for M C V F samples 119 Figure 3.31 Major element abundances in M C V F centres, as oxides, plotted against Differentiation Index (DI) (Thornton and Tuttle 1960). (a) S i 0 2 versus DI. (b) M g O versus DI. (c) CaO versus DI. (d) T i 0 2 versus DI. (e) A 1 2 0 3 versus DI. (f) P 2 0 5 versus DI 121 Figure 3.32 Total alkalies versus silica rock classification (after LeBas et al. 1986) for M C V F lava samples, with the compositions of nearby plutons (potential sources for xenoliths) for comparison 123 Figure 3.33 Locations of the intrusions used to determine contamination vectors for Figure 3.32 124 Figure 3.34 Selected trace element abundances in M C V F samples, normalized to primitive mantle (after Sun and McDonough 1989). (a) Ember Ridge, (b) Pali Dome East and Pali Dome West, (c) Cauldron Dome, (d) Ring Mountain, Ring Mountain Northwest, and Little Ring Mountain, (e) Slag H i l l and the Slag H i l l tuya. (f) Tricouni Southwest and Tricouni Southeast 125 xi Figure 3.35 Chondrite-normalized rare-earth element abundances for M C V F samples (after Sun and McDonough 1989). (a) Slag H i l l , the Slag H i l l tuya, and Little Ring Mountain, (b) Cauldron Dome, Pali Dome, and Mount Fee. (c) Slag H i l l and Cauldron Dome, representing the northern and central R E E trends, respectively, (d) Ring Mountain and Ring Mountain Northwest, (e) Ember Ridge, (f) Tricouni Southwest and Southeast 127 Figure 4.1 A typical subglacial dome, Ember Ridge Northwest 136 Figure 4.2 Quaternary volcanic deposits in the Garibaldi Volcanic Belt ( G V B ) . Glaciovolcanic landforms discussed in Chapter 4, excluding specific centres within the M C V F , are labeled 138 Figure 4.3 Typical flow-dominated tuyas. (a) Little Ring Mountain, south side, (b) Ring Mountain, north side 141 Figure 4.4 A typical impoundment feature, at Pali Dome east, (a) The upper surface of the flow, (b) The entire margin of the flow, (c) The lowest exposed part of the flow 144 Figure 4.5 Factors determining whether water accumulates or drains at M C V F subglacial volcanic vents 147 Figure 4.6 Section through an idealized glacier in the M C V F , showing the thickness of the firn, the crevassed layer, and the unfractured ice 149 Figure 4.7 Theoretical cavity pressure versus temperature change of magma 157 Figure 4.8 Maximum possible ice thicknesses to have overlain individual units within the M C V F , based on a comparison between their modern elevations and the maximum elevation reached by the Cordilleran ice sheet (-2300 m) 161 x i i Figure 4.9 Differences in heat budget between dacite and basalt 163 Figure 4.10 Maximum possible overlying ice thickness versus predicted thickness of ice melted 168 Figure 4.11 Initial stages of subglacial eruption of intermediate composition lava in steep topography 179 Figure 4.12 Subglacial eruption of intermediate composition lava in steep topography, continued from Figure 4.11. This figure depicts an eruption that does not breach the ice surface (thus forming a subglacial dome) 182 Figure 4.13 Subglacial eruption of intermediate composition lava in steep topography, continued from Figure 4.11. This figure depicts an eruption that ultimately breaches the ice surface, forming a flow-dominated tuya 184 Figure 4.14 Ice coverage over the M C V F at four times in the past. The distribution of modern alpine glaciers is shown 186 Figure 4.15 Impoundment of lava by large quantities of ice in steep terrain, (a) Lava flows downslope, onto and against the glacier margin and any glacial t i l l that marks the edge of the ice. (b) The still-molten portions of the lava flow continue to flow downslope, pushing the quenched margins ahead of them atop a base of breccia, (c) The lava flow continues to expand downslope. (d) After the ice is removed, an extremely unstable cl i ff is left behind 188 Figure 5.1 Controls on the morphology and internal stratigraphy of glaciovolcanic deposits in the G V B . (a) Factors related directly to eruptions, (b) Factors related to bedrock-vent-ice geometry 193 Figure 5.2 Controls on subglacial eruption. The presence or absence of water is the most important factor influencing eruptions 194 x i i i Figure A5.1 Wal l meltback rate versus wall deformation rate for hemispherical subglacial cavities, plotted on log-log axes 256 Figure A5.2 Equilibrium cavity radius (Equation 17) versus lava mass radius, for two ice thicknesses (200 m and 900 m), assuming cavities are at atmospheric pressure (0.1 MPa) 258 xiv Lis t of tables Table 2.1 Compositions and ages of Quaternary volcanic centres and fields of the G V B 9 Table 2.2 Summary and description of glaciovolcanic features in the G V B 12 Table 3.1 Summary of rock types, volumes, and ages of centres of the M C V F 26 Table 3.2 Summary of glaciovolcanic features at M C V F centres 28 Table 4.1 Summary of glaciovolcanic landforms and their attributes 133 Table 4.2 Symbols and subscripts used in calculations 154 Table 4.3 Constants used in calculations 155 Table 4.4 Effusion rates at volcanoes erupting lavas of intermediate composition 175 Table A5.1 Symbols and subscripts used in Appendix 5 calculations 251 Table A5.2 Constants used in Appendix 5 calculations 252 xv Acknowledgements I'd like to thank my supervisory committee, Kelly Russell, Cathie Hickson, and Jim Mortensen, for their inspiration, assistance, support, and patience through this long project. Financial support was provided by NSERC (Grant R89820 to Kelly Russell and R80659 to Cathie Hickson), the University of British Columbia (UBC), and the Geological Survey of Canada. Technical support was provided by both UBC and the Geological Survey of Canada (GSC). I gladly acknowledge the people who assisted with my field work, especially Dan Lui, my primary field assistant, as well as the other graduate students who gave freely of their time and energy: Margaret Harder, Nathalie Lefebvre, and Pat Hayman. Louise Fox kindly provided logistical support in Squamish and Vancouver. Interior supplied maps and helpful advice on the Mount Cayley area. Cathie Hickson, Kelly Russell, and a small group of South American geologists collected a series of rock samples in 1998 that were beneficial to my project. I am especially grateful to the people of the GSC Vancouver's GIS lab (Rob Cocking, Andrew Makepeace, Kaz Shimamura, and others) for their help, particularly Rob Cocking for his excellent work in the production of the Mount Cayley volcanic field (MCVF) map. Nathan Green provided assistance with both geochemistry and field work, John Clague answered many questions about Quaternary glaciations in southwestern British Columbia, Jack Souther discussed his observations and interpretations of the MCVF, Ken Hickey generated three-dimensional images of the MCVF, Hugh Tuffen answered questions about his research, and Tom Ullrich and Mike Villeneuve made repeated valiant attempts to generate whole rock Ar-Ar ages. I would also like to acknowledge the assistance and companionship of the other denizens of the Igneous Petrology Laboratory: Martin Stewart, Steve Quane, Alison Rust, Daniele Giordano, and Krista Michol, as well as former denizen Ben Edwards, the UBC Earth xvi and Ocean Sciences Department administrative staff (especially Alex Allen, who made sure I didn't miss UBC deadlines), and the other officemates, graduate students, friends, and Rock Dogs who ignored my quirks. To all of you, many thanks. Finally, I'd like to thank my family for both financial and moral support during my protracted education. I really appreciate it. xvii Parti. CHAPTER 1 Introduction Southwestern British Columbia's Quaternary Garibaldi Volcanic Belt (GVB; Figure 1.1) has been subjected to repeated continental and alpine glaciations, with intermittent volcanism superimposed upon the waxing and waning of ice. Thus, many of the GVB's volcanic centres show evidence for eruption in the presence of ice (glacio volcanism). In this thesis, I used field data from the Mount Cayley volcanic field (MCVF), located in the central GVB (Figure 1.2; Plate 1), to explore glaciovolcanic processes, with the intent of answering three primary questions: (1) What is distinctive about the glaciovolcanic deposits in the GVB relative to those in other settings? (2) What are the physical controls on glaciovolcanic eruptions in the GVB, and the eruptive processes that generate specific glaciovolcanic landforms? (3) What information can these glaciovolcanic landforms reveal about past ice thicknesses and distributions? Question (1) was addressed by mapping the MCVF at 1:20,000 during the 2 0 0 1 -2002 field seasons, and by conducting petrographic and chemical analyses of samples. This is the first detailed map of the MCVF (Plate 1). It is a beneficial contribution to 1 Figure 1.1 Tectonic setting of southwestern British Columbia's Garibaldi Volcanic Belt (GVB). Volcanoes or volcanic fields of the Cascade arc are indicated with black circles. Modified after James et al. (2000). Dashed lines show the depth to the top of the subducting plate (Fluck et al. 1997). The location of Figure 1.2 is shown with the rectangular box. 2 • Legend Quaternary centres with evidence for glaciovolcanism Quaternary centres without documented evidence for glaciovolcanism 0 k m 20 Figure 1.2 Quaternary volcanic deposits in the Garibaldi Volcanic Belt (GVB). The Franklin Glacier and Silverthrone complexes (north of this map) are not included. Location of this map with respect to the rest of British Columbia is indicated on Figure 1.1.) The distribution of volcanic rocks originating at Mount Cayley is based on Souther (1980). Limits of the two Mount Brew masses are from Green et al. (1988). The distribution of volcanic rocks from outside the Mount Cayley volcanic field (MCVF) is based on Mathews (1958), Green (1977), Lawrence (1979), Lawrence et al. (1984), Roddick and Souther (1987), Green et al. (1988), Green (1994), Hickson et al. (1999), Bye et al (2000), Hickson (2000), and Stewart et al. (2002). Salal Glacier volcanic complex Mount Cayley volcanic field Garibaldi Lake/ Mount Garibaldi volcanic field \ Monmouth Creek volcanic complex Watts Point volcanic complex H°>fe r Sound 3 Canadian geoscience because many Canadians are unaware of the presence of modern volcanoes within their province, and even the volcanological literature has neglected the northern Cascade Volcanic Arc in favour of the southern segments. The fact that Mount Cayley is still potentially active (Souther 1980; Jones and Dumas 1993; Hammer and Clowes 1996; Stasiuk et al. 2003) means that it could pose a hazard to nearby towns and logging and recreational areas. The map will aid in the recognition and mitigation of these hazards by delineating areas affected by previous eruptions and indicating the styles of eruption, information that could be used to forecast future volcanic activity. The Mount Cayley volcanic field (MCVF) contains numerous glaciovolcanic and subaerial volcanic deposits ranging in composition from basaltic andesite to rhyodacite. As such, it is an excellent locality at which to investigate GVB glaciovolcanic processes. My work shows that the GVB contains an abundance of subglacial domes, flow-dominated tuyas, and impoundment features, and has a paucity of primary fragmental volcanic material and pillow lavas in comparison to glaciovolcanic deposits in other settings. Question (2) was addressed by using the mapped distribution of volcanic deposits, with emphasis on features such as cooling joints, to generate two conceptual models, one for eruptive processes at subglacial domes and flow-dominated tuyas, and one for impoundment features. This was necessary because generally-accepted models for glaciovolcanic eruption do not adequately describe or explain most GVB glaciovolcanic deposits; the combination of steep topography, fluctuating ice, and dominantly intermediate compositions is not unique to the GVB, but it is markedly different from what is present in most other glaciovolcanic regions. 4 Question (3) was addressed by using field observations and the eruptive model for subglacial domes and flow-dominated tuyas to obtain information about ice thicknesses during glaciovolcanic eruptions that may be applied to similar deposits elsewhere. Glaciovolcanism results in landforms with morphologies and lithologic characteristics that depend upon a complex interplay of physical conditions during eruption, including topography, geometry and hydraulic characteristics of ice, quantity and location of meltwater, and eruption-related factors (e.g. magma composition) (e.g. Jones 1966, 1969; Bjornsson 1975; Allen 1980; Smellie et al. 1993; Smellie and Skilling 1994; Hickson et al. 1995; Werner et al. 1996; Gudmundsson et al. 1997; Hickson 2000; Lescinsky and Fink 2000; Smellie 2001, 2002; Edwards et al. 2002; Skilling 2002; Tuffen et al. 2002a, b). Glaciovolcanic processes are of interest in hazard assessments because subglacial eruptions have the potential to release large volumes of water catastrophically, as jokulhlaups, due to rapid ice melting, and because many glaciovolcanic features are subject to major landsliding. Glaciovolcanic processes are of interest in paleoclimate studies because the deposits produced can serve as paleoclimatic indicators, and because the distibution of glaciovolcanic products can potentially be used to constrain temporal and spatial linkages between volcanism and glaciation. No detailed study prior to this thesis project has focused on intermediate composition glaciovolcanism at a convergent margin, such as the GVB. This is significant because preliminary studies (e.g. Mathews 1951, 1952, 1958; Green 1977, 1981; Hickson 2000; Lawrence et al. 1984; Lescinsky and Sisson 1998; Lescinsky and Fink 2000; Bye et al. 2000) suggest that such glaciovolcanic deposits are distinct from those in other geological settings. 5 This thesis is organized into two parts. Part One (Chapters 1-3) provides an introduction to the GVB, presents the 1:20,000 map of the MCVF, and provides a description of the MCVF volcanic deposits (their lithological and petrographical attributes, as well as their eruptive histories, chemical characteristics, and probable ages). Part Two (Chapters 4-5) delineates the characteristic landforms of the GVB and their attributes, discusses how ice, topography, and magma influence GVB glaciovolcanic eruptions, determines a probable ice thickness range for non-ice-breaching eruptions, and presents conceptual models for subglacial to emergent and ice impoundment eruptions of intermediate composition lava in settings similar to the GVB. Several sections of this thesis have been published or presented at conferences. The following is a list of publications related to this work: Abstracts Kelman, M.C., Hickson, C.J., and Russell, J.K. 2003. Glaciovolcanism in British Columbia's Mount Cayley volcanic field: the interplay of magma, topography, and ice in a convergent margin setting. Geological Society of America 2003 Annual Meeting, Program with Abstracts, Seattle. Kelman, M.C., Hickson, CJ . , and Russell, J.K. 2003. The Mount Cayley volcanic field: Mapping glaciovolcanic features at a convergent margin. Geological Association of Canada - Mineralogical Association of Canada-Society of Economic Geologists 2003 Annual Meeting Program with Abstracts, Vancouver. Kelman, M.C., Russell, J.J., and Hickson, C.J. 2003. Glaciovolcanism in the Garibaldi Volcanic Belt: Topographic and compositional controls on morphology. Geological 6 Association of Canada - Mineralogical Association of Canada 2003 Annual Meeting Program with Abstracts, Vancouver. Papers Kelman, M.C., Hickson, C.J., and Russell, J.K. 2003. Quaternary glaciovolcanism along the Whistler corridor, in Geological Field Trips in Southern British Columbia, Geological Association of Canada, Cordilleran Section,Vancouver, British Columbia: 147-159. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2002. Effusive intermediate glaciovolcanism in the Garibaldi Volcanic Belt, southwestern British Columbia, Canada, in J.L. Smellie and M.G. Chapman (eds.), Volcano-Ice Interactions on Earth and Mars. Geological Society, London, Special Publications, 202: 195-211. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2002. Glaciovolcanism at Ember Ridge, Mount Cayley volcanic field, southwestern British Columbia, in Current Research, Part A; Geological Survey of Canada, Paper 2002-A15: 1-7. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2001. Preliminary petrography and chemistry of the Mount Cayley volcanic field, British Columbia, in Current Research, Part A; Geological Survey of Canada, Paper 2001-A11: 1-9. 7 CHAPTER 2 Garibaldi Volcanic Belt 2.1 INTRODUCTION The Cenozoic Garibaldi Volcanic Belt (GVB; Figure 1.1; Figure 1.2; Table 2.1) is the northernmost segment of the 1100 km long Cascade magmatic arc of the western United States (Green et al. 1988; Guffanti and Weaver 1988; Read 1990; Sherrod and Smith 1990; Hickson 1994). Quaternary volcanic activity in the Cascade arc is a result of the subduction of the Juan de Fuca Plate beneath the North American Plate (Green et al. 1988; Rohr et al. 1996). Located 300-400 km inboard of the subduction zone (Hickson 1994), the GVB stretches from Watts Point (located at the head of Howe Sound) northward to the Salal Glacier volcanic complex (Figure 1.2). The Franklin Glacier and Silverthrone volcanic complexes (Table 2.1) are occasionally classified as part of the GVB, although their magmatic styles are different than those of the rest of the belt (Green et al. 1988), and their eruptive histories and tectonic settings are poorly understood. The GVB is underlain by Mesozoic to Tertiary granitic and metamorphic rocks of the Coast Plutonic Complex (Roddick and Woodsworth 1977; Monger and Journeay 1994). The GVB has the lowest magma production rate and eruptive frequency in the Cascades (Scott 1990; Sherrod and Smith 1990). Rogers (1985) and Walcott (1993) suggested that northern Washington and southern British Columbia are being compressed relative to southern portions of the Cascade arc, and this may explain the lesser volumes of magma produced in the northern Cascades. The GVB is also distinctive in that, unlike the central and southern Cascades, 8 Table 2.1 Composition and ages of Quaternary volcanic centres and fields of the Garibaldi Volcanic Belt. Major centres and fields Composition Age Source* Silverthrone volcanic field basaltic andesite to rhyolite 1.0-0.4 M a 1 Franklin Glacier complex dacite, andesite 3.9-2.2 M a 1 Salal Glacier volcanic field alkali olivine basalt, hawaiite 1.0-0.6 M a 2 Mount Meager - Elaho Valley basalt to dacite 2.2 M a - 2.4 ka 3 ,4 ,5 Mount Cayley volcanic field basaltic andesite to rhyodacite 0.5 - 0.3 M a 3 Mount Garibaldi/Garibaldi Lake volcanic field basalt to dacite 1.3 M a - 10 ka 3, 6,7 Monmouth Creek complex basaltic andesite to dacite n.d.** 8 Watts Point volcanic centre dacite 130-90 ka 3 * 1: Souther & Yorath 1992, 2: Lawrence et al. 1984, 3: Green et al. 1988, 4: Leonard 1995, 5: Clague et al. 1995, 6: Mathews 1958, 7: Brooks & Friele 1992, 8: Green 1994. **n.d. = not determined repeated eruptions at central vents, resulting in large stratovolcanoes, are not the norm. Instead, volcanism tends to occur at centres that are clustered but not coincident, resulting in volcanic fields rather than large central volcanoes (Hickson 1994). The large-scale distribution of Miocene volcanic rocks in the region is controlled by northeast-trending fault systems (Coish and Journeay 1992), but Quaternary structures trend otherwise, and the distribution of Quaternary volcanic centres cuts across the northwest-trending structural grain of the region (Monger and Journeay 1994). Hickson (1994) proposed that major northwest-southeast structures in the vicinity of Mount Meager may produce structural weaknesses that are exploited by rising magma batches, and similar controls may exist for other GVB volcanic deposits. GVB volcanic deposits range from Pliocene to Holocene in age (Green et al. 1988) , and volcanic rocks range in composition from calc-alkaline and alkalic basalt to calc-alkaline rhyolite (Figure 2.1). Eruptions have ranged from small-volume effusions to voluminous explosive events. The most recent large eruption in the GVB, and Canada's youngest explosive eruption, occurred at Mount Meager at -2330-2360 BP and spread ash across southern British Columbia and into Alberta (Clague et al. 1995; Leonard 1995). Landforms include stratovolcanoes, discrete flows, spines, pyroclastic cones, domes, spires, and tuyas. Glaciovolcanic deposits have been recognized at most centres in the GVB, and are summarized in Table 2.2. As a volcanically active convergent plate margin, the GVB possesses high relief, steep slopes, and high uplift rates (up to 4 mm/year in the Coast Mountains; Holdahl et al. 1989) . Most volcanoes are perched at high elevations and deposits are commonly poorly consolidated or heavily fractured. The temperate climate of southwestern British 10 alkalies wt.% MgO wt.% Figure 2.1. Chemical composition of lavas of the MCVF and the rest of the Garibaldi Volcanic Belt (GVB). Data for the GVB are from Green (1977), Lawrence et al. (1984), Ke (1992), Stasiuk et al. (1996), and Stewart et al. (2002). (a) Total alkalies versus silica, after LeBas et al. (1986). (b) Tholeiitic-calc-alkaline classification of Irvine and Baragar (1971). 11 Table 2.2 Summary and description of glaciovolcanic features in the Garibaldi Volcanic Belt. Name Feature* Rock Type Dimensions in plan view (m) Maximum Thickness (m) Age / Glacial Period (absolute ages in Ma) Source** Tuber H i l l T basalt 2000 x 2000 <450 0.598±0.015; 0.731±0.037; 0.760±0.033 1,2 Ochre Mountain SGF? alkali olivine basalt 200x1000 ? ? 2 Logan Ridge I, SGF basalt 300 x 500 ? ? 2 Salal Glacier I hawaiite, alkali olivine basalt >1000x2000 ? 0.97±0.05; 0.59±0.05 2 Little Ring Mountain F D T andesite <1000 140 Fraser Glaciation this thesis Ring Mountain F D T andesite <360 380 Fraser Glaciation this thesis Ring Mountain Northwes' I basaltic andesite 500x1200 50 Fraser Glaciation this thesis Slag H i l l main mass L C andesite, basaltic andesite 1700x 5800 300 mostly Fraser Glaciation this thesis Slag H i l l tuya F D T andesite 400 x 540 90 Fraser Glaciation this thesis Upper Cauldron Dome FDT? andesite 1700x2600 310 mostly Fraser Glaciation this thesis Lower Cauldron Dome I andesite 1400 x 2500 120 Fraser Glaciation this thesis Pali Dome West I andesite 800 x 3000 100 Fraser Glaciation 4, this thesis Pali Dome East I, L C andesite 700 x 3000 120 mostly Fraser Glaciation 4, this thesis Ember Ridge (6 domes, ± 2 Mount Brew masses) S G D andesite range from 180-1100 140 Fraser Glaciation 3, 4, 6 this thesis Tricouni Southwest Unit I basaltic andesite 400-1600x3500 120 Fraser Glaciation this thesis Tricouni Southeast Unit (2 masses) I, SGD? andesite and dacite 670 x 1400 and 3000 x3000 70 and 180 Fraser Glaciation this thesis Whistler dump mass S G M ? ? ? Fraser Glaciation 5 Name Feature* Rock Type Dimensions in plan view (m) Maximum Thickness (m) Age / Glacial Period (absolute ages in Ma) Source** Cheakamus Valley flows SGF basalt 50-100 x (up to) 1600 10-20 Fraser Glaciation 6,7 Black Tusk L C dacite >1130x2350 430 1.3±0.1 to0 .17±0.04 6,7 Table F D T andesite 180x300 250 Fraser Glaciation 8 Mount Price (Barrier and Culliton Creek) I andesite 90-270 x 9000 ?? Fraser Glaciation 9 Mount Garibaldi L C dacite >5200 x 7800 1700 0.26±0.06 to Fraser Glaciation 6,7 Glacier Pikes S G D andesite ? ? Fraser Glaciation 7 Round Mountain S G D ? ? ? ? 7 Columnar Peak S G D dacite ? ? ? 7 Eenostuck Mass S G D basaltic andesite 180x800 90 Fraser Glaciation? 7 Monmouth Creek complex ? dacite, basaltic andesite 600x1000 >150 ? 7 Watts Point S G D dacite 600 x 700 <100 0.09±0.03;0.13±0.03 4, 10 *T = tuya, FDT = flow-dominated tuya, SGD = subglacial dome, SGF = subglacial flow, 1 = impoundment feature, L C = long-lived centre with intermittent evidence for glaciovolcanism, S G M = subglacial mound **1: Roddick & Souther 1987; 2: Lawrence 1979; 3: Kelman et al. 2001; 4: Souther 1980; 5: Hickson (pers. comm.); 6: Green et al. 1988; 7: Mathews 1958; 8: Mathews 1951; 9: Mathews 1952; 10: Bye et al. 2000 Columbia means that frost action and erosion by running water are important processes. Al l of the GVB volcanoes are adjacent to active glaciers, and all were to some extent influenced by past continental scale glaciations. As a result of these factors, GVB volcanic deposits are subject to rapid mass wasting, and many rock avalanches and debris flows have been documented at GVB volcanoes. 2.2 GLACIATION IN THE GARIBALDI VOLCANIC BELT The Quaternary Period (the last 1.8 Ma) has been a time of climatic fluctuation resulting in intermittent glaciations in the northern hemisphere. Isotopic and magnetic studies of deep-sea sediments indicate that there have been at least eight major climatic cycles in the last 800,000 years, and the colder portions of these were probably periods of widespread glaciation in British Columbia (Shackleton and Opdyke 1973, 1976; Clague 1991). It is generally believed that there have been two major glaciations in the Canadian Cordillera during the last 75-100,000 years, the Early Wisconsinan (>59 ka BP; Fulton 1984; Clague 1991) and Late Wisconsinan, locally referred to as the Fraser Glaciation (-30,000-10,000 BP; Fulton 1969; Mathews et al. 1970; Fulton 1971; Clague 1980,1981, 1991,1994; Friele and Clague 2001; Clague and James 2002). Evidence for the timing of glaciations prior to the Fraser event is scant, though there are dated ice-contact lava flows older than 100,000 years from the Garibaldi lake area (Green et al. 1988). Superimposed in time upon these climatic fluctuations are repeated volcanic eruptions (Figure 2.2). Radiocarbon dates and paleoecological studies suggest that cooling and subsequent glacier expansion marking the onset of the Fraser Glaciation commenced around 30,000-25,000 BP (Armstrong et al. 1965; Clague 1980, 1981; Hicock and 14 mean eustatic sea deviation from present day continental ice volume (10b km') Figure 2.2. Global climate change over the last 2.6 Ma, and the record of volcanism in the GVB. Appendix 1 provides the details (dates, errors, and sources) behind this figure. (a) The last 2.6 Ma of the oxygen isotope record (81 80), a proxy for global ice volume, based on the core from ODP Site 677 (Shackleton et al. 1990). Temperature increases with larger negative numbers. Multiple contemporaneous eruptions are indicated with numbers superimposed on the vertical lines. Eruptions during the first 0.1 Ma are not shown because there are so many; they are indicated in (b), which shows this time period in detail. Dated volcanic events are shown with vertical dotted lines. (b) The last 100,000 years' record of mean eustatic sea level changes, a proxy for global ice volume (solid red line, left axis), derived from the Pacific core VI9-30 isotope record and sea level data (Pillans et al. 1998), and deviation from present day continental ice volume (dashed blue line, right axis; Berger 1992), simulated by a climate model by Berger et al. (1988). The horizontal dashed line indicates zero line for sea level (left axis) and deviation from minor ice volume (right axis). Mean eustatic sea levels (solid line) that plot below this zero line represent times at which sea level was lower than it is today. Deviations from present day continental ice volume that plot below the horizontal line represent times at which ice cover was more extensive than it is today. The approximate timing of the Fraser (-30,000-10,000 BP; Fulton 1971; Clague 1980,1981, 1991, 1994; Clague and James 2002) and Salmon Springs (>59,000 BP; Fulton 1984; Clague 1991) Glaciations are shaded. Since no date was available for the start of the Salmon Springs Glaciation, it is shown to be the same length as the later Fraser event, however, its onset, shown as ~79,000 BP, is speculative. Dated volcanic events are shown with vertical green bars; the numbers indicate the number of events falling within the time period indicated by each bar. 16 Armstrong 1981; Clague 1991, 1994). This early period was dominated by alpine glaciation, and initial glacier growth was confined to mountainous regions (Clague 1976, 1977,1981; Ryder et al. 1991; Clague and James 2002). The Cordilleran ice sheet developed as glaciers over lowlands and plateaus thickened and coalesced (Clague 1994) (Figure 2.3), with ice extending south to latitude 47° (Clague 1991). Sea level diminished as glaciation progressed, and isostatic depression of the crust by 250 m or more occurred (Clague 1983). The ice sheet maximum thickness over southwestern British Columbia occurred between 14,000-15,000 BP (Clague et al. 1980; Clague 1981; Ryder et al. 1991; Clague and James 2002), and was above 2300 m elevation for much of southwestern British Columbia (Clague 1989; Ryder et al. 1991) (Figure 2.4). From this it can be inferred that ice more than 2000 m thick would have covered major valleys and ice hundreds of meters thick would have covered most ridges (Ryder et al. 1991); only the highest peaks would have protruded through the ice as nunataks. There are glaciovolcanic features in the Mount Cayley volcanic field (MCVF) and in the Mount Garibaldi region (Figure 1.2) as high as ~2200-2300 m, so it is probable that the local maximum ice thickness in this area was close to the regional value reported by Clague (1989) and Ryder etal. (1991). During ice sheet maxima, surface gradients of the ice sheet probably controlled the flow directions of the ice, which would commonly have been discordant to local topography (Ryder et al. 1991). However, during periods of more geographically restricted glaciation (Figure 2.3), ice flow would have been locally controlled. The direction and speed of ice flow influenced the development of both volcanic and sedimentary subglacial features and controlled which were preserved and which were 17 Figure 2.3. Style of advance and retreat of continental glaciers during the Fraser Glaciation. Modified slightly from Clague (1991). (a) Landscape prior to glaciation. The arrow indicates the regional topographic gradient. (b) Early alpine glaciation. Ice flow directions are controlled locally, by topography. (c) Coalescence of individual glaciers to form a continental ice sheet. Ice flow direction is more likely to be regionally controlled. (d) Early stages of ice retreat. (e) Final waning of glaciation; stagnant, isolated ice sheets remain in some valleys. The illustration does not show the erosional effects of the ice on the original topography. 18 Figure 2.4. Changing ice configurations during the Fraser Glaciation (Clague and James 2002). (a) Expansion of ice, to its maximum at 14,000-15,000 BP (Clague et al. 1980; Clague 1981; Ryder et al. 1991; Clague and James 2002). (b) Retreat of ice. 19 destroyed by erosion. For example, the preservation of fresh-appearing cirques whose formation is assigned to the early Fraser Glaciation is attributed to the short duration of the ice sheet phase and to the presence, during the glacial maximum, of stationary non-eroding ice in some valleys not aligned parallel to flow (Ryder et al. 1991). This is an important consideration in using extent of erosion as an age indicator in glaciovolcanic features; at the height of the Fraser Glaciation, the regional direction of ice flow in the central GVB was towards the southwest (Blaise et al. 1990; Clague 1994), so features located in valleys perpendicular to this trend may have experienced less erosion by the ice sheet. Studies of subglacial features such as roches moutonees, drumlins, eskers, and meltwater channels suggest that large areas of the subglacial bed were unfrozen (Ryder et al. 1991), a fact that also has implications for the development of glaciovolcanic features (e.g. it would influence the escape of eruption-derived meltwater along the glacier bed). Glacier decay began shortly after 14,000 BP (Clague et al. 1980; Clague 1981; Ryder et al. 1991; Porter and Swanson 1998; Clague and James 2002), and occurred by downwasting, stagnation, and complex frontal retreat, with uplands appearing through the ice cover first, while ice still remained in valleys (Figure 2.3). Detailed studies suggest that local glacier growth and decay were complex, with retreats, stillstands, and advances on fairly short timescales (Armstrong 1981; Clague et al. 1997; Clague and James 2002); remnant dead ice masses may have persisted in the Howe Sound region as late as 10,200 BP (Friele and Clague 2001). Deglaciation of southwestern British Columbia was complete sometime between 10,000-11,000 BP (Clague et al. 1980; Clague 1981; Ryder et al. 1991; Brooks and Friele 1992; Clague and James 2002). 20 2.3 INTRODUCTION TO THE MOUNT CAYLEY VOLCANIC FIELD The Mount Cayley volcanic field (MCVF; Figure 1.2, Figure 2.5, Plate 1), is located northwest and adjacent to the Garibaldi Lake volcanic field, and includes the centres at Little Ring Mountain, Ring Mountain, Slag Hill, Cauldron Dome, Pali Dome West, Pali Dome East, Mount Cayley, Mount Fee, Ember Ridge, Tricouni Southwest, and Tricouni Southeast. No glaciovolcanic features have been documented at Mount Cayley, and none were found at Mount Fee during the 2001 mapping. A l l other MCVF centres show evidence for eruption in contact with ice. A glacier currently covers the centre of the field, though comparisons between air photos from the 1960s with current ice configurations indicate that considerable glacial recession has occurred over the last 40 years. Mount Cayley itself is a composite pile of overlapping flows and pyroclastic deposits formed during at least three stages of subaerial activity (Souther 1980). It has a 3 1 * total estimated modern volume of 8 km but is substantially eroded . Individual volcanic centres of the MCVF are discussed in detail in Chapter 3. Basement rocks in the MCVF can be grouped into four assemblages (Roddick and Woodsworth 1977; Monger and Journeay 1994): The oldest (1) consists of a pendant of metasediments that include quartz mica schist, greenstone, amphibolite gneiss, and intensely deformed crystalline limestone. These metasediments are gradational with (2) a hornblende-rich complex of quartz diorite, diorite, and minor amphibolite units that are foliated and contain numerous mafic inclusions and dyke swarms. This hornblende-rich complex has sharp contacts with (3) a uniform, poorly foliated, medium-grained hornblende-biotite granodiorite (similar to that which underlies much of the northern 1 Hickson (1994) estimated the total volume to be as large as 13 km3. 21 123"21'00" 50°16'30" 123°18'00" 5Cri5'00" 50'13'30" 50°12'00" 50'10'30M 50°09'00" 50°0r3O" 50"06'00" 50W30" SOTO'OO" 50*01'30" 50°00'00" 49°58'30" 49"57'00" Figure 2.5. The Quaternary Mount Cayley volcanic field (MCVF). Numbers/colours represent individual centres, most of which include deposits from multiple eruptions. Volcanic geology of Mount Cayley is from Souther (1980). The distribution of Mount Brew deposits (the two southeast Ember Ridge masses) is from Green et al. (1988). The remaining volcanic rocks were mapped as part of this thesis. Contour interval = 100 m. 1 Tricouni Southwest 2 Tricouni Southeast 3 Ring Mountain 4 Pali Dome West 5 Little Ring Mountain 6 Slag Hill tuya 7 Ember Ridge 8 Cauldron Dome 9 Pali Dome East 10 Slag Hill 11 Mount Fee 12 Mount Cayley ice highways 50"06'00" 50°04'30" meters 2000 4000 50°01'30" SffOCOO" 49°58'30" 49"57'00" 123"21'00" 123"18W 123°15'00" 123"I2'00" 123°09'00" • ^ " d e W 22 GVB). The granodiorite is cut by (4) a pluton of pinkish white coarse- to medium-grained quartz monzonite. In the MCVF, a series of northerly-northwesterly trending fractures appears to be the only set of basement structures related to the volcanic rocks (Souther 1980); these are best developed close to Mount Cayley, where they are associated with hydrothermal alteration. However, similarly oriented fractures are also present to the north of Mount Cayley and to the south of Mount Fee. Souther (1980) suggested that the orientation of the intrusive spines on Mount Cayley and Mount Fee reflects the northerly to northwesterly trend of the GVB between Mount Cayley and Mount Meager. Like other volcanic complexes in the GVB, the MCVF has a history of landslides. The oversteepened and hydrothermally altered slopes of Mount Cayley, especially on the southwest side, experience regular rockslides and mixed rock and snow avalanches. This material is deposited and reworked in the narrow valleys to the southwest of the volcano, and may be periodically flushed out by floods and debris flows. Steep and poorly consolidated slopes at Mount Cayley have resulted in several major prehistoric (Brooks and Hickin 1991; Evans and Brooks 1991) and historic (Clague and Souther 1982; Evans 1986,1990; Jordan 1987; Lu 1988; Cruden and Lu 1992; Hungr and Skermer 1992; Evans et al. 2001) debris avalanches. There is also evidence for landslides at many smaller MCVF centres. Seismic imaging from Lithoprobe studies has revealed a strong mid-crustal reflector adjacent to Mount Cayley that shows a rapid transition to country rock (over <200 m) (Jones and Dumas 1993; Hammer and Clowes 1996). This is likely to be either a large, solidified, mafic sill-like intrusion, a magma lens, or a zone of saline fluid 23 produced by dewatering of the subducted slab (Jones and Dumas 1993; Hammer and Clowes 1996). Shallow earthquakes that have been reported since 1985 in the vicinity of Mount Cayley (Stasiuk et al. 2003) and the presence of at least three hot springs in the upper Turbid Creek valley and two hot springs in the Shovelnose Creek valley suggest that magmatic heat is still present. The main hot springs have extensive tufa and sinter deposits, while numerous cold seeps in the vicinity are precipitating bright red ferruginous ochre. Temperature measurements range from 18-40°C (Souther 1980). Geothermal exploration in the vicinity of Mount Cayley is ongoing. 24 C H A P T E R 3 Mount Cayley Volcanic Field 3.1 INTRODUCTION The purpose of this chapter is twofold: to describe the petrology, stratigraphy, and geochemistry of the volcanic rocks of the MCVF based on mapping conducted in 2001-2002, and to discuss the eruptive histories of individual volcanic centres of the GVB, with an emphasis on those centres at which glaciovolcanism occurred. A detailed 1:20,000 scale geological map is presented as Plate 1. Small maps scaled at or near 1:20,000, representing relevant sections of the comprehensive Plate 1 map, accompany text discussions for each centre in this chapter. A common legend accompanies Figure 3.3. Table 3.1 summarizes the rock types, volumes, and ages of each centre in the MCVF; centres listed in the table are ordered by age, as in the text. Table 3.2 lists the specific glaciovolcanic features of MCVF centres. Appendix 2 gives a list of sample locations and a detailed summary of petrography. 3.2 GEOLOGY AND STRATIGRAPHY OF INDIVIDUAL CENTRES 3.2.1 Centres erupted prior to the Late Wisconsinan (Fraser) Glaciation (>30,000 B P ) 2 2 The Fraser Glaciation occurred at -30,000-10,000 BP (Fulton 1969; Mathews et al. 1970; Fulton 1971; Clague 1980, 1981, 1991, 1994; Friele and Clague 2001). 25 Table 3.1. Summary of rock types, volumes, and ages of centres of the MCVF. Details about Mount Cayley are from Souther (1980). centre rock types ice-contact features? Estimated current volume (km3) age 2basis of age estimation Mount Cayley dacite to rhyolite flows, breccia, and tephra none documented 4.2 >2.7 ± 0.7 Ma to 0.31 ±0.05 Ma two K-Ar dates (Green et al. 1988) Mount Fee dacite flows, breccia, and tephra none documented 0.28 >50,000 y lack of ice-contact features, degree of erosion Slag Hill main mass andesite flows and traces of breccia and hyaloclastite yes 0.35 -30,000 to 10,000 y; Qvpau^ possibly older; Qvpau14 <10,000 y 3ice-contact features, degree of erosion Slag Hill tuya andesite flows yes 0.0094 15,000 to 12,000 y ice-contact features, degree of erosion Pali Dome East andesite flows, minor breccia, and traces of hyaloclastite yes 0.2 >50,000 to-10,000 y ice-contact features, degree of erosion Cauldron Dome andesite flows and traces of breccia yes 0.68 15,000 to 12,000 y 4ice-contact features, degree of erosion Ember Ridge andesite flows, and minor breccia and hyaloclastite yes total = 0.16 ave. = 0.020 15,000 to 12,000 y ice-contact features, degree of erosion Little Ring Mountain andesite flows and very minor breccia yes 0.021 15,000 to 12,000 y ice-contact features, degree of erosion Pali Dome West andesite flows and minor breccia yes 0.2 15,000 to 10,000 y ice-contact features, degree of erosion Ring Mountain andesite and dacite flows and minor breccia yes 0.64 15,000 to 12,000 y ice-contact features, degree of erosion Tricouni Southeast andesite flows and minor breccia yes 0.3 12,000 to 10,000 y ice-contact features, degree of erosion Tricouni Southwest basaltic andesite flows yes 0.13 12,000 to 10,000 y ice-contact features, degree of erosion 'Volumes are minima, since all centres have been subjected to varying degrees of erosion, and at many centres, material is buried beneath modern ice. 2Ice-contact features include fine-scale joints and joint orientations indicating cooling surfaces inconsistent with apparent paleotopography. If ice-contact features show minimal erosion, they are taken as evidence that units erupted during the most recent glaciation, the Fraser Glaciation (-30,000-10,000 BP). If the ice-contact features are at high elevations, they are inferred to have formed near the height of the Fraser Glaciation (12,000-15,000 BP). Low-elevation impoundment features are assumed to have formed during the waning stages of the Fraser Glaciation (12,000-10,000 BP). If ice-contact features are significantly eroded, they are taken as evidence that units erupted during a pre-Fraser glacial period. 'Three K-Ar dates are available (0.73 ± 0.07 Ma, 0.25 ± 0.03 Ma, and 0.60 ± 0.03 Ma; Green et al. 1988). 4 A K-Ar date is available (0.49 ± 0.08 Ma; Green et al. 1988). Table 3.2 Summary of glaciovolcanic features at MCVF centres. Cauldron Dome is broken into upper and lower sequences, Ember Pudge is broken into separate edifices (except for the two Mount Brew masses, which are grouped together), and the northwest flow at Ring Mountain is listed separately). unit 1dominant landform type ^ne-scale (3-30 cm) columnar joints horizontal columnar joints fine-scale planar joints pseudo-pillow joints fine-scale irregular joints sharp changes in joint orientation radiating/ fanning joints at flow margins hyaloclastite evidence for lava emplacement into fractured ice 'large-scale morphology Slag Hill main mass (Qvpaul) IMP, SGD? VC R R C R C VR C Slag Hill tuya (LQvpa7) FDT VC R R R C C VC Pali Dome East (LQvpaul) IMP C VR R R R VR C Upper Cauldron Dome (LQvpau2) U C VR R VR C C R V C Lower Cauldron Dome (LQvpau3) IMP C VR VR C R R VC Ember Ridge Northwest (LQvpal) SGD VC R VR C C VC V C Ember Ridge North (LQvpa2) SGD VC C R C C C R? VC Ember Ridge Southwest (LQvpa3) SGD VC C C C C C R VC Ember Ridge West (LQvpa4) SGD VC C VR C C R VR? V C unit dominant landform type ^ne-scale (3-30 cm) columnar joints horizontal columnar joints fine-scale planar joints pseudo-pillow joints fine-scale irregular joints sharp changes in joint orientation radiating/ fanning joints at flow margins hyaloclastite evidence for lava emplacement into fractured ice 3large-scale morphology Ember Ridge Southeast (LQvpa5) SGD C c R C C C C VC Ember Ridge Northeast (LQvpa6) SGD vc C c Little Ring Mountain (LQvpau5) FDT vc R R C VC VC vc Pali Dome West (LQvpau6) IMP vc R R VR R c c c Ring Mountain (LQvpau7) FDT vc VR VR C c c vc Ring Mountain northwest flow (LQvpa8) IMP c R VR R R R c Tricouni Peak Southeast (Lqvpadu) IMP, SGD?, U vc R R C C C c Tricouni Peak Southwest (Lqvpba) IMP vc C VC VC VC VR vc ' S G D = subglacial dome, FDT = flow-dominated tuya, IMP = impoundment feature, U - unclear eruption style (has obvious ice contact features, but stratigraphy and history poorly understood); landform types are discussed in detail in Chapter 4 2VC = very common, C = common, R = rare, VR = very rare 3Large-scale morphology as an indicator of glaciovolcanism (e.g. overthickened flow margins, edifice shapes inconsistent with paleotopography) Mount Cayley Mount Cayley lies in the central MCVF (Plate 1, Figure 3.1). Rising to an elevation of 2385 m at its principal summit, and to >2200 m at the lesser summits of Pyroclastic Peak, Wizard Peak, and Vulcan's Thumb, its northeastern margin abuts the southern end of an irregularly-shaped glacier that measures 5 by 9 km and trends slightly west of north. The south side of Mount Cayley terminates in a nearly vertical cliff >500 m high immediately above the Turbid Creek valley (Figure 3.2). Total volume of Mount Cayley is estimated at 4 km ; this was calculated by approximating the three-dimensional shape of Mount Cayley as a four-sided pyramid. Mount Cayley is a composite pile formed during at least three stages of activity prior to the Salmon Springs Glaciation (Souther 1980). The description that follows is based on the work of Souther (1980), unless otherwise noted, and there is no further interpretation given. Mount Cayley stage The oldest volcanic rocks at Mount Cayley represent the Mount Cayley stage, and comprise a composite pile of porphyritic dacite flows, tephra, and breccia. The lowest unit, EQvul (Souther's Unit 9), consists of a southwesterly-dipping prism of dacite and rhyolite flows and tephra cut by many dykes and sills. Many outcrops, especially the prominent southwestern cliff exposures, are stained a light yellow or red colour by hydrothermal alteration. EQvul is overlain by EQvpdul (Souther's Unit 10), which consists of massive flows of plagioclase-hypersthene-hornblende-phyric dacite up to 150 m thick that form the summit and north side of Wizard Peak and dip steeply away from 30 Figure 3.1. Mount Cayley Quaternary volcanic deposits (Souther 1980). The subglacial distribution of Mount Cayley units is not shown. The unit numbering system is new to this thesis. The contour interval is 100 m. 31 Qvpdu2 Q v u 2 Qvpdul Qvpyul Qvul I EQvpdl EQvpdul Figure 3.1. Legend1 Shovelnose stage plagioclase-hornblende-phyric dacite lava domes and conduits (Souther's unit 16) dacite lava flows and breccias (Souther's unit 15) Vulcan's Thumb stage biotite-phyric dacite cupolas and dykes (Souther's unit 14) bedded, variably consolidated tephra of unknown composition (Souther's unit 12) plagioclase-pyroxene-hornblende-biotite-phyric dacite lava flows and breccias (Souther's unit 13) Mount Cayley stage plagioclase-pyroxene-hornblende-phyric dacite intrusive spine (Souther's unit 11) plagioclase-pyroxene-hornblende-phyric dacite lava flows, necks, and breccias (Souther's unit 10) altered dacite and rhyolite lava flows, dykes, and tephra (Souther's unit 9) Other MCVF centres Ice ® Thermal springs Vents location of K-Ar dated sample (Green et al. 1988) ' A l l unit descriptions are from Souther (1980). 32 Figure 3.2. Photographs of the Mount Cayley volcanic complex, (a) View from a ridge west of Mount Fee, southwest of Mount Cayley. (b) View from a point on the ridge between the Turbid and Shovelnose Creek valleys, southwest of Mount Cayley; the summit of Mount Cayley is obscured by that of Pyroclastic Peak. 33 Mount Cayley. Mount Cayley itself is a narrow jagged ridge of similar intrusive dacite, EQvpdl (Souther's Unit 11). Intrusion of this spine was the final event in the Mount Cayley stage of activity. Vulcan's Thumb stage The Vulcan's Thumb stage commenced with the eruption of Qvul (Souther's Unit 13) , which includes the spire of Vulcan's Thumb, covers the top of Pyroclastic Peak, and rests on the steep surface of older Mount Cayley deposits and basement. Qvul consists of massive lava flows and blocky agglutinated breccias of plagioclase-hypersthene-hornblende-biotite-phyric dacite. It is K-Ar dated at 2.7 ± 0.7 Ma (Green et al. 1988). Qvul forms a portion of the ridge south of Wizard Peak, as well as the prominent pinnacles on the ridge at Pyroclastic Peak, which feature steeply-dipping cooling units up to 130 m thick. Qvul is overlain and flanked by unconsolidated or poorly consolidated bedded tephra, Qvpyul (Souther's Unit 12), that extends southwest in a lobe more than 4 km long and at least 1 km wide. The tephra is off-white, comprises ash to lapilli-sized fragments, and is heavily eroded to form vertical cliffs and ridges (Kelman, this thesis). Qvpyul also includes a 130 m thick section of blocky "tuff breccia" in the saddle between Mount Cayley and Wizard Peak, consisting of angular blocks of biotite-bearing dacite up to 50 cm across suspended in a white matrix of lapilli and ash. Extreme denudation has exposed Qvpyul and the underlying Mount Cayley stage lava units in the deep and precipitous canyon of Turbid Creek. The Turbid Creek canyon also contains three cupolas and several tabular masses of biotite-bearing porphyritic dacite, Qvpdul (Souther's Unit 14) ; their lithology suggests they are related to the Vulcan's Thumb dacite, Qvul. 34 Shovelnose stage The final stage of eruption at Mount Cayley, the Shovelnose stage, involved the eruption of a thick sequence of dacite flows, Qvu2 (Souther's Unit 15), extending down the Shovelnose and Turbid Creek Valleys almost as far as the Squamish River. The youngest Mount Cayley unit, Qvpdu2 (Souther's Unit 16), consists of two dacite domes in the upper Shovelnose Creek valley. The first dome has complex tiers of radiating small diameter columnar joints, is exposed as 400 m high cliffs, and is in intrusive contact with tephra of Qvpyul (Souther's unit 16) in the Shovelnose Creek valley. It is K-Ar dated at 0.31 ± 0.05 Ma (Green et al. 1988). The second dome, located to the north, is a steep-sided exogenous mass of plagioclase-hornblende-phyric dacite that is prominently columnar jointed at 15 cm spacing (Kelman, this thesis), includes a basal colonnade, and rests on blocky bedded tephra overlying basement. Mount Fee Mount Fee (Plate 1, Figure 3.3), located in the central MCVF southeast of Mount Cayley, consists of a narrow arcuate ridge of dacite lava flows and minor fragmental rocks. The north-south-trending limb of Mount Fee is a precipitous spine 1.5 km long and 0.5 km wide, while the east-west-trending limb is 2 km long and 0.5 km wide. It is lower and flat-topped along most of its length, descending eastward with a 760 m elevation difference between the west and east ends. There are four mappable units, all dacitic in composition: Qvdp, a small pyroclastic remnant; Qvdb, a sequence of breccias and minor associated lava flows; Qvpdl, porphyritic dacite flows that comprise the east-west limb; 35 Figure 3.3 Mount Fee Quaternary volcanic units. The contour interval is 20 m. Figure 3.3 Legend for indiv idual centre maps' &:*:•;.£:* oxidized material B volcanic breccia H hyaloclastite |69 columnar joint orientation <g> vertical columnar j oints | horizontal columnar j oints columnar joints <30 cm in diameter, attitude variable on smaller than mappable scale O columnar joints, attitude and size not defined 82 planar joint attitude or attitude of contact between flows B planar joints, attitude variable on a smaller than mappable scale 0 pseudopillow joints well-defined margin of subunit ^ „ ' inferred margin of subunit or inferred base of subunit/ice (used in cross-sections) (§) inferred vent location ( ^ r /~ r > landslide scar < m *iv primary or near-primary near-vertical cooling surface El location of K-Ar dated sample (Green et al. 1988) r i v e r volcanic rocks2 (colour varies between units and is as close as possible to colouring from Plate 1; lighter shading under ice) ice 'This legend applies to the small maps for all MCVF centres except Mount Cayley. The symbol for inferred margins of subunits is also used in most cross-sections. 2On each individual centre maps, the colours for that centre are as on the Plate 1 map, while other centres are in dark grey. Contact types for units boundaries are not shown on individual centre maps, but subunit boundaries are defined. 37 and Qvpd2, a coarsely porphyritic dacite lava with minor breccias that comprises the bulk of the main spine. Only the latter two units are volumetrically significant. Total current volume of Mount Fee is estimated at 0.3 km3, however, erosion has removed a considerable amount of material, especially from the two oldest units. Original eruptive volumes could have been several times larger. Unit Ovdp Qvdp comprises a small flat-lying lens of pyroclastic material, cropping out low on the western side of the main Mount Fee spine. It consists of dark brown, well-indurated, unwelded, crystal-rich, lithic-rich, matrix-supported lapilli tuff. Clasts are angular to subangular, 0.2-50 cm in size, and comprise nonvesicular plagioclase-hornblende-orthopyroxene-biotite dacite similar to the Qvpdl lava flows of the east-west limb of Mount Fee. Qvdp's stratigraphic position, minimal extent, and degree of erosion imply it is older than any of the other Mount Fee rocks The lava in the Qvdp clasts contains 15-20% plagioclase phenocrysts up to 2 mm in size, 1-2% microphenocrysts of hornblende that lack the opaque oxide haloes so common in most MCVF hornblendes, microphenocrysts of orthopyroxene, Fe-Ti oxides phenocrysts, biotite (possibly xenocrystic), and rare quartz-clinopyroxene xenoliths. The groundmass contains plagioclase, orthopyroxene, and clinopyroxene. Unit Ovdb A series of steeply east-dipping intercalated breccia-lava flow couplets is exposed on the eastern side of Mount Fee (Figure 3.4a), and patches of similar breccia are exposed 38 Figure 3.4. Photographs of Mount Fee. (a) The southern side of the east-west limb. Above the solid line is Qvpdl, the lava flow unit that comprises the bulk of that limb. Below the solid line are the lava breccia-lava flow couplets of Qvdb (L = lava, B = breccia), with approximate contacts highlighted with dashed lines. The small circle and arrow highlight a person, (b) Subvertical planar joints and steep irregular erosional forms on the south side of Qvpdl. (c) Detail of subvertical planar joints of Qvpdl. (d) The main spine of Mount Fee (Qvpd2) viewed looking south from the saddle between the limbs. on the north and southeast sides of Mount Fee. These breccias and lava flows are cut by the two units that comprise the bulk of Mount Fee, Qvpdl and Qvpd2. The breccias are monolithologic, matrix- to clast-supported, poorly sorted, imbedded, and locally oxidized, and contain subangular to subrounded ash- to block-sized fragments of the adjacent lava. On the east side of Mount Fee, where the breccias crop out in couplets intercalated with lava flows, they are also crudely bedded. The Qvdb lava is dark grey (fresh) to light grey/red-grey (weathered) nonvesicular plagioclase-homblende-orthopyroxene-phyric dacite3. Where Qvdb occurs as breccia-lava flow couplets, the thickness of each lava flow or breccia portion is 2-10 m. On the northern and southeastern sides of Mount Fee, only small quantities of massive breccia are present. The breccia at the north end of Mount Fee is similar to the breccias in the couplets east of Mount Fee but contains more angular fragments, more matrix, and is highly oxidized; flow-banding is present in samples from the rubble at the north end. The breccia on the southeast side is similar to that at the north end, but is less oxidized and more comminuted. Qvdb lava samples contain 5-15% plagioclase up to 4 mm in size, 1-10% hornblende microphenocrysts and pseudomorphs, 1-2% orthopyroxene microphenocrysts that have a strong pink-tan pleochroism4, and Fe-Ti oxide microphenocrysts. The groundmass contains plagioclase and clinopyroxene. Unit Qvpdl 3 "Weathered" colour refers to the colour produced by the ongoing modern weathering, not to any fossil weathered surfaces separating individual flows; this statement applies throughout Chapter 3, unless otherwise noted. 4 This pleochroism is distinctive amongst other Mount Fee orthopyroxenes, which are virtually colourless. 40 A sequence of dacite lava flows, Qvpdl, comprises the narrow, flat-topped, steep-sided east-west ridge and the northern third of the main limb of Mount Fee (Figure 3.4b). The contact between Qvpdl and Qvpd2 (the intrusive spine of Mount Fee) is located in the rubble and snow-covered saddle between the two limbs. Qvpdl consists of dark grey (fresh) to light grey/red-grey (weathered) porphyritic plagioclase-hornblende-orthopyroxene-biotite dacite lava occurring as one flow at least 150 m thick, overlying two or three thinner flows of the same lava. Vesicularity is almost nil. Along the east-west limb, planar jointing5 is well-developed and has 2-15 cm spacing; its orientation varies but is always near-vertical, resulting in erosion into steep pinnacles (Figure 3.4b, c). Poorly-developed platy joints with irregular attitudes and 5-10 cm spacing occur at the top of the east end of the ridge, then grade westward into irregular joints6 with >60 cm spacing along the top of the ridge and well-developed near-vertical platy joints along the flanks of the ridge, then into massive lava at the western end of Qvpdl. A single patch of white and grey tillite occurs on the western side of the saddle between the two limbs of Mount Fee. It consists of rounded to subangular sand to cobble-sized basement fragments and rare black fragments that may be volcanic, although they do not resemble any of the Mount Fee lava samples. The tillite is extremely bleached and contains small cavities lined with secondary minerals representing hydrothermal 5 The term "flaggy" is commonly used to describe a type of planar jointing, and appears in some glaciovolcanic literature, but its use is not clearly defined. Bates and Jackson (1990) define it as "splitting or tending to split into layers of suitable thickness for use as flagstones..." but give no definition specific to volcanic rocks. The term is not mentioned in McPhie et al.'s (1993) book on volcanic textures, nor is it mentioned in Lescinsky and Fink's (2000) paper, which summarizes jointing types in glaciovolcanic rocks. Given the vagueness in how it is typically used, it is not used in this thesis. The term "planar" is used instead. 6 The term "irregular" is used in this thesis to describe any jointing pattern that is not columnar, planar, or pseudopillow. 41 alteration. Qvpdl includes 15-20 % plagioclase phenocrysts up to 5 mm in size, 2-3% hornblende phenocrysts that are 1 mm in size, colourless microphenocrysts of orthopyroxene, rare biotite clots and pseudomorphs up to 1 mm, Fe-Ti oxide microphenocrysts, xenocrysts of quartz (±pyroxene), and granitic xenoliths. The groundmass contains plagioclase and orthopyroxene. Unit Ovpd2 Qvpd2 comprises most of the southern two-thirds of the main limb (spine) of Mount Fee. It forms a series of ridges and vertical towers that cut through the centre of the other Mount Fee units (Figure 3.4d). Most contacts are obscured by breccia or snow, but the geometry of Qvpd2 suggests that its contacts with the other units are nearly vertical, and Souther (1980) identified a nearly vertical contact between glassy lava of Qvpd2 and shattered granitic wall rock. Qvpd2 samples are grey (fresh) to red-grey (weathered) porphyritic plagioclase-biotite-orthopyroxene dacite that is massive to sparsely (>1 m) and irregularly jointed and locally oxidized, especially at the southern end. Vesicularity is virtually nil. A small patch of red, oxidized breccia occurs at the south end of the north-south limb. It is matrix-supported, contains subangular fragments of the adjacent Qvpd2 lava, and weathers with an irregular cauliflower-like surface. Qvpd2 contains 15-20% plagioclase phenocrysts up to 6 mm in size, 2-3% dark brown biotite up to 1 mm in size, microphenocrysts of orthopyroxene, quartz xenocrysts (lacking the fringe of clinopyroxene that is almost ubiquitous in MCVF samples), and granitic xenoliths. The groundmass has a distinctive sugary texture and contains 42 plagioclase, and orthopyroxene. Discussion No glaciovolcanic features are recognized at Mount Fee. Its elevation (2162 m) is such that the tops of its tallest spires would have protruded through the ice at all but the maximum extent of the Fraser Glaciation (Ryder et al. 1991; Clague 1994). The denuded state of the spine, and the small modern volumes of the oldest units, Qvdp and Qvdb, as well as the lack of any till beneath the contact with the basement (Souther 1980), suggest a pre-Salmon Springs age (>50,000 BP). Events were as follows, and are illustrated in Figure 3.5: (1) One or more explosive eruptions generated the pyroclastic deposit whose only remnant is the small outcrop of Qvdp. Considerable erosion followed. (2) Qvdb erupted onto steep slopes as a series of flows and autobreccias or mass flow deposits, building a cone of breccia-dominated material. The presence of small amounts of the Qvdb breccia on the north and southeastern margins of Mount Fee, and the geometry of the succeeding unit, Qvpdl, suggest that a substantial edifice was built; its hypothesized original extent is shown in Figure 3.5. The lack of erosion between the breccia-lava flow couplets, the uniformity of their dip angles, and the uniform appearance of all the couplets suggest repeated pulses of lava and debris that were closely spaced in 43 r o c k s u n r e l a t e d t o M o u n t F e e Figure 3.5 The development of Mount Fee. Al l units are dacitic. The oldest unit is the pyroclastic deposit, Qvdp, which exists only as a single small outcrop on the west side of Mount Fee. The original extent of Qvdp is unknown, hence it is not included in this series of sketches, (a) The second unit, Qvdb, is a series of intercalated breccias and lavas, which formed a steep-sided cone, (b) Following the emplacement and partial erosion of Qvdb, the series of lava flows of Qvpdl erupted from a vent on the north side of the Qvdb cone and flowed downhill to the east, (c) After further erosion, Qvpd2 was emplaced vertically, intruding through the older units; it may not have ever reached the surface. The sketch shows a time after the emplacement of Qvpd2, when erosion has exposed its highest levels, (d) Mount Fee at the present time. Erosion has removed most of the cone of Qvdb, leaving only remnants on the north, east, and southeast sides of the edifice. Erosion has also exposed up to 300 vertical meters of Qvpd2, resulting in the formation of a precipitous spine. 44 time7. (3) Qvpdl erupted as at least three pulses of lava from a vent on the Qvdb cone, V I , high on the north side of Mount Fee. The orientation of the east-west limb suggests that the eroding cone of Qvdb directed the Qvpdl lava flows to the north and east (Figure 3.5). Considerable erosion followed. (4) Qvpd2, the youngest unit at Mount Fee, intruded through all other units (Figure 3.5); the steepness of its contacts with the wall-rocks indicates a nearly vertical emplacement. The breccia at the south end is interpreted as a vent breccia of Qvpd2. The material that Qvpd2 intruded through (units Qvdb and Qvpdl) has, along the spine, been eroded away. The presence of traces of tillite overlying Qvpdl (the third unit) supports the hypothesis that at least two glaciations occurred following its emplacement, an early one from which the tillite was derived, and a later one that eroded it considerably; however, the timing of these glaciations relative to the emplacement of Qvpd2 (the fourth and final unit) is unclear. 3.2.2 Centres erupted during the Fraser Glaciation (-30,000-10,000 BP) Slag Hill Slag Hill is located in the northern MCVF, just south of Ring Mountain, and is 7 Souther (1980) postulated that this material was pyroclastic in origin. However, the lack of features to indicate emplacement while hot implies that the breccias are either autobreccias of the adjoining lava flows or mass flow deposits formed by failures of the steep slope between lava pulses. 45 partially flanked and covered by modern ice (Plate 1, Figure 3.6). It comprises a complex pile of andesite flows and minor fragmental rocks (the Slag Hill main mass, designated as Qvpaul), and an isolated andesite mass 900 m to the northeast, termed the Slag Hill tuya (LQvpa7). The Slag Hill tuya (LQvpa7) is ordered separately from the Slag Hill main mass (Qvpaul) on the Plate 1 map, because of the large average age discrepancy between the two; the tuya is clearly much younger than most parts of the main mass. However, they are discussed together here because they are clearly related, as a small amount of the Qvpaul lava from the main mass crops out beneath the tuya. Total present volume of the main mass is estimated at 0.4 km3. The tuya comprises 0.009 km 3 of material. A small isolated outcrop on the eastern side of the glacier (Figure 3.6) was included as part of Cauldron Dome in the original map of the Mount Cayley region (Souther 1980) but was reassigned to Slag Hill following the 2001-2002 field mapping8. Unit Qvpaul (Slag Hill main mass) The main mass of Slag Hill (Figure 3.6) consists of two prominent lobes, a round-topped, steep-sided east lobe and a flat-topped, steep-sided west lobe, with a deep gully separating the two (Figure 3.7). The east and west lobes merge, at the higher south end, into a narrow, intensely-jointed ridge bounded by ice. Al l lava samples are macroscopically similar, and are mapped as a single unit, Qvpaul. The eastern lobe has a rounded cross-section (Figure 3.7) and consists of at least two lava flows, Qvpaul • and Qvpauh. The largest joints at Slag Hill have 40-65 cm 8 The basis for reassignment of this outcrop is the three-dimensional geometry of Slag Hill, Cauldron Dome, and the glacier, even though it was not possible to resample the outcrop in question. 46 Figure 3.6. The Slag Hill main mass (Qvpaul) and the Slag Hill tuya (LQvpa7). The inferred subglacial distribution of Slag Hill is shown in paler orange. Lines of cross-section (A-A' and B-BO are for Figure 3.7. The contour interval is 20 m. 47 Figure 3.7 Two cross-sections of Slag Hill perpendicular to the long axis of the main mass, viewed looking south, and photographs. A-A' cuts through the high ice-flanked ridge of Qvpaul 3. B-B' cuts through the Slag Hill tuya and the north end of the Slag Hill main mass. Lines of section are shown in Figure 3.6. The legend applies to all succeeding cross-sections. (a) The high elevation ice-flanked ridge of Qvpaul 3, viewed looking east from below the crest of the ridge. Column orientations are highlighted with the white lines. Column widths are narrower than these lines (<30 cm). The photograph is taken at a ninety degree angle to the plane of the cross-section. The ice is not visible, since the image shows only the crest of the lava ridge. (b) Qvpaul „ at the base of the gully between the east and west lobes. Columnar joints at the bottom of the image are 60 cm in diameter. (c) The margin of the west lobe of Slag Hill, Qvpaul 3, viewed from partway up the gully between the east and west lobes. Surfaces are bulbous and near-vertical. Columns are poorly developed, 20-25 cm across, and near-vertical at the base of the cliff; many of the joints in the image are irregular. The person in the image, shown for scale, is marked with a circle and arrow for greater visibility. spacing and occur in Qvpaul i , near the floor of the gully between the lobes, where erosion has exposed lower portions of the east lobe; their orientation suggests a moderately sloping cooling surface inclined into the gully floor (Figure 3.7b). This lower flow extends into the gully and is partially overlapped by much younger flows of the west lobe (Qvpaul 3 ) . The east lobe is steep-sided and consists of at least two flows grouped as subunit Qvpaul. Small, irregularly oriented columnar joints (with spacing estimated at <35 cm) and rare planar joints occur at high elevations along most of the periphery of Qvpaul2, and its current shape approximates the original cooling surface. The western lobe (Qvpaul and the volumetrically minor Qvpaul4) has a different morphology than the east lobe, being flat-topped and lower in elevation (Figure 3.7). Its prominent eastern cliffs (Qvpaul 3) have steep bulbous surfaces and columnar joints with 20-25 cm spacing and vertical or locally irregular orientations (Figure 3.7c). Planar joints normal to columns, or planar joints in the absence of columnar joints, occur at several locations. A small (<5 m across) pod of hyaloclastite9 that merges into coherent lava is present on the steep slope at the top of the gully between the lobes and is part of Qvpaul (the upper west lobe). It is clast-supported; clasts are <0.1- 5 cm across, have jigsaw-fit texture, are petrographically similar to other Slag Hill lava samples, and have 1% rounded vesicles. The glassy matrix is largely altered. The high ridge south of the lobes (part of Qvpaul 3) meanders as it trends slightly west of north, with its summit standing up to 60 m above the adjacent ice. It narrows to <2 m wide at the crest (Figure 3.7a). The west side of this ridge pinches out abruptly <160 m from where it disappears under the edge of the 9 Throughout this thesis, the term hyaloclastite is used to describe a fragmental rock dominated by glass fragments generated during eruption; it is not used to imply any specific fragmentation process (e.g. quench fragmentation). 50 ice (Figure 3.6). Qvpaul4 comprises the small columnar-jointed flow atop the western lobe (Figure 3.6)10. It is <3 m high, has a gentle profile and low aspect ratio11, and is of limited areal extent. The southern end is buried by scree, and much of the surface is covered by rubble. Qvpaul, with the exception of the small flow atop the western lobe (Qvpaul 4 ) , is nearly homogeneous petrographically. Most samples of Qvpaul are black (fresh) to light grey/yellow-grey (weathered) clinopyroxene-plagioclase-hornblende-phyric andesite, and most are nonvesicular; a few have <2% ragged vesicles, and a single sample from the east lobe has -15% rounded vesicles. QvpauU, the small lava flow atop the western lobe, consists of fresh black orthopyroxene-plagioclase-hornblende-phyric andesite with 15% rounded vesicles. Qvpaul 1.3 contains 1-2% hornblende that is up to 1.5 mm in size and is commonly pseudomorphed by Fe-Ti oxides, 0-2% clinopyroxene up to 5 mm in in size, <1% plagioclase up to 2.5 mm in size, and <1% Fe-Ti oxide microphenocrysts. Granitic xenoliths and rounded quartz xenocrysts are present. The groundmass contains plagioclase and Fe-Ti oxides, usually contains clinopyroxene, and occasionally contains orthopyroxene. Qvpaul4 (the small flow atop the west lobe) samples are macroscopically similar to other Slag Hill samples, but possess <1% hornblende phenocrysts up to 1 mm (commonly pseudomorphed) and microphenocrysts of orthopyroxene and plagioclase. Clinopyroxene is absent. The groundmass contains plagioclase but no pyroxene. 10 This flow is not shown in Figure 3.7 due to its small size and thinness. 1 1 Aspect ratio is the ratio of the average vertical thickness of the landform to its horizontal extent. 51 Unit L0vpa7 (Slag Hill tuva) The Slag Hill tuya is a flat-topped, steep-sided mass of andesite 120 m high and 400 by 540 m in plan view (Figure 3.6, Figure 3.7). It is clearly related to the rest of Slag Hill, not only because of its physical proximity, but because lava similar to Qvpaul crops out near the base of the south margin of the tuya. For this reason, the tuya is included as part of the Slag Hill centre. However, all other Slag Hill tuya samples look different from Qvpaul and the tuya appears to be younger than much of the main mass; it is mapped as a separate unit, LQvpa7. LQvpa7 consists of dark grey (fresh) to yellow-grey (weathered) hornblende-plagioclase-phyric andesite that is rich in xenoliths, especially across the uppermost surface of the tuya, where granitic xenoliths more than 1 m long are present. Vesicles vary in abundance from up to 10% across the top surface. The saddle to the southwest of the tuya is prominently oxidized. Jointing is primarily columnar, with 10-30 cm wide columns that change orientation on a smaller than mappable scale. On the southern margin of the mass, columns bend and increase in diameter downwards, merging into large irregular joints that have -60 cm spacing. The centre of the top surface is coarsely jointed. Margins are eroded into spires, ridges, and gullies of finely and irregularly jointed glassy lava. LQvpa7 lava samples are petrographically distinct from those of Qvpaul (the Slag Hill main mass). LQvpa7 contains 10-14% plagioclase phenocrysts up to 2.5 mm in size, 5-15% hornblende phenocrysts that are up to 8 mm and is commonly pseudomorphed, and <1% microphenocrysts of Fe-Ti oxides. The only identifiable groundmass mineral is 52 plagioclase. Discussion The deposits of Slag Hi l l cover an elevation range of 700 m. Their volume and the range in morphological forms at different elevations suggest that Slag Hi l l is a long-lived complex with five or more discrete eruptions. Events were as follows: (1) Qvpaul i erupted as a flow or flows from an unknown vent somewhere beneath what is now the east lobe of Slag Hi l l . The eruptive environment is unknown, as is the original extent of the subunit; it may have covered the area where the Slag Hi l l tuya now sits but i f so, it was thin and subject to erosion. More probably (since a basement high separates the tuya from the main mass), a separate vent beneath where the tuya now sits erupted a small amount of lava that is similar to that in the main mass of Slag Hi l l . The age of Qvpaul i is unknown1 2. (2) Qvpaul 2, comprising at least two lava flows, erupted subglacially from a vent near V I (Figure 3.6) and did not breach the ice surface. Lava piled up beneath the ice, quenching against steeply-dipping cavity walls. Eruption occurred at or near the height of the Fraser 12 The Slag Hill main mass is designated as Qvpaul (Quaternary) and not assigned to the late Quaternary (as were the other MCVF samples of uncertain age) because of this lack of information about the age or eruptive environment of its oldest lava flows, even though the youngest Slag Hill flows appear to date from the end of the Fraser Glaciation or later. 53 Glaciation . (3) LQvpa7 erapted as a single flow from a vent beneath the Slag Hill tuya. It melted a hole in the ice, producing a flat-topped, steep-sided landform with abundant rounded xenoliths of varying lithology incorporated into its upper surface14. Earlier Qvpaul lava at the tuya was buried. (4) Qvpaul3 erupted from a vent somewhere along the southern Slag Hill ridge, probably near V2 (Figure 3.6). The eruption commenced beneath high-altitude alpine ice and melted through it; lava flowed down the slope to the north and south, forming the ridge that is today bounded by ice at high elevations. To the north, the lava spilled into the gully adjoining the east lobe and impounded against the retreating Cordilleran ice sheet to form the steep-sided, flat-topped west lobe. Minimal local water accumulations resulted in the formation of small quantities of hyaloclastite. Eruptions probably took place during the late Fraser Glaciation15. (5) Qvpaul4, a small subaerial flow on the west lobe, erupted from a vent near V3 (Figure 3.6). No interaction with ice occurred16, and the eruption took place during or after the 13 Green et al. (1988) obtained a K-Ar date of 0.73 ± 0.07 Ma for a sample from Qvpaul2; it is discussed in 3.4. The assignation of Qvpaul2 to the Fraser Glaciation is based on the small size of preserved joints and the lack of apparent erosion by glaciers. 14 This phenomenon (xenolith-rich upper surface) appears in other similar landforms in the MCVF. 15 Green et al. (1988) produced two K-Ar dates for Qvpaul3, 0.25 ± 0.03 Ma, and 0.60 ± 0.03 Ma; they are discussed in 3.4. 1 6 This lava flow's three-dimensional shape, lack of evidence for vertical or steep marginal cooling surfaces, and placement atop an obviously impounded lava flow at which there is no evidence for glacial overriding indicate that it did not interact with ice during eruption. 54 late Fraser Glaciation. Pali Dome East Pali Dome East (LQvpaul) is located in the central MCVF, north and northeast of Mount Cayley (Plate 1, Figure 3.8). It consists of a sequence of andesite lava flows and minor fragmental rocks that crops out on the east side of the glacier and continue beneath it. Flows have gentle profiles and coarse jointing at high elevations but commonly terminate in finely-jointed vertical cliffs at low elevations. Total present volume is at least 0.2 km , but a substantial volume of material may be buried beneath the ice. Although the Pali Dome East and Pali Dome West sequences on either side of the glacier were originally assumed to have issued from a common source and were grouped together by Souther (1980), the distribution of basement nunataks within the ice that separates them, the three-dimensional morphology of the volcanic rocks, basement, and ice, and the lithological differences between Pali Dome East and Pali Dome West suggest that they issued from separate vents. They are classed here as discrete volcanic units, although the name "Pali Dome" is retained for consistency. Unit LQvpaul Eastern Pali Dome (LQvpaul) comprises at least four flows of andesite plus minor breccia and hyaloclastite. Most lava samples, apart from variations in weathering, oxidation, vesicularity (varying from 0-2%), and jointing, are macroscopically similar. However, petrographic and erosional variations indicate multiple flows are present. Four subunits are defined: 55 OS Figure 3.8 Pali Dome East (LQvpaul) and Pali Dome West (LQvpau6). Lighter shades of yellow indicate the inferred subglacial distributions of volcanic rocks. The line of cross-section (A-B') for Figure 3.9 is shown. The contour interval is 20 m. LQvpaul i crops out as a rubbly mound the north end of Pali Dome East, at 2200 m, near the edge of the ice (Figure 3.8) . It is grey (fresh) to yellow-grey (weathered), nonvesicular, locally brecciated, planar-jointed, and flow-banded. Unlike LQvpaul 2, which overlies it, LQvpaul 1 contains no mafic phenocrysts. Hyaloclastite crops out at a single location in subunit LQvpaul 1. The hyaloclastite is matrix-supported and partially palagonitized. It consists of 1-30 mm fragments of the LQvpaul 1 lava in a pale brown matrix; portions are clast-supported and grade into coherent lava. Exposure of LQvpaul 1 is minimal. LQvpaul 1 has 1% plagioclase phenocrysts up to 3 mm in size, plus microphenocrysts of Fe-Ti oxides. The only identifiable groundmass mineral is plagioclase. LQvpaul 2 crops out near the north end of the complex (Figure 3.8). It drapes LQvpaul 1 and the basement and is parallel to the subjacent topography. It is pale grey (fresh) to red (weathered) and nonvesicular, is mottled red and grey or flow-banded, and has columnar joints with 70-90 cm spacing. Felsic and mafic xenoliths up to 1.5 cm are present, and angular xenoliths of LQvpaul 1 occur in some samples. LQvpauh is considerably eroded. LQvpaul2 has 5% plagioclase phenocrysts up to 2 mm in size, 1-3% hornblende phenocrysts up to 3 mm in size, and <1% microphenocrysts of clinopyroxene and Fe-Ti oxides. The only identifiable groundmass mineral is plagioclase. LQvpaul 3 crops out in the central part of Pali Dome East (Figure 3.8), overlying LQvpaul2, as a gently sloping flow or flows parallel to basement topography. It is grey (fresh) to red-grey or yellow-grey (weathered) and nonvesicular. Columnar joints are 40-57 60 cm in diameter, and a weak centimeter-scale planar jointing perpendicular to columns is evident. Rare mafic xenoliths are up to 1 cm across. LQvpaul3 contains 5% plagioclase phenocrysts that are up to 1 mm in size, <1% clinopyroxene phenocrysts, <1% orthopyroxene phenocrysts up to 1 mm in size, and <1% microphenocrysts of Fe-Ti oxides. Rare pseudomorphs up to 1 mm probably represent former hornblende phenocrysts; the occurrence of such partial or total pseudomorphs after hornblende in other MCVF lava samples is extremely common. The groundmass contains plagioclase and pyroxene, and has no glass. LQvpauU comprises the south end of Pali Dome East (Figure 3.8). It is probably underlain by LQvpaul 2, although most of the area around LQvpaul4 is covered in scree and/or snow. LQvpaul 4 crops out with flow surfaces similar to that of the underlying slope at high elevations, and as vertical cliffs up to 100 m high (or steep bulbous knobs) at lower elevations (Figure 3.9a, b). At high elevations, coarse columnar joints 10-30 cm in diameter have locally regular attitudes and chisel marks. At low elevations, there is intense fine-scale irregular or columnar jointing, with columns having 10-20 cm spacing. Columns vary in orientation on a smaller than mappable scale and have locally fanning geometries. At the base of the prominent low elevation cliffs are near-vertical columns 70-100 cm in diameter (Figure 3.9c); they decrease in size upwards, and become more irregular in orientation towards the top of the cliff. LQvpaul 4 is glassy and black (fresh) to grey or yellow-grey (weathered), and has 2-3% rounded vesicles proximally and <1% elongated vesicles distally. LQvpaul 4 contains 10-12% plagioclase phenocrysts near the top of the flow and 4% near the base of the flow; they are up to 1.5 mm. It also contains 8-10% hornblende in 58 Figure 3.9. The eastern half (B-B') of cross-section A-B' through Pali Dome East (LQvpaul) and Pali Dome West (LQvpau6; not shown). The small cross-section at the top left shows the complete line of section through Pali Dome East and West, as shown on Figure 3.8. The lava flow surface through which this section passes is LQvpaul4, although the section is probably underlain by LQvpaul2. The legend is the same as for Figure 3.7. (a) Finely jointed upper surface at the northeast end of the glassy cliffs shown in (b). (b) Glassy cliffs formed by impoundment of LQvpaul4 against valley-filling ice. The photos of (a) and (c) show details of the upper and lower margins of these cliffs, respectively. (c) Coarsely jointed lower portion of LQvpaul 4, exposed by erosion at the base of the cliffs from (b). the upper part of the flow and 2 % near the base of the flow; it is up to 2 mm, and is equant or stubby in the upper part of the flow and elongated near the base of the flow. Other phases include 1% orthopyroxene phenocrysts that are up to 1 mm in size, <1% Fe-Ti oxide microphenocrysts, and <1% clinopyroxene microphenocrysts. The groundmass contains plagioclase and pyroxene. Glass content is similar in proximal and distal samples from outcrop, but hand samples of scree from beyond the base of the cliffs are significantly glassier. Discussion Al l eruptions at Pali Dome East stem from a vent (VI) currently covered by ice (Figure 3.8, Figure 3.10). Events were as follows: (1) LQvpaul i erupted subglacially17 as one or more lava flows with an unknown proportion of hyaloclastite. So little of the unit is exposed that it is impossible to ascertain its original extent. However, because it is overlain by subaerial lava flows, it is inferred to have erupted during a pre-Fraser glacial period; this could have been the Salmon Springs Glaciation or an earlier glaciation. Significant erosion followed. (2) LQvpaul2 erupted subaerially, as a flow or flows that overrode LQvpaul i , Interpretation of the flow as subglacial is based on the presence, of the pod of hyaloclastite. 61 Figure 3.10 Three-dimensional views of Pali Dome East, its inferred distribution beneath the ice (lighter purple), and its inferred vent location. (Pali Dome West is also shown.) The pale green nunataks are basement. Cauldron Dome is bright green. Mount Cayley is pink. Scales are for the foregrounds of the images, and are approximate (since the views are perspectives), (a) A view from east of the glacier, (b) A view from the south, taken at a higher angle. Vertical exaggeration is 1.5x. 62 incorporating fragments of it 1 8. This occurred during a nonglacial period prior to the Fraser Glaciation. Significant erosion followed. (3) LQvpaul3 erupted subaerially, as one or more flows, draping the slope atop LQvpaul2. This occurred during a nonglacial period prior to the Fraser Glaciation, probably immediately before it, since erosion of LQvpaul 3 is less significant than for the preceding subunits. (4) LQvpaul4 erupted during the late Fraser Glaciation. At high elevations (now covered by ice), lava breached any ice that was present and flowed downslope to pond against the valley-filling remnants of the Cordilleran ice sheet. This resulted in coarse regular joints, gentle flow profiles, and low glass content at high elevations, and finer and more irregular joints, vertical cliffs, and more glass at low elevations. It is possible that two flows are present, one directed north and one directed south of the basement window at the south end of Pali Dome East (Figure 3.8, Figure 3.10), since the total volume is large, the southern tongue of LQvpaul4 is more heavily oxidized, and the southern tongue extends to lower elevations. Cauldron Dome Cauldron Dome, located in the central MCVF, consists of a flat-topped, elliptical, northeast-trending sequence of stacked andesite lava flows and minor fragmental rocks 1 8 Interpretation of LQvpaul2 and LQvpaul 3 as subaerial is based on their gentle tapering profiles, lack of fine-scale peripheral joints (which would indicate rapid cooling), and lack of horizontal or shallowly-plunging columnar joints (which would indicate vertical or steep cooling surfaces). 63 (LQvpau2) that is abutted at the west end by a lower elevation flat-topped, vertical-sided sequence of andesite flows (LQvpau3) (Plate 1; Figure 3.11). Upper Cauldron Dome (LQvpau2) is 2000 m wide and 2800 m long and has the morphology of a tuya, although its internal stratigraphy is unlike that of a typical tuya, and it is covered and flanked by ice. Lower Cauldron Dome (LQvpau3) is 1800 m long and tapers into the valley. Upper and lower Cauldron Dome together cover an elevation range of >600 m. Total present volume is estimated at 0.7 km . Unit LOvpau2 (Upper Cauldron Dome) Upper Cauldron Dome, Qvpau2 (Figure 3.11), consists of at least five black (fresh) to yellow-grey/red-grey (weathered) andesite flows stacked to a total thickness of 400 m (Figure 3.12b)19. Jointing is dominantly columnar, with 10-80 cm spacing. The steep margins of all of LQvpau2 are oxidized or weathered to red or yellow. The contact with lower Cauldron Dome, LQvpau3, is scree-covered. The stratigraphically lowest subunit, LQvpau2i, is plagioclase-olivine-orthopyroxene-phyric and is >100 m thick. Its thickness makes it likely that it comprises multiple flows, but its margin is buried by debris and the lava flows of lower Cauldron Dome. LQvpau2i's vesicularity is 3-5%, but rubbly lava with higher vesicularity, plus minor breccia, occurs at the northern end; poorly developed pseudopillow joints occur at two locations along the western margin. At the northern end, these pseudopillows grade into columnar joints with <25 cm spacing, irregular joints, and pods of breccia <2 m 19 Only the top and bottom subunits were sampled. The middle units (LQvpau22^t) are inaccessible although indirect observations can be made based on their exposure in the south-facing cliff (Figure 3.12). 64 Figure 3.11. Cauldron Dome, and the inferred subglacial distribution (lightest blue) of upper Cauldron Dome, LQvpau2. The line of cross-section, A-A' , for Figure 3.12, is shown. Boundaries of subunits are not shown. The contour interval is 20 m. Figure 3.12 Figure 3.12. Cross-section A-A' of Cauldron Dome, along the long axis of the sequence. The legend is the same as for Figure 3.7. (a) The lower sequence, LQvpau3, viewed in a direction parallel to the plane of the cross-section (only the upper subunit, LQvpau32, is visible in the cross-section). The ridge in the foreground of the photo is finely jointed and represents the least eroded part of lower Cauldron Dome (LQvpau32); its upper surface and the upper surface of the rest of lower Cauldron Dome are highlighted here to make them more visible. The midground is occupied by the flat-topped main mass of lower Cauldron Dome; the contact between the two subunits is highlighted with a dashed line. (b) The upper sequence, LQvpau2, viewed from the southeast. Contacts between flows in the cliff face are highlighted with dashed lines. A small spine that obscures part of the foreground is outlined with a solid line in the photo but is not shown in the small diagram; it is an erosional remnant of LQvpau2, and does not actually cut across LQvpau22. across. The pseudopillow joints represent cooling that involved water or steam penetration into the lava (Lescinsky and Fink 2000) . The breccias, which are volumetrically insignificant, are clast-supported and consist of angular fragments of the adjacent lava up to 5 cm across; they are autobreccias of LQvpau2i. The second subunit is LQvpau22; it is <40 m thick, has an irregular contact with the underlying lava flow, and varies in thickness considerably. The third subunit, LQvpau23, is <70 m thick, and is limited in areal extent. Fourth in the sequence is LQvpau24, which is -50-130 m thick, is less intensely jointed than the other subunits, and may comprise more than one flow. The fifth and uppermost subunit, LQvpau2s, is plagioclase-clinopyroxene-phyric and nonvesicular, is <100 m thick, and dips gently to the northeast; at its periphery, joint spacing is <25 cm, and orientations vary on a scale smaller than that of the map. LQvpau25 probably consists of multiple lava flows. The lowest subunit, LQvpau2i, contains 1-10% plagioclase phenocrysts that are up to 2 mm in size, 1% olivine phenocrysts up to 1 mm in size, <1% orthopyroxene microphenocrysts, <1% microscopic pseudomorphs that probably represent former hornblende, and <1% microphenocrysts of Fe-Ti oxides. Rare rounded quartz xenocrysts and granitic microxenoliths are present. The groundmass contains extremely variable quantities of glass, and identifiable mineral are plagioclase and orthopyroxene. The highest subunit of upper Cauldron Dome, LQvpau25, contains 1-2% plagioclase phenocrysts that are up to 3 mm, 1-2% microscopic pseudomorphs that 20 Pseudopillow joints consist of curviplanar joints along which there are tiny normal joints similar to those in some pillows, though in pseudopillow joints, the lava flow still exists as a coherent whole. Pseudopillow joints have been recognized at subaqueously emplaced intermediate and silicic lava flows (McPhie et al. 1993) and in glaciovolcanic deposits in the American Cascades (Lescinsky and Fink 2000). 68 probably represent former hornblende, <1% clinopyroxene microphenocrysts, and <1% Fe-Ti oxide microphenocrysts. LQvpau2s contains a higher proportion of granitic xenoliths than LQvpau2- (the lowest subunit in the sequence), and orthopyroxene xenocrysts and quartz xenocrysts fringed with fine-grained clinopyroxene are also present. The groundmass contains plagioclase and clinopyroxene. Unit LQvpau3 (Lower Cauldron Dome) The lower sequence at Cauldron Dome, LQvpau3 (Figure 3.11), consists of at least two grey (fresh) to yellow-grey/red (weathered) plagioclase-orthopyroxene-phyric andesite flows stacked below the west end of upper Cauldron Dome (Figure 3.12). The lower flow (LQvpau3i) is ~80 m thick and the upper flow (LQvpau32) is -140 m thick and of greater areal extent21. The contact between them dips gently towards the southwest (into the valley). The top of LQvpau32 is flat, except for a low ridge along the southern edge (Figure 3.12a) that consists of a chain of mounds up to 100 m high, with columnar joints at 5-20 cm spacing and locally irregular column orientations. Rare planar joints are present on the top surface. Both LQvpau3 flows are dominated by vertical or near-vertical columnar joints with 35-80 cm spacing and are locally oxidized. The upper flow (LQvpau32) has <1-10% vesicles that consist of rounded to irregular near-spherical voids. In low-vesicularity samples, void space is primarily represented by fractures up to 0.35 mm across. 21 The upper flow, LQvpau32, is K-Ar dated at 0.49 ± 0.08 Ma (Green et al. 1988) but I believe this date to be erroneously old. (This is discussed in 3.4.) 69 LQvpau3 lava samples'" contain 3-15% plagioclase phenocrysts that are up to 3 mm, <l-5% orthopyroxene microphenocrysts, <1% Fe-Ti oxide microphenocrysts, and <1% microscopic pseudomorphs that are probably former hornblende. Three types of xenoliths are present: rounded quartz xenocrysts up to 1.5 mm across (with or without fringes of fine-grained clinopyroxene), polycrystalline aggregates of orthopyroxene + opaques + olivine (potentially fragments from upper Cauldron Dome, LQvpau2), and granitic fragments up to 8 mm. The groundmass contains plagioclase and orthopyroxene. Discussion Current and past ice distributions, overall morphology, and the steep, finely jointed marginal cliffs indicate that ice was a significant influence during Cauldron Dome eruptions, and both Cauldron Dome units were assigned to the late Quaternary based on this fact. The K-Ar date determined for lower Cauldron Dome (0.49±0.08 Ma; Green et al. 1988) places it the later Quaternary, but is hypothesized to be erroneously old because of the field evidence that implies eruption during the most recent major glaciation. Events were as follows: (1) The lowest subunit of upper Cauldron Dome, LQvpau2-, erupted as a lava flow or flows either against or beneath ice and in the presence of little or no ponded water, as implied by the lack of pillows or hyaloclastite. The vent was in the area shown on Figure 3.11, and was buried by succeeding flows. 22 Only the upper lava flow (LQvpau32) at lower Cauldron Dome was sampled. 70 (2) Subunits LQvpau22-5 erupted successively, as flows, from the same vent as LQvpau2-. Their eruptive environments are not known, but the overall steep profile of the edifice suggests they too were flanked or overlain by ice. The irregularity of the surfaces between them indicates that significant erosion occurred between the emplacements of the subunits. (3) The lower Cauldron Dome flows, LQvpau3i and then LQvpau32, erupted successively from a vent somewhere beneath the current flat upper surface of lower Cauldron Dome (this vent is not shown on Figure 3.11). Although lower Cauldron Dome sits at an elevation beneath upper Cauldron Dome, it does not underlie it stratigraphically but must have erupted later, because its lava flows impounded against ice at lower elevations and there is no evidence of glacial overriding that would imply the subglacial or ice-marginal eruption of upper Cauldron Dome after it. This observation was first made by Souther (1980). Both LQvpau3i and LQvpau32 impounded against valley-filling ice during the late stages of the Fraser Glaciation, hence their thickness. The amount of scree that surrounds the cliffs and the coarseness of most exposed jointing indicate that landslides have removed most primary cooling surfaces. Ember Ridge Eight discrete masses of porphyritic plagioclase-(± hornblende ± clinopyroxene ± orthopyroxene)-phyric andesite, aligned in a crescent shape between Mount Fee and Tricouni Peak, are collectively known as Ember Ridge (Figure 3.13, Plate 1). Souther 71 Figure 3.13. The eight lava masses comprising Ember Ridge. Lettered lines correspond to cross-sections in Figure 3.14 and Figure 3.16. Joint types and attitudes are not shown, in order to reduce clutter because of the scale at which the map is displayed. Vent locations, ice-contact cooling surfaces, and occurrences of breccia or hyaloclastite are shown. The contour interval is 20 m. (1980) mapped five of these masses, grouped them based on similarities in morphology and jointing characteristics, and hypothesized that they were coeval. An additional small northeastern mass was identified during the 2001 mapping. The two lava masses at Mount Brew were not mapped, but were grouped with the six Ember Ridge masses, based on their geographic location and elevations, and their reported shapes, compositions, and jointing characteristics (Green et al. 1988; Green, pers. comm.). The six mapped Ember Ridge masses clearly represent six separate vents, because their joint size distributions and joint orientations imply cooling surfaces inconsistent with them being erosional remnants of a more extensive lava flow sequence. Furthermore, the extensive preservation of small joints implies erosion has been minimal; a large volume of material would have to have been removed between the masses, yet there are no traces of volcanic rock or even float between them. The six domes mapped in 2001 are classed as distinct units based on mineralogical and chemical differences. The two Mount Brew masses are grouped as a single unit because they were not mapped for this project. Total present volume of the eight Ember Ridge units is estimated at 0.13 km , and individual masses range from 0.0007 km (Ember Ridge Northeast, LQvpa6) to 0.05 km3 (Ember Ridge Northwest, LQvpal). The Ember Ridge units23 comprise jagged to bulbous, steeps-sided lava piles whose overall joint orientations indicate that their shapes approximate their original cooling morphologies, which are commonly domelike (Figure 3.14, 3.15a, 3.15d). Smaller lobes or knobs across which a coherent jointing pattern is discernible are 23 The general discussion of morphology and joints that follows does not apply to the Mount Brew units, although the literature describing them suggests that they are similar to other Ember Ridge units. 73 Figure 3.14. Cross-section A-A' through Ember Ridge Southwest, LQvpa3, without vertical exaggeration, viewed looking south, showing columnar joint diameters and orientations approximated in the plane of the section. The original cooling surface approximates the current shape of the dome, although erosion has diminished the aspect ratio slightly. The joint measurement on the west side was estimated from the end of the cliff. The joint diameter on the far eastern end was estimated from top of the cliff. Photographs are approximately in the plane of the section and are realistically oriented. Figure 3.15. Photographs of Ember Ridge. (a) Ember Ridge Northwest. This mass is the most dome-shaped of the Ember Ridge masses. Cooling surfaces on the left side of the mass, determined from joint orientations, parallel the surface of the mass. Columnar joint diameter increases from 60 cm at the base of the mass to 20 cm at the summit. (b) Tiny columnar joints (<5 cm diameter) at the summit of Ember Ridge North. The pencil in the centre of the image is 16 cm long. (c) Ember Ridge North, the highest of the Ember Ridge domes, its upper surface consists of irregular ridges and spires, some of whose jointing characteristics suggest they are near-primary. (d) Ember Ridge Southeast. Cooling units are draped over the west-facing hillside and parallel the hillside along much of the lower margin. (e) Pseudopillow joints at Ember Ridge West. (f) Intercalated lava flows and breccia at Ember Ridge Southeast. Contacts between coherent lava and breccia are highlighted with dashed lines, and rock types are indicated with letters (B = breccia, L = lava). The columnar jointing in the vertical screen of lava in the right foreground (arrow) is horizontal. The indicated scale applies to the ridge in the right foreground (the perspective is distorted, with the elevation of the highest visible point ~200 m above the point from which the photograph was taken. 76 typically <50 m in diameter. Horizontal dimensions of the six Ember Ridge masses range from 180 m to 1100 m, heights range from 140 m to 300 m, and thicknesses range from 30-140 m. Most masses display smaller-scale spires, knobs, gullies, and ridges (Figure 3.15c) whose joint sizes suggest that erosion has been minimal. Al l six masses crop out atop ridges (above 1680 m) or drape the sides of ridges (Figure 3.16). Al l masses are dominated by fine-scale columnar jointing with spacing from <5-40 cm (averaging -25 cm), typically increasing in size towards the stratigraphically lowest portions of outcrops. Columns commonly change direction over short distances, and the smallest columns vary in orientation on a smaller than mappable scale. However, as column size increases, orientations become more regular. Planar jointing occurs at numerous locations and is superimposed perpendicular to columnar jointing, with spacing at <1-10 cm. Flow banding normal to columnar jointing, with 1-6 cm spacing, is less common. Jointing details specific to individual Ember Ridge centres are discussed below. Unit LOvpal (Ember Ridge Northwest) Ember Ridge Northwest (LQvpal; Figure 3.15a, Figure 3.16) drapes the western side of a basement high and is the most domelike of all the Ember Ridge masses. Exposed portions consist of a single cooling unit that is up to -100 m thick24. The contact with the underlying basement dips steeply to the west. The lava is dark grey (fresh) to yellow-grey (weathered) and contains <1% vesicles. 24 In comparison to some other glaciovolcanic features in the MCVF (e.g. Ring Mountain and Little Ring Mountain), the fraction of the total edifice exposed at the Ember Ridge centres is large, although the basement contact is still (with one exception, a single location at Ember Ridge Southeast) obscured by scree. 77 Ember Figure 3.16. Cross-sections of five of the mapped Ember Ridge masses. (Ember Ridge Southwest is shown in cross-section on Figure 3.14). The section lines are shown in Figure 3.13. B-B' shows Ember Ridge Northwest and Ember Ridge North, both of which were emplaced at high altitude on steep slopes. C -C meters shows Ember Ridge Northeast, a tiny lava mass draping a slope. D-D' shows Ember Ridge West, which is 0 200 400 600 unusual in that it lies on a comparatively flat surface, the top of a ridge, although it drapes the steep northern 1 1 1 side of the ridge (this is not apparent in the cross-section). E-E' shows Ember Ridge Southeast, which also _ _ lies on a flat slope (and, unlike all other Ember Ridge units, has significant quantities of breccia and shows V . J i . — Z X evidence for synvolcanic slumping of breccia). LQvpal contains 10% hornblende phenocrysts that are up to 2.5 mm, 1% plagioclase phenocrysts that are up to 4 mm and exist as two populations, <1% clinopyroxene phenocrysts that are up to 1 mm, and <1% microphenocrysts of Fe-Ti oxides. Granitic xenoliths comprise <1% of the rock. The groundmass contains plagioclase and pyroxene. Unit LQvpa2 (Ember Ridge North) Ember Ridge North (LQvpa2) (Figure 3.15c, Figure 3.16) drapes the top and eastern side of a ridge in a series of jagged forms and consists of one or more of lava flows up to 100 m thick. The summit retains the smallest columnar joints (<5 cm in diameter) of any of the MCVF deposits (Figure 3.15b). The contact with the underlying basement dips steeply to the east. Samples are black (fresh) to yellow-grey (weathered) and are nonvesicular. LQvpa2 contains 10% plagioclase phenocrysts that are up to 6 mm in size and 5% hornblende phenocrysts that are up to 1.5 mm in size and are commonly pseudomorphed. Rare rounded quartz xenocrysts occur. The only identifiable groundmass mineral is plagioclase. Unit LQvpa3 (Ember Ridge Southwest) Ember Ridge Southwest (LQvpa3; Figure 3.14) consists of one or more lava flows of up to 80 m thick sitting atop a basement ridge. Samples are dark grey (fresh) to yellow (weathered) and nonvesicular. Most jointing is columnar but a low, gently-rounded mound at the north end has prominent planar jointing with <1 cm spacing. 79 Ember Ridge Southwest is the only Ember Ridge mass at which hyaloclastite has been identified. It occurs in a sequence of pods up to 4 m across along the top of the ridge, intercalated along irregular contacts with coherent, jointed lava. The hyaloclastite is red-yellow and matrix-supported, contains 0.2 mm to 15 cm subangular to subrounded lava clasts whose lithology matches the adjacent lava, and is partially palagonitized. LQvpa3 lava samples contain 10-15% plagioclase phenocrysts that are up to 2 mm in size, 3-4% orthopyroxene microphenocrysts, 1% hornblende phenocrysts that are up to 2 mm and are commonly pseudopmorphed, and 1-2% microphenocrysts of Fe-Ti oxides. Xenoliths consist of granitic fragments. The only identifiable groundmass mineral is plagioclase. Some samples have up to 40% pristine glass. Unit LQvpa4 (Ember Ridge West) Ember Ridge West (LQvpa4) consists of a lava flow up to -60 m thick, sitting atop a ridge (Figure 3.16); the north side of the mass is steep, while the south side has a gentler profile. The basement contact is inferred to be horizontal, except on the north side where the unit drapes down the slope. The lava is dark grey (fresh) to pale yellow-grey (weathered) and has 3-15% vesicles. A zone of atypical jointing <4 m across, best described as pseudopillow jointing, occurs near the highest portion of the ridge of the mass (Figure 3.15e). It consists of hemispherical fractures up to 1 m in diameter, with irregular, poorly developed joints normal to their curved surfaces, weathered yellow-brown on the outer surfaces . 25 These joints are slightly different than typical pseudopillow joints in that the "tiny normal joints" perpendicular to the pseudopillow surfaces are not well-developed (McPhie et al. 1993). 80 Vesicularity is highest near the pseudopillow joints and lowest at the east end of the mass. A single pod of breccia <1 m across is present at Ember Ridge West, near the top of the ridge. It is a red, matrix-supported, breccia of high vesicularity <l-3 mm fragments of the adjacent lava. LQvpa4 contains 3-10% plagioclase phenocrysts that are up to 3 mm, 1-5% clinopyroxene that is up to 1 mm, 1% orthopyroxene that is up to 1 mm, and <1% microphenocrysts of Fe-Ti oxides. Rounded quartz xenocrysts, rare granitic xenoliths, and rare microcline xenocrysts also occur. The groundmass contains plagioclase and orthopyroxene. Unit LOvpau4 (Mount Brew) The Mount Brew masses consist of steep bulbous knobs of sparsely porphyritic black glassy orthopyroxene-phyric andesite (Green et al. 1988) that crop out on the eastern side of Mount Brew. The larger unit drapes over the topography, dipping eastward, while the smaller sits atop a basement high (Green et al. 1988). The length and low aspect ratio of the easternmost Mount Brew mass make it anomalous among the other Ember Ridge masses. Unit LOvpa5 (Ember Ridge Southeast) Ember Ridge Southeast (LQvpa5; Figure 3.15d) sits at comparatively low elevation relative to other Ember Ridge units, draping the western side of a ridge (Figure 3.16). It comprises multiple cooling units up to 60 m thick; the south end consists of a domelike mass of lava and some irregularly-shaped lava protrusions, whereas the north 81 end is a complex pile of breccias and lava flows. Lava samples are black (fresh) to orange or yellow-grey (weathered) and tend to be highly weathered. Vesicularity is 2-10%, comprises large irregular cavities, and is unevenly distributed. Only along the southern margin of the mass is the contact with the underlying basement directly visible; it consists of a mass of irregularly jointed weathered lava abutting highly weathered granite via a half-buried patch of fragmental material. Ember Ridge Southeast is distinctive among the Ember Ridge units because of the presence of significant quantities of breccia at the north end of the mass. The breccia occurs in pods 1-10 m wide intercalated with screens of lava >2 m wide at various steep to near-vertical angles (Figure 3.15f). The lava screens have well-developed columnar joints perpendicular to their margins, and these joints pass through the entire thicknesses of the screens; they are throughgoing. The breccias are monomictic, clast-supported, well to poorly consolidated, and locally bleached grey-white or weathered yellow; they consist of fragments of the adjacent lava, and erode to form precipitous gullies. Vertical screens or ridges of lava <2 m wide, with throughgoing horizontal columnar joints as described above, also occur in the absence of breccia, and are up to 5 m high (Figure 3.15f). LQvpa5 lava samples contain 20% plagioclase phenocrysts that are up to 2 mm in size, 1% hornblende phenocrysts that are up to 5 mm in size, and Fe-Ti oxide microphenocrysts. The only recognizable groundmass mineral is plagioclase. There is up to 30% pristine glass in some samples. Lava samples from the south end of the mass contain more glass and have lower vesicularity, whereas at the north end of the mass the reverse is true. 82 The fragmental material found near the basement contact is off-white to red-brown, matrix-supported, imbedded, and well-consolidated. Its matrix is cryptocrystalline. Its clasts are subrounded to subangular and 0.05-0.5 mm, and include lava fragments, quartz, hornblende, plagioclase, clinopyroxene (which has a distinctive green pleocbroism), opaques, and orthopyroxene, plus secondary mica and other fine-grained green alteration material. Those fragments identifiable as lava show trachytic-textured groundmass plagioclase surrounded by opaque or cryptocrystalline material. Al l samples are extremely weathered. The source of this fragmental rock is unknown; its volcanic component is different than that of adjacent volcanic rocks, and it predates the eruption of Ember Ridge. Unit LQvpa6 (Ember Ridge Northeast) Ember Ridge Northeast (LQvpa6) drapes the end of a basement ridge (Figure 3.16) and is the smallest of the Ember Ridge masses. It consists of a single flow of dark grey (fresh) to yellow-grey (weathered) andesite lava <40 m thick. Rounded vesicles are present at 1-2%. LQvpa6 contains 1% plagioclase phenocrysts that are up to 1 mm, <1% clinopyroxene microphenocrysts, and <1% orthopyroxene phenocrysts up to 1 mm in size. The groundmass contains plagioclase and pyroxene. Discussion Based on their sizes, shapes, jointing characteristics, lack of hyaloclastite, and degrees of erosion, the six mapped masses at Ember Ridge had similar eruptive 83 environments and represent eruptions from isolated point sources beneath ice . Al l six centres represent either single lava pulses or multiple eruptions, closely spaced in time, from single vents. Jointing characteristics and the lack of flat upper surfaces indicate that eruptions did not breach the ice surface. Edifice shapes and joint characteristics are (discounting the effects of erosion) due to some combination of: (1) lava flows exploiting the shapes of preexisting ice cavities or fractures, (2) eruption on steep or irregular surfaces, (3) syneruptive collapse of portions of domes to form breccias, and (4) disruption of earlier-formed cooling units by endogenous intrusion of later pulses of lava. The almost total lack of hyaloclastite indicates that meltwater did not accumulate at the subglacial vents. The scarcity of xenoliths in the uppermost lava flows of the Ember Ridge masses is in contrast to the abundance of xenoliths in the uppermost lava flows of Ring Mountain, Little Ring Mountain, and the Slag Hill tuya. This discrepancy is likely due to the fact that Ring Mountain, Little Ring Mountain, and the Slag Hill tuya melted through the upper layers of the ice, acquiring their loads of debris, whereas the Ember Ridge masses either did not melt up into this debris-rich ice, experienced endogenous eruptions (thus making it impossible for late-stage lava pulses to incorporate debris from the ice), or erupted within ice that carried a smaller proportion of debris. This third possibility is a necessity for Ember Ridge Southeast, where some lava flows appear to have taken the shape of fractures in the ice, thus implying both thin ice and exogenous eruption. However, the former two possibilities may apply at other Ember Ridge masses. 26 Souther (1980) first proposed that the five Ember Ridge domes he mapped were the result of lava erupting beneath ice and piling up directly over the vents. The two Mount Brew masses may also have formed in this way, although insufficient data are currently available to be certain. 84 Relative ages of the eight Ember Ridge units are not known, but similar overlying ice thicknesses and eruptive conditions are probably responsible for their morphological similarities. As there is no evidence for significant post-eruptive erosion by ice, the masses must have been emplaced at or after the height of the Fraser Glaciation, after which time any net changes in the ice thickness would have been decreases. Hence, they are ordered from highest elevation vent ("oldest") to lowest elevation ("youngest"). Their age with respect to other MCVF units erupted during the Fraser Glaciation is postulated based on the elevation of ice contact features and degree of erosion. Little Ring Mountain Unit LQvpau5 Little Ring Mountain (LQvpau5) is located at the northern end of the MCVF (Plate 1; Figure 3.17). It is a cylindrical edifice 700 by 900 m with a flat top and near-vertical sides at least 240 m high (Figure 3.18a), and comprises at least three stacked lava flows atop a basement high. The lowest flow is at least 80 m thick and extends downslope on the northern side of the edifice, while the two upper flows are each <80 m thick; the contact between them is visible only on the north side of Little Ring Mountain. The bounding scree slope covers 4/5 of the total height of the edifice, so it is probable that more flows are buried. Present volume is approximately 0.02 km . LQvpau5 samples comprise dark grey (fresh) to light grey (weathered) plagioclase-clinopyroxene-hornblende-phyric andesite lava. Jointing is pervasive and 85 Figure 3.17. Little Ring Mountain. The line of cross-section for Figure 3.18 is shown. The contour interval is 20 m. 86 Figure 3.18. Cross-section A-A' , through Little Ring Mountain, LQvpau5. The line of section is shown in Figure 3.17. (a) The southeast side of Little Ring Mountain. A flow contact is visible (highlighted) a short distance above the top of the scree slope. There are at least two lava flows above this contact, although they are not obvious except on the north side of Little Ring Mountain. A low erosional spine extends downslope on the right side of the image. (b) Columnar joints on the upper east side of Little Ring Mountain. Columns are nearly vertical in the upper part of the image, and change directions to subhorizontal at the lower left. Column diameter is ~20 cm. 87 usually columnar, with 12-80 cm spacing. Columns have near-vertical, horizontal, locally variable, or radiating orientations. On the east side of Little Ring Mountain, columns within 50 m of the summit are typically <20 cm and have locally irregular orientations; at lower elevations, exposed columns are typically >20 cm across and have more regular orientations, and there is at least one prominent section of horizontal columns, indicating a vertical cooling surface (Figure 3.18b). High on the north and west sides of Little Ring Mountain, there is a section of -80 cm diameter columnar joints that represents part of the uppermost flow, exposed by erosion. There is one occurrence of planar jointing (<2 cm spacing) on the western side of the edifice. Margins of flows are locally eroded into narrow ridges and spires, and several low, narrow, eroded spines extend radially from the bounding cliffs. Vesicles are irregular in shape. The upper surface of the mass has a high vesicularity (10%); this is underlain by a zone of low vesicularity (<1%) that is in turn underlain by a zone of high vesicularity (10%). LQvpau5 contains <l-2% plagioclase phenocrysts that are up to 1 mm and lack compositional zoning (unlike most MCVF plagioclase), 1-2% clinopyroxene phenocrysts up to 1 mm, and <1% hornblende phenocrysts that are up to 1 mm in size and are in most cases pseudomorphed (except on the top surface of Little Ring Mountain). Xenoliths include rounded polycrystalline quartz aggregates, polycrystalline aggregates of pyroxene ± quartz, and rounded quartz xenocrysts with fringes of fine-grained acicular clinopyroxene (or nearly total replacement by clinopyroxene); rare biotite xenocrysts are also present. Only the first two xenolith types are abundant, and these are only abundant on the upper surface of Little Ring Mountain, where they are macroscopically prominent. The groundmass contains plagioclase and pyroxene. The highest proportions of glass are 88 in samples from the upper surface of Little Ring Mountain. Discussion Little Ring Mountain is similar in form to the Slag Hill tuya and, like it, consists entirely of lava flows. However, Little Ring Mountain is larger in volume and represents more than one pulse of lava. It formed as subglacial eruptions melted a hole through the ice, resulting in lava flows repeatedly flooding this hole and impounding against ice on all flanks. As at the Slag Hill tuya, the abundance of crustal xenoliths on Little Ring Mountain's upper surface is probably due to till fragments melting out of ice above or adjacent to the vent. As Little Ring Mountain's uppermost flow was subaerial, its elevation gives clues to the height of the ice sheet at the time of the final eruption. Given that the maximum elevation of the top of the Cordilleran ice sheet for the region was 2300 m (Clague 1989; Ryder et al. 1991) and that Little Ring Mountain's summit is at 2147 m, the final eruption must have occurred at or near the glacial maximum. Pali Dome West Like Pali Dome East, Pali Dome West (LQvpau6; Figure 3.19), which crops out on the western side of the glacier, comprises multiple overlapping andesite flows and very minor fragmental rocks that commonly terminate in finely jointed cliffs. However, erosion is less extensive than for Pali Dome East, so Pali Dome West is probably significantly younger. Total present volume is at least 0.2 km3, but a substantial amount of material may be covered by the current glacier. 89 o Figure 3.19 Pali Dome East and Pali Dome West. Lighter shades of yellow show the inferred subglacial distribution of volcanic units. The line of cross-section (A-B') for Figure 3.21 is shown. The contour interval is 20 m. Unit L0vpau6 The western Pali Dome sequence (LQvpau6; Figure 3.19) consists of at least three flows of black/grey (fresh) to light grey/red-grey (weathered) plagioclase-biotite ± hornblende ± orthopyroxene-phyric andesite, divided into three subunits. LQvpau6i sits atop older Mount Cayley deposits at the south end of Pali Dome West and is its most eroded subunit. High elevation portions of LQvpau6i were recently uncovered due to retreat of the current glacier, hence, are mantled in rubble and patches of snow and ice. LQvpau6i terminates in a near-vertical cliff west of Mount Cayley (Figure 3.20a), most of whose primary cooling surfaces have been removed by erosion. LQvpau62 is less eroded than the underlying LQvpau6i, comprising one flow at low elevation and possibly two at higher elevations; all of LQvpau62 thins to the north. The west margin is a prominent cliff (Figure 3.21a). Columns along the outer margin of LQvpau62 vary in diameter from 15 to 60 cm, averaging ~20 cm along the periphery, and show a trend of increasing size proximally. Orientations vary, often on a smaller than mappable scale, and may be nearly horizontal. Column diameter also increases northward to -70 cm, and the cooling surface dips slightly towards the apparent vent. At the north end of LQvpau62, pseudopillow joints are present (Figure 3.21b) and the top surface of the flow is locally glassy. Vesicularity is <l-3%; where significant, it consists of elongate interconnected vesicles. Portions of LQvpau62 are rounded and glacially polished, probably by the ice tongue to the northeast, which likely overrode LQvpaul more than once during post-Fraser Glaciation climate fluctuations, though it has retreated over the past forty years. 91 Figure 3.20 Three-dimensional views of Pali Dome West, its inferred distribution beneath the ice (lighter purple), and its inferred vent location. (Pali Dome East is also shown.) The pale green nunataks are basement. Cauldron Dome is bright green. Mount Cayley is pink. Scales are for the foregrounds of the images, and are approximate (since the views are perspectives), (a) A view from east of the glacier. The ice-impoundment cliff of Lvpau6, is prominent at the top centre of the image (arrow), (b) A view from the south, taken at a higher angle. Vertical exaggeration is 1.5x. 92 93 Figure 3.21 The western half of cross-section A - B \ through Pali Dome West, LQvpau6. The small cross-section at the top left shows the complete line of section through Pali Dome East and West, as shown on Figure 3.19. The legend is the same as for Figure 3.7. (a) Ice-impounded lava flow of LQvpau62. Jointing is columnar, variable in orientation, and 15-60 cm in diameter. (b) Pseudopillow jointing at the north end of Pali Dome West, within LQvpau62. The margin of the pseudopillow and two of the small normal joints are highlighted with white lines to make them more visible. (c) Subaerial flow LQvpau63 at the north end of Pali Dome West. The contact with LQvpau62, which underlies it, is oxidized and brecciated. It is highlighted with a dotted line. 94 At the northern end of the exposed western part of Pali Dome, a third subunit, LQvpau63, overlies the northern end of LQvpau62 along an oxidized and brecciated contact (Figure 3.21c). Jointing in LQvpau63 is coarsely columnar, with columns up to 100 cm in diameter. Planar jointing and flow banding are superimposed perpendicular to columns. Columnar-jointed lava grades into irregularly jointed and then massive lava. There are 1-2% irregular interconnected vesicles. Total thickness is > 100 m. The second subunit at Pali Dome West, LQvpau62 , contains 10% plagioclase phenocrysts that are up to 7 mm in size, <1% biotite phenocrysts that are up to 1 mm in size, <1% Fe-Ti oxide microphenocrysts, <1% orthopyroxene microphenocrysts, and <1% pseudomorphs that are up to 2 mm and probably represent former hornblende. Xenoliths include granitic fragments and rounded and embayed quartz grains that lack the fringe of fine-grained clinopyroxene common to quartz xenocrysts at most MCVF centres. The groundmass contains plagioclase and orthopyroxene. The highest proportions of glass (up to 20%) are present in distal samples, and glass content diminishes to 5% proximally. The youngest lava flow at Pali Dome West, LQvpau63, is petrographically similar to LQvpau62, but its biotite is fresher, there is more phenocrystic orthopyroxene, and there are traces of clinopyroxene in the groundmass, which is otherwise cryptocrystalline. Discussion Events at Pali Dome West were as follows: 27 Samples of the oldest flow, LQvpau6i, were not available, hence there is no petrographic description, but LQvpau6i is macroscopically similar to LQvpau62. 95 (1) LQvpau6i erupted as a lava flow from vent V2 (Figure 3.19), a location that is currently beneath the ice. The timing of this eruption, with respect to glacial events, is unclear. It is possible that the vertical cliff near the contact with the Mount Cayley units (Figure 3.18a) is a result of lava impoundment against ice, and this would suggest that eruption occurred during the late Fraser Glaciation. However, the extent of erosion makes it difficult to determine the timing and nature of events. Deposits were later overridden by ice. (2) LQvpau62 erupted as one or more flows from vent V2. As with the preceding subunit, eruptions probably penetrated thin alpine ice and then flowed subaerially downslope to impound against the waning Cordilleran ice sheet (at lower elevations than for LQvpau6i). This occurred during the late Fraser Glaciation. (3) LQvpau63 erupted as a lava flow, penetrating thin alpine ice before flowing downslope to the west. It was probably constrained on its south margin by alpine ice, based on the overall great thickness of the flow. However, the terminus was not impounded by ice. Thus, this occurred after ice had downwasted to below 1960 m (during the late Fraser Glaciation or after the end of the Fraser Glaciation). Erosion has been minimal except where fluctuating alpine ice has eroded and polished outcrops at the north end of Pali Dome West. Ring Mountain 96 Ring Mountain28, located in the northern MCVF (Plate 1; Figure 3.22; Figure 3.23), is a cylindrical, flat-topped edifice -1760 by 2300 m across, with steep sides at least 500 m high and a large reentrant in the southern side. It consists of at least five stacked andesite flows grouped collectively as LQvpau7. The base of Ring Mountain and up to 4/5 of its total height are buried in scree. A small discrete flow to the northwest, LQvpa8, is different in composition, but clearly originated from a vent on the lower slopes of Ring Mountain. Present volume of the Ring Mountain tuya is estimated at 0.6 km3, while the volume of the northwest flow is -0.009 km3. Unit LQvpau7 (Ring Mountain tuya) Ring Mountain, LQvpau7, consists of stacked black (fresh) to yellow/red (weathered) plagioclase-clinopyroxene-hornblende-phyric andesite lava flows (Figure 3.23). The uppermost flows are coarsely jointed and separated by layers of oxidized scoria or breccia, and crustal xenoliths are abundant on the top surface. Vesicularity is dominantly low (<l-2%) and consists of interconnected vesicles or fractures up to 0.3 mm wide that form a loose gridlike pattern. Columnar jointing is common and typically has 10-50 cm spacing; columns have near-vertical, horizontal, locally variable, or radiating orientations. Rare planar, hackly, or irregular joints are also present. On Ring Mountain's margins, especially its western side, outcrops contain highly variable, fine-scale jointing and are locally eroded into many small spires, ridges, gullies, and knobs (Figure 3.23a). 28 Souther (1980) referred to Ring Mountain as "Crucible Dome". Both names are informal, but the former is used here because it is more common. 97 Figure 3.22. Ring Mountain (LQvpau7) and the small flow to the northwest (LQvpa8). The line of section for Figure 3.23 is shown. The contour interval is 20 m. Figure 3.23 Figure 3.23. Cross-section A-A' through Ring Mountain (LQvpau7) and Ring Mountain Northwest (LQvpa8). The line of section is shown in Figure 3.22. The legend is the same as for Figure 3.7. Contacts between the individual flows that comprise Ring Mountain are not shown because they are impossible to follow around the entire edifice. (a) Ring Mountain, viewed from below and to the southwest. Spires with fine-scale jointing are indicated with an arrow. It is unlikely that these shapes represent primary morphology, but they are the remnants of the finely jointed perimeter of Ring Mountain that has elsewhere been eroded away. (b) Ring Mountain Northwest (LQvpa8). White lines indicate the attitudes of horizontal columnar joints. (c) Ring Mountain Northwest (LQvpa8). The margin of the flow has coarse vertical columnar joints whose attitudes are indicated with white lines in the centre of the image. LQvpau7 is petrographically homogeneous, apart from slight variations in glass content, texture, and vesicularity. Samples contain 1-3% plagioclase phenocrysts that are up to 5 mm in size, <1% clinopyroxene phenocrysts that are up to 2 mm in size, <1% hornblende phenocrysts that are up to 2 mm in size and almost ubiquitously pseudomorphed, and <1% Fe-Ti oxide microphenocrysts. Four types of xenoliths are present: rounded quartz xenocrysts that may be fringed with fine-grained clinopyroxene, aggregates of plagioclase-pyroxene-opaques (potentially cognate), granitic fragments, and polycrystalline plagioclase aggregates. The latter two types occur only in samples from the upper surface of Ring Mountain, but there they are abundant and may be tens of centimeters across. The groundmass contains plagioclase and clinopyroxene. Discussion The Ring Mountain tuya, LQvpau7, is similar in morphology and internal characteristics to both the Slag Hill tuya and Little Ring Mountain, but is considerably larger in volume than both. It formed by successive eruption of at least five flows from a single vent. Early eruptions were subglacial and are hypothesized to have resulted in irregular piles of flows similar to the Ember Ridge masses, but the ice surface was rapidly breached. Thus, later eruptions involved repeated flooding of an open cylinder in the ice, and this is the explanation for the overall shape of Ring Mountain. The uppermost flows erupted subaerially, flanked by ice ; this is evidenced by their flat-topped morphology, restricted areal extent, oxidation, and brecciated tops. Ring Mountain's summit is at 2192 29 As with Little Ring Mountain and the Slag Hill tuya, the abundance of xenoliths on the top surface suggests that englacial or supraglacial debris was incorporated into the uppermost flow at the time of eruption, and this event may be common to all ice-breaching features in the GVB. 101 m, indicating that for its upper flows to have erupted in contact with a flanking Cordilleran ice sheet, the final eruptions must have occurred at or near the glacial maximum, when ice was at 2300 m (Clague 1989; Ryder et al. 1991), and this is a further indication that the uppermost flows are unlikely to have been subglacial. Unit LQvpa8 (Ring Mountain northwest flow) The isolated lava flow of LQvpa8 crops out -900 m northwest and downslope of Ring Mountain (Figure 3.22). The separation between LQvpau7 and LQvpa8 (the Ring Mountain tuya) is a primary feature and not the result of erosion. Lqvpa8 is 1100 m by 500 m, up to 50 m thick, and covers an elevation range of -400 m (Figure 3.23). It consists of grey (fresh) to yellow-grey (weathered) plagioclase-olivine microphyric to phytic basaltic andesite. At high elevations, LQvpa8 forms vertical cliffs up to -5 m high. Columnar joints have 15-40 cm spacing and horizontal (Figure 3.23b), steep, or locally variable orientations that grade into irregular or massively jointed blocks with 30-40 cm spacing. At lower elevations, column size is 15-60 cm, and patches of irregular or planar joints are present. Flow margins have a gentle aspect ratio, and the flow thins to 2-7 m thick adjacent to the basement contact, where columnar joints are vertical (Figure 3.23 c). Vesicularity is low to moderate (<l-4%) and consists of ragged voids up to 5 cm across. Some samples are flow-banded; bands are -0.15 mm across and filled with fine-grained opaque material (probably a weathering product). LQvpa8 contains <1% plagioclase phenocrysts that are up to 2 mm in size, <1% olivine phenocrysts up to 1 mm in size, <1% microphenocrysts of orthopyroxene, and 102 <1% microphenocrysts of Fe-Ti oxides. Rare xenocrysts contain mixtures of plagioclase and mafic minerals. The groundmass contains plagioclase and pyroxene. Discussion LQvpa8 erupted from a vent low on the flank of Ring Mountain (Figure 3.22). The fact that LQvpa8 has ice-contact features at 1200 m (Figure 3.23c) but not at 940 m suggests that the lava flow exploited preexisting drainage channels in stagnant ice whose distribution was patchy at the time of eruption. Thus, eruption must have occurred after the Fraser Glaciation's glacial maximum but before the complete disappearance of ice. Tricouni Southeast South of Tricouni Peak are andesite and dacite lava flows that form two non-overlapping sequences grouped as a single unit (LQvpadu; Figure 3.24)30. They consist of an areally extensive mantle of lava flows that crop out as multiple low cliffs and mounds on heavily vegetated slopes, and a small discrete lava knob with an estimated present volume of 0.05 km . Unit LQvpadu LQvpadu consists of undivided black (fresh) to brown-grey (weathered) plagioclase-( ± hornblende ± orthopyroxene)-phyric to nearly aphyric andesite and dacite flows. The larger LQvpadu mass consists of at least five lava flows (divided into subunits 30 The two masses are grouped as a single unit due to their proximity, their possession of glaciovolcanic features at elevations that imply similar ages, and the macroscopic similarity of most lava samples. 103 Figure 3.24. Tricouni Southeast. Lettered lines A' -A" and B ' - B " indicate the cross-sections of Figure 3.25; the lines continue off this map to the west and onto Figure 3.26 to end in A and B, respectively, and correspond to the cross-sections of Figure 3.27. The contour interval is 20 m. LQvpadui-4; Figure 3.24). The smaller, high aspect ratio mass of lava flows adjacent to Freeman Lake (Figure 3.24) consists of at least two anomalously thick lava flows with vertical margins (LQvpadu5 and LQvpadu6). A l l subunits have minor quantities of autobreccia. In the larger mass, columnar joints are abundant and have 5-80 cm spacing in which the variability of orientation increases as joint size decreases; columns have locally radiating, bending, horizontal, or variable orientations (Figure 3.25a, c). Small columns have irregularly-shaped cross-sections (Figure 3.25b), and locally degrade into irregular jointing. Rare planar jointing is present. LQvpadui is a large western flow of andesite that drapes topography with a modern thickness of at least 100 m. It is the most eroded of the LQvpadu subunits, so the original thickness was greater. The low elevation terminus of LQvpadu- includes cliffs up to 30 m from which a large amount of material has been eroded. Some fine joints are preserved. There are traces of weathered hyaloclastite at lower elevations. The hyaloclastite is matrix-supported, massive, and partially palagonitized, and contains angular to subrounded 1-12 mm fragments of slightly glassy lava that is similar to nearby samples of coherent lava. It includes a high (~1%) proportion of xenoliths, which comprise a wide range of lithologies, are more rounded than the lava fragments, and represent till that was incorporated into the hyaloclastite. LQvpadu2 is a large andesite lava flow that drapes over LQvpadui and the underlying topography. It is at least 100 m thick and forms low, finely-jointed cliffs at low elevations. Weathered breccia (rounded fragments of various lithologies, plus a minor volcanic component), occurs on the eastern margin, along the volcanic-basement 105 Figure 3.25. Two cross-sections through Tricouni Southeast (LQvpadu). The lines of cross-section are shown in Figure 3.24. C - C runs approximately north-south. D-D' runs approximately northwest-southeast and shows both discrete lava masses. (a) Lava knob of LQvpadu4. The joint direction changes sharply near the top of the mass. Columnar joints are 25-50 cm in diameter. (b) Columnar joints in LQvpadu^ Most columnar joints along the east side of LQvpadi^ are poorly developed or have variably shaped cross-sections, and many joints are irregular. (c) Fanning columnar joints near Freeman Lake, in LQvpadu6, indicating impoundment against ice. contact. LQvpadu3 comprises two smaller lava flows, a lower dacite and an upper andesite, that overlie LQvpadui and the proximal portions of LQvpadu2. The contact between the two flows is obscured by scree, but geochemical differences between samples make it unlikely that a single eruptive event is responsible. Total thickness is <40 m. LQvpadu3 is finely-jointed and crops out with steep slopes up to 20 m high. LQvpadu4, a small knob of high silica andesite with vertical surfaces up to 10 m high (Figure 3.25a, c), sits directly atop LQvpadui. LQvpadui contains <1% plagioclase phenocrysts that are up to 1 mm, <1% hornblende microphenocrysts, and <1% microxenoliths or xenocrysts of quartz ± plagioclase. The groundmass contains plagioclase and pyroxene. The fraction of pristine glass is as high as 25% in some samples. Vesicularity is 3-7%. LQvpadu2 contains 1-5% plagioclase phenocrysts that are up to 1 mm in size, <1% green-brown hornblende microphenocrysts, and <1% rounded quartz xenocrysts. The groundmass contains plagioclase and orthopyroxene. Vesicularity is 5%. The andesite flow31 of LQvpadu3 contains 1-2% plagioclase microphenocrysts, 1-2% hornblende microphenocrysts, and <1% rounded polycrystalline quartz xenoliths. The only identifiable groundmass mineral is plagioclase. LQvpadu4 contains 10% plagioclase phenocrysts that are up to 5 mm in size, 1% orthopyroxene that is up to 1 mm in size, 1% hornblende up to 3 mm that is commonly totally pseudomorphed, <1% clinopyroxene microphenocrysts, and <1% Fe-Ti oxide 31 The dacite flow was not examined petrographically. It differs from the andesite flow in that it is more phyric, containing 4% plagioclase that is up to 4 mm; otherwise, the two lava flows are macroscopically similar. 108 microphenocrysts. Xenoliths and xenocrysts comprise <1% of the rock and include olivine crystals, olivine-orthopyroxene aggregates, rounded quartz crystals, and granitic fragments. The groundmass contains plagioclase and and pyroxene. The two flows of the discrete mass (LQvpadus and LQvpadu^) form a high aspect ratio knob (Figure 3.25). LQvpadus comprises a pile ~60 m high with coarse columnar joints exposed at its margins. Columnar joints are vertical and >60 cm in diameter and have a superimposed perpendicular planar jointing that has <l-3 cm spacing. Samples were not examined petrographically but are pale grey, weathered, microporphyritic, and macroscopically similar to both LQvpadue lava flows. LQvpadu6 consists of a lower elevation mass with knobs above and below Freeman Lake, comprising two lava flows with steep, finely-jointed margins that commonly have bending or fanning columns (Figure 3.25c). On the northern margin of LQvpadufi, the contact with the underlying soil is horizontal and is weathered and brecciated. At the same location, there is a patch of laminated sand to clay-sized sediment which includes till balls and prominent dropstones up to 8 cm across; dropstones are rounded and of various lithologies or, rarely, comprise angular fragments of the nearby lava. This sediment is waterlain. Immediately above Freeman Lake, however, the contact between LQvpadu6 and the underlying basement is steep and consists of a brecciated zone in which rounded clasts of basement and angular clasts of the lava are mixed. Within the coherent lava adjacent to the contact, there are abundant heterolithic xenoliths up to tens of centimeters across, unresorbed, and commonly rounded. These are probably till fragments. LQvpadu6 lava samples contain <1% plagioclase phenocrysts up to 1 mm in size 109 and <1% orthopyroxene microphenocrysts. In samples from adjacent to Freeman Lake, plagioclase is smaller and more inclusion-rich, and rare hornblende is present; it is up to 2.5 mm and is pseudomorphed. The groundmass contains plagioclase and pyroxene, and is partially replaced with secondary minerals. Vesicles comprise 5% of the rock, form elongated strings, and are up to 1 mm. The waterlain sediment from the margin of the upper LQvpadu^ flow is a bedded to laminated till-volcaniclastic breccia. Clasts include (1) polycrystalline quartz with undulose extinction, (2) quartz-feldspar-mica aggregrates, (3) discrete rounded quartz crystals, (4) lava clasts similar to the lava of LQvpadus, and (5) unidentified lava fragments (containing 3-5% plagioclase, 2% long bladelike green-brown hornblende, and Fe-Ti oxides in a cryptocrystalline groundmass). Lava fragments are angular and are <3.5 mm in size. Discussion All eruptions at Tricouni Southeast took place during the late Fraser Glaciation, when the elevation of the top of the downwasting Cordilleran ice sheet was <1400 m (significantly less, in some cases), and minor readvances of the ice sheet may have occurred between some eruptions. Events were as follows: (1) LQvpadui erupted from a vent near the location VI (Figure 3.24), and flowed southwards. It was impounded against ice, as indicated by the thickness of the terminus, the presence of traces of hyaloclastite, and the preservation of some fine-scale jointing. 110 (2) LQvpadu.2 erupted from vent VI (Figure 3.24) and flowed downslope to the southeast, covering the eastern margin of LQvpadui. It impounded against ice on its eastern and southern margins. Erosion and minor readvance of ice followed. (3) The two lava flows of LQvpadu3 erupted successively from vent VI and flowed downslope towards the south atop LQvpadui. Both impounded against small quantities of ice. (4) LQvpadu- erupted from a low-altitude vent, V2 (Figure 3.24). The vent was covered with such thin ice that it was quickly penetrated. (5) LQvpadus erupted from vent V3 (Figure 3.24). Its age is unknown relative to LQvpadu^, though it clearly predates LQvpadul6. Its thickness and restricted extent indicate confinement by ice, although significant erosion has removed the finely-jointed margins. (6) LQvpadu6 erupted from vent V4 near Freeman Lake (Figure 3.24). The vent was probably subaerial but flows impounded against ice above Freeman Lake. Lava probably flowed around the north end of Freeman Lake, then impounded against low elevation ice . The adjacent waterlain sediment probably represents ponding of meltwater behind glacial debris; its dominantly fine grain size and lamination indicate a low-energy 32 Because so much of the ground north of Freeman Lake is covered by rubble and there is a ~40 m difference in impoundment elevation above and below the lake, it is possible that two lava flows comprise LQvpadu<;, although petrographic differences are not great enough to support this contention. I l l depositional environment, and the larger dropstones imply rafting of debris by ice. Tricouni Southwest Tricouni Southwest (LQvpba) is located south of Tricouni Peak (Figure 3.26) and consists of an elongated basaltic andesite flow that forms the east wall of a north-south trending gully up to 200 m deep near the head of High Falls Creek (Figure 3.27). The flow ranges in width from 1900 m at the distal south end, where it spreads east-west, to -400 m at the proximal north end. The west gully wall consists of more gently sloping basement, and the lava-basement contact lies along the bottom of the gully, though it is obscured by scree. The Tricouni Southwest flow (LQvpba) occurs in at least two locations as a poorly consolidated breccia atop LQvpadu, so LQvpba is clearly the younger unit. Unit LQvpba LQvbpa is black (fresh) to yellow-grey/light grey (weathered) olivine-plagioclase-phyric basaltic andesite. It has a complex vertical succession of columnar and irregular joints exposed in the west-facing cliff of the gully. Columns have 10-50 cm spacing (dominantly -25 cm) and orientations that are locally variable, radiating, vertical, or horizontal (Figure 3.27a, b). A zone of vertical columnar joints is typically present near the base of the flow (Figure 3.27a), and is succeeded upwards by variably oriented columnar or irregular joints whose average size decreases towards the top of the flow. Flow-banding whose spacing is 1-2 cm is perpendicular to columnar jointing. Vesicularity is low (<l-5%), and some vesicles are elongated parallel to the flow 112 Figure 3.26. Tricouni Southwest. A-A' and B-B' are cross-section lines for Figure 3.27; they go off the edge of the map to the east and are continued on Figure 3.24 (the map of Tricouni Southeast), ending as A" and B " . The contour interval is 20 m. 113 Figure 3.27. Cross-sections through Tricouni Southwest (Lqvpba). A - A " is through the narrowest and steepest part of Tricouni Southwest and shows the prominent cliff of the High Falls Creek gully. B-B" shows the gentler profile of the southeastern margin of the lava flow. (Tricouni Southeast [LQvpadu] appears on the right half of both cross-sections.) Lines of section are indicated in Figure 3.24 and Figure 3.26. (a) Columnar joints along the bottom of the gully. The joint orientation shifts abruptly from vertical to nearly horizontal in the bottom third of the image. Joint orientations are highlighted in white. Columns in the top half of the image are <30 cm across. (b) The entire wall of the gully, viewed from the gully floor. The lowest exposed columnar joints are vertical. This is abruptly succeeded by a zone of smaller, more irregular columnar joints that either vary in orientation on a scale of less than 5 m or are approximately horizontal. The upper portions of the cliff are less accessible to observation, but most jointing in the top half of the cooling unit is fine in scale (<30 cm spacing) and either columnar or irregular. 114 direction. Vesicularity is highest on the southeast margin of the flow, which thins rather than truncating abruptly. At the lowest end of the gully (south) is a narrow spire of lava ~50 m high and <5 m across, with prominent horizontal jointing; it is a primary cooling feature. LQvpba contains 2% plagioclase microphenocrysts, 1-3% olivine phenocrysts up to 3 mm in size, <1% clinopyroxene phenocrysts up to 1 mm in size, and <1% rounded quartz xenocrysts. The groundmass contains plagioclase and clinopyroxene. Discussion LQvpba erupted from a vent as shown on Figure 3.26 and flowed downslope into the gully to the south, impounding against ice along the length of the gully (Figure 3.28). At the south end, lava filled fractures or crevasses in the ice, though most of these spire-shaped cooling features have been removed by erosion. To the east, LQvpba spread out with a gentle profile and may have been mostly if not totally subaerial. This interpretation (impoundment of the west edge of the lava flow and not the east edge) is based on the relative profiles and degrees of erosion of the two margins of the flow. Had the eastern margin of LQvpba been erupted into tunnels in ice or against flanking ice, it would probably have been thicker, and it is likely that some of the resulting ice-contact features would have been preserved atop the gently sloping basement, given that small-scale, intensely jointed primary cooling features are preserved along the steeper western margin of LQvpba. However, exposure of the eastern margin of LQvpba is poor, so the style of eruption is not absolutely clear. The reason for the discrepancy in morphology between the east and west sides of 115 Figure 3.28. Three-dimensional view of Tricouni Southwest (LQvpba) from the southwest. Tricouni Southeast (LQvpadu) appears in the right of the image. Vertical exaggeration is 1.5x. The scale applies only to the foreground, since the image is in perspective. the LQvpba lava flow is probably the aspect of the slopes: the High Falls Creek gully, where impoundment occurred, runs north-south and is deep, so it would have retained ice longer than the south-facing broad open slope on the east side of LQvpba, where there is minimal evidence for impoundment. 3.3 GEOCHEMISTRY 3.3.1 Introduction Representative whole rock samples from the MCVF were analyzed for major, minor, trace, and rare earth (REE) elements, and these analyses were used primarily to discriminate different units and subunits. Data are tabulated in Appendix 3, and a description of analytical methods is given in Appendix 4. Unless otherwise noted, symbols in figures are larger than the estimated analytical precision. 3.3.2 M a j o r elements Analyzed magma compositions in the MCVF range from basaltic andesite to dacite (Figure 3.29) and fall on the calc-alkaline trend (Figure 3.30). Souther (1980) and Green et al. (1988) reported rhyodacite and rhyolite lava flows at Mount Cayley, and tephra units at Mount Cayley are probably rhyolite, but analyses of these are not available. MCVF samples fall in the centre of the compositional range for GVB rocks, which range from basalt to rhyolite. There is a reported compositional gap separating two groups of lavas having ~48-51% Si02 and 54-65% Si0 2 at Garibaldi Lake, Mount Meager, and Mount Cayley (Green 1977; Green and Watters 1977; Green 1981), but this 117 16 35 40 45 50 55 60 65 70 75 Si0 2 wt.% Figure 3.29 Total alkalies versus silica rock classification (after LeBas et al. 1986) for MCVF lava samples. The compositional range is less than that for Garibaldi Volcanic Belt lava samples as a whole. Eight data are from Green (unpublished data), one datum is from Mathews (1958), and the remainder were analyzed for this thesis project. FeO* wt.% Figure 3.30 A F M diagram for MCVF samples. Eight data points are from Green (unpublished data), one is from Mathews (1958), and the rest were analyzed for this thesis. Al l MCVF samples fall in the calc-alkaline trend, and this is typical for GVB lava samples. Tholeiitic-calc-alkaline classification is from Irvine and Baragar (1971). 119 is not apparent in analyses of MCVF samples; all of which have >53% SiC«2 . Figure 3.31 shows bivariate diagrams of selected major elements in MCVF samples, plotted against Differentiation Index (DI). Differentiation Index is calculated as the sum of normative nepheline + quartz + albite + orthoclase + kalsilite + leucite (Thornton and Turtle 1960). Individual MCVF centres show ranges of up to 6 wt.% SiC»2 (at Tricouni Southeast, whose samples also show the greatest variation in most other major elements). In general, SiC<2, MgO, and CaO correlate strongly with DI, while values for P 2 O 5 are low and do not correlate systematically with DI. No major element trends coincident with location (e.g. from north to south) or with probable age are apparent (although Tricouni Southwest and Ring Mountain Northwest, two of the youngest units, are compositionally similar and are the only basaltic andesites in the entire MCVF). The most differentiated samples analyzed are from Mount Fee. Although many samples possess hornblende phenocrysts with opaque rims (probably Fe-Ti oxides or pseudomorphing by opaques) and many are coloured red (presumably by oxidation), most iron in MCVF samples is in the ferrous state. This implies that the chemical effects of weathering have not been substantial. The effects of xenolith assimilation on MCVF lava samples are potentially problematic. Most contain rounded quartz xenocrysts34, most contain other crustal xenoliths in minor quantities, and some (the Slag Hill tuya, Ring Mountain, Little Ring 33 However, if any of the Cheakamus Valley basalts, which lie near the eastern boundary of the MCVF and whose source is unknown, were assigned to the MCVF, the gap would be present (the MCVF is defined as a distinct "field" based primarily on its clustering of centres in the topographically high area around Mount Cayley, rather than on some definite tectonic or volcanological distinction). 34 Such xenocrysts are also common in the Garibaldi Lake region (Green 1977). 120 3C 40 50 J-JJ 60 70 80 30 40 50 DI 60 70 80 50 J-JJ 60 50 m 60 Legend » Tricouni Southwest and Ring Mountain Northwest 4 Tricouni Southeast • Mount Fee • All other MCVF centres Figure 3.31 Major element abundances in MCVF samples, as oxides, plotted against Differentiation Index (DI) (Thornton and Turtle 1960). DI = the sum of the normative Ne + Qtz + Ab + Or + Ks + Let. (Mineral abbreviations are from Kretz [1983].) (a) Si0 2 versus DI. (b) MgO versus DI. (c) CaO versus DI. (d) Ti0 2 versus DI. (e) A1203 versus DI. (f) P 20 5 versus DI. 121 Mountain, and Tricouni Southeast) contain significant quantities of crustal xenoliths, predominantly granitic in composition, that were either derived from overlying and adjacent ice or incorporated from unconsolidated debris overridden by flows. The small size and partially digested nature of many xenoliths made it impossible to mechanically separate all xenolithic material from samples sent for analysis. Hence, a certain amount of contamination of lava samples is likely, especially at the four xenolith-rich centres, and this material is likely to be dominated by locally-derived rock fragments (either from wall rocks or from glacial till). If it is postulated that most xenoliths came from nearby bedrock units, contamination vectors can be calculated for MCVF samples. Figure 3.32 shows the directional effects of contamination by Late Jurassic-Early Cretaceous intrusive material (from nearby plutons) on MCVF samples. Figure 3.33 shows a map with the locations of the plutons that are probable xenolith sources. The most likely potential sources for contamination from bedrock are located upstream of the MCVF with respect to regional ice flow direction (the Whistler intrusion and the Callahan Creek intrusion). Contamination from these sources may explain the high alkali content of some Slag Hill lava samples relevant to adjacent samples. 3.3.3 Trace and rare-earth elements Figure 3.34 shows trace element compositions for MCVF samples, normalized to primitive mantle according to the method of Sun and McDonough (1989). In general, MCVF samples have high LILE/HFSE (large-ion lithophile elements / high field strength elements) and are strongly Nb-depleted, which is typical for subduction zone magmas (Gill 1981; Pearce 1983; Ryerson and Watson 1987; Hofmann 1988; Hawkesworth et al. 122 o + o to 16 14 12 10 8 6 4 -2 -35 foidite picro-basalt basalt basaltic andesite 40 4 5 5 0 55 60 65 7 0 75 SiO, Legend MCVF samples with significant macroscopic xenolith contamination MCVF samples without significant macroscopic xenolith contamination Ashlu Creek intrusion Cloudburst pluton Whistler intrusion Callaghan Creek intrusion Figure 3.32 Total alkalies versus silica rock classification (after LeBas et al. 1986) for M C V F lava samples, with the compositions of nearby plutons (potential sources of xenoliths) for comparison. The M C V F samples with the highest degree of obvious xenolith contamination are Ring Mountain, Little Ring Mountain, and the Slag Hi l l tuya (black circles). The Whistler intrusion and Callaghan Creek intrusion are the most likely source for xenolithic material, based on their locations and the probable flow direction of ice in the M C V F , though the Ashlu Creek intrusion and Cloudburst pluton, as well as more distant unidentified sources, are also possible. The arrows show possible contamination vectors for M C V F rocks, with the small arrows representing extremes and the large arrow representing the average (these were determined visually from the data). Contamination of M C V F lava samples by these bedrock sources wil l drive bulk rock compositions in the directions shown by the arrows. Eight of the M C V F data are from Green (unpublished data), one datum is from Mathews (1958), and the rest were analyzed for this thesis. Intrusive rock compositions are from Cui and Russell (1995), and the locations of these intrusions are shown in Figure 3.33. 0 10 20 30 Kilometres Figure 3.33 Locations of the intrusions used to determine contamination vectors for Figure 3.32, within a generalized terrane map and geological map of the southern Coast Belt (from Cui and Russell 1995 [after Wheeler et al. 1991; Journeay 1990; Journeay and Northcote 1992]). Locations of the intrusive samples used are: (7) Ashlu Creek intrusion, (8) Cloudburst pluton, (9) Whistler intrusion, (10) Callaghan ^ S Creek intrusion. The box shows the ^ Q ' ^ outline of the MCVF map. The Whistler and Callaghan Creek intrusions are the most likely contamination sources for most of the MCVF because they are located upstream of the MCVF with respect to regional ice flow direction. 124 1000 - i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — i — r Ember Ridge i i i i i i i i i i i i i i i i i i i i i i i i _ Cs Rb BaTh U Nb K LaCePb Pr Sr P NdZrSmEuTi Dy Y YbLu 1000 e n S o > £ • c a , o P 100 - 1 1 1 1 1 1 1 1 1 Kb) 1 1 1 1 1 1 1 1 1 1 I 1 1 ; Pali Dome East : Pali Dome West j \ / \ u i C ^ * ^ . — i i i i • t \ z V — w j Cs Rb BaTh U Nb K U C c Pb Pr Sr P Nd ZrSmEuTi Dy Y Yb Lu 1000 n—I—I—1—I—r~i—I—1—I—r -( c ) ~i—i—i—i—i—i—i—i—r Cauldron Dome 1 i i i i i i i — i — i i i i i i i i — i — i t — i i — i i — CsRbBaTh U Nb K U C e P b P r Sr PNdZrSmEuTi Dy Y YbLu 1000 § E 100 a > 'S 10 5 1—i—i—i—i—i—i—i—i—r—i—i—i—i—i—i—i—i—i—i—i—r ( d ) • Ring Mtn Ring Mtn Northwest-Little Ring Mtn •( I 1 1—I 1—I 1 1—I 1 1 1 1—i—1—! I 1—I 1 1 I L _ CsRbBaTh U Nb K LaCe Pb Pr Sr P NdZrSmEuTi Dy Y YbLu 1000 1 L _ J I I I I I I I I 1 I I I L _ l U _ l 1 I I I L_ Cs Rb BaTh U Nb K LaCc Pb Pr Sr P Nd ZrSmEuTi Dy Y Yb Lu g 100 t-Q. o o 10 t T — i — i — i — i — i — i — i — i — i — — i — i — i — i — i — i — i — i — r (f) — Tricouni Southwest - Tricouni Southeast •f 1—i— I—I—I—|—I—I— i—I—I—i—I—;—i—i—i—I—I—J—J—I—L_ CsRbBaTh U Nb K UCePb Pr Sr P NdZrSmEuTi Dy Y Yb U Figure 3.34 Selected trace element abundances in MCVF samples, normalized to primitive mantle (after Sun and McDonough 1989). (a) Ember Ridge, (b) Pali Dome East and Pali Dome West, (c) Cauldron Dome, (d) Ring Mountain, Ring Mountain Northwest, and Little Ring Mountain, (e) Slag Hill and the Slag Hill tuya. (f) Tricouni Southwest and Tricouni Southeast. 125 1993; Thirlwall et al. 1994). Ba/La is high, a characteristic of arc-derived rocks (Hildreth and Moorbath 1988). Of all MCVF centres, Ember Ridge (Figure 3.34a) shows the largest variation in trace elements (and also in rare-earth elements; Figure 3.35b) and this, coupled with petrography, is the reason the Ember Ridge masses were broken out as separate units, in spite of their sirnilarities in morphology and outcrop-scale characteristics, and their apparent similarities in age. Pali Dome and Cauldron Dome (Figure 3.34b, c) both show little within-centre-variation in trace elements, except for the LILEs Cs and Rb. Ring Mountain, Ring Mountain Northwest, and Little Ring Mountain (Figure 3.34d) are more variable. Slag Hill and the Slag Hill tuya are reasonably similar in trace element makeup (Figure 3.34e). Tricouni Southwest shows little within-flow variation, while Tricouni Southeast shows significant variation (Figure 3.34f). This reflects the fact that Tricouni Southeast formed by repeated discrete pulses of lava, whereas Tricouni Southwest is a single flow. Rare-earth element (REE) abundances in MCVF samples are shown in Figure 3.35, chondrite-normalized according to the method of Sun and McDonough (1989). A l l samples show enrichment of LREEs (light rare-earth elements) relative to HREEs (heavy rare-earth elements). None show europium anomalies. Most MCVF samples fall into two groups: Slag Hill, the Slag Hill tuya, and Little Ring Mountain constitute the northern group and show the most pronounced LREE enrichment (Figure 3.35a), while Cauldron Dome, Pali Dome, and Mount Fee samples constitute the central group, and have the least pronounced LREE enrichment (Figure 3.35b). The trends of the two groups are compared in Figure 3.35c. Ring Mountain is 126 La Ce Pr NdPmSmEu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr NdPmSm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu Figure 3.35 Chondrite-normalized rare earth element (REE) abundances for MCVF samples (after Sun and McDonough 1989). (a) Slag Hill, the Slag Hill tuya, and Little Ring Mountain, (b) Cauldron Dome, Pali Dome, and Mount Fee. (c) Slag Hill and Cauldron Dome, representing the northern and central REE trends, respectively, (d) Ring Mountain and Ring Mountain Northwest, (e) Ember Ridge, (f) Tricouni Southwest and Tricouni Southeast. 127 LREE-enriched and steeper in slope than samples from the central MCVF trend, while Ring Mountain Northwest has a slope similar to the central MCVF trend but is enriched relative to it (Figure 3.35d). Ember Ridge samples also do not follow either of the groups' trends, but show great variability in REEs (Figure 3.35e), just as they do in other trace elements (Figure 3.34a). Tricouni Southwest and Tricouni Southeast display similarity in some samples (Figure 3.35f), though in general, Tricouni Southwest is more similar to the northern trend and Tricouni Southeast is more similar to the central trend. 3.4 AGE Mount Cayley units are ordered according to the map made by Souther (1980). Two Mount Cayley samples yielded K-Ar dates of 2.7±0.7 Ma and 0.31±0.05 Ma (Green et al. 1988), indicating that Mount Cayley is Neogene to Quaternary in age, and the degree of erosion and lack of glaciovolcanic features support these dates. Mount Fee has not been dated but, like Mount Cayley, is very eroded and lacks glaciovolcanic features, therefore it is also inferred to be much older than most MCVF deposits. The remaining volcanic centres in the MCVF, all of which have evidence for glaciovolcanism, are significantly younger than Mount Cayley or Mount Fee. Samples from the Slag Hill main mass and lower Cauldron Dome were K-Ar dated (Green et al. 1988), and yielded results of 0.73±0.07 Ma, 0.60±0.03 Ma, and 0.25±0.03 Ma for Slag Hill, and 0.49±0.08 Ma for lower Cauldron Dome. However, these dates are inconsistent with the preservation of many glaciovolcanic features (e.g. fine-scale jointing). Although Green et al. (1988) stated that incorporation of excess radiogenic argon in these analyses was likely to have been minimal, because the K-Ar dates were supported by 1 4 C data and 128 stratigraphic relationships, recent shrinking of the glacier has made it clear that the Slag Hill samples dated at 0.6 and 0.25 Ma are from the same lava flow, Qvpauh (although the third Slag Hill date, 0.73 Ma, does come from LQvpauh, which is older). There are no 1 4 C dates available for the MCVF to provide additional evidence. Therefore, the possibility that these dates are inaccurate is not unreasonable, and I believe that the K-Ar dates for Slag Hill and lower Cauldron Dome are erroneously old and that most glaciovolcanic features in the MCVF date from the Fraser Glaciation or (rarely) the Salmon Springs Glaciation. The most likely reason for the K-Ar dates to be inaccurate is the analysis of samples containing microxenoliths of older material. Attempts were made to obtain whole-rock Ar-Ar dates for a suite of MCVF samples, but were unsuccessful due to low potassium content, sample youth, and other analytical difficulties. These absolute dating attempts are still ongoing. In the absence of absolute dates, stratigraphy may be used to date volcanic deposits relative to one another, but most MCVF volcanic deposits occur as discrete centres separated from each other by basement. Thus, MCVF centres are ordered based on (1) the elevation of ice impoundment features inferred to have formed during the waning of the Cordilleran ice sheet from 14,000-10,000 B.P. (Clague et al. 1980; Clague 1981; Ryder et al. 1991; Brooks and Friele 1992; Porter and Swanson 1998; Clague and James 2002), (2) the elevation of undenuded glaciovolcanic surfaces, with those at higher elevations assumed to have formed before those at lower elevations (since ice would have disappeared from high elevations before it disappeared from valleys)35, and (3) the overall degree of erosion 35 While low-altitude glacial erosion and glaciovolcanic features are likely to be due to interactions with the Cordilleran ice sheet during glacial periods, high-altitude features may be far more recent, because 129 (e.g. the extent of preservation of fine-scale joints, and the presence or absence of scouring or striations indicative of erosion by glaciers). they may be influenced by alpine ice. For example, a lava flow at Pali Dome West (LQvpau63) has been polished by ice, and this is clearly due to the action of a nearby tongue of ice that has retreated during the last forty years. 130 Part II. CHAPTER 4 Glaciovolcanic Features and Processes in the Mount Cayley Volcanic Field 4.1 INTRODUCTION Intermediate composition glaciovolcanic landforms in the MCVF all share one significant attribute: they lack features recording lava-water interaction. This suggests that MCVF subglacial eruptions are well-drained, an uncommon condition in most glaciovolcanic eruptions, though it is a condition implied by intermediate composition glaciovolcanic deposits in the rest of the GVB (Mathews 1951,1952,1958; Green 1977; Lawrence 1979; Lawrence et al. 1984; Green et al. 1988; Green 1994; Bye et al. 2000; Hickson 2000). Drainage of meltwater from subglacial volcanic vents depends principally upon topography, ice geometry, and ice characteristics, but also on magma composition. In the GVB, topography is extreme, ice geometries have varied greatly throughout Quaternary time, and many volcanic units are intermediate to silicic in composition; these factors explain the lack of evidence for lava-water interaction. Magmatic properties such as total erupted volume, viscosity, effusion rate, effusion style (endogenous versus exogenous), and volatile content influence subaerial eruptions and their products. It is likely that they also play roles in governing the nature of glaciovolcanic eruptions. In the GVB, this is evidenced by basaltic glaciovolcanic deposits that, unlike their intermediate composition counterparts, do show evidence for lava-water interaction (Mathews 1958; Green 1977; Lawrence 1979; Lawrence et al. 1984; Green et al. 1988; Green 1994; Hickson 2000). 131 In this chapter, I briefly summarize the landforms produced by intermediate composition glaciovolcanism in the MCVF. I then address the questions: (1) How does meltwater drain from MCVF subglacial volcanic vents? (2) What controls whether or not MCVF subglacial eruptions breach the ice surface? (3) What were the ice thicknesses over MCVF subglacial volcanic vents at the start of eruptions? (4) How do magmatic factors affect glaciovolcanic eruptions in the MCVF? I conclude with two qualitative models for intermediate composition glaciovolcanic eruptions in steep topography. The first model applies to subglacial to emergent eruptions that result in subglacial domes or flow-dominated tuyas, depending on whether or not the ice surface is breached. The second model applies to subaerial eruptions in which lava is impounded against ice. 4.2 PRODUCTS OF WELL-DRAINED INTERMEDIATE COMPOSITION GLACIOVOLCANISM 4.2.1 Introduction Landforms resulting from glaciovolcanism are summarized in Table 4.1. In the GVB, three landforms are indicative of intermediate composition glaciovolcanism: subglacial domes, flow-dominated tuyas, and impoundment features. Pillow lavas are absent in all three landform types, and hyaloclastite is very rare. Pillows and 132 Table 4.1 Summary of glaciovolcanic landforms and their attributes. Those that dominate the MCVF are in boldface. Landform1 Morphology Internal characteristics Present in the GVB? Vent-ice geometry Water influential during eruptions2 Water scarce during eruptions3 tuya flat-topped, steep-sided mountain pillow lavas overlain by (or mixed with) hyaloclastite, overlain by flat-lying lava flows yes subglacial then subaerial (but surrounded by ice) x subglacial mound conical mound pillow lavas overlain by hyaloclastite yes subglacial only x tindar (aka hyaloclastite ridge) steep-sided ridge hyaloclastite ± pillow lavas ± lava flows no subglacial ± subaerial (but surrounded by ice) x pillow sheet thick, areally extensive deposit pillow lavas no subglacial only X esker-like flow low, steep-sided ridge lava flows yes subglacial ± subaerial (but surrounded by ice) X remanent cliff after supraglacial eruption4 vertical cliff with large debris fan downslope lava flows (primary fragmental material is also possiible) yes supraglacial X subglacial dome steep-sided, dome-shaped mound lava flows yes subglacial X flow-dominated tuya flat-topped, steep-sided mound or mountain lava flows yes subglacial then subaerial (but surrounded by ice) X impoundment feature steep lava flow margin or terminus ± anomalous flow thickness lava flows (primary fragmental material is also possible) yes subaerial and flanked by ice X 'Repeated eruptions under changing ice distributions over long time periods result in hybrid features that have characteristics of more than one type of glaciovolcanic landform. The Slag Hill main mass, which experienced subaerial, subglacial, and impoundment eruptions, is an example of this. 2Lava-water interaction features are pillow lavas and hyaloclastite. 3Deposits without lava-water interaction features may comprise lava flows and/or fragmental units generated by primary magmatic explosivity. 'Remanent cliffs represent supraglacial eruptions. Since most material was deposited atop the ice, it is fragmented and removed by deglaciation. A debris fan will form proximally, and cliffs marking the point at which eruptions began depositing material atop the ice may be left distally. Unlike the cliffs of impoundment features, these remanent cliffs will not have jointing features indicating a vertical cooling surface and extremely rapid cooling, and the flow is less likely to be anomalously thick. hyaloclastite result from interactions between water and erupting lava and are abundant in basaltic subaqueous, submarine, and glaciovolcanic rocks. They also occur in intermediate composition volcanic rocks; examples include the Mount Read Volcanics in Tasmania, the Matsuzaki Volcanics, Green Tuff Belt, and Yoshida Formation in Japan, and the Nihotupu Formation in New Zealand (McPhie et al. 1993). Although columnar joints can form in subaqueous settings, most intermediate composition examples are associated with pillows or hyaloclastite (e.g. McPhie et al. 1993), and the many almost ubiquitously columnar jointed lava flows present at MCVF edifices have no pillows and have an almost total lack of hyaloclastite. Furthermore, Lescinsky and Fink (2000) suggested that well-developed columnar joints in lava flows that have interacted with ice represent cooling involving little or no water penetration of the volcanic pile. The absence of pillows and hyaloclastite in MCVF glaciovolcanic deposits cannot simply be ascribed to erosion, since there are glaciovolcanic deposits of similar age elsewhere in the GVB that contain significant proportions of pillows and hyaloclastite (Lawrence 1979; Lawrence et al. 1984; Hickson 2000; Hickson, pers. comm.). Additionally, there are no indications of significant glacial erosion at any of the MCVF units interpreted as having erupted during the Fraser Glaciation (e.g. glacial striations), and there are fine-scale near-primary cooling features preserved at many edifices; these 36 Hyaloclastite is defined as "all vitroclastic tephra produced by the interaction of water and hot magma or lava" (Fisher and Schmincke 1984). It presence is universally taken as an indication that water was present during eruption. However, usage of the term implies no specific fragmentation mechanism (e.g. cooling-contraction). Pertinent literature discussions on water-related lava fragmentation mechanisms and their products are extensive (e.g. Sigvaldason 1968; Peckover et al. 1973; Honnorez and Kirst 1975; Fisher and Schmincke 1984; Yamagishi and Dimroth 1985; Kokelaar 1986; Smellie et al. 1993; Batiza and White 2000; White and Houghton 2000; Skilling 2002). 134 features would not have been preserved if erosion was so efficient as to remove all hyaloclastite. Although it is probable that small quantities of hyaloclastite were buried by later lava flows, it is unlikely that much hyaloclastite was ever present. The lack of pillows and hyaloclastite in the MCVF is a primary attribute of the deposits. This lack indicates that most glaciovolcanic eruptions took place in environments without standing water. Exceptions within the GVB are mainly basaltic. 4.2.2 Subglacial domes Subglacial domes are steep-sided lava masses with rounded or irregular upper surfaces; they are commonly dome-shaped (Kelman et al. 2002 a,b) (Figure 4.1) . They represent subglacial eruptions that did not breach the ice surface during formation. Eruptions involved almost no direct lava-water contact, as evidenced by the absence of pillows and the almost total absence of hyaloclastite. Diagnostic features are: • edifice shapes that are inconsistent with the apparent paleotopography, • joint orientation patterns that are inconsistent with the apparent paleotopography, • abundant fine-scale (<25 cm) columnar joints, indicating rapid cooling (e.g. Peck and Minakami 1968; Ryan and Sammis 1978; Budkewitsch and Robin 1994; Long and Wood 1986; DeGraff and Aydin 1987; Reiter et al. 1987; Aydin and DeGraff 1988; DeGraff et 37 In volcanology, the term dome is commonly used to describe high viscosity masses of intermediate to rhyolitic lava erupted effusively as dome-shaped features at or near a vent, which may grow passively for long periods but also display intermittent explosivity (Newhall and Melson 1983). Such domes commonly form at volcanoes subsequent to or between major explosive eruptions (Mount St. Helens is one example). However, the word dome is not used here in the same sense, and is not intended to imply anything about long-term volcanic behaviour or anything not included in the definition above; it was chosen to indicate overall morphology. 135 6 0 0 m Figure 4.1 Atypical subglacial dome, Ember Ridge Northwest. The sketch indicates the general orientation and relative spacing of columnar joints (long lines) and irregular joints (short lines), though the spacing is not to scale. Joint orientations along the top ridge vary on a scale too small to show in this image but are commonly perpendicular to the surface of the ridge or form locally radiating arrangements. Numbers indicate approximate columnar joint diameter in cm. There is no hyaloclastite. 136 al. 1989; DeGraff and Aydin 1993; Grossenbacher and McDuffie 1995), • pronounced changes in joint orientations over distances of <5 m, and/or radially-oriented masses of joints near flow margins, and • a scarcity or lack of hyaloclastite. Most examples of subglacial domes also contain: • other fine-scale joints (planar, irregular, or pseudopillow). Al l the subglacial dome samples studied have groundmasses in which microlites rather than glass dominate, in spite of having cooled in proximity to ice. No pillow lavas have been identified at any subglacial dome. Most of the features that indicate lava-ice interaction (e.g. jointing) are likely to be rapidly removed by erosion. The best examples of subglacial domes within the MCVF are the andesite masses of Ember Ridge, each of which is less than 0.04 km 3 in volume and is up to 150 m thick. Other examples within the GVB are basaltic andesite to dacite lava flows at Round Mountain, Glacier Pikes, and the Eenostuck Mass (Figure 4.2; Mathews 1958; Green 1977; Green et al. 1988). Examples outside the GVB include rhyodacite and rhyolite domes at South Sister and North Sister in the United States Cascades (Lescinsky and Fink 2000). Features that may be subglacial andesite domes have also been identified at Volcan Sollipulli in Chile (Gilbert et al. 1996). The lava mass that began growing during the 2004-2005 eruption at Mount St. Helens is also arguably a subglacial dome, although it has not yet been described in detail. 137 • Legend Quaternary centres with evidence for glaciovolcanism Quaternary centres without documented evidence for glaciovolcanism 0 k m 20 I I Figure 4.2 Quaternary volcanic deposits in the Garibaldi Volcanic Belt (GVB). Glaciovolcanic landforms of the GVB discussed in Chapter 4, excluding centres within the MCVF, are labeled: 1. Salal Glacier volcanic complex, 2. Barrier flow, 3. Culliton Creek flow, 4. Table, 5. Glacier Pikes, 6. Eenostuck Mass, and 7. Round Mountain. Location of this map with respect to the rest of British Columbia is indicated on Figure 1.1. The distribution of volcanic rocks from outside the MCVF is based on Mathews (1958), Green (1977), Lawrence (1979), Lawrence et al. (1984), Roddick and Souther (1987), Green et al. (1988), Green (1994), Hickson et al. (1999), Bye et al (2000), Hickson (2000), and Stewart et al. (2002). I I (N) B 50°35' 138 Subglacial domes differ in several important respects from subaerial domes. First, subaerial domes typically have extremely rough surfaces on the decimeter to meter scale, strewn with blocks that may be greater than 5 m across (Fink and Anderson 2000), whereas subglacial domes have minimal debris on their upper surfaces, with most talus correlating in size to the spacing of joints (<30-40 cm). Subaerial domes also possess surface features such as compressional ridges that indicate the lava was extremely viscous, and surface fractures and textures that result from brittle deformation at a variety of scales (Fink and Anderson 2000). Subglacial domes lack these features, and have morphologies that appear to have been influenced more by the surrounding ice than by physical properties of the lava. Subaerial domes are commonly crystal-rich (e.g. Cas and Wright 1987), whereas subglacial domes within the GVB have low phenocryst contents coupled with high groundmass microlite contents (it is not clear whether this is common to subglacial domes elsewhere). Additionally, many subaerial domes have vesicular carapaces underlain by dense interiors, whereas subglacial domes have low overall vesicularities and no apparent zoning or layering of vesicles. Especially significant is the fact that although subaerial domes may display a range of jointing types, there is little literature documenting well-developed columnar jointing at subaerial dome margins, except in cases where contact with ice is suspected (Lescinsky and Fink 2000). Subglacial domes, however, have intense, ubiquitous fine-scale jointing that is most commonly columnar, and peripheral columnar joints have orientations that indicate steep or vertical cooling surfaces. Finally, subaerial domes commonly show evidence for being part of longer-lived cycles of effusive and explosive eruption (e.g. Rose 1972; Cas and Wright 1987; Nakada et al. 1999; Barmin et al. 2002; Sparks and Young 2002), whereas 139 subglacial domes are not associated with pyroclastic rocks. 4.2.3 Flow-dominated tuyas Flow-dominated tuyas are flat-topped, steep-sided volcanic edifices that lack the internal stratigraphy found in classical basaltic tuyas (e.g. Mathews 1947; Jones 1966, 1969; Werner et al. 1996; Hickson 2000) (Figure 4.3). They result from subglacial eruptions that ultimately breached the ice surface, with late-stage eruptions remaining bounded by ice, allowing lava flows to stack up. Flow-dominated tuyas probably represent a continuation of the eruptive process that produces subglacial domes. These deposits record eruptions involving almost no direct lava-water contact, as evidenced by the scarcity of pillows and hyaloclastite. Diagnostic features are • edifice shapes that are inconsistent with the apparent paleotopgraphy, • joint patterns that are inconsistent with the apparent paleotopography, • stacked, primarily flat-lying lava flows, and • a scarcity or absence of hyaloclastite. Flow-dominated tuyas also commonly include: •fine-scale (<25 cm) columnar joints at the lateral margins, indicating rapid cooling, • pronounced changes in joint orientations over distances of <5 m, and/or radially-oriented masses of joints near flow margins, and 140 (a) Figure 4.3 Typical flow-dominated tuyas. (a) Little Ring Mountain, south side (highlighted below). Individual flow units are difficult to distinguish in this image due to distance and lighting, but the overall shape is apparent. The x indicates the upper vertical-sided portion of Little Ring Mountain, which comprises at least two lava flows. The y indicates a mixture of outcrop and scree. The area marked with z consists entirely o f talus, (b) Ring Mountain, north side. Sufficient erosion has occurred that several flow units are distinguishable in the vertical cliff. Relative spacing and orientations of joints are indicated schematically. Columnar joints are shown with long lines, and irregular joints are shown with short lines. Numbers indicate diameters of columnar joints or spacing between irregular joints, in cm. Note that the dips in flows appear far steeper than they actually are because the photo is taken from a position below them. •other fine-scale joints (planar, irregular, or pseudopillow). As with subglacial domes, samples from flow-dominated tuyas have groundmasses in which microlites rather than glass dominate. No pillow lavas have been identified at any flow-dominated tuya. Most of the features that indicate lava-ice interaction (e.g. jointing) are likely to be rapidly removed by erosion. Examples of flow-dominated tuyas within the MCVF are Ring Mountain, Little Ring Mountain, and the Slag Hill tuya. Al l are andesitic. They range in volume from 0.009 to 0.6 km 3 and in height from 90 to 400 m. An example from the rest of the GVB is the Table (Figure 4.2). I identified a single example of a flow-dominated tuya from outside the GVB, Fifty-two Ridge in the Wells Grey-Clearwater volcanic field (Hickson 1987; Hickson et al. 1995)38. It is the only known basaltic flow-dominated tuya. Flow-dominated tuyas share both similarities and differences with rhyolitic tuyas. Detailed investigations of the rhyolitic tuya at southeast RauSafossafjoll, Torfajokull, Iceland (Tuffen et al. 2002a), show many cases where lava flows take the shape of surrounding ice, displaying jointing patterns similar to those seen in ice-contact flows in the GVB. These rhyolitic subglacial or ice-contact lava flows commonly show no evidence for the presence of water during eruption, although some of them have perlitic alteration that suggests interaction with lava while the lava was still hot. However, unlike flow-dominated tuyas in the GVB, the Rau5afossafj6ll tuya has evidence for vigorous 38 Fifty-two Ridge has a higher proportion of hyaloclastite and a lower aspect ratio than other flow-dominated tuyas (it is pancake-shaped) but is still dominated by flows, the uppermost of which were subaerial. Its morphology, unusual for a basaltic subglacial volcano, is attributed to high vent elevation and thin overlying ice (Hickson 1987; Hickson et al. 1995). 142 explosive subglacial eruptions that were probably driven by magma-water interaction rather than by degassing of magmatic volatiles; although the role of water appears to be diminished in these rhyolitic eruptions (relative to "classic" basaltic subglacial erupions), there is more extensive lava-water interaction than at the flow-dominated tuyas of the GVB. 4.2.4 Impoundment features Impoundment features form when lava erupts from a subaerial vent and flows downslope until it encounters ice or when lava flows along open channels in ice (Mathews 1952; Lescinsky and Sisson 1998; Lescinsky and Fink 2000) (Figure 4.4). Diagnostic features of ice-impounded lava flows are: • unusually thick, vertically-faced flow termini or flanks that are inconsistent with the apparent paleotopography, and • abundant fine-scale (<25 cm) columnar joints, indicating rapid cooling. Impoundment features also commonly include: • joint orientation patterns that are inconsistent with the apparent paleotopography, • pronounced changes in joint orientations over distances of <5 m, and/or radially-oriented masses of joints near the margins of flows, • other fine-scale joints (planar or irregular), and • local irregularities in primary flow morphologies, such as spires and ridges up to -20 m 143 Figure 4.4 A typical impoundment feature, at Pali Dome East, (a) The upper surface of the flow. Note that the photograph is taken looking towards the margin of the flow (towards the valley), and the box on (b) is indicated merely to show location. Most of the columnar joints in the image are oriented directly towards the viewer, hence appear as a series of polygonal forms. Joints commonly vary in orientation over distances of <1 m. Patches of irregular joints are present. Numbers indicate approximate columnar joint diameters, in cm. Samples are glassier than those at the base of the flow, (b) The entire margin of the flow; upslope of the cliff, the profile of the flow is far less steep, (c) The lowest exposed part of the flow. Note that the smaller joints above the coarse joints are much further from the viewer, hence, the transition from coarse to fine joints is not as sharp as it appears; there is an intermediate zone not visible in the photograph. Below the coarse joints is an extensive apron of glassy scree. 144 in scale. As with other glaciovolcanic landforms in the GVB, samples from impoundment features have groundmasses in which microlites rather than glass dominate. There are no examples of pillows or hyaloclastite at any MCVF impoundment feature, but damming of meltwater by ice or glacial sediments is a possibility. There is evidence for ponded water adjacent to the smaller mass at Tricouni Southeast; the apparent paleotopography cannot explain this accumulation, suggesting that ice or a short-lived dam of glacial sediment was responsible. Most of the features that indicate lava-ice interaction (e.g. jointing) are likely to be rapidly removed by erosion. Examples of ice-impounded lava flows within the MCVF are the Tricouni Southwest flow, the two oldest flows at Pali Dome West (Lvpau6i and LQvpau62), the youngest flow at Pali Dome East (LQvpaul 4), lower Cauldron Dome, and one of the Slag Hill subunits (Qvpauh). Examples from elsewhere in the GVB (Figure 4.2) include the Barrier and Culliton Creek flows (Mathews 1952) and cliffs at the Salal Glacier volcanic complex (Lawrence 1979; Lawrence et al. 1984). An ice-impounded "sintered ash-flow tuff' occurs at the Salal Glacier volcanic complex (Lawrence 1979; Lawrence et al. 1984). Outside the GVB but within North America, impoundment features have been recognized at the Wells Grey-Clearwater volcanic field (Hickson et al. 1995; Hickson 2000), Llangorse volcanic field (Harder et al. 2003), and Hoodoo Mountain (Kerr 1948; Souther 1991; Edwards 1997; Edwards et al. 2002), as well as in the United States Cascade Range (Lescinsky and Sisson 1998; Lescinsky and Fink 2000), and the Aleutian Islands (Byers et al. 1947; Robinson 1947). Outside North America, examples include 145 flows at Volcan Sollipulli in Chile (Gilbert et al. 1996) and the McMurdo Volcanic Group in Antarctica (Wright 1980). 4.3 DRAINAGE AT MCVF SUBGLACIAL VOLCANIC VENTS 4.3.1 Introduction Based on the lack of pillows and hyaloclastite, volcanoes in the MCVF erupt under and against ice but not into standing water. This fact implies that all meltwater is efficiently drained. In order to remove water from a subglacial volcanic vent, physical pathways must be available and hydraulic conditions must direct the water away from the vent (Figure 4.5). 4.3.2 Drainage pathways Water can drain from subglacial volcanic vents in a variety of ways: flow through permeable ice, flow through permeable substrate (Kiver and Steele 1975; Tuffen et al. 2001, basal sheetflow (Smellie et al. 1993), basal flow involving hydraulic lifting of the glacier (Gu3mundsson et al. 1997; Smellie 2002), flow through englacial or subglacial veins or tunnels (Bjornsson 1975; Nye 1976), and supraglacial flow (Smellie 2002). In the MCVF, the two pathways whereby subglacial meltwater may potentially drain efficiently are highly permeable ice and englacial or subglacial tunnels39. Glacier ice is vertically zoned in terms of structure and permeability (Figure 4.6). 39 Escape of meltwater through floating of the glacier could remove a large amount of water from a subglacial vent, but would require an initial accumulation of water, and there is no evidence for this at any MCVF centre; the topography does not provide any barriers behind which water could accumulate. 146 (a) A re drainage N O pathways available? ^ Y E S (b) A re pathways fi l led with water only, or water + ai r? (c) A I R + W A T E R 1 W A T E R - — — , N O (d) Is P >PB? 1 (f) Does the ice slope towards the vent? Does topography slope away from the vent? N O I S m (g) Can the bedrock slope away from the vent overcome ice surface slope effects? N O Y E S water accumulates Figure 4.5 Key | bedrock • ice H vent cavity air 147 Figure 4.5 Factors determining whether water accumulates or drains at MCVF subglacial volcanic vents. Pc = cavity pressure at the vent. Pg = glaciostatic pressure (pressure resulting from the weight of the overlying ice; Pg - pgh). Red lines indicate conditions that will result in the accumulation of water at subglacial vents. Green lines indicate conditions that will promote drainage. At MCVF vents, water may behave differently at different stages of eruption (e.g. early during eruptions, when meltwater discharge is high, subglacial pathways will be water-filled, whereas after meltwater discharge declines, pathways may be filled with a mixture of water and air). (a) Drainage pathways include subglacial/englacial tunnels, permeable ice, and permeable substrate. (Hydraulic lifting of the glacier and supraglacial flow require specific circumstances and are not considered on this diagram.) Most water is drained by subglacial/englacial tunnels. (b) Water-filled subglacial tunnels may be at various pressures, and water in them moves according to the hydraulic gradient within the ice. (c) If subglacial tunnels are partially air-filled, they are likely to be at atmospheric pressure (0.1 MPa) (Cutler 1998). In this case, water drains according to topography (Scenario 1 from the text). (d) Pc can exceed Pg in subglacial eruptions because of the density differences between lava, ice, and water. This could potentially lead to floating of the glacier, although the location of MCVF vents on bedrock highs would diminish the likelihood of this occurring.. (e) When Pc>Pg, water will flow towards regions of lower pressure (away from the venf)(Scenario 3 in the text). (f) If Pc<Pg, a depression forms in the ice above the vent, and water is likely to flow towards it. If the ice surface slopes away from the vent, water will drain (since at all MCVF vents, the bedrock slopes away from the vent). (g) A bedrock slope (away from the vent) of eleven times that of the glacier surface slope is required to overcome the effect of ice sloping towards the vent (Nye 1976), but steep enough bedrock may still make drainage of meltwater possible (Scenario 2 in the text). 148 glacier surface E s-a a E s-Bb E t © firn 430-830 kg/m3 V i crevasse layer no crevasses 830-917 kg/m3 •tured ice c c 0 \ — p unfra< < ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ n 3 65 cr ST 3 •q 3 re 6S re" glacier undergoing extension glacier under neutral tension or compression Figure 4.6 Section through an idealized glacier in the MCVF, showing the thickness of the firn layer, the crevassed layer, and the unfractured ice. The total thickness of ice is based on the known regional maximum elevation of the Cordilleran ice sheet and the elevations of glaciovolcanic deposits. Not to scale. Modified from Smellie (2001). In a very thin glacier, the entire thickness may be permeable, allowing water to drain continuously, whereas as ice thickness increases, there is a greater potential for trapping water at the vent (Smellie et al. 1993; Smellie and Skilling 1994; Smellie 2001). 149 A layer of snow overlies firn which overlies ice. The upper thirty meters of this ice can behave brittly, a behaviour that supports crevasse formation. Firn and snow are permeable, whereas uncrevassed glacier ice below the pressure melting point is not. This means that lower portions of thick glaciers will function as aquicludes and will not transmit water unless subglacial or englacial conduits are present. In order for leakage of water through permeable portions of a glacier to play a major role in glaciovolcanic drainage, the glacier must be less than 150 m thick (Smellie and Skilling 1994; Smellie 2001, 2002). For subglacial dome eruptions, this is implausible, because eruptions commencing under such thin ice would breach the ice surface40. Hence, leakage of water through permeable portions of glaciers could not have been the primary route whereby water was removed from vents in the MCVF. Even in cases where ice attained a thickness of <150 m above the vent during eruption, it is unclear whether permeable ice would be able to remove water from vents rapidly enough to ensure that eruptions remained dry. Subglacial or englacial conduits were probably the main route by which water was drained. There is no difficulty in initiating development of a subglacial drainage system at MCVF vents. Subglacial cavities may be generated by ice passing over bedrock obstacles41, contact with flowing or heated water, contact with steam, contact with heated eruption materials, or conduction of geothermal heat from bedrock. Since the Cordilleran ice sheet was warm-based (e.g. Ryder et al. 1991) and thus at the pressure melting point, 40 This is demonstrated in the calculation in 4.4.2. 41 Ice melts due to high pressure on the upstream side, refreezing takes place on the lower pressure downstream side (Weertman 1957; Lliboutry 1958 a, b), and the small cavities thus formed may be interconnected (Walder and Hallet 1979). 150 even a small increase in temperature at its base could bring about melting. However, it is tunnels, not cavities, that are needed to carry away the water. The most likely mechanism whereby these small cavities could evolve into tunnels is by downslope melting due to the circulation of heated meltwater. This is plausible because even in non-volcanic settings, melting rates observed for downward-sloping passages are commonly high (Shreve 1972; Rothlisberger 1972; Lliboutry 1983; Cutler 1998). At Mount Rainier, Washington, fumarolic heat alone has been shown to result in the development of a long-lived drainage system with many downward-sloping passages (Kiver and Steele 1975). 4.3.3 Hydraulic potential Introduction Water did not accumulate at MCVF subglacial vents. There are three possible scenarios that could explain this (Figure 4.5): (1) Drainage conduits and vent cavities were at atmospheric pressure and partially filled with air. Such conditions, when coupled with topography sloping away from vents, would support efficient drainage of meltwater during eruptions (Figure 4.5c)42. (2) Drainage conduits and vent cavities were filled with water at less than or equal to glaciostatic pressure, but the topography sloped away from vents steeply enough that drainage was still possible (Figure 4.5g). 42 It is generally assumed that tunnels partially filled with water are at atmospheric pressure (Cutler 1998) and, in this situation, hydraulic potential depends only on elevation. 151 (3) Drainage conduits and vent cavities were fdled with water at pressures greater than glaciostatic (Figure 4.5e), forcing water away from vents. Scenario 1 Conditions whereby vent cavities reach atmospheric pressure (0.1 MPa) obviously apply to flow-dominated tuyas at some point during eruptions. Such conditions probably also apply to the later stages of eruptions that do not breach the ice, either because of thinning of ice over the vent or because drainage pathways equilibrate with atmospheric pressure at a distance from the vent. Evidence for the former includes lava bodies at Ember Ridge Southeast whose shapes and joint characteristics imply they filled steep-sided, narrow cavities that were probably fractures in the ice4 3; since only the upper 150 m of a glacier is able to deform brittly, the presence of lava flows that take the shapes of crevasses implies that the glacier was <150 m thick immediately over the vent. Scenario 2 Drainage via low pressure water-filled conduits due to steep topography depends upon the relative slopes of the glacier surface and underlying bedrock, because as ice gets thinner, glaciostatic pressure is reduced, so water flows from regions of thick ice to regions of thin ice (Shreve 1972; Bjornsson 1988). In observed subglacial eruptions, a 4 3 These isolated lava screens should be distinguished from lava screens that are surrounded by breccia; the latter could have formed within breccia-filled cavities, whereas the former must have cooled against ice, unless they were once surrounded by breccia that later eroded totally (a situation that seems improbable, given that many breccia samples are well-indurated, and at least one of these lava screens occurs more than 150 m away from the nearest breccia outcrop). 152 depression forms in the ice over a subglacial vent (GuSmundsson et al. 1997; Jonsson et al. 1998; Alsdorf and Smith 1999), promoting the flow of water towards the vent. Only if bedrock topography is extreme is it possible for it to overcome the effects of glacier surface slope (Bjorasson 1988; Nye 1976; Gudmundsson et al. 1997); a bedrock slope eleven times that of the glacier surface slope is required (Nye 1976). This scenario is difficult to evaluate without knowing the paleo-ice thickness and geometry, but the fact that MCVF vents are at high altitude, with bedrock sloping steeply away from them, makes it a possibility. Scenario 3 If vent cavities and drainage conduits near a vent are filled with water at pressures greater than glaciostatic, the glacier may be floated (in which case water would drain along the base) or water will be forced away from the vent via all available pathways. Such a situation could arise due to eruption-related volume changes, because a specific volume of ice is replaced with the water melted from it plus some quantity of lava (Figure 4.5e)44. This positive pressure situation will promote drainage that is as rapid as pathways can accommodate. Addressing this scenario, I have calculated the cavity pressures that could theoretically be produced during subglacial andesite eruptions, using the method of Hoskuldsson and Sparks (1997). Volumes of ice melted are determined by the calculations in 4.4.2. Table 4.2 lists all symbols and subscripts used. Table 4.3 lists all 44 The concept of positive pressure at subglacial cavities resulting from density differences betweeen magma, water, and ice was first discussed in detail (for subglacial rhyolite eruptions) by Fteskuldsson and Sparks (1997). 153 Table 4.2 Symbols and subscripts used in calculations. symbol explanation units Q thermal energy J T temperature °C AT temperature change/difference °c V volume m 3 AV volume change m P pressure MPa m mass kg r density kgm' 3 c heat capacity J k g 1 0 C L latent heat of fusion J k g 1 0 C h ice thickness m a area m b bulk modulus for water GPa subscript explanation m magma i ice w water e eruption f final g glaciostatic c cavity 154 Table 4.3 Constants used in calculations. constant value source n 917 kg m"3 Lf 3.35 x 105 Jkg"1 Hoskuldsson and Sparks 1997 Lm 2.09 x 105 Jkg"1 Hoskuldsson and Sparks 1997 Cw 4195.7 Jkg"1°C"1 Hoskuldsson and Sparks 1997 Cm 1130 Jkg"1 °C"1 calculated using data from Neuville et al. 1993 b 2.05 GPa Blake 1981 g 9.81 m s* T, 0°C constants used. Cavity pressure based on volume changes is calculated with: AV Pc~Pg w " Since AV - V w + V m - V;, and P g = pgh, we can rearrange to solve for P c b+Pigh, (2) Calculated cavity pressures for MCVF volcanoes are strongly positive and are shown in Figure 4.7. However, the calculation assumes that all water is retained at the vent, hence does not reflect the probable conditions during most stages of MCVF eruptions. Furthermore, such overpressures would never exist in a real system because high pressure and elevated water temperatures would promote formation of drainage conduits, and ice would deform in response to significant overpressures45. Pressure evolution at subglacial vents At the onset of subglacial volcanism, we can expect the glacier to be heated and melted. Before the first eruption of lava, subglacial cavities will be filled with water at pressures lower than glaciostatic (Pg) due to the density differences between water and 45 HOskuldsson and Sparks (1997) estimated that cavity overpressures of tens of MPa or higher would not be achieved at subglacial volcanoes. 156 1000 -4 KS 1 10 Legend ice thickness = 100 tn ice thickness 900 m 400 500 600 700 Te - T f for magma (°C) 800 900 1000 Figure 4.7 Theoretical cavity pressure versus temperature change of magma (eruption temperature minus final temperature, Te - Tf). The calculation assumes that effective pressure depends on glaciostatic pressure and any volume changes due to the differing densities of magma, water, and ice. It is also assumed that all water remains at the vent. ice. This condition promotes temporary retention of water and explains the small quantities of hyaloclastite that occur at some subglacial domes. Because such hyaloclastite units are formed first, they are likely to be buried by later deposits. However, the introduction of magma should make P c rise to greater than P g, as drainage systems are not yet fully developed and are unable to deal with the high meltwater discharge rates (Scenario 3). Once drainage pathways are sufficiently well-developed to accomodate meltwater discharge rates, P c should peak. The subsequent drop in meltwater discharge rates will lead to a drop in pressure in subglacial tunnels. This occurs because, while conduits can generally adjust to changes in temperature regime by melting over a time scale of hours, contraction of passages due to ice deformation takes days or weeks (Benn and Evans 1998), with the rates of conduit closure depending on the thickness of the ice and the water pressure in the conduits. A situation in which cavity pressure is less than or equal to glaciostatic and drainage pathways are water-filled, yet steep topography still promotes drainage (Scenario 2), may occur. Eventually, the vent and/or drainage system makes contact with the atmosphere. After this occurs, the vent cavity and drainage conduits will be partially filled with air and at atmospheric pressure. Because the topography slopes away from the vents, water will continue to drain (Scenario 1). This evolution of pressure in glaciovolcanic cavities is analogous to the evolution of pressure in nonvolcanic subglacial settings, where water pressure in subglacial tunnels varies with discharge. Early in the ablation season, discharge is low and water pressure is high because tunnels are expanding to catch up with drainage, whereas in late summer, 158 discharge and runnel size peak, and water pressure declines thereafter, resulting in shrinking tunnels that are only partially filled with water (Cutler 1998). Eruption of most MCVF lava flows after subglacial tunnel size and meltwater output have peaked and tunnels are partially filled with water and air at atmospheric pressure would explain why there is so little evidence for lava-water interaction. 4.4 ABILITY OF ERUPTIONS TO BREACH THE ICE SURFACE 4.4.1 Introduction Subglacial eruptions that ultimately breach the ice surface and continue to erupt through open chimneys in the ice produce different landforms (flow-dominated tuyas or tuyas46) than eruptions that never breach the ice (subglacial domes or subglacial mounds47). These ice-breaching landforms provide precise information (within tens of meters) about the elevation of the ice sheet surface at the time of eruption. Their presence also indicates that heat was lost to the atmosphere (rather than to ice or meltwater), that final eruptions occurred at atmospheric pressure, and that supraglacial meltwater flow was possible. Thus, the assumption (based on field observations) that flow-dominated tuyas breached the ice but subglacial domes did not is a crucial one. It can be quantitatively investigated. 46 The word tuya was first used by Mathews (1947) to describe flat-topped, steep-sided volcanoes typically comprising pillows overlain by hyaloclastite overlain by lava flows. Most examples are basaltic (Mathews 1947; Jones 1966, 1969). 47 The term subglacial mound was coined by Hickson (1987) to describe subglacial volcanoes that comprise steep-sided, cone-shaped piles of hyaloclastite, with or without pillow lavas at the base. A subglacial mound is in effect a tuya without the cap of lava flows, and represents a smaller eruptive volume relative to the thickness of overlying ice. 159 4.4.2 Quantities of ice melted I addressed the question of how much ice is melted during MCVF subglacial eruptions by calculating the quantities of heat released by the cooling of andesite lava and the concomitant quantities of ice melted48. If we calculate the ice volumes that can be melted by specific volumes of cooling lava and assume ice was melted only directly above vents, we can derive ice thicknesses that were melted. These calculated thicknesses are minimum ice thicknesses which must have overlain the vents at the commencement of eruption, since the ice surface was not breached. The maximum modern elevation attained in the central GVB by the Cordilleran ice sheet, as indicated by glaciovolcanic features and glacial erosion features, is approximately 2300 m (Clague 1989; Ryder et al. 1991). This means that the high altitudes of vents in the MCVF minimize the quantities of ice available for melting. Assuming that post-glacial isostatic rebound at all points in the MCVF was equal, a comparison of the modern elevations of glaciovolcanic features with the 2300 m ice sheet elevation maximum provides constraints on the maximum ice thicknesses that could have overlain subglacial domes and flow-dominated tuyas at the time of eruption. These constraints range from 300-900 m (Figure 4.8). They can be compared to the ice thicknesses that were melted through (determined by the following calculation). Table 4.2 lists all symbols and subscripts used. Table 4.3 lists all constants used. 48 The volumes of lava, ice, and water calculated here are used in the cavity pressure calculations of 4.3.3. 160 123°2V00 50°16'30" 50°15'00" so'is'so" 50°12'00" 50°10'30" 50 o09'00" 50°07'30" 50°06'00" SOWSO" 50°03'00" 50°01'30" 50°00'00" 49°58'30" Figure 4.8 Maximum possible ice thicknesses to have overlain individual units within the M C V F , based on a comparison between their modern elevations and the maximum elevation reached by the Cordilleran ice sheet (-2300 m). It is assumed that the relative elevations between points on the land surface of the M C V F has not changed since the Fraser Glaciation maximum. 300 m 600 m 900 m 1200 m 1500 m 1800 m >1800m glaciovolcanic centres (tP non-glaciovolcanic centres o Ti km 50°06'00" 50°04'30" 50°03'00" 50°01'30" 50°00'00" 49 o58'30" 49°57'00" 123°21'00 123°18'00 123"15'00" 123"12'00 123*091)0 123°06'00 161 The following assumptions are also made: (1) The magma is andesite. The consideration of composition is crucial, because there are differences in the heat budgets attending the eruption of mafic and felsic magmas. These arise because of their differences in eruption temperature (Te), and the magnitude of the gap between T e and the calorimetrically defined glass transition temperature (Tg) (e.g. Allen 1980; H6skuldsson and Sparks 1997; Kelman et al. 2001; Smellie 2002; Tuffen et al. 2002b). The value of T g marks the transition from the melt to the glassy state, and is an important limiting value for magmatic processes because above it, rates of nucleation, crystallization, and vesiculation can compete with most eruptive time scales, whereas below T g, they may be suppressed, eliminating potential latent heats of crystallization (L) (Figure 4.9)49. This means that a dacite or andesite will release less heat over its cooling history than a basalt (Figure 4.9). The fact that glaciovolcanic deposits with significant evidence for lava-water interaction are present in the GVB, but only in rocks of basaltic composition (Mathews 1958; Lawrence 1979; Lawrence et al. 1984; Hickson 2000), implies that composition plays a role in determining whether water accumulates or drains from subglacial vents. (2) The magma has an eruption temperature (Te) of 1000°C. Using MELTS (a model for the computation of multi-component phase equilibria in silicate melts; Ghiorso and Sack 1995), I calculated a liquidus temperature of 1183°C for an andesite sample (from Ember 49 Latent heats of vitrification are essentially zero and, thus, the transition from melt to glass does not contribute to the heat budget. 162 (a) 60 H Partial < T ^ Crystallization < ' 9 < > L + S -Cooling / L. Glassy Pile — - s * * r 1 i 1 i 1 > I , , < 1200 1000 800 600 400 200 0 T(°C) (b) 60 o E 40 X < cation \ I ^ Cooling Glassy Pile — I — r -s / 1 *r 6^ I l 1 l l i l l ! 1200 1000 800 600 400 200 0 T(°C) (c) 0.8 H ffi I 0.6 H 0.4 H 0.2 H T g -basalt R= 0.45 T 1200 1000 800 600 T(°C) 400 200 Figure 4.9 Differences in heat budget between dacite and basalt, (a) Heat released (AHJ by a basalt erupted at 1200°C undergoing 30% crystallization before reaching the glass transition temperature (Tg = 800°C) and then cooling to ambient ice temperature. Total heat released is a combination of sensible (S) and latent heat of crystallization (L). (b) A H R for dacite erupted at 1125°C that reaches its Tg (1000°C) without crystallization and then cools within the ice. The basalt curve (dashed) is shown for reference, (c) The ratio of heat released by dacite versus basalt mapped as a function of temperature. At the calorimetric T g, the dacite will have released 45 % of the total heat released by the basalt at an equivalent temperature. Over the entire path, dacite releases 70-80% of the total heat released by basalt. (Dacite is used rather than andesite in order to emphasize the difference, even though most M C V F rocks are andesite, not dacite.) Modified from Kelman et al. (2001). 163 Ridge Southwest) whose silica content was near the mean for MCVF samples. This temperature provided a starting point, although actual eruption temperatures would be significantly lower: Typical values for andesite pumice are 940-990°C (Carmichael et al. 1974), while values for crystal-rich andesite samples from the Soufriere Hills volcano in Montserrat (from the 1995- present eruption) range from 880-1050°C (Murphy et al. 2000). Thus, a value of 1000°C was adopted for the calculation. (3) The ice is at 0°C, and the lava cools to 0°C during the time frame of interest. This means that the calculated volume of ice melted is a maximum. (4) Ice is melted only directly over the vent. This is not realistic because some heat is expended melting drainage pathways around the vent, but it too means that the calculated ice thicknesses are maxima. (5) Heating of meltwater ranges from 0-50°C. If meltwater is removed soon after it is generated, this would minimize its potential for being heated. GuSmundsson et al. (1997) postulated that meltwater at the 1996 Gjalp subglacial eruption was heated to 15-20°C, and Hoskuldsson and Sparks (1997) predicted that water temperature within a growing meltwater cavity averaged 5-50°C. In the MCVF, it is likely that meltwater is heated most significantly early during eruptions, when drainage pathways cannot accomodate discharge, and that it is heated to a lesser degree after discharge has peaked. (6) Latent heat plays a role in cooling. This is assumed because most intermediate 164 composition glaciovolcanic rocks in the MCVF have less than 30% glass, often substantially less, in spite of the rapidity of their cooling. Therefore, a latent heat factor is added to calculations. (7) Magma pulses are spaced sufficiently close in time that the separate flows can be considered en masse. This is probably realistic, because the overall geometry of the large flow-dominated tuyas suggests repeated flooding of lava into open, vertical-walled cylinders. Thermal energy released by a mass of cooling lava, Q m , is calculated by: Qm=mJcm[T-Tf]+Lm) (3) If all this energy goes into melting ice, the mass of ice melted, in kg, is: m = L, •m (4) However, if some of the energy is applied to heating meltwater, then: Qm=0, •melt ice ,+ Q, heat water (5) 165 Qm=[ 7J (6) Equation 6 is rearranged to solve for the mass of ice melted: Li+(cwA Tw) (7) Thus, the volume of ice melted is: V = Qn PiiLi+iCnATJ] (8) These volumes of ice melted are translated into ice thicknesses that could be penetrated. Volume of ice directly over a vent is given by: (9) where am is the area above the upper surface of the edifice. It is calculated by measuring thetwo dimensional surface area for each edifice directly from the map. Combining Equations 8 and 9, we can calculate ice thickness, h,, with: (amPi)[LMcwATw)] (10) 166 Figure 4.10 shows the thicknesses of ice that could potentially be melted at six subglacial domes and three flow-dominated tuyas, and how these compare to the maximum possible thickness of the Cordilleran ice sheet above each edifice (from Figure 4.8). The most important conclusion to be drawn from Figure 4.10 is that the flow-dominated tuyas have no difficulty generating enough heat to breach the maximum possible thickness of the Cordilleran ice sheet several times over. This is true regardless of the degree of meltwater heating. It is unrealistic to assume that no heat was lost to the formation and maintenance of subglacial drainage channels, was released after the magma was solid, or was lost to the atmosphere or bedrock. However, even if only 60% of the heat given off by cooling the lava to ambient temperature is applied to melting ice directly over the vents (e.g. cooling from 1000°C to 400°C), this is still more than sufficient heat for the flow-dominated tuyas to breach the Cordilleran ice sheet at its maximum thickness (Figure 4.10). A second conclusion to be drawn from Figure 4.10 is that for four of the subglacial domes, breaching of the ice surface at the Fraser Glaciation maximum is not possible under any reasonable cooling scenario. For the other two subglacial domes, it is possible only under certain conditions (Figure 4.10). This does not prove that the subglacial domes could not have breached the Cordilleran ice sheet surface, only that it is unlikely or impossible that they did so during eruptions at the height of the Fraser Glaciation. If eruptions occurred after this maximum, breaching of the ice would potentially be energetically plausible. 167 Figure 4.10. Legend 0 v. 1 0 200 400 600 800 1000 1200 1400 1600 1800 2000 predicted ice thickness melted (Equation 10) (m) Figure 4.10 Maximum possible overlying ice thickness (known from the maximum elevation attained by the Cordilleran ice sheet) versus predicted thickness of ice melted (calculated with Equation 10). Points falling within the grey shaded area represent eruptions that could not have breached the ice surface at the maximum extent of the Fraser Glaciation. The bars represent ice ranges that could have been melted, assuming that 60-100% of all heat went into melting ice directly above the vents. The left end of each bar represents cooling from 1000°C to 400°C (and therefore, less melting of ice) and the right end of each bar represents cooling from 1000°C to 0°C (more melting of ice). Heating of meltwater to 25°C is assumed. If meltwater were heated only minimally, bars would elongate to the right (meaning that since heat was not carried from the system, more ice would be melted, and ice breaching would be more likely). If meltwater were heated to 50°C, the bars would elongate to the left, less ice would be melted, and the chance of eruptions breaching the ice surface would be lessened. Numbers 1 -6 (black) are subglacial domes, and 7-9 (blue) are flow-dominated tuyas. For most cooling scenarios, subglacial domes are unable to penetrate the full thickness of the Cordilleran ice sheet, whereas flow-dominated tuyas have no difficulty in doing so. 1. Ember Ridge Southeast. 2. Ember Ridge Northeast. 3. Ember Ridge North. 4. Ember Ridge West. 5. Ember Ridge Southwest. 6. Ember Ridge Northwest. 7. Little Ring Mountain. 8. Slag Hill tuya. 9. Ring Mountain (small inset map). Ring Mountain is located far to the right of the other edifices because it is so much larger in volume, hence, its ice melting potential is shown on the small inset chart. 4.5 ICE THICKNESS OVER SUBGLACIAL DOMES 4.5.1 Introduction Ice is the source of meltwater. It also serves as a barrier controlling the distribution of volcanic deposits as well as the behaviour (e.g. release versus retention) of meltwater. Its thickness and permeability structure control water pressure at subglacial vents. Finally, it operates as an agent of erosion. Thus, the difference between a volcano erupting beneath 1000 m of ice and 200 m of ice is significant. 4.5.2 Ice thickness based on assumptions about ice cavity size It is possible to estimate ice thicknesses contemporaneous with eruption if we assume that at some point during eruption, temporarily stable cavities were present prior to the injection of lava. The existence of such cavities at active subglacial volcanic vents requires that rates of ice cavity wall meltback (which depend on heat transfer rates) equal rates of ice cavity closure by ductile deformation (which depend on ice thickness). If we calculate heat transfer rates for lava masses and ice deformation rates for cavities of sizes that could have accomodated these lava masses, we can potentially obtain a more precise range of ice thicknesses for subglacial dome eruptions. However, we do not know the cavity pressures during eruption (and they probably changed over time). Additionally, we do not know the lava effusion rates, the rates at which subglacial drainage conduits develop, or the proportion of heat that goes into melting drainage conduits (as opposed to that which goes to melting the vent cavity). 170 Finally, many subglacial domes erupted as multiple lava pulses, and are large enough in total volume that it is unlikely that a cavity the size of the entire mass was present prior to eruption. Thus, the above method of estimating ice thickness is probably not appropriate for most subglacial domes in the MCVF, though it may be suitable for small examples (like Ember Ridge Northeast) or for small discrete lobes of larger subglacial domes. Appendix 5 gives the details of such a calculation. However, the results of it do not aid in narrowing the range of ice thicknesses that could have overlain subglacial domes at the time of eruption. 4.5.3 Ice thickness based on field observations and ice quantities melted Because of the many unknown variables in subglacial eruptions, the most reasonable estimates of ice thickness over subglacial domes are based on a combination of field observations and calculations of ice quantities melted. The calculation in 4.4.2 introduced a range of ice thicknesses that MCVF subglacial dome eruptions could theoretically have melted through (-150-800 m). This range depends on edifice size, degree of meltwater heating, and the fraction of heat that melted ice directly over the vent (as opposed to heat used to melt drainage channels around the vent). We can assume a wide range of initial ice thicknesses and degrees of meltwater heating and still not breach the ice surface. If we apportion 60% of all heat to melt ice directly over the vent and assume 25°C meltwater heating (Figure 4.10, the leftmost end of the bar for each edifice), the upper end of this range decreases to -500 m. This means that at least 150-800 m of ice covered each subglacial dome at the start of eruption (because of the melting of the ice and the upward growth of the volcano, the 171 final ice thicknesses would have been less). This range is still very broad. For Ember Ridge North and Ember Ridge Northwest, deposits are at such high elevations that in order for the eruptions to have not breached the ice surface, initial ice thicknesses must have been at or near the Fraser Glaciation maximum. At Ember Ridge North, this was 640 m. With 60% of heat apportioned to melting ice directly over the vent and 25 °C meltwater heating, 490 m of overlying ice could have been melted through, and the ice over the vent would have been thinned to 160 m by the end of the eruption. Similarly, at Ember Ridge Northwest, the overlying ice thickness at the Fraser Glaciation maximum was 535 m, a thickness of 402 m could have been penetrated by melting, and a final overlying ice thickness of 133 m would have remained. This means that at both Ember Ridge North and Ember Ridge Northwest, the final ice thicknesses over the vents were near or within the thickness range (<150 m) that is potentially permable and likely to deform brittly. We can generalize to the other Ember Ridge subglacial domes based on Ember Ridge North and Ember Ridge Northwest: By continuing with the assumptions of 60% of all heat going to melt ice directly over vents and 25°C meltwater heating, and assuming that eruptions just barely breached the ice surface, then initial ice thicknesses had to have been at least 110-490 m, with 110 m representing Ember Ridge Northeast, 490 m representing Ember Ridge North, and all other subglacial domes falling between the two. By adding the possibility that final ice thicknesses were up to 150 m, this range extents to 100-640 m. There is limited field evidence to support this generalization at Ember Ridge Southeast, in the form of lava bodies that appear to have filled vertical fractures in ice. Thus, it is probable that all MCVF subglacial dome eruptions commenced under -100-172 650 m of ice. 4.6 MAGMA-RELATED INFLUENCES 4.6.1 Introduction The heat available for melting ice relates directly to the total volume of magma erupted. The properties of that magma are also important; composition-related magma properties (such as liquidus temperature, heat capacity, glass transition temperature, viscosity, and volatile content) control eruption style and rate, and influence heat transfer to the surroundings. Because variations in magma properties affect subaerial eruptions, it is reasonable to assume that they also influence glaciovolcanic eruptions, and the fact that basaltic glaciovolcanic deposits in the GVB do show evidence for lava-water interactions supports this contention. In the MCVF, however, the compositional range is small, and most glaciovolcanic deposits are andesitic. There are no obvious morphological or textural differences between samples that can be correlated with composition. Therefore, it is difficult to evaluate the importance of magma-related factors to glaciovolcanism, although it is clear that large differences in composition (basaltic versus andesitic lavas) are important. 4.6.2 Total erupted volume The volume of lava erupted is the only obvious difference between subglacial dome and flow-dominated tuya eruptions. MCVF subglacial domes range in volume from 7 x 105 m 3 to 4 x 107 m 3, while flow-dominated tuyas range in volume from 9 x 106 m 3 to 173 6 x 108 m 3. A logical assumption is that larger volume eruptions are more likely to breach the ice surface, as are eruptions under thin ice (this is shown graphically in Figure 4.10). The smallest flow-dominated tuya (LQvpa7, the Slag Hill tuya) is at an elevation that indicates that at the height of the Fraser Glaciation it was overlain by no more than 375 m of ice; had ice above the vent been thicker, the eruption would not have breached the ice surface, and a subglacial dome would have formed. If initial eruptions at subglacial domes and flow-dominated tuyas are truly identical, then the buried portions of flow-dominated tuyas should resemble subglacial domes. This hypothesis could be tested at a flow-dominated tuya with substantial excavation by erosion. However, no such example has yet been identified. 4.63 Effusion rate and style, viscosity, and volatile content The eruption style for MCVF glaciovolcanism is almost entirely effusive. Effusion rate is a potentially significant but unknown factor in MCVF subglacial eruptions. The large variations in total erupted volume and in numbers of lava pulses make it possible that instantaneous or time-averaged effusion rates vary between centres. This would impact rates of heat transfer to the ice. Typical andesite and dacite effusion rates range from 0.05 m /s to 40.3 m /s (Table 4.4). If a subglacial dome with the same volume as Ember Ridge Northwest were emplaced at the lowest of the rates in Table 4.4 (0.05 m /s), it would take 153 days to emplace the entire mass. If the highest of the eruption rates from Table 4.4 were used (40.3 m/s), it would take only 4.6 hours. Thus, lower effusion rates diminish the possibility of ice breaching because they decrease melting rates relative to deformation 174 Table 4.4 Effusion rates at volcanoes erupting lavas of intermediate composition. location effusion rate (m /s) lava composition comment reference Unzen (Japan) 4.6 dacite start of 1991-1995 eruption Nakada and Motomura (1999) Santiaguito (Guatemala) 0.6-1.9 dacite 3-5 year rapid growth periods Rose (1987) Santiaguito (Guatemala) 0.16 dacite 10-12 year slow growth periods Rose (1987) Mount St. Helens (USA) 1.4-40.3 dacite short-term growth rates Anderson and Fink (1990) Soufriere Hills (Montserrat) 0.2 andesite late 1995 Sparks etal. (1998) Merapi (Indonesia) 0.05-0.32 andesite 1994-1995 Hammer et al. (2000) rates; ice can potentially recover the vents as fast as it is melted away. Lower effusion rates should also impede the development of subglacial drainage systems. Although effusion rates for MCVF eruptions are not known and there are no textural or petrological differences that suggest any differences in eruptive behaviour between centres, effusion rate cannot be ruled out as a factor that controls whether the ice surface is breached. Eruptions at flow-dominated tuyas were clearly exogenous, as evidenced by their flat upper surfaces and stacking of lava flows. There is also evidence for exogenous eruption at some subglacial domes and impoundment features (Ember Ridge Southeast and Tricouni Southwest), in the form of vertical screens or spires of lava whose joint orientations and sizes suggest that the lava was injected into fractures in ice. However, it is possible that some growth at subglacial domes was endogenous: At the three MCVF flow-dominated tuyas, there are abundant crustal xenoliths on the top surfaces, but comparatively few at lower levels in the edifices. These xenoliths are interpreted as till fragments derived from the surface or upper layers of overlying or adjoining ice, where debris is likely to be concentrated (e.g. Benn and Evans 1998; Owens et al. 2003). There is no concentration of xenoliths in any part of subglacial domes, suggesting that either subglacial domes did not melt up into the upper portions of the ice or that some dome growth was endogenous. Whether an eruption is endogenous or exogenous is significant, because endogenous eruptions are likely to involve lower heat transfer rates than exogenous eruptions, and are thus less likely to breach the ice surface. Viscosity is important to glaciovolcanic eruptions because, along with confining ice, it controls final edifice morphology. High magma viscosities are likely to result in high aspect ratio features even in the absence of flanking ice. The low glass content and 176 high crystallite content of MCVF sample groundmasses is likely due to degassing-induced crystallization. This phenomenon would increase viscosity (Sparks and Pinkerton 1978; Swanson et al. 1989; Cashman 1992; Geschwind and Rutherford 1995; Sparks 1997; Jaupart 1998; Hammer et al. 1999; Melnik and Sparks 1999; Voight et al. 1999). However, in the MCVF, edifice morphologies and the accompanying joint orientations indicate that confining ice was more important than viscosity in influencing deposit morphologies. Another important consideration with respect to magma viscosity is the possible decoupling of volatiles from low-viscosity magma during ascent, resulting in an initial burst of gas-driven melting50. The possibility of this occurring cannot be evaluated at MCVF vents, but its occurrence would accelerate drainage system development and make early equilibration of subglacial cavity pressure with atmospheric pressure more likely, thus promoting drainage and increasing the likelihood of dry eruptions. In summary, the compositional range of MCVF lava samples is small, and there are no differences in sample texture, groundmass crystallinity, vesicularity, or outcrop-scale characteristics that can be correlated with whether or not a unit breached the ice surface or with any other morphological feature. Although this is not proof that viscosity, effusion rate, and volatile content were identical for subglacial domes and flow-dominated tuyas, it suggests that total erupted volume relevant to ice thickness is the primary control on glaciovolcanic edifice morphology. Style of effusion (endogenous versus exogenous) may potentially play a minor role. 50 This is inferred as the cause of early rapid melting of ice at the 1969 eruption at Deception Island, Antarctica (Smellie 2002). 177 4.7 MODELS FOR INTERMEDIATE COMPOSITION GLACIOVOLCANISM 4.7.1 Introduction Below, I present two qualitative models to explain intermediate composition glaciovolcanic landforms in high relief settings. The first (4.7.2) applies to subglacial to emergent eruptions that generate subglacial domes and flow-dominated tuyas. The second (4.7.3) applies to subaerial eruptions in which lava is impounded against adjacent ice. The models are based primarily on observations from the MCVF but can be applied to the rest of the GVB as well as to other settings with significant topographic relief. 4.7.2 Subglacial to emergent eruptions Prior to eruption, heating at the base of the ice causes a cavity filled with meltwater to form (Figure 4.1 la). How prolonged and extensive this preemptive heating is determines the size of the vent cavity when the first magma erupts. Cavity pressure, P c, is negative (less than glaciostatic pressure, Pg) prior to the introduction of the first lava, because water's density is lower than that of ice. Hence, a depression forms in the ice above the heat source as soon as the melting of ice commences, and meltwater does not immediately drain (Bjornsson 1988). Initial eruptions are into a cavity filled with water51 at a pressure less than P g (Figure 4.1 lb), and consist of unknown proportions of lava flows and hyaloclastite (and 5 1 The cavity must be filled with water because it is assumed that there is no initial contact with the atmosphere. 178 original ice surface Figure 4.11 Initial stages of subglacial eruption of intermediate composition lava in steep topography. Drainage tunnels are not illustrated, but drainage of water is schematically depicted with arrows. The size of the depression in the ice above the vent is exaggerated for visibility. (a) Initial heating of the base of the ice and formation of a meltwater-filled cavity. Cavity pressure ( P c ) < glaciostatic pressure ( P g ) . A depression forms in the ice surface. (b) Initial intrusion of lava into the water-filled subglacial cavity. As lava is added to the cavity, P c begins to rise. Water begins to drain. P c peaks at some pressure greater than P g . 179 possibly pillows). The initial hydraulic gradient will be towards the vent due to the depression in the ice surface above it. Because there is so little evidence for lava-water interaction at subglacial domes and flow-dominated tuyas, it is likely that this situation does not occur for any protracted time period. As eruptive materials enter the subglacial cavity, P c begins to rise. Water at some point begins to drain away from the vent; when this occurs depends on the cavity pressure, the steepness of ice sloping towards the vent, and the steepness of bedrock topography sloping away from the vent. Even while P c is less than glaciostatic, sufficiently steep topography may promote meltwater drainage (Bjornsson 1988). Most water drains through englacial and subglacial conduits. However, it is likely that the developing subglacial drainage system will initially be insufficient to cope with the high meltwater discharge rates, unless ice is very thin, the eruption occurs near an ice margin, or a well-developed drainage system is present prior to eruption due to geothermal heating (as at Mount Rainier, for example; Kiver and Steele 1975). Lava continues to enter the cavity, ice continues to melt, and P c continues to rise until drainage conduits are sufficient to accomodate meltwater discharge. P c eventually peaks at some pressure that is probably greater than P g but much less than 10 MPa . After this point, the drainage system is able to remove meltwater as rapidly as it is produced. If initial ice thickness is less than 150 m, P c will not rise above atmospheric (0.1 MPa) because the vent cavity and drainage conduits will have an early connection with the atmosphere. However, if there is an initial accumulation of water, there will be a 5 2 The glacier is unlikely to be floated by any cavity overpressures because vents are located on ridge tops. 180 hydrostatic pressure imposed by its depth. Water will be able to escape supraglacially or through permeable ice, and the thinness of the ice will promote more rapid development of subglacial and englacial drainage conduits . However, for most subglacial domes, initial ice thickness had to have been greater than 150 m. Subglacial dome eruptions Lava continues to erupt into a cavity filled with water and fragmental material but, after drainage conduits are sufficient to accomodate the maximum meltwater discharge rates, pressure should continue to decrease towards glaciostatic. At some point, contact with atmospheric pressure is made, either by tlmining ice above the vent or via distal meltwater drainage conduits. After this occurs, subglacial drainage tunnels are likely to be partially filled with air, and further meltwater drainage will follow the underlying topography (Figure 4.12b). Further eruptions consist of lava flows emplaced within a relatively dry environment. Under ice thinned to <150 m thick, brittle failure of the ice may also lead to emplacement of lava units in crevasses. There may be collapse of unstable portions of the lava mass to form synvolcanic breccias, and further intrusion of lava through these breccias (as occurred at Ember Ridge Southeast). In subglacial dome eruptions, individual lava flows are likely to be smaller or more widely spaced in time than in flow-dominated tuya eruptions, and there may be a greater degree of endogenous effusion. Endogenous effusion may also promote the preservation of some hyaloclastite deposits on the top surface of some subglacial domes, 5 3 Drainage through permeable ice or via supraglacial overflow, and the significance of ice thickness for drainage, are discussed in detail in Smellie and Skilling 1994; Smellie 2002). 181 (a) © V original ice surface Figure 4.12 Subglacial eruption of intermediate composition lava in steep topography, continued from Figure 4.11. This figure depicts an eruption that does not breach the ice surface (thus forming a subglacial dome). Drainage tunnels are not illustrated, but drainage of water is schematically depicted with arrows. (a) Lava continues to erupt into a subglacial water-filled cavity. After cavity pressure has peaked and then begun to drop, subglacial meltwater conduits become able to accomodate discharge rates. Steep topography makes it possible for water to continue to drain, even though the cavity pressure may no longer be greater than glaciostatic. (b) Contact is made between the subglacial vent cavity, drainage conduits, and atmosphere. Lava continues to erupt into the subglacial cavity, which now contains air (and probably a small amount of water) at atmospheric pressure (0.1 MPa). Meltwater drains steadily, so the cavity is relatively dry. The depression in the ice surface above the vent continues to grow. Final ice thickness is less than or equal to -150 m, and ice may fail brittly over the vent. (c) Subglacial dome after the removal of ice. Legend air i c e water lava basement rock unconsolidated fragmental material hyaloclastite drainage of meltwater 182 however, most early-formed fragmental material is likely to be buried by lava flows or located near the base of the edifice (Figure 4.12 b, c). After ice is removed, a steep-sided edifice comprising finely jointed lava flows and volumetrically minor fragmental units is left behind (Figure 4.12c): a subglacial dome. Flow-dominated tuya eruptions For subglacial eruptions that ultimately penetrate the ice, the initial effusions and development of the drainage system and evolution of subglacial cavity pressure are similar to what occurs at subglacial domes (4.12a, 4.12b). However, the total volume of lava erupted is larger relative to the overlying ice thickness. Thus, the ice surface is breached. Figure 4.13a shows a time prior to ice breaching, after P c has peaked. After the ice surface is penetrated, lava continues to erupt through an open chimney in the ice (Figure 4.13b). Some heat is lost to the atmosphere. Drainage pathways are partially filled with air at atmospheric pressure, and have sufficient capacity to remove water from the vent area. Flows are subaerial and impounded on all sides by flanking ice and fragmental material, although variations in eruption rate and ice flow rate may result in some open space being intermittently present on the flanks of the flow-dominated tuya. The timing between lava pulses is probably short. After the ice is removed (Figure 4.13c), a steep to vertical-sided edifice comprising stacked, finely jointed lava flows is left behind. The upper surface lacks the flow folds seen in large subaerial flows of similar composition because the lava was 183 Figure 4.13 Subglacial eruption of intermediate composition lava in steep topography, continued from Figure 4.11. This figure depicts an eruption that ultimately breaches the ice surface (thus forming a flow-dominated tuya). Drainage tunnels are not illustrated, but drainage of water is schematically depicted with arrows. (a) As in 4.12a, cavity pressure peaks and then drops, as subglacial meltwater conduits become able to accomodate discharge rates. Steep topography makes it possible for water to continue to drain, even though the cavity pressure is no longer greater than glaciostatic. Following this, contact with atmospheric pressure (0.1 MPa) is made (as in 4.12b). Lava continues to erupt into a subglacial cavity that contains air and some water at atmospheric pressure. Water drains according to topography. The overlying ice gets thinner. (b) Lava continues to erupt into an open chimney in the ice. There is space present between the ice and the edge of the flow-dominated tuya (arrow x), which wil l intermittently fill with breccia and ice fragments. The space wil l increase in size when contact with lava is made but wi l l decrease in size between magma pulses due to flow of ice towards the vent and to brittle failure of the ice that flanks the vent. Water drains continuously. A l l eruptions are lava flows. (c) Flow-dominated tuya after the removal of ice. Legend air ice water lava basement rock unconsolidated fragmental material hyaloclastite drainage of meltwater 184 confined by ice and no free surface was available. The confinement by ice may also explain why there is no evidence for vents; they are buried. Any fragmental material is likely to be buried by later flows and the flanking scree apron that may comprise more than half the height of the flow-dominated tuya. 4.7.3 Impoundment eruptions Impoundment features are of interest because they can be used to map paleo-ice distributions and determine relative or absolute times of deglaciation. They are likely to be common in intermittently glaciated regions with extreme topography like the GVB, because they form during periods of restricted ice cover, such as during the downwasting of continental ice sheets or when high altitude vents are covered by thin alpine glaciers. Figure 4.14 shows four hypothesized paleo-ice distributions for the Cordilleran ice sheet in the MCVF. The extreme bedrock relief and high altitude vents promote ice sheet geometries during glacial recession that are likely to result in impoundment features. Two impoundment geometries are possible: Either lava flows downslope and impounds against ice sheet margins, or lava erupts under a thin glacier, breaches the surfaces, then continues to flow in subaerial channels constrained by ice. Unconsolidated glacial debris may also play a role in the damming of lava flows. However, it is unlikely that any large impoundment feature (e.g. lower Cauldron Dome) formed by impoundment against piles of unconsolidated glacial debris because of the inherent instability of such debris. Based principally on observations at the MCVF, I have developed a qualitative 185 2 km Figure 4.14 Ice coverage over the MCVF at four times in the past (shading). The distribution of modern alpine glaciers is shown (black outline); they are currently receding, thus probably covered a greater area during periods of deglaciation. (a) the height of the Fraser Glaciation, when local ice stood at -2300 m. (b) some time after the Fraser maximum, when local ice stood at 2000 m. (c) some time after (b), when local ice stood at 1800 m. (d) some time after (c), when local ice stood at 1400 m. (d) _ r ~ T L A % • • — a >^ 2 km Legend Modem distribution I I not covered by ice of Quaternary volcanic rocks: \Z3 c o v e r e d by i c e paleo-ice coverage j older basement ^VV_ modern alpine glacier margins model describing the first impoundment scenario, in which large quantities of lava pond against valley-filling ice (Figure 4.15): Eruption commences with lava flowing downslope and making contact with ice and any associated till (Figure 4.15a). This melts the ice but also quenches the lava, and results in the formation of a solid lava barrier against which continued effusions of lava pool. Meltwater is channeled away along the edge of the ice. Meltwater may also accumulate behind short-lived dams of glacial debris (as is indicated by fine-grained, laminated, dropstone-rich sediments associated with the LQvpadu6 flow at Tricouni Southeast). However, the lack of hyaloclastite associated with any impoundment feature suggests that large quantities of water are not typically retained during impoundment eruptions. There are two possibilities for what occurs next. Either the initial solidified lava barrier is pushed forward upon a mass of breccia to make room for molten lava to pile up behind it (Figure 4.15b, c), or the initial frozen lava barrier remains stationary and lava piles up behind it, eventually pouring over the top to again make contact with the adjacent ice (not illustrated). In both cases, the lava barrier will increase in height with time. The former scenario is more likely for large volume eruptions resulting in high cliffs, such as those comprising the lower sequence at Cauldron Dome. However, even if the lava barrier is pushed forward as shown in Figure 4.15, lava can still spill over the top of the solid barrier and get injected into fractures in the ice. This results in the formation of irregularly-shaped, finely-jointed spires and ridges like those at the southern margin of the Tricouni Southwest impoundment feature. Such small-scale features have a very low preservation potential. 187 Figure 4.15 Legend molten lava ice solid lava basement •Sit**.'*?. autobreccia and/or glacial till columnar joints Figure 4.15 Impoundment of lava by large quantities of ice in steep terrain. (a) Lava flows downslope, onto and against the glacier margin and any glacial till that marks the edge of the ice. Ice is melted, and meltwater is channeled away along the edge of the glacier, through open pathways and through water-saturated till. The distal portions of the lava flow are quenched against the ice, and the lava spreads out laterally, piling up behind the ice. Autobreccia develops at the terminus of the lava flow, and lava rolls over the top of this, resulting in breccia at the base of the lava flow. (b) The still-molten portions of the lava flow continue to flow downslope, pushing the quenched margins ahead of them atop a base of breccia. Lava piles up behind the barrier but the barrier is also physically pushed downslope as ice continues to melt. At the terminus of the lava flow, there is extensive fragmentation of the cooling lava, and the earliest-formed cooling surface is probably destroyed oo and incorporated as autobreccia. This breccia, probably with an added till component, is trapped between the lava flow margin and the ice. (c) As with (b), the lava flow continues to expand downslope. It also thickens vertically. The totally solid portion of the lava flow extends further uphill, encompassing a larger proportion of the total flow. (d) After the ice is removed, an extremely unstable cliff is left behind. Terminal joints may be partially removed by erosion (illustrated), or completely removed. After the ice is removed, an extremely unstable cliff remains. Terminal joints are likely to be partially or totally removed by erosion within a few thousand years. This exposes sequences of coarse vertical columnar joints. Both geometries of impoundment (downslope [Figure 4.15] and ice-channeled) have occurred in the MCVF. Downslope impoundment is best exemplified by the cliffs at Pali Dome East, Pali Dome West, and Lower Cauldron Dome. Impoundment by channeling ice is the likeliest explanation for the morphology of the long narrow ridge flanked by the modern alpine glacier at the south end of Slag Hill (Qvpaul3). This latter type of impoundment feature can occur on any glaciated volcano and does not require continental scale ice sheets, thus should be found in a wider range of glaciovolcanic settings. However, ice-channeled lava flows are likely to be immediately recovered by ice and eroded if there is continued glaciation after their eruption. 190 CHAPTER 5 Conclusions and Future Work 5.1 CONCLUSIONS This thesis investigates glaciovolcanism in the Mount Cayley volcanic field (MCVF) of southwestern British Columbia's Garibaldi Volcanic Belt (GVB). The core of the thesis is a 1:20,000 volcanological map of the MCVF, which was used in conjunction with field observations, geochemical data, and petrographic examinations to investigate the eruptive processes that produce specific landforms. This is the first study to focus solely on intermediate composition glaciovolcanic deposits in a convergent margin setting54. My most important conclusion is that the behaviour of meltwater during glaciovolcanic eruptions (accumulation versus drainage) is crucial, and the main control on this is bedrock-vent-ice geometry. Thin ice coupled with high altitude vents makes it probable that glaciovolcanic deposits will lack evidence for lava-water interaction. Thus, glaciovolcanism in the GVB is very different than in most other glaciovolcanic settings. My other conclusions are: • The landforms characteristic of the GVB are subglacial domes, flow-dominated tuyas, and impoundment features. Al l lack pillows and have negligible quantities of hyaloclastite, indicating that water did not accumulate at vents during eruptions. 5 4 Other studies of intermediate composition glaciovolcanic rocks have focused on specific landform types or on features such as jointing (e.g. Lescinsky and Sisson 1998), rather than on general glaciovolcanic processes. 191 • Eruption-related factors control the quantity of meltwater produced at subglacial volcanic vents, and the rate at which it is produced (Figure 5.1a). • Bedrock-vent-ice geometry controls the development of drainage pathways and the evolution of hydraulic conditions (Figure 5.1b). High altitude vents minimize ice thicknesses above subglacial volcanoes and this, coupled with steep topography that slopes away from vents, promotes meltwater drainage. Meltwater production rates also influence drainage pathway development and the evolution of hydraulic conditions. • Eruptions involving little or no direct lava-water interaction require drainage pathways, hydraulic conditions that promote water flow away from the vent, and meltwater drainage rates that equal or exceed meltwater production rates. If any of these conditions are not fulfilled, water will be trapped at the vent and deposits will show evidence for lava-water interaction (Figure 5.2). • The pressure in a subglacial vent cavity and its adjacent drainage conduits is probably less than glaciostatic prior to the first eruption of lava, but rises as intermediate composition lava is added to the cavity. Cavity pressure peaks at some pressure higher than glaciostatic, but diminishes once drainage conduits are sufficient to accomodate the highest meltwater production rates. Contact with the atmosphere is made at some point during eruption, after which time subglacial drainage conduits are partially air-filled and are at atmospheric pressure. 192 5 « a O u © 1>J (a) ERUPTION-RELATED FACTORS Ext r i ns i c In t r ins ic • eruption rate • viscosity • eruption style • eruption temperature • eruption • latent heat contribution volume • volatile content and behaviour (b) BEDROCK-VENT-ICE G E O M E T R Y FACTORS B e d r o c k •topographic relief • vent elevation Ice • distribution • thickness • properties (e.g. permeability) Meltwater Production quantity rate Drainage Potential pathway development hydraulic conditions Figure 5.1 Controls on the morphology and internal stratigraphy of glaciovolcanic deposits in the GVB. Ultimately, the most important control on edifice morphology and internal stratigraphy is whether or not meltwater is retained at the vent (Figure 5.2). (a) Factors related directly to eruptions. Together, these factors control rates and amounts of meltwater produced. Rates of meltwater production also influence drainage pathway development and the evolution of hydraulic conditions. (b) Factors related to bedrock-vent-ice geometry. Together, these factors control the development of drainage pathways and the evolution of hydraulic conditions at and around the vent. drainage pathways unavailable OR hydraulic conditions promote meltwater retention OR melting rate exceeds drainage rate ERUPTION INTO WATER drainage pathways available AND hydraulic conditions promote meltwater drainage AND drainage rate exceeds melting rate "DRY" ERUPTION Figure 5.2 Controls on subglacial eruption. The presence or absence of water is the most important factor influencing eruptions. The conditions required to get eruption into water, or eruption in the absence of water, are shown. If any of the former conditions are fulfilled, eruptions will be into water. Only if all of the latter conditions are fulfilled will eruptions be dry. • Composition influences style of glaciovolcanism, because the only GVB glaciovolcanic deposits that show evidence for extensive lava-water interaction are basaltic. However, no obvious differences in deposits that can be correlated with composition exist within the basaltic andesite to dacite range. • There are no systematic textural or penological differences between lava samples from subglacial domes and flow-dominated tuyas. The two landforms result from similar eruptive events; the principal difference is that during flow-dominated tuya eruptions, the ice surface is breached, while in subglacial dome eruptions, all volcanism is subglacial. Flow-dominated tuyas are likely to form when vents are at high altitude (hence under thin ice) and/or when total eruption volumes are large. Subglacial domes are more likely to form when vents are at low altitude, when total eruption volumes are small, or when eruptions involve a single lava pulse. •The pattern of waxing and waning of ice in a landscape characterized by high relief and high-elevation volcanic vents results in many impoundment eruptions. Hence, impoundment features should be very common in settings that, like the GVB, have high relief and temperate climates. Impoundment features are extremely unstable and are subject to rapid erosion, including major landslides. • Flow-dominated tuyas and impoundment features can be used to directly reconstruct paleo-ice distributions because the top surfaces of flow-dominated tuyas mark the 195 approximate top surfaces of ice sheets, and the termini or lateral margins of impoundment features mark the edges of ice sheets. Subglacial domes can be used as paleo-environmental indicators simply because they show that ice was present. Subglacial domes can also be used to estimate minimum initial ice thicknesses during eruption by calculating the quantities of heat released from cooling lava masses and converting these to ice thicknesses melted above vents. Based on this reasoning and on field observations that indicate late-stage eruptions into ice thin enough to deform brittly, most non-ice breaching subglacial eruptions in the MCVF commenced under -100-650 m of ice. 5.2 FUTURE WORK I would like to take the models I developed for the MCVF and test them at other glaciated convergent margins with extreme relief. I predict that glaciovolcanic deposits in such settings will include subglacial domes, flow-dominated tuyas, and impoundment features similar to those in the GVB. Outstanding issues to be addressed by such testing include obtaining more precise estimates for the ice thickness ranges under which subglacial domes form, investigating the potential for damming of lava or ponding of water behind moraine deposits in impoundment eruptions, clarifying the importance of eruption rate, and investigating the importance of volatile content and behaviour. Examinations of glaciovolcanic deposits in settings without extreme topography would aid in determining the relative importance of lava composition in shaping landforms. Examinations of lava-ice interaction features on Mars would aid in determining the importance of external factors such as ambient temperature, gravity, and atmospheric pressure; the fact that atmospheric pressure on Mars is much lower than on Earth 196 (-0.00067 MPa, compared to 0.1 MPa on Earth; Schofield et al. 1997) has important implications for subglacial eruptions that establish contact with the atmosphere55. I also believe there would be merit in conducting analogue experiments to go with my qualitative models, using polyethylene glycol wax (PEG) injected beneath ice5 6, in order to study the development of drainage conduits around a subglacial heat source in brittle ice. Experiments could be extended to "thicker" ice by applying a load to a block of ice during injection of PEG, thus allowing the ice to deform ductilely as a thicker ice sheet would do. This would provide a better understanding of the physics of subglacial eruptions. Finally, when commencing this research project, one of my original goals was to en explore the linkage between deglaciation and volcanism in the GVB. I compiled a database of ages for glacial and volcanic events in the GVB early during this project. Urifortunately, the lack of absolute dates for volcanic events left me with no platform to rigorously test the hypothesis that deglaciation triggers volcanism. Thus, a future project I envisage is the collection of precise age dates in order to look at correlations between glacial and volcanic events. 55 Relevant discussions of glaciovolcanism on Mars are already available (e.g. Hodge and Moore 1994; Wilson and Head 1994; Chapman et al. 2000; Greeley et al. 2000; Head and Wilson 2002). 56 Similar experiments were done by H6skuldsson and Sparks (1997) to examine the shapes of developing subglacial cavities. 57 Genetic links between deglaciation and volcanism were first investigated for Iceland (Jull and McKenzie 1996; Mclennan et al. 2002). 197 References Allen, C.C. 1980. Icelandic subglacial volcanism: thermal and physical studies. Journal of Geology, Vol. 88: 108-117. Alsdorf, D.E., and Smith, L.C. 1999. Interferometric SAR observations of ice topography and velocity changes related to the 1996 Gjalp subglacial eruption, Iceland. International Journal of Remote Sensing, Vol. 20: 3031-3050. Anderson, S.W., and Fink, J.H. 1990. The development and distribution of lava textures at the Mount St. Helens dome, in Lava Flows and Domes: Emplacement Mechanisms and Hazard Implications, J.H. Fink (ed.), IAVCEI Proceedings, Vol. 2: 25-46. Armstrong, J.E. 1981. Post-Vashon Wisconsin glaciation, Fraser Lowland, British Columbia. Geological Survey of Canada Bulletin 322: 34p. Armstrong, J.E., Crandell, J.R., Easterbrook, D.J., and Noble, J.B. 1965. Late Pleistocene stratigraphy and chronology in southwestern British Columbia and northwestern Washington. Bulletin of the Geological Society of America, Vol. 76: 321-330. Aydin, A., and DeGraff, J.M. 1988. Evolution of ploygonal fracture patterns in lava flows. Science.Vol 239: 471-476. 198 Barmin, A., Melnik, O., and Sparks, R.S.J. 2002. Periodic behavior in lava dome eruptions. Earth and Planetary Science Letters, Vol. 199: 173-184. Bates, R.L., and Jackson, J.A. 1990. Glossary of Geology, 3rd edition. American Geological Institute: 788p. Batiza, R., and White, J.D.L. 2000. Submarine lavas and hyaloclastite. in Encyclopedia of Volcanoes. H. Sigurdsson (ed.). Academic Press: 361-381. Benn, D.I., and Evans, D.J.A. 1998. Glaciers and Glaciation. Wiley, London: 734p. Berger, A.L. 1992. Astronomical theory of paleoclimates and the last glacial-interglacial cycle. Quaternary Science Reviews, Vol. 11: 571-581. Berger, A.L., Fichefet, T., Gallee, H., Marsiat, I., and Tricot, C. 1988. Testing the astronomical theory with a time dependent climate-ice sheets model. EOS (American Geophysical Union Transactions), Vol. 69:1236. Bjornsson, H. 1975. Subglacial water reservoirs, jOkulhlaups, and volcanic eruptions. Jokull, Vol. 25: 1-14. Bjornsson, H. 1988. Hydrology of Ice Caps in Volcanic Regions. Societas Scientarium Islandica, Reykjavik: 139p. 199 Blaise, B., Clague, J.J., and Mathewes, R.W. 1990. Time of maximum Late Wisconsin glaciation, west coast of Canada. Quaternary Research, Vol. 34: 282-295. Blake, S. 1981. Volcanism and the dynamics of open magma chambers. Nature, Vol. 289: 783-785. Brooks, G.R. and Friele, P. A. 1992. Bracketing ages for the formation of the Ring Creek lava flow, Mount Garibaldi volcanic field, southwestern British Columbia. Canadian Journal of Earth Sciences, Vol. 29: 2425-2428. Brooks, G.R., and Hickin, E.J. 1991. Debris avalanche impoundments of Squamish River, Mount Cayley area, southwestern British Columbia. Canadian Journal of Earth Sciences, Vol. 28:1375-1385. Budkewitsch, P., and Robin, P.-Y. 1994. Modelling the evolution of columnar joints. Journal ofVolcanology and Geothermal Research, Vol. 59: 219-239. Bye, A., Edwards, B.R., and Hickson, C J . 2000. Preliminary field, petrographic, and geochemical analysis of possible subglacial, dacitic volcanism at the Watts Point volcanic centre, southwestern British Columbia, in Current Research Part A; Geological Survey of Canada, Paper 2000-A20: 1-9. 200 Byers, F.M., Hopkins, D.M., Wier, K.L., and Fisher, B. 1947. Volcano investigations on Umnak Island 1946, Progress of investigations in 1946, part 3. United States Geological Survey Alaskan Volcano Investigative Report, Vol. 2: 19-53. Carmichael, I.S.E., Turner, F.J., and Verhoogen, J. 1974. Igneous Petrology. McGraw-Hill, New York: 739p. Cas, R.A.F., and Wright, J.V. 1987. Volcanic Successions. Allen and Unwin, London: 528p. Cashman, K.V. 1992. Groundmass crystallization of Mount St. Helens dacite, 1980-1986: a tool for interpreting shallow magmatic processes. Contributions to Mineralogy and Petrology, Vol. 109: 431-449. Chapman, M.G., Allen, C.C, Gudmundsson, M.T., Gulick, V . C , Jakobsson, S.P., Lucchitta, B.K., Skilling, LP., and Waitt, R.B. 2000. 'Fire and ice': Volcanism and ice interactions on Earth and Mars, in Environmental Effects on Volcanic Eruptions: From Deep Oceans to Deep Space, J.R. Zimbelman and T.K.P. Gregg (eds.), Kluwer Publishing, New York: 39-73. Clague, J.J. 1976. Quadra Sand and its relation to the Late Wisconsin Glaciation of southwest British Columbia. Canadian Journal of Earth Sciences, 13, 803-815. 201 Clague, J.J. 1977. Quadra Sand: a study of the late Pleistocene geology and geomorphic evolution of coastal southwest British Columbia. Geological Survey of Canada Paper 77-77,24p. Clague, J.J. 1980. Late Quaternary geology and geochronology of British Columbia. Part 1: radiocarbon dates. Geological Survey of Canada, Paper 80-13: 28p. Clague, J.J. 1981. Late Quaternary geology and geochronology of British Columbia. Part 2: summary and discussion of radiocarbon-dated Quaternary history. Geological Survey of Canada, Paper 80-35: 41p. Clague, J.J. 1983. Glacio-isostatic effects of the Cordilleran ice sheet, British Columbia, Canada, in Shorelines and Isostacy. D.E. Smith and A.G. Dawson (eds.). Special Publication - Institute of British Geographers, Vol. 16: 321-343. Clague, J.J. 1989. Chapter 1 in Quaternary Geology of Canada and Greenland. R.J. Fulton (ed.). Geological Survey of Canada, Geology of Canada, No. 1: 40-42. Clague, J.J. 1991. Chapter 12 in Geology of the Cordilleran Orogen in Canada. H. Gabrielse and C.J. Yorath (eds.). Geological Survey of Canada, Geology of Canada, No. 4: 419-434. Clague, J.J. 1994. Quaternary stratigraphy and history of south-coastal British Columbia. 202 in Geology and Geological Hazards of the Vancouver Region, Southwestern British Columbia. J.W.H. Monger (ed.)- Geological Survey of Canada Bulletin, Vol. 481: 181-192. Clague, J.J., Armstrong, J.E., and Mathews, W.H. 1980. Advance of the Late Wisconsin Cordilleran Ice Sheet in southern British Columbia since 22,000 yr. B.P. Quaternary Research, Vol. 13: 322-326. Clague, J.J., Evans, S.G., Rampton, V.N., and Woodsworth, G.J. 1995. Improved age estimates for the White River and Bridge River tephras, western Canada. Canadian Journal of Earth Sciences, Vol. 32: 1172-1179. Clague, J.J., and James, T.S. 2002. History and isostatic effects of the last ice sheet in southern British Columbia, in Ice Sheets and Sea Level of the Last Glacial Maximum. P.U. Clark and A.C. Mix (eds.). Quaternary Science Reviews, Vol. 21: 71-87. Clague, J.J., Mathewes, R.W., Guilbault, J.-P., Hutchinson, I., and Ricketts, B.D. 1997. Pre-Younger Dryas resurgence of the southwestern margin of the Cordilleran ice sheet, British Columbia, Canada. Boreas, Vol. 26: 261-278. Clague, J.J., and Souther, J.G. 1982. The Dusty Creek landslide on Mount Cayley, British Columbia. Canadian Journal of Earth Sciences, Vol. 191: 524-539. 203 Coish, R.A., and Journeay, J.M. 1992. The Crevasse Crag volcanic complex, southwestern British Columbia: structural controls on the geochemistry of arc magmas, in Current Research Part A; Geological Survey of Canada, Paper 91-1 A: 95-105. Cruden, D.M., and Lu, Z.Y. 1992. The rockslide and debris flow from Mount Cayley, British Columbia, in June 1984. Canadian Geotechnical Journal, Vol. 29: 614-626. Cui, Y., and Russell, J.K. 1995. Magmatic origins of calc-alkaline intrusions from the Coast Plutonic Complex, southwestern British Columbia, in The Southern Canadian Cordillera Transect ofLithoprobe. Cook, F A . (ed.). Canadian Journal of Earth Sciences, Vol. 32: 1643-1667 Cutler, P.M. 1998. Modelling the evolution of subglacial tunnels due to varying water input. Journal of Glaciology, Vol. 44: 485-497. DeGraff, J.M., and Aydin, A. 1987. Surface morphology of columnar joints and its significance to mechanics and direction of joint growth. Geological Society of America Bulletin, Vol. 99: 605-617. DeGraff, J.M., and Aydin, A. 1993. Effects of thermal regime on growth increment and spacing of contraction joints in basaltic lava. Journal of Geophysical Research, Vol. 98: 6411-6430. 204 DeGraff, J.M., Long, P.E., and Aydin, A. 1989. Use of joint-growth directions and rock textures to infer thermal regimes during solidification of basaltic lava flows. Journal of Volcanology and Geothermal Research, Vol. 38: 309-324. Edwards, B.R. 1997. Field, kinetic, and thermodynamic studies of magmatic assimilation in the Northern Cordilleran Volcanic Province, northwestern British Columbia; Ph.D. thesis, University of British Columbia, Vancouver, British Columbia: 324p. Edwards, B.R., Russell, J.K., and Anderson, R.G. 2002. Subglacial, phonolitic volcanism at Hoodoo Mountain volcano, northern Canadian Cordillera. Bulletin of Volcanology, Vol. 64: 254-272. Evans, S.G. 1986. Landslide damming in the Cordillera of Canada, in Proceedings, symposium on landslide dams; processes, risks, and mitigation. R.L. Schuster (ed.). American Society of Civil Engineers, Geotechnical Special Publication, Vol. 3:111-130. Evans, S.G. 1990. Massive debris avalanches from volcanoes in the Garibaldi Volcanic Belt, British Columbia. Program with Abstracts, Geological Association of Canada-Mineralogical Association of Canada Annual Meeting, Vol. 15: A38. Evans, S.G. 1992. High magnitude-low frequency catastrophic landslides in British Columbia. Open-File - British Columbia Ministry of Energy, Mines, and Petroleum Resources, 1992-15: 71-98. 205 Evans, S.G., and Brooks, G.R. 1991. Prehistoric debris avalanches from Mount Cayley volcano, British Columbia. Canadian Journal of Earth Sciences, Vol. 28: 1365-1374. Evans, S.G., Hungr, O., and Clague, J.J. 2001. Dynamics of the 1984 rock avalanche and associated distal debris flow on Mount Cayley, British Columbia, Canada; implications for landslide hazard assessment on dissected volcanoes. Engineering Geology, Vol. 61: 29-51. Fink, J.H., and Anderson, S.W. 2000. Lava Domes and Coulees, in Encyclopedia of Volcanoes, H. Sigurdsson (ed.), Academic Press, San Diego, CA: 307-319. Fisher, R.V., and Schmincke, H.-U. 1984. Pyroclastic Rocks. Springer-Verlag, New York: 472p. Fluck, P., Hyndman, R.D., and Wang, K. 1997. Three-dimensional model for great earthquakes of the Cascadia subduction zone. Journal of Geophysical Research, Vol. 102: 20,539-20,550. Friele, P.A., and Clague, J.J. 2001. Readvance of glaciers in the British Columbia Coast Mountains at the end of the last glaciation. Quaternary International, Vol. 87: 45-58. Fulton, R.J. 1969. Glacial lake history, southern Interior Plateau, British Columbia, 206 Geological Survey of Canada, Paper 69-37: 14p. Fulton, R.J. 1971. Radiocarbon geochronology of southern British Columbia. Geological Survey of Canada, Paper 71-37: 28p. Fulton, R.J. 1984. Quaternary glaciation, Canadian Cordillera, in Quaternary Stratigraphy of Canada - a Canadian Contribution to IGCP Project 24. R.J. Fulton (ed.). Geological Survey of Canada, Paper 84-10: 39-48. Geschwind, C.-H., and Rutherford, M.J. 1995. Crystallization of microlites during magma ascent: the fluid mechanics of 1980-1986 eruptions at Mount St. Helens. Bulletin of Volcanology, Vol. 57: 356-370. Ghiorso, M.S., and Sack, R.O. 1995. Chemical mass transfer in magmatic processes; IV, A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid-solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology, Vol. 119:197-212. Gilbert, J.S., Stasiuk, M.V., Lane, S.J., Adam, C.R., Murphy, M.D., Sparks, R.S.J., and Naranjo, J.A. 1996. Non-explosive, constructional evolution of the ice-filled caldera at Volcan Sollipulli, Chile. Bulletin of Volcanology, Vol. 58: 67-83. Gill, J.B. 1981. Orogenic Andesites and Plate Tectonics. Springer, Berlin: 390p. 207 Glen, J.W. 1952. Experiments on the deformation of ice. Journal of Glaciology, Vol. 2: 111-114. Glen, J.W. 1955. The creep of polycrystalline ice. Proceedings of the Royal Society of London, Vol. 228: 519-538. Greeley, R., Bridges, N.T., Crown, D.A., Crumpler, L., Fagents, S.A., Mouginis-Mark, P.J., and Zimbelman, J.R. 2000. Volcanism on the Red Planet: Mars, in Environmental Effects on Volcanic Eruptions: From Deep Oceans to Deep Space, J.R. Zimbelman and T.K.P. Gregg (eds.), Kluwer Publishing, New York: 75-112. Green, N.L. 1977. Multistage andesite genesis in the Garibaldi Lake area, southwestern British Columbia; Ph.D. thesis, University of British Columbia, Vancouver, British Columbia: 246p. Green, N.L. 1981. Geology and petrology of Quaternary volcanic rocks, Garibaldi Lake area, southwestern British Columbia. Geological Society of America Bulletin, Vol. 92: 697-702. Green, N.L. 1994. Mechanism for middle to upper crustal contamination; evidence from continental-margin magmas. Geology, Vol. 22: 231-234. 208 Green, N.L., Armstrong, R.L., Harakal, J.E., Souther, J.G., and Read, P.B. 1988. Eruptive history and K-Ar geochronology of the late Cenozoic Garibaldi volcanic belt, southwestern British Columbia. Bulletin of the Geological Society of America, Vol. 100: 563-579. Green, N.L., and Watters, B.R. 1977. Quaternary calc-alkaline volcanism in southwestern British Columbia. Abstracts with Programs - Geological Association of Canada, Vol. 2: 21. Grossenbacher, K.A., and McDuffle, S.M. 1995. Conductive cooling of lava: columnar joint diameter and stria width as functions of cooling rate and thermal gradient. Journal of Volcanology and Geothermal Research, Vol. 69: 95-103. Gudmundsson, M.T., Sigmundsson, F., and Bjornsson, H. 1997. Ice-volcano interaction of the 1996 Gjalp subglacial eruption, Vatnajokull, Iceland. Nature, Vol. 389: 954-957. Guffanti, M. , and Weaver, C.S. 1988. Distribution of Late Cenozoic volcanic vents in the Cascade Range: volcanic arc segmentation and regional tectonic considerations. Journal of Geophysical Research, Vol. 93: 6513-6529. Hammer, J.E., Cashman, K.V., Hoblitt, R., and Newman, S. 1999. Degassing and microlite crystallization during the pre-climactic events of the 1991 eruption of Mount Pinatubo, Phillippines. Bulletin of Volcanology, Vol. 60: 355-380. 209 Hammer, J.E., Cashman, K.V., and Voight, B. 2000. Magmatic processes revealed by textural and compositional trends in Merapi dome lavas. Journal of Volcanology and Geothermal Research, Vol. 100: 165-192. Hammer, P.T.C. and Clowes, R.M. 1996. Seismic reflection investigations of the Mount Cayley bright spot; a midcrustal reflector beneath the Coast Mountains, British Columbia. Journal of Geophysical Research, Vol. 101:20,119-20,131. Harder, M. , Russell, J.K., Anderson, R.G., and Edwards, B.R. 2003. Llangorse volcanic field, British Columbia, in Current Research Part A; Geological Survey of Canada, Paper A6: 1-10. Hawkesworth, C.J., Gallagher, K., Hergt, J.M., and McDermott, F. 1993. Mantle and slab contributions in arc magmas. Annual Review of Earth and Planetary Sciences, Vol. 21: 175-204. Head, J.W., and Wilson, L. 2002. Mars: a review and synthesis of general environments and geological settings of magma-H20 interactions, in J.L. Smellie and M.G. Chapman (eds.), Volcano-Ice Interaction on Earth and Mars. Geological Society, London, Special Publications, Vol. 202: 27-57. Hickson, C.J. 1987. Late Cenozoic rocks of the Wells Gray - Clearwater area, British 210 Columbia; Ph.D. thesis, University of British Columbia, Vancouver, British Columbia: 357p. Hickson, C.J. 1994. Character of volcanism, volcanic hazards, and risk, northern end of the Cascade magmatic arc, British Columbia and Washington state, in Geology and Geological Hazards of the Vancouver Region, Southwestern British Columbia. J.W. Monger (ed.). Geological Survey of Canada Bulletin 481: 231-250. Hickson, C.J. 2000. Physical controls and resulting morphological forms of Quaternary ice-contact volcanoes in western Canada. Geomorphology, Vol. 32: 239-261. Hickson, C.J., Moore, J.G., and Calk, L. 1995. Intraglacial volcanism in the Wells Gray-Clearwater volcanic field, east central British Columbia, Canada. Canadian Journal of Earth Sciences, Vol. 32: 838-851. Hickson, C.J., Russell, J.K., and Stasiuk, M.V. 1999. Volcanology of the 2350 B.P. eruption of Mount Meager volcanic complex, British Columbia, Canada; implications for hazards from eruptions in topographically complex terrain. Bulletin of Volcanology, Vol. 60: 489-507. Hicock, S.R., and Armstrong, J.E. 1981. Coquitlam drift: a pre-Vashon glacial formation in the Fraser lowland, British Columbia. Canadian Journal of Earth Sciences, Vol. 18: 1443-1451. 211 Hildreth, W., and Moorbath, S. 1988. Crustal contributions to arc magmatism in the Andes of Central Chile. Contributions to Mineralogy and Petrology, Vol. 98: 455-489. Hodges, C.A., and Moore, H.J. 1994. Atlas of Volcanic Landforms on Mars, United States Geological Survey Professional Paper 1534. Hofmann, A.W. 1988. Chemical differentiation of the Earth: the relationship between mantle, continental crust, and oceanic crust. Earth and Planetary Science Letters, Vol. 90:297-314. Holdahl, S.R., Faucher, F., and Dragert, H. 1989. Contemporary vertical crustal motion in the Pacific Northwest, in Slow Deformation and Transmission of Stress in the Earth, H.C. Cohen and P. Vanicek (eds.). American Geophysical Union, IUGG Geophysical Monograph 49: 17-29. Holman, J.P. 1997. Heat Transfer, 8th edition. McGraw-Hill, New York: 696p. Honnorez, J., and Kirst, P. 1975. Submarine basaltic volcanism: morphometric parameters for discriminating hyaloclastites from hyalotuffs. Bulletin of Volcanology, Vol. 39: 1-25. Hooke, R.L. 1981. Flow law for polycrystalline ice in glaciers; comparison of theoretical 212 predictions, laboratory data, and field measurements. Reviews of Geophysics and Space Physics, Vol. 19: 664-672. Hooke, R.L. 1998. Principles of Glacier Mechanics. Prentice-Hall, New Jersey: 248p. Hoskuldsson, A., and Sparks, R.S.J. 1997. Thermodynamics and fluid dynamics of effusive subglacial eruptions. Bulletin of Volcanology, Vol. 59: 219-230. Hungr, O., and Skermer, N.A. 1992. Technical Tour No.l : Debris torrents and rockslides, Howe Sound to Whistler corridor, in Technical Tours Guidebook, May 7, 1992, Geotechnique and Natural Hazards: A Symposium Sponsored by the Canadian Geotechnical Society and the Vancouver Geotechnical Society, Vancouver, British Columbia: 4-46. Irvine, T.N., and Baragar, W.R.A. 1971. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences, Vol. 8: 523-548. James, T.S., Clague, J.J., Wang, K., and Hutchinson, I. 2000. Postglacial rebound at the northern Cascadia subduction zone. Quaternary Science Reviews, Vol. 19: 1527-1541. Jaupart, C. 1998. Gas loss from magmas through conduit walls during eruption, in The Physics of Explosive Volcanic Eruptions. J.S. Gilbert and R.S.J. Sparks (eds.). Geological Society, London, Special Publications, Vol. 145: 73-90. 213 Jones, J.G. 1966. Intraglacial volcanos of southwest Iceland and their significance in the interpretation of the form of the marine basaltic volcanoes. Nature, Vol. 212: 586-588. Jones, J.G. 1969. Intraglacial volcanoes of the Laugarvatn region, south-west Iceland, I. Journal of the Geological Society of London, Vol. 124: 197-211. Jones, A.G., and Dumas, I. 1993. Electromagnetic images of a volcanic zone. Physics of the Earth and Planetary Interiors, Vol. 81: 289-314. J6nsson, S., Adam, N. , and Bjornsson, H. 1998. Effects of subglacial geothermal activity observed by satellite radar interferometry. Geophysical Research Letters, Vol. 25: 1059-1062. Jordan, P. 1987. Impacts of mass movement events on rivers in the southern Coast Mountains, British Columbia: summary report. Environment Canada, Water Resources Branch, IWD-HQ-WRB-55-87-3. Journeay, J.M. 1990. Structural and tectonic framework of the southern Coast Belt, British Columbia, in Current Research Part E; Geological Survey of Canada, Paper 90-1E: 183-197. Journeay, J.M., and Northcote, B.R. 1992. Tectonic assmeblages of the eastern Coast 214 Belt, southwest British Columbia, in Current Research Part A; Geological Survey of Canada, Paper 92-1A: 215-224. M l , M. , and McKenzie, D. 1996. The effect of deglaciation upon mantle melting beneath Iceland. Journal of Geophysical Research, Vol. 101: 21,815-21,828. Ke, P. 1992. A new approach to mass balance modeling: applications to igneous petrology; M.Sc. thesis, University of British Columbia, Vancouver, British Columbia: 153p. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2001. Preliminary petrography and chemistry of the Mount Cayley volcanic field, British Columbia, in Current Research Part A; Geological Survey of Canada, Paper A l l : 1-9. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2002a. Glaciovolcanism at Ember Ridge, Mount Cayley volcanic field, southwestern British Columbia, in Current Research Part A; Geological Survey of Canada, Paper A15: 1-7. Kelman, M.C., Russell, J.K., and Hickson, C.J. 2002b. Effusive intermediate glaciovolcanism in the Garibaldi Volcanic Belt, southwestern British Columbia, Canada, in J.L. Smellie and M.G. Chapman (eds.), Volcano-Ice Interaction on Earth and Mars. Geological Society, London, Special Publications, Vol. 202: 195-211. 215 Kerr, F.A. 1948. Lower Stikine and western Iskut River areas. Geological Survey of Canada Memoir 248. Kiver, E.P., and Steele, W.K. 1975. Firn caves in the volcanic craters of Mount Rainier, Washington. Bulletin of the National Speleological Society, Vol. 37: 45-55. Kokelaar, P. 1986. Magma-water interactions in subaqueous and emergent basaltic volcanism. Bulletin of Volcanology, Vol. 48: 275-289. Kretz, R. 1983. Symbols for rock-forming minerals. American Mineralogist, Vol. 68: 277-279. Lawrence, R.B. 1979. Garibaldi Group volcanic rocks of the Salal Creek map area (northern half), southwestern British Columbia. B.Sc. thesis. University of British Columbia, Vancouver, British Columbia: 83p. Lawrence, R.B:, Armstrong, R.L., and Berman, R.G. 1984. Garibaldi Group volcanic rocks of the Salal Creek area, southwestern British Columbia: alkaline lavas on the fringe of the predominantly calc-alkaline Garibaldi (Cascade) Volcanic Arc. Journal of Volcanology and Geothermal Research, Vol. 21: 255-276. LeBas, M.J., LeMaitre, M.W., Streckeisen, A., and Zanettin, B. 1986. A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of 216 Petrology, Vol. 27: 745-750. Leonard, E.M. 1995. A varve-based calibration of the Bridge River tephra fall. Canadian Journal of Earth Sciences, Vol. 32: 2098-2102. Lescinsky, D.T., and Fink, J.H. 2000. Lava and ice interaction at stratovolcanoes: use of characteristic features to determine past glacial extents and future volcanic hazards. Journal of Geophysical Research, Vol. 105: 23,711-23,726. Lescinsky, D.T., and Sisson, T.W. 1998. Ridge-forming, ice-bounded lava flows at Mount Rainier, Washington. Geology, Vol. 26: 351-354. Lliboutry, L. 1958a. Frottement sur le lit et mouvement pas saccades d'un glacier. Comptes Rendus Hebdomadaires des Seances de V'Academie des Sciences, Vol. 247: 228-230. Lliboutry, L. 1958b. Contribution a la theorie du frottement du glacier sur son lit. Comptes Rendus Hebdomadaires des Seances de V'Academie des Sciences, Vol. 247: 318-320. Lliboutry, L. 1983. Modifications to the theory of intraglacial waterways for the case of subglacial ones. Journal of Glaciology, Vol. 29: 216-226. 217 Long, P.E., and Wood, B.J. 1986. Structures, textures, and cooling histories of Columbia River basalt flows. Geological Society of America Bulletin, Vol. 97: 1144-1155. Lu, Z. 1988. Rock avalanches on Mount Cayley, British Columbia; M.Sc. thesis, University of Alberta, Edmonton, Alberta. Mathews, W.H. 1947. "Tuyas", flat-topped volcanoes in northern British Columbia. American Journal of Science, Vol. 245: 560-570. Mathews, W.H. 1951. The Table, a flat-topped volcano in southern British Columbia. American Journal of Science, Vol. 249: 830-841. Mathews, W.H. 1952. Ice-dammed lavas from Clinker Mountain, southwestern British Columbia. American Journal of Science, Vol. 250: 553-565. Mathews, W.H. 1958. Geology of the Mount Garibaldi map-area, southwestern British Columbia, Canada. Bulletin of the Geological Society of America, Vol. 69: 161-178. Mathews, W.H., Fyles, J.G., and Nasmith, H.W. 1970. Postglacial crustal movements in southwestern British Columbia and adjacent Washington state. Canadian Journal of Earth Sciences, Vol. 7: 690-702. Mclennan, J., Jull, M. , McKenzie, D., Slater L., and Gronvold, K. 2002. The link between 218 volcanism and deglaciation in Iceland. Geochemistry, Geophysics, Geosystems, Vol. 3: 25p. McPhie, J., Doyle, M. , and Allen, R. 1993. Volcanic Textures: a guide to the interpretation of textures in volcanic rocks. University of Tasmania, Centre for Ore Deposit and Exploration Studies, Launceston, Tasmania, Australia: 198p. Melnik, O., and Sparks, R.S.J. 1999. Nonlinear dynamics of lava dome extrusion. Nature, Vol. 402:37-41. Monger, J.W., and Journeay, J.M. 1992. Guide to the geology and tectonic evolution of the southern Coast Belt. A field guide to accompany the Penrose Conference on The Tectonic Evolution of the Coast Mountains Orogen, Vancouver. Geological Survey of Canada, Ottawa. Monger, J.W., and Journeay, J.M. 1994. Guide to the geology and tectonic evolution of the southern Coast Mountains. Geological Survey of Canada Open-File Report 2490. 77p. Murase, T., and McBirney, A.R. 1973. Properties of some common igneous rocks and their melts at high temperature. Geological Society of America Bulletin, Vol. 84: 3563-3592. Murphy, M.D., Sparks, R.S.J., Barclay, J., Carroll, M.R., and Brewer, T.S. 2000. 219 Remobilization of andesite magma by intrusion of mafic magma at the Soufriere Hills volcano, Montserrat, West Indies. Journal of Petrology, Vol. 41: 21-42. Nakada, S., and Motomura, Y. 1999. Petrology of the 1991-1995 eruption at Unzen: effusion pulsation and groundmass crystallization. Journal of Volcanology and Geothermal Research, Vol. 89: 173-196. Nakada, S., Shimizu, H., and Ohta, K. 1999. Overview of the 1990-1995 eruption at Unzen Volcano. Journal of Volcanology and Geothermal Research, Vol. 89: 1-22. Neuville, D.R., Courtial, P., Dingwell, D.B., and Richet, P. 1993. Thermodynamic and rheological properties of rhyolite and andesite melts. Contributions to mineralogy and Petrology, Vol. 113: 572-581. Newhall, C.G., and Melson, W.G. 1983. Explosive activity associated with the growth of volcanic domes. Journal of Volcanology and Geothermal Research, Vol. 17: 111-131. Nye, J.F. 1953. The flow of ice from measurements in glacier tunnels, laboratory experiments, and the Jungfraufirn borehole expedition. Proceedings of the Royal Society Series A, Vol.219: 477-489. Nye, J.F. 1976. Water flow in glaciers: jokulhlaups, tunnels, and veins. Journal of Glaciology, Vol. 17: 181-207. 220 Owen, L.A., Derbyshire, A., and Scott, C H . 2003. Contemporary sediment production and transfer in high-altitude glaciers. Sedimentary Geology, Vol. 155:13-36. Parrish, R.R. 1983. Cenozoic thermal evolution and tectonics of the Coast Range of British Columbia: 1. Fission track dating, apparent uplift rates, and patterns of uplift. Tectonics, Vol. 2: 601-631. Paterson, W.S.B. 1994. The physics of glaciers, 3rd edition. Pergamon Press, Oxford: 480p. Pearce, J.A. 1983. Role of the sub-continental lithosphere in magma genesis at active continental margins, in Continental Basalts and Mantle Xenoliths. C.J. Hawkesworth and M.J. Norry (eds.). Shiva, Natwich, England: 230-249. Peck, D.L., and Minakami, T. 1968. The formation of columnar joints in the upper part of Kilauean lava lakes, Hawaii. Geological Society of America Bulletin, Vol. 79: 1151-1166. Peckover, R.S., Buchana, D.J., and Ashby, D.E.T.F. 1973. Fuel-coolant interactions in submarine volcanism. Nature, Vol. 245: 307-308. Pillans, B. Chappell, J., and Naish, T.R. 1998. A review of the Milankovitch climatic beat: template for Plio-Pleistocene sea-level changes and sequence stratigraphy. 221 Sedimentary Geology, Vol. 122: 5-21. Porter, S.C, and Swanson, T.W. 1998. Radiocarbon age constraints on rates of advance and retreat of the Puget Lobe of the Cordilleran ice sheet during the last glaciation. Quaternary Research, Vol. 50: 205-213. Read, P.B. 1977. Meager Creek volcanic complex, southwestern British Columbia. Geological Survey of Canada, Paper, 77-la: 277-281. Read, P.B. 1990. Mount Meager Complex, Garibaldi Volcanic Belt, southwestern British Columbia. Geoscience Canada, Vol. 17: 167-170. Reiter, M. , Barroll, M.W., Minier, J., and Clarkson, G. 1987. Thermomechanical models for incremental fracturing in cooling lava flows. Tectonophysics, Vol. 142: 241-260. Robinson, G.D. 1947. The 1945 eruption in Tulik (Okmok) Caldera, Umnak Island, Alaska. Journal of the Washington Academy of Sciences, Vol. 37: 368-369. Roddick, J.C, and Souther, J.G. 1987. Geochronology of Neogene volcanic rocks in the northern Garibaldi Belt, British Columbia. Radiogenic age and isotopic studies; Report 1. Geological Survey of Canada, Paper, 87-2: 21-24. Roddick, J.A. and Woodsworth, G.J. 1977. Geology of Pemberton map area, British 222 Columbia. Geological Survey of Canada, Open File 482. Rogers, G.C. 1985. Variations in Cascade volcanism with margin orientation. Geology, Vol. 13: 495-498. Rohr, K.M.M., Govers, R., and Furlong, K.P. 1996. A new plate boundary model for the Pacific-North America-Juan de Fuca triple junction, in Slave-Northern Cordillera Lithospheric Evolution (SNORCLE) Transect and Cordilleran Tectonics Workshop, University of British Columbia, Lithoprobe Report 50: 213-214. Rose, W.I. 1972. Pattern and mechanism of volcanic activity at the Santiaguito lava dome, Guatemala. Bulletin of Volcanology, Vol. 36: 73-94. Rose, W.I. 1987. Volcanic activity at Santiaguito volcano, 1976-1984. in The Emplacement of Silicic Domes and Lava Flows. J.H. Fink (ed.) Geological Society of America Special Paper, Vol. 212: 17-28. Rothlisberger, H. 1972. Water pressure in intra- and subglacial channels. Journal of Glaciology, Vol. 62: 177-203. Ryan, M.P., and Sammis, C.G. 1978. Cyclic fracture mechanisms in cooling basalt. Geological Society of America Bulletin, Vol. 89: 1295-1308. 223 Ryder, J.M., Fulton, R.J., and Clague, J.J. 1991. The Cordilleran Ice Sheet and the glacial geomorphology of southern and central British Columbia. Geographic Physique et Quaternaire, Vol. 45: 365-377. Ryerson, F.J., and Watson, E.B. 1987. Rutile saturation in magmas: implications for Ti-Nb-Ta depletion in island-arc basalts. Earth and Planetary Science Letters, Vol. 86: 225-239. Schofield, J.T., Barnes, J.R., Crisp, D., Haberle, R.M., Larsen, S., Magalhaes, J.A., Murphy, J.R., Seiff, A., and Wilson, G. 1997. The Mars Pathfinder Atmospheric Structure Investigation/Meteorology (ASI/MET) Experiment. Science, Vol. 278: 1752-1758. Scott, W.E. 1990. Patterns of volcanism in the Cascade Arc during the past 15 000 years. Geoscience Canada, Vol. 17: 179-187. Shackleton, N.J., and Opdyke, N.D. 1973. Oxygen-isotope and palaeomagnetic stratigraphy of equatorial Pacific core V28-238: oxygen isotope temperatures and ice volumes on a 105 year and a 106 year scale. Quaternary Research, Vol. 3: 39-55. Shackleton, N.J., and Opdyke, N.D. 1976. Oxygen-isotope and palaeomagnetic stratigraphy of Pacific core V28-239, late Pliocene to latest Pleistocene, in Investigation of Late Quaternary Paleoceanography and Paleoclimatology. R.M. Cline and J.D. Hayes 224 (eds.). Geological Society of America Memoir, Vol. 145: 449-464. Shackleton, N.J., Berger, A. and Peltier, W.R. 1990. An alternative astronomical calibration of the Lower Pleistocene timescale based on ODP Site 677. Transactions of The Royal Society of Edinburgh: Earth Sciences, Vol. 81: 251-261. Sherrod, D.R., and Smith, J.G. 1990. Quaternary extrusion rates of the Cascade Range, northwestern United States and southern British Columbia. Journal of Geophysical Research, Vol. 95: 19,465-19,474. Shreve, R.L. 1972. Movement of water in glaciers. Journal of Glaciology, Vol. 11: 205-214. Sigvaldason, G.E. 1968. Structure and products of subaquatic volcanoes in Iceland. Contributions to Mineralogy and Petrology, Vol. 18: 1-16. Skilling, I.P. 2002. Basaltic pahoehoe lava-fed deltas: large-scale characteristics, clast generation, emplacement processes and environmental discrimination, in Volcano-Ice Interaction on Earth and Mars. J.L. Smellie and M.G. Chapman (eds.). Geological Society, London, Special Publications, Vol. 202: 91-113. Smellie, J.L. 2001. Lithofacies architecture and construction of volcanoes erupted in englacial lakes: Icefall Nunatak, Mount Murphy, eastern Marie Byrd Land, Antarctica. 225 Special Publication of the International Association ofSedimentologists, Vol. 30: 9-34. Smellie, J.L. 2002. The 1969 subglacial eruption at Deception Island (Antarctica): events and processes during an eruption beneath a thin glacier and implications for volcanic hazards, in Volcano-Ice Interaction on Earth and Mars. J.L. Smellie and M.G. Chapman (eds.). Geological Society of London, Special Publications, Vol. 202: 59-79. Smellie, J.L., Hole, M.J., and Nell, P.A.R. 1993. Late Miocene valley-confined subglacial volcanism in northern Alexander Island, Antarctic Peninsula. Bulletin of Volcanology, Vol. 55: 273-288. Smellie, J.L., and Skilling, LP. 1994. Products of subglacial volcanic eruptions under different ice thicknesses: two examples from Antarctica. Sedimentary Geology, Vol. 91: 115-129. Souther, J.G. 1980. Geothermal reconnaissance in the central Garibaldi Belt, British Columbia, in Current Research Part A; Geological Survey of Canada, Paper 80-1 A, 1-11. Souther, J.G. 1991. Hoodoo Mountain, in Volcanoes of North America. C.A. Wood and J. Kienle (eds.), Cambridge University Press, New York: 128-129. Souther, J.G., and Yorath, C.J. 1992. Neogene Assemblages, in Geology of the Cordilleran Orogen in Canada. H. Gabrielse and C.J. Yorath (eds.): 373-401. 226 Sparks, R.S.J., 1997. Causes and consequences of pressurization in lava dome eruptions. Earth and Planetary Science Letters, Vol. 150: 177-189. Sparks, R.S.J., and Pinkerton, H. 1978. Effects of degassing on rheology of basaltic lava. Nature, Vol. 276: 385-386. Sparks, R., Steven, J., Young, S.R., Barclay, J., Calder, E.S., Cole, P.D., Darroux, B., Davies, M.A., Druitt, T.H., Harford, C , Herd, R., James, M. , Lejeune, A.M. , Loughlin, S., Norton, G., Skerrit, G., Stasiuk, M.V., Stevens, N.S., Toothill, J., Wadge, G., and Watts, R. 1998. Magma production and growth of a lava dome of the Soufriere Hills Volcano, Montserrat, West Indies; November 1995 to December 1997. Geophysical Research Letters, Vol. 25: 3421-3424. Sparks, R.S.J., and Young, S.R. 2002. The eruption of Soufriere Hills Volcano, Montserrat: overview of scientific results, in The Eruption of Soufriere Hills Volcano, Montserrat, from 1995 to 1999. T.H. Druitt and B.P. Kokelaar (eds.). Geological Society, London: 45-69. Stasiuk, M.V., Hickson, C.J., and Mulder, T. 2003 The vulnerability of Canada to volcanic hazards. Natural Hazards, Vol. 28: 563-589. Stasiuk, M.V., Russell, J.K., and Hickson, C.J. 1996. Distribution, nature, and origins of 227 the 2400 BP eruption products of Mount Meager, British Columbia: linkages between magma chemistry and eruption behaviour. Geological Survey of Canada Bulletin 486: 27p. Stewart, M.V., Russell, J.K., and Hickson, C.J. 2002. Revised stratigraphy of the Pebble Creek Formation, British Columbia; evidence for interplay between volcanism and mountainous terrain, in Current Research Part A; Geological Survey of Canada, Paper 2002-E3: 1-7. Sun, S.S., and McDonough, W.F. 1989. Chemical and isotopic systematics of oceanic basalts; implications for mantle composition and processes. Geological Society Special Publications, Vol. 42: 313-345. Swanson, S.E., Naney, M.T., Westrich, H.R., and Eichelberger, J.C. 1989. Crystallization history of Obsidian Dome, Inyo Domes, California. Bulletin of Volcanology, Vol. 51: 161-176. Tbirlwall, M.F., Smith, T.E., Graham, A.M. , Theodorou, N. , Hollings, P., Davidson, J.P., and Arculus, R.J. 1994. High field strength element anomalies in arc lavas: source or process? Journal of Petrology, Vol. 35: 819-838. Thompson, P.H. 1968. Remnant magnetism in the Garibaldi volcanics near Squamish, British Columbia; B.Sc. thesis, University of British Columbia, Vancouver, British 228 Columbia: 19p. Thornton, CP. , and Tuttle, O.F. 1960. Chemistry of igneous rocks I. Differentiation index. American Journal of Science, Vol. 258: 664-684. Tuffen, H., Gilbert, J.S., and McGarvie, D.W. 2001. Products of an effusive subglacial rhyolite eruption: Blahnukur, Torfajokull, Iceland. Bulletin of Volcanology, Vol. 63: 179-190. Tuffen, H., McGarvie, D.W., Gilbert, J.S., and Pinkerton, H. 2002a. Physical volcanology of a subglacial-to-emergent rhyolitic tuya at Raudafossafjoll, Torfajokull, Iceland, in J.L. Smellie and M.G. Chapman (eds.), Volcano-Ice Interaction on Earth and Mars. Geological Society, London, Special Publications, Vol. 202: 213-236. Tuffen, H., Pinkerton, H., Gilbert, J.S., and McGarvie, D.W. 2002b. Melting of the glacier base during a small-volume subglacial rhyolite eruption: evidence from Blahnukur, Iceland. Sedimentary Geology, Vol. 149: 183-198. Voight, B., Sparks, R.S.J., Miller, A.D., et al. 1999. Magma flow instability and cyclic activity at Soufriere Hills Volcano, Montserrat, B.W.I. Science, Vol. 283: 1138-1142. Walcott, D. Neogene tectonics and kinematics of western North America. Tectonics, Vol. 12: 326-333. 229 Walder, J.S., and Hallet, B. 1979. Geometry of former subglacial water channels and cavities. Journal of Glaciology, Vol. 23: 335-346. Weertman, J. 1957. Deformation of floating ice shelves. Journal of Glaciology, Vol. 3: 38-42. Werner, R., Schmincke, H.-U., and Sigvaldason, G. 1996. A new model for the evolution of table mountains: volcanological and petrological evidence from Herdubreid and Herdubreidart8gl volcanoes (Iceland). Geologische Rundschau, Vol. 85: 390-397. Wheeler, J.O., Brookfield, A.J., Gabrielse, H., Monger, J.W.H., Tipper, H.W., and Woodsworth, G.J. 1991. Terrane map of the Canadian Cordillera. Geological Survey of Canada, Map 1713 A. White, J.D.L. and Houghton, B. 2000. Surtseyan and related phreatomagmatic eruptions, in Encyclopedia of Volcanoes. H. Sigurdsson (ed.). Academic Press: 495-511. Wilson, L., and Head, J.W. 1994. Mars: Review and analysis of volcanic eruption theory and relationships to observed landforms. Reviews of Geophysics, Vol. 32: 221-263. Wright, A.C. 1980. Landforms of McMurdo Volcanic Group, southern foothills of Royal Society Range, Antarctica. New Zealand Journal of Geology and Geophysics, Vol. 23: 230 605-613. Yamagishi, H., and Dimroth, E. 1985. A comparison of Miocene and Archean rhyolite hyaloclastites: evidence for a hot and fluid lava. Journal of Volcanology and Geothermal Research, Vol. 23: 337-355. 231 Appendix 1. Compilation of volcanic events used in Figure 2.2. 'Location Date used in Figure 4 Measured or estimated date(s) Dating method Source Mount Meager 2360 BP 2360 BP radiocarbon; varves Clague et al. 1995; Leonard 1995 Ring Creek flow (MGVF) 9800 BP between 10,650±70 BP and 9360±160 BP bracketing radiocarbon dates Brooks and Friele 1992 Dalton Dome (MGVF) 8000-10,000 BP post-Fraser Glaciation field observations Mathews 1958; Green et al. 1988 Paul Ridge (MGVF) 8000-10,000 BP post-Fraser Glaciation field observations Mathews 1958 Slag Hill (Qvpaul4) 10,000-12,000 BP late/post-Fraser Glaciation field observations this thesis Tricouni Southwest (MCVF) 10,000-12,000 BP late Fraser Glaciation field observations this thesis Tricouni Southeast (MCVF) (6 eruptions) 10,000-12,000 BP late Fraser Glaciation field observations this thesis Ring Mountain Northwest (MCVF) 10,000-12,000 BP late Fraser Glaciation field observations this thesis Atwell Peak stage (MGVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews 1958 Eenostuck mass (MGVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews 1958; Green et al. 1988 Glacier Pikes (MGVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews 1958; Green et al. 1988 Table (GLVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews 1951 Barrier/Culliton Creek flows (GLVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews, 1952 Cheakamus Valley basalts (youngest) (GLVF) 10,000-12,000 BP late Fraser Glaciation field observations Mathews 1958 Pali Dome West (LQvpau6) (MCVF) (3 eruptions) 10,000-12,000 BP late Fraser Glaciation field observations this thesis Pali Dome East (LQvpaul4) (MCVF) 10,000-12,000 BP late Fraser Glaciation field observations this thesis Lower Cauldron Dome (MCVF) (2 eruptions) 10,000-12,000 BP late Fraser Glaciation 2field observations this thesis Ring Mountain (MCVF) 12,000-15,000 BP Fraser Glaciation field observations this thesis Upper Cauldron Dome (MCVF) (5 eruptions) 12,000-15,000 BP Fraser Glaciation field observations this thesis Ember Ridge (MCVF) (8 eruptions) 12,000-15,000 BP Fraser Glaciation field observations this thesis Little Ring Mountain (MCVF) 12,000-15,000 BP Fraser Glaciation field observations this thesis Slag Hill tuya (MCVF) 12,000-15,000 BP Fraser Glaciation field observations this thesis Slag Hill (Qvpaul2.3) (2 eruptions) 12,000-15,000 BP Fraser Glaciation 2field observations this thesis 'Location Date used in Figure 4 Measured or estimated date(s) Dating method Source Pali Dome East (LQvpaul,) (MCVF) >65,000 BP >50,000 BP field observations this thesis Cheakamus Valley basalts (oldest) (GLVF) 65,000 BP >34,200 BP; 0.05±0.05 Ma 3 C, K-Ar, and field observations Green etal. 1988 Watts Point 0.09 0.09±0.03 Ma; 0.13±0.03 Ma K-Ar Green etal. 1988 Black Tusk (GLVF) 0.09 0.09±0.08 Ma; 0.04±0.04 Ma K-Ar Green etal. 1988 Elaho valley flows 0.09 Ma 0.09±0.06 Ma K-Ar Green etal. 1988 Capricorn complex (Mount Meager) 0.1 Ma 0.1±0.02 Ma K-Ar Green etal. 1988 Plinth complex (Mount Meager) 0.11 Ma 0.11±0.02Ma K-Ar Green etal. 1988 Cinder Cone (GLVF) 0.11 Ma 0.11±0.03 Ma; 0.04±0.04 Ma K-Ar Green etal. 1988 Elaho valley olivine basalt flows 0.14 Ma 0.14±0.1 Ma K-Ar Green etal. 1988 Devastation Glacier (Mount Meager) 0.15 Ma 0.15±0.1 Ma K-Ar Green etal. 1988 Black Tusk (GLVF) 0.17 Ma 0.17±0.04Ma K-Ar Green etal. 1988 Black Tusk (GLVF) 0.21 Ma 0.2H0.04 Ma K-Ar Green et al. 1988 Cheekye stage (MCVF) 0.26 Ma 0.26±0.16 Ma; 0.22±0.22 Ma; 0.26±0.13Ma K-Ar Green etal. 1988 Mount Price (GLVF) 0.3 Ma 0.3±0.2 Ma K-Ar Green etal. 1988 Mount Cayley (MCVF) 0.31 Ma 0.3U0.05 Ma K-Ar Green etal. 1988 Silverthrone Mountain 0.4 Ma 0.4±0.1 Ma K-Ar Green etal. 1988 Pylon complex (Mount Meager) 0.5 Ma 0.5±0.1 Ma K-Ar Green etal. 1988 MGVF - Round Mountain complex 0.5 Ma 0.51±0.04Ma; 0.46±0.02 Ma; 0.67±0.04 Ma K-Ar Green etal. 1988 Salal Glacier volcanic field -southern flow 0.59 Ma 0.59±0.05 Ma K-Ar Lawrence et al. 1984 Mashiter paleovalley sequence (MGVF) 0.5 Ma <0.75±0.02 Ma magnetic polarity Thompson 1968 Silverthrone Mountain 0.75 Ma 0.75±0.08 Ma K-Ar Green etal. 1988 'Location Date used in Figure 4 Measured or estimated date(s) Dating method Source Pylon complex (Mount Meager) 0.9 Ma 0.9±0.2 Ma K-Ar Green etal. 1988 Salal Glacier volcanic field -northern flow 0.97 Ma 0.97±0.05 Ma K-Ar Lawrence et al. 1984 Pylon complex (Mount Meager) 1.0 Ma 1.0±0.1 Ma; 0.7±0.4 Ma K-Ar Green etal. 1988 Black Tusk (GLVF) 1.1 Ma l.l±0.06Ma K-Ar Green etal. 1988 Mount Price (GLVF) 1.2 Ma 1.2±0.1 Ma K-Ar Green etal. 1988 Black Tusk (GLVF) 1.3 Ma 1.3±0.1 Ma; 1.3±0.07Ma K-Ar Green etal. 1988 Mount Meager 1.9 Ma 1.9±0.2 Ma K-Ar Green etal. 1988 Mount Meager 2.0 Ma 2.0±0.1 Ma K-Ar Green etal. 1988 Mount Meager 2.2 Ma 2.2±0.1 Ma K-Ar Read 1977; Evans 1992 Franklin Glacier 2.2 Ma 2.2±0.1 Ma K-Ar Green etal. 1988 'MGVF = Mount Garibaldi volcanic field, MCVF = Mount Cayley volcanic field; GLVF = Garibaldi Lake volcanic field 2K-Ar dates by Green et al. (1988) are older than the ages postulated based on field observations. 3C = bracketing radiocarbon date(s) Appendix 2a. MCVF sample locations, given as UTM coordinates. sample number coordinates centre MK-01-00-10 476040 5565118 Ring Mountain Northwest MK-01-00-11 476085 5565229 Ring Mountain Northwest MK-01-00-12 476085 5565229 Ring Mountain Northwest MK-01-00-13 476820 5564884 Ring Mountain Northwest MK-01-00-6 476237 5565237 Ring Mountain Northwest MK-01-00-7 476216 5565081 Ring Mountain Northwest MK-01-00-9 476245 5565081 Ring Mountain Northwest MK-01-10-2a 484358 5536703 Tricouni Southeast MK-01-10-2b 484414 5536661 Tricouni Southeast MK-01-10-3 484176 5536550 Tricouni Southeast MK-01-10-4 484453 5535756 Tricouni Southeast MK-01-10-5a 484255 5535645 Tricouni Southeast MK-01-10-5b 484309 5535550 Tricouni Southeast MK-01-10-6 484508 5535680 Tricouni Southeast MK-01-10-7 483757 5536278 Tricouni Southwest MK-01-10-8 484426 5535191 Tricouni Southeast MK-01-10-9 483705 5534968 Tricouni Southwest MK-01-11-2 481478 5546081 Ember Ridge West MK-01-11-3a 481264 5546118 Ember Ridge West MK-01-11-3b 481264 5546118 Ember Ridge West MK-01-12-3 481791 5547049 Ember Ridge Northwest MK-01-13-1 481626 5544259 Ember Ridge Southwest MK-01-13-2 481869 5544199 Ember Ridge Southwest MK-01 -13-3a 481906 5543694 Ember Ridge Southwest MK-01-13-3b 481863 5543666 Ember Ridge Southwest MK-01-14-1 482207 5544385 Ember Ridge Southwest MK-01-15-2 481660 5547349 Ember Ridge Northwest MK-01-16-2a 481102 5554934 Pali Dome East MK-01-16-2b 481102 5554934 Pali Dome East MK-01-16-2c 481102 5554934 Pali Dome East MK-01-16-3 481040 5554311 Pali Dome East MK-01-17-1 480998 5554006 Pali Dome East MK-01-17-2 481069 5553635 Pali Dome East MK-01-17-3 481238 5553703 Pali Dome East MK-01-18-2 483363 5547332 Ember Ridge North MK-01-19-1 482852 5547719 Mount Fee MK-01-19-2 482862 5547170 Ember Ridge North MK-01-20-1 482773 5548454 Mount Fee MK-01-20-2 482708 5548552 Mount Fee MK-01-20-3 482338 5548482 Mount Fee MK-01-20^a 482135 5548510 Mount Fee MK-01-20-4b 482088 5548547 Mount Fee MK-01-20-5 482144 5548075 Mount Fee MK-01-20-6 482505 5547705 Mount Fee MK-01-20-7 482607 5547876 Mount Fee MK-01-2-1 478317 5547751 Mount Cayley MK-01-21-1 484014 5548487 Mount Fee MK-01-21-3 483056 5548570 Mount Fee 235 sample number coordinates centre MK-01-2-2 478424 5547770 Mount Cayley MK-01-22-2a 482186 5549241 Mount Fee MK-01-22-2b 482186 5549241 Mount Fee MK-01-2-3 478424 5547770 Mount Cayley MK-01-23-1 484506 5546646 Ember Ridge East MK-01-23-2 481431 5551606 Mount Cayley MK-01 -2-4 478168 5547858 Mount Cayley MK-01-24-2 477385 5553634 Pali Dome West MK-01-24-3 477646 5554126 Pali Dome West MK-01-24-4a 477761 5554435 Pali Dome West MK-01-24-4b 477681 5554209 Pali Dome West MK-01-24-4C 477761 5554435 Pali Dome West MK-01-24-4d 477878 5554441 Pali Dome West MK-01-24-4e 477836 5554479 Pali Dome West MK-01 -24-4f 477878 5554441 Pali Dome West MK-01-24-5 477503 5554150 Pali Dome West MK-01-25-2 476634 5555242 Cauldron Dome MK-01-25-3 476727 5555170 Cauldron Dome MK-01-25-4 476955 5555159 Cauldron Dome MK-01-25-5 476764 5554989 Cauldron Dome MK-01-25-6 475977 5554919 Cauldron Dome MK-01-25-7 475827 5555131 Cauldron Dome MK-01-26-1 476150 5556752 Cauldron Dome MK-01-26-2a 476710 5557152 Cauldron Dome MK-01-26-2b 476710 5557152 Cauldron Dome MK-01-26-3 476713 5556610 Cauldron Dome MK-01-27-2 479794 5561325 Slag Hill tuya MK-01-27-2b 479640 5561072 Slag Hill tuya MK-01 -27-3a 479966 5561048 Slag Hill tuya MK-01-27-3b 479966 5561048 Slag Hill tuya MK-01-27-3C 479891 5561070 Slag Hill tuya MK-01-27-4a 479979 5561230 Slag Hill tuya MK-01-27-4b 479913 5561235 Slag Hill tuya MK-01-27-4C 479968 5561161 Slag Hill tuya MK-01-27-4d 479918 5561163 Slag Hill tuya MK-01-28-1 478347 5561084 Slag Hill MK-01-28-2a 478017 5560626 Slag Hill MK-01-28-2b 478017 5560626 Slag Hill MK-01-28-3a 478115 5559941 Slag Hill MK-01-28-3b 478005 5560064 Slag Hill MK-01-28-4a 478109 5559293 Slag Hill MK-01-28-4b 478207 5559160 Slag Hill MK-01-28-4C 478233 5559115 Slag Hill MK-01-28-4d 478129 5559173 Slag Hill MK-01-29-1 a 478729 5563340 Ring Mountain MK-01-29-1 b 478729 5563340 Ring Mountain MK-01-29-1 c 478729 5563340 Ring Mountain MK-01-29-2 478560 5563780 Ring Mountain MK-01-29-3 478381 5563820 Ring Mountain MK-01 -29-4 477310 5563296 Ring Mountain 236 sample number coordinates centre MK-01-30-1 478778 5558651 Slag Hill MK-01-30-2 478664 5559700 Slag Hill MK-01-31-2a 477550 5569721 Little Ring Mountain MK-01-31-2b 477550 5569721 Little Ring Mountain MK-01-31-3 477808 5569793 Little Ring Mountain MK-01-31-4 477673 5570224 Little Ring Mountain MK-01-31-5 477054 5569584 Little Ring Mountain MK-01-32-1 478164 5553309 Pali Dome West MK-01-32-2a 478324 5552552 Pali Dome West MK-01-32-2b 478700 5552510 Pali Dome West MK-01-32-3a 478260 5553395 Pali Dome West MK-01-32-3b 478309 5553550 Pali Dome West MK-01-4-3 483920 5543482 Ember Ridge Southeast MK-01-4-4 483973 5543637 Ember Ridge Southeast MK-01-4-5 483879 5543520 Ember Ridge Southeast MK-01-5-1 483804 5543611 Ember Ridge Southeast MK-01-5-2 483786 5543639 Ember Ridge Southeast MK-01-5-3 483787 5543732 Ember Ridge Southeast MK-01-7-2 482396 5534819 Tricouni Southwest MK-01-8-2 482337 5534848 Tricouni Southwest MK-01-9-4 483225 5536585 Tricouni Southwest MK-01-9-5 483192 5537071 Tricouni Southwest MK-01-9-6 483192 5537613 Tricouni Southwest MK-02-7-2 487965 5535037 Tricouni Southeast MK-02-7^ 487559 5535487 Tricouni Southeast MK-02-8-4 487361 5535568 Tricouni Southeast MK-02-8-5a 487489 5535784 Tricouni Southeast MK-02-9a-1 485719 5536410 Tricouni Southeast MK-02-9a-3 484673 5538159 Tricouni Southeast MK-02-9a-4 485394 5537476 Tricouni Southeast MK-03-10-6a 484527 5535625 Tricouni Southeast MK-03-10-€b 484361 5535588 Tricouni Southeast MK-03-10-6C 484492 5535345 Tricouni Southeast F-lake 487664 5535155 Tricouni Southeast 1HHB-98-KR-2 477912 5559878 Slag Hill HHB-98-KR-5 478186 5563350 Ring Mountain HHB-98-KR-6 478133 5563283 Ring Mountain HHB-98-KR-7 478096 5563230 Ring Mountain HHB-98-KR-8-1 477525 5569777 Little Ring Mountain HHB-98-KR-8-2 477467 5569707 Little Ring Mountain HHB-98-KR-8-3 477424 5569788 Little Ring Mountain HHB-98-KR-8-4 477481 5569858 Little Ring Mountain HHB-98-MC-1A 477997 5559599 Slag Hill 1Samples with the prefix HHB were collected in 1998 by C.J. Hickson and J.K. Russell. 237 Appendix 2b Petrography of nonfragmental units of the MCVF. unit 1 phenocrysts zgroundmass 3xenocrysts/ xenoliths 4voids Sequence of crystallization 6Qvdp 15-20% eu-anh 0.25-2 mm PI (common sieve texture/zoned sieve texture, rare embayed margins, rare comp zoning); 1-2% eu-sub 0.15-0.75 mm Hbl (no oxide haloes); <1% sub 0.05-0.3 mm colourless Opx; <1% 0.4 mm Bt 70% glass + crypto; trachytic PI; Opx; Fe-Ti; Cpx. 0.20- 0.6 mm rounded Qtz (± Cpx fringe) non Opx + Fe-Ti, PI, Hbl (Bt unknown, poss xeno) 7Qvdb 5-15% eu-anh 0.01-4.25 seriate mm PI (common inclusions/zoned inclusions, common sieve texture, common cont/disc/osc zoning, common embayed rims, rare opaque incl); 1-10% eu-sub 0.2-3 mm Hbl (oxide-vfg haloes or pseudo); 1-2% eu-sub 0.01-0.15 mm Opx (strong pink-tan pleo); <1% sub-eu 0.05-0.15 mm Fe-Ti 60% crypto + glass; PI; Cpx; Fe-Ti PI aggregates and granitic fragments (up to tens of cm); rounded Qtz ± Cpx fringe; Pl-Px-oxide aggregates (poss. cognate) non Fe-Ti, Opx + PI, Hbl + PI Qvpdl 15-20%0.1-5 mm seriate PI (common inclusion zoning, sieve texture, disc/cont comp zoning); 2-3% sub-anh 0.25-1.25 mm Hbl (oxide-dominated haloes or oxide-vfg pseudo); <1% 0.05-3 mm colourless Opx; ?1% anh 0.25-1.25 mm Bt (partial-total pseudo by Fe-Ti); <1% eu-anh 0.05-0.15 Fe-Ti 50-60% crypto + glass; crudely trachytic PI; Opx; Fe-Ti rounded Qtz (± Cpx fringe); Qtz ± Opx aggregates; PI; granitic fragments non, <1% microfractures Fe-Ti, PI + Opx, Hbl (Bt unknown) Qvpd2 15-20% eu-sub 0.1-6.25 mm seriate PI (common cont/osc comp zoning, rare incl, rare sieve texture); 2-3% anh 0.25-1.2 mm dark brown Bt (partial oxide-vfg replacement); 1% 0.005-0.75 mm seriate colourless Opx sugary-textured crypto; PI; Opx; Fe-Ti 0.5-0.75 mm rounded Qtz; rare granitic fragments non, <1% fractures Opx, PI, PI + Bt Qvpaul ,.3 1-2% eu-anh 0.2-1.5 mm Hbl (oxide-vfg haloes or pseudo); 0-2% eu-sub 0.05-1.75 mm (rarely 5 mm) Cpx (common clots); ?1% eu-anh 0.15-2.5 mm PI (common sieve texture, clots, and incl; rare cont comp zoning); <1% 0.05-0.15 mm eu-anh Fe-Ti 5-20% glass + crypto; trachytic PI; Fe-Ti; Cpx (common); Opx (rare) 0.1 mm to macro granitic fragments; ?2.5 mm rounded Qtz 0-2%; rarely 15% Cpx, Hbl (PI unknown) Qvpaul4 <1% eu-sub 0.15-0.45 mm Opx; <1% 0.1-0.35 mm PI in 2 size groups; <1% sub-anh 0.5-1.4 mm Hbl (oxide-vfg pseudo) 15% glass + crypto; trachytic PI; Fe-Ti 15% Opx, Hbl (PI unknown) unit phenocrysts 2groundmass 3xenocrysts/ xenoliths 4voids Sequence of crystallization LQvpa7 10-14% eu-sub 0.1-2.5 mm PI (common cont/disc comp zoning and sieve texture); 5-15% eu-sub 0.1-8 mm Hbl (oxide haloes or total pseudo); <1% eu 0.1-1.8 mm Fe-Ti 30-50% glass + crypto; crudely trachytic PI; Fe-Ti granitic fragments up to >1 m ?10% Fe-Ti, PI, PI + Hbl LQvpaul, 1% eu-sub 0.25-3 mm PI (rare cont/disc comp zoning); <1% sub-anh 0.05-0.2 mm Fe-Ti 40% crypto; trachytic PI non Fe-Ti, PI LQvpaul2 5% eu-anh 0.1-2.25 mm PI (common sieve texture, incl, and clots ± Hbl); 1-3% sub-anh 0.15-3 mm Hbl (oxide-vfg haloes); <1% eu-anh 0.1-0.5 mm Cpx; <1% sub 0.05-0.2 mm Fe-Ti 10% glass + crypto (rarely 60% crypto); crudely trachytic PI; Fe-Ti LQvpaul1 fragments and felsic-/mafic-dominated fragments up to 1.5 cm non Fe-Ti, PI, PI + Hbl (Cpx unknown) LQvpaul3 5% eu-anh 0.15-1 mm seriate PI (common incl and sieve texture, rare disc comp zoning); <1% sub Cpx (rarely as incl in PI); <1% eu-anh 0.25-1 mm Opx (common clots); <1% anh ?0.1 mm Fe-Ti (common as incl in PI); <1% eu diamond-shaped pseudo of oxides and vfg (probably former Hbl) 15% crypto; trachytic PI; Px; Fe-Ti <1% ?1 cm mafic-dominated fragments non Cpx + Fe-Ti, Opx, PI, Hbl LQvpaul4 10-12% (top) to 4% (base) sub-anh 0.15-1.5 mm PI (common sieve texture and incl); 8-10% (top) to 2% (base) eu 0.15-2 mm Hbl (oxide-vfg haloes); 1% sub 0.25-1.25 Opx (common clots ± PI); <1% sub-anh ?0.15 mm Fe-Ti; <1% anh 0.15-0.75 Cpx ?40% glass + crypto; trachytic PI; Px; Fe-Ti 2-3%-proximaliy; <1% distally Cpx + Fe-Ti, PI + Opx, PI + Opx + Hbl LQvpau21 1-10% sub-anh 0.2-2 mm PI (common disc/cont comp zoning, common clots with Ol); 1% eu-sub 0.1-1.25 mm Ol; ?1% sub 0.15-0.75 Opx; <1% eu 0.15-0.5 diamond-shaped pseudo of oxides and vfg (probably former Hbl); <1% anh 0.05-0.35 mm Fe-Ti extremely variable proportions of glass + crypto; trachytic PI; Opx; Fe-Ti rounded Qtz; ?0.4 mm granitic fragments 3-5% (rarely higher) Ol, Ol + Fe-Ti + Opx, Ol + Fe-Ti + Opx + PI, PI LQvpau25 1-2% anh 0.75-3 mm PI (common embayments, incl, sieve texture, cont comp zoning, and clots); 1-2% diamond-shaped 0.05-0.75 mm pseudo of oxides and vfg (probably former Hbl); <1% eu-sub 0.2-0.45 seriate Cpx (rare replacement by Px, Fe-Ti, vfg); <1% 0.1-0.16 mm seriate Fe-Ti; 20-30% crypto ± glass ± Fe-Ti; trachytic PI; Cpx granitic fragments; rounded Qtz (with Cpx fringe); <1% Opx non unknown unit 1 phenocrysts 2groundmass 3xenocrysts/ xenoliths 4voids 5sequence of crystallization LQvpau32 3-15% 0.1-3 mm PI (2 groups: [1] sub-anh and sieve texture, [2] eu-sub, larger, disc/cont/sector comp zoning, ind, embayments); <1-5% eu-anh 0.1-0.75 mm Opx (common clots with PI); <1% eu 0.1-0.8 mm Fe-Ti; <1% sub 0.1-0.9 mm pseudo of oxides (probably former Hbl) 5-20% glass + crypto; trachytic PI; Opx; Fe-Ti ?1.5 mm rounded Qtz (± Cpx fringe); clots of Opx + Fe-Ti + Ol (possible fragments from LQvpau2 t); granitic fragments up to 8 mm 0-10% Opx, Opx + Fe-Ti, Opx + Fe-Ti + PI + Hbl LQvpal 10% eu-sub 0.2-2.5 mm Hbl (common clots with PI; oxide-vfg haloes); 1% 0.6-4 mm PI (2 groups: [1] eu-sub, rare incl, comp zoning, [2] eu-anh, common incl, common sieve texture); <1% sub-anh 0.1-1 mm Cpx; <1% sub-anh 0.05-0.15 mm Fe-Ti 30% glass + crypto; crudely trachytic PI; Fe-Ti; Px granitic fragments <1% Fe-Ti, Cpx, PI, PI + Hbl LQvpa2 10% eu-sub 0.5-6 mm seriate PI (common sieve texture, incl, and cont/disc comp zoning, clots up to 5.5 mm); 5% eu-sub 0.15-1.5 mm Hbl (oxide-vfg haloes and pseudo) 45% glass + crypto; trachytic PI; Fe-Ti rounded Qtz non PI + Hbl, PI LQvpa3 10-15% eu-anh 0.1-2 mm PI (common cont comp zoning, embayments, and incl; rare sieve texture); 3-4% eu-sub 0.06-0.75 mm Opx (common clots); 1% sub-anh 0.15-2 mm Hbl (oxide-vfg haloes and pseudo); 1-2% anh 0.1-0.2 mm Fe-Ti 15-40% glass + crypto; trachytic PI; Fe-Ti granitic fragments non Opx + Fe-Ti, Opx + PI, PI + Hbl LQvpa4 3-10% eu-sub 0.1-3 mm PI (common sieve texture, comp zoning, and zoning of inclusions); 1-5% eu-sub 0.1-1.25 mm seriate Cpx; 1% sub-anh 0.05-1 mm Opx (clots with Cpx and PI); <1% eu-sub 0.02-0.08 mm seriate Fe-Ti ?15% glass + crypto + Fe-Ti; trachytic PI; Opx rounded Qtz, granitic fragments, microcline 3-15% Cpx + Fe-Ti, Opx + Fe-Ti, PI LQvpa5 20% eu-sub 0.1-1.75 mm seriate PI (common clots and cont/disc comp zoning); 1% sub-anh 0.15-5.25 mm Hbl (oxide-vfg haloes); <1% sub-anh 0.1-0.2 mm Fe-Ti glass (up to 30%) + crypto; trachytic PI 2-10% Fe-Ti, PI, Hbl LQvpa6 1% eu-sub 0.25-1 mm PI (common sieve texture, incl, and comp zoning); <1% sub-anh 0.14-0.3 mm Cpx; <1% eu-sub 0.25-1 mm Opx 40% crypto; trachytic PI; Px; Fe-Ti 1-2% PI, Cpx + Opx unit 1 phenocrysts 2groundmass 3xenocrysts/ xenoliths 4voids Sequence of crystallization LQvpau5 <1-2% sub-anh 0.15-1 mm PI (common sieve texture and embayments); 1-2% sub-anh 0.15-1.25 mm Cpx (common clots); <1% sub-anh 0.1-1.25 Hbl (oxide-vfg pseudo or haloes) 35-40% glass + crypto; trachytic PI; Fe-Ti; Px granitic fragments; rounded Qtz (± Cpx fringe); rounded Qtz aggregates; Bt <1-10% Cpx, PI + Cpx (Hbl unknown) LQvpau62 10% eu-anh 0.25-7 mm PI (common disc/cont comp zoning, incl, sieve texture, and clots); <1% anh ?1.25 mm Bt (oxide-vfg haloes); <1% eu-sub 0.05-0.6 mm Fe-Ti; <1% eu 70.35 mm Opx; <1% sub-anh 0.15-2 mm oxide-vfg pseudo (probably former Hbl) 720% glass + crypto; PI; Fe-Ti; Opx granitic fragments, rounded Qtz <1-3% Fe-Ti, PI, PI + Bt (Hbl and Opx unknown) LQvpau63 10% eu-anh 0.25-7 mm PI (common disc/cont comp zoning, incl, sieve texture, and clots); <1% anh 71.25 mm Bt (minor oxide-vfg haloes); <1% eu-sub 0.05-0.6 mm Fe-Ti; 1% eu 70.35 mm Opx; <1% sub-anh 0.15-2 mm oxide-vfg pseudo (probably former Hbl) 720% glass + crypto; PI; Fe-Ti; Opx; Cpx granitic fragments, rounded Qtz 1-2% Fe-Ti, PI, PI + Bt (Hbl and Opx unknown) LQvpau7 1-3% anh 0.15-5 mm PI (common sieve texture; rare incl, embayments, comp zoning); <1% sub-eu 0.05-1.75 mm Cpx (rare clots); ?1% eu-anh 0.05-1.75 mm Hbl (pseudo by oxides-vfg); <1% anh 70.25 mm Fe-Ti 15-25% glass + crypto; trachytic PI; Cpx; Fe-Ti PI aggregates and granitic fragments (up to tens of cm); rounded Qtz ± Cpx fringe; Pl-Px-oxide aggregates (poss. cognate) <1-2% unknown LQvpa8 <1% anh 0.25-2 mm PI (rare disc comp zoning); <1% anh 0.15-1 mm Ol; <1% 70.06 mm Fe-Ti; <1% eu-anh 70.15 mm Opx 10-15% glass + crypto; crudely trachytic PI; Px; Fe-Ti granitic fragments <1-4% unknown LQvpadu, 71% eu-anh 0.15-1.4 mm seriate PI (rare cont comp zoning, incl, and zoning of incl); 71% sub-anh 0.05-0.75 mm Hbl (oxide-vfg pseudo) 20-70% glass + crypto; trachytic PI; Px; Fe-Ti Qtz; Qtz-PI aggregates 3-7% unknown LQvpadu2 1-5% eu-sub 0.2-1.4 mm seriate PI (common zoning of incl); <1% anh 0.1-0.7 mm green-brown Hbl 5-10% glass + crypto; trachytic PI; Fe-Ti; Opx rounded Qtz 5% unknown unit 1 phenocrysts 2g round mass 'xenocrysts/ xenoliths 4voids 'sequence of crystallization LQvpadu3 (1 of 2 flows) 1-2% sub-eu 0.02-0.04 mm seriate PI (rare incl); 1-2% sub 0.05-0.5 mm Hbl (mostly pseudo by oxide-vfg) 75% crypto + glass; crudely trachytic PI; Fe-Ti rounded Qtz aggregates unknown LQvpadu4 10% eu-anh 0.25-5 mm seriate PI (common incl, sieve texture, and cont/disc comp zoning); 1% sub-eu 0.15-1 mm Opx; 1% sub-anh 0.15-3 mm Hbl (common oxide-vfg pseudo); <1% sub-anh 0.15-0.75 mm Cpx; <1% sub-anh 70.1 mm Fe-Ti 35-45% glass + crypto; crudely trachytic PI; Fe-Ti; Px Ol; Ol-Opx aggregates; Qtz (with Cpx fringe); granitic fragments Opx, PI + Hbl, Cpx LQvpadu6 71% eu-sub 0.25-1.2 mm PI (common incl); 71% 0.2-0.4 mm Opx (common as clots with PI); <1% eu-sub 0.1 2.5 mm Hbl (common oxide pseudo) 5-10% crypto; trachytic PI; Fe-Ti; Px; secondary material granitic fragments up to tens of cm 5% Opx, PI (Hbl unknown) LQvpba 2% sub-anh 0.1-0.75 mm PI (common comp zoning and incl); 1-3% sub-anh 0.15-3 mm Ol (common clots); <1% sub 0.25-1 mm Cpx (common clots) 10-25% glass + crypto; Fe-Ti; trachytic PI; Cpx rounded Qtz (with Cpx fringe) <1-5% Cpx, PI + Ol 'Mineral abbreviations (Kretz 1983): Pl = plagioclase, Hbl = hornblende, Opx = orthopyroxene, Cpx = clinopyroxene, Bt = biotite, Ol = olivine. Other abbreviations: Fe-Ti = iron-titanium oxides, eu = euhedral, sub = subhedral, anh = anhedral, comp = compositional, cont = continuous, disc = discontinuous, osc = oscillatory, incl = inclusions, vfg = very fine-grained material, pseudo = pseudomorphs/pseudomorphism, pleo = pleochroism. Seriate refers to phenocrysts whose size ranges down into groundmass sizes. Base/top refers to sample position within a single lava flow. 2A11 MCVF lava samples have groundmasses that are some combination of (1) identifiable crystals, (2) unidentifiable crystallites, (3) cryptocrystalline material (non-isotropic material in which crystals or crystallites cannot be distinguished), (4) opaque material, and (5) glass. The percentage of pristine glass is estimated where possible; otherwise, the intermixed groundmass phases are grouped as "cryptocrystalline material" (= crypto). Abbreviations: Px = pyroxene (variety unknown). 3Granitic fragments = xenoliths with some mixture of Qtz, Pl, Kfs, and mafic minerals. Abbreviations: non = nonvesicular. Proximal/distal refers to sample position relative to the vent. Abbreviations: poss = possible, xeno = xenocryst/xenolith "Description applies to the lava clasts within the pyroclastic deposit 'Description applies to the lava flows in the dominantly breccia unit, and the lava clasts within the breccia. Appendix 3. Major, trace, and rare-earth element chemistry of M C V F sam pies sample HHB-98-KR-8-2 MK-01-31-2a HH8-98-KR-5 MK-01-29-1 a MK-01-29-2 MK-01-00-13 HHB-98-KR-2 MK-01 -28-2b centro LRlng LRlng Ring Ring Ring RingNW Slag Slag Major Elements Si02 61.2 61.42 62.72 61.43 62.79 53.91 58.63 60.33 TiOj 0.705 0.711 0.61 0.614 0.596 0.933 0.644 0.797 AljOa 17.49 17.37 17.38 17.31 17.16 17.62 18.09 17.64 FeTotal 4.94 4.93 4.75 4.80 4.63 7.93 5.83 5.11 Fe 20 3 0 2.18 0 2.01 1.58 2.89 0 2.60 FeO 0 2.48 0 2.51 2.74 4.53 0 2.25 MnO 0.082 0.084 0.081 0.087 0.087 0.126 0.101 0.085 MgO 2.96 3.09 2.76 3.16 2.88 6.06 4.28 2.94 CaO 6.18 6.34 5.62 5.98 5.69 8.55 6.29 6.88 NajO 4.26 4.44 4.49 4.38 4.43 3.87 4.68 4.18 KjO 1.43 1.47 1.66 1.57 1.64 0.90 1.01 1.47 PjOs 0.264 0.268 0.232 0.223 0.216 0.317 0.254 0.316 HPJO- 0.2 0 0.08 0 0 0 0.26 0 H[2]0+ 0.5 0 0.09 0 0 0 0.56 0 La 0.42 0.01 0.11 0.54 0 0 0.55 0.21 Trace Elements A g 0 0.2 0 0.2 0.1 0.1 0 0.2 Ba 479 513 643 577 552 462 427 504 Be 0 0.6 0 0.5 0.5 0.5 0 0.5 Bi 0 0 0 0 0 0 0 0 Cd 0 0 0 0 0 0 0 0 Co 0 23 0 21 21 35 0 23 Cr 51 53 57 63 54 117 38 33 Cs 0 0.24 0 0.29 0.28 0.12 0 0.16 Cu 0 38 0 29 32 44 0 38 Ga 20.1 19.00 19.2 19.00 19.00 19.00 19.1 21.00 Hf 6.2 3.90 6.2 3.00 3.20 3.10 5.2 3.80 In 0 0 0 0 0 0 0 0 Mo 0 0.6 0 0.5 0.6 0.3 0 0.5 Nb 3.4 3.70 4.1 3.30 3.20 4.70 4.7 3.70 Ni 0 38 0 42 37 92 0 26 Pb 8.5 5 7.7 8 5 3 7.9 5 Rb 20.9 19.00 24.5 19.00 19.00 9.70 12 14.00 Sb 0 0 0 0 0 0 0 0 Sc 16 11.0 17 10.0 10.0 17.0 11 10.0 Sn 0 1.0 0 1.0 1.5 1.3 0 1.0 Sr 1300.8 1244 1088.7 1015 1032 1491 1020.7 1821 Ta 0 0.24 0 0.19 0.25 0.24 0 0.21 Te 0 0 0 0 0 0 0 0 Th 3 2.90 1.5 2.20 2.00 1.80 0 2.40 Tl 0 0.11 0 0.10 0.13 0.08 0 0.11 U 0 0.97 0 0.78 0.76 0.56 0 0.78 V 0 95 0 92 88 146 0 102 Y 11.3 12.00 12 12.00 12.00 18.00 12.7 12.00 Zn 0 71 0 68 66 95 0 74 Zr 155.7 150.0 132.5 116.0 120.0 119.0 92.1 140.0 Rare Earth Elements La 11 21.0 10 16.0 15.0 25.0 3 25.0 Ce 49 45 64 29 23 56 52 53 Pr 0 6.20 0 4.70 4.40 7.60 0 7.70 Nd 0 25.0 0 19.0 18.0 30.0 0 30.0 Sm 0 4.20 0 3.40 3.40 4.60 0 4.60 Eu 0 1.30 0 1.00 1.00 1.50 0 1.40 Gd 0 3.20 0 2.90 2.70 3.80 0 3.20 Tb 0 0.42 0 0.39 0.38 0.52 0 0.40 D y 0 2.30 0 2.20 2.10 2.80 0 2.20 Ho 0 0.42 0 0.41 0.40 0.58 0 0.40 Er 0 1.10 0 1.10 1.00 1.50 0 0.97 Tm 0 0.18 0 0.17 0.16 0.23 0 0.15 Yb 0 1.10 0 1.10 1.10 1.60 0 1.00 Lu 0 0.18 0 0.18 0.17 0.25 0 0.16 243 Appendix 3. Major, trace, and rare-earth element chemistry of M C V F samples sample MK-01-28-4b MK-01-30-1 MK-01-27-2a MK-01-26-2b MK-01-26-3 MK-01-26-6 MK-01-16-3 MK-01-17-2 centre Slag Slag Slag tuya CD Upper CO Lower CD Lower Pall East Pall East Major Elements SI0 2 62.17 62.55 57.00 58.60 60.57 58.44 59.71 58.23 TiC-2 0.625 0.622 0.721 0.674 0.616 0.626 0.676 0.683 A I A 17.77 17.81 18.74 18.26 17.77 17.47 18.82 18.98 FeTotal 4.82 4.78 6.05 6.20 5.38 5.53 5.59 5.82 FejOa 1.90 2.16 3.03 1.59 1.53 1.80 2.20 2.48 FeO 2.63 2.36 2.72 4.15 3.46 3.35 3.05 3.00 MnO 0.089 0.089 0.107 0.123 0.100 0.104 0.098 0.104 MgO 2.35 2.35 3.41 3.66 2.80 2.89 2.74 2.75 CaO 5.81 5.66 6.58 6.41 5.62 5.70 6.26 6.51 NajO 4.39 4.39 4.94 4.65 4.67 4.46 4.95 4.85 K 2 0 1.61 1.60 1.23 1.02 1.22 1.11 1.13 1.06 PTQS 0.274 0.272 0.423 0.266 0.234 0.228 0.263 0.278 H[2]0- 0 0 0 0 0 0 0 0 H[2]0+ 0 0 0 0 0 0 0 0 LOI 0.14 0.07 0.86 0.32 1.07 3.46 0 0.89 Trace Elements Ag 0.1 0 0 0 0 0.2 0.1 0.1 Ba 566 606 710 426 552 497 547 495 Be 0.6 0.6 0.7 0.6 0.6 0.5 0.6 0.5 Bi 0 0 0 0 0 0 0 0 Cd 0 0 0 0 0 0 0 0 Co 19 17 23 30 20 24 24 30 Cr 25 27 19 64 39 38 21 23 Cs 0.25 0.14 0.19 0.15 0.34 0.30 0.07 0.22 Cu 29 26 39 33 25 29 25 28 Ga 20.00 20.00 20.00 19.00 17.00 18.00 18.00 19.00 Hf 3.50 3.50 2.90 2.60 2.50 2.30 2.70 2.60 In 0 0 0 0 0 0 0 0 Mo 0.6 0.6 0.7 0.6 0.6 0.6 0.6 0.7 Nb 3.70 3.70 5.80 3.40 3.40 3.10 3.80 3.70 Ni 16 16 31 49 27 26 32 36 Pb 5 5 8 4 4 3 4 6 Rb 17.00 17.00 11.00 9.40 15.00 13.00 12.00 12.00 Sb 0 0 0 0 0 0.3 0 0 Sc 7.9 7.9 9.0 13.0 11.0 11.0 10.0 11.0 Sn 1.0 0.6 4.4 0.8 0.6 0.8 0.6 1.0 Sr 1357 1310 1294 735 714 687 889 942 Ta 0.21 0.22 0.30 0.18 0.24 0.22 0.22 0.23 Te 0 0 0 0 0 0 0 0 Th 2.40 2.30 2.40 1.10 1.30 1.10 1.40 1.30 n 0.13 0.05 0.07 0.03 0.10 0.09 0.05 0.08 U 0.81 0.82 0.81 0.44 0.57 0.51 0.59 0.54 V 86 85 107 106 90 92 106 107 Y 12.00 11.00 13.00 16.00 14.00 14.00 13.00 14.00 Zn 71 72 85 82 75 73 73 76 Zr 132.0 133.0 117.0 98.0 90.0 85.0 104.0 103.0 Rare Earth Elements La 20.0 19.0 24.0 10.0 10.0 9.7 14.0 13.0 Ce 37 39 47 28 21 26 29 32 Pr 5.70 5.50 6.10 3.40 3.10 3.00 3.70 3.80 Nd 23.0 22.0 24.0 15.0 14.0 14.0 16.0 16.0 Sm 3.70 3.70 4.00 3.20 3.00 2.90 3.10 3.10 Eu 1.10 1.10 1.20 1.10 0.92 0.93 1.00 1.00 Gd 2.80 2.80 3.00 3.10 2.80 2.80 2.80 2.80 Tb 0.37 0.37 0.41 0.46 0.41 0.42 0.41 0.41 Dy 2.10 2.00 2.40 2.80 2.50 2.40 2.30 2.40 Ho 0.40 0.39 0.44 0.54 0.48 0.48 0.46 0.45 Er 1.00 0.97 1.10 1.40 1.30 1.30 1.20 1.20 Tm 0.16 0.15 0.18 0.23 0.20 0.20 0.18 0.19 Yb 1.10 1.10 1.10 1.50 1.30 1.40 1.30 1.30 Lu 0.17 0.17 0.19 0.25 0.22 0.21 0.21 0.20 244 Appendix 3. Major, trace, and rare-earth element chemistry of MCVF samples sample MK-01-24-2 MK-01-24-4c MK-01-20-2 MK-01-21-1 MK-01-22-2C HHB-98-KR-13 MK-01-18-2 HHB-98-KR-15 centre Pall West Pall West Fee main Fee north Fee north Ember North Ember North Ember West Major Elements SIOj 6 Z 3 8 62.69 67.26 65.23 64.86 61.83 62.93 57.42 TiOj 0.543 0.540 0.418 0.491 0.489 0.577 0.500 0.705 Al 2 0, 17.33 17.30 16.18 16.65 16.71 17.99 17.95 18.41 FeTotal 5.01 4.89 3.73 4.41 4.36 4.88 4.58 6.26 FejOj 1.97 2.05 1.74 1.97 4.28 0 1.79 0 FeO 2.74 2.55 1.79 2.20 0.07 0 2.51 0 MnO 0.100 0.097 0.082 0.090 0.089 0.087 0.085 0.106 MgO 2.61 2.51 1.77 2.17 2.31 2.4 2.37 4.31 CaO 5.29 5.13 3.96 4.55 4.65 5.55 5.30 7.17 Na20 4.44 4.37 4.46 4.44 4.50 4.53 4.88 4.44 K 2 0 1.46 1.57 1.93 1.78 1.69 1.44 1.18 1.06 P*0, 0.205 0.201 0.166 0.191 0.177 0.187 0.155 0.24 H[2]D- 0 0 0 0 0 0.3 0 0.12 H[2jO+ 0 0 0 0 0 0.49 0 0.01 LOI 0.83 0.82 0 0.02 0.11 0.9 0.19 0 Trace Elements Ag Ba 0 0 0.1 0 0.1 0 0.2 0 575 693 828 781 708 555 264 493 Be 0.6 0.6 0.8 0.7 0.7 0 0.5 0 Bi 0 0 0 0 0 0 0 0 Cd 0 0 0 0 0 0 0 0 Co 21 19 17 16 17 0 20 0 Cr 35 36 33 35 40 85 29 40 Cs 0.37 0.50 0.42 0.28 0.37 0 0.12 0 Cu 26 25 16 20 23 0 28 0 Ga 18.00 17.00 16.00 17.00 17.00 18.8 19.00 19.1 Hf 2.50 2.90 3.10 3.30 2.90 5.5 2.00 5.5 In 0 0 0 0 0 0 0 0 Mo 0.7 0.7 0.6 0.7 0.4 0 0.4 0 Nb 3.50 3.90 4.50 4.20 4.10 5.2 1.70 5.2 Ni 24 24 19 23 26 0 16 0 Pb 4 5 5 5 6 7.7 2 7.1 Rb 20.00 21.00 29.00 25.00 23.00 21.2 8.30 14.6 Sb 0 0 0 0 0.5 0 0 0 Sc 10.0 10.0 7.1 8.6 8.9 15 9.0 13 Sn 0.7 0.6 0.7 0.7 0.8 0 0.5 0 Sr 716 696 577 637 641 758.8 771 909.3 Ta 0.22 0.24 0.30 0.26 0.26 0 0.15 0 Te 0 0 0 0 0 0 0 0 Th 1.60 1.80 2.40 2.10 2.00 0 0.86 0 Tl 0.13 0.17 0.15 0.15 0.07 0 0.06 0 U 0.75 0.82 1.10 0.86 0.80 0 0.52 0 V 82 81 56 70 53 0 78 0 Y 13.00 14.00 13.00 14.00 13.00 14.4 8.70 14.9 Zn 69 69 55 62 61 0 66 0 Zr 93.0 108.0 116.0 129.0 109.0 102.6 66.0 101.8 Rare Earth Elements La 12.0 13.0 15.0 15.0 13.0 5 5.5 7 Ce 24 18 24 32 22 33 19 34 Pr 3.30 3.60 3.80 4.00 3.60 0 1.80 0 Nd 14.0 15.0 15.0 16.0 14.0 0 8.4 0 Sm 2.70 2.90 2.90 3.10 2.90 0 1.90 0 Eu 0.87 0.88 0.79 0.86 0.81 0 0.70 0 Gd 2.60 2.70 2.50 2.80 2.60 0 2.00 0 Tb 0.37 0.40 0.37 0.40 0.37 0 0.28 0 Dy 2.20 2.40 2.20 2.30 2.20 0 1.60 0 Ho 0.43 0.48 0.42 0.47 0.44 0 0.30 0 Er 1.20 1.20 1.10 1.20 1.10 0 0.76 0 Tm 0.18 0.19 0.18 0.20 0.18 0 0.12 0 Yb 1.20 1.30 1.30 1.40 1.30 0 0.78 0 Lu 0.20 0.21 0.22 0.22 0.20 0 0.13 0 245 Appendix 3 . Major, trace, and rare-earth element chemistry of M C V F samples sample MK-01-11-2 MK-01-12-3 MK-01-13-3a MK-01-14-1a MK-01-23-1 MK-01-10-7 MK-01-10-9 centre Ember West Ember NW Ember SW Ember SW Ember NE TrlcSW TrlcSW Major Elements SK)2 59.41 58.63 60.91 62.93 58.69 54.07 56.66 TiOj 0.639 0.715 0.581 0.515 0.776 1.006 0.837 Al 20, 17.82 18.06 17.74 18.28 18.45 17.84 17.07 FeTotal 5.43 5.50 4.95 4.63 5.44 8.14 6.38 FejOi 1.96 2.76 1.93 1.70 2.09 2.79 2.55 FeO 3.12 2.47 2.72 2.64 3.02 4.82 3.44 MnO 0.102 0.100 0.094 0.087 0.099 0.139 0.112 MgO 3.59 3.65 2.52 1.57 2.83 5.27 5.62 CaO 6.13 6.20 5.66 5.17 6.06 8.54 7.91 Na20 4.56 4.84 4.40 4.83 4.57 3.80 3.92 K 20 1.30 1.25 1.44 1.54 1.41 0.89 1.21 P2O5 0.260 0.296 0.183 0.216 0.267 0.302 0.308 HPJO- 0 0 0 0 0 0 0 HPJCH 0 0 0 0 0 0 0 LOI 0.86 0.83 1.68 0.33 1.48 0.28 0.08 Trace Elements Ag 0.1 0.1 0 0.1 0.1 0.2 0.1 Ba 609 644 565 618 665 391 478 . Be 0.6 0.7 0.6 0.6 0.6 0 0.6 Bi 0 0 0 0 0 0 0 Cd 0 0 0 0 0 0 0 Co 21 25 23 28 19 39 30 Cr 74 53 32 15 19 93 215 Cs 0.23 0.21 0.34 0.35 0.25 0.11 0.13 Cu 32 25 28 30 38 40 36 Ga 18.00 18.00 18.00 19.00 19.00 18.00 18.00 Hf 2.80 2.90 2.60 2.90 3.40 3.30 3.30 In 0 0 0 0 0 0.06 0 Mo 0.7 0.6 0.7 0.8 0.8 0.4 0.4 Nb 4.50 5.20 3.10 3.40 4.90 3.80 4.90 Ni 58 66 27 0 23 59 56 Pb 21 9 4 5 5 3 5 Rb 14.00 12.00 19.00 20.00 15.00 9.20 14.00 Sb 0 0 0 0 0 3.5 0 Sc 12.0 11.0 10.0 6.2 12.0 18.0 15.0 Sn 5.3 1.2 0.7 0.7 1.0 4.7 1.0 Sr 880 1090 737 779 894 1369 1309 Ta 0.25 0.28 0.25 0.22 0.28 0.27 0.27 Te 0 0 0 0 0 0 0 Th 1.70 2.00 1.60 1.80 1.70 1.80 2.10 Tl 0.08 0.08 0.13 0.14 0.11 0.04 0.09 U 0.62 0.71 0.72 0.73 0.64 0.56 0.74 V 103 105 93 73 112 138 121 Y 13.00 12.00 13.00 13.00 19.00 19.00 14.00 Zn 74 78 68 68 72 87 79 Zr 113.0 116.0 100.0 115.0 133.0 125.0 129.0 Rare Earth Elements La 16.0 19.0 12.0 13.0 18.0 25.0 24.0 Ce 31 36 23 22 28 58 48 Pr 4.10 5.20 3.10 3.50 5.10 7.50 7.00 Nd 17.0 21.0 14.0 15.0 21.0 31.0 27.0 Sm 3.20 3.60 2.80 3.00 4.10 5.10 4.30 Eu 1.00 1.10 0.86 0.92 1.20 1.70 1.30 Gd . 2.80 2.80 2.70 2.70 3.70 4.10 3.40 Tb 0.40 0.39 0.40 0.39 0.54 0.60 0.47 Dy 2.30 2.20 2.20 2.30 3.10 3.30 2.50 Ho 0.45 0.42 0.45 0.44 0.63 0.68 0.49 Er 1.20 1.10 1.10 1.20 1.70 1.70 1.30 Tm 0.19 0.18 0.18 0.19 0.25 0.27 0.20 Yb 1.30 1.20 1.30 1.30 1.70 1.80 1.30 Lu 0.21 0.18 0.20 0.21 0.27 0.29 0.21 246 Appendix 3. Major, trace, and rare-earth element chemistry of M C V F samples sample MK-01-9-6 MK-01-9-6 MK-01-9-6 MK-01-10-2D MK-01-10-3 centre Trie SW dup Trie SW dup Trie SW dup Trie SE TricSE Major Elements Si0 2 54.11 54.20 54.06 58.04 64.42 TiOj 0.988 0.983 0.992 0.725 0.481 Al 20 3 17.81 17.84 17.77 18.53 17.42 FeTotal 8.05 8.05 8.05 6.15 4.42 Fe 20 3 3.01 2.86 2.92 2.54 1.79 FeO 4.54 4.67 4.62 3.25 2.36 MnO 0.130 0.132 0.135 0.111 0.093 MgO 5.45 5.45 5.44 3.44 1.61 CaO 8.53 8.53 8.55 6.55 4.61 Na20 3.93 3.86 3.83 4.66 4.69 K 20 0.85 0.86 0.87 1.12 1.79 P 2 O 5 0.294 0.295 0.295 0.300 0.225 H[2]0- 0 0 0 0 0 H[2JO+ 0 0 0 0 0 LOI 0.21 0.22 0.22 0.58 0.45 Trace Elements Ag 0.2 0.2 0.1 0.2 0.1 Ba 379 387 403 514 655 Be 0 0 0 0.6 0.7 BI 0 0 0 0 0 Cd 0 0 0 0 0 Co 33 32 32 29 22 Cr 95 99 96 50 14 Cs 0.12 0.12 0.12 0.19 0.39 Cu 33 33 32 38 16 Ga 18.00 18.00 19.00 19.00 18.00 Hf 3.10 3.20 3.20 2.40 2.90 In 0 0 0 0 0 Mo 0.5 0.4 0.5 0.6 0.8 Nb 4.00 4.00 3.90 3.70 3.60 Ni 61 63 62 44 0 Pb 6 3 3 5 6 Rb 9.00 9.00 9.00 11.00 21.00 Sb . 0 0 0 0 0 Sc 17.0 17.0 17.0 12.0 5.5 Sn 1.0 0.9 0.7 0.8 0.8 Sr 1393 1434 1349 985 857 Ta 0.24 0.24 0.23 0.19 0.25 Te 0 0 0 0 0 Th 1.80 1.80 1.80 1.20 2.10 Tl 0.04 0.04 0.04 0.08 0.16 U 0.55 0.58 0.55 0.45 0.85 V 134 135 134 106 71 Y 19.00 19.00 19.00 15.00 11.00 Zn 82 88 84 86 66 Zr 123.0 124.0 120.0 92.0 112.0 Rare Earth Elements La 24.0 24.0 24.0 14.0 14.0 Ce 65 63 56 21 21 Pr 7.40 7.40 7.50 4.10 3.80 Nd 30.0 30.0 30.0 18.0 16.0 Sm 5.10 4.90 5.00 3.40 2.70 Eu 1.60 1.60 1.60 1.10 0.85 Gd 4.20 4.00 4.20 3.10 2.30 Tb 0.60 0.59 0.57 0.45 0.34 Dy 3.40 3.30 3.30 2.50 1.90 Ho 0.65 0.67 0.67 0.52 0.39 Er 1.70 1.70 1.70 1.30 1.00 Tm 0.26 0.27 0.27 0.21 0.17 Yb 1.80 1.80 1.80 1.40 1.10 Lu 0.28 0.28 0.28 0.22 0.19 247 Appendix 4: Analytical methods Major and trace element geochemical analyses are given in Appendix 3. Samples selected for geochemical analyses underwent initial processing by the author at the University of British Columbia Department of Earth and Ocean Sciences. Samples with minimal evidence for weathering were crushed to fragments <2 cm in a jaw crusher. The chips were sorted manually, with visibly weathered fragments and fragments containing macroscopic xenoliths excluded. A tungsten carbide ball mill was then used to reduce the fragments to a powder small enough to pass through a 200 mesh nylon screen. Most geochemical analyses were of samples collected by the author, and were performed at the Analytical Chemistry Subdivision of the Geological Survey of Canada, in Ottawa. Analyses were of fused beads prepared from ignited samples, using x-ray fluorescence (XRF). Total iron was determined by XRF, but FeO was determined by ammonium metavanadate titration. Trace element analyses were by inductively coupled plasma emission spectrometry (ICP-ES) and inductively coupled plasma mass spectrometry (ICP-MS). A further discussion of analytical methods is available at: http://www.mcan.gc.ca/gsc/mrcl/labs/chem_e.html, the web page for the Geological Survey of Canada Analytical Chemistry Laboratories. Eleven analyses were of samples collected by C.J. Hickson and J.K. Russell in 1998. These were performed at McGill Geochemical Laboratories, by XRF analysis of fused beads. A further discussion of analytical methods is available at: http://www.eps.mcgill.ca/~geochem/x-ray.htm, the McGill Geochemical Laboratory's analytical procedures web page. The program IgPet was used to generate some of the geochemical plots. The 248 program MELTS (version 5.0) was used to calculate liquidus temperatures for selected samples. 249 Appendix 5: Calculations pertinent to ice thickness over subglacial domes Appendix 5a. Ice cavity wall meltback rates I have calculated rates of ice melting around subglacial domes using a method similar to that of Tuffen et al. (2002b). Table A5.1 lists all symbols and subscripts used. Table A5.2 lists all constants used. Ice is assumed to be at 0°C. For the purposes of calculation, I treat all cooling as heat transfer as conductive, ignoring heat transfer by convecting steam and meltwater. The effects of chilled rind development and thermal fracturing (jointing) are also ignored; Tuffen et al. (2002b) estimated that they were approximately equal and thus might cancel each other. The conductive heat flux (qm) from an isothermal hemispherical mass is given by: (11) where S is a shape term (Holman 1997) and Wcm is a correction for the latent heat of crystallization (included because most samples have a low glass content). T e is the temperature difference between the magma and its surroundings. This was taken to be the eruption temperature, since it is assumed that the surrounding ice is at 0°C. This heat flux is used to calculate rates of ice melting by assuming that all thermal energy goes into melting ice. Hence, the volume of ice melted per second, VY is: 250 Table A5.1 Symbols and subscripts used in Appendix 5 calculations. symbol explanation units Q thermal energy J q conductive heat flux W T temperature °c AT temperature change/difference °c V volume m 3 AV volume change _ 3 m V volume change per second m P pressure MPa AP effective pressure = P g - P c MPa m mass kg r density kgm"3 c heat capacity J kg"1 °C k thermal conductivity W m"1 °C"1 L latent heat of fusion Jkg'1°C S shape factor = 27tr (Holman 1997) r radius m f rate of change of radius m s"1 i e shear strain rate s-1 t shear stress Pa A ice parameter (Glen 1955) s"1 kPa 3 B ice parameter (Nye 1953) Pas 1 ' 3 n Glen's flow law constant (Glen 1955) h ice thickness m a area m b bulk modulus for water GPa subscript explanation m magma i ice w water e eruption f final eq equilibrium 9 glaciostatic c cavity 251 Table A5.2 Constants used in Appendix 5 calculations. constant value source n 917 kg m"3 Li 3.35 x 105 J kg"1 Hoskuldsson and Sparks 1997 L-m 2.09 x 105 Jkg1 Hoskuldsson and Sparks 1997 C m 1130 Jkg"1 °C~1 calculated using data from Neuville et al. 1993 1.26 Wm"1 °C"1 Murase and McBirney 1973 B 5.3 x107 Pa s1/3 Paterson 1994 n 3 Glen 1955; Hooke 1981; Paterson 1994 252 (12) The meltback rate (rj1) of the ice walls is the volume rate of ice melting (V;') divided by the surface area ^Tcr 2 for a hemispherical cavity): • V, r ' = - J ^ (13> Combining Equations 12 and 13, we get an expression for the meltback rate of the ice walls: 2nriLiPi This is an instantaneous meltback rate (the rate of melting of an infinitesimally thin shell of ice), since the cavity radius and rate of change of radius are changing together. Appendix 5b. Ice cavity wall deformation rates For cavity pressures greater than glaciostatic, water will be forced away from the vent by any available pathways, and cavities and conduits will remain open even without input of further heat. However, if cavity pressure is less than glaciostatic, cavities will shrink (Glen 1952,1955; Shreve 1972; Rothlisberger 1972; Cutler 1998), unless cavity 253 wall meltback rates equal or exceed cavity wall deformation rates. I have calculated deformation rates for the walls of hemispherical cavities at the base of the ice, using a method similar to that of Tuffen et al. (2002b). In addition to the assumptions listed in Appendix 5a, it is assumed that cavities are at atmospheric pressure (0.1 MPa) 5 8. It is also assumed that ice is incompressible and deforms ductilely in accordance with Glen's flow law (Glen 1955): e'=ATn (15) in which e' is the shear strain rate, x is the shear stress, A is a parameter determined by ice properties59, and n is Glen's flow law constant60. Studies of boreholes and cavities in ice have yielded relationships between cavity shape, cavity radius, and ice thickness (Nye 1953, 1976; Cutler 1998) . Specifically, for a hemispherical cavity: AP nB (16) where rc' is the deformation rate of the roof of a cavity of radius rc beneath a glacier of 58 Since there are no indications that cavities were water-filled and only rare indications that cavities contained significant quantities of fragmental material, it is assumed that cavities were at least partially filled with air at atmospheric pressure (0.1 MPa). 59 The parameter A depends on ice temperature, crystal orientation, impurity content and perhaps other factors and has been determined experimentally by many authors. A discussion of this is available in Paterson (1994). 60 Glen's flow law constant, n, is empirically determined. A large body of data supports adopting a value of 3 for this parameter (Hooke 1981). 254 thickness h and B is an ice viscosity parameter61. This relationship should hold true as long as cavity radius is much less than ice thickness (in situations where ice deformation is ductile, not brittle). Cavity wall deformation rates (Equation 16) are plotted against cavity wall meltback rates (Equation 14) and shown in Figure A5.1. If cavity wall closure rate is less than cavity wall meltback rate, lava is unlikely to reflect the shape of the ice cavity, or will fill only the bottom portion of the cavity. However, if wall deformation rate exceeds wall meltback rate, lava flows are likely to be moulded against ice on both lateral and upper surfaces (as is implied by lava flow morphologies and joint orientations at the Ember Ridge subglacial domes). Appendix 5c. Equilibrium cavity sizes Effective cavity pressure (AP) is defined as P g-P c. I have used a range of effective cavity pressures, reflecting different possible ice thicknesses, to calculate cavity radii for which wall meltback rates equal wall deformation rates. I compared these equilibrium cavity sizes to the actual sizes of lobes of the Ember Ridge domes and, in the case of Ember Ridge Northeast, the entire mass, because it represents a single small magma pulse. From this I determined a range of ice thicknesses beneath which moulding against ice is probable. The same method was used to estimate ice thickness above subglacial rhyolite flows at the Icelandic volcano Blahnukur (Tuffen et al. 2002b). Combining Equation 16 with Equation 14 produces an expression for equilibrium 61 The parameter B depends on pressure, temperature, water content, etc., and is experimentally determined. A discussion is available in Hooke (1998). 255 Ut ON -5 -4 -3 - 2 - 1 0 1 log (wall deformation rate) Figure A5.1 Wall meltback rate versus wall deformation rate for hemispherical subglacial cavities 5-20 m in radius, plotted on log-log axes. Wall meltback rate is the rate of cavity enlargement through melting of ice and is calculated from Equation 14. Wall deformation rate is the rate of cavity closure by ductile deformation of ice, and is calculated from Equation 16. Individual lines represent different ice thickness scenarios (in m). It is assumed that cavity pressure, Pc = atmospheric (0.1 MPa). Therefore, the 200 m line corresponds to an effective pressure (AP = Pg - Pc) of 1.7 MPa. AP values range up to 8 MPa for 900 m thick ice. In all cases, cavity closure rates exceed cavity wall meltback rates, implying that lava flows are moulded against ice. However, deformation rates are so much higher than meltback rates that it is unlikely that this calculation is valid for all MCVF subglacial domes; such large cavities could not have been maintained, especially under thin ice. More likely, the growth of cavities kept pace with the emplacement of lava, and most units were emplaced as mulitple lava pulses rather than a single mass. cavity radius: qJnB)n A P^ITCLJP, 1/3 (17) Calculated equilibrium cavity radii are shown in Figure A5.2 for cavity pressures equal to atmospheric (0.1 MPa) and cavities up to 20 m in radius. If stable cavities do exist prior to lava effusion, there is no energetic prohibition against moulding the lava against the ice walls, and moulding against ice only becomes difficult when ice thickness becomes low enough that it is liable to deform brittly; in this case, these equilibrium cavity size calculations would no longer be valid, and lava flows could take the shape of crevasses. However, Figure A5.2 shows that lava masses are typically larger than the calculated equilibrium cavity radii. Thus, it seems likely that stable cavities do not exist prior to lava effusion. What is clear from Figures A5.1 and A5.2 is that increased eruption temperatures or jointing of lava will promote more rapid heat transfer and more rapid ice melting, thus diminishing the probability that lava flows are moulded by the ice walls of cavities. For the same conditions, the formation of chilled rinds, the presence of air in cavities (rather than water), decreased effusion rates, or greater degrees of endogenous eruption will all inhibit heat transfer, decrease rates of ice melting, and increase the probability that lava flows are moulded by the ice walls of cavities. 257 to U l 00 u o B 3 10 8 H 6H 2 4 H 2 H 0 increased heat transfer rates . decreased heat transfer rates 200 m 900 m 1 1 1 4 6 8 lava mass radius (m) 10 12 14 16 Figure A5.2 Equilibrium cavity radius (Equation 17) versus lava mass radius, for two ice thicknesses (200 m and 900 m), assuming cavities are at atmospheric pressure (0.1 MPa). Increasing ice thickness (hence increasing the difference between Pg and Pc) increases the likelihood of moulding against ice. If the actual sizes of subglacial domes or their discrete lobes, as measured in the field, are representative of the size of stable cavities that existed prior to eruption, moulding by ice is a necessity. However, this is unlikely to be true because equilibrium cavity radii are smaller than the actual lava masses. Furthermore, most subglacial domes comprise multiple flows. Finally, increasingly thin ice (as is implied by field observations) makes it difficult to maintain large subglacial cavities without brittle failure of cavity roofs occurring. Therefore, it is unlikely that stable cavities the size of most observed lava masses were present prior to eruption. It is more probable that cavities grew to keep pace with eruptions. The concept for this plot is from Tuffen et al. (2002). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0052392/manifest

Comment

Related Items