Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Late cenozonic geology of the southern Rocky Mountain trench, British Columbia Clague, John Joseph 1973

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A1 C55.pdf [ 40.55MB ]
Metadata
JSON: 831-1.0052380.json
JSON-LD: 831-1.0052380-ld.json
RDF/XML (Pretty): 831-1.0052380-rdf.xml
RDF/JSON: 831-1.0052380-rdf.json
Turtle: 831-1.0052380-turtle.txt
N-Triples: 831-1.0052380-rdf-ntriples.txt
Original Record: 831-1.0052380-source.json
Full Text
831-1.0052380-fulltext.txt
Citation
831-1.0052380.ris

Full Text

LATE CENOZOIC GEOLOGY OF THE SOUTHERN ROCKY MOUNTAIN TRENCH, BRITISH COLUMBIA by JOHN JOSEPH CLAGUE A.B., Occidental College, 1967 M.A., University of California, Berkeley, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS OF THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Geological Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November 1973 li In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Geological Sciences The University of British Columbia Vancouver 8, Canada Date November 7, 1973 i i i ABSTRACT Geologic studies which have provided information on the late Cenozoic history of the southern Rocky Mountain Trench, British Columbia include: (1) the St. Eugene Formation and the origin of the southern Rocky Mountain Trench, (2) the stratigraphy and correlation of Quaternary sediments, (3) the patterns of glacier flow and the origin of late Wisconsinan t i l l , and (4) the sedimentology and paleohydrology of late Wisconsinan outwash. (1) The Tertiary history of the southern Rocky Mountain Trench is inferred from a study of the distribution, stratigraphy, fabric, lithologic composition, structure, and palynology of the Miocene St. Eugene Formation. This unit consists of flood-plain and fan facies and represents the uppermost Tertiary sediments f i l l i n g block-faulted basins in the Trench. Sediment deposition was probably contemporaneous with faulting, but at least 600 m of displacement along the east side of the Trench postdates deposition of the St. Eugene Formation. The Miocene southern Rocky Mountain Trench was similar in morphology and position to the present Trench, but the Miocene climate was more temperate than the present climate of the area. Palynological results indicate that during the Miocene southeastern British Columbia had abundant summer precipitation and mild, moist winters. (2) Exposed Pleistocene sediments correlate with deposits of the Pinedale Glaciation and the preceding interglaciation. Deposits of three stades and two interstades of the Pinedale Glaciation are recog-iy nized. During the earlier interstade the floor of the Trench in south-eastern British Columbia was deglaciated, whereas during the later interstade residual ice apparently remained locally along the center of the Trench-CS) Pinedale glacier-flow patterns and t i l l genesis in the southern Rocky Mountain Trench were determined through a study of glacial landforms, t i l l fabric, and t i l l composition. A review of the ways in which directed t i l l fabrics originate indicates that t i l l associated with drumlins accumulated by lodgement of particles due to subglacial pressure melting. Landforms and t i l l fabrics document one major shift in the pattern of glacier flow near the end of glaciation. In mountain-ous areas t i l l composition i s less sensitive as an indicator of such shifts than t i l l fabric. T i l l composition instead reflects the dominant pattern of ice flow during glaciation. T i l l constituents decrease with distance from their bedrock sources because of progressive deposition, dilution through sediment mixing, and breakage and abrasion during transport. Other factors which may affect the distribution and relative abundance of constituents in Trench t i l l s include reworking at the ice-sediment interface and transport of constituents englacially and subglacially by meltwater. (4) Late Wisconsinan channeled outwash i s coarse, poorly sorted, shows large-scale cross-bedding, and was deposited in channel-bar complexes of high-energy rivers. Peak discharges calculated from channel morphometry and maximum particle size are larger than maximum discharges attributable to uninterrupted summer runoff. Many channels transmitted '•VP-peak discharges during jbkulhlaups from glacial lakes in tributary valleys. An empirical relationship between total volume discharged during documented jbkulhlaups and corresponding maximum instantaneous discharges is applied to one Pleistocene glacier-dammed lake in the study area to show that discharges equal to or larger than those calculated from channel morphometry were attained during jbkulhlaups. Some applications of these geologic studies include the pre<-diction of flood magnitudes from self-dumping glacier-dammed lakes, the determination of the source of ore clasts or minerals in t i l l , and a preliminary assessment of the groundwater resources of the southern Rocky Mountain Trench. v i TABLE OF CONTENTS page ABSTRACT i i i TABLE OF CONTENTS v i LIST OF TABLES :x LIST OF FIGURES x i LIST OF APPENDICES x v i ACKNOWLEDGMENTS :;xvii CHAPTER ONE: INTRODUCTION , 1 PREVIOUS GEOLOGIC WORK 2 BEDROCK GEOLOGY 4 SURFICIAL GEOLOGY 6 Introduction 6 Deposits 7 St. Eugene Formation 8 Interglacial Sediments 8 Older Drift 10 Inter-drift Sediments 16 Younger Drift 18 Postglacial Sediments 42 LITERATURE CITED 44 CHAPTER TWO: THE ST. EUGENE FORMATION AND THE DEVELOPMENT OF THE SOUTHERN ROCKY MOUNTAIN TRENCH 47 ABSTRACT 47 INTRODUCTION 48 DISTRIBUTION AND CHARACTER OF THE ST. EUGENE FORMATION 49 Colluvium and Fanglomerate 51 Stratified S i l t and Sand 53 Coarse Gravel with Minor Sand Interbeds 53 page PROVENANCE OF ST. EUGENE FANGLOMERATE 54 Analytical Procedure 54 Fan-fabric Analysis 54 Method 55 Results 56 Clast Lithology Analysis 57 Method 57 Results 58 Discussion 58 DEFORMATION OF THE ST. EUGENE FORMATION 60 ST. EUGENE PALEOFLORA AND PALEOCLIMATOLOGY 61 Analytical Procedure 62 Method 62 Results 63 Discussion 64 Age Assigned by Comparison with Similar Floras 64 Paleoecology Inferred from Ecological Requirements of Extant Counterparts 76 Paleogeography Inferred from Physical Characteristics of Sediments 86 NEOGENE DEFORMATION IN SOUTHEASTERN BRITISH COLUMBIA 88 CONCLUSIONS 100 LITERATURE CITED 102 CHAPTER THREE: GLACIER-FLOW PATTERNS AND THE ORIGIN OF LATE WISCONSINAN TILL 109 ABSTRACT 109 INTRODUCTION 110 DISTRIBUTION AND CHARACTER OF TILL 112 GLACIER-FLOW PATTERNS 114 Analytical Procedure 116 T i l l - f a b r i c Analysis 117 Englacial transport 119 Deposition 120 Post-depositional reorientation 121 v i i i page Method 123 Results 124 Heavy Mineral Analysis 126 Method 129 Results 130 Clast Lithology Analysis 132 Method 132 Results 133 Discussion .134 CONCLUSIONS 166 LITERATURE CITED 174 CHAPTER FOUR: SEDIMENTOLOGY AND PALEOHYDROLOGY OF LATE WISCONSINAN OUTWASH 179 ABSTRACT 179 INTRODUCTION 180 DISTRIBUTION AND CHARACTER OF OUTWASH 181 Sedimentary Textures 181 Sedimentary Structures 182 Pebble Lithologies and Provenance 183 PALEOHYDROLOGY '. 185 Paleodischarge Determinations from Channel Morphometry 187 Paleodischarge Determinations from Discharge Records of Proglacial Rivers 198 Discussion of Results 206 CONCLUSIONS 210 LITERATURE CITED 212 CHAPTER FIVE: GEOMORPHIC HISTORY 215 LITERATURE CITED 233 CHAPTER SIX: APPLICATIONS OF GEOLOGIC KNOWLEDGE 235 MAGNITUDE OF JOKULHLAUPS 235 MINERAL EXPLORATION 238 :ix page GROUNDWATER . .. , , .. , 240 LITERATURE CITED 247 CHAPTER SEVEN: SUGGESTIONS FOR FURTHER RESEARCH 249 TILL FABRICS AND TILL GENESIS 249 ORIGIN OF DRUMLINS 250 COLLECTION OF FABRIC DATA 250 DECREASE OF TILL CONSTITUENT WITH DISTANCE FROM SOURCE 251 QUATERNARY STRATIGRAPHY 252 SEDIMENT CONSOLIDATION 252 GROUNDWATER RESOURCES 253 LITERATURE CITED 254 APPENDICES 255 '.X LIST OF TABLES Bedrock units in and adjacent to the study area. Palynomorph assemblages from the St. Eugene Formation. Comparison of the St. Eugene microflora and Miocene microfloras from British Columbia and the northwestern United States. Relation between orientation of fabric elements in t i l l and t i l l genesis. Comparison of till-fabric maximum clustering axes and the trends of associated glacial lineations. Peak velocities and discharges determined from channel morphometry and maximum particle size. Relation of glacier area and meltwater discharge during periods of maximum melt. Jbkulhlaup data. xi LIST OF FIGURES Frontispiece, Figure 1. Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Index map of the Rocky Mountain Trench and the study area. Locality index map. The southern Rocky Mountain Trench, British Columbia. Generalized geologic map of part of southeastern British Columbia. Surficial geology, southern Rocky Mountain Trench, British Columbia. Map of Bouguer gravity in the southern Rocky Mountain Trench, and gravity profiles and inferred geologic sections across the Trench. Figure 7. Thickness of Quaternary sediments. Figure 8. Composite columnar section of late Cenozoic sediments exposed in the southern Rocky Mountain Trench. Figure 9. The St. Eugene Formation. Figure 10. Interglacial sediments, Sand Creek. Figure 11. Chronology and correlation of late Pleistocene events in the Pacific Northwest and the Rocky Mountains of the United States. Figure 12. Older drift. Figure 13. Attitudes of beds, faults, and fractures in deformed outwash of the older drift. Figure 14. Inter-drift sediments. Figure 15. Younger drift. page i 11 13 15 17 in pocket 21 23 25 27 29 31 33 35 37 39 x i i page Figure 16. Postglacial sediments and landforms. 41 Figure 17. Index map showing outcrop areas of the St. Eugene 69 Formation. Figure 18. Stratigraphic sections of the St. Eugene Formation. 71 Figure 19. The St. Eugene Formation. 73 Figure 20. Fabrics of paraglacial a l l u v i a l fans. 75 Figure 21. Relation between fan-fabric strength and differences 77 in the fabric results for samples of size 60 and 20. Figure 22. Fan fabrics from the St. Eugene Formation. 79 Figure 23. Percent of clasts of the following lithologies in 81 St. Eugene fanglomerate: laminated greenish-gray a r g i l l i t e (source i s largely the Roosville and Gateway Formations), red-purple a r g i l l i t e and quartz arenite (largely of the Phillips Formation), and mafic igneous rocks (Purcell volcanics). Figure 24. The fabric of mudflow gravel. 83 Figure 25. Deformation of the St. Eugene Formation. 85 Figure 26. Plant microfossil sample sites from the St. Eugene 87 Formation. Figure 27. Selected plant microfossils from the St. Eugene 91 Formation. Figure 28. Location of some Miocene palynomorph assemblages in 95 British Columbia and the northwestern United States. Figure 29. Relation between age of flora and percent of f o s s i l 97 genera s t i l l living near their f o s s i l l o c a l i t i e s for mega- and microfloras from the western United States and British Columbia. Figure 30. Block diagram showing the proposed model of Tertiary 99 sedimentation controlled by block faulting in the southern Rocky Mountain Trench, British Columbia. x i i i Figure 31. Index map showing trends of glacial lineations and the distribution of d r i f t . Figure 32. Late Wisconsinan t i l l in the southern Rocky Mountain Trench. Figure 33. Features which indicate the direction of ice flow immediately prior to deglaciation of the southern Rocky Mountain Trench. Figure 34. Relation between t i l l - f a b r i c strength and differences in the fabric results for samples of size 60 and 20. Figure 35. Selected fabrics from the younger t i l l . Figure 36. Magnitude and direction of plunge of t i l l clasts. Figure 37. Comparison of t i l l - f a b r i c axes of maximum clustering and associated glacial lineations. Figure 38. T i l l - f a b r i c axes of maximum clustering determined at vertical intervals of 1 to 5 m through the younger drift. Figure 39. Relation between percent of a constituent in t i l l and distance from the bedrock source. Figure 40. Sediment sample si t e s . Figure 41. Distribution of selected bedrock lithologies. Figure 42. Amphibole distribution in t i l l of the southern Rocky Mountain Trench. Figure 43. Garnet distribution in t i l l of the southern Rocky Mountain Trench. Figure 44. Distribution of mafic igneous pebbles in t i l l of the southern Rocky Mountain Trench. Figure 45. Distribution of Blairmore Group clasts in d r i f t of the southern Rocky Mountain Trench. page 135 137 139 143 145 147 149 153 155 157 159 161 163 165 167 xiy Figure 46. Two phases of glacier flow near the end of the Pinedale Glaciation. Figure 47. Patterns of glacier coalescence off tributary valleys in the Cordillera. Figure 48. Distribution of outwash and meltwater channels, Rocky Mountain Trench, southeastern British Columbia. Figure 49. Cumulative size frequency curves of outwash samples. Figure 50. Sedimentary structures in outwash at gravel p i t s . Figure 51. Percent of clasts of the following lithologies in outwash and t i l l : laminated greenish-gray a r g i l l i t e (source is largely the Roosville and Gateway Formations), red-purple a r g i l l i t e and quartz arenite (largely of the Phillips Formation), and mafic igneous rocks (Purcell volcanics and intrusions). Figure 52. Fourth-order trend surface map of percent laminated greenish-gray a r g i l l i t e in t i l l . Figure 53. Meltwater channels. Figure 54. Topographic profiles across a major meltwater channel. Figure 55. Relation of glacier area, A, and maximum instantaneous discharge, Q. Figure 56. Relation of total volume drained during jOkulhlaup and peak water discharge. Figure 57. Pattern of ice flow i n southeastern British Columbia and southwestern Alberta during the Pleistocene. Figure 58. Distribution of existing glaciers in southeastern British Columbia and southwestern Alberta. Figure 59. Map of the Cordilleran Ice Sheet and related features in southeastern British Columbia, northwestern Montana, and northern Idaho. page 169 171 189 191 193 195 197 199 201 205 209 219 221 223 Figure 60. Comparison of a contemporary ice sheet and the Cordilleran Ice Sheet during the maximum Pinedale advance. Figure 61. A contemporary analogue of the Pleistocene glacierized Rocky Mountain Trench. Figure 62. Comparison of contemporary and ancient landforms produced by glaciers which remained active during recession rather than stagnating. Figure 63. A contemporary analogue of Pleistocene ice-dammed lakes in the southern Rocky Mountain Trench. Figure 64. Relation of cumulative volume drained during jOkulhlaups and instantaneous water discharge. Figure 65. Meltwater channels in the vicinity of Cranbrook, British Columbia. Figure 66. Meltwater channel complex northwest of Bull River; mean water discharge records of Norbury Creek and Bull River. page 225 227 229 231 241 243 245 xvi LIST OF APPENDICES page APPENDIX 1. Stratigraphic sections. 256 APPENDIX 2. Southern Rocky Mountain Trench radiocarbon dates. 258 APPENDIX 3. Size frequency data for selected t i l l samples. 259 APPENDIX 4. Fan-fabric sites and results. 260 APPENDIX 5. Contoured diagrams of selected t i l l fabrics. 262 APPENDIX 6. Till-fabric sites and results. 263 APPENDIX 7. Heavy mineral and clast lithology data. 266 APPENDIX 8. Textural parameters of outwash underlying late glacial meltwater channels in the southern Rocky Mountain Trench. 274 xvii ACKNOWLEDGMENTS A study of the surficial geology of the southern Rocky Mountain Trench began in 1969 with funds provided by the Water Investi-gations Branch of the British Columbia Department of Lands, Forests, and Water Resources. Continued research (1970-73) was supported by the National Research Council of Canada through Postgraduate Scholarships and Grant A-1107. Thanks are extended to W.H. Mathews for guidance, criticism, and encouragement during a l l phases of this study. My ideas were formulated, in part, through conversations and correspondence with W.C. Barnes, M.A. Church, A. Dreimanis, R.J. Fulton, P.L. Gordy, R.E. Kucera, L.M. Lavkulich, G.B. Leech, J.E. Reesor, G.E. Rouse, J.M. Ryder, 0. Slaymaker, and P.N. Sprout. M.A. Church suggested the method of paleodischarge determination from channel morphometry. I am grateful to P. Barnes of the British Columbia Department of Highways for drillhole and sediment texture data; to F.R. Edmunds of Cominco Ltd. for data on the thickness of semiconsolidated and unconsoli*-dated sediments; to P.L. Gordy, G.W. Graff, and G. Steward of Shell Canada Ltd. for seismic records and drillers' logs; to D.M. Mark for a computer program which analyzes axial orientation data; to A. Post for some of the photos used in this thesis; and to J.A. Westgate for the identification of tephra samples. Earlier versions of the thesis were reviewed by W.C. Barnes, xvlii M.A. Church, R.E. Kucera, L.M. Lavkulich, W.H. Mathews, B.C. McDonald, G.E. Rouse, 0. Slaymaker, and D. Smith. A.C. Clague assisted in the field and typed the manuscript. Also gratefully acknowledged are the services and research facilities provided by the Department of Geological Sciences at the University of British Columbia. A 'A 1 CHAPTER ONE: INTRODUCTION Research on the surficial geology of the southern Rocky Mountain Trench was undertaken in order to (1) determine the origin and evolution of this remarkable valley, (2) test the applicability of certain tech^ -niques in the investigation of glacial and other deposits in cordilleran regions, and (3) provide geologic knowledge applicable to man's develop-ment of the region. The organization of the report is as follows. This chapter describes the geologic and geographic setting of the Rocky Mountain Trench in southeastern British Columbia. Previous geologic work in the area is briefly reviewed and the distribution and character of bedrock, semiconsolidated and unconsolidated sediments discussed. Chapters 2, 3, and 4 each present in detail one aspect of the Cenozoic geology of the southern Rocky Mountain Trench. Each documents a particular develop-mental stage in the history of the Trench; together they provide a record of the evolution of the Trench and of the processes which have formed and modified i t . Tertiary block faulting and structurally con-trolled sedimentation are discussed in Chapter 2, Pleistocene glaciation in Chapter 3, and late Pleistocene meltwater erosion and sedimentation in Chapter 4. The geomorphic history of the area is reviewed in Chapter 5, and applications of geologic information gained during the study which relate to man's future activities in the region and elsewhere are considered in Chapter 6. Suggestions for further research are out-lined in a concluding chapter. 2 PREVIOUS GEOLOGIC WORK The Rocky Mountain Trench is a major topographic depression, extending from northwestern Montana through British Columbia almost to the Liard River (Fig, 1), The part of the Trench occupied by the Kootenay River is a flat—bottomed, somewhat sinuous valley 3 to 16 km wide, bounded by the Rocky Mountains to the east and the Purcell Mountains to the west (Figs, 2 and 3). Major tributaries to the Kootenay River include St. Mary, Bull, and Elk Rivers, Most of the Trench within the study area is between 760 and 910 m in elevation. The Rocky Mountain Trench has interested earth scientists since i t was recognized as a major physiographic form. The surficial geology, however, has received l i t t l e attention. Early references to the geology of the region occur in the writings of Bauerman (1884) and Dawson (1885, 1890). Dawson (1885, p. 28-31 and 146-157) commented on the geomorphology of the Trench. Daly (1912) investigated the bedrock and structure of the southern Rocky Mountain Trench during his geologic explorations through the Cordillera. He briefly discussed glaciation of the Galton Range and the Purcell Mountains (p. 584-588). Shortly thereafter, Schofield (1915, 1920) mapped the geology of the Cranbrook area and commented on the genesis of the Rocky Mountain Trench. He described sediments predating the last glacial advance; these he termed the St. Eugene silts and assigned to the "St. Eugene interglacial (?) epoch," Hollick (1914, 1927) assigned a Tertiary age to fossil plants from the St, Eugene silt s , and Berry (1929) compared the St. Eugene flora to similar Miocene floras in 3 Washington and Idaho. The Miocene age of the St. Eugene s i l t s was accepted by Rice (1937) in his -memoir on the bedrock and structure of the Cranbrook-Kimberley area. However, the St. Eugene strata as defined by Rice are only the lower part of the unit specified by Schofield. Unconsolidated deposits are discussed brie f l y (p. 24-25) and are shown on the geologic map as undifferentiated Recent and Pleistocene s i l t , sand, gravel, and glacial d r i f t . Rice (1936) stated that glacial erosion in the Cranbrook area was minor and that deglaciation was accompanied by stagnation of large parts of the glacier in the Trench. The geology of the Trench and adjacent areas between 49° and 50°15'N latitude was mapped and discussed by Leech (1954, 1957, 1958, 1959a, 1959b, 1960, 1966). Quaternary deposits are shown as undifferen-tiated t i l l , gravel, sand, s i l t , and alluvium on Leech's geologic maps. Price (1962) mapped the geology of Elk Valley. The British Columbia Department of Agriculture conducted a s o i l survey of the Upper Kootenay Valley (Kelley and Sprout, 1956), The s u r f i c i a l geology is discussed in the survey report, and s o i l maps indi-cate the different s o i l parent materials and late Quaternary landforms. From gravity data Garland and others (1961) and Thompson (1962) proposed the existence of basins of Cenozoic sedimentation at Fort Steele, Jaffray, and Waldo, containing up to 430, 1100, and 1500 m of semi-consolidated to unconsolidated sediments respectively. Depth estimates based upon seismic refraction data (Lamb and Smith, 1962) are in general agreement with the gravity depth estimates. 4 Crickmay (1964) reviewed previous work bearing on the origin of the Trench and concluded that ice was a minor agent compared to running water in excavating the valley. He also stated, without supporting evidence, that the St. Eugene silts are Pliocene in age. Pardee (1910, 1942, 1950), Hershey (1912), Davis (1921), Alden (1953), Richmond (1965), and Richmond and others (1965) described the landforms and unconsolidated sediments of northwestern Montana and dis-cussed the geomorphic history of the region. Alden (1948), Pardee (1948) , and Johns (1970, p. 6-12) discussed the surficial geology of the Kootenai Valley area south of 49°N latitude. The present study is based upon field work completed during the summers of 1970, 1971, and 1972. Mapping was done on Department of Mines and Technical Surveys topographic maps of the Columbia River Basin, Upper Kootenay River Area, printed in 1951. These maps have a 20-foot (6.1-m) contour interval and a scale of 1:31,680. Aerial photographs supplied by the Surveys and Mapping Branch of the British Columbia Department of Lands, Forests, and Water Resources facilitated mapping. The photos were flown in 1968 and 1969 and are at a scale of 1:31,680. BEDROCK GEOLOGY Bedrock within the study area is composed of sedimentary rocks of Precambrian to Cretaceous age, volcanic rocks of Precambrian age, and intrusive igneous rocks of Precambrian and Early Cretaceous (?) age (Fig. 4, Table 1). 5 Stratified rocks of Precambrian age in the Canadian Cordillera have been broadly subdivided into the Purcell (older) and Windermere (younger) Groups. The Purcell Group is informally subdivided into upper and lower divisions, each of which consists of several formations, Purcell rocks crop out over large portions of the Purcell Mountains. The west front of the Rocky Mountains from near Premier Lake to Sand Creek is largely Lower Purcell rocks overlain by the Siyeh Formation, The Galton Range to the south is underlain mainly by Upper Purcell rocks, In general, Purcell strata on the east side of the Trench strike north to northwest parallel to the mountain front and dip east to northeast. Strata in the Purcell Mountains also tend to strike north to northwest, but folding and faulting complicate the geology. The Moyie intrusions are diorite and quartz gabbro s i l l s and dikes intruded dominantly into the Aldridge and Kitchener Formations of the lower division of the Purcell Group. These intrusions, especially common in St. Mary Valley, are of Precambrian age and may be related in origin to andesitic lavas at the top of the Kitchener and Siyeh Formations. Windermere strata crop out along the west side of the Trench north of 50°N latitude and are particularly common in the Purcell Mountains northwest of the study area. On the east side of the Trench these rocks are limited to small fault slices near 50°15'N latitude. Rocks of Paleozoic age crop out over large areas of the Hughes and Lizard Ranges of the Rocky Mountains, but are much less common in the Purcell Mountains. The west escarpment of the Rocky Mountains north of Lussier River and between Sand Creek and Elko consists of Paleozoic rocks 6 underlain by Upper Purcell strata, Cambrian rocks crop out at the north end of the Galton Range near the crest. Paleozoic rocks are also present along the floor of the Rocky Mountain Trench north of Premier Lake, between Kimberley and Cranbrook, from the vicinity of Bull River to Sand Creek, and at Roosville. Rocks of Mesozoic age include sedimentary rocks which crop out in Elk Valley and granitic rocks which occur as batholiths, stocks, and dikes intruding Precambrian and Paleozoic strata. The largest intrusion is a batholith at the headwaters of Skookumchuck Creek. Other intrusions within the Trench and too small to be shown on the geologic map (Fig. 4) include those near Bull River, on Kootenay Indian Reserve, on Sheep Mountain, and near Elko. SURFICIAL GEOLOGY Introduction Thick deposits of semiconsolidated and unconsolidated sediments are limited to the major valley bottoms and are of Tertiary and Quaternary age; the drift cover is thin or absent on the uplands (Fig. 5). Cenozoic sediments in the Trench as determined from gravity (Garland and others, 1961; Thompson, 1962) and seismic (Lamb and Smith, 1962) data are as much as 1500 m thick; in some places, however, bedrock crops out across nearly the complete width of the Trench. Three deep basins in the Trench are separated by bedrock with only thin sediment cover (Fig. 6). The thickness of Quaternary deposits has been estimated from deep stratigraphic exposures along the major river valleys in the Trench and from seismic records where low-velocity Quaternary sediments are d i s t i n -guishable from intermediate-velocity Tertiary deposits (Fig, 7), In general, Quaternary deposits are about 100 m or less i n thickness along the sides of the Trench, but are locally much thicker along the Trench axis. Deposits Despite the genetic complexity of many sediments, i t is conven-ient to classify them according to transportational agency. They may be further subdivided on the basis of depositional medium. In the Rocky Mountain Trench the following deposits are recognized: Transportational agency ice ice and water ice and water water water water wind mass movement Listed in order of decreasing age, the known deposits in the ^anglomerate is consolidated and semiconsolidated a l l u v i a l fan sediment. Depositional medium ice water (flowing) water (standing) water (flowing) water(transitional from flowing to standing) water (standing) air Deposits t i l l , erratics glaciofluvial sediments glaciolacustrine sediments alluvium (f l u v i a l and fan deposits1) deltaic sediments lacustrine sediments loess, dune sand landslide debris, colluvium 8 southern Rocky Mountain Trench include the following: the St. Eugene Formation, interglacial sediments, older drift, inter-drift sediments, younger drift, and postglacial sediments. Stratigraphic sections are presented i n Appendix 1. A generalized composite section is shown in Figure 8. St. Eugene Formation The St. Eugene Formation (Fig. 9), exposed beneath Quaternary sediments along St. Mary River, Elk River, and Gold Creek, consists of the following sediment types: (1) colluvium and fanglomerate, (2) s t r a t i f i e d s i l t and sand, and (3) coarse gravel with minor sand interbeds. Details of stratigraphy are presented in Chapter 2. Interglacial Sediments Interglacial sediments (Fig. 10) crop out along Sand Creek (49°21'25"N, 115°17'05"W) where they are overlain by t i l l ; the base of the unit is not exposed. The sediments consist of about 7 m of inter-bedded clay, s i l t , sand, and gravel ranging in color from very light gray to yellowish gray (N 8 to 5Y 8/1).2 Fragments of wood, of which the largest is about 0.5 m in length, are abundant. The lower part (4 m) of the interglacial unit at Sand Creek is gravel consisting of well-rounded pebbles and cobbles in a s i l t and sand Colors are from Rock-color Chart (Goddard and others, 1963). 9 matrix. Most clasts are 1 to 8 cm in maximum size; many are weathered, in contrast to those i n the overlying d r i f t which are fresh. Both parallel and cross-stratification are present, but are inconspicuous because variation in clast size within the gravel i s slight and because clay, s i l t , and sand interbeds are uncommon. The upper part (3 m) of the interglacial unit at Sand Creek is clay, s i l t , and fine-grained sand with weak parallel s t r a t i f i c a t i o n . Fine-grained sediment also occurs as rare interbeds and lenses within the underlying gravel. The gravel facies i s interpreted as a nonglacial, high-energy channel deposit, while the fine-grained facies represents either an overbank or lacustrine deposit. Abundant local vegetation supplied woody detritus to the accumulating sediment. The Sand Creek interglacial sediments are believed to correlate with Olympia Interglacial deposits of the coastal Pacific Northwest (Armstrong and others, 1965; Crandell, 1965) and with deposits of the Bull Lake-Pinedale interglaciation of the Rocky Mountains of the United States (Richmond, 1965) (Fig. 11). Wood from sand directly below t i l l yielded a radiocarbon date of 26,800 ± 1000 years (GX-2032; see Appendix 2). No other Rocky Mountain Trench sediments which definitely correlate with the Sand Creek interglacial sediments have been found. Some s i l t , sand, and gravel exposed in the walls of the major river valleys may be interglacial in origin. These sediments, however, differ 10 from those at Sand Creek in that they lack visible organic detritus and commonly contain dropstones; the gravel fraction is coarser, more poorly sorted, and consists of less weathered clasts. Thus, i t is thought that these sediments were deposited in a proglacial environment after deposi-tion of the interglacial unit at Sand Creek. Older Drift Older drift (Fig. 12) crops out along major valleys in the study area where i t overlies either bedrock or sediments of the St. Eugene Formation. More commonly, however, the base of the unit is not exposed, The glacial deposits consist of t i l l (maximum observed thickness 23 m) and outwash (maximum observed thickness 46 m). The former is a massive diamicton of clay, s i l t , sand, and clasts of a range of sizes (Appendix 3). It is light gray to very light gray (N 7 to N 8) except where clasts of grayish-orange and light-brown St. Eugene colluvium or fanglomerate are incorporated in the lower part of the t i l l . Clasts are subangular to subrounded, commonly striated, and consist of a variety of rock types cropping out in the mountains flanking the study area. The older t i l l 3 is overlain either by coarse outwash, inter-drift s i l t and sand, or younger t i l l . In the last case, the contact between the two t i l l s is generally unrecognizable because of similarities in texture and color. Where overlain by s i l t and sand, the t i l l grades upward into pebbly silts by a decrease in number of clasts. d T i l l of the older drift is referred to as the older or lower t i l l ; t i l l of the younger drift is called the younger or upper t i l l . 11 Figure 1. Index map of the Rocky Mountain Trench (darkened) and the study area (stippled). Position and extent of Rocky Mountain Trench from Holland (1964). 12 Outwash consists mainly of sandy gravel, its color differing with lithologic composition. Clasts are well rounded and generally lack glacial striations, Outwash occurs as lenses in older t i l l , as horizontal and cross-stratified beds, and as broadly folded beds locally cemented by calcium carbonate. The latter sediments are of particular interest because they are widely distributed and are thought to represent a thick supraglacial outwash mantle which deformed competently during melting of underlying stagnant ice. Strata commonly dip 5 to 30° with dip components perpendicular to the axis of the valley in which the unit is exposed (Fig, 13). High-angle normal and reverse faults and fractures cut the outwash and are oriented parallel to the walls of what must have been major meltwater discharge routes. Similar faults in glaciofluvial sediments have been shown to originate by adjustments in the sediment pile as masses of associated ice melt (McDonald and Shilts, in press). Broadly folded, cemented outwash passes downward into cross-stratified, uncemented sand and gravel; in places i t grades upward into uncemented inter-drift s i l t and sand or is overlain across an angular unconformity by younger drift. The older drift is considered to be Wisconsinan in age and to correlate with early deposits of the Fraser Glaciation and with sediments of the early stade of the Pinedale Glaciation (Fig. 11). Clasts in older drift are fresh in comparison to those in the Sand Creek interglacial unit; thus, the former unit is thought to be younger than the latter. The overlying inter-drift sediments were probably deposited in a pro-glacial environment during a relatively short interval between major ice advances. Since the Olympia Interglaciation was a nonglacial interval 13 Figure 3. The southern Rocky Mountain Trench, British Columbia. A. Floor and east wall of the Trench south of Elk River (BC 577:38). B. Trench floor and Kootenay River flood plain viewed toward the southeast from near Fort Steele (BC 898:50). C. Floor and west wall of the Trench north of St. Mary River (BC 583:68). D. Rocky Mountain Trench viewed toward the south from near 50°N latitude (BC 896:19). 1 5 Figure 3 16 of major duration (Armstrong and others, 1965; Fulton, 1971), the inter-drift sediments were most likely deposited during an interstade within the last glaciation. A rounded clast of wood from the older t i l l yielded a radiocarbon date of > 36,000 years (GX-2031; see Appendix 2). The wood was probably eroded from older sediments and redeposited in t i l l . Inter-drift Sediments Inter-drift sediments (Fig. 14) up to 61 m thick crop out along the valley walls of Kootenay River and i t s tributaries. They overlie older glacial deposits and are overlain by younger t i l l and outwash. The sediments are very light-gray to white (N 8 to N 9) s i l t , sand, and minor gravel. No organic matter has been found in these sediments. Individual strata range from laminae to beds more than a meter in thickness. Rhythmic bedding is uncommon, and many exposures appear massive. Soft-sediment deformation, although uncommon, has locally produced contorted beds overlain by undisturbed strata, load casts, flame-like projections of fine-grained sediment into overlying s i l t and sand, and coherent blocks of sediment that slumped toward the center of the depositional basin. Also present are high-angle faults and fractures, clastic dikes intruded from below, and channel scour-and-fill structures. Isolated pebbles and cobbles in s i l t and sand are present near the base and top of the unit and were probably deposited from ice rafts. Inter-drift s i l t and sand were deposited in one or more lakes which formed on the floor of the Rocky Mountain Trench during a period 17 Figure 4. Generalized geologic map of part of southeastern British Columbia (compilation from Leech, 1957, 1959a, 1960; Price, 1962). The following are the formations and groups included in the composite map units: map unit (1) Fort Steele, Aldridge, and Creston Formations; (2) Kitchener and Siyeh Formations, Dutch Creek Formation (includes Roosville, Phillips, and Gateway Formations), and Mount Nelson Formation; (3) Toby Formation and Horsethief Creek Series; (4) Eager, Cranbrook, and Burton Formations; (5) Elko and Jubilee Formations; (6) McKay Group, Wonah and Glenogle Formations; (7) Beaverfoot-Brisco Formation; (8) Fairholme and Rundle Groups, Harrogate, Burnais, Alexo, Palliser, Banff, and Exshaw Formations; (9) Rocky Mountain Formation; (10) Fernie and Blairmore Groups, Spray River and Kootenay Forma-tions. The dotted line is the boundary of the geologic map. 18 of glacier recession. The transition between these sediments and under-lying older outwash is gradational, with dropstones, gravel lenses, and sediment deformation (p. 12) increasing downward; thus, deposition of inter-drift s i l t and sand began in a proglacial environment. Near the top of the inter-drift unit there i s an upward increase in dropstones related to the readvance of the Cordilleran Ice Sheet indicating that the upper part of the unit was also deposited in a proglacial environment. Dropstones and other indicators of nearby glaciers are absent in the middle part of the unit, so i t is not known whether these sediments are glaciolacustrine or simply lacustrine in origin. The absence of visible organic matter throughout the unit may indicate that vegetation was never firmly established during this nonglacial interval. Inter-drift sediments are tentatively correlated with deposits of the interstade separating the early and middle stades of the Pinedale Glaciation (Fig. 11). Younger Drift Younger glacial deposits (Fig. 15) crop out over large portions of the floor of the Rocky Mountain Trench and i t s tributary valleys. These sediments are late Wisconsinan in age and are correlated with deposits of middle and late Pinedale age. They include t i l l , glacio-f l u v i a l sand and gravel, and glaciolacustrine clay, s i l t , and sand. The t i l l (up to 44 m thick) is massive diamicton similar in color and texture to t i l l of the older drift (Appendix 3), The younger 19 Table 1. Bedrock units in and adjacent to the study area (from Leech, 1954, 1958; Price, 1962). Era System or Series Croup or Formation Lithologies MESOZOIC and (?) CENOZOIC 1 Lower Cretaceous and (?) later Monzonite, quartz monzonite, granodiorite Not in contact Lower Cretaceous Blairmore Group • Shale, mudstone, sandstone, conglomerate Unconformity MESOZOIC Lower Cretaceous and Jurassic Kootenay Formation Shale, siltstone sandstone conglomerate, coal MESOZOIC Juras ic Fernie Group Shale, siltstone sandstone limestone MESOZOIC Disconformity Trias Spray River Formation Shale, siltstone dolomltic siltstone Unconformity Permian and Pennsylvanian Mississippian • Upper Devonian • Middle Devonian Middle Devonian or earlier Middle Devonian or earlier and (?) Upper Devonian Rocky Mountain Foraatlo Rundle Group Banff Formation Exshaw Formation Palliser Formation Alexo Formation Falrholme Group Harrogate Formation Burnais Formation Basal Devonian unit Siltstone, sandstone, dolomitlc sandstone, sandy dolostone and limestone, chert Limestone, crlnoldal and cherty In part Limy siltstone, limeptone, s i l t y limestone, cherty limestone Shale, limestone Upper member: argillaceous limestone; Lower (main) member: limestone Sandstone, sandy and argillaceous limestone Upper part: shale, limestone; Lover part: limestone, stromatolitlc and coralline in part; dolostone Shale, limestone, argillaceous limestone Limestone, dolostone, gypsum Shale, sandstone, dolomltic sandstone, conglomerate, dolostone sandy dolostone Unconformity Middle(?) and Lower Silurian, Upper Ordovlclan Beaverfoot-Brlsco Formation Limestone, dolostone; thin shale near top; sandstone and conglomerate locally at base Disconformity? Upper or Middle Ordovlclan Wonah Formation Unconformity? HiddleC?) and Lower Ordovlclan Lover Ordovlclan and Upper Cambrian Upper and/or Middle Cambrian Glenogle Formation McKay Group Jubilee and Elko Formations Shale, siltstone, limestone Shale, limestone, intraformational limestone-conglomerate Dolostone Disconformity? Middle Cambrian Middle(?) and Lover Cambrian Lower Cambrian Burton Formation Eager Formation Cranbrook Formation Shale, sandstone, conglomerate, limestone Shale, siltstone, sandstone, limestone Sandstone, conglomerate Unconformity Horsethief Creek Series A r g i l l i t e , conglomerate Windermere Toby Formation A r g i l l i t e , conglomerate Unconformity Upper Purcell or ( ) later Moyie intrusions Diorite; quartz gabbro Intrusive contact between Moyie intrusions and some Upper Purcell strata Upper Purcell-Lover Purcell -Mount Nelson Formation Dutch Creek _ Formation 'Roosville Formation Phillips Formation .Gateway Formation Kitchener Formation* Siyeh Formation* (west of Trench) (east of Trench) Creston Formation Aldridge Formation Aldridge Formation I (east of Trench) (west of Trench) 1 - , I Fort Steele FormatL L (east of Trench) A r g i l l i t e , sandstone, dolostone A r g i l l i t e , siltstone, sandstone, limestone, dolostone A r g i l l i t e , siltstone, sandstone A r g i l l i t e , dolomitlc a r g i l l i t e , siltstone, dolomltic siltstone sandstone, dolostone A r g i l l i t e , dolomltic a r g i l l i t e , sandstone, dolostone, sandy dolostone; andesitic lava and tuff at top A r g i l l i t e , sandstone, argillaceous sandstone A r g i l l i t e , sandstone, argillaceous sandstone A r g i l l i t e , dolomitlc a r g i l l i t e , sandstone, argillaceous sandstone, dolostone •Kitchener and Siyeh Formations are only in part equivalent. 20 t i l l is somewhat more clayey and silty near its base, probably due to the incorporation of underlying fine-grained inter-drift sediments. Clasts are subangular to subrounded, striated, and of a variety of rock types cropping out in and near the study area. The t i l l is calcareous except where derived entirely from the Purcell Mountains. A detailed study of t i l l composition is presented and discussed in Chapter 3. T i l l is underlain by either advance outwash gravel or inter-drift s i l t and sand. At some localities inter-drift sediments grade upward into t i l l without a sharp break. More commonly, there is a pronounced unconformity between the two. Some thick t i l l exposures may include both older and younger t i l l s with no intervening stratified sediments. Glaciofluvial sediments, consisting mainly of sandy gravel, occur locally beneath the younger t i l l (advance outwash), as lenses within the t i l l (englacial and subglacial outwash), and as channel and valley train deposits overlying the t i l l sheet (recessional outwash). Thickness of glaciofluvial sediments is highly variable, but is greatest along the margins of the Trench off tributary valleys and near the axis of the Trench along what is now the valley of Kootenay River, These were the sites of major meltwater drainage during and immediately following glaciation. A bedrock channel along the east side of the Trench just south of Elko contains 40 m of outwash. South of Sheep Mountain, very poorly sorted subglacial outwash is as much as 52 m thick and occupies much the same stratigraphic position as t i l l in the central part of the Trench. Where younger t i l l is thickest, underlying outwash is thin or absent. The spotty occurrence of advance outwash is due either to restriction to meltwater channels or subsequent erosion by ice. Figure 6. Map of Bouguer gravity in the southern Rocky Mountain Trench, and gravity profiles and inferred geologic sections across the Trench (from Thompson, 1962). 22 Figure 7. Thickness of Quaternary sediments. Areas where bedrock is at or near the surface inside the dotted line are stippled; areas of thick Quaternary sediments inside the dotted line are unpatterned. The d r i l l -hole data were provided by the Water Resources Branch of the British Columbia Department of Lands, Forests, and Water Resources. Seismic records were supplied by Shell Canada Ltd. 23 24 Likewise, recessional outwash is thickest beneath meltwater channels, valley trains, and kame terraces which dissect the drumlinized t i l l plain on the Trench floor. Gravel outwash, deposited in high-energy, proglacial river channels, consists largely of pebbles and cobbles and a matrix of sand. Clasts are well rounded, and sorting and stratification vary with amount of water-working prior to deposition. The more poorly sorted and weakly stratified glaciofluvial gravels probably represent proximal ice-contact deposits, worked by water to only a minor degree. Other gravels are better sorted and stratified, reflecting greater fluvial transport. Imbrication in the latter shows that meltwater drainage was in a southerly direction, much as the present drainage. Sand outwash, representing low-energy river or backwater deposits in meltwater channels, is most common as a thin layer over gravel outwash along the axial part of the Trench. Some discontinuous sand beds are associated with gravel outwash beneath and within the younger t i l l . The sedimentology and paleohydrology of recessional glaciofluvial sediments are considered in detail in Chapter 4. Glaciolacustrine clay, s i l t , and sand occur along the Trench margins and in tributary valleys at two stratigraphic levels within the younger drift: (1) i n t r a - t i l l glaciolacustrine sediments, and (2) supra-t i l l glaciolacustrine sediments. 25 i at I sS O u g| 5 £ ul Z <« i •n-.&.o:<V9.-' a • • • • • • • • a • • • • • • • • trtf • • • • • • • • Q.*0 O Cf.JD C]#.0 • • »p • •.•fl'.p •**• q»M*.p •" ££££ ENVIRONMENT OF DEPOSITION c i PROGLACIAL GLACIAL 2 PROGLACIAL PROGLACIAL GLACIAL PROGLACIAL PROGLACIAL and (?) NONGLACIAL PROGLACIAL, SUPRAGLACIAL GLACIAL NONGLACIAL NONGLACIAL Glaciolacustrine, glaciof luvial Glaciolacustrine Glaciof luvial Glaciof luvial Glaciolacustrine and (?) lacustrine Glaciof luvial Fluvial and (?) lacustrine Fluvial Locustrine, fluvial Mass movement and mudflow deposition COLLUVIUM, FANGLOMERATE !•••••[ STRATIFIED GRAVEL h/o.O-.'^UV-l TILL CLAY, SILT, SAND • RADIOCARBON DATE — 26,800 ± 1000 'GLACIOLACUSTRINE SEDIMENTS PRESENT IN TRIBUTARY VALLEYS. 2SEDIMENTS PRESENT ALONG TRENCH MARGINS. ^SEDIMENTS PRESENT IN ROCKY MOUNTAIN TRENCH NEAR TRIBUTARY VALLEYS. Figure 8. Composite columnar section of late Cenozoic sediments exposed in the southern Rocky Mountain Trench. Column heights are proportional to the maximum exposed thickness of each unit. 26 Figure 9. The St. Eugene Formation. A. Colluvium, Elk River. B and C. Fanglomerate, Elk River. D. Laminated organic-rich s i l t , sand, and gravel, St. Mary River. E. Laminated organic-rich s i l t , St. Mary River. F. Coarse gravel, St. Mary River. 27 Figure 9 28 Figure 10. Interglacial sediments, Sand Creek. A. Interbedded sand and gravel overlain by s i l t and sand. B and C. Interbedded sand and gravel containing fragments of wood. D. Gravel consisting of well-rounded pebbles in a s i l t and sand matrix. E. Cross-bedded gravel. F. Contact between interglacial sediments and overlying t i l l . 2 9 F i g u r e 10 30 I n t r a - t i l l glaciolacustrine sediments are massive and rhythmi-cally bedded clay, s i l t , and sand occurring along the margins of the Rocky Mountain Trench stratigraphically above inter-drift sediments, but within the younger t i l l . The sediments crop out near St. Mary Valley (up to 16 m thick) along the west side of the Trench, and near Elk Valley (up to 28 m thick) along the east side, but are absent as a distinct unit in central Trench exposures. Instead, at the same s t r a t i -graphic level are discontinuous i n t r a - t i l l lenses of ice-contact sand and gravel, and coarse channel f i l l . Although the glaciolacustrine unit along the Trench margins cannot be continuously traced laterally into the coarse outwash pockets at the center of the Trench, i t is believed that the two are genetically related, the former representing deposits in water ponded behind sediment or ice dams, and the latter deposits beneath or near remnant ice masses in the central part of the Trench. I n t r a - t i l l glaciolacustrine sediments were deposited in ice-marginal lakes during an interval of glacier recession. This interval i s tentatively correlated with the interstade separating the middle and late stades of the Pinedale Glaciation (Fig. 11). Trench-marginal glaciolacustrine deposition during this interval contrasts with Trench-wide deposition of inter-drift sediments during an earlier interstade when the entire floor of the Rocky Mountain Trench in the study area was temporarily free of ic e . Supra-till glaciolacustrine sediments were deposited in ice-dammed lakes in the major tributary valleys during f i n a l deglaciation of the southern Rocky Mountain Trench. The sediments, in general, are 31 RADIOCARBON YEARS B.P. SOUTHWESTERN BRITISH COLUMBIA WESTERN WASHINGTON ROCKY MOUNTAINS .BRIDGE RIVER ASH FALL. -ST. HELENS Y ASH FALL-•MAZAMA ASH FALL-10,000—r^-, o < I o 20,000-Sumas Stade Everson Interstade Vashon Stade Evans Creek Stade Z o § o Ul 3 Ul Z a! 30,000— OLYMPIA INTERGLACIATION 40,000— < z o 10 Younger glacial episode Nonglacial interval Older glacial episode Z O § .MAZAMA ASH FALL-Late stade Interstade -GLACIER PEAK ASH FAll -Middle stade Interstade Early stade INTERGLACIATION Late stade 'Second glacial episode Nonglacial interval First glacial _ episode Nonglacial interval Early stade Figure 11. Chronology and correlation of late Pleistocene events in the Pacific Northwest and the Rocky Mountains of the United States (from Crandell, 1965; Richmond, 1965). Figure 12. Older drift. A. T i l l , Elk River. B. Glaciofluvial gravel partially cemented by calcium carbonate, near confluence of Kootenay and Elk Rivers. C. Deformed glaciofluvial sand and gravel overlain by younger t i l l , Sand Creek. D. High-angle faults and fractures in outwash, St. Mary River (see also Fig. 13). Sand and gravel are partially cemented by calcium carbonate. 33 34 Figure 13. Attitudes of beds, faults, and fractures in deformed outwash of the older drift. Top: index map showing the location of the deformed outwash and the attitudes of beds. Bottom right: equal-area projection of poles to fault, fracture, and bedding planes. Bottom left: contoured diagram of poles to fault and fracture planes (contours approximately 3-8-13% per 1% area). 35 49"36'N 40°34'N K ILOMETERS FAULT AND FRACTURE LOCALITIES F FAULT OR FRACTURE w • STRATIFICATION " - " ^ . • Figure 13 36 Figure 14. Inter-drift sediments. A and B. Contact between inter-drift sediments and younger t i l l , Elk River. C. Basal inter-drift sand and dropstones, St. Mary River. D. Thick-bedded s i l t with a blocky fracture pattern, Elk River. E and F. Interlaminated s i l t and sand, Elk River. The uppermost s i l t and sand laminae in F were scoured and incorporated as clasts in the overlying sand. Figure 14 38 Figure 15. Younger drift. A. Advance outwash gravel underlying younger t i l l , Elk River. B and C. T i l l , St. Mary River. D. I n t r a - t i l l sand, St. Mary River. E. I n t r a - t i l l gravel, St. Mary River. F. Supra-till rhythmically-bedded s i l t and sand, St. Mary River. G and H. Gravel underlying late glacial meltwater channel, Kootenay River near Kikomun Creek. 39 40 Figure 16. Postglacial sediments and landforms. A. Eolian sand overlying outwash gravel, Kootenay River near Kikomun Creek. B. Al l u v i a l fan, near Grasmere. C and D. Mudflow gravel of an al l u v i a l fan built out onto the present Kootenay flood plain south of Wasa Lake. Mazama 0 tephra is the light layer at the tip of the shovel in D. E. Postglacial canyon of Bull River. F. Kootenay River flood plain and alluvium, near Kikomun Creek. 41 Figure 16 42 rhythmically laminated and contain dropstones. They exceed 50 m in thickness in St. Mary Valley and occur at least as high as 980 m above sea level. In Elk Valley, late glacial lake sediments crop out to about 1360 m above sea level. Silt and fine sand were also deposited in shallow ponds on the floor of the Trench during deglaciation. Although widespread, the sediments are thin (1 to 15 m in thickness). Deltas of coarse gravel formed where meltwater streams entered late glacial lakes and ponds. Postglacial Sediments Postglacial deposits (Fig. 16) include loess, dune sand, collu-vium, and alluvium. Loess and eolian sand up to 1 m thick form a surface mantle over much of the study area. Surface sand in some meltwater channels and drained ponds has been reworked by wind into small dunes. Talus cones, alluvial fans, and thin colluvial mantles occur along the margins of the Trench and major river valleys (Fig. 5). Degradation by rivers and streams began during or shortly after deglacia-tion. The rivers have since cut down through sediments and bedrock alike and have left multiple terraces bordering their valleys. Beneath the terraces interstratified sand and gravel are inset on older sediments. These terrace sediments were deposited during an interval of rapid downcutting following deglaciation and are thus very late Pleistocene or early Holocene in age. The main evidence for this is: (1) the occurrence of 6600-year-old Mazama 0 tephra (glass analyses by J.A. Westgate, Department of Geology, University of Alberta) in alluvial 43 fans graded to alluvium of the present Kootenay River flood plain, and (2) the presence of kettled terraces at Wasa Lake which are only 5 to 10 m above the present flood plain (Figs. 2 and 5). 44 LITERATURE CITED Alden, W.C, 1948, Pleistocene glaciation, in Gibson, Russell, Geology and ore deposits of the Libby Quadrangle, Montana: U.S. Geol. Survey Bull. 956, p. 49-61. 1953, Physiography and glacial geology of western Montana and adjacent areas: U.S. Geol. Survey Prof. Paper 231, 200 p. Armstrong, J.E., Crandell, D.R., Easterbrook, D.J., and Noble, J.B., 1965, Late Pleistocene stratigraphy and chronology in south-western British Columbia and northwestern Washington: Geol. Soc. America Bull., v. 76, p. 321-330. Bauerman, Hilary, 1884, Report on the geology of the country near the forty-ninth parallel of north latitude west of the Rocky Mountains: Canada Geol. Survey Rept. Prog. 1882-83-84, rept. B, p. 1-42. Berry, E.W., 1929, The age of the St. Eugene s i l t in the Kootenay Valley, British Columbia: Royal Soc. Canada Trans., ser. 3, sec. 4, v. 23, p. 47-48. Crandell, D.R., 1965, The glacial history of Washington and Oregon, in Wright, H.E., Jr., and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 341-353. Crickmay, C.H., 1964, The Rocky Mountain Trench: a problem: Canadian Jour. Earth Sci., v. 1, p. 184-205. Daly, R.A., 1912, Geology of the North American Cordillera at the forty-ninth parallel: Canada Geol. Survey Mem. 38, 857 p. Davis, W.M., 1921, Features of glacial origin in Montana and Idaho: Assoc. Am. Geographers Annals, v. 10, p. 75-147. Dawson, G.M., 1885, Preliminary report on the physical and geological features of that portion of the Rocky Mountains between l a t i -tudes 49° and 50°30': Canada Geol. Survey Ann. Rept,, v. 1, rept. B, p. 1-169. 1890, On the later physiographical geology of the Rocky Mountain region in Canada, with special reference to changes in elevation and to the history of the glacial period: Royal Soc. Canada Trans., ser. 2, sec. 4, v. 8, p. 3-74. Fulton, R.J., 1971, Radiocarbon geochronology of southern British Columbia: Canada Geol. Survey Paper 71-37, 28 p. 45 Garland, G.D., Kanasewich, E.R., and Thompson, T.L., 1961, Gravity measurements over the southern Rocky Mountain Trench area of British Columbia: Jour. Geophys. Research, v. 66, p. 2495-2505. Goddard, E.N., Trask, P.D., DeFord, R.K., Rove, O.N., Singewald, J.T., Jr., and Overbeck, R.M., 1963, Rock-color chart: Boulder, Colorado, Geol. Soc. America. Hershey, O.H., 1912, Some Tertiary and Quaternary geology of western Montana, northern Idaho, and eastern Washington: Geol. Soc. America Bull., v. 23, p. 517-536. Holland, S.S., 1964, Landforms of British Columbia, a physiographic outline: British Columbia Dept. Mines and Petroleum Resources Bull. 48, 138 p. Hollick, Arthur, 1914, A preliminary report by Mr. Arthur Hollick of the New York Botanical Garden, upon plants from the Pleistocene deposits: Canada Geol. Survey Summ. Rept., 1913, p. 133-135. 1927, The flora of the Saint Eugene si l t s , Kootenay Valley, British Columbia: New York Bot. Garden Mem., v. 7, p. 389-465. Johns, W.M., 1970, Geology and mineral deposits of Lincoln and Flathead Counties, Montana: Montana Bur. Mines and Geology Bull. 79, 182 p. Kelley, C.C., and Sprout, P.N., 1956, Soil survey of the upper Kootenay and Elk River valleys in the East Kootenay district of British Columbia: British Columbia Soil Survey Rept. 5, 99 p. Lamb, A.T., and Smith, D.W., 1962, Refraction profiles over the southern Rocky Mountain Trench area of British Columbia: Alberta Soc. Petroleum Geologists Jour., v. 10, p. 428-437. Leech, G.B., 1954, Canal Flats, British Columbia: Canada Geol. Survey Paper 54-7, 32 p. 1957, St. Mary Lake, British Columbia: Canada Geol. Survey Map 15-1957. 1958, Fernie map-area, west half, British Columbia: Canada Geol. Survey Paper 58-10, 40 p. 1959a, Canal Flats, British Columbia: Canada Geol. Survey Map 24-1958. 1959b, The southern part of the Rocky Mountain Trench: Canadian Mining Metall. Bull., v. 52, p. 327-333. 46 1960, Fernie, west half, British Columbia: Canada Geol. Survey Map 11-1960. 1966, The Rocky Mountain Trench: Canada Geol. Survey Paper 66-14, p. 307-329. McDonald, B.C., and Shilts, W.W., 1973, Interpretation of faults in glaciofluvial sediments, in Jopling, A.V., and McDonald, B.C., eds., Glaciofluvial and glaciolacustrine sedimentation: Soc. Econ. Paleontologists and Mineralogists Spec. Pub. (in press). Pardee, J.T., 1910, The Glacial Lake Missoula: Jour. Geology, v. 18, p. 376-386. 1942, Unusual currents in Glacial Lake Missoula, Montana: Geol. Soc. America Bu l l . , v. 53, p. 1569-1600. 1948, Physiography, in Gibson, Russell, Geology and ore deposits of the Libby Quadrangle, Montana: U.S. Geol. Survey Bu l l . 956, p. 61-67. 1950, Late Cenozoic block faulting in western Montana: Geol. Soc. America Bu l l . , v. 61, p. 359-406. Price, R.A., 1962, Fernie map-area, east half, Alberta and British Columbia: Canada Geol. Survey Paper 61-24, 65 p. Rice, H.M.A., 1936, Glacial phenomena near Cranbrook, British Columbia: Jour. Geology, v. 44, p. 68-73. 1937, Cranbrook map-area, British Columbia: Canada Geol. Survey Mem. 207, 67 p. Richmond, G.M., 1965, Glaciation of the Rocky Mountains, in Wright, H.E., Jr . , and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 217-230. Richmond, G.M., Fryxell, Roald, Neff, G.E., and Weis, P.L., 1965, The Cordilleran Ice Sheet of the Northern Rocky Mountains, and related Quaternary history of the Columbia Plateau, in Wright, H.E., J r . , and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 231-242. Schofield, S.J., 1915, Geology of the Cranbrook map-area, British Columbia: Canada Geol. Survey Mem. 76, 245 p. 1920, The origin of the Rocky Mountain Trench, B.C.: Royal Soc. Canada Trans., ser. 3, sec. 4, v. 14, p. 61-97. Thompson, T.L., 1962, Origin of the Rocky Mountain Trench in southeastern British Columbia by Cenozoic block faulting: Alberta Soc. Petroleum Geologists Jour., v. 10, p. 428-437. 47 CHAPTER TWO: THE ST. EUGENE FORMATION AND THE DEVELOPMENT OF THE SOUTHERN ROCKY MOUNTAIN TRENCH ABSTRACT The Tertiary history of the southern Rocky Mountain Trench is inferred from a study of the distribution, stratigraphy, fabric, litho-logic composition, structure, and palynology of the Miocene St. Eugene Formation in southeastern British Columbia. The St. Eugene Formation consists of flood-plain and fan facies and represents the upper part of up to about 1500 m of sediments which accumulated in the proto-Rocky Mountain Trench upon cessation of Laramide deformation and after i n i t i a t i o n of extension and block faulting in the eastern Cordillera during Eocene or early Oligocene time. Deep Tertiary basins i n the southern Rocky Mountain Trench are bounded on the east and west by high-angle faults parallel to the Trench margins and on the north and south by faults transverse to the trend of the Trench. Block faulting of a half-graben style was probably contemporaneous with sediment deposition, but at least 600 m of displacement on the east boundary fault postdates deposition of the St. Eugene Formation. Although there is no present seismic activity along the Rocky Mountain Trench north of latitude 49°N, Holocene fault scarps and earthquakes in a zone along the Rocky Mountains of the United States attest to the continuation of block faulting south of 49°N. 48 The St. Eugene microflora includes at least 39 genera of ferns, gymnosperms, and anthophytes. Phytogeographic reconstruction based upon the habitats of extant counterparts indicates floral elements growing on poorly drained lowlands, adjacent slopes, and montane uplands; thus, there was moderate to high relief in southeastern British Columbia during St. Eugene time. The climate apparently was temperate, with warm summers, mild winters, and abundant, uniformly distributed precipitation. This contrasts with the present climate of the southern Rocky Mountain Trench which is semiarid with hot summers and cold winters, and suggests that the mountain barriers which presently restrict cool, moist, Pacific maritime air masses to the coast were lower during the Miocene, or that the polar seas were relatively warm. INTRODUCTION The Rocky Mountain Trench is a long, narrow, intermontane valley that extends northwest from Montana through British Columbia. The origin of this major topographic feature has been long debated; various theories, summarized by Eardley (1951, p. 317-318), North and Henderson (1954), and Crickmay (1964), include the formation of the Trench by erosion without faulting and by erosion along a fault zone. Recent work (Garland and others, 1961; Lamb and Smith, 1962; Thompson, 1962) documents the presence of deep, sediment-filled, bedrock depressions in the southern Rocky Mountain Trench. Leech (1966) has emphasized that these basins are tectonic in origin, having formed by half-graben block faulting following Cretaceous to Paleocene (?) thrusting. The present configuration of the Trench results from sediment infilling of these tectonic depressions, 49 glaciation, and Holocene erosion. These conclusions are supported by evidence for Cenozoic block faulting in northwestern Montana (Pardee, 1950; Johns and others, 1963; Johns, 1970) and Tertiary sedimentation controlled by block faulting in the Flathead Valley parallel to and east of the southern Rocky Mountain Trench (Price and Wise, 1959; Barnes, 1963; Price, 1966). The continuity of the Trench for over 600 km indicates that i t is located along a major zone of crustal weakness, perhaps an ancient continental margin (Thompson, 1962; Price and Mountjoy, 1970; Berry and others, 1971). The nature of the structural control undoubtedly differs along the length of the Trench; although block faulting characterizes the southern part, other types of deformation may have affected the region of the Trench to the north. Further information on the Cenozoic tectonics of southeastern British Columbia has been gained by a study of late Tertiary sediments in the Rocky Mountain Trench. This paper describes these sediments, termed the St. Eugene silts by Schofield (1915), and discusses their implications regarding late Tertiary tectonics, paleogeography, and paleoclimatology. DISTRIBUTION AND CHARACTER OF THE ST. EUGENE FORMATION The St. Eugene silts were defined by Schofield as s i l t , sand, and gravel underlying t i l l of the last glaciation along St. Mary River between Wycliffe and St. Eugene Mission. So defined, the St. Eugene silts occur widely in the Rocky Mountain Trench and are exposed in the major 50 river valleys beneath late Wisconsinan drift, but the unit is a grouping of diverse geologic-climatic subunits of different ages. Schofield assigned the sediments to the "St. Eugene interglacial (?) epoch," although he thought a "pre-Glacial" age possible. A collection of fossil leaves from stratified s i l t near the base of Schofield's St. Eugene unit was examined by Hollick (1914, 1927), who thought the flora to be Tertiary in age. Berry (1929) placed the St. Eugene flora in the Miocene Epoch. Rice (1937) then redefined the St. Eugene silts as s i l t , sand, and gravel of Miocene age underlying a l l glacial and interglacial deposits. He lists four localities, a l l along St. Mary River in the vicinity of Wycliffe and St. Eugene Mission, where these sediments are exposed. The unit is here raised to formational rank in recognition of equivalent exposures along St. Mary River, Elk River, and Gold Creek (Fig. 17). Stratigraphic sections including the St. Mary River type locality sections are presented in Figure 18. The St. Eugene Formation unconformably overlies bedrock of Precambrian and Paleozoic age along the margins of fault-bounded basins, conformably (?) overlies sediments of Tertiary age near the basin centers, and is overlain by Pleistocene sediments. The following types of sediment constitute the St. Eugene Formation: (1) colluvium and fanglomerate, (2) stratified s i l t and sand, and (3) coarse gravel with minor sand interbeds (Fig. 19). In general, colluvium and fanglomerate are the oldest exposed sediments, and coarse gravel the youngest. A palynomorph assemblage from stratified s i l t and sand of the St. Eugene Formation is considered to be Miocene in age. 51 Colluvium and Fanglomerate Colluvium and fanglomerate are the lowest exposed sediments of the St. Eugene Formation. Colluvium crops out above bedrock along the valley walls of St. Mary, Kootenay, and Elk Rivers; fanglomerate is present above bedrock and colluvium along Gold Creek and Elk River (Figs. 17 and 18). St. Eugene colluvium is over 10 m thick in some St. Mary River exposures, and fanglomerate is at least 18 m thick at some Elk River sites. Colluvium consists of poorly sorted, weakly stratified, angular debris in a clay-silt matrix. Most of the sediment comprises clasts of one or a few lithologies which crop out nearby. For example, St. Eugene colluvium east of St. Eugene Mission consists of shale, siltstone, and sandstone derived from nearby Paleozoic bedrock. The colluvium is rock debris carried mainly by gravity to the base of hills or ridges. Fanglomerate (i.e., indurated alluvial fan sediment) also is a debris accumulation at the foot of a slope, but is deposited from mudflows or flowing water. Like the colluvium, i t is poorly sorted, consisting of clasts up to 1 m in diameter in a clay- and silt-rich matrix. Stratifica-tion is defined by discontinuous s i l t and sand beds. Many of the larger clasts are well rounded, but granules and small pebbles are commonly angular to subrounded. Sorting and clast roundness increase upward in fanglomerate exposures, presumably due to increased water-working of sediments during transportation and deposition. The only observed sedi-mentary structures other than discontinuous parallel stratification are 52 channels cut into, and fil l e d with, fanglomerate. Colluvium and fanglomerate are distinguishable from other sedi-ments in the Trench by their colors. Hues are generally brown, orange, or yellow (5YR, 10YR, 5Y), and values are above 5.1 In contrast, glacial deposits are light gray to very light gray (N 7 to N 8). The coloring of the colluvium and fanglomerate is thought to be due to the presence of iron oxide formed during weathering. Lengthy or intense weathering is also indicated by probable paleosols, rotten clasts, and secondary (?) clay. About 3 km northeast of St. Eugene Mission colluvium crops out along the north valley wall of St. Mary River and is overlain by St. Eugene s i l t and sand. Below the contact is a probable paleosol. The uppermost 2 m of colluvium are light brown to moderate orange pink (5YR 6/4 to 5YR 8/4), whereas the underlying colluvium is yellowish gray or grayish orange (5Y 7/2 or 10YR 7/4). This difference may be due to more intense weathering of shale and siltstone clasts in the upper part of the unit. Soil horizons in the paleosol are not well developed; eluvial and il l u v i a l zones and horizon structures, i f present, are indistinct. Brownish-gray (5YR 4/1) cemented zones (pans) up to 3 cm thick are laterally discontin-uous with wavy boundaries. Although present throughout the vertical range of the exposure, the largest number and best developed pans are in the upper few meters of colluvium. These soil characteristics are similar to those of some soils which form in subtropical regions (for example, see Bennema, 1963). Colors are from Rock-color Chart (Goddard and others, 1963). 53 Stratified Silt and Sand The middle unit of the St. Eugene Formation is stratified s i l t and sand. At Gold Creek and St. Mary River these sediments overlie colluvium and fanglomerate and are overlain by coarse St. Eugene gravel or drift; s i l t and sand are interstratified with fanglomerate along Elk River (Figs. 17 and 18). The unit is over 18 m thick in some St. Mary River exposures, but is less than 4 m thick along Gold Creek and Elk River. The unit consists of gently dipping beds of very light-gray to bluish-white (N 8 to 5B 8/1) s i l t and sand with abundant plant remains in places. Interbeds of well-sorted, well-rounded gravel are uncommon, although along St. Mary River the unit is locally underlain by such gravel. Parallel stratification is the most common sedimentary structure; large-scale trough cross-stratification is rare. The sediment probably accumulated in lakes or backwater ponds on flood plains, with the gravel layers representing higher energy channel deposits. Coarse Gravel with Minor Sand Interbeds The uppermost unit of the St. Eugene Formation is coarse gravel. It overlies fine-grained St. Eugene sediment across a sharp, unconformable (?) contact in eastern St. Mary River exposures and at Gold Creek, but correlative gravel is absent elsewhere (Figs. 17 and 18). However, simi-lar but finer gravel underlies s i l t and sand both northeast and northwest of St. Eugene Mission. Also, St. Eugene fanglomerate exposed along Elk 54 River is characterized by better sorting and higher clast roundness near the stratigraphic top of the unit, where i t approaches the coarse gravel in texture. Maximum observed thickness of the unit is about 35 m. The unit consists of well-sorted, stratified gravel with minor sand layers. Gravel clasts, which are as large as 1 m in diameter, are well rounded and highly weathered (most are stained by iron oxide, many are rotten). A matrix of medium- to coarse-grained sand surrounds the clasts. The sediment was deposited in river channels under high-energy conditions. PROVENANCE OF ST. EUGENE FANGLOMERATE Analytical Procedure Provenance of the St. Eugene Formation was determined from the fabric and lithologic composition of fanglomerate. Fan-fabric Analysis Elongate or plate-shaped particles transported in a viscous medium can assume a preferred orientation which is a function in part of the nature of the medium, particle shape, the terrain over which the particles are transported, and other external interactions with the medium. For sedimentary particles, preferred orientations developed during transport may be altered during or after deposition. The occurrence of oriented fabrics in t i l l is well documented, although an understanding of the means by which glacier ice produces such 55 fabrics is s t i l l incomplete. Nevertheless, t i l l fabrics have been used extensively to determine the flow directions of former glaciers (e.g., Holmes, 1941; West and Donner, 1956; Wright, 1957; Andrews and Smith, 1970; Lineback, 1971). Clasts can also become strongly oriented by flowing water as indicated by the common occurrence of clast imbrication in river channels. Lindsay (1968) has shown on theoretical grounds that strongly directed fabrics are also produced by mudflows. The extremely poor sorting, abundant clay and s i l t matrix, and discontinuous, weak stratification indicate that much of the St. Eugene fanglomerate was deposited by mudflows. Late glacial and early postglacial alluvial fans are prominent landforms in the interior valleys of southern British Columbia. These paraglacial fans (Ryder, 1971) formed from reworked drift deposited by streams and mudflows during final deglaciation of the Interior. Although differing somewhat in origin, St. Eugene fanglomerate is texturally simi-lar to sediment of the paraglacial alluvial fans. Method. The fabric of a paraglacial alluvial fan in south-western British Columbia was determined as a model for St. Eugene fanglom-erate (Fig. 20). The fan was selected because of continuous exposures of mudflow gravel along an arc about the source, and because the surface morphology of the fan defined the transport direction of the sediment. At each of four fabric sites, 2 to 4 m below the surface of the fan, the trend and plunge of 60 elongate pebbles {alb > 2, b/c - 1, a > 1 cm) were 56 measured.2 Data were plotted on an equal-area net and contoured according to the method reviewed by Ragan (1968, p. 79-82). A spherical probability distribution applicable to three-dimensional data was used to determine axes of maximum and minimum clustering for each fabric (Mark, 1973). Each data set was tested for randomness by using significance tests based on this distribution (Anderson and Stephens, 1971). The fabric of St. Eugene fanglomerate at Gold Creek and Elk River was determined by this procedure. Each fabric consists of 20 or 60 individual measurements; i t was found that mean vectors of smaller samples are similar to those of larger samples from the same populations for strongly oriented fabrics, but generally not for weak fabrics (Fig. 21). Andrews and Smith (1970, p. 520-521) have shown that for their till-fabric samples of size 25, 50, and 100, the smaller samples exhibit the main characteristics of the larger samples, and standard deviation does not increase with decreasing sample size. Results. The data from the paraglacial alluvial fan are presented in Appendix 4, and the results in Figure 20. Mean vectors are approximately parallel to the direction of transport and plunge up-fan. The mean vector of fabric no. 1 is parallel to the transport direction of a small fan just east of site no. 1; the gravel was probably deposited by mudflows originating at the apex of this fan. The data from St. Eugene fanglomerate are shown in Appendix 4, zThe long, intermediate, and short axes of clasts are called the a, b, and c axes respectively. 57 and the results, including maximum clast size, are plotted in Figure 22. Mean vectors from Elk River sites trend and plunge northeast toward the north end of the Galton Range. Gold Creek fan fabrics are weaker than those along Elk River and trends of mean vectors are variable. Clast Lithology Analysis The Purcell Mountains in the vicinity of Gold Creek and the Galton Range are underlain mainly by Precambrian sedimentary and igneous rocks (Fig. 23). Paleozoic sedimentary rocks crop out north of these areas. The sedimentary rocks are quartz arenite, s i l t i t e , argillite, in part dolomitic, dolostone, and minor limestone; the igneous rocks are mainly andesitic lavas. In the Galton Range these rocks strike north parallel to the range front and dip moderately to the east. Igneous rocks crop out in a single zone at elevations ranging from 910 to over 1500 m. Along the west margin of the Trench, near Gold Creek, sedimentary and igneous rocks are folded and offset along a series of steeply dipping, northwest-striking faults. Method. Clast lithologies were determined from random samples of 100 pebbles (£> axis 16 to 32 mm) collected from fresh exposures of St. Eugene fanglomerate. Similar lithologies are found in a number of formations, and some formations occur over large areas, thus i t is commonly impossible to determine the bedrock source for pebbles in fan-glomerate. Some lithologies, however, are useful for provenance study because they are restricted to one or a few bedrock units which crop out over a limited area. Rock types which crop out in the vicinity of fan-58 glomerate exposures and which are suitable for provenance study include andesite; grayish-red-purple quartz arenite, s i l t i t e , and argillite; and laminated greenish-gray argillite. The occurrence and relative abundance of these lithologies in relation to their present bedrock distribution indicate the probable source of the fanglomerate and provide evidence for Neogene deformation within the study area. Results. Percentages of laminated greenish-gray argillite, red-purple clastic sedimentary rocks, and mafic igneous rocks are presented in Figure 23. These lithologies represent the majority of fanglomerate clasts, the remainder being light-colored argillite probably derived from the same formations as the laminated greenish-gray argillite. The litho-logic composition of fanglomerate at Gold Creek is similar to that of bedrock cropping out to the west and north; fanglomerate along Elk River is lithologically most similar to bedrock cropping out to the east and northeast. Dolomitic argillite and dolostone which crop out on Sheep Mountain were not found as clasts in fanglomerate exposures to the immediate south. Clasts eroded from Paleozoic and Mesozoic rocks were not found. Mafic igneous clasts, although common in fanglomerate at Gold Creek, are absent in Elk River exposures. Discussion Although clast orientations parallel to flow were consistently observed in the case of the paraglacial alluvial fan (Fig. 20, Appendix 4), there are areas where fabric maxima transverse to the flow direction are to be expected (Fig. 24). Boulton (1971, p. 52-54) has shown that, in the 59 main body of a flow, longitudinal extension and vertical settling result in alignment of particles parallel to the flow direction; the dip of the a-b planes is dependent on the shape of the underlying bed. This fabric is replaced at the nose of the flow by one resulting from longitudinal compression and characterized by transverse a-axis orientations and a-b planes dipping up-flow. Since sediment deposited near the nose of most a l l u v i a l fan mudflows is volumetrically subordinate to sediment of the main body of the flow, fabrics taken randomly in a mudflow unit w i l l generally yield maxima parallel to flow, as was the case for the Fraser Canyon a l l u v i a l fan (Fig. 20). Transverse maxima can be recognized in a mudflow unit of unknown source i f the spatial fabric pattern of the unit is determined. From the results of geophysical surveys across the southern Rocky Mountain Trench, Garland and others (1961), Lamb and Smith (1962), and Thompson (1962) have inferred the presence of several fault-bounded basins containing up to 1500 m of semiconsolidated and unconsolidated sediments. The St. Eugene Formation represents the uppermost exposed Tertiary sediments of this basin f i l l . The geophysical results and those from fabric and clast lithology studies indicate that fanglomerate of the St. Eugene Formation was deposited in these basins and adjacent to uplands from which i t s constituent clasts were eroded. Transport was to the southwest and west on the east side of the Trench, and probably to the south, southeast, or east on the west side of the Trench. These results, as well as the coarseness of the sediment, suggest that re l i e f was moderate to high and that the physiography of the area during deposition of the St. Eugene Formation was broadly similar to that of the 60 present. The absence of mafic igneous clasts from fanglomerate along Elk River is noteworthy, because andesite crops out on the adjacent flank of the Galton Range. This is explained by upward displacement of the southern Rocky Mountains relative to the Rocky Mountain Trench after deposition of the St. Eugene Formation (p. 92). DEFORMATION OF THE ST. EUGENE FORMATION The St. Eugene Formation is only mildly deformed (Fig. 25). In fact, i t might be mistaken for a Quaternary unit were i t not for a detailed study of its fabric (p. 54), composition (p. 57), minor structures, and palynomorph assemblages (p. 61). Fanglomerate on the east side of the Trench strikes N50-80°E and dips gently southeast. The dip direction is approximately parallel to the adjacent front of the Galton Range. It is thought that these attitudes are the result of deformation rather than deposition, because fan-fabric and clast lithology analyses indicate a source to the northeast of the depositional site rather than north to northwest as would be the case i f the beds were deposited with the attitudes described. Furthermore, Sheep Mountain, the only nearby bedrock high which could have supplied detritus in a south to southeast direction, is underlain by bedrock which differs in lithology from clasts in the fanglomerate. Fanglomerate and interbedded s i l t and sand at Gold Creek strike about N45°E and dip gently southeast. Here the strata are offset by faults 61 which strike northeast and dip steeply northwest (Fig, 25). The rake of slickensides is steep indicating oblique-slip offset with the dip-slip component of movement greater than the strike-slip component. Stratified s i l t and sand, and coarse gravel along St. Mary River strike north to northeast and dip gently east to southeast. beds rich in fossil plant remains. Hollick (1927) identified 18 plants from fossil leaf and fruit impressions collected near St. Eugene Mission. His floral l i s t , modified according to LaMotte (1952) where necessary, includes the following: According to Hollick, the flora indicates a warm temperate climate such as that of the southeastern United States today. Berry (1929) concluded that the flora was late Tertiary in age on the basis of similarities with Miocene floras of the northwestern United States. ST. EUGENE PALEOFLORA AND PALEOCLIMATOLOGY Stratified s i l t and sand of the St. Eugene Formation include Alnus sp. Betula ulmoides Cocculus heteromorpha Fagus borineviliensis Fagus sp. Ficus canadensis Carpites interglacialis Carya egregia Monocotyledon Passiflora canadensis Platanus dissecta Quercus kootenayensis Quercus sp. Tilia (?) incertae Vaccinium pseudocorymbosum Vitis alia 62 Analytical Procedure The St. Eugene Formation was analyzed for plant microfossils in order to: (1) determine the assemblages that existed in the area during deposition of the St. Eugene Formation; (2) compare the assemblages from the different outcrops; (3) assess the Miocene age assigned to the unit; (4) compare the St. Eugene microflora with similar floras elsewhere in British Columbia and the northwestern United States, and thus infer the late Tertiary climate and geography of the area. Method Samples were collected from outcrops of the St. Eugene Formation along St. Mary River (four sites), Elk River (three sites), and Gold Creek (one site) (Fig. 26). At each site the total thickness of exposed s i l t and sand was sampled by carefully cleaning the exposure face, then trenching the sediments. Spot samples were also collected from fossil-iferous beds at the Gold Creek and St. Mary River sites. No organic-rich layers are present in the relatively thin s i l t and sand beds at the three Elk River sites. Plant microfossils were recovered from sediment samples by means of standard palynological techniques (outlined in Brown, 1960; Kummel and Raup, 1965). Briefly, this involved: (1) elimination of mineral matter by treatment in hydrochloric acid, then hydrofluoric acid; (2) removal of cellulose by boiling in a solution of acetic anhydride and sulfuric acid (acetolysis); (3) elimination of carbonized exteriors of 63 microfossils with nitric acid. Pollen and spore concentrates were then alkalized, stained with safranine, and mounted on glass slides for further examination. Spores and pollen were identified to genus where possible; certain genera are represented by more than one species, Taxa were listed in two general categories, palynomorph present and palynomorph common. Results ' The palynomorph assemblages from the St. Eugene Formation are summarized in Table 2. Of the localities sampled, three St. Mary River sites and the Gold Creek site yielded plant microfossils. Samples from thin sand interbeds in fanglomerate and coarse gravel contained no palynomorphs. No major differences were found among floral assemblages at various sampling sites, except that the number of genera recognized is less in samples with low pollen and spore content. The most common palynomorphs are found in a l l samples; the rarer ones occur sporadically. These observations suggest that the fossil pollen rain was similar at sampling sites along Gold Creek and St. Mary River, and that the strata of these two areas are of the same age. At least 39 genera of ferns, gymnosperms, and anthophytes are represented (Fig. 27). An examination of Table 2 indicates that the dominant palynomorphs are spores of the Polypodiaceae-Dennstaedtiaceae group, pollen of Pinaceae (Cedrus, Picea, and Pinus), and pollen of Alnus and Betula. Less common woody elements include the Taxodiaceae (.Glypto-strobus, Metasequoia, Sequoia, and Taxodium), other members of Pinaceae 64 (Abies, Pseudotsuga, and Tsuga), and a variety of dicotyledons (Acer, Aesculus, Carpinus, Carya, Castanea, Corylus, Fraxinus, Ilex, Juglans, Liquidambar, Myrica, Nyssa, Platanus, Pterocarya, Quercus, Salix, Tilia, and Ulmus-Zelkova). The following families and genera are also present: Caprifoliaceae, Chenopodiaceae, Compositae, Ephedra, Ericaceae, Gramineae, Pachysandra-Sarcococca, Podocarpus, and various ferns. Discussion Fossil floras provide information as to the age of associated sedimentary strata, the climate existing during growth of the plants, and phytogeography near the depositional site. These factors are often interdependent, therefore, caution must be exercised in interpretation of microfloras. For example, climate at the site of growth is in part determined by regional topography and elevation. Nevertheless, certain conclusions can be drawn regarding the age and paleoecology of the St. Eugene Formation by utilizing the following approaches: (1) the St. Eugene assemblage can be compared with dated floras of similar composition from the same paleoprovinces; (2) the climatic and edaphic requirements of members of the St. Eugene flora may be deduced from the modern habitats of similar plants; (3) paleogeography may be inferred from the physical characteristics of the St. Eugene sediments. Age Assigned by Comparison with Similar Floras Penny (1969) and Leopold (1969) have reviewed microfloras of late Mesozoic and Cenozoic age. In middle and high latitudes of the 65 Northern Hemisphere, Paleogene floras are strikingly different from the Neogene floras which are of particular interest here. In general, late Tertiary floras are less rich in species and contain a higher proportion of extant taxa than those of early Tertiary age from the same region. Certain groups such as the Compositae first appear in the fossil record in the Neogene. Neogene floras are more closely related to the modern floras living near the fossil locality than are Paleogene floras. Also, late Tertiary floras are more provincial in distribution than those of early Tertiary age. These differences are due to the progressive evolution and migration of floral elements in response to a gradual deterioration of climate and differentiation of topography during the Cenozoic. Early Tertiary floras show marked similarities throughout the Northern Hemisphere and commonly grew under tropical or subtropical climatic conditions (Penny, 1969). In later floras, the palms and broadleaf evergreens are replaced by warm temperate, deciduous anthophytes and conifers. In the late Tertiary, cool temperate taxa and herbs are important floral elements. Although many late Tertiary megafloras from the Pacific North-west have been described (references in Chaney, 1959; Axelrod, 1964), detailed palynological studies are few. Table 3 illustrates the similarity between the St. Eugene and some other microfloras of Miocene age in British Columbia and the northwestern United States which are located in Figure 28. Because microfloras and megafloras from the same fossil locality often differ (in part because of relative differences in pollen, leaf, and fruit productivity and the relative transportability and preservation of these materials), direct comparisons are not made with Miocene megafloras. 66 Piel (1969, 1971) has described a succession of middle to late Tertiary pollen floras in central British Columbia. The Miocene assem-blage, indicative of a warm temperate climate with abundant summer rain-f a l l , contains essentially the same taxa as the St. Eugene Formation. The deciduous broadleaf element which characterizes Piel's Miocene unit is subordinate to the coniferous element in the succeeding Mio-Pliocene assemblage which Piel believes to have existed under a cool temperate climate. Also clearly different is the subtropical to warm temperate Oligocene flora which includes many genera not found in younger rocks of the area (e.g., Diervilla, Engelhardtia, Psilastephanocolpites, Sciadopitys, and Sigmopollis). Martin and Rouse (1966) have discussed a late Miocene or early Pliocene microflora from the Queen Charlotte Islands off the coast of British Columbia. Generically, the flora is remarkably similar to that of the St. Eugene Formation (Table 3), although gymnosperms (specifically Cedrus, Picea, and Pinus) are relatively more abundant in the St. Eugene flora. Fern spores, especially the Polypodiaceae-Dennstaedtiaceae complex, are abundant in both floras. The microfossil assemblage from the Queen Charlotte Islands apparently grew on a coastal lowland under a relatively humid, mild temperate climate. Eocene and Miocene microfossil assemblages from the Whatcom Basin, southwestern British Columbia and northwestern Washington have been described by Hopkins (1966, 1968). The younger assemblage includes the following dicotyledons and gymnosperms: Acer, Alnus, Carpinus, Carya, Castanea, Cedrus, Engelhardtia, Fagus, Glyptostrobus, Ilex, Juglans, 67 Keteleeria, Liquidambar, Metasequoia, Momipites, Picea, Pinus, Pterocarya, Quercus, Salix, Tilia, Taxodium, and Ulmus-Zelkova. Again the similarity with the St. Eugene Formation is marked. Absent are many of the charac-teristic Eocene palynomorphs including Anemia, Azolla, Cicatricosisporites, Pistillipollenites, and Platycarya. The Miocene flora grew on the Whatcom Basin lowland and upland basin margins under a temperate to warm temperate climate. Microfossil assemblages from five localities in south-central British Columbia are considered by Mathews and Rouse (1963) to be late Miocene or early Pliocene in age. Associated volcanic rocks at two localities yielded K-Ar dates of 10 ± 2 and 12 and 13 million years. These floras are somewhat less diverse than the St. Eugene flora, contain a higher percentage of total conifer pollen and a lower percentage of spores (especially the family Polypodiaceae), and lack Betula and Cedrus, two common constituents of the St. Eugene Formation. These floras, then, appear slightly younger than the St. Eugene microflora, although phyto-geographic factors such as nearness to uplands may in part explain the differences. The St. Eugene flora is generally similar to Middle and Upper Miocene floras from Mascall, Blue Mountains, Stinking Water, Sucker Creek, and Trout Creek of eastern Oregon (Chaney, 1959; Graham, 1965). Tuffs associated with the last two have been dated at 16.7 and 13.1 million years respectively. Differences among the floras are attributed in large part to phytogeographic variables and, to a lesser extent, age. 6 8 Figure 17. Index map showing outcrop areas of the St. Eugene Forma-tion. The relative ages of two or more sediment types which crop out at one site are indicated by diagonal lines separating the older (to the right of the line) from the younger (to the l e f t of the l i n e ) . 69 70 Gray (1964, her Table 1) has compiled Tertiary microfloral lists for Washington, Oregon, Idaho, and California. The St. Eugene microflora is closely related to the Miocene assemblages from these states (Table 3) and shows less similarity to both Oligocene and Pliocene floras. Somewhat more distant from southeastern British Columbia is the Miocene Kilgore fossil locality in northern Nebraska (MacGinitie, 1962). Nevertheless, the Kilgore and St. Eugene floras are similar, the major difference between the two being the comparative rarity of gymnosperm pollen from Nebraska. In summary, the palynomorph assemblage of the St. Eugene Formation can be considered Miocene in age on the basis of similarities with other microfloras from western North America assigned to this epoch by means of vertebrate and invertebrate fossils and radiometric age dates. The correctness of this age assignment is further indicated by the degree of modernity of the flora. Wolfe and Barghoorn (1960) have shown a close relation between mega- and microfloral age and the proportion of fossil genera s t i l l living near their fossil localities in western North America (Fig. 29). The progressive modernization of floras during the Tertiary presumably resulted from gradual cooling and topographic differentiation. The closeness of the relationship in Figure 29 based on widely separated floras is somewhat surprising and suggests that local climatic and topo-graphic considerations were of secondary importance in comparison to age in characterizing the degree of floral modernity. The St. Eugene assem-blage with about 44% native genera is determined to be of middle Miocene Figure 18. Stratigraphic sections of the St. Eugene Formation. T i l l unconformably overlies the St. Eugene Formation at sites 1, 3, and 4, and outwash at sites 2 and 5. 72 Figure 19. The St. Eugene Formation. A. Colluvium, Elk River, B. Fanglomerate, Elk River. C. Laminated organic-rich s i l t , St. Mary River, D and E. Coarse gravel, St. Mary River. F. Organic-rich s i l t (light layer) overlain by coarse gravel and underlain by fanglomerate, Gold Creek. 73 Figure 19 Figure 20. Fabrics of paraglacial alluvial fans. Top: location and morphometry of the studied alluvial fans. Fan-fabric sample sites and axes of maximum clustering at each site are shown. Bottom: contoured diagrams of fabric data (contours approximately 2-5-8% per 1% area; each fabric consists of 60 measurements). Photo numbers BC 4246-180 and -181. Sample statistics are presented in Appendix 4. 76 age and is grouped with such floras as the Sucker Creek (41%), Mascall (40%), and Trout Creek (47%). Paleoecology Inferred from Ecological Requirements of Extant Counterparts The paleoecology of fossil plant assemblages traditionally has been assessed by determining the distribution and ecological requirements of the extant counterparts of fossil genera. This approach was first used in studies of fossil leaves by such workers as Brown (1934), Chaney (1936, 1938, 1959), Axelrod (1941), and Dorf (1959, 1963), and later was applied to palynological studies by, among others, Traverse (1955), Rouse (1962), Hills (1965), Stanley (1965), Martin and Rouse (1966), and Rouse and others (1970). In such studies i t is generally assumed that fossil plants had environmental requirements similar to their extant counterparts. This assumption has certain limitations (Wolfe and Hopkins, 1967; Wolfe, 1971; Hopkins and others, 1972), but its application here is supported by the observation that the St. Eugene flora consists of associations of plants that today occur together in restricted climatic provinces. Also, although most modern genera have wide geographic and ecological ranges, plant associations are more restricted in their distribution and ecological requirements. The St. Eugene flora resembles the modern flora of the eastern and southeastern United States, especially of those areas between the Appalachian Mountains and the Mississippi Valley where oak-hickory-walnut-elm-beech forests border upland areas dominated by mixed deciduous and coniferous elements. A similar resemblance between Miocene floras of the 50 £ 30 50 -40 -L _|_ 10 _L 30 20 Azimuth 30 40 10 Plunge DIFFERENCE BETWEEN AXES OF MAXIMUM CLUSTERING FOR SAMPLES OF SIZE 60 AND 20 (in degrees) • one point £ two coincident points Figure 21. Relation between fan-fabric strength and differences in the fabric results for samples of size 60 and 20. Theta is the standard scattering angle around the axis of maximum clustering (in general, the greater the directional strength of the fabric, the smaller is theta). Theta is plotted against the difference in orientation of the clustering axes of the larger and smaller samples. Differences between samples of size 60 and 20 are small where fabrics are strongly directed. Figure 22. Fan fabrics from the St. Eugene Formation. Right: index map of sample sites. Clustering axes at each site were determined from the sum of 2 or 3 samples with maximum lateral spacings of about 100 m. Left: contoured diagrams of fabric samples (contours approximately 2-5-8% per 1% area; each fabric consists of 60 measurements). Sample statistics are presented in Appendix 4. F A N FABRICS Figure 22 80 Columbia Plateau and the modern floras of the eastern and southeastern United States has been noted by Chaney (1959). The presence of Taxodium and other lowland genera indicate the presence of swamps or poorly drained lowlands in southeastern British Columbia. In contrast, the strong representation of Pinaceae suggests a nearby montane region. Mixed deciduous hardwood and conifer forest probably occupied the slopes between these two habitats. The St. Eugene flora is also similar to the extant flora of warm temperate parts of central and southeastern China. Asian genera no longer native to North America include Cedrus, Glyptostrobus, Metaseguoia, and Pterocarya. Numerous other St. Eugene genera are associated with these Asiatic trees under climatic conditions close to those existing in southeastern North America. Thus, the St. Eugene plant community consists of plants which now occur as two separate floral elements subject to the same summer-wet, temperate climate (Chaney, 1948, 1959). If the climatic requirements of Glyptostrobus and Taxodium are about the same today as during the Miocene, the lowland habitats in southeastern British Columbia during the Miocene probably had the following climate: average annual precipitation, 100 to 150 cm, evenly distributed through the year; mild winters with temperatures seldom below freezing and warm summers (Fullard and Darby, 1967). If upland slopes were climatically similar to the Appalachian-Mississippi Valley region, the Miocene mixed deciduous hardwood-conifer forest received similar amounts of precipitation, with the mean temperature between 10 and 18°C. In contrast, the present climate of the floor of the southern Rocky Mountain Trench according to the KBppen classification 81 Figure 23. Percent of clasts of the following lithologies in St. Eugene fanglomerate: laminated greenish-gray argillite (source is largely the Roosville and Gateway Formations), red-purple argillite and quartz arenite (largely of the Phillips Formation), and mafic igneous rocks (Purcell volcanics). No mafic igneous clasts were found in fanglomerate exposed along the east side of the Trench. Bedrock geology from Leech (1958, 1960). 82 is Bsk (Continental Cold Semiarid) (Krajina, 1965). Average annual precip-itation at Newgate (1918-1954) and Cranbrook (1916-1954) is 35 and 37 cm respectively and is distributed approximately evenly through the year. Mean temperatures at Newgate and Cranbrook are 6 and 5°C respectively. The climate i s truly intemperate, with average summer temperatures above 16°C and average winter temperatures well below freezing (Kelley and Sprout, 1956). The present climate is controlled in part by the southward movement of cold continental air from the Arctic and by the northward movement in the summer of warm, dry air from the interior of the United States. Low precipitation in part results from isolation of the area from the moderating influence of the Pacific by a series of north- to northwest-trending mountain ranges. Mathews and Rouse (1963) have suggested that the humid, warm climate necessary to sustain the Miocene floras of south-central British Columbia might have resulted i f the Coast Mountains were much lower during the late Tertiary than at present. Another factor responsible for the humid, temperate climate may have been the existence of a warm polar sea during the Tertiary (Pie l , 1971, p. 1895). Under such conditions and with atmospheric circu-lation similar to that of the present, warm, moist air would move south through British Columbia instead of the cold, dry air which is a factor controlling the present distribution of vegetation. High precipitation would result as these warm, moist air masses were driven against mountain ranges by westerly winds from the Pa c i f i c . The presence of late Tertiary mountain ranges in southeastern British Columbia is shown by regional tectonic evidence and by the compo-sition of the St. Eugene f l o r a . The Rocky Mountains originated during oo CO STONES WITH a-b PLANES DIPPING IN DIRECTION OF SHORT LINE; LONG LINE SHOWS a-AXIS TREND. STONES WITH VERTICAL a-b PLANES; LONG LINE SHOWS a-AXIS TREND. STONES WITH HORIZONTAL a-b PLANES; LONGEST DIMENSION Of ELLIPSE SHOWS a-AXIS TREND. DIRECTION OF TRANSPORT. Figure 24. The fabric of mudflow gravel. A. Small debris flow in Elk Valley. B. Mudflow clast orientations. 00 •p-Figure 25. Deformation of the St. Eugene Formation. A, B, and C. Faulted St. Eugene strata, Gold Creek. Faults strike northeast and dip steeply northwest. The rake of slickensides is also steep (B and C). Overlying Quaternary sediments are not offset. D. Gently tilted St. Eugene fanglomerate, Elk River. The fanglomerate dips gently to the south and southeast, but the source of the sediment, the precipitous east wall of the Trench, is east and northeast of the depositional site. 8 5 86 the Laramide Orogeny which probably began in the late Cretaceous and ended in the early Oligocene (Russell, 1951, 1954b; Douglas and others, 1968, p. 464). Epeirogenic uplift continued from the Oligocene until late Pliocene, with many areas of the Rockies uplifted 1500 m or more during this interval (Cook, 1960). Diverse habitats for St, Eugene floral elements also indicate moderate to high relief in southeastern British Columbia during the Neogene; for example, Glyptostrobus and Taxodium were probably limited to lowlands near the depositional site, whereas the strong representation of conifers such as Abies and Picea suggests the presence of montane uplands. Paleogeography Inferred from Physical Characteristics of Sediments Moderate to high relief near the depositional site is also suggested by the physical characteristics of the St. Eugene Formation and the Eocene-Oligocene Kishenehn Formation (Price, 1962; Johns, 1970) in the Flathead Valley to the east. Gravel and fanglomerate of the St. Eugene Formation were deposited by rivers from high-gradient tributary valleys and by mudflows from adjacent mountain fronts (Fig. 30). Barnes (1963, p. 70) has concluded that the Kishenehn Formation accumulated in large part in flood-plain lakes, swamps, and river channels; deposition of coarse conglomerate marginal to the broad flood plain was by mudflows from bordering uplands. 87 P L A N T M I C R O F O S S I L S A M P L E SITES SAMPLE SITE: PLANT MICROFOSSILS COMMON PLANT MICROFOSSILS RARE OR ABSENT CHANNEL SAMPLE J • o ST. EUGENE FORMATION—SEDIMENT TYPES: STRATIFIED GRAVEL SILT AND SAND 1 : l : III i COLLUVIUM OR FANGLOMERATE K-^.j-j Figure 26. Plant microfossil sample sites from the St. Eugene Formation. Columnar sections show locations of microfossil channel samples. 88 NEOGENE DEFORMATION IN SOUTHEASTERN BRITISH COLUMBIA The St. Eugene Formation provides evidence for Tertiary tectonic activity in the eastern Cordillera. The restricted occurrence of the St. Eugene Formation in the Trench and the derivation of the strata from adjacent uplands indicate that the Trench was already a major physiographic form by the Miocene. Alden (1953) has suggested that the Purcell and adjacent trenches developed by erosion and faulting prior to the Miocene. This is deduced from the fact that lake beds were deposited in the southern Purcell Trench and later covered by Columbia River basalts. Alden interpreted some bedrock benches in major valleys and basins of western Montana and Idaho as remnants of Pliocene valley bottoms produced during regional uplift prior to glaciation. Most of the major British Columbia drainage routes were established by the Miocene as evidenced by occurrences of Miocene sediments along the Fraser Valley and its tribu-taries near Quesnel and Prince George (Piel, 1969). The Miocene age of the St. Eugene Formation, however, represents only a minimum age for the formation of the southern Rocky Mountain Trench as a major physiographic form and sedimentation trough. The occurrence of deep structural basins with Tertiary f i l l s in the Trench indicates that unexposed sediments overlying Precambrian and Paleozoic bedrock in these basins may be older than Miocene. About 2000 m of late Eocene and early Oligocene sediments crop out in the Flathead Valley to the east in a similar stratigraphic and structural setting (Russell, 1954a; Barnes, 1963; Price, 1966). These sediments unconformably overlie strata deformed 89 Table 2. Ralynomorph assemblages from the St. Eugene Formation. Assem-blage no. 2 includes microfossils from samples collected at two adjacent loca l i t i e s (see Fig. 26). P = palynomorph present; C = palynomorph common. „ , u Gold Creek* St. Mary River* Palynomorph taxa (1) (2) (3) Division Lycopodophyta Lycopodium P P Selaginella P Division Pterophyta Deltoidospora P Osmunda P P P Polypodiaceae-Dennstaedtiaceae c c P (includes Laevigatosporites) Triplanosporites P P P Division Coniferophyta Abies P P P Cedrus C P C Cupressaceae-Taxaceae P P P Ephedra P Glyptostrobus P P P Metasequoia P P P Pice a c c C Pinus c c c Podocarpus p ? Pseudotsuga-Ijarix p p p Sequoia ? ? Taxodium p p p Tsuga p p p Division Anthophyta Acer p p p Aesculus p Alnus c c c Betula p c c Caprifoliaceae p ? Carpinus-Ostrya p-Carya p p p Castanea ? Chenopodiaceae-Amaranthaceae p p p Compositae p ? Corylus ? p Ericaceae p p p Fraxinus p p Gramineae p p Ilex p p p Juglans p p p Liquidambar ? p Hyrica p ? Nyssa p Pachysandra-Sarcococca p Platanus 1 Pterocarya p p p Quercus p ? Salix ? p P Tilia p Ulmus-Zelkova p p P *Location of sample sites: (1) 49°04'N ,. 115014'W; (2) 49°35'N, 115°49'W; (3) A9°36'N, 115°44'W 90 Figure 27. Selected plant microfossils from the St. Eugene Formation. A—Ly co podium', B—Osmunda; C—Polypodiaceae-Dennstaedtiaceae; D— Triplanosporites; E—Cedrus; F—Ephedra; G—Glyptostrobus; H—Picea; I—Tsuga; J—Aesculus; K and L—Alnus; M—Betula; N—Carya; 0—Compositae; P—Ilex; Q—Juglans; R—Liquidambar; S—Nyssa; T—Pterocarya; U— Pachysandra-Sarcococca; V—Ulmus-Zelkova; W and X—unidentified t r i -colpate pollen. Bar length represents 10 um. 91 92 during the Laramide Orogeny and, therefore, establish an upper age for this event. Late Cretaceous and early Tertiary postorogenic sediments accumulated in intermontane troughs along the northern Rocky Mountain and Tintina Trenches (Eisbacher, 1972). Sedimentation in the southern Rocky Mountain Trench was probably controlled by fault-block topography. The eastern margin of the Trench coincides with a major normal fault (Leech, 1966). Coarse fanglomerate derived from the east and northeast occurs on the downthrown side of the fault (p. 59). Similar occurrences of fanglomerate on the west side of the Trench may be associated with a series of steeply dipping, north- to northwest-striking faults north of 49°N latitude. Thompson (1962) has concluded that the deep (up to 1500 m) bedrock basins beneath the Trench were formed by normal faulting. Presumably these basins are bounded by faults transverse to the trend of the Trench. East-west faults striking beneath the unconsolidated sediments in the Trench have been mapped by Leech (1960). At least some faulting postdates deposition of the St. Eugene Formation. Strata along Gold Creek are offset by northeasts-striking dip-sl i p faults. The clast composition of fanglomerate adjacent to the normal fault on the east margin of the Trench indicates a minimum of 600 m of post-St. Eugene displacement. Mafic igneous clasts are absent from this fanglomerate (elev. 760 to 910 m) despite the bedrock occurrence along the adjacent mountain front of a mafic lava as high as 1500 m (p. 57). Rather, the fanglomerate consists of lithologies present in bedrock units stratigraphically above the lava. This suggests that units below and 93 Table 3. Comparison of the St. Eugene microflora and Miocene microfloras from British Columbia and the northwestern United States. The floras are located in Figure 28. Palynomorph t a x a , S t . Eugene F o r m a t i o n ( i ) (2) (3) (4) (5) (6) (7) (8) (9) ( 1 0 ) * ( 1 1 ) * ( 1 2 ) * ( 1 3 ) * (14) D i v i s i o n Lycopodophyta Lycopodiura X X X X X X Selaginella X D i v i s i o n P t e r o p h y t a Deltoidospora X X X Osmunda X X X X X X P o l y p o d i a c e a e - D e n n s t a e d t i a c e a e X X X X X X X X Laevigatosporites X X X X Triplanosporites X D i v i s i o n C o n i f e r o p h y t a AA AA AA ** AA ** *A AA AA Abies X X X X X X X X X X X X Cedrus X X X X X X X X X X Cupressaceae-Taxaceae X X X X X X X X X X X Ephedra X X X X x X X Glyptostrobus X X X X Metasequoia X X X X Picea X X X X X X X X X X X X X X Pinus X X X X X X X X X X X X X X Podocarpus X X X Pseudotsuga-Larix X X X X X X X X Sequoia X X Taxodi um X X X X Tsuga X X X X X X X X X x X X D i v i s i o n A n t h o p h y t a Acer X X X X X X X X X X Aesculus X X X Alnus X X X X X X X X X X X X X X Be tula X X X X X X X X X X X C a p r i f o l i a c e a e X X X Carpinus-Ostrya X X X X X X X X X X Carya X X X X X X X X X X X X X X Castanea X X X X X X X Chenopodiaceae-Amaranthaceae X X X X X x Compositae X X X X Corylus X X X X X X X x X X E r i c a c e a e X X X X X X X X X Fraxinus X X X X X X X X Gramineae X X X X X X Ilex X X x' X X X X X X X Juglans X X X X X X X X X X X X X X Liquidambar X X X X X X X X X X X X X Myrica X X X X X X Nyssa X X X X X X X Pachysandra-Sarcococca X X P i a t a n u s X X X X X X X X P t e r o c a r y a X X X X X X X X X X X X X X Quercus X X X X X X X X X X X X x X Salix X X X X X X X X X X X X Tilia X X X X X X X X X X Ulmus-Zelkova X X X X X X X X X X X x X Number o f genera o r f a m i l e s 2 5 1 5 11 10 6 5 6 9 8 12 9 5 abse n t from S t . Eugene Fm. (1) S o u t h - c e n t r a l B r i t i s h C o l u m b i a , M i o - P l i o c e n e ( P i e l , 1969). (2) S o u t h - c e n t r a l B r i t i s h C o l u m b i a , Miocene ( P i e l , 1969). (3) S o u t h - c e n t r a l B r i t i s h C o l u m b i a , Miocene o r e a r l y P l i o c e n e (Mathews and Rouse, 1963). (4) Queen C h a r l o t t e I s l a n d s , B r i t i s h C o l u m b i a , M i o - P l i o c e n e ( M a r t i n and Rouse, 1966). (5) Whatcom B a s i n , B r i t i s h C o l u m b i a , Miocene ( H o p k i n s , 1966). (6) Sucker C r e e k , Oregon, Miocene (Graham, 1965). (7) M a s c a l l , Oregon, Miocene (Chaney, 1959). (8) S t i n k i n g W a t e r s , Oregon, Miocene (Chaney, 1959). (9) B l u e M o u n t a i n s , Oregon, Miocene (Chaney, 1959). (10) Western Oregon, composite Miocene m i c r o f l o r a ( G r a y , 1964). (11) E a s t e r n Oregon, composite Miocene m i c r o f l o r a ( G r a y , 1964).-(12) Washington, c o m p o s i t e Miocene m i c r o f l o r a ( G r a y , 1964). (13) Idaho, composite Miocene m i c r o f l o r a ( G r a y , 1964). (14) K i l g o r e , N e b r a s k a , Miocene ( M a c G i n i t l e , 1962). * M i c r o f l o r a i n c l u d e s o n l y woody angiosperms and gymnosperms. **Genera o f T a x o d i a c e a e not i d e n t i f i e d . ***Gyntnosperms and an g i o s p e r m s . Figure 28. Location of some Miocene palynomorph assemblages in British Columbia and the northwestern United States. (1) South-central British Columbia, Mio-Pliocene (Piel, 1969); (2) south-central British Columbia, Miocene (Piel, 1969); (3) south-central British Columbia (Mathews and Rouse, 1963) ; (4) Queen Charlotte Islands (Martin and Rouse, 1966); (5) Whatcom Basin (Hopkins, 1966); (6) Sucker Creek (Graham, 1965); (7) Mascall (Chaney, 1959); (8) Stinking Waters (Chaney, 1959); (9) Blue Mountains (Chaney, 1959); (10) western Oregon (Gray, 1964); (11) eastern Oregon (Gray, 1964); (12) Washington (Gray, 1964); (13) Idaho (Gray, 1964); (14) Kilgore (MacGinitie, 1962). Figure 28 96 including the lava were not being eroded in this area during St. Eugene time and therefore were lower than the fan apices. Subsequently, these units were elevated relative to the fanglomerate at least 600 m along the bounding normal fault. Strata originally dipping west to southwest off the Galton Range scarp were mildly deformed; they now dip south to southeast. Again there is a similarity between these conditions and those occurring somewhat earlier in the Flathead Valley, where conglomerate and breccia were deposited in a structural basin on the downthrown side of the Flathead fault (Price, 1966). Displacement was simultaneous with deposition as the sequence of lithologies that occur as clasts is opposite that of the normal stratigraphic succession of bedrock litho-logies in the hanging wall of the fault. Cenozoic normal faults are widespread across the Cordillera. Pardee (1950) and Alden (1953) have described block-faulted basins in northwestern Montana which are underlain by Cenozoic sediments similar to the St. Eugene Formation. Both fan and flood-plain facies are present, and interbeds of ash are locally common. Much of the strata are faulted and tilted. Alden has concluded that during deposition there was considerable relief in northwestern Montana resulting from block faulting superimposed on epeirogenic uplift. Uplands and basins were in about the same positions as they are now, and there is no evidence of widespread peneplanation in the area following deposition of the basin sediments. Faulting has persisted to the present in the Rocky Mountains of the United States (Woollard, 1958; Milne and others, 1970), but the zone of seismicity dies out near the International Boundary. Figure 29. Relation between age of flora and percent of fossil genera s t i l l living near their fossil localities for mega- and microfloras from the western United States and British Columbia (from Wolfe and Barghoorn, 1960, their Fig. 1). About 44% of the fossil genera of the St. Eugene Formation are s t i l l extant in the general area of the unit. loor— 90 -2 80 -HI Z 7 0 -ui o 6 0 -ui > E 5 0 -5 tj 3 0 -Ul O. 2 0 -10 -oL-Microflora of St. Eugene Formcition 8> o 1 EOCENE OLIGOCENE A G E O F F L O R A MIOCENE PLIOCENE 98 In summary, deposition of St. Eugene and other basin sediments in the eastern Cordillera postdates the major compressional deformation of the Rocky Mountains during the Laramide Orogeny. This deformation, which culminated i n the development of the Columbian Foreland Thrust and Fold Belt, may have resulted in part from subduction along the Pacific margin of the Cordillera (Wheeler and others, 1972). Except off southern British Columbia and Washington, underthrusting ended by the Oligocene and was replaced by lateral shifting of plates along transform faults off the coast and by isostatic u p l i f t elsewhere. Uplift of the Cordillera was accompanied by extension and block faulting which produced the major basins and trenches of the Rocky Mountains in southern Canada and the United States. Sediment deposition was controlled by faulting along the margins of grabens and half-grabens. In places, u p l i f t , extension, and faulting were accompanied by the widespread and abundant extrusion of lavas and the intrusion of epizonal plutons (Wheeler and others, 1972). Faulting continued after deposition of the Miocene St. Eugene Formation in southeastern British Columbia and has continued to the present i n the Rockies south of 49°N. The Rocky Mountain Trench (south of 50°N) , then, consists of tectonic depressions formed by Cenozoic half-graben block faulting superposed on late Cretaceous to early Tertiary allocthonous fold and thrust structures. These depressions have been modified by sediment i n f i l l i n g , glaciation, and postglacial effects (Leech, 1966). The remarkable continuity of the Trench from 47°N to beyond 59°N and i t s trend across structures produced during the Laramide Orogeny suggest that i t delineates a structural boundary of continental proportions. This is FLOOD-PLAIN FACIES FAN FACIES Figure 30. Block diagram showing the proposed model of Tertiary sedimentation controlled by block faulting in the southern Rocky Mountain Trench, British Columbia (diagram modified from a lecture sketch by W.M. Davis in King, 1959, his Fig. 86). 100 supported by abundant geophysical evidence which shows the Trench to be a major crustal boundary. Between 49° and 56°N the Trench is the western limit of broad, high-amplitude aeromagnetic anomalies found over the Rockies and the Plains (Haines and others, 1971). Geomagnetic depth-sounding results indicate that the Trench marks a transitional zone between a conductive, hydrated lower crust to the west and a resistive lower crust to the east (Caner, 1970). Gravity data show the craton ending at the Trench north of 50°N, although to the south the craton extends west of the Trench farther into the Cordillera. Finally, seismic results show that the lithosphere may thin toward the Cordillera (Wickens and Pec, 1968). Berry and others (1971) have suggested from this geophys-ical evidence that the Rocky Mountain Trench between 50° and 56°N is the margin of the Precambrian craton. They thus agree with Price and Mountjoy (1970) who consider the Trench to mark a Paleozoic hinge zone along the craton margin. CONCLUSIONS (1) The St. Eugene Formation consists of flood-plain and fan facies deposited in a depression similar in morphology and position to the present southern Rocky Mountain Trench. The flood-plain facies includes both high-energy river gravels deposited off tributary valleys and shallow-lake or slack-water s i l t and sand. The fan facies consists of talus and fanglomerate deposited along the margins of the proto-Rocky Mountain Trench and derived from adjacent fault-bounded uplands. (2) Up to 1500 m of sediment are present in a series of 101 structural basins underlying the Trench and bordered by high-angle faults transverse to the Trench trend. The Miocene St. Eugene Formation includes only the uppermost strata of these basin f i l l s , and i t is probable that deeper sediments are in part Oligocene and perhaps Eocene in age. Sedi-ments of Upper Eocene and Oligocene age occur in the structurally com-parable Flathead Valley to the east and provide an upper age limit for the Laramide Orogeny in the southern Canadian Rockies. Compressional deformation which produced the Rocky Mountain fold and thrust belt during the late Cretaceous and early Tertiary was followed by a period of isostatic uplift and extension accompanied by block faulting in the Cordillera. This change in tectonics may be due to the cessation of subduction along the northern Pacific margin and its replacement by lateral displacement of plates along transform faults. The southern Rocky Mountain Trench formed by block faulting of a half-graben style. Deposition of sediment in the Trench and erosion of the uplands were probably contemporaneous with faulting, although 600 m or more of dis-placement along the east boundary fault occurred after deposition of the St. Eugene Formation. (3) Palynological results substantiate conclusions based on the texture, structure, and lithologic composition of the St. Eugene Formation. The microflora includes elements which grew on poorly drained lowlands and montane uplands. The climate was humid with abundant summer precipi-tation and mild, moist winters; i t was thus more temperate than the present climate of the area. Factors which may have controlled the Miocene climate include a relatively warm polar sea and lower mountain barriers between southeastern British Columbia and the Pacific Ocean. 102 LITERATURE CITED Alden, W.C, 1953, Physiography and glacial geology of western Montana and adjacent areas: U.S. Geol. Survey Prof, Paper 231, 200 p. Anderson, T.W., and Stephens, M.A., 1971, Tests for randomness of directions against equatorial and bimodal alternatives: Dept. Statistics, Stanford Univ., Tech. Rept. 5, 19 p. Andrews, J.T., and Smith, D.I., 1970, Statistical analysis of t i l l fabric: methodology, local and regional variability (with parti-cular reference to the north Yorkshire t i l l c l i f f s ) : Geol. Soc. London Quart. Jour., v. 125, p. 503-542. Axelrod, D.I., 1941, The concept of ecospecies in Tertiary paleobotany: Natl. Acad. Sci. Proc, v. 27, p. 545-551. 1964, The Miocene Trapper Creek flora of southern Idaho: California Univ. Pubs. Geol. Sci., v. 51, 181 p. Barnes, W.C, 1963, Geology of the northeast Whitefish Range, northwest Montana [Ph.D. thesis]: Princeton, New Jersey, Princeton Univ., 163 p. Bennema, J., 1963, The red and yellow soils of the tropical and subtropi-cal uplands: Soil Sci., v. 95, p. 250-257. Berry, E.W., 1929, The age of the St. Eugene s i l t in the Kootenay Valley, British Columbia: Royal Soc. Canada Trans., ser. 3, sec. 4, v. 23, p. 47-48. Berry, M.J., Jacoby, W.R., Niblett, E.R., and Stacey, R.A., 1971, A review of geophysical studies in the Canadian Cordillera: Canadian Jour. Earth Sci., v. 8, p. 788-801. Boulton, G.S., 1971, T i l l genesis and fabric in Svalbard, Spitsbergen, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 41-72. Brown, CA. , 1960, Palynological. techniques: Baton Rouge, Louisiana, CA. Brown, 188 p. Brown, R.W., 1934, The recognizable species of the Green River flora: U.S. Geol. Survey Prof. Paper 185-C, p. 45-77. Caner, B., 1970, Electrical conductivity structure in western Canada and petrological interpretation: Jour. Geomagnetism and Geoelectri-city, v. 22, p. 113-129. 103 Chaney, R.W., 1936, The succession and distribution of Cenozoic floras around the northern Pacific basin, in Goodspeed, T.H., ed., Essays in geobotany in honor of William Albert Setchell: Berkeley, California Univ. Press, p. 55-85. 1938, Paleoecological interpretations of Cenozoic plants in western North America: Bot. Rev., v. 4, p. 371-396. 1948, The bearing of the living Metasequoia on problems of Tertiary paleobotany: Natl. Acad. Sc i . P r o c , v. 34, p. 503-515. 1959, Miocene floras of the Columbia Plateau, Part I, Composi-tion and interpretation: Carnegie Inst. Washington Pub. 617, p. 1-134. Cook, H.J., 1960, New concepts of late Tertiary major crustal deforma-tions in the Rocky Mountain region of North America: Proc. 21st Internat. Geol. Cong., Copenhagen, pt. 12, p. 198-212. Crickmay, C.H., 1964, The Rocky Mountain Trench: a problem: Canadian Jour. Earth Sci., v. 1, p. 184-205. Dorf, Erling, 1959, Climatic changes of the past and present: Michigan Univ. Mus. Paleontology Contr., v. 13, p. 181-210. 1963, The use of f o s s i l plants in palaeoclimatic interpretations, in Nairn, A.E.M., ed., Problems in palaeoclimatology: London, Interscience Publishers, p. 11-31. Douglas, R.J.W., Gabrielse, Hubert, Wheeler, J.O., Stott, D.F., and Belyea, H.R., 1968, Geology of Western Canada, in Douglas, R.J.W., ed., Geology and economic minerals of Canada: Canada Geol. Survey Econ. Geol. Rept. 1, p. 365-488. Eardley, A.J., 1951, Structural geology of North America (2nd ed.): New York, Harper and Row, Inc., 743 p. Eisbacher, G.H., 1972, Tectonic framework of Sustut and Sifton Basins, British Columbia: Canada Geol. Survey Paper 72-1, pt. A, p. 24-26. Fullard, Harold, and Darby, H.C., eds., 1967, The university atlas (12th ed.): London, George Philip and Son, Ltd., 176 p. Garland, G.D., Kanasewich, E.R., and Thompson, T.L., 1961, Gravity measurements over the southern Rocky Mountain Trench area of British Columbia: Jour. Geophys. Research, v. 66, p. 2495-2505. Goddard, E.N., Trask, P.D., DeFord, R.K., Rove, O.N., Singewald, J.T., J r . , and Overbeck, R.M., 1963, Rock-color chart: Boulder, Colorado, Geol. Soc. America. 104 Graham, Alan, 1965, The Sucker Creek and Trout Creek Miocene floras of southeastern Oregon: Kent State Univ. Bu l l . , v. 53, no. 12, 147 p. Gray, Jane, 1964, Northwest American Tertiary palynology: the emerging picture, in Cranwell, L . M . , ed., Ancient Pacific floras, the pollen story: Proc. 10th Pacific Sci. Cong., Honolulu, Univ. Hawaii Press, p. 21-30. Haines, G.V., Hannaford, W., and Riddihough, R.P., 1971, Magnetic anomalies over British Columbia and the adjacent Pacific Ocean: Canadian Jour. Earth Sci., v. 8, p. 387-391. H i l l s , L.V., 1965, Palynology and age of early Tertiary basins, interior British Columbia [Ph.D. thesis]: Edmonton, Alberta Univ., 188 p. Hollick, Arthur, 1914, A preliminary report by Mr. Arthur Hollick of the New York Botanical Garden, upon plants from the Pleistocene deposits: Canada Geol. Survey Summ. Rept., 1913, p. 133-135. 1927, The flora of the Saint Eugene s i l t s , Kootenay Valley, British Columbia: New York Bot. Garden Mem., v. 7, p. 389-465. Holmes, CD., 1941, T i l l fabric: Geol. Soc. America B u l l . , v. 52, p. 1299-1354. Hopkins, W.S., J r . , 1966, Palynology of Tertiary rocks of the Whatcom basin, southwestern British Columbia and northwestern Washington [Ph.D. thesis]: Vancouver, British Columbia Univ., 184 p. 1968, Subsurface Miocene rocks, British Columbia—Washington, a palynological investigation: Geol. Soc. America B u l l . , v. 79, p. 763-768. Hopkins, W.S., J r . , Rutter, N.W., and Rouse, G.E., 1972, Geology, paleoecology, and palynology of some Oligocene rocks in the Rocky Mountain Trench of British Columbia: Canadian Jour. Earth Sc i . , v. 9, p. 460-470. Johns, W.M., 1970, Geology and mineral deposits of Lincoln and Flathead Counties, Montana: Montana Bur. Mines and Geology Bul l . 79, 182 p. Johns, W.M. , Smith, A.G., Barnes, W.C, Gilmour, E.H., and Page, W.D., 1963, Geologic investigations i n the Kootenai-Flathead area, northwest Montana: No. 5, western Flathead County and part of Lincoln County: Montana Bur. Mines and Geology B u l l , 36, 68 p. Kelley, C C , and Sprout, P.N., 1956, Soil survey of the upper Kootenay and Elk River valleys in the East Kootenay d i s t r i c t of British Columbia: British Columbia Soil Survey Rept. 5, 99 p. 105 King, P.B., 1959, The evolution of North America: Princeton, New Jersey, Princeton Univ. Press, 189 p. Krajina, V.J., 1965, Biogeoclimatic zones and classification of British Columbia, in Krajina, V.J., ed., Ecology of western North America: Dept. Botany, British Columbia Univ., v. 1, p. 1-17. Kummel, Bernhard, and Raup, David, eds., 1965, Handbook of paleontological techniques: San Francisco, W.H. Freeman and Co., Inc., 852 p. Lamb, A.T., and Smith, D.W., 1962, Refraction profiles over the southern Rocky Mountain Trench area of British Columbia: Alberta Soc. Petroleum Geologists Jour., v. 10, p. 428-437. LaMotte, R.S., 1952, Catalogue of the Cenozoic plants of North America through 1950: Geol. Soc. America Mem. 51, 381 p. Leech, G.B., 1958, Fernie map-area, west half, British Columbia: Canada Geol. Survey Paper 58-10, 40 p. 1960, Fernie, west half, British Columbia: Canada Geol. Survey Map 11-1960. 1966, The Rocky Mountain Trench: Canada Geol. Survey Paper 66-14, p. 307-329. Leopold, E.B., 1969, Late Cenozoic palynology, in Tschudy, R.H., and Scott, R.A., eds., Aspects of palynology, an introduction to plant microfossils in time: New York, John Wiley and Sons, Inc., p. 377-438. Lineback, J.A., 1971, Pebble orientation and ice movement in south-central Illinois, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 328-334. Lindsay, J.F., 1968, The development of clast fabric in mudflows: Jour. Sed. Petrology, v. 38, p. 1242-1253. MacGinitie, H.D., 1962, The Kilgore flora, a late Miocene flora from northern Nebraska: California Univ. Pubs. Geol. Sci., v. 35, p. 67-158. Mark, D.M., 1973, Analysis of axial orientation data, including t i l l fabrics: Geol. Soc. America Bull., v. 84, p. 1369-1374. Martin, H.A., and Rouse, G.E., 1966, Palynology of late Tertiary sedi-ments from Queen Charlotte Islands, British Columbia: Canadian Jour. Botany, v. 44, p. 171-208. Mathews, W.H., and Rouse G.E., 1963, Late Tertiary volcanic rocks and plant-bearing deposits in British Columbia: Geol. Soc. America Bull., v. 74, p. 55-60. 106 Milne, W.G. , Smith, W.E.T., and Rogers, G.C, 1970, Canadian seismicity and microearthquake research in Canada: Canadian Jour. Earth Sci., v. 7, p. 591-601. North, F.K. , and Henders on, G.G.L., 1954, The Rocky Mountain Trench, in Scott, J.C., and Fox, F.G., eds., Guidebook, 4th annual field conference: Alberta Soc. Petroleum Geologists, p. 82-100. Pardee, J.T., 1950, Late Cenozoic block faulting in western Montana: Geol. Soc. America Bull., v. 61, p. 359-406. Penny, J.S., 1969, Late Cretaceous and early Tertiary palynology, in Tschudy, R.H., and Scott, R.A., eds., Aspects of palynology, an introduction to plant microfossils in time: New York, John Wiley and Sons, Inc., p. 331-376. Piel, K.M., 1969, Palynology of middle and late Tertiary sediments from the central interior of British Columbia, Canada [Ph.D. thesis]: Vancouver, British Columbia Univ., 157 p. 1971, Palynology of Oligocene sediments from central British Columbia: Canadian Jour. Botany, v. 49, p. 1885-1920. Price, R.A., 1962, Fernie map-area, east half, Alberta and British Columbia: Canada Geol. Survey Paper 61-24, 65 p. 1966, Flathead map-area, British Columbia and Alberta: Canada Geol. Survey Mem. 336, 221 p. Price, R.A., and Mountjoy, E.W., 1970, Geologic structure of the Canadian Rocky Mountains between Bow and Athabasca Rivers—a progress report, in wheeler, J.O., ed., Structure of the Canadian Cordillera: Geol. Assoc. Canada Spec. Paper 6, p. 7-25. Price, R.A., and Wise, D.U., 1959, Flathead, British Columbia and Alberta: Canada Geol. Survey Map 1-1959. Ragan, D.M., 1968, Structural geology, an introduction to geometrical techniques: New York, John Wiley and Sons, Inc., 166 p. Rice, H.M.A., 1937, Cranbrook map-area, British Columbia: Canada Geol. Survey Mem. 207, 67 p. Rouse, G.E., 1962, Plant microfossils from the Burrard Formation of western British Columbia: Micropaleontology, v. 8, p. 187-218. Rouse, G.E., Hopkins, W.S., Jr., and Piel, K.M., 1970, Palynology of some Late Cretaceous and Early Tertiary deposits in British Columbia and adjacent Alberta, in Kosanke, R.M., and Cross, A.T., eds., Symposium on palynology of the Late Cretaceous and Early Tertiary: Geol. Soc. America Spec. Paper 127, p. 213-246. 107 Russell, L.S., 1951, Age of the Front-Range deformation in the North American Cordillera: Royal Soc. Canada Trans., ser. 3, sec. 4, v. 45, p. 47-69. 1954a, Mammalian fauna of the Kishenehn Formation, southeastern British Columbia: Natl. Mus. Canada Bull. 132, p. 92-111. 1954b, The Eocene-Oligocene transition as a time of major orogeny in western North America: Royal Soc. Canada Trans., ser. 3, sec. 4, v. 48, p. 65-69. Ryder, J.M., 1971, The stratigraphy and morphology of para-glacial alluvial fans in south-central British Columbia: Canadian Jour. Earth Sci., v. 8, p. 279-298. Schofield, S.J., 1915, Geology of the Cranbrook map-area, British Columbia: Canada Geol. Survey Mem. 76, 245 p. Stanley, E.A., 1965, Upper Cretaceous and Paleocene plant microfossils and Paleocene dinoflagellates and hystrichosphaerids from northwestern South Dakota: Bulls. Am. Paleontology, v. 49, p. 175-384. Thompson, T.L., 1962, Origin of the Rocky Mountain Trench in southeastern British Columbia by Cenozoic block faulting: Alberta Soc. Petroleum Geologists Jour., v. 10, p. 428-437. Traverse, Alfred, 1955, Pollen analysis of the Brandon lignite of Vermont: U.S. Bur. Mines Rept. Inv. 5151, 107 p. West, R.G., and Donner, J.J., 1956, The glaciations of East Anglia and the East Midlands: a differentiation based on stone-orientation measurements of the t i l l s : Geol. Soc. London Quart. Jour., v. 112, p. 69-91. Wheeler, J.O., Aitken, J.D., Berry, M.J., Gabrielse, Hubert, Hutchison, W.W., Jacoby, W.R., Monger, J.W.H., Niblett, E.R., Norris, D.K., Price, R.A., and Stacey, R.A., 1972, The Cordilleran structural province, in Price, R.A., and Douglas, R.J.W., eds., Variations in tectonic styles in Canada: Geol. Assoc. Canada Spec. Paper 11, p. 1-81. Wickens, A.J., and Pec, K., 1968, A crust-mantle profile from Mould Bay, Canada to Tucson, Arizona: Seismol. Soc. America Bull., v. 58, p. 1821-1831. Wolfe, J.A., 1971, Tertiary climatic fluctuations and methods of analysis of Tertiary floras: Palaeogeography, Palaeoclimatology, Palaeo-ecology, v. 9, p. 27-57. Wolfe, J.A., and Barghoorn, E.S., 1960, Generic change in Tertiary floras in relation to age: Am. Jour. Sci., v. 258-A, p. 388-399. 108 Wolfe, J.A., and Hopkins, D.M., 1967, Climatic changes recorded by Tertiary land floras in northwestern North America, in Hatai, Kotora, ed., Tertiary correlations and climatic changes in the Pacific: Proc. 11th Pacific Sci. Cong., Tokyo, v. 25, p. 67-76. Woollard, G.P., 1958, Areas of tectonic activity in the United States as indicated by earthquake epicenters: Am. Geophys. Union Trans., v. 39, p. 1135-1150. Wright, H.E., Jr., 1957, Stone orientation in Wadena drumlin field, Minnesota: Geog. Annaler, v. 39, p. 19-31. 109 CHAPTER THREE: GLACIER-FLOW PATTERNS AND THE ORIGIN OF LATE WISCONSINAN TILL ABSTRACT Late Wisconsinan glacier-flow patterns and t i l l genesis in the southern Rocky Mountain Trench, British Columbia were determined through a study of glacial landforms, t i l l fabric, and t i l l composition. At least one major shift in the pattern of glacier flow occurred near the end of glaciation. An earlier stage when major tribu-tary glaciers were coalescent with the trunk glacier in the Rocky Mountain Trench was followed by recession of tributary glaciers and invasion of side valleys by the trunk glacier. Final recession of the trunk glacier occurred with no major halts and without major stagnation of the terminus. A review of the ways in which strongly directed t i l l fabrics originate indicates that t i l l associated with drumlins accumulated by lodgement of particles due to subglacial pressure melting against a planar substratum and not by the melt-out process (slow melting in situ at the base or surface of stagnant ice masses) nor by post-depositional reorientation caused by glacier overriding. Some fabrics from the lower part of the younger t i l l sheet suggest either complex patterns of ice flow similar to those at junctions of many modern valley glaciers, or mass movement of supraglacial t i l l (flowtill) associated with an interstadial interval within the late Wisconsinan glaciation. 110 Compositional analyses are less sensitive as indicators of glacier-flow patterns than till-fabric results in the southern Rocky Mountain Trench. This is apparently the result of two processes: (1) t i l l may have been reworked at the ice-sediment interface; and (2) sediment may have been transported to a subglacial depositional site by a mechanism other than the flowing ice itself, specifically by melt-water via an englacial "plumbing" system, followed by lodgement through pressure melting. Nevertheless, t i l l composition reflects the dominant pattern of ice flow during glaciation. Most pebbles and fine sand in the t i l l s of the Rocky Mountain Trench are of local origin. Local constituents decrease in relative amounts away from the mouths of tributary valleys from whence they were derived. These decreasing gradients cannot be due entirely to breakage and abrasion during transport, but rather are likely explainable by progressive deposition and dilution through sediment mixing. Compositional differences in t i l l are pronounced off tributary valleys, but become less distinct to the south. This probably results from lateral mixing of sediment due to shifts in the zone of coalescence of trunk and tributary glaciers with changes in relative ice flux from the two. INTRODUCTION The existence of a former glacier is indicated by t i l l , a poorly sorted sediment deposited directly from ice and characterized by a variety of mineral and rock types and, commonly, subangular clasts, I l l some striated sand grains, pebbles, and coarser clasts, and a pronounced directional fabric. Lenses of sorted, water-worked sediment are common in t i l l , and the t i l l unit itself commonly overlies or contains striated pavements. T i l l is a widespread sediment in areas glaciated during the Quaternary, but, as indicated by Goldthwait (1971), is most common and thickest in the outer part of each glaciated area, where multiple t i l l layers record successive glacial advances. T i l l studies to date have been primarily concerned with the characterization and correlation of strati-graphic sequences. Common methods used in such studies include strati-graphic position of t i l l , associated landforms, textural and lithologic analyses, color, weathering, and sometimes fabric analysis. Recently, increased emphasis has been placed on using the above methods to gain insight into the mode of origin of t i l l (review in Goldthwait, 1971). Such investigations are complemented by those of glaciologists and others concerned with how sediment is entrained, transported, and deposited by ice (Goldthwait, 1951; McCall, 1952; Weertman, 1961; Kamb and LaChapelle, 1964; Boulton, 1970a, 1970b, 1971; Nobles and Weertman, 1971). T i l l of late Wisconsinan age in the Rocky Mountain Trench of southeastern British Columbia was studied to determine the pattern of glacier flow in the Trench and its major tributaries; to assess the appli-cability of certain methods of t i l l investigation in the Cordillera, specifically fabric and lithologic analyses; and to determine the mode of origin of the t i l l . 112 DISTRIBUTION AND CHARACTER OF TILL Unconsolidated and partially consolidated deposits mantle the bedrock floor of the southern Rocky Mountain Trench and consist of up to 1400 m of Tertiary sediments overlain by Quaternary sediments locally thicker than 200 m. Deposits of three glacier advances are recognized. These advances are tentatively correlated with stades of the Pinedale Glaciation of the Rocky Mountains in the United States (Richmond, 1965), and with the Fraser Glaciation in the coastal Pacific Northwest (Armstrong and others, 1965; Crandell, 1965). Interglacial sediments which contain wood dated at 26,800 ± 1000 years B.P. (GX-2032) underlie drift of the early stade. Glacier recession between the middle and late advances was of short duration and extent; glaciolacustrine sediments were deposited only along the margins of the Rocky Mountain Trench, and apparently residual ice remained in the center of the Trench. In contrast, during the interstade between the early and middle glacier advances, the Rocky Mountain Trench within the study area was completely deglaciated, and sediments were deposited in one or more lakes on the floor of the Trench. Deposits of the early stade are informally called older drift (also lower drift); those of the middle and late stades are referred to as younger drift (also upper drift, Wycliffe t i l l ) . T i l l of latest Wisconsinan age is the dominant surface deposit on the floor of the Rocky Mountain Trench (Fig. 31). Called the Wycliffe t i l l by Schofield (1915), i t is a massive, unsorted deposit of clay, s i l t , sand, and clasts ranging widely in size (Fig. 32, Appendix 3), and is 113 light gray to very light gray (N 7 to N 8) in color. The clasts are subangular to subrounded, commonly striated, and consist of a variety of rock types present i n the Purcell and Rocky Mountains flanking the map area. The t i l l i s calcareous except where derived entirely from the Purcell Mountains. In general, patches of noncalcareous t i l l in the Trench are limited to the lower slopes of the Purcells. Kelley and Sprout (1956, p. 27) have reported noncalcareous t i l l off St. Mary Valley as far east as Wycliffe. Lenses of water-worked sand and gravel are present within the t i l l sheet. Horizontally st r a t i f i e d lake s i l t s from 3 to 15 m thick occur within the Wycliffe t i l l along the margins of the Trench and may record an interval of glacier recession between the middle and late stades of the Pinedale Glaciation. Where exposed in the walls of major postglacial valleys, Wycliffe t i l l ranges i n thickness from a few meters to more than 40 m, but is most commonly 10 to 30 m thick. The surface of the t i l l sheet slopes gradually to the south down the Trench (approximate slope = 0.7 m/km). The base also has a down-Trench slope, although there is substantial local r e l i e f on the interface between the t i l l and underlying sediments. Immediately prior to deglaciation, Wycliffe t i l l covered the entire Trench floor. This constructional surface was modified by l a t e r a l , proglacial, and subglacial meltwater as the glaciers receded. The t i l l , thus, was reworked locally into outwash; i t was removed completely along the major river valleys during the Holocene Epoch. 114 GLACIER-FLOW PATTERNS The Glacial Map of Canada (Prest and others, 1967) records observations of glacial striae, drumlins, other glacial lineations, and meltwater channels. In many cases, these indicate the centers of ice accumulation and regional patterns of ice movement in British Columbia during the Wisconsinan Glaciation. Major centers in both the Columbia and Rocky Mountains fed the glacier that was flowing southeast through the Rocky Mountain Trench into Montana. The following is my interpre-tation of the events that were occurring during this time. At the climax of the last major Pleistocene glaciation, the mountains bordering the southern Rocky Mountain Trench were almost completely buried by ice. Erratics and striae along the Rocky Mountain front just north of Bull River indicate the glacier reached an elevation of at least 2260 m there. On the west side of the Trench, Mount Baker (elev. 2210 m) has a rounded, streamlined crest resulting from glacier overriding. Daly (1912) concluded that at the peak of glaciation the average level of ice in the Galton Range and Purcell Mountains near 49°N latitude was 2230 m. The Cordilleran Ice Sheet was thus about 1500 m thick over the Trench at the International Boundary. In the Rocky and Purcell Mountains only the higher peaks projected above the glacier surface. Ice in the ranges bordering the Trench on the east discharged . . .in a general southward direction and was largely concen-trated in the Wigwam Valley and in the Rocky Mountain Trench, though some of the ice flowed eastward into the North Flathead glacier before beginning its southward journey. (Daly, 1912, p. 585) 115 Landforms on the Trench floor record directions of ice movement during the closing phase of the last glaciation and also provide informa-tion on the manner in which deglaciation occurred. Drumlins and modified drumlins are the dominant landforms developed on Wycliffe t i l l (Fig. 33). In general, these drumlins trend south and southeast parallel to the margins of the Rocky Mountain Trench; this indicates, as one might expect, that topography exerted primary control on the direction of glacier flow. The dominance of drumlinized topography and the rarity of ice-stagnation features such as eskers, hummocky ground moraine, kames, kame moraines, and kettles suggest that deglaciation occurred by uniform recession of an active terminal zone without large-scale stagnation of any part of the glacier. Most kettles occur in the recessional outwash underlying outwash plains, kame terraces, and meltwater channels, rather than in Wycliffe t i l l . Apparently, there were no major standstills or readvances during deglaciation, as recessional and end moraines are absent on the Trench floor from at least the International Boundary to the north end of the study area, a distance of about 150 km. One possible exception is a hummocky ridge just north of Bull River that may represent part of a moraine deposited by the Bull Valley glacier. Although drumlins provide a reasonably complete picture of the pattern of last ice flow, other directional indicators must be used off major tributaries. This is because Wycliffe t i l l at the mouths of these tributary valleys was eroded by meltwater entering the Trench, and subsequently was removed during early Holocene valley cutting. Fortunately, near these tributary mouths there are bedrock surfaces which were subject to glacial erosion throughout glaciation. On these surfaces, 116 the last major direction of glacier flow is indicated by striations, crescentic gouges, and crescentic fractures (Figs. 31 and 33). This additional information shows that the glacier flowing down the Trench fanned out to occupy the lowland in the Kimberley-Cranbrook-Fort Steele area and bulged up into Elk Valley. This implies that tributary glaciers in St. Mary and Elk Valleys had receded prior to deglaciation of the Trench floor, a conclusion corroborated by stratigraphic evidence for the existence of late glacial lakes in these valleys. These glacial deposits and landforms are time transgressive, being younger toward the north. For example, the late glacial lake impounded by Trench ice in St. Mary Valley is at least in part younger than a comparable lake in Elk Valley. The extent to which these features transgress time is not known, as there is at present only one radiocarbon date providing a minimum limiting age for deglaciation of the southern Rocky Mountain Trench at one point (10,000 ± 140 years B.P.; l o c a t i o n — 51°28'55"N, 117°13'25"W; GSC-1457; Fulton, 1971). On the basis of other limiting radiocarbon dates from southern British Columbia, i t has been speculated that the ice margin in the Trench was at 49°N latitude about 12,000 to 13,000 years ago (Prest, 1969). Analytical Procedure Further information on the pattern and mechanics of glacier flow has been obtained from an investigation of the fabric, heavy mineral composition, and clast lithologies of the late Wisconsinan t i l l . The pattern of glacier flow as determined independently from drumlins, 117 striations, and other flow indicators (p, 115) provides an internal check by which the applicability of these methods in glaciated areas of the Cordillera may be evaluated. Till-fabric Analysis The first major quantitative work on the fabric of t i l l was done by Richter (1932). Since that time t i l l fabrics have been exten-sively employed as indicators of the flow directions of former glaciers. It was found that elongate clasts preferentially trend parallel to flow and plunge up-glacier, although fabrics with transverse maxima were also recognized. Holmes (1941) showed that clast shape in part determines whether the fabric maximum is parallel or transverse to the flow direction. However, the mechanisms by which clasts in t i l l become oriented remained poorly understood, even though numerous hypotheses were advanced on empirical and theoretical grounds (Richter, 1936; Holmes, 1941; Sitler and Chapman, 1955; Glen and others, 1957; Harrison, 1957; Wright, 1957). Several authors have commented on the internal variability of t i l l fabric. West and Donner (1956) and Young (1969) concluded that horizontal variation is small in comparison to vertical variation, thus making i t possible to differentiate major ice advances. Harrison (1957) and Kauranne (1960), on the other hand, found considerable variation in short lateral distances, although horizontal variation is less than vertical variation between levels. In perhaps the most detailed study of within-and between-site variability, Andrews and Smith (1970) concluded that within-site precision estimates are low and similar to estimates on cross-beds, whereas between-site precision estimates over short lateral 118 and vertical distances vary from low to high. The treatment and assess-ment of till-fabric data improved with the application of three-dimensional analytical techniques and statistical tests of significance (Andrews and Smith, 1970; Mark, 1973). At present, however, there is a general lack of uniformity in till-fabric investigation and analysis procedures (Andrews, 1971), at least in part due to lack of consensus on the origin of the t i l l fabric. There is an absence of standardization in data collection (geometry and size of clasts; limiting ratios of a, b, and c axes;1 axes or planes measured; aspect of sampling site; sample size), sampling control (measure of within-site variability of fabrics), data portrayal (various rose diagrams, contour diagrams, and stereograms), and quantitative analysis of data (two- and three-dimensional methods with a variety of associated statistics including mean direction, dispersion about the mean, and tests of randomness). Despite the above difficulties, a better understanding of how sediment is entrained in a glacier, transported, and deposited is being provided by glaciologists. With regard to t i l l fabrics, the work of Boulton (1968, 1970a, 1970b, 1971) is particularly important. Boulton has studied the transport and fabric characteristics of sediment within Spitsbergen glaciers and the fabric of t i l l freshly deposited by the glaciers. "He has recognized that t i l l fabric may be due to one or a combination of the following: (1) orientation of sediment particles during transport, (2) acquisition of preferred orientation during 1The long, intermediate, and short axes of t i l l clasts are called the a, b, and c axes respectively. 119 deposition, and (3) post-depositional reorientation resulting from mass movement or glacier overriding. Each of these is considered below and the evidence provided by Boulton (1971) and others briefly reviewed. Englacial transport. It has been shown both on theoretical and empirical grounds that randomly oriented particles immersed in a shearing, viscous or quasi-viscous fluid develop a preferred orientation. Jeffrey (1922) concluded that ellipsoidal particles rapidly develop a long-axis distribution with a peak parallel to the flow direction. He pointed out, however, that energy dissipated by an ellipsoid is minimal if the long axis is in the transverse position. Thus, there may be a reorientation from the parallel to the transverse position over a long period of time, although the exact way in which this occurs is not understood. These conclusions are substantiated by observations of the englacial fabric by Boulton (1971). He found that the a axes of rod-shaped clasts and the a-b planes of blade-shaped clasts l i e in the foliation plane. The orientation of the a axes within this plane is dependent upon the local stress conditions in the glacier; in the terminal zone of compressive flow, a-b planes and ice foliation dip steeply up-glacier, and a axes are transverse to flow; in zones of extending flow, on the other hand, a-b planes are sub-horizontal, and a axes of blades and rods are parallel to flow. In zones where there is neither marked compression nor extension, preferred orientation of a axes of blades and rods is weak or absent. Where shear planes dipping up-glacier intersect the ice foliation, blades are reoriented so that their a-b planes are parallel to the shear surface. The a axes of both rods and blades are parallel to the direction of movement along these surfaces, 120 and therefore parallel to the direction of ice flow, Deposition. An englacial fabric may be inherited or modified depending on the manner of t i l l deposition. Boulton has suggested that there are three major processes of t i l l deposition and thus three groups of t i l l : lodgement t i l l , melt-out t i l l , and flowtill. Lodgement t i l l is deposited either from active ice by melting under pressure at the till-glacier interface, or from a zone of debris-laden stagnant ice separating the active glacier from the sediment substratum. Melt-out t i l l is deposited subglacially or supraglacially by the in~situ melting-out from stagnant ice of englacial debris. Flowtill originates when supraglacial melt-out t i l l undergoes mass movement, Boulton (1971) observed the deposition of lodgement t i l l by pressure melting of active ice against obstructions. The orientation of plastered-on elongate particles is dependent upon the geometry of the substratum. Where this is non-transverse, parallel-to-flow a-axis orientations result, but dip directions of a-b planes and plunge directions of a axes are limited by the slope direction of the surface. Boulton (1970a) has described a subglacial t i l l deposited by the melting of debris-rich stagnant ice overridden by active ice along a dScollement surface. The t i l l has a fabric inherited from englacial transport, but altered along shear planes produced after deposition by glacier overriding. Melt-out t i l l forms in the terminal zone of a glacier and is likely to eventually undergo mass movement and thus become flowtill. The fabric of melt-out t i l l reflects that of the englacial debris which is 121 concentrated by the melting of interstitial ice. The amount of alteration of the englacial fabric depends on the percentage of ice eliminated during melting; where ice content is high, fabric modification is substantial. This modification mainly involves a decrease in the angle of dip of blades which in the terminal zone are aligned in the englacial state parallel to the up-glacier-dipping foliation. When supraglacial t i l l moves under the influence of gravity, any pre-existent fabric related to ice flow will be replaced by one related to mass movement forces. Fabric elements both parallel and transverse to the resulting sediment transport direction develop, but, since this direction is rarely related to the direction of ice movement, the resulting fabric yields l i t t l e information on the glacier-flow pattern. Post-depositional reorientation. Fabric reorientation can occur along closely spaced shear zones or by viscous flow in water-saturated t i l l as i t is overridden by active ice. MacClintock and Dreimanis (1964) and Ramsden and Westgate (1971) have described fabrics modified by glacier overriding. The a axes of elongate particles are reoriented parallel to the direction of glacier movement. Fabric reorientation can also result from postglacial mass movement (e.g., landslides, creep). Table 4 summarizes the relationships between till-fabric elements and processes of glacier flow and t i l l deposition. Considering the intricacy of interrelationships between these factors, i t is not surpris-122 ing that within-site variability is large and that fabric interpretation is fraught with d i f f i c u l t i e s . Nevertheless, certain conclusions can be made. If the direction of ice movement responsible for a particular t i l l is independently known, i t may be possible to infer the mode of origin of the t i l l . T i l l characterized by blades with a-b planes dipping steeply up-glacier and rods with a axes transverse to flow indicates preservation by melt-out of an englacial compressive fabric produced in the terminal zone of a glacier. If a-b planes of blades are similarly oriented, but the a axes of rods are parallel to flow and dip steeply up-glacier, an englacial fabric produced along shear planes in the terminal zone and inherited during melt-out is indicated. A variety of interpretations may be placed on strongly directed fabrics with gently dipping elongate particles. For example, they may arise by the weakening of a steeply dipping englacial fabric during melt-out, by subglacial lodgement through pressure melting, by preserva-tion of an englacial fabric i n zones of extending flow, or by post-depositional reorientation along sub-horizontal shear planes under an active glacier. In most of these cases, maxima are parallel rather than transverse to flow. Intuitively, one might expect that most t i l l would originate in one or more of these ways; thus, i t is not surprising that most fabric diagrams for t i l l from areas of ground moraine are charac-terized by a gently dipping girdle with either a single maximum on the girdle or, less commonly, with two maxima 90° apart (Holmes, 1941, p. 1313; Harrison, 1957, p. 283; Kauranne, 1960, p. 87; Ramsden and Westgate, 1971, p. 339). The latter situation might originate by the melt-out of a composite englacial fabric in which rods in the terminal area of the 123 glacier are parallel to flow in shear zones and transverse to flow along foliation planes. Strong fabrics which are neither parallel nor transverse to the direction of ice flow may indicate deposition as flowtill or alteration of a pre-existent fabric by post-depositional and postglacial mass movement. Method. The fabric of late Wisconsinan t i l l in the southern Rocky Mountain Trench was investigated to supplement information on the pattern of ice flow and the origin of the t i l l . Fabrics were determined at 3 sites from the lower t i l l and at 26 localities as near as possible to the stratigraphic top of the upper t i l l , but below the zone of active frost disturbance and pedogenesis. At several localities i t was necessary to sample well below the strati-graphic top of the upper t i l l because of erosion of ground moraine by late glacial meltwater. Sampling was limited to natural exposures along major river valleys and to fresh road cuts. At several sample sites, where a substantial vertical section of the t i l l sheet is exposed, fabrics were measured at 1 to 5 m vertical intervals. At each sampling site an inclined or vertical exposure free of slumping and 1 m high by 2 to 10 m wide was cleared. The trend and plunge of 20 or 60 elongate pebbles (a/b > 2, b/c - 1, a > 1 cm) taken from the cleared exposure were recorded. A l l clasts which met the elongation criteria were measured. Optimum sample size was determined by comparing the mean orientations of fabrics of size 20 and 60 taken from the same sites (Fig. 34). Mean orientations were nearly the same for t i l l samples with strong fabrics, 124 but in general were quite different for weaker fabrics, Andrews and Smith (1970, p. 520-521) have shown that for till-fabric samples of size 25, 50, and 100, the smaller samples yield essentially the same results as the larger samples, and standard deviation does not increase with decreasing sample size. Internal variability in t i l l fabric was evaluated by selecting a group of control sampling sites where the direction of ice flow is known from such criteria as drumlins and bedrock striations. Fabrics determined at these sites were compared with the trends of the associated glacial lineations. Data were plotted on an equal-area net and contoured according to the method reviewed by Ragan (1968, p. 79-82). A spherical probability distribution applicable to three-dimensional axial data was used to determine axes of maximum and minimum clustering for each fabric (Mark, 1973). Each data set was tested for randomness by using significance tests based on this distribution (Anderson and Stephens, 1971). Results. Examples of typical till-fabric patterns are shown in Figure 35 and Appendix 5. Most of the t i l l fabrics show one of the following patterns recognized by Ramsden and Westgate (1971, p. 339) as common in ground moraine: (1) sub-horizontal girdle with a single maximum on the girdle, or (2) sub-horizontal girdle with two maxima 90° apart on the girdle. For 3120 pebbles measured, the modal plunge angle is 5°, the mean, 20°, and the median, 14° (Fig. 36). Within the control group of samples, there is a close relationship between fabric maxima and trends 125 of associated glacial lineations (Fig. 37, Table 5). In a l l except two cases, the axis of maximum clustering is within 24° of the direction of glacier flow defined by glacial lineations. One of the two exceptions represents an a-axis maximum transverse to glacier flow (fabric no. 1); the other fabric has a maximum 59° from the trend of the associated drumlin (fabric no. 15). The fabric maximum of the latter is parallel to and aligned with Bull Valley (Fig. 37) which carried a major tributary glacier during late Wisconsinan time. It is possible that this fabric, measured 1 to 2 m below the drumlin surface, represents a parallel fabric produced by the Bull Valley tributary glacier; i f so, the oblique drumlin trend originated slightly later by flow down the Trench following recession of Bull Valley glacier. Evidence showing that tributary glaciers receded before the active glacier in the Trench is presented on p. 116. The close correspondence between fabric maxima and trends of other indicators of glacier flow shows that internal variability at fabric sites is low and that fabrics in the Trench are characterized by sub-horizontal girdles with a axes generally parallel to flow. Although there is no systematic plunge direction of fabric maxima, 69% of parallel-to-flow fabrics have axes of maximum clustering which plunge up-glacier (43% of individual clasts plunge up-glacier, 30% down-glacier, and 27% have transverse orientations). Axes of clustering and associated measures of s t a t i s t i c a l significance for each t i l l fabric are presented in Appendix 6. Axes of maximum clustering for nonuniform samples are plotted in Figures 37 and 38. Figure 37 presents a l l fabric results from the lower t i l l and from near the top of the upper t i l l , including the control group; Figure 38 126 presents results of fabrics determined vertically through the upper t i l l . Heavy Mineral Analysis Heavy mineral concentrates from t i l l have long been used to differentiate t i l l sheets and determine the sources of the glaciers which deposited them (e.g., Dreimanis and others, 1957; Dreimanis, 1961; Bayrock, 1962; Willman and others, 1963). Most such studies have been made in regions of low relief such as central North America. Recently, studies of t i l l composition, including heavy mineralogy, have yielded information on the entrainment and transport history of sediment by glaciers (Dreimanis and Vagners, 1971; Gross and Moran, 1971). For example, Dreimanis and Vagners (1971) have identified "terminal grades," or limiting grain sizes produced by glacial comminution for a variety of heavy minerals. The proportion of a particular mineral in its terminal size fraction is then a measure of the length of transport of the t i l l matrix. T i l l s in the southern Rocky Mountain Trench contain a charac-teristic suite of heavy minerals. Since the provenance of certain of these heavy minerals is known, i t has been possible to obtain information on the pattern of glacier flow in the Trench, which complements that provided by t i l l fabrics, and on the mechanisms of entrainment, transpor-tation, and deposition of sediment by ice. The amount of a given mineral or rock constituent in t i l l is dependent on a number of factors, the most important being: (1) abundance of outcrop of the mineral or rock at the base of the glacier upstream from 127 the depositional site (Harrison, 1960, p. 443; F l i n t , 1971, p, 180); (2) erodibility of the source rock (F l i n t , 1971, p. 180); (3) durability of the rock or mineral in transport (Holmes, 1952; Dreimanis and Vagners, 1971; F l i n t , 1971, p. 181); and (4) distance of transport (Gravenor, 1951; F l i n t , 1971, p. 181). The fourth factor i s related to durability, dilu-tion, and progressive deposition away from the source area. Since the turbulent mixing characteristic of rivers is absent in glaciers, and since diffusion of coarse sediment is probably negligible, sediment dilution by mixing within glaciers is probably limited to shear zones in terminal areas. However, dilution may occur at the till-glacier inter-face as pressure melting releases a mix of sediment eroded from different bedrock sources. Dilution by this mechanism would be expected at the base of two coalescent glaciers with lithologically distinct sediment loads. As the boundary between the two glaciers shifted laterally due to variations in relative ice flux, sediment at the t i l l - g l a c i e r interface previously deposited by one glacier would be diluted with sediment released by pressure melting from the other. Even beneath a single unidirectional glacier, lateral dispersion of sediment may occur, as shown by dispersion of indicator lithologies in fan-shaped patterns down-glacier from point sources (Dreimanis, 1956; Goldthwait, 1968) and by flow divergence around basal obstructions (Carol, 1947; Lliboutry, 1959). The amount of a particular constituent in t i l l i s , of course, proportional to the amount present at the sole of the glacier during deposition. Near the source area, a high percentage of a particular rock type or mineral at the base of the glacier would result through the pressure-melting depositional process in lodgement t i l l which also has a high content of 128 that constituent. As this constituent is deposited, less is available at the glacier base for deposition farther from the source. Thus, both ice and lodgement t i l l contain progressively less of a mineral or rock type as distance from the source increases. This assumes that the constituent is transported near the base of the glacier rather than in an englacial position above the zone of active deposition. This general relationship may be expressed mathematically: d N « N V (1) dx * Nx is the amount of the constituent at the glacier base or in t i l l at a distance, x, from the source. If a constant, k, is introduced: Eq. (2) can be integrated (at x = 0, Nx = em = N , where m is a constant of integration): Nx = Noe kX (3) The constant, k, is a measure of the rate as a function of distance at which deposition occurs. This depositional rate constant may be determined by glacier dynamics or by size and physical properties of the transported constituent. Data collected by Gravenor (1951) and Dreimanis (1956) show the relationship between percent of various t i l l constituents and distance from source (Fig. 39). If mathematical expressions of the form of Eq. (3) are fitted to the data, high correlation coefficients are obtained. 129 Method. Bulk t i l l samples for heavy mineral analysis were collected at 136 sites in the southern Rocky Mountain Trench and major tributary valleys (Fig. 40). Of these, 117 were collected from canyon walls, road cuts, and surface excavations near the stratigraphic top of the upper t i l l . These samples were taken below the soil horizon of calcium carbonate accumulation. An additional 16 samples were collected vertically through the t i l l sheet at several localities. The remaining samples (nos. 18, 23, and 99) are from the lower t i l l exposed along St. Mary and Elk Rivers. Each sample is a composite of 1 to 2 kg of t i l l taken from two 1 m2 plots separated approximately 3 to 5 m laterally. Recovery of the heavy mineral fraction was accomplished through standard procedures (Krumbein and Pettijohn, 1938) of disaggregation, wet sieving, and gravity separation in bromoform (density = 2.89). The fine sand (105-210 urn) fraction was selected for quantitative analysis because i t represents the largest size fraction consisting almost entirely of monomineralic grains. Minerals were identified optically and by magnetic separation and x-ray diffraction methods. Grain mounts were then prepared, and 500 grains on each of two glass slides were point-counted for each sample. Recounts of the same sample yielded similar results. Data were recorded both as volume percent of total heavy minerals of fine sand size and as volume percent of the total fine sand fraction. The areal distribution of certain minerals near the top of the t i l l sheet was determined by trend-surface statistical techniques. A computer program modified from that of O'Leary and others (1966) was used 130 to process the data. Results. The following heavy minerals are recognized in Rocky Mountain Trench t i l l s : pyrite, sphalerite, galena, hematite, ilmenite, r u t i l e , goethite (pseudomorphic after pyrite), magnetite, dolomite, apatite, garnet, zircon, kyanite, staurolite, sphene, epidote group, tourmaline, amphibole family (hornblende, ac t i n o l i t e ) , muscovite, b i o t i t e , and chlorite.2 Altered s i l i c a t e grains, mainly amphibole, are common in a l l samples. Several rare minerals were not identified. Several minerals have restricted bedrock sources and are therefore of value in determining glacier-flow patterns. For example, in the fine sand fraction amphibole is almost entirely restricted to, and magnetite most common i n , mafic s i l l s and dikes of Precambrian age, which are most common in the St. Mary Valley area (Fig. 41). Tourmaline is an alteration mineral associated with the Sullivan ore body near Kimberley, but i t s frequency in t i l l is very low. Garnet and staurolite are rare minerals in southern Rocky Mountain Trench bedrock which has only been metamorphosed to a maximum of the greenschist grade (Wheeler and others, 1972, their Fig. 16). These minerals are present in the contact zones of granitic bodies in the Purcell Mountains (J.E. Reesor, written commun., 1971). However, t i l l s in tributaries draining such granitic bodies (e.g., samples 159 and 160, Appendix 7) contain very l i t t l e garnet and staurolite. It is thought that the source of these two minerals i s regionally metamorphosed rocks of the staurolite and kyanite zones which crop out extensively along the 2P a r t i a l separation of muscovite, b i o t i t e , and chlorite in bromoform. 131 Trench north of Kinbasket Lake (latitude 52°N), Percentages of abundant heavy minerals and some less common ones with restricted bedrock sources are presented in Appendix 7. Adjacent samples at a site exhibit l i t t l e variation in heavy mineral content indicating that internal variability is slight. Although there are some differences in heavy mineral percentages among vertical samples through the t i l l sheet at a station, there is no systematic variation with depth. The heavy minerals may be placed into two groups on the basis of their source and areal distribution: (1) local—minerals eroded from bedrock sources within the study area and transported short distances; (2) foreign—minerals from outside the study area transported relatively long distances. The first group includes such minerals as magnetite, dolomite, and amphibole; the second, garnet and staurolite. Figure 42 shows the areal distribution and trend of amphibole from the top of the t i l l . The local bedrock source, mafic dikes and s i l l s in St. Mary Valley, is readily apparent. Of particular note are (1) the rapid decrease of amphibole in t i l l away from the source, and (2) the presence of bands of amphibole-poor t i l l parallel to the Trench trend south of two major tributaries (Elk Valley and the Kootenay-Lussier Valleys). These bands become diffuse with distance down-Trench from the tributary mouths. Similar results are provided by magnetite and other opaque minerals. Figure 43 shows the areal distribution and trend of garnet. Although the fitted surface does not have as high a correlation coeffi-132 cient as that for amphibole, some observations can be made: (1) the garnet content in t i l l is highest at the north end of the study area, which is in agreement with the postulated northern source for the mineral; (2) an area of low-garnet t i l l off St. Mary Valley separates areas with higher garnet percentages to the north and south; and (3) as in the case of amphibole and other minerals of local origin, bands of garnet-poor t i l l parallel to the trend of the Trench occur south of Elko and the Kootenay-Lussier Valleys. Clast Lithology Analysis Clast lithologic analyses of t i l l have been widely used to determine glacier-flow patterns (Milthers, 1942; Dreimanis, 1956), to differentiate t i l l s (Howard, 1956), and to infer the mechanism of sedi-ment transport and the cause of the down-glacier decrease in clast frequency (p. 126) (Holmes, 1952; Anderson, 1955). Method. Clast lithologies were determined from samples of 100 pebbles (2> axis 16 to 32 mm) sieved from several kilograms of t i l l and outwash. In addition to t i l l samples collected from near the top of the t i l l sheet, a suite of vertical samples was taken at several sites (Fig. 40). Most samples analyzed for heavy mineral content were also analyzed for clast lithologies. There are few distinctive bedrock lithologies which are restricted in distribution in the study area. Granitic rocks, although distinctive, are distributed too widely to be of value as provenance 133 indicators. Mafic igneous rocks of Precambrian age, although widespread, are most abundant in St. Mary Valley (Fig, 41), Precambrian laminated greenish-gray argillite and red-purple argillite, s i l t i t e , and quartz arenite are most abundant along the east side of the Trench south of Elko, Paleozoic carbonates crop out extensively along the east side of the Trench near Elko and at the north end of the study area. Finally, distinctive chert-pebble conglomerate and chert arenite of Mesozoic age are restricted to Elk Valley. Results. Percentages of the above mentioned rock types (in the pebble fraction of t i l l and outwash) are presented in Appendix 7. Adjacent t i l l and outwash samples are similar in composition; as was true for heavy mineral analyses, there is no systematic variation in lithology with depth at multiple sampling sites. The areal distribution of clasts from the top of the t i l l sheet shows that local bedrock is the dominant component in t i l l . Mafic igneous pebbles are common only near their bedrock source at the mouth of St. Mary Valley (Fig. 44), and their frequency decreases rapidly with increasing distance from this source. The pattern for mafic igneous clasts is nearly identical to that for amphibole (Fig. 42). Since both were derived from the same source, their history of entrainment, transport, and deposition must have been similar. Other clast lithologies also show peaks in frequency immediately down-glacier from their bedrock sources: laminated greenish-gray argillite is common only along the east side of the Trench south of Elko, dolostone is common at the north end of the study area, and limestone both at the north end of the study area and 134 near Elko. Foreign clasts are rare in Rocky Mountain Trench t i l l . Durable Mesozoic rocks which crop out only in Elk Valley are present, however, as conglomerate and sandstone pebbles. The area of occurrence of Elk Valley chert-pebble conglomerate and chert arenite in Trench t i l l is shown in Figure 45. Their occurrence north of the mouth of Elk Valley is perhaps due to the spillover of ice from Elk Valley into the Trench via Sand Creek or to the lateral transfer of clasts englacially by meltwater. Discussion The dominance of drumlins as surface landforms on t i l l in the southern Rocky Mountain Trench indicates that t i l l deposition during the closing phases of glaciation occurred subglacially beneath active ice rather than by supraglacial melt-out or mass movement of supraglacial t i l l . Fabric characteristics establish further limits on the mode of t i l l deposition. Most fabrics, including those taken in control areas below drumlin surfaces, display sub-horizontal axes of maximum clustering parallel to the direction of glacier flow (p. 124). Such fabrics are produced at the active glacier-till interface by subglacial melt-out of overridden sediment-charged ice or by lodgement resulting from pressure melting (Table 4; Boulton, 1970a). Alternatively, similar fabrics are produced by shear or viscous flow following deposition of the t i l l , with the necessary force provided by glacier overriding (Ramsden and Westgate, 1971). However, post-depositional fabric reorientation has not occurred to any significant depth in the southern Rocky Mountain Trench, as 135 Figure 31. Index map showing trends of glacial lineations and the distri-bution of drift (alluvium, present along major rivers, is not differentiated from drift in this diagram). Relatively thick t i l l is present over much of the Trench floor, but is thin or absent on the uplands. The dotted line is the limit of differentiation of bedrock and unconsolidated sediments. Distribution of bedrock and unconsolidated sediments from Leech (1958, 1960). Figure 32. Late Wisconsinan t i l l in the southern Rocky Mountain Trench. A. Younger t i l l overlying sand deposited during the early interstade of the Pinedale Glaciation, Elk River (dotted line at contact). B and C. Younger t i l l , St. Mary River. 137 138 evidenced by sharp vertical changes in fabric orientation resulting from shifts in ice-flow direction. Had fabric reorientation occurred, the t i l l fabric would have been homogenized vertically at any one site. It is probable, then, that much of the t i l l accumulated by lodgement against a planar substratum. The exact mechanism by which this occurs is not known, but the following are possibilities: (1) particles may become individually lodged against an obstacle already present on the floor (Goldthwait, 1971, p. 16; Post and LaChapelle, 1971, their Fig. 29); (2) basal layers of debris-rich, moving ice melt when the combination of geothermal heat and heat produced by sliding exceeds that conducted away from the base of the glacier (Nobles and Weertman, 1971); calculations indicate that several centimeters of ice per year may be melted by these sources of heat (Robin, 1955; Weertman, 1963); (3) basal debris-rich ice may become immobile beneath the overriding glacier. A l l of these mechanisms involve basal sliding, and thus a glacier not frozen to its bed (Weertman, 1961). The record of glacier flow variations shown by fabrics through the t i l l sheet suggests that t i l l deposition occurred either progressively or sporadically throughout glacial occupation of the area, and was not limited to the terminal phase of glacier recession when large masses of englacial debris might be freed by rapid bottom melting as suggested by Goldthwait (1971, p. 17). The presence of a drumlinized t i l l surface further argues against the latter depositional mechanism. T i l l fabrics successfully document at least one major shift in the pattern of late Wisconsinan ice flow in the southern Rocky Mountain Figure 33. Features which indicate the direction of ice flow immediately prior to deglaciation of the southern Rocky Mountain Trench. A and B. Drumlins. C. Bedrock striations. The direction of flow was toward the left in A and C, and toward the right in B. Aerial photo number BC 5353-101. 140 Trench (Fig. 46). Combined till-fabric, striation, and drumlin data show the pattern of last flow (Fig. 37) to be parallel to the Trench margins except where ice fanned out in the Kimberley-Cranbrook lowland and bulged up into St. Mary, Elk, and probably Bull Valleys. Tributary glaciers at this time were isolated from active ice in the Trench. This stage was preceded by a period when tributary glaciers were confluent with the trunk glacier, as indicated by fabrics off St. Mary, Elk, and Bull Valleys with orientations parallel to the flow directions from these tributaries. These fabrics occur below those with down-Trench orientations. Tentative correlations of fabrics in the upper part of the t i l l sheet at vertical sampling localities along St. Mary River have been made on the basis of fabric and stratigraphic similarities (Fig. 38). Glaciolacustrine sedi-ments below the lowest fabric at site no. 19 correlate with i n t r a - t i l l glaciofluvial gravel (elev. 820 m) near site no. 24. Fabrics above these stratified sediments exhibit similar patterns at each station. Thus, correlation's made on the basis of fabric similarity are in agreement with the stratigraphic evidence. At the three sites (nos. 23, 24, and 25) off St. Mary Valley where fabric data could be gathered low in the t i l l sheet, correlations are more tenuous. Some fabrics display axes of maximum clustering either transverse to flow or at an acute angle to the overall down-Trench transport direction. It is possible that the latter repre-sent flowtill fabrics formed during melt-out accompanying the onset of the interstade between the middle and late stades of the Pinedale Glaciation. If this interpretation is correct, no fabric correlation either between sites or with respect to the glacier-flow direction would be expected. This explanation requires glacier stagnation which should Table 4. Relation between orientation of fabric elements in t i l l and t i l l genesis. ORIENTATION OF FABRIC ELEMENTS GENESIS OF FABRIC Englacial transport Deposition Post-depositional reorientation a axes of rods and blades p a r a l l e l to flow direction. a axes of rods plunge up-g l a c i e r ; a-b planes of blades dip up-glacier. a axes of rods and a-b planes of blades sub-hori z o n t a l . P a r a l l e l position attained by elongate objects i n a shearing medium (Jeffrey, 1922). Shear along planes i n terminal zone of g l a c i e r . Extending flow. Melt-out (preservation of e n g l a c i a l f a b r i c ; magnitude of plunge of a axes dependent on debris content of ice p r i o r to melt-out). Lodgement by pressure melting against an obstruction.* (Fabric may be produced by dragging of c l a s t s at base of g l a c i e r . ) Shear within t i l l by overriding g l a c i e r . a axes of rods and blades transverse to flow di r e c t i o n . a axes of rods sub-h o r i z o n t a l ; a-b planes of blades dip up-glacier. Reorientation from p a r a l l e l position to transverse position where there i s minimum energy d i s s i p a t i o n (Jeffrey, 1922), Compressive flow i n terminal zone of g l a c i e r . Melt-out (preservation of e n g l a c i a l f a b r i c ; magnitude of dip of a-b planes dependent on debris content of ice p r i o r to melt-out). Lodgement by pressure melting against an obstruction.* (Fabric may be produced by r o l l i n g of clasts at base of gla c i e r . ) a axes of rods and a-b planes of blades unrelated to flow. Neither marked compressive nor extending flow. Melt-out (destruction of e n g l a c i a l f a b r i c ) . Mass movement of supraglacial t i l l ( f l o w t i l l ) . P o s t g l a c i a l creep or landslide. *Subglacial fabrics produced by pressure melting depend on the configuration of the g l a c i e r - t i l l i n t e r f a c e ; where this interface i s planar, p a r a l l e l fabrics are dominant. 142 be indicated in the stratigraphy at these l o c a l i t i e s . Although there are lenses of water-worked sand and gravel and local occurrences of crude st r a t i f i c a t i o n in the lower part of the t i l l sheet, their regional significance i s unknown. An alternative explanation is that the pattern of ice coalescence off tributary valleys preceding the interstade was geometrically complex as a result of rapid changes in the flux of ice out of the tributaries. T i l l fabric at any one site might then reflect local anomalies of ice flow rather than the regional down-Trench flow direction. Examples of such distorted flow patterns are common where tributaries join trunk glaciers in the presently glaciated Cordillera (Fig. 47). Whatever the explanation, the fabric pattern in the lower part of the upper t i l l sheet is more variable than that near the top in relation to ice-flow pattern. Heavy mineral and clast lithology results, as well as the results of t i l l - f a b r i c analyses, show the dominant pattern 'of ice flow in the southern Rocky Mountain Trench: a south- to southeast-flowing trunk glacier augmented by coalescent tributary glaciers. Compositional differences are most distinct just south of the mouth of each major tribu-tary where sediments of Trench and side-valley provenance occur as more or less distinct entities. Farther down the Trench, lateral differences in composition become less distinct, probably as a result of lateral mixing of sediment englacially or subglacially. This mixing may be due to lateral shifts in the position of the zone of coalescence of Trench and tributary glaciers with changes in relative ice flux between the two. Temporal fluctuations in the pattern of glacier flow, although DIFFERENCE (in degrees) 1 — I V 10 20 DIFFERENCE (in degrees) 50 I | 40 30 - |50 40 t • J . 10 -II- ± 70 80 -I L 30 10 Plunge 20 30 Azimuth DIFFERENCE BETWEEN AXES OF MAXIMUM CLUSTERING FOR SAMPLES OF SIZE 60 AND 20 (in degrees) one point two coincident points Figure 34. Relation between till-fabric strength and differences in the fabric results for samples of size 60 and 20. Theta is the standard scattering angle around the axis of maximum clustering (in general, the greater the directional strength of the fabric, the smaller is theta). Theta is plotted against the differ-ence in orientation of the clustering axes of the larger and smaller samples. The histograms show that most of the smaller samples have clustering axes near those of the paired larger samples; trend differences (left) are mostly less than 20°, and differences in plunge magnitude (right) are mostly less than 10°. 144 Figure 35. Selected fabrics from the younger t i l l . Also shown is a contoured diagram of 156 axes of maximum clustering representing a l l till-fabric results from the southern Rocky Mountain Trench. Contours approximately 2-5-8% per 1% area. Additional till-fabric contoured diagrams are shown in Appendix 5, and a l l till-fabric sample statistics are presented in Appendix 6. 145 TILL FABRICS 1 TILL-FABRIC SAMPLE NUMBER AND SITE • TILL-FABRIC AXIS OF MAXIMUM CLUSTERING 146 Figure 36. Magnitude and direction of plunge of t i l l clasts. A. Curve shows the cumulative percent of clasts plunging less steeply than values listed on the abscissa; a l l till-fabric data are included. B. Rose diagram shows direction of plunge of t i l l clasts (the arrow points down-glacier); only till-fabric data for which the direction of glacier flow is independently known are included. 147 — r — r— 1 1 1— — T " 99.5 l 99 — 98 - -95 r-290 -* 80 h S 50 3 40 - -30 - -20 -10 nc3120 i i 1 1 1 i 1 10 20 30 40 50 60 70 PLUNGE ( ° ) A. MAGNITUDE OF PLUNGE n = 960 B. DIRECTION OF PLUNGE Figure 36 148 Figure 37. Comparison of till-fabric axes of maximum clustering (shown by lines) and associated glacial lineations (shown by arrows) (see also Table 5). Till-fabric sample statistics are presented in Appendix 6. 149 Table 5. Comparison of till-fabric maximum clustering axes and the trends of associated glacial lineations (see also Fig. 37). Fabric no. Location Elevation* (m) Number ofs observations CD T i l l f a b r i c -azimuth of axis of maximum clustering (2) Associated drumlins or striations— up-glacier trend* Difference in orientation between (1) and (2) in degrees 1 49°06'20"N, 115°05'30"W 850 60 72 344 86 AA 49o10'25"N, 115°11'10"W 821 20 336 345 9 6 49°12'00"N, 115°15'25"W 820 60 10 346 24 7 49°12'35"N, 115°07'20"W 880 60 171 346 5 8A 49°12'40"N, 115°10'15"W 829 20 170 342 8 10 49°16'00"N, 115°11'50"W 820 60 345 345 0 11 49°17'25"N, 115°08'10"W 900 60 36 27 9 12 49°18'00"N, l i s ' i e ^ ' v * 820 60 145 333 8 13 49°20'35"N, 115°17'10"W 840 60 150 326 4 14 49°23'20"N, 115°15'35"W 870 60 341 327 14 15 49°24'25"N, 115°22'45"W 870 60 14 315 59 16 49°27*20"N, 115°04'10"W 1000 60 351 344 7 17 49°34'25"N, 115°40'25"W 870 60 315 311 4 19A 49°35*15"N, 115°48'40"W 919 20 32 28 . 4 23A 49°36'45"N, 115"42'35"W 864 60 334 331 3 24A 49°36'50"N, 115<>40'05"W 840 60 328 319 9 28 49°42'25"N, 115°49'05"W 910 60 200 11 9 29 49°59'20"N, 115°45'20"W 900 60 317 339 22 o *Approximate. 151 shown by fabric analyses, are not recorded in t i l l composition. For example, although fabrics, glacial lineations, and stratigraphy show a late shift in glacier flow resulting from recession of tributary glaciers out of the Trench (p. 116 and 140), heavy minerals and clasts near the top of the t i l l sheet off these tributaries are, in large part, of side valley rather than Trench provenance (Figs. 42 and 44). In detail, the upper part of the t i l l sheet is characterized by an increase in tributary lithologies and a decrease in Trench lithologies toward the mouths of side valleys. Even very near the tributary sources, however, t i l l contains a minor component of Trench provenance. The heavy mineral and clast lithology results, then, suggest some local, subglacial reworking or mixing of sediment. Localized reworking of previously deposited t i l l would complicate an earlier flow pattern by indicating different flow directions at what appears to be the same stratigraphic level in the t i l l ; this may explain some of the apparently anomalous fabrics of Figure 38. Mixing may also explain the apparent homogeneity of t i l l composition, particularly if the mechanism of mixing is independent of shifts in directions of glacier flow. One such possibility is the introduction to the subglacial position of a mix of sediment via a system of englacial meltwater conduits. Sediment transported to the glacier base by meltwater might be deposited as lodgement t i l l ; its vertical distribution in t i l l would be sensitive to availability of supply and only indirectly or not at a l l to shifts in flow direction. Areal distri-bution would reflect proximity to bedrock source and direction of englacial meltwater transport. The pattern of englacial "plumbing" would necessarily evolve with glacier flow. There was meltwater within or at the base of Figure 38. Till-fabric axes of maximum clustering determined at vertical intervals of 1 to 5 m through the younger drift. Each open and closed circle represents one fabric; circles without attached lines are fabrics which are statistically uniform at the 10% level. The maximum clustering axis of each nonuniform fabric is shown by a line which ends in a circle at the elevation of the sample (north-south lines are parallel to the short edge of the page). The words gravel and silt refer to sediments which overlie and underlie the t i l l . A question mark indicates that the base of the t i l l is unexposed. The word top indicates that the stratigraphic top of the t i l l is present. Tentatively correlated fabrics from sites 19, 23, 24, 25, and 26 are labelled the same letter (x, y, or z) . Till-fabric sample statistics are presented in Appendix 6. 154 the active Trench glacier because lenses of water-worked sand and gravel are present within otherwise massive t i l l . If this explanation is correct, the distribution of lithologies in the t i l l may be, in large part, controlled by meltwater and, to a lesser extent, by active-ice transport. Two other observations, which are enigmatic if active-ice transport is sought as their sole explanation, are explained by the above hypothesis: (1) Garnet-staurolite distribution. Garnet and staurolite are thought to have been transported by ice from their source in the Trench north of about 52°N latitude (p. 130). Both minerals decrease in abundance southward in the area studied here. Off St. Mary Valley there is a pronounced garnet-staurolite low flanked to the north and south by higher frequencies. Although one might anticipate a decrease in these two minerals due to the supply of garnet-staurolite-poor sediment by the St. Mary tributary glacier, their increase south of St. Mary Valley would be unexpected unless they were transported englacially in a position or manner separate from minerals of St. Mary Valley provenance. This would be achieved i f much of the local sediment was transported in a short distance to the subglacial depositional position by meltwater, while the long-traveled foreign sediment was transported by the host ice. In this manner, locally derived sediment would greatly dilute garnet and staurolite southeast and east of St. Mary Valley, but the normal englacial load of garnet and staurolite would not be diluted with the short-traveled meltwater load farther south. (2) Chert-pebble conglomerate and chert arenite distribution. The occurrence in Trench t i l l of distinctive clastic rocks cropping out only in Elk Valley is shown in Figure 45. Pebbles Figure 39. Relation between percent of a constituent in t i l l and distance from the bedrock source. A. Crystalline clasts, southwestern Ontario (data from Gravenor, 1951). B. Iron ore clasts, southern Ontario (data from Dreimanis, 1956). 156 occur in relatively low abundance at the top of the t i l l sheet on both sides of the Trench south and west of Elk Valley, despite the fact that associated t i l l fabrics and landforms indicate a terminal phase of flow down-Trench with no contributions of ice from tributary valleys. These anomalous observations are reconciled i f the pebbles were transported down and across the Trench via meltwater through ice tunnels to near the base of the glacier, where they were deposited as lodgement t i l l . The occurrence of these pebbles at the very top of the t i l l sheet must be due to reworking of older lodgement t i l l . Alternatively, the pebbles might have been reworked from sediments older than late Wisconsinan t i l l , but this leaves unexplained their Trench-wide distribution and restricted northward occurrence. It is generally argued that breakage and abrasion during trans-port are the controlling factors in percentage decreases of minerals and clasts down-glacier from their sources (Holmes, 1960; Dreimanis and Vagners, 1971; Drake, 1972). Certainly abrasion is indicated by striated and faceted clasts common in t i l l , and clast size decreases with distance from source. But caution must be exercised in conclusions of this sort based on lithologic studies of specific t i l l size fractions. Losses within a size fraction due to breakage would normally be compen-sated by additions due to breakage of larger clasts. Indeed, at the "terminal grade" of breakage (p. 126; Dreimanis and Vagners, 1971) for a particular mineral, the additions from larger size fractions might result in net increases of that mineral with distance of transport. Thus, breakage will not necessarily result in the decrease in abundance of a mineral or rock type within a specific size fraction with increasing trans-157 158 Figure 41. Distribution of selected bedrock lithologies; unpatterned areas inside the solid line represent other bedrock lithologies. Laminated greenish-gray argillite and red-purple argillite, s i l t i t e , and quartz arenite are most common on the east side of the Rocky Mountain Trench south of Elko. Mafic igneous rocks are most abundant in St. Mary Valley. Bedrock geology modified from Leech (1958, 1960) 159 160 Figure 42. Amphibole distribution in t i l l of the southern Rocky Mountain Trench. At sites where two or more samples were collected, the mean amphibole content is shown. Contours are from the fourth-degree trend surface (r 2 = 0.88). Values are expressed as percent of fine sand fraction multiplied by 102. 161 162 Figure 43. Garnet distribution in t i l l of the southern Rocky Mountain Trench. At sites where two or more samples were collected, the mean garnet content is shown. Contours are from the sixth-degree trend surface (r 2 = 0.57). Values are expressed as percent of fine sand fraction multiplied by 102. 163 164 Figure 44. Distribution of mafic igneous pebbles in t i l l of the southern Rocky Mountain Trench. At sites where two or more samples were collected, the mean content of mafic igneous pebbles is shown. Contours are from the fourth-degree trend surface (r 2 = 0.72). Values represent percent of t i l l pebbles. 165 166 port. Where such decreases are observed, dilution by sediment mixing at the glacier-till interface (p. 127 and 151) or progressive deposition of sediment away from its source (p. 127) should be considered as con-trolling factors. Both factors have been shown to strongly influence mineral and clast lithology distributions in Rocky Mountain Trench t i l l . Although not directly evaluated, breakage and abrasion may be important as well. CONCLUSIONS (1) Deposits of three glacial advances have been identified in the southern Rocky Mountain Trench, and are tentatively correlated with deposits of the Pinedale Glaciation of the Rocky Mountains in the United States. (2) T i l l fabrics, landforms, and stratigraphic evidence show at least one major shift in the pattern of glacier flow near the end of glaciation in the southern Rocky Mountain Trench. An earlier period when major tributary glaciers coalesced with the trunk glacier in the Trench was followed by a period when tributary glaciers receded and active ice in the Trench bulged into the side valleys. The later phase was followed by orderly recession of the trunk glacier with neither major halts nor terminal stagnation and by drainage of ice-dammed lakes in tributary valleys. (3) T i l l fabrics are applicable in cordilleran areas and pro-vide information on directions of glacier flow and t i l l genesis. Strongly oriented fabrics both transverse and parallel to flow are produced during 167 Figure 45. Distribution of Blairmore Group clasts in drift of the southern Rocky Mountain Trench. Topographic contour interval 2000' (610 m). Outcrop area of Blairmore Group from Price (1962). NO a BLAIRMORE « CLASTS ELKOI BLAIRMORE CLASTS PRESENT 49fyo'N BLAIRMORE GROUP OUTCROP AREA OF BLAIRMORE GROUP. BOUNDARY BETWEEN DRIFT WITH BLAIRMORE CLASTS AND DRIFT WITHOUT BLAIRMORE CLASTS. 9 I N K I L O M E T E R S Figure 46. Two phases of "glacier flow near the end of the Pinedale Glaciation. A. During the earlier phase tributary glaciers were confluent with the trunk glacier in the southern Rocky Mountain Trench (arrows show directions of ice flow). B. During the later phase tributary glaciers were isolated from active ice in the Trench. The trunk glacier spread out across the lowland in the vicinity of Cranbrook and bulged into side valleys. Unpatterned area is lower than 1220 m (4000') in elevation. 169 M,0QoSll vO M,00oSll 170 Figure 47. Patterns of glacier coalescence off tributary valleys in the Cordillera. Top: coalescent glaciers flowing as separate ice streams parallel to the valley walls (Shakes Glacier, Alaska; photo by Austin Post). This flow pattern was probably common in the southern Rocky Mountain Trench during Pleistocene glaciations, as indicated by the occurrence off major tributary valleys of compositionally distinct bands of t i l l parallel to the Trench walls. Bottom: complex distortions of flow due to a surge by a tributary glacier (Susitna Glacier, Alaska; photo by Austin Post). 171 172 englacial transport, during deposition, and by shear and viscous flow within already-deposited t i l l . A consideration of the most common situa-tions in which t i l l originates leads to the conclusion that certain fabric patterns are to be more expected than others, namely, elongate elements defining a gently dipping girdle with either a single maximum parallel to flow or, less commonly, two maxima 90° apart. In areas of the Rocky Mountain Trench where the direction of glacier flow is independently known, rods in t i l l are parallel to flow and tend to plunge gently up-glacier. These fabrics were produced by lodgement of clasts due to subglacial pressure melting against a planar substratum, not by post-depositional reorientation nor by the melt-out process. Some fabrics from the lower part of the younger t i l l below interstadial sediments suggest either complex patterns of ice flow or mass movement of supra-glacial t i l l ( f l o w t i l l ) . In most areas, maximum information on t i l l origin and the direction of ice flow can be obtained by measuring the a axes of both rods and blades and the a-b planes of blades. A standardized procedure, incorporating Andrews's (1971) suggestions, should be adopted. (4) Heavy mineral and clast lithology analyses are less sensi-tive as indicators of ice-flow patterns than t i l l - f a b r i c results in the southern Rocky Mountain Trench because t i l l may have been reworked at the ice-sediment interface and because sediment may have been transported to it s subglacial depositional site by a mechanism other than the flowing ice i t s e l f . The distribution and relative abundance of pebbles and minerals in relation to their bedrock sources are compatible with (a) the 173 meltwater transport of part of the locally derived sediment to its subglacial position via a system of conduits in ice, followed by lodge-ment through pressure melting, and (b) limited reworking of t i l l at the glacier sole. Whatever their transportational history, most of the pebbles and fine sand In Rocky Mountain Trench t i l l are of local origin; garnet and staurolite, however, probably were transported by ice down the Trench at least 300 km from north of Kinbasket Lake. Observed rapid decreases in relative amounts of certain heavy minerals and pebbles away from their sources at the mouths of tributary valleys cannot be entirely due to breakage and abrasion en route, but are explained instead by progressive deposition and dilution through sediment mixing. Compositional differences in t i l l are most pronounced off tributary valleys and become less distinct in a down-glacier direction. This probably results from lateral mixing of sediment englacially or subglacially due to short-term shifts of the zone of coalescence of Trench and tributary glaciers caused by changes in relative ice flux between the two. 174 LITERATURE CITED Anderson, R.C., 1955, Pebble lithology of the Marseilles t i l l sheet in northeastern Illinois: Jour. Geology, v. 63, p. 228-243, Anderson, T.W., and Stephens, M.A., 1971, Tests for randomness of directions against equatorial and bimodal alternatives: Dept. Statistics, Stanford Univ., Tech. Rept, 5, 19 p. Andrews, J.T., 1971, Methods in the analysis of t i l l fabrics, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 321-327. Andrews, J.T., and Smith, D.I., 1970, Statistical analysis of t i l l fabric: methodology, local and regional variability (with particular reference to the north Yorkshire t i l l c l i f f s ) : Geol. Soc. London Quart. Jour., v. 125, p. 503-542. Armstrong, J.E., Crandell, D.R., Easterbrook, D.J., and Noble, J.B,, 1965, Late Pleistocene stratigraphy and chronology in southwestern British Columbia and northwestern Washington: Geol. Soc. America Bull., v. 76, p. 321-330. Bayrock, L.A., 1962, Heavy minerals in t i l l of central Alberta: Alberta Soc. Petroleum Geologists Jour., v. 10, p. 171-184. Boulton, G.S., 1968, Flow t i l l s and related deposits on some Vestspits-bergen glaciers: Jour. Glaciology, v. 7, p. 391-412. 1970a, On the deposition of subglacial and melt-out t i l l s at the margins of certain Svalbard glaciers: Jour, Glaciology, v. 9, p. 231-245. 1970b, On the origin and transport of englacial debris in Svalbard glaciers: Jour. Glaciology, v. 9, p. 213-229. 1971, T i l l genesis and fabric in Svalbard, Spitsbergen, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 41-72. Carol, Hans, 1947, The formation of roch.es moutonnees: Jour. Glaciology, v. 1, p. 57-59. Crandell, D.R., 1965, The glacial history of Washington and Oregon, in Wright, H.E., Jr., and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 341-353. Daly, R.A., 1912, Geology of the North American Cordillera at the forty-ninth parallel: Canada Geol. Survey Mem. 38, 857 p. 175 Drake, L.D., 1972, Mechanisms of clast attrition in basal t i l l : Geol. Soc, America Bull., v, 83, p. 2159-2166. Dreimanis, Aleksis, 1956, Steep Rock iron ore boulder train: Geol, Assoc, Canada Proc, v. 8, p. 27-70. 1961, T i l l s of Southern Ontario, in Legget, R.F., ed., Soils in Canada, geological, pedological, and engineering studies: Royal Soc. Canada Spec. Pub. , 3, p. 80-96:. Dreimanis, Aleksis, Reavely, G.H., Cook, R.J.B., Knox, K.S., and Moretti, F.J., 1957, Heavy mineral studies in t i l l s of Ontario and adjacent areas: Jour. Sed. Petrology, v. 27, p. 148-161. Dreimanis, Aleksis, and Vagners, U.J., 1971, Bimodal distribution of rock and mineral fragments in basal t i l l s , in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 237-250. Flint, R.F., 1971, Glacial and Quaternary geology: New York, John Wiley and Sons, Inc., 892 p. Fulton, R.J., 1971, Radiocarbon geochronology of southern British Columbia: Canada Geol. Survey Paper 71-37, 28 p. Glen, J.W., Donner, J.J., and West, R.G., 1957, On the mechanism by which stones in t i l l become oriented: Am. Jour. Sci., v. 255, p. 194-205. Goldthwait, R.P., 1951, Development of end moraines in east-central Baffin Island: Jour. Geology, v. 59, p. 567-577. 1968, Surficial geology of the Wolfeboro-Winnipesaukee area, New Hampshire: Concord, New Hampshire Dept. Resources and Econ. Development, 60 p. 1971, Introduction to t i l l , today, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 3-26. Gravenor, CP., 1951, Bedrock source of t i l l s in southwestern Ontario: Am. Jour. Sci., v. 249, p. 66-71. Gross, D.L., and Moran, S.R., 1971, Grain-size and mineralogical gradations within t i l l s of the Allegheny Plateau, in Goldthwait, R,P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 251-274. Harrison, P.W., 1957, A cla y - t i l l fabric: its character and origin: Jour. Geology, v. 65, p. 275-308. 1960, Original bedrock composition of Wisconsin t i l l in central Indiana: Jour. Sed. Petrology, v. 30, p. 432-446. 176 Holmes, CD. , 1941, T i l l fabric: Geol. Soc. America B u l l . , v. 52, p. 1299-1354. 1952, Drift dispersion in west-central New York: Geol. Soc. America Bu l l . , v. 63, p. 993-1010. 1960, Evolution of till-stone shapes, central New York: Geol. Soc. America Bull., v. 71, p. 1645-1660. Howard, A.D., 1956, Till-pebble isopleth maps of parts of Montana and North Dakota: Geol. Soc. America B u l l . , v. 67, p. 1199-1206. Jeffrey, G.B., 1922, The motion of ellipsoidal particles immersed in a viscous f l u i d : Royal Soc. London P r o c , ser. A, v. 102, p. 161-179. Kamb, W.B., and LaChapelle, E.R., 1964, Direct observation of the mechanism of glacier sliding over bedrock: Jour. Glaciology, v. 5, p. 159-172. Kauranne, L.K., 1960, A s t a t i s t i c a l study of stone orientation in glacial t i l l : Finlande Comm. Geol. Bu l l . 188, p. 87-97. Kelley, C.C., and Sprout, P.N., 1956, Soil survey of the upper Kootenay and Elk River valleys in the East Kootenay d i s t r i c t of British Columb i a : British Columbia Soil Survey Rept. 5, 99 p. Krumbein, W.C., and Pettijohn, F.J., 1938, Manual of sedimentary petrography: New York, D. Appleton-Century Co., Inc., 549 p. Leech, G.B., 1958, Fernie map-area, west half, British Columbia: Canada Geol. Survey Paper 58-10, 40 p. 1960, Fernie, west half, British Columbia: Canada Geol. Survey Map 11-1960. Lliboutry, Louis, 1959, Une th£orie du frottement du glacier sur son l i t : Annales Geophys., v. 15, p. 250-265. MacClintock, Paul, and Dreimanis, Aleksis, 1964, Reorientation of t i l l fabric by overriding glacier in the St. Lawrence Valley: Am. Jour. Sci., v. 262, p. 133-142. Mark, D.M., 1973, Analysis of axial orientation data, including t i l l fabrics: Geol. Soc. America Bu l l . , v. 84, p. 1369-1374. McCall, J.G., 1952, The internal structure of a cirque glacier, report on studies of the englacial movements and temperatures: Jour. Glaciology, v. 2, p. 122-131. Milthers, Keld, 1942, Ledeblokke og Landskabsformer i Danmark: Danmarks Geol. Unders., ser. 2, no. 69, 137 p. 177 Nobles, L.H., and Weertman, Johannes, 1971, Influence of irregularities of the bed of an ice sheet on deposition rate of t i l l , in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 117-126. O'Leary, Mont, Lippert, R.H., and Spitz, O.T., 1966, FORTRAN IV and MAP program for computation and plotting of trend surfaces for degrees 1 through 6: Kansas Geol. Survey Computer Contr. 3, 48 p. Post, Austin, and LaChapelle, E.R., 1971, Glacier ice: Seattle, Washington Univ. Press, 110 p. Prest, V.K., 1969, Retreat of Wisconsin and Recent ice in North America: Canada Geol. Survey Map 1257A. Prest, V.K., Grant, D.R., and Rampton, V.N., 1967, Glacial map of Canada: Canada Geol. Survey Map 1253A. Price, R.A., 1962, Fernie map-area, east half, Alberta and British Columbia: Canada Geol. Survey Paper 61-24, 65 p. Ragan, D.M., 1968, Structural geology, an introduction to geometrical techniques: New York, John Wiley and Sons, Inc., 166 p. Ramsden, John, and Westgate, J.A., 1971, Evidence for reorientation of a t i l l fabric in the Edmonton area, Alberta, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 335-344. Richmond, G.M., 1965, Glaciation of the Rocky Mountains, in Wright, H.E., Jr. , and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 217-230. Richter, Konrad, 1932, Die Bewegungsrichtung des Inlandeises, rekonstruiert aus den Kritzen und Langsachsen der Geschiebe: Zeitschr. fur Geschiebeforsch., v. 8, p. 62-66. 1936, Gefugestudien im Engebrae, Fondalsbrae, und ihren Vorlandsedimenten: Zeitschr. fur Gletscherk., v. 24, p. 22-30. Robin, G. de Q., 1955, Ice movement and temperature distribution in glaciers and ice sheets: Jour. Glaciology, v. 2, p. 523-532. Schofield, S.J., 1915, Geology of the Cranbrook map-area, British Columbia: Canada Geol. Survey Mem. 76, 245 p. S i t l e r , R.F., and Chapman, CA. , 1955, Microfabrics of t i l l from Ohio and Pennsylvania: Jour. Sed. Petrology, v. 25, p. 262-269. Weertman, Johannes, 1961, Mechanism for the formation of inner moraines found near the edge of cold ice caps and ice sheets: Jour. Glaciology, v. 3, p. 965-978. 178 1963, Profile and heat balance at the bottom surface of an ice sheet fringed by mountain ranges: Internat. Assoc. Sci. Hydro-logy, Gen. Assembly of Berkeley, 1963, Pub. 61, p. 245-252 [also available as U.S. Army Cold Regions Research and Eng. Lab., Research Rept. 134 (1964), 7 p.]. West, R.G., and Donner, J.J., 1956, The glaciations of East Anglia and the East Midlands: a differentiation based on stone-orientation measurements of the t i l l s : Geol. Soc. London Quart. Jour., v. 112, p. 69-91. Wheeler, J.O., Aitken, J.D., Berry, M.J., Gabrielse, Hubert, Hutchison, W.W., Jacoby, W.R., Monger, J.W.H., Niblett, E.R., Norris, D.K. , Price, R.A., and Stacey, R.A., 1972, The Cordilleran structural province, in Price, R.A., and Douglas, R.J.W., eds., Variations in tectonic styles in Canada: Geol. Assoc. Canada Spec. Paper 11, p. 1-81. Willman, H.B., Glass, H.D., and Frye, J.C., 1963, Mineralogy of glacial t i l l s and their weathering profiles in I l l i n o i s . Part 1. Glacial t i l l s : I l l i n o i s Geol. Survey Circ. 347, 55 p. Wright, H.E., J r . , 1957, Stone orientation in Wadena drumlin f i e l d , Minnesota: Geog. Annaler, v. 39, p. 19-31. Young, J.A.T., 1969, Variations in t i l l macrofabric over very short distances: Geol. Soc. America B u l l . , v. 80, p. 2343-2352. 179 CHAPTER FOUR: SEDIMENTOLOGY AND PALEOHYDROLOGY OF LATE WISCONSINAN OUTWASH ABSTRACT Late Wisconsinan ground moraine in the Rocky Mountain Trench, southeastern British Columbia, is dissected by meltwater channels which formed during the final retreat of the Cordilleran Ice Sheet, Outwash underlying the channels is coarse, poorly sorted, shows large-scale cross-bedding, and was deposited in channel-bar complexes of high-energy proglacial rivers. Length of transport of outwash gravel from t i l l source to channel depositional site is relatively short. The coarse fraction of t i l l is also of local derivation, as lithologies are similar to nearby bedrock lithologies. Peak discharges calculated from channel morphometry and maximum particle size were 10,000 to 20,000 m3/sec, larger than estimated maximum discharges of several thousand m3/sec attributable to summer runoff. Many channels transmitted peak discharges during jOkulhlaups from glacial lakes in tributary valleys. An empirical relationship between total volume discharged during documented jOkulhlaups and corresponding maximum instantaneous discharges is applied to Glacial Lake Elk to show that dis-charges equal to or larger than those calculated from channel morphometry were attained during jOkulhlaups. Mean discharge excluding jOkulhlaups for most of the active channels is estimated to be less than 1000 m3/sec; the ratio of maximum 180 instantaneous to mean discharges was about 10:1 but may have exceeded 20:1 in some channels. At low and moderate stages rivers had braided patterns within individual meltwater channels, but during peak flows channel bars were submerged. INTRODUCTION The Rocky Mountain Trench acted as a major channel for meltwater during retreat of the Cordilleran Ice Sheet from southern British Columbia. Wasting ice in the Trench and in the flanking Rocky and Columbia Mountains contributed meltwater which flowed south past 49°N latitude (Fig. 48), Deposits and landforms of the southern Rocky Mountain Trench record late glacial events. The flat valley floor is largely ground moraine with drumlins parallel to the Trench margins. This t i l l plain is traversed by meltwater channels, many of which are underlain by outwash. Large valley trains and kame terraces occur along the margins of the Trench (Fig. 48). The purposes of this paper are: (1) to characterize outwash and associated landforms produced during glacial recession in one area of the Canadian Cordillera and (2) to outline the paleohydrology of certain meltwater channels from the physical characteristics of the channels and from an empirical model based upon discharge records of proglacial rivers. 181 DISTRIBUTION AND CHARACTER OF OUTWASH The recessional outwash consists mainly of sand and gravel underlying meltwater channels or forming valley trains and kame terraces. On the east side of the Trench between Elko and Red Canyon Creek is a large terraced and channeled outwash plain. Individual meltwater channels trend south and southwest from this outwash surface towards the axis of the Trench. Sedimentary Textures The recessional outwash is poorly to very poorly sorted (Folk, 1968, p. 46) and coarse, with l i t t l e material larger than cobble-size or smaller than sand (Fig. 49, Appendix 8). Although lenses of sand are present in most exposures, bulk samples of outwash consist of from over 50% to more than 90% gravel. The sediment is much coarser than the modern alluvium of Kootenay River. In general, size frequency distributions are strongly fine-skewed to fine-skewed (Folk, 1968, p. 47) and range from very platykurtic to very leptokurtic (Folk, 1968, p. 48). Plots of skewness vs. deviation and skewness vs. kurtosis show the sediments to have textural attributes similar to river sediments but unlike other sediment groups (Friedman, 1961; Moiola and Weiser, 1968). Coarse channeled outwash near the Kootenay River is overlain by s i l t y sand (Fig. 48). The fine sediment probably accumulated during late stages of channel flow as backwater deposits where the channels intersected the ancestral Kootenay River. 182 Sedimentary Structures Common sedimentary structures in the outwash include parallel bedding, planar and trough cross-stratification, and current ripple lamination (Fig, 50). Small-scale cross-lamination, common in sand layers, is in part related to the migration of current ripple marks during deposition (Allen, 1963, p. 107-108) and in part to channel scour and f i l l . Large-scale cross-bedding predominates in gravel layers, Beds dip from a few degrees up to 30° and presumably originated through depo-sition at the site of migrating avalanche faces of bars and by channel f i l l following scour. Pebbles and cobbles within gravel beds are commonly imbricated with individual clasts inclined upstream. Surface bed forms of channels could not be adequately evaluated because of surface modification following channel abandonment. Changes include loess deposition on outwash surfaces and pedogenesis. Nevertheless, large bars and scour channels are preserved. Bars are mainly longitudinal and are flanked by scour channels. Some terminate with steep bar-avalanche faces. Most outwash surfaces are terraced and marked by a network of indistinct braided channels, or bear no visible bed forms. Probable large-scale mega-ripples with wave lengths up to about 25 m were observed on one outwash surface. These sedimentary structures and bed forms are characteristi-cally associated with fluvial and glaciofluvial deposits (Harms and Fahnestock, 1965). Similar structures occur in high-energy channel-bar complexes of contemporary braided rivers such as Donjek River, Yukon Territory (Williams and Rust, 1969). 183 Pebble Lithologies and Provenance Local differences in bedrock lithology enabled a study to be made of provenance of clasts in t i l l and outwash. Bedrock in and adjacent to the study area is composed of sedimentary and igneous rocks of Precambrian to Early Cretaceous (?) age (Leech, 1958, 1960). Similar lithologies are found in a number of formations, thus i t is commonly impossible to determine the bedrock source for pebbles and cobbles in outwash. Some rock types, however, are useful for provenance study. Lower Cretaceous chert-rich rudite, arenite, and si l t i t e crop out in Elk Valley. Clasts of these rocks are locally abundant in Elk Valley outwash, but are only minor constituents of outwash in the Rocky Mountain Trench, Other lithologies helpful in determining outwash provenance are mafic igneous rocks, grayish-red-purple quartz arenite, s i l t i t e , and argillite, and laminated greenish-gray argillite. These rocks are a l l of Precambrian age. Laminated greenish-gray argillite crops out extensively on the east wall of the Trench south of Elko but is less common in the Rocky Mountains to the north (Fig. 51). A Precambrian formation of red-purple argillite crops out mainly along the mountain front south of Elko, but also is present north of the study area. Mafic igneous rocks occur on the east side of the Trench north of Bull River and south of Elko. Pebble lithologies were determined from samples of 100 clasts (mean diameter 16 to 32 mm) sieved from several kilograms of outwash and t i l l . Two clusters of sample sites on the east side of the Trench were chosen, with adjacent bedrock lithologies differing at each cluster locality (Fig. 51); each cluster includes 5 outwash samples and 5 t i l l 184 samples. The purpose of the sampling scheme was to answer the following questions. Do the cluster localities differ in their clast lithological make-up, and, i f so, are these differences related to differences in bedrock lithologies? How far was the gravel fraction of t i l l transported? How much of the gravel was eroded from bedrock i n the immediate vi c i n i t y of the sampling site and how much was transported down the Trench from exposures far to the north? Do outwash samples differ from t i l l samples within a cluster, and what implications might this have for the length of transport of outwash clasts? Percentages of laminated greenish-gray a r g i l l i t e , red-purple clastic sedimentary rocks, and mafic igneous rocks are presented in Figure 51. Student's t-test (at 1% level of significance) shows both greenish-gray and red-purple a r g i l l i t e to be significantly more abundant in the southern cluster than in the northern cluster. Mafic igneous clasts are significantly more abundant at northern sample sit e s . The pebble component of t i l l i s thus lithologically similar to bedrock to the east and northeast. Most of the clasts have been trans-ported on the order of 10 km or less, rather than hundreds of kilometers. Paleozoic limestone, a major rock type immediately east and north of Elko and an important constituent of t i l l and outwash at Elko, i s a subordinate lithology in sediments at the southern sample locality about 17 km to the south. The areal distribution of green a r g i l l i t e clasts (Fig. 52), based on 28 t i l l samples, further indicates the similarity between pebble and local bedrock lithologies and supports the idea of short transport distances. Outwash on the floor of the Rocky Mountain Trench formed when 185 meltwater eroded ground moraine. The volume of t i l l gravel eroded by meltwater is about the same as the volume of gravel deposited by proglacial streams. Selective sorting during glaciofluvial transport and areal variations in t i l l composition would result in lithological differences between adjacent deposits of outwash and t i l l i f transport of the gravel fraction by meltwater was far. However, t i l l and outwash are not signi-ficantly different (5% level) in lithology at either sampling cluster, The derivation of outwash from t i l l and the lithological similarity between adjacent t i l l and outwash samples thus indicate short transport of outwash gravel. PALEOHYDROLOGY During the retreat of the Cordilleran Ice Sheet, large amounts of meltwater were discharged from the glacier front. The large meltwater channels, their outwash, and underfit streams and valleys in the Rocky Mountain Trench indicate that hydrologic conditions were different from the present. Hydrologic conditions during deglaciation were probably similar to those existing today in heavily glaciated cordilleran areas such as parts of Alaska, Yukon Territory, and British Columbia. Most temperate mountain glaciers there have receded since the Neoglacial maximum. Landforms uncovered during deglaciation and ice-marginal glaciolacustrine features are similar to late Pleistocene features in the Trench. Modern proglacial streams are commonly braided, transport coarse debris, have relatively steep gradients, and exhibit extremes of discharge. Ice-dammed lakes may drain rapidly during jOkulhlaups, and proglacial rivers then have peak water discharges tens of times greater 186 than average (Bradley and others, 1972; Mathews, in press), Meltwater channels are conspicuous features of the present landscape in the southern Rocky Mountain Trench. The channels vary in morphology and include the following types, (1) Narrow, relatively deep channels cutting across drumlins. These channels are generally discontinuous, in many places terminating against unmodified drumlins or kettle holes. There is generally l i t t l e or no outwash underlying their floors. They are thus strictly erosional features. Their anomalous morphology and association with ice-contact features suggest these channels were carved when the surrounding landscape was at least partly mantled by stagnant ice. Closely related to these are "serpentines" (Mannerfelt, 1945, p. 95-97, 212-214), small, sinuous channels situated on the flanks and tops of topographic highs. (2) Strictly lateral, single-walled channels. Such channels are ice-marginal erosional features formed before the bottom of the Trench was free of ice. The floors of many are bedrock, but in places have a thin cover of glaciofluvial gravels. (3) Wide, relatively shallow, double-walled channels underlain by outwash. Some are several hundred meters wide and many kilometers long, but most are short segments truncated by younger channels to the north and northwest (Figs. 48 and 53). They tend to be flat-floored except in areas of bars and scour channels (Fig. 54), Ice-marginal kame terraces and remnants of older, higher proglacial meltwater channels flank some of the channels. Most of the channels are cut in t i l l and 187 contain an outwash f i l l which, is generally a few meters in thickness but in places is thicker than 15 m. Cross-stratification and pebble imbrica-tion show that the outwash was deposited from water flowing to the south, southwest, and southeast away from the ice front. The distance between the active ice front and the depositional site was probably short, as most meltwater channels are pitted with kettles. Paleodischarges in meltwater channels may be calculated from empirically derived equations involving channel morphometric parameters and outwash texture. Dury (1965), for example, formulated equations relating meander morphometry and discharge at bankfull stage, and Schumm (1968) showed that channel morphometry varies with type and amount of sediment load as well as with discharge. Alternatively, since meltwater discharge is related to a number of basin parameters such as glacierized area and the number and volume of self-dumping ice-dammed lakes, paleodis-charges may be estimated from basin morphometry in the southern Rocky Mountain Trench. Paleodischarge Determinations from Channel Morphometry In a non-cohesive gravel-bed river, the size of the largest moving particle may be a measure of the critical tractive force, x c. This is expressed by the Shields relationship, x c = 890l>, where D is grain diameter in m, and x c is in newtons/m2 (Shields, 1936). This is valid for fully turbulent flow (i.e., generally, for D > 0.005 m). Church (1970, his Pig. 11.1) shows that for some flume, canal, and river data different constants apply in the above relationship, because of the variable effects 188 Figure 48. Distribution of outwash and meltwater channels, Rocky Mountain Trench, southeastern British Columbia. The dotted line south-east of Elko is the limit of differentiation of Quaternary sediments. Areas of rock outcrop and near-surface rock from Leech (1960). 189 UNDIVIDED QUATERNARY SEDIMENTS I 6 | ALLUVIUM ALLUVIAL FAN DEPOSITS OUTWASH SILT AND SAND OUTWASH GRAVEL TILL BEDROCK AT OR NEAR THE SURFACE BLUFFS OF UNCONSOLIDATED SEDIMENTS d, k SURFACE DUNED, KETTLED MARGIN OF MELTWATER CHANNEL OR TERRACE DRUMLIN STRIATIONS (DIRECTION OF ICE FLOW KNOWN, UNKNOWN) j ~ K . C R O W S N E S T C P A S S M O R R I S S E T \ KIIOMETERS i Figure 48 i o k m CONTOUR INTERVAL 1000' (305 m) 2 4 6 8 KILOMETERS 190 of imbrication and packing patterns of channel sediments and, to a lesser extent, suspended sediment concentration and viscosity as controlled by water temperature. For Baffin Island sandar: T c = 1800D (1) A similar constant applies to other gravel-paved rivers (Church, 1970, his Fig. 11.1). Outwash on Baffin Island sandar and on Rocky Mountain Trench outwash plains is similar in texture and structure. In both areas channel sediment is characterized by pebble imbrication and other oriented clast packing patterns, resulting in stable bed conditions under shear stresses in excess of the critical value predicted from the Shields relationship. These similarities suggest that the constant of Eq. (1) may yield more realistic estimates of critical tractive force than the standard Shields constant. Critical tractive force is also related to the slope of the energy grade line, s, and flow depth, d, by the formula: T C = pgsd (2) Combining Eqs. (1) and (2): d = 0.184D/s (3) (mks units) The Manning flow resistance equation is: i ? 2 / 3 s l / 2 v = (4) (mks units) Substituting d from Eq. (3) for R (hydraulic radius, R, is approximately equal to d for very wide channels) in Eq. (4) yields: Figure 49. Cumulative size frequency curves of outwash samples. The graphs relate grain sizes and weight percent of the sediment coarser than those sizes. A. Gravel pit 1, 49°13'N, 115°16'W (Appendix 8). B. Gravel pit 4, 49°13'N, 115°09*W (Appendix 8). 192 Figure 50. Sedimentary structures in outwash at gravel pits. A. Large-scale cross-bedding; exposure is about 5 m high and is parallel to axis of meltwater channel. B. Dipping gravel beds of A. C. Lensing of sand and gravel near margin of secondary channel; exposure is about 3 m high and is approximately normal to channel axis. D. Channel scour and f i l l ; trend of channel is normal to exposure E. Cross-laminated sand and parallel-bedded gravel; exposure is about 5 m high. F. Cross-laminated sand of E. 193 Figure 50 194 Figure 51. Percent of clasts of the following lithologies in outwash and t i l l : laminated greenish-gray a r g i l l i t e (source is largely the Roosville and Gateway Formations), red-purple a r g i l l i t e and quartz arenite (largely of the Phillips Formation), and mafic igneous rocks (Purcell volcanics and intrusions). Undifferentiated Purcell igneous rocks crop out on the east wall of the Trench near the northern sample cluster. At each of the two sampling clusters are five outwash and five t i l l samples. Bedrock geology of the east side of the Trench from Leech (1958, 1960). 195 Figure 51 196 0.323P2/3 v = C5) The resistance coefficient, n, is estimated from the Strickler equation (Chow, 1959, p. 205-206; Church, 1970, p. 467): n = 0.038D1/6 (6) (mks units) This equation yields resistance coefficients based upon the size of bed roughness elements, but not form resistance elements such as channel cross section and alignment. Combining Eqs. (5) and (6) yields velocity in terms of particle size and slope of the energy grade line; multiplying by Eq. (3) results in an equation for specific discharge, which when multiplied by channel width, w, leads to peak discharge, Q, during periods of flow capable of moving the largest outwash particles: Q = 1-56D3/2„ ( 7 ) s7 / 6 D was determined from the mean 2>-axis diameters of the 25 largest outwash clasts found at gravel pits along the meltwater channels, Very large, glacially striated clasts, thought to have been either ice rafted or incorporated in outwash as lag boulders from t i l l without glaciofluvial transport, were excluded in the calculation of D. Energy gradients were calculated from topographic maps with a 20-ft (6.1 m) contour interval and by surveying. To minimize the effect of differential isostatic rebound after channel abandonment, gradients were measured where channel directions are closest to the axis of zero differential 197 Figure 52. Fourth-order trend surface map of percent laminated greenish-gray argillite in t i l l ; r 2 = 0.96. Contours are drawn only on the floor of the Rocky Mountain Trench. Bedrock on the east wall of the Trench at and south of Elko is mainly gray and green argillite of Precambrian age. Laminated green argillite is less common in bedrock northwest of Elko. 198 u p l i f t . Channel widths were measured from 1:31,680 topographic maps and represent average values over sections of channel for which gradients were determined. Velocities and discharges for several of the larger meltwater channels are presented in Table 6. Larger channels carried discharges of about 20,000 m3/sec. Paleodischarge Determinations from Discharge Records of Proglacial Rivers Total meltwater runoff from glacierized basins in the Cordillera is a function of numerous variables including climate, ice surface area, and basin morphometry. Proglacial meltwater discharge may be interrupted by storage and intermittent release of water in self-dumping ice-dammed lakes. On the other hand, meltwater discharge produced by melting without storage in self-dumping glacial lakes i s continuous. In the latter case average discharge for a maximum flow period is closely related to the area of ice upstream from a proglacial river. The data in Figure 55, derived from topographic maps and water supply records for proglacial rivers, indicate a power relationship between peak discharge and glacier-ized area. The equation is of the form Q = aA^, where a and b are constants. Specific equations determined for periods of maximum flow are: Mean of maximum year of discharge: Q = 0 . 2 6 A 0 , 9 8 coefficient of determination =0.92 Mean of maximum month of discharge: Q = 0.43-fl1*06 coefficient of determination = 0.95 199 PS »^:1P^^^-' ' 3 1 t , 1 KM. 1 — z Figure 53. Meltwater channels. A. Photo stereogram of t i l l plain cut by meltwater channels, 49°06'-09'N, 115o05'-ll'W (BC 5353-052 and -053). B. Floor and east wall of one of the large meltwater channels of A. 200 Peak discharge: Q = 2.!UQ-99 coefficient of determination = 0,82 As the water discharge records are of limited and variable duration, derived Q values are somewhat less than long-term maxima produced by ablation. The small scatter of the data points of Figure 55 indicates, however, that derived discharges are similar in magnitude to long-term maxima for individual rivers; variance attributable to divergence of the two is relatively small. This is in part explained by observations that discharge differences between glacier runoff events with large recurrence intervals are proportionately smaller than differences between events with small recurrence intervals (Church, 1970, his Fig. 42). Maximum discharges can be determined i f the total glacier area contributing runoff to meltwater channels is estimated. In the case of paleochannels in the southern Rocky Mountain Trench such an estimate is only an approximation because the limits of ice contributing meltwater to channels south of Elko are not known. A major meltwater channel leaves the Trench at Cranbrook (approximately 49°30'N, 115°45'W) and probably discharged much of the meltwater originating north of Cranbrook. No other channels leave the Trench between 49°30' and 49°00'N, thus, channels on the east side of the Trench floor south of Elko were being fed by the Trench glacier at least as far north as Cranbrook. The estimate of glacierized area, assuming various positions of the glacier terminus between Elko and the 49th parallel, and varying degrees of tributary deglaciation, is then greater than 1000 but less than 4000 km2. The resulting estimates of average discharges for maximum periods of flow Figure 54. Topographic profiles across a major meltwater channel. Flat-floored parts of the channel (profile C-D) are separated by segments with longitudinal bars and secondary scour channels (profiles A-B and E-F). o 202 are listed in Table 7. Normal summer melt produced discharges of several thousand m3/sec, much smaller than the maximum value of about 20,000 m3/sec determined from channel morphometry. It is thought that the larger dis-charges occurred during jokulhlaups, which are floods resulting from the breaching of ice-dammed lakes. This is supported by the distribution (Fig. 48) and size of outwash in relation to former ice-dammed lakes in Trench tributary valleys, and by the presence of large longitudinal bars and probable mega-ripples (p. 182) on the surfaces of meltwater channels. Many tributary valleys were largely deglaciated when ice s t i l l f i l l e d the Trench. Ice-dammed lakes were present in the valleys of Wigwam, Elk, and St. Mary Rivers, and perhaps others. As the Trench glacier receded additional ice-dammed lakes formed in tributary valleys to the north. Recession of the trunk glacier probably caused progressive lowering and weakening of each ice dam until breached during jbkulhlaups. Many historic ice-dammed lakes drain catastrophically during a critical period of trunk glacier recession. For example, annual jSkulhlaups from Vatnsdalur, Iceland occurred only after 1898 during a period of rapid recession of the impounding glacier that began in 1887 (Thorarinsson, 1939, p. 223-224). Although glacier-dammed Summit Lake, British Columbia has existed since at least the first decade of this century, i t has drained catastrophically only since 1961 (Mathews, in press). Also, as Knik Glacier, Alaska receded during the 1930's, 40's, and 50's there was a trend toward earlier annual release of the water impounded by the glacier in Lake George (Stone, 1963, p. 34-35). These observations support the idea of a close relationship between the occurrence of jbkulhlaups and the strength of the ice dam. Table 6. Peak velocities and discharges determined from channel morpho-metry and maximum particle size. L o c a t i o n D i a m e t e r (b a x i s ) o f l a r g e s t c l a s t s (m) C h a n n e l g r a d i e n t A v e r a g e c h a n n e l w i d t h (m) V e l o c i t y T =890D c ( m / s e c ) T =1800D c P e a k d i s c h a r g e T =890D c ( m 3 / s e c ) T =1800D c 4 9 ° 0 0 ' - 0 5 ' N 1 1 5 ° 0 3 ' - 0 5 ' W 0 . 0 9 0 . 0 0 5 5 4 0 4 6 4 . 0 0 0 1 0 , 0 0 0 4 9 " 0 4 ' - 0 9 ' N 1 1 5 ° 0 5 ' - 0 9 ' W 0 . 0 9 0 . 0 0 4 6 3 0 4 7 6 , 0 0 0 2 0 , 0 0 0 4 9 ° 0 6 * - 1 0 ' N 1 1 5 ° 0 8 ' - 1 2 ' W 0 . 0 9 0 . 0 0 3 5 2 0 4 7 6 , 0 0 0 2 0 , 0 0 0 4 9 ° 1 3 * - 1 5 ' N 1 1 5 ° 0 8 ' - 1 0 ' W 0 . 1 0 0 . 0 0 9 3 4 0 4 6 1 , 0 0 0 4 , 0 0 0 4 9 ° 1 5 ' - 1 8 ' N 1 1 5 ° 1 1 ' - 1 4 ' W 0 . 1 0 0 . 0 0 5 5 7 0 4 7 4 , 0 0 0 1 0 , 0 0 0 204 Glacial Lake Elk was representative of lakes trapped in tribu-tary valleys by ice in the Rocky Mountain Trench, Active ice pushed into lower Elk Valley to maintain a dam near Morrissey. Glacial Lake Elk initially drained eastward at Crowsnest Pass (elev. 1357 m). Later the lake discharged to the southeast at an elevation of 1204 m near Morrissey. At this stage the lake had an area of about 200 km2 and a volume of more than 20 km3. After the ice south of Morrissey receded below the elevation of the Morrissey outlet, Glacial Lake Elk drained to the southwest towards Elko, thence south along the margin of the Trench into lower Wigwam Valley. A major bedrock gorge southeast of Elko (Fig. 48) was occupied and excavated by meltwater. Large amounts of greenish-gray argillite were eroded from the walls of the gorge, transported, and deposited on a wide depositional terrace north of Wigwam River near its mouth. The tremendous influx of debris into Wigwam Valley forced Wigwam River against the south wall of the valley. The combined meltwater then flowed southwest into the Trench. The size of outwash clasts downstream from the bedrock gorge indicates a major flood, probably produced by a jdkulhlaup from Glacial Lake Elk. During such meltwater floods, the terminus of the Trench glacier must have been south of Elko. Many of the channels in the southern part of the study area may have formed or been modified by jbkulhlaups from Glacial Lake Elk, An estimate of the maximum instantaneous discharge of a flood from Glacial Lake Elk may be made. Figure 56 shows the relationship between maximum instantaneous flood discharge and lake volume drained. For documented limno-glacial jbkulhlaups the equation of the regression line (r 2 = 0.96) is Q = 75V0*67, where Q is the peak discharge in m3/sec 205 / / 1 10 io2 io3 GLACIER AREA (km2) Figure 55. Relation of glacier area, A, and maximum instantaneous discharge, Q; r 2 = 0.82. Data are from water discharge records and topographic maps. Glacier or proglacial river: 1—Ram, Alberta; 2—Sentinel, British Columbia; 3—Woolsey, British Columbia; 4— Peyto, Alberta; 5—Wolverine, Alaska; 6—Athabasca, Alberta; 7— Lemon, Alaska; 8—Gulkana, Alaska; 9—Eklutna, Alaska; 10—Berendon, British Columbia; 11—Nellie Juan, Alaska. 206 and V is the volume drained (x 106 m3). If Glacial Lake Elk at the 1204-m stage drained completely, the maximum discharge of the resulting flood would be about 60,000 m3/sec. For only partial emptying, and for jbkulhlaups during the later stages of the lake, maximum discharges would be less. Meltwater channels west and northwest of Elko were not formed by floods from Lake Elk. Tributary valleys to the north, such as that of St. Mary River and possibly those of Bull River and Sand Creek, supported ice-dammed lakes somewhat younger than Glacial Lake Elk. Jbkulhlaups from these lakes were probably important in the formation of some meltwater channels within the study area. Flood waters probably traveled subglacially and englacially from the breach in the ice dam to the terminus of the Trench glacier. In many documented jbkulhlaups the distance between the source and the surface discharge point at the glacier terminus is many kilometers (Marcus, 1960; Mathews, 1965). Discussion of Results Peak discharges calculated from channel morphometry and outwash size were probably 10,000 to 20,000 m3/sec for several large meltwater channels (Table 6). Ablation would have been sufficient to account for the largest discharges only if a very large glacierized area were contri-buting meltwater which was then funneled into a single proglacial channel (Table 7). The most likely value for glacier area results in peak dis-charges from ablation of several thousand m3/sec, lower than maximum Table 7. Relation of glacier area and meltwater discharge during periods of maximum melt. Discharge estimates are determined from limiting values of glacierized area with the glacier terminus south of Elko. Mean of maximum Mean of. maximum Maximum year of discharge month of discharge instantaneous discharge Regression equation Q = 0.26A°-3Q Q = 0.43A1'06 0 = 2.JU0-99 Coefficient of determination 0.92 0.95 0.82 Q* (m3/sec) for maximum .glacier area 103 low 103 low 103 high Q* (m3/sec) for minimum glacier area 102 low 102 high 103 low •Minimum and maximum order-of-magnitude estimates of discharge calculated from limiting values of glacierized area, A (km2). Low = order of magnitude of Q x 1, 2, 3, or 4; high = order of magnitude of Q x 5, 6, 7, 8, or 9. 208 values determined from channel morphometry. Also, since a major melt-water channel trended along what is now the valley of the Kootenay River, and since this channel was active throughout deglaciation, in contrast to channels entering i t from the east and northeast (Fig. 48), total meltwater was divided between at least two channels at any given time during deglaciation. Direct runoff from the glacier was periodically augmented by large volumes of water released from ice-dammed lakes. Jbkulhlaups with peak discharges up to 60,000 m3/sec swept down the Trench from Glacial Lake Elk. Lake Elk was probably smaller than 20 km3 (the volume upon which the peak estimate was based) when channels south of Elko formed. However, a jokulhlaup of 20,000 m3/sec could result from a lake of 4 km3, much smaller than Glacial Lake Elk at the 1204-m stage. Jbkulhlaups of magnitude IO1* to 105 m3/sec compare in size to many Icelandic volcano-glacial floods (Thorarinsson, 1953; Rist, 1955) and floods from Lake George, Alaska (Stone, 1963; U.S. Geological Survey, 1969; Bradley and others, 1972), but are smaller than the late Pleistocene jbkulhlaups of Lake Bonneville (Malde, 1968) and Lake Missoula (Bretz, 1925, 1969; Pardee, 1942; Bretz and others, 1956; Baker, 1971). Peak discharges were many times larger than long-term mean discharges. Mean discharge excluding jbkulhlaups for proglacial rivers in the study area was probably less than 1000 m3/sec, even for years of exceptional melt (Table 7). The ratio of maximum instantaneous discharge to mean discharge was about 10:1 and may have exceeded 20:1 in channels affected by jbkulhlaups. Thus, during peak flows, bars on relatively 209 Figure 56. Relation of total volume drained during jBkulhlaup and peak water discharge; r2 = 0.96. Dashed lines indicate 95% confidence interval for estimates of peak discharge (residuals are assumed to be normally distributed). Ice-dammed lakes: 1—Strupvatnet (Whalley, 1971); 2—Ekalugad Valley (Church, 1970); 3—Demmevatn (Strom, 1938); 4—Gjanupsvatn (Arnborg, 1955); 5—Vatnsdalur (Thorarinsson, 1939); 6—Tulsequah (Marcus, 1960); 7—Summit (Mathews, 1965 and in press); 8—Graenalon (Thorarinsson, 1939); 9—George (Stone, 1963); 10— Missoula (Bretz, 1925; Pardee, 1942). 210 flat-floored channels were submerged, whereas at low and moderate stages, rivers had braided patterns within individual meltwater channels. CONCLUSIONS (1) Channeled outwash in the southern Rocky Mountain Trench was deposited by proglacial rivers during the final retreat of the Cordilleran Ice Sheet. (2) Large channels, coarse sediment, and sedimentary structures such as large-scale cross-stratification indicate deposition in channel-bar complexes of high-energy rivers. (3) Outwash was formed by the erosion of t i l l ; fines were removed and gravel transported a relatively short distance. The coarse fraction in t i l l is also of local derivation. (4) Peak discharges calculated from channel morphometry and maximum particle size were 10,000 to 20,000 m3/sec, larger than estimated maximum uninterrupted discharges such as would have resulted from direct summer runoff. (5) Many channels probably discharged peak volumes during jOkulhlaups from glacial lakes in tributary valleys. From an empirical relationship between total volume discharged during documented meltwater floods and corresponding maximum instantaneous discharge, i t is estimated that Glacial Lake Elk at one stage drained with discharges up to 60,000 m3/sec. Smaller floods during lower stages of the lake and jbkulhlaups from other lakes to the north may have formed many of the meltwater 211 channels and outwash deposits south of Elko. (6) Mean discharge for meltwater channels in the southern Rocky Mountain Trench was probably less than 1000 m3/sec, and the ratio of maximum instantaneous to mean discharges was about 10:1 and may have exceeded 20:1 in some channels. 212 LITERATURE CITED Allen, J.R.L., 1963, The classification of cross-stratified units, with notes on their origin: Sedimentology, v. 2, p. 93-114. Arnborg, Lennart, 1955, Hydrology of the glacial river Austurflj&t: Geog. Annaler, v. 37, p. 185-201. Baker, V.R., 1971, Paleohydrology and sedimentology of Lake Missoula flooding in eastern Washington [Ph.D. thesis]: Boulder, Colorado Univ., 146 p. Bradley, W.C, Fahnestock, R.K. , and Rowekamp, E.T,, 1972, Coarse sedi-ment transport by flood flows on Knik River, Alaska: Geol. Soc. America Bull., v. 83, p. 1261-1284. Bretz, J.H., 1925, The Spokane flood beyond the channeled scablands: Jour. Geology, v. 33, p. 97-115, 236-259. 1969, The Lake Missoula floods and the channeled scabland: Jour. Geology, v. 77, p. 505-543. Bretz, J.H., Smith, H.T.U., and Neff, G.E., 1956, Channeled scabland of Washington: new data and interpretations: Geol. Soc. America Bull., v. 67, p. 957-1049. Chow, V.T., 1959, Open-channel hydraulics: New York, McGraw-Hill Book Co., 680 p. Church, M.A., 1970, Baffin Island sandar, a study of arctic fluvial environments [Ph.D. thesis]: Vancouver, British Columbia Univ., 1066 p. Dury, G.H., 1965, Theoretical implications of underfit streams: U.S. Geol. Survey Prof. Paper 452-C, 43 p. Folk, R.L., 1968, Petrology of sedimentary rocks: Austin, Texas, Hemphill's, 170 p. Friedman, G.M. , 1961, Distinction between dune, beach, and river sands from their textural characteristics: Jour. Sed. Petrology, v. 31, p. 514-529. Harms, J.C, and Fahnestock, R.K., 1965, Stratification, bed forms, and flow phenomena (with an example from the Rio Grande), in Middleton, G.V., ed., Primary sedimentary structures and their hydrodynamic interpretation: Soc. Econ. Paleontologists and Mineralogists Spec. Pub. 12, p. 84-115. Leech, G.B., 1958, Fernie map-area, west half, British Columbia: Canada Geol. Survey Paper 58-10, 40 p. 213 • 1960, Fernie, west half, British Columbia: Canada Geol. Survey Map 11-1960. Malde, H.E., 1968, The catastrophic late Pleistocene Bonneville flood in the Snake River Plain, Idaho: U.S. Geol. Survey Prof. Paper 596, 52 p. Mannerfelt, CM., 1945, Nagra glaciomorfologiska formelement och deras vittnesbbrd om inlandsisens avsm^ltningsmekanik i svensk och norsk fjallterrang: Geog. Annaler, v. 27, p. 1-239. Marcus, M.G., 1960, Periodic drainage of glacier-dammed Tulsequah Lake, British Columbia: Geog. Rev., v. 50, p. 89-106. Mathews, W.H., 1965, Two self-dumping ice-dammed lakes in British Columbia: Geog. Rev., v. 55, p. 46-52. 1973, The record of two jbkulhlaups: Internat. Assoc. Sci. Hydrology, Symposium on the hydrology of glaciers, Cambridge, September 1969 (in press). Moiola, R.J., and Weiser, D., 1968, Textural parameters: an evaluation: Jour. Sed. Petrology, v. 38, p. 45-53. Pardee, J.T., 1942, Unusual currents in Glacial Lake Missoula, Montana: Geol. Soc. America Bu l l . , v. 53, p. 1569-1600. Rist, Sigurjon, 1955, Skeioararhlaup 1954 (the hlaup of Skeifrara 1954): Jbkull, v. 5, p. 30-36. Schumm, S.A., 1968, River adjustment to altered hydrologic regimen— Murrumbidgee River and paleochannels, Australia: U.S. Geol. Survey Prof. Paper 598, 65 p. Shields, A., 1936, Anwendung der Aehnlichkeitsmechanik und der Turbulenzforschung auf die Geschiebebewegung: Preuss. Versuchsanst. fur Wasserbau und Schiffbau Mitt. 26. Stone, K.H., 1963, The annual emptying of Lake George, Alaska: Arctic, v. 16, p. 26-40. Strom, K.M., 1938, The catastrophic emptying of a glacier-dammed lake in Norway 1937: Geologie der Meere und Binnengewasser, v, 2, p. 443-444. Thorarinsson, Sigurdur, 1939, The ice-dammed lakes of Iceland with particular reference to their values as indicators of glacier oscillations: Geog. Annaler, v. 21, p. 216-242. 1953, Some new aspects of the Grimsvbtn problem: Jour. Glaciology, v. 2, p. 267-275. 214 U.S. Geological Survey, 1969, The breakout of Lake George: U.S. Geol. Survey non-technical leaflet, 15 p. Whalley, W.B,, 1971, Observations of the drainage of an ice-dammed lake— Strupvatnet, Troms, Norway: Norsk Geog, Tidsskr., v. 25, p. 165-174. Williams, P.F., and Rust, B.R., 1969, The sedimentology of a braided river: Jour. Sed, Petrology, v. 39, p, 649-679. 215 CHAPTER FIVE: GEOMORPHIC HISTORY Geophysical evidence (p. 100) indicates that the Rocky Mountain Trench i s a structural boundary of continental proportions, possibly marking a Paleozoic hinge zone along the margin of the Precambrian craton. Evolution of the southern Trench as a major physio-graphic feature began after a period of late Cretaceous and early Tertiary compressional deformation which produced the Rocky Mountain fold and thrust belt. This was followed in the Eocene by a period of isostatic u p l i f t and extension accompanied by block faulting in the Cordillera, The change in tectonics may be due to the cessation of subduction along the northern Pacific margin and i t s replacement by lateral displacement of plates along transform faults. The southern Rocky Mountain Trench formed by block faulting of the half-graben style. Structural basins along the axis of the Trench served as the depositional sites for elastics eroded from the adjacent uplands. Although the age of the oldest sediments is not directly known, sediments of Upper Eocene and Oligocene age occur in the structurally comparable Flathead Valley to the east. Sedimentation continued into the Miocene with deposition of the St. Eugene Formation, which includes the uppermost part of as much as 1400 m of Tertiary strata. The St. Eugene Formation consists of flood-plain and a l l u v i a l fan facies; the former includes both high-energy river gravel deposited off major tributary valleys and shallow-lake or slack-water s i l t and sand; the latter consists of talus and fanglomerate deposited along the margins of the Rocky Mountain Trench and derived from adjacent fault-bounded uplands. The Miocene climate 216 was more temperate than the present climate of the area; summer precipi-tation was abundant, and winters were moist and mild. Deposition of sediment in the Trench and erosion of the uplands were probably contempo-r raneous with faulting, although 600 m or more of normal dip'-slip displace-ment along the east margin of the Trench occurred after deposition of the St. Eugene Formation. There is a hiatus in the geologic record of the southern Rocky Mountain Trench between deposition of the St. Eugene Formation and the Wisconsinan Glaciation. Although southeastern British Columbia is not now a seismically active area, block faulting may have continued with decreasing intensity perhaps into the Quaternary. Young fault scarps in western Montana and a belt of recent earthquakes parallel to the northern Rocky Mountains in the United States indicate that block faulting is s t i l l occurring south of 49°N. It is possible, then, that additional strata of similar character to the St. Eugene Formation accumulated in the southern Rocky Mountain Trench during late Tertiary and early Quaternary time. If so, these strata either have been removed by Quaternary glacial and fluvial erosion or are concealed beneath Quaternary deposits, or both. It is noteworthy that outcrops of St. Eugene strata are few as a result of both these factors; strata are exposed largely on the protected down-glacier flanks of bedrock highs and along the few deep postglacial valleys where rivers have cut through the Quaternary section. Alternatively, i t is possible that significant sediment accumu-lation on the floor of the Trench ceased after deposition of the St. Eugene Formation, although this is considered unlikely because of major post-St. Eugene block faulting. However, if the structural basins were 217 already filled, and i f there was no rise in local base level, additional sediment eroded from the surrounding uplands might have been transported out of the study area. It is believed that the presence or absence of Pliocene or early Quaternary sediments will be determined only by drilling the sediment f i l l underlying the floor of the Rocky Mountain Trench. Whatever the extent of Pliocene and early Quaternary sedimenta-tion, the St. Eugene Formation was gently deformed and eroded prior to the Wisconsinan Glaciation. Tectonic deformation of Wisconsinan glacial deposits was not observed. During each glaciation, the eastern Cordilleran Ice Sheet formed through the coalescence of large valley and piedmont glaciers in the Rocky and Columbia Mountains and flowed south through outlet valleys such as the Rocky Mountain Trench and the Purcell Trench to the west (Fig. 57). The source areas of the ice sheet were probably the high alpine areas which presently support glaciers (Fig. 58). The glacier flowing through the Rocky Mountain Trench was augmented by tributary valley glaciers in southern British Columbia and Montana, and reached south of 47°30'N at the peak of glaciation (Fig. 59). Near the maximum of each advance, the glacier flowing south through the Purcell Trench dammed the Clark Fork River at Lake Pend Oreille to form Glacial Lake Missoula. Multiple glaciation in British Columbia during the Pleistocene Epoch is well documented (for example, see Armstrong and Tipper, 1948; Armstrong and others, 1965; Fulton, 1972; Fulton and Halstead, 1972), but the stratigraphic record for pre-Wisconsinan glaciations is poor and is 218 largely known from d r i l l records or by comparison with the more complete glacial record in Washington, Idaho, and Montana. Wisconsinan glaciers covered nearly a l l of British Columbia and removed or buried most of the record of earlier glaciations. Deeply oxidized d r i f t of one pre-Bull Lake (pre^-Wisconsinan) glaciation has been identified in northwestern Montana (Richmond and others, 1965, p. 234), but correlative deposits were not found in the southern Rocky Mountain Trench in British Columbia. Three t i l l s representing stades of the Bull Lake (early Wisconsinan) Glaciation are separated by lacustrine deposits of Glacial Lake Missoula south of Flathead Lake in northwestern Montana (Richmond and others, 1965). The terminus of the late stade is the Mission moraine (Alden, 1953). In this area, glaciers advanced farther south during the Bull Lake Glaciation than during the Pinedale (late Wisconsinan) Glacia-tion (Fig. 59). However, no deposits of early Wisconsinan age have been identified in the southern Rocky Mountain Trench. The Bull Lake Glaciation was followed by a major nonglacial interval, referred to as the Olympia Interglaciation in the Pacific Northwest. Fulton (1971), in a paper on the radiocarbon geochronology of southern British Columbia, has concluded that the Olympia Interglacia-tion began more than 52,000 years ago and ended about 19,000 years ago, A nonglacial sequence in the Purcell Trench spans the interval from 43,800 ± 800 to 25,840 ± 320 years B.P. Three major depositional cycles within the Olympia Interglaciation have been identified—early and late periods of aggradation when base level was higher than at present, and an 219 Figure 57. Pattern of ice flow in southeastern British Columbia and southwestern Alberta during the Pleistocene. The southern Rocky Mountain Trench was one of several outlet valleys of the Cordilleran Ice Sheet in southern British Columbia (from Prest and others, 1967). 220 intervening time of low base level characterized by stream erosion with continued sedimentation in lake basins (Fulton, 1968). During the Olympia Interglaciation, as at present, the major sites of sedimentation were valleys located between h i l l y or mountainous upland areas. Channel, overbank, and a l l u v i a l fan sediments accumulated on the valley bottoms, and lacustrine and deltaic sediments were deposited in lake basins. The climate in the Interior during at least part of this period was sufficiently warm and moist to support forests and large vertebrates (Fulton, 1971, p. 5). Olympia Interglacial deposits in the southern Rocky Mountain Trench include channel, overbank, and, probably, lacustrine sediments. Wood from these sediments yielded a radiocarbon age of 26,800 ± 1000 years (GX-2032). This nonglacial interval terminated with the invasion of the Rocky Mountain Trench by ice at the onset of the Pinedale or Fraser Glaciation. As in earlier advances, the trunk glacier was augmented by local alpine glaciers. Drainage was diverted along the margins of the advancing ice sheet. Eventually, ice covered the Trench as well as the flanking mountains, with only the higher peaks projecting above the glacier surface (Fig. 60). Erratics and striae on the east side of the Trench just north of Bull River indicate the trunk glacier reached an elevation of at least 2260 m there. Daly (1912) concluded that the elevation of the ice surface in the Galton Range and Purcell Mountains near the International Boundary was 2230 m. The Cordilleran Ice Sheet was thus about 1500 m thick over the Trench near the 49th p a r a l l e l . It 221 Figure 58. Distribution of existing glaciers in southeastern British Columbia and southwestern Alberta (from Prest and others, 1967). 222 terminated in a lobate front marginal to Glacial Lake Missoula near the present Flathead Lake (Fig. 59). Three Pinedale stades are recognized in northwestern Montana (Richmond, 1965; Richmond and others, 1965), The Poison moraine (Alden, 1953) south of Flathead Lake is a compound moraine consisting of drift of the early and middle stades. Moraines at Kalispell and Coram were deposited by late Pinedale glaciers. In the southern Rocky Mountain Trench in British Columbia, deposits of three late Wisconsinan ice advances are recognized and are tentatively correlated with the three stades of the Pinedale Glaciation. At the beginning of the early stade, ice advanced over weathered bedrock and sediments on the Trench floor; weathered clasts of various units were incorporated in the basal t i l l of this advance. At its maximum advance the trunk glacier reached a terminal position at the Poison moraine. As the early stade ended, portions of the glacier in the Trench apparently became stagnant and were buried beneath thick accumula-tions of glaciofluvial sand and gravel along routes of meltwater discharge and beneath glaciolacustrine s i l t and sand elsewhere. The glaciolacustrine sediments are in part contemporaneous with and in part younger than the coarse outwash. With melting of the underlying stagnant ice blocks, these sediments were broadly folded and faulted. It is not known whether the fine-grained sediments accumulated in one or a number of lakes occupying the floor of the Trench; neither is the cause of impoundage known, although possibilities include ice or sediment dams and differential isostatic depression of the surface of the Trench in a northerly to 223 c M CO 4-1 OJ T J td C cu n) Xi +J T J 3 C o Q co 6 Xi C O • r l - H Pi CO OJ g r l O 3 r l 4 J LH Cd w cu M-l O Xi T J cd OJ T j C H CU Xi 4 J l-l o cu u T j c CU TJ cu d •C cd cn CD cd o c M cd cd cu r H U CU •r l T J r l O C J CU 60 r l 4 J O C r l <4-l - H O •> CU Cd , G 3 1 -3 io r H \ 0 • o cn CTI U r H m xi -CU CO CO r l "H U 3 4 J CU 6 0 - r l • H H * J Fu PQ O mm*) i« 7»J . , T 114°W f:..< 5?/ bilk * w T5o Pe-> PEND OREILLE : -GLACIAL LAKE COEUR D'ALENE V 2 GLACIAL LAKE ! MISSOULA :: N *o ^ 20 40 KILOMETERS - J V 1156W t-"j>I ' GLACIERS • PINEDALE GLACIATION (LATE, MIDDLE, EARLY STADES) t S ' ^ ' t S * ' ' BULL LAKE GLACIATION (LATE, EARLY STADES) (sooo) ft PLEISTOCENE GLACIAL LAKES AREAS NOT COVERED BY CORDILLERAN ICE SHEET OR GLACIAL LAKES (LOCAL ALPINE GLACIERS NOT SHOWN) NUNATAK AREAS AREAS ABOVE CORDILLERAN ICE SHEET - MAXIMUM PINEDALE ADVANCE (IN BRITISH COLUMBIA AREA ABOVE 2440m ALTITUDE IS SHOWN) CONTOURS ON SURFACE OF CORDILLERAN ICE SHEET - MAXIMUM PINEDALE ADVANCE 224 northwesterly direction due to the presence of an ice cover to the north. Ice-rafted clasts were deposited with the fine sediments early and late in the interstade, Glaciolacustrine deposition ended when glaciers of the middle stade advanced down the Trench. Once again, the maximum advance of the trunk glacier was to the Poison moraine. Recession from the Poison moraine may correlate with an interval of ice-marginal glaciolacustrine deposition along the sides of the Trench and subglacial or proglacial glaciofluvial sedimentation in the center of the Trench. It is thought that ice locally remained in the center of the Trench throughout the interstade between the middle and late stades. During the late stade, ice advanced as far south as Kalispell (Fig. 59). During final deglaciation of southeastern British Columbia, the glacier in the Rocky Mountain Trench became increasingly confined by the flanking mountains. Spillover of ice from the Trench into the parallel-trending Flathead and Bull Valleys and from Elk Valley into the Trench which occurred during glacial maxima ceased as the uplands appeared through the ice cover (Fig. 61). As in other areas of southern British Columbia, ice disappeared from the mountainous uplands before the adjacent valleys were deglaciated (Dawson, 1889; Nasmith, 1962; Fulton, 1967, 1969), However, the shrinking glacier in the southern Rocky Mountain Trench remained active throughout deglaciation and did not stagnate as did the early Pinedale glacier. The presence of drumlinized ground moraine and the absence of end moraines and major ice stagnation features such as kettle and kame topography and eskers indicate that the trunk glacier receded by downwasting and orderly frontal retreat without major glacial standstills (Fig. 62). Tributary glaciers retreated far into their Figure 60. Comparison of a contemporary ice sheet and the Cordilleran Ice Sheet during the maximum Pinedale advance. A. Antarctic Ice Sheet (U.S. Geological Survey photo number TMA 892 F33). B. Maximum extent of the Cordilleran Ice Sheet over part of southeastern British Columbia. Nunatak areas are black and glacier-covered areas white. 226 Figure 61. A contemporary analogue of the Pleistocene glacierized Rocky Mountain Trench. A. The glacier i s confined to a wide, linear valley by flanking mountains, much as was the Trench glacier immediately before and after maximum glacier advances (Bagley Ice Field, Alaska; photo by Austin Post). B. The southern Rocky Mountain Trench and flanking mountains viewed toward the south from near 49°45'N (BC 896:106). 227 Figure 61 228 valleys prior to the disappearance of the trunk glacier. The latter blocked the side valleys, thus impounding lakes in which clay, s i l t , sand, and deltaic gravel accumulated (Fig. 63). Drainage from these lakes and from the receding trunk glacier was south beneath and along the margins of the glacier (Glacial Lake Elk at an early stage overflowed to the east via Crowsnest Pass). Periodically, glacier-rdammed lakes drained rapidly to produce jbkulhlaups, and many of the large meltwater channels on the floor of the Rocky Mountain Trench were formed or modified by these floods. As the ice surface was lowered, ice-marginal channels at lower elevations were progressively occupied and abandoned. The courses of major streams entering the Trench reflect this phase of deglaciation. Streams such as Gold Creek and Elk River swing abruptly to the southeast along the Trench margins before crossing the valley to join Kootenay River. These streams were forced southeast by ice covering the Trench floor and became entrenched in their courses before ice disappeared from the valley bottom. Some streams, however, failed to entrench along southeast courses. Sand Creek, for example, flowed along the Trench margin to near Elko, but, as the trunk glacier receded, this ice-marginal route was abandoned for a more direct westerly course. Recessional outwash on the floor of the Rocky Mountain Trench includes kame terrace gravel, and sand and gravel underlying ice-frontal meltwater channels carved across the t i l l plain. The sequence of late glacial events occurred at different times along the length of the Trench. For example, while glaciofluvial gravel was being deposited near 49°N, t i l l was being lodged on the Trench floor farther north. Limiting radiocarbon dates for deglaciation are not Figure 62. Comparison of contemporary and ancient landforms produced by glaciers which remained active during recession rather than stagnating. A. Drumlinized ground moraine uncovered during the recent retreat of Muir Glacier in Alaska (photo by Austin Post). B. Drumlinized ground moraine left during the retreat of the late Pinedale glacier from the southern Rocky Mountain Trench in British Columbia (BC 5298-261). The flow direction is from the labelled corner of each photo. Figure 63. A contemporary analogue of Pleistocene ice-dammed lakes in the southern Rocky Mountain Trench. A. Lake Tulsequah, British Columbia shortly after i t emptied during a jbkulhlaup. The lake is impounded by a lobe of Tulsequah Glacier (photo by Austin Post). B. Pleistocene Glacial Lake Elk dammed by a lobe of ice flowing up Elk Valley from the Rocky Mountain Trench. The lake periodically drained during jbkulhlaups. Darkened areas are above an elevation of 1830 ra (6000'). 232 available for the study area. However, a bog-bottom date of 10,000 ± 140 years from the Rocky Mountain Trench at latitude 51°29'N (GSC-1457; Fulton, 1971), provides a minimum deglaciation date. The oldest postglacial date from southeastern British Columbia is 11,000 ± 180 (GSC-9095 Fulton, 1971), which was obtained from freshwater marl in the Columbia River Valley at latitude 49°30'N. During the Holocene, rivers entrenched the unconsolidated sedi-ment mantle on the Trench fl o o r , in places carving canyons in bedrock. Much of this downcutting occurred within a few thousand years after deglaciation, as evidenced by kettles present on some lower river terraces and by Mazama 0 tephra (age 6600 years B.P.) interstratified with mudflow gravels deposited on the present Kootenay River flood plain. Holocene deposition is limited to (1) small talus cones and a l l u v i a l fans along the margins of the Trench and river valleys, (2) a thin mantle of loess, (3) small sand dunes, and (4) flood-plain sediments in the major river valleys. Most of these deposits probably formed by the reworking of d r i f t during a short span of time immediately following deglaciation. 233 LITERATURE CITED Alden, W.C., 1953, Physiography and glacial geology of western Montana and adjacent areas: U.S. Geol. Survey Prof. Paper 231, 200 p. Armstrong, J.E., Crandell, D.R., Easterbrook, D.J.,,and Noble, J.B., 1965, Late Pleistocene stratigraphy and chronology in southwestern British Columbia and northwestern Washington: Geol. Soc. America Bu l l . , v. 76, p. 321-330. Armstrong, J.E., and Tipper, H.W., 1948, Glaciation in north-central British Columbia: Am. Jour. Sci., v. 246, p. 283-310. Daly, R.A., 1912, Geology of the North American Cordillera at the forty-ninth parallel: Canada Geol. Survey Mem. 38, 857 p. Dawson, G.M., 1889, Glaciation of high points in the southern interior of British Columbia: Geol. Mag., decade 3, v. 6, p. 350-352. Fulton, R.J., 1967, Deglaciation studies in Kamloops region, an area of moderate r e l i e f , British Columbia: Canada Geol. Survey Bu l l . 154, 36 p. 1968, Olympia Interglaciation, Purcell Trench, British Columbia: Geol. Soc. America Bu l l . , v. 79, p. 1075-1080. 1969, Glacial lake history, southern Interior Plateau, British Columbia: Canada Geol. Survey Paper 69-37, 14 p. 1971, Radiocarbon geochronology of southern British Columbia: Canada Geol. Survey Paper 71-37, 28 p. 1972, Stratigraphy of unconsolidated f i l l and Quaternary develop-ment of North Okanagan Valley: Canada Geol. Survey Paper 72-8, pt. B, p. 9-17. Fulton, R.J., and Halstead, E.C., 1972, Quaternary geology of the southern Canadian Cordillera: Guidebook, Field Excursion A02, 24th Internat. Geol. Cong., Montreal, 49 p. Nasmith, Hugh, 1962, Late glacial history and s u r f i c i a l deposits of the Okanagan Valley, British Columbia: British Columbia Dept. Mines and Petroleum Resources Bu l l . 46, 46 p. Prest, V.K., Grant, D.R., and Rampton, V.N., 1967, Glacial map of Canada: Canada Geol. Survey Map 1253A. Richmond, G.M., 1965, Glaciation of the Rocky Mountains, in Wright, H.E., Jr . , and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 217-230. 234 Richmond, G.M., Fryxell, Roald, Neff, G.E., and Weis, P.L., 1965, The Cordilleran Ice Sheet of the Northern Rocky Mountains, and related Quaternary history of the Columbia Plateau, in Wright, H.E., Jr., and Frey, D.G., eds., The Quaternary of the United States: Princeton, New Jersey, Princeton Univ. Press, p. 231-242. 235 CHAPTER SIX: APPLICATIONS OF GEOLOGIC KNOWLEDGE S u r f i c i a l geology studies in the southern Rocky Mountain Trench have provided information on the origin and evolution of the Trench, on the applicability of certain procedures of t i l l investigation in mountain-ous regions, on t i l l genesis, and on paleohydrologic determinations in the glaciofluvial environment. These sc i e n t i f i c contributions are outlined in Chapters 2, 3, and 4. Additional geologic information relevant to man's activities in and outside the study area is presented in this chapter. General applications of geologic knowledge outside the southern Rocky Mountain Trench include the prediction of flood magnitude from self-dumping glacier-dammed lakes and the determination of the source of ore clasts or minerals in t i l l . The groundwater resources of the study area are assessed in the f i n a l section of this chapter. MAGNITUDE OF JOKULHLAUPS1 Catastrophic floods from glacier-dammed lakes (jOkulhlaups) have resulted in l i f e and property losses in such areas as Iceland and Alaska. As glacierized areas are increasingly populated, i t becomes ever more important to determine the hazards posed by specific glacier-dammed lakes. Jbkulhlaup prediction w i l l be possible only after an understanding i s gained of the causes of these floods and the factors ^Discussion adapted from Clague, J.J., and Mathews, W.H., 1973, The magnitude of jSkulhlaups: Jour. Glaciology (in press). 236 affecting their magnitude. Analyses of the hydrographs for two successive jbkulhlaups from glacier-dammed Summit Lake, British Columbia (Mathews, in press) show that instantaneous water discharge, £>t, is related not to time, t, since the start of each flood but to the volume of water, Vt, released from the lake during this time. For a l l but the i n i t i a l stages of the flood, discharge can be expressed by a formula of the form; Qt = K(Vt)b (1) (in which for Summit Lake K - 0.72 and b = 1.5 i f Qt is expressed in m3/sec and Vt in m3 x 106) . Equations of this same form apply to jbkulhlaups from five other ice-dammed lakes, although the coefficient and exponent differ for each lake (Table 8). Hydrographs for these floods, based on calculated or measured discharges plotted against cumulative volume lost , instead of time, are shown in Figure 64. Values of K and b for the six examples show large variations, with extreme values found at two British Columbia lakes, Summit and Tulsequah. Possible factors influencing these values include: the head loss, H, between the high water mark in the reservoir and the toe of the ice dam where the escaping water generally emerges; the distance, L, from reservoir to point of emergence; the depth, Z>, of the reservoir at the dam; and the capacity, V^^, of the reservoir (Table 8), A plot °f vmax against the ratio H/L for the data of Table 8 indicates, as one might expect, that small reservoirs are in general impounded by dams with large height-to-length ratios. Likewise, K displays a negative relation-237 ship to VMAX, albeit a weak one. For a small dam the passage of 10° nr of water [for this volume, QT = K in Eq, (1)] can be expected to generate a larger leak, a higher discharge, and hence a higher value of K than is the case for larger ice dams with generally smaller values of H/L. Thus, the size of the dam may control both the available reservoir storage and the coefficient of Eq. (1); accordingly, K and VMAX are not wholly independent of one another. However, the interrelationships of VMAX, K, and H/L a l l show large scatter, far beyond that of the individual points on the rising curves of the hydrographs, and none of the other factors investigated can be clearly associated with variations in values of K and b. In Figure 64 the peak discharge, Qmax* occurring at the end of each jbkulhlaup is designated by a dot. Curiously, a l l peak discharge values f a l l close to a common line despite the variations in slope and position of the individual hydrographs. Peak discharges for floods from four additional ice-dammed lakes are presented in Table 8, and Figure 56 shows the resultant relationship between maximum flood discharge (in m3/sec) and available water storage (in m3 x 106) for the ten lakes. The data points cluster about a line represented by: Qmax = 7 5 ( T /r o a x)° - 6 7 (2) (r 2 = 0.96) This relationship is a remarkably good one considering that the data include peak discharges measured at varying distances from the toes of ice dams and are derived from lakes of widely differing size and other 238 characteristics. More data from other jbkulhlaups and from other ice-dammed lakes are clearly needed to clar i f y and explain the relationships, but in the meantime the log-log plot of Qt against Vt is a useful tool for investi-gating individual jbkulhlaups, and Eq. (2) offers an empirical basis for estimating possible maximum discharges from self-dumping ice-dammed lakes. These two expressions may thus be helpful in evaluating hazards to l i f e and property posed by ice-dammed lakes. MINERAL EXPLORATION Many ore deposits in North America and Europe have been scoured beneath glaciers and buried by d r i f t . An increasingly important tool in locating such ore deposits is the tracing of ore clasts and minerals in dr i f t back to their bedrock sources (e.g., Dreimanis, 1956; Lee, 1965). The down-glacier fan-shaped pattern of many boulder trains helps to locate bedrock sources. The observed exponential decrease of t i l l constituents with distance from point sources (p. 155) may also be of use in mineral exploration. In the case of a mineral or rock type eroded from a restricted bedrock source and transported near the base of a glacier, the exponential decrease with distance may be due to one or a combination of the following: progressive deposition, breakage and abrasion, dilution. Where dilution by locally derived sediment has been negligible, the decrease of a constituent with distance can be expressed by: Nx = N0e~kx (3) Table 8. JOkulhlaup data: K and b, coefficient and exponent of Eq. (1), p. 236; #> height of lake surface above toe of ice dam; L, distance from lake to toe of dam; D, depth of lake at ice dam; Vmax, reservoir storage; QmaXy maximum instantaneous water discharge. The jbkulhlaups are historic except from Lake Missoula which is of late Pleistocene age. Floods resulting from volcanic activity are excluded. Some of the data were compiled from topographic maps and air photographs. Lake Year K b H (m) L (km) H/L D (•) "max (106m3) Qmax (m3/sec) Reference Strupvatnet, Norway 1969 88 0.84 186 1 0.19 29* 2.6** 150 Whalley, 1971 Ekalugad V a l l e y , B a f f i n Island 1967 46 0.91 120 2 0.06 120 4.8 200 Church, 1972 Demmevatn 1937 — • ~ 406 0.14 79 11.6 1,000 Strom, 1938 Gjanupsvatn, Iceland 1951 30+ 0.72+ 167 5 0.03 20t T 20 370 Arnborg, 1955 Vatnsdalur, Iceland 1898 ~ -- 372 10 0.04 188 120 3,000 Thorarinsson, 1939 Tulsequah Lake, B r i t i s h Columbia 1958 150 0.49 210 8 0.03 73 229 1,556 Marcus, 1960 Summit Lake, B r i t i s h Columbia 1965, 1967 0.72 1.5 620 12 0.05 200 251 3,260 Mathews, 1965 Graenalon, Iceland 1939 24 0.77 535 19 0.03 230 1,500 5,000 Thorarinsson, 1939 Lake George, Alaskattt 1958 — — 40 9 0.004 40 1,730 10,100 Stone, 1963 Lake Missoula, Montana Pleistocene — — 640 — — 610 2 x 106 1.87 x IO6 5 Bretz, 1925; Pardee, 1942 *Depth to bedrock knob l i m i t i n g magnitude of JOkulhlaup = 13 m. **Volume stored i n lake = 4.6 x 106 m3; water released i n JOkulhlaup = 2.6 x 106 m3. ***Very approximate. t C o e f f i c i e n t and exponent determined from hydrograph of JOkulhlaup of June 1951; other data pertain to JOkulhlaup of October 1951. T+Lowering of lake surface during JOkulhlaup. tttDrains at surface along ice margin. ^max estimated for flood wave at Wallula Gap i n southeastern Washington. 240 (where N is the amount of the constituent at the glacier base or in t i l l at a distance, x, from the source; NQ is the amount at x = 0; and Jc is a constant reflecting the rate as a function of distance at which deposition or breakage occurs). Thus, knowing NQ and k, and determining N at two or more points along the glacier-flow path, one can calculate x, which serves to approximately locate the ore deposit. The constants, NQ and ky may be estimated from the frequency distribution of other t i l l constit-uents of known source. If breakage i s the factor controlling the dis-tribution, however, constituents of similar durability should be compared, and the "terminal grades" of minerals ( i . e , , minimum grain size below which breakage i s rare; Dreimanis and Vagners, 1971) taken into account. Also, with increasing distance of transport, sediment may be transferred from a subglacial to an englacial position, thus becoming unavailable for immediate deposition. Since the above relationship i s based on the assumption that transport occurs near the base of the glacier, Eq. (3) may not be applicable to the frequency distribution of a constituent far from i t s source. Nevertheless, the central part of the distribution should yield reliable estimates of distance to source. GROUNDWATER Annual precipitation on the floor of the southern Rocky Mountain Trench is less than water losses due to evapotranspiration, but there is a continual influx of surface water and groundwater from the adjacent water-positive mountainous areas. The water table is high throughout the area. Kettle lakes fluctuate in size seasonally but rarely dry up; springs and seepage occur along the walls of the major river valleys. VOLUME (m3) Figure 64. Relation of cumulative volume drained during jbkulhlaups and instantaneous water discharge. Equations are of the form Qt = iC(vt)^, in. which the coefficient, K, and exponent, for each flood are listed in Table 8. The peak discharge for each jbkulhlaup is indicated by dot. E—Ekalugad Valley; Gj—Gjanupsvatn; Gr—Graenalon; St—Strupvatnet; Su—Summit Lake; T—Tulsequah Lake. 242 The water requirements of the rural population are satisfied from shallow wells and by direct extraction from rivers, streams, and lakes. Cranbrook, however, is using large amounts of groundwater from uncon-solidated deposits. The large volume of semiconsolidated and unconsolidated sedi-ments in the Trench may represent a sizable groundwater reservoir. Although the character of sediments beneath the St. Eugene Formation is not known, these sediments f i l l structural basins up to 1500 m deep. Most St. Eugene sediments and Pleistocene clay, s i l t , and diamictdn have low permeabilities and probably form partial barriers to the movement of groundwater. Sand and gravel exposed in the valley walls along Kootenay River may be the uppermost sediments of a permeable f i l l underlying the center of the Rocky Mountain Trench. Outside this central zone, permeable units of Pleistocene age are relatively thin and discontinuous. Older outwash, although probably not an important aquifer, is permeable and may thus affect groundwater movement. If there i s channel continuity beneath Wycliffe t i l l , permeable proglacial outwash of the younger drift may transmit abundant groundwater. Further information on the distribution and specific yields of these potential aquifers can be obtained through a limited program of d r i l l i n g and pumping tests. The most accessible source of groundwater, however, i s late Wisconsinan outwash gravel underlying the major melt-water channel s on the Trench floor (Fig. 5). The gravel is extremely permeable and, in general, is underlain by relatively impermeable clay, s i l t , and diamicton. 243 Figure 65. Meltwater channels in the vicinity of Cranbrook, British Columbia. The large south- to southwest-trending channel is underlain in part by water-bearing glaciofluvial gravel. Two production wells (darkened circles) have a combined maximum design capacity of 1500 U.S. gpm (0.095 m3/sec), and the channel transmits a natural flow twice as large (Brown and others, 1970). 244 Figure 66. Meltwater channel complex northwest of Bull River (top); mean water discharge records of Norbury Creek (1959-1970) and Bull River (1914-1970) (bottom). Norbury Creek is fed by groundwater from the channel complex. Gauging stations on Norbury Creek and Bull River are indicated by triangles. Geology in part from Kelley and Sprout (1956) and Leech (1960). 245 Figure 66 246 The City of Cranbrook is extracting groundwater from the gravel of one large meltwater channel (Fig. 65). Two production wells are rated at a combined maximum design capacity of 1500 U.S. gal/min (0.095 m3/sec), and i t has been estimated that the channel in which the wells are founded carries a natural flow twice as large (Brown and others, 1970). Permeability of the aquifer gravel is very high; from coefficients of transmissibility reported by Brown and McNichol (1969), hydraulic conductivity at the two wells is determined to be about 10,000 and 100,000 gal/day/ft2 (0.0004 and 0.004 m3/sec/m2). The groundwater contains between 400 and 500 ppm total dissolved solids of which the bulk is Ca^ and HC03~, with lesser amounts of Mg"1-1" and SO^-. Many other late glacial meltwater channels represent potential groundwater sources. An example is the north-south trending channel located on the east side of the Trench north of Bull River (Figs. 5 and 66). This channel is recharged by abundant surface flow; the safe yield can be estimated from the mean discharge of the groundwater-fed stream draining the channel—about 1.0 m3/sec. A comparison of the hydrographs of this stream (Norbury Creek) and rivers fed by surface runoff (e.g., Bull River) (Fig. 66) shows: (1) the more uniform dis-charge of Norbury Creek, with Bull River characterized by high late spring and summer maxima and low base flows during the remainder of the year; (2) the occurrence of the Norbury Creek peak discharge about one to two months after the Bull River peak. This difference represents the average transit time of groundwater through the outwash gravel of the meltwater channel. 247 LITERATURE CITED Arnborg, Lermart, 1955, Hydrology of the glacial river Austurfljot: Geog, Annaler, v. 37, p, 185-201.. Bretz, J.H., 1925, The Spokane flood beyond the channeled scablands: Jour, Geology, v. 33, p. 97-115, 236-259, Brown, W.L., and McNichol, R.T., 1969, Groundwater development, the Corporation of the City of Cranbrook: Report by Robinson, Roberts, and Brown Ltd. for Associated Engineering Services Ltd., Vancouver, British Columbia, 11 p. Brown, W.L., McNichol,. R.T. , and Dakin, R.A. , 1970, Production wells, the Corporation of the City of Cranbrook, B.C.: Completion report by Robinson, Roberts, and Brown Ltd. for Associated Engineering Services Ltd., Vancouver, British Columbia, 9 p. Church, M.A., 1972, Baffin Island sandurs: a study of Arctic fluvial processes: Canada Geol. Survey Bull. 216, 208 p. Dreimanis, Aleksis, 1956, Steep Rock iron ore boulder train: Geol. Assoc. Canada Proc, v. 8, p. 27-70. Dreimanis, Aleksis, and Vagners, U.J., 1971, Bimodal distribution of rock and mineral fragments in basal t i l l s , in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 237-250. Kelley, C.C., and Sprout, P.N., 1956, Soil survey of the upper Kootenay and Elk River valleys in the East Kootenay district of British Columbia: British Columbia Soil Survey Rept. 5, 99 p. Lee, H.A., 1965, Investigation of eskers for mineral exploration: Canada Geol. Survey Paper 65-14, pt. 1, 17 p. Leech, G.B., I960, Fernie, west half, British Columbia: Canada Geol. Survey Map 11-1960. Marcus, M.G., 1960, Periodic drainage of glacier-dammed Tulsequah Lake, British Columbia: Geog. Rev., v. 50, p. 89-106. Mathews, W.H., 1965, Two self-dumping ice-dammed lakes in British Columbia: Geog. Rev., v. 55, p. 46-52. 1973, The record of two jokulhlaups: Internat. Assoc. Sci. Hydrology, Symposium on the hydrology of glaciers, Cambridge, September 1969 (in press) . Pardee, JIT., 1942, Unusual currents in Glacial Lake Missoula, Montana: Geol. Soc. America Bull., v. 53, p. 1569-1600. 248 Stone, K.H., 1963, The annual emptying of Lake George, Alaska; Arctic, v. 16, p. 26-40. StrOm, K.M., 1938, The catastrophic emptying of a glacier^dammed lake in Norway 1937: Geologie der Meere und BinnengewSsser, v, 2, p. 443-444. Thorarinsson, Sigurdur, 1939, The ice-dammed lakes of Iceland with partis cular reference to their values as indicators of glacier oscillations: Geog. Annaler, v. 21, p. 216-242. Whalley, W.B., 1971, Observations of the drainage of an ice-dammed lake— Strupvatnet, Troms, Norway: Norsk Geog. Tidsskr., v, 25, p. 165-174. 249 CHAPTER SEVEN: SUGGESTIONS FOR FURTHER RESEARCH Research projects often produce more questions than answers. Such questions, however, invite additional research which expands the basic body of scientific knowledge. In effect, advances in science result from individuals extending and modifying the conclusions of their colleagues. The following are some suggestions for further research based upon my studies in the Rocky Mountain Trench. TILL FABRICS AND TILL GENESIS Boulton (1971) has discussed the relationships that exist between till-fabric elements and t i l l genesis (p. 118, Table 4). In an extension of Boulton's work, I have interpreted the depositional history of Pleistocene t i l l from till-fabric data. This method of determining the genesis of Pleistocene t i l l might be further tested in another mountainous area where the direction of glacier flow is known and where there were no coalescent tributary glaciers which might have produced complex t i l l fabrics. In such an area the orientations of blade-shaped clasts in t i l l should yield information on the transportational and depositional history of the sediment (see Table 4). 250 ORIGINS OF DRUMLINS Drumlins have been extensively studied, but their origin is s t i l l debated (for a review and references, see Gravenor, 1953; Smalley and Unwin, 1968). There are, however, two main hypotheses as to how drumlins form: by subglacial deposition and by subglacial erosion of pre-existent materials. The dynamics of drumlin formation might be understood better i f drumlin fabrics were determined. Since t i l l fabrics may provide information both on t i l l genesis and the direction of ice movement during deposition, fabric data collected at a number of sites on a drumlin might indicate whether or not t i l l deposition and drumlin formation were contemporaneous. If the drumlin formed by the erosion of pre-existent t i l l , fabrics might be unrelated to drumlin morphometry. COLLECTION OF FABRIC DATA One major impediment to the use of fabric techniques in sedi-ment investigations is the large amount of time required for data collec-tion and analysis (for example, several hours are commonly required to measure and process a single fabric sample of 50 clasts). Although more rapid macroscopic techniques (Dreimanis, 1959) and microscopic methods (Dapples and Rominger, 1945) have been devised, a l l require that the orientations of individual grains or clasts be determined. Because elongate grains and clasts in sediments with directed fabrics are nonrandomly oriented, these sediments are anisotropic in terms of prop-erties such as thermal conductivity, r e s i s t i v i t y , permeability, shear 251 strength, and magnetic susceptibility. If any of these anisotropics are instrumentally detectable, a rapid technique would then be available for measuring the direction and strength of particle alignment in sediments. DECREASE OF TILL CONSTITUENT WITH DISTANCE FROM SOURCE Mineral grains and clasts decrease in abundance in t i l l with increasing distance from their bedrock sources. Explanations for this relationship include breakage and abrasion during transport, progressive deposition of sediment away from i t s source, and dilution by sediment mixing at the g l a c i e r - t i l l interface (p. 127). The importance of each of these factors can best be evaluated by determining the amount of the t i l l constituent as a function of distance to source for each till size fraction. The functional relationships among these three variables (amount of constituent, distance to source, and grain size) would pre-sumably depend upon the factors (breakage and abrasion, progressive depo-s i t i o n , and dilution) operative at the glacier base. For example, breakage would produce a decrease in frequency of the largest clasts away from the source, but might result in a net increase with distance of a constituent at i t s "terminal grade" (minimum size below which breakage rarely occurs). Other patterns would result where the controlling factor is progressive deposition or dilution. Obviously, breakage, abrasion, progressive deposition, and dilution may occur simultaneously to produce rather complex relationships between constituent abundance and distance to source. Explanations for such patterns require data from a l l t i l l size fractions rather than from a single grade. 252 QUATERNARY STRATIGRAPHY The stratigraphy of Quaternary deposits in the Rocky Mountain Trench between latitudes 49° and 50°10'N has been determined, and correla-tions have been made with deposits of the Flathead Lake area in north--western Montana. A study of Quaternary sediments and glacial landforms in the Trench north of 50°10'N might provide additional information on the extent of glacier recession during each of the two Pinedale inter-stades, and on the pattern and timing of deglaciation. SEDIMENT CONSOLIDATION The thickness of sediment or ice which once covered a clay or s i l t unit may in some places be determined from bulk density measurements on the clay or s i l t . The relationship between bulk density and burial depth, however, is not a simple one, because such factors as post-depositional sediment desiccation, pore pressure during compaction, and variable clay and s i l t mineralogy must be evaluated. Clay and s i l t occur at five stratigraphic levels in the southern Rocky Mountain Trench. Inasmuch as the maximum burial depths of these units are known, the applicability of consolidation techniques may be determined. The uppermost, or late g l a c i a l , s i l t has not been overridden by ice; s i l t within the younger drift and inter-drift s i l t have been compacted by the weight of overlying sediments plus about 1500 m of ice; fine-grained sediments of the Olympia Interglaciation and the St. Eugene Formation have been buried at a minimum by the sediments presently overlying them and by about 1500 m of ice. 253 GROUNDWATER RESOURCES Additional information on the distribution of groundwater and units of semiconsolidated and unconsolidated sediment in the Rocky Mountain Trench can be obtained from a program of drilling and pumping tests. 254 LITERATURE CITED Boulton, G.S., 1971, T i l l genesis and fabric in Svalbard, Spitsbergen, in Goldthwait, R.P., ed., T i l l : a symposium: Columbus, Ohio State Univ. Press, p. 41-72. Dapples, E.C., and Rominger, J.F., 1945, Orientation analysis of fine-grained clastic sediments: a report of progress: Jour. Geology, v. 53, p. 246-261. Dreimanis, Aleksis, 1959, Rapid macroscopic fabric studies in d r i l l -cores and hand specimens of t i l l and t i l l i t e : Jour. Sed, Petrology, v. 29, p. 459-463. Gravenor, CP., 1953, The origin of drumlins: Am. Jour. Sci,, v. 251, p. 674-681. Smalley, I.J., and Unwin, D.J., 1968, The formation and shape of drumlins and their distribution and orientation in drumlin fields: Jour. Glaciology, v. 7, p. 377-390. 255 APPENDICES APPENDIX 1 STRATIGRAPHIC SECTIONS E52 COUUVIUM, FANGLOMERATE E " t l w *GX-2031 >36,000 (wood) T i l l 4 — * YOUNGER DRIFT** 3 — * INTER-DRIFT SEDIMENTS* 2 — *OlO£R DRIFT* 1 — INTERGLACIAL SEDIMENTS 5 — ST. EUGENE FORMATION R — BEDROCK 4 5 4 > »«?c£ 4 .a;p.K; p0§ 3 p i 10 ms 12 11 l i s K i CT* 13 4 p i 2 6 , 8 0 0 ± 1 0 0 0 (wood! 14 | • H tf n tj,-tf-t*;o . 1 •.•tl" "dr-Jl APPENDIX 1 (Continued) STRATIGRAPHIC SECTIONS APPENDIX 2. Southern Rocky Mountain Trench radiocarbon dates. A l l samples were collected by J.J. Clague and analyzed by Geochron Laboratories, Inc. Dates are based on a half l i f e for C11+ of 5570 years and are referenced to the year A.D. 1950. Laboratory sample no. Material Location Stratigraphic position of sample Radiocarbon age (years B.P.) GX-2031 wood 49° 09 '45"N, 115°13'25"W older t i l l >36,000+ GX-2032** wood 49°21 '25"N, 115°17'05"W Olympia Interglacial sediments GX-2033*** peat 49°23 '25"N, 115°18'20"W channeled outwash at top of younger drift 19,100±850 GX-2034*** wood 49°27 '50"N, 115°27'50"W channeled outwash at top of younger drift 14,0001750 *Sample is part of a clast of wood in t i l l . The wood was probably reworked from sediment older than the t i l l . Sample pretreatment—removal of foreign organic material, immersion in hot dilute HCI and in hot dilute NaOH. **Sample was collected about 0.5 m below contact with t i l l . Sample pretreatment—removal of foreign organic material, immersion in hot dilute HCI and in hot dilute NaOH. ***Samples GX-2033 and GX-2034 are clasts of peat and wood from permeable glaciofluvial gravel. Many cobbles and pebbles in the gravel have rinds of precipitated CaC03. The peat and wood possibly were reworked from sediments older than the gravel, and probably have been contaminated since outwash deposition. Sample pretreatment—concentration of organic material, removal of foreign organic material, immersion in hot dilute HCI. +Count rate of sample could not be distinguished from background rate. 259 A P P E N D I X 3. Size frequency data for selected t i l l samples. Only that-portion of each sample finer than -2<j> was analyzed. The sediment was sieved at half-<J> intervals from -2<j> to 4<f>, and pipette samples were collected at one-<j> intervals from 4<f> to 12<j). A l l t i l l samples are from the younger drift, except nos. 1 and 5 which are from the older drift. Sample nos. 6 and 7 are from the same site; no. 6 was collected from the top of the t i l l and no. 7 from the base just above inter-drift s i l t s . Sample locations: 1 — 4 9 ° 1 1 ' 5 5 " N , 1 1 5 ° 0 8 ' 0 0 " W ; 2 — 4 9 ° 1 2 ' 4 0 " N , 1 1 5 ° 1 0 ' 1 5 " W ; 3 — 4 9 ° 2 3 1 2 0 " N , 1 1 5 ° 1 5 ' 3 5 " W ; 4 — 4 9 ° 2 7 ' 2 0 " N , 1 1 5 ° 0 4 ' 1 0 " W ; 5 — 4 9 ° 3 6 ' 1 5 " N , 1 1 5 ° 4 2 ' 15"W; 6 and 7 — 4 9 ° 3 6 1 5 0 " N , 1 1 5 ° 4 0 ' 0 5 " W ; 8 — 4 9 ° 3 7 ' 0 5 " N , 1 1 6 ° 1 3 ' 5 0 " W ; 9 — 4 9 ° 3 7 ' 2 5 " N , 1 1 5 ° 5 6 ' 2 0 " W ; 1 0 — 4 9 ° 5 3 ' 4 0 " N , 1 1 5 ° 4 1 ' 1 5 " W ; 1 1 — 5 0 ° 0 8 ' 1 0 " N , 1 1 5 ° 4 3 ' 2 5 " W . 0 GRAIN SIZE SAND CLAY SILT APPENDIX 4. Fan-fabric sites and results. FAN FABRIC SITES INDEX MAP OF FAN-FABRIC AREAS Samples from an alluvial fan of Quaternary age: 1 ~4 Samples from fanglomerate of Miocene age (St. Eugene Formation): 5 - o MAP OF FAN-FABRIC SITES IN THE SOUTHERN ROCKY MOUNTAIN TRENCH APPENDIX 4 (Continued) Axis of maximum Axis of minimum Fabric Location Elevation* Number of clustering** clustering** no. (m) observations p\ 8l ••«3 3^ 9rl s3 Paraglacial alluvial tan. Fraser Canyon, British Columbia 1 50°46'00"N, 12i°50'10"W 420 60 94 2 33.5 0.696+ 0 62 71.2 0. .104+ 2 50°46'05"N, 121°50'2O"W 400 60 42 13 39.4 0.597+ 208 77 71.5 0. .101+ 3 50°46'10"N, 121°50'25"W 400 60 74 1 41.5 0.561+ 336 83 70.5 0, .111* 4 50°46'20"N,'121°50"15"W 400 60 117 13 39.9 0.588+ 247 71 69.3 0. ,125t St. Eugene fanglomerate, Rocky Mountain Trench, British Columbia 5s 49°04'20"N, 11S°14'20"W 750 5A C 5B 5C 6§ 49011'55"N, 115°08'00"W 790 6A 6B 75 49''11,55"N, 115°08'35"W 770 7A 7B 8s 49°12'15"N, 115°07'45"W 810 8A 8B 8C 140 8 3 46.5 0. .473+ 249 84 69. .0 0, .129+ 20 318 8 41, .8 0, .555++ 75 74 73, .9 0, ,077+ 60 28 10 45. .5 0, .491f 245 78 67. .7 0, ,144+ 60 186 0 45. .9 0. ,485f 279 82 70. ,7 0. ,109f 120 29 8 38. ,0 0, ,621+ 238 80 70. .1 0, .116+ 60 33 12 30. ,8 0. .738t 169 73 74. .1 0. .075+ 60 14 0 43. ,4 0. ,528+ 285 79 68. ,1 0. ,139f 120 76 7 39. ,8 0. .591+ 291 82 78, ,4 0. ,040+ 60 49 1 41. .4 0. .563+ 303 86 77. ,9 0. ,044+ 60 86 11 35. .0 0. ,672t 288 79 79. 8 0. ,031+ 180 55 8 42. .7 0. ,540+ 266 80' 70. ,4 0. ,113+ 60 61 4 45. .8 0. .486+ 307 81 68.4 0. ,136+ 60 66 19 40. 5 0. .578+ 248 71 71. 3 0. ,103+ 60 43 2 39. ,8 0. . 590t 217 88 74. .0 0. .076+ *Approximate. **Axes calculated by the "eigenvalue" method for the analysis of axial orientation data (Mark, 1973). The two vectors are orthogonal. A = azimuth; P = plunge; 6 " standard scattering angle around the associated vector; S = statistic against which the null hypothesis that the population is uniform may be tested (Anderson and Stephens, 1971). ^Population is nonuniform, 99% confidence limits. "^'Population is nonuniform, 95% confidence limits. ^Composite fabric of two or more individual fabrics taken at intervals of several meters to about 100 m laterally. 262 TILL FABRICS TILL-FABRIC SAMPLE NUMBER AND SITE * -J (upper till, lower till) D " TILL-FABRIC AXIS OF MAXIMUM CLUSTERING APPENDIX 5, Contoured diagrams of sele 2-5-8% per 1% area. Additional t i l l - f a cted t i l l fabrics, Contours approximately brie diagrams are shown in Figure 35. 263 TILL-FABRIC SITES TILL-FABRIC SAMPLE NUMBER AND SITE: UPPER TILL (single sample, multiple sample) • F] LOWER TILL (single sample) 2 A 5 0 l-l I I I I S O U T H E R N R O C K Y M O U N T A I N T R E N C H CONTOUR INTERVAL 2000' (610m) 5 10 15 20 Kilometers 14 „13 12 45' APPENDIX 6. Till-fabric sites and results. 264 APPENDIX 6 (.Continued) „ , , c Axis of maximum Axis of minimum Fabric . , Elevation Number of , . * 4 . , * * Location , . , , c lus ter ing* * c lus ter ing*" n o - <m> observations , _ fl> _ , „ , 1 A 9 ° 0 6 ' 2 0 " N , 115 °05 '30"W 850 60 72 9 39.2 0.599+ 257 81 67.5 0.146+ 2 A 9 ° 0 9 ' A 5 " N , 115 °13 '25"W 730 60 27 16 42.0 0.553 f 247 70 69.8 0.120+ 3§ A 9 ° 1 0 ' 1 5 " N , 1 1 5 ° 1 1 ' 1 5 " W 790-817 419 165 2 46.3 0.476+ 47 86 65.6 0.171+ 3A 817 20 302 12 33.7 0.692+ . 139 78 73.8 0.078+ 3B 815 20 335 17 37.9 0.623+ 106 65 69.0 0.128++ 3C 813 20 148 2 31.1 0.734+ 238 32 71.1 0.105+ 3D 8L2 59 276 27 47.9 0.449+++ 101 63 67.2 0.150+ 3E 810 20 301 7 38.8 o.car 207 24 65.0 0.178 3F 808 20 21 10 43.7 0.523+++ 281 46 67.1 0.151++ 3G 807 20 185 16 39.7 0.593+ 298 53 64.0 0.193 3H 805 20 57 6" 46.8 0.468 315 65 68.3 0.137++ 31 80 A 20 164 11 32.6 0.709+ 278 65 73.3 0.092+ 3J 801 20 310 1 41.8 0.555' + 44 80 75.0 0.0671" 3K 800 20 309 19 AA.2 0 . 3 1 5 n + 122 71 71.5 0.101+ 3L 798 20 342 19 41.0 0.569T+ 106 59 72.1 0.094+ 3M 795 20 0 0 A4.4 0.511 269 73 67.8 0.142++ 3N 793 20 155 10 48.8 0.433 301 78 66.1 0.16A++ + 30 792 20 32 7 35.1 0.670+ 296 43 73.9 0.077+ 3P 791 60 180 15 43.7 0.523 f 329 72 71.0 0.106T 3Q 790 20 173 22 44.5 0.509 320 65 66.3 0.161'++ 4? A 9 ° 1 0 ' 2 S " N , 1 1 5 ° 1 1 ' 1 0 " H 818-821 60 167 3 35.1 0.670+ 68 70 69.7 0.121+ 4A 821 20 336 1 32.5 0.711+ 67 61 69.0 0.128'+ AB 819 20 165 8 40.3 0.581+ 60 62 65.4 0.173. AC 818 20 179 4 28.9 0.766+ 76 73 76.6 0.054T 5 4 9 ° 1 1 ' 5 5 " N , 115 °08 '00 "W 790 60 355 8 43.6 0.524 r 199 81 67.2 0.150+ 6 49''12'00"N, 115 °15 '25 "W 820 60 10 12 42.1 0.551+ 248 68 64.1 0.191+ 7 4 9 ° 1 2 ' 3 5 " N , 115 °07 '20 "W 880 60 171 2 38.6 0.610+ 296 86 70.5 0.112T 85 A 9 ° 1 2 ' 4 0 " N , 115 °10 '15 "W 819-829 140 351 2 A1.8 0.556+ 254 69 67.6 0.145+ 8A 829 20 170 5 32.3 0.714T 332 84 69.8 0.119+ 8B 828 20 32 27 37.A 0.631+ 267 48 67.5 0.1467+ 8C 825 20 339 7 35.8 0.658+ 248 1 69.3 0.125++ 8D 823 20 153 8 33.4 0.697' 285 78 70.6 0.111T 8E 822 20 38 18 45.3 0.494 251 69 70.7 0.109+ 8F 821 20 209 6 45.4 0.493 321 73 71.2 0.104+ 8G 819 20 349 11 35.5 0.663+ 220 72 70.8 0.108+ 9 4 9 ° 1 5 ' 3 5 " N , 1 1 5 ° 0 5 ' 5 5 " U 9A0 60 59 12 43.1 0.534+ 215 77 68.9 0.130+ 10 4 9 ° 1 6 ' 0 0 " N , 1 1 5 ° 1 1 ' 5 0 " W 820 60 345 1 44.9 0.502+ 253 65 66.9 0.153T 11 A9"17'25"N, U5 o08'10"W 900 60 36 8 35.6 0.661+ 147 67 69.9 0.118+ 12 A 9 ° 1 8 ' 0 0 " N , 115 °16 '35 "W 820 60 145 3 40.5 0.578+ 241 67 66.4 0.160+ 13 A 9 ° 2 0 ' 3 5 " N , 115 °17 '10 "W 8A0 60 150 3 39.4 0.597+ 53 64 71.9 0.096+ IA A 9 ° 2 3 ' 2 0 " N , 1 1 5 ° 1 5 ' 3 5 " W 870 60 341 2 31.0 0.734+ 225 84 .76.4 0.056+ 15 4 9 ° 2 4 ' 2 5 " N , 115°22'A5"W 870 60 14 14 36.1 0.652+ 231 72 72.4 0.091+ 16 A 9 ° 2 7 ' 2 0 " N , i l 5 ° 0 A ' 1 0 " W 1000 60 351 3 47.1 0.463++ 249 73 65.3 0.174+ 17 4 9 ° 3 4 - 2 5 "N, 115*40'25"W 870 60 315 13 39.0 0.605+ 173 74 67.8 0.142+ 18 A9'3A'A5"N, 115°A7'00"W 890 60 23 12 46.2 0.480T 229 76 68.1 0.139+ 195 4 9 ° 3 5 ' 1 5 " N , 115"A8'A0"W 907-919 100 150 5 42.3 0.547+ 265 78 68.5 0.134+ 19A 919 20 32 5 36.7 0.642+ 293 63 70.1 0.116+ 19B 916 20 150 12 30.9 0.736+ 253 49 72.6 0.090+ 19C 913 20 154 ' 4 38.4 0.614+ 302 85 66.1 0.164+++ 19D 910 20 133 6 35.8 0.658+ 276 82 68.6 0.133++ 19E 907 20 301 4 44.1 0.516+++ 195 76 71.6 0.100+ 20 A 9 ° 3 6 ' 0 0 " N , m ' s i u o ' ^ 870 60 193 18 44.5 0.509+ 356 71 70.5 0.111+ 21 A9 036'15"N, 115°A2'15"W 820 60 315 13 39.1,_ 0.602'; 169 75 70.8 0.109+ 22 4 9 ° 3 6 ' 1 5 " N , 115*52'55"W 870 60 39 11 36.4 0.647+ 211 79 71.7 0.099+ 235 4 9 ° 3 6 ' 4 5 " N , 115"42'35"W 820-86A 440 0 6 45.9 0.484T 136 82 68.8 0.131+ 23A 864 60 334 8 36.4 0.648+ 100 76 73.4 0.082+ 23B 861 20 342 8 34.3 0.683+ 120 80 75.8 0.060+ 23C 859 20 27 15 42.6 0.541++ 248 70 73.9 0.077+ 2 3D 857 20 326 10 38.5 0.612+ 208 70 68.6 0.133++ 23E 855 20 156 5 35.1 0.669 f 13 83 72.2 0.094+ 23F 853 20 349 4 37.6 0.628+ 236 79 73.4 0.082+ 23G 851 20 144 1 42.3 0.546 + T 236 70 70.9 0.108+ 23H 849 20 352 8 41.6 0.559++ 146 81 71.2 0.104+ 231 8A6 20 337 8 38.5 0.612+ 219 74 69.8 0.119+ 23J 843 20 217 4 36.8 0.641+ 128 8 66.9 0.154+++ 23K 840 20 20 7 42.2 0.549++ 127 66 70.3 0.113+ 23L 838 20 196 18 47.2 0.462 21 72 68.3 0.137++ 23M 835 20 75 8 35.9 0.657+ 167 6 67.9 0.141++ 23N 832 20 13 7 41.8 0.555++ 117 61 67.8 0.142++ 230 830 20 3 ; 3 38.2 0.618+ 135 78 76.0 0.058+ 265 APPENDIX 6 (Continued) „ . , * „ , , Axis of maximum Axis of minimum Fabric t „. Elevation Number of , . ** . . ** Location , , . „. clustering clustering no. (m) observations A ; P j 8 j S j A j P j B j S j 231' 23<J 23R 23S 23T 245 24A 24B 24C 2 AD 24E 24F 24G 24H 241 24J 24K 24L .'24M 24N 240 24P 24Q 24R 25§ 25A 25B 25C 25D 25E 26^  26A 26B 26C 26D 26E 27 28 29 49°36'45"N, 115*/,2'35"W 49*36'50"N, 115°40'05"W 49°37'15"N, 115°37'05"W 49<'37'25"N, 115°56'20"W 49<,37'S5"N, 115"57'30"W 49°42'25"N, 115°49'05"W 49°59'20"N, 115°45'20"W HZH 20 2 CO 11 41 . 1 0. V.')'1 >' 158 50 t,<) .1 0 . I.20+ 826 20 5 8 34 .3 0.682+ 172 81 72 .7 0. ,089 + 824 20 284 16 39 .6 0.595+ 38 56 67 .7 0, .143++ 822 20 59 7 38 .8 0.608+ 174 74 69 .5 0, .123+ 820 20 231 14 20. 2 0.880+ 135 24 78 .0 0. 043+ 810-840 481 345 14 45 .1 0.499+ 209 71 67 .8 0. 142+ 840 60 328 10 32 .0 0.719+ 187 77 72 .4 0. ,092+ 839 20 317 12 34 .8 0.674+ 74 66 74 .1 0. •°"L 838 20 "308 1 37 .9 0.622+ 214 78 67 .6 0, .145++ 837 20 103 2" 36 .4 0.648+ 358 82 66 .6 0. ,158+++ 836 20 331 13 45 .8 0.487 210 66 71 .9 0. ,096+ 834 20 351 21 44. .2 0.515+++ 218 60 66 .7 0. .156+++ 833 20 24 26 35 .6 0.661+ 150 49 69 .5 0. ,123+ 831 20 42 30 34 .6 0.678+ 237 59 75 .0 0. 067+ 829 60 36 15 38, .0 0.621T244 73 68 .4 0. ,135+ 826 20 319 19 33, .0 0.704+ 154 71 72 . 5 0. ,090+ 824 20 324 20 29 .9 0.752' 220 32 69 .9 0, .119+ 822 20 20 3 42 .8 0.539++ 282 73 68 .5 0. ,134++ 820 20 168 6 33 .3 0.699+ 69 55 71 .0 0, .107+ 818 20 290 2 41, .7 0.558++ 196 62 66 .9 0. ,154+++ 816 20 344 15 35, .4 0.664+ 247 19 67 .5 0. ,147++ 814 21 26 23 46, .5 0.473 263 53 65 .2 0. ,176 812 20 358 20 43, .6 0.524+++ 226 60 72 .1 0. ,094+ 810 60 34 22 42, .3 0.547+ 219 68 66 .8 0. ,155+ 795-805 180 300 12 49, .4 0.423+ 143 77 63 .6 0. 198+ 805 60 55 1 40, .9 0.571+ 324 56 63 .5 0. 199+ 804 20 223 13 42, .4 0.546T+ 54 76 70 .9 0. ,107+ 802 20 313 23 44, ,7 0.506 115 66 66 .8 0. 155+++ 800 20 142 8 44, .9 0.501 15 76 67 .2 0. ,150++ 795 60 321 20 38. .4 0.613+ 213 38 67 .8 0. 143+ 909-917 100 91 2 40. .9 0.571+ 185 66 67 .8 0. ,143+ 917 20 84 6 45. .6 0.489 180 41 68 .5 0. .134++ 915 20 80 14 28. ,2 0.777+ 282 75 73 .1 0. 084+ 913 20 269 2 31. .0 0.735+ 0 22 69 .6 0. 122+ 911 20 313 11 37. ,8 0.624+ 209 48 70 .4 0. 113+ 909 20 242 7 43. ,9 0.520+++ 114 79 72 .5 0. 091+ 910 60 65 20 37. .0 0.638+ 219 68 67 .9 0. 142+ 910 60 200 26 38. .2 0.618+ 69 54 72. .0 0. 096+ 900 60 317 12 28. ,4 0.774+ 87 73 75 .9 0. 059+ *Approximate. **Axes calculated by the "eigenvalue" method for the analysis of axial orientation data (Mark, 1973). The two vectors afe mutually orthogonal. A ° azimuth; P = plunge; 6 » standard scattering angle around the associated vector; S = statistic against which the null hypothesis that the population is uniform may be tested (Anderson and Stephens, 1971), "fpopulatioryis nonuniform, 99% confidence limits. "^Population is nonuniform, 95% confidence limits. +++Pop ulation is nonuniform, 90% confidence limits. SComposite fabric of three or more individual fabrics taken at intervals vertically through the t i l l sheet. 266 Ir-SCOO' 15' II5"00 S O U T H E R N R O C K Y M O U N T A I N T R E N C H CONTOUR INTERVAL 2000' (610 m) ,0 5 10 15 20 Kilometers 115^122 • JO 121 JO 1,4 ' i f 0 _ l i o 83 82 A—Pebble samples B—Heavy mineral samples C—Pebble and heavy mineral samples / 0 4 9 -15' SEDIMENT SAMPLE SITES SAMPLE SITE AND NUMBER i • A B c UPPER TILL •'"B'» « " * ' • multiple sample • B a • • LOWER TILL single sample A OUTWASH f"**, multiple sample o O 49-00' II6000' 1 45' 30' APPENDIX 7. Heavy mineral and clast lithology data. 267 APPENDIX 7 (Continued). Clast lithology data. * * « * « <u -v +-V in *J C 0J-I-B D. 3 0J V •r* rt *H at rt -!->• o rS *-> O n +• at Xi rt ^ l £j rt O0-I- 014-0 C c Si u (0 at at u i at -H O at at c c o U u O0 O D T3 oo * J a. o X. w CL t-1 c c 0) •H u Ct ±> 1 c rt ra —i N d •H rH 3 —« s *j CO to a •w u u H U H C *-l CL "—t to' VI CU co > C U 00 ^ -H U O0 a -H 1 -H at o at S cj at (fl 1^1 <u c •o U 60 rt C at oo -o OO E •H rt W 0 u rt £. O c rt 3 o ai u o »J w w O • X O o rt o rt cr u O rt cd ro —] a o percent of pebbles 1 4 9 ° 0 0 ' 2 0 " N 1 1 5 ° 0 3 ' 4 0 " W 870 T 0 0 0 0 40 4 17 4 35 2 4 9 ° 0 0 ' 5 0 " N 115 °07 '55 "W 800 T 0 3 2 0 44 5 6 2 38 3 4 9 ° 0 2 ' 1 5 " N 115 °03 '50 "W 900 0 0 5 0 0 44 6 0 0 45 4 4 9 ° 0 2 ' 2 0 " N 115 °12 '45 "W 770 T 0 3 3 0 24 7 9 5 49 5 4 9 ° 0 2 ' 3 5 " N 115 °07 '10 "W 810 0 0 0 0 0 50 7 6 1 36 6 49 °O4 '05"N 115 °10 '35 "W 790 T 0 0 0 0 50 4 12 3 31 8 4 9 ° 0 4 ' 2 5 " N 115 °14 '25 "W 780 T 0 12 0 0 16 7 9 3 53 9 4 9 ° 0 5 ' 1 5 " N 115 °06 '15 "W 840 O 0 0 1 0 42 1 8 4 44 10 4 9 ° 0 6 ' 2 0 " N 115 °05 '30 "W 850 T 0 0 1 0 60 4 15 1 19 11 4 9 ° 0 7 ' 2 5 " N 115 °08 '35 "W 820 0 0 1 0 0 52 6 8 4 29 12 4 9 ° 0 7 ' 3 0 " N 115 °08 '35 "W 820 T 0 0 0 0 42 5 11 4 38 13 4 9 ° 0 7 ' 3 0 " N 115°15 '45"W 780 0 0 1 2 0 25 6 16 2 48 14 4 9 ° 0 7 ' 4 5 " N 115 °14 '30 "W 760 T 0 0 2 0 28 3 18 3 46 16 4 9 ° 0 9 ' 3 0 " N 115 °07 '40 "W 860 0 0 0 0 0 73 1 5 2 19 18 4 9 ° 0 9 ' 4 5 " N , 115 °13 '25 "W 730 T 0 1 0 1 16 6 12 6 58 19 49"10'00"N 115 °13 '50 "W 780 T 0 1 6 0 19 1 25 5 43 20B 4 9 ° 1 0 ' 1 5 " N 115 °11 '15 "W 817 T 0 2 3 0 34 5 25 3 28 20C 4 9 ° 1 0 ' 1 5 " N 115°11 '15"W 812 T 0 2 0 0 34 5 19 6 34 20D 4 9 ° 1 0 ' 1 5 " N 115°11 '15"W 800 T 1 2 1 0 29 6 17 3 41 20E 4 9 ° 1 0 ' 1 5 " N , 115 °11 '15"W 790 T 0 1 0 ' 0 43 9 19 5 23 21 4 9 ° 1 0 ' 5 0 " N , 115 °11 '00 "W 810 0 0 2 0 1 19 3 23 ,6 46 22 4 9 ° 1 1 ' 4 5 " N , 115 °17 '40 "W 920 T 0 1 3 0 24 4 18 5 45 23 4 9 ° 1 1 ' 5 5 " N , 115 °08 '00 "W 790 T 1 0 3 0 34 4 18 2 48 24 4 9 ° 1 2 ' 0 0 " N , 1 1 5 ° 1 5 ' 2 5 " W 820 T 0 3 2 0 17 2 23 3 50 25 4 9 ° 1 2 ' 2 5 " N , 115 °14 '40 "W 780 0 0 1 0 1 5 3 11 6 73 26 4 9 ° 1 2 ' 3 5 " N ) 115°07 '20"W 880 T 0 0 0 0 33 6 24 2 35 27A 4 9 ° 1 2 ' 4 0 " N , 115 °10 '15"W 829 T 0 2 6 0 15 5 29 5 38 27B 4 9 ° 1 2 ' 4 0 " N , i i5°io ' i5"w 819-821 T 0 4 1 0 18 3 25 4 45 29 4 9 ° 1 3 ' 2 0 " N , 115 °12 '15 "W 810 0 1 1 6 0 16 3 27 4 42 30 4 9 ° 1 3 ' 3 5 " N , 115"16'25"W 810 T 0 1 0 0 12 1 40 7 39 31 4 9 ° 1 3 ' 4 0 " N , 115 °12 '45 "W 800 T 0 0 1 0 13 3 28 2 53 32 4 9 ° 1 4 ' 5 0 " N , 115"05'55"W 960 0 0 1 0 0 51 9 3 1 35 33 4 9 ° 1 5 ' 3 5 " N , 115"05"55"W 940 T 0 0 1 0 64 1 20 2 12 34A 4 9 ° 1 5 ' 4 5 " N , 115 °05 '50 "W 920 0 0 0 2 0 39 6 27 6 20 34B 49 °1 .5 ' 45 "N , 115 °05 '50 "W 880 O 0 2 4 0 54 4 19 1 16 35 4 9 ° 1 6 ' 0 0 " N , 115olL'50"W 820 T 0 0 1 0 7 1 46 7 38 36 49 ° ]6 ' no "N , 115 °18 '05"W 770 O 0 2 0 1 4 1 15 9 68 38 4 9 ° 1 6 ' 5 0 " N , 115 °05 '50 "W 940 T 1 0 3 0 . 68 1 19 2 6 39 4 9 ° 1 6 ' 5 5 " N , 115 °10 '10 "W 850 0 0 0 0 0 6 1 20 1 72 40 4 9 ° 1 7 ' 0 5 " N , 115 °06 '15 "W 940 O 0 0 0 0 8 2 43 3 44 41 * 4 9 ° 1 7 ' 1 0 " N , 115 °07 '25"W 900 0 0 0 0 0 8 4 33 4 51 42 4 9 ° 1 7 ' 2 5 " N , 115 °08 '10 "W 900 T 0 1 1 0 33 3 34 2 26 43 4 9 ° 1 7 ' 5 0 " N , 115 °06 '15 "W 940 0 0 1 2 2 10 2 27 6 50 44 4 9 ° 1 7 ' 5 0 " N , 115*07'50"W 910 0 0 1 0 0 6 3 52 8 30 45 4 9 ° 1 7 ' 5 5 " N , 115 °06 '45 "W 950 0 0 0 0 0 12 3 53 6 26 46 4 9 ° 1 8 ' 0 5 " N , 1 1 5 ° 0 7 , 2 0 " W 900 0 0 0 0 0 7 1 31 4 . 57 47 4 9 ° 1 8 ' 2 0 " N , 115 °14 '20 "W 820 T 0 4 0 1 10 3 32 5 45 48 4 9 ° 1 8 ' 4 0 " N , 115"10'50"W 860 T 0 0 0 0 3 1 31 5 60 49 4 9 ° 1 9 ' 3 5 " N , 115 °17 '25 "W 810 0 0 2 0 0 7 4 27 9 51 50 4 9 ° 1 9 ' 5 0 " N , 115 °01 '05 "W 960 0 0 0 6 1 10. 5 31 5 42 51 4 9 ° 2 0 ' 2 0 " N , 115 °17 '45 "W 800 0 4 6 0 0 4 0 27 0 59 52 4 9 ° 2 0 ' 3 5 " N , 115 °00 '55"W 970 0 0 4 2 0 10 4 44 15 21 54 4 9 ° 2 0 ' 4 5 " N , 115 °00 '15"W 970 0 0 0 i 0 15 2 30 5 47 55 4 9 ° 2 1 ' 0 0 " N , 115 °20 '55 "W 830 T 0 6 0 1 9 3 25 5 51 56 49< ,21'05"N, 115 °12 '00 "W 870 T 0 1 0 0 8 3 29 3 56 57 49"21'10"N, 1 1 5 ° 0 0 ' 4 0 " W 970 0 0 0 12 0 0 0 44 5 39 58 4 9 ° 2 1 ' 1 5 " N , 115 °59 '55 "W 950 0 0 0 29 0 ' , 0 0 0 0 71 59 49'23'20"N, 115 °15 '35 "W 870 T 0 0 0 0 4 5 28 9 54 60 49*23'25"N, 115 °18 '20 "W 840 0 0 0 . 0 . 2 10 0 32 5 51 62 4 9 ° 2 3 , 5 0 " N , 115*23'35"H 820 T 1 2 0 1 7 2 35 1 51 64 4 9 ° 2 4 ' 1 5 " N , 115°0'1 '00"W 1020 T 0 0 48 0 0 0 5 10 37 65 4 9 ° 2 4 ' 5 5 " N , 115 °23 '50 "W 810 0 0 0 0 0 2 7 11 10 70 66 4 9 ° 2 5 ' 0 5 " N , 115 °08 '45 "W 1070 T 0 0 0 0 23 0 41 0 36 67 4 9 ° 2 5 ' 1 5 " N , 1 1 5 ° 0 2 ' 4 0 " W 970 0 0 0 9 0 0 0 32 3 56 68 4 9 ° 2 5 ' 4 0 " N , 115o24'40"W 810 T 0 2 0 1 3 2 28 8 56 268 APPENDIX 7 (Continued). Clast lithology data. 1 « V * * CJ T3 -t-"a" o cj cj c l-i 4J o •rt •rt C0 i-i CU ft iJ CO •+-U +-o a a o a o •rt .merit t lgnt Ji u vt <u cj *o a. p c OQ u U CJ u •rt | D I cj -C " W -rt •rt CJ kj -rt Lone-tone' CO .merit t •ti u *J t—i Vt ™ •rt *J i-l C *rt O. 0) Vt 1 CO U > at TO CO *rt a c -o 0  ca c O CO •a oo s •rt CO W o w tn o X u u ca a co cr cj O co ai co .-> o o percent of pebbles 69 49°25'45"N 115°16'40"W 910 0 0 0 0 1 13 1 33 0 52 70 49°27'20"N 115°04'10"W 1000 T 0 0 5 0 0 0 79 . 2 14 71 49°27'35"N 115°30'50"W 850 T 3 6 0 0 4 0 18 8 61 72 49°27'50"N 115°27'45"W 790 0 0 2 0 2 6 2 20 10 58 73 .49°28'55"N 115°25'40"W 840 T 12 10 0 0 8 4 7 5 54 74 49°29'10"N 115°21'30"W 900 0 0 0 0 0 1 5 54 23 17 75 49*29'10"N, 115°33'00"W 840 T 0 1 0 0 6 4 5 9 75 76 49°30'00"N 115°02'40"W 1080 T 0 0 0 0 0 0 0 0 100 77 49°30'50"N 115°47'25"W 950 T 0 6 0 1 13 1 1 1 77 78 49°30'55"N 115*31'50"W 820 T 1 5 0 1 8 2 5 7 71 80 49°31'25"N, 115°36'35"W 830 T 1 9 0 0 10 2 9 5 64 81 49°31'50"N, 115°28'25"W 900 T 1 5 0 1 8 4 9 4 68 82 49°32'20"N, 115°39'00"W 850 T 0 4 0 0 21 0 4 3 • 68 83 49°32'20"N, 115°44'00"W ' 920 T 1 8 0 1 17 2 3 2 66 84 49°33'10"N, 115*33'35"W 860 T 1 3 0 0 2 '2 11 14 67 87 49°34'05"N, 115°35'55"W 870 T 0 6 0 1 2 4 10 13 64 88 49°34'10"N, 115°19'40"W 870 T • 0 5 0 0 13 0 10 12 60 90 49°34'45"N, 115°47'00"W 890 T 1 4 0 0 8 2 3 7 75 91 49°34'55"N, 115*47'10"W 860 0 0 12 0 1 7 3 3 0 74 92 49°35'00"N, 115°44'35"W 890 T 0 11 0 0 7 2 6 0 74 93 49°35'05"N, 115°47'00"W 820 0 0 1 0 5 3 3 9 8 71 94 49°35'15"N, 115*34'20"W 830 T 0 5 0 0 9 3 13 5 65 95A 49°35'15"N, 115°48'40"W 919 T 0 11 0 2 12 0 1 5 69 95B 49*35'15"N, 115°48'40"W 913 T 0 7 0 1 11 3 4 3 71 95C 49°35'15"N, 115*48'40"W 870 T 1 7 0 2 7 2 4 1 76 97 49°36'00"N, 115°51'40"W 870 T 0 4 0 2 5 5 1 2 81 99 49°36'15"N, 115°42'15"W 820 T 0 4 0 0 7 3 13 8 65 100 49°36'15"N, 115°52'55"W 870 T 1 12 0 0 3 1 0 1 82 101 49°36'30"N, 115°18'00"W 1060 T 0 0 0 0 0 0 39 0 61 102A 49°36'50"N, 115°40'05"W 840 T 1 2 0 0 9 4 6 0 78 102B 49*36'50"N, 115°40'05"W 837 T 0 6 0 3 6 1 7 14 63 102C 49°36'50"N, 115°40'05"W 833 T 2 . 5 0 0 4 1 15 15 58 102D 49°36'50"N, 115°40'05"W 829 T 1 3 0 0 0 0 25 12 59 102E 49°36'50"N, 115°40'05"W 824 T 1 5 0 0 4 0 12 17 61 102F 49°36'50"N, 115°40'05"W 822 T 0 2 0 4_ 4 4 11 18 57 102G 49°36'50"N, 115°40'05"W 820 T 0 3 0 2 7 4 9 5 70 102H 49°36'50"N, 115°40'05"W 816 T 1 0 0 1 7 1 15 18 57 1021 49°36'50"N, 115°40'05"W 810 T 0 4 0 0 5 0 13 23 55 103 49°36'50"N, 115°43'55"W 850 T 0 7 0 0 9 1 8 0 75 105 49°37'00"N, 116*14'20"W 1020 T 4 8 0 0 9 0 0 0 79 106 49°37'05"N, 116°13'50"W 1020 T 2 20 0 0 5 0 0 0 73 108 49°37'15"N, 115*37'05"W 805 T . 1 2 0 0 9 1 10 10 67 109A 49°37'25"N, 115°56'20"W 913 T 0 19 0 2 7 1 0 0 71 109B 49*37'25"N, 115*56'20"W 909 T 0 17 0 0 7 2 0 0 74 110 49°37'50"N, 115°57'20"W 920 T 0 44 0 0 4 0 0 0 52 111 49°37'50"N, 115°59'45"W 910 T 0 27 0 0 4 0 0 0 69 112 49°38'00"N, 115°41'25"W 850 T 0 4 0 1 3 3 13 12 64 113 49°38'00"N, 115°47'00"W 880 T 0 5 0 2 11 3 12 6 61 114 49°38'15"N, 115°58'15"W 1010 0 0 4 0 3 14 2 2 0 75 115 49038'55"N, 115°56'30"W 1030 T 0 33 0 0 2 1 0 0 64 116 49°39'20"N, 115*36'40"W 910 T 0 6 0 0 2 2 12 8 70 117 49°39'25"N, 115°39'25"W 820 0 0 11 0 0 3 0 7 4 75 118 49*39'30"N, 115°52'20"W 980 T 0 11 0 1 11 4 0 1 72 119 49'40'10"N, 115°47'15"W 870 T 1 3 0 1 9 1 11 4 70 120 49*40'20"N, 115°34'25"W 1040 T 0 0 0 0 6 0 0 0 94 121 49°40'45"N, 115°38'55"W 870 T 0 2 0 1 5 4 9 4 75 122 49°4r20"N, 115°41'00"W . 790 0 0 5 0 0 2 0 10 9 74 123 49'41'20"N, 115°57'40"W 1120 T 0 11 0 2 8 6 2 0 71 124 49°41'25"N, 115°34'05"W 1110 T 0 0 0 0 0 0 0 0 100 125 49*41'30"N, 115°43'25"W 850 T 0 2 0 0 3 3 12 18 62 127 49°41'50"N, 115°52'30"W 1010 T 0 ' 5 0 2 9 1 4 9 70 128 49°42'35"N, 115°39'00"W 920 T 0 6 0 0 5 1 2 10 76 129 49*43'20"N, 115°42'35"W 800 0 • 1 0 0 0 7 1 9 10 72 130 49°43'55"N, 115°50'20"W 970 T 0 9 0 • 1 16 8 0 0 66 131 49*44'00"N, 115°45'40"W 870 T 0 6 0 0 3 2 17 14 58 269 APPENDIX 7 (Continued), Clast lithology data. C 01 -1-(0 . • H a nj u M w a ^  M T* I ti (0 t-H N O I-H *J t-t (T3 I-i 3 O a ra tr u t/J ' H c <-t CJ H-« percent of pebbles 133 49°46'00"N, 115°42'10"W 830 T 0 0 0 0 3 0 26 24 47 134 49°47'25"N, 115°39'40"W 970 T 0 3 0 0 7 1 11 14 64 135 49°48'30"N, 115"47'30"W 890 T 2 4 0 1 9 5 9 11 59 137 49°50'35"N, 115°40'10"W 860 T 0 0 0 0 2 0 29 22 47 138 49°51'10"N, 115°42'05"W 870 T 0 1 0 • 0 5 2 22 16 54 139 49°51'40"N, 115°47'10"W 860 T 0 10 0 0 12 2 10 12 54 140 49°53'40"N, 115°41'15"W 880 T 0 2 0 0 4 4 19 14 57 141 49°54'40"N, 115°47'20"W 840 T 0 1 0 2 9 1 12 9 66 142 49<'56'00"N, 115*41'35"W 840 0 1 0 0 11 2 5 7 7 67 143 49°56'10"N, 115°41'00"W 900 T 0 1 0 0 6 0 29 33 31 144 49°56'35"N, 115°45'00"W 860 T 0 1 0 0 1 0 20 38 40 145 49°57'00"N, 115°47'25"W 850 T 0 0 0 11 2 3 10 7 67 146 49°58'10"N, 115°42'10"W 910 T 0 1 0 0 1 0 18 40 40 147 49°59'20"N, 115"45'20"W ' 900 T 0 1 0 0 4 0 30 16 49 148 50°00'05"N, 115°40'05"W 970 T 0 0 0 1 0 2 19 52 26 149 50°01'30"N, 115°48'25"W 840 T 1 2 0 8 2 7 7 20 53 150 50°02'40"N, 115"45'05"W 890 T 0 0 0 0 4 1 40 28 27 151 50°05'10"N, 115°45'20"W 890 T 1 1 0 1 0 0 39 25 33 152 50°06'25"N, 115°48'20"W 860 T 1 0 0 21 16 0 11 9 42 154 50°07'45"N, 115°37*50"W 1150 T 0 0 0 0 1 0 7 73 19 155 50°08'00"N, 115°58'55"W 1060 T 0 2 0 0 39 4 0 0 55 156 50°08'10"N, 115°43'25"W 1060 T 0 0 0 1 1 1 37 26 34 157 50°08'15"N, 900 T 0 0 0 0 4 2 30 17 47 158 50°09'20"N, 115°50'55"W 920 T 0 0 0 2 4 7 9 20 58 Approximate. **T, t i l l ; 0, outwash. ***Andesite to quartz gabbro (Purcell extrusives and intrusives)—Precambrian. ^Blairmore Group—Mesozoic. "^Toby Formation and Horsethief Creek Series, Windermere System—Precambrian. "*""*"+Mainly Upper Purcell System—Precambrian. ^Mainly Paleozoic. .^Includes shale, s i l t i t e , quartz arenite, and argillite (dolomitic in part) of Purcell System—Precambrian; shale, s i l t i t e , and sandstone—Paleozoic. 270 APPENDIX 7 (Continued). Heavy mineral data. heavy minerals Lon heavy minerals [on « i 1 Ul Location Elevation (m)* Percent by weight of In fine Band fractJ Percent by volume of in fine sand fractl Magnetite Non-magnetic opaqueB* Total opaquea | Dolomite c a o c o td N a o 3 <fl Sphene Amphibole « « V £ O Magnetite Non-magnetic opaques Total opaques Dolomite V c k. a CJ a M N Staurollte Sphene Amphibole Others Anphlbole/garnet percent of heavy mineral separate of percent of total fine sand fraction by fine sand fraction by volume* volume (x 10z)** 1 49*00'20"N 115*03'40"W 870 0.8 0.6 6 47 53 3 1 T T T 12 31 3 26 29 1 1 T T T 7 22 13 2 49*00'50"N 115*07'55"W 800 1.0 0.6 8 49 57 2 2 0.1 0.3 0.2 13 25 5 32 37 1 1 T 0.2 0.1 8 13 7 4 49*O2'20"N 115*12'45"W 770 2.4- 1.6 13 42 55 3 2 0.1 0.4 0.5 11 28 20 66 86 5 4 0.2 0.6 0.8 18 45 5 6 49*04'05"N 115*10'35"U 790 1.3 0.9 3 44 47 7 4 T 0.4 0.6 11 30 3 39 42 7 •3 T 0.4 0.5 10 27 3 7 49*04'25"S 115*13'55"W 750 7.7 4.8 69 2 1 T 0.1 0.2 7 21 330 7 4 0.5 1.0 34 103 9 8 49*04'25"N 115*14'25"W 780 4.4 2.8 11 56 67 5 1 T 0.1 T 7 20 30 160 190 14 3 0.3 19 54 7 10 49*06'20"N 115*05'30"W 850 0.9 0.6 52 7 1 T 0.1 0.2 6 34 31 4 1 T T 0.1 4 20 7 12 49*07'30"N 115*08'35"W 820 0.8 0.5 46 4 2 0.2 0.8 0.2 20 27 23 2 1 0.1 0.4 0.1 10 13 9 14 49*07'45"N 115"14'30"W 760 1.5 1.0 4 39 43 17 2 T 0.4 0.6 12 25 4 40 44 18 2 T 0.4 0.6 13 22 5 15 49*08'55"N 115*08'30"W 840 0.2 0.1 58 T 3 0.3 0.5 1.0 12 24 7 T T T T 0.1 1 2 4 17 49*09'35"N 115*17'20"w 890 1.6 1.2 35 11 5 0.1 0.6 0.3 17 31 41 12 6 0.1 0.7 0.4 20 40 3 18 49*09'45"H 115*13'25"W 730 1.5 1.0 43 8 5 T 1.0 0.6 11 31 44 8 5 T 1.0 0.6 11 30 2 20A 49*10'15"N 115*11'15"H t t t 1.5 1.1 34 26 3 T 0.2 0.3 10 26 39 29 3 T 0.2 0.3 11 27 4 20B 49*10'15"N 115*11'15"W 817 1.2 0.8 6 41 47 3 5 0.2 1.0 0.5 14 30 5 34 39 3 4 0.2 0.8 0.4 12 21 3 20C 49*10'15"N 115*11'15"W 812 1.0 0.7 3 47 50 6 3 T 0.7 0.5 13 27 2 32 34 4 2 T 0.5 0.4 9 20 5 20D 49*10'15"N 115°11'15"W 800 1.4 1.0 3 31 34 7 6 T 1.3 0.9 19 32 3 31 34 7 6 T 1.3 0.9 19 32 3 20E 49*10'15"N 115*11'15"U 790 1.1 0.8 3 29 32 5 7 0.3 0.7 0.8 15 39 2 24 26 4 6 0.3 0.6 0.7 13 29 2 22 49*11'45"N 115*'.7'40"U 920 1.0 0.7 4 32 36 4 5 0.1 0.5 0.5 22 32 3 22 25 3 4 T 0.4 0.4 16 21 4 23 49*11'55"N 115*08'00"W 790 1.0 0.6 2 59 61 5 1 0.1 0.3 1.2 2 29 1 37 38 3 1 T 0.2 0.8 1 16 2 24 49*12'00"N, 115*15'25"W 820 1.6 1.2 6 26 32 3 8 0.2 1.1 1.2 21 34 7 30 37 4. 9 0.2 1.3 1.4 25 42 3 26 49*12'35"N, 115*07'20"W 880 1.1 0.7 4 61 65 5 1 0.3 0.1 0.3 8 20 3 41 44 3 T 0.2 T 0.2 6 17 16 - 27A 49*12'40"N, 115*10'15"W 829 1.5 1.0 3 44 47 4 2 0.2 0.4 0.7 10 36 3 45 48 4 3 0.2 0.4 0.7 11 33 4 27B 49*12'40"K 115*10')5"U 819-821 1.4 1.0 3 46 49 7 3 0.1 0.5 0.2 13 27 3 44 47 7 3 0.1 0.5 0.2 13 29 5 28 49"12'55"N 115*11'40"U 820 0.9 0.6 67 1 2 0.2 0.3 0.7 8 21 37 1 1 0.1 0.2 0.4 4 16 4 30 49*13'35"H 115°16'25"w 810 1.5 1.1 4 25 29 6 7 0.2 1.6 1.3 22 33 4 28 32 7 8 0.2 1.7 1.4 24 36 3 33 49"15'35"N 115*05'55"W 940 0.5 0.3 56 8 1 T T 0.2 4 31 19 2 T T T T 1 8 3 35 49*16'00"S 115*11'50"W 820 1.2 0.9 3 40 43 2 4 0.1 0.3 0.2 15 35 36 1 3 T 0.3 0.2 12 37 4 37 49*16'40"N, 115*10'35"W 860 0.4 0.2 63 1 3 0.3 0.2 0.5 9 23 15 T 1 T T 0.1 2 2 3 38 49*16'50"N 115*05'50"W 940 0.3 0.2 60 3 1 T 0.1 0.3 2 34 13 1 T T T T 1 5 2 42 49°17'25"N 115*08'10"W 900 0.6 0.4 3 57 60 9 1 T 0 0.1 3 27 1- 21- 22 4 T T 0 T 1 13 5 47 49*18'20"N 115*14'20"W 820 1.0 0.7 4 45 49 1 6 0.1 0.8 1.2 13 29 2 31 33 1 4 T 0.5 0.8 9 22 2 48 49"18'40"N 115*10'50"W 860 2.1 1.4 3 67 70 3 I 0.1 0.1 ' T 5 21 3 91 94 5 2 0.1 0.1 T 7 32 5 53 49°20'35"N 115*1"7'10"W 840 1.5 1.0 43 8 6 0.2 0.5 0.9 10 31 43 8 6 0.2 0.5 0.9 10 31 2 55 49*21'00"N, 115°20'55"H 830 1.9 1.4 5 24 29 4 8 0.4 0.6 0.4 25 33 7 33 40 5 11 0.5 0.8 0.5 33 49 3 56 49*21'05"N 115*12'00"W 870 2.0 1.3 2 70 72 4 1 T T 0.3 3 20 3 88 91 6 2 T T 0.4 4 27 3 59 49*23'20"N 115*15'35"W 870 1.8 1.2 2 42 44 4 3 0.2 1.0 1.5 15 31 3 51 54 5 4 0.2 1.2 1.8 18 36 5 61 49*23'40"N 115*20'50"W 830 1.4 1.0 30 2 8 0.5 1.0 0.6 17 41 31 2 8 0.5 1.0 0.6 17 40 2 62 49*23'50"N, 115"23'35"W 820 1.5 1.1 4 25 29 5 9 0.1 1.1 0.9 17 38 5 27 32 5 10 0.1 1.2 1.0 19 42 2 63 49*24'10"H, 115*26'10"W 860 1.0 0.7 31 3 10 0.3 1.8 1.1 20 33 21 2 7 0.2 1.2 0.8 13 25 2 64 49*24' 15'H, 115*01'00"W 1020 0.7 0.5 1 73 74 T T 0.2 0 T T 26 T 34 34 T T T 0 T T 16 2 66 49*25'0S"N, 115*08'45"M 1070 1.9 1.1 T 82 82 7 T 0.1 0 0 T 10 T 92 92 8 T 0.1 0 0 1 9 4 68 49*25'40"N, 115*24'40"U 810 1.2 0.9 5 23 28 4 8 0.2 1.0 1.1 18 40 4 20 24 3 7 0.2 0.9 1.0 16 38 2 70 49*27'20"N, 115*04'10"W 1000 0.6 0.4 48 2 T I 0 0.1 T 50 T 19 19 1 T T 0 T T 20 71 49"27'35"N, 115*30'50"W 850 1.9 1.4 3 27 30 10 6 T 0.9 1.1 17 35 5 37 42 14 8 T 1.2 1.5 24 49 3 73 49*28'55"N, 115*25'40"W 840 4.0 2.7 1 51 52 9 4 T 0.4 0.1 7 28 4 139 143 25 11 1.1 0.3 18 72 2 75 49*29'10"K, 115"33'00"W 840 2.0 1.5 4 25 29 9 6 T 1.5 0.9 19 35 6 37 43 13 9 T 2.2 1.3 27 54 3 76 49*30'00"N, 115*02'40"W 1080 0.3 0.2 54 0 0 T 0 0 T 46 10 0 0 T . 0 0 0 10 77 49*30'50"N, 115*47'25"W 950 3.8 2.9 3 16 19 2 1 T T 0.7 41 36 9 47 56 7 4 2.0 118 103 34 78 49*30'55"N, 115*31'50"W 820 2.1 1.6 5 27 32 6 5 T 1.1 0.6 16 39 8 41 49 10 8 1.7 0.9 24 66 3 79 49*31'00"N, 115*36'05"W 830 2.5 1.9 24 0 4 0.1 0.4 1.1 36 34 46 0 7 0.2 0.8 2.1 67 67 10 SO 49*31'25"N, 115*36'35"W 830 2.6 1 .9 4 21 25 4 3 0.2 0.3 0.8 33 34 8 40 48 7 6 0.4 0.6 1.5 64 62 11 81 49*31'50"N, U5*28'25"U 900 2.9 2.1 4 24 28 4 5 0.1 1.0 0.6 17 44 9 51 60 9 11 0.2 2.1 1.3 35 91 3 82 49*32'20">1, 115*39'00"W 850 2.4 1.8 4 27 31 4 2 T 0.5 0.9 23 39 7 48 55 7 4 0.9 1.6 42 69 10 83 49*32'20"N, 115*44'00"W 920 4.3 3.3 4 15 19 4 1 T 0.4 0.7 35 40 13 51 64 14 3 1.3 2.3 116 129 39 84 49*33'10"N, 115°33'35"W 860 2.9 2.1 8 23 31 6 5 0.1 0.7 0.9 19 37 16 49 65 13 11 0.2 1.5 1.9 40 77 4 85 49*33'45"N, 115*41'45"W 920 4.8 3.5 28 6 3 T T 1.1 27 35 100 22 10 3.9 95 119 10 86 49*33'55"N, 115*30'35"W 870 3.3 2.5 27 16 3 T 0.2 0.4 11 42 66 39 6 0.5 1.0 28 109 4 87 49*34'05"K, 115*36'55"U 870 3.1 2.3 7 15 22 4 4 0.2 0.5 1.1 32 36 17 35 52 8 8 0.5 1.2 2.6 73 65 9 88 49*34'10"N, 115*19'40"H 960 2.3 1.5 1 64 65 19 T 0.2 0 0 T 16 2 96 98 29 T 0.3 0 0 T 23 89 49*34'15"N, 115*48'10"U 920 6.9 5.3 23 5 1 T T 1.3 33 37 121 27 7 6.9 174 194 24 271 APPENDIX 7 (Continued). Heavy mineral data. n e IM V c B r minerals 1 a V lion ' heavj :ion « « o « 'racl CJ 3 tr « 3 cr •£ TJ o. o P. O tl B o c c e 0 5 IK a V c > Xi o a B V CJ tl c CJ 3 cr a a. CJ « CJ « tt «J B CJ 3 er nj g. 4J « CJ a cc 0J CJ 1 « svati B V c tl CJ B CJ e ? o . | c :con o u 3 CJ fi a V tt c S 1 a 6 o CJ B a o 0 3 CJ c tt Jhlbc ihibc 0 ~1 M CJ Cu tt c «J X o o H 0 o «J C3 VI B o a X O O H a a o N a. t/i % o s percent of heavy mineral separate of percent of total fine sane fraction by fine sand fraction by volume1" volume U 1 0 ' ) " 90 49"34'45"N 115*47'00"W 890 4 0 3 0 3 27 30 5 1 T 0.3 0.3 18 45 10 81 91 13 3 0.9 0.9 54 137 20 92 49*35'00"N 115*44'35"W 890 6.5 5 0 6 16 22 3 2 0.1 T 2.1 37 34 32 77 109 16 8 10.5 187 169 23 94 49*35'15"N 115*34'20"W 830 2 3 1 7 4 22 26 8 7 T 0.8 1.1 14 43 7 37 44 13 11 1.3 1.9 24 75 2 95A 49"35'15"N 115*48'40,,W 919 4 7 3 5 .3 27 30 4 1 T 0 0.9 22 42 9 94 103 15 5 0 3.2 79 145 16 95B 49*35'15"N 115"48'40"U 913 4.8 3 5 4 29 33 3 1 T 0.1 0.3 21 42 14 103 117 12 2 0.4 1.1 74 143 35 95C 49*35'15"N 115*48'40"W 870 6 1 4 5 6 23 29 1 2 T 0.3 0.7 31 36 25 106 131 3 7 1.4 3.2 140 164 21 96 49*35'20"N, 115*40'10"W 840 6 4 4 7 30 9 1 T 0.1 0.8 30 29 140 42 6 0.5 3.8 142 136 23 97 49*36'00"N 115*51'40"W 870 3 7 2 8 3 30 33 11 1 0 0.1 0.5 14 38 7 82 89 35 1 0 0.3 1.4 37 116 27 98 49*36'10"N U5*39'30"W 830 2 3 1 8 25 23 2 T 0.4 0.6 19 30 44 40 3 0.7 1.1 34 57 11 99 49"36'15"N 115*42'15"W 820 2 3 1 7 5 20 25 10 5 0 0.4 1.2 18 40 9 35 44 17 8 0 0.7 2.1 31 67 4 100 49*36'15"K 115*52'55"W 870 10 3 8 2 4 t l 15 1 T T 0 0.7 44 39 30 93 123 7 3 0 5.8 365 316 111 101 49*36'30"N 115*18'00"W 1060 1 2 0 8 1 43 44 39 T 0 0 0.2 1 16 1 36 37 33 T 0 0 0.2 T 10 2 10 2A 49*36'50"N, 115*40'05"W 840 3 0 2 2 9 15 24 6 5 T 0.5 1.7 28 35 20 33 53 13 12 1.1 3.7 61 76 5 102B 49"36'50"N, 115*40'05"W 837 2 5 1 9 7 19 26 6 5 T 0.3 1.2 27 34 12 35 47 11 9 0.6 2.2 49 71 5 102C 49*36'50"N 115°40'05"W 833 2 4 1 8 6 17 23 7 5 0.1 0.4 1.2 29 34 11 30 41 12 8 0.2 0.7 2.1 51 65 6 102D 49*36'50"N 115*40'05"W 829 2 5 1 9 7 20 27 6 5 0.1 0.6 1.3 25 35 13 36 49 11 9 0.2 1.1 2.4 46 71 5 102E 49*36'50"N 115*40'05"W 824 2 1 1 5 9 21 30 6 5 T 1.1 1.0 24 33 14 32 46 9 8 1.7 1.5 36 48 4 102F 49*36'50"N 115*40'05"W 822 2 2 1 6 9 23 32 6 8 T 0.4 0.7 21 32 14 37 51 9 12 0.6 1.1 33 53 3 102G 49*36'50"N 115*40'05"W 820 2 1 1 5 8 22 30 5 6 T 1.1 1.5 24 32 12 33 45 7 9 1.6 2.2 35 50 4 102H 49*36'50"N, 115*40'05"W 816 2 0 1 4 9 22 31 6 8 0.1 1.0 1.6 21 31 13 31 44 8 12 0.1 1.4 2.2 29 43 3 1021 49*36'50"K 115*40'05"W 810 1.9 1 4 9 23 32 5 6 0.1 1.1 1.0 21 34 12 31 43 7 9 0.1 1.5 1.4 30 48 3 103 49*36'50"H 115*43'55"U 850 3 5 2 6 6 17 23 2 2 T 0.1 1.1 37 35 16 44 60 5 6 0.3 2.9 96 90 18 104A 49*36'55"N 115*49'40"W 960 4 7 3 4 33 2 1 T 0.1 0.3 38 26 112 7 3 0.3 1.0 130 87 42 104B 49*36'55"N 115*49'40"W 960 5 4 4 0 34 4 1 T 0.1 0.3 31 30 134 15 4 0.4 1.2 124 121 28 105 49*37'00"N 116*14'20"W 1020 6 5 5 1 7 9 16 0 T T 0 2.1 50 32 36 47 83 0 1 0 10.7 255 160 250 106 49*37'05"N 116*13'50"W 1020 6 8 5 3 5 11 16 1 T 0.1 T 1.4 52 29 27 60 87 3 1 0.5 7.4 277 154 521 107 49*37'10"N, 115*37'00"« 800 3 3 2 5 29 9 3 T 0.1 0.1 12 47 72 21 7 0.3 0.3 29 120 4 108 49*37'15"N 115*37'05"W 805 3 2 2 4 3 26 29 9 4 0.1 0.2 0.2 12 45 6 61 67 20 10 0.2 0.5 0.5 28 114 3 109A 49*37'25"N 115'56'20"W 913 12 1 9 7 2 11 13 T 1 T T 0.3 53 33 22 107 129 4 13 2.9 510 311 41 109B 49*37'25"N 115*56'20"W 909 12 9 10 4 2 9 11 1 1 0 T 0.4 54 33 . 25 95 120 6 12 0 4.2 559 339 49 110 49*37'50"N, 115*57'20"W 920 13 6 11 2 2 7 9 T. 1 0 0 0.6 58 31 19 77 96 1 6 0 0 6.7 643 367 115 111 49*37'50"N 115*59'45"W 910 16 0 13 0 4 9 13 T 1 T 0 0.7 55 30 56 115 171 1 7 0 9.1 716 396 111 112 49*38'00"N 115"41'25"W 850 2 3 1 6 10 24 34 5 5 T 0.8 2.2 23 30 16 40 56 9 7 1.3 3.6 38 45 5 113 49*36'00"N 11S*47'00"U 880 2 2 1 6 4 21 25 8 5 0.1 1.1 1.0 24 36 7 34 41 12 8 0.2 1.8 1.6 39 56 5 115 49*38'55"N 115°56'30"U 1030 12 8 10 4 1 9 10 T 1 0 T 0.5 54 34 6 99 105 1 14 0 5.2 566 349 42 116 49*39'20"N 115'36'40"W 910 3 1 2 4 1 16 17 4 1 0.2 T T 28 50 3 39 42 9 3 0.5 66 119 25 118 49*39'30"N, 115*52'20"W 980 7 3 5 8 2 13 15 1 . 1 0.1 0.1 0.2 48 35 11 75 86 8 8 0.6 0.6 1.2 278 198 37 119 49*40'10"N, 115*47'15"W 870 2 3 1 7 6 25 31 6 3 T 0.4 0.6 20 39 10 43 53 10 4 0.7 1.0 33 68 8 120 49*40'20"N. 115*34'25'U 1040 7 3 5 3 T 41 41 19 T 0.1 0 0 1 39 1 219 220 103 1 0.5 0 0 3 202 3 121 49°40'45"N, U5*38'55"U 870 2 1 1 5 2 31 33 7 2 0.2 0.5 0.7 19 38 3 47 50 11 3 0.3 0.8 1.1 30 54 11 123 49*41'20"N 115"57'40"W 1120 5 2 4 0 3 18 21 2 2 0.2 0.3 0.9 41 33 12 71 83 8 9 0.8 1.2 3.6 161 133 19 124 49*41'25"N, 115*34'05"U 1110 9 1 6 7 T 38 38 9 0 0 0 0 T 53 1 252 253 61 0 0 0 0 2 354 125 49"41'30"N, 115*43'25"W 850 1 3 0 9 5 33 38 8 5 T 0.9 0.4 14 34 4 29 33 7 4 T 0.8 0.4 13 32 3 126 49*41'45"N, 115*48'55"U 910 1 8 1 3 36 7 3 0.1 0.5 1.1 20 32 46 9 4 0.1 0.6 1.4 25 44 6 127 49*41'50"N, 115*52'30"U 1010 1 6 1 1 5 32 37 11 5 T 0.6 1.0 15 30 5 36 41 12 5 T 0.7 1.1 16 34 3 128 49*42'35"N, U5*39'00"U 920 2 4 1 8 2 19 21 7 1 0.6 T 0.1 20 50 4 34 38 13 2 1.1 0.2 36 90 22 130 49*43'55"N 115°50'20"W 970 3 2 2 4 3 20 23 10 2 T 0.3 0.7 16 48 8 47 55 24 5 0.7 1.7 40 114 8 131 49*44'00"N, 115*45'40"W 870 1 7 1 3 6 22 28 9 6 0.2 0.9 1.2 18 37 8 28 36 12 8 0.3 1.1 1.5 23 48 3 132 49*45'45"N, 115*46'40"W 880 0 9 0 6 36 2 10 0.1 1.4 0.7 16 34 21 1 6 T 0.8 0.4 10 21 2 133 49*46'00"N, 115*42'10"U 830 1 1 0 8 4 27 31 10 14 0.2 2.2 1.3 .6 35 3 22 25 8 11 0.2 1.8 1.1 5 28 1 134 49*47'25"N, 115*39'40"U 970 2 9 2 1 1 33 34 8 3 T 0.2 T 17 38 3 71 74 16 5 0.4 37 78 7 135 49*48'30"N, 115*47'30"U 890 1 9 1 4 12 16 28 4 8 T 0.9 1.2 28 30 16 23 39 5 11 T 1.3 1.7 39 43 4 136 49*49'50"N 115*41'55"« 850 1.0 0 7 1 61 62 5 4 T 0.3 0.2 5 23 1 41 42 3 3 T 0.2 0.1 4 18 2 137 49*50'35"N 115*40'10"W 860 1 8 1 2 2 51 53 7 7 T 0.9 0.2 5 27 2 63 65 9 9 T 1.1 0.2 7 29 1 138 49*51'10"N, 115"42'05"U 870 0 9 0 6 3 33 36 8 12 0.1 1.5 0.5 9 33 2 21 23 5 8 T 1.0 0.3 6 17 1 139 49*51'40"S, 115*47'10"U 860 2 4 1 8 9 12 21 12 7 T 1.0 1.7 26 31 17 21 38 22 12 1.8 3.1 46 57 4 140 49*53'40"N, -115*41'15"W 880 0 8 0 6 4 34 38 5 13 0.1 2.0 0.3 7 35 2 19 21 3 7 T 1.1 0.2 4 24 1 141 49*54'40"N. 115°47'20"U 840 1 8 1 3 4 27 31 13 5 T 1.1 1.2 17 32 6 34 40 17 7 T 1.4 1.5 22 41 3 143 49*56'10"N. 115*41'0O"W 900 0 8 0 5 4 43 47 8 6 0.1 0.9 0.3 5 33 2 23 25 4 3 T 0.5 0.2 3 14 1 144 49*56'35 , ,N. 115*45'00"W 860 1 8 1 4 5 22 27 15 9 T 0.9 0.8 16 .31 6 30 36 21 12 T 1.2 1.1 22 47 2 272 APPENDIX 7 (Continued). Heavy mineral data. co a * u «J u a a B a ' heavy :lon '. heavy :ion « « « o u M w U B *. 00 o u a « u | opaquec opaquec e B * i > . u s Ci 5 » g B CS * " is j* «t « V cr a S e « it « cr e e . e « 0 B £ «v B >M ce 3 o « B o « o X < • 3 ce J 0 u B o e 0 jt m o x ! • u O .J > u u e t. «t e. U E U T« 01 a. B ce a X i B O a o t* | o Ct B M e u t* . N 3 a CO V JG a. C/l jl 1 0> £ O B ce & c B O z e o H O Ct c « u u u s • CO « JS B. CO ji % ot X o X 1 percent of heavy mineral aeparate ol percent of total fine sand fraction by fine sand fraction by volume* volume (x 10O** its 49*57'00"1» 115*47 25"W 850 2.6 1.8 5 30 35 15 12 Q.l 2.4 0.8 9 26 9 54 63 27 22 0.2 4.3 1.5 17 45 1 146 49*58'10"N 115*42 10"W 910 1.1 0.8 2 27 29 11 10 T 2.1 1.0 15 32 1 23 24 9 8 I 1.7 0.8 13 23 2 147 49*59'20"H 115*45 20"U 900 2.2 1.6 5 26 31 14 6 0.2 1.3 1.6 18 28 9 42 51 23 10 0.3 2.1 2.6 29 42 3 148 50*00'05"N 115*40 05"W 970 0.7 0.5 3 36 39 10 11 0.3 1.1 0.7 11 27 1 18 19 5 5 0.2 0.5 0.3 5 15 1 149 50*01'30"H 115*48 25"U 840 2.0 1.5 4 27 31 30 3 T 1.2 0.6 10 24 5 40 45 43 4 T 1.7 0.9 15 40 4 150 50"02'40"N 115*45 05"M 890 1.2 0.8 6 28 34 9 7 T 1.3 1.2 15 32 5 23 28 7 6 T 1.1 1.0 12 25 2 151 50*05'10"N 115*45 20"W 890 1.8 1.4 6 24 30 14 3 T 0.4 1.0 24 28 8 32 40 19 4 I 0.5 1.4 33 42 9 152 50*06'25"N 115*48 20"W 860 3.3 2.3 2 40 42 25 8 0.2 1.3 T 3 20 4 92 96 58 19 0.5 3.0 8 45 1 153 50*07'35"N 116*01 05"W 1090 4.4 3.2 4 36 40 2 T T T 0.1 33 25 13 112 125 7 I 0.3 105 83 154 50*07'45"N 115*37 50"U 1150 0.3 0.2 3 34 37 11 6 0.3 1.7 0.6 7 36 1 8 9 3 1 T 0.4 0.1 2 4 1 155 50*08'00"H 115*58 55"W 1060 4.3 3.1 6 28 34 17 T 0 T 0.1 10 39 19 88 107 52 1 0 0.3 31 119 50 156 50*08'10"N 115*43 25"W 1060 1.2 0.8 5 37 42 14 2 0.3 1.0 0.8 13 27 4 30 34 11 2 0.2 0.8 0.7 11 20 6 157 50*08'15"N 115*46 55"W 900 2.5 1.7 12 33 45 10 2 0.1 0.4 1.3 15 26 20 57 77 17 4 0.2 0.7 2.2 26 43 7 158 50*09'20"N 115*50 55"U 920 2.0 1.5 4 24 28 27 4 6.1 0.3 0.7 17 23 6 36 42 41 5 0.2 0.5 1.1 26 34 5 159 50*50'10"N 116*39 05"W 1360 2.8 2.0 4 42 46 7 T 0.2 0 T 1 46 7 83 90 14 T 0.4 0 1 95 5 160 50*51'50"N 116*34 15"W 1160 2.3 1.8 3 27 30 50 T 0.3 0 T T 20 5 47 52 88 T 0.5 0 1 38 2 A^pproximate. **Hainly goethite (pseudomorphic after pyrite) and llmenite. ***Hainly altered silicate grains, but includes minerals of low abundance such as apatite, kyanite, epidote, and tourmaline. P^ercentages calculated from point counts of 500 mineral grains on each of tvo glass slides; for zircon, staurolite, and sphene, T<0.1Z; for other minerals listed, T < 0.5Z. •*""*Tor lircon, staurolite, and sphene, T<0.001X; for other minerals listed, T < 0.005Z, ••"I"-"Sample from upper part of late Wisconsinan till sheet; elevation of sample not known. 273 APPENDIX 7 {Continued)• Correlation matrix for heavy mineral and clast lithology data. The upper number of each pair is the coefficient of correlation; the lower number is the probability that the correlation is not significant (F-test). Longitude Total heavy mineralo Magnetite Non-magnetic opaques* Total opaques Dolomite Garnet Staurolite Sphene Amphibole Granitic Mafic igneous** Chert-pebble conglomerate and sandstone*** Dark a r g i l l i t e , a r g i l l l t e -and quartz-pebble conglomerate7 Greenish-gray argillite+t Red-purple a r g i l l i t e t t Hrae8tone't"''+ Dolostone-'"'"-' Total carbonates''"'",' 0 81 0 00 0 13 0 49 0 14 0 00 0 04 0 45 0 66 0 67 0 00 0 00 0 03 0 16 0 59 0 26 0 74 0 09 0 00 0 01 0 04 0 25 0 67 0 45 0 98 0 69 0 00 0 00 0 00 0 00 0 43 0 31 0 13 -0 12 0 48 0 37 0 00 0 00 0 12 0 21 0 00 0 00 0 19 0 28 0 19 0 18 - 0 13 - 0 01 0 03 0 03 0 00 0 03 0 05 0 16 0 88 0 75 0 23 0 23 - 0 13 0 02 - 0 23 - 0 14 0 07 0 82 0 01 0 01 0 15 0 80 0 02 0 12 0 47 0 00 a 09 0 48 0 67 0 80 0 11 0 29 -0 08 0 25 0 08 0 30 0 00 0 00 0 00 0 25 0 00 0 34 0 00 0 38 0 07 0 43 0 92 0 66 0 30 0 38 - 0 14 0 20 -0 13 0 71 0 45 0 00 0 00 0 00 0 00 0 00 0 11 0 02 0 14 0 00 0 01 0 10 0 06 0 11 0 16 0 17 0 06 0 14 0 09 0 13 0 01 a 89 0 30 0 54 0 28 0 10 0 06 0 55 0 14 0 35 0 17 0 85 0 06 0 48 0 84 0 57 0 30 0 42 - 0 15 0 19 - 0 12 0 66 0 89 0 11 0 56 0 00 0 00 0 00 0 00 0 00 0 10 0 04 0 22 0 00 0 00 0 24 0 16 - 0 28 - 0 11 - 0 13 - 0 08 - 0 10 - 0 12 -0 18 - 0 16 - 0 11 - 0 08 -0 05 - 0 10 0 09 0 00 0 24 0 19 0 44 0 30 0 21 0 05 0 11 0 25 0 43 0 63 0 28 0 31 0 23 0 00 - 0 03 0 06 0 07 0 37 0 46 0 53 -0 05 - 0 05 0 01 -0 07 - 0 06 0 00 0 02 0 92 0 77 0 52 0 49 0 00 0 00 0 00 0 64 0 62 0 86 0 47 0 55 0 59 - 0 53 - 0 24 - 0 14 - 0 08 - 0 18 -0 15 - 0 36 - 0 27 -0 25 - 0 18 - 0 06 -0 19 0 00 - 0 09 0 00 0 00 0 01 0 15 0 43 0 05 0 10 0 00 0 00 0 01 0 06 0 52 0 04 0 93 0 34 0 34 - 0 24 - 0 19 - 0 05 - 0 10 - 0 09 - 0 04 - 0 04 0 05 -0 18 - 0 18 -0 02 -0 12 -0 07 0 03 0 36 0 00 0 01 0 04 0 63 0 32 0 34 0 65 0 65 0 63 0 05 0 05 0 83 0 20 0 44 0 72 0 00 0 13 - 0 49 - 0 50 - 0 44 - 0 38 - 0 42 - 0 22 - 0 14 - 0 06 - 0 36 - 0 40 - 0 13 - 0 45 0 02 - 0 13 0 01 - 0 13 o 17 0 00 0 00 0 00 0 00 0 OO 0 02 0 14 0 58 0 00 0 00 0 17 0 00 0 84 0 18 0 86 0 16 0 55 0 19 -0 32 -0 25 - 0 37 - 0 35 -0 02 0 14 0 19 - 0 21 -0 26 - 0 07 - 0 28 - 0 03 0 03 - 0 34 - 0 23 0 00 0 05 0 00 0 01 0 00 0 00 0 79 0 14 0 05 0 02 0 01 0 49 0 00 0 74 0 75 0 00 0 01 0 24 - 0 23 - 0 54 - 0 46 - 0 49 - 0 51 - 0 17 - 0 01 0 08 - 0 38 -0 43 - 0 13 - 0 48 - 0 01 - 0 07 - 0 19 - 0 23 0 01 0 01 0 00 0 00 0 00 0 00 0 07 0 87 0 44 0 00 0 00 0 17 0 00 0 90 0 46 0 04 0 01 0.17 0.07 0.81 0.00 0.72 0.00 at c - o. ti JS to a. c *Mainly goethite (pseudomorphic after pyrite) and llmenite. **Andesite to quartz gabbro (Purcell extruslves and intrusives)—Precambrian. ***Blairmore Group—Mesozoic. +Toby Formation and Horsethief Creek Series, Windermere System—Precambrian. tlMainly Upper Purcell Syatetn—Precambrian. tttMainly Paleozoic. 274 APPENDIX 8. Textural parameters of outwash underlying late glacial meltwater channels in the southern Rocky Mountain Trench. S t a t i s t i c a l p a r a m e t e r s , * , , uj u „ «. r G r a v e l : s a n d : ( s i l t S a m p l e * L o c a t i o n Kd^ « z o J S A j . * c + c l a y ) IA - 3 . 8 - 2 . 8 2 . 8 * 0 . 5 0 0 . 9 0 7 4 : 2 5 : 1 IB - 3 . 4 - 2 . 3 3 . 2 $ 0 . 4 2 0 . 6 8 6 4 : 3 4 : 2 IC 4 9 ° 1 3 ' N - 4 . 4 - 3 . 1 3 . 3 * 0 . 5 3 1 . 0 6 7 6 : 2 2 : 2 ID 1 1 5 ' 1 6 ' W - 4 . 2 - 3 . 4 2 . 8 * 0 . 4 8 1 . 4 0 8 1 : 1 6 : 3 IE - 4 . 0 - 2 . 7 3 . 2 * 0 . 5 4 0 . 7 6 6 9 : 2 8 : 3 I F - 4 . 2 - 3 . 0 3.3<p 0 . 5 5 1 . 5 9 8 0 : 1 5 : 5 l ( a v c . ) - 4 . 0 - 2 . 9 3 . 1 $ 0 . 5 0 1 . 0 7 7 4 : 2 3 : 3 2A - 3 . 3 - 2 . 5 2 . 2 $ 0 . 5 0 0 . 7 2 6 9 : 2 9 : 2 2B - 4 . 2 - 3 . 0 2 . 6 $ 0 . 6 2 1 . 1 0 8 1 : 1 7 : 2 2C „ - 1 . 3 - 1 . 9 2 . 8 $ - 0 . 2 0 0 . 6 5 5 2 : 4 6 : 2 2D i i ^ T i i u - 3 - 2 _ 3 - ° 2 - 5 * ° - 1 9 0 l 7 1 7 4 : 2 4 : 2 2R " - 2 . 2 - 2 . 4 2 . 7 $ - 0 . 0 5 0 . 6 9 6 0 : 3 8 : 2 2V - 3 . 1 - 2 . B 2 . 6 $ 0 . 1 8 0 . 7 5 6 8 : 3 0 : 2 2G - 4 . 1 - 3 . 6 2 . 7 $ 0 . 3 2 0 . 9 3 8 1 : 1 7 : 2 2 ( a v e . ) - 3 . 1 - 2 . 7 2 . 6 $ 0 . 2 2 0 . 7 9 6 9 : 2 9 : 2 3A - 3 . 2 - 2 . 7 2 . 4 $ 0 . 4 0 0 . 9 4 7 5 : 2 3 : 2 3B *<>••<; • N " 3 - 8 " 3 - 8 1 - 5 * ° - 0 9 1 - 6 4 9 4 : 5 : 1 3C i i e « i 7 i u ~ 4 ' A " 4 a 2 , ° * 0 , 4 6 2 ' 0 3 8 9 : 9 : 2 3D - 4 . 1 - 3 . 8 2 . 2 $ 0 . 4 0 1 . 6 0 8 9 : 8 : 3 3E - 4 . 0 - 2 . 6 3 . 1 $ 0 . 5 6 0 . 9 5 7 5 : 2 4 : 1 3 ( a v e . ) - 3 . 9 - 3 . 4 2 . 2 $ 0 . 3 8 1 . 4 3 8 4 : 1 4 : 2 4A - 1 . 2 - 1 . 5 2 . 0 $ - 0 . 1 5 ' 0 . 8 3 5 2 : 4 6 : 2 4B - 2 . 3 - 2 . 3 1 . 8 $ 0 . 0 2 0 . 9 5 7 7 : 2 2 : 1 4C - 2 . 1 - 2 . 3 2 . 0 $ - 0 . 1 1 0 . 7 9 6 9 : 3 0 : 1 4D - 1 . 5 - 2 . 0 2 . 2 $ - 0 . 2 5 0 . 9 3 5 9 : 3 9 : 2 4E 4 9 * 1 3 ' N - 4 . 3 - 3 . 7 2 . 0 $ 0 . 4 8 0 . 9 4 8 7 : 1 2 : 1 4F 1 1 5 * 0 9 ' W - 3 . 3 - 3 . 3 1 . 7 $ 0 . 1 0 1 . 0 6 9 0 : 9 : 1 4G - 4 . 0 - 3 . 8 1 . 7 $ 0 . 3 2 1 . 3 2 9 1 : 8 : 1 4H - 3 . 2 - 3 . 2 2 . 0 $ 0 . 0 8 1 . 0 3 8 6 : 1 3 : 1 41 - 2 . 5 - 2 . 6 2 . 1 $ - 0 . 0 6 1 . 0 1 7 7 : 2 2 : 1 4 J - 2 . 9 - 2 . 9 2 . 2 $ 0 . 0 5 0 . 8 2 7 9 : 2 0 : 1 4 ( a v e . ) - 2 . 7 - 2 . 8 . 2 . 0 $ 0 . 0 5 0 . 9 7 7 7 : 2 2 : 1 5A - 2 . 6 - 2 . 4 1 . 7 $ 0 . 1 0 1 . 4 9 8 3 : 1 6 : 1 5B 49° 1 0 ' N - 4 . 5 - 4 . 1 1 . 7 $ 0 . 4 9 1 .11 9 1 : 8 : 1 5C 1 1 5 * 0 8 ' W - 4 . 1 - 3 . 7 1 . 9 $ 0 . 3 7 0 . B 8 8 8 : 1 1 : 1 5D - 4 . 9 - 4 . 4 1 . 6 $ 0 . 5 8 1 . 2 9 9 2 : 7 : 1 5 ( a v e . ) - 4 . 0 - 3 . 7 1 . 7 $ 0 . 3 9 1 . 1 9 8 9 : 1 0 : 1 * 4 t o 10 s a m p l e s from 5 g r a v e l p i t s . G r a i n s i z e a n a l y s e s by B r i t i s h C o l u m b i a Department of H i g h w a y s . " S t a t i s t i c a l p a r a m e t e r s a r e m e d i a n (Md*), G r a p h i c Mean ( » z ) , I n c l u s i v e G r a p h i c S t a n d a r d D e v i a t i o n ( o j ) . I n c l u s i v e G r a p h i c Skewness ( S * j ) and G r a p h i c K u r t o s i s ( % ) ( F o l k , 1 9 6 8 ) . " d ^ and Mz e x p r e s s e d i n $. A 116°oo' 115V 50 oo- h50°oo' Ik 49oo' 49°oo/ 116°oo 45 30 115W FIGURE 5 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0052380/manifest

Comment

Related Items