Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Lake sediments as records of palaeoenvironmental change : Kwoiek Creek, Coast Mountains, British Columbia Souch, Catherine Jane 1990

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

Item Metadata

Download

Media
831-UBC_1990_A1 S68.pdf [ 10.63MB ]
Metadata
JSON: 831-1.0100482.json
JSON-LD: 831-1.0100482-ld.json
RDF/XML (Pretty): 831-1.0100482-rdf.xml
RDF/JSON: 831-1.0100482-rdf.json
Turtle: 831-1.0100482-turtle.txt
N-Triples: 831-1.0100482-rdf-ntriples.txt
Original Record: 831-1.0100482-source.json
Full Text
831-1.0100482-fulltext.txt
Citation
831-1.0100482.ris

Full Text

LAKE SEDIMENTS AS RECORDS OF PALAEOENVIRONMENTAL CHANGE KWOIEK CREEK, COAST MOUNTAINS, BRITISH COLUMBIA By CATHERINE SOUCH B.A. (Hons.) University of Cambridge 1982 M.Sc. University of Br i t i s h Columbia 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Geography) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June 1990 ©CATHERINE SOUCH 1990 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of G e O A g - f t P * ^ The University of British Columbia Vancouver, Canada Date , 'S /o 6 / q>o  DE-6 (2/88) i i ABSTRACT It has been suggested that the dominant controls on alpine sediment transfers during the Holocene Epoch relate to climate change, specifically paraglacial sedimentation and Neoglacial activity. Alpine lakes with appropriate geometry and hydraulic conditions trap a high proportion of sediments inflowing from their surrounding drainage basins. Thus alpine lake sediments have the potential to yield a comprehensive, integrated signal of drainage-basin geomorphic ac t i v i t y through time, which may be interpreted as a proxy record of Neoglacial ac t i v i t y . This study i s concerned with the interpretation of alpine lake sediments in glacierized drainage basins as records of Neoglacial a c t i v i t y . It adopts an ex p l i c i t l y geomorphological approach that integrates an under-standing of the drainage basin sedimentary system, specifically sediment sources and transfers, with the interpretation of lake sediment deposits and extends existing models of alpine sedimentary response down-valley, away from the immediate proglacial environment. A down-valley sequence of four valley bottom lakes, Kha, Klept, Kokwaskey and Kwoiek, within the Kwoiek Creek watershed, southeastern Coast Mountains of British Columbia, were studied. Sub-bottom sounding and multiple cores from each lake allowed identification of lake-wide changes i n sediment input through time; i n addition terrain mapping and character-isation of sediment sources provided a framework within which to identify the sources of the lake sediments and their fluctuations through time. Preliminary characterization of the sediments broadly separated organic and c l a s t i c components. Detailed laboratory analyses revealed organic matter content to be a good inverse indicator of sedimentation rates. Grain size analyses revealed three distinct textural populations. i' i i i Graphical partitioning of the cumulative grain size distributions identified each fraction for further analysis. The provenance of the coarsest and intermediate fraction was determined through SEM surface texture analysis of a s t a t i s t i c a l l y representative number of grains. The coarsest fraction was derived from localized col l u v i a l sources. The intermediate fraction was derived from gla c i a l sources and strongly f i l t e r e d downsystem. The finest fraction was characterised as glacial in origin because of consistent trends in i t s v a r i a b i l i t y at the drainage basin scale through time. Fluctuations i n the total influx of the intermediate and finest fractions are interpreted as a proxy record of Neoglacial activity in the watershed. Analysis of persistence in the sedimentation data indicates history of the order 100 yrs, which i s interpreted as an index of the relaxation time of sedimentary stores. Basal dates on the sediments provide the earliest dates for deglaciation in the southern Coast Mountains, suggesting that extensive areas of southwestern British Columbia were ice free prior to 11 500 B.P. Three phases of Neoglacial ac t i v i t y centred 6000 to 5000 B.P., 3500 to 2900 B.P. and post 750 B.P are suggested by increased sedimentation rates for glacially-derived material. When compared with reconstructions from a pollen study conducted within the watershed and regional chronologies reported i n the literature, there i s remarkable consistency. The major advantage of the lake sediment approach as developed in this study i s the continuity and apparent sensitivity of the derived proxy records. These records permit a consideration of both the magnitude and frequency of palaeoenvironmental change, specifically Neoglacial activity, at one s i t e . Such a record has not been found elsewhere in British Columbia, where discontinuous t e r r e s t r i a l records have been used. iv TABLE OF CONTENTS ABSTRACT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ACKNOWLEDGEMENTS CHAPTER 1 INTRODUCTION 1.1 The alpine sediment system as a framework 1.2 Alpine sedimentary records 1.3 Objectives CHAPTER 2 THE STUDY SITE 2.1 Selection of the Kwoiek Creek watershed 10 2.2 The catchment 12 2.3 Geology 13 2.4 Physiography and glacial history 13 2.5 Contemporary climate 16 2.6 Vegetation 19 2.7 Contemporary geomorphology and sediment transfer processes 21 2.8 Anthropogenic act i v i t y 24 CHAPTER 3 COLLECTION OF FIELD DATA AND PRESCREENING OF CORES 3.1 The lakes 26 3.1.1 Morphometry 26 3.1.2 Vertical density structure and circulation 33 3.2 Selection of core sites 35 3.3 Lake sediment sampling 36 3.4 Laboratory procedures 40 i i i v v i i v i i i x 1 5 8 V CHAPTER 4 INTERPRETATION OF THE LAKE SEDIMENTARY RECORD 4.1 Organic matter content: an inverse indicator of sedimentation rate 45 4.2 Characterization of the cl a s t i c component 51 4.3 Sediment source analysis: SEM analysis of the coarse fractions 59 4.3.1 Methodology 63 4.3.2 Objective c l a s s i f i c a t i o n of the deposits 71 4.3.3 Classification of lake deposits 87 4.4 Sediment sources: the s i l t - f r a c t i o n 89 4.5 Summary 93 CHAPTER 5 TEMPORAL CHANGES IN SEDIMENTARY RESPONSE 5.1 Temporal v a r i a b i l i t y in sedimentation rates 96 5.2 Changes i n the glac i a l sediment signal through time 111 5.3 Characteristics of temporal v a r i a b i l i t y 119 5.4 Summary 123 CHAPTER 6 REGIONAL CORRELATION 6.1 Deglaciation 125 6.2 Holocene Epoch: Palynological records 130 6.2.1 Palynology Kwoiek Creek watershed 132 6.3 Holocene Epoch: Glacial chronologies 136 6.4 Significance of results from the Kwoiek Creek watershed 142 CHAPTER 7 SUMMARY AND CONCLUSIONS 7.1 The methodology 146 7.2 Palaeoenvironmental reconstructions 148 REFERENCES CITED 151 v i APPENDIX I LITHOSTRATIGRAPHY OF THE CORES I. 1 Characterization scheme 162 APPENDIX II CHRONOLOGY II. 1 Radiocarbon chronology 198 11.2 Tephra identification 198 II.2.1 Methodology in this study 201 11.3 Dated anthropogenic effects 208 APPENDIX IV DATA IV.1 Grain size data 210 v i i LIST OF TABLES 2.1 Summary of selected climate statistics 17 3.1 Morphometric characteristics of lakes studied 31 3.2 Concentrations of suspended particulate matter in the waters of gl a c i a l lakes 37 4.1 Assessment of v a r i a b i l i t y of cla s t i c and organic sediment influx 50 4.2 Summary of modal stati s t i c s for grain size populations by lake 56 4.3 Surface textures used in SEM analysis 67 4.4 Relative abundance of surface textures for samples of known origin 70 4.5 Characteristic abundances of surface feature categories on quartz grains from varied environments of modification 84 4.6 Difference s t a t i s t i c s for comparison of known sedimentary environment signatures determined form the Kwoiek Creek watershed with literature values 86 4.7 Characteristics of lake sediment samples c l a s s i f i e d 88 5.1 Characteristics of the detailed sedimentation records for each lake 107 6.1 Deglacial dates determined in other studies in southern Coast Mountains 128 6.2 Radiocarbon dates obtained in this study 129 6.3 Neoglacial chronologies for Coast Mountains 138 AII.l Radiocarbon dates obtained i n the study 199 All.2 Element concentrations for Mazama and Bridge River tephra 204 All.3 The "A" s t a t i s t i c for comparison of Mazama and Bridge River tephra 206 All.4 Examples of I s t a t i s t i c for unknown samples 207 All.5 Date of logging road construction past each lake 209 v i i i LIST OF FIGURES 1.1 Schematic representation of main sediment transfer patterns and storages i n alpine watersheds 2 2.1 Location map of the Kwoiek Creek watershed 11 2.2 Bedrock geology Kwoiek Creek watershed 14 2.3 Location of climate stations 18 2.4 Vegetation cover Kwoiek Creek watershed 20 2.5 S u r f i c i a l geology Kwoiek Creek watershed 22 3.1 Bathymetry of Kha Lake and coring locations 27 3.2 Bathymetry of Klept Lake and coring locations 28 3.3 Bathymetry of Kokwaskey Lake and coring locations 29 3.4 Bathymetry of Kwoiek Lake and coring locations 30 3.5 Thermal profiles for the lakes 34 3.6 Suspended sediment profiles for the lakes 36 4.1 Relation between sedimentation rate and organic matter content of varved sediments for Kokwaskey Lake 48 4.2 Textural triangle for lake bottom sediments 52 4.3 Example of log-probability plot 54 4.4 Characteristics of sediment sub-populations 57 4.5 Origin of quartz grain surface features 62 4.6 Photographs of selected grains to i l l u s t r a t e surface textures 68 4.7 Relative contribution of individual surface texture to difference s t a t i s t i c s : a) Within group b) Between group c) Extreme range 74 4.8 Difference s t a t i s t i c s for samples from known sedimentary environments 76 4.9 Cumulative distribution of difference s t a t i s t i c s 78 4.10 Difference s t a t i s t i c s - pooled glacial and co l l u v i a l 80 ix 4.11 Difference s t a t i s t i c s - pooled glacial and c o l l u v i a l within and between group difference s t a t i s t i c s 81 4.12 Classification of known subaqueous deposits 83 4.13 Example of XRD reflections of s i l t - s i z e d fraction 92 4.14 Methodology to isolate glacial sediment signal 94 5.1 Standardised sedimentation data for Kha Lake 98 5.2 Standardised sedimentation data for Klept Lake 99 5.3 Standardised sedimentation data for Kokwaskey Lake 100 5.4 Standardised sedimentation data for Kwoiek Lake 101 5.5 Fi l t e r i n g of sedimentation data: Kha Lake 103 5.6 Fil t e r i n g of sedimentation data: Klept Lake 104 5.7 Fi l t e r i n g of sedimentation data: Kokwaksey Lake 105 5.8 Fi l t e r i n g of sedimentation data: Kwoiek Lake 106 5.9 Downstream attenuation of absolute rates of sedimentation and v a r i a b i l i t y 110 5.10 Variations through time in proportions of g l a c i a l l y derived fractions 113 5.11 Variations through time in fine sediment influx: proxy record of glac i a l a c t i v i t y Kwoiek Creek watershed 114 5.12 Serial correlation coefficients for sedimentation data 121 6.1 Locations of previous palaeoenvironmental reconstructions in southern B r i t i s h Columbia 126 6.2 Pollen p r o f i l e for Kwoiek Lake 134 6.3 Schematic representation of Holocene Epoch g l a c i a l and pollen signals 143 AI.l Lithostratigraphy of cores 163 AII.l Photomicrographs of Mazama and Bridge River tephra 202 X ACKNOWLEDGEMENTS I would like to express my appreciation to the many people who generously contributed their time and expertise to this study. Dr. Olav Slaymaker, my supervisor provided advice and guidance throughout my graduate studies. His good humour and encouragement to always see the broader context have been much appreciated. I would like to thank the members of my committee for their constructive comments on drafts of the thesis. In addition, Dr. M. Church for his expertise and c r i t i c a l insights in the analysis of the laboratory data; Dr. J. Ryder for her assistance in many practical aspects of the study, through her example in the f i e l d she taught me most of what I know about the Quaternary Geomorphology of southern B.C.; and Dr.J. Clague for his practical assistance and advice with many aspects of the work. I have been given much specialised assistance with f i e l d equipment and laboratory procedures. I would like to thank Peter Lewis for his i n i t i a l assistance with the RTT system; Mike Church and Bob Gilbert for their persistence with sub bottom sounding; Dr. G.E. Rouse for providing both f a c i l i t i e s and guidance for pollen work; and John Knight who was particularly helpful with the SEM analyses. In addition I would like to thank my f i e l d assistants and the many "volunteers" who assisted with the f i e l d component of the research, the feminist reading group and the "occupants" of the geomorphology offices for their contributions in very different ways! Funding for this research has been provided to Dr. Slaymaker by Natural Sciences Engineering Research Council of Canada. Personal funding was provided through University of British Columbia Graduate Fellowhips and Teaching and Research Assistantships in the Department of Geography. In particular I would like to thank Helen Cleugh and Gary Barrett for many hours of discussion on a l l topics; and Sue Grimmond, for her advice and assistance in a l l aspects of this study and most of a l l for her unwavering support and encouragement. 1 CHAPTER 1 INTRODUCTION This study i s concerned with the interpretation of alpine lake sediments in glacierized drainage basins as records of palaeoenvironmental 2 3 conditions over timescales of 10 -10 yrs. The study adopts an e x p l i c i t l y geomorphological approach which integrates an understanding of the drainage basin sedimentary system, specifically sediment sources and transfers, with the interpretation of lake sediment deposits. The emphasis i s both methodological, the development of techniques and analyses to document and interpret changes i n the palaeo-sediment system through sediment source and sink analysis; and substantive, i n i t s interpretation of a variety of sedimentary signals to obtain a greater understanding of Holocene palaeoenvironmental history of southern Br i t i s h Columbia. 1.1 The alpine sediment system as a framework for timescales of 10^ yrs The movement of sediment through a drainage basin may be envisaged as a cascade of material moved episodically between storages by a suite of transfer processes. The "alpine sediment system" provides a conceptual framework within which to view the associated processes of weathering, erosion, transport and deposition, emphasising sediment transfer processes and incorporating g l a c i a l , f l u v i a l , aeolian and slope processes. A highly schematic representation highlighting sediment sources and storage points i s presented in Figure 1.1 (adapted from Church, 1980). As a framework i t enables integration between processes (climatological, biological, geomorphological, hydrological) and between places (supraglacial, g l a c i a l , subglacial, proglacial) (Gurnell & Clark, 1987). Alpine sediment transfers vary in time, the nature of the v a r i a b i l i t y depending on temporal scale. At the Holocene timescale the dominant controls relate to climate change, specifically paraglacial sedimentation Figure 1.1 Schematic representation of main sediment transfer patterns and storages i n alpine watersheds (adapted from Church, 1980). Emphasis on sediment sources and sinks. The transfer pathways are greatly simplified. Fluvial transport V • A Sediment sources (mode of erosion) Transport mode Sediment sink (mode of deposition) / \ Unstable Stable for short periods Stable for long periods 3 from late Wisconsinan glaciation and Neoglacial activity (Church, 1980). During the second half of the Holocene Epoch many mountain areas, particularly in mid-latitude and mid-altitude locations, have undergone repeated periods of glac i a l advance and retreat (Grove, 1988), with associated contrasts i n the glacio-hydrological (Rothlisberger & Lamb, 1987) and sediment systems (Johnson, 1982). The general phasing of changes in a glaciated sediment system has been codified by Church and Ryder (1972) in the framework of paraglacial sedimentation. This model focuses attention on the importance of considering sediment dynamics and interpreting the sedimentary record against a background of glac i a l forcing and relative chronology. The underlying assumption of the paraglacial concept i s that glaciation represents a fundamental change in the ter r e s t r i a l erosional environment such that major quantities of sediment are produced in the form of glacial d r i f t . This material may have attained s t a b i l i t y within the glac i a l depositional environment, but i s unstable with respect to the f l u v i a l environment that succeeds the glacier spatially and temporally. Thus the proglacial and postglacial rivers evacuate the glacial sediment at a rate that i s far in excess of the "normal" material supply expected i n nonglacial environments. For periods up to several thousand years, sediment yield of large regions i s unrelated to the concurrent primary production of debris by non-glacial processes (Church & Slaymaker, 1989). Thus a change from glacial to non-glacial conditions in i t i a t e s a series of significant changes as the various components of the system adjust at different rates to the altered and altering environment. Inevitably there w i l l ensue marked changes i n the rate and pattern of sediment production and transfer, reflecting the sensitivity of the system i n a phase of disequilibrium, that should be 4 recorded i n the deposits i n sediment stores. Over the Holocene Epoch, at 2 3 timescales of the order 10 to 10 y r s , sedimentary records i n g l a c i e r i z e d a l p i n e drainage basins should provide a continuous palaeoenvironmental record that r e f l e c t v a r i a t i o n s i n Neoglacial f o r c i n g of the sediment cascade and p a r a g l a c i a l reworking. Downstream from the immediate p r o g l a c i a l environment i t becomes important to consider the f a c t that changes i n sediment p r o p e r t i e s r e f l e c t both a d i r e c t g l a c i a l e f f e c t and a secondary geomorphic c o n t r o l , namely other sediment t r a n s f e r processes and sediment traps operative i n the drainage basin (see Figure 1.1). Sediment sources other than g l a c i a l meltwater streams may cont r i b u t e s i g n i f i c a n t l y to the sedimentary record as a consequence of sub a e r i a l weathering and episodic mass movement events. In steep t e r r a i n , mass movements (slumps, earthflows, creep, debris avalanches etc) provide sediment to the f l u v i a l system where entrainment by flowing water may occur. In other instances, more ra p i d flow phenomena (debris flows, debris torrents etc) are s i g n i f i c a n t sources of material f o r a channel system. These introduce considerable v a r i a b i l i t y to the sediment record at timescales 10^-10* y r s . Interpretation of sediment records i n terms of palaeoenvironmental conditions i s p o s s i b l e only i f the f u l l range of processes governing sedimentation at appropriate temporal and s p a t i a l scales i s understood. This permits the sedimentary s i g n a l s of i n t e r e s t to be i s o l a t e d , and models of sedimentary response with s u i t a b l e f o r c i n g functions, boundary conditions and lags can be formulated. The nature of the v a r i a b i l i t y of al p i n e sediment t r a n s f e r s , and more generally, the e f f e c t of reworking of g l a c i a l and c o l l u v i a l deposits on down-valley sedimentation over 2 3 timescales of the order 10 -10 yrs has been l i t t l e considered, and i s the 5 focus of this study. 1.2 Alpine sedimentary records The late Quaternary stratigraphic record of British Columbia i s largely a product of brief depositional events that were separated by long periods of non-deposition and erosion (Clague, 1986). At the broadest regional scale glaciation has been characterised by deposition and non-glacial periods by non-deposition and erosion. When viewed in a spatial framework, geologically significant sedimentation i.e. that which produces deposits capable of being preserved i n the stratigraphic record, has been restricted to particular parts of the B r i t i s h Columbia landscape. The main te r r e s t r i a l sediment "sinks" have been valleys and coastal lowlands, while mountains and uplands have been areas of erosion (Clague, 1986). As a consequence, the stratigraphic record of the Holocene Epoch in alpine environments is restricted to lakes and mountain valleys. Many investigators (see for example the pioneering work of de Geer (1912), Antevs (1922); and more recently Schytt (1963), Cewe and Norbbin (1965), Norbbin (1973), Karlen (1976, 1981), Davis et a l (1979), Leonard (1986 a,b) and Zelinski & Davis (1987)) have suggested that the glacial sedimentary records i n alpine lakes exceed the sensitivity, resolution and continuity of the t e r r e s t r i a l record upon which Quaternary stratigraphers have traditionally r e l i e d . The basic premise of such studies i s that lakes with appropriate geometry and hydraulic conditions trap a high proportion of inflowing particulates from their surrounding catchments, thus yielding a comprehensive, integrated signal of drainage-basin scale geomorphic act i v i t y (Oldfield, 1977). The location and size of a lake relative to i t s contributing area, i n addition to the hydraulics of the system, dictate 6 the sensitivity and resolution of the record. When considered in this context lake sediments may be viewed as the most significant sediment store in the watershed sediment cascade. If their study i s combined with a consideration of sediment sources and transfer processes, lake sediments provide an opportunity to determine temporal v a r i a b i l i t y of sediment movement in a catchment in terms of both volume and provenance, thereby providing a basis for integrated insight into variations of the alpine sediment system at several timescales. Attempts to model the sedimentary response of alpine lakes have been largely restricted to the immediate proglacial environment. Many studies have been conducted on controls of annual, and to a lesser degree, daily sediment yield variations, for example Ostrem (1975), Perkins & Sims (1983), Weirich (1985) and Gilbert & Desloges (1987). These investigations stress the importance of hydroclimatic factors, specifically runoff, which i s largely a function of snow and ice melt rates. At longer timescales 1 2 (10 -10 yrs), available data indicate that sedimentation rate variations bear a close relation to ice extent, suggesting that these longer term changes are of value for reconstructing glacial activity (Karlen, 1976, 1981; Leonard, 1986 a,b). At these timescales, rates of glacial erosion have much more influence on downvalley sedimentation rates, and the a v a i l a b i l i t y of meltwater for sediment transport i s less of a limiting factor (Leonard, 1981). Given the l o g i s t i c a l constraints of obtaining long sediment records i n proglacial environments, Karlen (1976) suggests that i t i s expedient to study deposits i n lakes with a lower input of gla c i a l sediment. These may be lakes in drainage basins where the area covered by glaciers i s small i n comparison with the area where the glacial sediment w i l l eventually settle, or downvalley lakes where there has been some 7 attenuation of the g l a c i a l s i g n a l of i n t e r e s t . Such downvalley lakes are the focus of t h i s study. Lake sediments " i n s i t u " are a consequence of complex i n t e r a c t i o n s between mechanisms of deposition, resuspension, chemical and biogenic transformation, and longer term diagenesis. Moreover sediment accumulation i n open lakes i s a function of the r e l a t i o n between input and l o s s through the outflow: t h i s i s i n turn r e l a t e d to sediment type, residence time and lake morphometry. Changes i n the mode of sediment input to a lake may a f f e c t the d i s p e r s a l of incoming sediment and have profound e f f e c t s on deposition patterns. These may a r i s e , f o r example, as a consequence of changes i n the d i s t r i b u t a r y pattern on a d e l t a , or d i f f e r e n t modes of sediment d e l i v e r y (see for example, Smith et a l . , 1982). A l t e r a t i o n s i n the r e l a t i v e contributions of overflows or underflows r e s u l t from changes i n the p h y s i c a l limnology and s t r a t i f i c a t i o n regime of a lake, p o s s i b l y as a consequence of changes i n the bathymetry of the lake due to sedimentation and i n f i l l i n g . Thus i n developing a d e s c r i p t i v e , downvalley, drainage basin scale model of lake-sediment character i t becomes necessary to consider two scales of c o n t r o l s : those operating on sediment sources i n the watershed ( i . e . d r i v i n g forces of t r a n s f e r processes and r e s i s t a n c e of the t e r r e s t r i a l m a t e r i a l s ) ; and those r e l a t e d to the nature of the d e p o s i t i o n a l environment. The approach proposed i n t h i s study to resolve and understand 2 3 these controls at timescales of 10 -10 y r s , i s to document sedimentary changes at a drainage basin scale by studying a number of lakes i n a downvalley sequence, with m u l t i p l e cores taken from each lake ( c . f . Dearing, 1982). This approach permits basin-wide chronologies to be constructed ensuring that l o c a l i s e d changes not be misinterpreted. 8 1.3 Objectives It i s the contention of this study that fluctuations in the g l a c i a l l y driven sedimentation signal and i t s attenuation by non-glacial inputs at the drainage basin scale can be identified within the framework of the alpine sediment system, by analysis of sediment source, pathway, and lake-sediment sink. Characteristic physical and chemical processes in sediment production, transfer and storage impart distinctive traits to the sediments which, i f identified, can be used to reconstruct, quantify and interpret flows of material. Thus an understanding of the palaeo-sediment system can be attained, which i s important not only for a contingent understanding of the functioning of the contemporary alpine sediment system, but i s fundamental to the use of alpine sedimentary records for palaeoenvironmental reconstructions. Specifically this study involves: 1. Analysis of drainage basin sediment sources in conjunction with the lake sedimentary record i n order to develop a methodology to document and interpret changes in characteristics and processes of sedimentation in a chain of alpine/sub-alpine lakes during the Holocene Epoch in southern British Columbia. This permits isolation of the glacial sediment signal of interest from co l l u v i a l , f l u v i a l and aeolian sediment inputs. 2. Development of a descriptive model for the Holocene Epoch of lacustrine sedimentary response to fluctuations i n Neoglacial forcing i n downvalley lakes in alpine/subalpine watersheds at timescales of the 2 3 order 10 to 10 yrs. From the model of paraglacial sedimentation i t would be expected that the rate of sedimentation and the relative contribution of gla c i a l sediment has fluctuated through time, decreasing 9 down-system, with an associated decrease in sensitivity and increase i n lag time of sedimentary response. 3. Comparison of these patterns/processes with a local palynological study conducted within the basin, and glacial and climatic history derived from literature sources, i n order to assess their consistency with more conventional lines of evidence. The structure of the thesis i s designed to highlight these objectives. Chapters 2 and 3 provide contextual information in terms of the study area, identification of sediment sources, and establishment of continuity in the contemporary sediment cascade. Chapter 4 outlines the methodology developed to interpret the sedimentary signal, which identifies distinct components of the lake sediment record and documents their source. Chapter 5 presents the temporal trends and develops a descriptive model of palaeoenvironmental change which is assessed for consistency with regional data from more conventionally used lines of evidence in Chapter 6. Chapter 7 presents the methodological and substantive conclusions of the study. Appendix I presents the lithostratigraphies of lake cores and their cross-correlation, Appendix II the techniques used to develop a chronology, and Appendix III l i s t s data and results. Details of methods are introduced into the text where appropriate rather than being separted into a separate chapter i n order to make the arguments for the sequencing of the analyses undertaken i n the study clearer. 10 CHAPTER 2 THE STUDY SITE This chapter presents the r a t i o n a l e f or the s e l e c t i o n of the study area, i t s physiography, geology, vegetation, hydroclimate and contemporary geomorphology. The dominant sediment sources and t r a n s f e r s that d e l i v e r sediment to each of the lakes studied are i d e n t i f i e d i n order to provide the framework pursued i n Chapter 4 f o r the c h a r a c t e r i z a t i o n of sediment sources. 2.1 Se l e c t i o n of the Kwoiek Creek watershed Not a l l lakes provide equally continuous or s e n s i t i v e records of environmental change. Therefore the s e l e c t i o n of the study s i t e was a c r i t i c a l step i n the i n v e s t i g a t i o n . The Kwoiek Creek watershed ( l a t i t u d e 50°03'N to 50°12'N, longitude 121°35'W to 122°00'W) (Figure 2.1) was i n i t i a l l y selected f or three reasons. F i r s t , within the watershed there are two chains of lakes that are g l a c i a l l y fed, and through which, i n p r i n c i p l e , i t i s p o s s i b l e to document changes i n sedimentation downstream from g l a c i a l to n o n - g l a c i a l environments. Second, approximately 10% of the catchment i s c u r r e n t l y g l a c i e r i z e d , and g l a c i e r s would be expected to have existed throughout the Holocene Epoch (as w i l l be demonstrated l a t e r ) . T h ird, the drainage basin i s r e l a t i v e l y a c c e s s i b l e as a consequence of extensive logging a c t i v i t y i n the l a s t two decades. This study required a watershed with a strong g l a c i a l s i g n a l . For t h i s reason the southern p o r t i o n of the drainage basin, fed by Kwoiek G l a c i e r , was selected for i n v e s t i g a t i o n (Figure 2.1). I t was a l s o important that sedimentation be continuous over the Holocene Epoch with no unconformities i n the lake sediment record. This i s more l i k e l y i n lakes constrained at t h e i r o u t l e t by a rock s i l l , than i n those dammed by a moraine, or a c o l l u v i a l or a l l u v i a l fan. Fans may have blocked the v a l l e y , and thus Figure 2.1 The Kwoiek Creek watershed 11 12 formed the lake, at any stage during the Holocene. Furthermore they may have been susceptible to episodic failure resulting i n lake drainage, erosion of the sediments, and unconformities in the record. In addition, the resolution and length of the sedimentary record encompassing the Holocene is important, given the l o g i s t i c a l constraints of retrieving long cores and the d i f f i c u l t i e s of working in very deep lakes, within a mountain watershed. Ideally, lakes with steep-sides and f l a t bottoms, conditions conducive to minimum post-depositional slumping, and lakes which do not mix to their base, thereby minimising post-depositional disturbance of the sediments, are desirable. For a l l these reasons the chain of lakes Kha, Klept, John George and Kwoiek was selected. Kokwaskey Lake was also studied in an attempt to document the signal from the Haynon, Chochiwa, Kokwaskey Lake chain (see Figure 2.1) and to isolate this chains effect on the sedimentary record of the d i s t a l Kwoiek Lake. 2.2 The catchment 2 The Kwoiek Creek watershed i s a 250 km drainage basin, located on the eastern side of the Coast Mountains (Figure 2.1). Approximately 40% of the watershed currently i s above tree line and 10% i s glacierized. Kwoiek Creek flows a linear distance of 32 km, west to east, originating in the Kwoiek and Chochiwa glaciers, with a number of smaller nonglacial tributaries, the most important of which are shown on Figure 2.1. The topography of the watershed i s characteristic of the eastern Coast Mountains. It has high relative r e l i e f , with most summits and ridge crests above 2000 m. The highest peak i s Skihist Mountain (2944 m), on the divide with the Stein watershed; the lowest point, approximately 150 m, the valley floor at the confluence with Fraser River. The valley-side slopes are steep, average gradient 31°, (0.6 mm 1) (determined randomly using a 13 1 cm grid overlay) although they may be much steeper over short distances, with vertical rock faces common. 2.3 Geology Bedrock of the Lytton map area was mapped by Duffell and McTaggart (1952) and Monger (1980-1982). The generalised geology for the Kwoiek Creek watershed i s shown in Figure 2.2. Much of the Kwoiek Creek basin i s underlain by coarse grained granodiorite of the Coast Plutonic Complex. The granodiorite i s not homogeneous, but varies with regard to mineralogy, content and lithology of inclusions (granite, schist, and others), and degree of fo l i a t i o n . It i s typically massive with widely spaced joints. Mechanical weathering produces large, blocky fragments and extremely coarse debris. In some places sheet structures resulting from unloading are well developed and give rise to smooth, slightly convex rock faces upon which exfoliation i s occurring, for example, the northern valleyside in the Kwoiek Creek watershed, upstream from Fraser River. Areas of the drainage basin are underlain by low-grade metamorphosed st r a t i f i e d rocks. The largest of which are belts of northwest-southeast striking rocks of the Relay Mountain Group, consisting primarily of thinly bedded or laminated, sli g h t l y dipping phy l l i t e , a r g i l l i t e , conglomerate and greywacke, and low grade greenschist of the Bridge River complex (Figure 2.2) (for f u l l description see Monger 1982). In most places rounded ridges and summits have developed on these rocks, but some fresh glacial forms persist, for example, Kwoiek Needle. 2.4 Physiography and regional glacial history The extensive plateaus north of Stein Mountain, as well as the accordant summits to the northwest of the catchment are r e l i c t s of the original 14 Figure 2.2 Bedrock geology of the Kwoiek Creek watershed (Source information: Monger, 1982; GSC Open F i l e 980, Ashcroft) 15 uplifted land surface of low re l i e f that was dissected by late Tertiary rivers (Mathews, 1968). Evidence of glac i a l erosion within the catchment is marked. Most valleys are typically g l a c i a l troughs, with cirques, horns, aretes and tarns common above 1800m. At the onset of Fraser glaciation the Kwoiek i c e f i e l d area probably was an ice accumulation and dispersal zone. Advancing glaciers flowed eastward to Fraser River valley. During the glac i a l maximum, however, ice was deflected to the south or southeast across the Coast Mountains by a broad zone of southerly and southeasterly flowing ice that occupied the Fraser and Thompson plateaus. The presence of erratics i n the Coast Mountains near Kwoiek Creek drainage basin indicates the surface of Pleistocene ice sheets at elevations up to 2300 to 2500 m (Duffell and McTaggart, 1952). At this time the flow directions were controlled more by ice sheet surface topography than by the underlying valleys (Davis & Mathews, 1944). Because of their proximity to the ice divide the glaciers in the v i c i n i t y of the Kwoiek watershed during Pleistocene glaciations were relatively incapable of extensive erosion or transportation at the time of the glacial maximum. Deglaciation in this area probably was characterised by recession of active valley glaciers, not downwasting, which predominated further from the source (Fulton, 1971). A more precise chronology of deglaciation i s discussed later in the context of new data collected as part of this study (section 6.1). Glacial fluctuations during the Holocene Epoch have been documented for a number of locations in the southern Coast Mountains (Ryder and Thomson, 1986) based on moraine chronologies (see more detailed discussion in Chapter 6). Evidence for an early Neoglacial advance (Garibaldi Phase) approximately 5000 years B.P. in southwestern British Columbia has been ) \ 16 found only in the mountains of the Garibaldi Park region. Investigators have documented a second advance between 3000 and 1800 years ago (see for example, Ryder & Thomson, 1986). This varied in intensity, duration and timing between regions but i t i s evident throughout the coastal region from Alaska to Oregon. Ice advances that commenced about 1000 years B.P., with maximum positions attained in the nineteenth century, were the most pronounced during the Holocene Epoch in most regions. A regionally synchronous response of glaciers seems to have occurred during recession from these L i t t l e Ice Age maxima, beginning 100 to 350 years ago. Only the moraines of the most recent Neoglacial event are found in Kwoiek Creek watershed. The Holocene Epoch Neoglacial fluctuations are documented in greater detail i n Chapter 6 (Table 6.3). 2.5 Contemporary climate The watershed l i e s in the lee of the Coast Mountains, climatically intermediate between the mountain and interior systems. The relative a r i d i t y of the eastern side of the Coast Mountains dampens the glacial a c t i v i t y seen in the Pacific Ranges further to the west. Precipitation decreases eastward and increases with elevation. No climate data have been collected from within the watershed but the range of precipitation and temperature can be estimated from stations i n the region (see Table 2.1; locations shown on Figure 2.3). Extensive glaciers and snow fields l i e in the southwest of the Kwoiek Creek watershed, where a relatively large neve 2 with several glacier tongues occupies an area of approximately 26 km . Winter frontal precipitation predominates. Compounding the effects of synoptic-scale circulation are local and regional topographic controls on temperature and precipitation. No river gauges are operative within the watershed. The hydrology i s 17 Table 2.1 A summary of selected climate s t a t i s t i c s for stations closest to the Kwoiek Creek watershed Station Mean T Max T Min T Rain Snow Total S.D. (°C) 1 (°C) 2 (°C) 3 (mm) (mm) / \ 6 (mm) (mm)7 Alta Lake 5.7 22.9 -7.7 836.9 608.6 1420.3 226.1 (50 09° N 122 59° W 59 m) 8 Bralorne 4.1 22.7 -12.0 375.3 271.3 636.3 132.1 (50 47° N 122 49° W 1015 m) Chilliwack 10.2 24.3 -1.5 1750.7 129.4 1880.4 349.0 (49 07° N 122 06° W 6 m) Hells Gate 9.2 26.8 -4.8 1009.1 188.5 1198.9 201.8 (49 47° N 121 27° W 122 m) Hope 9.7 24.4 -3.1 1539.5 192.6 1715.8 346.9 (49 22° N 121 29° W 39 m) Lytton 10.0 29.4 -6.3 326.8 162.5 467.3 80.4 (50 14° N 121 34° W 175 m) Pemberton Meadows 7.0 26.2 -9.0 638.3 283.4 990.2 184.0 (50 27° N 122 56°W 223 m) 1 Mean annual temperature - average of the 12 monthly values 2 Mean monthly maximum - mean of a l l daily maximum temperatures recorded for a particular month (in a l l cases data for July) 3 Mean monthly minimum - mean of a l l daily minimum temperatures recorded for a particular month (in a l l cases data for January) 4 Rainfall (mm) 5 Snowfall - water equivalent (mm) 6 Total precipitation (mm) 7 Standard deviation of precipitation 8 Station location - Latitude, Longitude, Elevation Data from Environment Canada (1982) Canadian Climate normals, 1951-1980 Figure 2.3 Location of climate stations referred to i n Table 2.1 1. Alta Lake 2. Bralorne 3. Chiliwack 4. Hells Gate 5. Hope 6. Lytton 7. Pemberton Meadows 19 strongly influenced by spring snowmelt, modulated by storage i n the lakes, the levels of which fluctuate approximately 0.3 - 0.5 m over the course of a year. Discharge peaks in late April to July due to snowmelt. An estimate of maximum snowmelt, assuming a l l available radiant energy i s u t i l i s e d (further assumptions outlined by Church, 1988), i s of the order 43 mm/day, or a discharge 75 m"^  s 1 at the head of Kwoiek Lake. This i s undoubtedly a very high estimate for the catchment given the steep slopes and tree 3 -1 cover. An estimate of 50 m s may be more r e a l i s t i c . A mean annual flood 3 - 1 3 - 1 of 22.5 m s and a maximum daily flow of 97.5 m s for flows entering Kwoiek Lake were computed using regional envelope curves for the Cascade Mountains (Church, personal communication). The curves for the Cascade Mountains were selected to be more representative than those for the Coast Mountains because of the location of the Kwoiek Creek watershed i n the lee of the Coast Mountains. Undoubtedly these values are overestimates for Kwoiek Creek because of the storage effects of the lakes. 2.6 Vegetation Three major biogeoclimatic zones can be identified within the watershed (see Figure 2.4). Source information i s derived from Ministry of Forests, Research Branch 1:125 000 map sheets of biogeoclimatic zones, and Ministry of Forests, Inventory Branch, 1:20 000 Forest Cover maps. In the valley bottom subcontinental Interior Douglas Fir (Pseudotsuga menziesii) predominates (IDF zone) with Engelmann spruce (Picea engelmannii), ponderosa pine (Pinus ponderosa), lodgepole pine (Pinus contorta), mountain hemlock (Tsuga mertensiana), western hemlock (Tsuga  heterophylla), birch (Betula) and Populus, with an understorey of Ribes, Acer circinatum, Paxistima myrsinites, Arctostaphylos uva-ursi, and Ceanothus velutinus. At mid elevations subcontinental Engelmann spruce and 20 Figure 2.4 Vegetation cover of the Kwoiek Creek watershed (Source information: M i n i s t r y of Forests, Forest Service Research Branch, Biogeoclimatic units, J u l y 1980, Lytton Sheet, 1:125 000). 21 subalpine f i r (Abies lasiocarpa) predominates (ESSF zone), with lodgepole pine and Douglas f i r , and an understorey of Vaccinium, Ribes, Rhododendron, Equisetum, Juniperus, Valerian sitchensis, Paxistima  myrsinites, Veratrum v i r i d e . Tree line elevation i s approximately 1700 m, but i t varies with changes in aspect, slope s t a b i l i t y and microclimate and may be as high as 2000 m on south-facing slopes. At the highest elevations in the basin there i s an association of alpine tundra/ englemann spruce/ subalpine f i r parkland, with Lupinus arcticus, Potentilla f l a b e l l i f o l i a , P u l s a t i l l a occidentalis, Phyllodoce empetriformis, Erigeron peregrinus, Valerian sitchensis. Logged slopes are revegetated primarily by shrubs, herbs, bryophytes and young conifers, while logged stream banks of valley fla t s support a dense cover of black cottonwood (Populus balsamifera) and alder (Alnus rubra) interspersed with young conifers and an understorey of shrubs and herbs. 2.7 Contemporary geomorphology and sediment transfer processes As a prerequisite to interpreting changes in lake basin sedimentation the sediment sources and processes of sediment transfer within the watershed were examined. A map of s u r f i c i a l materials is presented (Figure 2.5), adapted from Ryder (1976), from which i t is possible to identify primary sediment sources and transfer processes delivering sediment to the lakes. Map units (see Figure 2.5) are delimited according to the character of the s u r f i c i a l materials. The emphasis i s thus on sediment stores and their genesis. Materials are divided according to the mode of origin (genesis) and surface expression in accordance with the system developed by the B.C. Ministry of Environment (Ryder & Howes, 1984). Essentially four units are identified on Figure 2.5: rock outcrops/cliffs; talus/debris cones; c o l l u v i a l blankets and veneers; and moraine/till. F i g u r e 2.5 S u r f i c i a l g e o l o g y o f t h e Kwoiek C r e e k w a t e r s h e d ( a d a p t e d f r o m R y d e r , 1976) 22 23 Rock outcrops/cliffs refers to bedrock outcrops or sites where rock i s covered by only a thin mantle of d r i f t or colluvium. Colluvium includes most postglacial accumulations of mass wasted materials. This unit i s subdivided into talus/debris cones and blankets/ veneers on the basis of surface expression. The talus/debris cones are composed of coarse angular fragments which have either fallen or crept downslope and usually have an obvious source above. In the Kwoiek Creek watershed the veneers/blankets of c o l l u v i a l material typically occur on steeper (20° to 33°) slopes than those covered by t i l l . The local mantle of t i l l i s of local provenance. Generally i t i s discontinuous and restricted to valley-sides at relatively low elevation and gradients, but in many places i t underlies avalanche deposits and rubbly postglacial colluvium on slopes of intermediate elevation and steepness (for example, in the v i c i n i t y of Kha and Klept Lakes). Granitic rocks underlying the headwaters of the watershed have given rise to coarse grained t i l l s and moraines that are pale grey to white in colour, with subrounded boulders and cobbles i n a matrix of gritty sand and minor s i l t . L i t t l e Ice Age moraines at the head of the Kwoiek Creek watershed are composed of blocks up to 1 m in diameter. V a l l e y - f i l l includes gla c i a l , g l a c i o - f l u v i a l and a l l u v i a l facies. Prominent late glacial recessional moraines are not present. In the v i c i n i t y of creeks, especially on the north side of the watershed, ice-contact gravels are evident. In other locations crude bedding indicates material of glacio-fluvial origin. Given the resolution of the terrain map drawn in this study these are not mapped as a separate unit and are found primarily i n those areas mapped as moraine/till. The a l l u v i a l materials are moderately well sorted and consist chiefly of bedded gravel with i n t e r s t i t i a l sand. Clasts are rounded or subrounded and vary in size from pebbles to boulders, with sand and minor s i l t interbedded. The stream channels are floored with cobbles and gravels. The s u r f i c i a l materials sustain poor s o i l development with orthic regosols to poorly developed eutric brunisols present. Three generalised sources of sediment for the lakes can be identified: recent ( L i t t l e Ice age) glacial deposits; i n situ weathered bedrock (eluvium) which may be mobilised (colluvium); and, "older" c l a s t i c material (for example, t i l l , g l a c i o - f l u v i a l deposits) which mantle the hillslopes and i n f i l l gullies, which were deposited during the late Pleistocene Epoch. These materials may be mobilised by mass wasting (rockfall/rockslide, avalanche, debris flow, gully wash, s o i l creep) and f l u v i a l processes, and delivered to the f l u v i a l system and thus to the lakes. Evidence of aeolian transport and deposits i s rare. An important common characteristic of a l l the sediment sources is their coarse sand to s i l t texture. Storage of material occurs in small fresh morainal ridges close to present glacier termini, c o l l u v i a l deposits, Fraser Glaciation d r i f t , floodplains and partly i n f i l l e d lakes. Associated with each sedimentary environment are specific chemical and physical processes that impart distinct characteristics to the sediments. If identified, these can be used to reconstruct, quantify and interpret provenance of material. This approach i s exploited i n Chapter 4 in order to characterise the sources and transfer pathways of sediments deposited in the lakes. 2.8 Anthropogenic a c t i v i t y The major anthropogenic disturbance within the catchment that has influenced the flux of c l a s t i c sediment has been logging by BC Forest 25 Products (Fletcher Challenge) since 1971. The chronology of logging road construction i s presented i n Appendix I I , as i t provides a means of dating disturbed sediments within the lakes. 26 CHAPTER 3 COLLECTION OF FIELD DATA AND PRESCREENING OF CORES This chapter presents information on the contemporary physical limnology and sedimentation patterns i n the lakes, the selection of sampling sites to obtain the sediment cores, the coring procedures, and the methodology subsequently used to analyse the cores. The methodology followed was similar for a l l lakes; any specific differences are noted. 3.1 The lakes Physical sedimentation in small mountain lake basins i s influenced by a large number of variables. These include basin morphometry, discharge and density of inflow, quantity and v a r i a b i l i t y of sediment input, circulation and v e r t i c a l density structure of the water body, presence or absence of winter ice, and biological factors such as burrowing and sediment-ingesting organisms. Data, in various forms, were collected for each of these characteristics i n order to describe the contemporary limnology of the lakes and to learn something of present day dispersal and sedimentation patterns, and the nature of the continuities in the contemporary sediment system. 3.1.1 Morphometry The bathymetry of each of the lakes was mapped with a 50 kHz RayJeff MX2550 echo sounder. Multiple transects were run in each lake, with navigation by "dead reckoning" between known points identified on aerial photographs. Intersecting transects were made wherever possible to cross check p r o f i l e locations by comparing similar bottom features and configurations. The morphometric characteristics of each of the lakes are presented i n Figures 3.1 to 3.4 and Table 3.1. The areas and volumes of the lakes shown in Table 3.1 were determined by planimetry of the e 3.1 Bathymetry of Kha Lake and c o r i n g l o c a t i o n s F i g u r e 3.2 Bathymetry of Klept Lake and c o r i n g l o c a t i o n s K l e p t L a k e Figure 3.3 Coring and Ekrnan sampling s i t e s w i t h i n Kokwaskey Lake. Problems with sounding equipment prevent bathymetry of the lake from being determined. The values next to each of the core s i t e s i n d i c a t e water depths determined with a plumb l i n e . Table 3.1 Physical characteristics of the lakes Kha Klept Kokwaskey Kwoiek Lake Lake Lake Lake Latitude 50 Longitude 122' 3 2 Area (xlO m ) Max length (m) Max width (m) Max depth (m) Mean depth (m) Volume (xl0 3m 3) Altitude* (m a.s.l.) 2 Contributing area (km ) # Residence time (h) 08'00"N 50°08'00"N 52'45"W 122°51,45"W 83.5 89.4 440.0 460.0 290.0 230.0 10.6 9.7 4.9 4.5 415.7 405.1 1120 1120 54.0 62.0 7.7 7.4 50°07'00"N 50O07'30"N 122°50'00"W 122O45'10"W 406.6 244.6 1279.0 1062.5 460.0 300.0 - 7.1 - 2.7 - 646.4 1050 835 42.0 152.25 _ 7.9 This number i s rather low because of the extensive deltaic area included in the computations + At lake outlet Minimum residence time based on mean annual flood derived from regional envelope curves 1965-1984 (see section 2.5). The mean annual flood was used for residence time computations because most sediment enters the lakes during the spring freshet. N.B. The values for Kokwaskey Lake are incomplete because of problems with the sounding equipment. Given the apparent size of the lake the residence time would be expected to be longer than for the other 3 lakes. 32 bathymetric maps. The residence time computations are based on the discharges calculated for the mean annual flood obtained from regional envelope curves (section 2.5). In alpine systems of the Coast Mountains of southern British Columbia most sediment i s moved through the f l u v i a l system during the spring freshet. Given the way the mean annual flood was calculated there are some inherent uncertainties in the calculations. However, what is important in this study i s the order of magnitude of the residence times for each of the lakes, i.e., that for most of the year they are of the order of hours to days and not longer. Simple calculations of settling times based on Stokes' law indicate that for Kwoiek Lake, assuming a non-stratified water column, material >16 pm would settle out during the mean annual flood. Finer material would be deposited during the rest of the year when residence times are greater. Kwoiek and Kokwaskey lakes are impounded at their outlets by bedrock s i l l s , and Klept and Kha by fans which encroached across the valley. These fans are now relatively inactive, with no evidence of recent deposition on their now well vegetated surfaces. A l l the lakes are aligned along the valley topographic axis, which has important implications for orientation relative to prevailing katabatic winds and thus mixing and dispersal patterns. In cross section, a l l of the lakes are relatively steep sided with a f l a t floor. They are a l l composed of one major basin, although in Kwoiek Lake there i s a pronounced longitudinal trough (see Figure 3.4) which i s important i n controlling the distribution of sediments. In two of the lakes, Kwoiek and Kokwaskey, a well developed headwater delta occupies the western end. In a l l of the lakes, with the exception of Kokwaskey which i s fed through both the Chochiwa Glacier chain and the Kwoiek Creek chain, the majority of the water and sediment enters through Kwoiek Creek, with small side streams having no discernible effect on lake temperature or turbidity profiles. 3.1.2 Vertical density structure and circulation A l l the data presented on the physical limnology of the lakes are intended solely for descriptive purposes. The emphasis i s on the order of magnitude of the various characteristics, not the specific values per se. Although collected in a standard manner the data were not collected on a continuous basis, but when trips were made into the f i e l d between 1986 and 1988. There are no marked differences between the years. Data are presented for 1986 to i l l u s t r a t e important trends. Thermal st r a t i f i c a t i o n plays an important role i n the distribution of fine sediment entering any lake because i t influences the pattern i n which inflowing water mixes with the lake water, thereby affecting the subsequent transport and deposition of a l l but the coarsest bedload. The highly seasonal and weather-dependent nature of glacial-river discharge, temperature and suspended sediment concentration variations, together with the normal seasonal evolution of lake thermal structure, results in changing and often complex mixing patterns at different times of the year. Temperature profiles are documented and summarised for the lakes of the Kwoiek Creek watershed (Figure 3.5). A l l the lakes may be c l a s s i f i e d as cold dimictic or polymictic. In general, the lakes have weak thermal stra t i f i c a t i o n during the summer months (least well developed i n Kha lake and not developed u n t i l late i n the summer in Klept Lake), becoming isothermal i n the f a l l and spring, resulting in f u l l lake turnover. The lakes are ice covered during winter (November to A p r i l ) . The temperature profiles at this time are unknown. Figure 3.5 Thermal profiles for the lakes of the Kwoiek Creek watershed. Data from 1986. Note different depth scales. a) Kha Lake 1: June 28 2: July 17 3: August 12 4: October 2 b) Klept Lake 1: June 16 2: July 17 3: August 12 4: October 2 'errpero(ure (C) o* » o Tcmoerolure 'C) c) Kokwaskey Lake 1: June 17 2: July 16 3: August 14 4: October 3 d) Kwoiek Lake 1: June 15 2: July 15 3: August 13 4: October 4 35 Thermal effects on water density are usually overwhelmed by turbidity differences during maximum melt periods. In large lakes, the intervals of strongest underflows are often triggered by high suspended sediment loads. In small lakes with small inflows, strong underflows may be triggered by anomalously high suspended sediment loads supplied suddenly to the stream, for example, as a consequence of bank collapse or colluvial a c t i v i t y . For the lakes of the Kwoiek Creek watershed sediment inflow was not monitored on an annual basis. A knowledge of the sediment within the water column was obtained through v e r t i c a l profiles at a number of stations within each lake, illustrated i n Figure 3.6. The data are quite variable, with a strong attenuation in concentration downsystem. Ranges of depth-integrated values are presented in Table 3.2. When compared with data for g l a c i a l , alpine lakes the values are generally low (see Table 3.2). This i s consistent with the fact that the lakes studied in the Kwoiek Creek watershed are not ice contact lakes and other sedimentary stores exist within the watershed. There i s a strong, seasonal, May to July, snow-melt pulse of sediment into the system, documented i n the measurements and visually observed in the f i e l d . 3.2 Selection of core sites Likens & Davis (1975) introduced the term "sediment focusing" to describe the phenomenon of greater sediment accumulation in the deepest part of lakes. Much work has been directed towards predicting where the sediment w i l l be focused (see for example Hilton, 1985), but this has met with limited success. A strategy of sub-bottom sounding and multiple coring enables definition of lake-wide changes i n sedimentation patterns which permits an assessment of the extent to which observed variations represent large scale changes in sediment input, rather than localised 36 Figure 3.6 Suspended sediment profiles for the four lakes under investigation. Data from 1986. Note different scales a) Kha Lake 1: June 28 b) Klept Lake 1: June 16 2: August 12 2: August 12 3: October 2 3: October 2 c) Kokwaskey Lake: 1. June 17 d) Kwoiek Lake (Proximal) 1: June 15 2. August 14 2: October 4 3. October 3 (Distal) 1: June 15 2: October 4 Sediment (mg I) Table 3.2 Maximum recorded concentration of suspended particulate matter in the waters of g l a c i a l lakes (adapted from Gilbert & Desloges 1987). Lake Concentration Reference (mg l " 1 ) Tasikutaaq Lake, Baffin Island Hector Lake, Alberta Bow Lake, Alberta Lower Waterfowl Lake, Alberta Ape Lake, B.C. (before draining)* Garibaldi Lake, B.C. Lillooet Lake, B.C. Lake Wakatipu, N.Z. Stewart Lakes, Baffin Island* Cascade Lake, Washington* Hazard Lake, Yukon Unnamed lake, Purcell Mtns, B.C. Sunwapta Lake, Alberta Malaspina Lake, Alaska This study + Kha lake - Max - Min Klept lake - Max - Min Kokwaskey lake - Max - Min Kwoiek lake - Max - Min 5.7 Lemmen (1984) >8.0 Smith et a l (1982) 15.0 Smith et a l (1982) 17.2 Smith et a l (1982) 28.7 Gilbert & Desloges (1987) 30.0 Mathews (1956) 49.0 Gilbert (1975) 57.0 P i c k r i l l & Irwin (1983) 109.0 Gilbert et a l (1985) 110.0 Campbell (1973) 120.0 Liverman (1980) 140.0 Weirich (1985) 380.0 Gilbert & Shaw (1981) 723.0 Gustavson (1975) 22.2 13.8 21.2 12.7 19.1 17.3 9.0 6.2 * ice contact lake + from water profile data presented in Figure 3.6 max: maximum integrated value documented in measurements; min: minimum integrated value documented i n measurements Obviously these do not represent the true maximum and minimum values for the year, rather the range for the period of measurements. 38 changes. Thus as a precursor to coring, bathymetric sounding with a 50 kHz echo sounder, and sub-bottom sounding at 3.5 and 7 kHz were conducted with the objective of describing the bathymetric and sedimentary characteristics of each lake. There were two reasons for determining the sub-bottom stratigraphy of the lakes. F i r s t , to provide a basis for selecting coring sites in zones of greatest accumulation and potentially greatest resolution, and second, once the cores were obtained, to permit cross-correlation of results between point cores, where distinct sediment layers could be observed in both cores and sub-bottom traces. Sub-bottom sediment profiles were obtained using a Raytheon RTT-1000A, portable, low-frequency sonar. Unfortunately for reasons that could not be resolved the sub-bottom sounding was largely unsuccessful. The results are not pursued further in this study. 3.3 Lake sediment sampling Surface samples were collected with an Ekman sampler at multiple sites i n each of the lakes (Figures 3.1 - 3.4) i n order to document spatial v a r i a b i l i t y in the facies and to determine coring sites. An undisturbed core from each surface sample was preserved in a 0.12 m diameter t i n and allowed to become p a r t i a l l y dry to strengthen and consolidate the sediment before shipment. The s i l t y nature of most of the sediments allowed approximately the upper 20 cm to be recovered. Based on a consideration of the patterns of sedimentation determined by the Ekman samples and bathymetry, multiple cores were taken from each lake (for locations see Figures 3.1 - 3.4). In this study any sample referred to with the prefix E was taken with an Ekman sampler, C i s from a core, with the core/Ekman number referenced to Figures 3.1 to 3.4, and the depth 39 from which the sample was taken indicated. The abbreviation KWK refers to Kwoiek Lake, KOK to Kokwaskey Lake, KLP to Klept Lake and Kha i s represented in f u l l , KHA. The corer that was used i s a modified Wright (1967) square-rod piston corer built at UBC's Department of Geophysics & Astronomy, loaned by Dr. G.K.C. Clarke. The corer uses one of two 2.375 inch (6.03 cm) outer diameter (OD), 2.25 inch (5.72 cm) inner diameter (ID), 1/16 inch (0.16 cm) wall stainless steel tubes, 1.20 m long, one with a tempered steel serrated cutting edge, and one with a tempered bevelled cutting edge. Another similar core tube made of clear plexiglass, 2.25 inch (5.72 cm) ID and 1.20 m long which can be used in place of the stainless steel tubes proved useful for coring the surface sediments. Drive rods are 5 foot (1.52 m) lengths of 1 5/16 inch (3.33 cm) diameter zirconium-magnesium joined with coarsely threaded stainless steel couplings. These are standard rotary d r i l l rods selected for r i g i d i t y and light weight. A T-extension handle was made to f i t these. A d r i l l i n g platform 8' by 6'(2.44 m by 1.83 m) was built from 2 pieces of plywood and styrofoam floats. A d r i l l i n g hole cut in the centre enabled work around the corer. Four anchors were deployed from the raft, one at each corner, to ensure a stable, immobile platform. Casing was made from 4 inch (10 cm) diameter ABS piping cut into 1, 2, and 5 foot (0.30, 0.61 and 1.52 m) lengths and f i t t e d with ABS couplings so i t could be assembled to the appropriate length for each d r i l l hole. Samples were removed i n one metre sections from a continuous borehole. A careful record was made of drive lengths and the length of core recovered. Discrepancies may be due to compression of sediment during the coring procedure, sediment lost from the corer bottom, and/or bore hole i n f i l l i n g between drives. At each site 40 two cores were taken within 2 m of each other. This meant that enough material was collected for a l l the analyses, and no part of the record was lost. The latter was achieved by ensuring that the depths of breaks in the core were not the same i n adjacent cores i.e. that the drive lengths were different. The longest core obtained was 4.24 m (laboratory measurement) from Kokwaskey Lake (K0K5). In total 60 cores were collected from 30 sites (i.e. 2 at each " s i t e " ) . The locations of each site are shown on Figures 3.1 to 3.4. 3.4 Description of the cores After extrusion in the f i e l d and measurement of their lengths, the cores were wrapped in saran wrap and aluminum f o i l and transported in a core box to the laboratory where they were s p l i t length-wise. The cores were then air dried with careful observation during this process of stratigraphic units, contact types, sedimentary structures, texture, colour and other distinctive features, for example, inclusions of organics. Some cores were dried for 2 to 3 weeks before details of a l l structures and units became evident.- Similar facies were repeated i n the cores in a l l of the lakes. As a precursor to selecting cores for more detailed analysis, a systematic characterization was undertaken based on a r i g i d l y defined lithofacies scheme. Visual appraisal of grain size, bedding and sedimentary structures was used to subdivide the sedimentary sequence as outlined by Eyles et a l . (1983) and adapted by Schmok (1986). This scheme and detailed logs of the cores are presented i n Appendix I. X-radiography, a fast and non-destructive scanning and recording technique, was undertaken to find sediment structures that may not be vi s i b l e i n the dried core. The radiographs were made in the Department of Radiology of the University of Bri t i s h Columbia with instrument settings 41 100 kV, 400 mA, phototimer 1.2 ms, 40" FFD, QUI screens. The technique is particularly valuable for studying the upper portions of the cores ( 0 - 2 0 cm) which had a high water content. Radiograph observations were used to supplement the visual observations and are incorporated into the core descriptions (Appendix I). Central to any study documenting and interpreting changes through time is chronology. Radiocarbon dating, tephra identification and dated anthropogenic events in the sedimentary records were used to date the lake sediments. Appendix II outlines details of the techniques and results. A l l radiocarbon dates referred to i n this study are uncorrected C-14 dates. A l l the cores exhibit changes in their depositional records, the major features of which are alternations between massive and laminated s i l t s , with incorporations of organic matter. The majority of the deposits are layers of massive sediments composed of particles ranging i n size from fine sands to fine s i l t s . This facies may represent periods when the sedimentation rate was sufficiently low that the layers were deposited without internal structure. Alternatively any structures may have been destroyed subsequently by bioturbation. The thinly laminated sediments consist largely of darker coloured s i l t with laminations of coarser, lighter coloured s i l t and fine sands. These deposits were observed in most cores. Detailed observation of these structures revealed no apparent sorting within the laminations. These deposits seem to represent periods of s l i g h t l y greater sedimentation rates with pariodic introductions of coarser material into the lakes. The more regular laminations, the rhythmites, observed only in the upper protions of Kokwaskey Lake sediments, are identified as varves (Appendix II). There i s no evidence within any of the cores for flow structures, deformed beds or erosional 42 unconfomities. The boundaries between facies are largely gradational. The sediments of Kha Lake are predominantly laminated s i l t s . The basal sediments, which prevented a longer record from being obtained, are coarse sands and granules with particles up to 5 mm i n diameter. A radiocarbon date of 2350 + 110 yrs B.P. (S-2936) was obtained on wood incorporated in laminated s i l t s 17 cm above the contact between the coarse sands and overlying laminated s i l t s . From the core obtained at site Kha2 (Figure 3.1) approximately 4.05 m of sedimentation has occurred in the last 2350 years. Above the laminated s i l t s more massive s i l t s with occasional beds of coarser material predominate. The sediments from 1.5 m to approximately 0.5 m depth are laminated s i l t s , with massive s i l t s overlying these. The upper 10 cm of record i s distinct with many beds of coarser material. These are interpreted as the effect of logging and road construction i n the v i c i n i t y of the lake. Klept Lake i s characterised by more massive deposits with frequent inclusions of organic material. Spatially the records are more variable than i n Kha Lake, with many coarse beds of local origin. A basal date of 9640 + 380 years B.P. (S-3011) was obtained on wood incorporated in massive s i l t s approximately 17 cm above a contact with coarse sand. This i s overlain by 2.26 m of fine sand/ s i l t s which have been deposited during the Holocene Epoch. Two coherent tephra layers were identified in the sediments of Klept Lake, the lower at approximately 1.48 m depth, identified as Mazama tephra, the upper, approximately 0.52 m depth, identified as Bridge River tephra. The sediments of Kokwaskey Lake are more uniform spatially although variable with depth. The basal sediments are laminated blue-grey clays. Organic material contained within these has been dated at 11,485 + 185 43 years B.P. (S-2935), with approximately 4.15 m of sediment deposited over the postglacial period. The blue-grey clays are overlain by coarser sandy/silt deposits, which in the lower portions of the core are massive, becoming laminated i n the mid-section and increasingly finer towards the surface. Two clearly defined tephras are evident in the Kokwaskey Lake sediments, the lower Mazama tephra, the upper Bridge River tephra. The uppermost sediments, encompassing approximately the last 250 years, are varved (see Appendix I I ) . The deposits in Kwoiek Lake would appear to be more uniform with depth. The basal sediments are coarse sands with a transition to fine s i l t s mid-core, to laminated s i l t s more recently. Approximately 2.49 m of sediment has accumulated over the postglacial period. The basal date on these sediments i s 12,255 + 770 years B.P. (S-3010). There i s no Mazama tephra but Bridge River tephra is evident within the upper 0.60 m of sediments. The changes in sedimentary characteristics and their significance in each of the lakes through time are considered in much greater detail in subsequent chapters. The cores within each lake were cross-correlated on the basis of their lithostratigraphy. Given the focus of this study on basin-wide changes in 2 3 sedimentation over timescales 10 -10 years, preliminary screening of the data entailed discarding any beds which were obviously of local origin, for example graded beds due to localised mass wasting events at the edges of the lakes, and beds that only existed in one core. Specific cores were then selected for further study. These are identified where appropriate in the subsequent text. The cores were selected because they offer the greatest resolution and temporal coverage. However the shorter and/or younger cores have an important function in establishing spatial v a r i a b i l i t y of sedimentary characteristics and processes. 45 CHAPTER 4 INTERPRETATION OF THE LAKE SEDIMENT RECORD The basic premise of this study i s that during the Holocene Epoch 2 3 variations i n the alpine sediment system, at timescales 10 -10 years, have been largely controlled by fluctuations in Neoglacial a c t i v i t y and are recorded in the sediments of alpine lakes. This chapter presents a series of analyses to interpret the clastic component of the lake sedi-ments in order to gain insight into the nature of the sediment sources and transfer processes delivering sediment to each of the lakes. These analyses and interpretations are based on lake wide sedimentary charac-t e r i s t i c s that were identified for each of the lakes. The f i r s t portion of this chapter presents data and interpretations of the organic component as an index of the volume of sediment influx through time; the second part presents textural analyses of the clastic component in order to identify different c l a s t i c sediment populations, which through source id e n t i f i c a -tion are shown to contain distinct palaeoenvironmental information. These analyses are used i n Chapter 5 to document and interpret changes in the palaeo-sediment system through time, and to develop a chronology of Holocene palaeoenvironmental changes for the Kwoiek Creek watershed. 4.1 Organic matter contentt an inverse indicator of sedimentation rates Several workers have suggested that an inverse relation exists between cla s t i c sedimentation rate and organic carbon content of sediments in glacial lakes (Karlen, 1976; Surgenor,1978, Davis et a l . , 1979, Leonard, 1981). Similarly Oldfield (1977) notes that i n non-glacial lakes total carbon content of sediments, reflecting organic carbon content, bears an inverse relation to total sediment input. This relation may be used to establish a strong co-variation between the two variables, and therefore permit organic matter content, which i s easily measured, to be used as a 46 surrogate index of sedimentation rate. It i s proposed that the variations i n clastic sediment input into the lakes in the Kwoiek Creek watershed w i l l be closely related either to changes in rate of release of sediments by the glaciers, or to rates of re-entrainment of g l a c i a l , glacio-fluvial or coll u v i a l deposits by meltwater streams. In either case such material is relatively deficient in organic material and changes in input of this material would li k e l y be reflected inversely by changes in concentration of organic material, either autochthonous or allochthonous in origin. The relation may be complicated by the fact that the input into the lake of organic material from other sources may also be variable. However, the c r i t i c a l assumption i s that the organic input is less variable than c l a s t i c sediment input, and thus variations i n the organic content primarily reflect changes in cla s t i c sedimentation. In order to evaluate this assumption two independent tests were conducted. F i r s t , by using the upper sediments in Kokwaskey Lake, which are varved (see Appendix II) and therefore provide annual resolution of sedimentation rates; and second, comparing integrated values between dated tephra horizons within the cores of the lakes. Because of the number of samples that had to be analyzed, loss on ignition was used as a measure of total organic content i n this study (combustion of the organic material in a muffle furnace for 3 hours at 450°). Samples of known volume, which represent approximately 1 cm depth of core, were used to determine both loss on ignition and minerogenic bulk density (mass c l a s t i c material per unit volume). The results for the organic matter are reported as percent by dry weight. For two cores from the varved sediments of Kokwaskey Lake, samples were 47 taken for organic content determination every centimetre and compared with sedimentation rates in two cores K0K3 and K0K5 (see Figure 3.3). An average of 10 varves were represented in 1 cm of core, resulting in a temporal resolution for the test of the order 10 1 yrs over a time period of approximately the last 250 years, the depth of core for which the individual varves could be resolved. It was expected that organic matter content and sedimentation rate would be negatively correlated. If there i s a constant influx of organic material per year then the greater the annual sedimentation rate the lower the organic content per unit depth; conversely the lower the sedimentation rate the greater the organic content per unit depth. No significant variations i n minerogenic bulk density are evident in the upper-sections of the cores studied, so this effect was neglected. The most recent part of the record with greater sedimentation rates due to anthropogenic activity within the catchment was not used in this analysis. The data are plotted in Figure 4.1 (K0K3 *,K0K5 +). Regression lines and equations were derived independently for the two cores analyzed in order to account for spatial v a r i a b i l i t y i n the organic content. Of interest i s the consistency of the relation within a core. 2 Significant (a< 0.01) negative correlations (r = 0.90 and 0.81; n= 21 and n=23 respectively) were obtained for the two cores: Core K0K3: S.R. = -0.92 (O.M.) + 3.10 Core K0K5: S.R. = -0.72 (O.M.) + 2.68 S.R. sedimentation rate (mm yr 1) O.M. organic matter content (loss on ignition % by weight) In down-core tests there may be significant serial correlation. In order to account for this effect the residuals were tested for f i r s t order 48 Figure 4.1 Relation between sedimentation rate and organic matter content of varved sediments for Kokwaskey Lake serial autocorrelation, r (K0K3 r^O.15, n=20, not significant a 0.01; K0K5 r^O.53, n=23, significant a 0.01). The significance of serial correlation i s considered further i n section 5.3. To ensure the consistency of this relation over the longer time periods of interest in this study, a second test was conducted. Clastic and organic sedimentation rates were determined for each of the lakes from a l l of the cores that penetrated tephra layers. Comparison of c l a s t i c and organic sedimentation rates during the known time intervals between the lower and upper tephra layers, and the upper tephra and the present, provides a means of assessing which component of sedimentation remains more nearly constant over timescales of 10^ yrs. The upper-most portions of the cores, which record the effects of logging within the catchment were not included in this phase of the analysis. Clastic sedimentation rates were determined from the weighted average of minerogenic bulk density data for the cores between layers of known age. Organic sedimentation was determined for the same core sections by multiplying the total sedimentation rate for each section of the core by i t s organic content (percent by weight). This provides sedimentation rates for both -2 -1 the clastic and organic material which can be expressed as mg cm yr The data on influx rates and ratios of their v a r i a b i l i t y through, time are presented i n Table 4.1. The ratios of the v a r i a b i l i t y in c l a s t i c sedimentation are much greater than those for the organic material. This lends further support to the hypothesis that clastic sedimentation rates have varied more than organic sedimentation rates in the lakes of interest and thus organic matter content of the lake sediments can be used as an inverse index of rates of sedimentation. The flux of organic material does vary through time. However, i t i s l i k e l y that i t does so inversely with 50 Table 4.1 Assessment of v a r i a b i l i t y of clastic and organic sediment influx over timescales of the order of thousands of years Clastic / -2 -1, (mg cm yr ) Organic / "2 - I s (mg cm yr ) Klept Lake (KLP1) 0-2400 2400-6800 * Variability 0.0240 0.0186 1.29 (0.78) 0.0021 0.0019 1.11 (0.90) Kokwaskey Lake (K0K5) 0-2400 0.0429 2400-6800 Variability 0.0325 1.32 (0.76) 0.0018 0.0021 1.14 (0.88) * Variability i s defined as the ratio of the influx for the specified lake 0-2400 B.P. divided by that for 2400-6800 B.P. The value i n brackets i s the influx 2400-6800 divided by that 0-2400. N.B. The values reported are point influx rates calculated from individual cores. They do not represent areally weighted values of sedimentation rates in the lakes Only the tephra horizons were used as chronological markers in this analysis to reduce potential errors because of dating uncertainty with the radiocarbon dates. Hence values for Kha and Kwoiek Lakes are not presented. 51 glacier activity, thereby enhancing the patterns of interest in this study. The relation between organic matter and sedimentation rates i s used in Chapter 5 to document temporal v a r i a b i l i t y in sediment influx i n each lake. 4.2 Characterization of the cl a s t i c component Grain size characteristics of selected cores (Kha CI and C2? Klept CI and C6: Kokwaskey C3 and C5; Kwoiek C6 and C15) and surface samples from a l l the Ekmans, were determined through a combination of standard wet sieving techniques and use of a SediGraph 5000 analyzer (see Stein, 1985 for description). Samples of approximately 5-6 g, representing 1 cm depth of core were ligh t l y crushed with a pestle and mortar to break up large aggregates. To deflocculate the sediment, each sample was placed in approximately 50 ml of sodium hexametaphosphate solution and l e f t for 24 hours with periodic s t i r r i n g . Immediately prior to analysis each sample was placed in an ultra sonic bath to ensure complete disaggregation of material. The samples were wet sieved at 1/2<P intervals down to 63 pm, and then analysed using the SediGraph from 88 nm to 1 pm. The results were combined and cumulative grain size distributions plotted (see examples in Appendix III). The most common sediments are s i l t s , clayey s i l t s and clays with a clearly distinct group of deposits composed of sand and s i l t (Figure 4.2). From inspection of the cumulative distribution plots (see Appendix III) i t i s evident that many of the sediments are polymodal. Histograms and cumulative curves reveal complicated distributions with the preponderance of curves (samples) consistently showing breaks of slope around 4<t> and 6<P (see examples in Appendix III). Numerous attempts have been made to relate a particular cumulative log 52 Figure 4.2 Textural triangle for lake bottom sediments, Kwoiek Creek watershed. 53 p r o b a b i l i t y curve shape to a s p e c i f i c environment (see for example, S l y et a l . , 1983). Associated with t h i s have been attempts to quantify curve i n t e r p r e t a t i o n by u t i l i s i n g s t a t i s t i c a l parameters to characterise the curve and b i v a r i a t e p l o t s of s t a t i s t i c a l measures to d i s t i n g u i s h environments. However, given the polymodal nature of many of the g r a i n s i z e d i s t r i b u t i o n s i n t h i s study, such analyses of curve shape and s t a t i s t i c a l measures (such as skewness and kurtosis) may r e f l e c t only the r e l a t i v e magnitude and separation of modes and thus be of l i t t l e i n t e r p r e t i v e value. An a l t e r n a t i v e approach to the problem i s to separate the constituent populations and to r e l a t e each of these to sediment transfer processes or sources. I t has been suggested (see review i n Middleton, 1976) that when cumulative curves are p l o t t e d on a l o g - p r o b a b i l i t y scale, log-normal d i s t r i b u t i o n s appear as a s t r a i g h t l i n e , and the component d i s t r i b u t i o n s of bimodal and polymodal sediments appear as s t r a i g h t l i n e segments. There are three basic approaches to p a r t i t i o n i n g polymodal p r o b a b i l i t y curves: a n a l y t i c a l , numerical and graphical (Clark, 1976). Harding's (1949) gr a p h i c a l method provides a straightforward approach to p a r t i t i o n i n g d i s t r i b u t i o n s with up to four modes. The method i s best described by S i n c l a i r (1974). I n f l e c t i o n points on l o g - p r o b a b i l i t y p l o t s i n d i c a t e the approximate proportions of, or the f r a c t i o n ( f ) of, the t o t a l mixture that each mode represents. For example i n Figure 4.3, a case of a bimodal d i s t r i b u t i o n , the arrow i n d i c a t e s the point of i n f l e c t i o n at approximately the 50th cumulative p e r c e n t i l e , i n d i c a t i n g the presence of 50% of a coarser population C, and 50% of a f i n e r population F. When graphical p a r t i t i o n i n g i s conducted on curves p l o t t e d on l o g - p r o b a b i l i t y paper t h i s approach assumes that each subpopulation has a log-normal d i s t r i b u t i o n s . 54 Figure 4 .3 Example of separation of log-probability textural curves. This plot il l u s t r a t e s partitioning of a bimodal distribution with "Coarse" and "Fine" textural sub-populations intermixed. See text 55 An extensive literature exists in f l u v i a l geomorphology that supports this assumption for f l u v i a l l y transported sediment. Given the prescreening of the data to remove deposits of obviously local origin most materials remaining were f l u v i a l l y transported from their source. This i s especially true for the material of interest within the lakes which i s g l a c i a l l y -derived. This material must have been f l u v i a l l y transported for some distance in order to be deposited in the lakes and thus the assumption of log-normality would seem to be r e a l i s t i c . Thus graphical partitioning of log-probability grain size plots was conducted for samples (cores KHA2, KHA5, KLP1, KLP5, K0K3, K0K5, KWK14, KWK15). When the component populations were separated, i t became apparent that three well-defined populations are present, referred to subsequently as coarse (C) intermediate (I) and fine (F). These three component distributions were then analyzed for their mean, standard deviation and range. The higher order s t a t i s t i c s , skewness and kurtosis, were not computed because of the greater uncertainty in the separation of the t a i l s of the distributions which f a l l within the region of greatest graphical overlap. The average values for each core and their v a r i a b i l i t y are presented in Table 4.2. It can be seen quite clearly that-,there are three well defined modes: a coarser fraction (2.0 to 5.00 , range defined +2 standard deviations, medium to fine sand) was found in a l l lakes, although this population in Klept Lake i s finer and less well sorted; an intermediate fraction (3.0 to 7.0 <t>, fine sand to medium s i l t ) in Kha Lake and Klept Lakes; and a finer fraction (6.0 to 9.0<p, fine s i l t ) i n a l l lakes. The mean characteristics of each of the sediment populations for each lake are shown schematically in Figure 4.4. In a l l lakes unimodal deposits exhibit the same characteristics as the finest fraction. There appear to be no systematic 56 Table 4.2 Summary of average c h a r a c t e r i s t i c s of each of the three g r a i n s i z e populations by lake. C i n d i c a t e s coarse mode, I intermediate mode, and F f i n e s t mode Mode D 50 (<*) S.D. Kha Lake C I F 3.39 5.30 7.37 Klept Lake C 4.03 I 5.40 F 7.42 Kokwaskey Lake C 3.14 F 7.28 Kwoiek Lake C 3.50 F 7.87 0.31 0.62 0.71 0.75 0.69 0.56 0.44 0.82 0.27 0.93 Sorting (<t>) S.D sorting (<P) 0.81 0.78 0.90 0.14 0.25 0.33 8 7 23 1.09 0.70 0.78 0.46 0.19 0.31 15 8 22 1.18 0.53 0.44 0.23 7 32 0.76 0.79 0.15 0.30 7 29 Figure 4.4 Charactersti.es of sediment sub-populations. Synthetic grain-size curves to i l l u s t r a t e the mean distribution of each sediment sub-population from each lake. Kha Lake Sand • S i l t • Clay i \ i i \ < • ' \ \ i > 1 ; i . * • • ,* • > i > • r I j / \ 10 Grain size (<P) Klept Lake Sand S i l t Clay Kokwaskey Lake 0 1 2 3 4 5 6 7 8 9 10 Grain size (») S i l t Clay 7 8 9 10 Grain size (*) Kwoiek Lake Sand S i l t Clay 0 1 2 3 4 5 6 7 8 9 10 Grain size (<t>) 58 changes i n either the mean size or the sorting of the coarsest fraction down-system. There i s a slight fining of the finest mode but no systematic trend in sorting. The f i r s t step in the interpretation of the physical significance of the breaks in the log-probability plots was to ensure that they are not a consequence of the analytical procedures used. As outlined above, every effort was made in the laboratory to ensure that this was not the case. Each sample was sieved over the sand/silt break and the SediGraph analysis started at 88 pm to ensure overlap. In the sedimentological literature considerable discussion has been generated over the interpretation of different textural modes (see for example, Visher, 1969; Shea, 1974; Middleton, 1976). In general the size "breaks" or subpopulations may be related to either sources, mechanical breakage, or hydraulic sorting, either during transport or deposition. In general in the lakes of the Kwoiek Creek watershed, sediment i s deposited through settlement from suspension. No current structures, which are the only unequivocal evidence of underflows, are evident in the deposits (see Appendix I). This does not preclude weak underflows but suggests other transport mechanisms, such as overflows and interflows, and settling from suspension. Thus the differences i n textural modes are unlikely a result of different depositional processes within the lakes. A basic premise of this study i s that during the Holocene Epoch there have been significant fluctuations in the relative contributions of different sediment sources. It i s argued that the different sediment subpopulations are deposited concurrently and are thus a reflection of these sources. This hypothesis i s supported by the fact that even when viewed under the binocular microscope no distinct beds with sorting are 59 v i s i b l e which would i n d i c a t e hydraulic s o r t i n g , a transport c o n t r o l . Furthermore there i s l i t t l e temporal synchroneity i n the deposition of the coarsest f r a c t i o n i n the d i f f e r e n t lakes ( t h i s i s discussed i n greater d e t a i l i n Chapter 5). It would be expected that at the s p a t i a l scale of the Kwoiek Creek watershed high magnitude h y d r o l o g i c a l events, whether r e l a t e d to snowmelt or synoptic events, would a f f e c t the e n t i r e watershed and thus there would be temporal synchroneity i n terms of sedimentary c h a r a c t e r i s t i c s , e s p e c i a l l y of the coarser f r a c t i o n , i f hydrologic/ hydraulic processes were the c o n t r o l . Thus the argument of a source c o n t r o l over sediment c h a r a c t e r i s t i c s i s pursued. The next major sections (4.3 and 4.4) present the r e s u l t s of two techniques to determine the sources of the d i f f e r e n t sediment f r a c t i o n s : surface texture a n a l y s i s of the f i n e s t sand f r a c t i o n using the scanning e l e c t r o n microscope, which encompasses the two coarser modes; and analyses of the mineralogy of the f i n e s i l t - s i z e d f r a c t i o n , the f i n e r mode, i n order to document and i n t e r p r e t changes i n sources through time. These provide a foundation f o r the i n t e r p r e t a t i o n of the f l u c t u a t i o n s i n the f r a c t i o n s through time presented i n Chapter 5. 4.3 Sediment source a n a l y s i s : SEM a n a l y s i s of the coarse f r a c t i o n s One of the objectives i n t h i s study i s to determine the sources of sediment deposited i n the lakes, s p e c i f i c a l l y to d i s c r i m i n a t e between sediment derived o r i g i n a l l y from g l a c i a l sources, sediment from contemporary c o l l u v i a l sources, and sediment from o l d s u r f i c i a l deposits (of Fraser g l a c i a t i o n ) , a l l of which may have been reworked f l u v i a l l y (see Figure 1.1 and section 2.7 for d i s c u s s i o n of contemporary sediment sources and t r a n s f e r mechanisms). I t i s proposed that one technique with which 60 this can be achieved involves characterization of the surface features of quartz sand grains, commonly termed surface textures, within the lake deposits using the scanning electron microscope (SEM). The basic premise underlying work on grain surface texture i s that given necessary time within distinct sedimentary environments characteristic chemical and physical processes modify the surface texture of sediment. The contention i s that the modifying environment i s sufficiently unique to impart distinctive suites of features of the surface on the sedimentary particles (Bull, 1981). Essentially two branches of the work have evolved. First, those studies which are concerned with shape analysis of individual grains, primarily with the two-dimensional representation and numerical summary of grain outline using techniques such as Fourier analysis (Dowdeswell, 1982; Ehlrich, 1987). The basic premise of such work i s that grain shape i s extremely sensitive to abrasion. As a consequence rapid changes occur in roundness and angularity, which may thus be useful indices of sedimentary history. The second branch of work i s concerned with specific surface characteristics, either presence/absence or intensity of development. The latter may be defined either as the number of grains exhibiting an effect, or as the percentage of the vi s i b l e area covered on a single grain by a certain characteristic. Examples are found in the work of Margolis & Kennett (1971), Higgs (1979) and Bull (1985). This second stream of research has developed from the studies i n which investigators were concerned with the association of single characteristics with a particular, unique environment (for example, Gravenor, 1979). At this stage of development the technique f e l l somewhat into disrepute as many "red herrings", as they have subsequently been termed, were published. 61 These arose largely because i t was not recognised that i t i s not the transport pathways/ sedimentary environments per se but rather the processes operating within them that impart the surface features, and that such processes can be replicated in seemingly disparate environments. For example, V-shape pits are formed as a consequence of grain to grain impacts, which may occur whether the medium of transport i s air or water, and thus may be found on both water and wind transported grains. Similarly chattermarks, which were formerly thought to be indicative of g l a c i a l -transported grains, have been shown to have mechanical or chemical origin (Bull et a l . , 1980). Consequently individual surface textures are rarely, i f at a l l , useful indicators of palaeoenvironment. However, certain combinations remain highly diagnostic (Margolis & Kennett, 1971). This i s the basic premise of the approach adopted in this study. Furthermore, environments and their associated processes are not 100% efficient i n marking grains. Some grains may go through a transport cycle without receiving any new surface features. Therefore i t i s not inconsistent that modified and unmodified grains can co-exist in an environment. Hence, a number of grains, not a single grain, should be used to characterise an environment. The processes that modify surface features within any environment can be broadly categorised into mechanical abrasion, chemical precipitation, and chemical solution. Figure 4.5 is a schematic representation of the sequential origin of quartz grain surface features (adapted from Trewin, 1988). Such a framework may be considered in a temporal context, i.e. the cycle of weathering and erosion, or a spatial context, for example in this study, the routing of sediment from source to lake-sediment sink. Surface textures observed on grains can be inherited from the source rock, altered 62 Figure 4.5 Origin of quartz grain surface features (adapted from Trewin, 1988) STAGE PROCESSES FEATURES SOURCE AREA WEATHERING REGIME TRANSPORT REGIME DEPOSITION REGIME BURIAL DIAGENESIS UPLIFT & EXPOSURE Size range, shape & surface features grains in relation to primary igneous & metamorphic rocks Chemical & physical conditions which affect the grain Intensity, duration, number of modes of transport & sequencing of importance Primary grains -•Crystal growths & intergrowth surfaces Inherited characteris-tics Chemical Physical Solution features etching, precipitates v Conchoidal cleavage Blocky fractures Ice ^ Crushing Water r> Abrasion A WindL»>Impact Conchoidal fractures sharp edges, upturned plates Scratches, chatter marks, grooves Wind upturned plates Water V-shaped pits Tendency to uniformity during long transport Is final transport phase preserved or modified by precip-itation & solution -> Chemical -•> Solution & precipitates iOrganic How changes occurred r-»» Mechanical by burial solution \ -*> Chemical & physical effects from ingest-ion by organisms -•Fractured & distorted grains, compaction scratched Chemical Solution surfaces overgrowths, amorphous s i l i c a Large over growths & solution surfaces _^  Subsequent cycle _Inherited diagenetic characteristics 63 by subsequent weathering and transportation, or subsequently derived by alteration as a consequence of depositional conditions i . e . diagenesis. The controlling factors are both endoscopic, pertaining to grain size, shape and particularly microstructure (Gomez et a l . , 1988), and exoscopic, pertaining to the environmental conditions of weathering and transport. In general surface texture SEM work has not developed to a stage whereby suites of features can be used to define processes of sediment transfer in any environment. Thus in order to use surface characteristics diagnostically to c l a s s i f y the origin and subsequent transportational history of sediments, a methodology needs to be followed whereby materials are taken from known environments and then compared to material of unknown source, enabling the textural modification to be identified and inferences made about processes (Bull, 1981). 4.3.1 Methodology In order to control the variety of influences on grain surface texture, the approach taken in this study was to characterise the surface features of a constant lithology, igneous monocrystalline quartz, from materials of known transportational history, residing in known sedimentary environments, and to use these "signatures" to develop a scheme whereby unknown lake deposits could be c l a s s i f i e d . This attempt to minimise li t h o l o g i c a l l y inherited characteristics was possible because of the relative uniformity of the bedrock geology in the contributing areas to the lakes (see section 2.3). This i s evaluated further below. As a f i r s t step three replicate samples were collected from each of six known sedimentary environments within the watershed, resulting i n a total of 18 samples. These sedimentary environments were identified i n accordance with the generalised c l a s t i c sediment routing scheme (presented 64 in Figure 1.1; Figure 2.5). A l l the samples collected were surface samples from Quaternary s u r f i c i a l deposits, the origin of which was inferred from location, morphology, stratigraphy and sedimentology. A l l were of granodiorite provenance. The deposits sampled were: 1. Contemporary g l a c i a l : Two samples were taken from the contemporary Kwoiek Glacier with no attempt to differentiate subglacial, englacial or supraglacial material, and one from L i t t l e Ice Age moraines 2. Glacio-fluvial: Two samples from material from streams draining the Kwoiek Glacier and one from well sorted glacio-fluvial L i t t l e Ice Age deposits 3. Colluvial: the products of mass wasting on the hillslopes in the upper portion of the catchment. One sample was taken from weathered in situ bedrock (eluvium), one from a talus slope, and a third from a debris flow deposit 4. Colluvio-fluvial: products of mass wasting that have entered the f l u v i a l system i. e . samples from non-glacial fed streams 5. Deltaic: modern delta (2 from Kwoiek, 1 from Kokwaskey) to characterise an internal lake source with f l u v i a l history to assess whether there i s subaqueous imprint on the suite of surface textures 6. "Old" s u r f i c i a l materials: the deposits of Fraser Glaciation ( 2 from t i l l and 1 from a glacio f l u v i a l deposit), a l l in the v i c i n i t y of Kwoiek Creek. The last category i s differentiated from the other environments by age. It was expected that the sediments would exhibit features indicative of longer exposure to subaerial weathering processes, specifically chemical 65 s o l u t i o n and p r e c i p i t a t i o n . Although t h i s i s a f a i r l y crude subdivision these 6 units are thought to c h a r a c t e r i s e the p o s s i b l e sources f o r the sediment adequately (see d i s c u s s i o n i n Chapter 2 of c l a s t i c sediment sources and pathways). As stated i n Chapter 2 the evidence f o r aeolian deposits within the v a l l e y bottom i s rare. In essence what i s important i n t h i s study i s to d i f f e r e n t i a t e categories 1 and 2 i . e . recent g l a c i e r derived m a t e r i a l from that sediment derived from a l l other sources. The approach taken was to sieve the f i e l d sample to separate the f i n e sand, 63-250 pm, g e n e r a l l y 63-125 pm. This f r a c t i o n was selected to be representative of the two coarser f r a c t i o n s , i d e n t i f i e d from the i n i t i a l g r a i n s i z e c h a r a c t e r i z a t i o n of the sediments (section 4.2). I t was important to ensure that the s i z e f r a c t i o n selected be present i n a l l environments studied, such that the r e s u l t s from the known sedimentary environments be d i r e c t l y comparable with that work on the lake sediments, and that the size range i s representative of the sediments i n general i . e . that i t constitutes a s i g n i f i c a n t proportion of the grain s i z e d i s t r i b u t i o n (see s e c t i o n 4.2). Furthermore, most SEM work has been conducted on sand siz e d grains, f o r a v a r i e t y of p r a c t i c a l reasons i n terms of sample preparation, and because such grains are g e n e r a l l y believed to be transported independently and not as aggregates and would thus be expected to e x h i b i t c h a r a c t e r i s t i c s of p a r t i c u l a r t r a n s p o r t a t i o n pathways. Therefore the r e s u l t s from t h i s study could be compared with those i n the l i t e r a t u r e . The separated f r a c t i o n was b o i l e d i n 10% HCI for 20 minutes on a hot p l a t e to remove carbonates and organic stains, and subsequently dispersed i n sodium hexametaphosphate to ensure that a l l grains were disaggregated. 66 A sonic bath was not used for dispersal since this might create new surface textures i n the laboratory. The material was then cleaned in d i s t i l l e d water and air dried. Under a binocular microscope, 50 monocrystalline quartz grains were selected. Quartz was chosen because of i t s relative resistance to mechanical and chemical breakdown and i t s abundance. A minimum of 30 grains for each sample (following the recommendations of Tovey & Wong, 1978) were mounted on a SEM stub and coated with gold. Each stub was randomly coded and viewed non-sequentially on the SEM (Semko Nanolab 7, in the Department of Geological Sciences, University of British Columbia). To ensure that there were no variations in inherited characteristics sub-samples of the monocrystalline quartz grains were taken and used to make grain-mounts which were thin-sectioned and studied under polarised light with a petrographic microscope. In a l l cases the quartz grains analyzed were monocrystalline, unstrained, with l i t t l e evidence of f l u i d inclusions. The presence or absence of 34 characteristics were noted (see Table 4.3) for each grain on each stub. Although many investigators advocate the use of multiple characteristics, the actual surface textures used vary between studies. The textures observed in this investigation were selected after a literature search and preliminary observations of the samples from the Kwoiek Creek watershed to see the range of features evident. In the st a t i s t i c a l analysis of the data some features were found to be of l i t t l e discriminating value and were omitted from further consideration (see later discussion). Photographs of selected grains are presented i n Figure 4.6 to i l l u s t r a t e the range of features observed. Krinsley and Doornkamp's (1973) "Atlas of quartz sand surface features" was used extensively i n the early stages of this work to identify and define the surface features. Table 4.3 Surface textures used in SEM analysis 67 Number Surface texture Mechanical features 1 Complete grain breakage 2 Edge abrasion 3 Breakage blocks (<10 pm) 4 Breakage blocks (>10 pm) 5 Conchoidals (<10 pm) 6 Conchoidals (>10 pm) 7 Straight steps 8 Arcuate steps 9 Parallel striations 10 Imbricate grinding 11 Adhering particles 12 Fracture plates 13 Meandering ridges 14 Straight scratches 15 Curved scratches 16 Mechanical V-pits 17 Dish-shaped concavities Morphological features 18 Rounded 19 Subrounded 20 Subangular 21 Angular 22 Low re l i e f 23 Medium rel i e f 24 High rel i e f Chemical features 25 Oriented etch pits 26 Anastomosis 27 Dulled surface 28 Solution pits 29 Solution crevasses 30 Scaling 31 Carapace 32 Amorphous s i l i c a 33 Euhedral s i l i c a 34 Chattermarks number used in text to reference the individual surface textures Relative abundance recorded: 1 absent (< 1 grain) 2 present (2-7 grains) 3 common (8-21 grains) 4 abundant (> 22 grains) 68 Figure 4.6 Photographs of selected A: 1 Breakage bl o c k s ; 2 Edge a b r a s i o n (sub-rounded o u t l i n e ) ; 3 Conchoidal f r a c t u r e s C: 1 Breakage blocks; 2 Edge a b r a s i o n E: 1 P r e c i p i t a t e of s i l i c a p r e d a t i n g g r a i n breakage to i l l u s t r a t e surface textures B: 1 Angular o u t l i n e ; 2 Adhering p a r t i c l e s D: 1 Adhering p a r t i c l e s ; 2 Conchoidal f r a c t u r e s 3 P r e c i p i t a t e of s i l i c a ; 4 Breakage block 5 Impact p i t s F: 1 S o l u t i o n hole; 2 Adhering p a r t i c l e s 69 For each stub ( i . e . each sample) the presence/absence data were summed and the percentage occurrence of each of the features for the 30 grains calculated and categorised for relative abundance (see footnote to Table 4.3). When conducted for the known sedimentary environments this provides a "finger-print" of surface textures for the major sedimentary sources and transfers of the Kwoiek Creek watershed. The results for the individual samples and the modes for each environment are presented i n Table 4.4. The environments exhibit some notable differences. The c o l l u v i a l deposits contain very angular grains exhibiting breakage blocks and conchoidal fractures with l i t t l e edge abrasion. Chemical features are present which indicate grain alteration by percolating water. The colluvio-fluvial deposits show minor evidence of f l u v i a l action, slightly more rounded than the c o l l u v i a l deposits but otherwise very similar. The glacial grains exhibit few effects of water rounding. They are angular with high r e l i e f , breakage blocks, conchoidal fractures, randomly oriented striations and semi-parallel step like features. There i s very l i t t l e precipitation or solution, and any that exists i s a consequence of grain surface disruptions and/or f l u i d inclusions and the relative ease of alteration at these disrupted layers. The glacio-fluvial grains show alteration from their parent state, the glacial source, similar to that of the colluvial to c o l l u v i o - f l u v i a l transition, with more pronounced edge abrasion and rounding; grains lack the high relief and sharp angular outlines typical of grains of primary glacial origin. The deltaic grains are similar to those from the glacio-fluvial deposits which are most probably their source (see later discussion). The most distinct grains are those from the "old" s u r f i c i a l deposits, with a relative abundance of chemical solution and precipitation features. These are undoubtedly a Table 4.4 Relative abundance of surface textures for samples of known origin. F i r s t three samples for each sedimentary environment are modal abundance for each stub i.e. the individual samples, x i s the mode of the modes i.e. i t incorporates 90 determinations. ST C CF G GF D OSM 1 2 3 X 1 2 3 X 1 2 3 X 1 2 3 X 1 2 3 X 1 2 3 X 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 2 2 2 2 3 3 3 2 3 3 3 4 4 4 4 4 4 4 4 4 3 3 3 3 3 3 3 3 3 3 3 3 4 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 4 4 3 3 3 3 2 3 3 3 4 4 4 3 3 3 3 3 3 3 3 3 4 3 3 5 4 4 4 4 4 4 4 4 4 4 4 4 4 3 4 4 4 4 3 4 4 3 3 3 6 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 4 4 4 4 4 4 4 4 4 7 3 3 3 3 3 2 2 2 3 2 3 3 3 2 3 3 3 3 3 3 2 3 3 3 8 4 3 3 3 3 2 3 3 3 3 2 3 3 2 2 2 3 2 3 3 3 3 3 3 9 1 1 1 1 1 1 1 1 1 1 1 1 1 3 2 2 1 2 2 2 1 1 1 1 10 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 1 1 1 1 1 1 1 1 11 3 3 3 3 3 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 4 3 4 4 12 4 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 4 4 13 2 2 2 2 3 2 2 2 1 2 2 2 1 2 2 2 1 2 2 2 2 2 2 2 14 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 15 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 1 1 1 1 1 1 1 1 16 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 1 2 2 2 1 1 1 1 17 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 18 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 1 1 1 1 1 1 1 1 19 1 1 1 1 2 3 2 2 1 2 2 2 2 1 2 2 2 3 3 3 3 3 3 3 20 1 2 2 2 2 2 2 2 4 3 3 3 3 3 3 3 4 3 3 3 4 3 4 4 21 4 4 4 4 3 3 4 3 2 3 3 3 3 3 2 3 1 2 2 2 1 1 1 1 22 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 2 2 23 2 3 2 2 2 2 2 2 3 2 2 2 4 3 3 3 4 3 3 3 4 3 4 4 24 4 3 3 3 4 3 3 3 3 4 4 4 2 3 3 3 2 3 3 3 1 1 1 1 25 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 26 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 27 2 2 3 2 2 2 2 2 1 1 1 1 2 2 2 2 2 3 3 3 1 1 1 1 28 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 3 3 3 29 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 3 3 3 30 2 2 2 2 1 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 2 3 3 3 31 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 32 3 4 4 4 4 4 4 4 3 2 2 2 4 3 3 3 3 3 3 3 4 4 4 4 33 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 34' 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 3 3 1 Surface texture number refers to Table 4.3 C Colluvial CF Colluvio-fluvial G Glacial GF Glacio-fluvial D Deltaic OSM Old s u r f i c i a l materials 71 function of time and increased contact with percolating s o i l water. It i s conceivable that certain of these features, for example diagenetic overgrowths, may be destroyed during transport. However, solution effects, for example pits and crevasses (surface textures 28 & 29) are unlikely to be destroyed. Thus the sediments derived from this source and transported would be expected to remain distinct from those derived from other sediment sources. 4.3.2 Objective c l a s s i f i c a t i o n of the deposits In order that the lake sediments of unknown origin be c l a s s i f i e d according to source some objective technique needs to be devised to assign unknowns to specific classes. Margolis & Kellner (1969) b r i e f l y describe a s t a t i s t i c a l evaluation of their binary data (present (1) absent (0)), whilst Bull (1978) attempted analysis of similar data sets by linear and multiple discriminant analysis. Culver et al (1983) undertook an evaluation of operator variance and differentiation between known environments using canonical variate analysis. This technique i s concerned with the structure within the known groupings i.e. which surface textures contribute most to the characteristics of a particular environment, whereas multiple discriminant analysis i s more concerned with the allocation of unknown samples to mutually exclusive categories. The problem arises that the relative abundance data compiled in this and i n previous studies satisfy few of the requirements for the application of standard parametric s t a t i s t i c s such as those previously used, which require the variables to be normally distributed, independent, and for the number of samples to exceed the number of characteristics. With this i s mind a Euclidean difference s t a t i s t i c (D..), which uses the relative abundance data compiled in this study, is proposed to discriminate 72 o b j e c t i v e l y between samples, where: D = (X r. - X ) 2 / n J i = l J X ^ i s the r e l a t i v e abundance f or each of the c h a r a c t e r i s t i c s ( i ) f o r the sample of the known reference environment X.. i s the r e l a t i v e abundance for each of the c h a r a c t e r i s t i c s ( i ) f o r the i i J sample to be c l a s s i f i e d n i s the number of c h a r a c t e r i s t i c s used This s t a t i s t i c provides a means to assess the d i f f e r e n c e s between c h a r a c t e r i s t i c s f o r i n d i v i d u a l samples and groups. An unknown sample i s compared with signatures f o r known environments and i s c l a s s i f i e d to that environment for which the d i f f e r e n c e s t a t i s t i c i s minimised i . e . Min ID ^I j = l , 6 . The signatures take the form of the mode of the abundance categories f o r the 34 c h a r a c t e r i s t i c s f o r the r e p l i c a t e samples taken from each of the known sedimentary environments sampled i n t h i s study (as i n Table 4.4). Given the r e l a t i v e abundance, category data used i n SEM studies, a root mean square d i f f e r e n c e s t a t i s t i c i s p r e f e r r e d to a mean absolute d i f f e r e n c e s t a t i s t i c because i t gives greater weighting to those di f f e r e n c e s greater than one category. This helps reduce p o t e n t i a l problems of subtle d i f f e r e n c e s i n a number of categories between samples, p o s s i b l y as a consequence of contamination by a small number of exotic grains, r e s u l t i n g i n i n c o r r e c t c l a s s i f i c a t i o n s . The d i f f e r e n c e s t a t i s t i c permits a consideration of the r e l a t i v e importance of the i n d i v i d u a l surface c h a r a c t e r i s t i c s i n terms of t h e i r v a r i a b i l i t y between r e p l i c a t e samples and t h e i r d i s c r i m i n a t i n g value between the d i f f e r e n t environments. It i s important to e s t a b l i s h the ph y s i c a l basis f o r the d i s s i m i l a r i t i e s between environments i n order to 73 establish that they are not solely a function of sampling strategy and unrepresentative signatures. Furthermore, some surface textures appear to be conservative, exhibiting l i t t l e variation between the different sedimentary environments and thus can be removed from further consideration. The f i r s t step in the analysis was to identify these to improve the sensitivity of the s t a t i s t i c and to determine the optimal combination of characteristics to discriminate between sediments from the known environments. The analysis was conducted in two stages. The f i r s t stage was to assess the degree of v a r i a b i l i t y for each of the characteristics within the known environments. This was achieved by comparing each sample within a known environment against the mode for that environment and assessing the relative contribution (%) of each of the characteristics to the difference stati s t i c s (Figure 4 . 7 a ) . This provides an index of the s t a b i l i t y of each of the characteristics within each environment and identifies which characteristics are the most variable and therefore the least reliable in defining a particular environment. The second stage was to assess the relative contributions of the individual textures to the v a r i a b i l i t y between environments. This was assessed in two stages. F i r s t , by comparing the modes of the six groups against one another and looking at the relative contributions of each surface characteristic (Figure 4 . 7 b ) , and second, by comparing the individual samples against one another to assess the maximum between environment v a r i a b i l i t y (Figure 4 . 7 c ) . This second comparison i s referred to subsequently as extreme range v a r i a b i l i t y . In general those characteristics that vary most within environments are also those that vary most between environments. In terms of differentiating between environments the most significant surface textures Figure 4 . 7 Relat ive contr ibut ion of individual surface texture to difference s t a t i s t i c s : a) Within group 7 6 b) Between group 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 c) Extreme range i 1 7 1 61 —j 5 1 1 " T T : T i T T T T T T T 7 7 T T i T T T T T T T T T T T T T T T T I 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 75 relate to morphology, edge abrasion, adhering particles, solution pits and crevasses. Cumulatively characteristics 1, 3, 6, 10, 17, 18, 25, 26, and 31 (see Table 4.3) contribute <10% to the difference s t a t i s t i c and were omitted from further consideration. A l l subsequent analyses were conducted with the reduced number of surface characteristics (n=25). Further reduction of the number of characteristics while increasing the differences between some paired comparisons, decreased the discriminatory power of the s t a t i s t i c for the reference samples overall. In order to assess the confidence with which an unknown sample can be cl a s s i f i e d a number of independent tests were made. The f i r s t was to use the difference s t a t i s t i c to compare the samples from known environments against one another. This allows an assessment of both within (i.e. comparison of individuals against modes for each sedimentary environment) and between (i.e. modes for each environment compared against one another) environment v a r i a b i l i t y . In addition extreme range stati s t i c s were computed as outlined above, for which the 18 individual samples were compared against one another to assess minimum and maximum v a r i a b i l i t y within and between environments. The extreme range stati s t i c s were computed because the number of replicates for each environment (n=3) i s small, with the modal scores for some of the characteristics relatively unstable (Figure 4.7a). The results from these analyses are presented in Figure 4.8. The solid lines represent the range of stati s t i c s for the mode comparisons, the dashed lines the extreme range comparisons. Ideally there would be a clear separation of the s t a t i s t i c s , with the within environment difference statistics clustered close to zero, i.e. no differences, and the between environment values clustered much higher, with no numerical overlap between the two clusters. However, this i s the case only for the 76 4.8 Difference s t a t i s t i c s for samples from known sedimentary environments a) Colluvial b) Colluvio-Fluvial c) Glacial d) Glacio-fluvial e) Deltaic (internal lake source) -• f) Old S u r f i c i a l materials - deposits of Fraser glaciation • • #-# 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 • • Within group modal comparisons t 0 Between group modal comparisons Extreme range comparisons. Individual samples within and between groups respectively 77 samples taken from the o l d s u r f i c i a l m a t e r i a l s . In order to view the degree of d i s s i m i l a r i t y / overlap more c l e a r l y cumulative frequency p l o t s of the d i f f e r e n c e s t a t i s t i c s are p l o t t e d i n Figure 4.9. This graph i l l u s t r a t e s the cumulative % of the d i f f e r e n c e s t a t i s t i c s f o r paired comparisons l e s s than a p a r t i c u l a r value, for the within and between environment comparisons with both modal and i n d i v i d u a l sample ( i . e . extreme range) comparisons p l o t t e d separately. For each set of comparisons the d i f f e r e n c e s t a t i s t i c s were ca l c u l a t e d , ranked i n ascending order and the cumulative % of d i f f e r e n c e s t a t i s t i c s equal to or l e s s than a c e r t a i n number c a l c u l a t e d . These numbers were then p l o t t e d . Some of the sample comparisons r e s u l t i n the same d i f f e r e n c e s t a t i s t i c . These points can be i d e n t i f i e d by steeper l i n e s connecting them. Thus the number of points a c t u a l l y p l o t t e d on the ordinate scale equals, or i s l e s s than, the number of sample comparisons f o r each case of i n t e r e s t . Obviously the number of c a l c u l a t e d d i f f e r e n c e s t a t i s t i c s f o r within environment modal comparisons i s l e s s than f o r the i n d i v i d u a l sample ( i . e . extreme range) comparisons, hence the lower number of points p l o t t e d . The within group s t a t i s t i c s c l u s t e r c l o s e r to 0 than the between group s t a t i s t i c s . Rather than just i l l u s t r a t i n g the range l i m i t s of the d i f f e r e n c e s t a t i s t i c s , evident i n Figure 4.8, the cumulative curves enable an assessment of the p r o b a b i l i t y of overlap. The v e r t i c a l l i n e drawn on the graph i n d i c a t e s the lowest values obtained for between group comparisons, which i n t h i s example i s the same f o r both modal and extreme range comparisons. From these analyses f o r the Kwoiek Creek watershed i t i s proposed that the lowest of these values represents the minimum d i f f e r e n c e s t a t i s t i c that w i l l be obtained when comparing a sample against a signature f or an environment from which i t i s not derived. Therefore i t 78 M - within; Samples from each environment compared with the mode for that environment M - between: Modes from each environment compared against one another E.R.- within: Individual samples from each environment compared against replicates from that environment E.R.- between: Individual samples from one environment compared against those from other environments 79 i s proposed when classifying unknown samples, guidelines be imposed whereby a sample i s cl a s s i f i e d to that environment for which the difference s t a t i s t i c i s minimised and the value must be less than 0.44 (when n=25) or an indeterminate result should be recorded. From the comparisons of known sedimentary environments outlined in this section using modal comparisons this would result in 69% of samples being correctly classified and 31% not c l a s s i f i e d . When the individual data are studied, the differences for the comparisons of the glacial and glacio-fluvial environments, and the col l u v i a l and colluvio-fluvial environments are seen to be amongst the smallest, and are in the graphical region of overlap. From the histogram plots earlier, and the descriptive comparison of the grains in the different environments outlined above, i t i s clear that the f l u v i a l reworking of the parent glacial and co l l u v i a l grains introduces only subtle differences in characteristics. As the main thrust in this study i s the identification of g l a c i a l l y derived grains regardless of f l u v i a l reworking (which must occur at least for the gl a c i a l l y derived material in di s t a l lakes) the second stage in the analysis was to pool the glacial and glac i o - f l u v i a l data, and the co l l u v i a l and colluvio - f l u v i a l data, and to conduct a comparison between these two pooled groupings. These comparisons are presented in Figure 4.10 as cumulative distributions. These graphs indicate a much clearer differentiation of the environments with only 20% overlap (threshold D.. 0.67) on the cumulative curves for modal comparisons. When the within group s t a t i s t i c s are broken down further (Figure 4.11), i t can be seen that the variance within the colluvial environments i s greater than that within the glacial environments (see both ER and modal comparison lines). This indicates that greater 80 F i g u r e 4.10 Difference s t a t i s t i c s - pooled g l a c i a l and c o l l u v i a l Figure 4.11 Difference s t a t i s t i c s - pooled gla c i a l and c o l l u v i a l within and between group difference s t a t i s t i c s 82 confidence can be attained when classifying glacial as opposed to c o l l u v i a l sediments. For the c o l l u v i a l deposits one sample in particular, the i n situ eluvium, i s most distinct and tends to bias the results, enhancing the apparent differences. However, as material can be delivered into the lakes by rockfalls etc, i.e. directly from such a source, this sample was retained in the analysis. It i s not considered to be s u f f i c i e n t l y distinct to create a separate category. As an independent test of this scheme within the watershed, and in order to evaluate the additional influence of subaqueous post-depositional conditions on the lacustrine samples, signatures of sediments of known co l l u v i a l origin within the lakes (inferred from stratigraphic characteristics such as graded beds at the margins of the lakes) were compared with the t e r r e s t r i a l signatures (Figure 4.12). These results offer support for the use of this approach in the c l a s s i f i c a t i o n of lake deposits of unknown origin. A l l samples were correctly classified, although i t i s important to caution that this may provide a sense of over-confidence as such samples are the ones most l i k e l y to be derived from a single source and probably exhibit the strongest source signal. In order to test the generality of the characteristics and c l a s s i f i c a t i o n procedure, data were compiled from the literature, primarily from the work of Bull (see Table 4.5), who has promoted this multi-parameter relative abundance approach, to compare against the data for the environments in this study. Such an approach i s fraught with problems as a consequence of: variation i n the selection of grains between different investigators; identification and nomenclature of specific features; and the great v a r i a b i l i t y i n both inherited characteristics (i.e. microstructural controls) and the processes operative in different Figure 4.12 Classification of known subaqueous deposits (n=25) A l l samples are clasified as colluvial 1 • H Glacial reference • •• Colluvial reference • • • ' ' ' ' ' ' , , , ' • .0 0.2 0.4 0.6 0.8 • G l a c i a l reference comparisons • C o l l u v i a l reference comparisons • » Within environment variance (modal) 1.0 1.2 i 1 Within environment variance (extreme range) Table 4.5 Characteristic categories of abundance of surface features found on quartz grains from various environments of modification (Bull, 1985). ST Colluvial Glacial Fluvial Colluvio- Glacio- Pedogenic Fluvial Fluvial 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 1 2 3 4 3 4 2 2 1 1 3 4 3 1 1 1 1 1 1 2 4 1 3 3 2 1 3 3 1 1 1 3 2 1 3 2 4 4 4 4 2 3 2 3 3 4 4 3 2 1 1 1 1 2 4 1 2 4 1 1 1 1 1 1 3 3 1 2 2 3 3 3 3 4 1 1 1 1 3 1 1 1 1 4 1 3 3 2 2 2 3 1 2 1 4 1 1 1 1 4 1 2 2 3 3 4 2 4 1 2 1 1 3 3 3 1 1 2 1 1 2 4 3 2 3 3 1 1 1 2 1 1 1 3 1 1 3 4 3 4 3 4 2 3 2 3 2 4 4 2 2 3 1 3 3 3 1 3 3 2 1 1 2 1 1 1 1 3 1 1 1 1 2 4 2 3 1 1 1 1 3 1 1 1 1 1 1 3 3 2 2 2 4 2 4 3 4 4 3 3 4 4 1 2 Surface texture number refers to Table 4.3 85 environments, nominally classified as the same. For example, the colluvial environments of Bull are ancient slopes of Swaziland, with a very different weathering regime to that of the Coast Mountains of British Columbia. The comparisons are reported i n Table 4.6. It i s important to note that for these comparisons the f u l l range of surface textures was used i.e. n=34. As would be expected, differences do exist. As such at this stage i t i s cautioned that the classification scheme developed in this study i s restricted to within-watershed discrimination of materials, although i t s u t i l i t y has yet to be tested elsewhere i n the Coast Mountains, or in alpine areas with a similar lithology. From this f i r s t stage i t was concluded that subtle differences in the combinations of surface characteristics are apparent in the known sedimentary environments of the Kwoiek Creek watershed which enable fine-sand material to be cl a s s i f i e d according to source. The analyses on the known sedimentary environments in this study indicate that these characteristic surface textures mostly are imparted during the weathering phase of the sediment1s history and do not seem to be destroyed or altered during subsequent f l u v i a l transport. This i s undoubtedly a consequence of the spatial scale of this investigation and the distance the grains are transported. If the objective i s to minimise the number of incorrect classifications i . e . minimise Type I errors, then to clas s i f y an unknown deposit within the Kwoiek Creek watershed, using the 25 surface characteristics employed in this analysis, a difference s t a t i s t i c , < 0.67 should be obtained when compared to a reference sample. In the case of the known sedimentary environments analyzed i n this study this would lead to >80% of the samples being correctly c l a s s i f i e d , and <20% not cla s s i f i e d . Table 4.6 D i f f e r e n c e s t a t i s t i c s f o r comparison of known sedimentary environment signatures determined f o r the Kwoiek Creek watershed with l i t e r a t u r e values of B u l l (1985) Sedimentary environment Difference s t a t i s t i c C o l l u v i a l 0.65 C o l l u v i o - F l u v i a l 0.80 G l a c i a l 0.94 G l a c i o - F l u v i a l 0.99 Old S u r f i c i a l m a t e r i a l s / 1.40 Pedogenic sand N.B. i n t h i s example n=34 4.3.3 Classification of the lake deposits The objective of the SEM work was not to use i t as a technique to characterise and describe samples down selected cores but to determine i f the coarse and intermediate grain size fractions have a clear source signal such that fluctuations in these components could be used to make inferences about changing sediment sources through time. Thus the individual samples selected were specifically chosen from deposits where only one of these fractions i s present i.e. to reflect either the coarse or intermediate sediment fraction, such that the respective sources of the two fractions could be identified with certainty. Samples approximately 1 cm in depth were taken from cores from a l l lakes at variable depths to ensure there had been no changes in the characteristics of the different fractions through time. The samples are described i n Table 4.7. The difference s t a t i s t i c s for the comparisons of the lake sediments of unknown origin are presented in Table 4.7. Of the 20 samples analyzed 7 are cl a s s i f i e d as gl a c i a l i n origin, 9 are colluvial, and 4 are indeterminate. No notable differences i n classifications arose as a consequence of selecting the coarser or finer sand grains from an individual sample for analysis i.e. 125-250 Mm compared to 63-88 nm. The coarse fraction of samples from the lower 2 lakes, Kokwaskey and Kwoiek, was cla s s i f i e d as of c o l l u v i a l or indeterminate origin. From the upper two lakes, Kha and Klept, the coarse fraction was cla s s i f i e d as colluv i a l i n origin and the middle fraction glacial in origin. There are two probable explanations for the indeterminate results. It i s important to stress these are the results for a bed and not for individual grains. F i r s t , given the low rate of sedimentation in the lakes and absence of sedimentary structures to identify deposits of individual 88 Table 4.7 Lake sediment samples c l a s s i f i e d as to origin using SEM Sample Fraction 1 D .-Colluvial D.-Glacial Source KHA.C2.20 I 0.78 0.49 Glacial KHA.C2.70 C 0.49 0.73 Colluvial KHA.C2.90 c 0.52 0.71 Colluvial KHA.C2.128 I 0.71 0.45 Glacial KHA.C2.220 I 0.75 0.45 Glacial KHA.C2.268 I 0.79 0.47 Glacial KLP.C1.18 I 0.82 0.37 Glacial KLP.C1.27 1 0.72 0.64 Glacial/ Indeterminate KLP.C1.93 c 0.43 0.79 Colluvial KLP.C1.98 c 0.45 0.79 Colluvial KLP.C1.111 I 0.73 0.51 Glacial KOK.C5.60 c 0.65 0.67 Indeterminate KOK.C5.155 c 0.53 0.75 Colluvial KOK.C5.240 c 0.49 0.77 Colluvial K0K.C5.315 c 0.43 0.82 Colluvial KWK.C15.40 c 0.72 0.68 Indeterminate KWK.C15.70 c 0.75 0.67 Indeterminate KWK.C15.135 c 0.31 0.77 Colluvial KWK.C15.145 c 0.67 0.81 Indeterminate/ Colluvial KWK.C15.185 c 0.45 0.79 Colluvial 1 C - coarsest fraction ; I - intermediate fraction events, many sedimentation events are represented i n each unit of a n a l y s i s . Thus the r e s u l t s are probably i n d i c a t i v e of the i n t e g r a t i o n of events and sources i . e . sediments of mixed o r i g i n . Second, the basal sediments i n Kwoiek and Kokwaskey Lakes, given t h e i r coarse nature, are probably deposits r e f l e c t i n g c o l l u v i a l , p a r a g l a c i a l sedimentation. The residence times of material on the slopes at t h i s time, which were deposited by the r e t r e a t i n g Fraser g l a c i a t i o n i c e , were short; thus any signal c h a r a c t e r i s t i c of the o l d s u r f i c i a l materials, as determined today, would be poorly developed. These r e s u l t s i n d i c a t e that the coarsest f r a c t i o n i n many of the samples i s c o l l u v i a l i n o r i g i n . The intermediate f r a c t i o n i s g l a c i a l i n o r i g i n . When the r e s u l t s from the d i f f e r e n t lakes are compared they i l l u s t r a t e a strong f i l t e r i n g of the intermediate g l a c i a l s i g n a l down-system, as would be expected. The f r a c t i o n i s deposited i n Kha and Klept lakes and does not reach the two more d i s t a l lakes. I t i s important to stress that the deposits analyzed here are ones selected to represent lake-wide changes i n sedimentation i . e . deposits of an obviously l o c a l i s e d , c o l l u v i a l o r i g i n were discarded. The presence of g l a c i a l l y derived sediment at various depths i n the sediments of Klept Lake supports the contention that g l a c i e r s have existed i n the Kwoiek Creek watershed throughout the Holocene Epoch. An i n t e r p r e t a t i o n of the s i g n i f i c a n c e of the changes i n source of the coarser f r a c t i o n s through time i s presented i n Chapter 5 a f t e r consideration of the source and v a r i a b i l i t y of the f i n e s i l t - s i z e d f r a c t i o n . 4.4 Sediment source a n a l y s i s ; The s i l t - f r a c t i o n The f i n e s t g r a i n - s i z e f r a c t i o n (medium to f i n e s i l t ; s ection 4.2) which 90 constitutes most of the sediment i s not amenable to the same form of SEM surface texture analysis (see problems outlined by Krinsley, 1978). The size range of this fraction (presented in section 4.2) suggests that i t may be "glacier rock flour" and thus be the sedimentary signal of specific interest in this study for reconstructions of glacial a c t i v i t y . A systematic examination of glacial rock flour by Keller & Reesman (1963) from a wide range of lithologies reported that particle sizes characteristically ranged 12 pm down i.e. the finest s i l t and clay. This generally corresponds to the fine sub-population identified in this study. In the lakes studied the lower size limit of this fine fraction may well be truncated as a consequence of the short residence times of the water and sediment in the lakes such that there i s insufficient time for the finest sediment to settle out (see calculations in section 3.1). The i n i t i a l approach taken to characterise the source of the fine s i l t -sized material was by analysis of i t s mineralogy. An attempt was made to relate this to source characteristics, following a similar methodology of source-sink characterization outlined i n the previous section for the SEM work. Samples of the 32 to 4 pm (5 to 8 <r>) size fraction from samples of known genesis (see l i s t i n g for SEM work) were ground into powder, fixed on unoriented mounts and analyzed on a Philips PW 1730 X-ray diffraction instrument, Department of Oceanography, U.B.C. using standard procedures. The dominant minerals were identified on the basis of the intensity of reflections using standard tables. No clear distinctions between the mineralogy of c o l l u v i a l environments and recent glacial deposits were evident. Both exhibited a predominance of quartz and feldspars in f a i r l y regular proportions, with no minor minerals which would seem to be diagnostic of origin. One of the three contemporary 91 colluvial samples has a greater abundance of chlorite and phyllosilicate minerals, specifically vermiculite and smectite, which are present i n only trace amounts in the gl a c i a l samples studied. This i s undoubtedly a reflection of the greater subaerial exposure of the slope deposits and i s consistent with the relative abundance of the incipient chemical features that were observed in the SEM work. This would suggest that i f phyllosilicates are observed in the s i l t fraction of the lake sediments beyond trace amounts, the material was derived from a col l u v i a l not a glacial source. However, i f they are not present no conclusions can be drawn concerning provenance or history. The "old s u r f i c i a l materials" exhibited a much greater degree of weathering, with a greater abundance of smectite and vermiculite, and are the only deposits with appreciable kaolinite present. The distinctiveness of the Fraser Glaciation deposits (OSM) permits the differentiation of material derived from them and from material of contemporary c o l l u v i a l and glacial sources. The results from the lake sediments are ambiguous (see an example i n Figure 4.13). There i s no clear evidence for more than trace abundance of any minerals other than quartz and feldspar. This i s not surprising given the bedrock geology, at least i n terms of the contributing areas for the lakes under investigation, and the low residence times of many s u r f i c i a l materials before being deposited in the lakes. Previous studies have indicated that variations i n the mineralogy of the non-clay c l a s t i c fractions of most lake sediments generally reflect variations i n drainage basin geology (Jones & Bowser, 1978). Superimposed on this factor are the effects of selective settling and winnowing of finer grained sediments (Jones & Bowser, 1978). The absence of a strongly weathered s i l t fraction in the lake sediments suggests that there has been l i t t l e reworking and 92 Figure 4.13 Example of XRD reflections of s i l t - s i z e d fraction (lake sediment sample KWK C15-5) Intensity Feldspar Quartz 93 deposition of the older s u r f i c i a l materials. Rather, the material deposited in the lakes appears to have been derived from contemporary glac i a l and colluvial deposits. However, using this technique i t i s not possible to identify confidently the source of the fine fraction. One of the explicit reasons for choosing to work at a drainage-basin scale with a number of lakes was to permit a consideration of basin-wide as compared to localised, sedimentation patterns. This provides an alternative approach to identifying the source of the fine fraction by looking at the spatial and temporal coherence in i t s fluctuations. This approach i s pursued in Chapter 5 through analyses which consider the coherence in the fluctuations of the s i l t - s i z e d signal down-valley, i . e . between lakes, through time in conjunction with a consideration of the coherence of the c o l l u v i a l signal as identified by the SEM technique. These are used to isolate the gl a c i a l sedimentary signal of specific interest i n this study. 4.5 Summary From the analyses presented i n this chapter i t i s concluded that potentially important palaeoenvironmental information i s contained in the c l a s t i c component of alpine lake sediments, both in terms of their rate of influx and provenance. Figure 4.14 i s a schematic representation of the methodology adopted in this study to isolate this information. Following i n i t i a l characterization of the sediments into c l a s t i c and organic components, grain size analysis and subsequent partitioning of the cumulative probability curves revealed three distinct sub-populations that were mixed either prior to or during deposition. The technique of graphical partitioning allows "unmixing" and provides an opportunity to examine each population by i t s e l f i n order to gain further insight into 94 Figure 4.14 Schematic representation of methodology pursued i n this study to interpret palaeo-sediment record SEDIMENT FRACTIONS Sub-bottom sounding/ collection cores Definition lake-wide sedimentary characteristics Characterisation Lithostratigraphy Cla s t i c fraction Organic fraction Textural analysis & graphical partitioning Loss on ignition Coarse SEDIMENT SOURCE IDENTIFICATION Colluvial derived sediment Glacia l derived sediment Synchroneity basin-wide changes Glacial derived sediment Rate of sediment influx PALAEO INTERPRETATION Proxy signal glacial a c t i v i t y Holocene Epoch 95 origin. The provenance and history of the coarse and intermediate fractions can be determined by SEM surface texture analysis of a s t a t i s t i -c a l l y representative number of grains and objectively cl a s s i f i e d as to origin using a Euclidean difference s t a t i s t i c . It has been shown that the coarse fraction in each of the lakes i s derived from localised col l u v i a l sources and the intermediate fraction in Kha and Klept lakes i s of glacial origin. The s i l t grains are less easy to characterise using source-sink matching characteristics which are remarkably uniform across the basin i n terms of weathering products. Further consideration of the origin of the lake sediments i s presented in Chapter 5 by investigating the coherence of the s i l t signal down-valley through time. CHAPTER 5 TEMPORAL CHANGES IN SEDIMENTARY RESPONSE This chapter presents information on the fluctuations through time of sedimentation rates, identification of the glacial component and a consideration of i t s v a r i a b i l i t y through time. This provides the basis for a model of Holocene Epoch glacial a c t i v i t y in the Coast Mountains of B r i t i s h Columbia which i s assessed for regional consistency in Chapter 6. 5.1 Temporal v a r i a b i l i t y i n sedimentation rates The inverse relation between organic content and clastic sedimentation rates that was established in section 4.1 i s used as the basis for inferring changes in sedimentation rates through time in each of the lakes. Multiple cores from each of the four lakes were taken to represent the sedimentary environments within each of the lakes (see discussion in section 3.4). These cores are identified on each of the subsequent figures. Subsamples were taken at 5 cm intervals and analyzed for organic matter content (see section 4.2). A single core from each lake was then selected for more detailed analysis at 1 cm intervals. Any horizons which contained large fragments of organic material, for example those used for C-14 dating, were omitted from the analysis. The cores analysed at the lower temporal resolution (sample every 5 cm) were used to ensure that the results from the single cores selected for detailed analysis were representative of lake-wide changes through time. On average 1 cm of core represents 6 yrs of sedimentation in Kha Lake; 40 yrs in Klept Lake; 30 yrs in Kokwaskey Lake; and 60 yrs in Kwoiek Lake. For each lake the data from each core are transformed to a standardised index: I = - ( ( X j/x) - 1.0) I i s the sedimentation index x. i s organic matter content for the sample 97 x the mean organic content for the core A value greater than 0 (i.e. a low organic content) indicates above average sedimentation rates, a value less than 0 ( i . e . a high organic content) below average rates, and 0 the average rate. This standardisation has the effect of accentuating fluctuations through time within the environments rather than variations in magnitude attributable to spatial variations in influx within each of the lakes. The raw standardised data for the individual cores show generalised trends but are quite "noisy", with numerous outliers (see "spikes" on Figures 5.1 to 5.4). The potential origins of such outliers are numerous, but for the most part they are attributable to high magnitude cl a s t i c or organic sedimentation events. Hence they are not representative of integrated sedimentation 2 episodes at the timescale of 10 years that are of interest in this study. For this reason the data were f i l t e r e d to enable low frequency features to remain unchanged while damping higher frequency variations. This was achieved by using a 5 point running median. A running median rather than running mean was selected because i t i s not influenced by one or two outliers in a section of a core, but requires more persistent departures for a trend to be established. A 5 point running median was selected as i t maintains the resolution of the data at a scale appropriate for this study, and more importantly, i t i s at least 3 points greater than any deposit that can be attributed to one event even in the lower portions of the record (maximum thickness of graded beds at the base of Kokwaskey 1.7 cm, see core logs in Appendix I) . In order to discriminate further the low frequency trends in sedimentation regime from higher frequency variations, a 21 point weighted running mean was Figure 5.1 Standardised sedimentation data for Kha Lake a) KHA.C2 (1 cm) b) KHA.C4 (5 cm) c) KHA.C3 (5cm) d) KHA.C5 (5 cm) 99 Figure 5.2 Standardised sedimentation data for Klept Lake a) KLP.C1 (1 cm) b) KLP.C5 (5 cm) c) KLP.C6 (5 cm) d) KLP.C2 (5 cm) 0.8 0.4 0.4 0.4 Figure 5.3 Standardised sedimentation data for Kokwaskey Lake Figure 5.4 Standardised sedimentation data for Kwoiek Lake a) KWK.C15 (1 cm) b) KWK.C14 (5 cm) c) KWK.C6 (5 cm) 1.0 1.0 1.0 placed through the data, with weights which decrease outward in each direction from a central weight (Hartwig & Dearing, 1983). The results of this are presented in Figures 5.5 to 5.8. It can be seen quite clearly from these figures that systematic changes i n sedimentation rates have occurred over the timescale of the Holocene in a l l of the lakes. The trends from the cores analysed in greatest detail (i.e. 1 cm resolution) would appear to be representative of lake-wide changes. The subsequent discussion focuses on their interpretation. The degree of va r i a b i l i t y in the sedimentation rate records may be assessed through computation of the variance for each data set, or the complacency by the standard deviation. These st a t i s t i c s were computed for the raw sedimentation data and the raw data for the post 2400 yrs B.P. period (see Table 5.1) and are interpreted below. The sedimentation record from Kha Lake i s relatively short, with a basal date of 2350 +110 B.P. Two maxima are evident in the sedimentation record, one i n the upper sediments, at 0.7 m, and the other at the base of the core, peaks at 3.25 and 3.90 m (see Figures 5.1 and 5.5). The upper, more organic sediments may well be attributable to anthropogenic effects, specifically road construction and logging in the immediate v i c i n i t y of the lake. The record in Klept Lake (Figures 5.2 and 5.6) i s much more "noisy". However, overall the range i n magnitude in the sedimentation index and variance i s smaller than that for Kha Lake. The lake sediment record indicates very high clastic sedimentation rates recently with low rates prior to this. A road was constructed next to Klept Lake, but logging and widespread removal of timber and disturbance of the vegetative cover has not occurred. For much of the Holocene the lake sediments exhibit a remarkably complacent record. The lake has not recorded recent (last 2400 years) changes Figure 5.5 Standardised sedimentation data for Kha Lake (KHA.C2) a) Raw standardised data b) 5-point running median c) 21-point weighted mean 104 Figure 5.6 Standardised sedimentation data for Klept Lake (KLP.C1) a) Raw standardised data b) 5-point running median c) 21-point weighted mean a) b) c) e 5.7 Standardised sedimentation data for Kokwaskey Lake (K0K.C5) a) Raw standardised data b) 5-point running median c) 21-point weighted mean 106 Figure 5.8 Standardised sedimentation data for Kwoiek Lake (KWK.C15) a) Raw standardised data b) 5-point running median c) 21-point weighted mean a) b) c) 0 i.O -0.8 0 0.6 -0.4 0 0.6 107 Table 5.1 Characteristics of the detailed sedimentation records for each of the lakes (Mean=0) Variance Standard (n) deviation Original data Kha (KHA.C2) 0.081 0.285 410 Klept (KLP.C1) 0.047 0.216 258 Kokwaskey (K0K.C5) 0.127 0.356 429 Kwoiek (KWK.C15) 0.162 0.403 272 Post 2400 B.P. Kha (KHA.C2) 0.081 0.285 410 Klept (KLP.C1) 0.053 0.231 55 Kokwaskey (K0K.C5) 0.073 0.269 80 Kwoiek (KWK.C15) 0.069 0.262 57 108 in sedimentation as pronounced as in Kha Lake just upstream, undoubtedly as a consequence of Kha's formation (see section 3.10) and f i l t e r i n g effect. Sedimentation rates in the lake do seem to decrease after 2400 B.P. (see Figure 5.6 at 55 cm depth) possibly as a consequence of Kha Lake's formation, however, a decrease i s evident in a l l lakes within the watershed at this time. With the exception of the recent period the only pronounced high sedimentation rates are at the base of the core ca. 9640 +380 B.P. Ac-cumulation rates in terms of the depth of sediment in the lake in the pre 2400 B.P. period compared to the post 2400 B.P. period are remarkably low when i t is considered that i t i s proposed that Kha Lake did not come in to existence until 2350 B.P. and that f i l t e r i n g of the material did not start u n t i l this time. This indicates that either the basal date on the sediments cored in Kha Lake does not indicate the date of formation of the lake, that some depositional environment has always existed at the Kha Lake site (i.e. the fan that now constrains Kha Lake may have extended across the valley impounding the lake and subsequently been eroded on more than one occasion during the Holocene), or that Klept Lake has never functioned as a very good sedimentary trap. Whatever the reason, i t can be concluded that i n comparison to the other lakes in the watershed Klept Lake i s not a very sensitive site, exhibiting a relatively complacent record, lowest variance, with input from local slopes dominant. ^ The sediments of Kokwaskey Lake (Figures 5.3 and 5.7) seem to provide a particularly sensitive record of changes in sedimentation, with consistent departures from the mean recorded in different cores. In general the late Holocene i s characterised by high sedimentation rates with two notable increases in the last 3000 years, the peak at 15 cm and one just before 2400 B.P., with minor excursions around 4900 B.P. Minima in the mid to early 109 Holocene (centred ca. 6800 years ago) are especially conspicuous. The lower part of the record is characterised by relatively high rates of sedimentation. Sedimentation rates for Kwoiek Lake (Figures 5.4 and 5.8; Table 5.1) are the most variable. Rates peaked before 12 255 +770 yrs B.P. and have been high from 5035 +1875 yrs B.P. to the present, with two major peaks, one just prior to 2400 years ago and the other after 1240 +245 yrs B.P. Figure 5.9 presents summary plots of the temporal trends in sedimentation rates for each of the lakes i n a downstream sequence. The vertical axis i s standardised i n terms of depth to better i l l u s t r a t e fluctuations i n the absolute rates of sedimentation. The greatest rates of sedimentation are observed i n Kha and Kokwaskey Lakes. From the depths of accumulation shown on Figure 5.9 i t can be seen that absolute rates of sedimentation have decreased from Kha to Klept to Kwoiek Lake. Kokwaskey has a second major source from Chochiwa glacier and the Haynon Lake chain, complicating interpretation of i t s signal. This down-system decrease evident from Kha to Kwoiek Lake would suggest a dampened response to Neoglacial forcing, which is consistent with the expected pattern of sedimentation from the paraglacial model as outlined in section 1.3. In order to assess the degree of v a r i a b i l i t y through time for each lake the sedimentation indices must be defined and compared for similar time periods. The average value, the reference, w i l l vary depending on the time period selected i.e. w i l l be different for the last 2400 years as compared to that for the f u l l Holocene Epoch. Because the record for Kha Lake encompasses only the last 2400 years, the mean sedimentation rate for the last 2400 years was determined for each core and comparable sedimentation indices computed. The results are presented i n Table 5.1. Over the f u l l Holocene Epoch the degree 110 Figure 5.9 Downstream a t tenuat ion of absolute rates of sedimentation and v a r i a b i l i t y (Standard depth sca le , d i f f e r e n t time va r i a t e ) a) Kha Lake b) K l e p t Lake c) Kokwaskey Lake d) Kwoiek Lake of v a r i a b i l i t y in the sedimentation records of Kokwaskey and Kwoiek lakes is greater than that for Kha and Klept lakes. This study has documented that there have been marked changes in sedimentation from the early to latter parts of Holocene which are not represented in the Kha lake deposits because they only encompasses the post 2400 B.P. period. Klept Lake has already been described as relatively complacent. Hence i t i s d i f f i c u l t to evaluate any attenuation of v a r i a b i l i t y over the f u l l postglacial period. For the post 2400 B.P. period i f the record for Klept Lake is ignored there would seem to be a general decrease in v a r i a b i l i t y downsystem. Although there i s general agreement between the records in the lakes in terms of highs and lows in the sedimentation rates i.e. lower values i n the earlier part of the Holocene, higher in the latter, there are important differences in the details of the records. One example i s the markedly different patterns in sedimentation ca. 6000 B.P. in Klept and Kokwaskey Lakes. The above discussion has focused on the total sediment influx. As demonstrated in Chapter 4, distinct sediment fractions are being introduced into the lakes, not a l l of which are derived from the headwater glaciers. Thus i f these records are to be used as a proxy of Neoglacial activity within the catchment i t i s important to establish the glacial-derived component. The next section isolates and interprets fluctuations in the glacial sediment signal through time. 5.2 Changes in the glacial sediment signal through time As identified i n section 4.2 three distinct sediment populations are being introduced into the lakes. The coarsest fraction i s either of collu v i a l or indeterminate origin (see SEM results, section 4.3), while i t i s suggested that the intermediate and fine fractions are the glacial sediment of interest in this study. The graphical separation procedures outlined in section 4.2 112 were used to identify the relative proportions of each textural fraction for each sample of those cores that were studied intensively for changes in sedimentation rates. The temporal v a r i a b i l i t y of the "intermediate" and "fine" fractions combined, as % composition of each sample, i s presented in Figure 5.10. When the proportions of the fine and intermediate fractions are used to weight the sedimentation index to derive a new index of the v a r i a b i l i t y of rates of glacial sediment influx, through time, more interesting trends emerge (see Figure 5.11). In Figure 5.11 the vertical axis has been transformed from a depth to a temporal scale based on calculations of the influx of organic material. If, as outlined in sections 4.1 and 5.1, i t i s assumed that the influx of organic material has remained rel a t i v e l y constant through time, then an expected rate of organic influx can be defined per unit time using the tephra horizons and radiocarbon dates. This average organic influx rate can be used to define units of similar temporal duration i.e. to partition the core into units of similar organic content which, given the assumptions outlined above, were deposited in a given period of time. If combined with information on changing bulk density of the sediments this methodology can be used to document changes in the influx of c l a s t i c sediment through time at a given temporal resolution. The time periods chosen in this set of analyses were 10 years for Kha Lake, and 50 years for Klept, Kokwaskey and Kwoiek Lakes, consistent with the temporal resolution of the analyses (i.e. 1 cm in each of the cores). During preliminary analysis of the organic matter content data i t was noted that organic matter influx rates change dramatically in the earlier part of the Holocene, undoubtedly as a consequence of changes in t e r r e s t r i a l vegetation (see Chapter 6 ) . This suggests that the rates of organic influx should not encompass this period 113 Figure 5.10 Variations i n proportions of g l a c i a l derived fractions with depth a) Kha Lake b) Klept Lake c) Kokwaskey Lake d) Kwoiek Lake % Intermediate t f i n e f r a c t i o n % Intermediate fc f i n e f r a c t i o n o o o o 2400 -9640 +380 % Intermediate I f i n e f r a c t i o n 5 r -2400 - 4900 +325- -/ -3 M •o ° 6800 - - /- --11 485 +185' % intermediate fc f i n e f r a c t i o n -1240 +245-•-- 2400- -12 255 +770 114 Figure 5.11 Temporal v a r i a b i l i t y i n g l a c i a l - d e r i v e d sediment. Note the standard temporal scale on the ordinate. The v a r i a t e i s a r e l a t i v e s c a l e that should only be compared down-core, not between lakes (see d i s c u s s i o n i n section 4.1). a) Kha Lake b) Klept Lake c) Kokwaskey Lake d) Kwoiek Lake Composite Kha Laka - - - Klapt Laka Kokvaskay Laka Krolak Laka and therefore should not be defined as an average for the entire postglacial period. Ideally they would be defined for the early/mid Holocene period to the present. The average values used in the construction of Figure 5.11 are based on different periods for each of the lakes depending on the dates available. An average value i s defined for the only dated period in Kha Lake to 2350 B.P., for the sediments down to Mazama tephra in Klept and Kokwaskey Lakes, and to Bridge River tephra for Kwoiek Lake. This relatively short period was chosen for Kwoiek Lake because of the large error bars associated with the radiocarbon dates in this lake. These values are used to develop a chronology back to approximately 10,000 B.P. (based on information on pollen spectra presented in Chapter 6 indicating relative stability in the catchment's vegetation by this time). The magnitude of possible errors associated with the chronology are considered further below. Sedimentation rates per unit time were then calculated from the influx of clastic material for these "standard" time intervals. A 21-point running mean (see section 5.1) was placed through the data to better i l l u s t r a t e the low frequency changes in sedimentation regime that are of interest i n this study. From the records displayed i n Figure 5.11 i t would appear that the rates of gla c i a l sediment influx in a l l lakes are f a i r l y constant until ca. 7000 B.P. At that time sedimentation rates i n Kokwaskey and Klept Lakes show a gradual increase. In Klept Lake this i s followed by a minor, well defined peak of ca 1000 years duration, approximately 6000 to 5000 B.P. Rates in a l l three lower lakes show a pronounced increase ca 4000 to 3000 B.P., peaking ca 3500 years B.P. in Klept Lake, 3000 B.P. i n Kokwaskey Lake and 2900 B.P. in Kwoiek Lake. A second major increase i s evident i n the last 1000 years. The chronology of these later events can be seen best in the higher resolution record of Kha Lake. The high sedimentation rates of the earlier 3000 - 2000 year event 116 would seem to have fallen to lower levels by 1500 years B.P., starting to ri s e almost immediately, reaching a maximum ca 300 - 250 years ago. The synchroneity of events recorded in each of the lakes i s best seen in the f i n a l plot i n Figure 5.11 where the trends for the individual lakes are superimposed. Attention should be directed to the relative magnitude of fluctuations down the individual cores not between cores (see explanation in section 5.1). Although there i s good general correspondence between the records from the four lakes i t i s apparent that differences do exist in the details of the chronologies of the individual records. For example the timing of the mid Neoglacial phase which peaks ca 3500 years B.P. in Klept Lake, 3000 B.P. in Kokwaskey Lake and 2900 B.P. i n Kwoiek Lake. The question i s raised as to whether these are real and reflect an attenuation of the Neoglacial signal downsystem or whether they are an artefact of the way the chronology has been derived for the records presented in Figure 5.11. It i s possible to estimate potential errors in the organic matter influx rate chronology using dates that have not been used to define the original average influx rates i.e. to predict where the "independent" dated horizon would be expected to be found in the core and to compare this with i t s actual stratigraphic position. The most reliable chronological horizons are the tephra deposits. The radiocarbon dates obtained i n the upper portions of the cores in this study have very large error bars (see Table AII.l) and were not used. The Bridge River tephra was not used i n the definition of average organic influx rates in either Klept or Kokwaksey Lakes, so i t i s possible to use i t s position to estimate the order of magnitude of l i k e l y errors. The position of Bridge River tephra i n Klept Lake i s predicted to occur four sedimentation units above where i t actually occurs i.e. an error of 200 years, suggesting that events i n the upper portion of this record may be dated of the order of 200 years too old. In Kokwaskey Lake the Bridge River tephra i s two units below i t s actual position i . e . events may be dated of the order 100 years too young. This indicates that the confidence that can be ascribed to the chronology presented in Figure 5.11 i s of the order + 200 years, at least for the latter part of the Holocene. This suggests that there may well be some attenuation of the Neoglacial sediment signal as the differences in the lake chronologies, for example for the Neoglacial phase ca. 3500 to 2900 B.P., are greater than 200 years, although the difference in timing may not be as pronounced as indicated i n Figure 5.11. When the gla c i a l sediments trends are compared with those of total sedimentation rates (Figure 5.9) i.e. a l l sources combined, a number of important differences are evident (see for example the high sedimentation rates ca 170 cm depth in Kokwaskey in Figure 5.9 not evident ca 5000 B.P. in 5.11). It i s proposed that the trends in Figure 5.11 provide a proxy record of Neoglacial ac t i v i t y over the Holocene Epoch. Some of the finer material i s undoubtedly derived from episodic mass wasting. However, textural data on the s u r f i c i a l materials derived from plutonic rocks of the Coast Mountains (see Clague, 1989; Figure 1.10) and of the Stein River watershed (Ryder personal communication) indicate a predominance of sandy material with a low proportion of fines. The important point i s that the mass wasting events, at least as evidenced by the coarsest fraction, are not occurring at a basin-wide scale, whereas there i s a coherent sedimentary signal apparently imparted by glacial activity in the watershed's headwaters. Thus some confidence can be assumed in interpreting the fine-lake sediment record as a palaeo-record of glacial sediment supply. However, i t i s important to stress that the samples analysed in detail were deliberately chosen not to reflect localised inputs of collu v i a l material. 118 Therefore the sampling strategy deliberately selects against this fraction and any interpretations of the col l u v i a l signal through time should be considered accordingly. The temporal v a r i a b i l i t y of colluvial a c t i v i t y at the drainage basin scale as recorded by lake sediments i s worthy of further study. This section has demonstrated that the signal in Kha lake i s dominated by glacial sediment, evidenced by the close correspondence of total and fine-fraction sedimentation rates (compare Figure 5.9 with 5.11). Kha Lake provides a high resolution sedimentation record for the last 2400 years, with peaks in sedimentation rates in the period post 2350 +110 and i n the recent period. The record i n Klept Lake i s smoothed considerably when the col l u v i a l l y derived material i s removed, with three peaks 6000 - 5000 B.P., 4000 - 3000 B.P. and the period post 2000 B.P evident. In Kokwaskey Lake there are two fine sediment peaks one ca. 3000 years B.P. and the other in the last 750 years. In Kwoiek Lake there are two similar peaks i n glacial-derived sediment, one slig h t l y later than in Kokwaskey, centred on 2900 B.P. and the last 1000 years, which seems to be of slightly lesser magnitude but longer duration. The implications of these fluctuations in terms of a record of Neoglacial a c t i v i t y in the Coast Mountains of southern British Columbia are considered further in Chapter 6. It i s interesting to compare the results from the detailed analysis of the glacial sediment signal with the lithostratigraphies of" the cores to see whether certain facies are associated with periods of greater Neoglacial activity (compare Figure 5.11 with cores logs in Appendix I ) . There would appear to be no single facies that i s consistently associated with periods of greater Neoglacial a c t i v i t y . In Kha Lake periods of greater glacial sediment supply are broadly associated with laminated s i l t 119 deposits. In Klept Lake variations in the sediment lithostratigraphy i s more a function of inputs of coarse material from c o l l u v i a l sources, however, periods of greater glacial sediment influx are associated with massive s i l t s , while other periods with more massive sands. In Kokwaksey Lake the uppermost varved sediments correspond to the increased sediment supply associated with the L i t t l e Ice Age. Periods of lower glacial sediment input are associated with coarser facies. In Kwoiek Lake the most recent period of increased glacial sediment influx i s associated with laminated s i l t s , while earlier phases of Neoglacial a c t i v i t y correspond to facies of massive s i l t . It can be concluded that although phases of Neogacial activity are broadly correlated with variations in the texture of the deposits in each of the lakes, the core lithostratigraphies cannot be interpreted dir e c t l y as a proxy of Neoglacial a c t i v i t y . Thus the detailed analyses of the sediment sources and lake sediment deposits conducted in this study are a necessary prerequisite i n identifying phases of Neoglacial activity from the lake-sediment record. 5.3 Characteristics of temporal v a r i a b i l i t y The above discussion has focused on the smoothed general trends in the sedimentation data through time. Church (1980) discusses temporal v a r i a b i l i t y in environmental data in the framework of "trend", "persistence" and "intermittency". Trend includes cyclic and quasi-cyclic behaviour through time and i s concerned with the general patterns documented above. Persistence occurs when a particular value of a sequence constrains adjacent values. This i s usually the result of continuity or storage constraints in physical systems and may be studied via correllograms or by investigating the Markovian properties of the event sequence (Yevjevich, 1972). Intermittency refers to the (non-serial) 120 tendency for like values to be grouped i n a sequence, and is of value in considering the time distribution of extreme events. Given the damping of the Kwoiek Creek system because of the lakes, and the prescreening of the data for any apparently anomalous high magnitude events of local origin, the issue of intermittency i s not considered further. The degree of persistence in each of the records indicates the dominant temporal scales of "history" i n sedimentation records. Autocorrelation or serial correlation describes the linear dependence among successive values of a series that are a given lag apart. In order to analyse persistence in a record i t i s necessary to separate the random and non-random elements of a time series (Matalas, 1963). Trend must be eliminated from the non-random component before studying the oscillatory behaviour of the time series. Analyses for persistence were conducted on the raw glacial-derived sedimentation data defined for constant time units outlined above (10 years for Kha Lake, 50 years for the other three lakes) using the organic influx data (i.e. the raw data which make up Figure 5.11). Trend was removed from these data by subtracting a 21-point running mean and the residuals analysed for different levels of serial correlation. Figure 5.12 presents the data on the se r i a l correlation coefficients for the core i n each of the lakes that was studied in d e t a i l . The upper line on each graph in Figure 5.12 represents the seria l correlation coefficients calculated, and the lower line the expected values i f the process i s a simple Markovian process (see further discussion below). If a time series i s randomly distributed once trend has been removed the serial correlation coefficients between values for a l l orders w i l l be zero. Two features of the plots are of particular interest. The f i r s t i s the 121 Figure 5.12 Serial correlation coefficients for the data. The upper l i n e in each diagram indicates the values obtained for successive correlation values, the lower line the expected values i f the sequence i s random (see text for explanation). a) Kha Lake b) Klept Lake 400 800 Time Time c) Kokwaskey Lake d) Kwoiek Lake « 0.3 o 0.2 400 BOO 400 800 Time Time s t a t i s t i c a l l y significant f i r s t order ser i a l correlation in a l l of the lakes i.e. the values for adjacent sedimentation rates (KHA r^O.35, n=410; KLP r^O.42, n=220; KOK r^O.44, n=410; KWK r^O.36, n=198; a l l significant at a=0.01). This indicates strong history in the sedimentation system at the timescale of the sampling interval of the analyses. The f i r s t and second ser i a l correlation coefficients (representing 50 and 100 years of sedimentation) are s t a t i s t i c a l l y significant in Kwoiek, Kokwaskey and Klept Lakes, while the f i r s t 15 coefficients (representing 150 years of sedimentation) are significant i n Kha Lake (a=0.01). The second feature of interest i s the deviations of the serial correlation coefficients in a l l 4 lakes from those expected from a simple Markovian process. If a sequence of values i s non-randomly distributed in time, then each value repeats some of the information of the previous value, which in this study i s what i s simulated by the simple first-order Markovian process i.e. the f i r s t order coefficient i s successively multiplied for increasingly higher orders to generate an "expected" serial correlation coefficient that i s due solely to the f i r s t order serial correlation in the data. These values are displayed graphically as the lower line on each of the graphs. Speculatively i t could be suggested that the deviations represent true persistence in the data and these separations have physical significance in terms of glacial-driven sedimentation processes, for which a temporal scale can be defined. For example they may be related to storage effects within the glacial-sediment system affecting sediment supply and release of material into the f l u v i a l system. The data for Klept and Kokwaksey Lakes exhibit a very close correspondence to the f i r s t order Markovian decay, indicating only f i r s t order persistence i n the data i.e. a "history" of the order 50 years. For 123 Kwoiek Lake i f the second order correlation coefficient i s also incorporated into the model (see the dashed line) a much better f i t i s obtained, suggesting history over timescales of the of 100 years. Speculatively, these fluctuations may be interpreted in terms of relaxation times of sediment stores within the sediment cascade related to Neoglacial events. They may relate to the time for the sediment stores to attain s t a b i l i t y in the proglacial environment, for example stabilization of moraines, of the order 100-150 years. Conversely the data may indicate changes in the organic flux and relate to the st a b i l i t y of the terrestrial ecosystem, although the pollen data presented in the next chapter demonstrate that at the watershed scale the te r r e s t r i a l vegetation would appear to have been relatively stable over the Holocene Epoch. 5.4 Summary The data presented in this chapter indicate that there i s a coherent, sensitive and continuous record of glacial sediment supply i n the clastic component of the down-valley lakes in the Kwoiek Creek watershed. Isolation of the g l a c i a l sediment signal through direct sediment source identification and description of sedimentation patterns which are present throughout the Kwoiek Creek watershed provides a basis for partitioning the g l a c i a l l y derived portion of the record. When knowledge of the source i s combined with a consideration of rates of sediment influx, a record of glacial activity emerges. In general the correspondence between lakes i s strong when the finest size fraction i s considered in conjunction with sedimentation rates, with v i r t u a l l y a l l the changes which appear in one lake appearing in cores spanning the same time periods from other lakes. The sedimentary signal i s not one of simple selective deposition away from the source. The degree of 124 continuity i n the system has varied through time, and i t i s proposed, i s dependent on the degree of glacial forcing. For example, the earliest peak in glacial-derived sediment in Klept Lake ca 6000-5000 B.P. i s not recorded i n any detail in the Kokwaksey and Kwoiek lakes, whereas the two subsequent phases of Neoglacial activity, which regional evidence suggests were of a greater magnitude (see Chapter 6 for f u l l discussion) are pronounced in a l l of the lakes. It is proposed that v a r i a b i l i t y in glacial sediment supply in the Kwoiek Creek watershed i s dominated by events of Neoglacial and Glacial order, superimposed on which are the much shorter term events e.g. mass wasting evident i n the raw sedimentation data, that were removed from the data in this study. The chronology of the proposed Neoglacial fluctuations i s considered i n greater detail in Chapter 6. 125 CHAPTER 6 REGIONAL CORRELATION Independent evaluation of the chronology of Neoglacial a c t i v i t y presented i n Figure 5.11 f o r consistency with regional palaeoenvironmental conditions i s problematic. Any d i s c u s s i o n i s r e s t r i c t e d because of the incompleteness of many of the records obtained from other s i t e s i n southern B r i t i s h Columbia and ambiguities i n t h e i r i n t e r p r e t a t i o n . Furthermore, regional comparisons have inherent problems given the macro-scale nature of the con t r o l s on climate change which modify mean mid-latitude c i r c u l a t i o n patterns, which i n turn can y i e l d very d i f f e r e n t synoptic scale c l i m a t i c responses within a "region" (see d i s c u s s i o n by Yarnel, 1982). However, despite these problems i t i s important to compare the records of major c l i m a t i c f l u c t u a t i o n s over the Holocene Epoch, i n southern B.C. and at the study s i t e , i n order to assess the consistency of the chronology i n f e r r e d from the lake sediment record. This chapter presents proxy c l i m a t i c data, p o l l e n , macrofossils, g l a c i a l chronologies and tree l i n e studies, from the Coast Mountains of B r i t i s h Columbia and elsewhere i n the P a c i f i c Northwest, f o r comparison. The locations of a l l s i t e s are shown on Figure 6.1. The structure of the chapter i s chronological. The f i r s t s e ction i s concerned with the pattern of r e g i o n a l d e g l a c i a t i o n and the im p l i c a t i o n s of the basal dates obtained i n t h i s study, the second with palaeobotanical and palaeogeomorphological evidence f o r p o s t g l a c i a l environmental change, and the t h i r d section i s a consideration of the implications of the lake-sediment data derived i n t h i s t h e s i s f o r Holocene g l a c i a l chronologies i n southern B r i t i s h Columbia and the nature of the framework f o r future research i n g l a c i e r i z e d a l p i n e watersheds. 6.1 Deglaciation Studies to date have shown that the C o r d i l l e r a n Ice Sheet reached i t s maximum extent i n southern B r i t i s h Columbia approximately 14 000 to 14 500 126 Figure 6.1 Locations of previous palaeoenvironmental reconstructions in southern British Columbia discussed in text. The ecotone boundary between coastal and interior systems is shown as a dashed line. Kwoiek Creek shown as a large circle. Pollen sites indicated by s o l i d square: 1. Marion fc Surprise lakes; 2 Squeah fc Pinecrest Lakes; 3. Fishblue Lake; 4. Phair fc C h i l h i l Lakes; 5. Finney Lake; 6. Horseshoe lake. Glacial sites indicated by t r i a n g l e : 1. Mt. Garibaldi fc Sentinel glacier; 2. Sphinx glacier; 3. Bridge Gl a c i e r ; 4. Gilbert glacier; 5. Tiedemann glacier; 6. Franklin glacier; 7. K l i n a k l i n l glacier; 8. Jacobsen glacier; 9. Borealis glacier; 10. Ape g l a c i e r ; 11. Purgatory glacier; 12. Glacier B; 13. Scud g l a c i e r . Deglaclatlon sites indicated by open square: 1. Chiliwack River valley years B.P. (see references i n Armstrong, 1981; Clague, 1981, 1989; Ryder & Clague, 1989). Deglaciation was in progress by 13 500 years B.P. and the ice sheet wasted rapidly (Clague, 1981). The regional pattern of deglaciation was complex, but in general, the peripheral glaciated areas and highlands became ice free f i r s t . Active glaciers are thought to have remained longest in some mountain valleys, but probably coexisted with remnants of dead ice i n the valleys and plateaus of the B.C. interior. By about 9 500 - 10 000 years B.P. the glaciers in the region were no more extensive than they are today (Fulton, 1971; Clague, 1981). Deglaciation dates derived from basal dates on bogs and lake sediments in the southern Coast Mountains are presented in Table 6.1. Some of these dates have been questioned as being anomalously old due to old-carbon effects. The basal radiocarbon dates obtained i n this study (Table 6.2) suggest that extensive areas of the mountains of southern B.C. were ice free prior to 11 500 yrs B.P. The significance of the dates from the Kwoiek Creek watershed i s that they are from the floor of a mid-elevation mountain valley, downstream from contemporary alpine glaciers. As such they provide information on the status of valley glaciers within the Cordilleran system. The two oldest dates are from the two lowest lakes in the valley, 12 255+770 yrs B.P. (S-3010) (Kwoiek, elevation 835 m) and 11 485+185 yrs B.P. (S-2935) (Kokwaskey, elevation 1050 m). Taking the most conservative estimate of deglaciation (date less 2 standard deviations) these dates indicate that the middle and lower portions of the catchment were ice free and vegetation was established prior to 11 000 years B.P. with sedimentation occurring i n the lower lakes from this time on. The dates are younger with elevation, suggesting an upvalley sequence of deglaciation (Table 6.2). However, they are not consistently taken from the same relative stratigraphic position i n 128 Table 6.1 Deglacial dates determined i n other studies in southern British Columbia Author Site (elevation) Date Hebda (1982) Mathewes (1973) Saunders et a l (1987) Mathewes et a l . (1972) Ryder & Thomson (1986) R.J. Fulton in Lowdon et a l (1971) Finney Lake 13 170 +870 (Hebda pers. comm) Marion Lake (305 m) 12 350 +190 (1-5950) Surprise Lake (540 m) 11 230 +230 (1-5816) Slesse Creek (259 m) 11 900 +120 (GSC 3306) (Chilliwack valley glacier) Tamihi Slide (310 m) 11 200 +90 (GSC 4041) (Fraser Lowland piedmont lobe) Pinecrest Lake (320 m) 11 000 +170 (1-5346) 11 430 +150 (1-6057) Squeah Lake (205 m) 11 140 +260 (1-6058) Bridge glacier (1390 m) Tiedemann glacier (825 m) 9 810 +160 (S-1570) 9 510 +150 (GSC-939) Table 6.2 Radiocarbon dates obtained i n t h i s study Date Max Min Sample ID S i t e l o c a t i o n M a t e r i a l & elevation dated 12 255 +770 13 795 10 715 S - 3010 Kwoiek Lake Engelmann Spruce (835 m) cone & m u l t i p l e needles 11 485 +185 11 855 11 115 S - 2935 Kokwaskey Granular a l g a l Lake m a t e r i a l (1050 m) 9 640 +380 10 400 8 880 S - 3011 Klept Lake Cone & m u l t i p l e (1120 m) needles * Maximum and minimum dates are computed based on two standard d e v i a t i o n s 130 the lake sediments for significance to be attached to this i.e. they are from different depths above the g l a c i a l / postglacial sediment contact (see section 3.4, and core logs in Appendix I) . These results are consistent with those presented by Mathewes et a l (1972) and Mathewes & Rouse (1975) for Pinecrest and Squeah lakes to the south, i n the Fraser canyon near Yale. Mathewes et al's dates (11 000 +170 yrs B.P. (I-5346), 11 140 +260 yrs B.P. (1-6058) and 11 430 +150 yrs B.P. (1-6057) were obtained at an elevation of approximately 205 m a . s . l . The dates from the present study are at higher elevations and closer to the contemporary alpine gl a c i a l l i m i t . Hence at the position of the lakes, the last ice advance and subsequent retreat must have occurred prior to this time, indicating that deglaciation was more widespread within the mountain system and that i t may have been progressing at a faster rate earlier than was conventionally thought. These dates have important implications for the modelling of regional deglaciation in the southern portion of the province, specifically the nature of the retreat/ disintegration of the Cordilleran ice sheet in mountain valleys. For example, the COHMAP (Cooperative Holocene Mapping project) (1988) simulations have boundary conditions for 12 ka which include a f u l l Cordilleran ice sheet and for 9 ka extensive areas of ice in the Coast Mountains. New evidence for the influence of earlier and more rapid deglaciation should be carefully evaluated i n such simulations. 6.2 Holocene Epoch: Palynological records In general pollen records provide a more complete record of long term changes i n climate in the Pacific Northwest than do glacial records. Although vegetation i s probably less sensitive to short-term climatic fluctuations than are glaciers, vegetation records provide a continuous, although somewhat 131 smoothed proxy record of change. Several sites in coastal and interior in southern British Columbia, have yielded informative, reasonably well dated palynological records, with supporting evidence from plant macrofossils, molluscs and sediment stratigraphy. A l l of the study sites l i e within the maximum Cordilleran ice limi t , and none i s older than 12 500 yrs. The records are reviewed in detail by Mathewes (1984) and Ritchie (1987). The locations of the previous studies conducted within the coastal ecotone are shown in Figure 6.1 A consistent feature throughout the Pacific Northwest is the high frequency (up to 90% of total pollen) of diploxylon Pinus pollen between ca. 12 000 and 10 000 years B.P, indicating a forest, or parkland forest, dominated by lodgepole pine (Barnosky, 1984). These early records are d i f f i c u l t to interpret i n terms of a climatic signal given that macroclimatic effects are complicated by species migration, natural succession of plant communities, and s o i l development. Between 10 500 yr B.P. and 10 000 yr B.P. a dramatic change i s apparent in many pollen diagrams from south coastal Br i t i s h Columbia with an abrupt increase in Pseudotsuga menziesii and Alnus spp. (Mathewes, 1973; Mathewes & Rouse, 1975). This increase i s interpreted as reflecting a transition from cool, moist conditions which prevailed during deglaciation to a dry, warm climate. The evidence for this in the v i c i n i t y of the Kwoiek Creek watershed i s particularly strong at Pinecrest and Squeah Lakes (see location on Figure 6.1), with Douglas-fir, grasses, bracken f i r and spikemoss reaching peak frequencies at ca. 8620+135 B.P. (Mathewes & Rouse, 1975). Palynological indications of increasing wetness on the coast in the latter part of the Holocene Epoch begin before 5000 yr B.P. Increasing frequencies of western hemlock and cedar type pollen (probably Thuja) ca. 6800 yr B.P. 132 (Mazama tephra) have been reported extensively in coastal regions of the Pacific Northwest (see for example, Mathewes, 1973; Hansen & Easterbrook, 1974; Mathewes & Rouse, 1975; Leopold et a l . 1982; Heusser, 1983). Palaeoclimatic reconstructions using numerical transfer-functions, however, do not indicate late Holocene cooling or an increase i n wetness (Mathewes & Heusser, 1981), rather more or less consistent conditions over the last 6000 years, i l l u s t r a t i n g problems with the sensitivity of palynological data in coastal southern British Columbia. 6.2.1 Palynology: Kwoiek Creek watershed Pollen results of a core from one site, in Kwoiek Lake (KWK C.15), are presented to outline the history of postglacial vegetation in the Kwoiek Creek watershed. This provides a basis for comparison with the history determined from the physical characteristics of the lake sediments and for comparison with regional pollen signals. Lake deposits were chosen in this study because preservation of pollen i s generally better than in peats (Faegri and Iversen, 1964). Furthermore previous attempts to core bogs i n the headwaters of the Kwoiek Creek watershed have found only thin accumulations of peat (Ryder, personal communication). Laboratory preparation followed a standard procedure. Samples of known mass and volume were boiled in 5% KOH, screened, treated in HF and acetolysis solution, and stained with safranine. Lycopodium tablets were added to each sample in a concentration that produced about 1 trace spore to 4 f o s s i l grains, i n order to calculate pollen concentrations. Pollen residues were mounted in flotex resin and examined at magnifications of 400 and lOOOx ( o i l immersion) Approximately 250 te r r e s t r i a l grains were t a l l i e d for each sample. This i s a lower number than usually t a l l i e d because of the low pollen concentrations within the sediments. The samples were selected after the 133 chronology for the core had been determined, to provide regular temporal coverage hence they do not have a constant spacing. Pollen identifications were based on the reference collections of Dr. G.E. Rouse (Departments of Botany and Geology, University of Br i t i s h Columbia) and published atlases and keys (for example, Bassett et a l . , 1978; Moore & Webb, 1978). Grains that could not be identified from these sources were t a l l i e d as unknown; grains that were broken or hidden, or that had deteriorated beyond recognition were recorded as indeterminate. Because these numbers are small they are combined. The category Pseudotsuga may incorporate some Larix as the two are d i f f i c u l t to differentiate. Differentiation of haploxylon pine (Pinus albicaulis) from diploxylon pine (Pinus contorta or Pinus ponderosa) pollen was attempted based on the ornamentation of the grains; haploxylon pines have warts on the leptoma, whereas the diploxylon pines have no ornamentation i n that area (Ting, 1965). However, many of the grains, particularly in the lower portion of the record, are broken, consequently the presence or absence of such features i s not clear, and the two are combined. The pollen grains of Abies and Picea were not identified to the species level because the size range method outlined by Hansen (1947) i s problematic (Mathewes, 1973). Cupressaceae pollen was separated into Chamacaepyris and Juniperus with the aid of reference slides. Given the v a r i a b i l i t y in sedimentation rates (documented in Chapter 5) absolute pollen concentrations, are not very meaningful unless they can be determined with confidence for known periods of time (Faegri & Iversen, 1964), so individual palynomorphs are presented as a percentage of total pollen. The results are presented in Figure 6.2. The solid shading.on the plots represent the actual percentages, the lines the percentages x 10, so they are more clear for those species not so abundant. Present day vegetation patterns for the watershed are summarised in section 1 3 4 I. Figure 6.2 Pollen diagram for Kwoiek Lake 135 2.6 and Figure 2.5. Much attention has been directed to the zonation of pollen diagrams (Birks & Birks, 1980). Given the vastly differing pollen productivity and dispersal mechanisms of different plants, standard grouping techniques, for example principal components analysis, are inappropriate. Because the primary objective of this study was to interpret the pollen record in the context of i t s regional consistency, zonation of the pollen diagram was undertaken visually to enable easier comparison with literature data where similar visual zonation has been used. Kwoiek Lake i s not as sensitive as the Pinecrest and Squeah sites, (Mathewes St Rouse) mentioned above, which are located in the climatically transitional zone of the Fraser Canyon, where the vegetation is moisture stressed and very sensitive to changes in effective precipitation. However, the overall pattern evident i n the Kwoiek Lake pollen diagram i s consistent with the Pinecrest and Squeah Lake records (the closest previous study). Emphasis i s directed primarily to the conifer and angiosperm pollen as a record of watershed-scale changes in vegetation. The basal zone i s characterised by i n i t i a l high percentages (maximum 64%) of Pinus (probably Pinus contorta). At this time Picea and Tsuga Mertensiana relatively are high (indicating a "coastal s i t e " ; Barnosky, 1984). Alnus constitutes approximately 20% of the pollen, although this i s i t s lowest value for the postglacial period. L i t t l e else i s present in Zone I. At the base of zone II i n i t i a l l y low absolute pollen concentrations rise and there i s a transition i n the vegetation composition, with increases in Abies, Pseudotsuga, Alnus and Betula. This i s entirely consistent with other palynological studies in the area, and indicates a successional transition to more shade-tolerant conifers and the hypsithermal assemblage. At the same 136 time pollen of Myrica, Graminae and Pteridium exhibit an increase. There are no marked mid-Holocene transitions indicative of the end of the Hypsithermal interval evident i n other studies in the region. The transition from Zone II to Zone III i a associated with a gradual decrease in Pseudotsuga, an increase in Tsuga heterophylla and most markedly an increase in Thuja chamaecyparis. Corylus appears at this time (consistent with patterns documented by Mathewes, 1973; Mathewes & Rouse, 1975; King 1980). The rise of Thuja and a decrease in Douglas F i r at this time may be indicative of increased wetness. Alternatively i t may reflect a decreased frequency of f i r e (Mathewes & Rouse, 1975), which in turn may be related to wetness. Zone III i s characterised by relatively stable levels of Pinus, Picea, Abies, Pseudotsuga, Thuja and Alnus. The increase i n Alnus, Betula and Rosaceae, at the expense of the conifers, at the very top of the cores undoubtedly indicate the effect of anthropogenic disturbance in the catchment. Throughout the core many conifers, for example Pinus (after the i n t i a l colonisation of other conifers), Tsuga mertensiana, Pseudotsuga etc, are complacent and give only a general indication of climate change. Although the lower reaches of the catchment are xeric, the contributing area to the lakes extends over an extensive area where the species have a wide altitudinal range and the conifers are thus not sensitive indicators of climatic change. However, the most important feature of the pollen diagram i n the context of the thesis i s that the changes that i t does show are consistent with the regional patterns documented in the literature, indicating that the Kwoiek Creek watershed functioned lik e other sites in the southern Coast Mountains. 6.3 Holocene Epoch: Glacial chronologies Although the record of Holocene glacier fluctuations in the Coast Mountains i s fragmentary, recent studies have documented a consistent regional picture in terms of timing of events, despite important differences in the relative magnitudes of events between sites. The radiocarbon dates from these studies are reported in Table 6.3 and the sites are located in Figure 6.1. The dates are from Neoglacial lateral moraines, and sites upvalley from Neoglacial end moraines, where in some places in situ sheared tree stumps provide dates for glacial overriding (Ryder & Thomson, 1986). The dates are broadly categorised into three Neoglacial phases. Over the past twenty years several authors have proposed that significant glacier expansion occurred in various parts of the Canadian Cordillera between 6 and 9 ka B.P. (see for example, Beget, 1983; and Osborn & Luckman, 1988 for a review of the Canadian Cordillera). Luckman and Osborn (1979) have discussed significant problems associated with the dating control of these studies. In addition, theoretical climatic reconstructions for example, Kutzbach & Guetter (1982), COHMAP (1988), and pollen data indicate that the early Holocene Epoch was quite unfavourable for the accumulation of glacier ice. No convincing evidence for such advances i s reported for the southern Coast Mountains. A phase of glacier expansion about 6000-5000 C-14 years BP is indicated by dates from g l a c i a l l y overridden growth-position tree stumps in Garibaldi Park and from roots on a nunatak on Mount Breakenridge (see dates in Table 6.3; sites o r i g i n a l l y described by Mathews, 1951). Glacier transported wood fragments of similar age from a small glacier 6 km southeast of Bridge Glacier (5500 +70 years B.P., Blake, 1983) and Garibaldi Park (6170 +150 years B.P.) may have been derived from trees overridden by this advance. Older transported wood fragments from the same areas can be interpreted only in terms of a relatively high tree line . No end moraines are associated with this advance probably because i t s terminal positions were overridden during Table 6.3 Gl a c i e r chronologies f o r Coast Mountains broadly categorised by date Location Lat & Long C-14 Date Lab Elev a t i o n Reference (N) (W) (yr B.P.) no. (rn a.s . l . ) E a r l y Neoglacial expansion I. Overridden i n s i t u stumps Mt G a r i b a l d i 49°52' 122°59' 5260+200 Y-140 1860 Stuiver et a l (1960) Sentinel 49 54' 122 59' 5300+70 GSC-2027 1510 Lowdon & Blake G l a c i e r (1975) Mt 49°44' 121°57' 5950+140 GSC-760 2134 Lowdon Breakenridge (1968) I I . G l a c i a l l y transported wood fragments Nr. Bridge 50°48' 123°25' 5500+70 GSC-3219 1935 Blake (1983) G l a c i e r Sentinel 49°53' 122°59' 6170+150 GSC-1477 1670 Lowdon & Blake G l a c i e r (1973) Sphinx 49°55' 122°58' 7640+80 GSC-1993 1650 Lowdon & Blake G l a c i e r (1975) Mid Neoglacial expansion I. Moraine bog/ Peat l i t t e r Tiedemann 51°21' 124°56' 2250+130 GSC-948 825 Fulton (1971) G l a c i e r Tiedemann 51°21' 124°56* 2940+130 GSC-938 825 Fulton (1971) G l a c i e r G i l b e r t 50°53' 124°11' 2040+40 S-1572 1450 Ryder & Thomson Gl a c i e r (1986) I I . Overridden/ Transported wood Tiedemann 51 019' 124°58' 3345+115 S-1470 980 Ryder & Thomson Gl a c i e r (1986) Table 6.3 (cont.) 139 Tiedemann Glacier 51 19' 124 58' 2355+60 S-1471 990 Ryder Sc Thomson (1986) Gilbert Glacier 50 53" 124 11' 2220+75 S-1459 1430 Ryder & Thomson (1986) Gilbert Glacier 50 53' 124 11' 2175+75 S-1461 1450 Ryder & Thomson (1986) Gilbert Glacier 50 53' 124 11' 3415+70 S-1462 1460 Ryder St Thomson (1986) Tide Lake 56 16' 130 03' 2730+170 GSC-1372 650 Lowdon Sc Blake (1973) Jacobsen Glacier 52 03' 126 04' 2470+50 GSC-4155 1370 Desloges Sc Ryder (in press) Late Neoglacial ( L i t t l e Ice Age) Klinaklini Glacier Klinaklini Glacier Franklin Glacier Bridge Glacier Bridge Glacier Sphinx Glacier Scud Glacier Scud Glacier Glacier B 51°19' 125°49' 400+45 S-1566 (Tree root from palaeosol) 51°19' 125°49' 900+40 S-1567 (Tree stump in growth position) 51°16' 125°26' 835+45 (Tree root from palaeosol) 50°49* 123°34* 680+50 S-1463 (Log close to growth position) 50°49' 123°34' 540+45 S-1571 (Root from palaeosol) 49°55' 123°29' 460+60 Y-347 (Wood in growth position) 57°19* 131°24' 455+65 S-2297 (Tree stump from growth position) 57°19' 131°24' 625+140 S-2298 (Tree stump in growth position) 57°04' 131°02' 595+60 S-2296 (Ah material from palaeosol) 530 Ryder Sc Thomson (1986) 400 Ryder St Thomson (1986) S-1568 1170 Ryder Sc Thomson (1986) 1750 Ryder Sc Thomson (1986) 1750 Ryder & Thomson (1986) Barendsen et a l (1957) 670 Ryder (1987) 670 Ryder (1987) 1140 Ryder (1987) Table 6.3 (cont.) Purgatory G l a c i e r 52°10' (Wood -126°21' roots) 480+50 GSC-4191 595 Desloges & ( i n press) Ryder Purgatory G l a c i e r 52°10' (Forest 126°21' . l i t t e r 630+65 S-2978 upper palaeosol) 595 Desloges & ( i n press) Ryder Purgatory G l a c i e r 52°10' (Wood) 126°21' 1110+70 S-2976 595 Desloges & ( i n press) Ryder Purgatory G l a c i e r 52°10' 126°21' 785+70 S-2977 595 Desloges & ( i n press) Ryder Purgatory G l a c i e r 52°10' (Wood) 126°22' 460+50 GSC-4030 533 Desloges & ( i n press) Ryder Ape Lake 52°05' (Wood) 126°10* 770+60 GSC-4028 1395 Desloges & ( i n press) Ryder Jacobsen G l a c i e r 52°04' 126°08' (Palaeosol) 400+60 S-2979 1495 Desloges & ( i n press) Ryder Borealis G l a c i e r 50°10' (Wood) 126°07' 20+60 GSC-4163 1418 Desloges & ( i n press) Ryder Laboratories: S - Saskatchewan Research C o u n c i l GSC - Geological Survey of Canada Y - Yale WAT - Waterloo late Neoglacial time. This episode i s referred to by Ryder and Thomson (1986) as the "Garibaldi phase", a phase rather than an advance because there i s no apparent subsequent episode of contraction. This expansion may well correspond to the transition from warm and dry conditions of the early to mid Holocene, commonly called the xerothermic interval and documented in the pollen records (section 6.2.1), to the cooler moister conditions of the late Holocene. The Garibaldi phase can be correlated with the Dome Peak expansion (South Cascade glacier, Miller 1969) and possibly the Gamma Peaks advance at Glacier Peak (Beget, 1984) i n the northern Cascade Mountains of Washington state. In the Canadian Cordillera the mid-Holocene Epoch transition to cooler, wetter conditions appears to be time-transgressive, approximately 6000 years B.P. i n the Coast Mountains and 3000-4000 years B.P. i n the Rockies (Osborn & Luckman, 1988). Evidence of a mid-Neoglacial phase, the Tiedemann advance, i n the western British Columbia, ca. 3300 to 1900 years B.P. comes from Tiedemann and Gilbert glaciers in the southern Coast Mountains; Jacobsen Glacier, Bella Coola; and from Tide Lake in the northern Coast Mountains (Ryder & Thomson, 1986; Ryder, 1987; Desloges & Ryder, in press). Tiedemann Glacier was more extensive than during the L i t t l e Ice Age for about two millennia, from 3345 to 1300 years B.P., with a maximum at about 2300 years B.P. An advance of the Gilbert glacier commenced before 2200 years B.P. and culminated ca. 1900 years B.P., when i t s size approximated i t s late Neoglacial extent. Middle Neoglacial advances have been well defined both north and south of the study area. For example, in the St. Elias Mountains (Denton & Karlen, 1977), and on Mount Rainier (Crandell & Miller, 1964) mid-Neoglacial moraines l i e just beyond those of the late Neoglacial. A late Neoglacial. advance i s common to a l l glacierized mountains although i t s time span i s variable (Table 6.3). Reliable dates from overridden growth-position tree remains indicate that in the Coast Mountains, the late Neoglacial advance commenced before 900 B.P. and expansion continued u n t i l i t reached i t s maxima. Multiple and overridden moraines indicate that minor fluctuations in ice positions probably occurred close to the time of the maximum (Ryder & Thomson, 1986; Ryder, 1987). Dates obtained from dendrochronology and iichenometry indicate that most glaciers in the Canadian Cordillera began to recede from maximum L i t t l e Ice Age positions at various times during the eighteenth, nineteenth and twentieth centuries (Ryder, 1989). Rates of recession have decreased markedly in the last few decades and some glaciers have advanced (Osborn & Luckman, 1988). In summary, Neoglacial advances within the Coast Mountains began by 6 ka BP and glaciers were more extensive over most of the last 4 ka than earlier during the Holocene. The recent climatic amelioration (i.e. since the late Neoglacial maximum) has had a greater cumulative effect than any previous climate fluctuation over the Holocene Epoch (Ryder, 1987). The Holocene trends are summarised diagrammatically i n Figure 6.3 with similarly highly schematic trends from the palaeobotanical records. 6.4 Significance of results from Kwoiek Creek watershed The chronology of glacial a c t i v i t y presented in this study i s entirely consistent with those reviewed i n this chapter (see Figure 6.3). There i s no evidence in the sedimentary records of any of the lakes for an early Holocene phase of glacial activity. Sedimentation rates in the early postglacial are high but the characteristics of the sediments indicate material of indeterminate origin, probably derived from young Fraser Glaciation deposits, not the alpine glaciers at the head of the Kwoiek Creek watershed (see SEM results i n section 4.3.3). This i s the period of paraglacial sedimentation 143 Figure 6.3 Schematic representation of Holocene Epoch g l a c i a l and pollen signals Composite g l a c i a l chronology Climatic trends from palaeobotanical data Chronology from lake sediment (Ryder £ Thomson, 1986) (Mathewes, 1984) record, Kwoiek Creek Fluctuations i n influx of g l a c i a l derived sediment: a Mean July T Mean Annual proxy Neoglacial chronology ( C) Precipitation 144 documented by Church & Ryder (1972). It i s d i f f i c u l t to assign a precise date to the end of this period but data from Kwoiek and Kokwaskey Lakes (Figure 5.9) indicate a decrease in sedimentation rates soon after deglaciation to much lower rates, probably within a period of a 1000-2000 years. The exact chronology is d i f f i c u l t to resolve because of a lack of dateable material in the lower portions of the cores and the inappropriateness of extending the assumptions of constant influx of organic material to this time to derive a chronology. Changes in the sedimentation regime of the lakes are evident ca. 6800 B.P. (Figure 5.11), postdating Mazama tephra deposition. These changes are most prominent in Klept and Kokwaskey Lakes. There i s , however, no evidence of increased deposition of g l a c i a l l y derived sediments in Kwoiek Lake at this time. It i s possible that this mid-Holocene "phase" of glacial a c t i v i t y increased glacial sediment supply, which was strongly f i l t e r e d down-system. The two most recent episodes of Neoglacial activity, documented in the literature are evident in a l l of the lake sediment records. The f i r s t i s centred ca 3500 B.P. in Klept Lake, 3000 B.P. in Kokwaskey Lake and 2800 B.P. in Kwoiek Lake. The second occurs in the last 750 years (postdating 1240 +245 B.P.), with greatest sedimentation rates approximately 400 years ago documented for Kha Lake. In terms of relative a c t i v i t y ( i . e . strength of sedimentary response) these two events would seem to be comparable in magnitude, and are much stronger than any other events over the postglacial period. However, the most recent episode appears to be of longer duration. The high resolution Kha Lake record indicates that within each of these Neoglacial periods there are fluctuations i n sediment supply to the f l u v i a l system (see Figure 5.11) which may reflect periods of advance, stillstands and subsequent advance. For example, for the most recent Neoglacial phase 145 there i s a peak in sedimentation ca 400 years B.P. and a second rise within the last 100 years. When compared with regional data i t may be speculated that the 400 B.P. peak corresponds to the period just before maximum L i t t l e Ice Age conditions, and that i n the last century to the maximum rates of glacial recession. This would suggest that Neoglacial driven sediment yi e l d is at a maximum just after the Neoglacial peak and during the subsequent phase of retreat, consistent with the notions of paraglacial sedimentation. The major advantage of the lake sediment approach as used in this study i s that i t provides a continuous record, enabling a consideration of both the magnitude and frequency of postglacial changes at a site. At no site within the Coast Mountains i s there t e r r e s t r i a l evidence for a l l three Neoglacial phases evident in the lake sediment record for the Kwoiek Creek watershed. In fact there are only two or three sites with evidence for the two most recent Neoglacial phases because the most recent L i t t l e Ice Age advance was generally more extensive than earlier events and therefore destroyed a l l evidence of their occurrence. Furthermore the lake sediment record may be easier to date than the t e r r e s t r i a l record. This i s particularly true in an environment such as the Kwoiek Creek watershed where the lakes of interest are below tree-line and consequently there were frequent incorporations of organic material in the lake sediments which can be dated, thereby providing a chronology against which to interpret changes. The implications of both the methodology and the substantive findings for southern British Columbia are considered further in the f i n a l chapter. 146 CHAPTER 7 SUMMARY & CONCLUSIONS This research p r o j e c t has sought to make two basic contributions. F i r s t to develop a methodology to i n t e r p r e t changes i n the Holocene Epoch 2 3 palaeo-sediment system over timescales 10 -10 years, which makes e x p l i c i t the l i n k between sediment sources and lake-sediment sinks. This permits a more c r i t i c a l i n t e r p r e t a t i o n of the lake sedimentary record, such that changes i n the co n t r i b u t i o n of g l a c i a l sediment through time can be i d e n t i f i e d with confidence. This provides a l e v e l of understanding that goes beyond previous d e s c r i p t i v e studies that s o l e l y document changing c h a r a c t e r i s t i c s i n the sediment sink through time. Second to provide a continuous record of palaeoenvironmental conditions, s p e c i f i c a l l y g l a c i a l a c t i v i t y , f o r the southern Coast Mountains of B r i t i s h Columbia. 7.1 The methodology The a l p i n e sediment system i s proposed as a framework f o r studying a l p i n e sediment t r a n s f e r s through time. Previous studies have documented that the dominant controls over Holocene p r o g l a c i a l sediment transport 2 3 rates over timescales 10 -10 years r e l a t e to g l a c i a l and Neoglacial a c t i v i t y . Sediment a v a i l a b i l i t y and supply together with meltwater stream capacity are the c o n t r o l l i n g f a c t o r s . Superimposed on these factors are the e f f e c t s of episodic mass wasting and high magnitude hydrological events. Previous attempts to model the lake-sedimentary response have been l a r g e l y r e s t r i c t e d to the immediate p r o g l a c i a l environment. This study extends that work downvalley to lakes with lower sedimentation rates and presents a methodology to separate the g l a c i a l sediment signal from c o l l u v i a l sediment t r a n s f e r s . This i s achieved through sediment source-sink matching at the drainage basin scale by documenting the synchroneity 147 of trends down-system i n successive lakes, i . e . attenuation of the g l a c i a l s i g n a l . I t i s demonstrated that the organic matter content of the sediments ex h i b i t s a strong inverse r e l a t i o n with sedimentation rates and can be used to i n f e r rates of i n f l u x through time, and thus to provide a chronology f o r changes i n sedimentation regime. Textural a n a l y s i s of the sediments in d i c a t e s that g r a i n - s i z e d i s t r i b u t i o n s are polymodal. Graphical p a r t i t i o n i n g of l o g - p r o b a b i l i t y grain s i z e p l o t s was used to separate three constituent populations which are r e l a t e d to s p e c i f i c sediment sources. Surface texture a n a l y s i s using the Scanning Electron Microscope (SEM) was used as a way of i d e n t i f y i n g the source of the coarse and intermediate lake sediment populations. A s t a t i s t i c a l approach was developed i n which 30 grains from r e p l i c a t e known samples were characterised f o r 34 surface textures and o b j e c t i v e l y c l a s s i f i e d as to source using a Euclidean d i f f e r e n c e s t a t i s t i c . Those c h a r a c t e r i s t i c s that have d i s c r i m i n a t i n g value were i d e n t i f i e d and were used to discriminate the source of material of unknown o r i g i n . C o l l u v i a l derived sediments, g l a c i a l derived sediments and deposits derived from the older s u r f i c i a l materials of Fraser G l a c i a t i o n were d i f f e r e n t i a t e d . Analyses of known sedimentary environments i n d i c a t e d that surface textures are established l a r g e l y during the weathering phase of the sediment's h i s t o r y and not g r e a t l y a l t e r e d or destroyed during subsequent transport. Although the g e n e r a l i t y of the d i f f e r e n t i a t i n g c h a r a c t e r i s t i c s remains to be determined f o r other environments, the methodology i s appropriate f o r use i n a wide range of sediment routing studies. M i n e r a l o g i c a l a n a l y s i s of the f i n e s i l t f r a c t i o n was l e s s s u c c e s s f u l . No 148 clear distinction could be made between co l l u v i a l and recent glacial deposits although the older s u r f i c i a l materials exhibit a clear signal. The similar trends i n the temporal v a r i a b i l i t y of the finest fraction throughout the drainage basin are used as evidence that the finest fraction represents the glacial signal of interest derived from the catchment's headwaters. When the proportions of the fine and intermediate fractions are used to weight sedimentation rates, an index of the influx of gl a c i a l sediment in each of the lakes through time results. The record i s smoothed with resolution of the order 50 years. The degree of variance and absolute rates of sedimentation decrease downsystem as would be expected. Analyses of the sedimentation time series indicate persistence in the record of the order 100-150 years which i s speculated to represent the relaxation time of sedimentary stores within the watershed. The absence of strongly weathered material in the recent sediments of each of the lakes indicates l i t t l e contemporary reworking and deposition of older Fraser Glaciation deposits. Rather, material that i s being moved in the f l u v i a l system i s derived from contemporary gla c i a l and colluv i a l sources. Thus the sedimentary signal in the lakes of the Kwoiek Creek watershed i s not simply one of selective deposition away from the source, but of dif f e r e n t i a l input of sediment from distinct sources and through distinct transfer pathways. It i s concluded that downvalley lake sediments do provide a sensitive and continuous record of changes i n glacial driven sedimentation rates, provided the sediment from different sources i s identified. 7.2 Palaeoenvironmental reconstructions The basal dates obtained in this study suggest that extensive areas of the mountains of southern B.C. were ice-free prior to 11 500 B.P. The 149 oldest dates obtained are from the two lowest lakes in the watershed, 12 255 +770 yrs B.P. (S-3010) (Kwoiek, elevation 835 m) and 11 485 +185 yrs B.P. (S-2935) (Kokwaskey, elevation 1050 m). Taking the most conservative estimate of deglaciation (date less 2 standard deviations) these dates indicate that the middle and lower portions of the catchment were ice free prior to 11 115 years B.P. and that sedimentation has occurred in the lower lakes from this time. From the records i t would appear that the rates of glacial sediment influx in Klept, Kokwaskey and Kwoiek Lakes are low and f a i r l y constant until ca. 7000 B.P, providing no evidence for early Holocene g l a c i a l activity. Approximately 7000 B.P., sedimentation rates i n Kokwaskey and Kwoiek Lakes show a gradual increase, while in Klept Lake a minor, well defined peak of ca 1000 years duration occurs 6000 to 5000 B.P. indicating greater glacial sediment supply and a phase of renewed glacial activity^ the Garibaldi Phase. Glacially driven sedimentation rates i n a l l three lower lakes show a pronounced increase ca 4000 B.P., peaking ca 3500 to 2900 years B.P. in the different lakes. This phase can be correlated with the regional Tiedemann Neoglacial advance. There is a third major peak i n the last 750 years, correlated with the L i t t l e Ice Age. The chronology of the two latte r events can be seen best in the high resolution record of Kha Lake. The high sedimentation rates of the earlier 3000 - 2000 year event would seem to have fallen to lower levels by 1500 years B.P., starting to r i s e again about 750 year B.P., reaching a maximum ca. 400 years B.P. and again i n the last 100 years . There i s consistency in the timing of the events i n each of the lakes, lending support to both the argument that they represent Neoglacial activity, and to the use of the organic influx calculations as a basis for deriving a chronology of 150 changes. The primary advantage of lake-sediment records i s their continuity, and as demonstrated in this study, their apparent sensitivity to Neoglacial forcing. At no sites within the Coast Mountains i s there t e r r e s t r i a l evidence for a l l three Neoglacial phases and at only one or two sites i s there evidence for more than one event, because the recent L i t t l e Ice Age has destroyed evidence of v i r t u a l l y a l l earlier Neoglacial ac t i v i t y . In general, because of lower rates of sedimentation, i t i s easier to obtain f u l l Holocene Epoch sediment records from downvalley rather than from proglacial lakes. The temporal resolution of these downvalley records is not as great as those of proglacial sediments, but events which lasted of 2 the order of 10 years can be resolved, and such resolution i s appropriate for interpreting changes at timescales of the order 10 3 years. Furthermore because such lakes are below treeline there are more frequent inclusions of organic material which make the records easier to date, thereby providing a chronology against which to interpret changes. Wider application of this lake-sediment approach would permit questions relating to the synchroneity/ diachroneity of palaeoenvironmental change to be addressed, which w i l l lead ultimately to a greater understanding of the mechanisms and nature of the forcing function, namely climatic change. REFERENCES CITED Antevs E. 1922. The recession of the last ice sheet in New England. American Geographical Society Research Series, 11, 120p. Armstrong J.E. 1981. Post Vashon Wisconsin glaciation, Fraser Lowland, British Columbia. Geological Survey of Canada, Bulletin 322, 34p. Bacon C.R. 1983. Eruptive history of Mount Mazama and Crater Lake caldera, Cascade Range, USA. Journal of Volcanology & Geothermal Research, 18, 57-115. Barendsen G.W., Deevy E.S. Jr . & Gralenski I.J. 1957. Yale natural radiocarbon measurements III. Science, 126, 908-919. Barnosky C.W. 1984. Late Pleistocene and early Holocene environmental history of southwestern Washington state, U.S.A. Canadian Journal of Earth Sciences, 21, 619-629. Bassett J., Crompton C.W. & Parmalee J.A. 1978. An atlas of airborne pollen grains and common fungus spores of Canada. Research Branch Canada Department of Agriculture, Monograph 18, 321p. Beaudoin A.G. & King R.H. 1986. Using discriminant function analysis to identify Holocene tephras based on magnetite composition: a case study from Sunwapta pass area, Jasper National Park. Canadian Journal of Earth Sciences, 23, 804-812. Beget J.E. 1983. Radiocarbon dated evidence of worldwide early Holocene climate change. Geology, 11, 389-393. Beget J.E. 1984. Tephrochronology of late Wisconsin deglaciation and Holocene glacier fluctuations near Glacier Peak, N. Cascade Range, Washington. Quaternary Research, 21, 304-317. Birks H.J.B. & Birks H.H. 1980. Quaternary palaeoecology. E. Arnold Press, London, 289p. Blake W. Jr. 1983. Geological Survey of Canada Radiocarbon Dates XXIII. Geological Survey of Canada, Paper 83-7, p.20. Bull P.A. 1978. A s t a t i s t i c a l approach to scanning electron microscope analysis of cave sediments. In Whalley W.B. (ed.), SEM Analysis i n the study of sediments, GeoAbstracts, Norwich, 212-226. Bull P.A. 1981. Environmental reconstruction by electron microscopy. Progress i n Physical Geography, 6, 368-397. Bull P.A. 1985. Procedures in environmental reconstruction by SEM analysis. In Sieveking G. & Hart M.B. (eds.), The Scientific study of Flint and Chert. Cambridge University Press, 221-226. Bull P.A, Culver S.J. & Gardner R. 1980. Chattermarks as palaeo-152 environmental indicators. Geology, 8, 318-322. Campbell W.J. 1973. Structure and inferred circulation of South Cascade Lake Washington, USA. International Association of Sc i e n t i f i c Hydrology Publication, 95, 259-262. Cewe T. & Norbbin J. 1965. Tarfalajakka: Ladtjojakka och Ladtjojaure. Vatten foring, slamtransport och sedimentation. Ymer 1-2, 85-111. Church M. 1980. Records of recent geomorphological events. In Cullingford R.A., Davidson D.A. & Lewin J. (eds.), Timescales in Geomorphology. Wiley & Sons Ltd., Chichester, 13-29. Church M. 1988. Floods i n cold cliamtes. In Baker V.R., Kochel R.C. & Patton P.C. Flood Geomorphology, Wiley, 205-229. Church M. & Gilbert R. 1975. Proglacial f l u v i a l and lacustrine environments. In Jopling A.V. & McDonald B.C. (ed.), Glaciofluvial and glaciolacustrine sedimentation. Society Economic Paleontologists. Special Publication, 23, 22-100. Church M. & Ryder J.M.R. 1972. Paraglacial sedimentation: a consideration of f l u v i a l processes conditioned by glaciation. Geological Society America Bulletin, 83, 3059-3072. Church M. & Slaymaker 0. 1989. Disequilibrium of Holocene sediment yield in glaciated B r i t i s h Columbia. Nature, 337, 452-454. Clague J.J. 1981. Late Quaternary geology and geochronology of British Columbia. Part 2. Summary and discussion of radiocarbon-dated Quaternary history. Geological Survey of Canada, Paper 80-35, 41p. Clague J.J. 1986. The Quaternary stratigraphic record of Brit i s h Columbia: evidence from episodic sedimentation and erosion controlled by glaciation. Canadian Journal of Earth Sciences, 23, 885-894. Clague J.J. 1989. Introduction (Chapter 1). In Quaternary Geology Canada and Greenland. Fulton R.J. (ed.), Geological Survey of Canada, 1, 19 -22. Clark M.W. 1976. Some methods for s t a t i s t i c a l analysis of multimodal distributions and their application to grain-size data. Mathematical Geology, 8, 267-282. COHMAP 1988. Climatic changes in the last 18,000 years: Observations and model simulations. Science, 241, 1043-1052. Cormie A.B. 1981. Chemical correlation of volcanic ashes for use as stratigraphic markers i n archaeology. Unpublished M.A. thesis, Simon Fraser University, Burnaby, British Columbia. Cormie A.B. & Nelson D.E. 1983. Energy-dispersive X-ray fluorescence analysis as a rapid method for identifying tephras. Quaternary Research, 19, 201-211. 153 Crandell D.R. & Miller R.D. 1974. Quaternary stratigraphy and extent of glaciation in the Mount Rainier region Washington. U.S. Geological Survey Professional Paper 847. Culver S.J., Bull P.A., Campbell S., Shakesby R.A. & Whalley W.B. 1983. Environmental discrimination based on quartz grain surface textures: a s t a t i s t i c a l investigation. Sedimentology, 30, 129-136. Davis P.T., Upson S. Sc Waterman S.E. 1979. Lacustrine sediment variations as an indicator of late Holocene climatic fluctuations, Arapaho Cirque, Colorado Front Range. Geological Society of America Abstracts with Programs, 11, 410. Davis N.F.G. & Mathews W.H. 1944. Four phases of glaciation with ill u s t r a t i o n s from southwestern Br i t i s h Columbia. Journal of Geology, 52, 403-413. de Geer G. 1912. A geochronology of the last 12 000 years. International Geological Congress 1910, Comte rendu 11, 241-253. Dearing J. 1982. Core correlations and total sediment. In Berglund B.E. (ed.) Palaeohydrological changes i n the temperate zone in the last 15000 yrs. Subproject B, Lake and Mire environments, IGCP 158B, 1-21. Denton G.H. & Karlen W. 1977. Holocene glacial and tree-line variations in the White River valley and Skolai Pass, Alaska and Yukon Territory. Quaternary Research, 7, 63-111. Desloges J.R. & Ryder J.M. (in press). Neoglacial history of the Coast Mountains near Bella Coola, Br i t i s h Columbia. Canadian Journal of Earth Sciences. Dowdeswell J.A. 1982. Scanning electron micrographs of quartz sand grains from cold environments examined using Fourier shape analysis. Journal of Sedimentary Petrology, 52, 1315-1323. Duffell S. & McTaggart K.C. 1952. Ashcroft map-area, British Columbia. Geological Survey of Canada, Memoir 262, 122p. Ehlrich R., Kennedy S.K. & Brotherhood CD. 1987. Respective roles of fourier and SEM techniques in analyzing sedimentary quartz. In Marshall J.R. (ed.), Clastic particles: Scanning electron microscopy and shape analysis of sedimentary and volcanic clast. Von Nostrand Reinhold Company, New York, 292-301. Environment Canada 1982. Canadian climate normals. Eyles N., Eyles CH. & Miall A.D. 1983. Lithofacies types and vertical p r o f i l e models: an alternative approach to the description and environmental interpretation of gla c i a l diamict and diamictite sequences. Sedimentology, 30, 393-410. Faegri K. & Iversen J. 1984. Textbook of pollen analysis (2nd ed.), Hafner, New York. 154 Fulton R.J. 1971. Radiocarbon dated chronology of southern B r i t i s h Columbia. Geological Survey Canada, Paper 71-37, 28p. Gilbert R. 1975. Sedimentation i n Lillooet lake, B r i t i s h Columbia. Canadian Journal of Earth Sciences, 12, 1697-1711. Gilbert R. & J.R. Desloges. 1987. Sediments of ice-dammed, self dumping Ape Lake, British Columbia. Canadian Journal of Earth Sciences, 24, 1735-1747. Gilbert R. & Shaw J. 1981. Sedimentation in proglacial Sunwapta Lake, Alberta. Canadian Journal of Earth Sciences, 18, 81-93. Gilbert R., Syvitski J.P.M. & Taylor R.B. 1985. Reconnaissance study of paraglacial Stewart Lakes, Baffin Island, D i s t r i c t of Franklin. In Current Research part A. Geological Survey of Canada Paper LA, 505-510. Gomez B., Dowdeswell J. & Sharp M. 1988. Microstructural control of quartz sand grain shape and texture: Implications for discrimination of debris transport pathways through glaciers. Sedimentary Geology, 57, 119-129. Granar L. 1956. Dating of recent f l u v i a l sediments from the estuary of the Angerman river (1859-1950). Geologiska Foreningen Forhaudlingar Stockholm, 78 20-39. Gravenor CP. 1979. The nature of the late Paleozoic glaciation i n Gondwana as determined from an analysis of garnets and other heavy minerals. Canadian Journal of Earth Sciences, 16, 1137-1153. Grove J.M. 1988. The L i t t l e Ice Age. Methuen, London, 498p. Gurnell A.M. & Clark M.J. 1987. Glacio-fluvial sediment transfers: An alpine perspective. Wiley & Sons, Chichester, 524p. Gustavson G.C 1975. Sedimentation and physical limnology in proglacial Malaspina Lake, southeastern Alaska. In Jopling A.V. & McDonald B.C. (eds.), Glaciofluvial and Glaciolacustrine Sedimentation. Society of Economic Paleonotologists and Mineralogists, Special Publication 23, 249-263. Hansen H.P. 1947. Climate versus f i r e and s o i l as factors i n postglacial forest succession i n the Puget Lowland of Washington. American Journal of Science, 245, 265-286. Hansen B.S. & Easterbrook D.J. 1974. Stratigraphy and palynology of late Quaternary sediments in the Puget Lowland Washington. Geological Society America Bulletin, 85, 587-602. Harding J.P. 1949. The use of probability paper for graphical analysis of polymodal frequency distributions. Journal Marine Biological Association, U.K., 28, 141-153. 155 Hartwig F. & Dearing B.E. 1983. Exploratory data analysis. Quantitative applications in the Social Sciences, 16, 83p. Hebda R. 1982. Postglacial history of grasslands of southern B.C. and adjacent regions. In Nicholson A.C., McLean A. & Baker T.E. (eds.), Grassland Ecology and Classification. Symposium Proceedings British Columbia, Ministry of Forests, 157-191. Heusser C.J. 1977. Quaternary palynology of the Pacific slope of Washington. Quaternary Research, 8, 282-306. Heusser C.J. 1983. Vegetation history of the northwestern U.S. including Alaska. In Porter S.C. (ed.), Late Quaternary environments of the U.S. Vol. 1, University Minnesota Press, Minneapolis, Minnesota, Chapter 1, 239 - 258. Higgs R. 1979. Quartz-grain surface features of Mesozoic-Cenozoic sands from Labrador and western Greenland continental margins. Journal of Sedimentary Petrology, 44, 151-157. Hilton J. 1985. A conceptual framework for predicting the occurrence of sediment focusing and sediment redistribution in small lakes. Limnology & Oceanography, 30, 1131-1143. Johnson H. W. 1982. Interrelationships among geomorphic interpretations of the stratigraphic record, process geomorphology and geomorphic models. In Thorn C.E. (ed.) Space and Time in Geomorphology. George Allen & Unwin, London, UK, 219-241. Jones B.F. & Bowser C.J. 1978. The mineralogy and related chemistry of lake sediments. In Lerman A. (ed.), Lakes: Chemistry, Geology and Physics, 179-234. Karlen W. 1976. Lacustrine sediments and tree-line variations as indicators of climatic fluctuations i n Lappland, northern Sweden. Geografiska Annaler 58A, 1-34. „ Karlen W. 1981. Lacustrine sediment studies: a technique to obtain a continuous record of Holocene glacier variations. Geographiska Annaler 63A, 273-281. Keller W.D. & Reesman A.L. 1963. Glacial milks and their laboratory simulated counterparts. Geological Society of America Bulletin, 74, 61-76. King M. 1980. Palynological and macrofossil analyses of lake sediments from the Lillooet area, Br i t i s h Columbia. Unpublished M.Sc. Thesis, Simon Fraser University, Burnaby, British Columbia. King R.H., Kingston M.S. & Barnett R.L. 1982. A numerical approach toward the classification of magnetites from tephra i n southern Alberta. Canadian Journal of Earth Sciences, 19, 2012-2019. Klassen R.A. & Shilts W.W. 1982. Subbottom p r o f i l i n g of lakes of the Canadian Shield i n Current Research, Part A, Geological Survey of Canada, Paper 82-1A, 375-384. Krinsley D.H. 1978. The present state and future prospects of environmental discrimination by scanning electron microscopy. In Whalley W.B. (ed.), Scanning electron microscopy in the study of sediments, Geoabstracts, Norwich, 169-179. Krinsley D. & Doornkamp J.C. 1973. Atlas of quartz sand surface textures. Cambridge University Press, 91p. Kutzbach J.E. & Guetter P.J. 1982. The sensitivity of monsoon climates to orbital parameter changes 9000 yr B.P. Experiments with NCAR Circulation model. In Berger A., Imbrie J., Hays J., Kukla G. & Slatzman B. (eds). Milankovitch and Climate, Reidel, Dordrecht, 801-820. Lavkulich L. 1981. Methods manual Pedology Lab, Department of Soil Science, University of Br i t i s h Columbia, Vancouver, 121 p. Lemmen D.S. 1984. Sedimentation i n glacially-fed lake: Tasikutaak Lake, Cumberland Peninsula, Baffin Island, N.W.T. Unpublished M.Sc Thesis, Queen's University, Kingston, Ontario. Leonard E. 1981. Glaciolacustrine sedimentation and Holocene gla c i a l activity, northern Banff national Park, Alberta. Unpublished PhD Thesis, University of Colorado, 27lp. Leonard E.M. 1986a. Varve studies at Hector Lake, Alberta, Canada and the relationship between gl a c i a l a c t i v i t y and sedimentation. Quaternary Research 25, 199-214. Leonard E.M. 1986b. Use of lacustrine sedimentary sequences as indicators of Holocene glacial history, Banff National Park, Alberta, Canada. Quaternary Research, 26, 218-231. Leopold E.B., Nickman R., Hedges J.I. and Ertel J.R. 1982. Pollen and lignin records of late Quaternary vegetation, Lake Washington. Science 218, 1305-1307. Likens G.E. & Davis M.B. 1975. Postglacial history of Mirror lake and i t s watershed in New Hampshire, U.S.A. Internationale Verein Theor Agnew Limnologie Verhein, 19, 982-993. Liverman D.G.E. 1980. Sedimentology and drainage history of a glacier dammed lake, St. Elias Mountains, Yukon Territory. Unpublished M.Sc. Thesis, University of Alberta. Lowdon J.A. & Blake W. J r . 1968. Geological Survey of Canada radiocarbon dates VII. Radiocarbon 10, 207-245. Lowdon J.A. & Blake W. Jr. 1973. Geological Survey of Canada radiocarbon dates XIII. Geological Survey of Canada, Paper 73-7, 61 pp. Lowdon J.A. & Blake W. Jr. 1975. Geological Survey of Canada radiocarbon dates XV. Geological Survey of Canada, Paper 75-7, 32 pp. Lowdon J.A., Robertson I.M. & Blake W. Jr. 1971. Geological Survey of Canada radiocarbon dates XI. Radiocarbon 13, 255-324. Luckman B.H. & Osborn G.D. 1979. Holocene glacier fluctuations in the middle Canadian Rocky Mountains. Quaternary Research, 11, 52-77. Margolis S.V. & Kellner E. 1969. Quantitative palaeoenvironmental determination of ancient sands using electron microscopy and d i g i t a l computer techniques. Geological Society Abstracts with Programs, 7, 142-143. Margolis S.V. & Kennett J.P. 1971. Cenozoic g l a c i a l history of Antarctica recorded i n sub Atlantic deep-sea cores. American Journal of Science, 271, 1-36. Matalas N.C. 1963. Autocorrelation of r a i n f a l l and streamflow minimums. U.S. Geological Survey, Professional Paper 434-B, lOp. Mathewes R.W. 1973. A palynological study of postglacial vegetation changes in the University research Forest, southwestern British Columbia. Canadian Journal of Botany, 51, 2085-2103. Mathewes R.W. 1984. Palaeobotanical evidence for climatic change in southern B.C. during the late-glacial and Holocene time. Syllogeus, 55, 397-422. Mathewes R.W. & Heusser L.E. 1981. A 12 000 year palynological record of temperature and precipitation trends i n southwestern British Columbia. Canadian Journal of Botany, 59, 707-710. Mathewes R.W. & Rouse G.E. 1975. Palynology and palaeoecology of postglacial sediments from Lower Fraser canyon of British Columbia. Canadian Journal of Earth Sciences, 12, 745-756. Mathewes R.W. & Westgate J. 1980. Bridge River tephra: revised distribution and significance for detecting old carbon errors in radiocarbon dates of limnic sediments in southern B.C. Canadian Journal of Earth Sciences, 17, 1454-1461. Mathewes R.W., Borden C.E. & Rouse G.E. 1972. New radiocarbon dates from the Yale area of the lower Fraser River canyon, British Columbia. Canadian Journal of Earth Sciences, 9, 1055-1057. Mathews W.H. 1968. Geomorphology, southwestern Br i t i s h Columbia. Mathews W.H.(ed.), Guidebook for geological f i e l d trips i n southwestern Br i t i s h Columbia, Department of Geology, University of b r i t i s h Columbia, Report 6, A p r i l 1968, 18-23. Mathews W.H. 1951. Historic and prehistoric fluctuations of alpine glaciers in the Mount Garibaldi map area, southwestern Br i t i s h Columbia. Journal of Geology, 59, 357-380. 158 Mathews W.H. 1956. Physical limnology and sedimentation in a glacial lake. Geological Society America Bulletin, 67, 537-552. Middleton G.V. 1976. Hydraulic interpretation of sand size distributions. Journal of Geology, 84, 405-426. Miller CD. 1969. Chronology of Neoglacial moraines in the Dome Peak area North Cascade Range, Washington. Arctic & Alpine Research, 1, 49-66. Monger J.W.H. 1982. Geology of the Ashcroft Map area, southwestern Br i t i s h Columbia. In Current Research, Part A. Geological Survey of Canada Paper 82-1A, 293-297. Monger J.W.H. 1982. geological Survey of Canada Open F i l e 980, Ashcroft. Moore P.P. & Webb J.A. 1978. An il l u s t r a t e d guide to pollen analysis. Chaucer Press Limited, Suffolk, 133p. Mullineaux D.R., Hyde J.H. & Rubin M. 1975. Widespread late glacial and postglacial tephra deposits from Mt St Helens Volcano, Washington. U.S. Geological Survey, Journal of Research 3, 329 - 335. Norbbin J. 1973. Vattenforing och slamtransport i Tarfala och Ladtjojakka 1960-1967. Unpublished Thesis, University of Stockholm. Oldfield F. 1977. Lakes and their drainage basins as units of sediment based ecological study. Progress i n Physical Geography, 1, 460-504. Osborn G. & Luckman B. H. 1988. Holocene glacier fluctuations in the Canadian Cordillera. Quaternary Science Reviews, 7, 115-128. Ostrem G. 1975. Sediment transport i n glacial meltwater streams. In Jopling A.V. & McDonald B.C. (eds.), Glaciofluvial and glaciolacustrine sedimentation. Society Economic Paleontologists & Mineralogists, Special Publication, 23, 304-320. Perkins J.A. & Sims J.D. 1983. Correlation of Alaskan varve thickness with climatic parameters and use i n paleoclimatic reconstruction. Quaternary Research, 20, 308-321. P i c k r i l l R.A. & Irwin J. 1983. Sedimentation in a deep glacier-fed lake: Lake Tekapo, New Zealand. Sedimentology 30, 63-75. Reading H.G. 1978. Facies. In Reading H.G. (ed.) Sedimentary environments, Blackwell, Oxford, 4-14. Reasoner, M.A. & Healy R.E. 1986. Identification and significance of tephras encountered i n a core from Mary Lake, Yoho National Park, British Columbia. Canadian Journal of Earth Sciences, 23, 1991-1999. Ritchie J.C. 1987. Postglacial vegetation of Canada. Cambridge University Press, 178 p. Rothlisberger H. & Lang H. 1987. Glacial hydrology. In Gurnell A.M. & Clark M.J. (ed.), Glacio-fluvial sediment transfers: An alpine perspective. Wiley & Sons, Chichester, 207-273. Ryder J.M. 1976. Terrain inventory and Quaternary Geology, Ashcroft, British Columbia. Geological Survey of Canada Paper 74-49, 17p. Ryder J.M. 1981. Terrain inventory and Quaternary Geology, Lytton, Br i t i s h Columbia. Geological Survey of Canada, Paper 79-25, 20p. Ryder J.M. 1987. Neoglacial history of the Stikine-Iskut area, northern Coast Mountains, British Columbia. Canadian Journal of Earth Sciences, 24, 1294-1301. Ryder J.M. 1989. Holocene glacier fluctuations (Canadian Cordillera). In Fulton R.J. (ed.), Quaternary Geology Canada and Greenland. Geological Survey of Canada, 1, 74 - 76. Ryder J.M. & Clague J.J. 1989. Quaternary stratigraphy and history: Area of Cordilleran Ice Sheet, Br i t i s h Columbia. In Fulton R.J. (ed.), Quaternary Geology Canada and Greenland. Geological Survey of Canada, 1, 48-58. Ryder J.M. & Howes D.E. 1984. Terrain Information: A user's guide to terrain maps in British Columbia. B.C. Ministry of Environment, Surveys and Resource Mapping Branch, 16p. Ryder J.M. & Thomson B. 1986. Neoglaciation of the southern Coast Mountains of British Columbia: chronology prior to the late Neoglacial maximum. Canadian Journal of Earth Sciences, 23, 273-287. Saunders I.R., Clague J.J. & Roberts M.C. 1987. Deglaciation of Chilliwack River valley, British Columbia. Canadian Journal of Earth Sciences, 24, 915-923. Schmok J. 1986. Sedimentology and chronology of Neoglacial Lake Alsek, Yukon Territory. Unpublished M.Sc. Thesis, Department of Geography, University of British Columbia, 182p. Schytt V. 1963. Glaciarernas l i v . Svenska Tuistforeningens Arsskrift 1963, 144-158. Shea J.H. 1974. Deficiencies of c l a s t i c particles of certain sizes. Journal of Sedimentary Petrology, 44, 985-1003. Shilts W.W. & Farrell L.E. 1982. Subbottom pr o f i l i n g of Canadian Shield Lakes: implications for interpreting the effects of acid rain. In Current Research, Part B, Geological Survey of Canada Paper 82-1B, 209-221. Sinclair A.J. 1974. Selection of threshold values in geochemical data using probability graphs. Journal of Geochemical Exploration, 3, 129-149. Sly P.G., Thomas R.L. & Pelletier B.R. 1983. Interpretation of moment 160 measures from water l a i n sediments. Sedimentology, 30, 219-233. Smith N.D. 1978. Sedimentary processes and patterns in a glacier-fed lake with low sediment input. Canadian Journal of Earth Sciences, 15, 741-756. Smith N.D., Vendl M.A. Sc Kennedy S.K. 1982. Comparison of sedimentation regimes in four glacier-fed lakes of western Alberta. In Davidson-Arnott R., Nicking, W. & Fahey B.D. (eds.). Research in Glacial, Glacio-fluvial and Glacio-lacustrine systems. Proceedings of the 6th Guelph Symposium on Geomorphology, 1980. Geobooks, Norwich, 203-238. Stein R. 1985. Rapid grain-size analyses of clay and s i l t fraction by SediGrpah 5000D: comparison of Coulter Counter and Atterburg methods. Journal of Sedimentary Petrology, 55, 590-615. Stuiver M., Deevey E.S. & Garlenski L.J. 1960. Yale natural radiocarbon measurements V. American Journal of Science, Radiocarbon Supplement, 2, 49-61. Surgenor J.W. 1978. Lacustrine sediments as indicators of fluctuations of Riukojietha Glacier, Lappland, Sweden. Unpublished M.A. thesis, University of Maine, Orono, 54p. Ting W.S. 1965. The saccate pollen grains of Pinaceae mainly of California. Grana Palynology, 6, 270-289. Tovey N.K. & Wong K.Y. 1978. Preparation, selection and interpretation problems in scanning electron microscope studies of sediments. In Whalley W.B. (ed.) Scanning Electron Microscopy in the study of sediments, GeoAbstracts, Norwich, 181-200. Trewin N. 1988. The use of the SEM i n sedimentology. In Tucker M (ed.). Techniques in Sedimentology, Blackwell, Oxford, 229-274. Walkley A. 1946. A c r i t i c a l examination of a rapid method for determining organic carbon in soi l s - effects of variations in digestion conditions and inorganic s o i l constituents. Soil Science, 63, 251-264. Weirich F.H. 1985. Sediment budget for a high energy glacial lake. Geografiska Annaler, 67A, 83-99. Westgate J.A. Sc Gorton M.P. 1981. Correlation techniques i n tephra studies. In Self S. Sc Sparks R.S.J, (eds.), Tephra Studies, Reidel, Dordrecht, 73-94. Wilcox R.E. 1965. Volcanic ash chronology. In Wright H.E. Jr. Sc Frey D.C. (eds.) the Quaternary of the United States. Princeton University Press, Princeton, New Jersey, 307-316. Wright H.E. Jr. 1967. A square-rod piston sampler for lake sediments. Journal of Sedimentary Petrology, 37, 975-976. Visher G.S. 1969. Grain size distributions and depositional processes. Journal of Sedimentary Petrology, 39, 1074-1106. 161 Yarnel B. 1982. The relationship between synoptic scale atmospheric circulation and glacier mass balance in southwestern Canada. Unpublished Ph.D. Thesis, Simon Fraser University, Burnaby, B r i t i s h Columbia. Yevjevich V.M. 1972. Stochastic processes in hydrology. Water resources Publications, Fort Collins, Colorado. Zelinski G. & Davis P.T. 1987. Late Pleistocene age of type Temple Lake Moraine, Wind River, Wyoming, U.S.A. Geographie Physique et Quaternaire, XLI, 397-401. 162 APPENDIX I LITHOSTRATIGRAPHY OF THE SEDIMENTS Ideally, a facies i s a distinct sediment deposit that forms under certain conditions of sedimentation, reflecting a particular process or environment (Reading, 1978). The Kwoiek Creek lake sediments are cla s s i f i e d on the basis of sedimentary structures, lithostratigraphy, sedimentary properties and grain size analyses as outlined in section 3.5. This i s an adaptation of the scheme developed by Eyles et a l . (1983) and ut i l i s e d by Schmok (1986) in a similar context. The objective i s the systematic documentation of the sedimentary sequences based on visual appraisal to allow prescreening and removal of deposits that are not representative of changes in sedimentation patterns throughout the lakes. For representation purposes the sediments are categorised into six major sedimentary units: laminated and massive s i l t s , laminated and massive fine sands, rhythmites, and laminated blue-grey clays, with inclusions of organic matter, ash and larger particles (stones) recorded separately. The more detailed characteristics of each of the deposits are discussed in Chapter 4. Only the composite core log i s presented for each site (see Chapter 3 for explanation of coring strategy). 7 163 Core: Kha 1 10 2o_ 3 0^*" p<t<*a4<*«**v*>**<t 50. 60. / / / /./ t t t t t t t l 40- .v.v.v/.v.v.v. 7Q_ 80_ • r • r • • t ' I • Sedimentary units Organic matter Fine sand Fine sand/ s i l t s VM Silts •:->»}) Laminated silts \ \ \ • • • , X \ V Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90_ 100 164 Core: Kha 2 10 20-70-40-5cu wmfwi 80_ Sedimentary units Organic matter Fine sand Fine sand/ s i l t s &•& siits v>>-> Laminated silts Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90-100 Core Kha 2 (cont.) O t-: 2,4 K tk. 4 0 -60_ 70_ I 2350 +110 Sedimentary units Organic matter Fine sand Fine sand/ silts Silts Laminated silts Laminated sands Ash Organic matter & silts Stones 60_ Laminated blue-grey clays Core: Kha 3 Sedimentary units Organic matter Fine sand Fine sand/ silts SH: Silts Laminated A silts \ \ \ ' * • \ \ \ Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 167 Core: Kha 4 r.v r » 2C_h r - * r - ' 40-50-60_ SO-m Sedimentary units Organic matter Fine sand Fine sand/ si l t s 8)8 snts \ \ \ ' • • I \ s \ Laminated silts Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90_ iOO Core: Kha 5 168 9 0 -Sedimentary units Organic matter Fine sand Fine sand/ s i l t s i l l 5iHs Laminated t ^ J silts VTT \ % N N V V Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays i 0 0 Core : Kha 6 10 !•: Sedimentary units Organic, matter Fine sand LV Fine sand/ silt s — r • SSfl Silts 4 0 - ; v !•••' r"-5 0 j I-"-' 7CU.V™™.-." T s s Laminated silts Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays Core: Klept 1 170 40- •: 20-m$WM 30-2400 50 -6800 ;*;*v Sedimentary units Organic matter Fine sand Fine sand/ si l t s mil Silts Laminated silts Laminated sands 60- v \ x ' • • \ \ \ Ash 70-Orosnlc matter & silts Stones 80- Laminated blue-grey clays 90-10Q_ 171 Core: Klept I 204i 30-: 40-:  50-f 9640 +380 60-Sedimentary units Organic matter Fine sand Fine sand/ si l t s Silts Laminated silts Laminated sands XXX • • \ X X Ash 70-Oroonlc matter & silts Stones 80- Laminated blue-grey clays 90-100 Core: Klept 2 172 50-2400 70-Sedimentary units Organic matter \ % \ • • • N \ \ Fine sand Fine sand/ s i l t s Silts Laminated silts Laminated sands Ash Organic matter & silts Stones 80- Laminated blue-grey clays 90-100_ Core: Klept 3 173 104* 2 0 J ; : i : ^ ;l 404 50-' . . . ... . . . mmmmm wwwwwww 2400 Sedimentary units Organic matter m j \ N X X X \ Fine sand Fine sand/ silts Silts Laminated silts Laminated sands Ash 70-Organfc motter & slits Stones 80- Laminated blue-grey clays 90-Core. Klept 4 174 10 fMMt' 20_; £ o 30-~7 4Q_ 50- .v/.v.v.v.-.v.v 60_.v.v.y.>v.-.>y 70--'•mi vl'X'Xv 2400 Sedimentary * units Organic matter Fine sand Fine sand/ si l t s Silts x \ x Laminated silts Laminated sands Ash Organic matter & silts Stones 80-Laminated blue-grey clays 90_ 100 175 Core: Klept 5 2a 60-70-80-I M I I I H I I I I I Sedimentary units Organic matter Fine sand Fine sand/ s i l t s \ \ % X N N Laminated silts Laminated sands Ash Orrjanic matter & silts Stones Laminated blue-grey clays 90_ 100 176 Core: Klept 6 10 20_:; 50-60-70 30-111 I I 1 1 ; 1 1 t 1 1 ft**: 2400 23 6800 Sedimentary units Organic matter Fine sand Fine sand/ si l t s ••im Silts Laminated tv^d silts \ % x • s \ \ \ 80-Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90_ 100 Core: Kokwaskey 1 4QJ 50J 20Jmmwm Sedimentary units Organic matter Fine sand Fine sand/ s i l t s Silts Laminated silts Laminated sands • / . \ \ \ • * ( \ \ \ Ash 60- Organlc matter & silts 70- Stones 80-Laminated blue-grey clays iOO C o r e : K o k w a s k e y 2 ! « 1 i l l l i l I 6Q4::::>:::::::v::::::::::: I SO-SO. 178 Sedimentary units • Organic matter Fine sand Fine sand/ SS:- s i l t s X-X-> Laminated silts Laminated sands \\\ Ash | j p | Organic i l l matter & silts Stones rr^n Laminated >X* blue-grey clays 100 rr -rn Laminated b lue-grey clays 90 Rhythmites 100, Core: Kokwaskey (cont.) 180 20-; 30-40 50_ 60-70. * * * * * * * Sedimentary units Organic matter Fine sand Fine sand/ silt s T-T->q • • /• \ \ x ' * • \ \ \ 80_ Silts Laminated silts Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90-100 C o r e : K o k w a s k e y 4 40-Sedimentary units Organic matter Fine sand Fine sand/ silt s V: Silts Laminated si Us Laminated sands 60- \ \ \ / • \ \ \ Ash 70-• Organic matter & silts Stones 30- Laminated blue-grey clays 90-182 Core: Kokwaskey 5 1 2 3 4 Sedimentary • 1 i i iv .v.'.1.'.'.'.'.'.v.'i t . ••.) units 90_ 100 Core: Kokwaskey (cont.) 10 30-60-70_ 40-:;:;:;:;:!:;: . 11 485 +185 Sedimentary units Organic matter Fine sand Fine sand/ s i l t s Silts ;- Laminated 3 silts \ \ N \ X \ 80_ Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90-100 C o r e : K w o i e k 2 10->: 20-:;: 30-40-Sedimentary units Organic matter Fine sand Fine sand/ silt s Silts Laminated silts 50-60-70-80-90-Laminated sands IN «• V I N N N ' • • N N N Ash Organic matter & silts Stones Laminated blue-grey clays 100-Core: Kwoiek 3 2i 3' 40-50-60-70-80-90-100. Sedimentary units • Organic matter Fine sand Fine sand/ s i l t s Silts Laminated silts = Laminated = ' sands S ^ A s h |—I Organic I matter & Stones rrpn Laminated ;*» blue-grey clays 186 C o r e : K w o i e k 4 80- Laminated blue-grey clays 90-187 Core: Kwoiek 5 10-f: Sedimentary units Organic matter 20-:-: Fine sand 30->W: Fine sand/ silts ::•:•:•:•: Silts «> ° 40-f: Laminated silts 50-:::::;:::::::::::::>:::::: Laminated sands \ \ X • • • \ \ N Ash 70-? Organic matter & silts Stones 80- Laminated blue-grey clays 90-188 C o r e : K w o i e k 6 80- Laminated blue-grey clays 90-189 Core : K w o i e k 7 £ 104;: 204-: 304i: • >.f.>.C.Nfi 2400 a 40-E • * *' 1 II Sedimentary units Organic matter Fine sand Fine sand/ s i l t s mA Silts Laminated silts Laminated sands 60-' f 1 XXX • • .' x \ \ Ash Organic matter & silts Stones 80- Laminated blue-grey clays 90-Core: Kwoiek 8 lO-:i £ 30-:;:;:;;:::^  2400 t • I • I . 1 .1 • I • I 60. Sedimentary units Organic matter Fine sand Fine sand/ silts :;:•:•:;: Silts Laminated silts X <. x| X X X ' • • X X X Laminated sands Ash 70-Oroanlc matter & silts Stones 80- Laminated blue-grey clays 90-191 Core : K w o i e k 9 m 20-|;: t:!:::::x::::::S::::: 1. 30-| 4) O 40-i 2400 Sedimentary units Organic matter Fine sand Fine sand/ s i l t s m& Silts Laminated silts Laminated sands 6 a ' / * \ \ \ • • • \ \ \ Ash 70-® Oroanlc matter 6V silts Stones 30- Laminated blue-grey clays 90-Core : K w o i e k 10 192 304: 404 50-f 604' j.i.HMmnni 70-Sedimentary units Organic matter Fine sand Fine sand/ si l t s Silts Laminated silts Laminated sands Ash Organic matter & silts Stones 80- Laminated blue-grey clays 90-193 Core: Kwoiek I 1 j o - - : - : 20--x £ 304: u a. 40- S 6 0 - S : ^ ^ : ^ 70-80-i n i i n l i H M Sedimentary units Organic matter Fine sand Fine sand/ silt s Silts Laminated silts Laminated sands <. <. % \ \ N * * K X \ Ash Organic matter & silts Stones Laminated blue-grey clays 90-194 C o r e : K w o i e k 12 10-:; 0> 40-* so-; 60-> Sedimentary units Organic matter (.<. v y • s / \ \ V X • \ \ \ 70-Fine sand Fine sand/ si l t s Silts Laminated silts Laminated sands Ash Organic matter & silts Stones 80- Laminated blue-grey clays 90-Core Kwoiek 13 195 10-J 404 504: 60-fc: 704i 1 1 1 1 1 i l l I t i l l * 2400 Sedimentary units Organic matter Fine sand Fine sand/ si l t s W:S Silts X X x X X X f * X X X Laminated silts Laminated sands Ash Organic matter & silts Stones 80- Laminated blue-grey clays 90-196 Core: Kwoiek 14 £ Cv {•: t-: •:• 20-ii 30- mm& a 40J. 2400 Sedimentary units Organic matter Fine sand Fine sand/ s i l t s m< Silts f • \ \ X f / \ \ \ Laminated silts Laminated sands Ash Organic matter & silts Stones Laminated blue-grey clays 90-197 Core: Kwoiek 15 20.-: E 30- v 2 40-50- i V',".' iiji i i i i i l i l i 60-2400 70-80-• i i i i i i i i i i v,v,v,v Sedimentary units Organic matter i i i 255 +770 Fine sand Pine sand/ silt s Silts Laminated silts Laminated sands Ash • Organic matter & silts Stones Laminated blue-grey clays 90-APPENDIX II CHRONOLOGY I. 1 Radiocarbon chronology Organic material submitted for radiocarbon dating was extracted from cores i n the lab, cleaned of mineral material and immediately sealed. The samples were then sent to the Saskatchewan Research Council Radiocarbon Laboratory where ro u t i n e preparation procedures were followed which involved p r e s o r t i n g to remove v i s i b l e r o o t l e t s , treatment with sodium hydroxide to remove the soluble organic f r a c t i o n s , and treatment with HCI to remove the sodium hydroxide and any carbonates due to leaching. The dates obtained are presented i n Table A I I . l . The materials that have been dated were i d e n t i f i e d as s p e c i f i c a l l y as p o s s i b l e (Table A I I . l ) . With the exception of the basal sample from Kokwaskey (S-2935) they are remnants of t e r r e s t r i a l f l o r a : cones, needles and pieces of wood. I t i s assumed that no post-depositional mixing or contamination of the organic m a t e r i a l occurred, and so a l l dates are int e r p r e t e d as maxima for the u n i t s i n which they are enclosed, although they may provide minimum dates f o r events within the watershed, for example d e g l a c i a t i o n . Discussion of the s i g n i f i c a n c e of the dates i s incorporated within the body of the t h e s i s . I I . 2 Tephra I d e n t i f i c a t i o n In a number of cores two d i s t i n c t tephra layers were observed: the lower at approximately 2/3 depth i n the cores, a f i n e r grained and a white-yellow colour (10 YR 7/3); the upper, approximately 1/4 depth, much coarser and grey (10 YR 7/1). It was important to e s t a b l i s h the i d e n t i t y of these tephras i n order to be able to use them as t i m e - s t r a t i g r a p h i c markers i n conjunction with radiocarbon dates to provide a chronology of sedimentation events. 199 Table AII.l Radiocarbon dates obtained in the study. Stratigraphic positions and core logs i n Appendix I Date Max Min Sample ID Site location Material dated 12 255 +770 13 795 10 715 S - 3010 5 035 +1875 8 785" 1 285 S - 2970 1 240 + 245 1 730 750 S - 2987 Kwoiek lake Kwoiek lake Kwoiek lake Engelmann Spruce cones & needles Wood Wood 11 485 +185 11 855 11 115 S - 2935 4 900 +325 5 550 4 350 S - 2988 Kokwaskey Kokwaskey Granular algal material Wood 9 640 +380 10 400 8 880 S - 3011 Klept 1 380 +390 2 160 600 S - 2989 Klept Cones & needles Wood 2 350 +110 2 570 2 130 S - 2936 Kha Wood Calculations based on + two standard deviations 200 Three Holocene volcanic eruptions distributed tephra over south-western Bri t i s h Columbia: Mazama (6800 BP) (Bacon, 1983), Mt. St. Helens Yn (3400 BP) (Mullineaux, 1975) and Bridge River (2400 BP) (Mathewes & Westgate, 1980). The mapped extent of fall-out plumes suggests that Mazama and Bridge River tephras are present i n the Kwoiek Creek watershed. Hence the primary effort was directed towards differentiating these two tephras, although careful comparisons were made against literature values for other tephras to ensure that no erroneous identifications were made. Many techniques have been proposed for the identification and differentiation of tephras: optical methods involving the determination of the refractive index of glass (e.g. Wilcox, 1965); bulk chemical analysis (e.g. Westgate and Gorton, 1981); and, chemical analysis of the glass (e.g. X-ray emission; emission spectrographs; neutron activation analysis; electron microprobe; thermomagnetic properties; X-ray fluorescence). Electron microprobe analysis of glass encased titano-magnetites i s usually the preferred method because i t can be conducted on single shards of glass and thus minimises problems of contamination. However, i t i s unsuited for the differentiation of Mazama and Bridge River tephras because their compositional fields of titano-magnetites are similar (King et a l . , 1982; Beaudoin and King, 1986). In a study of the Pacific Northwest tephras Cormie & Nelson (1983) conclude that X-ray fluorescence (XRF) analysis of bulk glass separates provides a reliable method for rapidly identifying and differentiating between Mazama, Mount St. Helens Yn and Bridge River tephras. They found that the <62 Mm size-fraction was chemically very similar to the glass separates. In addition, i n areas where only the Mazama, St Helens Yn and Bridge River tephras occur they found that i t i s possible to simplify the 201 procedure even further. The chemical treatments (with HCI and NaOCl to remove organic stains, metal oxides and carbonates) can be omitted for clean samples as they have l i t t l e effect on the concentrations of Y, Zr and Nb which are different in these three tephras (Cormie, 1981). The Mazama tephra can be distinguished by i t s high Zr concentration, from direct analysis of the <62 um. The potassium concentration in Mount St. Helens Yn i s characteristically low and may be useful for distinguishing this tephra from Bridge River. However, problems with potassium due to contamination and weathering effects may be greater. II.2.1 Methodology in this study X-ray fluorescence analysis of both minor and major element concentrations was used i n this study i n conjunction with microscope observations of shard morphology to identify and differentiate the tephras. The tephras were tentatively identified on the basis of features observed in freshly s p l i t cores and by distinctive properties discernible from standard petrographic techniques, including colour, texture, glass shard habit, and phenocryst assemblages. Photomicrographs (Figure AII.l) show the distinct nature of the glass shards. The lower tephra has glass shards which are commonly thin, bubble-wall fragments. The upper deposit has typically chunky shards and displays lineated gas vesicules. These morphologies are very similar to those documented by Reasoner and Healy (1986) for Mazama and Bridge River tephra, respectively, deposited in Mary Lake in the Canadian Rockies. In a l l cases the samples were dry sieved through a 62 um sieve and the smaller size fraction was collected for analysis. The rationale for this 2 0 2 Figure All.2 Photomicrographs of glass shards of Bridge River and Mazama tephra (shards approximately 150 pm in length) a) Bridge River tephra b) Mazama tephra stems from the findings of Cormie and Nelson (1983), who state that by removing the larger fraction most of the phenocrysts are also removed, which are the major contaminants. It i s assumed that the < 62 nm fraction i s composed primarily of glass. In this study this was confirmed by viewing the samples < 62 nm fraction analyzed under an optical microscope. X-ray fluorescence analysis of a series of standards, samples of known origin and unknown tephras were analyzed for major (Fe, Mn, T i , Ca, Ca, K, Si , A l , Mg oxides) and minor (Nb, Zr, Y, Sr, Rb, Pb, Zn, Cu, Ni, Co, Cb, Mn, T i , V, Cr, Ba) element concentrations using a Philips PW 1400 XRF machine, in the Department of Oceanography, The University of British Columbia. The raw data were used to calculate means and standard devia-tions for each element concentration for the known tephras. These are reported with the values from Cormie & Nelson's study in Table All.2. Following Cormie & Nelson (1983) a simple s t a t i s t i c a l measure of d i f -ference, the "A s t a t i s t i c " , was computed to determine whether the relative concentrations of the elements were sufficiently different from tephra to tephra to differentiate the samples. Such identification requires that the differences between the averaged concentrations for the different tephras be large compared to the v a r i a b i l i t y of one tephra between sites: A . . = I x . - x .1 (AII.l) — - 1 e,i e , ] l ei,3 2(S . + S .) e,i e,] where x . and x ,. are the average relative concentrations of element e e,l e n . , . . J for tephras l and j S . and S . are the corresponding standard deviations e, i e,j The A coefficient serves to identify those elements that most usefully characterise the tephra. For any element, i f A > 1 then individual samples from 2 tephras can be correctly identified with a probability > 95% using Table All.2 Element concentrations for Mazama and Bridge River tephra Element Mazama Bridge River X S.D. X S.D. Majors Si 67.93 1.15 67.64 2.36 Al 14.33 0.18 14.70 0.18 Ti 0.53(0.25) 0.04(0.03) 0.56(0.20) 0.07(0.03) Fe 2.89(1.28) 0.47(0.24) 3.01(1 • 17) 0.65(0.23) Ca 2.20(1.39) 0.37(0.71) 2.29(1 .52) 0.03(0.75) Mg 0.88 0.32 0.98 0.33 K 2.68(2.16) 0.28(0.56) 2.54(2 .51) 0.24(0.62) Na 4.57 0.24 4.21 0.74 Mn 0.06 0.02 0.07 0.007 P 0.11 0.01 0.17 0.014 S Not adequately calibrated 2 Minors Ba 822.20 135.70 803.05 85.77 Co 8.70 3.62 8.80 4.95 Cr 20.27 7.74 36.15 30.19 Cu 36.88 5.49 29.58 9.92 Mn 542.18 64.82 639.80 182.29 Na 4.18 0.40 3.66 0.89 Nb 7.58(12) 1.04(3) 10.10(16) 0.42(4) Ni 14.00 4.18 7.00 7.21 Pb 50.60 55.70 19.05 3.04 Rb 50.60(50) 5.04(6) 42.60(45) 0.71(5) Sr 336.63(260) 59.67(19) 349.00(290) 33.09(27) V 47.23 14.56 65.50 16.40 Y 25.13(21) 1.63(4) 18.30(13) 0.71(4) Zn 59.05 7.31 82.35 28.35 Zr 217.75(219) 17.92(18) 160.45(136) 9.83(17) 2 oxide % _ concentrations ppm x mean S.D. one standard deviation values in brackets are those reported by Cormie Sc Nelson (1983) 205 that element alone. The A c o e f f i c i e n t requires that the elemental concentrations be normally d i s t r i b u t e d . This was v e r i f i e d by Cormie (1981). The A - c o e f f i c i e n t s f o r the known samples run i n t h i s i n v e s t i g a t i o n are reported i n Table A l l . 3 . Zirconium (Zr) was found to be a us e f u l d i s -criminator, as was Yttrium (Y). The A s t a t i s t i c can be adapted, to an I s t a t i s t i c (Cormie & Nelson, 1983), so that an unknown sample be i d e n t i f i e d , by q u a n t i t a t i v e comparison with the known tephras: I . . = Ix . - x .1 (Al l . 2 ) ei,D 1 e , i e,] 1 2(Se .) i 3 where x g,^ i s the concentration of element e i n the sample i xe » j i s the average concentration of element e i n the known tephra j S . i s the standard d e v i a t i o n of element e f o r tephra j e, ] An I value < 1 in d i c a t e s a sample ( i ) should be c l a s s i f i e d as reference tephra ( j ) , a value > 1 in d i c a t e s that the two are d i s s i m i l a r . For each sample the I values are scanned element by element rather than combining the d i f f e r e n t elements i n t o a s i n g l e i d e n t i f i c a t i o n s t a t i s t i c , because the concentrations of c e r t a i n elements can be extremely s e n s i t i v e to low l e v e l s of contamination and weathering e f f e c t s . Abnormal I values i n d i c a t e that the samples should undergo a d d i t i o n a l treatments. Such information may not be apparent i n a combined s t a t i s t i c . Examples of the r e s u l t s of the I - s t a t i s t i c s f o r data c o l l e c t e d i n t h i s study, and i d e n t i f i c a t i o n s based on these are presented i n Table A l l . 4 . These r e s u l t s are consistent with the pre l i m i n a r y i d e n t i f i c a t i o n s of the tephra u n i t s as Bridge River and Mazama tephra within the cores. In a l l cases the lower deposit was i d e n t i f i e d as Mazama and the upper as Bridge River, with no evidence of reworking and m u l t i p l e d e p o s i t i o n at any s i t e . 206 Table All.3 The A s t a t i s t i c for comparison of Mazama and Bridge River tephra Element "A" s t a t i s t i c Mazama/B.R. Si 0.004 Al 0.51 Ti 0.14 Fe 0.05 Ca 0.11 Mg 0.08 K 0.14 Na 0.18 Mn 0.09 P 1.25 Ba 0.04 Co 0.005 Cr 0.21 Cu 0.24 Mn 0.20 Na 0.20 Nb 0.86 Ni 0.13 Pb 0.26 Rb 0.70 Sr 0.007 V 0.30 Y 1.46 Zn 0.32 Zr 1.03 Values >1.00 indicate the elements which are most useful in discriminating between Mazama and Bridge River tephra Table All.4 For Mazama Examples of I-statistics for "unknown samples C8 C9 CIO C l l C15 Al 1.38 0.25 1.08 0.11 P 2.50 1.00 3.50 0.50 Nb 1.36 0.13 0.18 1.87 0.63 Y 1.94 0.60 0.70 2.16 0.30 Zr 1.41 0.29 0.05 1.72 0.28 For Bridge River Al 0.36 0.77 0.44 1.14 P 0.35 1.43 0.36 1.79 Nb 0.35 3.33 3.45 0.83 1.43 Y 0.35 6.20 6.41 0.14 5.49 Zr 0.35 2.38 3.01 0.22 3.43 Ident. B.R. M* M B.R. M some ambiguity with Al and P, yet other elements show strong di s crimination 208 II.3 Dated anthropogenic e f f e c t s In a l l the lakes changes i n the sediments of some of the cores f o r the most recent period i s evident. The uppermost 2-8 cm of core are characterised by coarser, laminated deposits i n d i c a t i n g l o c a l i s e d inputs of coarser sediment. This i s believed to be a consequence of a c c e l e r a t e d erosion within the catchment, p r i m a r i l y as a consequence of logging road construction. This change i n the record can be used to provide a datum within the surface sediments and was used to e s t a b l i s h the f a c t that the rhythmites i n the upper p o r t i o n of the Kokwaskey Lake sediments are annual deposits i . e . varves. Information on the logging h i s t o r y , s p e c i f i c a l l y the date of road construction adjacent to each lake, was provided by B.C. Forest Products (Fletcher Challenge) and i s reported i n Table A l l . 5 . These are assumed to date the obvious recent change i n the records. Table All.5 Date of logging road construction past each lake Lake Date Kwoiek 1971 Kokwaskey 1974 John George 1977 Klept 1977 Kha 1979 210 APPENDIX III Examples of cumulative grain size curves SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t C M ) . 01 001 1.0 Rangai Lowar L i m i t - .001 mm. +10.00 phi. Upper L i m i t - .088 mm. +3.50 phi. Volght of glvan i n t e r v a l - 37.4 g. D r. m •0.5 £ 0.4 i a L U. Dominant el ave -Reoeeelve e l o v a -. 008 mm. .062 mm. Haani Madion Var1anco St. Dav. Skew Momant- +7.08 p h i . Graphic- +7.09 p h i . - +8.81 p h i . - 1.33 p h i * Moment- +1.24 ph i . Craphlo- - I . 17 phi. Moment- +1. IS Craphlo- -.35 Momant- 3. 40 Craphlo- 1.21 +7.00 ph +4.00 ph Pe r o e n t l l e e F i n e r O i l " Dl«.o > 0 a o -Daao -D73.0 -D H D -DflS.0 -Doo.0 -+8.55 phi. +8. 39 phi. +7.70 phi. +8. 81 phi. +8. 27 phi. +8.05 phi. +5.34 phi. +4. 48 phi. S l a v s S i z e <phl> SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t <mffl> .1 .01 .001 Rongei Lowar L i m i t - .001 mm. +10.50 p h i . Upper L i m i t - .354 mm. +1.50 p h i . Weight o f given I n t e r v a l - 19.7 g. "0.4 £ X r-L a c c • +1 u a L Dominant s l a v e -Receeelva e l e v e -.004 mm. .250 mm. Maam Madlon Variance St. Dev. Skew Momant- +7. 81 p h i . Graphic- +7.70 p h i . - +7.88 p h i . - 2.01 phi". Momant- +1. 42 p h i . Graphic- -1. 13 p h i . Moment- -. 27 Graph 1 o- -. 03 Momant- 4. 41 Graphic- 1. 41 +8.00 phi '2.00 phi P e r o e n t l l e e F i n e r • s.o -Din.0 -023.0 -D5O.0 -Dra.0 -D K D -Oca.0 -Doo.0 -+B. 97 phi. +9.84 p h i . +8.38 p h i . +7.88 phi. +8. 08 p h i . • f l . 58 phi. •S. 23 phi. +2. 88 p h i . S i ova S l z a <phl) SEDIMENTATION ANALYSISt S i z e D i s t r i b u t i o n P l o t Cmm) .01 1.0 I 1 • 1 • ' ' H L 0. B X O) •- 0. 7 -0 U0.3-• L o.o : ••3.5 Rangsi Lowor Unit- .001 an. +9. 50 phi. Upper Limit- . 08B no, +3.50 phi. Weight of glvan Interval- 17.4 g. c a x 0.4 i +1 0 I U. +5. 5 +B. 5 S I Q V Q S i z e Dominant alava -Raceaelvo «1ova-Meoni Madlan Variance St. Oav. Ska. Momant-Graph 1 c-Moment-Craphlc" Moment- + 1. 12 Graph lo- -.03 Moment- 3. 31 Graphic- 1.31 00S mm. -7.50 phi 082 mi. -4.00 phi •7. 30 phi. +7.27 phi. +7.25 phi. .81 phi*. +. 00 phi. -.87 phi. Paroontllao Finer D 3.0 ' Oia.o ' DUO ' OSLO 1 D 7 9 . 0 ' Dw.o ' OgaLo ' •oa.0 • +0.05 phi. +8. 15 phi. +7.75 phi. +7.25 phi. +8. 75 phi. +8. 41 phi. ••5.88 phi. •5. 57 phi. <phl> SEDIMENTATION A^JALYSISl S i z e D i s t r i b u t i o n P l o t (mm) .1 .01 . I i I I i i • ' i 1 i i ' i •5. 5 +8. S +7. S Rangei Loiar Llnlt- .004 mm. +B.00 phi. Upper Limit- .354 mm. +1.50 phi. •sight of glvan Interval" 16.3 g. •onlnant el eve -Raooaolve alave- . 008 nm. .250 no. Median Variance St. Oav. -* Kurtoale Moment- +8. 4C phi. Graphic- +8.22 phi. - +7.08 phi. - 2.28 phi* Moment- +1.51 phi. Graphlo- -1.60 phi. Moment- -1.02 Graphlo- +.80 Momant- 2.04 Graphlo- 1. 87 +7.50 phi .00 phi Paroantlie* F1 nar D 9.0 < OlB.0 < 029.0 < 090.0 * Omo ' Ou.o 009.0 ' Oea.0 < +7. 40 phi. +7.30 phi. +7.31 phi. +7.08 phi. •8.31 phi. +4. 18 phi. +2.04 phi. +2. 38 phi. S l a v e S i z e (phi) S E O I M K N T A T I O N ANALYST S i S I Z Q D i s t r i b u t i o n P l o t (mm) 1 .1 .01 I I I . 001 1.0 Rangei Lower L i m i t -Uppar L i n i t -Weight of glvan I n t e r v a l - 35.5 g. .001 mm. M O . 00 phi. 1.414 nm. -.50 p h i . a 0. 4 i Oomlnant s i e v e -Recessive e l eve-Median Variance St. Dev. Moment' C r a p h l c Moment-Graph 1c-Moment-Graphlc-Moment-Graphlc-PeroentIlea F1nor . 003 mm. +8. SO phi 1. 000 mm. +0. 00 phi •7.00 p h i . +0. 83 p h i . +7. SB p h i . 5.20 phi*. +2.2B p h i . -2.B7 p h i . -. 87 +.59 2. 19 1.05 D 3.0 -Dlfl.0 -023,0 -DJD.0 -OTJ.0 -DM.0 -009.0 -Doo.0 -+9. 33 p h i . +8.97 p h i . +8. 87 p h i . •7.88 p h i . +8.08 p h i . +3. 83 p h i . +2.88 p h i . +1. 42 p h i . S i e v e S i z e <phi> S E D I M E N T A T I O N A N A L Y S I S i S i z e D i s t r i b u t i o n P l o t (mn) 1 .1 .01 I n i i • i l i Lu 001 1.0 Rangei Lo«er L i m i t - .001 mm. +10.00 phi. Upper L i m i t - 1.414 mm. -.50 p h i . Weight of given i n t e r v a l - 14.4 g. Dominant s i e v e - . Recessive s i e v e - 1. Meani Median Varlanoe St. Dev. She« Momant-Oraphlc-Moment-Grophlo-Moment-Graphl c° Moment-Graphlo-P e r o e n t l l e s F i n e r 0 3.0 - *B. 84 p h i . +fl. 12 p h i . +8.80 p h i . +8.01 p h i . +5.08 p h i . +3. 82 p h i . +2. 83 p h i . •2.22 p h i . Oie.0 -023. o » 030.0 -Dn.0 -014.0 -DoS.0 -OSBLO « 003 mm. +8.50 phi 000 mm. +0.00 phi +7. 05 p h i . +8.98 phi. +8.01 p h i . 3.28 phi*. •2.29 p h i . -2.85 p h i . -. 43 +.58 I. 88 .75 SEDIMENTATION ANALYSIS! S l z a D i s t r i b u t i o n P l o t (mm) .01 .001 Rangei Lover L i m i t - .001 mm. +10.00 p h i . Upper L i m i t - . 088 mm. +3. SO ph i . Weight o f glvon I n t e r v a l - 20.1 g. Dominant s i e v e - .002 mm. Reoaeslvo a l a v e - .002 Meant Median Varlanoa St. Oav. Ska. Kurtoele Moment- +8.SB p h i . Graphic- +B. 50 p h i . - +8.88 p h i . - . 18 phi', Moment- +.42 ph i . Graphic- -. 34 p h i . Momant- +. 53 Graphic- *. 30 Moment- 1. 88 Graphlo- 1.2S +0.00 phi +4.00 phi Pa r o e n t l l e e F i n e r D 8.0 Dis.0 -D29.0 -DSLO -Dmo -DM .0 -Dos.0 -DOLO -'8.08 p h i . +8.00 p h i . +8.84 p h i . +8.88 p h i . +8.45 p h i . +8.22 p h i . +7.80 p h i . +7.48 p h i . S l a v e S i z e <pht) SEDIMENTATION ANAL.YSISi S i z e D i s t r i b u t i o n P l o t <mm3 .1 .01 1. 0 i 1 - 1 I * I I I i—' ' 1  001 1.0 Rangei Lover L i m i t - .001 mm. +10.00 p h i . Upper L i m i t - .354 mm. +1.50 ph i . Weight of glvan I n t e r v a l - 16.2 g. Dominant e l a v a Reoeealva e l a v a -.008 .250 Maam Median Varianoe St. Oav. Skew Kurtoele Moment- +S. 74 p h i . Graphlo- +5.81 p h i . - +6.75 p h i . - 3.79 phi*. Moment- +1. 05 p h i . Graphlo- -2.08 p h i . Momant- -. 01 Graphlo- +.87 Momant- 1. 55 Graphlo- . SS +7.50 phi +2.00 phi Pero e n t t l a a F i n e r 0 5.0 -Dio.0 -023.0 -OBO.0 -Dmo -0u.o -Oos.0 -Doe.0 -+7. B8 phi. +7.42 p h i . +7.30 phi. •8.75 p h i . +3.70 phi. +3.28 p h i . +2.83 p h i . +2. 10 phi. S i e v e S i z e <phl> SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t <•«> .1 . 01 . 0 0 1. 0 -T—'—-rr1 I ' l l ' I—I 1 1 L 1. 0 Rangei Lower U n i t - .001 nun. +10.00 phi. Upper L i m i t - .354 mm. +1.50 phi. ( e i g h t of given i n t e r v a l - IB. 1 g. • X c Oomlnant a leva Receaelva elevi Meant Median Variance St. Dev. Shew . 004 mm. . 250 mm. Moment- +8.82 phi Graphic- +6. 52 phi - +7.58 phi - 3. 14 phi Moment- +1.77 phi Graphlo- -1. Bl phi Momant- -. 85 Graphic- +. 84 Momant- 2.82 Graphlo- 1.70 +8. 00 ph +2.00 ph Peroant11aa F1nor D 9.0 - +8. 13 phi. +7.00 phi. +7.82 phi. +7.50 phi. +8.57 phi. +4. 08 phi. +2. 03 phi. +2.37 phi. Dia.0 •23.9 030.0 •19.0 DM.0 D03.0 Oua.0 S l a v s S i z e (phl> SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t CmnO 1 .1 .01 .001 X Ol - O. 7 TI a c a 0 . 5 +i 8 a 3 L U_ 0 . 2 -* 1 * • • * 1 K0K310 i n •rrfT - l . S +.5 +2. S +4. 5 S i e v e S i z e +8.5 Rangei Lower L i m i t - .001 nm. +10.50 p h i . Upper L i m i t - 2.828 nm. -1.50 p h i . Height o f given I n t e r v a l - 84. S g. Dominant el a v e - .123 nm. +3.00 phi Recaealva e l e v a - .022 am. +5.50 phi 0. 4 ^ •as a l, Median Variance St. Dev. Ku r t o e l a Moment- +2.82 p h i . Graphlo- +2.82 p h i . - +2.85 p h i . - 2.84 phi*. Moment- +1.82 p h i . Graphlo- -1.01 p h i . Moment- +2.92 Graphlo- +.05 Momant- 0. IS Graphlo- 1. B4 Pa r o a n t l l o e F i n e r 0 9.0 1 Dio.0 1 Dzs.0 ' D90.0 1 On.0 1 004.0 1 Doa.0 1OOOLO 1+8.07 p h i . +3.61 p h i . +3.28 p h i . +2.85 p h i . +1.01 p h i . +1.5B p h i . +.01 p h i . +. 10 p h i . <phl> 2.15 SEOIMENTATION ANALYSIStSize D i s t r i b u t i o n P l o t (mm) 1 .1 .01 .001 J^_|—I 1 hill l-i^ J < U-i i I I i—i 1 I I I I I I i—i 1 1- i.Q D V c a 0.5 -*) o c o u 1)0.3-• L K W K 1 4 1 0 0 Rangei Lower Limit- .001 mm. +10.00 phi. Upper Limit- 5.657 mm. -2.50 phi. Weight of given Interval- 52.5 g. c • r. 0.4 £ 0 0 L Dominant eleve -Receeelve elove-Meam Median Variance St. Dev. Skew Moment-Graph 1 c-Moment-Grophtc-Moment-GraphIc-Moment-Graphlc-Poroantllee Finer 0 s.0 - +8.88 phi. +7.38 phi. +6.25 phi. +3.53 phi. +2.88 phi. +2.88 phi. +2.33 phi. + 1.72 phi. 125 mm. +3.00 phi 000 mm. -2.00 phi +4. 48 phi. +4.52 phi. +3.53 phi. 4.88 phi". +2. 18 phi. -2.34 phi. +1.21 -.84 2.89 .77 Dift.o -OJ5.0 -DSHD -Dn.o -DS4.0 -DfJ.0 -Dfle.0 -•1.5 -3. S S i e v e S i z e (phi) SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t ^ . 01 001 1.0 Rangei Lower Limit- .001 mm. +10.00 phi. Upper Limit- .088 mm. +3.50 phi. Weight of given Interval- 31.5 g. +7.50 phi +4.00 phi -*) •0. 7 rD) Domi riant elova • a RecoootvQ eiova-HBoni Moment" -0. B X A Graphic" Median C Var 1 ancti -•0.5 I St. Dav. Homent" Graph!o* L SKe« Moment" a r Graphic" -0. 4 on Fll Kurtoei* Momont" Graphic" •0 . 3 *J 0 • Paroentllaa Finer L D 9.0 - +8.99 phi. -0.2 •i«.o - +0. 19 phi. +7.31 phi. .73 phi*. +. 88 phi. -. 78 phi. -1. 32 -. 14 3.84 1. 13 OJS.0 -050,0 -Dre.0 -Du.0 -On.0 -Doo.0 -+7.88 phi. +7.31 phi. +8.83 phi. +8. 84 phi. +8. 14 phi. +5.88 phi. SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t (mm) .1 .01 .001 Rangei Lo«er L i m i t - .001 nm. +10.00 phi. Upper L i m i t - .707 no. +.50 phi. Weight of given I n t e r v a l - 10. B g. +> u • L U. Dominant e l a v a -Reaeoelve «1eve-Median Variance St. Dev. Skew Kur t o e l e Monent-Graphlc-Moment-Graphlo-Monant-Craphio-Manent-.Graphlc-003 044 +7.25 phi +0.84 phi +8. 08 phi 4. 43 phi +2. 10 phi -2. 48 phi -.83 +.74 2.59 2.03 8. SO phi + 4.50 phi P e r o e n t l l e e F i n e r 0 9.0 Dle.o •25.0 ' 090.0 • n o 0u.i Oos.0 Ooa.0 < +9.87 p h i . +8.69 p h i . +8. 42 p h i . +8. 06 p h i . +7.08 p h i . +3.78 phi. +2.93 phi. +2.25 p h i . +3. S +4. S +5. S S i e v e S i z e SEDIMENTATION ANALYSISi S i z e D i s t r i b u t i o n P l o t 1 . 1 . 0 1 Rangei Lower L i m i t - .003 mm. +8.50 p h i . Upper L i m i t - 1.414 mm. -.50 phi. Weight o f given i n t e r v a l - 45 g. Dominant e l a v a Raoaselva e l a v a -Maant Madlan Variance S U Oav. Skew Ku r t o e l e .011 mn. .044 nm. Moment- +4. 53 phi Graphlo- +4. 37 p h i . - »3. 74 phi. - 5.02 phi*. Moment- +2.24 phi. Graphlo- -2.29 phi. Momant- +. 34 Craphlo- -.41 Momant- 1.71 Graphlo- . 74 +6.50 phi •4.50 phi P a r o e n t l l a e F l n a r D 9.0 Oia.0 -D29.0 -Dsao -Omo -DM. a -DOSLO -Owo.0 -+7.87 p h i . +6. 08 p h i . +0. 42 p h i . +3.74 p h i . +2.71 p h i . +2. 40 p h i . +1. 18 p h i . *. 13 p h i . S i e v e S i z e <phl> 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100482/manifest

Comment

Related Items